cursodearameo.com.ar - aramaic studies volume 3

3.1 (2005) 3-14]DOI: 10.117/1477835105053512

THE ROLE OF ABRAHAM IN TARGUM ISAIAH*

Alberdina Houtman

Theological University Kampen

Abraham is one of the most important representations of Israel.^ It istherefore rather surprising that in the biblical prophetic literature hedoes not play a very important part.^ Only seven times is his namementioned, of which four are in the book of Isaiah, i.e. 29.22, 41.8,51.2, and 63.16.^ Apart from these four cases, there are no less thaneight extra references to Abraham in the base text of TgJon Isaiah, i.e.5.1, 29.23, 41.2, 43.12, 46.11, 48.15, 48.16, 51.2." Moreover, there arereferences to Abraham in the Tosefta Targums to 10.32, 33.7, 41.2 and

*The investigations were supported by the Foundation for Research in the Fieldof Humanities, which is subsidized by the Netherlands Organization for ScientificResearch. I thank my colleagues Harry Sysling and Jan Wim Wesselius for someuseful remarks.

The biblical text used in this article is MS Leningrad according to the Bar IlanResonsa Project version 9. The Aramaic text is MS 2211 according to the editionof A. Sperber.

^ J. Ribera Florit, 'Conceptos a traves de los cuales se desarroUa la hermeneuticadel Targum de Ezequiel', in: J. Targarona Borras & A. Saenz-Badillos (eds.), JewishStudies at the Turn of the Twentieth Century, I. Biblical, Rabbinical, and MedievalStudies (Leiden [etc.]: E.J. Brill, 1999), pp. 188-98, at p. 190.

^I restrict the term prophetic literature here to the Latter Prophets.According to John Van Seters, Abraham only started to play a role in the sacredhistory of Israel in the exilic period. Until that period, the emphasis was on theExodus tradition. Whereas Jeremiah and Ezekiel still held to the importance of theExodus tradition as the theme of divine election, with Deutero-Isaiah the attentionshifted towards the covenant with Abraham. See J. Van Seters, Abraham in Historyand Tradition (New Haven [etc.]: Yale University Press, 1975), pp. 264-65. Thesingle occurrence of Abraham in Primo-Isaiah is generally regarded as a gloss.

^The other occurrences are Jer. 33.66, Ezek. 33.24, and Mic. 7.20.'*J.C. de Moor, A Bilingual Concordance to the Targum of the Prophets, IX.

Isaiah (3 vols.; Leiden [etc.]: E.J. Brill, 2002), I, pp. 5-6. The other three casesmentioned there, 10.32 and two occurrences in 33.7, do not belong to the base

© SAGE Publications (London, Thousand Oaks CA, and New Delhi), 2005

4 Aramaic Studies 3.1 (2005)

66.10. Obviously, the meturgemans recognized the developing importof the patriarch for the sacred history of Israel. In what follows, wewill investigate why the meturgemans inserted references to Abrahamwhere his name did not occur in the biblical text, and in what roles theycast him.^ We differentiate between the base text of Targum Jonathanand the Tosefta Targums, because we may assume that the base text ofTargum Jonathan is the product of an editorial process and thereforemore consistent in its theological outlook. The Tosefta Targums onthe other hand may be either remnants of a more ornate PalestinianTargum tradition or the product of an ongoing living Targum traditionand may therefore represent different views.

The calling of Abraham

Let us start with the beginning of the Abraham narrative. According toGenesis 12, God revealed Himself to Abram, summoning him with thewords 'Go forth from your native land and from your father's house,to the land that I will show you'. This migration would become thebeginning of the sacred history of the people of Israel within their land.The meturgeman saw a reference to this event in the first part of Isa.41.2, where the Hebrew text reads: tinb ini^-^p' p'K mrao Tm 'Q.' Thewords TUn 'roused' and insn|T 'He called him' in combination withthe word group miDD, point to someone being called from the east.The word pi'H 'righteousness' is interpreted by the meturgeman as areference to the pn^, 'the righteous one' par excellence, which is Abra-ham.* In this way the meturgeman arrived at the translation Ti'N ]D

text of Targum Jonathan but to the corpus of Tosefta Targums; J.B. Van Zijl,A Concordance to the Targum of Isaiah (Aramaic Studies, 3; Missoula: ScholarsPress 1979), p. 2, also includes 10.32. The inclusion of 10.32 by both De Moor andVan Zijl, is due to the edition of Sperber on which both concordances are based.Sperber presented the elaborate version as if it were the base text, whereas in factit is a marginal reading.

^R. Kasher, Targumic Toseftot to the Prophets (in Hebrew; Sources for the Studyof Jewish Culture, 2; Jerusalem: World Union of Jewish Studies, 1996), passim.

®See also B.D. Chilton, The Clory of Israel: The Theology and Provenience ofthe Isaiah Targum (JSOT supplement series, 23; Sheffield: JSOT Press, 1983), pp.46-48.

''This verse is generally interpreted as a reference to (I!yrus. However, becausethe rabbis were hesitant to praise people who were suspect in their eyes, they werehappy to reinterpret the verse. See I. Heinemann, mJSn "311 (Jerusalem: Magnes,1953/54), pp. 154-55.

®See e.g. Cant. R. 3.5; 4.5 'Abraham was the head of all the righteous'.

HouTMAN Abraham in Targum Isaiah

nn-as xnjion '':':3 'Who brought Abra-ham openly from the East,^ the chosen one in righteousness truly?^"He brought him to his place'. According to this interpretation, Godselected Abraham to serve Him in the land that He would give him.

The second part of Isa. 41.2 tells how God delivered up nationsto him and trampled sovereigns under foot. Once the identificationof Abraham was made, it was most obvious that this reference to asuccessful war concerned the conflict with the kings of Genesis 14. Thiselement is worked ont as follows in a Tosefta Targum to Isa. 41.2.^^

The governors and the kings gathered to wage war against Abraham thechosen. And they fell before him and he divided himself against them bynight. And he killed four of their kings and they were eleven camps. AndAbraham the righteous thought and said in his heart: All the peoplesof the world will say "he captured the kingdoms", that my house wasfilled from their captives. A vow in the name of the Most High, that Iwill not take from their captives one shoestring, except what the youthshave eaten. -^ In truth and in truthfulness [...] And I tried him andsanctified him with ten trials. And he did not put off the matter. AndI told him: don't fear from your enemies: I will be with you and I willincrease you and multiply you and I will receive your prayer and all yoursons will be numerous as the stars of heaven and as the sand of the seaand as the dust of the land. And he believed the Memra of the LORD aswas written: 'Who brought Abraham openly from the East, the chosenrighteous truly? He brought him to his place' etc.

It can hardly be a coincidence that this verse of Isaiah is part of thehaftarah reading to parashat lekh lekha which deals with both thecalling of Abraham and his war against the kings.

Returning to the election of Abraham from all the people in theworld, it seems that his qualities were so outstanding that the choicewas not difficult. In Isa. 51.2ba the meturgeman translated the Hebrewtext rnsnp nnx"'D as •'Dn' s'? rr'nTip Ka'pun 'xn^n' Dmnx mn in n s 'For

same identification is made in e.g. Midr. Ps. 77 and Yal. S. Isa. §447.Moreover, in TgNeof Lev. 22.27 Abraham is referred to as 'the man from the East'.

'"The first and second Rabbinic Bibles, the Antwerp Polyglot and Cod. Urbinas1 read K'p""i;i 'the righteous' instead of Kp"I 'righteousness. This would producethe translation 'truly the most righteous'.

'^Genizah fragment, Jerusalem 557.4^ 4°, 6b. See R. Kasher, Targumic Toseftot,pp. 159-60.

14:23


Abraham was only one, unique in the world/'' I brought him near toMy worship'. Just as in Isa. 41.2, the concept of calling is interpreted as'bringing near'. Apparently, God's call is so irresistible, that 'to hearis to go'. The link between the word iriK and Abraham is based onEzek. 33.24aP where it is stated DiTiDX HT! "iriK. By way of analogy, theword nnx is also here connected with Abraham. The last expression,'I brought him near to My worship', occurs also in the Targum to Isa.48:16, where in addition the time and the reason for the mission arespecified: ^^nbth HTiTip Dmnx jonQ 'n'^mo s^anu itDisnKi piia 'From thetime that the nations departed from My fear, from there was Abraham.I brought him near to My worship'.

Covenant

Another important episode in the biblical story of Abraham occurs inGenesis 15, where we find the story of the so-called 'covenant betweenthe pieces'. In this intriguing story, God confirmed the assignment ofthe Land of Israel to Abraham and his offspring in a covenant. At thesame occasion, God foretold to Abraham the slavery in Egypt. Themeturgeman found two places in the biblical text of Isaiah that in hisview referred to this story, i.e. the beginning of Isa. 43.12 and Isa.48.15. In both cases, it concerned a cryptic Hebrew text. Let us seehow he worked this out.

MT:

TgJon: n'p-IS N]N 'n'O'? TH^Jl XQ

MT: - - . . . . _ -TgJon: x'inn j'D n'b n'Q^pT KQD Dn:iaa jiDn'

I foretold to Abraham your father what would come in the future and Idelivered you from Egypt, just as I swore to him between the pieces.

same expression 'NTIT' Drnns mn in occurs also in Ezek. 33.24 and Mai.2.15. It may be that the meturgeman felt the need to translate the Hebrew word insby means of two different words to indicate the different connotations of the word,namely 'one in number' (from which a large offspring would arise), and 'singular,unique' (in righteousness). In rabbinic literature the word TFI" means someone whois devoted to a particular scrupulous life. For instance in b. Ta'an. 10b the question'who is called a yahid' is answered by 'whoever is worthy lo be appointed managerof a community'.

HOUTMAN Abraham in Targum Isaiah 7

In the view of the meturgeman the cryptic formulation of the Hebrewtext was a m^p ~\~n, a 'short path'/' ' which allowed him to add theinformation that was necessary for the right understanding of the text.The combination of the words 'm^n 'I foretold' and 'nuBim 'I delivered'offered him a sound opportunity to connect the text with Genesis 15. ^The second reference to the covenant according to the meturgemanoccurs in Isa. 48.15

MT:TgJon: jiDiDX cniDX UV D'p nntj

MT: lDTITgJon: n-mis

I, I by my Memra decreed a covenant with Abraham your father, andalso elevated him; I have brought him up to the land of the House of MyShekhinah and I caused to prosper his way.

Here, it is obviously the word vnsip that allowed a connection with theAbraham story. Because in the biblical text of Isa. 51.2 the word combi-nation vnsip occurs with explicit reference to Abraham, the meturge-man was justified by way of analogy in interpreting the same expressionhere also as referring to Abraham. The Hebrew text says that God hasspoken, Tiim, and since God's word is rehable and irrevocable, it hasthe force of a covenant.

Abraham as a model for Israel

In the course of the tradition, Abraham was developed as a corporatepersonality. What was said about Abraham, implicitly also concernedIsrael.^^ This process had its roots in the biblical text. In Isa. 41.8,amongst others, it says 'But you, Israel, My servant, Jacob, whom I

^''Nr. 9 of the hermeneutic rules of Eliezer ben Yose ha-Gelih. See e.g. D.I. Brewer,Techniques and Assumptions in Jewish Exegesis before 70 CE (Texte und Studienzum Antiken Judentum, 30; Tiibingen: J.C.B. Mohr), p. 229.

^^The explicit reference to the 'covenant between the pieces' also occurs in TgJonMic. 7.20 and TgJon Ezek. 16.3.

'•'B. Ego, 'Abraham als Urbild der Toratreue Israels: TraditionsgeschichtlicheUberlegimgen zu einem Aspekt des biblischen Abrahamsbildes', in F. Avemarie&; H. Lichtenberger (eds.). Bund und Tora: Zur theologischen Begriffsgeschichtein alttestamentlicher, fruhjiidischer und urchristlicher Tradition (WissenschaftlicheUntersuchungen zum Neuen Testament, 92; Tubingen: J.C.B. Mohr, 1996), pp. 25-40, at p. 35.


have chosen, seed of Abraham My friend'.^^ This text is translated lit-erally in the Targum.^* The idea that the election of Israel is the resultof the special relationship between God and Abraham, as propagatedin this text, is transposed to other parts of Targum Isaiah. In the inter-pretation of the famous song of the vineyard of Isaiah 5, for example,the meturgeman compared the people of Israel to a vineyard that hasgrown out from the seed of God's friend Abraham. In verse 1, it says 'Iwill sing now for Israel-which is like a vineyard, the seed of Abraham,My friend'.

In Isa. 51.1-2 the people of Israel are addressed as pursuers of jus-tice, and Abraham is held up as an example to them. The Targumremoves any doubt as to the meaning. Let us compare the two ver-sions.

MT: iQ'3n mn' - 'cpna pi:i

TgJon: I' DnOX 'V Dip ]D jS' X p^l KQCip '3T1

MT: nmp] - -nn nnpo 'PNI uraiin

MT: - i n x o DD' 'pinn m o b^i

TgJon: mn i n '-ix jOHKn^jn n-iB3i

MT: inmxi lnDinxT - vnxip

TgJon: n-iTJOxi nTonDT 'jn'ps'p n r a n p

51.1 Listen to My Memra, you who pursue truth, you who seektion from before the LORD: Consider that as hewn stone from the rockyou were hewn, and as the chip from an empty pit you were carved.51.2 Look at Abraham your father and at Sarah who brought you forth.For Abraham was only one, unique in the world, I brought him near toMy worship, and I blessed him and made him many.

The parallelism between the two verses makes clear that Abraham isthe rock from which the people of Israel were hewn, and Sarah is thepit out of which they came forth.^^ In this way the listeners learn that,

"See also II Chron. 20.7b and Ps. 105:6.^^The Hebrew 'DHK, an active participle with object suffix, is translated accord-

ingly as 'Dm. Both forms are often rendered in English as 'My friend'. See, however,the discussion by M. Goshen-Gottstein 'Abraham-Lover or Beloved of God', in J.H.Marks & R.M. Good (eds.). Love and Death in the Ancient near East: Essays inHonor of Marvin H. Pope (Guilford: Four Quarters Pub. Co, 1987), pp. 101-104.

^^The pit is nsed as a symbol for the womb (see Radak's commentary on thisverse). In b. Yeb. 64b this verse is connected with Sarah's barrenness. Obviously

HouTMAN Abraham in Targum Isaiah 9

being considered a 'chip off the old block', they are supposed to followthe example of their worthy ancestors.

There is yet another aspect in which Abraham served as a model forthe people of Israel, namely in relation to the exile. Just as Abrahamwas called from the east, so centuries later, the people of Israel werecalled to return from the Babylonian exile, back to the Promised Land.This correspondence is pointed out in the Targum to Isa. 46.11a.

MT: £D':J - - niTQD - NipTgJon: b'bp fjiiJD 'nbn nsn'N':' umiaa

MT: -TgJon: '-i'n3 Dm^K 'jn Kp'm :;-i!<Q

46.11 Saying to gather the exiles from the East to bring in openly as aswift bird from a distant land the sons of Abraham, My chosen.

Obviously, the expression Ti:iiJ 2?'X 'man of My counsel' in the Hebrewtext occasioned here the introduction of Abraham, an identificationthat is also known from midrashic literature.^°

In times of oppression, Abraham could serve as a sign of hope.In Isa. 29.22, God promised Jacob deliverance from his distress. TheHebrew text can be translated as 'Assuredly, thus said the LORD tothe House of Jacob, Who redeemed Abraham: No more shall Jacobbe shamed, no longer his face grow pale'. The designation of God asthe Redeemer of Abraham seems to imply that He saved Abrahamfrom some unspecified danger. According to an ancient and widespreadtradition, Abraham was condemned by Nimrod to be killed by fire, butwas miraculously dehvered by God. ^ This tradition aippears to be builton popular etymology, interpreting "ilX as 'fire' in Gen. 15.7, where God

the meturgeman wanted to express the same notion by the addition of the word1P'"1 'empty'.

^°E.g. Gen. R. 54.1 and Ag. Ber. chapters 21 and 81. In b. 'Abod. Zar. 19a thereis a slightly different version, in which Abraham is identified with the 'man whohas not followed the counsel of the wicked' of Ps. 1.

2'Gen. R. 38.13, 44.13; b. Pes. 118a, b. 'Erub. 53a; Ps.-Philo, LAB 6.14-18. Thelegend also found its way into the Christian church and Islam. In addition to thewritten sources, the tradition occurs in religious art. See J. Gutmann, 'Abraham inthe Fire of the Chaldeans: a Jewish Legend in Jewish, Christian and Islamic Art',Fruhmittelalterliche Studien: Jahrbuch des Instituts fiir Fruhmittelalterforschungder Universitat Munster 7 (1973), pp. 342-52; K. Appel, 'Abraham als DreijahrigerKnabe im Feuerofen des Nimrod', Kairos 25.1-2 (1983), pp. 36-40. See also G.Vermes, Scripture and Tradition in Judaism (Studia Post-Biblica, 4; Leiden: E.J.Brill, 1961), pp. 85-90.


announced Himself as 'I am the LORD who brought you out from Urof the Chaldeans'.• • This hiterpretation occurs in all the PalestinianTargums to Gen. 15.7. ^ Within a Tosefta Targum to Isa. 10.32 wefind a short reference to this tradition. This Tosefta, that describesthe push of Sennacherib to Jerusalem, connects this occurrence withthe hostile treatment that befell Abraham in the past. It says 'And sothey came to Abraham our father, when they threw him in the midstof the burning furnace'. The legend of Abraham in the furnace and hismiraculous rescue is used here as an illustration from history.

One final aspect of Abraham as a role model should be mentioned.Both in Judaism and in Christianity, Abraham is put forward as amodel of faith. This is based on Gen. 15:6, where it is stated 'Andbecause he put his trust in the LORD, He reckoned it to his merit'. 'This theme does not occur in the base text of Targum Isaiah, butthe difficult Hebrew of Isa. 33.7 brought about a Tosefta Targum inwhich the topic is elaborated.^^ The Hebrew text of this verse reads]VDT ~iQ nt>^ 'Duba n:in lp:;: D'PKIK ]n. This might be translated as'Behold, their valiant ones cry without; the ambassadors of peace weepbitterly' (ASV). Because of the double meaning of "[SbQ, which mayrefer to either a human messenger or an angel, the meturgeman feltfree to associate both the problematic DbxiK and the n^b^\D ^DKba withthe angelic world.• ^ This turns the verse into a depiction of angels indistress. In consequence, the meturgeman had to look for a reason fortheir distress and found it in the planned sacrifice of Isaac. He renderedthe verse accordingly as follows.

Behold, when I revealed Myself to Abraham their father and said thatI would give him Isaac, he believed My word. And after that, at thesecond time, when I told him to offer him up as a burnt offering, hedid not tarry, but went and built an altar on Mount Moriah and offeredhim up as a burnt offering. Then rose all the heavenly angels, angels

22l. Heinemann, m:sn 'Dm, p. 18.^''Moreover there are references to it in Ps.Jon. and TgNeof. to Gen. 11.28, 11.31,

and 16.5 (also FragTgs). See C.T.R. Hayward, 'Inconsistencies and Contradictionsin Targum Pseudo-Jonathan; The Case of Eliezer and Nimrod', JSS 37.1 (1992),pp. 31-55, at pp. 49-50.

• ''In the NT this verse is cited or alluded to in Rom. 4.3; Gal. 3.6; James 2.23.^^See J. Stenning, The Targum of Isaiah, pp. 225-226; R. Le Deaut, Le nuit

pascale (Rome; Institut Biblique Pontifical, 1963), pp. 177-78; R. Kasher, TargumicToseftot, pp. 158-59.

• See R. Kasher, 'Angelology and the Supernal Worlds in the Aramaic Targumsto the Prophets', JSJ 27 (1996), pp. 168-91, esp. pp. 176-79.

HoUTMAN Abraham in Targum Isaiah 11

of fear, angels of trembling, and cried outside their enclosure, saying:Lord of the world, is this not the righteous Abraham, because of whoserighteousness the world was founded? And whom you gave a son aftera hundred years and whom you told: in Isaac I will call you sons? Andeven the angels of peace who stand in the camp of the Shekhinah wereweeping bitterly, until My mercy turned over and I spared him.

The element of faith also occurs in the Tosefta Targum to Isa. 41.2cited above in the section on the calling of Abraham. In this Tosefta itis said about Abraham, after God promised him numerous offspring,that 'he believed the Memra of the Lord'. It is remarkable that thisaspect of faith is not reflected in the base text of Targum Jonathanbut occurs only in these two Tosefta Targums.

The exclamation of the angels, that because of Abraham the worldwas founded, brings us to the last section of this paper.

The Merit of the Fathers

The so-called doctrine of merits is an old and widespread conceptwithin the Jewish religion. ' According to this idea, righteous menand women obtain merits during their lifetime. These merits do notonly benefit themselves, but also their fellow-creatures and even fol-lowing generations. Within this group of the righteous, the patriarchstake pride of place, with Abraham as the greatest among them. ® Abra-ham's righteousness is considered to exert a beneficial influence on thefate of the people of Israel and even of the world at large, as we saw inthe Tosefta Targum in the preceding section.

General references to the merits of the fathers can be found inTgJon Isa. 43.7, 62.6, and 64.4. ^ Apart from these general references,

^''See E.g. S. Schechter, Some Aspects of Rabbinic Theology (New York; Schocken,1909), pp. 170-98; A. Marmorstein, The Doctrine of Merits in Old Rabbinic Litera-ture (Publication, 7; London; Jews' College, 1920); E.E. Urbach, The Sages: TheirConcepts and Beliefs, I. (Jerusalem; At the Magnes Press, ^1979; reprint 1987),pp. 483-511; Encyclopaedia Judaica, vol. 16, cols. 976-978, s.v. Zekhut Avot; M.Remaud, A cause des peres: le "merite des peres" dans la tradition juive (Louvain;Peeters, 1997).

• Of Abraham it is said for instance that he was tested with ten trials. Each ofthe trials, which he withstood, resulted in some miracle for his descendents. SeeARN 33.

• 43.7 All this because of your fathers, upon whom. My name is called, whomI have created for My glory. I prepared their exiles and also worked miracles forthem.

TgJon Isaiah also explicitly mentions the blessed role of Abraham. In29.23 it says:

For when he^" sees the prodigies that I will work^^ for his sons, thebenefits of Abraham for his offspring after him upon their land, in theirmidst, they will sanctify my name and say, the Holy One of Jacob isHoly and about the God of Israel they will say. He is strong.

The insertion of Abraham into this verse is obviously triggered by thepreceding verse, where God is mentioned as the one who redeemedAbraham. ^

Gould it be that this tendency in the exegesis of TgJon Isaiah meansthat the meturgemans spoke with one voice on this question, or shouldwe perhaps look for signs that point in a different direction? Accord-ingly, we shall turn our attention to the biblical text of Isa. 63.16 andits targumic rendering.

MT: - - ij' N - - - - nns o

TgJon: ]']3 bs 3X0 ]'X'3O iit>V "[Onm XIH nx '"IX

MT: X7 7K"IS'1 - 13UT' X7 DmnX 'DTgJon: x"? "pxnej'i D'"iiiQa xjpox Kb Dmnx '"ix

MT: - - mn' nnx - . . .

TgJon: i<:bv -[nnm T nx x iman ions XJ*?

62.6 Truly, the deeds of your righteous fathers, city of Jerusalem, are set andpreserved before Me every day and every night continually, they do not cease.The remembrance of your good deeds is spoken of before the Lord unceasingly.64.4 Met before You are the deeds of our righteous fathers, who rejoiced doing yourwill in justice and in righteousness; in the way of your goodness and your mercythey were mindful of Your fear. Truly, every time there was anger from, before Youon us because we sinned, by them, by the deeds of our righteous fathers from of oldwe were saved.Besides these references in Targum Jonathan, there is a, passing remark to the'deeds of Abraham, Isaac and Jacob, Moses and Aaron' in a Tosefta Targum toIsa. 66.10.

^"The singular 'he' is difficult to understand, because it is not clear to whomit refers. Moreover, it makes the verse as a whole incoherent. However, althougha plural would have been more logical, all the witnesses have the reading 'rmnnninstead of ]in'tnQD. Chilton translates the singular as it is. Stenning translates a plu-ral, mentioning in brackets that literally it should be a singular. Ribera translatesit as a plural without pointing out the problem.

•'^According to the variant reading TDiJXT in the printed editions and MS Or1474.

above, p. 9.

HOUTMAN Abraham in Targum Isaiah 13

M T : -[QS D'PIIJQ 1J'?X:I - -

TgJon: -\m KD'PIJD XDpHD ]']3 ':5:J y

The biblical text can be translated as 'Surely You are our Father:though Abraham regard us not and Israel recognize us not, You, OLORD, are our Father'.

At first sight, this seems to place Abraham and Israel in a negativelight. The meturgeman, however, saw this differently. In his view, theverse is meant rather as homage to God's unrivalled compassion, thanas an indictment of the fathers. He translates:

63.16 Surely, You are the One whose compassion for us is greater than[that of] a father for sons. For Abraham did not bring us up from Egypt,and Israel did not work wonders for us in the desert. You, O Lord, arethe One whose compassion for us is greater than [that of] a father forsons; 'our Redeemer from everlasting' is Your name.

In targumic literature, the idea of God as a father is quite popular,especially when it concerns supplication.^^ In many cases, the Targumuses the expression 'Father' even when the Biblical text does not. How-ever, where the Bible does use the expression, as in the present case,the meturgeman is inclined to stress the allegorical meaning.^^

The targumic version of the text does not diminish Abraham's im-portance, but rather puts it into perspective by comparing it to that ofGod. Abraham is still depicted as a father, but it is made clear that itis God Himself, in His abundant compassion, who brought His peopleout of Egypt.^^ As it is stated in Deut. 4.37 'And because He lovedyour fathers. He chose their offspring after them; He Himself, in Hisgreat might, led you out of Egypt'. It is God personally who led thepeople of Israel out of Egypt. The motivation for this act, however,was a consequence of His love for the fathers. It may be that the tar-gumic version of Isa. 63:16 is meant as a warning against improper heroworship, but it is certainly not a refutation of the doctrine of merits.

M. McNamara, Targum and Testament. Aramaic Paraphrases of the He-brew Bible: A Light on the New Testament (Shannon: Irish University Press, 1972),pp. 115-19.

^''E. Levine, The Aramaic Version of the Bible: Contents and Context (BZAW,174; Berlin [etc.]: De Gruyter), p. 50.

^^See W.A.M. Beuken, 'Abraham meet van ons niet' Jesaja 63.16 (oration uponhis accession to office as Professor of the Exegesis of the Old Testament at theCathohc University of Nijmegen; Nijkerk: G.F. Callenbach b.v., 1986), p. 5.


Summary and conclusions

In Targum Jonathan, Abraham is put forward as a comforting exampleof God's acts in history as well as a shining example to be followed.

God's calling of Abraham and the subsequent covenant are stressedbecause of the relevance for his descendents, the people of Israel. Godbrought Abraham to the Promised Land so that he and his descendentsmight worship there the one living God.

In times of distress, the remembrance of God's loving attitude to-wards Abraham in the past could be used as a comforting assurancethat He would certainly also take care of his descendents.

The idea of the beneficial influence of the merit of the fathers isindeed endorsed by the meturgemans of Isaiah, though they seem tobe conscious of the related danger of hero worship.

In two Tosefta Targums Abraham's faith is highlighted, an elementthat is remarkably absent in the base text of Targum Jonathan.

In conclusion, we can say that the meturgemans of Isaiah, in spiteof some differences, generally agreed on the role of Abraham as one ofthe most important representations of Israel, and used different aspectsof his life as a didactic instrument to instruct and encourage theiraudience.

3.1 (2005) 15-42]DOI 10.117/1477835105053513

IN THE TARGUM OF THE PROPHETS*

G.W. Lorein

Leuven

Introduction

In his article 'OfFenbaren wird sich das Reich Gottes' K. Koch startswith the statement that God's Kingdom is the dominating conceptof the New Testament and that the Semitic roots for this conceptare hmited, except in the Targum (Tg) of the Prophets, especially theprophetae posteriores. As this concept is not demanded by their sourcetext, but is the proper initiative of the meturgemanim, he states thatthe revelation of God's Kingdom is typical of the theology of TgJon andthat, whenever the word xmn'pa occurs, we are to think of this topic.This ought to be studied again. A second point that needs an investiga-tion is Koch's statement that 'Reich' is the best translation for xniD'pa;but can we bring this problem resolution by providing a 'best transla-tion' ? And would it be the most frequent appropriate translation?^

"Research for this publication included a five month stay with the team prepar-ing the Bilingual Concordance to the Targum of the Prophets at the Kampen The-ological University in 1998. This was made possible by a research grant of the'Nederlandse Organisatie voor Wetenschappelijk Onderzoek (NWO)'.

Special thanks are due to Professor J.C. de Moor, with whom I discussed theframework of this article, and to Dr. E. van Staalduine-Sulman and Dr. W.F.Smelik, with whom I discussed some of the specific problems it addresses.

Additional Abbreviations:KTg: Targum according to Kimchi;PalTg: Palestinian Targum (according to MS Reuchlin);BCTP: J.C. de Moor (ed.), A Bilingual Concordance to the Targum of the

Prophets (Leiden: E.J. Brill, 1995-);TAB: M. McNamara (ed.), The Aramaic Bible (Wilmington: Glazier; Edinburgh:

Clark, 1987-);'o'pa: (one of) the four derivatives of ""7(2 in Hebrew: riD'bD, rmbD, HD' GD, mD' DD^K. Koch, 'Offenbaren wird sich das Reich Gottes', NTS 25 (1979), pp. 158-65

(158-59).

© 2005 SAGE Publications (London, Thousand Oaks CA, and New Delhi)


I do not want to doubt what Koch is actually writing in the irest ofhis article, but in this article I want to work in the other direction: howdoes the Targum use the word smD' Q? Does it imjDort its own theologywhen it uses this word? Or: what is the relation t;o its Masoretic; Text(MT)? In order to test the correctness of Koch's proposition, we mustsurvey all the occurrences of the word NmD'PQ, and not only those textsthat attract our attention by their rich theology and their differencesfrom MT. Only then will we be able to know whether TgJon had specificideas in mind when he used this word. This task has become mucheasier since the publication of the Bilingual Concordance to the Targumof the Prophets and Kasher's collection of Toseftot for the same partof Scripture.^

Although there seems to be a major difference between TargumJonathan and other types of Targum to the Prophets (Palestinian,Toseftot, those mentioned only by Kimchi), I have decided to studyall of the available Targumic material. Indeed, the 'other' Targums,being so few, could be studied only with difficulty in their own right.Nevertheless, we will always consider Targum Jonathan as the standardtype and the other Targums as additional evidence.

Fortunately, KmD' a is the only Aramaic word to be treated,'' where-as there are four in the Hebrew MT: noi' Q,'' niD' Q, HDQQ and m 'pno.Unfortunately, even this one Aramaic word occurs about 327 times in

Cf. B.D. Chilton, 'Regnum Dei Deus Est', SJT 31 (1978), pp. 261-70, and 0 .Camponovo, Konigtum, Konigsherrschaft und Reich Gottes in den fruhjiidischcnSchriften (OBO, 58; Freiburg: Universitatsverlag, 1984), pp. 417-32.

^At the time of writing, many of the texts concerned existed in a digital formin Kampen, but the Minor Prophets, the variants of Isaiah, Ezekiel and the MinorProphets (according to A. Sperber, The Bible in Aramaic based on Old Manuscriptsand Printed Texts, III. The Latter Prophets according to Targum, Jonathan [Leiden:E.J. Brill, 1962]) and the Toseftot collected by R. Kasher, Targumic Toseftot to theProphets (Jerusalem: World Union of Jewish Studies, 1996), pp. 65-225, had to beread 'manually'.

^Of course, in studying the idea of dominion, it would also be interesting tosurvey the noun "j'PD and the verb "]'PD; but, on the one hand, that would haveoverloaded this article, and, on the other hand, we may presume that studying thisderivative will give us a good impression of TgJon's thinking about this topic.

'*The word HDl'PQ, too, is mentioned in Imperial Aramaic, but a writing errorseems to account for its lexicographical existence (NDDi' n being written instead ofsmD' D; see J. Hoftijzer and K. Jongeling, Dictionary of the North-West SemiticInscriptions [Handbuch der Orientalistik, I 21.1; Leiden: E. J. Brill, 1995], s.v.). Atany rate, the word does not occur in Targum Aramaic.

LOREIN NHID' Q in the Targum of the Prophets 17

Tg.^ On the presumption that the average reader does not wish towade through my observations on all these texts, I have decided tocluster them.

I will propose several groups, which extend, broadly speaking, fromvery literal translation to very arbitrary innovation. I have tried toarrange the groups in a logical sequence. However, perfection cannotbe reached, because we have to deal here with on the one hand purelyformal phenomena, which reach the surface easily, and on the otherhand rather conceptual considerations, which are rather hidden in theexegetical ideas of the meturgeman. In general, we can say that theTargum is a simplifying translation, trying to avoid all difficulties, lin-guistic as well as theological.^

Literal Translation

Although TgJon has innovated a lot, there are also 109 texts wherea derivative of "]'PQ in MT has been translated with xms' iQ in Tg. Inmost instances, the translation is just literal. The actual translationin English differs, as in most categories. In some instances, we haveto point to a shift.^ In some other instances, an interpretive develop-ment has taken place. These texts will be treated when we discuss theinnovations of Tg.^

Example: Josh. 11.10bMT For Hazor formerly was the capital of all those kingdoms.TgJon For Hazor formerly was the capital of all those kingdoms.

This example is not exciting, being just what it is: an example of aplainly literal translation.^

^An exact count is difficult to obtain; what about variants? Errors? I haveincluded the first problematic category in my count, but not the second.

''For the formal phenomena, see E. Van Staalduine-Sulman, The Targum ofSamuel (SAIS, 1; Leiden: E.J. Brill, 2002), pp. 90-91. For the more fundamentalconsiderations, see this work, pp. 94-128, and P.S. Alexander, 'Jewish AramaicTranslations of Hebrew Scriptures', in M.J. Mulder and H. Sysling (eds.), Mikra(CRINT, 2.1; Assen: Van Gorcum; Philadelphia: Fortress Press, 1988), pp. 217-53(226-28), where no. 3, 'Doublets', should be completed with J.C. de Moor, 'MultipleRenderings in the Targum of Isaiah', JAB 3 (2001), pp. 161-80.

' See next section.^V. infra, pp. 30-31."Same type in Josh. 10.2; 13.12 (]E3 > 'p2 [Baxavotta] > |[JinQ; cf. L. Smolar

and M. Aberbach, Studies in Targum Jonathan to the Prophets [New York: Ktav,


Literal Translation with Some Clarification

In this section, the translation is not completely literal; TgJon has nottried to change the meaning, only to clarify it. A first way to accomplishthis goal is the resolving of metaphors. °

Example: 1 Sam. 15.28aMT The LORD has torn the kingship over Israel from you today.TgJon The LORD has removed the kingship over Israel from you today.

The metaphor of 'tearing' has been changed to 'removed'."

1983), p. 115), 21, 27, 30, 31; 1 Sam. 10.16 (but HK > '7.-J), 18, 25; 11.14; 13.13,14; 20.31; 24.21; 27.5; 2 Sam. 3.10, 28; 5.12; 7.12, 13, 16 (for this chapter, theliteral translation is remarkable, the Messiah not being intriaduced: Van Staalduine-Sulman, Targum of Samuel, pp. 527-29); 12.26 (D3D is the most Hteral translation ofIDb that exists: cf. BCTP); 16.3 (N3S is the standard translation of Hebr. " N, as inthe New Testament [e.g. Mk. 14.36] and frequently in Targum and Talmud Aramaic:cf. M. Jastrow, A Dictionary of the Targumim, the Talmud Babli and Yerushalmi,and the Midrashic Literature [New York: Choreb; London: Shapiro, 1886-1903],S.V.; however, it is different from Biblical [Dan. 5.13] and Qumran [lQapGen. ii24] Aramaic: cf. J.A. Fitzmyer, A Wandering Aramean. Collected Aramaic Essays[SBLMS, 25; Missoula: Scholars, 1979], p. 117 n. 25 [134-35], but we should bearin mind that a word in the realm of affection can have a particular evolution, andthat &e6? has no vocative in classical Greek); 1 Kgs 1.46; 2.12, 15 {bis), 22, 46; 5.1(rns is too literal a translation); 9.5; 10.20; 11.35; 12.21; 18.10 {bis); 2 Kgs 11.1;14.5; 19.15 (for -|3a 13 rcb, v. B.D. Chilton, The Clory of Israel: The Theologyand Provenience ofthe Isaiah Targum [JSOTSup, 23; Sheffield: JSOT, 1983], pp.6-7), 19; 25.25; Isa. 9.6; 13.4; 19.2 {bis); 37.16, 20; 47.5; Jer. 1.10, 15; 10.7; 15.4;18.7, 9 (TgJon should have distinguished more clearly between vs. 7 and vs. 9, ifKimchi were to be right: see P. Churgin, Targum Jonathan to the Prophets [YaleOriental Series Researches, 14; New Haven: Yale, 1927], p. 121 n. 12; now we cannotconclude that there is any difference between MT and TgJon, especially when weaccept the sg. reading with MSS Or. 1473 and Or. 1474; it has to be agreed that itis not clear to which people[s] reference is made); 24.9; 26.1; 27.1, 8; 28.1, 8; 29.18;34.1, 17; 41.1; 49.28, 34; 51.20; 52.31; Ezek. 17.13, 14; 29.14, 15; 37.22; Amos 6.2;7.13; Nah. 3.5; Zeph. 3.8 (with the pi. of MSS Montefiore and Or. 1474); Hag. 2.22{bis).

^°Cf. infra, pp. 25-28, for the situation where smD'7D itself is the result of thesolving of a metaphor. See further, e.g., J. Ribera-Florit, 'The Use of the De-rash Method in the Targum of Ezckiel', in CA. Evans (ed.). The Interpretation ofScripture in Early Judaism and Christianity. Studies in Language and Tradition(JSPSup, 33; Sheffield: Academic, 2000), pp. 406-22 (421).

^'Same metaphor is resolved in 1 Sam. 28.17; 1 Kgs 11.11, 13, 31 (in this verse[as in 1 Sam. 15.28] the metaphor has been called for by the same verb, but then inthe hteral sense of the word, in 1 Kgs 11.30 [as in 1 Sam. 15.27): Van Staalduine-Sulman, Targum of Samuel, p. 333); 14.8.

LOREIN xm^'po in the Targum of the Prophets 19

In other texts TgJon has tried to clarify the text by adding words,without departing from the meaning of MT.

Example: Isa. 13.19MT And Babylon, the ornament of kingdoms, the proud splendour

of the Chaldaeans, will be . . .TgJon And Babylon, which formerly was the joy of kingdoms, the glo-

rious praise of the Chaldaeans, will be . . .

To make a clear distinction between past and future, TgJon adds anadverb of time. At the same time, the apposition is converted to arelative clause.^^

Literal T'ranslation with Some Shift

In this section we will have to consider texts where the translation islargely literal, but not totally. From a methodological viewpoint this isa difficult section, because dividing it clearly from the sections 'LiteralTranslation with Some Clarification' and 'Innovation with 'D' Q in MT'is somewhat arbitrary. Nevertheless, a special category seems necessaryto keep the two other categories mentioned more homogeneous.

Example: 1 Sam. 14.47aMT When Sanl had seized royal power over Israel.TgJon When Saul had prospered in the royal power over Israel.

The verb nb)^ seems to provoke some semantic shift. TgJon refers to aphase in the life of a usurper or the like that comes after the actualphase described by MT: firstly, one seizes power;^^ only later one can

Other metaphors resolved in: 1 Kgs 11.34; Isa. 23.11, 17 (the 'normal' resolu-tion of the metaphor—which was to be avoided anyway: cf. Smolar and Aberbach,Studies, pp. 49-54; E. Levine, The Aramaic Version of the Bible. Contents and Con-text [BZAW, 174; Berhn: de Gruyter, 1988], pp. 101-102—was not possible, sinceTyre is the subject; in this verse, the Antwerp Polyglot offers a second occurrence,probably by mistake).

^^Same type in 1 Sam. 18.8 (TgJon gives at least a possible explanation of theelliptical MT); 1 Kgs 12.26; 2 Kgs 15.19 (on which verse TAB is too bold in itsn. 18); Isa. 10.10 (]n'7D'l is clearly lectio facilior; it would be difficult to explainsn'73 as a Pa'el); 14.16; 17.3 (TgJon tries to solve MT'S syntactical problem); 34.12(TgJon tries to solve MT's textual problem); 60.12; Jer. 25.26; Hos. 1.4 (both MTand TgJon speak about the 'dynasty' of Israel, but TgJon avoids the interpretationof 'territory'); Amos 9.8.

^''Or: one becomes king; in that case, there would also be a semantic shiftfrom 'kingship' to 'royal power'. According to Van Staalduine-Sulman, Targumof Samuel, p. 318, TgJon uses the word to avoid the idea of Saul carrying out acoup, but this is not an explanation for the situation in 2 Sam. 16.8.


be successful with it. Formally, TgJon uses a preposition where MT hasa direct object. ^

A very special case is the situation where God's KniD' Q has beencombined with the verb '" J. This combination can also occur when MTshows a form of "pD, or even when there is no derivative of "pa at allin the background.^^

Obad. 21bMT And to the LORD will be the kingship.TgJon And (then) will be revealed the royal power of the LORD upon

all the dwellers of the earth.

We have to recognize that MT is rather brief. About what is MT speak-ing? This becomes clear in Tg: it is not an undetei-mined kingshi]) thatin one way or another comes to God, but that which is permanentlythe case, that is, the fact that God is king will be revealed in the end.In that situation, the sense 'royal power' seems to be more appropriate.By the addition 'over all the dwellers ofthe earth' this text seems to geta universalistic flavour. Nevertheless, it does not make clear whetherthe nations will be subjected to this royal power (as beneficiaries ornot), or whether they will only see it from a distance; or, in gram-matical terms, whether KUIK 'nn' bD '7V is an adjunct to NHID' Q or to'b^nm. Gonsequently, TgJon could have been still clearer, but what ismore important, MT can also be interpreted in a universalistic way.This means that TgJon cannot be considered as a clear theologicaldevelopment.^^

situation in 2 Sam. 16.8 {TAB stays too close to MT in its translation);similar situation in 1 Kgs 21.7. In Ezek. 16.13, there is a problem in a text withnb:i too.

i^V. infra, pp. 21, 34.^''C. van Leeuwen, Obadja (De Prediking van het Oude l?estament; Nijkerlc: Cal-

lenbach, 1993), ad loc. (without reference to TgJon). In that case, the influence ofTgJon Zech. 14.9 would have been less important than CJhilton, Clory of Israel,p. 79, suggests; of course, this does not void Chilton's chronological reconstruction(see pp. 78-81).

The Targum of this verse has some pluses: KTg Obad. 21 repeats a standardformula which adds no information to the main text; TosTg Obad. 21 (Kasher,Targumic Toseftot, no. 137 1. 11) points out that God's dominion applies alreadyin this world, and not only in the world to come. In this Tosefta (1. 6) there is alsoa rather strange digression with a secular, territorial use of smD':'Q.

See for parallels T. Mos. 10.1 (written in Palestine in the first years of the firstcentury CE: T.C. Vriezen and A.S. van der Woudej, Oudisraelitische en vroegjoodseliteratuur [Ontwerpen, 1; Kampen: Kok, 10th edn, 2000], p. 473) and 2 Bar. 39.7

LOREiN sniD' a in the Targum of the Prophets 21

Noun or Verb "['PQ in MT

In these cases, we do not have a derivative of "j' Q in MT, but "["Q itself.Thus we have only a change of word category, which does not seem tomake much difference at the semantic level.

Example: 2 Kgs 24.12bMT And the king of Babylon took him in the eighth year of his being

king.TgJon And the king of Babylon took him in the eighth year of his reign.

In translating the Hebrew verb ~\bn by the Aramaic noun xmobn TgJonhas not changed the meaning. '

However, it is not always as simple as this example might suggest. Inthree cases the simple MT "fp^ (with God as subject) has been replacedby the 'reveahng of the smD'PD of the Lord'.

Example: Isa. 24.23baMT For the LORD of hosts is king on Mount Zion and in Jerusalem.TgJon For the royal power ^ of the LORD of hosts will be revealed on

Mount Zion and in Jerusalem.

Obviously, there is a difference in wording between MT and Tg, but ifwe read commentaries on MT (which are commenting just on MT, notTg), we see that MT is interpreted by commentators in just the sameway as TgJon interprets it. Therefore, it is not necessary to perceivehere an important development in interpretation.^^

(written in Palestine in the first years of the second century CE: Vriezen and Vander Woude, Oudisraelitische en vroegjoodse literatuur, pp. 480, 483.

^^Same type in 2 Sam. 3.39; 1 Kgs 11.14 (MS Or. 1471 and Antwerp Polyglot); 2Kgs 25.27 (MS Or. 1472), 28 (MS Montefiore and Leiria Edition; the other witnessesare easier [S'D'PQ] , but the variant reading could be understood as meaning hostagesof the different dynasties that had revolted against Babylon: thus H.A. Brongers, / /Koningen [De Prediking van het Oude Testament; Nijkerk: Callenbach, 1970], adMT); Isa. 45.1 (MS Or. 1474; I would prefer the reading ]'3'7Q of the other witnesses,because, contrary to normal use in TgJon, the metaphor ['opening the hips'] hasbeen preserved in TgJon).

In Isa. 57.9 the consonants '['PD have played a part in TgJon's choice for KDlDba,but other factors are important there too.

*Cf. KHTOJ in Edd. Bomberg and the Antwerp Polyglot.^"A. Schoors, Jesaja (De Boeken van het Oude Testament; Roermond: Romen,

1972), ad loc. Cf. J.J. Collins, 'The Kingdom of God in the Apocrypha and Pseude-pigrapha', in idem. Seers, Sybils and Sages in Hellenistic-Roman Judaism (JSJSup,54; Leiden: E. J. Brill, 1997), pp. 99-114 (112-13). According to Camponovo, Konig-tum, p. 419, the wording of TgJon is determined by the presence of the idolaters in


Literal Translation without '3*70 in MT

The title of this section might appear strange; an example may helpto clarify.

Example: Jer. 22.18bMT They will not lament for him: 'Woe, lord! and. Woe, his majesty!'TgJon They will not lament: 'Woe about the king! and. Woe about his

royal dignity!'

nin means 'majesty', and xniD' D can have about the same meaning,'royal dignity'. Therefore, TgJon is innovating formally by introducingxmD'pa, but conceptually the meaning is not changed at all.

This means that TgJon can present xniD' D where there is no 'Dba inMT, as a literal translation or, in other cases, as a clarification accordingto the plain meaning of the passage.^"

Metonymic Use

The most frequent form of metonymy involving xniD' Q comes whenour term is used, not for a territory, but for the inhabitants of thatterritory. The best translation in this instance is 'kingdom', as thesame metonymy occurs with this English word. In the texts mentionedin this section, we can be fairly sure about this meaning, because theword used in MT clearly indicates a population.

Here we find such obvious background words as "i:, Dx'p and na, butalso nriDOQ and 30'.

the first half of the verse. In general, he says (see p. 410) that "po has been replacedby KUID'PD to avoid the idea that a human action would be attached immediatelyto God. The previous text (2 Kgs 24.12b) shows that a broader context has to betaken into account.

Similar cases: Isa. 52.7 (E.J. Youngf, The Book of Isaiah, III. Chapters 40 through66 [Grand Rapids: Eerdmans, 1972], ad loc; on this verse, Gamponovo, Konigtum,p. 421, agrees with me); Mic. 4.7 (cf. R. le Deaut, La nuit pascale [AnBib, 22; Rome:IBP, 1963], pp. 140, 276; Gamponovo, Konigtum, p. 426; this Targum passage seemsto be the most recent of the series: both Ghilton, Clory of Israel, pp. 79-81, andLe Deaut, La nuit pascale, p. 276, defend this position, with different arguments);Zech. 14.9 (A.S. van der Woude, Zacharia [De Prediking van het Oude Testament;Nijkerk: Gallenbach, 1984], ad loc; here the universalistic flavour is plainly presentin MT: cf. Ghilton, Clory of Israel, p. 78).

^"Same type in 1 Kgs 15.13 ('kingship' of the queen mother); Mic. 4.8 (MS Mon-tefiore; 'royal power');

clarification in 2 Sam. 7.18 (MS Reuchlin); Isa. 41.9.

LOREIN sms'^a in the Targum of the Prophets 23

Example: DU, Isa. 61.9aMT Their seed will be known among the nations;

and their progeny amidst the peoples.TgJon Their children will be raised among the peoples;

and the children of their children amidst the kingdoms.

In Tg, 'be known' has been particnlarized to 'be raised', and twovery common techniques have been applied: resolution of the meta-phor ('children' instead of 'seed') and replacement of a more difficultword ('the children of their children' instead of D'K:iK:i).

Why has D'Dan not been translated by S'QniJ? That word was alreadyused in the first half of the verse, as a translation of D^^}. This happensrather frequently. As TgJon prefers not to use the same word in thesame verse twice, and as S'QDi; is the most obvious translation of D'lJ(which has no closer equivalent in Aramaic), smiD' a is used as a 'second

e' ^ translation when K'QQiJ has already been used.^^Therefore, it is clear that the translation of WV, D'ox':' or CQV by

needs no further attention, and that in these cases Kmo' Q isused with the metonymical sense of 'populations'.^^

f. the use of K' K if ^3 has already been used earlier in the verse.phenomenon occurs in Isa. 2.4; 11.10; 17.12 {TAB's comment seems less

convincing); 25.7; 33.3 (the reading of MS Reuchlin is to be rejected for this reason);34.1; 41.15; 43.4, 9; 49.7, 22; 55.4; 60.2; 61.9; 62.1; Jer. 10.25; 51.58; Hab. 2.5, 13;Zech. 8.22; 9.10.

^•'Same type with '13 in Isa. 2.2 (accepting the reading of MSS Reuchlin andOr. 1474), 4; 11.10; 14.26 (accepting the reading of MSS Reuchlin, Montefiore andOr. 1474); 25.7 (accepting the reading of MSS Montefiore, Or. 1474 and Paris);33.3 (the meturgeman has avoided an anthropomorphism, but his rendering lacksclarity: Whose strong deeds are meant?); 49.7; Ezek. 31.17; Zech. 8.22 (MS Reuchhnseems to have misread ]nD'PDl as ]"3'7m, while MS Montefiore and Edd. Bomberg havethe same fault, but with the—grammatically necessary—adaptation of ]2~i2~i intop - i m ) ; 9.10.

Same type with DX'p in Isa. 17.12, 13; 34.1 (the plural of the MSS Montefiore,Or. 1474 and Paris and the Edd. Bomberg is to be preferred); 41.1; 43.4, 9 (thealternation of the tenses has been removed); 49.1; 55.4; 60.2; Hab. 2.13 (somewitnesses [MSS Montefiore and Or. 1474 and Kimchi] and the use of the 3 pi. fem.form seem to imply the plural NmiD'PD, but the combination of these argumentsmakes it rather complicated).

Same type with DV in Isa. 24.13; 49.22; 61.9; Mic. 4.1; Hab. 2.5. In Isa. 7.8stands indeed for CSJ, but the situation is different: not plural (as in all cases above)but singular, and perhaps not even metonymic (v. infra, n. 77). Admittedly, in Isa.49.22 and Mic. 4.1 we find the singular too in MS Or. 1474 (for Mic. 4.1 also inMSS Reuehlin and Montefiore and in the Edd. Bomberg, but always without b'D;the combination of TAB, ad loc, 'all the kingdoms' has no foundation), which


One problem remains, that is to say, the combinationIf the component S'DQ:J already indicates 'population', what in thatcase is the meaning of xmo' Q? This use indicates that the meturgemanwas not obsessed by his use of xmo' Q in a metonymical way, i.e., withthe meaning of 'population'. Furthermore, it has to be observed thatin these texts we have a phenomenon similar to that discussed in aprevious paragraph: HDlX had to be translated by KiJIX, so, this wordcannot be used for 'f)K; therefore, that word had to be translated byanother word, in casu K'DQU. In this way we get the combination

, which seems somewhat overloaded.

Example: US", Jer. 48.18aMT Descend from glory so that^^ you will dwell in thirst, dweller of

the daughter of Dibon.TgJon Descend^^ from the glory so that ' yon will dwell in destruction,

kingdom of the assembly of Dibon.

The image of thirst has been replaced by the ab:3tract noun 'destruc-tion', as often in Tg. ra has become xnCjS, which does not occur ex-clusively for Israel, but also for heathen nations.^* When we considerthat DC' as well as SDID' Q (in the metonymic sense) can designate in-habitants, this translation poses no problem.^^

There is still another situation where metonymy plays its part, al-beit in the MT background, as in the case of MT n'^ with the meaning

probably has to be interpreted in a particularistic way. Here also. Or. 2211 seemsto be preferred.

^''The combination occurs in Isa. 23.17; Jer. 25.26; Ezek. 17.24; 34.5, 8; Zech.2.4; 5.9.

f. P. Joiion, Crammaire de I'hebreu biblique (Rome: IBP, 1923), § 116a, f.imp. 2 sg. fem. from nrij.use of an indirect volitive is not described for the Aramaic language, but

see, for an asyndetic counterpart, S. Segert, Altaramdische Crammatik (Leipzig:Enzyklopadie, 1975), § 6.6.6.4.4., and for Rabbinic Hebrew M. Perez Fernandez,An Introductory Crammar of Rabbinic Hebrew (trans. J. Elwolde; Leiden: E.J.Brill, 1997), § 30.4 (but the example does not illustrate the rule); although thereare many m,orphological studies on later Aramaic, a syntactical overview is stilllacking.

2»Cf. Jer. 46.19.^"Same type with n c in Jer. 46.19 (exile equipment: a basket with a minimal

amount of tools, pottery, a mat and a light: W.L. HoUaday, Jeremiah, II [Herrneneia;Minneapolis: Fortress, 1989], p. 331; TAB ad Ezek. 12.3).

Same type with nnSBD in Jer. 10.25; 25.9; Zech. 14.18; Nah. 3.4 (prostitutionrelated to religious affairs, either because understood as a metaphor, or out ofprudery; the two mechanisms do not exclude each other).

LOREIN xmn' Q in the Targum of the Prophets 25

of dynasty; this is really a case of metonymy. On the other hand, TgJonKniD'pa is not: the meaning 'dynasty' is too close to the idea of "[':'Q tospeak about metonymy.

Example: 1 Kgs 11,38cMT I will be with you, and I will build for you a trustworthy house

as I built for David,TgJon My Memra will be as your help, and I will establish for you a

firm dynasty as I established for David,

The typically biblical phrase DU n^n/ elvai [X£T(X, ° which can easilybe misunderstood, is explained by 'be as your help',^^ The image ofbuilding a house, which clearly indicates the institution of a dynasty,has been replaced by a still clearer expression,^^ This transition from aconcrete image to a clearer abstract expression often occurs in TargumJonathan,^^

Resolved Metaphors

For this section, I have selected texts where a metaphor is used inMT, but where TgJon simply says what the metaphor means and, forthat reason, introduces xm^bn because he aims at clarity. Althoughdifferent background words are involved here, the usual practice is thatmetaphors for (mainly foreign) kingdoms are translated by

Example: nbinn, Isa, 47,1aMT Descend so that you will dwell on dust, virgin daughter Babylon.TgJon Descend so that^* you will dwell on the dust, kingdom of the

assembly of Babylon.

The first half has been translated almost literally; the status emphati-cus, however, repairs the lack of the article with is:;. In the second half,the prophet does not speak about different kinds of young women, butabout a city, which he represented in a metaphorical way. TgJon does

. 39.2-3; Acts 7.9; cf. Hag. 1.13.the Memra, cf. Levine, Aramaic Version, pp. 59-60.

^^Consequently, TAB (ad loc.) is wrong in saying that TgJon 'stresses the king-dom, rather than the dynasty'. MT rr3 as well as TgJon xmD' a mean 'dynasty'!

^^Same type in 1 Sam. 2.35 (v. infra, n. 74); 25.28; 2 Sam. 7.11 (cf. VanStaalduine-Sulman, Targum of Samuel, pp. 23, 526: just a clarification, whereasthe New Testament [Lk. 1.32; Heb. 1.5] is messianic), 27; 1 Kgs 2.24,

The situation is different in Amos 6,11 (v, infra, n, 37),^''For this translation, v, supra (p, 24) ad Jer, 48,18,


not translate metaphors literally, but uses plain language: a kingdomand an assembly, as synonyms for 'a city'. Although we may think im-mediately of 'Israel' while hearing about 'daughters', this is not alwaysthe case, ^ so that we do not need to see a mechanically operatingmeturgeman behind this text.^^ As to the meaning, then, TgJon doesnot innovate. Regrettably the alliteration with 2 is lost in Tg, butliterary effect was not Tg's main concern.^^

Some other metaphors also demand the translation xmo'pn, butwithout indicating a population.

f. Isa. 47.1, 5 (about the Chaldaeans); Jer. 46.11, 24 (about Egypt); 48.18(about Diban [Moab]; v. supra, p. 24); 50.42; 51.33 (about Babylon).

TAB.went before literary devices: see Van Staalduine-Sulman, Targum of

Samuel, p. 129.Same type with n' ina in 2 Kgs 19.21; Isa. 37.22 (I cannot find any explanation

for the plural of MS Reuchlin; accordingly, the manuscript shift between sg. and pLin general must not be taken too seriously); Jer. 46.11.

Same type with riD in Isa. 47.1, 5 (but cf. infra n. 52); Jer. 46.24.Same type with n:)33 in Isa. 41.15.Same type with n''n in Ezek. 34.5, 8, 28.Same type with (n)3iJ in Ezek. 19.Hay; 31.10 (but intei'pretation and lexicon are

not clear: cf. W. Bacher, 'Kritische Untersuchungen zum Prophetentarguni. Nebsteinem Anhange iiber das gegenseitige Verhaltnis der pentateuchischen Targumim',ZDMC 28 [1874], pp. 1-72 [p. 52 n. 2], though not convincing), 14.

Same type with n"?:!? in Jer. 46.20.Same type with yv in Ezek. 17.24 {quater; in the first two statements, TgJon

restricts himself to re solving the metaphors; in the last, he becomes more resoluteby identifying the trees and by recycling the metaphors ai5 similes; in mm PiDbo wewould have expected the status emphaticus: cf. Segert, Altaramdische Crammatik,§ 6.3.3.1.5.; but cf. also Perez Eernandez, Crammar of Rabbinic Hebrew, § 11.7,where he seems to suggest that a construct state can be determined by a relativeclause).

Same type with pp in Zech. 2.1, 2b, 4 (bis).Same type with mi in Jer. 49.36.Same type with mo in Isa. 49.23 (semantically, this is not a variant of I'PD [cf.

supra, pp. 21], considering the image of nursing).Same type with "is in KTg Ezek. 19.2.Similar type with n'3 in Amos 6.11 {bis; kingdom, i.e., territory; N.B.: the same

interpretation has been given to MT as to TgJon—rightly or wrongly: see C. vanLeeuwen, Amos [De Prediking van het Oude Testament; I^Iijkerk: Callenbach, 1985],ad loc.—, so that it is not necessary to speak of an interpretational development).

Hab. 3.8 ("inj; Antwerp Polyglot) is to be considered as an error: the suffix ]^r^, aswell as the word 'armies' itself, indicates that the reading yoba is to be preferred.

Zech. 2.3 (tDin; MS Montefiore) is also an error, as the explanation of the visioncomes too early; the other witnesses maintain the image in this verse.

LOREIN Kno' n in the Targum of the Prophets 27

Example: Unt, Ezek. 30.24aMT And I will strengthen the arms of the king of Babylon.TgJon And I will strengthen the royal power of the king of Babylon.

Here the part of the body that indicates the strength of a man hasbeen replaced by what it signifies: royal power. *

Example: T3, 2 Sam. 21.17bMT You will not go ont any more with us for war, and yon will not

extinguish the lamp of Israel.TgJon You will not go out any more with us for war, and you will not

extinguish the dynasty of Israel.

With the 'lamp' something more extensive than just the individualDavid is meant; consequently, ~i] is not a metaphor for 'king', but for theroyal institution, the dynasty.^^ By maintaining the verb ('extinguish'),TgJon has resolved the metaphor only partially,^" contrary to normalusage, where the metaphor is resolved completely, although sometimesrecycled as a simile. ^

type with n''?"! in Ezek. 17.23 (N.B,: in Ezek. 17.22-23, there is no in-terpretational shift, since MT as well as TgJon are messianic: cf. G.C. Aalders,Ezechiel [Commentaar op het Oude Testament; Kampen: Kok, 1955], pp. 291, 295;S.H. Levey, The Messiah: an Aramaic Interpretation [Monographs of the HebrewUnion College, 2; Cincinnati: Hebrew Union College, 1974], p. 79 [pace pp. 86-87,102, although Levey seems right in observing the differences between TgJon Ezek.and the other Targumim of the prophets]; B.D. Chilton, 'The Targumim and Ju-daism of the First Century', in J. Neusner and A.J. Avery-Peck [eds.], Judaismin Late Antiquity, HI. Where we stand: issues and debates in Ancient Judaism II[Handbuch der Orientalistik, 141.3.2; Leiden: E. J. Brill, 1999], pp. 115-50 [142-43]).

Same type with :JTIT in Ezek. 30.25 [bis; cf. E. Dhorme, L'emploi metaphoriquedes noms de parties du corps en hebreu et en akkadien [Paris: Gabalda, 1923], pp.138-41; and now also Ribera-Elorit, 'Use ofthe Derash Method', p. 413).

Same type with p p in 1 Sam. 2.10 and Jer. 48.25 (cf. Dhorme, L'emploim,etaphorique, pp. 34-41).

^ Cf. the argumentation of C. J. Goslinga, Het tweede boek Samuel (Commentaarop het Oude Testament; Kampen: Kok, 1962), p. 400 (who nevertheless uses theword 'king'), and, above all, the texts mentioned in the note at the end of theparagraph.

^° A Tosefta-Targum of this passage exists, but the wording of this phrase is aboutthe same (v. Kasher, Targumic Toseftot, no. 69 1. 63).

"• Same type with "I": in 1 Kgs 11.36; 15.4; 2 Kgs 8.19.Similar type in Isa. 60.20, where also a light (i.e., CQD) indicates permanent

kingship, that is, in human terms, a dynasty. As this text speaks about Cod, thetranslation 'dynasty' is, of course, out of the question.

Same type with n^D in Amos 9.11a.Same type with niQli in Ezek. 17.22 (cf. supra n. 38).


Example: ~\ii\ Ezek. 29.3bMT For he said, 'Mine is my stream and I have made it for ^ myself.'TgJon For he said, 'Mine is the dominion and I have conquered it.'

I have used here the equivocal term 'dominion', b(3cause in this passage(Ezek. 29.3-10) ~IX' has been translated four times by sm ' Q, so thatit seems necessary to find a translation that fits the whole passage.In vs. 4, however, the translation of the second "iX' by SDpin, plus theuse of the verb ba2, point to the meaning 'royal power'; on the otherhand, in vs. 10 this translation is impossible. By interpreting xmo' Q asa territory,''^ the four texts can be treated in the same way. A solutionthat fits the four occurrences is to be preferred, even when one of thefour (vs. 4) remains problematic.

It also happens that an image is replaced by a simpler expressionin such a way that we cannot indicate simple one-to-one relationshipsbetween the Hebrew and the Aramaic.

Example: Isa. 10.19bMT And a boy will write them down.TgJon And they will be eonsidered as a weak kingdom.

While not a single word has remained, the thought has not beenchanged. ^

Clarifieations Introducing

In some texts TgJon formally introduces xmo' Q, but he does not intendto change the meaning of the text, but only to reach his goal: clarityfor the audience.

Example: 2 Sam. 23.1bMT and the word of the man who was raised on highTgJon and the word of the man who was raised to royal dignity

suffix: cf. B.K. Waltke and M. O'Connor, An Introduction to BiblicalHebrew Syntax (Winona Lake: Eisenbrauns, 1990), §§ 10.2.li, 16.4f.

''^A 'kingdom', but not in the sense of population; although E33 is mostly usedof people, it can also be used of territory, which seems to be preferable hei'e.

One might wonder whether this passage is treated in the right section, as nK' canbe considered, not as a metaphor, but as metonymical for Egypt (pars pro toto).

^^Same type in Judg. 9.9, 11, 13 (cf. W.F. Smelik, The Targum of Judges [OTS,36; Leiden: E. J. Brill, 1995], p. 523); Isa. 44.28.

LOREIN smo'pn in the Targum of the Prophets 29

Poetical language has been explained for the uneducated audience, lestthey think that David was physically lifted up.'*^

Expansion: Epitheton Ornans

Very often, xniD'pa is introduced as an epitheton ornans, mostly with

'OIID. It is possible that one MT occurrence could canse a whole series of

TgJon occurrences,''^ but this is not necessarily the case. It is clear that

a throne has much to do with a dynasty; therefore, the use of xniD' a

in this section can be compared with the meaning of 'dynasty'.''^

'DIID: for example, 1 Kgs 1.48b

MT . . . Who has given today someone sitting on my throne, and myeyes are seeing (it)!

TgJon . . . Who has given today a son sitting on my royal throne, andmy eyes are seeing (it)!

TgJon has added yi—because Dm is more pregnant than xar, ~a—to

make the sentence easier to understand,''^ and smD'PQ—as an epitheton

ornans. One might also have expected a replacement of 'TV, but this

was not necessary: 'TV is not a metaphor and is clear enough.''^

type in 2 Sam. 23.5 (clarification of the intention of the n—\2; we mightwonder whether David's dynasty was to reign or David personally, in an eschato-logical perspective); 1 Kgs 22.31 (only in MS Or. 2371 and Ed. Bomberg; ratherclumsy addition under the influence of 1 Kgs 20.1, 16; cf. W. Bacher, 'KritischeUntersuchungen', p. 51); Isa. 16.4; 47.7 {]m'?a is gen. to the adj. ns'pn; cf. for Rab-binic Hebrew Perez Fernandez, Grammar of Rabbinic Hebrew, § 11.4A); Jer. 10.7(here Camponovo, Konigtum, p. 423 n. 53, sees a relation with the idolaters too);48.4, 38; 49.4; Ezek. 31.6, 12; Zech. 2.2a (MS Reuchlin; obviously an error, as theclarification comes too early; the other witnesses translate MT literally; cf. vs. 3).

• ^E.g. 1 Kgs 1.47 for the whole chapter."^Of course, the situation is different when God is the subject of the idea of

'being king': Jer. 49.38.''^Churgin, Targum Jonathan, p. 72, presumes this has happened under the in-

fiuence of 1 Kgs 3.6, but it has to be observed that a grammatically undetermined,but in the context individual 'sitting one' is not easily understood immediately.

*^Same type in 2 Sam. 3.10 (MS Or. 1472); 7.16; 14.9; 1 Kgs 1.13, 17, 20, 24,27, 30, 35, 37, 47 (only once, and that where it suits best); 2.4, 12 (MSS Or. 1472,2364, 2371, Montefiore etc.), 24, 45; 5.19; 8.20, 25; 9.5; 10.9; 2 Kgs 10.3, 30; 15.12;Isa. 14.13 (TgJon puts the royal throne at another—but not an easier—place thanMT does, perhaps to avoid even the notion of astrology: cf. Smolar and Aberbach,Studies, p. 196); Jer. 33.17; 49.38; Jona 3.6 {TAB's translation 'his royal throne'is lighter and better than, e.g., 'the throne of my kingdom' used by TAB in Jer.49.38; E. Levine, The Aramaic Version of Jonah [New York: Sepher-Hermon, 1981],p. 87, describes the situation very well by speaking about a 'somewhat greaterspecificity'); also in TosTg Ezek. 1.1 {bis).


p': for example, Isa. 47.2bMT Take away your veil, strip off your robe, take away the cover of

your thigh, cross rivers.TgJon Take away your royal glory, your rulers are broken, scattered are

the people of your armies, they are taken away like ° the river.

The images have been explained; although Xip' occurs mostly as anattribute of God, it is also used of passing, human glory. ^ Moreover,

has been added.^^

Innovations with 'D'PQ in MT

These are three texts in MT with a derivative of "pn, which, in Tg,show deviations from MT that are too important to catalogue under'Literal Translation with Some Shift'.

Example: Ezek. 16.13bMT And thou didst prosper unto kingship.TgJon And ye did prosper and rule over all the kingdoms.

The most obvious change is from singular to plural,^^ but th(3 mostimportant is the change in meaning from HDl' Q to sniD' Q, i.e., from'kingship' to 'kingdom'.^^

Isa. 62.3 represents a unique case, because only there does MT usea derivative of "[" n (HDi'pa), whereas TgJon does not.

^°According to BCTP, the Hebraism IQD occurs also in variants of 1 Sam. 16.7and Jer. 3.22. Other witnesses read the plainly Aramaic 'aD, 'like the waters of theriver'.

^^Chilton, Glory of Israel, p. 76.s^Same type in Isa. 47.5 (MSS Reuchlin, Montefiore, Niiremberg, Edd. Bomberg;

with this reading, niD'PD is a genitive with "ip"; the other witnesses—without np"—consider niD'PQ as vocative [cf. supra, n. 37]); Ezek. 32.7 ('royal glorious splendour').

Similar type in 1 Sam. 10.7 (but with interpretative development; v. infra, pp.36-38); 2 Sam. 12.27 (TgJon has the 'royal city' for MT 'Water City' by lack oftopographical knowledge); 1 Kgs 11.14 (MS Or. 1471, Antwerp Polyglot); Ezek.30.21 (but lacking in MS Or. 1474, Edd. Bomberg and Kimchi).

^^ TAB'S translation departs from TgJon.^*Same type in Mic. 4.8 (MT subject, TgJon preposition; MT David's royal dignity

returns for the fugitives of the North, living in the new quarter of Jerusalem: D.Schibler, Le livre de Michee [Commentaire Evangelique de la Bible; Vaux-sur-Seine:Edifac, 1989], ad loc), TgJon rule comes to the kingdom of Jerusalem).

In Jer. 51.27, TgJon refers to other kingdoms than MT (cf. Smolar and Aberbach,Studies, p. 118, 121-22).

LOREIN xniD'pQ in the Targum of the Prophets 31

Isa. 62.3MT And you shall be a royal diadem in the hand of your God.TgJon And you shall be the diadem of praise before your God.

Obviously, mention of the hand is considered too anthropomorphic,but, what is more important, it is considered necessary to avoid themerest idea of Zion occupying a royal position vis-a-vis the Almighty, ^because TgJon tries to prevent all theological problems. Therefore,while TgJon likes the word Kms' o, he had to drop it here. Of course,Zion's adopting an attitude of praise presented no problem.

Innovations with a Background Word

In a number of places, TgJon replaces a word in MT by xm ' a, whileat the same time innovating at the interpretational level. It shouldbe understood that we can speak about innovation only when TgJondoes more than simply explain MT; in other words, in order to bementioned in this section, TgJon has to depart from the meaning of MT,by allegorizing or by a small shift in sense (more substantial alterationwill be mentioned in another section).

In this category TgJon is confronted with the problem that a trans-lator cannot ignore problematic passages, although the commentatorcan. He is bound to translate, even when he ends up rather explainingor keeping in line with his theology. Here we meet an explanation thatpushes a metaphor too far, in other words, allegorizing, which is a kindof theology.

Example: Ezek. 31.16bMT And in the nether world all the trees of Eden consoled each other,

the choice and best of Lebanon, all the drinkers of water.TgJon And in the nether world all the kings of olden times consoled

each other, the rulers and the rich in possessions, all those who^ royal power. ''

MT simply reminds Egypt that even large empires not only rise, but alsofall. TgJon tries to find an explanation for every detail of the image.The trees of Eden have become kings of olden time, the synonymous

course, that was not the intention of MT, either.^^Whatever may be the vocalisation of nni) (v. Sperber, Bible in Aram,aic, ad

loc).^'' TAB is wrong in translating 'all who serve the state' (probably misled by the

Hebr. 13:) 'serve', which in Aramaic would be n' S or BOB).


'choice and best ofthe Lebanon'^* have become two different categories,

the powerful and the rich.^^

We find a special case in Ezek. 7.7, 10, where TgJon has not alle-

gorized, but has tried to explain an unknown Hebrew word.

Ezek. 7.10MT Behold the day, behold she is coming; the sephira has come out,

the rod has sprouted forth, arrogance has blossomed.TgJon Behold the day of retribution, behold she is coming; the royal

power has been revealed, the ruler's rod has sprouted forth,wickedness has appeared.

Whereas the Septuagint had solved the problem of the difficult word

rn''S:i by skipping it, TgJon tried to interpret this word. The day has

been explained by adding 'retribution';™ the strange phenomenon of

the female participle in MT has been preserved in TgJon (indicated in

the translation by the use of 'she'). The word i^^tl)^^ can be considered

as a translation of MT nCDQ,^ but its occurrence might also have been

influenced by a search for parallelism with xniD' 'Q. ^ The rare ]nT has

been translated by the frequent K:Jm~i. The main question is why the

•'' Mostly, Lebanon is rendered by 'the Temple' (for the cedar wood), but thetranslation given here is possible, too, especially in Ezekiel: R.P. Gordon, Studies inthe Targum to the Twelve Prophets: From Nahum to Malachi (VTSup, 51; Leiden:E.J. Brill, 1994), p. 52; Ribera-Florit, 'Use of the Derash Method', p. 410 n. 10(with the remark that it is already implicit in Scripture); cf. G. Vermes, Scriptureand Tradition in Judaism (SPB, 4; Leiden: E.J. Brill, 2nd edn, 1973), pp. 26-27.

^"Same type in PalTg 1 Sam. 10.22 (D' H has been replaced frequently by ID'^Q,under the infiuence of TgJon 2 Sam. 7.18: Van Staalduine-Sulman, Targum ofSamuel, p. 281); Isa. 49.7 (The meturgeman based himself on the interpretationof the Lord's Servant as the collective wandering Jew—instead of an individualSuffering Servant; his Servant was not despised by the people, but dispersed amongthe peoples. Having in mind the plural D'l:, the translation smis'^a was obvious tohim.); Jer. 51.26; Ezek. 13.13; 31.14 {bis); Joel 2.25 (cf. C. van Leeuwen, Joel [DePrediking van het Oude Testament; Nijkerk: Callenbach, 1993], pp. 34-35); Zech.4.7a (the future ^ has been replaced by the theocratic 3: cf. A. Sperberf, The Biblein Aramaic based on Old Manuscripts and Printed Texts, IVB. The Targum andthe Hebrew Bible [Leiden: E.J. Brill, 1973], p. 87; TAB [the volume on the MinorProphets—by Gathcart and Gordon—always gives more illuminating commentsthan most other volumes] gives a quite complete discussion of this text [but thereference should be to Sir. 50.26]; cf. Levey, Messiah, p. 98). Also in KTg Ezek.32.14 (if my interpretation of Sperber's ]inmiDn'7 as an error for pnmiD'pnC ) is right;cf. Vermes Scripture and Tradition, p. 160).

"''Which word should accordingly have been italicized in TAB.Jastrow, Dictionary, s.v.

Amos 3.12.

LOREIN xmn' n in the Targum of the Prophets 33

(probably) negative m'SIi has been replaced by xmD' 'a,'' which is al-most certainly positive, as God's royal power seems to be meant in thiseschatological context.^''

Another question is where this Targum is to be placed in Chilton'schronological reconstruction. Chilton himself considers it as early, be-cause it is universalistic, but the elliptical wording rather points to adate when the idea was already so common that an ellipse was easy tounderstand.^^ If TgJon had wanted to avoid a universalistic interpreta-tion, he should have taken extra measures in vs. 7 instead of translatingliterally Ni iK'DTr, which in any case has to be regarded as collective.

Innovations without a Background Word

The same situations occur when there is no background word.

Example: Isa. 38.12bMT As a weaver I have rolled up my life; from the thrum He will cut

me off.TgJon My life has been rolled up as a stream of high banks; from my

royal glory am I uncovered.

The main goal of TgJon was to give a simplified version of the MT meta-phor. In this case, however, he has chosen not to use only straightfor-ward language, but offers another image in the first half of the verse.In the second half, it seems to be just a simplification, but more isinvolved: the idea of royalty is absent in MT.^^

Here again, I have to mention some texts where xniD' Q has beencombined with the verb "'PJ, and, in a certain sense, the distance fromMT is smaller than in the previous section, so that we have to wonder

"^Probably Isa. 28.5, with its positive context, has played a pai't in TgJon'sinterpretation (cf. TAB ad loc).

'^''Chilton, Glory of Israel, p. 79; J. Ribera, 'The Image of Israel accoi-ding toTargum EzekieP, in K.J. Cathcart and M. Maher (eds.), Targumic and CognateStudies: Essays in Honour of Martin McNamara (JSOTSup, 230; Sheffield: Aca-demic, 1996), pp. 111-21 (p. 118 n. 26); Camponovo, Konigtum, p. 424.

''^Chilton himself supposes that this abbreviation could have originated in anyperiod. Chilton, Glory of Israel, p. 79.

""Same type in Isa. 42.7 (limits the reference to the house of Israel [cf. Acts26.18!]; cf. J.C. de Moor, '"Van wie zegt de profeet dit?" Messiaanse apologetiekin de Targumim', in: H.H. Crosheide et al. [eds.], De knechtsgestalte van Ghristus:Studies door collega's en oud-leerlingen aangeboden aan prof. dr. H.N. Ridderbos[Kampen: Kok, 1978], pp. 91-110 [102]); 43.2; Ezek. 19.11aP, 14 (cf. Ribera, 'Imageof Israel', p. 115); Hos. 13.15; Zech. 6.5.


whether there is only a small shift in meaning, or whether we have tospeak about a real innovation.

Example: Isa. 31.4bMT So will the LORD of hosts descend to fight on mount Zion and

on its height.TgJon So will the royal power of the LORD of hosts be revealed to settle

on mount Zion and on its height.

TgJon avoids anthropomorphism, which is just normal practice for Tg.xniD^Q looks like a conceptual innovation, but it is another way to shieldthe LORD from a physical action, and the MT K3i: may have suggestedthe idea of 'royal power'.^'' On the other hand, the MT verb XD i, 'fight',has been rendered by •'"itc, 'settle', which is a more appropriate term for'royal power'. Lastly, MT views the near future, while TgJon seems tobe more eschatologically oriented. As a result, we may say that TgJondoes not break with MT at any one particular point, but the numberof shifts makes it an innovation in any

Additions

In this section I will discuss texts where Tg clearly adds something,without any textual base in MT. From a formal angle, this is the groupwhere Tg departs most from MT; but from a conceptual angle, analteration might be more surprising than an addition.

Obviously, additions do not eome out of the blue. Additions mayhave no textual base in MT, but they nevertheless have some logicalconnection with their context. There even exists a 'twilight zone' be-tween rather extended developments of the text ('alteration') and textswithout a basis in, but nevertheless a connection with, MT ('addition').

Example: 1 Sam. 15.29MT

TgJon And if you say, 'I will return from my sins and I will be forgiven,so that I can exercise the royal power, I and my sons, over Israelfor ever', . . .

Here we find the reasoning Saul might have deve^loped after Samuel'smessage, in order to continue to exercise royal power and safeguard

. Chilton, Clory of Israel, p. 77.type in Isa. 31.5 (MSS Reuchlin and Montefiore); 40.9.

LOREIN xniD' 'Q in the Targum of the Prophets 35

his dynasty.®^ In this way, even though this text has no basis in MT atall, it nevertheless fits as a logical explanation after vs. 28 and as anintroduction to vs. 29 in MT.™

On the other hand, many additions constitute real digressions. Insuch situations, TgJon will doubtless have had some reason for takingthat path, but the connection between the phrase with xniD' D and theoriginal text will sometimes remain beyond our horizon.

Prom the point of view of translation technique, it is of coursepointless to cite an example here. Moreover, all senses of NmD' Q arerepresented, as well as a whole range of referents. That is why I will giveonly one example which is important for the revealing of God's Kno' D.

Example: TosTg Hos. 12.4MT

TosTg And (tben) will be revealed the royal power of the LORD uponall the dwellers of the earth.

We have met this phrase earlier.^^ What is interesting is that we alsohave this universalistic variant in a Tosefta.''

two ideas go together in this case, but the combination with the verb 13iJbrings us to the translation given supra.

^°Same type in 2 Sam. 23.5 (in fine); Isa. 17.6 ('remnant' already mentionedin MT); 28.9 (answer to first half of the verse); Amos 9.11b; also in KTg 1 Sam.15.17 (arbitrary explanation, but occurring more often: see TAB); PalTg 1 Kgs 2.9(arbitrary explanation of this verse, but cf. Bacher, 'Kritische Untersuchungen', p.12); KTg Obad. 21; TosTg Obad. 21; TosTg Zech. 2.15 (particularistic 'correction'of the universalistic first half of the verse, influenced by Zeph. 3.8).

'•'V. supra, p. 20.^^Chilton, Glory of Israel, does not mention this text (provided by R. Kasher,

Targumic Toseftot, no. 133 11. 17-18, and by Y. Komlosh, 'D'amnn '7tD T '2m', 'TO,b2r -ISO [1957-'58], pp. 466-81 [475-76]; with thanks to Dr. A. Houtman for a copyof this almost irretrievable article).

Same type of addition in Sperber's edition of 1 Sam. 17.43 1 (the—theologicallyimportant—verb n i ' points to 'kingdom' in the territorial sense; cf. E. vanStaalduine-Sulman, 'The Aramaic Song of the Lamb', in: J.C. de Moor and W.G.E.Watson [eds.]. Verse in Ancient Near Eastern Prose [AOAT, 42; Kevelaer: Butzon& Bercker; Neukirchen-Vluyn: Neukirchener Verlag, 1993], pp. 265-92 [277]); 2 Sam.19.30 (Ed. Leiria); 1 Kgs 2.36 (MS Reuchlin: PalTgj; 10.18 (MS Montefiore and Ed.Leiria), 20; Isa. 41.2; 66.23 {bis; this Tosefta is only mentioned by P. Grelot, 'Deuxtosephtas targoumiques inedites sur Isai'e LXVF, RE 79 (1972], pp. 511-43 [536-37];the first occurrence in this Tosefta is interesting for the history of the interpreta-tion of Ex. 19.6: cf. M. McNamara, Targum and Testament. Aramaic Paraphrases ofthe Hebrew Bible: A Light on the New Testament [Shannon: Irish University Press;Grand Rapids: Eerdmans, 1972], pp. 148-59); Zech. 4.7 (MS Montefiore, which hasalso the form '" Jn"! earlier in this text); 11.7 (of course, Ezek. 37.15-28 is in the


Alterations

In this last group, Tg truly goes astray according to our exegeticalstandards. A possible reason for this is that the meturgeman did notunderstand MT, and while a commentator can always close his eyes tothe thorniest problems, a translator has to take a position.

Example: Ezek. 28.14aMT With an anointed cherub, which inhibited the access, I have

placed you.TgJon You are a king anointed for kingship and I have given you great-

ness.

While for MT we can offer only a provisional translation,''^ TgJon hasgiven a completely understandable rendering, basing himself upon whathe did recognize in MT. ID'PO seems to be introduced mainly as a figuraetymologica after ~pa, which itself seems to be a rendering of 3T1D. ^

Another explanation in many such instances is Tg's explicit desireto change MT for theological reasons. Much has already been writtenabout messianic texts,^^ and for this reason I will take my example forthis paragraph from the prophetae priores.

background of this text, but TgJon brings his interpretation into the text and de-velops this in his own way; cf. C.L. & E.M. Meyers, Zechanah 9-14 [AB; New York:Doubleday, 1993], pp. 262-64); and in Kasher's Toseftot edition of 1 Kgs 2.1 [bis);2 Kgs 4.1 {bis); Isa. 41.2 (in view of the form n:J3lK and the context, this is mostprobably a copyist's error; cf. also the main text of TgJon Isa. 41.2); 66.243, 3;Ezek. 1.1 {bis, if we reckon the three variants edited by Kasher [125S, 125:i, 1251]as only one Tosefta); Obad. 21.

^ Cf. B. Tidiman, Le livre d'Ezechiel, II (Commentaire Evangelique de la Bible;Vaux-sur-Seine: Edifac, 1987), pp. 66-69, and, in general i'or this passage, W. Zim-merli, Ezechiel, II (BKAT; Neukirchen-Vluyn: Neukirchener Verlag, 1969), p. 688.

^^Same type probably in Isa. 57.9 (cf. supra n. 17); Jer. 51.58 (cf. Hab. 2.13);Zech. 4.7b. Perhaps 1 Sam. 2.35 is also just an imperfect translation, but in thiscase rather based on a careless rendering of MT than on not understanding MT; itcould, however, also be an interpretational development (cf. S. Aalen, '"Kingdom"and "House" in Pre-Christian Judaism', NTS 8 [1961-62], pp. 233-40 [234-35], buta link of the two words by the common meaning 'dynasty' would have been moreto the point; R.P. Gordon, 'Translational Features of the Peshitta in 1 Samuel',in: K.J. Cathcart and M. Maher [eds.], Targumic and Cognate Studies: Essays inHonour of Martin McNamara [JSOTSup, 230; Sheffield: Academic Press, 1996],pp. 163-76 [173]).

''^J.J. Brierre-Narbonne, Exegese Targumique des Propheties Messianiques(Paris: Geuthner, 1936); Chilton, Clory of Israel, pp. 86-96; De Moor, '"Van wiezegt de profeet dit?'"; H. Hegermann, Jesaja 55 in Hexapla, Targum und Peschitta(BFCT, 2.56; Giitersloh: Bertelsmann, 1954); Levey, Messiah, pp. 33-103, 142-44;Levine, Aramaic Version, pp. 199-215; etc.

LOREIN xmD' D in the Targum of the Prophets 37

Example: 1 Sam. 10.7bMT As for you, do what your hand finds, for God is with you.TgJon Prepare for you royal utensils, for the Memra of the LORD is as

your help.

TgJon correctly does not like the idea that leaders are allowed to dowhatever comes to hand, and to avoid even the slightest possibilitythat the audience would suspect that this may be meant here, he con-sciously changes the meaning of MT by making TgJon speak aboutroyal garments and appearance.'^^

Certainly, all the texts that could be mentioned in this section areinteresting, but they do not tell much more about how Tg arrivedat the use of smD' D. ' Moreover, these interpretational developments

would be the sense of ]'3KD, according to Smolar and Aberbach, Studies,pp. 33-34.

For the general idea, v. TAB.For 'God is with you', v. supra p. 25.''^Same type in 1 Sam. 2.4 (The positive appreciation of the Hasmonaeans is not

normal [Levine, Aramaie Version, pp. 167-68; Smolar and Aberbach, Studies, pp.66-67], but occurs nevertheless also in other places: Levine, Aramaie Version; Smo-lar and Aberbach, Studies, p. 65; see also P.S. Alexander, 'From Poetry to Histori-ography: The Image of the Hasmoneans in Targum Canticles and the Question ofthe Targum's Provenance and Date', JSP 19 [1999], pp. 103-28; it could be used as adating argument, but is more problematic than K. Koch, 'Das apokalyptische Liedder Profetin Hanna. 1 Sam 2,1-10 im Targum', in W. Zwickel [ed.], Biblische Wel-ten. Festsehrift fiir Martin Metzger zu seinem 65. Geburtstag [OBO, 123; Freiburg:Universitatsverlag; Gottingen: V&R, 1993], pp. 61-82 [78-79], seems to imply.); Isa.7.8 (is TgJon less radical than MT in speaking about a—well structured—kingdominstead of speaking about a people?); 8.6 (the Waters of Shiloah—very strangely—considered as the political-military resources; cf. Levine, The Aramaic Version ofthe Bible, p. 23); 17.1; 46.3; 49.9; 53.3; 62.1 (TgJon makes a compromise betweenthe elliptic MT and a really concrete rendering; the main change is of course thedisappearance of God's rest: was there a problem with God [not] resting on theSabbath?); Ezek. 30.22 (two synonyms instead of two different categories); 31.4(MT iJOn and TgJon sniD' D both indicate a sphere of influence); Hos. 10.9 (by in-terpreting ~i}2i as regarding the crowning of Saul, TgJon starts to speak about thedynasty; perhaps the development has begun with a double translation of IDV ['tostand' and ; 'to resist']); Mic. 4.8 (the introduction of KrHD O is based on the secondhalf of the verse; cf. Le Deaut, Nuit paseale, pp. 276-77; Levey, Messiah, pp. 4, 92;Gamponovo, Konigtum, p. 426); Hab. 1.11 (cf. Gordon, Studies, p. 85; Smolar andAberbach, Studies, p. 155 n. 164); 3.17 (because MT is too poetical, TgJon describesthe history of Israel, in which the kingdom of Babylon plays its part, too; cf. LeDeaut, La nuit paseale, p. 360 n. 71); Zech. 5.9 (cf. Levine, The Aramaie Version ofthe Bible, p. 167); and also PalTg Isa. 21.5 (cf. Bacher, 'Kritische Untersuchungen',p. 20, for the insertion of the names).


ought to be placed within a diachronic framework, subject-matter foranother article.' *

General Remarks

Before concluding, it is useful to notice that the occurrences of xmD' aare not spread evenly across TgJon. Might this be due to the work ofdifferent meturgemanim? Or did a single meturg(3man translate with-out aiming at an absolutely consistent use of terms?

The metonymical sense of xniD' D is the one which most frequentlyoccurs where TgJon innovates, but the other senses of the word areobviously known too. In fact, the suffix nv produces an abstract noun,with the basic meaning of 'kingship' (purely transposed deverbativenoun vis-a-vis the verb ~p!2 'to be king'). All other meanings are toone degree or another metonymical, i.e., they represent a semanticshift, from abstract non-metonymical to concrete purely metonymical:kingship, royal power, royal dignity, dynasty, genitive used as we doadjectives and epitheton ornans, reign, realm (territory), kingdom andespecially its inhabitants.^^ So we cannot speak about 'the best trans-lation'. Nor have I found relations between translation techniques andactual English translations. Native speakers will not have felt thesedifferences in meaning explicitly; for them smD' a was just one singleconcept, albeit with variations.

The manuscripts seem to have had many problems in distinguishingbetween singular and plural. This phenomenon may have a theologicalexplanation, where the singular xm 'pn refers to Israel, while the plu-ral NmiDbn refers to the nations. Generally, however, lack of attentionseems to be the reason, so that we are allowed to choose the readingthat fits best in a given context.*"

course, much preparatory work has already been done, partly by the Kam-pen database of biblical texts cited or referred to in intertestamental and rabbinicaltexts.

'' Cf. E. Lipinski, Semitic Languages: Outline of a Comparative Grammar (Ori-entalia Lovaniensia Analecta, 80; Leuven: Peeters, 2nd edn, 2001), § 29.46 {'oftenan abstract or collective meaning'); J. Vergote, De rededelen als grondslagen vande algemene grammatica ([Leuven], n.d.), p. 31 (transposed nouns tend to a moreconcrete meaning); cf. Aalen, '"Kingdom" and "House"', pp. 233-36, 240. Withthanks to Professor W. Clarysse for helpful discussion in this matter.

The expression of 'ceasing to be a ID'PD' (cf. Isa. 7.8; 17.1, 3; Ezek. 29.4; Hos. 1.4;but cf. also Jer. 48.42!) still awaits discussion.

Isa. 2.2; 34.1; 37.22; 49.22; Amos 9.8; Mic. 4.1; Hab. 2.13. Cf. snpra n. 37.

LOREIN xmri'PQ in the Targum of the Prophets 39

Conclusions

A statistical table can help in formalizing conclusions. Inevitably, thedifferent categories used in this study are somewhat arbitrary; also,grouping the occurrences according to the biblical books may not re-flect accurately the division of labour among the different meturge-manim. On the other hand, we shall not be too far from the truth

Lit. Tr.Lit. + Clar.ShiftNoun/VerbOther Lit.Meton.Metaph.Clarif.Ep. Orn.Innov. w.

Innov. wo.

Addit.Alter.

S

Josh. Judg. Sam.

0.3 - 0.3

0.1

0.0

0.1

0.1 0.1

0.1

0.1

0.1

0.3 0.1 0.8

Kgs

0.4

0.2

0.1

0.0

0.1

0.5

-

-

0.1

-

1.5

Isa.

0.1

0.1

0.1

0.3

0.2

0.0

.81

0.1

0.1

0.1

1.1

Jer.

0.4

0.0

0.0

0.1

0.2

0.0

0.0

0.0

0.0

0.8

Ezek.

0.1

0.0

-

0.0

0.4

0.0

0.0

0.2

0.0

0.1

0.9

Hos.JoelJon.Zeph.

0.0

0.0

-

-

.

0.0

.

0.1

0.0

0.0

0.3

AmosMic.Nah.Hag.Zech.

0.1

0.0

0.1

0.1

0.2

.

0.0

0.1

0.1

0.1

0.9

Obad.

1

-

-

-

-

3

-

4

Hab.

-

0.7

-

0.7

1.3

0.2

0.1

0.0

0.1

0.2

0.1

0.0

0.1

0.1

0.1

0.9

if we assume a 'MT book ~ TgJon book' relation.*^ In order not tooverload this table, I have gone a bit further: after the computation,I have grouped some thinly populated categories,*^ and those minorprophets '* that stayed within the average range. Because the raw fig-

this category one case has disappeared vis-a-vis MT.® I would strongly dissuade the reader from re-calculating this table; results will

be different for at least one or two units.^''During a presentation of my research it was suggested to me that I should di-

vide the translation techniques into two main groups: those implied by the Aramaiclanguage and those chosen by the individual meturgeman. It would be very inter-esting if we could make such a distinction, but I am afraid that this is impossible;these are just common TgJon translation techniques.

^''This does not mean that I assume that all the minor prophets have beentranslated by the same meturgeman!

40 Aramaic Studies 3,1 (2005)

ures can easily give wrong impressions (presenting a book with onechapter along with a book with scores of chapters), I have, for eachbook, divided the occurrences by the number of chapters,^^

It will be clear how the minor prophets have been grouped: afirst group of minor prophets is clearly below the average, and a sec-ond group meets the average. At the end of the table, Obadiah andHabakkuk have to be mentioned separately because they present veryhigh scores. It must be taken into account, however, that they containonly one chapter (with three additions!) and four chapters respectively.

Judges and Malachi (which has zero occurrences) are clearly be-low the average, while Isaiah (with 19 occurrences of the metonymicaltype), Zechariah and Kings (with many literal translations and 23 epi-theta ornantia) join Obadiah and Habakkuk in the champions' league.

Jeremiah has the largest number of completely literal translations,which is especially remarkable for a prophet.

Contrary to my preliminary impression, Ezekiel just barely meetsthe average, mainly because of its large number of resolved metaphors(19).

In the second table, I try to make clear to what degree the Tar-gums have innovated, in totals, with occurrences per chapter betweenbrackets.

As far as Targum Jonathan is concerned, there are no innovationsat all in Joshua or in some of the minor prophets (including Obadiah),The meturgemanim of Judges, Samuel, Jeremiah and some other minorprophets have not been very active either. Kings (with his epithetaornantia), Isaiah (with his metonymical activity) and Ezekiel (with hisresolved metaphors) present the largest corpora, while Habakkuk andZechariah have one occurrence or more per chapter.

In the third column, we see that, relatively, Obadiah (namely vs,21) has attracted the most attention from targums different from Tar-gum Jonathan; but, surprisingly, the largest number come from Kings,which is also second in percentage terms. However, this third column isvery difficult to evaluate, because we have no idea how far the toseftotmentioned really represent original targumic activity.

The total number of occurrences seems to imjjly that the conceptof xmo'pQ was very important for Tg, However, this is not true. Aswe know, the main duty of TgJon is to give a translation which isclear for the audience, who does not understand the MT, and which

*^0,0 does not mean 'zero', but 'less than '0,05',

LOREIN xniD'PD in the Targum of the Prophets 41

Josh,Judg,Sam,Kgs,Isa.Jer,Ezek,Hos,JoelAmosObad,JonahMic,Nah,Hab.Zeph,Hag,Zech,Mai,

E

MT

7-

2226152271-31-11-12--

109

(0,3)

(0,4)(0,6)(0,2)(0,4)(0,1)(0,1)

(0,3)

(1)

(0,1)(0.3)

(0,3)

(1)

(0.3)

Tg

-3

1835541833214-1414--

16-

192

(0.1)(0.3)(0,7)(0,8)(0.3)(0,7)(0,1)(0,3)(0.4)

(0,3)(0,6)(0.3)(1,3)

(1,1)

(0,5)

Tos-,

--386-41--3------1-

26

Pal-, KTg

(0,1)(0.2)(0.1)

(0,1)(0.1)

(3)

(0,1)

(0.1)

does not put them onto wrong ideas. So TgJon always looks for easierexpressions than MT, and as soon as a word has been incorporatedin the TgJon vocabulary, it will have a greater frequency than its MTcounterparts. Even where the combination "Ti NHID' Q '" nn (or thelike) occurs,*^ TgJon does not really develop a theology of his own;normally, he expresses in plainer (and less anthropomorphic)^"" wordsthe intention of MT, The fact that this expression occurs with somefrequency in Tg, is a result of the tendency just noted,*^ TgJon wasnot preoccupied with the idea: in that case he should have spoken more

a, 24,23; 31,4, 5 (MSS Reuchlin and Montefiore); 40,9; 52,7; Obad, 21; Mic,4,7; Zech, 14,9; and also in TosTg Hos, 12,4, Partly in Ezek, 7,7, 10; and also inTosTg Zech, 4,7; KTg Obad, 21,

God's dominion is also mentioned in Isa, 60,20; Jer, 10,7; 49,38; and also in TosTgIsa, 66,23, but there the construction is clearly different,

® Le Deaut, Nuit pascale, p. 276,^^Le,, incorporation —» frequency,Cf, R,P, Gordon, 'Targum as Midrash: Contemporizing in the Targum to the

Prophets', in Proceedings of the Ninth World Congress of Jewish Studies. Panel


often about it, for example, in Isa, 2,2 and Ezek, 37,22, where the waywas open, but where TgJon did not use the opportunity.

The reader will have noticed that in this context I have trans-lated 'VT xrnD'pQ with 'royal power' or 'dominion'. Indeed, I departhere from the translation of Koch ('Reich'), Chilton ('kingdom') andLohfink ('Konigreich, Staat'),*^ I have not coined the translation 'royalpower' for this context only; in my opinion, it is also necessary in otherplaces,^" Camponovo moves in the same direction as I do, with histranslation 'Herrschaft',^^ At any rate, we have to keep in mind that atranslation is our problem: for native speakers, it is just KniD' O

In conclusion, although the meturgemanim have arrived at the wordKniD' Q via many different translation techniques and in situations thatdemand different English translations, the idea was not central to theirthinking. Where the wording seems to deviate from MT, very often themeaning does not.

Sessions. Bible Studies and Ancient Near East (Jerusalem: World Union of JewishStudies, 1988), pp, 61-73 (63),

^^Koch, 'Offenbaren wird sich das Reich Gottes', p, 159; Ghilton, 'Regnum DeiDeus Est', p, 265 and passim; N, Lohfink, 'Der Begriff des Gottesreiches vom Al-ten Testament hergesehen', in: J. Schreiner (ed,), Unterwegs zur Kirche: Alttesta-mentliche Konzeptionen (Quaestiones disputatae, 110; Freiburg: Herder, 1987), pp,80, 35 (not a targumic, but nevertheless an inspiring, article),

3°Judg, 9,9, 11, 13; 1 Sam, 14,47; 15,29; 1 Kgs 2,12, 46; 12,26; 15.13; 21,7; 2 Kgs14,5; Isa, 8,6; 17,3; 44,28; Jer, 48,4, 25, 38; Ezek, 17,23; 19,11, 14; 30,21, 22, 25{bis); 31,4, 6, 12; Mic, 4,8; Hab, 1,11; and also in PalTg Isa, 21,5,

^^Gamponovo, Konigtum, pp, 410, 428-31,

[AS 3.1 (2005) 43-58]DOI: 10.117/1477835105053514

Two NEW INCANTATION BOWLS FROM ROME (ITALY)*

Marco Moriggi

Universita degli Studi di Catania

1. The Collection

There are no large collections of incantation bowls in Italy. We nowhave knowledge of the collection honsed in the 'Museo Nazionale d'ArteOrientale' in Rome and another in possession of Ms. Louise Michail,an antique dealer in Milan.^ The specimens presented here are in theirturn kept in a private archive near Rome and belong to Prof. PaoloCosta.

*This contribution is a shortened and revised version of the paper 'A New Col-lection of Incantation Bowls from Rome (Italy)' read at the 'ARAMITH. AramaicLexicography' Conference (Sheffield, 23-25 July 2002). The author would like toexpress his gratitude to the owner of the collection, Prof. Paolo Costa (Univer-sita di Bologna) and his wife. Dr. Germana Graziosi, for their kind permission tostudy the bowls here presented. The author wishes also to thank Dr. Siam Bhayro(Yale University), organizer of the 'ARAMITH' Conference, for his kindness andthoughtful help. Prof. Shaul Shaked (The Hebrew University of Jerusalem) in ac-knowledgment of his precious support and useful suggestions. Prof. Fabrizio A.Pennacchietti (Universita di Torino) for his indispensable assistance at every mo-ment of both the research and the work on the text of this contribution. Thanksare due to Dr. Ezio Albrile for his constant support as regards cultural and biblio-graphical problems. Dr. Ruth Henderson (Universita di Torino), English languageconsultant, Mr. Stefano Dorato for computer assistance.

^M. Moriggi, 'Aramaean Demons in Rome: Incantation Bowls in the MuseoNazionale d'Arte Orientale', EW 51 (2001), pp. 205-228; Idem, 'Peculiarita hn-guistiche in una coppa niagica aramaica inedita', in P. Fronzaroli et al. (eds), Pro-ceedings of the 10th Italian Meeting of Hamito-Semitic (Afroasiatic) Linguistics(Quaderni di Semitistica; Florence: forthcoming). In the present paper I hese ab-breviations are used: n.e. = no editor; Aclr = Acta Iranica; BaM = BaghdaderMitteilungen; EW = East and West; JSAI = Jerusalem Studies in Arabic and Is-lam; Me = Mesopotamia; StEL = Studi Epigrafici e Linguistici sul Vicino Orienteantico.



The collection is made up of six bowls, all well preserved. One ofthem was broken, but has been completely restored. The bowls arennmbered from no, 1 to no, 6. Here is a short description of each one:

1) Hemispherical bowl, The bowl (diameter 165 mm, height 70mm)is almost completely covered by a stiff calcareous glaze, A text is writ-ten spirally from the inner bottom to the rim of the basin. This text, 12lines long, is enclosed within a line drawn on the rim, Aramaic squarescript text,

2) Hemispherical bowl. This specimen (diameter 160 mm, height70mm) is better preserved than the preceding one. Only a small partof the surface is covered by a soft pale-brown fiatina. At the innerbottom of the basin a demon is drawn. An Aramaic square script textdisplayed spirally runs from the demon's head tci the internal rim ofthe bowl. The text is 10 lines long. About the demon one may notethat, despite the small size of the drawing, it clearly presents a creaturewith long, curved horns coming out of a shroud which covers the head.On the head short hairs are stylised and, under a line (head-band?),there are two large eyes. The body is represented very simply, as onecan recognize arms and legs, but no other detail,^

3) Hemispherical bowl. This bowl (diameter 170 mm, height 55mm)is characterized by a greyish patina covering the bottom of the basin.The text is arranged in three sectors around the inner centre of thebowl. Each sector faces the others at right angles: this arrangement istypical of Mandaic bowls. There is a circle at the inner bottom andanother on the rim. This bowl contains a Mandaic formula,^

4) Hemispherical bowl. This one is the restored specimen and mea-sures 150 mm in diameter and 65 mm in height. Despite the severalbreaks, the restorer was skilful in putting the pieces back in the correctposition. The surface is clean, without any glaze. It is quite certain that

typology of the bowls is described according to the criteria establishedby E.C.D, Hunter, 'The Typology of the Incantation Bowls: Physical Features andDecorative Aspects', in J,B, Segal, Catalogue of the Aramaic and Mandaic Incan-tation Bowls in the British Museum (London: British Museum Press, 2000), pp,163-188,

^The iconography of this drawing fits with the general trends common to Ara-maic square script (Jewish Aramaic) bowls as defined by E.CD, Hunter 'Typology',pp, 173-176; Eadem, 'Who are the Demons? The Iconography of Incantation Bowls',StEL 15 (1998), pp, 95-115.

• Bowl no, 3 presents many doubtful readings and its study has proved to be avery hard task. It will be presented in another publication.

MORIGGI TWO New Incantation Bowls 45

this is a pseudo-script bowl. Large marks imitating letters (perhapsMandaic ones) are visible all over the basin. They are untidily written.

5) Hemispherical bowl. This bowl (diameter 155 mm, height 65mm) shows deep grooves on the inner surface. These are due to thefact that all these artefacts were wheel-turned. As in the case of bowlno. 4, we may consider this specimen a pseudo-script bowl: some fadedlines of a pseudo-text are barely visible in the basin.

6) Hemispherical bowl. This specimen (diameter 163 mm, height52mm) shows a badly scratched surface. The basin is studded withblack spots. What may be picked out are only shadows of letters, prob-ably pseudo-letters.

2. The Texts

Bowl no. 1: Transcription.^

[...] 'O ]m '0'^ ]Qi

]m

m n'

[...]

p

]vt "a

nnnm [cnin (l)i [...] (2)

] { } ][...] (3)]'Q'nn |in[...] x (4)

p^ [...] (5)

(6)

'nn(7)

nsiJ(-3-i) Konm [...]K I D ]nD[...] xanm [...] sann j^nnn (8)

Konm [...] [...] xanm (9)[...]!

^Pseudo-script bowls are present in many collections, but only in small num-bers. The first attempt to explain their existence was that they were forgeries whichmagicians sold to people without any education. This theory has been recently chal-lenged by another, according to which pseudo-script bowls have a magical signifi-cance as do the others, but it is hard for us to understand what this significance was.Gf. Sh. Shaked, 'Parole des anges, parole des demons: a propos des coupes magiquesde la Babylonie sassanide', in n.e., Compte Rendu de la soixante-treizieme sessionannuelle du Gomite de I'Union Academique Internationale (Bruxelles: SecretariatAdministratif de l'UAI, 1999), pp. 17-33 (19-20).

"In this and in the following transcription these symbols are used: (x) = uncer-tain reading, [x] = reconstructed reading, [.), [..] = one or two letters missing; (...]= three or more letters missing; {x} = scribal omission; <x> = editorial addition;(?) = doubtful translation; (...) = sequence without meaning.

D'nm (xon s^ao) ms3 xmno'si •'-i(D)n(S) n^ ]m ]rT -a NT ]'<n>m n^rrpi n^n'm (nnn) naom ICD)']! Dnnm (li)

Translation(1) Sealed and countersealed is the house [,,,] (2) and the threshold [,,,](3) [,,,] the seed of this Yzy' son of Zywn and Myhmhyd daughter ofQ'qy, his wife (4) their [,,,],

sealed and countersealed are from evil sorceries and from evil demonsand from evil plagues [,,,] (5) [,,,] and from curses and invocations andfrom evil idols and from female goddesses and from wrathful and pow-erful amulet-spirits,

(6) Sealed and countersealed are Yzy' son of Zywn and Myhm-hyd daughter of Q'qy, and their house and their threshold and theirproperty,

sealed they are in their entrance, sealed they are (7) in their exit,sealed in their beds and in their bedchambers, sealed from [,,,]

in the name of the three magical bonds and in the name of theseven (8) seals,

the seal [,,,] and the powerful seal [...]kryn son of [,,,] and the fourthseal of Nybryzy' son of Nyry (9) and the fifth seal [,,,] and the sixthseal [,,,] and the seal of the sun and the seal-ring of the moon, theseventh (?) [,,,] (10) [,,,] Amen Amen Selah,

Bound are and armed are and (magically) equipped are all cursesand invocations and pressed and subjugated are curses and demons,idols and goddesses in the name of the pure heaven.

And sealed (11) and countersealed and protected and doubly pro-tected are his son and his house and his property of this Yzy' son ofZywn and Myhmhyd daughter of Q'qy and Rwy' son of Zywn [...]daughter of Brdwd [,,,] (12) which they have, yes and Amen Selah,

Notes on howl no. 1Line 3) Myhmhyd. Feminine name, short for Myhr 'nhyd 'Mihr-Anahid',This name is not attested in other bowls in its entirety, but we some-times meet names beginning with Myhr-. They are cited in two Mandaicand in one Syriac bowl and they refer to a man,^ The combination of

J,A, Montgomery, Aramaic Incantation Texts from Nippur (Publicationsofthe Babylonian Section, 3; Philadelphia: University Museum, 1913), pp, 231-233(bowl no, 34); V,P, Hamilton, Syriac Incantation Bowls (Ann Arbor, MI: University

MORIGGI Two New Incantation Bowls 47

the male and the female deity is odd. Only the individual componentsMihr and Anahid occur in Middle Iranian onomastics.^

Line 4) sypty: 'plagues'. Loanword from Akkadian sibtu 'plague'.This word is attested as an Akkadian loan since the 9th century B.C.in the Tell Fekherye inscription (1. 23).^

Line 5) ptkry: 'idols'. Iranian loanword. In many texts ofthe magicalcorpus from Mesopotamia the term ptkry seems to be used as theIranian synonym of the Semitic (e.g. Mandaic) 'I'hy' 'gods', formingthe counterpart to the feminine demon group 'ystrt' 'goddesses'.^"

Line 5) hwmry: 'amulet-spirits'. Another etymology was proposedfor this substantive by S. Niditch. She supposed a derivation of the termfrom the root hmr ('to ferment') and translated hwmry as 'furies'."The origin of the possible connection between amulets and spirits is tobe located in the Mandaic tradition where: 'Die Mandaer sehen die Za-uberwirkung gewisser Gegenstande in den sie bewohnenden Damonenund benennen diese gradezu mit dem Namen jener; so brauchen sie[...] S'lDin [...] "Kiigelchen, Wirbel" (zu Amulettzwecken getragen) alsNamen gewisser boser Geister'.^^ The word hwmry' has been renderedas 'amulets' by Ch. Miiller-Kessler when studying a Mandaic lead roll(BM 132947: Is. 21-25). ^

Line 6) qynhwn: 'their property'. Probably a scribal omission for

Microfilms International, 1971), pp. 120ab, 156; Segal, Catalogue, pp. 125-126 (bowlno. 095M).

*Ph. Gignoux, Iranisches Personennamenbuch, II. Mitteliranische Personenna-men (Wien: Verlag der Osterreichischen Akademie der Wissenschaften, 1986), pp.42, 123; F. Justi, Iranisches Namenbuch (Marburg: N.G. Elwert'sche Verlagsbuch-handlung, 1895), pp. 15b, 207b-208a.

^J.C. Greenfield and A. Shaffer, 'Notes on the Akkadian-Aramaic BilingualStatue from Fekherye', Irag 45 (1983), pp. 109-116; Ch. Miiller-Kessler, 'Die Za-uberschalensammlung des British Museum', AfO 48/49 (2001/2002), pp. 115-145(140b).

"Cf. Sh. Shaked, 'Bagdana, King of the Demons, and Other Iranian Terms inBabylonian Aramaic Magic', Aclr 25 (1985), pp. 511-525.

^^S. Niditch, 'Incantation Texts and Formulaic Language: A New Etymology for"hwmry"'. Or 48 (1979), pp. 461-471.

^^Th. Noldeke, Manddische Grammatik (Halle: Verlag der Buchhandlung desWaisenhauses, 1875), p. 76 (n. 1). See also E.S. Drower and R. Macuch, A MandaicDictionary (Oxford: Oxford University Press, 1963), pp. 135b-136a.

i^Gh. Miiller-Kessler, 'Interrelations between Mandaic Lead Rolls and Incan-tation Bowls', in T. Abusch and K. van der Toorn (eds.), Mesopotamian Magic:Textual, Historical, and Interpretative Perspectives (Ancient Magic and Divination,1; Groningen: Styx, 1999), pp. 197-209 (200). Note that this scholar states that 'themeaning of hwmry' is not clear in this context' (n. 27).


qynynhwn (cf. Segal bowl no. 023A: Is. 1, 11). '' In the quoted bowl J.B.Segal translates qynynhwn as 'their property'. An analogous expressionis found in a Syriac text edited by Naveh and Shaked (no. 10: 1. 7). ^

Line 7) mypqhwn: 'their exit'. Here is the counterpart to m'[wlhwn]'their entrance', to be restored in the hne above. See the Mandaic par-allels mn 'ylwn wmn n 'pqwn 'from their entrance and from their exit'(bowl no. 078M: 14, restored) and 'ylh wn'pqh (bowl no. 088M: 13). ^The root npq is often present in incantation bowls. ^ We have many ex-amples of bowls buried under thresholds, i.e. where the entrance (andthe exit) of a house are physically located.^* In a bowl text publishedby Montgomery we find a direct allusion to demons 'qui logent dansleurs maisons et marchent sur leurs seuils' (no. 6: Is. 3-4). ^

Line 7) 'rsyhwn: 'their beds'. The expression is also found in an-other Aramaic square script text studied by Montgomery (no. 7: 1.17). ° The word is part of the account of the client's possessions andplaces to be protected from demons and can be compared to similarwords in Mandaic bowls, e.g. mn twhmh wmn ny 'kh wmn 'rsh wmnb 'ys 'dyh 'from his semen and from his sexual intercourse and fromhis bed and from his pillow' (bowl no. 18N97: 1. 9). ^ The bed, as thelocation connected to sexual relations between husband and wife, wasconsidered suitable for demons to hide in. This is the case of liliths,the female-demons always involved in the charms on bowls. A typicalexample of the lilith's behaviour, in which the succuba replaces thewoman of the house, is in a text recently published by Ch. Miiller-

l, Catalogue, pp. 63b-64b.J. Naveh and Sh. Shaked, Amulets and Magic Bowls: Aramaic Incantations

of Late Antiquity (Jerusalem & Leiden: The Magnes Press & E.J. Brill, 1985), p.277a.

'^See Segal, Catalogue, pp. 105ab, 116b-117a; Miiller-Kessler, 'Die Zauber-schalensammlung', pp. 130b, 134b.

^^Naveh and Shaked, Amulets, p. 274a; J. Naveh and Sh. Shaked, Magic Spellsand Formulae: Aramaic Incantations of Late Antiquity (Jerusalem: The MagnesPress, 1993), p. 270b.

i^R. Venco Ricciardi, 'Trial Trench at Tell Baruda (Choche)', Me VIII/IX(1973/1974), pp. 15-20 (19).

i^J.N. Epstein, 'Gloses babyloarameennes', REJ LXXIII (1921), pp. 27-58 (34).^"Montgomery, Aramaic Incantation Texts, p. 147.^^E.C.D. Hunter, 'Two Mandaic Incantation Bowls from Nippur', BaM 25 (1995),

pp. 605-620 (613-615).

Kessler. In this story Bguzan-Lilit enters the house and the bed of aman causing him misfortune and sorrow, ^

Line 7) myskbyhwn: 'their bedchambers'. The word is logically con-nected with the preceding one. Other occurrences are found in Mont-gomery's nos, 7 (1, 7), 8 (1, 5) and 9 (1, 8), W,H, Rossell prefers formyskb' the meaning 'couch, bed',^^

Lines 7-8) bswm (tlth) <'>sryn wbswm s{b'h) htmyn: 'in the nameof the three magical bonds and in the name of the seven seals', Asimilar sequence is in another Aramaic square script bowl (Naveh andShaked's no, 14: 1, 5) that reads: htym bsb'h htmyn wmhtm btlt' swrynrb{rbyn) 'sealed by seven seals and firmly sealed by three large walls',^''

Line 9) symsh: 'the sun'. The form syms' is present in Naveh andShaked's no, 7 (1, 7 Aramaic square script) and no, 26 (1, 5 Syriac),^^Magic texts such as the one under discussion here tend to have - ' forthe ending, but sometimes the -h slips through. The use of an -h toexpress /-a/ of the masculine singular emphatic state is well attested inthe language of Aramaic square script texts on bowls, as already statedby scholars, H, Juusola thinks that this trait (common to Nedarim andGeonic Aramaic) should be connected with the 'conservative nature'of the language of 'Jewish Babylonian Aramaic' incantation bowls, ^One may notice that this text shows here symsh for hm' 'heat (?) <ofthe sun>' in the parallel passage of Naveh and Shaked's bowl no, 14(1, 6: bhm' b 'zqt' dsyhr'). The word hm' is also attested in a Hebrewpassage inserted in a magic bowl from Nippur studied by Ch, Miiller-

h, Miiller-Kessler, 'The Story of Bguzan-Lilit, Daughter of Zanay-Lilil;', JA OS116.2 (1996), pp, 185-195,

^'''W,H, Rossell, A Handbook of Aramaie Magical Texts (Shelton Semitic Series,2; Ringwood Borough, NJ: Shelton College, 1953), p, 151a.

^"•Naveh and Shaked, Magie Spells, pp, 113-114, One frequently comes across themagic number 7 in connection with seals, knots, tongues and demons in the frame-work of Mesopotamian magic, see Ch, Miiller-Kessler, 'Aramaische Beschworungenund Astronomische Omina in nachbabylonischer Zeit: Das Fortleben mesopotamis-cher Kultur im Vorderen Orient', in J, Renger (ed,), Babylon: Focus mesopotamis-cher Geschiehte, Wiege friiher Gelehrsamkeit, Mythos in der Moderne (Colloquiender Deutschen Orient-Gesellschaft, 2; Berlin: Saarbriicker Druckerei und Verlag,1999), pp, 427-443 (432, 442); Eadem, 'A Mandaic Gold Amulet in the BritishMuseum', BASOR 311 (1998), pp, 83-88 (85),

25Naveh and Shaked, Amulets, pp. 169-170; Naveh and Shaked, Magic Spells, pp,139-140,

^''H, Juusola, Linguistic Peculiarities in the Aramaic Magic Bowl Texts (Stu-dOr, 86; Helsinki: Finnish Oriental Society, 1999), pp, 31, 141; Naveh and Shaked,Amulets, pp, 31-32,

Kessler, who reads it bswm br' brsyt glgl hm' and translates it 'inthe name of the one who created in the beginning the crescent of the

Line 9) 'yzqt' dsyhr': 'the seal-ring of the moon'. An analogous se-quence appears in Naveh and Shaked's bowl no, 14 (1, 6: b'zqt' dsyhr','by the ring of the moon'). The seal- or signet-ring is a prevalent fea-ture in incantation bowls. Its uses have been clarified by B,A, Levinewith references to Jewish and Mandaic traditions,^^

Line 11) nyt,r: 'protected'. Probably a scribal error for ntyr.

ColaThe spell contained in bowl no, 1 can be divided into the followingcola:la) h[tym] wmhtm... ' [...]hwn (Is, 1-4): opening formula (people andthings to be protected by the exorcism);II) htymyn ... tqypt' (Is, 4-5): the object of exorcism;Ib) htymyn ... wqynhwn (1, 6): opening formula (repeated in the plu-ral);III) htymyn ... pwn [...] (Is, 6-7): places to which the exorcism is espe-cially devoted;IVa) bswm (tlth) <^>sryn wbswm s{b'h) htmyn (Is, 7-8): invocation tothe three bonds and the seven seals;IVb) htm' [,,,],,, 'mn 'mn [sl]h (Is, 8-10): detailed description of theseals;V) 'syryn ... {smy' dky') (1, 10): the object of exorcism;Ic) whtym... w'mn slh (Is, 10-12): opening formula (repeated in thesingular).

It is worth indicating parallels of some of these cola found in otherbowl texts: I) Hyvernat; Gordon F; Gordon E: Is, 1, 6; ^ Gordon no. 3:

Shaked, Magic Spells, pp, 113-114; Ch, Miiller-Kessler, 'The EarliestEvidence for Targum Onqelos from Babylonia and the Question of Its Dialect andOrigin', JAB 3 (2001), pp, 181-198 (195, n, 60), See also the discussion concerningthe etymology and the meaning of Syriac hmm,' '(intensie) heat' in Sh, Shaked,'Manichaean Incantation Bowls in Syriac', JSAI 24 (2000), pp, 58-92 (72),

^^Naveh and Shaked, Magic Spells, p. 113; B,A, Levine, 'The Language of theMagical Bowls', in J, Neusner, A History of the Jews in Babylonia, V (SP.B, XV;Leiden: E,J, Brill, 1970), pp, 343-375 (364-368),

•^^Hyvernat's bowl is presented synoptically with Gordon E and E in C,H, Gor-don, 'Aramaic Magical Bowls in the Istanbul and Baghdad Museums', ArOr VI(1934), pp, 319-334 (331-332), Hyvernat's and Gordon E bowls do not bear line-numbers.


1. l;30 Montgomery nos. 21: 1. 1; 22: 1. 1; 23: 1. 1; 30: 1. V?^ Geller AaronF: 1. 1;32 Smelik (9163 Allard Pierson Mus.): 1. 1; ^ Moriggi (Michailbowl): 1. 1.34 Cf j igQ gggj^i j Qg Q^5^. ig 3_4. Q jg^ . J ^. Qj^7^. J 7.35

Naveh and Shaked no. 15: Is. 1-4. "II) Hyvernat; Gordon no. 3: Is. 1-2; Montgomery nos. 21: Is. 1-2; 22: Is.1-2; 23: Is. 1-2; Segal nos. 014A: Is. 3-5; ^ 015A: Is. 4-6; 016A: Is. 2-6.Ib) Hyvernat.IVa) Gordon no. 3: 1. 3; Montgomery nos. 21: 1. 3; 22: 1. 3; 23: 1. 3;Naveh and Shaked no. 14: 1. 5. ^ Cf. also Hyvernat; Gordon F; GordonE: Is. 1-2, 6; Montgomery nos. 5: Is. 2-3; 19: 1. 4; ^ Naveh and Shakedno. 14: Is. 4-5.IVb) Naveh and Shaked no. 14: Is. 5-6; IM 62265: Is. 6-8.''° As regardsthe sentence whtm' dsymsh w 'yzqt' dsyhr', see Naveh and Shaked no.14: 1. 6 {bhm' b'zqt' dsyhr').

Texts beginning with htym wmhtm are common to the corpus ofAramaic square script incantation bowls. As a matter of fact T. Harvi-ainen stated that, among them, Montgomery nos. 21, 22, 23, 30, Gor-don no. 3 and Geller Aaron F 'do not reveal [...] Jewish features'.''^This seems to be true for the text presented here, too. Harviainenneither included such beginning formula in the group comprehendingthe ones he considered typically Mandaic, nor stated that it could beshared between the Jewish and the Mandaic cultural environments.^^

^°C.H. Gordon, 'Aramaic Incantation Bowls', Or X (1941), pp. 116-141, 272-284,339-360 (120-121).

^^Montgomery, Aramaic Incantation Texts, pp. 203, 221.^^M.J. Geller, 'Eight Incantation Bowls', OLP 17 (1986), pp. 101-117 (115-116).'^^K.A.D. Smelik, 'An Aramaic Incantation Bowl in the Allard Pierson Museum',

BO XXXV (1978), pp. 174-177 (177).''''Moriggi, 'Peculiarita linguistiche', forthcoming.^^Segal, Catalogue, pp. 57a-59b.^^Naveh and Shaked, Magic Spells, p. 115.•''^Miiller-Kessler, 'Die Zauberschalensammlung', p. 121a.^^Naveh and Shaked, Magic Spells, p. 113.^^Montgomery, Aramaic Incantation Texts, pp. 138, 195.^°Bowl IM 62265 is housed in the Iraq Museum (Baghdad). A study of this

specimen by A.H. Faraj and M. Moriggi is forthcoming.''^T. Harviainen, 'Pagan Incantations in Aramaic Magic Bowls', in M.J. Geller et

al. (eds.), Studia Aramaica: New Sources and New Approaches {JSS Supjjlement,4; Oxford: Oxford University Press, 1995), pp. 53-60 (57). Cf. Idem, 'Syncretisticand Confessional Features in Mesopotamian Incantation Bowls', in n.e.. AncientNear East and India (StudOr, 70; Helsinki: Finnish Oriental Society, 1993), pp.29-37 (35).

"•^Harviainen, 'Pagan Incantations', pp. 54-58 passim.

The text of bowl no. 1 shares many sections with other bowl texts,but it is characterized by interpolations like Ib, Ic and V which are notfound in the parallel texts. The magician mingled two (or three, seeIVa and IVb) formularies according to the principle that the scribesof incantation bowls 'relied on memory rather than a hard copy setbefore

Bowl no. 2: Transcription(xn)n'Oi<] [...] n n o s -iTNiJnDi ]'t-a rxt,(7-:i) xni-ox (l)

[...] D^lSDSn nm^'S D'lSOK TDS PK' 'iJ (2)

' s i xn I'ox XDIOQT sjnox sspin TON K3O>ID TON (4)X]-10K (SSpn) X3D10 TDX

-13 ]xi)s im ivt nn -iio-o ]oi orain'o (xpao (Tsmnxn ]a NDIOQT (5)

'x (6)

S'n'a n^Di sntrn'^i snnnp xncoibi (7)

mn n33i m s '33 '^DI snoi':' ':'3 ]n3'Q N3nom X3-IDS (8)

DX

(9)

n[...](D) '3X1 rr3 -lon'ST SID'S sin n3 [...] (lO)

Translation1) Bound is 'yl't in the mysteries and kh"zr bound is [...] bound is(2) 'yl't, bound is the supreme Aspades with Aspades [...] (3) [...] inthe bond of eternity, bound is the mystery of the sun in the mystery ofthe moon, bound is the mystery of the moon in the mystery of the sun(4) bound is the star, bound is the power. I bind and I strongly bind,bound is the mystery of the sun in the mystery of the moon, bound isthe powerful star.

I bind (5) and I strongly bind from b 'rtyz 'yd (...) and from sysnwyson of zywn and from pryn son of sysyn.

I bind and I strongly bind from them and from their houses andfrom (6) their threshold and from the sons they have and they willhave from this day and forever.

''•'D. Levene, 'Heal O' Israel: A Pair of Duplicate Magic Bowls from the PergamonMuseum in Berlin', JJS LIV (2003), pp. 104-121 (Ul) .


I bind and I strongly bind from them the curse of the open field andthe curse of the village and the distant curse (7) and the near curseand the curse of the house of the dead and the curse of the cemeteryand the curse of the open field and the curse of the village. I bind andI strongly bind from them all evil sorceries and powerful magical acts.(8) I bind and I strongly bind from them all curses of all the sons ofAdam and the daughters of Eve who are called sons of man. I bind andI strongly bind from them all evil sorceries and powerful magical acts.I bind and I strongly bind from them (9) all curses and invocationsand pain (?) and pain (fem.) (?) and [...] from them evil (plur.). I bindand I strongly bind from them (...) and [...]

10) [...] in it, that bond (by) which was bound the house of weapons

Notes on bowl no. 2Line 3) 'swr: 'bond, binding'. This form is a common variant of 'yswr'.It is also found in the Syriac bowl no. 16 (1. 4) published by J. Navehand Sh. Shaked.^''

Line 3) 'wl'm: 'eternity'. This word also means 'world, saeculum'.Our translation is based upon the context.''^ Such Hebrew-influencedspelhngs of 'wlm are frequently attested.

Line 3) 'syr rz ' dsymsh br{z)' dsyhr' 'syr rz ' dsyhr' brz ' dsymsh:'bound is the mystery of the sun in the mystery of the moon, bound isthe mystery of the moon in the mystery of the sun'. This sequence isrecalled in other incantation bowl texts where 'mysteries' and celestialbodies are concerned. See for instance br'zh rbh drqy'h 'by the greatmystery of the firmament' (Naveh and Shaked's bowl no. 14: 1. 7); brzsmyh 'by the mystery of heaven' (Naveh and Shaked's bowl no. 21: 1.

and Shaked, Magic Spells, p. 118.''^Montgomery, Aramaic Incantation Texts, p. 298a; Naveh and Shaked. Magic

Spells, p. 271b.'"'Naveh and Shaked, Magic Spells, pp. 113,128; The idea of 'binding' or 'burying'

opposite elements (sun-moon; earth-heaven) is also attested in a group of Syriacbowls where we read rz smy' bsmy' qbyr wrz 'r" b'r" qbyr ('the mystery ofheaven is buried in heaven and the mystery of the earth is buried in the earth').See V.P. Hamilton, Syriac Incantation Bowls, pp. 110a-110b (no. 9: Is. 4-5); T.Harviainen, A Syriac Incantation Bowl in the Finnish National Museum, Helsinki:A Specimen of Eastern Aramaic "Koine" (StudOr, 51.1; Helsinki: Finnish OrientalSociety, 1978), pp. 6 (no. HB: Is. 3-4), 8 (no. 1MB: Is. 3-4); Naveh and Shaked,Amulets, pp. 124-125 (no. 1: Is. 3-4).

54 Aramaie Studies 3.1 (2005)

Line 4) 'srn': 'I bind'. Cf. the sequence 'srn' lyk in an Aramaicamulet (no. A28: 35), studied by J. Naveh and Sh. Shaked.'''

Line 5) sysnwy. The same name was detected by CH. Gordon inan Aramaic square script bowl previously edited by J. Obermarm.^*

Line 5) sysyn. Feminine name well attested in incantation bowl

Line 5) btyhwn: 'their houses'. I translate as plural, since the suffixpronoun 3rd person masculine plural -yhwn.^'^

Line 6) dnyn: 'this'. This demonstrative occurs in another bowl(Naveh and Shaked's no. 25: Is. 4, 7). Variant of the more common

Line 6) dhr': 'the open field'. The translation is based upon a par-allel sequence found in Naveh and Shaked's bowl no. 2 (1. 5: bdhr'wybmt', 'in the open field and in the village').

Line 10) The sequence 'ysr' d'ytsr: 'bond (by) which was bound'finds its parallels in 'yswr' d'tsrw: 'bond (by) which were bound'(Montgomery's no. 4: 1. 4; Gordon's no. 11: Is. 13-15). ^ The word'ysr' is a variant of 'yswr' (see Naveh and Shaked's no. 4: 1. 2; Segalbowls nos. 007A: 1. 12; 008A: 1. 8).

The sequence byt z 'ny ('the house of weapons') corresponds toMandaic byt zyn' and byt z 'yn' found respectively in Lidzbarski's no.Ic: 1. 17 and Lidzbarski no. II (end).^^

ColaThe text of bowl no. 2 can be divided into the following cola:I) 'syrt' ... twqp' (Is. 1-4): opening formula (binding of entities andcelestial bodies).II) 'srn' wmsm' (1. 4): textual interpolation.I) 'syr ...(tqp') (1. 4): opening formula.

•*^Naveh and Shaked, Magic Spells, pp. 97-98.**J. Obermann, 'Two Magic Bowls: New Incantation Texts from Mesopotamia',

AJSL LVII (1940), pp. 1-31 (6); Gordon, 'Aramaic Incantation Bowls', p. 117.'' Cf. Montgomery, Aramaic Incantation Texts, p. 280b; E.M. Yamauchi, Mandaic

Incantation Texts (AOS, 49; New Haven, CT: American Oriental Society, 1967), p.371.

sORossell, Handbook, p. 38.^^Naveh and Shaked, Magic Spells, pp. 137-138. For occurrences and interpreta-

tions of dnyn, see Juusoia, Linguistic Peculiarities, pp. 103, 114.^^Montgomery, Aramaic Incantation Texts, p. 133; Gordon, 'Aramaic Incantation

Bowls', p. 273.^^M. Lidzbarski, Ephemeris fiir semitische Epigraphik, I (Giessen: J. Rickersche

Verlagsbuchhandlung, 1900-1902), pp. 94-95, 98-99.


II) 'sm' wmsm' ... sysyn (Is, 4-5): people to be protected,II) 'srn' wmsm' ... wl'lm (Is, 5-6): protecting the house and childrenof the clients.II) 'srn' wmsm' ... w'[..\ (Is, 6-9): the objects of exorcism {'sm'wmsm' repeated 6 times),III) bh hw' 'ysr' d'ytsr byt z'ny (m)[,,,] (1. 10): part of a closingformula (?),

This text is made up of two formulae which are joined to one an-other in the passage between lines 4 and 5 (the sequence 'sm' wmsm'in line 4 is to be considered an unintentional interpolation). The firstformula seems to have a Mandaic Vorlage (cf, the similar - longer -sequence in the Mandaic bowl no, 32: Is, 4-10, published by Yamauchi)but lacks the usual doxological introduction bswm' dhyy ', '' This is atypical phenomenon when a Mandaic formula is adapted in an AramaicSquare script text. According to T, Harviainen 'no Jewish Aramaicbowl text begins with the Mandaean formula "in the name of Life'",^^

The second formula consist of a recurrence of the two participles'sm' wmsm' introducing the names of clients, things and people tobe protected and finally the objects of exorcism,

A final consideration is due to the language of these two bowls.The traditional label assigned to the language of Aramaic Square scriptbowls is 'Jewish Babylonian Aramaic', Two other definitions have beenproposed in the last few years. The first, 'Eastern Aramaic Koine', wassuggested by T, Harviainen and intended to focus on the very closeconnections between the bowl texts written in Mandaic letters and theothers (Jewish Aramaic, Syriac), Along with this theory runs the ideaof 'an amalgamation of closely related dialects (Greek Koine)' develop-ing in 'an area (Central Mesopotamia) with a mixed population','^^ Thesecond proposal was introduced by Ch, Miiller-Kessler and Th, Kwas-

^^Yamauchi, Mandaic Incantation Texts, pp, 292-293,^^Harviainen, 'Pagan Incantations', p, 55, Cf, Harviainen, 'Syncretistic and Con-

fessional Features', pp, 34-35, The text of bowl no, 2 is characterized by otherMandaic features such as: bny 'dm wbnt hwh dmytqryn {bny 'ns') 'the sons ofAdam and the daughters of Eve who are called sons of man' (1, 8) and byt z 'ny'the house of weapons' (1, 10),

^^Harviainen, A Syriac Incantation Bowl, p. 27, The theory of 'Eastern Ara-maic Koine' was developed along with the work of T, Harviainen on Syriac andAramaic Square script bowls. For comments on 'Eastern Aramaic Koine', see L,Van Rompay, 'Some Remarks on the Language of Syriac Incantation Texts', in R,Lavenant (ed,), V Symposium Syriacum (OCA, 236; Rome; Pontificium IstitutumStudiorum Orientalium, 1990), pp, 369-381 (372-373),


man. The two scholars focused on the concept of 'Standard LiteraryBabylonian Aramaic', a language represented in the Talmud tractatesNedarim, Nazir, Kerithoth, Me'ilah, Tamid which is also found in the'majority of bowls studied so far', ' They considered that only a smallnumber of bowls were written in the 'Koine' and only one bowl in whatshould be called 'Talmudic Aramaic',^*

According to the linguistic features detectable in our text relatedto the criteria established by these scholars, we should define the lan-guage of these two bowls as 'Standard Literary Babylonian Aramaic',Actually they present the following features: preservation of gutturalsand of word endings (in contrast with weakening of the gutturals andapocopated forms of Talmudic Aramaic), singular masculine emphaticstate ending both in -h and - ' (in Eastern Aramaic Koine one shouldalways find - ' ) , 3rd person masculine imperfect introduced by y (inEastern Aramaic Koine one can also find n- or ^),^^

h, Miiller-Kessler and Th, Kwasman, 'A Unique Talmudic Aramaic Incanta-tion Bowl', JAOS 120 (2000), pp, 159-165 (159ab),

^^Miiller-Kessler and Kwasman, 'A Unique Talmudic Aramaic Incantation Bowl',p, 159ab,

^^For a list ofthe features distinguishing 'Talmudic Aramaic', 'Standard LiteraryBabylonian Aramaic' and 'Eastern Aramaic Koine' (also called 'Koine BabylonianAramaic') see Miiller-Kessler and Kwasman, 'A Unique Talmudic Aramaic Incanta-tion Bowl', pp, 159b-160b and (only as regards 'Koine') T. Harviainen, An AramaicIncantation Bowl from Borsippa: Another Specimen of Eastern Aramaic "Koine"(StudOr, 51,14; Helsinki; Finnish Oriental Society, 1981), pp, 23-24, A study ofthedifferences between 'Standard Literary Babylonian Aramaic' and 'Koine Babylo-nian Aramaic' is announced in Ch, Miiller-Kessler, A Handbook of Magic Bowls inKoine Babylonian Aramaic (Croningen; Styx, forthcoming).

[AS 3.1 (2005) 59-81)DOI 10.117/1477835105053515

THE RELATIONSHIP OE THE PESHITTA TEXT OESECOND SAMUEL WITH THE PESHITTA TEXT OF

FIRST CHRONICLES

Graig Morrison

Pontifical Biblical Institute, Rome

Introduction

When, in the course of interpreting the Hebrew Bible, the text criticperceives that the Masoretic Text has suffered corruption, recourse isoften made to the biblical versions. At times, the versions may preservereadings that reflect a Hebrew text different from the MT. Other times,the versions offer various interpretations for a difficnit Hebrew passage.But before individual readings from the versions are compared to theMT, the text critic must be aware of the character and translation tech-niques of the version in hand. In the case of the Peshitta. given that itsorigins and early textual history remain obscure, further considerationmust be given to the possibility that its translation was influenced bythe Septuagint or the Targum, These are some of the current questionsthat spur research into the character of the Peshitta as a translation.

Second Samuel offers its own particular challenges since several ofits episodes are retold in First Ghronicles. This fact raises the ques-tion as to whether, at the time of the Peshitta's translation or duringits transmission, episodes in the Peshitta of 2 Samuel were harmo-nized with corresponding passages in the Hebrew or Peshitta text of1 Ghronicl(>s, The related question, whether Peshitta 1 Ghronicles wasinfluenced by the Hebrew or the Peshitta text of 2 Samuel, is not underexamination.' These two questions should be kept separate, since a listof agreements between 2 Samuel and 1 Ghronicles can be deceiving if

'Thus, instances where Pesbitta 1 Chronicles departs from the Hebrew text ofClironicles and agree witb tbe Hebrew text and/or tbp Peshitta text of 2 Samuelare not considered. For example, tbe Peshitta text of I Ghron. 1G,3 adds

© 2005 SAGE Publications (London, Thousand Oaks C.- , and New Delhil

the shared readings are not first evaluated as examples of translationtechniques proper to the character of the Peshitta version of 2 Samuelor 1 Chronicles. When snch internal explanations are overlooked, oneruns the risk of drawing a conclnsion on the basis of agreements thatare patient of several explanations and are not necessarily examplesof borrowing. Each reading presented below is first considered in lightof the character of the Peshitta translation of 2 Samuel before it isproffered as evidence of borrowing.^

The harmonization of parallel passages in the Bible was given briefconsideration by M.P. Weitzman in a study of duplicate texts in thePeshitta. He ootisidered the question from the point of view of PoshittaChronicles and concluded that in 'Chronicles...where the translatorlooks to au earlier book hv always consults the Hebrew text; and al-though he shares with his predecessors an idiosyncratic understandingof certain Hebrew words and certain characteristic choices of expres-sion, there is no evidence that he consulted their translations at all."^This paper presents the instances in which Peshitta 2 Samuel departsfrom the MT and its reading agrees wdtli the Hebrew text or Peshittatext of 1 Chronicles in order to describe the relationship (if there isone) between the Peshitta translations of 2 Samuel and 1 Chronicles.''Does the Peshitta of 2 Samuel reveal traces of influence from Peshitta1 Chronicles?

At the outset of this study, several possibilities with varying de-grees of plausibihty are allowed for explaining the agreements presentedbelow. The translator of Peshitta 2 Samuel could have been influ-enced, consciously or unconsciously, by the Hebrew text of Chronicles

.xjr<r r^J i i oiiji in agreement witli the Hebrew and Peshitta texts of 2Sam. 6.19: in'3'7 D"N CVrr^D "p"!. But because the Peshitta text of 2 Samuel is notinvolved, this agreement is not considered.

^The study hy D, Englert {The Peshitto of Second Samuel [.IBL MonographSeries, 3; Philadelphia: The Society of Biblical Literature, 1949]). an importantcontribution for its time, does not go very far in revealing the character of thePeshitta of 2 Samuel.

•'M.P. Weitzman, The Syriac Version ofthe Old Testament (Cambridge: Cam-bri(]ge University Press, 1999), p. 2(12.

'Recently E, van Staalduine-Suiman has identified harmonizations between Tar-gum Samuel and Hebrew Chronicles. She writes: 'Only where the text of Samuelcan be clarified by the version of Chronicles within the boundaries of the iicceptedexegetical techniques [techniques that she identifies], TJon makes use of Chron-icles" (The Targum of Samuel [Studies in the Aramaic Interpretation of Scripture,1; Leiden: E.J. Brill, 2002], p. 133).

MORRISON Peshitta 2 Samuel and 1 Chronicles 61

or its Peshitta translation. Weitzman's hypothesis that the Peshittaof Chronicles (c. 200) was translated about fifty years after Samuel^make.-s it impossible for the younger Peshitta translation of 1 Chron-icles to have influenced the older Peshitta translation of 2 Samuel. Butthere is very little data to support his theory, however logical it mayseem to our minds. Moreover, it is not necessary to accept his timelineof translations in order to proceed with this reseai'ch since, if there isclear evidence that can demonstrate which book was translated first,it will emerge in the evaluation of the agreements below. Thus, thepossibility, however remote, that Peshitta 2 Samuel was influenced byPeshitta 1 Chronicles is, for the moment, allowed to stand. Similarly,with regard to influence during transmission, it is less plausible thatPeshitta 2 Samuel was influenced by the Hebrew text of Chronicles.More plausible is the possibility that Peshitta 2 Samuel was influencedby the Peshitta text of Chronicles. But again, at the outset of thisstudy, both possibilities are left open until the evidence presented be-low is evaluated.

Method

The instances where the Peshitta text of 2 Samuel departs from theMT were noted. These divergences were then compared with the MTand the Peshitta text of 1 Chronicles. Instances where the Peshitta of2 Samuel agrees with the Hebrew text or Peshitta text of 1 Chroniclesagainst the Hebrew text of 2 Samuel were identified and are discussedbelow, (Minor agreements, such as those of word order, were not con-sidered.) These divergent readings were then evaluated to determinewhether they witness to borrowing or influence from Chronicles. Ineach case, possible explanations for the Peshitta reading are proposed.The possibilities that the Peshitta translator of 2 Sanniel was influ-enced by the Hebrew/Peshitta text of 1 Chronicles or that a latercopyist, transmitting 2 Samnel, was influenced by the Peshitta or, lesslikely, the Hebrew text of 1 Chronicles is only considered after otherexplanations are discounted. The divergent Peshitta readings were alsochecked against the Greek text {the Cambridge edition and its appara-

^Weitzman, The Syriac Version ofthe Old Testam,f:nt, p. 258.

tus^) and the Targum^ as agreement with these two versions allows forother explanations, such as influence from the LXX or Targum on thePeshitta (a question still to be studied in 2 Samuel) or the possibilitythat the Hebrew Vorlage behind the Peshitta diverged from the MT.

Discussion of Identified Agreements

2 Sam.. 5.2 and 1 Chron. 11.2

2 Sam. 5,2'7Niw:'"'/r -[-yb rrnr. nnsiAnd you will be leader over Israel.

And you will be leader over my people Israel.

1 Chron. 11,2

And you will be leader over my people Israel,

And you will be leader over all the tribes of Israel.

The Peshitta text of 2 Sam. 5.2 adds »in-a- in agreement with the Hebrewtext of 1 Chron. 11.2 and against its Peshitta translation. The additionin the Peshitta text of 2 Sam, 5.2 can be explained as a harmonizationto the phrase \*lm*r<r »3aj^ C^N S'Tix "DlJTiN) that appears earlierin this same verse. There is no need to reach for the Hebrew t.ext of1 Chron, 11.2 to explain the addition of ^M_S.. Harmonization oftenaccounts for divergent readings in the Peshitta.*^

2 Sam.. 5.9 and 1 Chron. 11.7

2 Sam. 5.9

Then David lived in the stronghold and he called it the city of David.

^A.E. Brooke f.t al. (eds.). The Old Testament in Greek Aecording to ihe Textof Codex Vaticanus., I. The Octateueh. II. The Later Historical Books. 111,1, Esther,Judith, Tobit (Cambridge: Cambridge University Press. 1906-1940),

^A. Sperber (ed.), The Bible In Aramaic (4 vols,; Leiden: E.J. Brill. 19IJ2) andR, Le Deaut, Targum des Chroniques (Cod. Vai. Vrb. Ebr. 1) (AnBib. 51; Rome:Biblical Institute Press, 1971).

"CE, Morrison. The Character of the Syriac Version ofthe First Book, of Samuel(MPTL, 11; Leiden: E.J. Brill, 2001). pp. 34-49.

Then David lived in the stronghold,'-* that is Zion, and he called it thecity of David.

1 Chron. 11.7Ti l Ty ib-iKip p"'7y i:iO3 i"n HD'I

Then David lived in the stronghold. Therefore they called it the city ofDavid.

^cni aciin ^ t a rtjcn i i^S^ yCLicn .i r<>icUL=3 .i

Then David lived in the cities of Zion. Because of this they called themthe cities of David.

In 2 Sam. 5.9 the Peshitta adds ^cutn^ ,ro ,cn. 'that is Zion'.^° In 1Cliron. 1L7 the divergent Pesliitta reading also includes the additioti of'Zion'.^^ The reading in 2 Sam. 5.9 can be explained as a harmonizationwith 2 Sam. 5.7:

m T:; K-H jrii m:;c "K i n - '7-1Then David ca]jtured the stronghold of Zion. It is the city of David

Then David captured the stronghold of Zion. It is the city of David.

2 Sam. 5.23-24 and I Chron. IJ,.14-15

2 Sam. 5.23-24-[ycDD 'n"i [24] c-NDn ':'IOQ an'? nxn- cn-ins-'^s son n' yn N': -QK-- m--D ~'~ 'PND-I

UTdi^ njnc2 r.rn'^ -|':ab mn- N: ' m -3 j-nnn m n'SDnn "DK-in myi: ' p-nnThen David inquired of the Lord and he [the Lord] said, 'Do not go up;go around behind them and come to them opposite the halsain trees.When you hear the sound of the marching on the tops of the balsamtrees, then be attentive for then the Lord has gone forth ahead of youto strike the camp of the Philistines'.

[21] 71. AJ -\-.nn\

Then David inquired of the Lord and the Lord said to him. "Do notgo up. Instead, turn around to behind them and come against themopposite Bechim. When you hear the sound of walking on the top of the

Peshitta always reads AiO't-ai for MT m:iQ.LXX reads ev rf, -nz^ioyri, but the Old Latin has in omnem regionem Sion.

^'Though it is not in perfect agreement with 2 Sam. 5.9. this reading i,s includedfor the sake of completeness.

64 Armnaic Studies 3.1 (2005;

mountain of Bechiui. then be strong becatise then the Lord has goneforth before you to strike the camp of the Philistines'.

1 Chron. 14.14-15

r -n'l [15]

Again David inquired of God and God said to him, 'Do not go up afterthem. Go around them and come to them opposite the balsam trees.When you hear the sound of the marching on tlie tops of the balsamtrees, then you should go out for battle for God has gone forth ahead ofyou to sli'ikc the ranip of the

r<r'*a\,.T out.la [j ^ ^ ^

Again David inquired of the Lord and lie [the Lord] said to him, "Donot go up. Instead, go around tliom and go. attack them from the front.When you hear the sound of the shouting on tliR top of the mountain,then go forth at the opportunity because then the Lord goes forth beforeyou to destroy the camps of the Philistines.'

In 2 Sam. 5.23 and 1 Chron. 14.14 the Lord strategizes with Davidagainst the Philistines: David and his men are to go toward the 'bal-sam trees'. The meaning of ••i!Dn is disputed. The LXX takes it as atoponyni, xXau&jj.wvoq. derived from S/TO"!. Targum Jonathan rendersit with the generic 'trees' (N"':''S). The Peshitta in 2 Sam. 5.23 trans-fers the Hebrew consonants into Syriac resulting in a place name. Butin the subsequent verse the Hebrew phrase CKDnn 'mN"i3 is rendered•p.^rtf-,t r^+a\.t <TLt.-t= {'on the top of the mountain of Dechim').Among the versions, th( addition of the term rtf1ei\ is nnique to thePeshitta, but it also appears in 1 Chron. 14.15 which ignores C^xr nentirely.

The uniqueness of the Peshitta reading, and the fact that it can-not be explained internally within the Peshitta version itself {such as,for example, a harmonization), suggests that borrowing may have oc-curred. Bnt because the Peshitta text of 2 Sam. 5 24 is so different fromthat of 1 Chron. 14.15, borrowing at the time of translation seemsunlikely, It seems more reasonable to argue that one text influencedthe other during the transmission of the Peshitta text. The question,then, is the direction of the influence. A definitive response cannot begiven. Perhaps the Chronicles translator created the translation

tor C'xrzn '^VTQ to resolvo the problematic term C'Nmn thathe had omitted in the preceding verse and this translation influencedtlie transmission of the Peshitta text of 2 Sam. 5.23-24. But a morehkely explanation is that the Peshitta translator of 2 Samuel addedrtfici^i. assimilating the translation to the expression rtficL^i mjt.'t=which is (onnnon in the Peshitta (see Exod. 24.17; 34.2; Judg. 9.T: Ps.72.16, Isa. 2.2. and Mic. 4.1). The Syriac translator of 1 Chron. 14.15was tlien influenced by the Poshitta text of 2 Sam. 5.24.

2 Sam.. 7.6 and 1 Chron. 17.5

2 Sam. 7.6C-i:iOQ "^SIE" 'IHTN Th'JTl UTH^

from the day I brought up the people of Israel from Egypt

fioiu the (lav I brought iij) the people of Israel hoiu the land of Egypt

1 Chron. 17.5

from tbe day I brought up Tsracl

froui tbe day I brought up Israel froui the laud of Egypt

2 Sam. 7.6 adds f<i.1r«r in agreement with 1 Cbron. 17.5 and againsttheir respective Hebrew texts.''"^ This addition is characteristic of thePeshitta translation which can preflx the name of a territory withrti.*rtf as in 1 Chron. 5.10: -u'?;'? nn'C: .T%\\.I r<i.^r3.T r<i.-tr^ ^ yajr< (see also 1 Sam. 10.18 and 12.8). Thus, tlu^

addition oi r<i.irtr in 2 Sam. 7.6 can be considered a case of translationtechnique and need not indicate borrowing from 1 Cluou. 17.5.

2 Sam. 7.10 and 1 Chron. 17.9

2 Sam. 7.10irj rn" sbi rnnn "DOi T~ua;i1 will plant hiui aud he will live in hiw own ]>lac-e aud be will not be

disturbed agaiu.

of ^t-»3.T in 1 C'iii'oii. 17.r) asks whether the translator of PeshittaChronicles borrowed from the Hebrew or Peshitta text of 2 Samuel. This question,as noted in the introduction, requires separate consideration and can only bo an-swered once the character of PeHhit.ta Chronicles is better underetood.

66 ATumaic Studies 3.1 (2005]

.^n U r u .soAct r iAjca cniotrâ , ^

I will plant him and I will make him dwell in his place in peace and hewill not be disturbed again.

1 Chron. 17.9TV T;T N' T rnnn p'Di ir;-nuu:iI will plant him and he wili live in his own place aud lie will not bodisturbed again.

I will plaut a place for my people Tsraei aud I will uiake him dw(>ll andhe will live in his place aud he will not be disturJiw! again.

The Hebrew texts in this case are identical. Pesliitta 2 Sam. 7 10 in-terprets pm with jcncu'tx.pt'o 'I will make him dwell,' a translationtliat does not appear iu the LXX (xaTaoxT^voiaet,) or Targum ("TIEJ"!).The Peshitta of 1 Chrou. 17.9 renders rnnr. 'p'J'\ with mAvô.ts jaiun.But Peshitta 1 Chron. 17.9 also contains the reading .mcu-u.r'fo fromPeshitta 2 Sam. 7.10, The LXX and Targum translations in 1 Chron.17.9 arc identical with thos(> in 2 Sam. 7.10. The Peshitta Chroniclesreading, cnAiio.ta JAUO icncutx-pô. can be explained as a confla-tion. The original translation of rnnn ]DE?T in 1 Chron. 17.9 was ^ ^ acD^Lia.ta. Then a later copyist borrowed tODCutr.r 'o (consciously orunconsciously) from the Peshitta version of 2 Sam. 7.10. That the con-flation was generated by the translator of Chronicles seems unlikelysince, if the translator were responsible, one could reasonably expectthat he would have noticed that the Peshitta of 2 Sam. 7,10 is muchcloser to the Hebrew text of 1 Chron. 17.9 than is his translation.

2 Sam^. 6.6-7 and 1 Chron. 13.9-10

2 Sam. 6.0-7Tn- fjN'inn [7] ~p2n mcc "D in ins'i ••n'^N- |1~K*'7N KTI? nb^-i ]ID: ]nr-iy

They ranie to the threshing floor of Naeon. Uzzah stretched out his handto the ark of God and he seized it because the oxeu shook '* it. The Lordbecame angry with Uzzah and God struck him there . . . So he died therebeside the ark of God,

[7]

meaning of "aQD is uncUiar in this context and it receives various transla-tions in both ancient and modern translations.

M O R R I S O N Peshitta 2 Samuel and 1 Chronicles G7

They eanie to the firm threshing floors. Uzzah stretched out his handto the ark of the Lord aud he seized it because the oxen slipped. TheLord became angry with Uzzah and the Lord struck him there becausehe stretched out his hand. So he died there beside the ark of God.

1 Chron. 13.9-10mn- f]S"-in"i [10] "lpzn mm 'z ]i-iNn"-K ins'? IT--,N ^VJ rh-'D", "-"D ]-\r-v isn'i

They came to the threshing floor of Kidon. Uzzah stretched out his handto seize the ark because the oxen shook it. The Lord became angry withUzzah and he struck him because he stretched out his hand to the ark.So he died there i.>efore God.

r h\T<c\ [ I D ]

They eanie to the upper threshing fioors. Uzzah stretched out his handto support the ark because the oxen had begun to run to the threshingfloors. The Lord became angry with Uzzah and he struck him therebecatise he stretched out his hand to the ark. So he died there beforethe ark.

The reader wants to know why Uzzah was killed, or, more specifically.what does 'porr'?!' mean in 2 Sam. 6.7. The Peshitta translator renders'?Dn"'7:; with m.T.rtf ij^nr^.T A^ in agreement with 1 Chron. 13.10.The simplest explanation for the reading in 2 Sam. 6.7 is that thetranslator saw the phrase ouoir^ \ i . cn.r.r*' rtHwL^ \^art!a trom 2Sam. 6.6 on the explanation for God's action. Had he been influencedby 1 Cliroti. 13.10 it is reasonable to assume that he would have takenthe entire phrase. K'Aici.aua AJ^ m.T.,r« r om Y*""^"* ^ - from thatverse. Thtis an explanation internal tu the Peshitta translation of 2Samuel is preferred.

2 Sam. 7.15 and 1 Chron. 17.13

2 Sam, 7.151":a'7D -men ncs b'<KD z-jc Tncn "'.DXD

just as I removed [it] from Saul whom 1 removed from before you

just as 1 removed [it] from Saul who wa.s before you but I remo\ed himfrom before me

1 Chron. 17.13

just as I removed [it] from him who was before you

just as I removed it from Saul who was before yciii

There aro tw o agreements here. The Peshitta in 2 Sam. 7.15 addswyjw.TD ^ r<ami in agreement with 1 Chron. 17.13, resulting ina conflation of 2 Sam. 7.15 (-i';s'7Q 'mon nON ') and the text of 1Chron, 17.13 ("J'3D' ri'n ifflXQ). Perhaps the conflated text was alreadyin the Hebrew Vorlage behind the Peshitta, as McCarter suggests. ^ Itis also possible that the Peshitta translator of Samuel borrowed r ncn.Tt/y-rn.-ui ^ from the Hebrew text or the Peshitta text of Chroniclesor that it was introduced into 2 Sam. 7,15 by a copyist during trans-mission. The second agreement, the addition of AoniL ^ in 1 Chron,17.13, is not under consideration since the Peshitta of 2 Samuel doesnot depart from the Hebrew ^

2 Sam. 6.9 and 1 Chron. 13.12

2 Sam. 6.9niT |1-1S •'7K K13" 1'N

How ran the ark of tbe Lord come to uie?rC-t .T oxJO-tr licii ii.r^ rdlij r

How can I bring to myself the ark of the Lord?

Peshitta reading ,s^xo ^ for "["lEl' Q is not of direct concern to this paperas it does not appear in 1 Chron, 17.13. At least two possibilities can explain it.First, it could be a harmonization to the next verse where tlie Peshitta 2 Sam. 7.16reads »=n.To for ~|"]E'^. Second, the LXX reading sx npoowTitiu \XOD for ~[']E'^C [2 Sam.7.1G) together with the Peshitta reading cuuld point to a different Hebrew Vorlageor influence of the LXX on the Peshitta text of 2 Samuel. The relationship betweenthe Peshitta and the LXX in 2 Samuel requires further study.

'^P.K, McCarter, / / Samuel [AB, 9: Garden City NY: Doubleday. 1984), pp,194-95.

^''The Peshitta text of 1 Cin'on, 17,13 agrees witli tlie Targum of Chronicleswhich reads: "jmp |0 1'':'Q mn" bliX'O ]^ ^n'TJl riDD'n (Just as I removed [it] fromSaul who was king before you).

The reading in the Targum leaves open the ])ossibility that a variant HebrewVorlage is responsible for this reading in Peshitta 1 Chron, 17.13 (though no wit-ness to the LXX preserves the addition "from Saul'), If no-, a variant Vorlage, thenborrowing from 2 Sam. 7.15 could explain the addition of AnrOe. ^ in 1 Chron.17,13 but further research into the character of the Peshitta text of Chronicles isrequired.


1 Chron. 13.12Cn'^KH "UN HN •'7N SfDN "[Ti

How ran 1 l>rinf; to myself the ark of God?

How can I bring to myself the ark of the Lord.''

In 2 Satn. 6.9, David, terrified by Uzzah's death, exclaims: 'How canthe ark of the Lord come to me!' The Peshitta reads A r«f for HebrewX12' in agreement with 1 Chron. 13.12. The LXX (elaEAeuCTETcrx) andTgJon C?':;") follow the MT. The reading in 2 Sam. 6.9 cannot besatisfactorily explained as an examjile of translation technique and thereadings in tlio LXX and Targum do not support a variant Vorlage. It ispossible that the Peshitta translator of 2 Saniuel was influenced by theHebrew text of 1 Chron. 13.12. It is also possible that a later copyistof Pesliitta 2 Samuel harmonized the original translation in 2 Sam. 6.9

A with the Peshitta text of 1 Chron. 13.12.

2 Samuel 8 and 1 Chronicles 18: Aram and Edom

2 Samuel 8 recounts David's defeat of Aram of Damascus. The Peshittaconsistently renders Hebrew C")K with >jair^. 'Edoni", even thoughEdotn is not mentioned as one of David's conqnered territories in theHebrew text tmtil 2 Sam. 8.14. The LXX renders CiN with 6 Supogin 2 Sam. 8.5-6 but then reads 'Edom' for Helnew 'Aram' in 2 Sam.8.12 (MT: c~-NC; LXX: iv. xr^c, lSou[xat.a;;). The Targum, as expected,is faithful to the Ml' throughout 2 Samuel 8. Thus, the Peshitta aloneis consistent in reading CiK a,s -paxri. Peshitta 2 Samuel 8 shares thischaracteristic with 1 Chronicles 18. The name DiX appears in 1 Chron.18.5 (the LXX reads Supo; and Targum Chronicles reads CiS) and thePeshitta text of 1 Chron. 18.5 agrees with 2 Sam. 8.5:

2 Sam. 8.5-iiy'7 pom ciK snnThen Aram of Damascus came to help

Then Edom and DamtLscus came to Iielp

1 Chron. 18.5~^\^b pen-- c N N3nThen Aram of Damascus came to help

Then the men of Edoui and Damascus came to help

70 Armnaic Studies 3.1 (2005)

name •Aram' appears again in the next verse. Peshitta Chroniclesdiverges significantly from the Hebrew whereas Peshitta Samuel doesnot:

2 Sam. 8.6cnnu'? -m'7 ens Tim pac-i C",N2 c'2)i: n i DC!*IThen David set garrisons in Aram of Damascus and [the people of] Arambecame servants tu David.

Then David set governors in Edom and in Damascus and the Edoinitesbecame servants to David.

1 Chron. 18.6c ' l i r I".!"? ciN "-'1 p^Q~n cnxD T H CC"IThen David set [garrisons] in Aram of Damascus and [the people of]Aram became servants to David.

Then David set commanders in Damascus and the people of Damascusbecame servants to David.

Neither 1 Chroniclos 18 nor 2 Samuel 8 translates cnx with ysir^ (as, forexample, in Gen. 10.22 and Num. 23.7). 2 Sanni(>l consistently rendersC-iN with y^airf (see 2 Sam. 8.12, 13; 10.6, 8, 9, 11, 13, 14, 15, 16, 17,18, 19; 15.8). While both 1 Chronicles and 2 Samuel avoid translating•IS with -p-ir^, a comparison of 2 Sam. 8.6 with 1 Chron. 18.6 revealsthat the translators have various ways of handling the word niit. 1Chron. 18.6 simply ignores it. The reading TJO.IT^ for c~N in 2 Sam. 8.5is consistent with the translator's handling of this name throughout 2Samuel. Therefore this reading, though it agrees with 1 Chron. 18.5,can be explained internally within 2 Samuel aud n(!ed not be consideredau example of influence from 1 Chronicles.

2 Sam. 8.13 and 1 Chron. 18.12

2 Sam. 8.13'p'». "iDU n]iGO n'7C"N";3 C X T N in-cna inc n UD I T : CU"IDavid made a name for himself when he retm-ned from defeating Aramin the Valley of Salt, eighteen thousand men.

r<ClCD

And there David waged a battle when he returned after he defeatedEdom in the Valley of Salt. He slaughtered eighteen thousand men.

1 Chron. 18.12

And Abishai, the son of Zoruiah, defeated Edom in the Valloy of Salt,eighteen thousand men.

And Abishai, the .son of Zeruiah. who is the brother of Joab, slaughteredthe Edomites in the Valley of Salt, eight(>eii tlioiisKiid men.

The verb jitw in 2 Sam. 8.13 has no corresponding equivalent in theHebrew text and the Peshitta reading does not appear in the LXX orthe Targum. The reading in 2 Sam. 8.13 can be considered an innerSyriac harmonization with 2 Sam. 8.4 where the Peshitta reads j-tw forHebrew I'zh''. The translator needed a verb in this phrase and. jjerhapsnnconsciiHisly. inserted a recently nsed verb. It is also possible that thereading in 2 Sam. 8.13 is a eonflation of the Hebrew text in 2 Sam,8.13 (imDHD) and the Peshitta text in 1 Chron. 18.12 (-=tw), Bnt theinternal explanation, harmonization, is preferred.

2 Sam. 8.18 and 1 Chron. 18.17

2 Sam. 8.18

Benaiali, son of Jehoiada, and the Cheretliites and Pelctliites. And Da-vid's sons were priests.aam ^ j j i o t .TiO.l tXiaa r<u\ja A^Ct r<'ir<u A-V .V-l^CU l a r<'. l-iO

Bcuaia son of Yuyada was over the freemen and the slaves. David's sonswere leaders.

LXX:

And Baiiaias son of Jodae, coimselor, and the Cheletites and the Pheiitesand David's sons were leaders.

TgJon:

Benaia son of Yehoyada was appointed over the ai-chers and over thoslingers. David's sons were leaders.

1 Chron. 18.17

Beiiaiah, son of Jehoiada was over the Cherethites and Polethil.es. AndDavid's sons were loaders beside the king.

Benaia son of Yuyada was over the archers and over the sliiigors, David'ssons were leaders.

72 Aramaie Studies 3.1 (2005)

LXX:xai Bavata? uLo? ItoSae ini rou ;(Epey9L xal TO'J q;eXE8!)', xal ulol AauL^01. TipuTOL StaSo^ot Toij fiaaiXswi;

And Banaias sou of Jodae was over the Cherethites and the Phelitos andDavid's sons were the first successors of the king.

Targum Chronicles:nn-Q 'PUT •Q-.n: K-nsn " 'NC r.irf: NCIT "-I]C b'j' NPG^ -JT:; "U N;Qr2 yi-in- ID IT]31

N2'7CT K-iuc'? NrQip2 "ZfT i n "im N np*? "nn; n-y'7pi TTFCP |nn -aisBenayahu son of Yehoyada was appointed over thi) great Sauhedrin andover the small Sauhedrin. He was inquiring by meaus of Urim and Thum-mim. Based upon the decision of his mouth the arehers and slingers weregoing down into battle. David's sous occupied the first position in thekiug"s guard.

Peshitta 2 Sam. 8.18 reads ^.^lot for C'}n2 with Tg.Ion (]'3n2n) andthere is a similar reading in the LXX (auXapxat.). The Peshitta of 2Sam. 8.18 agrees with 1 Chron. 18.17 where ^^iot translates MTCJOXin. The MT of 2 Sam. 8.18 (and the Vulg., sacerdotes) is the lectiodifficilior as it disagrees with Ntmibers 3 which records that the levit-ical priesthood belonged to the sons of Aaron. Furthermore, David'saction in the MT pai'allels that of Jeroboam who. after establishingthe golden calves in Dan and Bethel, appointed i)riests who were notLevites (1 Kgs. 12.31). The LXX, Peshitta and Targum translators, per-haps conscious of this problem, offered various solutions. In this case,the Peshitta translator could have consulted eitliei: the Hebrew text orthe Peshitta translation of 1 Chron. 18.17 for the solution ^^^-taS. Butthe agreement among the versions to avoid a literal rendering of theHebrew suggests that the Peshitta translator too was acquainted withthis prevailing int('rpretation. Thus there is no reason to snp])ose thatthe Peshitta text in 2 Sam. 8.18 was directly influenced by 1 Chron.18.17.

2 Sam. 10.19 and 1 Chron. 19.19

2 Sam. 10.19

Then all the kings, servants of Hadadezer, saw

Then all the servants of Hadadezer saw

1 Chrou. 19.19

Then the servants of Hadadezer saw

Then the servants of Hadadrzer saw

The Peshitta of 2 Sam. 10.19 agroos with 1 Chroii. 19.19 hy omiftiiig••Dbon. The term 'kings' makes little sense in this verse and the trans-lator could have deliberately otnitted the word without eonsulting theHebrew or Peshitta text of 1 Chron. 19.19.^"

2 Sam. 12.31 and 1 Chron. 20.3

2 Sam. 12.31

The people who were in it [the city] he brought out and he set them withsaws, with iron picks, and with iron axes and he made them pass over tothe hrick kihi.'** Thus he was doing to all the cities of the Ammonites.Then Da\'id and nil fhr people returned to .Ieru.sa!eni.

.CV=n_i. , ' - ' I r ^ f c t o -r-nXrt-^ V T* v rtfj_^<DO w A . .rur-n-i

And the people who were in it [the city] he brought forth and he threwon them iron neck chains and shackles and he made them pass througha measuring instrument. Thus he did to all the cities of the Ammonites.Then David and all the people returned to Jerusalem.

1 Chrun. 20.3

The people who were in it [the city] he brought out and ho cut themwith saws, with iron picks, and with axes. Thus David was doing to allthe cities of the Ammouites. Then David and all the people returned toJerusalem.

translator can {•onsciously omit Hebrew words that are repetitive, con-trmlictory or that make little sense. Sec Morrison, The Character of the SyriaeVersion, pp. 78-82.

^"Reading the qere ]3'7Q3 for the keUh p'7C2. See D. Bartheleniy, Critique textuellede I 'Ancien Testament, I. Josue, Juges. Ruth. Samuel. Rois. Chroniques. Es-dras. Nehemie. Esther (OBO. . )n.l: Fribourg-Giittingen: Vandenlioeck fc Ruprecht.1982), p. 264.

Aud tlie people who were in it he brought them forth aud he boundthem with shackles and iron neck chains and with locks and in prisons.Ho bound them all. Thus he did to all the people who were found in thecities of the Ammonites. He killed no one, but he went aud made themdwell in the cities of the land of Israel. Then David aud all who werewith him returned to Jerusalem.

The Hebrew texts of 2 Sam. 12.31 and 1 Chron. 20.3 rec'ount David'streatment of the defeated populace of Kabbah. According to tlie Pe-shitta of 2 Sam. 12.31, David brought the citizens out in iron neckchains and shackles and made them pass through a measuring instru-ment. This interpretation does not appear in the LXX or the Targnm.The expression "to throw shackles on someone' appears in Ezek. 3.25where rC'iAjcx. vyA_i. ,,^^ I CD renders •'mni; ~i'^:i i]n] HJH. The shack-ling of conquered peoples is often the victor's final gesture in a militarycampaign. Zedekiah is c-arried off to Babylon in sl'ackles (2 Kgs. 25.7):^22 '~HT* C'~anj3 irriCN"; \ - - \ oA^ar^a *^^\*^- (T>\a>t<. '•' Thus.the context of this verse David's conquest of Rabbah—along withthe word ':'n3, 'iron", the material from which shackles can be made.^°has allowed the translator some license in describing the treatment ofRabbah's citizenry. The word f rtXao {related to Latin collar) is rarein the Peshitta, appearing cHily in 2 Sam. 12.31 and in 1 Chron. 20.3,The Hebrew |""in also appears only in these two iastances.

The MT in 2 Sam. 12,31 reports that the people were transferredto a brick kiln for labor. According to the Peshitta, David made thempass through an instrument for measuring. The Peshitta reading bringsto mind 2 Sam. 8.2 where David, having defeated the Moabites, metedout punishment by measuring the prisoners:

.-iT>3r.,r<'O ^

OOCDO fi..»-w\ •^\-i.i r<'\-Tin AX irnX . \^.> -. ' i

He defeated the Moabites and he measured them with cordy. He madethem he down on the ground and measured two cords for those to be

also Nali, :M0: r«*At\TV-. otatrCAir^ m. i - i o i and Isa. 45.14:

"With regard to rCiuicutaa, R, Payne Smith writes: "fortasHiy fuerit instrumentnnimensovium, quo captivi ad varia supplicionim genera distributi sint' ('PerhaiJs itwas a measuring instrument by which prisoners were selected for variou.s typesof punishment') (R. Payne Smitli, Thesaurus Syriacu.s [Oxford: Clarendon Press,1879], col, 2237).

executed and tlie full length of one cord for these to be kept alive. Thusthe Moabites became slaves, bearers of tribute, to David.

This verse seems to be in the background of the Peshitta's interpre-tation in 2 Sam. 12.31 in which David again 'measures' a conqueredpeople.

The Peshitta text of 1 Chrou. 20.3 offers a lengthy expansion withrespect to the Hebrew text. The binding of the citizens of Rabbah isaccomplished not only W\\\\ shackles and neck chains bvit also withlocks and prisons. (Peshitta Chronicles also insists that David killedno one, against the MT.) Part of the expansion in Peshitta 1 Chron.20..'3. that David bound the peo})le with shackles and iron neck chains(rtfi^ta.T K'lVonô nfW'-r,-^) corres]>onds to the Peshitta reading in2 Sam. 12.31 [r 'AAjtizan rî la .T r^ticma). This agreement suggests arelationship between these two readings. Both have arrived at the samelexical choice for the Hebrew term f "in and both rendered n~;Q withptfivlit (the translator in 2 Sarn. 12.31. after adjusting tiie word order,seems to have ignored '^n^n nni;Q3). The term n~i:D appears again in

1 Kings 7.9 where it is translated with ottjâ\,.^^ The term t^ticua.like "["in in the MT, appears only in these two verses in the entireOT Peshitta. The similar translations suggest that the 1 Chroniclestranslator was probably influenced by the handling of this text in 2Sam. 12.31. But given that the verlis are not tho same [t^nîrâ/\sx>T<a),that the word order in the shared reading is different, and that the otherparts of the expansions hi each text were not imjiorted into the other.namely, the 'measuring' in 2 Sam. in 12.31 and the "locks" and "prisons"in 1 Chron. 20.3. it seems best to stiggest that the borrowing wa.sprobably unconscious and that it occurred at the time of the translationof 1 Chron. 20.3.

2 Sam. 23.20 and 1 Chron. 11.22

2 Sam. 23.20':'N:i2pc C'';J;D-31 "n-D-s-p UTin-p in-]2iAnd Benaiah, son of Jehoiada, son of a warrior,"^ great of deeds, fromKabzt't'l

^^MT; n-;Q2 7\'r\-\v2 n"i; mnz rr" CZHN Ttvr'-^z {all thewi' valuable stones, cut tomeaaurr. sawed with a. .saw).

(all these good stones according to the measure of a cut stone thai wa.s iiwith a chisel).

the qere '7"n.

AJ—-in ^ icnci.iî^ ctam ^ T . °tx..i r - u •t-îV nf'^.iN. .^ i . n i tra r . i-^

Benaiah, son of Jehoiada, a warrior, whose deeds were marvelous, fromKabzeel.

1 Chron. 11.22

Benaiah. son of Jehoiada. son of a warrior, groat of deeds, from Kabztêl.

Benaiah, son of Jehoiada, a warrior, whose deeds were niarvelonw. fromthe town of Kabzeel.

The Hebrew phrase cbUD-nn, 'great of deeds' , ^ receives the same trans-lation in both 2 Sam. 23.20 and 1 Chron. 11.22, .meirtij oom ^. <*- °.»ttranslations that do not appear in the Taj-gmn. i,xx. or Vulgate. Thiscase seems to indicate influence of one text on the other but again thedirection of the influence cannot be determined.

2 Sam. 23.25 and 1 Chron. 11.27

2 Sam. 23.25mnn Kp"'7K "iinn HQCShammah the Harodite. Eliqa the Harodite;

Shamnia was from Tur Malko.

1 Chron. 11.27

Shammotli the Harorlte. Helez the Pelonite;

Shammotli from Tiir Malko. Helez from Pelton.

In 2 Sam. 23.25 the Poshitta reads r<£aA») ic\ ^ for "linn. Thephrase "l"inn Kp''7X is omitted.'' *' Peshitta Chronicles interprets '"innnwith the same reading as in Peshitta Samuel. In Peshitta 2 Sam. 23.11Shammah is said to })e from Tur Malko:

meaning of the expression iw somewhat unrerthin. THAT (IL p. 464)suggests that 'groB au Taten" is the preferred translation because of the antiquityof 2 Sam. 23.20. TkWAT (VI, j). 700) suggests "von Grolien Taten'.

^''This is perhaps a deliberate omission given that ' l inn Is repeated from thefirst part of the verse and the name Kp"':' is quite similar to the next name in thelist (23.26): |'':'n/^Xjj. But beeanse the phrase ""nnn Np''7K is abwent from the mostancient LXX mss, it is also possit)Ie that it wa.s not present in the HebrewIjehind the Peshitta.


After him Shammah, sou of Age. the Haratite.^^

After him Shammah, sou of Age, from Tur Malko arose.TgJon:

and after him came Shamuia, sou of Age, of the uiountaiu people.

The Hebrew words mnn and '"lnnrt have been interpreted in the Pe-shitta on the basis of Hebrew "in. 'mountain' as in the Targum (butthe Targnm does not include X-'7n). But the readingremains unique. The translation of 'l~nn nncj withrtfiiiM in 2 Sam. 23.25 seems to be a harmonization to 2 Sam. 23.11, (hi2 Sam. 23.33 Shannnah appears again aud is said to live on the Mountof Olives: '"linn nnD; f Avl'i ^ci\, ^ i rtfr ue..) Borrowing between 2Sam. 23.25 and 1 Chron. 11.27 s(H'nis ( nite ai)parent, though, again, thedirection of the borrowing cannot be determined. Perhaps the Peshittatranslator of 2 Samuel created this reading in 2 Sam. 23.11 and thenharmonized 2 Sam. 23.25 to it. This interpretation was then borrowedby the translator of Peshitta Chronicles or by a later copyist. However,borrowing from 1 Chronicles liy the Peshitta translator of 2 Sanmelcannot be rnled out.

2 Sam. 24.2 and 1 Chron. 21.2

2 Sam. 24.2

and I will know the emimeratiou of the people

iuid l)riug to me their census and tlie enumeration of the people

1 Chron. 21.2

and bring to uie tlmt I might kuow their uuniber

and come to me and I will kuow the nuuiber of the people

There are two readings of interest: (1) the Peshitta reading»i in 2 Sam. 24.2 (not found in the LXX, Targum. or Vulgate) agreeswith the Hebrew text of 1 Chron. 21.2 ("' X Mi'2rA) against its Peshittatranslation (,^a^ aAio). (2) The Peshitta reading cn_L=x.cijjci yCiaij_._L»in 2 Sam. 24.2 appears, at first glance, to he a conflation of n"E~Q from

^ ' i s text does nol appear in the MT of 1 Chronicles 11,

1 Chron. 21.2 with IDOG from 2 Sam. 24.2. Peshitta 2 Sam. 24.2 omits'nvT' entirely. But it is also possible that the Peshitta reading in 2 Sam.24.2 is an inner Syriac harmonization to the Peshitta of 2 Sam. 24.9:

Joab gave the nmnber uf the cciistis of the people.rÔ.^..! TO.3Z.auO rd-m-Hn âr^CL. , ^ r ^ O

Joab brought the census and the enumeration of the people.

In 2 Sam. 24.9 the Peshitta reads . ^ râ for "n'T. ^ The phraser ri -iT.cu*Q is also close to the phrase caj.=uccu>a ^pm i. l -yi in 2 Sam.24.2. tliough ^Qtn 1.1 -w is closer to 1 ChroJi. 21.2 CHDOr:. But the expla-nation that best accounts for the three divergent readings in this verse(cn 1 iT.cuto ôm 1. i- a. rÂurâ and the omission of "n^Tl) is that ofan inner Syriac harmonization with 2 Sam. 24.9 though, admittedly, itdoes not account for the 3rd m. pi. suffix on yO"n. i-« which remainscloser to 1 Chron. 21.2 DIDOD than 2 Sam. 24.9 (ison). Harmoniza-tion between a command and its fulfillment is a common translationtechnique in the ^'

2 Sam. 24.12 and 1 Chron.21.10

2 Sam. 24.12"Y'^D '?m] ' : : N d7zi

Three things I am offering to you

Three atrocities I will bring upon yon

1 Chron. 21.10

Three things I am offering to yon

Three atrocities I will bring upon yon

2 Sam. 24.12 and 1 Chron. 21.10 agree to add ^L^ . This addition isunique in the Peshitta and the influence eould be in either diro( tiou.

2 Sam. 24.17 and 1 Chron. 21.17

2 Sam. 24.17

^^'In 1 Chron. 21.5 the Peshitta reads r^îi..! rdLaacOiKi ^^'i- '-*• jr<rcu .=01*0following MT CUrrnpEQ -ISCQ'nS DKV ]n'l.

^'Morrison, The Character ofthe Syriac Version, pp. 46-47.

And these sheep, what have they done?

These innocent sheep, wiiat. iiave they done?

1 Chron. 21.17

And these sheep, what have they done?

riini these iiiiiocciit .sheep, what have tliey done?

2 Sam. 24.17 agrees with 1 Chron. 21.17 to add f<"Adi*i»Di\. The addi-tion is n(.)t. fonnd in the LXX, Targum, or Vulgate. Again, it suggestsinfluence in one direction or the other.

Summary

Nineteen agreements between Peshitta 2 Samuel and tlie Hebrew orPeshitta text of 1 Chronicles were identified. They can be summarizedas follows:

1. Agreements in which the reading iu 2 Samuel can be accounted forinternally as an example of the Peshitta'.s translation techniqne. Sixare harmonizations: 2 Sam. 5.2, 9; 6.7; 8.13 and the two readingsin 24.2. The agreement in 2 Sam. 7.6 can be attributed to a sharedtranslation technique. The omission of D'D'PQ" in 2 Sam. 10.19 canalso be attributed to the character of the Peshitta translation. Alsoin this category is the agreement in 2 Sam. 8.5.

2. The reading in 2 Sam. 8.18 (in agreement with 1 Chron. 18.17)witnesses to a widespread interpretation of this verse and cannt)tbe considered an example of influence.

3. There are four readings that seem to indicate influence or 1>orrow-ing, but tlie direction of the influence/borrowing cannot be deter-mined: 23.20, 25; 24.12 and 24.17.

4. Two readings, 2 Sam. 6.9 and 2 Sam. 7.15. suggest that the Pesliittatranslation of 2 Samuel was influenced by the Hebrew or Pesliittatext of 1 Chronicles. It cannot be determined whether the influenceoccurred at the time of translation or transmission.

5. In two causes (2 Sam. 5.24 ami 1 Chron. 14,15; 2 Sam. 12.31 and 1Chron. 20.3) the Peshitta translator of Chronicles waa hiflLiencedby the Peshitta text of 2 Samuel. The agreement in 2 Sam. 7.10 is

the result of a conflation in Peshitta Chronicles and suggests thata Chronicles copyist was infinenced by the Peshitta text of 2 Sam.7.10. Further research into the character of Peshitta Chronicles isnecessary before drawing any conclusion regarding the influence ofPeshitta 2 Samuel on Peshitta 1 Chronicles.

CoTiclusiori,

This study has focused on the question of whether the Peshitta trans-lation of 2 Saninel evinces traces of inflnence from the Peshitta textof 1 Chronicles. Two cases suggest that the Peshitta text of 2 Samnelwas infiucnccd either by the Hebrew text (at the time of translation)or by the Peshitta text (during transmission) of 1 Chronicles. Thereare hve cases that indicate infiuence of one text on the other, but thedirection ofthe influence cannot be determined with certainty. The ex-ampl(\s of borrowing nnist b(; viewed in light of the overall independentcharacter of the two translations. For example, in the Hebrew text of2 Sam. 23.8-39 (^1 Chron. 11.10-47) there are numerous text criti-cal problems. The Peshitta of Sanmel and the Peshitta of Chroniclesoffer different solutions for each of them except in one case (2 Sam.23.20 and 1 Chron. 11,22) where the two translations intersect againsttheir respective Hebrew texts. But this one agreement is dwarfed bythe number of instances in this episode that illustrate the independentcharacter of these two translations. Another example that illustratesthe independence of 2 Samuel from 1 Chronicles involves the transla-tion of the names 'Cherethites' and 'Pelethites' mentioned in 2 Sam.8.18 (cited above). They appear again in 2 Sam. 15.18; 20.7; 20.23; 1Kgs. 1.38. 44: and 1 Chron. 18.17.' * In 2 Samuel 'pn^n and "n' an areconsistently translatc<l r^'ir^ ('freeuicn') im<\ rtfxA^ ("slaves'), but in 1Kings TIDM and 'Vh^Ti arc rendered pijAnri ^.UC.TO ri'iLxDn ('archersand shngers') as in Peshitta Chronicles (see above 2 Chron. 18.17). ThePeshitta of 2 Samuel has its own translation for these terms unrelatedto the Hebrew. Peshitta or Targum texts of 1 Chronicles or the Peshittatext of 1 Kings. This unique reading in Peshitta 2 Samuel witnessesto its independence from the interpretation 'archers and slingers* thatwas known to the Peshitta translator of Kings aud Chronicles and tothe Targum translator. Furthermore, several agreements, such as the

' The first mention of the Cheretliites appears in 1 Sam, 1 0,14 (MT: n:: "arJS "~:sTllirrb ~i''DK'b^'] T l"" ) , The Peshitta omits it: rtf.iomj.i -nVi ^ _.Air<.i ^ r<ro.

MORRISON Peshitta 2 Samud and 1 Chronicles 81

agreement in 2 Sain. 5.24 aud 1 Chron. 14.15, appear within roatiingsthat are otherwise quite different. Thns. in general, Peshitta 2 Samueland Peshitta 1 Chronicles manifest a high degree of iudependenee fromeach other.

What tht'ii to make of l.he paucity of cayes that suggest influence?It seems reasonable to argne that the texts influenced each other notat the time of translation bnt during their transmission. At least threecenturies separate the translation period (the second half of the seeondcentury) from the first manuscript evidence. The conflate readings,in particular, suggest that, during these three centuries of copying,readiitgs from 2 Sanniol found their way into 1 Chronicles aud viceversa (perhaps, in some cases, unconsciously}. Thus, the Peshitta textsof 2 Samuel and 1 Chronicles reveal traces of infiuence of oue texton the other, even if a convincing explanation for how and when thatinflnence occurred remains uucertain.

[AS 3.1 (2005) 83-107]DOI: 10.117/1477835105053516

ETYMOLOGICAL NOTES ON NORTH-EASTERNNEO-ARAMAIC*

Hezy Mutzafi

Tel Aviv University

The lexical stock of North-Eastern Neo-Aramaic (NENA)^ exliibits alarge number of puzzling words of unknown or doubtful origin, andis fraught with tantalizing etymological conundrums awaiting solution.To select just a few examples from a small number of NENA dialects^—what might be the respective etyma of C.Urmi slg 'to pluck, extract,

'Akkadian, Kurdish and Syriac words are usually given without reference whenthey can readily be tracked down in the following dictionaries: Akkadian: AHw,CAD\ Kurdish: T. Wahby and C.J. Edmonds, A Kurdish-English Dictionary (Ox-ford: Clarendon Press, 1966), F.F. Omar, Kurdisch-Deutches Worterbuch (Kur-manci). (Berlin: Kurdische Studien Berlin im Verlag fiir Wissenschaft und Bil-dung, 1992), B. Rizgar, Kurdish-English English-Kurdish Dictionary (Kurmanci),(London: Lithosphere Printing, 1993), M.L. Chyet, Kurdish-English Dictionary(New Haven and London: Yale University Press, 2003); Syriac: R. Payne Smith,Thesaurus Syriacum (2 vols., Oxford: Clarendon Press, 1879-1901; henceforth:Thesaurus), J. Payne Smith, A Compendious Syriac Dictionary (Oxford: Claren-don Press, 1903), C. Brockelmann, Lexicon Syriacum (Halle: Max Niemyer, 2ndedn, 1928).

Notes on transcription: The vowels i, e, c and o are long; otherwise vowel lengthis marked only for long o versus short a and long u versus short u. Stress ispenultimate unless otherwise indicated. Superscript + preceding a word indicatesword-emphasis.

Abbreviations: Akk. = Akkadian, Ar. = Arabic, Azer. = Azerbaijani, C. = Chris-tian dialect of . . . (e.g. C.Urmi), J. = Jewish dialect of . . . (e.g. J.Urmi), JBA =.lewish Babylonian Aramaic, JPA = Jewish Palestinian Aramaic, Kurd. = Kur-dish, Mand. = Mandaic, OA = Older Aramaic (vis-a-vis Neo-Aramaic), Syr. =Syriac. See further under n. 2.

^NENA consists of dialects spoken (or originally spoken) east of the Tigris riverin Kurdistan, the plain of Mosul and Iranian Azerbaijan.

^The lexical data offered in this paper refer to the following Neo-Aramaicdialects:


uproot', J.Urmi +znqr 'to pour, strew', Herteviri qopina 'forehead',^Betanure swixa 'fool', A-Tiare hatamta 'very wealthy (man)','' S-Tiarexuta 'ford', or Baz pode 'snot', with cognates such as C.Urmi pudi,'Aqra poze and J.Sanandaj bole? None of these lexemes can be deci-sively identified as stemming from a foreign source;, nor do they appearto have an ascertained Aramaic ancestry. At least some of them maywell be genuine Aramaic words that have drifted so radically fromtheir forbears during the long course of Aramaic history, having beenreshaped by phonological, morphological, semantic, or accumulative,many-faceted processes, that their origins are no longer readily recog-nizable.

Such radical changes have indeed occurred in many NENA wordsof Aramaic provenance. Compare, e.g., J.Koy Sanjaq nhala 'ear' withits precursor attested in Syriac as r<f 'ciii-ii< a plural form of r<i.Trtf'ear',^ K-Tiare ponqisa 'bubble', a diminutive form (ending in -isa <*-ita) of Syr. r^'.TJOa 'ball' and A-Tiare 'asa 'cock's upper claw, spur'.

Christian NENA dialects; in Turkey: Baz, Hertevin, Jilu, MarBishu, Marga, Sat;Tiare cluster: A-Tiare = Ashitha, K-Tiare = Ko, S-Tiare = Sarspidho; Tkhumacluster: M-Tkhuma = Mazra; Iraq: Alqosh, 'Ankawa, 'Aqra, Aradhin, Baritle,Isnakh, Koy Sanjaq, Nerwa, Qaraqosh, Telkepe; Iran: Salamas, Sanandaj, Urmi.

Jewish NENA dialects; in Iraq: 'Amadiyya, Aradhin, Arbel, Barzan, Betanure,Dobe, Koy Sanjaq, Nerwa, Rustaqa, Sulemaniyya, Zakho; Iran: Kerend,Naghada, Sanandaj, Saqqiz, Urmi.

Additionally, there are references to Mlahso Neo-Mandaic, Neo-Western Aramaicand Turoyo.

Data on Neo-Aramaic dialects without a bibliographical reference are based on myinformants.

^O. Jastrow, Der neuaramdische Dialekt von Hertevin (Provinz Siirt) (SemiticaViva 3, Wiesbaden: Harrassowitz, 1988), p. 197.

''According to an informant, the adjective hapmta is after the name of 'a cer-tain Arab king who was very rich'. Could this word stem from the name of thepre-Islamic poet ^^LkJ'i y>iU- (Hatim of the tribe of Ta'iyy), renowned for his gen-erosity (see C. Huart, A History of Arabic Literature (Beirut: Khayats 1966), pp.23-24)? For Ar. h > Tiare h cf. Ar. o ' ^ > Kurd, heywan > Tiare hewan 'an-imal'. Cf. also J.Koy Sanjaq 'aflatun 'very clever (person, including woman)' <Kurd,, Ar. 'Plato' (and similarly in Y. Sabar, A Jewish Neo-Aramaic Dictionary.Dialects of Amidya, Dihok, Nerwa and Zakho, Northwestern Iraq (Semitica Viva28, Wiesbaden: Harrassowitz, 2002; henceforth: JNAD), p, 99a, s.v. ]1QK'73N).

^The historical process is, briefly, *'ednaha.ta > *'9nna,hata > *nhata > *nMla >*nhala > *nhala > *nhala; see the evidence in H. Mutzafi, 'The reflexes of the wordSJlS ('ear') in Eastern Neo-Aramaic: Etymology, Diversification and Innovation',in M. Bar-Asher and M. Florentin (eds.), Samaritan, Hebrew and Aramaic StudiesPresented to Professor Abraham Tal (in Hebrew), forthcoming.

MuTZAFi Etymological Notes 85

the direct precursor of which appears in S-Tiare as sdsa 'id.', and itsetymology is attested in Syr. r nooLi..

Rather far-reaching semantic shifts are also common in NENA di-alects, e.g. those evinced by J.Aradhin palsa 'Hsh scale' vis-a-vis Baz(and other dialects) palsa 'small coin' and Syr. r mNcta 'id.' (< Greek);^S-Tiare quddsta 'goat's dewlap' compared with 'Aqra quddsta 'earring'and Syr. rfS^hn 'earring, nose-ring';^ A-Tiare (among other dialects)qudasta 'small section of a bunch of grapes' compared with the samelatter Syriac word riSuthnf and K-Tiare sura'a 'bunch of grapes' com-pared with C.Aradhin sra'a 'udder' and Syr. i ^ i i - 'udder full of milk'.

Among NENA words with elusive, though not insoluble etymologiesare those harking back to fossilized two- or even three-word phrases orsentences, e.g. C.Aradhin l-bdttu6 'on an empty stomach', in which l-is historically derived from the preposition ^V 'on' and the form bottuOis a contraction of nicp n3 'he spent the night fasting', as in BiblicalAramaic (Dan. 6.19) and likewise in Syr. i\6\ ^ ; ^ ° C.Nerwa gunnita'morning' < *gu n'ihta < *gu n'ihtd < *gu ngihta 'at dawn';^^ as well

''In the sense of 'cock's spur' only in Manna, f<..i\i ri^^^'irf r<ijt\i(Syriac-Arabic dictionary entitled 'Guide to the Chaldean Aramaic Language';Mossoul, 1900), p. 554b, s.v. For the process *'sasd > S-Tiare sasa cf. in the samedialect *'mada > mada. 'baptism', 'raqa > raqa 'to flee, run'.

^Cf. Qaraqosh pdlsd 'money; scales of fish' (G. Khan, The Neo-Aramaic dialectof Qaraqosh (Leiden: E.J. Brill, 2002), p. 740) and Sabar, JNAD, p. 263a, s.v. NO' D:filsa 'penny', pi. folse 'flsh scales'.

*An informant described the goats kept by Tiare shepherds as having each apair of narrow dangling dewlaps, hence the resemblance to a pair of traditional longearrings. For the semantic relationship between dewlaps and jewels cf. Betanuretoga (< Kurd. < Ar. j j i ) 'necklace; goat's dewlap'.

^See also Sabar, JNAD, p. 274b, s.v. snDnp, NHCnp; and compare, moreover, S-Tiare totiOa 'large green or blue bead' with Syr. r^iL^cL^ '(single) grape; (small)bunch of grapes'.

^"See G. Krotkoff, A Neo-Aramaic Dialect of Kurdistan; Texts, Grammar andVocabulary (New Haven: American Oriental Society, 1982), p. 162; idem, 'Studiesin Neo-Aramaic Lexicography', in: A. Kort and S. Morschauser (eds.). Biblical andRelated Studies Presented to Samuel Iwry (Winona Lake: Eisenbrauns, 1985), pp.123-24 (127). A more transparent form is M-Tkhuma bdtwa,6 'having no breakfast,going hungry since the morning', as in hesta howan hgtwaO 'I am still going hungry(since the morning)'.

' ^See Sh. Talay, 'Ein Teufel im Kloster: Texte im christlisch-neuarama,ischenDialekt von Nerwa (Nordirak)', Mediterranean Language Review 14 (2002), pp.125-69 (126, n. 7). For the incorporation of gu 'in' as part of NENA nouns cf. Jilu'^gullQxta 'hollow of the hand' < *gu lahtd 'in the palm of the hand', Tiare gubba'breast-pocket' < *gu 'ubba 'in the bosom' (and cf. Turoyo 'ebo 'breast-pocket'

as hybrids such as J.Urmi carmena, (f. carmenta) 'squint-eyed' < Azer.geri 'squint-eyed^^ -|- Aramaic m- {< *man) 'from', ena {< *'ayna)'eye'; Jilu '^ponddlene 'eyeglasses', literally: 'trick on the eyes' < Kurd.fend 'trick' -I- Aramaic ai (< *'al) 'upon', ene (< *'ayne) 'eyes';^^ andJ.Sanandaj tregyane 'pregnant', literally: 'two souls' < Aramaic tre +Kurd, giyan 'soul' + Aramaic pi. ending -e.

Some NENA words of obscure etymology may well owe their originto the Aramaic of antiquity, yet their forbears have never found theirway into classical Aramaic literature and lexica.''' Thus the origin ofNENA syl 'to fuck', unrecorded in OA sources, can be rather safelydetermined as sU in view of Rabbinic Hebrew sU 'to commit lewdness'and the regular shift of geminate verbs to Il-y ones in NENA. ^ Perhapsthe almost ubiquitous NENA word quta (Turoyo quto) 'vulva' belongsto this category too, as well as pode (and cognates) 'snot' mentionedabove, and other words not necessarily obscene or unpleasant, whichpossibly stem from ancient Aramaic vernaculars that did not gain lit-erary status, e.g. in the case of qlw 'to (be) clean', qliwa 'clean', foundin a large number of NENA varieties, but absent in classical Aramaicidioms. ^

Often a cross-dialectal comparative study within the same dialect-group is indispensable for clarifying challenging etymologies, anci evena single dialect may furnish a missing link crucial for laying bare theorigin of an enigmatic word found in a kindred dialect. For instance,

in O. Jastrow, Laut- und Formenlehre des neuaramaischen Dialekts von Midin im,Tur 'Abdin (Wiesbaden: Harrassowitz, 4th edn, 1993), p. 178).

^^P. O'SuUivan et al, Azerbaijani-English Dictionary (Kensington, Md.: Dun-woody Press, 1994), p. 343a, s.v.

^^See S.E. Fox, 'The Three Brothers: A Folktale in t;he Jilu Dialect of Neo-Aramaic', forthcoming in Mediterranean Language Revievj, n. 68. For the Kurdishloanword fend see B. Rizgar, Kurdish-English English-Kurdish Dictionary, p. 78,S.V., H.J. Polotsky, 'Notes on Neo-Syriac Grammar', 705 16 (1996), pp. 11-48 (46,§5.3). Foreign f is regularly adapted as p in Jilu: Kurd, fend > pgnd, Kurd, tifiqin> tpaqa 'to meet', etc.

'•'This notion was already expressed by D.T. Stoddard, Gramm.ar of the ModernSyriac Language as Spoken in Oroomiah, Persia and in Koordistan (New Haven:B.L. Hameln, 1855 [= JAOS 5 (1856), pp. l-180a-hj), pp. 8, 180a; and recently bySabar, JNAD, p. 12.

i^For this etymology see Sabar, JNAD, pp. 12; 239b, s.v. b-'-O.^^ln H. Mutzafi, The Jewish Neo-Aramaic Dialect of Koy Sanjaq (Iraqi Kur-

distan) (Semitica Viva 32, Wiesbaden: Harrassowitz, 2004), p. 238b, s.v. qlw, Iconjecture that qlw might be a reflex of unattested *qlb 'to be or make tidy', adenominative of qaiba (Syr. t<=Ari) 'mould, pattern'.

MUTZAFI Etymological Notes 87

the origin of the S-Tiare word pdss- bound to the phrase dia poss- +pronominal suffix, meaning 'without one's intention' (as in twirali diaP9ssi 'I broke it unintentionally') would be practically unidentifiablewithout a comparison with the A-Tiare expression dia b'dswa 'unin-tentionally' (usually with a pronominal suffix, as in twirali dia b'sswi),comprised of the preposition b- and a form of the well-known Aramaicroot sby 'to wish, desire'.^^

In addition to the etymological remarks given above, in what fol-lows I shall propose etymologies for ten lexical items selected fromworks related to NENA, primarily dictionaries and glossaries. Furtheretymologies will be discussed in passing.

bola 'hair of the head'^^ and cognate forms and meanings

The word bola, referring to the hair of the head in J.Zakho and kin-dred Jewish dialects of northernmost Iraq, is cognate with the botan-ical term bola, boUa, bula or bulta in various rural Christian NENAdialects. The original denotation of bola (and cognates), as preservedin some conservative NENA dialects spoken by Christian peasants, e.g.C.Aradhin, Tiare and Tkhuma, is 'panicle', i.e. a tawny infiorescenceat the head of certain plants, resembling, in a way, a lump of untidydown or soft hair, or a small dishevelled broom. Plants with pani-cles in the NENA-speaking area are the reed {qanya, qaniOa, zaia),sorghum [xrowiye, xroriyye, etc.), millet {prage, daxna, garas) maize{xatte romaye, xotte samaye) and rice (razza).

Since the panicles of sorghum, millet and rice are grain-producingheads of crop like ears of wheat and barley, in certain Christian dialectsthe word bola (and cognates) has also assumed the denotation of 'ear ofcorn', and in some cases even ousted the common NENA word sibglta(and cognates) 'ear of corn' altogether. Thus in the CUrmi dialect-cluster bula denotes 'panicle, ear of corn' (still distinct from sa

0,M, Gewargis Ashitha, Hilqa de Leshana: Assyrian-Arabic Dictionary(Baghdad: Al-Maghreb, 1997), p, 427a: r<^^r<r, i < i ^ r ^ r X'.i, s,v, f<=i . The OAroot sby has no verbal forms in NENA, where for the sense of 'to wish, desire'a reflex of the Aramaic root b'y (mostly b'y or 'by) is employed instead, I aminclined to assume that the form 'aswa as part of the word b'aswa came into beingby analogy with words of the pattern qatia with ' as the first radical, e,g, A-Tiare'aspa (< *yezpa) 'borrowing', sqiJii xacci qamxa, b'gspa 'I borrowed some flour',S-Tiare 'srqa (< *'erqa) 'running, going quickly', zilli b'arqa 'I went quickly' (cf,Sabar, JNAD, pp. 98b, 101a, s,vv, KSOS,

Sabar, JNAD, p. 106a, s.v,

'ear of corn'),^^ and in C.Nerwa and 'Aqra bulta means 'panicle, earof corn',^° with no Sdbalta. In Baz bola means 'panicle; ear of corn',whereas sdble (sg. sdbla) means 'crop left on ground for the poor toglean'. In some other dialects sibdlta (and cognates) has assumed theadditional meaning of 'panicle' and then ousted bola (at least as abotanical term), e.g. in the case of Betanure subaWa 'ear of corn, pan-icle'. Yet most rural Christian dialects retain two distinct words for'panicle' and 'ear of corn', as the following illustrates:

S-TiareM-TkhumaTelkepe'Ankawa

paniclebolabulaboUabolla

ear of cornsibdlOa

siblasdbdltasubalOa

As panicles look somewhat like dishevelled hair at the top of reeds,sorghum etc., in the C.Urmi dialect-cluster bula is also used to denote'wisp of hair', and in Christian dialects of Hakkari, Turkey, bola or bUlais occasionally employed as an impolite or jocular word for the tresses ofa girl or woman, especially when unkempt, e.g. in Tiare (mother sayingto daughter angrily): 'elisso, sruqle bohxl 'comb your hair, Elishsho!',or 'elisso, 'an la garmatle pumnidx, garsanne bolaxl 'Elishsho, if youdon't shut your gob, I'll pull your hair!'. In J.Zakho and other Jewishdialects of northernmost Iraq, spoken mostly by townspeople, the wordbola has lost its botanical meaning, and refers to hair of the (human)head only. ^

As for the etymology, the prevailing theory has been Noldeke's sug-gestion that bula 'ear of corn' is derived from *sbola < K'piBC, attestedin Targum Job 24.24.^^ Noldeke's idea was followed by Maclean and,

^"in A.J. Oraham, Dictionary of the Stabilized and Enriched Assyrian Languageand English (Chicago: Consolidated Press, 1943), p. 45a, s.v. t<\'ci.rj, the denotations'blade (of leaf), spire' etc., nnknown from any dialect, axe probably a misnse ofterms for 'ear of corn' (and the same applies for Oraham, Dictionary, p. 492b, s.v.

'blade', etc.).f. the C.Nerwa pi. forms in Talay, 'Ein Tenfel im Kloster', p. 131, section 17:

bole, bulyaOa d-xstte.^^For the semantic connection between ear of corn and hair cf. French epi 'ear

of corn; tnft of hair, cow-lick' and Ar. iL—, 'ear of corn; moustache'.^^Th. Noldeke, Grammatik der neusyrischen Sprache am Urmia-See und in Kur-

distan (Leipzig: T.O. Weigel, 1868), p. 100.

later on, by Tsereteli.^^ This etymology is encumbered by the factthat, as is shown above, the original denotation of bola is 'panicle'rather than 'ear of corn', and by the phonologically unappealing re-construction *sdbbola > *sbola > bola, involving the very question-able elision of initial s, apparently unparalleled throughout the NENAdialect-continuum.

I believe that the key to this etymological puzzle lies in the di-alectal form boUa, found mainly in the plain of Mosul, that shouldbe considered older than the other cognates—bola, bula and bulta—since unconditional non-etymological gemination is virtually unknownin NENA dialects, whereas degemination is ubiquitous. Moreover, ina few cases the dialects of the plain of Mosul preserve etymologicalgemination that has been simplified in many other NENA vajieties,e.g. Telkepe, Alqosh, Baritle massaOa 'balance, scales', Telkepe, Baritlesappuda 'skewer', against degemination in Tiare, Tkhuma, Betanureetc. masaOa, sabuda. In light of the oldest cognate form in existence,boUa, the forbear of our word is in all likelihood a passive partici-ple of unattested pa"el of bit *mbulMa 'untidy, dishevelled' (a salientfeature of panicles), whence *mbolla (compare *mxulldla 'washed' >Tiare mxoUa, Telkepe, Betanure, C.Aradhin, etc. mxuUa). The changeof *mboUa into boUa can be readily explained as resulting from elisionof m in the initial cluster mb (compare, e.g., J.Zakho *mbaqqarta >baqarta 'question'). Thereupon degemination gave rise to bola in mostdialects. In dialects exhibiting the shift o > 0 in open syllables {*slota> +sJuta 'prayer', etc.), e.g. in C.Urmi, we find bula; and in C.Nerwaand 'Aqra the feminine form bulta is derived from bola (compare zora'small' - f. zurta, smoqa 'red' - f. smuqta in these dialects).

As for the connection of the root bll to untidy hair, consider thefact that in various NENA dialects the verbal root blbl 'to be mixedup; untidy, dishevelled (and related meanings)', derived from bll (thepostulated root of bola), is often linked with the hair of the head, e.g.in Betanure bola mbiilbdla 'dishevelled, unkempt hair', and likewiseBaz bola bulboUa (or kawsa bulballa), Marga parta mbolabla.^'^

^^Maclean, Dictionary of the dialects of vernacular Syriac (Oxford: ClarendonPress, 1901), p. 27b, s.v. r^a=i, K.G. Tsereteli, A Reader ofthe Modem AssyrianLanguage with a Dictionary (in Russian; Tbilisi: Tbilisi University Press, 2nd edn,1980), glossary, p. 200b, s.v. I<!SCL=.

^''And see A.J. Maclean, Dictionary, p. 153b, s.v.

bota 'edible seed, pip, pit of fruit'^^ and cognates

The form bota 'edible seed (especially roasted), pip, pit of fruit' inJ,Urmi and other Jewish dialects of Iranian Azerbaijan is cognate withbo'ta ('Amadiyya), boca (e,g, in Betanure, Tiare) and "'"buca (e,g. inCUrmi), in the same meaning, and with bu'ca (Alqosh) 'kernel of ed-ible seed, husked seed' and boya (J.Zakho) 'seed of melons, pumpkins,eaten as a snack'.^^ Curiously enough, all these cognates are stemmingfrom the same origin attested in JBA as xnuiD, Km3 'abscess', pi, "UID,'iJl3, "lD; '' and compare also Mandaic pi. N'la 'boils' and the form K]Knn'abscess',^* The cognates closest to the etymon are J.Urmi bota and'Amadiyya bo'ta, with a feminine ending -ta and a plural form boe andbo'e, respectively. In addition, Alqosh bu'ca still preserves the originalconsonant '. ^

This etymology is supported by the fact that in a very large numberof NENA dialects, in Syriac and in several adsti'atal languages, thereare words representing a metonymic link between the senses of 'grain,seed, berry' and 'pimple, pustule, boil, abscess, wart', all being granularor berrylike in shape. Examples from merely a fe;w NENA dialects:

Forbear NENA dialectsj- ' tsi 'seed, grain, pit' J,Zakho parsanta 'kernel, grain; pimple,

abscess'

^^According to I, Garbell, The Jewish Neo-Aramaic Dialect of Persian Azerbai-jan (Janua Linguarum, Series Practica 3; The Hague: Mouton, 1965), p, 300a, s,v,bota, and my informants, Garbell renders the word bota: i, roasted seed, ii, cru-cible, but two homophones should be distinguished under separate entries: bota 1,pi, boe 'edible seed' etc, < Aramaic (see below); and bota 2, pi, bote 'crucible'< Persian bota; see F,A, Steingass, Comprehensive Persian-English Dictionary(London: Lowe and Brydone 1892), p, 205, s,v,

26Sabar, JNAD, p, 106a, s,v, S'a,• See Sokoloff, A Dictionary of Jewish Babylonian Aramaic of the Talm,udic and

Geonic Periods (Ramat-Gan: Bar-Ilan University Press, 2002) (henceforth DJBA),p, 191b, s,v. sn ra , Nm: ,

^^See E,S, Drower and R, Macuch, A Mandaic Dictionary (Oxford: ClarendonPress, 1963), p, 55a, 57b, s,vv,

2^The J,Zakho oral translation of Hebrew niJ3r3i? 'boils, blains' (Exodus 9,9) asbu'e-bu'e (Sabar, The Book of Exodus in Neo-Aramaic, in the Dialect of the JewishCommunity of Zakho, including selected texts in other Neo-Aramaic dialects and aglossary (in Hebrew; Jerusalem: Magnes Press, 1988), pp, xxxvii, 113; idem, JNAD,p, 106b, s,v,), might be an ad hoc rendering based on the Hebrew word rather thana doublet of boye, sg, boya 'seed eaten as a snack'.

MuTZAFi Etymological Notes 91

tnflp 'type of ulcer''^° Alqosh sassa 'seed of cardoon; corn (of skin),wart'

f. of r^ajii 'grain' Isnakh xabta (pi. xabe) 'grain; pimple,

'mulberry; ulcer' Tiare tuOa 'mulberry; large wart''grain, berry'^^ Betanure ma'ta 'single, detached grape',

J.Zakho ma'ta 'redness or pimple in theeye',^^ C.Urmi +mata 'pimple, pustule',"•"mated '*"anvj 'a single grape', Qaraqoshpi. ma'a 'smallpox'^''

Pertinent to the semantic connection under consideration are, more-

over, JBA xnnr:: 'grape; infected pustule or boil', Syr. rtfiiaX , 'lentil;

freckle, pustule' and modern Persian dana 'grain, berry, seed, stone of

fruit, grape; pimple'^^

As for the development of reflexes of *bo'ta in NENA dialects,

'Amadiyya bo'ta exhibits a regular shift *' > ', J.Urmi bota evinces

a regular ehsion of *' (cf., e.g., *balo'ta > J.Urmi balota 'throat'),

and J.Zakho boya has arisen by back-formation from the pi. boye <

*bo'e.^^ The forms bu'ca in Alqosh and boca or buca in Betanure and

f. Akkadian sissu 'ulcer, abscess, protuberance, wart'.^ Likewise as regards habto, hapto in Turoyo dialects; see A. Tezel, Comparative

Etymological Studies in the Western Neo-Syriac (Turoyo) Lexicon (Acta Universi-tatis Upsaliensis: Studia Semitica Upsaliensia 18; Upsala: Elanders Gotab), p. 129,and habba in Iraqi Arabic; see D.R.Woodhead and W. Beene, A Dictionary of IraqiArabic: Arabic-English (Washington, D.C: Georgetown University Press, 1967), p.89b, s.vv. ha,bb, habba.

^^Cf. Biijlical kebrew, Isa. 48.19 n^D 'grain (of sand)', Hebrew of the Pales-tinian Talmud nvD 'kernel; small coin' (J. Levy, Worterbuch iiber die Talmudimund Midraschim (4 vols., Berlin: Benjamin Harz, 2nd edn, 1924), vol. 3, pp. 183-84); and in OA attested only in the transferred meaning of '(small) coin', e.g. Tar-gumic Aramaic of Onqelos-Jonathan type sn:Ja, KUD (G.H. Dalman, Aramdisch-neuhebrdisches Handworterbuch zu Targum, Talmud und Midrasch (Frankfurta/Main: Kauffmann, 1922), p. 247a, s.v.), and Syr. rsliiso. Proto-NENA *ma'tamust have retained a botanical meaning which was no longer retained in classicalAramaic languages.

^^Sabar, JNAD, p. 210b, s.v. KnKD 2. ma'ta was glossed by an informant drakomain Iraqi Arabic, i.e. 'trachoma'. Note that in modern Hebrew trachoma is callednj^nj, derived from ]'l)n3 'grain'.

•'''Khan, Neo-Aramaic dialect of Qaraqosh, p. 736.''^Steingass, Persian-English Dictionary, p. 501b, s.v.^^For another example of the rare shift *' > y in J.Zakho and closely related

Jewish dialects see Sabar, JNAD, p. 144b, s.v. nssm, dra'a (< *dra'a) 'yard, cubit',pi. dra'e, draye.


many Christian NENA varieties have evolved by pharyngealization oft due to partial assimilation to the preceding pharyngeal,^'' followedby alization, viz. *bo'ta > *bo'ta > *bu'ca > bu'ca, boca etc.' ^ Thepharyngeal is retained in Alqosh bu'ca due to the influence of a formerc in *bu 'ca, which at some point lost its emphasis by dissimilation from'.^^ This emphatic c is still present in the cognates boca and buca, whichin turn lost the pharyngeal '. Thus the load of contiguous pharyngealand pharyngealized consonants in *bo'ca or *bu'ca was resolved byneutralizing pharyngealization in the form bu'ca and by elision of ' inboca and buca.

"''cmy'' 'to be extinguished, to extinguish' and cognates

The verbal root '^cmy, represented by the verbs """cinaya 'to be extin-guished' and '^macmoye or ~^macmUyi 'to extinguish', and the cognatescrny and cmy, are typical of most NENA dialects of northernmost Iraqand Hakkari, Turkey, and of the Christian NENA dialects of the plainsof Urmi and Salamas in Iran.^^

Noldeke relates this verbal root either to Hehirew 7]2O 'to be extin-guished', or, rather, to Syr. cnLS>i:i 'to become dark"*^ (in fact mainly'to be bhnded'). Following Noldeke, Maclean suggested Syr. oisriA asa possible etymon,^^ and the same etymology is repeated by modernscholars, either with a question mark*'* or without.''^

Noldeke's suggestion is questionable, for in the NENA dialects ofnorthern Iraq and Hakkari, where the verbal root in question is cmyor the like, there are no examples of a palatalization of k yielding c,but only a rather small number of cases where k is palatalized into

•'^Admittedly, such an assimilation is quite rare in NENA.^^For t > c cf. *ntp > ncp 'to drip' in K-Tiare, *qtp > qcp 'to pluck' in Barzan,

among other dialects, *tmm > tym 'to close, shut' in Qaraqosh > cym 'id.' in otherNENA dialects.

^ Cf. Alqosh *so'a > *co'a > co'a 'smooth', as also in J.Koy Sanjaq; see Mutzafi,Jewish Neo-Aramaic Dialect of Koy Sanjaq, pp. 30-32, esp. 32.

''"See, inter alia, Sabar, JNAD, p. 132b, s.v. '-a-i.''^Elsewhere in NENA a reflex of OA d'k or nwh, depending on dialect, is used

for the same meanings, e.g. Telkepe d'x, Barzan tyx, J.Urmi nyx.^^Noldeke, Grammatik der neusyrischen, pp. 41, 237.''''Maclean, Dictionary, p. 134b, s.v. cniai, cn^su-*''See Sabar, JNAD, p. 132a, s.v. '-a-:. "'''^See Tsereteli, A Reader, glossary, p. 108b, Krotkoff, A Neo-Aramaic Dialect,

p. 163. In addition, in Oraham, Dictionary, p. 227a, the C.Urmi verb +cmaya ispseudo-etymologically spelled

non-emphatic c, primarily preceding e or i (e.g. Tiare cepa 'stone', cisa'bag'), and rarely preceding front a (e.g. Baz calwa 'dog', Tiare caka'tooth').

There can be hardly any doubt that the etymon is smy, infini-tive sma,ya, which, in addition to its original meaning 'to go blind', isrecorded by Maclean with the meaning 'to go out as a fire' concerningC.Urmi and dialects of Hakkari, and, similarly, 'to become dim (fire,light)' recorded by Tsereteli.''^ For the semantic connection betweenextinguishing and bhndness consider JBA "in: 'to extinguish' and fromthe same root I'm < Tn:* 'bhnd'."''

The phonological processes that have, in all likelihood, shaped theverbal root in question are *siny > *STny > crny. This reconstructionis buttressed by the occurrence of the following consonantal changes inNENA dialects:

Firstly, germane to the postulated process *STny > *smy is theoccurrence of a sporadic pharyngealization of m (often together withan adjacent r or 1) in NENA, including in dialects with the verbalroot ~^cmy or cmy, e.g. J.Zakho rnaya 'water', Betanure rnale 'hemay fill', qamaya 'first (sg.m.)', Marga mara 'spade, hoe', malxa 'salt'.The pharyngealization of m necessarily afi ects an adjacent s, as in K-Tiare *smadra > sammadra 'blossom ofthe vine', C.Aradhin *smora >sdrnmora 'squirrel', A-Tiare *mismara > basrnara '(metal) nail', Marga*mata > masa 'village'.

Secondly, the postulated change *smy > crny should be regardedas part of a sporadic shift of s to c, especially near sonorants, found ina large variety of NENA dialects, including those with the verbal rootcmy, e.g. *semsa (cf. Syr. rf^^) > J.Zakho (etc.) camca 'secretionfrom the eye, sleep', mrasa > *mrasa (possibly preserved in TelkepeIdbbi inrgsle 'I had severe chest pains' < *'my heart was crushed') >'Amadiyya (etc.) mraca 'to crush, squash', *may3s 'azze 'type of lizard'(lit. 'goat sucker')^* > *maysizzi > Telkepe maycizzi 'lizard', *sla.pa (cf.Syr. ^ ^ 'to bruise') > C.Aradhin (etc.) clapa 'to split',^^ *nisawsoye(cf. Syr. -'^a^, Telkepe swasa 'to chirp') > Betanure mcawcoye 'tochirp'.

Maclean, Dictionary, p. 227a, Tsereteli, A Reader, glossary, p. 148a.Sokoloff, DJBA, p. 263a, 276b. Cf. also Syr. t<ri.cnv>^'dim-sighted; blind',

f. Maclean, Dictionary, p. 152b, s.v. r^v^ ^ r < ^ , and apparently already inSyriac, see SokolofF, DJBA, p. 533b, s.v. s'7".' '

"•"Krotkoff, A Neo-Aramaic Dialect, pp. 123, 167.

There remains the cognate cmy with non-emphatic m in certainNENA dialects, e.g. Tiare and CAradhin, which does not tally wellwith the proposed process *smy > *smy > cmy. This cognate mightbe an offshoot of cmy that has arisen by dissimilation of *m from thecontiguous emphatic consonant c.

kokr 'to be or make sad^^ and cognates

The J.Zakho verbal root kokr, having the 3 sg.m. present base mkoksrand infinitive form mkokore, is cognate with Dobc; kwkr 'to be or makesad, upset, annoyed', J.Koy Sanjaq kfkr 'id.' and Naghada kfkr 'tobrood with disturbing thoughts'. The latter cognate is the most con-servative in terms of both form and content, for, surprising as it mayseem, the J.Zakho verbal root kokr is derived from fkr 'to think' (<Kurd. < Ar.).5i

The derivation fkr > kfkr with the second radical repeated be-fore the first is part of a well-documented proc(iss in various NENAdialects, with parallel examples such as J.Koy Sanjaq, ncl 'to pluck'> cncl 'to pluck all (e.g. feathers); to tear to small bits', C.Urmi npl'to fair > pmpl {*pnpl) 'to stagger, stumble' and Barzan nsl 'to drip,trickle' > snsl 'id.'.^^ The second radical o in J.Zakho results from acontraction of *aw < *af; compare Dobe kawkar 'he is sad, upset',J.Koy Sanjaq kafkar 'id.', and the C.Urmi verbal root tots, presentbase totas 'to search, investigate' < *twts, Hawtds < tfts, mtaftds asis still in J.Zakho < Ar. ji~lj.^^ The derivation fkr > kfkr involved asemantic change whereby fkr retained the sense of 'to think' and itsoffshoot kfkr gained the more limited sense of 'to brood with disturb-ing thoughts', as still in Naghada. "* A moderate semantic shift fromthis latter meaning to the meaning of 'to be or make sad or upset' haslater on affected the verbal root kfkr (J. Koy Sanjaq) and its offshootskwkr (Dobe) and kokr (J.Zakho). To sum up, the process that yielded

5°Sabar, JNAD, p. 182b, s.v. "l-D-l-D.^^The original meaning of Ar. fkr is obsolete in Dobe, J.Koy Sanjaq and Naghada,

where it presently means 'to understand', unlike J.Zakho fkr 'to reflect, contem-plate' (Sabar, JNAD, p. 263a).

^^See further in Sabar, 'The Quadriradical Verb in E)astern Neo-Aramaic Di-alects', JSS 27 (1982), pp. 149-176 (156 §5.4).

^ Cf. Stoddard, Grammar, p. 17 ^cijtivnii 'search ye' and Maclean, Dictionary,p. 207b, s.v. ot;Ai.aAv»).

"•See further in H. Mutzafi, 'Two Texts in Barzani Jewish Neo-Aramaic', BSOAS67/1 (2004), pp. 1-13 (4 n. 30).

J,Zakho kokr is: *fkr 'to think' > *kfkr 'to have disturbing thoughts'> *kwkr 'to be or make sad' > kokr 'id.'. Thus one of the oddest-looking verbal roots in NENA reflects a well-known, common Arabicverbal root, that has undergone a series of far-reaching phonological,morphological and semantic changes,

sibarGa 'bracelet ^^ and cognates

The Tiare word sibarda 'bracelet', and cognates such as suborOa orsibarta in some other NENA dialects, are of no attested OA ancestor.This word is a borrowing from Akkadian, more precisely Neo-Assyrian,sabirru 'bracelet (and related meanings)'. *^ The Akkadian femininegender is expressed morphologically in NENA, with a forbear suchas *sabirr3ta, later *sibirrata, apparently with vowel assimilation >*sibarr9ta > sibarOa.^'' The cognate form subarda in Betanuire andclosely related dialects such as 'Amadiyya^^ has u by partial assim-ilation to b, and in some other dialects *t became a stop t as part of aregular shift, e.g, MarBishn *sibarta > sibarta like *kabbta (preserved,e,g,, in Betanure) > MarBishu calabta 'bitch',

sibarOa is one of several NENA words of Akkadian origin unat-tested in OA, or closest in form or meaning to Akkadian, Among thesewords are Qaraqosh baxsima 'storeroom for grain in the roof, fromAkk. bit hasimi 'barn, storehouse' and most probably Qaraqosh raxisa'pile of straw' from Akk, rahisu 'pile of harvest produce, straw';^^ theNENA cognate words for 'ladder', such as sammalta {e.g. in Qaraqosh),sammalta (Betanure), simalta (Sat) and semalta (S-Tiare)—all moreclosely related to their Akkadian ancestor simmiltu 'stair(case)' thanSyr, r<rA[ flD or Mandaic sn' 'DDlO are;' ° NENA nps, npasa (in J,Zakho

e, inter alia, Gewargis Ashitha, Hilga de Leshana, p. 522b (precise pronun-ciation is according to informants),

^ CAD, S: p, 219, s.v. semeru, AHw III: p. 1250, s,v, sewerum.*sibarr3|a > sibsrto cf, *qarrsta 'cold weather', *'arra|a 'co-wife' > NENA

arfe, 'ar6a,f, the plural form of this cognate, subraOa, in Sabar, JNAD, p, 295b, s.v.

^^For both baxsima and raxisa see Khan, Neo-Aramaic Dialect of Qaraqosh, p,515; idem, 'Some Remarks on Linguistic and Lexical Change in the North EasternNeo-Aramaic Dialects', Aramaic Studies 1,2 (2003), pp, 179-90 (185),

^"Cf. S,A, Kaufman, The Akkadian Influences on Aramaic (Chicago: Universityof Chicago Press, 1974), p, 92, In Telkepe and 'Ankawa, however, the form issimbilta, ostensibly closer to Mandaic, but an independent development *simmilta> simbilta is quite possible.

nbs, nbasa) 'to card (wool, cotton)', which may well stem from Akk.napasu 'to comb and clean wool', whereas the Syriac cognate verb istwoii; ^ possibly Tiare six, slaxa 'to uproot, pluck' from Akk. salahu 'topull out, uproot'; and C.Urmi babla, Tiare bibla 'hly',^^ perhaps froman unattested Akkadian form related to babaltu 'a certain plant'.

+tata, tata 'cock's comb; crest of hoopoe'^^

+ tata and tata (depending on dialect), feminine in gender, are foundin a large number of Christian NENA dialects and in Betanure. Theorigin of this enigmatic word is *tagta < *tagta, an unattested feminineform of OA taga 'crown', e.g. JBA x:n, XJKn, Syr. r < ^ , rCi^ik, andMandaic X3Sn, from Middle Persian tag.^''

"•"tata and tata are regular reflexes of tagta: OA g shifted to ' inProto-NENA, and subsequently shifted to ' or was elided accordingto dialect.^^ There are still rehcs of ' stemming from g in NENA, e.g.'Ankawa sra'a 'oil lamp' < *sraga, 'Amadiyya (and closely related di-alects) Voya 'inside', from an OA form related to 1-gaw,^^ and Betanureya'isa 'wife's sister's husband, brother-in-law' < *'agisa.^'^ In dialects

^'Additionally, Ar. JiiJ 'to card' is in all probability a loanword from the sameunattested Aramaic source from which the NENA form has emerged.

' ^Gf. Oraham, Dictionary, p. 40a, Gewargis Ashitha, Hilqa de Leshana, p. 42a(and informants).

^^See Maclean, Dictionary, p. 324b, s.v. r<'&-i.A\, Oraham, Dictionary, p. 558b,s.v. r<rA\A\ (precise pronunciation is based on informants). The spelling r iLi.A\ with.i. in Maclean is meant to indicate word-emphasis, i.e. +tata; compare, e.g., p. 245b.r<xh\^ for + 'scca 'nine'.

''"See Sokoloff, DJBA, p. 1193b, s.v. wn, NJSn. The form taga 'crown, wreath'and the st.abs. form tag are still retained in Neo-Mandaic (R. Macuch, Handbookof Classical and Modem Mandaic (Berlin: W. de Gruyter, 1965), p. 498, s.v. crown,526, s.v. wreath; idem, Neumanddische Tcxte im Dialekt von Ahwdz (Wiesbaden:Harrassowitz, 1993), p. 441). Furthermore, the OA form taga still survives in NENAas taxa 'wreath', a doublet of tata occurring in Tiare wedding songs as an adaptedclassicism (Tiare has no phoneme g), e.g. in the shorl; repetitive song xatncnib-rese taxa (< *taga), kSlya-le qam buraxa, t-saqal razsid zuwaxa (< *zuwwaga)'our bridegroom, a wreath upon his head, is standing a,t the wedding ceremony,that he may receive the sacrament of marriage' (informant). Except for the poeticword taxa, the regular colloquial word for 'wreath' in Tiare is klila.

"^See Mutzafi, Jewish Neo-Aramaic Dialect of Koy Sanjaq, p. 39, no. 5.^^Apparently from *J-gawya < *l-gawwaya, 'to the inner part'.''Gf. Syr. '^•^-\f, an alternant of •^'^•\ with gg vs. NENA *g. Gf. also the g in

the Hebrew cognate O'M (Mishna, Kaufmann Ms., e.g. in Sanh. 3.7). Alternatively,the form ya'isa (and cognates such as G.Aradhin ya'isa) might have evolved froma vocative usage such as ya gisa 'brother-in-law!', ya gisi 'my brother-in-law!'.

in which the pharyngeal ', including ' < *g, has often triggered word-pharyngealization prior to its elision, e.g. in C.Urmi and Sat, we findthe cognate +tata; cf. C.Urmi """yaysa 'wife's sister's husband' < *ya'isa< *'a,gisa and Sat +xata (= Qaraqosh xa'ta) 'anagyris, a type of thorn-bush' < *xa'ta < *hagta (Syr. KT^ii^). In other dialects, in which ' waselided intervocalically without a trace, e.g. in Tiare and Tkhuma, wefind the form tata; compare, for instance, xata 'anagyris' in these di-alects. In Betanure the regular refiex of *' (< *g) is ' in all positions,hence one would expect ta'ta rather than tata as a refiex of *tagta. Theirregular refiex in Betanure may be explained as due to an infiuence orborrowing of neighbouring Christian NENA tata.

A very close parallel to NENA tata in terms of etymology and se-mantics is Mlahso Eastern Neo-Aramaic togo 'cock',^* undoubtedly asynecdoche of an earlier denotation *'cock's comb' < *'crown'. Com-pare, moreover, modern Persian taj 'crown, diadem; crest, tuft, combof bird.69

texon 'brother's wife'™ and cognates

The etymology of the word texon in the Jewish dialect of Saqqiz can beconfidently unravelled through a comparison with dialectal cognates,which are all restricted to certain other dialects of the Trans-Zab group,to wit the group of Jewish NENA dialects located east of the GreaterZab river:'' J.Sulemaniyya texona, Kerend 'atexona, J.Sanandaj'atxona, Dobe tiyaxona, Arbel and J.Koy Sanjaq tyaxona. ^These cognates of Saqqiz texon evince the final element xona or axona,representing the word 'axona 'brother', originally a diminutive form of'axa (still so in 'Amadiyya) < *'ahha. The short form xona, with eli-sion of the initial syllable, is common in Christian NENA dialects ofnorthernmost Iraq and Hakkari, but in the Trans-Zab group this form is

h. Talay, 'Die Geschiehte und die Spriiche des Ahiqar im neuaramaischenDialekt von Mlahso', in W. Arnold and H. Bobzin (eds), "Sprich doch mit deinenKnechten aramdisch, wir verstehen es!" 60 Beitrdge zur Semitistik, Festschrift fiirOtto Jastrow zum 60. Geburtstag (Wiesbaden: Harrassowitz, 2002), pp. 693-712(700 no. 65; 712).

''''Steingass, Persian-English Dictionary, p. 273a, s.v.™Y. Israeli, The Jewish Neo-Aramaic language of Saqqiz (Southem Kurdistan)

(in Hebrew; Doctoral Dissertation, Jerusalem: Hebrew University, 1998), p. 65.' ^For this dialectal group and some of its salient features see Mutzafi, Jewish

Neo-Aramaic Dialect of Koy Sanjaq, pp. 9-10.' • For J.Koy Sanjaq tyaxona see Mntzafi, Jewish Neo-Aramaic Dialect of Koy

Sanjaq, p. 248a, s.v.

confined to the word texona, 'atxona etc., the regular word for 'brother'being 'axona. In Saqqiz the earlier form *texona lost its final vowel firstin the vocative form, where, by rule, stress shifts to the first syllable ala Kurde. Just as in J.Sulemaniyya the vocative of texona is *r;exona> texon, with apocope of a due to the remoteness of stress, so is thevocative texon in Saqqiz a truncated form of *texona. By analogy withthe vocative form texon, the basic, non-vocative form *texona becametexon in Saqqiz.

In light of the cognates 'atexona and 'atxona, the first element ofthe word in question must be *'attat (Syr. ki^rtjP the construct formof *'attdta (Syr. r^ 'iur^) 'woman, wife'. In all cognates but J.Sanandaj'atxona the Iranian annexation particle (izafe) -i intervenes between areduced form of *'attat, 'wife of-' and the following form xona or axona'brother'.^* Thus, e.g., in Kerend *'ata-i 'axona (wife-of brother) >*'ata-i xona > *'atayxona > 'atexona.

The sequence tt of *'attat has undergone typical NENA degemi-nation following the vowel a, as, e.g. in Arbel *'attuqa > 'atuqa 'old(thing)', Telkepe *'attira > 'atira 'rich'. The final *t was ehded aspart of a sporadic feature in NENA—compare, e.g., Kerend *l-i)mmat> hmma 'when?'. In certain dialects the initial syllable of *'att.at hasalso been ehded, possibly due to the remoteness of stress, e.g. in Saqqiz*'ata-i 'axona > *ta-i xona > Hexona > texon—compare in the samedialect *'attuqa > *'atuqa > tuqa 'old'. In the province of Arbel allthat has remained from *attat, is the consonant !:: *t-i 'axona > Dobetiyaxona, Arbel, J.Koy Sanjaq tyaxona.

Fossilized refiexes of *attat 'wife of are found in several other com-pounds in the same Trans-Zab dialects:

'atebaba (Saqqiz, J.Sanandaj), tebaba (J.Sulemaniyya) <i baba 'stepmother', literally: 'wife of father'

temon (Saqqiz), temona (J.Sulemaniyya), tyamona (Arbel, J.KoySanjaq) < *'attat-i 'ammona 'wife of paternal uncle'

the vocalization in Syriac see T. Ando, rtf Icuc riuA.i i<'A\yi.(ri (Treasureof the Syriac Language; 2 vols., 1897, Mosul, Reprint: Ann Arbor 1978 with thetitle Assyrian Dictionary), p. 37a, s.v. r^isidur^.

^'^izafe is occasionally employed in certain Trans-Zab dialects, e.g. in Saciqiz (Is-raeli, The Jewish Neo-Aramaic language of Saqqiz, p. 140: sammas-i knista 'thesynagogue's beadle', etc.) and in Rustaqa (G. Khan, 'The Neo-Aramaic dialectof the Jews of Rustaqa', in Arnold and Bobzin (eds), "Sprich doch mit deinenKnechten aramdisch", pp. 396-409 (408-409).

M U T Z A F I Etymological Notes 99

tixalona (Arbel), tixalon (J.Koy Sanjaq) < *'attat-i halona 'wifeof maternal uncle'; compare in the same meaning Turoyotah(h)olo < *'attat h

The words texon and 'atebaba and their cognates have close paral-lels in JBA and Syriac:

Trans-Zab NENA JBA^^ Syriactexon, 'atxona etc. XriK nn K'atebaba, tebaba S3N nn'K

There are also close parallels of these Trans-Zab words in Kurdish andin many NENA dialects which replaced the construct form *attat_ by aninnovative construct form of baxta, the pan-NENA word for 'woman,wife':

Trans-Zabtexon, 'atxona. etc.'atebaba, tebaba

Baritlebaxtaxonabaxtawa

Tkhumabdtxona (!)bajctbaba

Betanurebaxaxonabaxbaba

Kurdish^*jinbirajinbav, bawejin

The fossilized construct form 'attat in words such as 'atxona,'atebaba and temona is the only vestige of the common Aramaic wordfor 'woman, wife' (Syr. r<ri(\^r^, JBA xnn'N, Knra'x) in the entire NENA-speaking area, where this word has been replaced by baxta. ^

l, Comparative Etymological Studies, pp. 15, 218, 255.ff, DJBA, p. 128b.

with a 3 sg. m. pronominal suffix: ,mastri Âur^ 'his brother's wife'(Payne Smith, Thesaurus, col. 287). ~

^^These Kurdish compounds include the words jin 'wife', bira 'brother' and bav(northern Kurd.) or baw (central Kurd.) 'father'. Unlike the Trans-Zab parallels,the Kurdish compounds lack izafe, yet in the word bawejin the vowel interveningbetween baw and jin is a type of annexation particle (See Wahby and Edmonds,Kurdish-English Dictionary, p. 43a, s.v. -e 3).

'''^The NENA word baxta was in all likelihood created by back formation fromrrrfii'itiLi, pi. of r<rA»A\jsj3 'hired weaveress' (for the pi. form see T. Audo, IÂISI-JXJrîrN.T, p. 76a; cf. the form baxta,9a as a plural of baxta 'woman, wife' in manyconservative NENA dialects. The original denotation of the antecedent of baxtawas probably 'female spinner, spinster', for spinning is one of the most importantoccupations of women throughout the distribution of NENA, where women spinintensively, 'whether standing, sitting, or walking' (E. Brauer, The Jews of Kur-distan (completed and edited by R. Patai; Detroit: Wayne State University Press,1993), p. 129, and similarly on p. 181).

tpr 'to catch fire, burn; be furious, rebuke, incite; cling', tpr 'to cling'

The verbal root tpi, in some dialects tpr or """tpr, is found in JewishNeo-Aramaic varieties with the basic meaning ol' 'to catch fire, burn,be ablaze', and in a few of these dialects (e.g. Arbel) it denotes 'to befurious' or 'to rebuke, chastise'.*° In some Christian NENA dialects tpior tpr, depending on dialect, denotes 'to stick to, chng' or 'to incite'.*^

Maclean maintains that the origin of tpr is Arabic jis> (i.e. >1> 'toleap'?) and also connects it to tapra 'nail';*^ and Sabar suggests severaloptional etymologies: tupra 'nail', J.Zakho and related dialects tir 'tobe weary, loathe < ^ 'to leap, drive away, ruin, etc.', OA t,fl 'attach,cling to', or an unattested OA loanword from Akkfidian tiparu 'torch'.*^Sabar also quotes an unpublished proposal by Polotsky, according towhich the origin of J.Zakho tpr 'to burn' might be \^^S\i\r< ~ i.°>\^^r<r'to be angry, enraged'.

I cannot see a more plausible etymology for both tpr and tpr thanthe 'etpd'el form of the verbal root pwr 'to seethe, boil'*'' attestedin Syriac as •u^^^pf ~ •l*3^^r<f 'to be angry, enraged',*^ as alreadyconsidered by Polotsky for J.Zakho tpr. This etymology is buttressedby the fact that in modern Aramaic and elsewhere there is a close

*"According to informants and Garbell, Jewish Neo-Aramaic dialect of PersianAzerbaijan, p. 293a; Khan, A Grammar of Neo-Aramaic: The Dialect of the Jewsof Arbel (Leiden: E.J. Brill, 1999), p. 559; Sabar, JNAD, pp. 12, 66, 174b.

^^See Stoddard, Crammar, p. 180b; Maclean, Dictionary, pp. 113b, 172b, 325a;Gewargis Ashitha, Hilqa de Leshana, p. 582b; and Talay, 'Ein Teufel im Kloster',p. 140, no. 6: tpira b-an-garme, rendered 'nagte an den Knochen', but literally:'clinging to those bones' (C.Nerwa informant). In addition, the verbal root tpr isknown to me from Tiare and Tkhuma informants. I assume that tpr 'to pierce aswith the nails; figuratively: to cope with' in D.T. Stoddai'd, Dictionary of ModemSyriac and English (Yale University Ms. AOS Rn St64ni, n.d.), p. 150, and similarly'to put claws into e tc ' in Maclean, Dictionary, p. 113b is a denominative of tapraor tupra 'fingernail, toenail' and is etymologically unrelated to the verbal I'oot tprin question; see the etymology proposed below.

^•^Maclean, Dictionary, p. 113b, s.v. 1'^\,.^^Sabar, JNAD, pp. 12, 66, 174b, s.v. "l-S-Q. As regards the latter suggestion,

note that the ascertained Akkadian word is diparu rather than tiparu {AHw, vol.2, p. 172; CAD, D, pp. 156-57).

®*Cf. Turoyo dialects fyr, pyr 'to boil over', derived either from Aramaic pwror from the Ar. cognate root jji (cf. Tezel, Comparative Etymological Studies, pp.68-69).

Brockelmann, Lexicon, p. 560b. In Payne Smith, Thesaurus, col. 3071:

semantic bond between the denotations 'to catch fire; kindle' and 'to

stick to, cling', as in shown in the following table:

Forebear: NENA dialects®^1 uyi^ 'to seize, catch' lwx 'to stick; to catch fire; kindle'

2 ,°>N 'to attach' tpy, tpy 'to stick; to catch fire'3 ^JiJ• 'to hang, stick, 'lq, 'lq, ylq 'to hang, stick; catch fire, kindle'^^

catch'Turoyo

4 ^diD 'to thrust, stick, qyO 'to meet, touch, hit; kindle'**fix'

Neo-Mandaic5 Mand. "["ID 'attach; srx 'burn, intr.; kindle; start engine'*^

catch or set fire'

Similar semantic connections in classical Aramaic and elsewhere

in Semitic languages are the Syriac expression rtfioj i^i*ir< 'the fire

blazed', literally: 'the fire seized'; Samaritan Aramaic \/3O3, 'etpa'el

'to be taken, to blaze'; JPA -v/SD], 'af'el 'to cause to take (etc.), to

signal by fire'; Hebrew ^y^p'7, pa'al 'to take', hitpa"el 'to burn', y/^n^,

pa'al 'to seize; to burn (subject: fire)', hiPil 'to set fire';^° and Akkadian

dialects exhibit only some of the following denotations.^'^Similarly in Arabic dialects, e.g. Anatolian ones, see S. Vocke and W. Waldner,

Der Wortschatz des anatolischen Arabisch (Erlangen, 1982), pp. 287-88, s.v. 'lq;and in Neo-Western Aramaic, see G. Bergstrasser, 'Glossar des neuaramaischenDialekts von Ma'lOla', Abhandlungen fiir die Kunde des Morgenlandes 15.4 (1921),p. 3, s.v. 'lq.

^^These meanings are distributed according to verbal stems: Neo-pa'ai qoysQ 'tomeet, touch, hit' and Neo-'af'ei maqdO 'to kindle' (informant).

*^Drower and Macuch, Mandaic Dictionary, p. 335b, s.v. SRK; Macuch, Hand-book, p. 508, s.v. kindle; idem, Neumandaische Chrestomathie mit grammatischerSkizze, kommentierter Ubersetzung und Clossar (Porta Linguarum Orientalium 18;Wiesbaden: Harrassowitz, 1989), p. 241, s.v srk; idem, Neumandaische Texte, p.423, s.v. srx; and cf. Syr. o \ t o 'to adhere, stick', •^•^•' - r^^cu Av 'to) 'fire blazedin the forest' (Audo, t<uA.t C 'AVSI.JB, p. 181).

most of these cases and others see Z. Ben Hayyim, 'Miscellanea', in M.Goshen-Gottstein et al (eds.). Studies on Hebrew and Other Semitic LanguagesPresented to Prof. Chaim Rabin on the Occasion of his Seventy Fifth Birthday(in Hebrew; Jerusalem: Academon Press, 1990), pp. 11-23 (12-15). See also PayneSmith, Thesaurus, cols. 114, 120; A. Tal, A Dictionary of Samaritan Aramaic (2vols., Leiden: E.J. Brill, 2000), pp. 530-31, s.v. 3DJ; M. Sokoloff, A Dictionary ofJewish Palestinian Aramaic of the Byzantine Period (Ramat-Gan: Bar Ilan Uni-versity Press, 1990), p. 353, s.v. 202.

ahazu(m), basically 'to take, seize', S-stem suhuzu(m) 'to kindle fire'(among other meanings).^^

The semantic relationship between clinging or seizing and kindlingmay be (at least partly) explained as due to the perception of fire as ifadhering to the object it consumes and seizing it; compare expressionsin NENA like nura lwixle, tpele or 'llqle b9d xa beOa 'fire broke out(lit.: 'chnged') at a certain house'; and compare Enghsh to catch fireand touchwood.

The denotation 'to be angry' of Syr. t*^A^r«' ~ t ja^A\r<r is orig-inally a figurative usage of a basic meaning like 'to be hot, boiling,seething, burning', which is still represented by tpr (or t:pr, ~^tpr) 'tocatch fire, burn, be ablaze' in many Jewish NENA dialects. The Syriacmeaning 'to be angry' itself is represented by tpr in a few Jewish di-alects, tpr in Christian NENA dialects is assumed to have undergonethe semantic shifts *'to catch fire' > *'to catch fire; to chng' > 'tochng'.

The integration of the t prefix of erstwhile t-stem verbs into tri-radical verbal roots as the first radical, assumed for *'dttdpir > tpir,y/tpr, has many parallel cases in NENA. Thus, for instance, the OA'etpd'el form of hpk 'to turn, overturn' has become tpx, denoting 'tospill, shed' in C.Urmi and 'to fall forward' in Tiare.^^

Even if in Syriac t h e ^ of the alternant •U.A i»r<r does not representa phonetic t but is a result of incorrect scribal usage, as maintained byNoldeke,^^ still the secondary t in NENA tpr, tpr etc. can be readilyexplained as due to assimilation of t to an adjacent emphatic r, viz.tpr > tpr > tpr (later tpr by dissimilation?), for a pharyngealization

^•'I am grateful to Dr. Matthew Morgenstern for drawing my attention to theAkkadian parallel.

• Cf. Tsereteli, 'Uber die Reflexivstamme in den modernen aramaischen Dialek-ten', RSO 39 (1964), pp. 125-32 (129-30), and see additional cases in the same ar-ticle. Already in JBA we find -[Ein < -jEin'X 'to reverse' (SokolofF, DJBA, p. 1225a).For the semantic shift 'to reverse' > 'to spill' cf. French verser 'to pour, shed;overturn'.

^^Th. Noldeke, Kurzgefasste Syrische Grammatik (Darmstadt: WissenschaftlicheBuchgesellschaft, 2nd edn, 1966), p. 120, §177B. An alternative explanation mightbe folk etymology linking 'l».ad\d\t< with i, °> 'fire' (Payne Smith, Thesaurus, col.3071), in all likelihood from Gk. nup (for Gk. u > Syr.^ cf. xivSuvo? > a><x±^x±o'peril', an alternant of n>n<nt< .r,- gee Brockelmann, Lexieon, p. 676a), whichspawned an unaspirated emphatic t by partial assimilation to the contiguousunaspirated emphatic p < Gk. unaspirated n, viz. *tpr > *tpr > tpr, tpr, '^tpr;cf. R. Voigt, 'Die emphatische p des Syrischen', in R. La\'enant (ed.). SymposiumSyriacum, VII (Rome: Pontificio Instituto Orientale, 1998), pp. 527-37.


of consonants by assimilation to r that had undergone 'spontaneous'pharyngealization is very common in NENA, e.g. Telkepe *tarba >tarba > tarba 'tallow', C.Koy Sanjaq *bat,ar > *batar > batdr 'after',Betanure *sarna > sarna > sarna 'spinning-wheel' and Alqosh *besra> pisra > pisra 'meat', to adduce jnst a few examples.®"*

According to the etymology proposed here, tpr is phonetically moreconservative than tpr, preserving the t of the t-stem non-emphatic, butsemantically more progressive, representing a semantic shift from 'tobe on fire, burn', a meaning still retained by tpr, to the sense of 'tostick to, chng'.

xaro'ta 'loofah bathing sponge', xr' 'to rub and wash thoroughly,especially with a xaro'ta',^^ and related meanings

The fact that in J.Zakho xaro'ta is a loofah bathing sponge and theverbal root xr' is related to the rubbing and scraping of the body(also of the floor or dishes) and washing thoroughly, especially witha loofah, has prompted S.E. Eassberg and, more recently, Y. Sabar,to assume that these words reflect (or may reflect) a metathesis of therare OA root rh' 'to wash', otherwise unattested in Eastern Aramaic.^^The meaning in J.Zakho is, however, unique to this dialect, whereas inall other dialects that retain the lexemes in question, e.g. 'Amadiyya,J.Aradhin and Atrnsh, xaro'ta (in Atrush xarota) is an oval smoothstone used for rubbing and polishing a floor made of soil after thefloor is sprinkled with water, in order to level and smoothen it, and xr'(inflnitive xra'a) refers to the implementation of such a xaro'ta.

Since Eassberg and Sabar were familiar only with the J.Zakho usageof xaro'ta and xr', they certainly had good reason to flnd OA rh' anattractive etymon. Yet based on the abovementioned findings fromother dialects, I should hke to suggest an alternative etymology, namely

f. the observation by E. Coghill, The Neo-Aramaic Dialect of Alqosh (DoctoralDissertation, Cambridge: University of Cambridge, 2003), p. 39, that in Alqosh inmost cases of non-original emphasis there is a rhotic consonant present in tlie word.

^^Based on M. Yona, nn^J-TnD-'Q-lK ]i'p-a {Aramaic-Kurdish-Hebrew Dictionary(2 vols.; Jerusalem, 1999), vol. 1, p. 193b, s.v. nK")?, p. 195a, s.v. KHilD; Sabar,JNAD, pp. 12, 31 (§6.1), 64 (§16.1), 199b, s.v. K-n-D, 200a, s.v. KnSTO, arid infor-mants.

' See S.E. Fassberg, 'A Contribution of Western Neo-Aramaic to Aramaic Lexi-cology; rht/rwt; and rh", JSS 45.2 (2000), pp. 277-91 (290); Sabar, JNAD, pp. 12,31, 64, 199b. For the distribution of this root in Aramaic see S.E. Fassberg, 'AContribution of Western Neo-Aramaic', pp. 288-90, but as for J.Zakho see below.

*hrg, attested in Syr. .iyiu, basically 'to rub, polish' and in Mandaicnn 'to rub, grate'.^'^ This etymology is preferable on semantic grounds,furnishing a common denotation from which both 'to rub and polishthe floor' and 'to rub (and wash) the body' have come into being, andthe assumed development *hrg (> *xr') > xr' reflects regular NENAconsonant shifts. Furthermore, Syr. 4 ^ " has acc^uired the additionalmeanings 'to wash, clean, cleanse', as is shown by t;he Syriac and Arabicequivalents of this verb given in Payne Smith, .A udo, and Manna.^*Thus it seems that already in Syriac the basic raeaning 'to rub' haddeveloped a secondary meaning of 'to rub the body while washing,bathing', a semantic change that has also occurred in J.Zakho.

Reflexes of the Aramaic verbal root rh' remain, therefore, unlcnownin Eastern Aramaic.

+yaya 'furrow^^ and cognates

'^yaya 'furrow' in the C.Urmi dialect-cluster and C.Salamas is cognatewith yaya in Tkhuma and ya'ya in Telkepe and nearby dialects.^""Maclean claims that this word is a loanword from Azer. yay, but Ihave found no such word for 'furrow' in available dictionaries.^°^ Bear-ing in mind that Maclean's Azerbaijani etymologies originate 'from oralquestioning at Urmi' (Maclean, Dictionary, p. xii), one might suspectthat in the case of the alleged Azerbaijani word yay Maclean was (inad-vertently) misinformed by Aramaic speakers. Likewise, Maclean claimsan Azerbaijani origin for genuine Aramaic xurga 'step-son', f. xurggOa,attested in JBA as xmn, Nnmn. ' ^ Furthermore. Maclean considered

^''Drower and Macuch, Mandaic Dictionary, p. 152b., s.v." See Payne Smith, Thesaurus, cols. 1366-1367; Audo, [^ITN.T rfhca^js, p. 370;

Manna, rtfuA.t t<r<.iai, p. 261a, all s.v. ^lu.^^See, inter alia, Maclean, Dictionary, 119a; Tsereteli, A Reader, p. 74/8 and

passim, glossary, p. 98a; Gewargis Ashitha, Hilqa de Leshana, p. 220a (precisepronunciation is based on informants).

Maclean, Dictionary, p. 119a, s.v. r<il one finds also the (dialectal?) form

the denotation 'furrow' in Azer. I have found only lagar, see O'SuUivan,Azerbaijani-English Dictionary, p. 198b, s.v.

^°^See Maclean, Dictionary, p. 95b; and Sokoloff, DJBA, p. 443a, where smn isclaimed to be of Hebrew origin, though the existence of *hurga > xurga in NENA,primarily in Christian dialects, points to an Aramaic sourc(? ofthe NENA word, andJBA SJnin may well have been a genuine Aramaic cognate oi the Proto-NEN A wordpresently reflected as xurga. For the Azer. parallel see O.I. Musayev, Azdrbaycanca--Ingilisca Liigat {Azerbaijani-English Dictionary, Baku: Qismat, 1998), p. 451a, s.v.ogey, Maclean, Dictionary, p. 5a, s.v.

the Azerbaijani word di§l3m3^°^ used in the NENA expression sate cayeddslama 'to drink unsweetened tea while having a lump of sugar in themouth', as if it were native Aramaic da-slama 'of peace' (1) °''

Even if in some Azerbaijani dialects such a word as yay exists for'furrow', this word must be a NENA loanword, for both the pharyn-gealization of the form "'"yaya and the glottal stop of the cognate ya'yalead us to reconstruct *ya'ya as the direct ancestor of our word. * ya'yaitself stems from *yagya (compare ''"tata above, p. 96) < *'agya <*'agya] compare JBA K JX, i<':^v 'furrow', the Mandaic pi. form X':x'furrows, ruts', the Targumic construct form rfJiiJ 'irrigation groove' inEzek. 17.7 and the Mishnaic Hebrew pi. form riVTW 'irrigation groovesdug around a plant'.^°^ In view of these OA and Hebrew cognates, thehistorical Semitic root of "'"yaya appears to be *\/'wg 'to be curved' -compare Ar. ^ry^.

"'"yaya is one of several lexemes which NENA shares with one orboth of its close congeners of lower Mesopotamia, JBA and Mandaic,but not with Syriac. To this hst belong, inter aha, NENA baJa (e.g.in Isnakh) or waia (e.g. in Tiare), found in 'szzad hala or 'avwsd wala(< *'drbe dd-bala) 'wild goats' in relation to JBA x'73, s' xn 'rural', TIJx' X n 'wild goats' and Mandaic K'?xn 'prairie, etc.';^°^ NENA t'l (andcognates) 'to play', whose antecedent bba 'id.' is attested in JBA andMandaic;^°^ NENA dialects (e.g. Qaraqosh, 'Ankawa) sagi < *saggi 'go(sg.)!', clearly related to JBA and Mandaic "30 'to walk, go,

O'Sullivan, Azerbaijani-English Dictionary, p. 106b.Maclean , Dictionary, p. 307a, s.v. i<taXr..Sokoloff, DJBA, p. 79a, s.v. K'JX, K'JIU; Drower and Macuch, Mandaic Dic-

tionary, p. 5a, s.v. AGA 2; Dalman, Handworterbuch, p. 307b, s.vv. njliJ, Kn"31iJ; M.A.Jastrow, A Dictionary of the Targumim, the Talmud Babli and Yerushalmi, andthe Midrashic Literature (2 vols.. New York: Verlag Choreb, pp. 1047-1048, s.vv.S'Jii), srrJW. The punctuation of tlTJiiJ is based on Mishna, M. Qat. 1.1, KaufmannMs.

Sokoloff, DJBA, p. 220b; Drower and Macuch, Mandaic Dictionary, p. 48a.Sokoloff, DJBA, p. 505; Drower and Macuch, Mandaic Dictionary, p. 180a.

The historical processes that have yielded NENA t'l, which can be summarizedas *tll > *tyl (> *t'l?) > t'l, are discussed in extenso in H. Mutzafi, 'On theEtymology of the Verbal Root t'l 'to play' and its Cognates in North-Eastern Neo-Aramaic', forthcoming in ZDMG.

^°*See Sokoloff, DJBA, p. 787a; Drower and Macuch, Mandaic Dictionary, p.317a (the Syr. hapax i^mr^ is a Mandaic loanword, see Brockelmann, Lexicon, p.458b). The OA verb '30 survives in NENA only in the imperative of the verb 'togo' (to the exclusion of NENA dialects where this imperative is xus < *rxos, e.g.C.Urmi), in suppletion with inffections of 'zl in the other tenses. Already Panoussi

NENA dialects (e.g. Telkepe) qadamta 'early morning', whose forbearis akin to Mandaic xnDNiSp 'id., JBA xnonp 'dawnV°^ and NENA di-alects (e.g. Betanure) syn 'to feel freezing cold (and related meanings)',historically derived from sun, attested in JBA as ]]:i 'to cool',^^" by theNENA rnle of geminate > Ily

Conclusion

In conclusion, most of the foregoing etymological discussions and re-assessments illustrate the crucial importance of comparative inter-dialectal data for elucidating the origins of NENA words whose for-bears cannot be detected by mere recourse to classical Aramaic andother potential source-languages. Furthermore, since NENA is not adirect descendant of Syriac^^^ (hence the term 'modern Syriac' is mis-leading), but of Proto-NENA, it is not surprising to find NENA wordsrelated to JBA and Mandaic and absent in Syriac (see bota, ^yayaabove, pp. 90, 104), or borrowed directly form Akkadian (see sibdrOa,p. 87). Many NENA lexical idiosyncrasies display far-reaching semanticshifts, e.g. bola *'untidy' > 'panicle' > 'hair of the head' or bota 'pus-tule, boil' > 'edible seed', and, conversely, retention of ancient mean-ings lost in classical Aramaic, as in t,pr, basicallj' 'to catch fire, burn'compared with Syriac 'ettapir 'to be enraged'. In some cases ancientvestigial forms are combined with recently importied foreign words andparticles, as in texon 'brother's wife' < *'attat + Iranian -i + 'axona,and the occurrence of hybrid compounds in NENA has to be bornein mind when trying to unravel the tricky etymologies of odd-lookingwords such as '^pondalene 'eyeglasses' < *Kurd. fend + 'al 'ayne 'trickon the eyes' (see p. 86). As a matter of course, the more we gather

('Ein vorlaufiges Verbglossar zum aussterbenden neuaramaischen Senaya-Dialekt',RSO 65 (1991), pp. 165-83 (179 n. 30) correctly connected the C.Sanandaj formsay 'go (sg.)!' to Mandaic ':iD—especially given that the pi. form of say is sagon(Panoussi, e-mail). In light of dialectal forms like sagi, the imperative form si innumerous NENA dialects appears to be a reflex of *s'i < *s'i < *sgi, rather thana phonetically unmotivated reflex of *zi (= JBA 'T, see Sokoloff, DJBA, p. 100a) <*zil.^"^Drower and Macuch, Mandaic Dictionary, p. 399a; Sokoloff, DJBA, p. 987a.iioSokoloff , DJBA, p. 968b.^^^See also bota (no. 2) above, and for further lexical and other isoglosses binding

NENA to JBA and Mandaic see H. Mutzafi, 'On the Etymology of the Verbal Roott'i, forthcoming in ZDMG, n. 42.

already recognized by Noldeke, Grammatik der neusyrischen, p. xxxv.


information on NENA dialects, the more we learn about the histori-cal background of the lexical stock of this evanescent group of modernAramaic.

[AS 3.1 (2005) 109-26]DOI 10.117/1477835105053517

THE SYRO-HEXAPLARIC HEADINGS OF THE PSALMSIN MANUSCRIPT 12T3*

Harry F. van Rooy

Potchefstroom Campus, North-West University, South Africa

Introduction

In a paper read at the meeting of the International Organization forSeptuagint and Cognate Studies in Oslo in 1998^ it was pointed outthat one manuscript of the Syro-Hexapla (12t3^) contains many uniquevariant readings in the headings of the Psalms. This manuscript con-tains the text of the Psalms according to the Peshitta, but has variantsfrom the Syro-Hexapla added between lines. In the case of the headingsof the Psalms, it does not contain any of the headings that do occur inmanuscripts of the Peshitta. The Psalms in Peshitta manuscripts donot have headings corresponding to the headings of the Hebrew origi-nal, but either do not have any headings at all, or headings unrelatedto the headings of the Masoretic text.^ In his study of the headingsof the Psalms in the East Syriac Church Bloemendaal distinguishes

*This paper is a revised version of a paper read at the meeting of the Inter-national Organization for Septuagint and Cognate Studies, in Boston, November1999.

^H.F. van Rooy, 'The Headings of the Psalms in Book 1 of the Psalms in theSyro-Hexapla', in B.A. Taylor (ed.), X Congress of the International Organizationfor Septuagint and Cognate Studies: Oslo 1998. (SBLSCS, 51; Atlanta: SBL, 2001),pp. 373-392.

• 12t3 is the number assigned to this manuscript in the edition of the Peshittapublished in Leiden. See Peshitta Institute, List of Old Testament Peshitta Manu-scripts (Preliminary Issue) (Leiden; E.J. Brill, 1961). For other Peshitta manu-scripts the number of this list is used as well, if that manuscript is mentioned inthe list.

•'See W. Bloemendaal. The Headings of the Psalms in the East Syriac Church(Leiden, E.J. Brill, 1961), p. 1.


n o Aramaic Studies 3.1 (2005)

four different groups of headings that do occur in manuscripts or edi-tions of the Psalms in Syriac, viz., the headings of the East Syriacchurch, headings related to those of the Codex Ambrosianus (the WestSyrian tradition), the headings in the editions of Sionita, Lee and thePolyglots, and manuscripts with a mixture of Elastern and Westernheadings.^

In his study of the Eastern tradition, Bloemendaal used eight manu-scripts, as well as three editions containing the East Syrian headings.^Only one of the eight manuscripts (6tl) is not classified by Walter as aNestorian manuscript.^ Ofthe other seven used by Bloemendaal Walterused five (12tl, 13tl-3 and 17t2) in his edition of the Psalms, exclud-ing only 9ml and the Manuscript Mingana 428. In his edition Walterclassified fourteen of the manuscripts that he used as Nestorian manu-scripts.^ Five of them were used by Bloemendaal in his edition of theEast Syrian headings. Ofthe other nine manuscripts, only 12t3 has theheadings of the Syro-Hexapla. The usual East Syrian headings occur in12t4, 13t4, 16t2, 17tl, 17t3 and 18>8dtl. 16t3 does not have headingsat all, while 16t6 has the East Syrian heading for Psalm 1, but eitherno headings (usually) or headings unrelated to the other traditions forthe other Psalms. One would expect the headings of the Eastern tra-dition in 12t3, but in their place this manuscript has the headings ofthe Syro-Hexapla, in full, not between the lines as is the case for thePsalms themselves. In this sense the headings in this manuscript areunique.

It became clear in the previous paper that the relationship be-tween this manuscript and the other manuscripts of the Syro-Hexaplaneeds more study, since this manuscript can not readily be classifiedwith any of the groups of the Syro-Hexaplaric Psalter distinguishedby Hiebert.® The Psalms in the Syro-Hexapla differ from the rest ofthe Syro-Hexapla in the sense that the Psalms do not refiect a truehexaplaric text. The debate about the original Greek Vorlage of theSyro-Hexapla is still not settled. The relationship between the differ-ent manuscripts of the Syro-Hexaplaric Psalter is another matter under

**Bloemendaal, Headings, pp. 2-3.^Bloemendaal, Headings, pp. 12-15.• D.M. Walter, The Book of Psalms (The Old Testament in Syriac, II.3; Leiden:

E.J. Brill, 1980), p. viii.^Walter, Psalms, p. xxvi.8R.J.V. Hiebert, The 'Syrohexaplaric' Psalter (SBLSCS 23; Atlanta: Scholars

Press, 1989).

VAN ROOY Syro-Hexaplaric Headings of the Psalms 111

discussion. Up to now the view of Hiebert is tlie most convincing. Hedistinguishes three groups of manuscripts, to which he gave the namesSyrPs, SyrPs* and SyrPs* . These manuscripts and the groups are dis-cussed in more detail in the next section. He regards SyrPs as a revisionof the Philoxenian Psalter, probably done by Paul of Telia. SyrPs*^ heregards as a different revision of the Philoxenian Psalter, perhaps byThomas of Harkel. SyrPs*^ reflects some of the features of SyrPs*, butis closely related to SyrPs. It could be a light revision of SyrPs basedon SyrPs^.^

A study must be made of all the variants in 12t3 (f according tothe system used by Hiebert), and not only those in the headings, todetermine the relationship between its version of the Syro-Hexapla enthe text in the other manuscripts. The headings must, however, beconsidered on their own, as the headings can in some cases reflect adifferent tradition from that contained in the Psalms themselves. Italso became clear from the previous study of the headings of the first40 Psalms that influences from other directions must also be taken intoconsideration with regard to the headings in 12t3, as in the case of theheading of Psalm 12, ° where an addition may be linked to the East-Syriac heading of Psalm 3. This manuscript also had unique variantsfor the Hebrew TMb in many instances.^^

This paper will present a survey of the variants in the headingsof the Psalms in different manuscripts of the Syro-Hexapla and thenlook in detail at all the headings in manuscript 12t3. ' This manuscriptcontains more than 60 unique variants. One must take in considerationthose cases as well where this manuscript agrees with the AmbrosianSyro-Hexapla when other manuscripts have variants. All the variantswere studied and the readings of 12t3 compared with that of all theother manuscripts. It is impossible to discuss all the variants in detail,but the variants that are important for the relation of 12t3 to the other

Cf. Hiebert, 'SyrohexaplaHc' Psalter, pp. 247-260.°The numbers of the Psalms are those of the Syro-Hexapla.

iiFor example, in Psahns 10 (11), 15 (14), 20 (19), 22 (21) and 33 (32) it omitsthe reference to David altogether. In Psalm 30 (29) it is the only manuscript reading.1.0.1.1 in stead of .r.a.tX. For other examples, cf. Van Rooy, 'The Headings of thePsalms', pp. 384-387.

'•^The following works were used for the different headings:Hebrew: K. EUiger & W. Rudolph, Biblia Hebraica Stuttgartensia (Stuttgart:

Deutsche Bibelgesellschaft, 1984); LXX: A. Rahlfs, Psalmi cum Odis (Vetus Tes-tamentum Graecum, 10; 3rd edn; Gottingen: Vandenhoeck & Ruprecht, 1979);Syro-Hexapla: Hiebert, 'Syrohexaplaric' Psalter.


manuscripts will be dealt with in more detail. In his study Hiebert didnot deal with the headings on their own, and with i:he number of uniquevariants in 12t3.

The Manuscripts Used in this Study

In his study of the Syro-Hexaplaric, Psalter Hiebert used ten manu-scripts of the Syro-Hexapla.' ^ One of his manuscripts (d) does not haveheadings. Of his manuscript k only a few fragments remain, contain-ing parts of Psalms 70, 73, 77 and 79, but unfortunately no headingis part of the text available. Hiebert regarded his manuscripts a-g asrepresentatives of his SyrPs, while h and j contain SyrPs* (h 1.5-27.6;j 1.1-27.6) and SyrPs*^ (from 27.7 onwards).^'' Manuscript k containsonly parts of the Psalter, but these correspond with SyrPs' .' ' ^ Thefollowing is a list of the manuscripts relevant to this paper:

a Milan Ambrosian Library, C. 313 Inf. This manuscript was also usedby Baars in his edition of the Syriac Apocryphal Psalms (9SH1)^^

b British Museum Additional Manuscript 14.434 f. 1-79 (8SH1)

c British Museum Additional Manuscript 14.434 f. 80-128

e Cambridge University Library Oriental 929 (14SH1)

f Baghdad Chaldean Patriarchate 211 (12t3)

g Vatican Library Borghiani siriaci 113 f. 1-135

h Baghdad Chaldean Patriarchate 1112 (15/12SH1)

j Paris, National Library Syr 9

f. Hiebert, 'Sjjrohexaplaric' Psalter, pp. 5-13 for a full discussion of the manu-scripts.

^''Hiebert, 'Syrohexaplaric' Psalter, pp. 247-260, distinguishes three different tra-ditions with regard to the Psalms in Syro-Hexaplaric manuscripts. SyrPs is therevision ascribed to Paul of Telia, while he ascribes SyrPs^ to Thomas of Harkel.gyj.pgb contains features of the previous two revisions.

^^Hiebert, 'Syrohexaplaric' Psalter, pp. 5-13.'^W. Baars, Apocryphal Psalms (The Old Testament in Syriac, IV.6; Leiden:

E.J. Brill, 1972). The notations in brackets following this aad some of the followingmanuscripts are those used by Baars in his edition of the Syriac Apocryphal Psalms.

"Baars, Apocryphal Psalms, p. ix, says that j (=his 13SH1) was edited (he onlytreats Psalm 151) to bring its linguistic usage in accordance with the style of theSyro-Hexapla. The same is not true of Psalm 151 in h, because this Psalm is partof the later addition to this manuscript. This later addition has the normal SyrPstext.

VAN ROOY Syro-Hexaplaric Headings ofthe Psalms 113

Manuscript g is a copy of f* and its variants will not be dealt withseparately in this discussion. Hiebert's Manuscript f is designated 12t3in the notation used for the Leiden Peshitta. In the list of manuscriptsused for the Leiden Peshitta^^ this manuscript is dated to AD 1126.It contains the Psalms in the version of the Peshitta, as well as theOdes and other texts.^° It is an important Nestorian manuscript.^^ Inaddition to the text of the Peshitta it has many important marginalreadings,^^ as well as the text of the Syro-Hexapla in interlinear fashionwhen the Syro-Hexapla differs from the Peshitta.^^

A Survey of Variants in the Different Manuscripts of theSyro-Hexapla

This discussion is directed at the variants in manuscript 12t3 (f ofHiebert). Consequently, the variants in the other manuscripts will notbe discussed extensively, except where they are important for the vari-ants in f (12t3).^'' Two Psalms do not have headings in the Syro-Hexapla,^^ while a number of Psalms do not contain any variants onthe headings.^^ In a previous study^^ I examined the headings of thePsalms in the first Book of the Psalms in the Syro-Hexapla. Thatstudy demonstrated that Hiebert's distinction between SyrPs on theone hand and SyrPs^ and SyrPs*^ on the other was also valid for theheadings in the Psalms in the first book, with the exception of manu-script f, the subject of this paper. The manuscripts of SyrPs (a, b, cand e) contain only a few variants, and none that are very important.His manuscripts h and j do contain a number of important variantsthat clearly distinguish them from SyrPs.

^^Hiebert, 'Syrohexaplaric' Psalter, pp. 9-10.^^Peshitta Institute, List, p. 31.•^"Hiebert, 'Syrohexaplaric' Psalter, p. 9.•^'Walter, Psalms, p. xxvi.^^Walter, Psalms, p. xxvi.^^Hiebert, 'Syrohexaplaric' Psalter, p. 8. This manuscript is described by Baars in

his dissertation as well. Cf. W. Baars, New Syro-Hexaplaric texts: edited, comm,entedupon and compared with the Septuagint (Leiden: E.J. Brill, 1968), pp. 12-13, 16, 29and 35. He did not, however, deal with the text of the Psalter in his work.

^''To facilitate references to Hiebert's work, the letter f will henceforth be usedfor this manuscript.

25psalms 1 and 2.26psalms 5, 22, 24, 25, 27, 42, 48, 57, 63, 65, 68, 69, 72, 77, 78, 80, 85, 89, 92, 94,

96, 97, 98, 100, 122, 123, 124, 125, 127, 129, 137, 138, 139, 141, 143, 146 and 150."See note 1.

When the headings of all the Psalms are taken into considerationthe picture differs to some extent from that in Book 1 of the Psalms.Manuscript b has only four variants in comparison with a ^ and chas only seven variants.^^ This demonstrates again that these threemanuscripts belong to the same tradition as far as the headings areconcerned.

The picture is somewhat different with regard to e. This raanu-script is a polyglot from the fourteenth century, containing the Psalmsand Odes in Greek, Hebrew, Arabic and in the version of the Syro-Hexapla.^° It has the same variants as b in Psalras 53 and 93. It hasthe same variants as c in Psalms 45 and 70. It shares some variantswith other manuscripts as well. In Psalm 45 it shares a variant with c,h and j , with f, j and h in Psalms 82, 119, 144 and 64, with f and h inPsalm 59, with h and j in Psalm 50 and with f in Psalms 38, 64, 71, 99,132 and 142 and with j in Psalm 107. Most of these variants are notreally important, e.g., related to seyame or different pointing (Psalms53, 59, 82, 93 and 119) or the spelling of proper nouns (Psalms 38, 50and 142). Important variants occur in Psalm 132, where e and f omitthe name of David and in Psalm 107, where e and j omit DâX^cn.

In addition to this e contains a large number of unique variants,with only a few in the first Book of the Psalms. Many of these variantsare also not very important:

1. The reading cruoXAcn (against the normal râiXct)) in Psalms 104,105, 106, 110, 111, 112, 113, 114, 115, 116, 117, 118, 134, 135, 145,147 and 148;

2. The spelling of proper nouns: jjr^loa for jjion in Psalms 41, 83,84, 86 and 87, t<^\^jo for i<.ieu» in Psalm 59, yinô.i* for •pahL.x^rfin Psalm 76 and .x.ar<r.-t for X,QI in Psalm 140 (cf. also Psalms 87

for Mc7i» and âxr^ for -ph^] and 88 in this regard).

3. Pointing: Psalms 45, 56, 87 and 120.

In a number of cases the variants are more important. Uniqueplusses occur in Psalms 30, 70 and 102. The plus in Psalm 30 is very

^®In Psalms 9 (agreeing with h), 53 (agreeing with e), 87 and 93 (agreeing withe); only 87 is unique.

^În Psalms 16, 37 (agreeing with f, h and j), 44, 45 (agreeing with e, h and j),47, 53 (agreeing with h) and 70 (agreeing with e); only 16, 44 and 47 are unique.

''"Hiebert, 'Syrohexaplaric' Psalter, p. 10.

interesting. It adds yO vau in Syriac characters at the end of the head-ing. This is not a Syriac word, but rather a Hebrew word (llTSn) inSyriac transhteration. It could be meant as a gloss to indicate whatthe Hebrew original of the Syriac r xcucna ivs) could have been. Thisword is a translation of a plus in the Septuagint. The Hebrew wordtransliterated in this manuscript had the meaning of anxiety or fearin later Hebrew. ^ It was probably not meant as an addition to theheading, but as a gloss to the Syriac. Such glosses appear frequentlybetween the lines of the Syriac.^^

At the end of the heading of Psalm 70 an addition occurs, statingthat the Hebrew does not have a heading. This addition also occurs inthe Lucianic recension of the Septuagint and Theodoret.^^ At the endof the heading of Psalm 102 e adds rîoôa, agreeing with a numberof Lucianic manuscripts. More Lucianic manuscripts have this additionbefore the name David.

In Psalm 62 e reads 'the desert of Judah', not of 'Edom' like allthe other manuscripts. In this instance e agrees with the Hebrew, aswell as with the majority of Lucianic manuscripts, Theodoret, S and55, as well as with the marginal note in a. In Psalm 87 the author ofthe Psalm is not called an Israehte, but rather an Oriental (rt'.ni.T^).

Manuscript e has a variant in the headings of Psalms 110 and 111.The other manuscripts have an addition to the heading of the Sep-tuagint, with a reference to the eTiKJxpocpT] of Haggai and Zechariah.This is rendered by .oAini. io>Ai«yit in the other manuscripts of theSyro-Hexapla. Manuscript e has a reference to the book or account ofHaggai and Zechariah (r<rAvax=iuiio). In Psalm 110 this addition oc-curs only in the Gallicanum and in Psalm 111 in the Gallicanum, R,the Latin of R and G and a small number of Lucianic manuscripts. InPsalm 131 the other Syriac manuscripts have a unique reading (.T.a.t^r» )Ai\o-iY.) while e has a reading (rliJMaso.T rtf'â^x.^) agreeing withthe Greek and Hebrew, as well as with a marginal note in a, f andg ascribing it to the Hebrew. The variant of e can thus be relatedto those marginal readings, but it could also perhaps be an influencefrom the Hebrew column of that manuscript on the Syriac. The situ-ation is similar in the case of another variant in the heading of Psalm

M. Jastrow, Dictionary of the Targumim, Talmud Babli, Yerushalmi andMidrashic Literature (New York: Ktav, 1992), p. 491.

''•^Hiebert, 'Syrohexaplaric' Psalter, p. 279.Rahlfs, Psalmi, p. 197.

59. In the Masoretic Psalm 60 the heading states at the end that Joabstruck Edom in the Valley of Salt. This reference to Edom is omittedby many witnesses to the Septuagint and all the other manuscripts ofthe Syro-Hexapla. Manuscript e has this reference, agreeing with theLatin of R, Augustine, the Vulgate, Theodoret and a number of Lu-cianic manuscripts. From this discussion it is evident that e contains alarge number of unique variants, frequently influenced by the Hebrewor the Septuagint, especially the Lucianic recension. It does, however,still belong to the same tradition as a, b and c, as can be seen by thelack of agreement with the variants in h and j .

The relatively large number of variants common to h and j supportsHiebert's distinction between SyrPs and SyrPs* and SyrPs'^. The largerfrequency of these variants in the first book of the Psalms also sup-ports the distinction between SyrPs* and SyrPs* . It is not necessaryto discuss all the variants in h and j for the purpose of this paper.

The most obvious example is the rendering of the Hebrew termn)i:ab. This term is traditionally understood as 'fiir die Musikmeister'.'^''In the Psalms in the first book of the Psalms it occurs in Psalms 4,5, 6, 8, 9, 10, 11, 12, 13, 17, 18, 19, 20, 21, 30, 35, 38, 39 and 40.The LXX translates this with zlq TO ze'koq in every instance. Thistranslation probably connected the Hebrew word to the noun n:ij, atemporal expression well known in the phrase n: ]' .' ^ The Greek phraseis omitted by fewer than 16 Lucianic manuscripts and 1219 in Psalm30 and by fewer than 16 Lucianic manuscripts and 55 in Psalm 40. TheSyro-Hexapla usually translates this with r&iAaj^ (at the end^^). Inonly two instances it has r so eut , in Psalms 5 and 6. Manuscript hand j have tOiAojt^ in the following instances: Psalms 4, 5, 6, 8, 9,10, 11, 12, 13 (only j , as this Psalm is missing from h), 17, 18, 19, 20and 21. In Psalms 30, 35, 38 39 and 40 they have ri\ajL3, as havemanuscripts a, b, c and e.

Although cases of the independent possessive pronoun (with sufiixadded to A^.T) and the proleptic suffix occur infrequently in the head-ings, h and j differ from the other manuscripts in the few instancesthat do occur.^''

^^L. Koehler & W. Baumgartner, Hebrdisches und Aramdisches Lexikon zumAlten Testament, III (3rd edn; Leiden: E.J. Brill, 1983), p. 676.

^^Koehler & Baumgartner, Lexikon, p. 676.^®See J. Payne Smith, A Compendious Syriac Dictionary (Oxford: Clarendon,

1976), p. 565.f. e.g. Psalms 7, 16, 17 and 51.

differs from SyrPs in the rendering of tic, TO TIXO? andagrees with SyrPs. There are, however, other variants wherediffer from SyrPs, in Psalms 30, 33, 52, 61, 64, 76, 87, 101

and 126, also supporting the distinction Hiebert made between SyrPs*and SyrPs^. Psalm 101 could serve as a good example. SyrPs readsas follows: •ti.rtf'a rd^n -pxa t<ocn -a^^dvso.i rCaj -^^^--V r^^ciii-'Ti^" ^ - h and j have 73.100 .rôcn ftN^nAv»nt rî^ r<*T.-i\ r^^cA-iTiA»<-i N-i t t r ^ rCm'isn. The difference is related to the a before ttrtf inSyrPs and before :D.TO in h and j . In this instance h and j agree withthe LXX against the other Syriac manuscripts.

Manuscripts h and j each have a number of unique variants as well.Many of them are not really important, such as iârî with an r in hin Psalms 6, 15, 34, 51, 52 and 53, and Aar<x with an r in j in Psalms17, 51, 53, 55, 56 and 58. Important variants occur in Psalm 11, whereh, j and f each have a unique variant. These variants will be discussedin the next section.

The following conclusions can be made on account of the precedingdiscussion of the headings of the Psalms in the different manuscripts:

1. Manuscripts a, b and c reffect the same tradition with regard tothe headings.

2. Manuscript e belongs to the same basic tradition as manuscriptsa, b and c, but does contain a relatively large number of variants.These variants very infrequently agree with manuscripts h and j ,and can thus not be related to the other traditions distinguished byHiebert (SyrPs^ and SyrPs*^). As indicated in the discussion above,some of these variants may be due to inffuence from the other threecolumns in this manuscript, especially from the Hebrew column,and from the side of the Lucianic recension of the Septuagint aswell.

3. Manuscripts h and j have a large number of variants in common,supporting Hiebert's distinction between three traditions in the dif-ferent manuscripts of the Syro-Hexapla.

Variants in f

Some of the variants that f shares with other manuscripts have alreadybeen discussed. Those that are important for the discussion of thevariants in f will be dealt with again.

Some of the variants are very important because they demonstratethat f is unique in many respects. With regard to r<»Ao.rii in SyrPsthe variant that appears frequently in h and j (ri cwr ) has alreadybeen noted, r^si^cuc^ also occurs in f in Psalms 9, 10, 18 and 19. Insome other instances f has a related, but somewhat different, variant. Ithas a unique reading, rCuj^ i<i»aAcuc , in Psalm 8. In Psalm 11 it hasa heading different from all the other manuscripts: r<;i>iXcvrX r<rini»D UDf<^oi y.'nhi \\'9i r<lijj-j.T. As far as the rendering of the term ZIQ TOTEXO? is concerned, it has rdLu-j.i ri ajt . The same rendering occursvery frequently in f, in Psalms 12, 13, 20, 21, 30, 35, 38, 39 and 40 inthe first book of Psalms. The noun r.^'--.' is etymologically relatedto the Hebrew noun riK], but has the meaning 'victory'.^^ The Syriacphrase of f means 'till the completion of the victory'.^^ This is a veryinteresting variant in f. It can be compared to some of the other Greekrenderings of this word. Aquila usually has TW viy.onoiS), Symmachushas iniv'fKLoq and Theodotion has zlq vlxoc, or elq TO VLXO?. TheGallicanum follows the LXX (in finem) while the Vulgate has victori.^°One can perhaps say that f shows some Hexaplaric influence in thisinstance. In the other books of the Psalms this variant occurs onlyin the second Book of the Psalms in Psalms 41, 43, 45, 46, 50 and61. This variant does not occur in f in the last three Books of thePsalter.

The spread of this variant is very important for defining the rela-tionship of f to the other manuscripts. As far as the rendering of theterm elq TO TCAO? is concerned, it shows an affinity with h and j intheir parts belonging to SyrPs*, but continues with the same variantwhen h and j switched to SyrPs'^ in the first book of the Psalms andin six instances in the second book of the Psalms. It also demonstratesknowledge of the Hebrew term translated by the LXX as etc, TO TEXO?,

and differently in some of the other Greek versions.It is impossible to discuss all the variants in f in detail in this paper.

A selection will be treated under a number of groups of variants thatoccur in this manuscript. Some of these groups are important for therelation of f to the other manuscripts and some are; not that important.

word appears in the East Syriac heading of Psalm 47 (=46 of the Syro-Hexapla), and also in the West Syrian headings of Psalm 44 (43) and 54 (53).

'^ J. Payne Smith, A Compendious Syriac Dictionary, p. 348.Origenes, Hexaplorum quae supersunt PG XVI (1) cols. 583-586 and 635-

638.

Omissions in fA number of omissions occur in f. In quite a number of instances thereference to David as author of the Psalm is omitted, in Psalms 19,95, 130 and 132. In Psalm 19 it agrees with h and j , in Psalm 130with a large number of witnesses to the Septuagint' ' and in Psalm 132with e and a number of witnesses of the Septuagint.''^ In the last twoinstances it follows inter alia the so-called Lucianic tradition. In Psalm95 the omission is unique to f. In this instance the Masoretic text has noheading at all. In the case of two Psalms it omits the heading r<:.a\^cn(106 and 115). In another three cases it omits rîeisoiio (Psalms 41,43 and 70). In Psalm 41 it agrees with the Masoretic text and manywitnesses of the Septuagint. In Psalm 70 the Masoretic text has noheading and the omission in f agrees with a number of witnesses to theSeptuagint, excluding the Lucianic tradition. In Psalm 43 it agrees withthe Masoretic text against the Septuagint. The omission of the nameof the prophet Nathan in Psalm 50 is unique. The heading of Psalm 50is indeed very short in f: t< r<r .TA .iâ.t^ rîasjvsa riiu-j.i tOaAojAj_aa.^trj S\j^ >*'--' rod\oA. It can be regarded as a summary of theusual heading.

With regard to these omissions, no fixed pattern of agreement witha certain group of witnesses can be determined.

Variants related to word order in fVariants related to two consecutive words or phrases changing positionoccur in Psalms 14, 35, 39 and 64. This is unique in the case of Psalms14 and 35. It agrees with a number of witnesses to the Septuagint inPsalm 39 and with one manuscript of the Septuagint (S) in Psalm 64.

Variants related to suffixes and the construction of the genitive in fIn a number of instances f has jjievn o ^ for jj'tAn.i t<Jis (e.g. Psalms41, 43, 47), but in the majority of instances it agrees with the othermanuscripts. In the case of Psalms 9 and 33 the proleptic suffix is addedto a noun before i.

Variants related to spelling, pointing and synonyms in fIn a number of instances f has variants in the spelling of personalnames, frequently related to the addition of vowel letters, like an t<in the name .T.OI^.I (cf. Psalms 3, 38, 51, 59 and 142). Psalm 7 has a

, Sa-2017, Ga, L ' , 1219'; cf. Rahlfs, Psalmi, p. 312., Sa-2017, L', 55; cf. Rahlfs, Psalmi, p. 314.

very interesting variant. The name Cush (OID in Hebrew) is renderedby XoucjL in the Septuagint. The other manuscripts have >x.au, but fhas >z.cL , agreeing with the Hebrew. Variants related to pointing (e.g.words with or without seyame) occur in Psalms 87, 93, 107, 119 and133. In some instances f has variants that are no more than the use ofsynonyms for the reading in the other manuscripts (Psalm 37, 45, 59and 66). There are also examples where f omitted the hexaplaric signslike an asterisk (e.g. Psalms 109 and 110).

Variants related to particles in fQuite a number of examples occur where f contains variants relatedto .1 and prepositions. In Psalm 3 it reads jit^. .-L for .ot^.T .dvsortf.In Psalms 8 and 9 it reads .\\^'w for .^Xu. It has .T.a.TX for .T.O.TI inPsalms 26, 109 and 121 and .x.o.i.i for x.a^\ in Psalms 29 and 90. Ithas prepositions in the place of .i in Psalms 29, 49 and 53, and .i for apreposition in Psalms 71, 74 and 93. It omits .i in Psalms 28, 118 and144, and adds .i in Psalm 131.

Important variants in fSome of the variants discussed above are very important for the eval-uation of the headings in f, in relation to the headings in the othermanuscripts, pointing to the uniqueness of f in many ways. Some ofthe variants are particularly important on account of their complexity.In this section ten of them will be discussed in more detail.

Manuscript f has a heading to Psalm 6 that differs from that foundin the other manuscripts in a number of respects. The other ones havethe following heading: r<ria»v») t<h^Q...i..^nh\

. f has a related but different heading:^^a. 1."a^ .°>\ii ri'^'tj.i*)! '*'-»-*»'Ai risocu. This unique vari-

ant is probably an explanatory expansion of the heading in the othermanuscripts.

At the end of the heading of Psalm 9 f has an important addition.The heading in the other manuscripts is (with a few minor variants):

To this f adds: miho^n A..i.o .i<\si cD^cu^a.A^ -^V^- This isa messianic addition from a typical Christian perspective (About theyouth of the son and his death). The heading in the Masoretic; texthas a number of text-critical problems, related especially to the wordniD'p J. If the text is read as two words mo bsi, it would indeed bepossible to hnk the addition in f with the Masoretic heading. It can.

however, be read as one word, as proposed by the critical note la inthe BHS (rriQ'?JJ''7i!). This reading could be the basis for the translationof the Septuagint (secrets, xpucpibiv), followed by the Syro-Hexapla. Itis interesting to note that the complicated Syriac word mA>rnAt>w.\ vthat is not listed by Payne-Smith, sounds somewhat like the Hebrewvariant mentioned above. The margin of a has a note that the headingof this Psalm by Aquila has a reading t<t=i.T ri' nwLA^.T. This makes itpossible that the reading in f could somehow be related to the headingin Aquila that could in its turn be related to a Hebrew Vorlage similarto the proposal of the BHS.

In Psalm 12 the LXX has no variants. Manuscript f has a longaddition at the end of the heading: 1 ^ .ai.iiisr) Tin\».-,rf' ^ x^r<riu.^. This addition can be compared to the East Syrian headingfor Psalm 3 in the Peshitta''^ for the reference to David's persecutionby Absalom and to the East Syrian heading of Psalm 13 (12 in theSeptuagint and Syro-Hexapla) for the reference to David's sin.'''' Theheading of f is therefore a combination of headings from the Syro-Hexapla and the East Syrian tradition.

In Psalm 11 f, h and j each has its own variant. The other manu-scripts have .i a.tX rtfievsovso '*'-'-^^.k .o,\,, r<toi\ojLa.The followingvariants appear in f, h and j :

f:

h:

Manuscripts f and h omit the name of David. The remainder of thevariant readings are all unique and the origin of the variants is un-known.

In Psalm 31 the other manuscripts have nf^aj^ft^jr.i .^.a.o^.Manuscript f inserts a synonym for this word and then explains theinsertion by adding the word that appears in the other manuscripts:

In Psalm 51 the other manuscripts follow the Septuagint by read-ing ii^\^r,.-,r^ Manuscript f agrees with the Masoretic text and withmarginal readings in a, b, c and j by reading o\.\-yi.»ir<r.

In Psalm 53 f has a heading deviating from that of the other manu-scripts. They have, with some minor variations:

''•'See Bloemendaal, Headings, p. 35.''''See Bloemendaal, Headings, p. 39.

am aS Aar^z^ OtSOrâ rCS^X aA\r<? . I ^ .lUThe heading in f is:

. The contents are basically the same, but the constructions differ.The heading of f is unique.

With regard to Psalm 55 Hieberf*^ says that the margin of f con-tains the same heading as in the other manuscripts. It is ascribed toAquila. The text has a heading taken from Aquila and Symmachus.The heading is as follows:

,°>\ll r^^Cl^l.i d \ ^ ^ rt^.Al v\<» ,IT<rt-\-,\ x^ rt'-V'^ v**! t<2ui.^^ .l> O.I.I

CICD .taO.! rtfocn p-"T oA^.I

This heading is indeed closely related to the headings ascribed toAquila and Symmachus in the Ambrosian Syro-Hexapla. The headingof Aquila is as follows:

rt*.AiV\<\ ,m,r»-\-i\ xa •<>.\«Tnr>Ti

The main difference is the omission of the reference to Gath in thesection of the heading of f agreeing with the heading of Aquila. Theheading of Symmachus is as follows:

ocn.,aj O.Xur^.t ,dvsot<

In the latter part of this heading in f a section was omitted and replacedwith a few words from the heading of Aquila.

In Psalm 103 f inserts rîoso vio before the remark that the Hebrewdoes not have a heading. This insertion is related to the reading of someLucianic manuscripts. Some witnesses to the Septuagint add 'a Psalm'before or after 'Of David'. The heading in the Syro-Hexapla agreeswith the addition in some Lucianic manuscripts.

In Psalm 108 f has a longish addition agreeing with the readingascribed to Aquila in the margin of a. The margin of a states thatAquila has the following heading for this Psalm: .'L.O.-T.I r^rôAt t- ^Nrîoio vi>D, agreeing exactly with the addition in f. The complete head-ing in f is

''^Hiebert, 'Syrohexaplaric' Psalter, p. 77.® waw has a supralinear stroke denoting an abbreviation for


This heading is a combination of the normal heading of the Syro-Hexapla and the heading of Aquila.

With regard to the headings in f, tlie following conclusions can bedrawn:

1. As far as the rendering of eL? xo xzkoq in the Septuagint is con-cerned, f shows some affinity with the reading in h and j , but con-tinues with this variant longer than they do. It demonstrates someknowledge of the Hebrew term used in the Vorlage of the Septuagintas well.

2. Omissions in f are mostly related to the omission of the name ofDavid or the word for 'A Psalm' No pattern of agreement with acertain tradition or group of witnesses can be determined.

3. In the case of some of the variants, f has unique readings, with theorigin of the variant unclear, such as in Psalms 6 and 53.

4. Some of the headings in f show affinity with some of the otherversions. Psalm 9 has an addition from a typical Christian perspec-tive that may be related to the heading of Aquila in some way.The heading of Psalm 12 is a combination of the heading of theSyro-Hexapla and a heading from the East Syrian tradition of thePeshitta. The heading of Psalm 55 is related to the headings ofSymmaehus and Aquila. In Psalm 108 the heading of f is a combi-nation of the normal heading of the Syro-Hexapla and the headingof Aquila.

The Spread of Variants in the Different Manuscripts

In the following table all the variants appearing in all the manuscriptsare given in order to show their appearance in the different traditions,especially in relation to manuscript f. Manuscript e does not have allthe headings, but usually agrees with manuscripts a, b, c and d. WhereHiebert has only one variant in a specific heading, the variant is referredto by the number of the Psalm. Where more than one variant occur,the different variants are distinguished by a, b, c etc. added to thenumber of the Psalm and in the order presented by Hiebert. Wherethe variant has suf added, it means that the variant is related to asuffix and not listed in Hiebert's apparatus.''^

'"^Hiebert, 'Syrohexaplaric' Psalter, p. 15-19.


f unique

P"S unique; anothervariant hjP"s unique; anothervariant efabcd (e where ex-tant); hj together aunique variantfabcdhj (e variant)fabcd; e a differentvariant; hj still an-other variantfabdhjd uniquefabcdh

fabcdj

fabcd (ehj differ)fabd (cehj differ)facdhj (be differ)fa*fe

flij

fehj

fehfchjfhfjfmg j^mg l^mg

f one variant, hj an-other variant

3c 6a 7b 8b 9b 9c 9e 11 12b 14b 14c 19b 21b 21c23b 26 29a 29b 30a 31b 32 33a 35 37a 38a 39a 39b40 41a 43 45a 45b 46 47b 49 50a 50b 50c 51c 5 Id51e(=a™s b^s c- s j™s) 53b 54a 55a 59a 61b 64b 6670a 74 89 90 93b 95b 96b 103a 103b 108 (= Aquilain a™g f"g) 109 118b 121 130 133 136 (corr in mg)52a

137a

3b 4 7a 7 suf 17a 17b 17c 17sufa 17sufb 20b 35 suf61c 86suf 126 142 suf

41b30b

16 31a 44 47a 70b (e agrees v/ith c)60 61a 62a 102a14a 23a 51b 53a 53c 55b 56b 58a 67 73 81 83a (jhas a correction) 87b (j has a correction) 128 (j hasa correction) 145b (j has a correction)6b 15 17d 34 52b 54c 58b 64a 95a 142 (h has acorrection) 14950c45c93a3638b (e* "*; e* has another variant) 71 (e s has an-other variant) 99b 132 142a3d 7c 9a 10a 18 19a 33b 110b (e has unique read-ing)64c (hj without .i) 82 (h has, i without dot) 11914459e37b8c 10b59b3a8a 12a 20a 21a 87f 131b


f one variant, j an-other variantf one variant, h an-other variantf one variant be, an-other variant, ch an-other variantf one variant, e an-other variantf one variant, e an-other variant, j an-other variantfacdje unique

e' s uniqueeh one variant, j an-otherejg uniquee'"s unique

13 28

51a

53d

87d 106 115 148a

59b

9d56a 59c 62b 70c 75 (e* correction by first hand) 76(correct reading in e™s, another variant in hj) 83b84 86 87a (e has a correction) 87c (hj have anothervariant) 87e (a™g b">g another variant) 88 96a 101(hj another variant) 102b 104 105 107b (correctionin margin) 110b (cf note at fhj) I l i a 111b (efhjsine notis) 112 113 114 116 117 118a 134 135 145a147 148b4* 131a (adsignatur ebr in a^g P s g™g)137b 140

91 110a96a

107a79 (with correction in g)54b 99a (correction)

Conclusions

With regard to Hiebert's SyrPs, with a as basic text, very few variantsappear in manuscripts b and c. These three manuscripts clearly reflectthe same tradition, also with regard to the headings. Manuscript b hasonly four variants and c has only seven. These two manuscripts do notshare any variants, confirming that these three manuscripts representthe same tradition. Manuscripts h and j represent a different tradition

112 113 114 116 117 118a 134 135 145a 147 have the variant cruoXiro for


than that of a, b and c with regard to the headings as well, as can beseen from the large number of shared variants.

Manuscript e goes its own way in many respects. It is closer to thetradition of a, b and c, but has a large number of unique variants.Many of these variants could perhaps be related to the fact that thismanuscript is a polyglot, with influence on the Syro-Hexapla columnfrom the side of the Hebrew and Greek columns. It frequently showsevidence of Lucianic influence. It does, however, still belong to thesame broad tradition as a, b and c.

The variants in f were treated in more detail in this paper. It hasquite a number of unique variants. Striking examples are the shortenedform of the heading of Psalm 50, the expanded heading of Psalm 6,the messianic addition at the end of the heading of Psalm 11, the EastSyrian addition to the heading of Psalm 12 and the headings influencedby Aquila and Symmachus in Psalms 55 and 108.

The same trends as in the first book of the Psalms can be distin-guished. Some of the variants show hexaplaric influence, others agreewith Aquila and Symmachus and influence from other Syriac tradi-tions of the headings to the Psalms can also be found. This manu-script is unique in the sense that it is a Peshitta manuscript with theSyro-Hexaplaric headings. The headings frequently demonstrate an in-dependent tradition as well. This study dealt only with headings of thePsalms. It would be interesting to see whether the interlinear glossesin the body of the Psalter in this manuscript have the same trends,but that must be the subject of a next study.

[AS 3.1 (2005) 127-31]10.117/1477835105053518

BIBLIOGRAPHY OF THE ARAMAIC BIBLE

Compiled by the staff of the Peshitta Institute, Leiden, and of the Targum Institute,Kampen. Please send relevant bibliographical material to [email protected],and if possible also an offprint to the Semitic Institute, Kampen Theological Uni-versity, POB 5021, 8260 GA Kampen, The Netherlands.

ADNA, Jostein, 'The Servant of Isaiah 53 as Triumphant and Inter-ceding Messiah: The Reception of Isaiah 52:13-53:12 in the Targumof Isaiah With Special Attention to the Concept of the Messiah', inBernd Janowski et al. (eds.). The Suffering Servant: Isaiah 53 in Jewishand Christian Sources (Grand Rapids: Eerdmans, 2004), pp. 189-224.[revised version of the author's 'Der Gottesknecht als triumphieren-der und interzessorischer Messias; die Rezeption von Jes 53 im Tar-gum Jonathan untersucht mit besonderer Beriicksichtigung des Messi-asbildes', in: Bernd Janowski and Peter Stuhlmacher (eds.), Der lei-dende Gottesknecht: Jesaja 53 und seine Wirkungsgeschichte (Tiibin-gen: J.G.B. Mohr (Paul Siebeck), 1996), pp. 129-158.]

Keywords: MessiahTexts: Isaiah 52-53

BERNSTEIN, Moshe J., 'A Jewish Reading of Psalms; Some Observa-tions on the Method of the Aramaic Targum', in: Peter W. Flint andPatrick D. Miller, Jr. (eds.). The Book of Psalms: Composition andReception (Leiden: Brill, 2005), pp. 476-504.

Keywords: Targum Psalms

BROCK, Sebastian P., 'A Neglected Revision of the Peshitta Psalter',in: Carmel McCarthy & John F. Healey (eds.). Biblical and Near East-ern Essays: Studies in Honour of Kevin J. Catheart (London: T & TClark International, 2004), pp. 131-142.

Keywords: PeshittaTexts: Psalms



COOK, Johann, 'Exegesis in the Septuagint', JNSL 30 (2004), pp. 1-19.Keywords: Peshitta

DEROUCHIE, Jason S., 'Circumcision in the Hebrew Bible and Tar-gums: Theology, Rhetoric, and the Handhng of Metaphor', Bulletinfor Biblical Research 14 (2004), pp. 175-203.

Keywords: circumcision

DYK, Janet W., 'Linguistic Aspects of the Peshitta version of 2 Kings18 and 19', in: Johann Cook (ed.), Bible and Computer: The Stel-lenbosch AIBI-6 Conference: Proceedings of the Association Interna-tionale Bible et Informatique 'From Alpha to Byte', University of Stel-lenbosch 17-21 July, 2000 (Leiden: Brill, 2002), pp. 519-543.

Keywords: PeshittaTexts: 2 Kings 18-19

ELLIOTT, J.K., review of Barbara Aland and Andreas Juckel (eds.).Das Neue Testament in syrischer (jberlieferung, IL Die paulinischenBriefe, Teil 3:l./2. Thessalonicherbrief 1./2. Tirnotheusbrief Titus-brief Philemonbrief und Hebrderbrief (ANTF, 32; Berhn: Walter deGruyter, 2002), NT 46 (2004), pp. 78-79.

EVANS, Craig A., 'Targumizing Tendencies in Matthean Redaction', in:Alan J. Avery-Peck et al. (eds). When Judaism and Christianity Began.Essays in Memory of Anthony J. Saldarini Volume I: Christianity inthe Beginning (Supplements to the Journal for the Study of Judaism85,1; Leiden: Brill, 2004), pp. 93-116.

Keywords: Targum and New Testament

FOLMER, Margaretha, review of Per A. Bengtsson, Passover in TargumPseudo-Jonathan Cenesis: The Connection of Early Biblical Eventswith Passover in Targum Pseudo-Jonathan in a Synagogue Setting(Stockholm: Almqvist & Wiksell, 2001), Aramaic Studies 2 (2004),pp. 263-265.

Keywords: Pseudo-Jonathan

FRANK, Yitzchak, Crammar for Gemara & Targum Onkelos (expandededition) (Jerusalem: Feldheim Publishers / Ariel Institute, 2003) [iSBN1-58330-606-4].

Keywords: Grammar, Targum Onkelos

GARCJ'A MARTINEZ, Florentino, 'Sammael in Pseudo-Jonathan andthe Origin of Evil', JNSL 30 (2004), pp. 19-41.

Keywords: Pseudo-Jonathan

GARCIA MARTINEZ, Florentino, 'Sodom and Gomorrah in the Targu-mim', in: Ed Noort and Eibert Tigchelaar (eds.), Sodom's Sin. Gene-

BIBLIOGRAPHY 129

sis 18-19 and its Interpretations (Themes in Bibhcal Narrative. Jewishand Christian traditions, 7; Leiden: Brill, 2004), pp. 83-96.

Keywords: SodomTexts: Genesis 18-19

GELSTON, A., review of Peter J. Williams, Studies in the Syntax ofthe Peshitta of I Kings (MPIL, 12; Leiden: E.J. Brill, 2001), VT 54(2004), p. 423.

Keywords: PeshittaTexts: 1 Kings

GREENBERG, Gillian, 'Indications of the Faith of the Translator in thePeshitta to the "Servant Songs" of Deutero-Isaiah', Aramaic Studies 2(2004), pp. 175-192.

Keywords: Peshitta, MessiahTexts: Deutero-Isaiah

HAELEWYCK, J.C., review of Johann E. Erbes, The Peshitta and theVersions. A Study of the Peshitta Variants in Joshua 1-5 in Relationto Their Equivalents in the Ancient Versions (Studia Semitica Up-saUensia, 16; Uppsala University: Uppsala, 1999), LM 117 (2004), pp.237-238.

HouTMAN, Alberdina, 'Different Kinds of Tradition in Targum Jona-than to Isaiah', in: P. van Reenen et al. (eds.). Studies in Stemmatology,II (Amsterdam/Philadelphia: Benjamins, 2004), pp. 267-281.

Keywords: stemmatology, Targum Isaiah

JENNER, Konrad D., 'CALAP and its Relevance for the Translationand Interpretation of the Syriac Bible: The Presentation of a Re-search Programme on the Computer Assisted Linguistic Analysis of thePeshitta', in: Johann Cook (ed.), Bible and Computer: The StellenboschAIBI-6 Conference: Proceedings of the Association Internationale Bibleet Informatique 'From Alpha to Byte', University of Stellenbosch 17-21July, 2000 (Leiden: Brill, 2002), pp. 681-699.

Keywords: Peshitta, computing

KEULEN, Percy S.F. van, 'A Case of Ancient Exegesis: The Storyof Solomon's Adversaries (1 Kgs. 11:14-25) in Septuaginta, Peshitta,and Josephus', in: Johann Cook (ed.), Bible and Computer: The Stel-lenbosch AIBI-6 Conference: Proceedings of the Association Interna-tionale Bible et Informatique 'From Alpha to Byte', University of Stel-lenbosch 17-21 July, 2000 (Leiden: Brill, 2002), pp. 555-571.

Keywords: Biblical figures: Solomon


KiRAZ, George. Comparative Edition of the Syriac Gospels, 4 vols.(Piscataway: Gorgias Press, 2004) [iSBN 1-59333-160-6].

Keywords: Syriac NT

KOSTER, Marinus D.,' "Translation or Transmission? That Is the Ques-tion": The Use of the Leiden O.T. Peshitta Edition', in: Matthias Au-gustin and Hermann Michael Niemann (eds.), 'Basel und BiheV: Col-lected Communications to the XVIIth Congress of the InternationalOrganization for the Study of the Old Testament, Basel 2001 (Frank-furt a.M.: Peter Lang, 2004), pp. 297-312.

Keywords: Peshitta

LEVENE, Dan, and Benno Rothenberg, 'Word-Smithing: Some Metal-lurgical Terms in Hebrew and Aramaic', Aramaic Studies 2 (2004), pp.193-206.

Keywords: Lexicography

LUND, J., The Old Syriac Cospel of the Distinct Evangelists, A Key-Word-In-Context Concordance (Piscataway: Gorgias Press, 2004) [iSBN1-59333-069-3].

Keywords: Syriac NT, concordance

OLSSON, Birger, 'The Ganticle of the Heavenly Host (Luke 2.14) inHistory and Gulture', NTS 50 (2004), pp. 147-166.

Keywords: Peshitta, SinaiticusTexts: Luke 2.14

OWENS, Robert J., review of Gilhan Greenberg, 'Translation Techniquein the Peshitta to Jeremiah (MPIL, 13, Leiden: E.J.Brill, 2002), CBQ66 (2004), pp. 123-125.

Keywords: PeshittaTexts: Jeremiah

OWENS, Robert J., review of Peter J. Williams, Studies in the Syntaxof the Peshitta of I Kings (MPIL, 12; Leiden: E.J. Brill, 2001), BiOr61 (2004), pp. 393-395.

Keywords: PeshittaTexts: 1 Kings

PEURSEN, Wido van, 'Morphosyntactic and Syntactic Issues in theSyriac Text of 1 Kings 1', in: Johann Gook (ed.), Bible and Computer:The Stellenbosch AIBI-6 Conference: Proceedings of the AssociationInternationale Bible et Informatique 'From Alpha to Byte', Universityof Stellenbosch 17-21 July, 2000 (Leiden: Brill, 2002), pp. 99-111.

Keywords: PeshittaTexts: 1 Kings 1

BIBLIOGRAPHY 131

PEURSEN, Wido van, 'The Peshitta of Ben Sira: Jewish and/or Chris-tian?', Aramaic Studies 2 (2004), pp. 243-262.

Keywords: Ben Sira, Peshitta

RIBERA FLORIT, Josep, 'Los targumes de Job; algunos aspectos tex-tuales y literarios', Estudios Biblicos 62,1 (2004) 77-86.

Keywords: Targum Job, Qumran Job Targum

ROOY, Harry F. van, 'Towards a Critical Edition of the Headings ofthe Psalms in the Different Syriac Traditions', in: Johann Cook (ed.),Bible and Computer: The Stellenbosch AIBI-6 Conference: Proceedingsofthe Association Internationale Bible et Informatique 'Prom Alpha toByte', University of Stellenbosch 17-21 July, 2000 (Leiden: Brill, 2002),pp. 545-554.

Keywords: Peshitta, headings of the Psalms

SMELIK, Willem F., 'How to Grow a Tree; Computerised Stemmato-logy and Variant Selection in Targum Studies', in: Johann Cook (ed.),Bible and Computer: The Stellenbosch AIBI-6 Conference: Proceedingsofthe Association Internationale Bible et Informatique 'Prom Alpha toByte', University of Stellenbosch 17-21 July, 2000 (Leiden: Brill, 2002),pp. 495-518.

Keywords: Stemmatology

STEC, David M., The Targum of Psalms (The Aramaic Bible: TheTargums, 16; Collegeville: Liturgical Press, 2004) [iSBN: 0-8146-5491-6].

Keywords: Targum Psalms, translation

SYSLING, Harry, 'Three Harsh Prophets. A Targumic Tosefta to Para-shat Korah', Aramaic Studies 2 (2004), pp. 223-242.

Keywords: Tosefta-TargumTexts: Numbers 16

TILLY, Michael, review of Matthias Henze, The Syriac Apocalypse ofDaniel: Introduction, Text, and Commentary (Tubingen: Molir Sie-beck, 2001), TLZ 129 (2004), pp. 771-773.

Keywords: PeshittaTexts: Daniel Apocalypse

[AS 3,2 (2005) 137-45]DOl 10.1177/1477835105059088

PARABLEPSIS IN THE CHRISTIAN PALESTINIANARAMA[C LECTIONARY:

T H E CASE OF MATTHEW 17.26

Tjitzc Baarda

Free University, Amsterdam

1. Introduction

The editors of the Talestinian-Syriac Lectionary of the Gospels' (usu-ally denoted as SyP ')^ detected many instances in which the threemanuscripts showed omissions due to homoioteleuton. In her 'List ofVariants in the Three Codices'.^ Agnes Smith Lewis noted for all threemanuscripts eleven instances of parablepsis. As for the separate codicesshe hsted more eases of this kind of parablepsis: for Codex A 25, forCodex B 24. and for Codex C no less than 64 cases.' Last year whenI was studying the textual problems of Mt. 17. 24-27 I found there'*another case of parablepsis in verse 26 which, as I will try to show, mayhe of special interest for textual critics of the New Testament.

In our critical New Testament editions SyP ^ is mentioned as apossible witness to the following Greek readings in verse 2G:'

1. ELTiovTo Se- - SO Tischendorf''^'^ ' ", The Greek New Testament.^

A.S. Lewis and M.D, Gibson, Thf. Palestinian Syriac Lectionary of the Gospels{London: Ke^an Paul. Trench, Trilbner, 1899).

^Lewis &: Gibson, Lectionary. pp, xix-lxiii.•'Lewis Hi Gibson, Lectionary, pp. xix-xxiii.•*The text ia preserved in Lesson LXXI {Lectionary. p. 82) which comprises Mt.

17.24-18.4.•"'Remarkably enough, Lewis did not pay attention to the textual form of the

Lectionary text of Mt. 17.26 in her list of variants.^A.F.C. Tischendorf, No'vum TestamentuTii Graece, I (Leipzig: Winter, 7th edn,

1859), p. 83 (syrhr); B. Aland, K. Aland. .1. Karavidopoulos, CM. Martini, and B,Metzger, The Greek New Testament (Stuttgart: Deutsche Bibelgesellschaft, United

© 2005 SAGE Publications (London, Thousand Oaks CA, and New Dellii)

138 Aramaic Studies 3.2 {2005;

2. Xiyti auTw- so Tregelles, Tischendorf ^ ^ *'' ", Horner.'

Which of these two assigiiiiieiits is correct? I had to answer this questionwhen I prepared a seminar paper' on the rather complicated 'knot' inthe textnal tradition of this verse. My conchision was, and is, thatneither of these readings was correct. In this short note I will make anattempt to give my Une of reasoning.

2. Scribal Parableps'is in Mt. 17.25-26

One of the dangers for scribes is parablepsis^ especially in cases of ho-moioteleuton. This is certainly true for the verses 25-26. which presentthe following discussion between Jesus and Peter:**

1. . . . Tcpoecp&aaEv auxov 6 lY]aoug2. TL. aoL S o x c i , Stp.wv: o l

3. aTTO TLVwv X(X[j.[iiavouCTLv

4. oiKO Twv ulwv auTwv r, ano TWV5. eiTiovTO^ §£• aTTO Twv aXXoxpLuv,

6. IcpT] auTu 6 "iTjaoui;' apa yt zkeu^epoi eloLv oi uloL

It is obvious that the eye of a copyist could easily wander from line 4to line 6. And, in fact, in a few manuscripts Hne 5 is omitted. Tregellesand Swanson^" mentioned the famous cursive 33; Von Soden^^ and

Bible Sorirties. 4th o<in. 1993), p. 66: SyrP' '; K. Aland, M. Black, C.M. Martini,B. Metzger, and A. Wikgren, The Greek New Testament (Stuttgart: United BibleSocieties, 3rd edn. 1975), p. 67: SyrP' ' with the caveat "vid {— videtur)'.

''S.P. Tregelles, The Greek New Testament. I (London: Bagster-Stewart, 1857).p. 64 (Syr.Hier.); C. Tischendorf, Noimm Te.stam.entum. Graece. I (Leipzig: Giesecke& Devrient, 8th edn, 1869), p. 104 (syr*^"); G. Horner, The Goptic Version of theNew Testament in the Southern Dialect, I (Oxford: Clarendon Press, 1911), p. 185app. (Syr(j)).

'^'Mat.thew 17:26 and the Problem of Parablepsis'. 58tl» General Meeting,Rlieinischp Friedrifh-WilhelniH-Universital. Bonn, July 29-Augiist 2. 2003 (Semi-nar 'Textual Criticism", Session of July 31). I hope to publish this communicationin the near future.

^This is the text generally accepted by the editors. For the division in cTiyoi Itook as an example l.he sense division of Codex Bezae.

^"R. Swanson (ed.), New TctitameM Greek Manuscripts, 1. Matthew (Sheffield:Sheffield Academic Press. 1995), p. 168-69.

"H. von Soden, Die Schrijten des Neuen Testaments. IL Text mit Apparat(Gottingen: Vandenhoeck und Rnprec:ht. 1913), p. 61.

BAARDA Parablepsis in the CPA Lectionary 139

o ]\jg 1241; Scrivener described Ms. 59 as having thesame phenomenon. "* Finally, Tregelles noted also the uncial X*, andit is true that its scribe also omitted line 5, but in this manuscript theerror was restored in the margin of the manuscript, either by the scribehimself or by a later corrector.

If the data given here are correct,^^ there are at leaat fouj- Greekmanuscripts which have an omission of this phrase due to parablepsis:Mss. X*, 33, 59, and 1241. an illustration of how easily the wording ofthis passage could suffer a homoioteleuton error. In this short study Iwill show that the 'Palestinian-Syriac Lectionary' presents a text whichresulted from a parablepsis fault.

3. Mt 17.25 in the 'Palestinian-Syriac Lectionary'

The relation between the three manuscripts (A, B, and C)' ''' is unclear,so that it is not possible to establish the original text of SyP ' on thebasis of a stemma. In many instances B and C differ from A, notonly in variant readings but also in errors. An interesting example isfound in verse 25a, where B aud C erroneously read ^"TJoK. 'saying'(plural), instead of the correct rendering found in Ms. A, i_K "lioK,that is, Xeyef vaL However, there are other cases in which A and Bagree against C, or A and C against B, or where all three texts differ.This suggests that there was a long tradition history of the so-called'Syro-Palestinian' text before it took form in these three manuscripts.In the following survey we will make an attempt at reconstructing the"original' text of the pertinent passage, in which I follow again thedivision in lines as presented above for the Greek text.

'^S.C.E. Legg, NouuTU Testamentum Graece, E'uangeliurn secundum Matthaeum(Oxford: Clarendon Press, 1940), in loco.

'•^F.H.A. Scrivener. Adversaria Grjtiea Sacra (Cambridge: Cambridg!^ UniversityPress. 1893). p. 69: '- XifZL usque ad aXXoTpLuv (6|i,0!.0T,) c' [N.B, c = Evan. 59.Conville and Caius College (No. 403). cf, ibid, p, xviii; in Von Soden's numbering:Ms. £ 372 (of the K^ Type).

'• I have assumed that the data given by these editors are correct, but theoret-ically one should have to check the manuscripts. For example, the third editionof The Greek New Testament {GNT) notes as fact that tlie Syro-Sinaitic Cospel{"omit Sy**') has omittetl the phra.se in qtiestion. However, all editions of this Syriacmanuscript tell us that this is not the case.

^^A = Vat. Syr. XIX (1030 CE); B and C are manuscripts of the Moria-stery ofSt. Catherine dated 1104 and 1118 respectively.

140 Aramaic Studies 3.2 (2005;

Tin o n . 1

K.4.AAJD m n - i ^ - uyA Kxm KJD .2

r<j< r<mi -n A nmi Ti m .3

, ^ m i m ^ .4

1. .. . (a) anticipated (b) the Lord Jesus (c) and said:2. 'What is (d) pleasing you, (e) Peter? The kings of the earth,3. from whom (are they) taking (f) tax or (g) census?4. From their sons or from (li) foreigners?'.

It may be useful to comment on this text and its variant readings:

Line 1(a)

1. Mss. B-C om. auTov: the omission of the suffix m- ('him' — auxov) isunique, but there is no reason to suppose that the pronoun was lackingin the Greek Vorlage.2. The reading Tinnn 'anticipated' in Ms. C (Pa'el Perf.) seems thepreferable reading; Ms. B (aLktu, Pe'al ptc. pass.) is a less likely read-ing,'^ because then one .should expect an immediately following TJLK

(not "UIKD) after the subject.3. The reading nuwui. 'anticipated him' (in Ms. A) is secondary, mostlikely borrowed from a Syriac text (cf. Sy"-* ''P' ).

(b) 'IvjaoGi;] ''Lord Jesus' (Mss. A-B-C), the addition of 'Lord' isstandard in SyP^l

(c) Xsywv] ^and said', is probably idiomatic, cf. Sy^'^P, Arm, Eth.

Line 2(d) x'l aoi Soxsi] The verb rCxm means both 'to be profitable'

(cf. Hebr. 4.2, for wcpeXrjCTEv) and 'to please'.^' In Sy'' ^ -A Kjjn Kjnis constantly used to express SOXEI (-1- dative). ' One might renderhere with something like 'which of the two possibilities given seemspreferable to you?'

(e) The address 'Peter' is standard in SyP* ^ as name of the apostle,even when the Greek reads

"'Cf. R. Payne Smith, Thesaurus Syriaciis. II, fasc. IX (Oxford: Clarendon Press,1893), rol. 3496 (ad Jri. 12.18 3Liiii 'vix refte').

''Cf. the connotation of Soxst (-1- dative) 'it is a pleasure t o . . . ' , mentioned inLSJ p. 442a (s.v. Soxew II, 4.b).

Mt. 18.12; 21.28: 22.17,42: 26.66: Lk. 10.36.e name ^ - -"v occurs only in Lk. 2.25, 28, 34; 3.20: cf. ^ - n * m in Mt 10,4

(C).

BAARDA Parablepsis m the CPA Lectionary 141

Line 3(f) TEXT) Y] xTJvCTov] Mss. A-C follow the Greek text. MH. B omits

'or census', "^ but this is a unique reading which may have been causedby a copyist's lapsus.

(g) XTjVCTOv] ^m k n [qison) so Mss. A and C: it is found also in Mt.22.17 (Lesson LXXVII. Mss. A-B-C), and iindem (Lesson CXLIII, Ms.C), wlierc Ms. A reads ^m^n [qenson) and Ms. B ,^fnfcfn {q'liisdnl).

Line 4(h) The text (Mss. A-C) agrees with the Greek text. Ms. B

slightly differs as if it reads the singular form.

4. Mt. 17.26a in the 'Palestinian-Syriac Lectionary'

Whereas verse 25 largely agrees with the Greek text, verse 26 seems todiffer from the generally accepted Greek text, for it reads:

^ Q_m .55. He then said: 'From the foreigners'.

The questit)n is now which Greek reading did the translator findin his Greek model? As we have seen in our introduction, there weretwo suggestions made by textual critics, namely that Syf*' read in theGreek source either dnovToq U- (with B 0 0281 1 700 892* 1675),or Xiyzi auxw* (with D, Old Latin d). In my view, SyP ' agrees withneither of these suggestions. In fact, it agrees with the reading 6 hiecpT] in Ms. N, which is as far as I see a unique reading in Greek textualtradition.^^ As to the first two words (6 Se = k'^ Qjn) tliere can be nodoubt. The verb TJILK might render either "Kiyzi, SITIEV, or £97], butsince the readings 6 SE AeyeL and 6 SE Elnev^^ do not occur in anyGreek manuscript, there can hardly be any doubt that 6 Se ecpY) wasthe text in the Vorlage of SyP^l The agreement between the Greek Ms.X and SyP ^ can be corroborated by the following consideration.

Soden {Schriften. p. 63) has noted this uniqtie omission ('pa''').^'C.F. Matthaei. Evangelium secundum Matthae.uvi, Graece et Latine (Riga:

Hartknoch, 1788), p. 270), mentioiiH Ma. 238 a« having the verb £97;, The GreekNew Testament^ mentions Ms. 1230 for this reading, but it is not clear whetherthey agree v/ith the text of Ms. S (6 SE I^YJ).

'• This latter reading is assnmed by Von Soden, Die Schriften des Neuen Tes-taineMs. 1.2 Die Textforrmm, A. Die Evangelien (Gottingen: Vaixlenhoeck <Rnprecht, 2nd edn, 1911), p. 1499.

142 Arammc Studies 3.2 (2005)

5. A Parablepsis in the 'Palestinian-Syriac'Translation of Mt. 17:26

The Greek text of verse 26b given above reads:

6. EcpT] auToi 6 'IYJOOUI;* apa yE sXeu^spoL E'-QLV QI ULOL

In Syi'^' we find the following translation of this sentence:

t< k^-\ ^Qxm ^"TLN KJLH ^ I l K K.QJT1 m nm k rCTJD mA n-ni< ^n -\

When then said to him the Lord .Tesus: 'Then^^ free are the sons'.'

This rendering is rather surprising, because one would expect after thesubordinate temporal clause a ])rincipal clause, but such a main clauseis lacking. The only possible solution is that we have here another c asoof parablepsis. The original text of SyP^ must have read here a quitedifferent text:

26a. rCiTjuu ^ T J L K n Q_m26b.26c. K A - L U , ^ i m ^"TN Kxi ] , o rC

26a. He then said: "From the foreigners,26b. but when he said: 'From the foreigners',26c. said to him the Lord Jesus: 'Then free are the sons'.

It is obvious that the eye of an early copyist wandered from the firstT-niK ('said') to the second TJILK ('said'), and thus created an awkwardsentence. This parablepsis error must have crept in textual tradition ata rather early stage of textual transmission, since this incorrect readingis found in all three extant manuscripts.

Now it becomes obvious that the original text of Sy'* ' was basedupon a reading that is found in the Greek Ms. N, which reads:

6 hk EcpY]' aTio Tuv aXXoTpLwv,^ hi' OLno TWV aXXoxpicov,

6 "Iyâou ' apa yz sXEU&EpoL elatv OL I

So the outcome of this study is surprising, because in Greek traditionthis text is preserved in Ms. N only, which deviates from most other

ptc. KOJn in combination wirii ôA renders apayc. cf. F. Schnlthes.s. Lex-icon Syropalaestinum (Berlin: Reimer. 1903), p. 49b.

^^N.B, Von Soden {Schrtften. 1.2 (A), p, 1499) c:onjecture.s as the nnderiyingGreek text ELTIOVTO^ SS aOxw xoij 'Irâryijz, (loco ecpv] XUTW 6 "ITÔOU;). but he doesnot observe that this would be an impossible text in view of the lack of an apodosis.

BAARDA Parablepsis in the CPA Leetionary 143

text forms that are attested in this verse in that it has two lines insteadof one line.

6. Testimony for 'Two Lines' in Secondary 'Alexandrian' Texts

Besides Ms. K there are only two other 'Alexandrian' manuscripts thathave two lines, namely L and C,^'' but they differ from the reading inMs. K in the first line:

Xcyst xuTO) 6 nirpo^' ino TWV aXXoTpLtov,

CLKOVTog Si- ano TWV aXXoTpLoiv, .. .

My conjecture is that this later 'Alexandrian" reading was a .secondarycorrection of the earlier Alexandrian reading preserved in Ms. N, beingan adajjtion to the Majority text.* * If this is, indeed, the case the textform which survived only in Ms. H may have had a broader testimonyin early times. For this assumption there are are other indications insome versional witnesses, namely the Bohairic and Ethiopic versions.

7. Testimony for 'Two Lines' in Bohairic Manuscripts

In several Bolmiric witnesses (A B G T D ' 2 D E G^ H 0 K O Hunt 18) ^we find the following reading. Affor Jesus's question. 'From the sons orfrom the strangers?' (NTOTOY NNiajHpi ujiN IJTOTOY Nthese manuscripts read:

1. Neoq AC nex iq . xe. NTOTOY

but lie said: 'from the foreigners'2. eT2i.qxoc JLG :xe NTOTOY

but when he (had) .said: -from the foreigners' . . .

Significant are the omission auTw and of the explicit subject in line 1.Moreover, tho; addition of AG in both lines matches the Greek particles(SE) in Ms. X. Therefore, there cannot be any doubt that this Goptic

C reads SIJIOVTOC Se aOrou, apparently a secondary reading,is text was apparently rendere{l in a gloss to the Coptic-Arabic Ms. A, 'And

Peter said: "from the foreigners", and when he said: "from the foreigners"." Cf.G. Horner, Goptic Version (...) in the Northern Dialect. I. Matthew and Mark(Oxford: Clarendon Press. 1898). p- 152 app. (Ms. A = Ev. Coptic-Arabic Ms.Oxford, Bodleian: Huntington 17).

•^7aNT^ (bo"'^^). Nestle-Aland^'. CNT^ (bol''); ef. Horner, Goptic Version. I. p.152 app.


text represents the same text form which is found in the Greek Ms.N. It is true that the Bohairif- tradition has also niannscripts with oneline. Two witnesses (J N) have only the second line, three witnesses (FC L) have only the first line. One might reason that the first of thesetexts was derived from a text as found in Greek Ms. B pc, the latterfrom a Greek text like Ms. D, as G. Horner does. * But in the lattercase {Neoq JLG ne.XA.ci) the Bezan text (XEYEL auxu) cannot be takeninto account, for it has Ae (— Hi) and omits 'to him". The only witnesswhich explains the reading is 6 SE £cp7] of Ms. K. The most reasonablesolntion is that the Bohairic tradition from the very beginning had twolines of which in some manuscripts due to parablepsis only one lineremained.

8. Testimony for 'Two Lines' in Ethiopic Texts

The same reading is also found in Ethiopic texts, ' as can be concludedfrom the data given in the new and magisterial Ethiopic edition ofMatthew pubhshed by my colleague Rochus Zuurmond,^° in whic h theB-rccension (based on the Mss. 12, 13, 14, 37,-" 38) ^ presents thefollowing reading:

and he said: 'from the forcigeners',and when he said: 'from the foreigners', . . .

The omission of both auTw and of the explicit subject makes it clearthat this Ethiopic B-text represents the Greek text of the Ms. N. Ofcourse, it is possible that the text of the B-recension was influenced bya Coptic text. However, since this recension as such was most likely theresult of adaptation to a Greek text, one cannot exclude the possibilitythat its reading was derived from a Greek text. It is most remarkablethat both the D-text and the E-text have preserved the reading of theB-text in the first line (without auxw and the explicit subject). Just

Goptic Version (...) in the Northern Dialect, I, p. 152 app.'8 (aeth™"^). GNT-^ (eth"-"), GWT* (eth"'«).

•^"Zinirmond, Novum Testamentum Aethiopice, III. The Gospel of Matthew{Wiesbaden: Harrassowitz Veriag, 2001), p. 183 {cf. p. 182, app.).

•'^Ms. 37 presents the same text, but suffered from a slight parablepsis due tohonioiarcton.

''^Zuuriiicmd, Novum Tentament^im Aethiopice. Ill, p. 10: this text is also foundin Walton's Polyglot, hut in the sectMid clause this latter text reads "and after hehad said to him .

BAARDA Parablepsis in the CPA Lectionary 145

as in the case of the Bohairic tradition it seems very likely that theselater recensions were based on the D-recension whose second line waseither deleted by a revision or rather omitted due to parablepsis.

9. Conclusion

The conclnsion is that the so-calloti Talestinian-Syriac version' has tobe listed besides other versional witnesses (the Bohairic version and theEthiopic B-recension) as a witness to the Greek text which is preservedin Ms. K. and probably in an ancestor of Mss. C and L. It is not theplace here to make a plea for the originality of this reading in Mt.17.26, as I did in the text-critical seminar at Bonn to which I referredearlier.^^ My only purpose was to give the reading of SyP*' the placewhich it deserves in the textual apparatus.

note 8.

[AS 3.2 (2005) 147-65]DOI: 10.1177/1477835105059089

SYRIAC MEDICAL TERMINOLOGY:SERGIUS AND GALEN'S PHARMACOPIA'

Siam Bhayro

Yale University

1. Introduction

In this paper, I propose to introduce the subject of Syriac medical ter-minology with particular reference to some interesting examples fromBL Add 14,661. Below I give a summary of previous scholarship on Syr-iac medical literature (§3) and then an introdnction to the translator ofBL Add 14,661, Sergins of Ra"s al-'Ayn (§4). Following this, I presentsome examples from BL Add 14,661, a Syriac translation of Galen'streatise on simple medicines, beginning with an example concerning'scurvy-grass' (§5),

In addition to the translation itself, BL Add 14,661 also containsSergins' own prologues and alphabetical lists of botanical terms. Theselists are of particular interest to the lexicographer and I present someexamples below (§6). But first, a few words on Galen's importance arein order.

'Tliis is an expanded form of the paper I presented at the 2002 Aramith Confer-enee on Aramaic lexicography (Sliefficld, UK). I would hke to thank the followingcolleagues who participated in that conference, and made helpful comments bothfollowing my paper and in subsequent correspondence: Sebastian Brock (Univer-sity of Oxford), Geoffrey Khan (University of Cambridge) and The.K)dore Kwas-man (University of Cologne). Please note the following abbreviationy: AP — I.Low, Aram(Eische Pflanzennamen (Leipzig; Verlag von Willielm Engelmann, 1881);BB — H. Bar Bahlul. Lexicon Syriacum. (ed. R. Duval; Paris: E Reipublicas Ty-pographaeo. 1888); Br. = C. Brockehnami, Lexicon Syriacum (BerHn: Verlag vonRenther k Reichard, 1895); Lane = E.VV. Lane. An Arabic- Engli.'ih Lexicon. I/l-8 (London: Williams k Norgate, 1863-1893); LS = H.G. Liddell aiid R. Scott, AGreek-English Lexicon (rev. H.S. Jones; Oxford: Clarendon Press, 9th edn., 1940);PS — J. Payne Srnifh, A Compendious Syriac Dictionary (Oxford: Clarendon Press.1903).

© SAGE Publications (London, Tlinnsaiid Oaks CA, and New Delhi), 2005


2. Galen's status in the Abbasid period

As an author, Galon (ciroa 129-200 CE) is probably to be reckoned asthe most prolific of his day. His influence in the West declined follow-ing his death,^ but his influence in the East reached far beyond LateAntiquity and the great Arab physicians of the Abbasid period were inawe of him.^ As Galen's works were the basis for medical treatises com-posed by Arab doctors, he continued to play an integral role in Islamicmedical education,'* and in their accounts of the history of medicine,Arab physicians ascribed to him the honour of resurrecting the properstudy of medicine.^ Ishaq ibn Hunayn's (d. circa 910 CE)LJ?UI « jb"'Ghronology of the Physicians' refers to Galen as being the "last of

^According to Niitton, Galen's Greek writings 'amount to ap])roximately 10per cent of all snrviving Greek literature before AD 350" tiee V. Nutton, AncientMedicine (London: Routledge, 2004). p. 390 (note 22), for this statistic, and pp.216-47 for the most up tu date introduction to Galen"s life, career and significance.See also L. Garci'a-Ballester, 'Galen's Medical Works in the Gontext of hi.s Biog-raphy', in J. Arrizabalaga et al. (eds), Galen and Galenism: Theory and MedicalPractice from Antiquity to the European Renaissance (Variorum Gollected StudiesSeries; Aldershot: Ashgate Variornm, 2002), pp. 1-5:1; and A.O. Whipple. The Roleof the Nestorians and Muslims in the History of Medicine (Princeton; PrincetonUniversity Press, 1967), p. 10. The collection of his works compiled by Kiihn reachesto twenty-two volumes—see Claudii Galeni, Opera omnia I-XX (ed. G.G. Kiihn;Leipzig: Prostat in officina iibraria Car. Cnoblochii, 1821-1833).

^On the contrasting fortunes of Galen, in the West and in the East, see O.Temkin, Galenism: Rise and Decline of a Medical Philosophy (Ithaca: Gornell Uni-versity Press, 1973). pp. 59-71. See also G. Strohmaier, 'Galen in Arabic: Prospectsand Projects', in V. Nutton (ed.), Galen: Problems and Prospects (London: TheWellcome Institute for the History of Medicine, 1981), pp. 187-96 (187).

For a summary of Arabic sources on Galen, see P. Johnstone, 'Galen in Arabic:The Transformation of Galenic Pharmacology', in V. Nntton (ed.), Galen: Problemsand Prospects (London: The Wellcome Institute for the History of Medicine, 1981),pp. 197-212 (198-200).

• For Galen's place in Islamic medical education, see E. Lieber, 'Galen in Hebrew:The Transmission of Galen's Works in the Mediaeval Islamic World', in V. Nutton(ed.), Galen: Problems and Prospects (London: The Wellcome Institnte for theHistory of Medicine, 1981). pp. 167-186 (170-180): D. Gutas, The '-Alexandria toBaghdad" Complex of Narratives: A Contribution to the Study of Philosophicaland Medical Historiography among the Arabs', Documenti e Studi sulla TradizioneFilosofica Medievale 10 (1999). pp. 155-93 (169-72).

•''The stories tend to begin with Hermes or Asclepius, who is identified withEnoch and Idrls. The first great physician is Hippocrates, but his death is followedby a dark age in which true medical knowledge is rejected. Six centuries afterHippocrates, Galen rediscovers Hippocrates' true medicine, and remains its bestexponent. For a fuller account of the Arabic histories of Greek medicine, see M.Meyerhof, 'Sultan Saladin's Physician on the Transmission of Greek Medicine to

BHAYRO Syriac Medical Terminology 149

the physicians'.'' Ibn Slna's (980-1037 CE) encyclopedic ^^UJI ,^[:Sk-JaJI ^ 'Ganon of Medicine' was heavily indebted to Galen J and, inits Latin translation, it remained the standard medical textbook inEuropean schools through to the seventeenth century.'* Galen's influ-ence spread beyond medical matters and is discernable even in termsof the genre or style of writing adopted by Arabic writers for theirautobiographical compositions.^

3. The study of Syriac medical texts

In view of the imi)ortance of the Galenic corpus, both in terms ofits size and the extent of its influence, tlie paucity of text editions ofGalen's works in its Greek. Arabic. Hebrew and Latin versions, as wellas accompanying studies on the medical lexicons of these languages,is rather surprising. ^ The situation is particularly disappointing in

the Arabs', Bulletin of the History of Medicine 18 (1945), pp. 169-78; see alsoJohnstone. 'Galen in Arabic;', pp. 198-200.

Such accounts prolsably demonstrate why Galen's works were preferred to thoseof Hippocrates. The eleventh centnry physician Ibn Ridwan wrote that Galen'refined the teachings of Hippocrates and made the art of medicine easy andcomprehensible'—for more on Galen's importance compared to Hippocrates, seeM.W. Dols. 'Syriac into Arabic: The Transmission of Greek Medicine'. ARAM 1(1989). pp. 45-52 (47).

*'F. Rosenthal. 'Ishaq b. Hunayn's Ta'rih al-atihbd''. OHens 7 (1954). pp. 55-80(75).

^Strohmaier. 'Galen in Arabic', p. 188. For an introduction to the Canon, seeH.D. Isaacs, "Arabic Medical Literature', in M.J.L. Young et al. (eds). Religion,Learning and Science in the 'Abbasid Period (CHAL; Cambridge: Cambridge Uni-versity Press, 1990), pp. 342-63 (356-58).

*'For a list of Latin translations of the Canon published after 1500, see N.G.Siraisi, Avicenna in Renaissance Italy: The Canon and Medical Teaching in ItalianUniversities after 1500 (Princeton. N.): Princeton University Press, 1987). pp. 361-66. For its primary status in European schools, see S.C. Nogales. 'Ibn Slna', inM.J.L. Young et al. (eds). Religion, Learning and Science in the 'Abbasid Period(CHAL; Cambridge: Cambridge University Press. 1990), pp. 389-404 (391); Isaacs,'Arabic Medical Literature', p. 358: Siraisi, Avicenna, pp. 43-124.

"This is mentioned by F. Rosenthal in his review of Galen: On Medical Expe-rience.^ ISIS 36 (1946). pp. 251-55 (251). and discussed further by M.J.L. Young.'Arabic Biographical Writing', in M.J.L. Young et al. (eds). Religion. Learning andScience in the 'Abbasid Period (CHAL; Cambridge: Cambridge University Press.1990), pp. 168-87 (183).

'"For Greek, only a few critical editions exist see. for example, the Corpusmedicoriim Graeeorum (Berlin: In at^dibus B.G. Teubneri. 1927-), whicii pnblishesother authors in addition to Galen. KiJhn's edition (see n. 1) is uncritical and

150 Aramaic Studies 3.2 (2005]

respect of the Syriac Galenic corpus, or even any Syriac medical liter-ature, particularly in view of the importance of the Syriac translationsin the transmission of Greek medical literature into the Islamic world. '

In 1870, Sacliau published three excerpts of Galen found in folios13-15 of BL Add 17.156. giving just the Syriac text.'^ Our presenttext, BL Add 14,66L was published by Merx in 1885.'^ Merx gave theSyriac text along with the e<iuiva!ent Greek text according to Kiihn'sedition. For the passages for which there is no Greek equivalent, suchas the translator's prologues, Merx provided a German translation.There is hardly any analysis, however, and Merx's article seems to havestimulated very little subsequent research. In 1899. Gottheil publishedfourteen folios from MS. Syriaque 325, giving the Syriac text and anEnglish translation with minimal notes.^^ Again, as with Merx's edi-tion of BL Add 14.661, Gottheil's efforts remain largely nnutilised.In 1913. Bndge published BL nis Oriental 9360, the soKâlled Bookof Medicines.^'^ This two-volume edition certainly did stimulate subse-quent research, particularly in a series of articles by Schleifer in whichthe dependency of much of the Book of Medicines upon Galen was es-

outdated. but very often all we have at our disposal. Regai'ding Arabic, the situationis worse (see footnote 19 below), with fewer text editions and most of the materialstill available only in manuscript form see. for example, the Corpus medicorumGraeeorum supplementurn orientate (Berlin: In aedibns Acadeniiae Scientiarum,1963-). The situation regarding the Hebrew sources is worse still, with no texteditions yet available. Regarding medical terminology, the most recent study ofLatin medical terminology is D.R. Lang.slow. Medical Latin in the Roman Empire(Oxford: Oxford University Press. 2000). For Araliic medical tenninology, there isnow M. Ulhnann. Worterbuch zu den giiechisch-aiubischen Ubersetzungen des 9.Jahrhunderts (Wiesbaden: Harrassowitz Verlag. 2002).

'Âs Dols states, 'The Syriac translations of Greek medical works were the vita!,althougli usually forgotten, links in the transmission of the texts into Arabic and.finbseqnently. their disseniinati<jn in Islamic society' - Dols. 'Syriac into Arabic', p.45.

^Ê.C. Sachau, Inedita Syriaca (V'ietma: K.K. Hof- und Staatsdruckerei, 187U;Hildesheim: G. Ohns, 1968), pp. \ - --ua. These folios contain excerpts that corre-spond to the following equivalent pa.ssages, referenced according to Kiihn's edition:f. 13 - I pp. 384-87. f. 14 - I pp. 369-72. f. 15 - VI pp. (i43-47.

•*A. Merx. "Proben d(T syrischen Ucbersetzung von Galemis" Schrift iiber {lieeinfachen Heilmittel'. ZDMG 39 (1885). pp. 237-305.

^''R.J.H. Gottheil. 'Contribntions to Syriac Folk-Medicine', JAOS 20 (1899), pp.186-205.

^Ê.A.W. Budge (ed.). Syrian Anatomy Pathology and Therapeutics or "TheBook of Medicines": The Syriac Text Edited from, a Rare Alanuscript with an En-glish Translation (London: Oxford University Press. 1913).

tabhslied."' At this point, it is worth repeating Gotthcirs observationthat the lexicographers Bar Ali and Bar Bahlul cite Galen.^''

Of the entire Syriac medical corpns, therefore, there are only fourtext editions and of these only two are fully translated. Furthermore,whilst reviewing the above sources, I recently discovered something ev-idently overlooked by previous analyses, namely that the third source,Gotthoil's publication of MS. Syriaque 325. is in fact another versionof part of the fourth source. Budge's Book of Medicines.^^ So we arereally left with three published sources, rather than four, and only oneof these publications has stimulated subsequent research of any depth.This means that we know very httle about both the development ofSyriac medical terminology and the relationship between the Syriacmedical lexicon and its Greek and Arabic counterparts.^^

^^J. Schleifer. "Zum syrischen Medizinbuch'. ZS 4 (1926). pp. 70-122. 161-95: 5(1927). p]). 195-237; 6 (1928). pp. 154-77, 275-99; idem 'Zuni syrischen Medizin-bnch\ RSO 18 (1940). pp. 341-72; 20 (1942-3), pp. 1-32, 163-210, 383-98; 21 (1946),pp. 157-82.

'''See Gottheil, Contributions, 186. For a more detailed discussion of these twotenth-century Nestorian lexicographers, and their sources, particularly the jixoaf^oiswj 'Explanation of Words' (once said to be composed by Hunayn), see G.EndresK, "Bilingual Lexical Materials in the Arabic Tra<lition of the Hellenistic Sci-ences', in J. Hamesse &: D. Jacquart (ed.s), Lexiques Bilingues dans les DomainesPhilosophique et Scientifique (Moyen Age-Renaissance) (Turnhout: Brepols Pub-lishers, 2001), pp. 161-73 (164-66, 168).

^**Gottheil divides his text into forty-four sections. GottheiTs §1 is found in Budgepp. 553-54 (text) and p. 657 (translation). Gottheil's text continues more or less thesame as Budge's, so that his §44 is found in Budge p. 561 (text) and p. 606 (trans-lation). Note, however, that in Budge's manuscript, Gottheil's §44 is a continuationof his §43, rather than a new prescription. I intend to provide a more thorough com-parison of these sources and to discuss the significance of this discovery elsewherein the future.

'^Interest in this field does seem to be increasing, however, as Ford's recent paperdemonstrates—see J.N. Ford, "Two Syriac Terms Relating to Ophthalmology andtheir Cognates', JSSt 47 (2002), pp. 23-38.

Regarding the Arabic medical corpus, the situation is still often frustrating.For example, in her recent edition of the Greek text of Galen's Art of Medicine,Veroniqne Boudoii discusses the existence of the Syriac fragments publishedby Sachan (see footnote 12 above) as weil as the later Arabic translation -seeVeroni<ine Boudon. Calien: Tome II: Exhortation a I'etude de la medecine: Artmedical. II (Paris: Les Belles Lettres, 2000), pp. 229-38. Both of these prove be-yond use, however, as Sachan's edition remains withont translation and analysis,and no text edition for the Arabic version exists—this despite Gottheil's statementin 1899, 'Sufficient material is at hand for an edition of portions of the Arabic

152 Aramaic Studies 3.2 (2005;

In view of this sorry situation, it is not surprising that Degen pro-posed the publication of a Corpus Medicorum Syriacorum and provideda useful introduction.^" But the Syriac cor])us still remains beyond ac-cess and analysis for the most jmrt. and our knowledge of the develop-ment of Syriac medical terminology and its relation to the Greek andArabic medical lexica remains minute. I intend to start publishing andanalysing the Syriar corpus, concentrating on providing text editionsof the manuscripts, each with a translation into English, concordance,lexicon and analysis of the medical terminology with reference to theequivalent Greek and Arabic terms. I propose to begin with BL Add14,661 as Merx's 1885 edition provides a useful starting point.

4. Two translators: Sergius and Hunayn

BL Add 14,661 contains three complete books of Galen—his sixth, sev-enth and eighth treatises on simple medicines De Simplicium Medica-mentorum Temperamentis et Facultatibus}^ The translation is bySergius (d. 536 CE), a sixth century Christian priest from Ra's al-'Ayn. Educated in Alexandria, Sergius worked in the fields of medicine,philosophy and theology, translating and composing introductions toGalen's medical works and the Pscndo-Dionysian Gorpus, in additionto composing two introductions to Aristotle's Logic.' • According toWright, our present manuscript, BL Add 14,661, dates to either thesixth or seventh eontury, so it is not too far removed from Sergiushimself.' '

Galen. It is a wonder that no one. as yet. has undertaken this work'! See Gottheil.Contributions, p. 186. Similarly, no text edition exists for Ibn Sma's ojjlill 'Ganon'.

• R. Degen. 'Ein Gorpus Medicorum Syriacorum'. Med.hist.J. 7 (1972). pp, 114-22; idem. 'Galen im Syrischen: Eine Ubersicht liber die syrische Uberlieferung derWerke Galens'. in V. Nutton (ed.), Galen: Problems and Prospects (London: Well-come Institute for the History of Medicine, 1981), pp. 131-66.

^^This equates with the following sections of Kuhn's edition: XI pp. 789-892, XTIpp. 1-158. See also Degen. 'Galen im Syrischen'. p. 146.

S. Brock. A Bri.ef Outline of Syriac Literature (Kottayam: St. Epbreni Ec-umenical Research Institute. 1997). p. 43; idem. "Greek and Syriac in Late An-tique Syria', in A.K. Bf)wnian and G. Woolf (eds). Literacy and Power in the An-cient World (Cambridge: Cambridge University Press, 1994), pp. 149-60 (156); W.Wright. A Short History of Syriac Literature (London: Adam and Charles Black.1894), pp. 88-93.

23\\;' Wright. Catalogue of the Syriac Manuscripts in the British Museum ac-quired since the ye.ar 1838, III (London: British Mnseum, 1872), p. 1187.


According to the thirteenth century Christian scholar Bar Hebraeus,Sergius was the first to translate Greek medical and philosophical worksinto Syriac i"*

And there were also excellent Syrian physiciauw, such a.s Sergius of Ra'nal-'Ayn who first translated medical texts from Greek into Syriac.

Accordingly, the thirteenth century Islamic biographer Ibn ablUsaybi'ah also ascribes this accomplishment to Sergins.^^ but his briefbiographical reference to Sergius is not so complimentary:^''

JiJI

SergiuH of Ra's al-'AynOne of the people of the city of Ra's al-'Ayii. He translated many booksand he was mediocre in translation.And Hunayn used to Improve his (Sergius') translation. When it (a trans-lation) is found with the improvement of Hmiayn, then it is the goodone.And what is found unimproved (by Hunayn) is mediocre.

So Sergius' translations are compared unfavourably to those of theninth century Nestorian Hunayn ibn Ishac} al-Ibadl.^' Hunayn, perhapsthe most celebrated of the translators from the 'Abbasid period, is

from E.A.W. Bndge (ed.), The Chronography of Gregory .Abfi'l Farajthe son of Aaron, the Hebrew Physician commonly hiown as Bar Hebraeus beingthe first part of his political history of the world (London: Oxford University Pre^,1932), II, p. 21'", col. ii, lines 2-5. Note, however, that a slightly different text i.sgiven in Budge. Book of Medicines, p. CLIV:

s, ".Arabic Medical Literating', p. ;J43.according to A. MUUer (ed.), Ibn AU Usaibi'a (d. 1270): 'Uyun al-anbd'

fl tabaqat al-atibbd\ I (Cairo: al-Matba'ah al-Wahblyah. 1882), p. 204. I would liketo thank Mr. Kevin Van Bladel for help with this Arabic source.

" For a brief introdnction to the life and work of Hunayn, .see L.E. Goodman,'The Translation of Greek Materials into Ara!)ic\ in M.J.L. Young et ai (eds).Religion, Learning and Science in the 'Abbasid Period (CHAL; Cambridge: Cam-bridge University Press, 1990). pp. 477-97 (487-91). See also Whipple, Role of theNestorians, pp. 24-29.


described by Isaacs as the 'Erasmus of the Arabic Renaissance'.^^ In

his Risala^ Hunayn states that he translated at least ninety-five Galenic

works into Syriac and a further thirty-nine into Arabic.^^

In Bar Hebraeus' account of how Hunayn restored his career after

an early setback, we again encounter this unfavourable comparison of

Sergins in respect of Hunayn, albeit from the lips of one of Hunayn's

contemporaries:^"

> * • " ' * ' . \ n .

And (Hunayn) departed weeping. And he went to the land of theRdw.dye\ and he wiis there until he had learnt the Greek language thor-oughly. And he was able to translate texts from Greek into Syriac, andfront Syriac into Saracen (i.e. Arabic). And he retnrned again to Bagh-dad as a Greek (or perhaps, in Greek apparel) and he entered beforeGabriel, the head of the physicians, son of Bokhtisho.^-^ And when he(Gabriel) had tested his (Hunayn's) knowledge, he (Gabriel) greatly hon-oured him (Hunayn) and he named him "Onr master Hunayn". And hesaid to those at hand, "If this one lives, the world will not leave anym e m o r i a l f o r S e r g i u s of R a ' s a ^ ' ^

t:s, 'Arabic Medical Literature", p. 344. According to Meyerhof and Sobhy,Huuayn translated in excess of 150 works (including his own translation of Galen'sDe Simplicium), and composed a further 100. the vast majority of which are nowlost—see M. Meyerhof and G.P. Sobhy. The Abridged Version of "The Book ofSimple Drugs" of Ahmad Ibn Muharmnad al-Chfiq by Cregorius Abul-Farag (Barhe-braeus): Edited from the only two known Manuscripts with an English Translation,Commentary and Indices, 1/1-4 (Cairo: Al-Ettemad Printing Press & PublishingHouse, I/I Alif 1932, 1/2 Ba & Gim 1937, 1/3 Dal 1938. 1/4 Ha & Waw 1940).I/I p. 12.

^"See G. Bergstra.s.ser (ed. and transl.). Hunain ibn Ishaq: Ube.r die .syrischen undarabischen Calen-Ubersetzungen (Leipzig: F.A. Brockliiius. 1925: Nendcln. Liecht-enstein: Kraus Reprint, 1966). Hunayu's translation activities are discussed in detailin S. Brock, 'The Syriac Background to Hunayu's Translation Techniques'. ARAM3 (1991). pp. 139-62 (140).

•"'Syriac text from Budge. Bar Hebraeus^ IL ]>. 53*", col. ii. lines 9-19.•*'Tbe scribe has erased an initial .t here.• • For more ou Gabriel bar BokhtiHlio', see Wright, Short History, pp. 214-15,

217.'*''0r perhaps, "...will not the world abandon tbe memory of Sergius of Ra's

al-'AynT.

BHAYRO Syriac Medical Teiminology 155

So it is clear that, to a certain extent, Hunayn's reputation is estab-hshed at the expense of the earlier Sergius. '*

The contrast most often drawn between Sergius and Hunayn inrespect of their translation techniques Ls that Sorgius is more hteraland Hunayn is more idiomatic''^ This is probably not true, but even ifit were, it should not be taken as a criticism of Sergius for two reasons.Firstly, as Rosenthal pointed ont. this literal approach to translationwas typical of the seventh century translation movement in general.^^Secondly, this more literal approach would mean that Sergius' Syriactranslations would have been closer to the Greek Vorlage than themore idiomatic translations and revisions of the later Hunayn and,hence, still of value.^^

There is evidence to suggest, however, that the 'hteral' label appliedto Sergius is too simplistic. In his Introduction to Aristotle's Categories,dedicated to Theodore. Sergius writes:'*^

When we were translating certain works of the doctor Galen from Greekinto Syriac, I used to translate, while you would write it down after me,correcting the Syriac wording, in accordance with the requirements ofthe idiom of thi.s language.

So it seems that there was an attempt on the part of Sergins to pro-duce an idiomatic translation. Indeed, Brock observes that Sergius andHunayn were similar in their approaches to translation, with both be-ing reader-orientated, and that the difference in the quality of theirproducts may he, in part, in the considerable lexicographical develoi>

^'*Indeed, Huuayn seem.s quite willing to do this himself, as his references toSergius in tbe Risala make c-lear sec Brock. 'Hunayn's Translation Techniques'.p. 150.

^^For example, see Goodman, 'Translation', p. 488.^^The fourteenth century scholar al-SafadT observed how the earlier translators

would render 'each Greek word by a single Arabic word of au exactly correspondingmeaning' whilst tboseof Hunayn's generation and later would grasp "in his mind themeaning of the whole sentence' and then render it 'by a corresponding sentence iuArabic, regardless of the congruence or lack of congruence of the individual words'.See Rosenthal's review of On Medical Experience, pp. 253-54.

^^Tbus Strobiiiaier states that Hunayn's 'translations are not so literal a-s toperruit a reconstruction of the Greek text where this is lost, but tbey reveal athorough understanding of its meaning'- Strohmaier. 'Galen in Arabic', pp. 187-88.

' Tbe Syriac text is si ill unpublished. Tbe Englisb translation is from Brock.Brief Outline, p. 202.


ment observable in Syriac during the three aiui a half centuries thatseparate them.'^

The unfavourable view of Sergius was probably motivated by anagenda that sought to promote the translations of Hunayn, somethingeasy to accomplish at the expense of the earlier Sergins. But Sergius'role in the transmission of Greek medical literature into the Islamicworld is crucial and can hardly be overstated. As mentioned above,Sergius' translations were the first translation of any Greek medicalwork into a Semitic language, a daunting task for even the most giftedtranslator.*^ Furthermore, according to Hunayn. Sergins translated theGreek (Alexandrian) canon, except for tw o items, as well as otherGalenic material into Syriac. Indeed, it seems that Sergius virtually cre-ated a Syriac canon of Galen's works, which was a modification of theGreek canon. ^ It is also worth noting that, as the above quote from Ibnabl Usaybi'ah implies, many of Hunayn's ^translations' were probablynot translations, bnt instead revisions of Sergius' earlier translations.'*^

Hunayn is rightly credited for the quality of his translations andrevisions, and also for his role in the creation of an Arabic medical vo-cabulary.*^ In this regard, it is noted how those of Sergius' generationtended to transliterate technical vocabulary rather than find equiva-lents in the target language.'*"* Thus, in our present text, there are listsof botanical terms in transliteration. But, as our discussion of thesetransliterated lists demonstrates (for example Dorycnium in §6, exam-ple 3), there is much to be learnt from even this aspect of the work ofSergius and his efforts certainly merit a close analysis. Furthermore, asBrock observes, the socio-linguistic context had changed greatly in theintervening three and a half centuries between Sergius and Hunayn,with Greek not retaining its former prestige among Hunayn's intended

, "Hunayu's Translation Tecbniqnes', p. 151.ls, 'Syriac into Arabic', p. 45.

r, 'Galen in Hebrew', pp. 173-75.r, 'Galen in Hebrew', p. 168.

and Sobby, The Book of Simple Drugs, I/I, pp. 12-13: Goodman,'Translation', p. 48S-89.

' Again quoting from ai-Safadl, 'There are no corresponding Arabic words for allGreek words; therefore, in this (literal) kiud of translation, many Greek expressionsremain as tbey are'—see Rosenthal's review of On Medical Experience, p. 254.


readership.''' Before we consider some of these transliterations, let usfirst examine one example from the main body of the text.

5. Sergius' translation of Galen's discussion of ^scurvy-grass'

Sergins provides the following sentence concerning 'scurvy-grass' fromBL Add 14,661. fol. 22a:

And also the leaves of this plant are astringent and bind wounds,in the likeness of the leaves of the lapathum of the field (i.e. wild lap-athum).

Merx identified the following section from the Greek version of Galen'sworks as corresponding to this Syriac excerpt:"*^Kiihn, vol. XI, p.854:

TJq Ttooti; ECTTI ^£V xal -ra cpuAXot axuTtTLxa TS xalTpaufxa-uwv otiOLa -zoic, TWV dypt-cov ovTa XaTia&tov.

Concerning the grass (SpETavLXYJ , the leaves are both astringent and(useful for) closing up wounds, like those of the lapathum of the field.

Thus we have Sergius' translation of Galen's discussion of the plantY; ppsTavvLXTj or r\ [3p£TTavt.x7 , Lat. Britannica., which, according toPliuy, may have got its name because it grew opposite the coast ofBritain:*^

In Gemiania trans Rhenum castris a Germanico Gaesare promotis mar-itirno tractu fans erat aquae dulcis solus, qua pota intra hiennium dentcsdeciderent compagesque in genibus solverentur. stomacacen medici vo-cabant et scelotyrben ea mala, reperta auxilio est herba quae appellaturbritannica, non nervis modo et oris m,alis salutaris, sed contra anginas

•*' 'Brock, "Hunayn's IVanslation Techniques', p. 147. It is also worth consideringthe peculiar difficulties encountered in the translation of pharmacological/botanicalworks, such as the need to identify the plant correctly, especially with plants notnative to the region of the translator—such things are discussed in detail by John-stone, 'Galen in Arabic', pp. 200-202. Tbus Sergins' decision to transliterate is notentirely beyond justification.

" ^Text according to Merx. 'Proben der syrischen Uebersetzung', p. 271. All otherquotes are my own readings of BM Or Add 14,661.

"^'Merx, 'Proben dor syriscben Uebersetzung', p. 271-72.' ' Text and translation from Pliny. Natural History: Books XXIV-XXVII (trans.

W.H.S. Jones: LCL 393; London: Harvard University Press. 2nd edn., 1980), pp.150-51.

158 Aramaic Studies 3.2 (2005'

quoque et contra serpentes. folia habct oblonga nigra, radicem, nigrajn.sueus eius exprimitur et e radice. florem vibones vocant, qui collectusprius quam tonitrum audiatur et devoratus securos in totum annum ametu anginae praestat. Frisi gens tum fida, in qua casira erant, mon-stravere illam, mirorque nominis causam, nisi forte confines oceano Bri-tanniae vtluti propinquae dicavere. non enim inde. appellaiam, quoniamibi plurima nasceretur, cerium est etiamtum Britannia libera.

When Gcrmaiiicus Caesar had moved forward his camp across the Rhine,in a maritime district of Germany there was only one source of fresh wa-ter. To drink it caused within two years the teeth to fall out and the useof the knee joints to fail. Physicians used to call these maladies stom-acace and scelotyrhe. A remedy was found in the plant called britannica,which is good not only for the sinews and for diseases of the mouth, butalso for the relief of quinsy and snake-bite. It has dark, rather long leaves,and a dark root. Its juice is extracted even from the root. The blossom iscalled vibones; gathered before thunder is heard, and swallowed, it keepsaway the fear of quinsy for a whole year. It was pointed out to our menby the Frisians, at that time a loyal tribe, in whose territory our camplay. Why the plant is so called I greatly wonder, unless perhaps, livingon the shore of the British ocean, they have no named the britannicabecause it is, as it were, a near neighbour of Britain. It is certain thatthe plant was not so named because it grew abundantly in that island:Britain was at that time an independent state.

Thus according to Pliny, it was used by Roman soldiers to heal soremouths and painful knees, prompting modern analysts to suggest itwas used in the treatment of scurvy, heiu-e the definition 'scurvy-grass' —a reference to any of the various cruciferous plants of the genusGochlearia used in the treatment of scurvy."^^

Of cotirse, the Greek text given above comes from Kuhn's edition ofGalen's w^orks. Despite our reliance on this nineteenth century edition,however, we can still establish useful lexical equivalences between theGreek and Syriac terminology if we bear in mind that later researchon the Greek manuscript sources may illuminate this further.

We also have at our disposal a very useful Arabic text, namely. BarHebraeus" abridgement of the twelfth century Andahisian physician Al-

the discussion in Meyerhof k. Sobby, Book of Simple Drags, 1/2 pp. 329-30;see also G. Hatfield, Memory, Wisdom and Healing: The Histort/ of Domestic PlantMedicine (Stroud: Sutton Pnblishing, 1999), p. 162, who mentions that variousspecies of thiy genns grow ai'ound tbe Britisb coastline as weil as on mainlandEurope.


Ghafiqi's Book of Simple Drugs:'^ The original Arabic version of thispharmacological textbook is now lost, although a Latin translationexists. The abridged version of Bar Hebraeus, although not a word forword translation, does preserve the salient points and is an importantsource for the Arabic pharmacological terminology. Obviously cautionhas to be exercised as this Arabic text is an epitome rather than atranslation.

For the example under discussion, the Arabic abridged text

BartaniqaAn astringent—it heals wounds.

With the Greek text of Kiihn, the Syriac text of Sergius, and the Arabictext of Bar Hebraeus, we can produce the following sample lexicalentries;

A11. bot. scurvy-grassGr, 1 [BpexavvcxTj or 7] ppeTTavcKT : Ar. ,_jA-jlJ?y; Lat.Britannica.Apparently any of the various cruciferous plants of thegenns Cochlearia used to treat scurvy [BL Add M.G613a (22a); cf. Kiihn Xl:854; MeySob, no, 155; Pliny NHXXV:20-21].

vb, ther. bindGr. xoXAaco; Ar. J_si.Pe. cause wounds to close up, BL Add 14,661 22a:

and they close up wounds {xaL.V [Kiihn XI:854]; OLP-I_^I j^j^ [MeySob, no.

155]),

n. ljot. cress or m o n k ' s r h u b a r bGr. TO Xa7ra&ov.Meaning uncertain; PS 146 gives 'lapathum, cress', i.e.general term for various cruciferous plants: for TOXajiaOov LS 1030 gives 'monk's rhnbarb\ i.e. dock species

^"Meyerhof & Sobhy publisbed only part of this, covering tbe letters Alif throughto Waw see note 28 atjove.

•"' Meyerhof & Sobby, Book of Simple Drugs. 1/2 p. 73,

B

c

Rumex patientia and Rmnex pseudoalpinus; BL Add14,661 22a: rtft.=i.i.i r A>—i.cu:rui.i r ^ a T ^ A»a»i.T3 in the like-

ness of the leaves of theM. of the field (ie. wildM.) (Sti-otaTfiic, Twv ayp^"^ ^'^'^'^ XandcSuv [Kiihn XI:854]).

Dr<3t\^ n. bot. leaf

Gr. TO CpuXXov.Of plant; e.g. in description of therapeutic use of

râ, BL Add 14,661 22a: . r o o a l ^ -3r<ror<j_ai *« r<im r^\n v^ and also the leaves

of this plant are astiingent [x%q, Tioa; eaTi fiiv xai. xa(puXXa axujtxLxa [Kiihn XI:854]).

Er<A»cui:?3 n. pa th , w o u n d {(f. rciiso to strike.)

Gr. TO Tpau[i.a; Ar. TJ^.(External) tissue damage; e.g. in description of thera-peiitic u.se of rtfoj-irtfipi^ts, BL Add 14,661 22a:re'd^ciioo and they close up wounds {xal xoXA7]xt,xijiaxcov [Kiihn XI:854]; o U I ^ I J-JU. [MeySob, no. 155j).

Fadj. ther. astringent (cf. -ai^ to astringe)Gr. CTTUTiTLXo?, '(\, 6v; Ar. ij^ti-causing tissue to contract thus reducing blood flow orother secretions: e.g. in description of therapeutic use ofr<a.î<\jf<\s, BL Add 14,661 22a: i<tiii.i .mcLa*! î<'notnjiur<r rtAî-so r^tn and also the leaves of ihis plant

are astringent (XT ^ Tioa? loxl [xev xal xa cpuXXa axuTixixa[Kiihn XI:854]; ^L i [MeySob, no. 155]).

Gn. bot. p lan tGr. 1 7r6a.in specific reference to a medicinal plaut; e.g. in descrip-tion of therapeutic uso of r«;£i-jrt%r<rt=. BL Add 14.661

also the leaves of this plant are astringent (xf,; noa; eaxlLLsv xal xa 'fuXXa axu7rTLxa [Kiihn XL854]).

Eath of the above (nitries gives the lemma, followed by a definition anda simple gloss. Next there is a list of lexical equivalences gleaned fromthe parallel sources, and finally a section of referenced examples. Evenfrom this one example, we can see how this kind of analysis of botanical

and medical terminology adds to our current lexical knowledge in anumber of respects.

Gonsider entry B above. According to PS 82b, the root jia.i Pe.'to cleave, adhere' etc. hai5 an Aph. use 'to make adhere', which in amedical context can mean 'to close a wound'.''' In our text, however,it is clearly the Pe. being used (^.ji=.i-Pe. m. pi. act. ptc), so itappears that, at least in Sergius' medical texts, the Pe. also has themeaning 'to fasten^to close up (a wound)'. So entry B adds to thelexical data for the root jia.-i.'*^ In terms of the lexical equivalences,the verb xoXXau means 'to glue or fasten together', with XOXATJTLXO

referring to the 'closing up {of a wound)' (LS 972), and the root J - :simply means 'to put right', and in a medical context means 'to healby causing a wound to close up or skin over' (Lane 914c). Thus thereader is furnished with two referenced lexical equivalences, the secondof which is not so obvious.

Example F refers the reader to the verb *3i_. whidi means 'topurge'. Both PS 485 and Br. 308 give the meaning 'to a^stringe'. Thesubstantive r<:ii-*^ is also listed (PS 295 and Br. 308b). Our en-try furnishes the reader with two lexical equivalences: the adjectiveazunxiKoc, 'astringent', derived from the verb tJTucpw meaning 'to drawtogether' (LS 1658), and the Ar. ^ L s , from the root ^jo^ meaning 'tograsp with tlie hand\ thus 'to contract' and therefore 'be astringent'(Lane 2481c-2482b, 2483l>-c). Finding these lexical equivalences wouldnot be an easy task. Again, as with the previous example, the citationprovides the reader with a context for the lexical equivalences.

Regarding the nouns, example A is a clear case of the Syr. andAr. terms being transliterations of the Gr./Lat. term. The referencesfurnish the reader with the parallel passages that provide these terms.With example C, however, we have a proper Syr. term rather thana transliteration. The identification of Syr. r ^L^^cunu with Gr. TOXarra&ov is well known (e.g. PS 146), but the above entry furnishesthe reader with a clear basis for this. As with examples B and F, theAr. equivalent in example E is not so obvious, being derived from theroot ^jfr 'to be destroyed' (Lane 480c-81a).

•'^Br. 65b-66a lists no equivalent medical usage.^•'lii giving the example from BL Add 14,661, the reader is informed of the source

for this particular use of the Pe., which should prevent unwarranted conclusionsbeing reached in other contexts and with other writers.

162 Aramaic Studies 3.2 (2005]

6. Sergius' transliterations of Greek botanical tenns

Sergius prefixes a short introduction, dedicated to tlie priest Theodore,to his translation of each of Galen's books. Foiiowing each introduction,he provides a hst of Syriac transhterations of Greek botanical terms.This list is of great interest for the lexicographer. The form of many ofthe entries reads ,7: which is y. with x being a Syriac transUteration ofa Greek botanical term, and y its suggested Syriac equivalent. Sergiusappears to be aware of the speculative nature of some of his suggestedSyriac equivalents as he sometimes uses the phrase x which is perhapsy (see Example 3 below). The following three examples illustrate thenature of the data obtainable from this list, and some of the problemsencountered in its analysis.

Example 1 - WormwoodBL Add 14.661, foi. 2b:

Woiynwood which is wild brkt'.

Syr, cpcuoYotartf' is a transliteration of the Gr. TO a[ip6Tovov, whiehrefers to "wormwood' (LS 3b). It is worth noting that the Greek nounis nenter, but Sergius transliterates it as though it were masculine.Sergius is identifying TO (x[ip6TOvov with 'wild r^hi^-i^'. But we needto be carefnl as this identification of TO a^poTovov with "wild pÂi^ta"is only Sergius' opinion, so an appropriate lexicon entry would need toshow

H.. 'i'^ r inu-T r<Avjit=j n, bot. wormwood?'-f;,";l :-, •?, Ace. to Sergius, to be identified with Gr. TO a[ip6Tovov.• : 'V . • Sergius: r<imj.i r<rA^t3 .rooAurî <»nja\,oti3r^ Wortn-

' ."':. -' wood which \s wild 1= [BL Add I4.GG1 2b].

Example 2—A locBL Add 14,661, foi. 3a:

Aloe which is aloe.

In this example, Sergius wants to list aloe by its Greek name(LS 72b). which he transliterates as r<t<cSr<r. Sergius also knows the

is accomplished in BB 20: . i s i ^ i i .ro.i f<W>.T rÂ^ta 00.^11:0 uyrtf

'according to SergiuH, wild '.a. wliicli is wormwood' (see PS 97Vj for this Persianloan word); see also AP 78-79.


Syriac word for aloe, so he inserts it following his usual 'which is'formula. The existing Syriac equivalent, however, is , a l i (AP 295; PS414b). which is itself probably a Greek loanword into Syriac. So we areleft wondering why Sergius chose to give what is effectively anothertransliteration.

In his introduction to the book, Sergius writes the following:BL Add 14,661, foi, 2v. 1-13:

mi

Then he (Galen) begins, moreover, in tlie order of the lettcrH/'' audteaches concerning every other material that is taken for a remedy; fromstones and from within the ground, and from flesh and from dung, andfrom blood and from the fur of beasts. Aud with these, he completes thediscourse of these six latter treatises. In this discourse, then, which wenow intend to translate from the Greek language iuto the Syriac tongue,it being tho sixth out of this entire work, but the first of the remainingones, the writer commences on the discussion, as we said, by settingdown in it all the plants whose name begins with the first letter, namelyalpha, up to the ninth, that is, iota. And with this' ^ he completes thesixth discourse.

Thus Sergius is at pains to show that he has translated this text directly fromGreek into Syriac and, in tho process of doing so, he has been challengedby the Greek botanical terms that were listed in alphabetical order froma to t,. It thus appears that Sergius is rather proud of his achievement intransliterating the list and hence preserving its Greek alphabetical order,something evidently still important in his period (see n. 45). A lexical entryfor such a case could look like this:

. the letters of the Greek alphabet.

. the ninth letter iota.

n. bot. aloeSergius' transliteration of Gr. 7] aXoY] (cf. ,ni_s. aloe).BL Add 14,661 3a: .CAJ. olL.A re' reTK-ciirer Aloe which is

ak)e.

This entry informs the reader of the more common Syriac term, the Greekequivalent and the reference for this particular transliteration. An entry forthe 'proper" Syriac term would also include this reference:

J,n\ V n. bot. aloe

Gr. : aXoY] (cf. r^rtfoAp^ a/oe).Sergius (BL Add 14.661 3a): .cvii. ^b^r^ r^râlr^ Aloe

which is aloe.

Example 3—Dorycniurn?PS l ib gives the following entry:

rîwr^ or r^jjor^ nl. dorycnium. an alkaline plant

Thus we are given two Syriac forms, nn equivalent Latin name and a prop-erty of the plant. Bnt from where does this identification of rtfijjcir*' withdorycnium come? In antiquity, both Dioscurides (SopuxvLOv) and Galen(SopuxvLStov) discussed it, but subsequent authors were unable to iden-tify it. Therefore, the Arabic sources simply transliterated it as o.^3j^ o""Ojrr^jji^ and it remained one of the few botanical terms for which no Arabicequivalent was found. '

Hunayu translated Dioscurides' discussion of the dorycnium into bothArabic and Syriac, and he also translated Galen's discussion into Arabic,so it is thought that the identification of r^âr^ with dorycnium comesfrom Hunayii. Thus Meyerhof and Sobhy state 'Hunain ibn Is-haq itloutifiedthe name doryknion with Syriac uhld s' mN'. ^ Bnt the following line fromSergius' list of transliterations sheds new light on this.BL Add 14,661, foi 4b:

Dorycnium which is perhaps rAuor^.

So we see that Meyerhof and Sobhy were mistaken on two counts. Firstly,the identification of r<dwor^ with dorycnium comes from Sergius and not

fc Sobhy, Book of Simple Drugs, 1/3 p. 514-15.k. Sobhy, Book of Sim.ple DTugs. 1/3 p. 515. Note that Low tUmbted

the accuracy of this identification (AP 42-43), wtiicli perhaps fits in witli Srrgius'own doubt (see below).


from Hunayn. Secondly, their -uhld x'rmx' should instead be 'uhldWe note also that Sergius himself was not entirely sure of the identificationof rAuflr^ with dorycnium (*--v 'perhaps').

With this new data, we can rewrite the PS entry as follows:

KrtfluoK', r^jjr^ n. hot. dorycnium?

Ace. to Sergius, perhaps to be identified with Gr. TOSopuxvLOv. TO SopuxvtSLOv; Ar, cij^jji- o^jji\ Lat.DoT'ycniuw..

Sergius: r Wir f ta^ .cnaAur-f.! ^rf. loio.i Dorycniumwhich is perhaps > [BL Add 14,661 4h; cf. Kiihu XI:864,MeySob, no. 247],

The PS entry implies that there is no doubt about the identification ofrA±*Or^ with dorycnium. but Sergius. the earliest attested source for thisidentification, is unsure and subsequent botanists are also not convinced.The above entry thus expresses this doubt and furnishes the reader with thepertinent reference. The lexical equivalences to the supposed identificationare also referenced.

7. Conclusioji,

Even from the few examples discussed in this paper, we have seen tlmt ananalysis of the botanical lists provided by Sergius is certainly worthwhile.As the discussion of doryenium demonstrates, modern scholarship may haveto reassess its perception of the significance of Sergius" accomplishments inrespect of the later Hunayn, We have also seen that, when constructing alexicon of Syriac botanical terms, it is important that each occurrence is fullyreferenced so that care is taken to distingiiish between certain and suggestedidentifications. Thus entries I and J are clearly certain in the identificationof the plant. Entries H and K make it cleai' that the identification is Sergius'suggestion, but entry K goes one step further and informs the reader thatSergins had some doubts about his identification. All of these sample entriesprovide full references and quotes, so that the envisaged lexicon also servesas a concordance.

Prior to researching Syriac medical terminology, it is essential that reli-able text editions and coneordances are produced. Following this, it shouldnot be too difRcult to construct a lexicon of Syriac medical terminology,with reference to the equivalent Greek, Arabic and Latin terms.

[AS 3.2 (2005) 167-86]DOI 10.1177/1477835105059090

GHOOSING A TEXTUAL BASIS FOR THE NEW ENGLISHANNOTATED TRANSLATION OF THE SYRIAC BIBLE*

Bas ter Haar Romeny

Faculty of Theology, Leiden University

As various projects for rendering the 'Bible of Alexandria' into French,English, German, and Italian were starting np or in progress, peoplepointed out the need also to translate the 'Bible of Edessa' into one ofthe modern languages: it was time to start thinking of a new translationof the Peshitta into Enghsh.^ The Leiden Peshitta Institute respondedto these calls by setting up an editorial board and organizing a seminaron the policy and guidelines of such a translation in 1999. With theformal support of the International Organization for the Study of theOld Testament (IOSOT), this editorial board has started inviting col-laborators for what has become known as the New English AnnotatedTranslation of the Syriac Bible (or NEATSB).

*The author's research is supported by the Netherlands Organisation for Sci-entific Research (NWO). A shorter version of this paper was read at the secondNEATSB seminar, entitled 'The Bible of Edessa', Leiden, 2 August 2004.

^It is indeed a new translation, which is intended to replace George M. Lamsa,The Holy Bible from Ancient Eastern Manuscripts Containing the Old and NewTestament Translated from the Peshitta, the Authorized Bible of the Church of theEast (Philadelphia: A.J. Holman Company, 15th edn., 1967). This translation isassimilated to the Hebrew text in quite a few places, and is not based on a reliabletext of the Peshitta. The latter point also applies to Andrew Oliver's lesser knownA Translation of the Syriac Peshito Version of the Psahns of David (Boston: E.P.Dutton k Co., 1861; New York: James Pott, 1867). I am most grateful to Dr. CraigMorrison (Rome) and Dr. Ignacio Carbajosa (Madrid) for drawing my attention tothe latter version and sending me a photocopy of this work.

^A full survey of the discussions on the editorial policy and procedures can befound in Konrad Jenner et al, 'The New English Annotated Translation of theSyriac Bible (NEATSB): Retrospect and Prospect', PIC 23, in Aramaic Studies 2(2004), pp. 85-106.



Among the main issues discussed at several meetings and at thefirst NEATSB seminar mentioned above, was the question of which textshould serve as the basis of the translation. Participants in the 1999seminar agreed that the 'BTR text as published in Vetus Testamen-tum Syriace^ (i.e. the Leiden edition), should serve as the basic text,'though only in the perspective of a well-founded text-critical and text-historical evaluation'. This presentation discusses the background ofthe decision taken in 1999, as well as its implications and the problemsconnected with it. How this decision should be implemented is stilla matter of discussion, so this paper will also outline possibilities forits implementation. Samples taken from Genesis and Samuel, amongother books, illustrate the importance of an evaluation of variants andof including a very concise critical apparatus in translation.

1. Three Options

The first discussions on the basic text for the translation took placein 1994, in the framework of deliberations on the possibility of a 'Stu-dent Edition of the Peshitta'. The late Jonas Greenfield had asked thePeshitta Institute several times to set up such an edition, which in hisopinion should present a vocalized Syriac text as well as an Englishtranslation. In this way one could meet the needs of students who wereinterested in biblical studies, but lacked sufficient knowledge of Syriac.In these first discussions on the Syriac text of the Student Edition, onwhich also the English translation would be based, three main optionswere evaluated and arranged in the following order of preference:

1. An edition of the text of the preferred manuscript for individualbiblical books. The relevant manuscripts in this option seemedlimited to 5bl, 5phl, 8bl, and 9al. Obvious clerical errors, it wasconceded, should be corrected.

2. The establishment of a critical text. The aim of this option shouldbe to establish the best and most original text on the basis oftext-critical and text-historical criteria.

3. The reproduction of the text presented in the Leiden edition.Ghoosing this option would result in a text which is close to, butnot identical with, Koster's BTR, the average text of the seventh-

T E R HAAR ROMENY Choosing a Textual Basis 169

and eighth-century manuscripts, representing the second phase inthe development of the text.^

Whichever of these alternatives was going to be chosen, it was clearfrom the outset that an edition of the text should be completed witha deliberate selection of variant readings from the critical apparatus ofthe Leiden edition and from the Syriac Fathers. The options may nowbe reviewed in more detail.

1.1. Choosing a Preferred ManuscriptWhat exactly do we mean when we speak of a 'preferred manuscript' ?This concept applies to cases in textual criticism where it is impossibleto determine the stemma of the tradition, and thus to find or recon-struct one manuscript as the archetype of the known tradition. In suchcases one has to establish the originality of readings on a variant-by-variant basis, using all available witnesses. One can determine a 'best'or 'preferred' manuscript, however. This is the one that, wherever achoice between variants is possible on the basis of intrinsic arguments,gives us the best text in most instances. In cases where we have nointrinsic arguments for a choice, where our judgment is helpless, weare allowed to follow this manuscript, simply because we may assumethat as it leads us in the right direction in, say, 60% of the determi-nate cases, it will also do so in those that cannot be determined. Theprobability of adopting the correct text is thereby somewhat improved.

Two things are important to keep in mind: (a) the establishment ofthe status of 'preferred manuscript' is done on the basis of a text-criticalevaluation of all variants; (b) the status of preferred manuscript is usedonly to make a choice in cases where there are no intrinsic argumentsto do so. Its authority can thus be invoked only in a very narrowsphere. Prom these two points it follows that once all variants have beenevaluated—a prerequisite for establishing a preferred manuscript—, itis the text that is the result of this evaluation that should be pubhshedin all determinate cases. Wherever a reasoned choice is possible, weneed to follow this choice instead of the preferred manuscript: afterall, why should we also follow this manuscript in the 40% of readingswhere we have reason to assume that it leads us astray? The policy of

^Marinus D. Koster, The PeshiUa of Exodus: The Development of its Text in theCourse of Fifteen Centuries (Studia Semitica Neerlandica, 19; Assen: Van Gorcum,1977), passim but especially pp. 1-5.


following a preferred manuscript in all cases cannot be defended fromthe perspective of text-critical method.

We may ask, then, why it was proposed to follow a 'preferred manu-script'. There are three possible explanations. First, it is sometimes de-cided to follow a single manuscript as an easy way out: an evaluationof one or two chapters of a biblical book may be seen as just enoughto prefer the manuscript that seems best in those chapters. Worse stillwould be taking a few very clear examples, and decide on that basis,instead of evaluating all readings. We can safely assume that this wasnot the strategy intended.

Second, a misunderstanding of what is decisive in textual criticismmay have played a role. A sample of such misunderstanding can befound in Marinus Koster's recent review of Michael Weitzman's intro-duction to the Peshitta.^ Here Koster suggests that it is possible toestablish the 'track record', or relative reliability, of a manuscript be-fore the evaluation of intrinsic arguments takes place. He suggests thatstudies of the transmission history connected with the Leiden edition,such as his own work on Exodus, should come first. It is in these stud-ies that the results of collating all extant manuscripts are summed up.This, Koster argues, 'guarantees a measure of objectivity that nevercan be achieved by solely proffering proof texts'. At this point, Kosterseems to assume that Weitzman's method, which is the standard ap-proach in textual criticism, consists of selecting a number of readings,on the basis of which conclusions are drawn. The reality is that thebasis of Weitzman's approach is 'to choose between the rival read-ings in every variant passage, i.e. at every point in the text where themanuscripts disagree'.^ The fact that Weitzman gives many samplesin his book does not mean that he has drawn his conclusions only onthe basis of these samples.

Another problem in what Koster writes, is the idea that summingup collation results sets the 'track record' of a manuscript. The reality

^Marinus D. Koster, 'A New Introduction to the Peshitta of the Old Testament',^ 5 1 (2003), pp. 211-46 (223-24), reviewing Michael P. Weitzman, The SyriacVersion of the Old Testament: An Introduction (University of Cambridge OrientalPublications, 56; Cambridge: Cambridge University Press, 1999). Though Kosternow no longer defends a diplomatic edition (private communication), he is stillreasoning very much in terms of the position and quality of a manuscript as awhole, as this publication shows. Together with the third reason mentioned below,this forms the background to his original point of view (see note 8).

^Weitzman, The Syriac Version, p. 264.


is that one cannot determine the reliability of a manuscript just bysumming up objectively; one has to interpret what one finds. As A.E.Housman has it:®

When you invoke the authority of a MS against the exercise of the judg-ment, you are inciting the creature to rebel against the creator, and youare sapping the very ground on which you stand.

Indeed, it was Koster himself who established the position of 5bl forExodus on the basis of intrinsic arguments. In the famous kespd-semdcase he rightly argued that later tradition had adapted the text to makeit suit semantic developments in the Syriac language. A large part ofhis thesis is devoted to a full and painstaking discussion of this case andothers, by which Koster demonstrates in a convincing way that manyvariants can best be explained as attempts by later copyists to makethe text more readable.^ Koster has grouped the variants together ina number of categories, which hides what he is doing; but when onesees through this, it is clear that the decision on the status or 'trackrecord' of 5b 1 is taken here, where variants are weighed on the basisof intrinsic arguments and not just summed up. It was this work thatlaid the basis of our present understanding of the text history of thePeshitta.

The way Koster speaks about the establishment of the reliability ofa manuscript as a whole and as the first step of the text-critical process,as well as his claim to objectivity, would perhaps be more fitting underdifferent circumstances, that is, if it had been possible to reconstruct astemma of the tradition. However, the tradition of the Peshitta is toocomplicated for this (as is that of most other Bible versions), if onlybecause of the interference caused by the fact that copyists knew partsof the text by heart and by the fact that new biblical manuscripts wereoften collated against one or more muster copies.

A third reason behind the idea of following a 'preferred manuscript'may have been the point of view that the original text cannot or can-not yet be established, and that the text of any particular manuscriptshould be seen as a witness to a certain phase in the tradition of thetext. By taking 7al, one publishes a witness to the second (BTR)

^A.E. Housman, 'The Editing of Juvenal' (Preface of 1905), in Collected Poemsand Selected Prose (ed. by Christopher Ricks; London: Penguin Books, 1988), pp.395-402 (398).

Koster, The Peshitta of Exodus, passim; see especially pp. 55-114 for the dis-cussion of the readings of 5bl.


phase, by taking 5b 1 or 9al one gets a picture of the first, more an-cient phase.^ This sounds reasonable. However, the hidden assumptionhere is that the various phases show some kind of uniformity. The veryexistence of 9al, the text of which belongs to the first phase, alreadyshows that a clear chronological compartmentalization is impossible.What is more serious is the fact that even the first phase was not ho-mogeneous.^ During this phase, we see a variety of texts: 5b 1 is oftencloser to the Hebrew text in Genesis, but it also has some variants thatare clearly of a secondary nature, leading the text further away fromthe Hebrew.^" Ephrem's Genesis text is somewhere halfway between5bl and 7al, and Aphrahat is even closer to 7al (and thus to the later'Textus Receptus' [TR] or standard text). One could well argue thatthe period up to the sixth century was the period in which most textforms came into existence.

Thus if one adopts 5b 1 or 9al, one cannot claim that they arerepresentative of the text in any given period. Furthermore, neitherof them is the archetype of the known tradition, so the only correctreason for adopting them would be that intrinsic arguments showedthat in a majority of cases their readings represent a more originaltext. However, once this has been established on the basis of a studyof all variants, as has to be done, there is no reason to adopt thesemanuscripts as a whole, as I have a argued already. It may be truethat in the process of judging variants the Urtext cannot be establishedwith absolute certainty, but the fact that one can discern a developmentshows that one can at least get closer to it.

*This seems to have been Koster's original stance; see Koster, The Peshitta ofExodus, p. 197, where he pleads for a diplomatic edition of 5bl, to reflect the flrststage. However, Koster is still thinking in terms of manuscripts as representatives ofa stage, as is clear from his recent article ' "Translation or Transmission? That is theQuestion?" The Use of the Leiden O.T. Peshitta Edition', in Matthias Augustin andHermann Michael Niemann (eds.), 'Basel und BibeV: Collected Communications tothe XVIIth Congress of the International Organization for the Study of the OldTestament, Basel 2001 (Bern: Peter Lang, 2004), pp. 297-312 (300). As we haveseen above, Koster's tendency to speak in terms of manuscripts rather than readingsalso played a role in his initial choice for a diplomatic edition.

^Thus Bas ter Haar Romeny, 'Techniques of Translation and Transmission in theEarliest Text Forms of the Syriac Version of Genesis', in Piet B. Dirksen and Arievan der Kooij (eds.). The Peshitta as a Translation: Papers Read at the II PeshittaSymposium Held at Leiden 19-21 August 1993 (MPIL, 8; Leiden: E.J. Brill, 1995),pp. 177-85.

^°Cf. Koster's own description of the position of 5bl in his 'Translation or Trans-mission?', pp. 306-307.


An additional point should be mentioned here. It is important torecognize that manuscripts covering more than one book of the Bible donot necessarily present a uniform text. Thus the 5bl text of Leviticus,Numbers, and Deuteronomy is in a later hand than the Genesis and Ex-odus text. Copying obscures the visible difference between the two sec-tions. Thus it is perfectly possible that many of our larger manuscripts,and especially pandects such as 7al, 8al,^^ and 9al, combine texts ofdiflFerent origins, ages, and qualities. For 9al, it is interesting to notethat Weitzman already recognized that the number of unique agree-ments with the Hebrew is much higher in Kings and Jeremiah than inother books,^^ and that Hayman demonstrated that in terms of suchreadings, 9al is outdone by 7pj2 in Numbers.^^ Brock also noted the'very mixed character' of 9al in Isaiah.^^

1.2. A Critical TextThe arguments just brought forward against choosing a single manu-script are persuasive in favour of a critical text. An editor restores on aword-for-word basis the text thought closest to the original text of thetranslator or translators. Wherever there is good reason to assume thatnone of the extant witnesses represents the original text, conjecturalemendation should not be shunned. Arguments heard against this ap-proach, such as 'we do not know enough about the textual history' or'we do not know whether there was a single translation at the basis ofthe tradition' are no longer valid. It is true that there are still manythings that we do not know for certain, but some work has been done.In the case of the Peshitta, there is no longer any reason to assume

this MS and its composite nature, see Konrad D. Jenner, De perikopen-titels van de geillustreerde Syrische kanselbijbel van Parijs (MS Paris, Biblio-theque Nationale, Syriaque 341): Een vergelijkend onderzoek naar de oudsteSyrisehe perikopenstelsels (doctoral dissertation Leiden 1994; Leiden: UniversitaireDrukkerij, 1993).

'^Michael P. Weitzman, 'The Originality of Unique Readings in Peshitta MS9ar , in Piet B. Dirksen and Martin J. Mulder (eds.). The Peshitta: Its Early Textand History. Papers Read at the Peshitta Symposium held at Leiden 30-31 August1985 (MPIL, 4; Leiden: E.J. Brill, 1988), pp. 225-58; idem. The Syriac Version,p. 283.

^^A. Peter Hayman, review of Marinus D. Koster, The PeshiUa of Exodus, JSS25 (1980), pp. 263-70 (267).

^''Sebastian P. Brock, 'Text History and Text Division in Peshit;ta Isaiah', inDirksen and Mulder (eds.). The Peshitta: Its Early Text and History, pp. 49-80(52).


multiple origins, thus the textual critic has a clear target.^^ There isalso consensus about the idea that the original translation kept veryclose to a second-century Hebrew text similar (but not identical) tothe basis of MT. ^ Errors in transmission and internal Syriac consid-erations, such as the wish to make the text more readable, influencedthe development away from that original translation which followedthe Hebrew rather closely.

Considering that the editor of a text is a person who has beenimmersed in the material for a number of years and is thus supposedto know most about it, it is a reasonable expectation on the part ofthe reader of the text to flnd the editor's reconstruction of the originaltext printed in the edition. The editor should not leave the work tothe—sometimes casual—reader. This does not exclude the fact thatnew ideas come up or new material becomes available. It is a basicpoint that a text edition gives the best text according to the materialand ideas of the moment; for the rest, there is always the principle diesdiem docet. A critical text does not mean a definitive text.

1.3. Following the Leiden EditionThe third option is to adopt the text of the Leiden edition. It was thisoption that was chosen by the committee when it met in February 1995.The background to this decision was a practical one. The initial planfor a Student Edition as suggested by Jonas Greenfleld was cancelledfor practical reasons. The edition of a translation and of a one-volumetext edition were from now on to be considered separate issues. As theone-volume text edition and the final decision on the nature of its textwere to be postponed, it seemed logical to take one's departure fromthe best existing printed text. Konrad Jenner writes:'"'

this point, see especially Weitzman, The Syriac Version, pp. 263, 308-309.f., among others, Arie van der Kooij, Die alten Textzeugen des Jesajabuches:

Ein Beitrag zur Textgeschiehte des Alten Testaments (OBO, 35; Freiburg: Univer-sitatsverlag; Gottingen: Vandenhoeck & Ruprecht, 1981), pp. 297-98; Brock, 'TextHistory and Text Division', pp. 52-53; Anthony Gelston, The Peshitta of the TwelveProphets (Oxford: Glarendon Press, 1987), p. 118; Piet B. Dirksen, 'The Old Tes-tament Peshitta', in Martin J. Mulder and Herman Sysling (eds.), Mikra: Text,Translation, Reading and Interpretation of the Hebrew Bible in Ancient Judaismand Early Christianity (GRINT, 2.1; Assen: Van Gorcum; Philadelphia: FortressPress, 1988), pp. 255-97 (258-59); Marinus D. Koster, 'Peshitta Revisited: A Re-assessment of its Value as a Version', JSS 38 (1993), pp. 235-68 (254); Weitzman,The Syriac Version, pp. 15-67.

et al, 'NEATSB', p. 92.


For solely practical reasons the text and apparatus of the planned trans-lation should basically represent that of the BTR-text as published inVetus Testamentum Syriace (the Leiden edition in progress). . . . Theestablishment of a critical text would require too long and too thorougha study of the theological and commentary literature.

The NEATSB seminar held in February 1999 seemed to go a stepfurther, though. According to the minutes, the participants agreed onthe following statement:^^

The authority of the BTR-text as published in Vetus Testamentum Syr-iace is generally accepted. So, this text will serve as basis for translation,though only in the perspective of a well-founded text-critical and text-historical evaluation. The evaluation of textual variants is a matter ofcommon sense.

Talking about the authority of an edition is of course not the same asaccepting its text for practical reasons. Let us have a closer look atwhat the choice for the Leiden edition implies.

The original idea of the Leiden edition was to print the basic text,usually 7al, without any changes, 'except for the correction of obvi-ous clerical errors that do not make sense', as the 1972 General Prefacestates.^^ This kind of approach has been termed 'irrational': it assumesthat wherever a scribe made a mistake, he produced an impossible read-ing.^" Perhaps 'arbitrary' would be a better word. The goal seems tohave been to publish a useable text, not one that goes back to the origi-nal as far as possible, nor one that can be assumed to have corrected allthe mistakes in 7al: only the readings deemed impossible are corrected.All arguments brought forward to defend this approach are based onthe idea that a critical text should be definitive. ' I have already dis-cussed this idea; in my opinion, it is not a particularly common oruseful approach to textual criticism. It would in fact make any criticaledition a virtual impossibility. One should perhaps consider that theproject was a child of its time: when this approach was chosen, thereadings of Qumran texts such as 4QSam^ were not yet available tothe public, and their significance was yet to be understood. What was

'^Jenner et al, 'NEATSB', p. 98.^^Peshitta Institute Leiden, The Old Testament in Syriac according to the Pe-

shitta Version: Ceneral Preface (Leiden: E.J. Brill, 1972), p. viii.2°Cf. Housman, 'The Editing of Juvenal', p. 396.^ip.A.H. de Boer, 'Towards an Edition of the Syriac Version of the Old Testa-

ment', PIC 16, in VT 31 (1981), pp. 346-57 (355).


clear, however, was that some of the late nineteenth-century and earlytwentieth critics had gone too far in emending MT and reconstruct-ing its text on the basis of the Septuagint, as they lacked a properappreciation of the translator's contribution to the text.

After the publication of the first fascicles of the edition in 1972 and1973, ' it appeared that the size of the second apparatus would be toolarge, making the undertaking impossible for financial and practicalreasons. Piet de Boer, then general editor, decided to omit all variantreadings occurring only in manuscripts younger than the twelfth cen-tury, and to widen the scope for emendations in the basic text. Thefirst decision can be defended if it can be demonstrated that the latermanuscripts all go back to existing earlier manuscripts. This is indeedthe case, in the sense that there is a general impression, based on full orsample collations, that these manuscripts do not carry unknown vari-ants that cannot be explained as inner-Syriac corruptions or changes.Still, it has been decided to publish the variants of manuscripts up toand including the fifteenth century in a separate volume.

The second decision, the introduction of a larger number of emenda-tions, was also connected with the wish to make the apparatus leaner,'thus facilitating the use of the edition and also its printing'.^^ Themain rule for emendations should be seen in this light : ^

Emendations were made also in those cases where the reading of themanuscripts chosen as the basic text of the edition is not supported bytwo or more manuscripts from the material used up to and includingthe tenth century. The printed text in these cases is chosen on the basisof a definite majority of the manuscripts dated to the tenth century orearlier.

Thus the choice was made for something between a diplomatic editionand a majority text. As the group of manuscripts used to emend 7alincludes manuscripts from the ninth and tenth century, we cannot saythat the result is a pure BTR text according to Koster's definition. In

^ In 1972 fascicle 6 of part IV of the edition appeared, containing the Canticlesor Odes, Prayer of Manasseh, Apocryphal Psalms, Psalms of Solomon, Tobit, and1 (3) Esdras; in 1973 fascicle 3 of the same part appeared, with the Apocalypse ofBaruch and 4 Esdras. Earlier, in 1966, a sample edition containing Song of Songs,Tobit, and 4 Esdras had been published.

^^De Boer, 'Towards an Edition', p. 356; cf. also De Boer's Preface, in PeshittaInstitute Leiden, The Old Testament in Syriac according to the Peshitta Version,LL Preface; Genesis-Exodus (Leiden: E.J. Brill, 1977), p. viii.

^' Boer, 'Preface', p. viii.


some cases, this approach may have drawn the text towards a laterTR, non-BTR, text.

The real picture is even more complicated. In addition to the mainrule, a number of exceptions were formulated.^^ First, it is stated, therule cannot be followed when there are only 'few' ancient manuscriptsavailable. Second, editors are allowed to retain the reading of 7al, oreven a reading not found in ancient manuscripts, when they prefer thisreading or, as De Boer said elsewhere, 'if they have special reasons todo so'.^^ Thus, depending on the choices of the editor, the text of theLeiden edition can be a mix of a diplomatic edition, a majority text,and a critical text. This can be illustrated by a number of examples.

In the edition of Genesis and Exodus, ' it was often the majorityprinciple that was decisive. Thus in Gen. 8.17 and Exod. 2.13, amongother places, original Peshitta readings found in 7al but supportedonly by 5bl were relegated to the apparatus. In Gen 8.17 this concernsonly the letter waw, the conjunction 'and'. Popular belief has it thatsuch instances are insignificant. However, in this case it is clear that theconjunction cannot have been written by someone who was translatingthe Hebrew text. The conjunction most likely suggests that iw-i A^ 'allflesh' is to be taken as the first part of an enumeration: 'every livingthing that is with you: of all flesh and birds and animals and everycreeping thing that creeps on the earth'. In the Hebrew text, however,every one of the categories following 103 'flesh', starting from 'birds',is introduced by the preposition bet, used here to specify items whichtogether constitute the collective 'all flesh'. ^ Thus the 'birds' comefirst in the enumeration, not 'all flesh': 'every living thing that is withyou of all flesh: birds and animals and every creeping thing that creepson the earth'. The translator must have realized this, but as he did nothave the same means of indicating specification, this remained unclearto some of his readers. The situation in Exod. 2.13 is simpler. Herethere is a small gap between the Hebrew expression meaning 'and he

Boer, 'Preface', p. viiiBoer, 'Towards an Edition', p. 356.

Institute Leiden, The Old Testament in Syriac, Ll. The Genesis edi-tion has been prepared by the editorial staff of the Institute and is 'based onmaterial collected and studied by T. Jansma' (p. xv); Exodus has been preparedby M.D. Koster.

^*For the preposition 3 in the sense of 'existing in' after collective words, seeWilhelm Gesenius and Emil Kautzsch Hebrew Crammar (2nd English edn, revisedin accordance with the 28th German edn (1909) by A.E. Cowley; Oxford: GlarendonPress, 1910), § 119i.

went out the next day' and the one meaning 'and behold two Hebrewmen fighting', into which the reader has to supply the idea of 'and hesaw'. The translator retained the conciseness of the Hebrew, but latertradition added the linking idea, making explicit what was imphcit.

The original readings just discussed can be found in the apparatusbecause the main rule says that a 7al reading not supported by atleast two other pre-eleventh-century manuscripts should be emended.Yet it appears that in Gen. 3.8, the support of 7h5 and 911 was notconsidered enough to keep 7ars singular form »jt ^p<ro 'he hid himself,which follows MT, in the main text. The form is the more difficultone, 'Adam and his wife' being the subject, and very probably reflectsthe original text. The reason why this reading was discarded is theunwritten rule that lectionaries do not count. These are liturgical textsthat are often adapted to their use in service or to the New Testament,or contain other unexpected changes. Personally I would say that thefact that they contain all sorts of unexpected variant readings does notalter the fact that elsewhere they might be the only witness to havepreserved the truth.^^

The arbitrariness that creeps into the application of the majorityprinciple can further be illustrated by the following two examples. InGen. 3.24, the reading r<itt..Tta\ AU.TM ^ 'at the east of Paradise',which follows the structure of MT ]yb DipQ more closely than the variantreading rioa^.Tta tui.i^n ^ , just made it into the main text: it issupported by 7al and two pre-eleventh-century manuscripts (7h5 and7k4). Here the rule leads to a correct result. In Gen. 4.10, however,the main text sports a reading of a clearly secondary nature, makingexplicit the subject and indirect object of the verb rtoo'* 'to say'. Theoriginal shorter reading found in 7al was discarded in spite of thesupport of 7k4, the first hand of 7h4, as well as a lectionary. In thiscase the fact that 7h4 was corrected (in fact only partly) towards themajority reading seems to have ruled this witness out of court.

The more critical approach is exemplified by Sebastian Brock's Isa-iah edition.^" Thus 7ars r^+a^» 'little' after r<iA\n 'and a child' in Is.

fact that 1011 has preserved a number of original readings for Exodus,independently of 5bl, had already been discovered by Koster {The Peshitta ofExodus, pp. 99-101, 186, 506-507), and can be confirmed for Genesis. In Genesis,13/913 appears to be based directly on 5bl—which means that if we had not had5bl, 13/913 would have been our only key to the original text of P Gen. 24.

''"Sebastian P. Brock (ed.). The Old Testament in Syriac according to the PeshittaVersion, III.l. Isaiah (Leiden: E.J. Brill, 1987).

10.19, though supported by two other MSS datable before the eleventhcentury, is not adopted. In the Genesis and Exodus edition, the rea-son why would be clear: one of the two supporting manuscripts is alectionary. For Brock it seems that something else also plays a role: heconsiders the 7al reading secondary; a small note identifies the plus asan adaptation to the text of Is. 11.6. In this case one could still say thatthere is at least a balance among the pre-eleventh-century MSS: it is6h3, 8al, and 9al versus 6h5, 7al, and 915. In Is. 4.4 this is not the case.Here the reading r^^.i. n. 'burning, fiery' is adopted, which among thepre-eleventh-century manuscripts finds support only in 912, lOdl, 1012,and a later hand in 8al (that is, two West Syrian lectionaries and tworepresentatives of TR). The discarded reading, r " !- ! 'heavy, hon-ourable', finds support in 6h3, 6h5, 7al, and 8al*—the most ancientwitnesses available in this case, in other words, the BTR text. Brock'schoice is right, however: one does indeed expect •^>^t.n. as a trans-lation for Hebrew "I:JD 'burning' rather than t<i\^^.ti^.^^ Both words,as adjectives to the word 'spirit', make sense and the difference, a dotunder or over the letter, is very small. We may therefore assume thatthe variant arose from a simple mistake.

Especially interesting are a few cases where Brock retains the 7altext even though it is not supported by at least two other manuscripts.Thus in Is. 65.12 5phl and 7al read ^• lo in conformity with MT butagainst the rest of the Syriac tradition. All other manuscripts, including9al, add the object: .^o^^in.^^ Had the same variant occurred inGenesis or Exodus, I am sure we would have found the secondaryreading in our main text. Thus it depends very much on the editorwhat kind of text we have. Yet even when we have an editor thatapplied thought to his choices instead of arithmetic, it would seemthat in practice an original reading only reached the main text if itwas supported by 7al or if a secondary 7al reading could easily bediscarded as a mistake.

2. A Compromise

From the point of view of text-critical method, it goes without sayingthat one would be hard pressed to defend the approach of the Leiden

form in MT is a pi'el infinitive: hence "IJ3D m~l is usually translated as'spirit of burning' or, if connected with a different verb "1^3, 'spirit of wiping out'.The Syriac translator might have read a qal participle, but it is not necessary toassume this. For the verb .iriL. as a translation of "IJJD, see for example Exod. 22.5.

also Brock, 'Text History and Text Division', p. 54.


edition.^^ De Boer himself conceded that 'it is difiicult to approve ofthe introduction of emendations in the basic text'.^^ One can only un-derstand De Boer's decisions if one considers his aim. The text shouldbe seen as a point of reference: it should be common enough to guaran-tee a concise apparatus. The main text as such has no status; togetherwith the apparatuses, it forms a do-it-yourself kit. As De Boer

The text printed in this edition—it must be stated expressis verbis—ought to be used in exegetical and textual study together with the ap-paratuses.

The reader cannot just quote the text, he should first go over theapparatus and do the work of the textual critic. De Boer himself wasvery unhappy with the way the main text of the edition came to bequoted as 'the Peshitta' without further ado. He admitted to have'underestimated the force of the printed text even among scholars',and says that his successors would not be to blame if they came upwith 'a system that that gives less occasion to misunderstanding'.^^

All this goes to show that the person who set the rules for theedition would have been horrified to hear that the 'authority of theBTR text as published in the Leiden edition' is 'generally accepted'.The edition is the most reliable source of information on the variousreadings available in the earlier tradition, but it is up to the reader tosift through this material and make the necessary text-critical choicesbefore using the text. As one of the main goals of the English trans-lation is to be a help to those whose knowledge of Syriac is wanting,we cannot convey again the task to be performed by the critic. Manyof our readers will want to know what we think the Peshitta is, as itis not reasonable to expect them to able to exercise judgement, noteven on the basis of a translated apparatus. I think I am not tryingto change the decision of 1999 here; I would just put more stress onthe second part of it, which states that the edition is the basis of thetranslation 'only in the perspective of a well-founded text-critical and

•'• It is worth remembering here once more that in 1972 Konrad Jenner andMaarten van Vliet, then collaborators at the Institute, had already argued in favourof a critical text. Unfortunately Piet de Boer and Wim Baars were not ready forthe idea yet.

Boer, 'Towards an Edition', p. 356.Boer, 'Preface', p. viiiBoer, 'Towards an Edition', pp. 356-57.


text-historical evaluation'.^'' I might add that the close reading neces-sary for a translation creates outstanding conditions for text-criticalevaluation. The question which should be dealt with now is how ourchoices are to be presented, and how much of the apparatus should bepresented in translation.

3. Practical Application

The task presented to us in a way by De Boer himself is clear: find asystem that gives less occasion to misunderstanding than the edition.During the first stage of the translation process one needs to create a'virtual critical text': one that is not printed in Syriac, but functionsas the basis of the translation. This is not a new thing: nearly all mod-ern translations of the Hebrew Bible are based on a virtual criticaltext. They do not just translate BHS, but adopt more or less readingsfrom Qumran manuscripts, the Septuagint, or other witnesses, wher-ever they feel that the Codex Leningradensis cannot or should not befollowed. The remaining issue is that of the presentation of the ren-dering of the virtual critical text: should it be the running text of thetranslation, and should rejected readings from the Leiden edition berelegated to the notes, as for instance in NRSV? Or should the runningtext have a one-to-one relationship with the main text of the Leidenedition, and should the notes point the reader to what we think is theoriginal reading?

I must say that I have always interpreted the 1999 decision in thelatter sense: the text of the translation is that of the Leiden edition,and the footnotes will give a rendering of a selection of the apparatus.This is not self-evident, however, as can be seen from Alison Salvesen'ssample text of 1 Sam. 2.1 as printed in the NEATSB presentation inAramaic Studies.^^ The Leiden edition with its apparatuses reads asfollows : ^

December 1995, Michael P. Weitzman was consulted about the StudentEdition and NEATSB. Though he could approve of the text of the Leiden edition, henevertheless argued for a text which would be more in line with his notion of theoriginal Peshitta, which he was developing at that moment in his book The SyriacVersion (see especially Chapter 6).

^^Jenner et al., 'NEATSB', pp. 103-106.^"P.A.H. de Boer (ed.), 'Samuel', in Peshitta Institute Leiden, The Old Testament

in Syriac according to the Peshitta Version, 11.2. Judges-Samuel (Leiden: E.J. Brill,1978).

t -1] ,aArin 7al

1 -] .cnit<= 7al 7k3' 1012 -» | .—-.V V..] s. sey 10k4

The full text of the verse in Salvesen's sample, together with her notes,reads as follows:

2.1 Hannah prayed and said, 'My heart is strong in the Lord, and myhorn has been exalted [in my God]. My mouth has been opened againstmy enemies, because you have made me rejoice in your salvation.

1 in my God] This phrase appears in LXX, MS 7al and a few other P MSS(see apparatus to the edition), but is absent in the rest of the P tradition. MTand 4QSam^ have 'in the Lord'. — is strong] Apparently this represents HebrewT' i), 'exult', the meaning of which was probably unknown to the translator of P,hence the resort to a verb with positive connotations appropriate to the context.— you have made me rejoice] The corresponding Hebrew is 'nnDB, 'I have rejoiced'.Perhaps the translator of P read the form as a Piel with first person suffix, ']nnDB.

The reading 'in my God' probably belongs to the original text ofthe Peshitta. The word may have dropped out because of parablepsis:the preceding word also ends in a yod.'^^ Still, Piet de Boer, editor ofPeshitta Samuel, decided to remove this 7al reading from the runningtext because, according to the main rule (with its unwritten supple-ment regarding lectionaries), the support of the lectionary 1012, thesecond hand of 7k3, and some later MSS is insufhcient. Salvesen keptthe reading in the running text, albeit between square brackets, andadded a factual footnote.

I would agree with Salvesen to have this reading in the running text;and I would also put other readings which ean be assumed to have be-longed to the original Peshitta there, including Salvesen's conjecturesfor 1 Sam. 2.8 and 9. Both are corrections of rather obvious mistakesthat occurred within the Syriac tradition. The next question, then, is

fact that MT reads 'in the Lord' and LXX 'in my God' could be used toargue that 7al added the phrase on the basis of the Septuagint. As it does notbelong to the translation style of the Peshitta to omit words in this way, I wouldsay it is slightly more probable that the Septuagint and the original Peshitta hada Vorlage Tl' S. A defective Vorlage without Tl' KD or mrrn is a theoretical thirdpossibility.


how we should draw attention to the fact that we are not followingthe text of the edition. Square brackets or hooks in the running text isone possibility. Personally, I would plead for a separate apparatus fortextual issues in the Syriac tradition. This is against the 1999 decisionto have textual and informative notes in one apparatus. Yet I think thenature of the present Peshitta edition and the choice of rendering a vir-tual critical text necessitate special attention to textual matters. Witha separate line, it will always be immediately clear that the translatorhas chosen a reading different from the edition, and anyone who wantsto know the translation of the edition will not need to excavate thisinformation from a group of notes on all kinds of issues. This would bemy text of the same sample:

2.1 Hannah prayed and said, 'My heart is strong in the Lord, and myhorn has been exalted in my God. My mouth has been opened againstmy enemies, because you have made me rejoice in your salvation.

1 in my God 7al P*] > Ed

1 in my God] This phrsise, dropped out through parablepsis in most of the Ptradition, reflects Hebrew 'n'^S, as does LXX. MT and 4QSam'^ have 'in the Lord'.— is strong] Apparently this represents Hebrew \'7V, 'exult', the meaning of whichwas probably unknown to the translator of P, hence the resort to a verb withpositive connotations appropriate to the context. — you have made me rejoice]The corresponding Hebrew is 'nriDB, 'I have rejoiced'. Perhaps the translator of Pread the form as a Piel with first person suffix, 'J

The textual apparatus should be as concise as possible. In general, itwill not be necessary, I think, to mention supporting manuscripts: usersof the translation who do not know Syriac or the manuscript traditionof the Peshitta will not need these details, others can easily find themin the edition. I have mentioned 7al here, because this is one of thecases where the edition goes against 7al. This is a matter that will needexplanation in the introduction. The support for a P* reading in oneof the Fathers needs to be noted, though, as this is new information,not yet available in the edition. For the same reason a somewhat longernote may be necessary in cases of conjectural emendations, as here thetranslator is introducing a reading without manuscript support.

Another practical issue concerns the extent of textual informationon the Syriac tradition that is to be covered in the notes. The 1999decision was that the annotations should provide information about'significant variant readings in the Peshitta manuscripts'. This leaves


some room for interpretation. In the situation sketched above, it is al-ways immediately clear which reading is most original according to thetranslator, which reading is found in the main text of the edition, andwhich reading is found in 7al if the edition does not follow this manu-script. In order to know what else we need, we should perhaps ask firstwhat exactly a significant variant is. Should we think of each variantthat is meaningful and translatable, even if it is a clearly secondaryreading? I think it will be more useful to think of significance in termsof text history. One could say that so far, we have covered the sup-posedly original reading and the BTR reading. It may be very helpful,especially for those working on later commentaries, to have also thereading of the later standard text, dubbed TR (Textus Receptus) byKoster. As the difference between BTR and TR is relatively small, thisshould not give too many extra readings; still, it means that the textform which was most widespread is also accessible through NEATSB.

In order to get an idea of the practical results of this approach, Ihave looked once more at the two sample texts from Samuel and at tenchapters of Genesis for which 5bl is extant. For the two Samuel textswe do not have the witness of 9al or any other unique carrier. Theresult is that I have only one more variant to add to the 7al readingthat is to be retained and the two conjectures. In 1 Sam. 1.9 thereis a TR reading to be reported, also supported by 6hl. In fact, I aminclined to adopt this reading in the running text: the Leiden editionhas a plus vis-a-vis MT that serves to place Hannah in the Temple, inline with the sequel; 6hl and TR have a shorter text, comparable toMT. The wording of the plus, t^.'tsD.i nniV.-i\ A>n\nnn 'and she wentup to the house of the Lord' is based on 1 Sam. 1.7. The necessity ofadding this kind of phrase is highlighted by the fact that the Old Greekalso has a plus here. It reads xoCi xaxecTTT) EVWULOV xupLou, based onV. 26 of the same chapter. Graig Morrison rightly concluded that thedifference in wording excludes the possibilities of infiuence from theSeptuagint on the Peshitta or an identical Hebrew Vorlage for the twoversions.^^ I would add that whereas the plus in the Septuagint goesback to its original text, being either a refiection of a longer Vorlage oran addition of the translator himself, the plus in the Peshitta traditioncan best be explained as a secondary development: the shorter text ismore difficult and in contrast to the Greek tradition, where one should

'' Craig E. Morrison, The Character of the Syriac Version of the First Book ofSamuel (MPIL, 11; Leiden: E.J. Brill, 2001), p. 130.

reckon with the possibility of revision after a Hebrew text, there is noobvious explanation for an omission.

For Gen. 8-9 I found 22 instances where I would not follow theedition. In 20 of these I would follow 5bl (either or not supported byother manuscripts),'*^ and in 2 cases a reading found only in Ephrem,that is, in Gen 9.6 and 9.22. In the former case I would read yA^••ui i<m\t< 'in the image of God he made' following the Hebrew,''^instead of .tiLnji. t<m\r< yA^ 'in the image of God he was made'. Atthe end of Gen. 9.22 I would add r<£oaxa 'outside' (or 'in the street', asEphrem interprets it),^'' a usual translation for Hebrew |"in3 'outside',omitted in all Syriac biblical MSS. Note that in these two and twofurther cases, ^ I do not follow 5bl. The 5bl readings in such casesare not listed in the translated apparatus, for the fact that a readingoccurs in 5b 1 is in itself no reason to assume it is more ancient. Finally,there is one reading specific to TR to be reported (Gen. 8.7). As weare dealing with a translation, I should add that by no means all ofthese readings will be noted. Only 12 of the cases were I do not followthe edition appear to be translatable.^^

4. Conclusion

Discussions about the basic text of the New English Annotated Trans-lation of the Syriac Bible started with three options. In 1995 and 1999,those in charge of the project were persuaded to follow the existingLeiden edition for practical reasons. At the same time, however, itwas stipulated that that this had to be done 'in the perspective of awell-founded text-critical and text-historical evaluation'. I have arguedthat the nature of the text of the edition forces us to stress especiallythe latter part of the decision. In line with how its first General Editor

n. 8.1, 5, 7, 9, 10, 12, 17 (three instanees), 18, 19, 20, 21; 9.10 (two instances),11, 15 (first variant), 16, 21, 22. One of the readings in Gen. 8.17 is supported by7al; the case of 8.7 by Eusebius of Emesa and that of 8.20 by Ephrem (in 9.21,however, Ephrem supports the text of the edition). In 9.5 (second variant) I wouldfollow the edition and 5bl against 7al.

''^R.M. Tonneau (ed.), Ephraem Syrus in Genesim et in Exodum commentarii(CSCO 152/Syr. 71; Louvain: Secretariat du CorpusSCO, 1955), p. 63 1. 3.

''''Tonneau (ed.). In Genesim et in Exodum,, p. 63 1. 13; for the interpretation, cf.Ephrem's comment on the same page.

^^Gen. 9.5, 15 (second variant)."''Gen. 8.1, 7, 10, 12, 17 (second variant), 18, 19, 20, 21; 9.6, 21, 22 (Ephrem).


intended the edition to be consulted, users—and thus also translators—should first do the job of the textual critic. The running text of thetranslation should therefore refiect a virtual critical text established onthe basis of the information present in the apparatus of the edition andour present knowledge of the Syriac Fathers. In view of the readershipof the translation, it seems best to keep the extent of what one couldcall the 'critical apparatus in translation' rather limited. It should givethe reading of the edition, if discarded by the translator, the reading of7al, if discarded by the Leiden edition, as well as the specific readingsof the later standard text (TR). If this apparatus is indeed presentedseparately from the other notes, the reader will always have a clearpicture of whether there are textual issues on the page and what thetext of P*, 7al, Ed and TR is.

The policy regarding the textual basis of the New English AnnotatedTranslation of the Syriac Bible advocated in this paper was acceptedby the General Editors of the translation in their meeting of 3 August2004.

[AS 3.1 (2005) 187-201]DOI: 10.1177/1477835105059091

DETERMINATION IN TARGUM JONATHAN TO SAMUEL*

Renaud J. Kuty

Leiden University

1. Introduction

Determination is often claimed not to function properly anymore inthe Aramaic dialect of TgOnq and TgJon. Scholarly views may differin detail but most agree that in that Aramaic dialect the status em-phaticus (the determinate form of the noun) has lost some or all ofits determining force, with the result that its uses and functions havebecome difficult to distinguish from those of the status absolutus (theindeterminate form of the noun).^ As a result of this process, both thest.emph. and st.abs. have come to compete for the status of unmarked

*The present article is an elaboration of a paper presented to the 4th Congressof the International Organization for Targumic Studies, held on July 29-30 2004 atLeiden University. A more detailed treatment of the subject will be presented inmy doctoral dissertation Studies in the Syntax of Targum Jonathan to Samuel, cur-rently in progress under the supervision of Prof. T. Muraoka at Leiden University.This dissertation is a part of a larger scientific project, 'Origin and History of Tar-gum Jonathan to the Prophets: Towards a Critical Edition', financed by the DutchOrganization for Scientific Research and headed by Dr J.W. Wesselius (KampenTheological University). The present study is based on the text of Samuel, whichis the corpus of my doctoral dissertation. References to the text of Samuel followthe format e.g. 11.4.5, which stands for 2 Sam. 4:5. The following abbreviations areused: st.emph. = emphatic state; st.abs. = absolute state; BH = Biblical Hebrew;TgOnq = Targum Onqelos; TgJon = Targum Jonathan; TgJon Sam. = TargumJonathan to Samuel. Finally, I wish to thank Prof. T. Muraoka and Mr MartinBaasten who read a draft of this paper prior to its publication and made manyhelpful comments.

^Cf. G. Dalman, Grammatik des jiidisch-palastinischen Aramaisch (Leipzig: J.C.Hinrichs Verlag, 1905), p. 188; M.Z. Kaddari, 'Studies in the Syntax of TargumOnqelos' (in Hebrew), Tarbiz 32 (1963), pp. 232-41; E.Y. Kutscher, 'Aramaic',in EncJud, 3 (1971), p. 268; A. Tai, The Language of the Targum of the Former



form of the noun, and the st.emph. is usually thought to have oustedthe st.abs. from that function.

The question of determination is an important argument in theongoing scholarly discussion on the origin and development of TgOnqand TgJon. Crucial though this question may be, however, as yet itdoes not appear to have been thoroughly investigated. It is the purposeof the present article to contribute to filling this lacuna. On the basis ofthe standard published edition of TgJon,^ the text of TgJon Sam. hasbeen subjected to a close scrutiny, so as to provide a detailed accountof the working of determination in TgJon Sam. and assess its allegedstate of collapse.

In what follows an attempt will be made to show that the assump-tion that determination in TgJon Sam. is not working properly any-more is largely unfounded. In order to define our terms, a distinctionwill be made between formal determination and semantic determina-tion, the former referring to form, the latter to meaning.^ A noun willbe considered semantically determinate when it fulfills at least one oftwo conditions: it is either cognitively available or used generically.

1. A noun is cognitively available when the entity it refers to isconsidered to be in some way available to the addressee, e.g.because:

Prophets and its Position within the Aramaic Dialects (in Hebrew; Tel Aviv: TelAviv University, 1975), p. 85; S.E. Fcissberg, A Grammar of the Palestinian Tar-gum, Fragments from, the Gairo Genizah (Atlanta: Scholars Press, 1991), p. 136; C.Miiller-Kessler, 'The Earliest Evidence for Targum Onqelos from Babylonia andthe Question of its Dialect and Origin', JAB 3 (2001), pp. 181-98 (188). As yet,the most detailed treatment of the matter is Kaddari, 'Studies'. Finally, Kutscher'sstatement in particular is noteworthy, cis he seems to have revised diametricallythe view he held one decade before on that very question (cf. E.Y. Kutscher, 'TheLanguage of the "Genesis Apocryphon". A Preliminary Study', in C. Rabin, Y.Yadin (eds.). Aspects of the Dead Sea Scrolls (ScrHie, 4; Jerusalem: Magnes Press,1958), p.lO, n. 42).

^Ms. Or 2210 (British Museum), as edited by A. Sperber, The Bible in Aramaic(Leiden: Brill, 1959-73). Though Sperber's methodology is to a certain extent ques-tionable, his edition of TgJon Sam. is by far the most reliable we have until now(cf. A. Houtman, 'Planning a New Targum Edition: Look before you Leap', JAB2 (2000), pp. 213-31).

^Thus in n.11.2 snns Rtm 'and he saw a woman', the noun Rnn« 'woman' willbe considered semantically indeterminate (for reasons to be discussed below) butformally determinate (i.e. it occurs in the st.emph., the primary way determinationis expressed in Classical Aramaic).

KUTY Determination in Targum Jonathan to Samuel 189

• It has been mentioned previously in the text (anaphora),e.g. 1.25.5 'and David sent ten young men, and David saidto the young men';

• It refers to an entity thought to be unique in the addressee'sknowledge of the world, e.g. 1.20.24 'the moon'; 1.13.7 'theJordan';

• It refers to an entity otherwise well-known to him, e.g. 1.19.22'and he came to the great cistern that is in Secu'',

2. Nouns used generically refer to a whole class or species. They canoccur either in the plural (e.g. 1.28.1 'the Philistines') or in thesingular (1.2.8 'the poor, the needy').

The position adopted in this study can be outlined as follows:

1. Determination in TgJon Sam. does function according to a largelyconsistent hnguistic system.

2. The apparent irregularity of use of the st.emph. and st.abs. is aconsequence of the fact that the linguistic system governing theiruse actually consists of two distinct (sub)systems:

(a) A linguistic system in which the distinction between st.emph.and st.abs. is fully maintained and carried out,

(b) A linguistic system in which the distinction between the twois neutralized to the advantage of the st.emph., that by andlarge proves to exert both functions.

3. Though these two systems are coexistent in the linguistic systemof the Aramaic of TgJon Sam., they are not intermingled. Inother words, they are not interchangeable, and a close scrutinyof our corpus reveals that each of them is used under distinctcircumstances.''

As a result of the above, the state of affairs witnessed in the Ara-maic of TgJon Sam. is twofold. In certain clearly defined circumstancesst.emph. and st.abs. are used properly, i.e. in agreement with the clas-

• The idea that two different systems of determination are co-existent in the Ara-maic of TgJon is consonant with Kaddari's views on the working of determinationin TgOnq (cf. Kaddari, 'Studies', pp. 235-41). In contrast to Kaddari's views, how-ever, the present study suggests that the use of these two systems is not arbitrary,but rather follows definite rules.


sical norm set by Old and Imperial Aramaic;^ in other, clearly definedcircumstances the st.emph. is the usual form, with the exception ofcertain specific areas of grammar where the st.abs. has maintained itssway. The foremost conclusion that can be drawn from this bipartitedistribution is that the Aramaic of TgJon Sam. displays a clear pref-erence for the st.emph., allowing it to encroach on the domains thatwere traditionally the prerogative of the st.abs.

In the conclusion of this article an attempt will be made to accountfor this peculiar state of affairs. In the section below the uses of thest.emph. and st.abs. will be discussed in some detail.

2. The expression of determination in TgJon Sam.

Probably one of the most interesting results produced by the presentstudy is the fact that determination does not work in the same way inthe singular and in the plural. Essentially, with regard to the expressionof determination our analysis of TgJon Sam. allows us to draw a sharpline between singular and plural nouns.

2.1 Nouns in the PluralA close scrutiny of our corpus demonstrates that in the plural the clas-sical distinction between st.emph. and st.abs. is systematically main-tained: when semantically determinate the noun occurs in the st.emph.,when semantically indeterminate it occurs in the st.abs., and this ap-plies equally to both masculine and feminine nouns. Countless exam-ples illustrate this state of affairs:^ 1.2.1 ]"n3:i yoi 'signs and mightydeeds'; 1.6.7 ]prD p^^\ 'milk-cows'; 1.14.48 fisn 'armies'; 1.16.21 ]'3I Va]'a bearer of weapons'; 1.19.4 ppn ]'Q:ns 'good words'; 11.3.2 pi 'sons';11.4.11 ]'JJ'O-l I'-Q] 'wicked men'; II.6.5 jni'D "IJX ^D 'all (manner ofinstruments made of) fir wood'; II.7.6 jii'TDl I'JDODD 'in tents and in

^Following Fitzmyer's widely accepted classification of the Aramaic dialects(J.A. Fitzmyer, 'The Phases of the Aramaic Language', in J.A. Fitzmyer, A Wan-dering Aramean: Collected Aramaic Essays (Missoula, MT: Scholars Press, 1979)pp. 57-84).

^Considering that in the Aramaic of TgJon Sam. the st.emph. tends to prevailover the st.abs. rather than the opposite the examples adduced throughout this ar-ticle will focus mainly on semantically indeterminate nouns. The instances selectedreflect various (morpho)syntactic contexts.

KuTY Determination in Targum Jonathan to Samuel 191

curtains'; II.20.3 ]Tr]b ym (lit.) 'women concubines'; 11.22.30 ]'DnD bÎ'S^pn 'all strong cities'.^

There are only a few unambiguous exceptions to the 'plural fac-tor'. In some instances, the awkard use of status in TgJon Sam. hasa parallel in BH with its occasional obscure use (or lack of use) ofthe article, e.g. 1.18.8 K'D' S nn' 'b) ] m TH"? Dn"" 'they have givento David ten thousands, and to me they have given [the] thousands^(BH mnm and D's'pxn respectively); 1.30.17 K'pT bs lTDil 'who rodeupon camels' (cf. BH D ' ajn); 11.1.22 ]n3: lino yb^'îp mn 'from (the)blood of (the) killed, from (the) fat of {the) warriors' (cf. BH wbbn andD^'-l^2). respectively). Such instances aptly remind us that determinationin Hebrew is not without problems of its own.® In addition, the notionof determination may not have been exactly identical in Hebrew andAramaic, which might explain certain discrepancies between the two.Ône should therefore be extremely cautious in assessing what is and isnot 'regular'.^" As a result, the number of problematic instances dropsdrastically.

Plurality, a morphosyntactic parameter, therefore appears to be asufficient condition to guarantee the observance of the classical distinc-tion, and one may note in passing that this fact alone is sufficient toinvalidate the claim that st.emph. and st.abs. do not work properlyanymore in TgJon.

2.2 Nouns in the SingularSingular nouns can be divided into two groups, depending on theirgrammatical behaviour when semantically indeterminate:

^Representative instances can be mentioned ad libitum, e.g. II.5.11 DTn...N"'pmD ]']33 ]']msn i-'^Dmsi ]-m ypd^ I'jmxi i'"i:]i j -ns ':;si Tin m'7 ]n:is -IK'and Hiram the king of Tyre sent messengers unto David, and wood of cedars andcarpenters who were trained to cut the wood and artisans who were trained inthe building of walls'. Cases of anaphora are particularly suitable to illustrate thisprinciple, e.g. 1.6.10 ]~\T\ i'mn TO.I'i 'and they took two cows\ continued in 1.6.12

SJ'IDKI 'and the cows went straight'; 1.25.5 T n nnsi I'D'"?!:; Sior T n n' Bl'and David sent ten young m,en, and David said to the young men'; 1.19.14

^b<2^ 'and Saul sent messengers', continued in the next verse 1.19.15 n' BlK'lJtS n" "PISD 'and Saul sent the messengers (again)'.

Â case in point is 'imperfect determination', whereby a noun, though cogni-tively unavailable, is formally determinate, cf. P. Joiion and T. Muraoka, A Gram-mar of Biblical Hebrew (Rome: Editrice Pontificio Istituto Biblico, 1996), §137m-o.

' Cf. A. Tal, Language, pp. 85-87.° All the more because one's understanding of determination is inevitably biased

by one's own linguistic background. Obviously, the use of the definite article inEnglish is not necessarily the same as the use of the st.emph. in Aramaic.


• Type A: Nouns that occur in the st.abs., e.g. p'Pin 'part, share', asin 11.20.1 Tnn p'pin KJ*? n 'p 'There is no portion for us in David';

• Type B: Nouns that occur in the st.emph., e.g. 3nn 'sword', asin 1.17.45 K3"im ""^V T\K nx 'you are coming against me with a

The question that has to be answered now is which noun belongs towhich Type. A close scrutiny of our corpus suggests that yet anothernominal category is influential in this respect: formal gender. It can beobserved that nouns formally marked as feminine (i.e. provided withthe feminine marker sn- in the st.emph.sg.) do not behave in the sameway as nouns that are not marked as feminine. For convenience's sake,we shall refer to the former as 'formally feminine' and to the latter as'formally masculine'.'^^

2.2.1 Formally feminine nounsNouns formally marked as feminine display an overwhelming tendencyto occur in the st.emph., irrespective of their own semantic determina-tion: the feminine nouns occurring frequently in our corpus display aclear preference for the st.emph. as against the st.abs., and the femi-nine nouns occurring only sporadically largely confirm this preference.Typical instances include: 1.7.9 Krh:} rfposi 'and he offered it up as aburnt offering^; 1.28.7 X^in Knnx '" IVI 'seek for me a woman whoknows...'; 11.3.7 snrn'p 'PIKD'71 'and Saul had a concubine'; II.6.8 T :J"inK]V2 Kn:J"in 'the Lord made a break on Uzzah'; 11.11.14 sm^K T'n inDi'and David wrote a letter'; 11.12.3 sniDD r]^'b mm 'and it was like adaughter to him'; 11.13.2 S'n mb^U2 'she was a virgin'; 11.17.17 nbtKI

'and a maid went'; 11.23.21 xn']-nQ nxiHim STDl 'and in the hand

^'It can therefore be said that in the singular Type A nouns maintain the classicalnorm of determination, whereas Type B nouns have dropped it for the benefit of thest.emph. (whereas in the plural all nouns maintain the classical distinction). As willbe seen shortly, however, certain linguistic contexts encourage specific behaviouralpatterns with regard to formal determination, enforcing—irrespective of semanticdetermination—the respect of the classical distinction between the st.abs. and thest.emph., or alternatively neutralizing it for the benefit of either. These factors willbe treated in §3 below. All instances adduced in §2.2 can be considered free of suchinfiuences.

• It must be noted that nouns referred to as formally masculine include notonly semantically masculine nouns (e.g. ~pQ 'king'), but also semantically femininenouns without formal feminine marker (e.g. Din 'sword').


of the Egyptian was a spear'.^^ Such a state of affairs strongly suggeststhat feminine nouns as a whole helong to Type B.

Exceptions, instances involving a feminine singular noun in thest.abs., are few in number. "* Some of them can be confidently ascribedto contamination.^^ As for the others, the question remains whether theuse of the st.abs. is merely the product of contamination or indicativeof a possible classification as Type A.

2.2.2 Formally masculine nounsIn contrast, the behaviour of formally masculine nouns with regard todetermination is more difficult to assess. With very few exceptions, nodistinct linguistic considerations can be identified that would conditionthe Type of masculine nouns. As a result, which of the two Types agiven item belongs to is unpredictable and must be determined onan individual basis, i.e. by studying its grammatical behaviour in itsactual occurrences.

Such an undertaking, however, entails difficulties of its own. Someformally masculine nouns are very poorly attested in TgJon Sam.,many of them occurring only once or twice in the whole corpus. Oth-ers are more widely attested, but it turns out to be difficult to findnouns that occur systematically either in the st.abs. or in the st.emph.when semantically indeterminate. Such a state of affairs should notcome as a surprise, however: inasmuch as two determination systemsresting on two widely different principles co-exist side by side in theAramaic of TgJon Sam. a certain amount of confusion—actually in-terference, or contamination—is inevitable. But inevitable though thiscontamination may be, the ambiguity it causes can make the evidenceinconclusive.

For these reasons, the state of affairs of the singular masculine nounsin TgJon Sam. is often not straightforward, and in not a few cases it

passage 1.6.14-15 ]t>v ip'ox ODB n'3 ^-\2i^ ... Krbi) lp'os sn-iin m 'andthey offered up the cows as a holocaust ... and the men of Beth-Shemesh broughtup holocausts' is a case in point, as the noun «rbv 'holocaust, burnt offering',semantically indeterminate in both verses but singular in the former and plural inthe latter, occurs in the st.emph. and st.abs. respectively.

^^Exceptions include: 1.2.36 ^ODT KSab 'for a eoin of silver'; II.7.11 -p D"p' oba'V 'the Lord will establish a kingdom for you'; 11.14.10 s' Q J^ 1Q"1 ]Q 'whoeverwill speak to you a word'; II.22.5 snKD XpV ']n3'pN 'distress surrounded me like awoman'. Obviously, instances in which the use of the st.abs. is due to the influenceof some factor (cf. §3 below) are not considered here.

i^f. §2.2.2 below.


has proven difficult to assign masculine nouns to Type A or Type B. ^In other cases, however, masculine nouns can be assigned to a Type.As the rest of this article will deal primarily with factors neutralizingthe preference for the st.emph. of Type B nouns, the listing belowfeatures, in alphabetical order, some of the masculine nouns that canbe assigned to Type B with some confidence, together with a few selectexamples:^^

"QJ 'man' (lat. vir): 1.9.16 X"a: imb n' OX 'I will send to you aman'; 1.17.4 liri'DDQ X13:i pS]l 'and a man from among them cameforth';2^^ 'sin, guilt, debt': 1.20.8 snn '3 n''K DS 'if there is sin in me';1.28.10 Kmn -\:v-\v^ un (lit.) 'if harm happens to you';DV 'day': 1.2.34 jin'Tin jlbopn' in saVD 'in one day the two of themwill be killed'; 1.27.1 '71X0T xnn m mv loanK xa'pi 'perhaps Iwill be given over one day in the hand of Saul';D'n'? 'bread, food': 1.30.11 son'? 7]'b DH'T 'and they gave himbread'; II.9.7 mns bv san"? b^D'^\ 'you will eat bread on my table';nma 'altar': 1.14.35 n' nip xnam 'nob 'to build an altar beforethe Lord', 11.24.18 xnniD n' mp D'pK 'erect before the Lord analtar';-pQ 'king': 1.8.5 K^bo st> ''m 'appoint for us a king'; 1.13.14 r^npti)r]'J2!) bv SD'PQ nno'? 'V 'and the Lord commanded him to be kingover his people';D ' llJ 'young man': 1.25.14 in iza^'b^V 'in ... '^'Tixbl 'and one youngman told Abigail'; 11.17.18 if.C'b'iV imn' xtm 'and a young mansaw them';]p"lS 'redemption, delivery': 1.11.9 K]p"lS ]^Db •'H'' inn 'tomorrowyou will have deliverance'; 1.19.5 ' XIC 'PD'? S3T N]p"iD 'T lnv^ 'andthe Lord worked a great victory for all Israel';D3ns 'word, thing' as in 1.3.11 bxno'n SQJns Tnv N]xn 'beholdI am doing a thing in Israel'; 1.21.3 XQ ns ']ip3 XDD 'the kingcommanded me a word'.

that sense, Miiller-Kessler ('Earliest Evidence', p. 188) is right in countingthe 'difficulty in distinguishing between the absolute and emphatic state' amongthe characteristic features of the Aramaic of TgOnq and TgJon, a description apterthan the categorical statements made by others.

^'''The attribution of these items to Type B is based on both qualitative andquantitative (i.e. statistic) criteria. Fuller details of the method will be given in mydissertation.


3. Factors

Reference has been made above to situations in which the use of thest.emph. or st.abs. with singular, semantically indeterminate nouns isnot the product of the Type to which they belong, but is rather theeffect of the context in which they find themselves. Essentially thesesituations can be explained by the existence of various linguistic factorsthat stand in the way of the otherwise unquestioned use of the st.emph.or st.abs., enforcing the respect of the classical distinction between thest.emph. and the st.abs., or alternatively neutralizing it for the benefitof either.

These factors can be classified into three groups, depending on thelevel of linguistic expression at which they operate: lexical, morphologi-cal and (morpho)syntactic. Lexical factors revolve around the notion ofidiom, a given noun occurring in the st.abs. or st.emph. because it is apart of a set phrase. The status of the noun can further be determinedby its form (morphological factors) or by the broader syntactic contextin which it occurs ([morpho]syntactic factors). Within the frameworkof the present study, four factors will be examined.^*

3.1 Word Pairs and Enumerations (lexical factor)In certain idiomatic expressions the status in which a given noun oc-curs is not open to variation, but rather pre-determined by the largercontext of the set phrase in which the noun is embedded.^^

This is the case for various word pairs (falling usually under thescope of hendiadys) and enumerations, in which the st.abs. appears tobe preferred with semantically indeterminate nouns, even when eitheror both of the nouns involved belong to Type B. Examples include1.15.3 snx -\V^ -ID:Q 'from man unto woman'; 1.22.19 pT 1V^ •''' l JD'from youth and unto suckling'; 1.24.12 ]rnoi mn 'guilt and corruption';1.27.9 Knsi -a: 'man and woman'; 11.22.47 Ipisi ']^pn 'strength andsalvation' .'^'^

®A more comprehensive treatment of these factors will be presented in my dis-sertation.

^^Similarly, in the English expression 'to kick the bucket' the direct object mustbe formally determinate (i.e. provided with the definite article) for the expressionto have its idiomatic reading 'to die'. The clause 'John kicked a bucket', though cer-tainly not ungrammatical, can only be interpreted literally (cf. S.C. Dik, The The-ory of Functional Grammar. Part 1: The Structure of the Clause (ed. K. Hengeveld;Berlin; Mouton de Gruyter, 1997), pp. 103-104).

^"1.15.3 and 1.27.9 show that the nouns involved in enumerations can be con-nected by simple coordination or by more complex prepositional structures.


3.2 Infinitives (morphological factor)Infinitive forms used substantivally appear to form a consistent, thoughpeculiar, category of nouns. The instances encountered in TgJon Sam.suggest that they occur in the st.ahs., as expected, when semanticallyindeterminate, as in 11.15.14 XDro S]*? Tin nb 'there will be no escapefor us'; but that they also occur in the st.abs. when unambiguouslysemantically determinate, as in 1.4.6 I'ln X3T iOD' bp 'this sound of thegreat shouting'; 1.18.19 ni'Di ran^nx 'the giving of Merab'; II.16.2 'nSDTl'pom 'the drinking of those grown weak'. ^

However, when infinitives are used as ordinary nouns rather thanverbal nouns, they do behave like other nouns, in that they belongto one of the two Types and behave accordingly. Thus Ticn 'feast,banquet' (as against 'to drink, the drinking'), which appears to be ofType B, e.g. 1.25.36 n n nn X'noQ ri^b xm 'and behold he had a feast inhis house'; 11.3.20 S^nca ... -\::iKb TM 12V^ 'and David made a banquetfor Abner'.

3.3 Adjectives / Participles used substantivally (morphological factor)When used on their own with the value of a substantive, adjectivesand participles respect as a rule the classical distinction, e.g. 1.2.32 ^b"irra^ 30 TT' 'there will not be an old [man) in your house'; 1.16.1 '''7:1XD'PD '^n!2b ... ICD ... 'there is revealed ... one fitting ... to be the king';1.18.23 ])::>'TV^ KT JTH 'is it a small {thing) in your eyes?'; 1.23.17 'inxI'ln"? "f? 'I will be second to you'; II.4.10 "IODQD 'like one bringing goodnews'; 11.15.19 nK 'b: DXI 'and if you are an exile'; 1.26.20 in 2;':'n 'aweak (man)'.^^

Sometimes, however, participles are used substantivally with suchconsistency that they cease to be felt as verbal forms and end up beingtreated as ordinary substantives, e.g. ^a~\ 'shepherd' (as against 'onewho pastures'), ID] 'guard, guardian' (as against 'one who watches /

2'ln the same vein, the irregularity of 11.23.7 s m Km irn ns'pjnN 'the revelationof the great court' bears witness to the strength of the preference for an infinitivein the st.abs.

• A few instances involving a semantically determinate element can be men-tioned: 1.2.8 X 'Dn m n ... SJ'DOD ... D^D 'He raises up the poor ... he exalts theneedy' (genericity); 11.12.2-3 DUIQ VD n'b WDOQ'^I ... ]S} nn KTn:?';' 'the rich (man)had sheep ... and the poor {man) had nothing ...' (cognitive availability). In ad-dition, in certain cases TgJon features the st.emph. where the Vorlage has anindeterminate form, e.g. 11.12.1 Sl'DOQ im SITID in 'one was rich and one waspoor'; n.14.2 sn'D 'Pi) s' DSnD ... snnSD 'like a woman ... mourning over a deadone'. It is not clear whether these should be considered exceptions or the result ofa slightly different interpretation of the Targumist.


guards / keeps'), CQDQ 'servant' (as against 'one who serves'). Theattestations encountered in our corpus suggest that "U") and ~IQ] areassigned to Type B (e.g. 1.17.34 -\12V mn K'U"i 'your servant was ashepherd'; 1.28.2 -jJ'iCX 'Cn"; KIB: 'I will make you guardian for myhead'), whereas CQOQ belongs to Type A (e.g. I . l .U caCD "H't n'nOQKI'T Dtp 'and I will hand over him, who will be a servant before theLord').23

3.4 Nouns modified by an Attributive Adjective / Participle (syntacticfactor)The presence of an attributive adjective or participle appears to en-courage the use of the st.abs. Essentially, one notes that in semanti-cally indeterminate noun phrases involving a noun and an attributiveadjective ('a wise woman') or an attributive participle ('an interestingbook'), the st.abs. is very frequent.

In the present study we will focus on formally masculine nouns:1.20.2 T:JI DJns 'a small thing'; 1.21.7 D'an D'n' 'hot bread'; 1.25.830 DV 'a good day'; II.13.3 DOn -Q: 'a wise man'; 11.14.25 TSD n3:i'a handsome man'; 11.18.20 IDO 12: 'a suitable man'; 11.18.27 DQ "Q:'a good man'; 11.22.17 ^'pn -pa 'a strong king'. Though the use ofthe st.abs. in noun phrases of this kind is by no means consistent, thenouns involved in the instances above nonetheless belong undisputablyto Type B^ and, significantly, no other reason than the presence of theattributive adjective or participle appears to motivate the use of thest.abs.

4. Conclusions

Prom this preliminary survey three major conclusions can be drawn:

1. In contrast to the common view, according to which determina-tion in TgJon is irregular, or is not used correctly anymore, it ispossible to find a largely consistent pattern for the distributionof the st.emph. and st.abs. in TgJon Sam.

2. The apparent irregularity is due to the fact that the system ofdetermination in TgJon Sam. is not a monolithic whole, butis rather founded on two different linguistic subsystems rooted

nouns "in and IQ] are originally peal participles, the noun E'DEa a paelparticiple. The extent to which this distinction is relevant for classifying the Typeof these nouns is uncertain,

f. §2.2.2.


in two different understandings of determination. In the first ofthese subsystems the distinction between st.emph. and st.abs.is fully maintained and carried out. In the other subsystem thisdistinction is neutralized to the advantage of the st.emph. which,by and large, ends up doing the job for both.

3. Though coexistent in the linguistic system of the Aramaic ofTgJon Sam., these two subsystems are not interchangeable, anda close scrutiny of our corpus reveals that each of them is usedunder distinct circumstances. It has been shown that two basicgrammatical distinctions must be made on the basis of numberand gender: as a rule plural nouns maintain the classical normof determination, feminine singular nouns do not and masculinesingular nouns are shared between those that do and those thatdo not.

In view of these facts, the question arises how such an unusual stateof affairs may have come into being. More specifically: how did thesetwo subsystems originate, and by what process could two so widelydifferent concepts of determination come to coexist, each with its owndomain of use, in one and the same Aramaic dialect.

These two determination subsystems, inasmuch as a linguistic anal-ysis of TgJon Sam. permits to retrieve them in their pristine state, tallywith what we are accustomed to understand in terms of the Western/ Eastern divide prevalent in Late Aramaic: one system shares withWestern Late Aramaic the respect for the distinction between st.emph.and st.abs., whereas the other system partakes of Eastern Late Ara-maic as regards the central position granted to the st.emph.

That the Aramaic of TgJon should be characterized by typical East-ern Aramaic features should come as no surprise. It is an etablishedfact that our text received its ultimate form in Babylonia, the territo-rial entity of Eastern Aramaic, and nowadays most scholars agree thatthis happened in the 3rd or 4th century CE, i.e. towards the beginningof the Late Aramaic period. The notion that the second determinationsystem of TgJon Sam. has its roots in Eastern Late Aramaic practicesis therefore highly plausible, and finds further support in the fact thatit shares with well-known Eastern Late Aramaic dialects such as Syriacand Mandaic not only the same basic principle (viz. the supremacy ofthe st.emph.), but also some of its limitations, i.e. linguistic contextsin which the st.abs. remains prevalent.^^

dissertation will amplify this statement.

KUTY Determination in Targum Jonathan to Samuel 199

The existence of the first determination system in the Aramaic ofTgJon Sam., however, is of much deeper significance for this study.Indeed, respect for the distinction between st.emph. and st.abs. undis-putably points outside the Eastern Late Aramaic area and period,which opens the way to a wide array of speculations. Taken on itsown, however, this point cannot be taken as a decisive argument infavour of a Western provenance of TgJon. Indeed, considering the longprehistory of TgJon, the first redactional stages of TgJon (i.e. 'proto-TgJon') are likely to have taken place before the advent of the LateAramaic period, i.e. at a time when the classical distinction betweenst.emph. and st.abs. had possibly not collapsed yet in the East. In the-ory at least, the observance of the distinction between st.emph. andst.abs. in TgJon could therefore have been a continuation of MiddleAramaic usage, irrespective of dialectal distinctions between East andWest, and the view that the whole redactional process of TgJon tookplace in Babylonia would not be weakened in the least.

However, if the Babylonian hypothesis is to be maintained certainessential aspects remain puzzling:

1. Even if the possible retention of the distinction between st.emph.and st.abs. in the Eastern Middle Aramaic dialects is a distinctpossibility, it can on no account be taken for granted. The gram-matical features usually considered typical for the Eastern LateAramaic dialects, among which the loss of the determinative forceof the st.emph. marker, are the product of a long and complexdevelopment, which can be traced back long before the adventof the Late Aramaic period.^^ As a result, any claim that theclassical distinction was actually retained in the East during theMiddle Aramaic period is unwarranted and should be treatedwith utmost caution.

2. Supposing that the distinction was actually retained, if TgJonwas entirely Eastern in provenance, originating in the MiddleAramaic period in Babylonia and enjoying there an uninterruptedgrowth over the years in the hands of subsequent generations ofscholars until its final redaction in the Late Aramaic period, onewould expect the determination system of the Aramaic in which it

f. E.Y. Kutscher, 'Aramaic', in T.A. Sebeok (ed.). Current Trends in Linguis-tics VI : Linguistics in South West Asia and North Africa (The Hague; Mouton,1970), p. 361; K. Beyer, Die aramdischen Texte vom Toten Mecr (Gottingen: Van-denhoeck & Ruprecht, 1984), pp. 97-98.


was composed to be much more consonant with the largely com-mon basis shared by the other Eastern Late Aramaic dialects.Crucially, if the Aramaic of TgJon was a native Eastern Aramaicidiom, developed continuously from the Eastern variety of Impe-rial and Middle Aramaic along lines similar to the other EasternLate Aramaic dialects,^^ one would not expect such a compart-mentation in its determination system, plural nouns maintainingto the letter the classical usage so unnatural for Eastern Aramaicwhereas singular nouns largely drop it to go their own (Eastern)way. Quite on the contrary, such a blind split in the determinationsystem of TgJon Sam. is strongly suggestive of a non-natural pro-cess. The presence, side by side within one and the same dialect,but each on its own territory, of two linguistic systems of whichthe notional and referential frameworks are so widely divergent,can hardly be the product of a natural diachronic evolution. Itis better explained by entertaining the possibility of an artificialamalgamation of these two systems.

In other words, the present study of the determination in TgJonSam. allows us to presuppose the superposition of (at least) two dis-tinct redactional layers in the composition of TgJon, which in turnpoints towards the existence of (at least) two distinct stages in its his-torical development: (1) an initial redactional process in the hands ofscribes for whom the distinction between st.emph. and st.abs. was afully active grammatical device; and (2) a revision process in the LateAramaic period in the hands of scribes native to the Eastern Aramaicarea, for whom this distinction was not a living linguistic reality anymore and for whom the st.emph. had become the primary form ofthe noun.^^ In the course of revising TgJon the scribes of the secondgeneration would then have updated the text of TgJon not only incontent, but also in form. Speaking of determination specifically, their

difference between literary and spoken idiom is immaterial here. Literaryidioms also evolve over time, and even though their evolution cannot properly betermed 'organic', it is no less subject than that of spoken idioms to the homogeneity,coherence and continuity of the linguistic environment.

^^This conclusion differs considerably from Kaddari's, who claims that the pres-ence of the two determination systems in the Aramaic of TgOnq is the result of anatural linguistic development inherent to Western Aramaic rather than the prod-uct of some Eastern influence (cf. Kaddari, 'Studies', pp. 240-41). In other words,according to Kaddari, the state of affairs witnessed in the Aramaic of TgOnq is tobe explained by diachrony, not by geography.


revision would have led them to partly adapt the expression of thedetermination of singular nouns according to their own understand-ing thereof. The reasons why only certain aspects of the determinationsystem would have undergone revision, and the criteria on which theseaspects would have been selected to the exclusion of the others, mostprobably involve both linguistic and extra-linguistic (e.g. theological)considerations, and therefore deserve a treatment more thorough thanthat presented here.

Attributing the second redactional stage to scribes native to theEastern Aramaic area in the Late Aramaic period is a fair assump-tion, considering our current understanding of the complicated historyof the transmission of TgJon. Beyond that point, however, we are nolonger on safe ground. Nonetheless, although one cannot afford to bedogmatic in this area, it must be emphasized that a contrario determi-nation in TgJon Sam. is strongly suggestive of an ultimately Westernprovenance for our text. Indeed, the views expounded above make theBabylonian hypothesis (as espoused by Geiger and others) unlikely,unless one is prepared and willing to subscribe to the two views (1)that the classical distinction between st.emph. and st.abs. was a fullyactive and exploited grammatical device in the Babylonian dialects ofthe Middle Aramaic period; and (2) that a first version of TgJon wascomposed in the Babylonian Rabbinic academies towards the turn ofthe Christian Era in an (as yet hypothetic) Eastern Middle Aramaic di-alect, that it disappeared from the literary and religious scene entirelyfor several centuries and that it eventually reappeared in the 3rd or4th century and was submitted to large-scale revision and adaptationto the cultural and linguistic customs of the time.

However, the Palestinian hypothesis has much to commend itselfas an alternative, for both linguistic and extra-linguistic reasons. Onthe one hand, the characteristic retention of the classical distinctionbetween st.emph. and st.abs. in the West provides a more plausibleexplanation for the presence of the first determination system in theAramaic of TgJon Sam. than the Babylonian hypothesis does. On theother hand, on purely geographic grounds, unlike the Babylonian hy-pothesis, the Palestinian hypothesis offers a rational explanation, i.e.a historical background, for the existence of the two putative redac-tional stages of TgJon, their amalgamation and the resulting linguisticheterogeneity in the determination system of TgJon.

[AS 3.2 (2005) 203-204]DOI: 10.1177/1477835105059093

ADDITIONAL NOTES TO THE ARAMAIC MAGIC BOWLBM 135563

Matthew Morgensteru

University of Haifa

In a recent issue of this jonrnal I republished an Aramaic magic howlBM 135563. In this brief note, I wish to present an alternative inter-pretation of one difficult word, and to provide further support for theinterpretation of another.

10. O'nnp 'no—In my article (p. 211) I hesitatingly followed Miiller-Kessler and Kwasman's interpretation and translated 'your trays (?)of flour'. However, as is clear from the entry in M. Sokoloff's dictio-nary, STIQ is really a greased dish for an oven, not a tray for Bour.Prof. Moshe Assis of Tel Aviv University has suggested to me that theform TIQ may be better understood as an apocopated form of the Qalmasculine singular participle from the root ]"nQ 'grind'. Accordingly,we would understand ''D'nQp TICD to mean 'the grinder of your Hour'.This interpretation better suits the context, and is also supported bythe many apocopated forms found in the language of the bowl.

11. ]3n']—In my article (pp. 219-20) I suggested that this word is de-rived from a phonetic variant of the root '7"3n, and that the same root isattested in a bowl from the Mousaeiff Collection published by Shaked. Ihave now found further evidence for the root from an important Rab-binic manuscript, MS Paris 1402 of Halachot Gedolot.^ In a citation

^M. Morgenstern, 'Notes on a Recently Published Aramaic Magic Bowl', AS 2(2004), pp. 207-22.

^M. Sokoloff, A Dictionary of Jewish Babylonian Aramaic of the Talmudic andGeonic Periods (Ramat Gan: Bar Ilan University Press; Baltimore: The John Hop-kins University Press, 2002), p. 499 s.v. K'ntD.

•'A linguistic assessment of this manuscript will appear in my forthcoming mono-graph. Studies in Babylonian Aramaic.



from b. BM 97a that appears on folio 30a lines 6-7 we readn'DI inn K '"QDU "PDX 'some say that it ate mice, fell ill an died'. The sametext appears in the Spanish manuscript of the tractate, MS Hamburg165, as n'm '?jn''K n^DS b^K.

The shift from het to he characterizes the early eastern manuscripts,''while the interchange of nun and lamed attests to the existence of thesecondary root already found in the magic bowl. This now provides an-other example of phonetic features of Babylonian Aramaic attested inthe magic bowls and preserved only in the best Rabbinic manuscripts.^

^I deal with this phenomenon in detail in my forthcoming work mentioned inn. 3. Meanwhile, see Y. Kara, Babylonian Aramaic in the Yemenite Manuseriptsof the Talmud (Jerusalem: Magnes Press, 1983, Hebrew), pp. 58-62.

^See my article, 'Linguistic Notes on Magic Bowls in the MoussaiefF Collection',BSOAS (forthcoming).

[AS 3.2 (2005) 205-13]DOI 10.1177/1477835105059094

RENDERING 'FLESH AND BONES':PAIR REVERSAL AND THE PESHITTA OF JOB 2.5*

David J. Shepherd

Briercrest College and Seminary

Over the last few years, the rather curious phenomenon of pair reversalin the Old Testament Peshitta has been noted with respect to variousbooks by A. Gelston\ R.A. Taylor^, M.D. Koster^ G. Greenberg^ andP.J. Williams.^ In returning to the topic in his recent monograph on theSyriac versions of the Gospels, the latter finds the translators' reversalof source text pairs to be widespread in these texts as well. Indeed,Williams' approach suggests the possibility that the phenomenon ofpair reversal in the Old Testament Peshitta may best be explainedby a combination of factors including the translator's idiom, differingperceptions of logical and chronological sequence and the desire to

'While any shortcomings in what follows remain the responsibility of the author,the present study has benefited from the questions, comments and collected wisdomof P.J. Williams and those attending The Bible of Edessa, a symposium convenedby the Peshitta Institute under the auspices of the meeting of the InternationalOrganization for the Study of the Old Testament in Leiden, 2004.

^A. Gelston, The Peshitta of the Twelve Prophets (Oxford: Clarendon Press,1987), p. 71. See also pp. 135-36, 149.

2R.A. Taylor, The Peshitta of Daniel (MPIL, 7; Leiden: E.J. Brill, 1994), pp.320-21. Similarly, p. 77: ' . . . it should be noted that there is in Syr of Daniel atendency to reverse the order of matched pairs.'

^M.D. Koster, The Peshitta of Exodus (SSN, 19; Assen: Van Gorcum, 1977),pp. 55-56, 583 n. 374.

""G. Greenberg, Translation Technique in the Peshitta to Jeremiah (MPIL, 13;Leiden: E.J. Brill, 2002), p. 53.

^P.J. Williams, Studies in the Syntax of the Peshitta of 1 Kings (MPIL, 12;Leiden: E.J. Brill, 2002), p. 155, notes pair reversals in the Peshitta of 1 Kgs 8.29and 8.59.


harmonize the word order of particular pairs.^ While Williams' workrepresents a significant step forward, his study does not afford him theluxury of grappling with the question of how one might know whethera particular pair reversal actually reflects, in his words, a 'natural'or 'preferred' word order in Syriac.'' What, we might wonder, are thetextual, historical and linguistic factors which conspire in any givencase to produce one word order rather than another? The present studytakes up a particular instance of pair reversal in the Peshitta of Jobwith a view to addressing these questions.

The beginning of the second chapter of Job finds Satan keen to pickup the gauntlet thrown down yet again by the LORD, as we can seefrom verse 4 and following:

''And Satan answered the LORD, and said. Skin for skin, yea, all that aman hath will he give for his life. ^But put forth thine hand now, andtouch his bone and his flesh (nO3 "7X1 ICi'îV "PX ::3l), and he will cursethee to thy face. Ând the LORD said unto Satan, Behold, he is in thinehand; but save his life. ^So went Satan forth from the presence of theLORD, and smote Job with sore boils from the sole of his foot unto hiscrown.

While the Satan's invocation 'Skin for skin' remains an obscure rhetor-ical flourish, the explicit physical affliction of Job we find narrated inverses 6 and 7 makes it clear that the LORD acquiesces to the Satan'sdemand to 'touch his bone and his flesh'.*

When we turn to the Peshitta of Job, we see that the translationdiverges from the Hebrew by rendering cnôtv^ or<f tnirosiX âtoo'touch his fiesh or his bone'. While the provision of at< 'or' in place ofa 'and' is patient of various explanations, our primary interest in the

^While P.J. Williams, Early Syriac Translation Technique and the Textual Criti-cism of the Greek Gospels (Texts and Studies 3.2; Piscataway: Gorgias Press, 2004),pp. 210-13, offers six possible explanations for the reversal of paired items in theSyriac Versions of the Gospels (Error, Relative clause in Second Position, Influenceof Tatian's Diatessaron, Preferred Order, Assimilation, Ad hoe Explanations) onlythe latter three seem germane to the present discussion. While this study offers anexplanation of the pair reversal in Job 2.5, instances of pair reversal such as thosefound in 1 Kings 8 (see n. 5 above) make it unlikely that this explanation will beable to account for all instances of pair reversal in the Peshitta.

' WiUiams, Early Syriac T'ranslation Technique, p. 211.*For a survey of various explanations of the expression 'skin for skin', see D.J.A.

Glines, Job ISO (Dallas; Word Books, 1989), pp. 43-45.

SHEPHERD Rendering 'Flesh and Bones' 207

Peshitta's rendering lies in its reversal of the Hebrew pair such thatSyriac rtoDt^'bone' now follows t<\sD=i 'flesh'.^

We may begin by observing that the Hebrew presents a word order(nS3--D:iJJ 'bone'-'flesh') here in Job 2.5 which is found elsewhere in theHebrew Bible only in the so-called 'relationship formula' identified byReiser in passages such as 2 Sam. 19.13, where David instructs Zadokand Abiathar to say to the men of Judah: DHX nsjm "QUiU DHK 'nx;'You are my kinsmen; you are my bone and my flesh.' ° Given theapparent incongruity of such a relational expression here in Job 2.5where the context seems, on the contrary, to be concerned with theaffliction of Job's own literal flesh and bones, one might well wonderwhether the Syriac translator has not attempted to eliminate any hintof this relational idiom by reversing the order of the pair in his Syriactranslation.^^ While this explanation is indeed possible and, on firstblush, perhaps even plausible, we will see that evidence from elsewherein the Peshitta renders it improbable.

It is perhaps predictable that in several instances of this idiom(Gen. 2.23; 29.14; and Judg. 9.2), the word order of the Hebrew ("103-D: :: 'bone'-'flesh') is reproduced faultlessly by the Peshitta. However,when we turn to the three appearances of this same expression in 2Samuel, we discover the very same pair reversal in the Peshitta ofSamuel that we encounter in the Syriac of Job. Whereas in the Hebrewof 2 Sam. 19.13, we have seen that David's message to the Israelites is,Dns nO31 'Q U nnK 'nx 'You are my kinsmen; you are my bone and myflesh,' in Syriac it is •»)'t\a .'tmao ^ ^ K T .unir 'You are my kinsmen

^Perhaps the Satan's estimation is that a God as powerful as the LORD wouldnot need to touch both Job's bone and flesh to prompt the anticipated curse. Itmay also be that the translator favoured ari 'or' over o 'and' in order to makesense of the fact that while Job's flesh is exphcitly afflicted in the verses whichfollow, his bones are not as obviously abused.

i°W. Reiser, 'Die Verwandtschaftsformel in Gen 2:23', TZ 16 (1960), pp. 1-4.Reiser's conviction is that this shorter idiom (found in Gen. 29.14; Judg. 9.2; 2Sam. 5.1; 19.13, 14; and 1 Chron. 11.1) was adapted by the Yahwist for use inGen. 2.23. W. Brueggeman, 'Of the Same Flesh and Bone (GN 2,23a)', CBQ 32(1970), pp. 532-42, concurs that the Genesis usage is derivative, and that it must beunderstood in light of what he sees to be the original idiom's primarily covenantalconnotations rather than blood connection or kinship.

^ Amongst those who hear in the present passage an intentional allusion toAdam's exclamation in Gen. 2.23, the suggestion of S. Meier, 'Job 1-2: a Reflectionof Genesis 1-3', VT 39 (1989), pp. 184-92, that 'bone and flesh' points to Job'swife, is to be preferred to that of Glines, Job 1-20, pp. 45-46, who sees Job's slainchildren as the object of the allusion.

and my flesh and my bone.' So too in the following verse David remindsAmasa: tsoi^a hur^ ,\ta=> r^m 'See, you are my flesh and bone' ratherthan reproducing the order (or rhetorical question) of the Hebrew:Tim nC3i 'D^:: Xl' n 'Are you not my bone and my flesh?' Finally inverse 1 of chapter 5, the Israelites come to David in Hebron, no longerproclaiming 13n]X -[icm -jn :; ^m 'See we are your bone and your flesh'but rather «y\^'C^p ^ tri'tm= r<rcn 'See we are your flesh and yourbone'.^^ That the Syriac translators insist on reversing the order ofthis pair not only in Peshitta Job, but also in Peshitta Samuel wherethe idiom is admirably integrated into contexts of covenant and/orkinship, suggests that we must look further afield for an explanationof this transposition in the Peshitta of Job.^^

Continuing our search requires a return to the English translationwith which we began. While the Authorized Version (1611) reproducesthe order of the Hebrew of Job 2.5 by supplying 'his bone and hisflesh', the much more recent New International Version (1984) reversesthe Hebrew pair in the very same way the Syriac does by rendering'his flesh and bones'. Why the NIV translators preferred 'his flesh andbones' to 'his bone and his flesh' will be quite obvious to any nativespeaker of the English language: for the same reason most say 'saltand pepper' instead of 'pepper and salt', or 'flsh and chips' insteadof 'chips and flsh'. Simply put, they sound better. "* In other words,while 'bone and flesh' does not collocate in this order in English, 'fleshand bones' certainly does—its separate parts functioning here as acompound synecdoche for the whole of Job's body. ^ That the NIV un-

n. 25 below for the textual variant (plural: oyioTi^) found in the 12alfamily of witnesses.

^^While Brueggeman, 'Of the Same Flesh and Bone', pp. 536-37, rightly observesthat the idiom must have the connotation of covenant rather than kinship in 2 Sam.5.1, he also admits that the blood-ties may well be in view in 2 Samuel 19 and inJudg. 9.2.

^''That word order is one component of a definition of collocation is suggested byY. Choueka, 'Looking for Needles in a Haystack or: Locating Interesting Expres-sions in Large Textual Databases', in Proceedings of the RIAO International Con-ference on User-Oriented Content-Based Text and Image Handling (Cambridge,MA: MIT Press, 1988), pp. 609-23, who defines collocation as ' . . . a sequence oftwo or more consecutive words, that has characteristics of a syntactic and semanticunit, and whose exact and unambiguous meaning or connotation cannot be deriveddirectly from the meaning or connotation of its components.'

^^E.W. Bullinger's 19th century compendium. Figures of Speech Used in the Bible(London: Eyre and Spottiswoode, 1898 [repr. Baker, 1987]) does not refer to 'fleshand bones' but does include the comparable collocation 'flesh and blood' (pp. 644-


derstands Job's body as the intended reference of 'his flesh and bones'in 2.5 is evident from the way it renders the relationship formula in 2Samuel. In chapter 5 verse 1, for instance, the children of Israel remindDavid not that they are his 'flesh and bones' but his 'flesh and blood'which is of course the most idiomatically appropriate way of expressingin Enghsh the kind of relationship or kinship suggested by the Hebrew"ira-D^:; collocation.^^

Of course the conclusion that 'flesh and bone/s' resonates in En-glish in a way that 'bone and flesh' does not, begs the question why.While acknowledging that the origins of idioms are often inscrutable,we may point to the answer suggested by the data of the Oxford En-glish Dictionary. According to this redoubtable resource, the colloca-tion 'flesh and bone/s' first appears in English courtesy of the Anglo-Saxon gospels and their rendering of Lk. 24.39: 'Gast naesb flaesc &ban'. ' It seems very plausible that this and Wycliffe's 14th-centuryEnghsh version of the end of Eph. 5.30 ('And we ben membris of hisbodi, of his fleisch, and of his boonys') may offer at least a partial expla-nation for why 'flesh and bone/s' sounds more natural than its inversein English: the expression has been encountered in this order over andover again in arguably the most influential text in the English languageover the last thousand years, the New Testament.^^ While these earli-est English versions of the New Testament and indeed all others before

45), perhaps because it is a more common idiom in both the New Testament (Mt.16.17; 1 Cor. 15.50; Gal. 1.16; Eph. 6.12; and Heb. 2.14) and in English. Bullingerunderstands 'flesh and blood' as one of four varieties of synecdoche of the part, inwhich a part of a thing is put for the whole thing. Specifically, 'Flesh and Blood isput for the human nature as distinct from the Divine Nature: or for the body ofman as animal, mortal, and corruptible.'

^"According to the Oxford English Dictionary (ed. 1989) the earliest occurrenceof 'flesh and blood' as a synecdoche for human corporeality in English is to befound in the Anglo-Saxon Gospels (c. AD 1000) rendering of Mt. 16.17: 'Hit be neonwreah flaesc ne blod'. The same collocation is used to denote 'near kindred' inCursor M. 4129 (Cott.) 'He es your aun fiess and blod.' (al300) and Shakespeare'sMerchant of Venice II. ii. 98 (1596).

^''OED 'c.lOOO Ags. Cosp. Luke xxiv.39 Gast naesb flaesc & ban.' (ed. Skeat,1871-87). For the prevailing influence of religion on the development of languagesee Mario Pei, The Story of Language (Philadelphia: J.B. Lippincott Company,1965), pp. 206-14.

^^J. Forshall and F. Madden (eds.). The Holy Bible, Containing the Old and NewTestaments, with the Apocryphal Books, in the Earliest English Versions Made from,the Latin Vulgate by John Wycliffe and his Followers (Oxford: Oxford UniversityPress, 1850). For further reading, see the expansive bibliography on the influence


Tyndale were of course versions of the Vulgate, it is also clear that theLatin faithfully followed and thus transmitted to the English, the wordorder found in the Greek, ^

Given Luke's facility with Greek, it would be strange if , , , TCveu[xa

CTapxa xal oCTxea oux ijzi... ' , , , a spirit does not have flesh and

bones,.. ' in Lk, 24,39 represented anything other than bona fide Greek

idiom,^° Indeed, the interference of precisely such a Greek idiom pro-

vides an explanation for a curious and previously inexplicable de-

parture in the allusion of Eph, 5,30 to the Septuagint of Gen. 2.23:

, , , OCTXouv ex xwv OCTXCMV [XOU xal CTap^ ex XT]^ aoi.py.6c, [xou,,.

'bone of my bones and flesh of my flesh',^^ While the latter follows the

Hebrew of Genesis ( , . , nE^Q nom 'D i JQ U)iS ... 'bone of my bones and

flesh of my flesh'), the influence of contemporary Greek idiom is the

most likely explanation for why either Paul or a later glossator reverses

the order of the Septuagint when it is cited in the letter to the Eph-

esians: ex XY]? CTapxo? auxou xal ex xwv oaxewv auxou 'from/of his

flesh and from/of his bones',^^ Indeed, the Greek preference for plac-

of the Bible on English literature in D,L, Jeffrey, A Dictionary of Biblical Traditionin English Literature (Grand Rapids: Eerdmans, 1992), pp, 929-36,

^^R, Weber et al. (eds,), Biblia Sacra iuxta Vulgatam Versionem (Stuttgart:Deutsche Bibelgesellschaft, 1983) ' , , , spiritus carnem et ossa non habet sicut mevidetis habere, , , ' (Lk, 24,39) ' , , , de carne eius et de ossibus e ius , , , ' (Eph, 5,30),For background on the Early English versions see G, Shepherd, 'English Versionsof the Scriptures before Wycliff', in G,W,H, Lampe (ed,). The Gambridge His-tory of the Bible, II (Gambridge: Gambridge University Press, 1969), pp, 362-86,and B, Metzger, 'The Anglo-Saxon Version', in his The Early Versions of the NewTestament (Oxford: Glarendon Press, 1977), pp, 443-55,

^°See J, Fitzmyer, The Gospel According to Luke I-IX (Garden Gity: Doubleday,1981), pp, 107-27, for an analysis of Luke's Greek and a nuanced endorsement ofJerome's early assessment that Luke was 'among all the gospel writers, the mostgifted writer of Greek' {inter omnes evangelistas graeci sermonis eruditissimus fuit,Ep. ad Damasum 20,4,4; GSEL, 54,108),

^^According to A, Rahlfs (ed,), Septuaginta id est Vetus Testamentum graeceiuxta LXX interpretes (Stuttgart: Wiirttemburgische Biblanstalt, 1935) and com-pared with A,E, Brooke et al. (eds,). The Old Testament in Greek, According tothe Text of Godex Vaticanus: Supplemented from Other Uncial Manuscripts, witha Gritical Apparatus Gontaining the Variants of the Ghief Ancient Authorities forthe Text of the Septuagint (Gambridge: Gambridge University Press, 1906-40),

^^See M, Barth, Ephesians: Introduction, Translation and Gommentary (GardenGity: Doubleday, 1974), pp, 720-38, and P, Rodgers, 'The Allusion to Genesis 2:23at Ephesians 5:30', JTS 41 (1990), pp, 92-94, for a discussion of the textual statusof this clause (attested in (K ) D F G (K) •F SK lat Ir) and its significance for thereading of Ephesians 5 (if original) or the history of the chapter's interpretation


ing CTap^ 'flesh' before OCTTEOV 'bone' in this collocation will come asno surprise to anyone familiar with the diction of Aristotle, who usesthe expression quite liberally in, for instance, his Metaphysics.'^^

In light of all this, what then can we say about the preferencewe see in both the Peshitta of Job and Samuel for t^tma 'Hesh' tocome before rsia't^bone'?^'' It seems quite possible that just as inEnglish 'flesh' sounds more natural before 'bone' than after it, so toor^tma sounded more natural to the ancient Syriac ear before rtfast^rather than after. Why this might have been the case is of courseimpossible to answer for certain but it is worth noting, I think, thatwhere 'flesh and bones' does appear in Syriac, it seems to do so oftenin connection with precisely the New Testament texts we have beendiscussing. So for instance, Ishodad of Merv agrees with the variousversions of the Syriac Gospels by supplying rtf^ii^a nfima 'flesh andbones', faithfully following the word order found in the Greek of Lk.24.39 and apparently imported into the Syriac of Job 2 and 2 Samuel. ®Unburdened by doubts regarding the end of Eph. 5.30 in the Greek,

(if a later gloss). While J.A. Bengel, New Testament Word Studies, 5.2 (GrandRapids: Kregel, 1971 [repr. ed. 1864]), p. 416, suggests that the reversal of theorder in Ephesians reflects the prioritizing of the new creation of Christ's 'flesh'over the old natural structure of the 'bones', it is interesting to note that Aquinas{Commentary on St. Paul's Epistle to the Ephesians [Albany: Magi Books, 1966])alludes to the parallel order of this pair in Lk. 24.39 in his explanation of theinversion in Eph. 5.30.

^^See for instance Arist. Metaph., 1058b (transl. H. Tredennick; Cambridge, MA:Harvard University Press; London: William Heinemann, 1933, 1989): 'For we arenow regarding "man" as matter, and matter does not produce difference; and forthis reason, too, individual men are not species of "man", although the flesh andbones (al aapxs? xal xi. OCTTS [ed. W.D. Ross; Oxford: Clarendon Press, 1924])of which this and that man consist are different.' See also 993al, 1034al, 1035a,1036b, 1058bl, 1070b.

• '*See M.P. Weitzman, The Syriac version of the Old Testament: An Introduc-tion (University of Cambridge Oriental Publications 56; Cambridge: CambridgeUniversity Press, 1999), pp. 164-205, for a helpful discussion of the relationshipbetween the various books of the Old Testament Peshitta. The fact that here bothP-Job and P-Samuel reverse the order of the pair found in the Hebrew text whileP-Genesis does not, is simply one more amongst a multitude of interconnectionswhich may be traced between the books of the Old Testament Peshitta.

• For the former, see M.D. Gibson, The Commentaries of Ishodad of Merv, 1 and3 (Cambridge: Cambridge University Press, 1911), ad loc. Regarding the latter,the work of G.A. Kiraz, Comparative Edition of the Syriac Cospels (Leiden: E.J.Brill, 1996), indicates that the Sinaiticus, Curetonianus, Peshitta and Harkleanversions speak with one voice on the matter. The appearance of the plural (oy.a) '!^'your bones') in later textual traditions of 2 Sam. 5.1 (12al family) points to the

the Syriac tradition cites it in its own tongue more widely and indeedmore frequently than any other part of Ephesians 5. ^ Indeed, if theoriginal translators of the Old Testament Peshitta were as familiar withthe Syriac versions of these New Testament passages as were the manycommentators and interpreters who would later cite them, then theanalogy of the English tradition may well be instructive. Perhaps infact, the Syriac translator or translators responsible for the Peshittaversions of Job and Samuel preferred r<;>3'ti 'bone' to follow '*'*m-.'flesh' for the very same reason that the English translator of the NIVdid centuries later, namely, because the Greek of the New Testamentitself had influenced the Syriac idiom in much the same way it wouldlater influence English and other European languages.

Of course, if the translation of the Old Testament Peshitta wasaccomplished by Jews in the vicinity of Edessa before they convertedto Christianity, as Weitzman suggests, this preference we see in someparts of the Old Testament Peshitta for irrtxaa 'flesh' to come beforet<aj'ti^bone' may well be httle more than a vestige of the linguisticlegacy of the Hellenism which flowed through Antioch to Edessa. ^ If,however, as Ter Haar Romeny quite convincingly argues, the EdessanJews produced the Old Testament Peshitta at some point after theirconversion, it seems quite plausible that when the proselytes turned totranslate the sacred scripture of their ancestral fathers, it was the writ-ings of their new-found faith which prompted them to reverse this par-ticular pair of words where they encountered them in Job and Samuel.^*Whichever explanation is finally found more persuasive, the presentdiscussion bears witness not only to the complex interplay of idiomand linguistic/literary influence which may lead to transpositions inthe Bible versions of any era, but also to the importance of attendingcarefully even to transpositions as apparently trivial as when, instead

possibility that it is not only the translation of the Old Testament Peshitta whichhas been influenced, but also its transmission.

^^B. Aland and A. Juckel, Das Neue Testament in syriseher Uberlieferung, 2. DiePaulinischen Briefe, 2 (Berlin: Walter de Gruyter, 1995), list no less than twenty-three citations amongst 14 different sources. In every case, the Syriac faithfullyfollows the order of the Greek.

^^Weitzman, The Syriac Version of the Old Testament, pp. 258-62.^*R.B. ter Haar Romeny, 'Hypotheses on the Development of Judaism and Ghris-

tianity in Syria in the Period after 70 CE' in: H. van de Sandt (ed.), Matthew andthe Didache: Two Documents from the Same Jewish-Christian Milieu? (Assen: VanGorcum; Minneapolis: Fortress Press, 2005), pp. 13-33.


of bone and flesh, the Syriac translators render the flesh before thebone.

[AS 3.2 (2005) 215-40]

DOI: 10.1177/1477835105059095

VOWELS IN THE TREES: T H E ROLE OE VOCALISATIONIN STEMMATOLOGY*

Eveline van Staalduine-Sulman

Theologische Univevsiteit Kampen

1. Introduction

Within the framework of the Kampen research projeet on the 'Originand History of Targuni Jonathan to the Prophets', a preliminary studyfor an edition of Targuni Samnel has been undertaken. This includesa stemmatological ytudy, which is the responsibility of the present au-thor.

Although we now know of 25 continuous manuscripts of TargumSamuel and hundreds of fragmentary manuscripts and haftai'ot read-ings, how they are related to one another is still largely unknown.Sperber published some witnesses to the Yemenite and the Westerntradition, but without evaluating the significance of the variants.^ TheBabylonian tradition has recently been publisiied.^ Important as thispublication is, it does not clarify the relationship between this tradi-tion and the other traditions. A reliable stemma of all the material is,therefore, still much needed.

Stemrnatology is the humanistic discipline that attempts to recon-struct the transmission of texts, especially texts in manuscript form, on

*This article is an expanded version of a paper read at the IOTS congress atLeiden, the Netherlands (2004). I wish to express my gratitude to the hearers fortheir questions and comments, to Alberdiiia Hontnian and Willem Snielik for theircorrections and additions, iind to Marjolein Tiirner-Prins for her positive infiiiciiceon my English.

^A. Sperber, The Bible in Aramaic Based on Old Manuscripts and Printed Texts,Vol. 2: The Former Prophets According to Targuni Jonathan (Leiden: E.J. Brill.1959).

^E. Martinez Boroliio, Tar<jum Jonutan de los profetas prtmeros en tradicionBabilonica, Vol. 2 : I-II Samuel (Maxirid : CSIC, 1987).

© SAGE Publications (London, Thousand Oaks CA, and New Delhi), 200,5

the basis of the relations between the various surviving manuscripts.Until recently, stemmatological methods were not applied to TargumStudies, and now that they are, the question arises what adaptationthe methods must nndergo before they can take account of all thepeculiarities of the Targuuis.**

backgToundBabv IonianYemenite

ItalianSefardi

Ashkenay.i

Editions

siglumE l

L22

L23L71L72Q2Q6R3H7K5L75ZlZ7DlG2JlL26S3BlB2

•manuscriptL 229 New York (Eb 1)2210 London2371 London1471 London1472 London2 Rav Yosef Kaparh coll.63/2 Rav Yosef Kapach coU.Reiifhlin 3 (Kennicott 154)Hebreu 75 ParisKennicott 5 [85] OxfordOpp Adds A" 75 Oxford1 Salamanea7542 MadridOr. FoL 1-4 (Keimicott 150)11 (883) GottweigEl f.6 (Kennifott 182) Jena26.879 LondonSolger 3.2" NnrenibergB'lbtia Rabbinica, VeniceBiblia Rabbinica. Venice

date

dated 146816th-17th centurydated 1589dated 1512/13dated 143116th 17th centurydated 1105/0614th 15th century13th centnryprobably early 14th centurydated 1532 Alfonso de Zairioradate<l 1533 Alfonso de Zainoravoc. 1455, nis prob. 14th c.14th-15th centnry13th-14th rentnry13th centurydated 129tBomberg 1515/17Boniberg 1524/25

This article joins this discussion by describing research into the roleof vocahsation within stemmatology. It was triggered by Smelik's re-mark about vocahsation that tlie nse of variant vocalization is fraughtwith difficulties. The systems of vocalization differ, the scribe did notalways vocalise the text himself, or. as seems likely in case of Alfonsode Zamora, he invented the vocalization himself.'** During earlier col-lations of Targuni texts I had the feeling that local vocalisation habitsexisted, even within one vocalisation system. Therefore, I set out tosolve the problems of comparison, the separate vocalisator and the in-vented vocalisation dnring a new process of collating. I therefore used

^See e.g. W.F. Smelik. Trouble in the trees! Variant selection and tree con-struction ilhistrated by the texts of Targuni Judges.' in: P. van Reenen et al. (eds.).Studies in Stemmatology II (Amsterdam, Philadelphia: John Benjamins, 2004), pp.167-204, also published in Aramaic Studies 1.2 (2003), pp. 247-87: A. Houtman,"Different kinds of tradition in Targuni Jonathan to Isaiah." in: Idem., pp. 2()9-83.

"'Smelik, "Trouble in the trees!' p. 172.

VAN STAALDUINE-SULMAN Vocalisation in Stemmatology 217

the (vocalised) manuscripts and included their vo(;alisation listed inthe table at p. 216.

2. Collation of Texts

Before adding the uncertain data of vocalisation to stemmatologicalresearch, one must establish a method to produce a 'safe' stemma ofthe manuscripts on the basis of the consonantal text. This stemma canserve to check the outcome of stemmas including vocalisation.

• A feasible start to the stemmatological research is to begin withthe continnous text tradition in order to have enough material tocreate reliability. Fragments and haftarot can be used in a laterstage to fill in the stemma.^ For this part of the collation, all theavailable coutinuous manuscripts and early editions have beenused. The Babylonian fragments are also included to provide apoint of comparison with a markedly different tradition.

• Because of the abundance of the material,*^ one must look for away to create a reliable stemma without having to scrutinise allthe textual material first. One possible way to do this is a sam-ple survey.'' For this project I have composed a sample of fivetheologically 'neutral' verses from each cluster of five chapters ofthe Books of Samuel. It is advisable to spread the sample acrossthe entire text because of the possibility of successive contamina-tion, i.e. the observation that some manuscripts have more than

^It is not necessary to exclude the litnrgi(?al iiaftarot from a stemma, becausea stemma only establishes interrelationships and does not .say anything about theoldest text. H. von Soden. Dii-. Schnften des Neuen Testam.cnts in ihrer altestenerreichbaren Textgestalt hergestellt auf Grund ihrer Textgeschichte (Berlin: Alexan-der Duncker, 1902, 1910; Gottingen: Vandenhoeck k. Riiprecht. 1913) exchided thelectionaries with New Testament texts. If these texts were connted as variants toestablish the most probable text, their nnmber wonid have forced Von Soden toedit a late-Byzantine, liturgical text of the New Testament, according to T. vanLopik, 'Tekstkritiek: telt het wegen of weegt het tellen?" NedThT 45 (1991), pp.101-106.

''I.e. Bible verses. The number of 25 complete manuscripts is small compared toNew Testament studies with e.g. 1785 manuscripts containing the Gospel of .lohn,cf. K. Wachtel, 'Kinds of variants in the manuscript tradition of the Greek NewTestament.' in; P. van Reenen et al. (eds.), Studies in Stemmatology II (AniKterdam.Philadelphia: John Benjamins. 2004), pp. 87-98 (96).

^A. Houtman, 'Textual Tradition of Targuni Jonathan of Isaiah', in: J. Tar-garona Borras, A. Saenz-Badillos. Jeioish Studies at the Turn of the TwentiethCentury, Vol. 1, Leiden 1999, 145-53.


one Vorlage, even within one Bil^lical book.* The sample of Tar-gum Samuel also includes a representative selection of haftarahreadings (indicated by h).

1 Samuel1:17-21 h8:11-1512:16-20 h16:7-1120:7-11 h23:14-1830:6-10

2 Samuel3:17-216:10-14 h10:9-1315:21-2519:21-2522:21-25 h

No citations are included, because research into the reliability ofthese citations and their textual tradition has not been fully done.Furthermore, 'since these texts were copied in later centuries bycopyists who normally knew their Bible text well, the citationswere "corrected" to conform to contemporary text forms, and asa result the citations do not represent the actual text used (...)at all'«The program Collate was used for the actual collation.'" Theoutcome was an unselected list of variant readings.

3. Selecting Varimits for StemmMology

In order to establish relationships between manuscripts, one must haveat least two variant readings, both witnessed in at least twomanuscripts.^^ The variant readings must be selected: only readingsreveahng a stemmatological relation eau be used. Other variants, e.g.those revealing the scribe's carelessness, nmst be excluded.

can be established by the so-called shock wave method, cf. E. Wattel andM.J.P. van Mnlken, 'Shock Waves in Text Traditions', in: P. van Reenen and M.J.P.van Mulken (eds.). Studies in Stemmatology (Amsterdam: John Benjamins, 1996).pp. 105-21.

"J.W. Wevers. 'Die Methode,' in: Das Gottinger Septuaginta-Untemehme-n(Gottingen: Vaiidenhoeck and R.u])recht. 1977), p. 14.

^"P.M.W. RobinHon, Gotlate: Collation of Large Textual Tmditions, Version 2(Oxford: University Centre for Humanities Computing, 1994).

^^B. Salemans, 'Varianten als bonwstenen van stemma's: een pleidooi voor een-vond en openheid bij het opstellen van tekststambomen," in: G.R.W. Dibbets andP.W.M. Wackers (eds.), Wat Duikers vent is dit! Opstellen voor W.M.H. van Hum.-melen (Wijhe: Quarto, 1989), pp. 319-43 (331).


• Variants must be takeu iu chunks that are as small as possible.The text may contain several variant readings in each biblicalverse. A plus can be treated as one variant reading, regardless ofthe number of words it contains.

• Singular readings are not .significant for genealogical relation-ships. ^ Having only 25 complete manuscripts of Targum Samuelwe caunot entirely exclude these singular readings for establish-ing the original reading.'^ They have a justified place within thecritical apparatus of an edition. Yet they are useless within stem-matology because they do not reveal relationship, only lack ofrelationship.

• The hot issue among stemniatologists is whether one must useall the variants or only a selection of variants, Aud if a selection,what are the selection criteria? The risk iu selecting variants isthat it may lead to a wrong or manipulated outcome. '* The riskin not selecting, however, is even greater: a Variantenfnedhof.^''Many variants do not reveal relationships, because copyists areinclined to correct texts according to their own view on text audlanguage, or according to the original Hebrew text in the case ofthe Targum. Every kind of text and every language must createits owu lists of non-revealing types of variants, because what ascribe allows himself to correct depends ou the kiud of text, andwhat a scribe is able to correct depends on the language. Forexample, the use of capitals iu medieval Dutch is non-revealing,

^^E..I. Epp, 'Toward the Clarification of the Term 'Textual Variant",' in: .l.K.Elliot (ed.). Studies in New Testament Language and Text: Essays in Honour ofGeorge D. Kilpatrick on the Occasion of His Sixty-Fifth Birthday (Leiden, 1976),pp. 153-73 (161). Yet some singnlar readings can be nseful as opposed to a singlegroup of readings, cf. E. Wattel, M.J.P. van Mulken, 'Weighted Formal Support ofa Pedigree", in: P. van Reenen and M. van Mnlken (eds.). Studies in Stemmatology(Amsterdam/Philadelphia: John Benjamins, 1996), pp.135-67 (155).

^^It is completely different within New Testament Stndies, since the chance thata singular reading among hundreds of manuscripts represents the original reading.is very small, cf. E.C. Colwell and E.W. Tiirie, 'Method in Classifying and Evalu-ating Variant Readings,' in: E.C. Colwelt (ed.), Stiidies in Methodology in TextualGriiicism of the New Testament (Leiden: E..]. Brill. 1969). pp. 96-105 (103).

^*A.A. den Hollander. De Nederlandse bijbelvert.aHnqen 1522-1545 (Nieuwkoop:diss., 1997), p. 135.

^^H.P.S. Bakker, Towards a Gritieal Edition of the Old Slavic New Testament:A Transparent and Heuristic Approach (Amsterdam: diss., 1996), p. 9. Cf. E. Tov,Textual Griticism of the Hebrew Bible (Minneapolis: Fortress, 1992). p. 301: 'Tex-tual Criticism does not proceed according to deniocratic rules".

220 Aramuic Studies 3.2 (2005)

but no issue at all in Aramaic {language); the introduction ofspeakers is uon-revealing in medieval Dutch, bnt is added veryconsistently in the Targum (a sacred text) and may therefore bereveahng.^^ Recent research tends to select very strictly: ^

- Scribal conventions, such as orthography and word bound-aries, seem to be iinrehable for steinniatology. " "Any scribecould adapt these things to his usage, without affecting thequality of the copy.' ^

- The use of variant vocalisation is fraught with difficulties. **Therefore, it must be examined whether vocalisation is areliable source for genealogical insight and which aspectsare the most reliable.

- Errors and dislocated readings do not reveal genealogicalrelationships,^^ because scribal errors are often typical andsusceptible to correction.^^ Smelik must therefore concludethat 'tliis category seems to yield the most unreliable resultsofair.^3

- Grammatical properties, like number, gender and state, seemto achieve more stable results, but still suffer from noise. "*

- Prepositions, copula and relativa do not seem to be entirely

- Substitutions, pluses, minuses and semantic shifts yield the

f. Salemans. 'Varianten als hoiiwstenen', p. 336.^''i follow the categories in Smelik. 'Trouble in the trees!' pp. 182-83.^^This is said for Dutch Vjy Saleumns, 'Varianteii als bouwstenen', p. 336; for

Targum Aramaic hy Smelik, 'Trouhle in the trees! pp. 185; Houtman, 'Differentkinds of tradition', p. 272. Epp, 'Toward the Clarifieation', p. 169 states that theorthogTaphy of names is revealing in New Testament Greek. This is not likely inAramaic-. Salemans p. 338 considers orthography of names non-revealing in Dutch.

^*'G. Miuk, Trohleuis of a highly <:oiitaminate<i tradition: the New Testament.Steimnata of variants as a source of a genealogy for witnesses," in: P. van Reenen etal. (eds.). Studies m Stemmatology II (Amsterdam, Philadelphia: John Benjamins,2004), pp. 13-85 (28).

2"Smelik, Trouble in the trees!' p. 172.2 Salemans, 'Varianten als bouwstenen" p. 324; cf. Epp, Toward the Clarifica-

tion,'' p. 160, who speaks of 'clear and demonstrable scribal errors'; cf. also p. 168., "Trouble in the trees!' p. 172., 'Trouble in the trees!', p. 185.

'Trouhle in the trees!' pp. 185-87. Cf. Houtmaii, 'Different kinds oftradition,' p. 272.

^^Snielik. Tro\ible in the trees!' p. 189. Cf. Ht)utman, 'Different kinds of tradi-tion,' p. 272.

VAN STAALDUINE-SULMAN Vocalisation in Stemmatnlogy 221

most convincing results. ** Naturally, substitutions reflectinga variant in the Hebrew text, haplography and dittographyshould not be taken into account.

- Word order is considered revealing, unless it is against gram-matical rules.^'' Changes in word order do not often occurin the Targnmim.

- Paratextual evidence like Tosofta Targnms nnist not be in-clnded in the procedure of stenimatolog>'. They are useful'as additional evidence in the reconstruction of the textualhistory'.^^

• In the entire process of selecting variants and producing a stemmaincidental contamination is the most powerful disturbance. Onecategory of contamination consists of identical errors which occurindependently.^^ Another category consists of identical correc-tions which occur independently. 'A scribe who knew his Hebrewtext would be quite unconsciously influenced by that knowledgeinto making changes sporadically in the copy.' **

4. Tlu: Three. Level Method

All the selected variants nuist now be examined with regard to theirinterrelationships. This is done by the Three Level Method, developedby Antonij Dees for the construction of a tree.^^ A. Houtman describedthe method as follows; '

, 'Trouble in the trees!' p. 189. Cf. Houtman, 'DifFerent kinds of tradi-tion,' p. 272: Salemanb, 'Varianten als bouwstenen/ p. 338.

^^Salemans, 'Varianten als bouwstenen,' p. 338.^^Houtman, 'Different kinds of tradition," p. 275.^ Cf. Zamora's minuses and pluses in Smelik, 'Trouble in the trees!' p. 178. See

also pp. 191-193.^"J.W. Wcvtirs, 'Die Methode', in: Das Gottingcr Septuaginta-Unttmehmen

(Gottingen, 1977), p. 17.•"A. Dees, 'Sur une constellation de quatre mamiscrits,' in; A. Dees et al. (eds.).

Melanges de linguistigue et de litterature offerts a Lein Geschiere par ses amis,colleges et eleves (Amsterdam: Rodopi, 1975), pp. 1-9: 'Considerations theoriquessur la tradition manusrrite du lai de rOrnbre,' Neophilologus 60 (1976), pp. 481-504;'Over stambomen en handschriften,' FoTura der Letteren 18 (1977), pp. 63-78. Somepros and ran.s are reflected upon in K.-H. Uthemann, 'Which Variants are Usefulin Discovering the Deep Structure of the Manuscript Tradition of a Text? Contra aKO-called Essentially Quantitative Approach,' in: P. van Reenen and M. van Mulken(eds.). Studies in Stemmatology (Amsterdam/Philadelphia: John Benjan^ins, 1996),pp. 249-61 (250-53).

'^Houtman, 'Different kinds of tradition,' pp. 272-73.

222 Aramaic Studies 3.2 (2005^

According to this method, in the first step the witnesses are clusteredinto subfamilies on purely quantitative grounds. In the second step, wit-nesses that might have been intermediary in the process of transmissionare identified. These two steps produce the chain of relationships thatunderlies the genealogical tree. At this stage, the nature of the rela,-tionships between the different members is settled, but not as yet theirdirection. This must be determined at the third step, where the pointof suspension, i.e. the root of the tree, is established ou the basis ofqualitative arguments. This last step is the most difficult one, and allpossible means must be employed to arrive at a well-founded decision,such as: assessment of the origins of the variants, palaeographical andcodicological data, and historical information.

Fii^st LevelAll the relevant variants are clustered and pi'ovided with a valuation.The selection of variants described above induces a valuation system,in which substitutions, semantic shifts, pluses and minuses—if not dueto clear scribal errors—are valued most. Grammatical and syntacticalchanges can bo used, but not valued so high. Other variants, like or-thography, prepositions or copulae, must either be omitted or be givena very low value,^^ Paratextual evidence, such as Tosefta Targums, il-histrations and masoretic notes, cannot be used either. The followingvaluation was used for the consonantal variants:

Criteria 1-6 include the consonantal text1 — strong semantic difference2 — strong omission or addition3 = weak semantic difference4 — weak omission or addition5 = syntactic difference6 = other differences, e.g orthography

counts for3.52.21.51.2-_34

Table 1: Valuation

^•'Houtman. 'Different kinds of tradition.' p. 272 describes her valuation systemas: orthography 0.5, minor syntaftical changes 1, substitutions, additions and omis-sions 2. An extensive discussion of weighting systems is given in M. Spencer et al..,•The effects of weighting kinds of variants,' in: P. van Reenen et al. (eds.). Studiesin Stemmatology II (Amsterdam, Philadelphia: John Benjamins, 2004), pp. 227-39with its final conclusion: 'Determining appropriate weightings in these fa.ses is anopen problem' (p. 238).

VAN STAALDUINE-SULMAN Voeahsation in Stcvmiatokxjy 223

The selected variants are used in four different methods to estab-lish relationships between the manuscripts. The outcome consists offour trees, each representing the network of relations between the wit-nesses. This procedure was developed by Evert Wattel (Free University,Amsterdam) and can be based on similarities or dissimilarities of themanuscri])ts, working with single or with [)airs of manuscripts.

• Clustering method• Skeleton method• Quadruple method• Pairing method

The pairing method seems to be the most reliable. The skeleton andthe quadruple methods do not differ much in results, although thequadruple method appears to be more accurate in handling a largeamount of data. Furthermore, the four rough versions provide a degreeof unrelatedness: for example, while the relations between the Yemenitemanuscripts are given values between 15 and 27 (indicating a closerelationship), the relations between the Yemenite family on the onehand and the Western family on the other hand are given values ofmore than 99 (indicating that they are related very loosely). In mt)stcases a compilation must be made on the basis of the entire outcome.

My first tree (cf. fig. 1) was based on consonantal variants, i.e.substitutions, semantic shift, additions and omissions. Because of thesmall number of data in this sample, the less-reveahng variants, suchas grammatical variants, were integrated, but on a very low valuation.Only if there was no revealing information at all. the less-revealingvariants were used to establish relationships.The limited data meantthat the connections of the separate branches were marginal, but theoverall tree was probable, although without historical orientation.

• The consonantal stemma shows the following division: Yemenite,Sefardi, Ashkenazi, Solger-Rabbinic Bibles group'*^ and a group oftwo manuscripts, viz. the Babylonian MS Eb 1 and MS Reuchhn.

•'"*These categories were actually counted as 0.001. to avoid relationship-lessmanuscripts.

•'^The Western text ran be divided into three categories: Sefardi, Ashkenazi andthe striuR of Solger and the Rabbinic Bibles. See also the outcome in Houtman,'Different kinds of tradition," p. 276.


- l1 04 t? 46

Figure 1: Tree based on consonantal variants

The similarity between the latter manuscripts may be coinciden-tal because of tlie small sample.^^


The Yemenite manuscripts show a very high internal coherence,even though two manuscripts from the seventeenth century areinvolved.^'' The numerical evaluation in the rough versions indi-cate that these manuscripts form a separate tradition which isrelatively far removed from the Western trachtions. The kinshipof MS 2210 and Kapach 63/2 is strong, just as the one of MSS2371 and 1472.The Ashkenazi category is subdivided into an early pair ofmanuscripts and a late pair. The issue of whether the time of ori-gin is coincidental requires further investigation. The early pairconsists of MSS 26.879 and El f.6.-* The late pair consists of MSSOr.Fol. 1-4 and 11 (883) Gottweig.^^Within the Sefardi group only the Zamora manuscripts reveal avery strong relationship.'*'Ms Solger together with the first and second Rabbinic Bible canbe discerned as a strong subgroup. By comparison of these textsA. Houtman proved concerning Isaiah that Codex Solger cer-tainly served as the Vorlage. of the first Rabbinic Bible.''

•'"Yet, there are more similarities between MS Reuchliii and the Babylonian tra-dition. Some vocalisation habits of MS Reuchlin correspond to the Babyloniantradition:1. A Shewa under the conjuctive Waw in "iCKl [e.g. in 1 Sam. 1.17);2. A Holem in "TD {e.g. 1 Sam. 1.21).

S. Morag, The Vocalization of Codex Reuchlinianus: Is the "Pre-Masoretic" BiblePre-Masoretic?' .]SSt (1959), pp. 217-37 (237), states that 'CR does not possessany major Babylonian feature which is not also a Palestinian feature.' The stemmaof the vocalisation will show that the vocalisation of CR is not especially linked tothe Babylonian tradition. On the contrary, it is (only slightly) linked to the Sefarfiitradition.

•' Cf. A. van der Heide, Tht Yemenite Tradition of the Targum of Lamentations{Leiden: E.J. Brill, 1981), 9-11. It is good to bear in mind that 4t is merely possibleto uncover the genealogical structure of the preserved texts' and that 'it is impossi-ble to find the genealogical relationship between manuscripts if most of them havenot survived'; so Mink, 'Probleiris,' p. 24.

''* is confirmed in Houtman, "Different kinds of tradition/ p. 276,is not confirmed in Houtman, 'Different kinds of tradition,' p. 276.

'**'Their numerical evaluation is 9, very related.^^A. Houtman, 'Targum Isaiah According to Felix Pratensis,' JAB 1 (1999), pp.

191-202. See also D. Shepherd, 'Before Bomberg: The Case of the Targum of Jobin the Rabbinic Bible and the Solger Codex (MS Niirnberg},' Biblica 79 (1998),360-79.


Second LevelIn the second level intermediary manuscripts must be discerned. Wesuppose concerning the first Rabbinic Bible that MS Soiger was usedas the Vorlage. Within the Yemenite group tlie first manuscript we en-counter, is MS Kapadi 2. The year of origin (1431) does not prohibit theconsideration of this manuscript as the Vorlage of the other Yemenitemanuscripts. Close scrutiny of the variants could lead to the conchisionwhether it could have been so in reality. The second manuscript in thisgroup is MS 1471, dated 1589. The manuscript can definitely not havebeen the Vorlage of the other four, beeause it is too late, although wecannot say anything about the text.Third LevelThe last step to be made is the establishment of a historical orien-tation: whieh manuseript most likely represents the eldest version andcan be taken as the origin of all the others? If we believe that the Baby-lonian tradition is the eldest, and that MS Eb 1 is the best representa-tive of this tradition.*^ the orientation is not difficult. The Babyloniantradition is then most original and from there two branches appearseparately: the Yemenite and the Western tradition. If Codex Reuch-linianus is really close to the Babylonian tradition, it is likely the eldestrepresentative of the Western branch.

5. Examining Contamination

After completing the orientated stemma, two examinations must becarried out with regard to contamination. A first category, incidentalcontamination, was discnssed at the end of section 3. and is importantfor the selection of variants. Two other forina of contamination canonly be checked afterwards."*''

• simultaneous contamination: one scribe used various manuscriptsat the same time and compiled a new text. In quiet times itis safe to assume that this kind of contamination stays withinthe tradition: 'The assumption is that, if contamination occurs,it emerges from those texts which were at the disposal of thescribe, i.e. texts in his direct environment, i.e. texts which are,for the most part, closely related with each other.'' ' The result

E. Martinez Borobio, Targum Jonatan de los profetas primeros en tradicionbabtlojiica (Madrid : CSIC, 1987}. p. 9.

Hollander, De Nederlandse bijbelvertalingen, p. 138.'Problems," p. 14.


is that contamination in a dense textual tradition, such as thetenth century New Testament manuscripts, occurs in very smallsteps and does not radieally affect the stemma. Contaminationappears to have a much greater impact on less dense textualtriulitions, such as the early New Testament manuscripts, becauseof the many missing links.^^ In these cases a highly contaminatedmanuscript will float within the trees and it will be impossible todetermine its actual place. In times that Jews were expelled fromcertain areas simultaneous contamination across the borders oftraditions would have occurred. Therefore, the term 'branch' ofthe tree nmst be taken with a pinch of salt.''^

• successive contamination: one scribe used two or more manuscriptssuccessively. This can be determined by the shoek wave method,in which various stages of the manuscripts are compared witheach other. If a manuscript changed from one to another Vorlage,the outeome of this method will clearly show it.*'' The shock wavemethod did not reveal any successive contamination within thesami")le of Targum Samuel.

6. Method of Ineluding Vocalisation

The use of variant vocalisation is said to be fraught with difficulties.The comparison of the simple and complex systems was made by ex-cluding the presence and place of the Shewa quiescens, Dagesh andother extra indicators (valuation 9)." ^ Two difficulties remained:'*^

• The scribe did not always vocalise the text himself, which is prob-ably true for MSS Or. Fol. 1-4, Solger 3.2°, Reuchlin 3, Kennicott5 [85], Jena El f.G, and Hebreu 75. *

. -Problems,' pp. 22-23.f. Smelik. -Trouble in the Tree.s!' 196.f. E. Wattel and M. van Mulken. 'Shock Waves in Text Traditions," in: P. van

Reenen and M.J.P. van Mulken (eds.). Studies in Stemmatology (Amsterdam: JohnBenjamins, 1996), pp. 105-121; an example is given in A.A. den Hollander, 'Howshock waves revealed successive contamination: A cardiogram of early sixteenth-century printed Dutch Bihles.' in: P. van Reenen et al. (eds.). Studies in Stemma-tology II (Amsterdam, Philadelphia: John Benjamins, 2004), pp. 99-112.

If these signs would have been included, it would, for example, divide theYemenite tradition into two categories: one not using these signs (MSS London2210, 2371 and Kapach 2) and the other using them (MSS London 1471, 1472 andKapach 63/2).

•'^Smelik, 'Trouble in the Trees!', p. 256.The name of the naqdan and the date of vocalisation is mentioned in tlie Jena


• The scribe or punctuator used a vocalisation system unknown tous, or invented the vocalisation himself and was not very con-sistent in doing so. which is probably true for MSS Reuchlin 3,Salamanca 1 and 7542 Madrid.

The first problem was solved by making two stemmas: one based onthe consonantal text only and the other based on the vocalisation. Asystem was invented for separating the data for the two stemmas. Thevocalisation variants were therefore given a different valuation system(7-9).

The second problem was solved by comparing pronunciation pat-terns {valuation 7) and not merely vocalisation (valuation 8). By study-ing the vocalisation systems of Codex Reuchlin and of Alfonso deZamora it is possible to detect whether these manuscripts really hada different pronunciation, or only different vowel signs. I consideredthose as orthographical data, which is not included in the consonantalstemma as well. All variants had to be evaluated at least three ^

• on the consonantal text regardless of grammar and orthography(valuation 1-6)

• on the pronunciation regardless of its exact voealisatioii and thepresence of Shewa or Dagesh (valuation 7)

• on the exact vocalisation regardless of the presence of Shewa orDagesh (valuation 8)

Example no. 1There is a variant in 2 Sam. 3.19 between rrn, n'2l and "2. WereAbner's actions right in the sight of "the House' of Benjamin, or inthe sight of 'the [people] of the House' of Benjamin, or in the sight of'the sons' of Benjamin? The variants in the following table are undereonsideration.^^

mann.scrij)t: a century after the original writing. The other manuscripts show signsof a separate naqdan, becaiLse thp vowels often correct the consonantal text. InCodex Reuchlin it is clear that the naqdan of the Aramaic text was different fromthe one of the Hebrew text, since a different set of vowel signs was used. Thedifference in writing the Qanies is the most clear feature.

^U made a valuation system including grammar, orthography and punctuation.Erst in order to make the system complete and secondly to avoid complete absenceof any relationship between manuscripts.

' The punctuation is transliterated as follows: E = Sere: e = Segoi: : = Shewa;.= Dagesh.


those of the housed.;bEytBl Dl G2H7.I1d.:bEy- 'S3d:bEytEl B2 L22 L23 L72 Q2d.:beytR3

houseb.EytL2G K5

bEytL71 Q6

sonsb.:nEyL75 Zl Z7

This resulted in the following sets of related and unrelatedmanuseripts.^'* The variants in the consonantal text are ineluded inthe calculations for the stemma. the variants in punctuation are not.

II 3.19 consonantsBl Dl G2 H7 Jl El B2 L22 L23 L72 Q2 R3 S3 L26 K5 L71 Q6 -L75 Zl Z7valuation 1 (major semantic variant: between 'house' and "sons';

II 3.19 consonantsBl Dl G2 H7 J l El B2 L22 L23 L72 Q2 R3 S3 ^ L26 K5 L71 Q6valuation 3 (minor semantic^^ variant: between 'house' and '[those] of thehouse")

11 3.19 punctuation not includedBl Dl G2 H7 Jl R3 S3 — El B2 L22 L23 L72 Q2valuation 9 (punctuation variant within the variant '[those] of the house'

II 3.19 punctuation - not includedL26 K5 ^ L71 Q6valuation 9 (punctuation variant within the variant 'house')

Example no. 2A second example consists of the word IQSI in 2 Sam. 3.21. It does nothave any consonantal variants, but at least two categories of pronun-

- marks an abbreviation.•'''*The variant between Segol in R3 and Sere in the rest of the MSS does not count.

because there is only one deviating MS.''•'One could also choose the "minor syntactic variant', as did Smelik. 'Trouble

in the trees!' p. 187. The choice of the 'minor semantic variant' is based on theassimiption that the variant ' [those] of the house' is a plain explanation of the col-lectively used word 'house*. See E. van Staalduine-Sulman, The Targmn of Sarnuel(SAIS 1: Leiden: E.J. Brill, 2002), p. 108.


ciation pattern exist. Tlie first pattern has a mute vowel in the secondsyllable, the seeond pattern has it in the first syllable.'*^

wa):amarBl B2 Dl G2 K5 L26 L75 S3 .11 ^wa):marL22 L23 L71 L72 Q2 QGwA)amarZlwa)amArZ7

w:)amarElw:)AmarH7w:)AmArR3

The abbreviation in MS Jl does not matter, because the voealisation iscomplete. The exact vocalisation of the Yemenite manuseripts does notmatch that of the Western tradition. However, the pronunciation iw thesame, since the Hatef Patah does not exist in the Yemenite vocalisationtradition. Knowing that Alfonso de Zamora did not use the Patah andthe Qames in a consisteTit way and noting that he almost consistentlyused the Patah instead of the Hatef Patah,'' ^ I concluded that the twovariants of Zamora represent the same pronunciation patterns as theother MSS in this category.

II 3:21 pronunciation patternBl B2 Dl G2 K5 L26 L75 S3 Jl L22 L23 L71 L72 Q2 Q6 Zl Z7 ^ El H7R3valuation 7 (two categories of pronunciallon)II 3:21 vocalisationBl B2 Dl G2 Kf) L26 L75 S3 Jl ^ L22 L23 L71 L72 Q2 Q6valuation 8 (the exact matching of vocalisation)'''''

The other category is more diverse, since MS H7 is very consistent inusing the Qames in this verbal form. Codex Reuchhn, however,*" is

'•''The punctuation is tran-sliteratcd a,s follows: A = Qamos; a — Patah; : = Shewa;:a — Hatef Patah.

^^Jl reads; wa);ama-. The - marks an abbreviation.•''' Almost consistently, for he rarely used the Hatef Patah in a verbal form, but

consistently used it in the interrogative prefix and in the word 'IX.•''"The variants of Zl Z7 El H7 and R3 are not taken into account, because

singular readings do not reveal relatedness. cf. n. 11.^Inconsistency in the use of Patah and Qanies is typically Sefardi. because in

these areas Hebrew lacked the distinction in pronunciation between Patah and


very inconsistent in its use of the Patah and the Qames. However, thiscategory is easily distinguished, because of its Shewa in the first sylla-ble. It gave rise, then, to the sets of related and unrelated mamiscriptslisted above.

Example no. 3A final example is the Greek word used for the Torah (1 Sam. 8:11). Ba-sically it is written NDQ], but the matres lectionis are added in variantways, which do not always match the vowels. A survey of place andpresence of the matres lectionis compared to the survey of the pro-nunciation will show the differences. Mss Dl G2 and L75 show thatthe voealisation is not always in accordance with the consonantal text.D r represents the consonantal reading, Dl" the correction, which wasprobably made by the punctuator.*^^

matres lectionisY -

niyniosA)Bl S3

niym.osA)QG

uiyni:sA)Dl"

Y W

niymowsA)B2 H7 K5L26R3niym.owsA)Jl

nymws)Dr

- wnimows A)L72

W ~

niwiuisA)*^^G2 '

nuwnisA)L23 L71 Q2Elnwum:sA)Zl Z7nwnis-*'''L75'

numsA)L22

There are five ways of writing the consonants of this word, but onlythree ways of pronouncing it. The pronunciation 'iiimsa' may be an

Qames. It is. however, also a frequent phenomenon in Ashkenazi Hebrew, see J.Oslzowy-Schlanger, Les manuscrits hebreux dans VAngUiterre me.dievale: etude his-torique et paleographique (Paris, Louvain: Peeters, 2003), p. 133.

''^The punctuation is transliterated as follows: A = Qanies; i — Hireq: o — Holem;u — Shiireci (supralinear) or Qibbus (sublinear); : — Shewa. In Western manuscriptsthe combination of 'nw' refers to the Shnreq coml)iiied to the Waw.

^^The consonants suggest the pronunciation of 'numsa', but the punctuator ig-nored the Waw and vocalised it as 'nimsa'.

''^The consonants suggest the pronunciation of 'numsa', l)nt the punituator ig-nored tlie Waw and vocalised it a.s "niinosa".


Ashkenazi variant of 'numsa", because the Sureq was pronounced as aHireq in Western and Gentral Ashkenaz.''*

pronunciationNIMOSAniyinosA)Bl S3niym.osA)Q6niymowsA)B2 H7 K5 L2G R3uiym.wosA)JlnimowsA)L72nim.osA-L75^

NIMSAniym:sA}''-^Dr'niwmisA)^^G2 '

NUMSAnuwmsA)L23 L71 Q2 Elnwuni:sA)Zl Z7numsA)L22niwni:sA)G2^nwms-

Finally, there is a difference in vocalisation because some MSS doublethe Mem. These three characteristics {matres lectionis, pronunciationand punctuation) led to the sets of related and unrelated manuscriptsin Table 2.

This method is of course open to criticism, but it gave me theopportunity to group the manuscripts, even if every single manuscripthas a different set of consonants and vowels. And looking back I cansay that the results support the method.

7. A Stemma of Pure Vocalisation

On the basis of the method mentioned earlier, Wattel made three stem-mas of the material, one of the consonantal texts, discnssed above, andtwo others of the vocalisation and pronunciation jjatterns respectively.The consonantal stemma was discussed in section 4. The stemma based

S. Morag, Pronunciations of Hebrew\ EJ, Vol. 13 (1971), p]). 1120-45,(1128 for Hebrew). The naqdan may have punctuated tiie word foneiically.

'' 'The consonants sugge.st the promiticiation of 'nimosa". but the punctuatorerased the Waw and vocalised it as 'nimsa'.

^••The consonants suggest the pronunciation of 'numsa', but the punctuator ig-nored the Waw and vocalised it as 'nimsa'.

three groups because a single variant does not indicate relationship.

VAN STAALDUINE-SULMAN Vocalisaiion in Stemmatology 233

I 8:11 consonants not includedBl S3 Q6 D r ^ B2 H7 K5 L26 R3 D r Jl ^ G2 L23 L71 L72 Q2 El ZlZ7 L75''valuation 6 {matres

I 8:11 pronunciationBl B2 S3 H7 K5 L26 R3 J l L72 Q6 L75" ^ G2 " DP' ^ L22 L23 L71 Q2El Zl Z7 LlTfvaluation 7 (pronunciation regardless of exact vocalisation)

I 8:11 punctuation - not includedBl B2 S3 H7 K5 L26 R3 L72 ^ Jl Q6valuation 9 (pvmctnation within the 'niniosa"-group. i.e. presence of theDagesh)I 8:11 punctuation not includedZl Z7 ^ L22 L23 L71 Q2 Elvaluation 9 (punctuation within the •nuinsa'-group, i.e. presence of theShewa)

Table 2; Related and Unrelated Manuscripts

on pure vocalisation proves to be unrealistic (cf. fig. 2). It consists offour groups, of which only the Eastern tradition represents reality:

• The first group consists of the Babylonian and Yemenitemanuscripts. The Babylonian MSS could have been the inter-mediate between the Yemenite tradition and the Western trarditions. Furthermore, it is striking that MSS 1472 and 2371 arevery closely linked and standing at the eud of the family, as intiie consonantal stemma. The outstanding relationship betweenthese two manuscripts could have been enhanced by the identicalwriting of the tetragrainmaton. The other Yemenite manuscriptsare linked differently from the consonantal steuima. Yet, the factthat a family of Yemenite manuscripts is discernable proves thatcopying within this tradition was not only a matter of consonants,but also of vowels. There must have been a scribal tradition ofvocalisation.''^

• The second group consists of three outsiders: MS Reuchlin andboth manuscripts made by Alfonso de Zamora. It can hardly becalled a group, for the Zamora manuscripts show evidence of some

**Cf. S. Morag, 'Pronunciations of Hebrew'. E.I Vol. 13 (1971). 1120-45 (1124for Hebrew).


a 102

.1-3 V* 54 26

25 4B

1 @B2

27 2 7 1.9 1,1 4 a 3;7 4;7 3 3

L?! @CI DI -126 'B? •L75 G2 K5

•L72 -123

Figure 2: Stemma of Pure Vocalisation

relatedness, while My Rouchlin is rather unrelated to these two.This group was predictable because of their uncommon way of


vocalisation. The fact that both MS Reuchlin and the two Zamoramanuscripts did not consistently use the Patah and the Qamesmust have caused incidental contamination.

• The third group consists of two Ashkenazi manuscripts and theSolger-Rabbinic Bibles string. These two Ayhkenazi manuscriptsare not linked in the consonantal stemma. The string of MS Solgerand the first Rabbinic Bibles is real, but it is given the otherway round. The vocralisation of the second Rabbinic Bible showsthe closest relationship with the Ashkenazi tradition, Solger theloosest.

• The fourth group is the most unlikely: The Sefardi MS Opp Adds4" 75 is linked to the Ashkenazi MS 11 from Gottweig, whilethe Sefardi MSS Kennicott 5 and Hebreu 75 are linked to theAshkenazi MS El f.6.

It is clear from these results that pure vocalisation is not very heli>fnl in building stenmias. The invented vocalisation systems cannot beincluded. The Western branch of the tree got muddled, and only theSolger group is clearly distinguishable. This leads to the conclusion thatthe Western copies of Targum Jonathan reflect the absence of a scribalnorm for vocalisation—even of a strict personal norm (Zamora).^^

8. A Stemma of Pronunciation Patterns

The stemma based on pronunciation patterns shows more interestingfeatures (cf. fig. 3). Again the basis of the stemma is unsure because ofthe small amount of evidence, but at least tlie families are discerniblein their real state.

• The Yemenite family is present, enlarged with the BabylonianMS Eb 1. The latter is again closer to the Western manuscripts

^^A similar yet atrouger conclusion was drav/ri with regard to French Percevalmanuscripts from the thirteenth and fourteenth century: 'When MSS. have madethe same itinerary, at approximately the same moment, we do not find essential dif-ferences in variation between one copyist copying the same exemplar and differentcopyists copying one and the same exemplar. Absence of a norm makes individ-ual variations so frequent that inde])endent copyists and one and the same copyistworking on the same text behave equally freely when copying', cf. L. Schosler.'Scribal variations: When are they genealogically relevant - and when are they tobe considered as instances of "mouvance"?' in: P. van Reenen et al. (eds.). Studiesin Stemmatology II (Amsterdam, Philadelphia: .John Benjamins. 2004). pp. 207-26(223).


J l L26

Fignre 3: Stemma of Pronunciation Patterns

than the Yemenite are. The Yemenite manuscripts are given inthe same order as in the consonantal stemma. Again MSS 1472and 2371 are very closely linked and standing at the end of the


family. " The only difference with the consonantal stemina is thefact that MSS 2210 and Kapach 63/2 are not linked together,although they are next to each other. This similar outcome isnot surprising, since even the pure vocalisation of this family isvery consistent.

• The Sefardi family is enlarged by MS Reuchlin, which is float-ing within the Sainiiel stemnia.' ^ This result calls for furtherinvestigation, as noted above. Whereas MS 0pp. Adds 4° 75 isclose to Zamora's consonantal text, it must give up this place toMS Kennicott 5 with regard to pronunciation patterns. It evenseems to stand alone with regard to pronunciation. The abso-lutely closest relationship is not surprisingly found betweenthe two manuscripts of Alfonso de Zamora, both in consonan-tal text and in pronunciation patterns. This relatedness is evengreater than the internal coherence of the Yemenite manuscripts.This means that alttiough he used his own vocalisation systemand was not very consistent in that, he was firmly rooted withinthe Sefardi pronunciation tradition. If Zamora invented his vocal-isation system, he only used it to write down the traditional Se-fardi pronunciation. This also means that the distinction betweenpronunciation and mere vocalisation is very helpful in stemma-tology.

• The Ashkenazi family is linked to the Solger group, although thelatter is still standing on its own. The Ashkenazi manuscripts aresubdivided into two categories, as in the consonantal stemma: anearly pair of manuscripts and a late pair. The early pair con-sists of MSS 26.879 and El 6." ^ The late pair consists of MSSOr.Fol. 1-4 and 11 (883) Gottweig.'^ The faet that MS Or. Fol. 1-

™The linked riiaiiuscriptH are valuated between 1 and 200: the closer to i the(•kxser the relationshi]) between the nianiiscripts. The Yemenite manuscripts arevahiated between 15 and 27 in the consonantal stemma and between 13 and 26in the pronunciation stemma, in contrast to the Sefardi branch with valuation.sbetween 34 and 58. Only the two Zamora textn show the closest relationship: 9 inthe consonantal stemma and 11 in the proiumciation stemma.

^' It is rather peruliar that Codex Reuchlin is linked t(j tlir Babylonian niaiius(-riptin the consonantal stemma, while the vocalisation also shows traces of Babylonianinfluences.

^^Which is confirmed in Houtnian, 'Different kinds of tradition,' p. 276.^•'Which is not confirmed in Houtman. 'Different kinds of tradition,' p. 276.

Time may have played a role in the formation of a real Ashkenazi tradition, sinceit is known that the Ashkenazi pronunciation of Hebrew only evolved in the 13th


4 was vocalised at least one century after its production has hadno impact on its place in the stemma. It was clearly vocalisedwithin the same tradition as the consonantal text. Although itoffers many corrections to the consonantal text, probably madeby the punctuator, there is no reason to believe that tbis punc-tuator used an entirely different VorlageJ^ The Solger group isvisible, but the only dissimilarity between tbe consonantal andthe pronunciation stemma is the place of the second RabbinicBible. Whereas tbe consonantal stemma suggests tbe most prob-able historical development, the pronunciation stemma suggeststhat the vocalisation of the second Rabbinic Bible is also influ-enced by other Western vocalisation traditions. Being printed inItaly it may bave undergone Asbkenazi influences. The place inthe stemma may also point to Sefardi influences. Thiti is not sur-prising, hecause tbe editor, Jacob ben Hayyim, was an exile fromSpain; liiy family had been driven out in the great expulsion of1492.

9. Conclusions

Comparison of the three stemmas—consonants, vocalisation and pro-nunciation patterns -leads to the conclusion that vocalisation can playa role in stemmatology. Pure vocalisation does not give reliable data,but pronunciation patterns do. The main families of a consonantalstemma correspond to those of a pronunciation stemma, except for thejjosition of Codex Reuchlin which nuist be studied in greater detail.

The vocalisation stemma shows that tbe Yemenite tradition is notonly very coherent in its consonantal tradition, but likewise in its vo-calisation. This must be due to a largely written tradition, since thestemma based on pure vocalisation also gave evidence of a consistentYemenite family.

Stemmatology of vocalisation and pronnciation was helpful in thecase of the second Rabbinic Bible, wbich appears to have been influ-enced by traditions other than tbe first Rabl)inic Bible only.

tPutury from a more Sefardi [jronunciatioii. The two early manuscripts may not befully Ashkenazi in this regard. See S. Morag, 'Pronunciations of Hebrew', EJ. Vol.13 (1971), pp. 1120-45 (1129-30 for Hebrew).

^'*0r he must have used a Vorlage which was very close to the consonantal Vor-iage; contra Snielik, The. Targum of Judges, p. 122; Van Staalduine-Sulman, TheTargum of Samuel, p. 54.


The fact or the assumption—that a vocalisation is not from thesame hand and/or does not originate in the same century as the conso-nantal text, has no dominant influence on the stemma. We suspected adifferent punctuator in MSS Or. Fol. 1-4, Solger 3.2*', Kennicott 5, JenaEl f.6 and Opp. Adds 4" 75, but these manuscripts did not vary exten-sively in the pronunciation stemma. We may safely assume, then, thatthe first difficulty with vocalisation—that of the separate punctuator—is not well-defined. Whether there is a separate punctuator is not theproblem, but whether the punctuator whoever he was—stood in thesame local tradition. This can only be established by making a pro-nunciation stemma.

Therefore, we must also assume that the Western vocalisation tra-dition was not a written, but primarily an oral tradition. Regardlessof the precise spelling, the prommciation patterns appear to be di-vided along the same linos as the consonantal patterns. As long asmanuscripts remained within the same school of scribes, their vocali-sation might differ, but the pronunciation patterns stayed the same.

Tlie fact that punctuators used an unknown vocalisation system—or even invented one—appears to have no dominant influence on thestemma either. We know that Alfonso de Zamora. fairly inconsistently,used his own system and that Codex Reuchlin has a vocalisation systemof its own. The position of the Zamora manuscripts is identical inboth the consonantal and the pronunciation stemmas, including theirnumerical valuation: the pronunciation patterns are as much related asthe consonantal texts. The position of Codex Reuchlin has moved from'intermediate' to Western. Both positions need further investigation.We may assume, then, that the second difficulty with vocalisation—that of the invented or inconsistent vocalisation system—is not as greatas we thought. Furthermore, we conclude that the distinction betweenpronunciation patterns and vocalisation appears to be valid.

To put it briefly, addition of vocalisation—or rather: pronunciationpatterns—to stemmatology may well be helpful in refining a stemma ofTargnmic manuscripts and in establishing whether the punctuator useda different Vorlage.'^^ The only two problems that remain are, first, theamount of time needed to collate manuscripts including vocalisation

same tenor is found in W.F. Smelik, 'How to Grow a Tree: ComputerisedStemnia,tology and Variant Selection in Targnni Stndiea', in: Johaitn Cook (ed.),Bible and Computer: The, SteAltnbosch AIBI-6 Conference: Proceedings of the As-sociation Internationale Bible, et InfoiinaHquf. " Frovi, Alpha to Byte\ University ofStellenbosch 17-21 July, 2000 (Leiden: E.J. Brill, 2002), pp. 495-518 (512-13).


and, secondly, the difficulty of comparing the findings with unvocalisedmanuscripts.

[AS 3.2 (2005) 241-83]DOI: 10.1177/1477835105059096

T H E LITERARY NATURE OF THE BOOK OF DANIELAND THE LINGUISTIC CHARACTER OF ITS ARAMAIC

Jan-Wim Wesselius

Theologisrhf Universitcit Kamprn

The biblical book of Daniel tells its readers about the hfe and vi-sions of Daniel, a .ludean courtier and visionary, and of three friends ofhis during various short periods iu the reigns of the Babylonian kingsNebuchadnezzar (in about 605-603 BCE, maybe also a little later) andBelshazzar (in about 541-539 BCE), an enigmatic king Darius, 'theMede' (539) and the Persian king Cyrus (in about 539-537). Daniel'slong life (according to the information provided by the book at leastca. 622-537) encloses the entire episode of the Exile of the Judeans andof the inhabitants of Jerusalem (about 587-539), though the Exile ismentioned only implicitly, if at all.

'Earlier versions of parts of thiti article have appeared as 'DanieF [in Dutch] inJ.P. Fokkt'lnian and W. Weren (eds.), De Bijbel Literair (Zoetermeer: Meinema,2003), pp. 251-261, and us "Hoe de valse beschuldiging een echte besehuldigirig wordt.De relatie tussen de besehuldigingen in Genesis 38-44 ^ ' Daniel 3 en 6, en de bedoel-ing van het boek DanieV [Dutch], in C. Hontman (ed.), De leugen regeert,... Valsebesehuldiging in de Bijbel en in de wereld van de Bijbel (Kampen: Kok, 2004),pp. 58-80. Some of the argnmeiits presented here are already in J.W. Wesselius,'The Writing of Daniel', in J.J. Collinti and P.W. Flint (eds.). The Book of Daniel:Cornposttion and Reception. 2 vols. (FIOTL 2; Leiden: Brill, 2001). pp. 291-310,id., "Discontinuity. Congruence and the Making of the Hebrew Bible', Scandina-vian Journal of the Old Testament 13 (1999), pp. 24-77, and in earlier publicationsquoted there. Detailed discussions of many issues touched upon only briefly herewill appear iu my forthcoming monograph. Language, Style and Structure in theBook of Daniel. The present article is meant to be a sucrinct. but ba-sirally com-plete presentation of my new model of tlie literary and linguistic nature of the Bookof Daniel. Biblical texts will be quoted according to the Revised Standard Version(RSV), with a more literal translation of the Hebrew and Aramaic where that isuseful for the argument.

' But it should be noted that according to the chronology of the book itself tlie

© SAGE Publications (London. Thousand Oaks CA. and New Delhi), 2005


The book is in its present form best described as a kaleidoscopicwork. The story is not told in one continuous text, but in ten episodeswhich exhibit great variation in language, style and literary genre. Themain division in the book is between six stories about Daniel and histhree friends Hananiah, Mishael and Azariah in the first six thapters,and four accounts of visions seen and told by Daniel in chs. 7-12 (10-12 clearly belong together as one unit). The first two chapters and theaccounts of the visions are dated by king and regnal year, the othersare dated mainly by the mention of the ruling monarch. The bookthus looks like a kind of dossier abo\it Daniel, with various documentsabout episodes in his life aud his visions in more or less chronologicalorder, with only loose connections between them, usually in the formof references to the events of earlier chapters.^

In Daniel 1 the readers are told that king Nebuchadnezzar of Baby-lon lays siege to Jerusalem and takes it in the third year of the Judeanking Jehoiakim, which according to Jer. 25.1 should be identical withNebvichaduezzar's accession year, the period between his actually be-coming king and the beginning of his first regnal year at the BabylonianNew Year festival in the autuuiu.'* On this occasion the vessels of theTemple and probably also Jehoiakim himself (the text is ambiguous)are brought to Babylon. This event probably corresponds with whatis narrated in 2 Chronicles 36.6, where it is noted that Nebuchadnez-zar 'bound him in fetters to take him to Babylon'. It is not entirely

Exile lasted for 70 years, wbicb would make Daniel's life and career even longer,see also below.

There is a vast arnounty of literature about the book of Daniel, of which I canmention only a few titles here: G. Ch. Aalders, Daniel [Dutch] (COT; Kampen: J.H.Kok, 1962); A. Bentzen, Daniel (HAT, 19; Tiibingen: Mohr, 2nd ed., 1952); J.J.Collins, Daniel: A Commentary on the Book of Daniel (Hermeneia; Mineapolis:Fortress Press. 1993); P. R. Davies. Damel (Sheffield: JSOT Press, 1985); .1. E.Goldingay. Daniel (World Biblical Commentary. 30; Dallas: Word, 1989); J. A.Montgomery, A Critical and Exegetical Commentary on the Book of Daniel (ICC;Edinburgh: Clark, 1926). Two collections of important articles: A.S. van der Woude(ed.). The Book of Daniel in the Light of New Findings (Lenven: Peetens. 1993)and Collins and Flint, The Book of Daniel: Composition and Reception. Specificallyabout the literary character of the parts of the book: S. Talmon, 'Daniel', in: R.Alter k F. Kermode (eds.), The Literary Guide to the Bible (Cambridge, Mass.:Belknap Press, 1987). pp. 343-56.

^Note that this veiy precise date is almost forced onto the reader, because Daniel2, at least three years after the events of the beginning of ch. 1 {Dan. 1.5), is datedto Nebuchadnezzar's second year (2.1). See already Bentzen. Daniel p. 17, whorightly noted that there is :io contradiction between the dates.

WESSELIUS The Literary Nature of the Book of Daniel 243

clear whether this is a third conquest of Jerusalem in the year 605 BCE(beside the well-known occasions of 598 and 587), or a purely literaryreflection of the episode in 2 Chronicles 36 (it is interesting to note thatthe ambiguity about .lehoiakim's fate is found there also)."* In Baby-lon, the story in Daniel continues, some of the Israelite princes wholive there (and at least some of which have apparently been broughtthere on the same occasion) are chosen to be educated in 'the lettersand language of the Chaldeans"; this last word always indicates theclass of professional scholars and dream interpreters in the book ofDaniel. Among them are the Judeans Daniel and his three companionsShadrach, Meshach and Abednego. They refuse to be contaminatedby the food of the court, which does not meet the standards of theirreligion (a connection with kashrut seems to be suggested, but is notnoted explicitly), and show that they can thrive on vegetables and wa-ter. Finally they are introduced to the king, and he is amazed by theirwisdom and knowledge.

In the second chapter the king has a dream, and, after they un-succesfully try to persuade him to tell the dream to them, he wantshis Chaldeans to tell him both the dream and its interpretation. Whenthey prove unable to do so he gives orders to kill all the wise menof Babylon, including Daniel and his companions. Daniel, however, isinformed about the dream and its interpretation through divine inspi-ration, and he tells them to the king. A huge statue of the materialsgold, silver, bronze, iron and iron mixed with clay, which the king sawin his dream, symbolizes the empires which will arise in the world fromthe Babylonians onward. In a situation of great political division thekingdom of God, represented as a great stone in the dream, will de-stroy all those empires. Deeply impressed, king Nebuchadnezzar giveshonour to Daniel s God and appoints him and his companions to highpositions.

The third chapter starts with another huge statue, this time a realgolden one which the king ordered to be set up, for which all his sub-jects must bow down on the penalty of being thrown into a fiery oven.Certain Chaldeans accuse Daniel's three friends (surprisingly, Danielhimself is absent from the story) of refusing to do obeisance to thestatue. In a conversation with the king the three men concede this,but persist in their refusal. The king gets very angry and orders hisservants to throw them into the oven. An angel, however, saves them

^Collins, Daniel: A Commentary, pp. 133-34.

244 Aramaic Studies 3.2 (20U5)

and when they come out of the oven unscathed, the king orders all hissubjects to respect the God of Daniel.

Daniel 4, couched as an edict or proclamation of the king to allthe nations of the earth, is the last chapter in which Nebuchadnezzarfigures. He has a dream of an enormous tree which is cut down, withonly its roots remaining in the earth, whic h his wise men are unable tointerpret, but Daniel explains it: the king will become insane and bedriven away from other humans, until the moment he recognizes thepower of God. Thus it happens and the king again orders obeisance toDani(>rs God.

The events of Daniel 5 take place on the very last day of tlie king-dom of Babylon. During a royal banquet king Belshazzar gets drnnkand orders to bring the vessels of the Temple in Jerusalem so that heand his courtiers can drink from them. At once a hand appears whichwrites an inscription on the wall of the palace. Daniel is able to readand interpret the inscription: it contains three names of units of weight,mina. shekel and half-shekel, which he interprets as referring to the endof the reign of Belshazzar and of the kingdom of Babylon. In that verynight the city is taken by the Medes and the Persians and Belshazzaris killed. The king of Persians and Medes, 'Darius the Mede', who isoften supposed to l)e personally referred to in the inscription,^ thenreceives the kingship over Babylon.

Many aspects of ch. 3 return in Daniel 6. Again there are certainunnamed Chaldeans who utter a denunciation, in this case that Danielkeeps praying to his God in spite of a royal edict, treacherously pro-posed by them to the king, that everybody is to refrain from making arequest to any human being or god for thirty days. Darius has no choicebut to let Daniel lie thrown into the lions' den. But when he coniesback to the den on the following morning, it turns out tliat Daniel hasbeen saved by an angel, and the king orders the accusers with theirfamilies to be thrown to the lions. Like Nebuchadnezzar before liim,king Darius now gives praise to Daniel's God.

The atmosphere of tlie prediction of Daniel 2 returns in ch. 7, datedto the first year of king Belshazzar. This time tlu^ riddle is not presentedin a dream of the king, but in one of Daniel himself, which is explainedby a heavenly being. Four animals symbolize again four empires, the

• K. Galling, "Die 62 Jahre des Meders Darius in Dan 6.1', ZAW 66 (1954), p.152: the sum of the weights mentioned in the inscription is probably 62 shekels,which number coincides with the age of Darius at tlie time he conquers Babylon(Dan. Tj.iil [liSV 6.1]).

WESSELIUS The Literary Nature of the Book of Darnel 245

last of which will have ten kings. The last of these kings will be so evil,that he will disturb even the order of nature itself. His power, however,will be ended through the coming of God's empire, when judgment andpower will be given to the 'Holy of the highest', probably a designationfor the people of Israel. Both here and for the following chapters we willpresent only an outline of the detailed visions and their explanations.

In the third year of Belshazzar Daniel has a new vision (Dan. 8),this time apparently not in the form of a dream. A ram with one hornis defeated by a he-goat, and from the latter's horn arise four others.after whi<'h from one of tliese comes a small horn, which does terriblethings and stops the service of sacrifices. The angel Gabriel gives theexplanation: the he-goat is the king of Greece, who will defeat the kingsof Persia and Media. The Greek empire will be divided into four parts,and at the head of one of these there will be a very evil king, who willeven oppose God himself, but will be utterly defeated in the end.

When Daniel, in the first year of 'Darius the Mede\ reads the wordsof the prophet Jeremiah that seventy years will pass after Jerusalem'sdestruction (Dan. 9.2), in a prayer he confesses sins on behalf of hisentire nation and begs for the restoration of Jerusalem. Again Gabrielappears and he explains that on the one hand the word of Jeremiah hasbeen fulfilled at the beginning of Daniel's prayer, but that on the otherhand this word refers to the remote future, not of seventy years but ofseventy 'weeks' of years, 490 years, until the days of an evil monarch,who will interrupt the sacrificial service.

The last episode in the; book takes up three complete chapters (1012). In the third year of Cyrus, the king of Persia, Daniel sees a visionof a celestial being. He falls asleep and when he wakes up, variousdivine beings speak with him, and explain the further eonrse of historyin great detail, of the Persian empire until the Greeks defeat it, andof the divided Greek empire again, but this time with the addition ofnumerous details about the evil king, who will interrupt the sacrificesand will commit all kinds of evil acts, until he will come to his endwithout human intervention. After the great oppression there will bea time of salvation for Daniel's people. Daniel is ordered to seal thebook until the time of the end, and is jsromised that he himself willrise again at that time (12.13).

It has often been noted that the book is much more interestedin the sequence of the great world em])ires. and in the time scheme ofevents in world history, than in sp(K;ifically Israelite history. Though noteverything is as clear a-s we would like it, the eschatological age which


is referred to time and again seems to be identical with the period ofoppression of the Jewish religion during the reign of the Seleucid kingAntiochus IV Epiphanes around 165 BCE. As noted from antiquityonwards, the description of second century BCE history is correct up tothe great persecution under Antiochus, after which an eschatologicalage is expected. The book as a whole thus presents itself as pointingto an unfulfilled expectation of the end of history and the coming ofGod's kingdom around that time on the basis of the predictions otDaniel in the sixth century BCE. Whether this means that the bookwas at least partly written with this expectation in mind, in the shortperiod between the persecution and Antiochus' death, as maintainedby most critical scholars,^ is outside of the scope of this literary andhnguistic study. The events in the book itself range from the accessionof king Nebuchadnezzar to the third year of king Gyrus; see below fora possible explanation of this time-frame.

In spite of the apparent unity of the contents of most of the bookof Daniel we see, as noted above, a bewildering variety of language,narrators and style in the twelve chapters of the book, which made alarge majority of scholars agree that the book was put together by oneor more redactors from various earlier materials, while they did notwant to interfere too deeply in the texts which they used.''

The book contains parts in two different languages: 1.1-2.4 (untilthe word n"'D"iK, 'in Aramaic') and chs. 8 12 are in Hebrew, the part inbetween is surprisingly in Aramaic. The Aramaic part of the book lookslike a separate unity, among other things because a clear concentricstructure can be recognized: predictions about the course of historyin 2 and 7, martyrs' stories in 3 and 6. and enigmatic predictions tothe eastern kings about the direct future of their reign in chs. 4 and5.* This division, however, is clearly different from the one accordingto content, where, as noted above, the narrative chapters 1 6 are incontrast with the visions in 7-12.

Between the chapters themselves, moreover, there are considerabledifferences also. The stories in 1, 2, 3, 5 and 6 are told by an anonymous

''E.g. CoUins, Daniel: A Cnmm.entary, pp. 61-71.^One of the few scholarK of H non-fnndainriitahstic background to defend the

basic unity of Daniel in modern times was H.H. Rowley, in hi.s 'The Unity of theBook of Danier, HUCA 23 (1950 1951), pp. 233-73: reprinted in his The. Semantof the Lord and Other Essays on the Old Testament (London: Athlone. 1952), pp.235-68: 2nd ed. (Oxford: Blackwell, 1965), pp. 247-80.

" A. Lenglet, 'La structure litteraire de Daniel 2-7', Biblica 53 (1972), pp. 169-90.


narrator. As noted above, ch. 4 is entirely in the form of a letter, orrather a proclamation, of king Nebuchadnezzar to all the nations ofthe earth. In ch. 7 Daniel's vision, which is told in the first person (T,Daniel'), is introduced by an anonymous narrator, but 8 and 9 startin the first person without any introduction, whereas the anonymousnarrator again introduces the first-person account of 10-12. Maybe theend of ch. 7. where Daniel tells about his reaction on the vision andits explanation, should be regarded as a transition to the first-personaccounts of 8 and 9: in this way we are reminded that Daniel is stillspeaking. Another remarkable feature is that in ch. 3 the main personis not Daniel, but his three friends, while no reason is given why Danielis absent from this chapter, or the companions from ch. 4 and the restof the book. By itself not problematic, but still remarkable is that thechapters are not in the exact chronological order: the events of 7 and8 (dated to Belshazzar's first and third years) precede those of ch. 5,which describes the last day of his reign.

In the personal names we also see a certain degree of variation. Inch. 1 the readers are told that Daniel and his companions Hananiali.Mishael, and Azariah receive Babylonian names, namely Bcltesha/.zar,Shadrach, Meshach, and Abednego, respectively, but the use of thesenames in the sequel seems to be rather arbitrary. The name of theking who takes Babylon for the Persians in 539 BCE is probably givenin two forms also, as Cyrus, 'the Persian', the historical name of thisking, which we also encounter in the book of Isaiah (44.28 and 45.1),beside the enigmatic Darius, 'the Mede', the descendant of Ahasuerus(Dan. 9.1), certainly not the famous king Xerxes, who ruled much later(486-465), but either a fictitious earlier Xerxes or the great Cyaxares,the great-grandfather of Cyrus according to Herodotus.^ There hasbeen an enormous amount of discussion about the supposed questionof the identity of this Darius, but we should probably turn the reasoningaround: in an even minimally coherent literary text there is very littlereason to suppose that Cyrus and Darius are different persons if wefind the dates Darius 1 (Dan. 9.1) and Cyrus 3 (Dan. 10.1) one afterthe other, and the events at the beginning of the reign of 'Darius theMede' seem to be those usually associated with Cyrus (see below).B.E. Colless has in my view convincingly presented the case for the

^Herodotus, Histories, I, 103-16. About the question of this Aha^uerus see, forexample, the balanced verdict in CollinK, Daniel: A Commentary, p. 348, and theearlier literature mentioned there.


identification of the two, though in my opinion it remains a mysterywhy Cyrus could be called 'Darius'. One can note, however, that inthis way the character of the book as a 'linear composed dossier' (seebelow) would be underscored once more. In this case the difficult verseDan. 6.29 [RSV 28] should probably be translated as: 'So this Danielprospered during the reign of Darius, namely the reign of Cyrus thePersian'.^^ It can be added that for some of the arguments about thebook of Daniel which are presented below, an identification of the twois helpful, though never essential: in all these cases more or less thesame reasoning would be valid in the case of 'Darius the Mede' as areal intermediate figure between Belshazzar and Cyrus.

It is also notable that there are a few important lacunae in the bookitself. It describes events during the reigns of kings at the beginningand at the end of the exile, but as noted above the captivity itself,which started with the three (or two) displacements of exiles which arementioned in the Hebrew Bible, in 605, 597 and 587 BCE, and endedthrough the famous edict of king Cyrus in his first year of office whichallowed the exiles to go home again (2 Chion. 36.22-23; Ezra 1.1-4)is passed over almost in silence. This would seem to go a long waytowards explaining why the image of king Nebuchadnezzar in Danielis not the negative one to be expected for the person who destroyedJerusalem and its Temple, but a view which can be compared to therather positive one in the book of Jeremiah, cfr. Jer. 27.6, 'Nebuchad-nezzar, my servant'. We shall see below that there is more to the roleof Nebuchadnezzar in the book of Daniel. The provenance of Danieland his comjjanions is left in the dark. Not only are we not told exactlyhow and when they arrived in Babylon. !)ut the names of their fathersare not even given, which is highly unusual in the world of the HebrewBible, and would be even more so for princes from the royal house ofJudah (Dan. 1.3 and 6). It is suggested at the beginning of the bookthat they came to Babylon with king Jelioiakiin, but this is not saidexplicitly, and we can only guess after their family relationship withthe last kings of Judah, though it seems indeed likely that in the book

' B. E. CoUesK, 'Cyrus the Persian as Darius the Mede in the Book of Daniel',JSOT 50 (1992), pp. 113-26; cf. also L.L. Grabbe, "Another Look at the Gestalt of"Darius the Mede" ". CBQ 50 (1988), pp. 198-213; D. J. Wiseman, 'Some HistoricalProblems in the Book of Daniel', in: D. J. Wiseman a.o. (eds.). Notes on SomeProblems in the Book of Daniel (London: Tyndaie Press, 1965), pp. 9-18. Notethat the reasoning proposed by Colless is mainly of a literary nature, whereaseai'licr authors usually attempted .some sort of historical explanation.


of Daniel the prediction to king Hezekiah is fulfilled, 'And some of yourown sons, who are born to you, shall be taken away; and they sliallbe eunuchs in the palace of the king of Babylon' (2 Kgs 20.18 and Isa.39.7).^1

In view of these observations, it is understandable that most schol-ars suppose that a complex history underlies this seemingly highlyirregular state of the text of the book of Daniel. John Collins' state-ment in the introduction to his and Peter Flint's collection of essayson Daniel. 'It is generally agreed that the tales in Daniel 1 6 are olderthan the visions in chapters 7 12. and are traditional tales that mayhave evolved over centuries', is characteristic of the position taken bymost researchers, and is only the starting point for the assumption ofa number of editorial processes.^^ It should be noted, however, that inview of the disagreement about nearly every aspect of the supposedredactions, Philip Davies' words about the problem of the languagesdeserve attention: 'The presence, and the distribution, of the two lan-guages in Daniel may be in the end inexphcable', and might well havebeen extended to the entire problem of composition and origin of thebook, though I will indeed attempt such an explanation here.^^

Derivation of Table of Contents

The unity of the book of Daniel, however, is surprisingly confirmedby the observation that the structure of the entire book mirrors thelayout of two other works, namely the story of the life of Joseph inGenesis 37-5U and the biblical book of Ezra (i.e. the first part of thebook Ezra-Neheniiah of the Hebrew canon). This is a complex andvery interesting literary strategy, which explains many of the problemsof the literary form of the book: most of the discontinuous featuresnoted above, which have traditionally been explained as the outcomeof a complicated process of redaction underlying the present form ofthe book, turn out to be literary reflections of comparable traits of theother two compositions.

Two remarks need to be made about this. Firstly it may be use-ful to point out that this is not an instance of a relatively vaguely

'^Collins, Daniel: A Commentary, p. 135.^^.I.J. Collins, 'Current IssiieH in tlie Study of Daniel', in: Collins and Flint (eds.).

The Book of DanifX pp. 1-15 (5).' P.R. Davies, Daniel, p. 35. A brief survey of various opinions about the j)rehis-

tory of the book in Collins, Daniel: A Commentary, pp. 24-38.


defined process of influence' of one text on another, but that we aredealing here with a highly sophisticated literary emulation. As I willdemonstrate, the author of Daniel wrote his book in such a way thatas a whole it refers to other literary works within and outside of theHebrew Bible, and within this framework contains numerous allusionsto various elements in these other works. Secondly, the existence ofnumerous agreements between Daniel and other works, especially Ezraand the life of Joseph, is hardly something new. Many of them, in fact,have been commented upon in the vast literature dealing with the bookof Daniel. What was completely new, by contrast, when I first pointedit out. is that these agreements occur exactly in linear order: this is anobservation which can be repeated and is hardly subject to doubt, andthus requires some sort of explanation in any case.

The change of main person in Daniel 3 agrees with the same phe-nomenon in Genesis 38, the story about Judah and Tainar which in-terrupts the story of Joseph. Important elements of the narrative ofJoseph's life, such as all the revelations of what is to happen in thenear and remote future and all the accusations uttered against themain persons, return in corresponding places in the book of Daniel;see Figure 1. The nature of the predictions seems to be comparablein all cases: the three episodes of double dreams which Joseph has orwhich are explained by him correspond with three riddles posed tothe Babylonian kings, two dreams and one enigmatic inscription. Notethat in Daniel all three thus assume the character of Pharaoh's doubledream in Genesis 41: an enigma presented by God to the king, which isexplained by the Israelite courtier. Both in Daniel and in Genesis theexplanation of these riddles has consequences within the cycle of sto-ries itself, whereas the explanation of the others becomes meaningfulonly after its completion.

While these elements constitute part of the narrative texture ofGenesis 37- 50 only, they are the main motives in the correspondingchapters in Daniel. All the elements which occur in pairs in Genesishave been fused into one only in Daniel, such as the 'double dreams'of Joseph, of the steward and the baker, and of Pharaoh, the sexu-ally tinted accusations of Genesis 38 and 39 and the accusations oftheft in 42 and 44. By contrast, the nightly prediction of Genesis 46appears to correspond with the two visions of Daniel 7 and 8. In thesubstance of these elements some systematic shifts can be observed.We already noted the three riddles which God pnt before the kings,corresponding with three different pairs of dreams in Genesis. The four


different act-usations in Genesis 38. 39, 42 and 44 are fused into a pairof highly similar accusations of refusal to couimit idolatry iu Daniel3 (corresponding with the pair in Genesis 38 and 39) and forbiddenadherence to Daniel's own religion in ch. 6 (corresponding with Gene-sis 42 and 44). Some shifting of motives and elements can be observed

Daniel 2 T1

1 ,

3

!

J:

4

Aij

' 5

A

~ 6

A^:V-rtol

Genesis 38 39 40 41 42 T 43 44 45 46 47 48 49 50

Predictions referring to the near fijfure. given through riddles which are to beexplained (tnainly dreams), with consequences in the slory itself

Predictions referring to the remote future only

Accusations coneenung situafion caused hy the accuser

Punishments connected with accusations

Kspecial position of ehapfer with regard to mam person; connectionwith preceding ehapfer through mirroring of interaction

Figure 1: Intertextual Connections between Daniel and Cenesis

between the parts which are now similar in Daniel. The punishmentswhich followed or threatened to follow the false accusations of Genesis38 and 39 return in Daniel 3 and G, namely the punishment of burningand of being thrown into a prison or 'pit, hole' (Gen. 40.15. cf. forthis expression also Ex. 12.29). The story about Nebuciiadnezzar's firstdream, in Daniel 2, formally corresponding to Genesis 37 with Joseph'stwo dreams, not unexpectedly has assiuned certain characteristics ofGenesis 41, with the dreams of Pharaoh.'*

Another aspect of the book of Daniel where emulation of Genesisseems likely is the length of time covered by Daniel's career in the book:twenty years of Nebuchadnezzar's reign until the fall of Jerusalem (hisaccession year plus the 19 of 2 Kings 25.8), seventy years of exile (Dan.

'""See. for rxainplo. G.G. Lahoiitr. •Genese 41 ot Daniel 2: Question d'origine',in: Van der Woude, The Book of Daniel in the Light of New Findings, pp. 271-84.


Beginning of Aramaic with the word "ramit

Aramaic documents (Ezra) and Aramaic narralive episodes / chapters (Daniel)

Narrative in letter

Chapters which intemipt the chronological order

Reaction of main person as transition to episode in the first person without introduction

Chapters where the main person starts to speak without introduction

— — Confession of guilt

Figure 2: Intertextual Connections between Daniel and Ezra

9.2) and three years of king Cyrus (Dan. 10.10), probably making atotal of 93, exactly the same number of years as in the long careerof .Joseph in Genesis (Genesis 37.2 and 50.22);''' 'Darius the Mede'does not fit easily in this scheme, perhaps another argument for hisidentification with Gyrus.

Also in the relation with Ezra we see a close agreement betweenchapters, often also with an increase in the significance and a system-atization of the corresponding elements; see Figure 2. The six Aramaicdocuments of Ezra 4-7 (Ezra and Daniel are the only two books with asizeable amount of Aramaic) return as the Aramaic chs. 2-7 in Daniel,while the beginning of the Aramaic also shows a close agreement: inboth cases the word rrnix, '(in) Aramaic', seemingly as an announce-ment of the language of a document or of speech, but in reality as theintroduction of a long stret(-h of text in the Aramaic language. Thisagreement, though very striking to everyone who reads the texts inthe original languages, has received preciously little attention; one of

s, 'Discontinuity, Congruence and the Making of the Hebrew Bible', p.37 n. 19.


the few to have dealt with it is J.E. Miller, who ascribed it to whathe called 'bilingual editing\^® The assumption of literary emulationprovides a clear and unambiguous explanation for this state of affairs.

The divisions in cliajjters (or rather in major episodes) seems tocorrespond in Ezra and Daniel also: in both cases we see six and fourcoherent episodes, which in Ezra describe the restoration of cult andTemple {1 G), and the mission of Ezra (7 10). respectively. In theepisode described in Ezra 5, a large part of the information aboutevents is provided in Tattenai's Aramaic letter in 5.8-17; in the corre-sponding Aramaic chapter Daniel 4 Nebuchadnezzar's edict in the formof a letter has become the only source for the episode. Both in Ezra andin Daniel chs. 5 and C are not in the right chronological order withinthe book: as noted above Daniel 5 and 6 properly belong after ch. 8,whereas Ezra 5 and G, dated to the reign of Darius I (522-486), belongbefore the parts of Ezra 4 which are dated to the kings Xerxes (4.6,ruled 486-461} and Artaxerxes (4.7-23, ruled 464-424). ' The transi-tion between chs. 7 and 8, formally from third-person account to thefirst person in both cases, is made through a personal reaction of themain persons Ezra and Daniel: 'Blessed be the Lord, the God of ourfathers, who put such a thing as this into the heart of the king, to beau-tify the house of the Lord which is in Jerusalem, • and who extendedto me his steadfast love before the king and his counselors, and beforeall the kings mighty officers. I took courage, for the hand of the Lordmy God was upon me, and I gathered leading men from Israel to goup with me' (Ezra 7.27-28), and "Here is the end of the matter. As forme, Daniel, my thoughts greatly alarmed me, and my color changed;but I kept the matter in my mind' (Dan. 7.28)'; these sentences alsomark the transition from Aramaic to Hebrew in both cases (though

'f'J.E. Miller, 'The Redaction of Daniel\ JSOT 52 (1991), pp. 115-24. The mainthesis of his ai'ticle, however, namely that the present-day Book of Daniel wasedited out of a complete book in Hebrew and another one in Aramaic, seems notvery likely.

^^See for a new view of the problems of the Aramaic documents in Ezra 4-7, whichlook like a single Aramaic text composed of documents in incorrect chronologicalorder, J.W. Wesselins, 'Ezra-Nehemia' [Dntcli] in Eokkeiman and Weren (eds.},De Bijbel Literair. pp. 215-24; cfr. also H.G.M. Willianison. 'The Composition ofEzra i-vi', JTS 34 (1983), pp. 1-30. who rightly concluded that these chapters mustbelong to the latest stage of the composition of Ezra-Neliemiah. Note also that, likethe Aramaic chapters of Daniel, the Aramaic episodes in Ezra 4-6 can be shownto exhibit a chiastic structure: S.C. MatzaL 'The Structure of Ezra iv-vi'. VT 50(2000), pp. 566-69.


the verses themselves are in different languages, Hebrew in Ezra andAramaic in Daniel). In chs. 8 and 9 of both books, Ezra and Danielstart speaking without further introduction, whereas they arc formallyintrodiK'ed again at the beginning of Daniel 10 and Ezra 10. Finally,both Ezra 9 and Daniel 9 contain a confession of guilt on behalf of thepeople of Israel. The conclusion seems justified, briefly said, that allthese elements in the book of Daniel caiuiot })e considered in isolationfrom the parallels in Ezra and Genesis any more.

It becomes clear that a number of issnes, which appeared likeformidable and to all a])pearance8 insoluble problems, can be explainedfrom the hterary nature of the book of Daniel without much effort.Thus even the perennial question why Daniel contains both Hebrewand Aramaic parts is answered by the observation that the distributionof these two languages constitutes a literary ennilation of the situationin Ezra. The same observation can be made for a number of other,hotly debated but never resolved, problems in the book of Daniel, suchas the change of main person in Daniel 3 or the changes of the personwho is speaking in Daniel 7-12.

All this means that the existence of various texts among the DeadSea scrolls dealing with Daniel or other Jewish courtiers may indicatethe literary background of the book of DanioL and may have providedthe inspiration for making such a collection of seemingly independentstories, but that the book of Daniel itself cannot be regarded as a moreor less random collection of such stories.^^ The intertextual referencesto the episodes of Joseph and Ezra, which are also related to Danielthrough the great hkeness of Ihe main persons in function and piety,provide an undeniable unity to the book, and show that many featuresof its parts are not accidental or the result of a historical developmentof the text.

Such observations about the dependence of certain works in theHebrew Bible as we made for Daniel are not unique to this book. Wemay, in fact, be dealing with the key to understanding much of theliterary texture of the books of the Hebrew Bil>le. Elsewhere I argued

"*See about thi.s literature, for example, P.W. Flint. 'The Daniel Tradition atQumran' in: Collins and Flint (eds.). The Book of Damel. pp. 329-67: E. Eshcl. To.s-Hible Sources of the Book of DanieF. ibid., pp. 387-94; S. White Crawford, •4QTalesof the Persian Court (4Qr)50a-e) and its Relation to Biblical Royal Courtier Tales.especially Esther, Daniel and Joseph', in: E.D. Herbert k. E. Tov (eds.). The Bibleas Book: The. Hehrew Bible and the Judaean Desert Diseoveries (New Castle DE:Oak Knoll Books, 2002), 121-37.


that the seemingly fragmentary nature of many of the books in theHebrew Bible, especially in the historical literature, is not to be ex-plained from the supposed prehistory of the text as we now have it,but represents a hitherto unrecognized, highly characteristic, literarygenre, namely the linear composed dossier (or linear literary dossier),a text which consists of a number of elements which are usually inroughly chronological order, and which show a considerable aiiiuunt ofdeliberately conceived discontinuity and even ontright contradiction,which was counterbalanced by several indicators of unity and conti-nuity. Many of these elements which bind the text together have longbeen recognized, such as unity of subject, references within the text orcommon terminology, and have caused nmnerous problems to scholarswho attempted to explain the text mainly from its supposed historicalbackground. Two of those indications of unity and continuity, however,have remained hidden until recently. The first one is the just-mentionedderivation of the layout of the complete work from a comparable otherwork, in most cases within the Hebrew Bible, but outside it in the caseof the so-called Primary History in Genesis-2 Kings, the global struc-ture of which appears to derive from the near-contemporary Greek-langnage Histories of Herodotus of Halicarnassus (composed ca. 450DCE}.'^ The second one, the use of striking and probably relativelyuncommon literary and linguistic forms for each one of a number ofcomparable narrative sitnations, will be discnssed below.

It may be useful to point out that the literary emulation in Daniel(and, one may add, in the Primary History) through its reflection ofthe overall structure of other works is relatively straightforward wlirnwe compare it to other instances in the ancient world, both in Jewishand in classical Greco-Ronian literature. A good oxainplo is the bookof Tobit, which ou good grounds has been noted to have intertextualconnections with the books of Genesis and Job in the Hebrew Bible,^°and with the story of Telemachus' journey in Homer's Odyssey,^^ but

'•'Wesselius, 'Discontinuity, Congruence and the Making of the Hebrew Bible';id.. The Origin of the History of Israel: Herodotus'" Histories as Blueprint for theFirst Books of the Bible (London: Sheffield Academic Press/Continuum, 2002).

'^"Dei)orah Diniant, 'Use and Interpretation of Mikra in the A])()crypha andPKeudepigraplia'. in: M.J. Mulder and H. Sysling (eds.), Mikra: Text. Transla-tion, Reading and Interpretation of the Hebrew Bible, in Ancient Judaism and EarlyChristianity (Assen and Philadelphia: Van Gorcuni, 1988; repr. Peabody, MA: Hen-drickson, 2004), pp. 379-419: 417-19.

• ^Carl Fries. 'Das Buch Tobit nnd die Telemachie'. Zeitschrift. fur Wis-senschaftliche Theologie lui (1911), pp. 54-87; Dennis R. MacDonald, 'Tobit and the


which does not exhibit the same rigid adhering to the table of con-tents of the works to be emulated. The Jewish literary world of latePersian and early Hellenistic times apparently appreciated such liter-ary works, which can be read and enjoyed without any knowledge oftheir intertextual links, but yield a tot of insight about their intentionand structure once their literary background is revealed. In this respectthey look very similar to various works of Greek and Latin literature,where the same observation can be made. '

We are evidently dealing in the book of Daniel, as in the other casesof the linear composed dossier, with a conscious literary strategy, whichwas not intended to deceive the reader (in that case the indications forthe unity of the text would have been entirely superfluous), but whichaimed to create the above-mentioned kaleidoscopic view of the eventswhich were to be described. The author performed a trne literaiy tourde force: he painted the life of Daniel and his visions in a series oftexts, which have been consciously made very different from each otherin various respects, so that they can be read independently from eachother and make the impression of deriving from various backgrotuids,while the author caught them all the same in one coherent frameworkthrough the systematic application of several narrative techniques, asdiscussed below, and through the intertextual character of the entirebook. This unitary character also makes it very unlikely that the visionscan be contrasted with each other as referring to different views of thefuture, originating in different times and circumstances, but makes italmost certain that they should be used to supplement each other.In other words, the highly detailed vision in chs. 10-11 is in the lastresort an expansion of the visions in chs. 2 and 7, and the corning ofGod's empire and the role of the people of Israel as recounted in thosechapters can safely be presumed for this last and greatest vision also.

Finally, it is important to outline the dual significance of this emula-tion of the structure of other works for the thesis proposed here. On theone hand this emulation is part of the entire network of sophisticated

Odyssey', in: Dennis R. MacDonald (ed.), Mime.'^i.H and Intertextuality in Antiquityand Chrifitiamty (Harrisbnrg, PA: Trinity Press International. 2001). pp. 11-40.Compare also in the same volume George W.E. Nickelsburg, "Tobit, Genesis, andthe Ody.sspy: A Complex Web of Intertextuality', pp. 41-55.

^^See, beside the literature mentioned below about the relationship between var-ious epic works in Latin and Greek, for example also Edmund C. Cueva. The Mythsof Fiction: Studies in the Canonical Greek Novels (Ann Arbor: University of Michi-gan Press, 2004).


intertextuahty which is described in this article and in the publicationsabout the Primary History and Ezra/Nehemiah referred to here. Onthe other hand, and in this stage of research in a way more impor-tantly, as noted above this copying of the table of contents of anotherwork in such a way that a very small set of short and miainbignousrules can describe the transformation from one work's overall structureto the other's^'* constitutes the observation of a fact about the textsand a problem for their interpretation. The chances that such a simplecongruence between two works would arise by accident are infinitelysmall, that such a congruence would then occur several times withinthe one small corpus of the Hebrew Bible is downright impossible. Ifone would for some reason reject the other observations and proposalswhich are made here about the book of Daniel, one would be facedeven more by the problem how to explain this basic and repeatableobservation.^^

Unusual Literary Forms as a Unifying Principle

Another interesting link between the Aramaic court stories in Daniel2-6 seems to belong to the stock instruments of the genre of the linearcomposed dossier in the Hebrew Bible also. It would seem, as notedabove, that in some cases the sense of iniity of the parts of the dossieris increased by providing an unusual and striking literary layout toevery instance of a number of comparable episodes. Thus it can be ob-served that in the Primary History the introduction of main persons inthe first eight books is in the vast majority of cases (the first humans,Abraham, Jacob, Joseph, Moses, Saul and David) marked by the use

^•'See the listing of tliese rules in my "DiMrontinuity, Congruence and the Makingof the Hebrew Bible', pp. 6G, 68-69, 71 and 75: 'The Writing of Daniel', pp. 306-307.

••i emphasize this point because some authors tend to turn a blind eye towardsobservations which do not fit in the framework of the discipline as they conc^eive it.Even if any or ail of a number of strange opinions attributed to me by B. Beckingin his review of my The Origin of thf. Hi.stoTi) of Israel {Review of Biblical Litera-ture.. Iittp://www.bo()kreviews.org/pdf/;i038_3303.pdf), of which 'Both Joseph andCyrus lived in exile before reaching a powerful position' [sic! not in my book andnot in reality] is a characteristic example, were to be found in my book, the basiccongruence of the two works as described on pp. 57-68, which is indicated to bepivotal to the book on p. viii-ix of the Preface, would still have to be dealt with;Becking for some reason refrained from referring to those central pages at all. Ofcourse any book can be c ailed "weakly [... ] argued". a.s Becking does at the end ofhis review, if one takes the liberty to omit its explicitly stated main argument fromconsideration.

258 Ammaic Studies 3.2 (2005)

of a peculiar duplication in the account of tlieir first described move-ment in space, with two alternative courses for the background or thecircumstances beiug provided, connected through ambiguous sentencesand contrasted by means of apparent or real contradictions.^''

In the book of Daniel we can observe that in all the stories ofDaniel 2 6, which deal either with accusations against the Judeans(Dan. 3 and 6} or riddles presented to the eastern kings (Dan. 2. 4and 5} important information, which in the story is already availableto the characters and plays a role in the development of the story, ispresented to the reader only once it becomes absolutely vital for theaction itself. Sometimes this postponement is functional within thestory, but usually it seems to have mainly the function of heighteningthe tension. Here and there this procedure looks somewhat forced, andin one instance, in Daniel 5, it makes the interpretation of the narrativevery difficult. In the case of the predictions (2, 4 and 5) the account ofthe contents of the riddle is postponed, with the accusations (3 and 6)both the nature of the accusation and the way in which the martyrsare saved are placed in a later position in the story.

In Daniel 2 it is not at once clear whether the king remembers thedream or has entirely forgotten it, nor whether he asks of his wise mento tell the interpretation only or the dream itself also, until he con-firms this in 2.9. The frequently recurring expression "the dream andits interpretation' can apparently be understood in either direction,as a hcndiadys with the meaning 'the interpretation of the dream', orliterally as referring to two different queries. However that may be.the dream is told only just before its interpretation by Daniel, in 2.31-

^•''J.W. Wesselius. 'Collapsing the Narrative Bridge', in: J. W. Dyk a.o. (ods.).Unless Some One. Guide Me. Festschrift for Kartl A. Deurloo {Maastricht: ShakerPublishing, 2001). pp. 247-55; id.. Towards a New History of Israel', Journal ofHebrew Scriptures [online journal: www.arts.ualherta.ca/JHS of www.purl.org/jhs]3 (2000 2001). article 2: PDF version p. 1-21; The Origin of thf. History of Israel:Herodotus' Histories as Blueprint for the First Books of the Bible (London: Sheffif IdAcademic Pross/Continuum, 2002); and my forthcoming book God's Eleetion andRejection: The Literary Strategy of the Historical Books at the Beginning of theBible. See about a possible West-Semitic background of surh ambiguity my "Lan-guage Play in the Old Testament and in Ancient Nortli-Wost Semitic Inscriptions:Some Notes on the Kilamuwa Inscription', in: R,P. Gordon k J,C. de Moor (ods.).The Old Testament in Its World: Papers Bead at the Winter Meeting. Januari2003, The Society for Old Testament Study, and at the .loint Meeting. July 2003.The Society for Gld Testament Study and Het Oudtestamentisch Werkgezelschapin Nederland en Belgil (Leiden etc.: Brill, 2005 [appeared 2004]) pp. 253-65.


35, while he must already have known it from verse 19 onwards. IuDaniel 3 the readers are at first not informed about the reaction of thethree men to the king's decree. Only after the Chaldeans" denunciationwe understand that they visibly refused to comply with it. Somethingcomparable happens with their salvation from the fire. The divine in-tervention must have taken placo at the very moment when they arethrown into the oven and the men who were to execute them die fromthe flames themselves (3.22), yet the readers are informed only whenthe king announces that he sees four persons walking around iu themiddle of the fire (3.25): the three men and the angel wlio saved them.In Daniel 4 the dream is put in Nebuchaduezzar's mouth first when hetells it to Dauiel (4.10-17). uot wheu he informs his wise men about it(4.7). Dauiel 5 presents formidable problems in this respect. The useof the expression 'the iuscriptiou aud its iuterpretatiou' causes greatuncertainty: does the king really ask for both, or are we deahug witha hendiadys again? Yet, especially in view of the sentence 'because allthe wise lneu of my kingdom are uot a))le to uiake known to me the in-terj)retation" (4.15, RSV 18), where nothing is said about auy difficultywith readiug the iuscriptiou itself, it appears very likely that readingthis text which eousisted of the common names of weights mina ( —60 shekels), shekel and half-shekel, posed no problem, and that it waskept from the reader only to conform to this literary pattern, raisingthe suspension of the story in the meantime, nntil Daniel reads it in5.25 and explains these ordinary uames of weights as a jirediction re-ferring to the end of the kingdom of Babylon.• ' Finally, iu ch. 6 we seethe same pattern an in Daniel 3. The real background of the deuunci-ation, the fact that Daniel always prayed three times a day and wouldcertainly not be stopped from doing this by the king's decree, is toldto the readers when his accusers make certain of this before telling itto the king (6.12), not as the background of their plans (0.8). Danielwas saved when he fell between the lions in the den (G.17, compareverse 23). but ouly on the following morning his answer to the kiug"squestion makes this clear to the readers (6.22). We have a uuique iu-dication that this procedure was experieuced as very strange in antiq-uity already. Thongh the translator of tlie original Greek translationof Daniel (which was replaced by the so-called Theodotion version in

^^I hope to discuss the problems of this chapter at length elsewhere, but for thetime being it suffices that it is possible, to read the chapter in this manner, and thatthis fits in a general literary trend in Daniel 2 6.


most Septuagint manuscripts) probably did uot understand the liter-ary nature of his original, he took such offense to the uuusual audsomewhat forced postijonenieut in these (;ases, that he attempted toremove most of them. The amount of intervention needed for this pur-pose accounts for much of the greater and smaller deviations from theMasoretic Text in chs. 2-6, which in turu makes it very likely that theMasoretic text is the original form of the text of the book of Daniel:there is a perfectly logical path from the complex Ma.soretic text to theeasier and, one could add, in the literary field sometimes undoubtedlysuperior. Septuagiut text ty])e, but uone the other way round.^^

Having established the iutertextual liuks with the book of Ezra audwith Geuesis, our eyes are opeued ouce more to the non-linguistic fac-tors determining the use of certain wortls, expressions aud grammaticalconstructions in the Aramaic of Daniel. On the grammatical level wecan observe that the use of the demonstrative pronoun "pH 'these",in Dan. 3.12. 13, 21, 22, 23, 27 and 6.6. 12, 16, 25, mainly servos todraw the twfi chapters together (it is used both for the accusers of theJudeans and for Daniel's three companions), but canuot be used asproof that this demonstrative pronoun was really used in the Aramaicspoken or written at the time of composition of the book of Daniel; inEzra by contrast, both this word aud the correspouding singnlar formsT^ (m.) and "]! (fem.) are used very frequently. This word should beadded to the well-known list of words and expressions connecting thetwo chapters, such as the dennnciation (3.8; 6.25); the trusting in God(3.28; 6.24); the role of the angel (3.28; 6.23); and the note that themartyrs were not hurt iu any way (3.25; 6.24).

The very frequent use of the construction 1.. .'l bziyfbD, 'in agree-ment with the fact that...' or 'because", with waw apodosis at the be-ginning of the second sentence, probably derives at least partly from apossible interpretation of the sentence nniJi NJn' n i^b^'U n' n""" ^2p~^2!NTnQ'7 N]':?- '-,N i^b KDbQ (Ezra 4.14), translating it as 'Now because("I '72y['^2i] we eat the salt of the palace, (i) it is not fitting for nsto witness the king's dishonour", iustead of the simple juxtapositionof the two clauses as iu, for example, the RSV: 'Now because we eatthe salt of the palace aud it is not fitting for us to witness the king'sdishonour'. Iu Dauiel, however, it has beeu extended into a c-omuiouly

in general about the relation Ix'tween the Masoretic Text, 'Theodotion', andthe Old Greek version of Daniel: A.A. Di Leila, 'The textual history of Septuagint-Daniel and Theodotion-Daniel'. in: CoUins and Flint (eds.). The Book of Daniel.^pp. 586-607.


used literary means of indicating refiection on events and slowing downof the speed of the narration; a sentence expressing the state of affairsis followed by a sentence which begins with "l ' ?R"' r ^ ' gives thereason or the circumstauces, and a final sentence beginning with l andreturuiug to the subject of the first seuteuce.' ' Also with regard to theexpression 'the dream/the inscription and its interpretation' in chs. 2,4 and 5 which we discnssed above, it is iuevitable to assuuie that thelingnistic usage umst have been infiueuced by literary cousideratious.Many features of the Aramaic of Daniel will have to be reconsideredin the light of these and comparable observations, and we must in auycase be very cautious with using them for linguistic information beforewe fully understand the literary aspects of the text we are dealing with.

This intertextual aspect of the use of language in Daniel is also rele-vant for one of the most frecjuently discussed passages in the book, thedialogue between king Nebuchaduezzar aud the three men in 3.14-18,where the text seems to suggest doubt about God's ability to save hisservants. The passage should be studied in the light of the observationthat in Daniel 2 and 3 there are two instances of the king threatening agroup of his subjects with certain death if they do not comply with hiswishes. He commands his wise men to tell his dream and its interpre-tation iu Dau. 2.5-6 aud 8-10. and orders Daniel's three tompauions toworship his statue iu 3.14-15. In both cases the uuusually frauk answerof the subjects echoes or mirrors the formal structure of the king'swords, after which he gets exceedingly angry and condemns them todeath. In Daniel 2 we see in both cases the construction 1.. .'I '75p"'7Dbetween two parallel sentences at the begiiming and end of the speechof the king and of the wise men. in Daniol 3 in both a positive and anegative conditional sentence around a simple questiou of tlie king andan answer by the three men, with the question aud the answer aboutworshippiug the king's statue at the beginning of the king's speechand the end of the men's. Moreover, both the first conditional sentenceof the king and the last one of the men have no apodosis, apparentlybecause it is so evident that it need not be said explicitly.

By echoing or mirroring the words of the king his subjects, whodo not have nuich to expect from him iu auy case, put themselves outhe same level with him, aud exhibit a liberty of speech which would

^^.1. W. Wesseiius, 'Language and Style in Biblical Aramaie: Observations on theUnity of Daniel II -VI', VT 38 (1988), pp. 194-209. See the appendix to this articlefor a list of these instances.


otherwise be uuthiukable in tlie relation with the king. ^ Apart fromthis psychological aspect, well known from many martyrs' stories, thisdouble echoiug iu two subsequent chapters seeuis to be an expandedversion of the same phenomenon in Genesis 37 aud 38, which we alreadynoted to be intertextually connected with Daniel 2 aud 3. Already inantiquity it was noted that the interaction of Joseph's brothers withtheir father .Jacob, with the repetitiou of the verb "ID: (hif.). 'to recog-nize', is rt sumed in the interaction between Judah and Tamar, wherewe also encounter this verb twice.^° See the table at p. 263, wheresome important Aramaic words have been added between brackets,Note that interestingly the agreement within the passages in Genesis37 and 38 and between them has largely shifted from a cougrueuce ofwords and contents to one of structure. Another curious literary shiftis to be found in the main subject couuectiug the chapters: deceptionof Jacob aud of Judah iu Genesis 37 38. a huge statue iu Daniel 2-3,so from a common theme to a COIUIIKJU uiaterial object as focus ofatteutiou.'^'

All this bears on my old proposal to understand the couditioualsentence in verse 17 as it is translated here, not translating 'H at itsbeginning as 'behold' (which would be highly uutisual in Aramaic inany case), which would iuiply the martyrs' certainty of being saved,nor stopping the couditioual sentence after 'fiery furnace,' which wouldimply doubt about God's ability, but certainty about his willingness tosave, nor translating TTK as '(our God) exists', but uuderstandiug thesentence which begins with ]n as a protasis following the a.podosis 'wedo not need to answer you...\ which nicely avoids all theological trapswhich could be iuvolved iu the idea of doubting or postulatiug God'sexistence or his ability or williuguess to save his followers:

^^Wesselius, 'Language and Style', pp. 204-208.•' See about the rabbinie traditions on Genesis 38, where this was already noted,

and where the parallel with Daniel 3 is also found (albeit probably not on literarygrounds), especially E.M. Menn, Judah & Tamar (Genesis 38) in Aneient JewishExegesis: Studies in Literary Form & Henneneuties (Leiden: E.J. Brill, etc. 1997),and Y. Zakovitch & A. Shinan, The Story of Judah and Tamar: Genesis 38 in theBible, the Old Versions and the Aneient Jewish Literature (in Hebrew; Jerusalem:Hebrew University, 1992}.

The agreement in subject wa,s already noted by the Cluircti Father Hippoly-tus of Rome in his commentary on Daniel, though not, of course, a-s a literaryphenomenon: In Danielem., ii, 15. Hippolytus assumed that Nebuchadnezzar, afterseeing the huge statue in his dream in Daniel 2, decided to ereet a statue for himselfin eh. 3.


Dan. 2.8-y. the king:I know with certainty that you are tryiug to gaintime.

Because ('1 b^p'bzi) you see that the word fromme is sure ^that if you do not make the dreamknown to me, there is but one .sentence for you,{') you have agreed to speak lying and corruptwords before me

till the time will change.

Genesis 37.32And they sent the long robewith sleeves and brought it totheir father, and said, "Thiswe have found; recognizenow whether it is your son'srobe or not".

Dan. 2.10-11, the Chaldeans:There is not a man on earth who can meet theking's demand.

To such a degree that ("T ^"ip'^D) no great andpowerful king has asked snch a thing of any ma-gician or enchanter or Chaldean "("I) the thingthat the king asks is difficult,

and none can show it to the king except thegods, whose dwelling is not with flesh.

Genesis 37:33And he recognized it, andsaid, 'It is my son's robe; awild bea*)t has devoured him;Joseph is without donbt tornto pieces'.

Dan. 3.14-15, the king:Is it true. O Shadrach, Meshach. and Abednego.that you do not serve nxy gods or wor-ship the golden image which I have set up?

""Now if (]n) you are ready when you hear thesound of the horn. pipe. lyre, trigon. harp, hag-pipe, and every kind of music, to fall down andworship the image which I have made... [well andgood]; but if you do not (N*? ]m) worship, youshall immediately be cast into a burning fieryfurnace;

and who is the god that will deliver you out ofmy hands?

Genesis 38.25As she was being broughtout. she sent [word] toher father-in-law, 'By theman to whom these be-long, I am with child.' Andshe said. 'Recognize nowwho.se these are, the signetand the cord and the staff'.

Dan. 3.16-18, the three men:O Nebuchadnezzar, we do not need to answeryou about this

' if (]n) our God whom we serve is able to de-liver us from the burning fiery furnace and willdeliver us out of your hand, O king. ' But if not(X7 jm) .... [nothing more is to be said, but inany case]

be it known to you, O king, that we will notserve your gods or worship the goldenimage which you have set up.

Genesis 38.26Then Judah recognizedthem and said, 'She is morerighteous than I, inasnnichas I did not give her to myson Shelah'. And he did notlie with her again.


'^Shadrach. Meshach, and Abednego answered the king, "O Nebudiad-nozzar, we do not need to answer you in this matter ' if our God whomwe serve is able to deliver us from the burning fiery furnace and willdeliver us out of your hand, O king. ^ But if not.... [nothing more is tobe said, but in any case] be it known to you, O king, that we will notserve your gods or worship the golden image which you have set up'.' '

This idea has been criticized by Tim Meadowcroft, in a way rightly, asin vacuo the other interpretations may well be considerod more likelythan the one proposed here. ''* In the presently recognized literary con-text, however, this inversion of the two parts of the conditional sen-tence, for which a ninnber of parallels in ancient Hebrew and Aramaictexts can be adducred, ^ creates the desired effect of completing themirroring of the king's words, which in the case of such conditional sen-tences is indeed only possible if one lets protasis and apodosis changeplaces, and for this reason the use of this less common constructionseems no real objection.

Contacts with the Greek cultural sphere

The existence of this literary procedure for writing the book of Danieland other books of the Hebrew Bible can hardly be subject to doubtany more. This literary strategy may look highly unusual at first sight,which is doubtlessly the reason why it remained undetected through-out two millennia of intensive occupation with the form and contentsof those books. It was, however, hardly unusual in the cultural back-ground of the ancient Mediterranean, albeit in a different literary con-text. Especially various poetical epic works in Greek and Latin wereconnected through sophisticated techniques of allusion and intertextu-ality. In comparison with those works, however, it is remarkable thatin the Hebrew Bible this technique was apparently used especially forwriting historical works, rather than for complex poetical compositionsof an epic nature,'''' once again stressing the fact that Hebrew prose nar-rative in many respects surprisingly takes the place of epic poems in

"•^Wessehus, 'Language and Style', ibid.^^T.J. Meadowcroft. Aramaic Danid and Greek Daniel: A Literary Comparison

(JSOTS, 198; Sheffield: Sheffield Academic Press, 1995), p. 149. His main argument,however, that 'n'S without a suffix does not serve as a copula elsewhere in BiblicalAramaic, is very weak in view of the limited amount of material which is available.

^^Wesselius. 'Language and Style", p. 205 u. 11.^•''See. for example, the contrasting treatment of the genres in Ronald S. Hen-

del, The Epic of the Patriarch: The Jacob Cycle, and the Naiiutive Traditions of

WESSELIUS The Literary Natum of the Book of Daniel 265

other Mediterranean cultures. The elements which are alluded to areon the whole also different: in the case of the Greek and Latin poeticcompositions mainly certain type-scenes and developments, within thedifferently structured proae texts of the Hebrew Bible certain episodes,often coinciding with present-day chapters, and a number of strikingcases of discontinuity. For the Primary History itself the picture ismixed: on the one hand there was the division of the work in ninebooks in groups of one, five and three deriving from and alluding tothe same division in the work of Herodotus, but on the level belowthat there is no precise division in distrete units, so that as charac-teristic anchors for allusion in the overall framework its author chosethe genealogy of the main family in relation to the events of the GreatCampaign in both works, and the course of causally connected casesof deception at the begirming of the work.'***

But another question is more pressing for the time being. What isthe reason why this technique was used simultaneously in Israel andin the world of Greek and Latin literature? Of course the answerswhich readily come to miiid are paiallel literary development, stimulusdiffusion in the eastern Mediterranean or a pedigree for this techniquewhich goes back very far into the past, from where it reached both theGreek and the Oriental world. But in the case of the Primary History itis apparent from the emulation of the structure of Herodotus' Historiesthat the contacts with the Greek world must have been much closerthan we usually think. wlii< h makes a direct derivation of tliis literarystrategy and a literary contact on a rather high level between the two( ultures very likely. We can now observe that the rareness of referencesto the Greeks and Greek culture, and likewise to the Persians, areprobably the result of liinitations wiiich the authors of the HebrewBible imposed on themselves, rather than on a supposed nnawarenessof them. This was not done in an attempt to obscure this background,for in that case the literary references to Greek literature would havebeen inai)i)ropriate, but apparently as part of a deliberate program toconcentrate on the national patrimony only.

Canaan and hrad (Harvard Semitic Monographs 42; Atlanta, Georgia: ScholarsPresK. 1987). and ahout the rplatioii between poetry and prose in general the ar-ticles in .J.C. de Moor and W.G.E. Watson (edy.). Verse, in Ancient Near EasternProse {Kevelat!r/Neukirchcn-Vluyn: Neukirchener Verlag, 1993).

^''See my 'Discontinuity, Congruence and the Making of the Hebrew Bible', pp.41-43, and The Origin of the History of Israel, pp. 35-41, and Fig. 2.1 on p. 61.


A closer look reveals that there is a literary contact between Danieland Herodotus also. I noted a number of similarities in an earlier articlealready, but they did not seem to fit easily in the general literarypicture of the agreements with Ezra and Genesis.^'' In this case, aswith the relationship between the Histories and the bibhcal PrimaryHistory, an analogy with the relation between certain Greek and Latinpoetic works proves highly illuminating, and appears to solve some ofthe earlier problems. There is an interesting parallel for this literarystrategy of allusion to both a classic work and an earlier emulation ofit in classical literature, which throws even more light on the literarynature and method of couiposition of the Book of Daniel. Daniien Nelishas recently demonstrated systematically that Vergil not only based hisAeneid on Homer's Iliad and Odyssey, a connection which has beenexhaustively discussed by G.N. Knauer,''^ and in a more incidentalway on various other works, but that he also leaned very heavily ona hellenistic Greek work, the Argonautica by Apollonins of Rhodos(3rd century BCE), which des{-ri})es the journey of the ship Argo withJa-son and his companions to fetch the Golden Fleece from Colchis onthe Black Sea. ^

It would seem that, just as Vergil recognized the literary dependenceof Homer's Iliad and Odyssey on one side, and Apollonius' Argonauticaon the other, and played with the similarities and differences for thecomposition of his own Aeneid. the author of Daniel recognized theallnsions to Herodotus' Histories in the story of Joseph, noted theagreements and differences between the two, and made his own choices,which not rarely moved his account closer to the Histories again thanthe Ufe of Joseph in Genesis 37-50 is.

In a famous passage in Histories I, 107-108, Herodotus tells hisreaders about the two dreams which the Median king Astyages hadabout the end of his reign. In the first dream he saw a stream of urinecoming from the vagina of his daughter Mandane, which inundated first

^^J.W. Wesselius, 'Analyses, Imitation and Kiuiilation of Classical Texts in theHebrew Bible', Dutch Studies-NELL 2 (1996), pp. 43-68.

•' See especially G.N. Knauer, Die Aeneis und Homer: Studien zur poetischenTechnik Vergils mit Listen der Homerzitate m der Aeneis (Gottingen: Vandenhoeck& Rupprecht, 1964, second ed. 1979).

•'^Daniien Nelis, Vergil's Aeneid and the Argonautica of Apollonius Rhodius(ARCA: Classical and Medieval Texts. Papens and Monographa, 39; Leeds: Frari(;i.sCairns, 2001). In general about these works and their relation: CR. Beye. AncientEpic Poetry: Homer, Apollonius, Virgil (Ithaca and London: Cornell UniversityPress, 1993).

the entire city and then all of Asia. His dream-interpreters explainedthe dreani, apparently as meaning that her offspring was to removehim from office. His rea( tion was to marry her off to a Persian namedCainbyses. The Persians being regarded as of lower rank than the Me-dians, he apparently did this in order to make her children less hkelyto snpplant him. Bnt within a year he had a second dream, in whicha vine grew from the vagina of his daughter {who was pregnant at thetime). finally covering all of Asia. His dream-interpreters gave the sameinterpretation. Then Astyages decided to do away with his grandsonright after his birth. Bnt his scheme went wrong and the young Cyrnsindeed dethroned him, bnt that story is of no direct importance for onrpnrpose here.

Elsewhere I demonstrated that it is very likely that these twodreams are reflected in a pecnliar way in the story of Joseph, namelyas tiiree pairs of dreams, of Joseph (Gen. 37), of the steward and thebaker (Gen. 40), and of Pharaoh (Gen. 41).^" The main differenceswith the situation in Genesis are that there are three pairs of dreamsin Genesis and only one in the Histories, while there is a considerableamount of time between the two dreams in Herodotus, but hardly anyin the three instances in Genesis, that both Astyages' dreams predictthe end of the dreamer's reign ratlier than his coming to power (Gen.37), restoration to dignity or execution (40) or a period of plenty andone of hunger for the dreamer's country (41). To onr amazement someof tlK'se elements of Herodotus' account are restored in Daniel in spiteof th(! linear correspondence with Genesis! Nebuchadnezzar's dreamsin Daniel 2 and 4 both describe the end of his reign, albeit for theremote future in ch. 2 and only temporarily in 4. and there is an entirechapter. Daniel 3, between them. Because of the fusion of the pairs,through making the king into the dreamer and through changing thethird resulting case, Daniel 5. into a prediction based on an enigmaticinscription, the nature of the dreams in the first half of the book ofDaniel has moved very close to that of Astyages' dreams in the Histo-ries. The actual contents of the two dreams are also much more similarin Herodotus and Daniel than in Genesis: in Daniel 2 the great stonewhich destroys the statue, having become a huge monntain, finally fillsthe entin; earth, like Mandanc's urine filling all Asia, in Daniel 4 theking is represented in the dream as an enormous tree, which can beseen until the ends of the earth, again very similar to the vine from

'*" The. On<p.n of the History of hmei. pp. 12 and .59.


Mandane's vagina which covered all Asia. Additionally, we seem to findallusions in Daniel 4 to the tree and the birds of the dreams in Genesis40.9-10 and 17, and to the wild beasts which are associated with Ne-buchadnezzar in Jer. 27.6 and 28.14. Of course the situation in Daniel2 and 4 is in many ways also very different from that in Histories I,107-108, but the literary return to Herodotns in Daniel is so clear thatthis cannot be an argument against the allusion, just as the differencesof the Aeneid from Homer, or from the Argonautica, or from both,cannot be an argument against literary dependence.

Something comparable is going on in Daniel 3 and 6, which we notedto be highly similar martyrs' stories. It can be noted that the story ofJudah and Tamar in Genesis 38 seems to allude to Gyrus' intendedburning of the defeated king Groesus of Lydia in Histories I, 86-88,related to him by marriage just as Taniar was connected to Judahthrough her marriage to two of his sons."'' But otherwise the connectionbetween the stories seems to lie only in the element of the somewhatunexpected punishment of burning, which finally is not carried out.But other elements of the story of Groesus' intended burning, suchas the person pronouncing the verdict being a king, who spoke withthe condemned person while the latter was already in the place ofexecution, his desire to stop the execution while not being able to doso, and the divine intervention to save the condemned person are absentfrom Genesis 38. These elements, however, return in Daniel 3 and 6.Darins 'the Mede', as noted above probably identical with king Cyrns,in Daniel 6 did not want to execnte Daniel bnt had to do so for legalreasons. He spoke with him while Daniel was already in the lions' den,and the divine intervention, of course, is found in both chapters. Again,it is not likely that these elements which draw Daniel and Herodotustogether across the intervening literary episode of the Primary History,would be present accidentally only.

There are three other possible instances of allusion to the Historiesin Daniel, not of the same weight as those discussed above, but stilllikely. Elsewhere 1 pointed out that the train of causally related casesof deception in the Histories, which underhe the great conflict betweenthe Greeks and the Persians under Darius and Xerxes, start with thegroup of Median young men who, on the orders of the Median kingGyaxares (Astyages' father), are to learn the Scythian language andtheir way of archery (I. 73), which episode corresponds with the story

'^^The Origin of the Histoi-y of Israel, 15 and 73-74.

WESSELIUS The Literary Nature of the. Book of Daniel 269

of Jacob's and Rachel's deception of the aged Isaac in the PrimaryHistory, which sets in motion both a chain of deceptions and a trainof causal relations which are at the beginning of the history of thenation of Israel and in particular of the tribe of Judah.''^ In Daniel.by contrast, the point of departure of the events in the book is theselection of the Israelite princes to learn 'the letters and language ofthe Chaldeans' (Dan. 1.4). Also in this case we appear to have anallusion to Herodotus reaching back across the direct example of thePrimary History.

The final prediction before the end of the life of Joseph, Jacob'ssayings about his sons in Genesis 49, can be noted to take the positionof the prediction about Darius, which Cyrus sees in a dream in the nightafter he enters the country of the Massagetes where he dies in battlesoon afterwards (I. 209). In the book of Daniel the final long predictionin chs. 10 12 appears to correspond with the place of Genesis 49 in theHrst place, but also may refer back to the Histories, especially since itis dated to the final year mentioned for Cyrus in the Book of Daniel,his third.

When the author of Daniel discarded the third 'donble dream' (ofPharaoh in Genesis 41) in favour of an oracle announcing the taking ofBabylon by 'Darius the Mede' (probably just another name for Cyrus,as noted above), he established anotlier parallel with the Histories.On one level this regards the piece of information that Babylon wassuddenly taken dnring a nightly banquet, on a more fundamental levelthe fact that the Mene Tekel inscription of Daniel 5 appears to give apersonal detail of Darius in hidden form, namely his age of 62 at thetime of his conquest of Babylon,*^ just as in Herodotus' account theorac'le of Delphi gave an enigmatic prediction of the figure of Cyrusbefore his conquest of Lydia in Histories I, 55: 'When comes the daythat a mule shall sit on the Median throne...'. It turns out that Cyrusis the one characterized as a 'mule' here, because as noted above hedescends from a mixed Median-Persian marriage (I, 95),

There are additional instances of similarity between the book ofDaniel and the description of Babylon in the first book of the Histories,

my 'De bedrogen bedrieger als oorsproiig van het gescliiedverhaal bijHerodotus en in de Bijbel' [Dutch], in: A. M. van Erp Taalman Kip & I. .L F.de Jong (eds.}. Schurken en schelmen. Cultuurhistorische verkenningen rond deMiddellandse Zee [Festschrift J. M. Breiner] (Amsterdam; Amsterdam UniversityPress. 1995), pp. 33-43. and Origin of the. History of Israel, pp. 35-41.

^ "Die 62 Jahre'.


which may or may not respresent cases of deliberate allusion, such asthe huge golden statue in L 183 and the story of Daniel 3. or thekilling of unsuccessful dream-interpreters in I, 128 and Daniel 2. bntthese cannot carry much weight for our argument, as we are dealingwith fairly common elements in those caaes.

It is remarkable that the five instances where the author of Danielapparently makes a part of his account closer to the Histories than thepassage in Genesis serving as its origin are found in the right chrono-logical order of the hrst book of the Histories. Of these five, the la.stthree episodes (enigmatic reference to identity of king, king willing butunable to save opponent from execution, and prediction of future justbefore end of reign} deal with Cyrus in both works, the first two, whichare connected with Cyaxares and Astyages in the Histories (young menwho are to be instructed in another culture and two dreams about endof reign), have been transferred to king Nebnchadnezzar in the bookof Daniel.

The Babylonian Exile and the Oriental kings

This realization that we are basically dealing only with the kings Neb-uchadnezzar and Cyrus (probably alias Darius the Mede, as notedabove, bnt in any case represented by this enigmatic figure) in thebook of Daniel, apart from the transitional hgnre of Belshazzar, al-lows ns to take a new look at the regularities pointed out first byA. Lenglet and recently resumed by J.P. Tanner.*^ As noted above,Lenglet demonstrated that Daniel 2-7 ha s a chiastic structure, withpredictions about the further course of history in four parts in 2 and 7,martyrs' stories in 3 and 6 and stories about God's power over kings in4 and 5, and rightly recognized that suf h chiastic structure is usuallyordered around a central passage, which is thus given a very promi-nent place within an episode or a book. His proposal that the entireepisode of the chs. 4-5 would constitute this centre, however, seems notvery likely, because the centre would be almost as large, if not largerthan, the chiastically placed parts on either side.*^ But if we reject thisoption, we are still faced by the question what is in the middle. Myproposal is that Daniel 2 7 has a virtual centre between chs. 4 and 5,

•*' A. Lenglet. 'La structure litteraire de Daniel 2 7'; J.P. Tanner. -The LiteraryStructure of the Book of Daniel', Bibliotli.cca Sacra 160 (.Tuly - September 2003),pp. 269-82.

"'•''Lenglet, 'La structure litteraire de Daniel 2 -7". pp. 185-87.


namely the taking of Jerusalem, the destruction of Solomon's Templeand the Babylonian exile. This is somewhat less unexpected than itmight look at first sight. Nebnchadnezzar and Cyrus are the kings whobegin and end the exile, respectively. Both are explicitly said to be inGod's service, Nebuchadnezzar especially when he is called God's 'ser-vant' in Jer. 27.6, Cyrus when he is even calkxl 'anointed' of the Lordin Isa. 45.1. When we take a closer look, we see that in the book ofDaniel both imdergo a learning process about Israel's God which leadsto their promulgating an edict about their allegiance to him, obhgingtheir subjects to honour and respect this God, Nebuchadnezzar in ch.4, which as noted above also tells the story of his madness inside theedict itself, and Cyrus/Darins in 6.27. The elements of this processare largely the structural elements which express the chiasmus notedby Lenglet, namely God's ability to save (chs. 3 and 6). his ability toexplain riddles (2, 4 and 5) and his power to humiliate or exalt kings (4and 5). In the case of Nebuchadnezzar this causes a climactic series ofstatements by the king through the chs. 2-4: though the recognition ofDaniel and his companions starts already in 1.20, his statements abouttheir God are found only in 2.47, 3.28-29 and 4.2^3 and 37; note that,as pointed out above, the entire ch. 4 is in the form of a royal edict.By contrast. Cyrus (I use the name instead of 'Darius' with the caveatexpressed above) only gives praise to God in his edict of 6.26-28, butthen the power of God to set up and depose kings and his ability toreveal secrets had been shown decisively at the end of the reign of hispredecessor Belshazzar in ch. 5. and both already concerned the personand reign of Cyrus rather than of Belshazzar; note also the continuityof Daniel's career under Belshazzar and Cyrus in 5.29 and 6.3. Cyrus'acknowledgment of the God of Israel thus covers the events of ch. 5also, though the exact wording of 6.26-28 refers to ch. 6 only.

It can also be noted that some crucial sentences in tlie edicts arenearly identical (corresponding parts have been underlined): 'KingNebuchadnezzar to all peoples, nations, and languages, that dwell inall the earth: Peace be multiplied to you! ^It has seemed good to meto show the signs and wonders that the Most High God has wToughttoward me. ^How great are his signs, how mighty his wonders! Hiskingdom is an everlasting kingdom, and his dominion is from genera-tion to generation' (Dan. 3.31-33; RSV 4.1-3). and "Then King Dariuswrote to all the peoples, nations, and languages that dwell in all theearth: Peace lx- inultipliefl to you! '^H make a decree, that in all myroyal dominion men tremble and fear before the God of Daniel, for

272 Ararnaie Studies 3.2 (2005)

Nebuchadnezzar Belshazzar & Darius (= Cyrus?)

King's oflicial proclamation aboutflie power of Daniel's God

Captivity

Humiliation and elevation ofkings, announced by enigmawhich is explained by Daniel

n Accusation, martyrdomE and miraculous deliverance

Dream of king or Daniel aboutfour empires and ihe commg

of the kingdom of God

Fignre 3: The Arammc Chapters 2-7 in Relation to the Oriental Kingsand Their Edicts, the Captivity and the Main Subjects of the Chapters

he is the living God, enduring for ever; his kingdom shall never bedestroyed, and his dominion shall be to the end. ^ Hc delivern and res-cnes, he works signs and wonders in heaven and on earth, he who hassaved Daniel from the power of the lions' (Dan. 6.25-27).

Nebnchadnezzar and Cyrus are the two human actors who, in spiteof their very different roles, execnte God's plan with his people Israelin connection with the Exile: Nebnchadnezzar, God's 'servant", by de-stroying Jerusalem and the Temple and taking the people into captiv-ity, Cyrus, his 'anointed', by allowing the exiles to retnrn to Jerusalemand to rebuild the Temple there. The book of Daniel explains howthey came to have these remarkable functions by telling its readershow through their experience with Daniel and his companions, observ-ing their revealing of secrets, their being saved from the danger arisingfrom their religious loyalty, and their predicting the exalting and de-posing of kings, they became convinecd of the power of Israel's God.


Both announce this change of heart in an edict to all their subjects. It isonly natural that between these key figures marking the beginning andend of the Babylonian captivity the Exile itself is to be sought, thoughthis traumatic event is only hinted at through the chiastic structure ofthe ch. s surrounding its virtual presence (see Figure 3).

It is hardly accidental that exactly on the threshold of the post-exilic era, in the first year of 'Darius the Mede' (the above-mentionedproposed identity with Cyrus helps, but is not essential), Daniel pon-ders about the prediction of Jeremiah in Jer. 25.11 and 29.10, thatseventy years would pass over the ruins (jf Jerusalem, and receives onthe one hand the assurance that the captivity is at an end (Dan. 9.23-25), and on the other hand is informed that Jeremiah's prediction alsomaps the future course of world history (9.25-27). Once we see the im-portance of the captivity for the stories in Daniel 1-6, we realize thatthe story at the beginning of Daniel 9 i.s the hinge ou which the en-tire book turns, which through its explanation of Jeremiah's predictionconnects the episode of the exile, the centre of the Aramaic part of thebook in chs. 2-7, and thus also the focus of the court stories in 2-6,with the course of world history as described in Daniel 2, 7, 8, 9 and10-12, two complementary aspects of the book of Daniel which wouldotherwise have remained unconnected.

Conclusion

Summarizing we can state that the book of Daniel may well be the mostintertextually determined and complex one among the books of theHebrew Bible. Its intricate narrative texture with its manifold allusionsis in the centre of the force field of the works of the earlier Israeliteliterature and the contact with Greek culture. It fits the role of thekings of the great oriental empires in the history of Israel and in God'splans, only hinted at in the books of Isaiah and Jeremiah, in a greatreligious framework of world history, and explains how they arrived atfulfilling this role. In this sense there is no break between the storiesin Daniel 1-6 and the visions in 7-12.

As is well known, the book is in a continual dialogue with most ofthe other books of the Hebrew Bible, aud often quotes them directlyor indirectly. We can now, moreover, add that its overall structure andmany of its details are determined by the stories about two personswho exhibit a great likeness to the fignre of Daniel, namely Josephand Ezra. This congrueuce with two other compositions determines


the course of the various stories to a high dcgrre. besides providingunity to a book whidi looks outwardly like a collection of ten differentaccounts of episodes of the life and predictions of Daniel; this charac teras a collection or dossier is reinforced by the variety in the namesof the persons figuring in them. The book can thus be described asbelonging to the literary genre of the 'linear literary dossier', or 'linearcomposed dossier': a book which looks like a collection of various piecesof hterature, but was in reality composed as a whole and should be readas a whole in its present form.

Consequently, it is hardly surprising that the unity of the book isalso confirmed by internal references and a number of connnon words,expressions and themes on the one hand, and by a common literaryprofile of a number of episodes on the other. At the same time, theauthor of Daniel used the passages of Herodotus which underlay cer-tain passages in the story of Joseph as alternative and supplementaryversions for vital episodes in his book. He chose certain lingnistic formsin the Aramaic part of his book for literary rather than linguistic rea-sons, but still managed to give his Aramaic chapters a characteristicand natural linguisti{' look and feel.

The author of Daniel, past master in Hebrew and Aramaic lan-guage and literature, clearly belonged to the same Hellenized circle ofintellectual Jewish readers and authors which had earlier produced thePrimary History and more or less at the same time brought forth thebook of Ezra-Nehemiah and the book of Tobit. The book of Daniel isnot a collection of material from different periods and situations, buta coherent literary and religious composition in which the changes oflangnage, of person and of style have a perfectly logical literary expla-nation.

WESSELIUS The Literary Nature of the Book of Danid 275

Appendix: I. • -' ' 5p''?3 and related idioms in Biblieal

One of the most noticeable features of the dialect(s) in which the Ara-maic parts of Ezra and Daniel have been written is the very frequentuse of the compound preposition '7DpbD. evidently composed of D. 'fand ' np, which is always spelled as two words (with a maqqef be-tween them) and vocalised as ':'3p"'p3.*'' While this compound is rarelyattested in approximately contem])oraneous documents outside of Bib-lical Aramaic,"*^ it appears 20 times in Daniel 2-6 (not once in ch. 7)and three times in the Aramaic parts of Ezra. It is never used as anindependent preposition in Biblical Aramaie, but we find it in twocompounds, the adverb r\T\ '73p"':'3 (seven times)'** and the conjunctionn ' 'np-' ? (Ezra 4.14 and 7.14, 13 times in Daniel 2-6).

It has always been implicitly assumed that •! '73p"'7D connects thesentence which precedes it to the following sentence, and has a ratherwide range of meaning: 'as, because, while, though". While the latterstatement is undoubtedly correct, the former may be somewhat lessevident than it would seem from the unanimity of the commentatorsabout it. We sliall discuss all the cases of •! ^"^p^b-^, and see that in

•""'The appendix of this article serves to elaborate one example for the statementsin its main body that certain stylistic and linguistic featnres .serve to connect theAramaic chapters of Daniel, and to counterbalance (together with certain ottiercommon features) the evident discontimiities of the book. In order not to burdenthe article with a lot of secondary literature, I refrained from discusHing everypassage in detail. This appendix reiterates a part of my article 'Language andStyle in Biblical Aramaic', with additions and corrections, and an update on theliterary and linguistic consequences.

^^The vocalization in one Genizah manuscript of the Palestinian Targum seemsto be k^loqbal or k'^loqbal; see the examples from Genesis 38.26 in the next note.

^ 'The only instances seem to he Henoch 14, 4 and 7; Murabba'at 72, 6: 4QAm-ram 1, 1 {partly restored); see the glossary in K. Beyer, Die aramdischen Textevom Toten Meer... (Gottingen: Vandenhoeck k Ruprecht, 1984), for bibliographi-cal details. Note that this fact is much more striking now that we have a relativewealth of material than formerly, when Biblical Aramaic constituted almost theonly testimony for the Aramaic of this period. In the most reliable te.stimonies forthe Palestinian Targum, the Genizah texts, this preposition is not very cominoneither: in this sizeable corpus it appears apparently only in M. L. Klein. GenizahManusenpts of Palestinian Targum to the Pentateueh (2 vols.; Cincinnati: He-brew Union College Press, 1986), pp. 88-89, Gen. 38.26. and 50-51, Gen. 30.38(=P. Kahle, Masoreten des Westens, II (Stuttgart: W. Kohlhammer. 1927: repr.Hildesheim: G. Olms, 1967). MS D. Gen. 38.26 [iwk-e b^p'^D] and MS E, Gen. 30.38

'•'Dan. 2.12,24; 3.7-8,22; 6.10; Ezra 7.17.


most instances it is used in the same syntactic and stylistic situation,at the })eginning of the second of three sentences which clearly belongtogether. The first sentence of the three is in itself quite sufficientfor describing the actual events. Of the two sentences which have beenadded, the first is introduced by 'l ' 3p"'73, the second by "i, the commonword for 'and'. These two sentences seem to serve for describing thecircumstances or for providing an elaboration of the description of themain act. The first additional sentence introduces a new point of view,whereas the last returns to the theme of the introductory sentence, inthis way completing and concluding the elaboration. This return caneither merely involve a logical connection, or be supplemented by anassociation between the two by means of the use of the same root.Daniel must say the dream in 4.15 because he is able to do so, andthe hons do not hurt him (•]i'73n) in 6.23 because he has done no harm(n'^lin). When we study all the instances of '1 ' ?p."'73 in Daniel 2-6 it appears that this formal pattern can be discerned in a majorityof the instances.^" In most instances the new interpretation which isproposed here can be demonstrated to be superior to the accepted one,in the others it is at least possible. We would have to assume that 1is a waw apodosis. Such a waw apodosis frequently occurs in BiblicalHebrew, ^ and also a few times in earlier Aramaic.^^ We shall look atthese passages in the order in which they are found in the Bible. In eachcase, I shall first give the Aramaic text and the translation as foundin the RSV, and then discnss alternative interpretations (deliberatelykept close to the wording in the RSv) for these passages.

Dan. 2.8-9:

•1 9

RSV: 'I know with certainty that you are trying to gain time, becauseyou see that the word from me is sure that if you do not make the tlreamknown to me, there is but one sentence for you. You have agreed to speaklying and corrupt words before me till the times change.'

•''"Only in Dan. 2.45, 3.29, 5.12 and 6.11 we seem to find unequivocal instancesof kol q"bel di merely connecting the two sentences which it is in between.

^^P. Jouon and T. Mixraxika, A Grammar of Biblical Hebrew (Ronae: PontificalBiblical Institute. 1991, pp. 628-30; compare Mtiranka"s discussion of the use of thiswayj in various types of sentences.

^^See the article by P. Grelot, 'Le waw d'apodose en arameen d'Egypte", Semitiea20 (1970), pp. 33-39.


Note the recurrence of the word jiu, 'time'. It would seem that theking first merely states the fact that the Chaldeans are trying to gaintime, then indicates that the reason for this attem])t is the strictnessof his connnaud, and tinally refines his accusation: they will tell himlies until circumstances change. It should be noted, however, that invv. 5-6, as is evident from the Chaldeans' answer, the sanction wasplaced upon not telling the interpretation of the dream. To avoid this,merely gaining time before having to tell it is of little use (note thatDaniel in verse 16 easily obtains a reprieve, whereas the Chaldeans donot even a.sk for it), but giving a preliminary, incorrect interpretation,which can be modified as circumstances change, certainly is. It is clear,therefore, that the translation runs much more smoothly if the sentencebeginning with 'l 'P?p;'73 is taken with the last instead of with the firstsentence:

'I know witli certainty that you are trying to gain time. BecauKC youhave seen that the word from me is sure that if you do not uiake thedream known to me, there is but one sentence for you, you have agreedto speak lying and corrupt words before me till the times change.'

It may even be possible to suppose that the causal relation betweenthe last two sentences is the other way round, which is certainly notexcluded by the use of n ' ^pfbs elsewhere in these chapters, and thatthe king announces this sanction on their not telling the dream be-cause he supposes that they will tamper with the interpretation if theyare told the dream first. Compare the use of the verb "n, 'to see', alsoafter '1 b^^'bD.. for describing something which is to be explained subse-quently in Dan. 2.41-42 (see below) and the structure of the Chaldeans'answer, where the second sentence seems to be depending on the thirdalso. This would give us a tentative translation of this passage as: 'Iknow with certainty that you are trying to gain time. For this reasonyou have seen that the word from me is sure that if yon do not makethe dream known to me, there is but one sentence for you, because youhad agreed to speak lying and corrupt words before me till the timeswould change.'

Dan. 2.10-11.

31 -pD'b^ n b2p~bD n'inn':' b-DV KDbo rbn

pb sDba zip n]-i


RSV: "There iy not a man on earth who can meet the king's demand;for no great and powerful king has asked such a thing of any magicianor enchanter or Chaldean. The thing that the king asks is difficult, andnone can sliow it to the king except the gods, whose dwelling is not withflesh.'

The first and the last sentences are obviously finked by means of theparallel expressions CJK TI'K'K'::', 'there is not a man', and x'? pnniI'n' N in':'.. .'n'N, 'and none.... except the gods'. A connection betweenthe sentence containing 'l '73p"'P3 and the preceding is very unlikelyindeed; there is no reason why a king would not be able to ask some-thing which is not possible. The comiection with the following sentenceresults in a much more natural translation. Exactly because this de-mand is so difficult that no one is able to fulfill it, no king has everthought it worthwile even to ask this thing: There is not a man onearth who can meet the king's demand; to such a degree that no greatand powerful king has asked such a thing of any magician or enchanteror Chaldean the thing that the king asks is difficult, and none can showit to the king except the gods, whose dwelling is not with flesh.' Notethat the Chaldeans' answer, as we observed earlier, thus echoes theking's speech in the preceding verses in a remarkable way: it exhibitsthe same literary structure, in which, apart from the peculiar use ofn b^p'bzi, the repetition of 'key words' at the beginning and end ofeach speech is notable. The king begins and ends with ]1V, 'time', theChaldeans with Ti'K ^b, 'there is no (one)'.

Dan. 2.40:

N'7nDDT v.bD b-^m pinn x'^ns -i bzp-bo N'^PSD ns'pn xian n-p'nn ID'PQT:;im pi - ]-':'X-'7D vv-n-'i

RSV: 'And there shall be a fourth kingdom, strong as iron, because ironbreaks to pieces and shatters all things; and like iron which crushes, itshall break and crush all these.'

Even in the translation it is evident that the syntactic relation betweenthe first two sentences is problematic; the second sentence merely elab-orates the theme of the power and destructiveness of iron and can easilybe connected with the last sentence. The first and third sentences bothstress the enormous destructive power of the fourth kingdom, and thefirst sentence would have been sufficient in itself to express what wasto be said about it. It seems much better to translate: 'And there shallbe a fourth kingdom, strong as iron. Just as iron breaks to pieces and

shatters all things it shall, like irou which crushes, break and crush allthese.'

Dan. 2.41-42:

mnn nybz IS' Q " nD ]in:ai -ins—i ^•DH ]in]Q

nspn mnn X-ID'P

RSV: 'And as you saw the feet ami toes partly of potter's clay and partlyof iron, it shall be a divided kingdom; l)ut. some of the firmness of theiron shall be in it, jiLst as you saw iron mixed with the miry clay. Andas the toes of the feet were partly iron and partly clay, so the kingdomshall be partly strong and partly brittle.'

Here the waw before the third sentence is lacking, but in other respectsthe situation is similar to the other passages which we are discussing.The same observations as with tlie preceding pa ssage are valid here.Both the first and the third sentence are concerned with the fourthkingdom. The fact that the writer only elaborates in this way uponthe iron kingdom and upon the kingdom of clay and iron c ertainlyindicates his csjjecial interest in these, more than in the precedingkingdoms. As in the preceding example, the seeming redundance inthis passage can be eliminated by the translation: 'And as you saw thefeet and toes partly of potter's clay and partly of iron, it shall be adivided kingdom; but some of the firmness of the iron shall be in it.Just as you saw iron mixed with the miry clay, and as the toes of thefeet were partly iron and partly clay, so the kingdom shall he partlystrong and partly brittle.'

Dan. 4.15 (vs. 18 in RSV):

13 y^np ]'7]bi<-mi -iRSV: 'And you, O Belteshazzar. declare the interpretation, becaxise allthe wise men of my kingdom are not able to make known to me theinterpretation, but you are able, for the spirit of the holy gods is in yon'.

It is clear that the two sentences which contain forms of the verb for'to be able' [bn'D/bD') are intimately connected, which was recognizedin the RSV by the insertion of 'bnt' between them. I think we can solvethis problem by assuming that 'i ' 5p"' 3 connects the two sentencesfolhjwing it, so that we may translate: 'And you. O Belteshazzar. de-clare the interpretation; while all the wise men of my kingdom are not

280 Aram.aic Studies 3.2 (2005)

able to make known to me the interpretation you are able, for the spiritof the holy gods is in you'.

Dan. 5.22-23:

bv^ 23

RSV: 'And yon his son, Belshazzar. have not hmnbled your hcai't, thoughyou knew all this, but yon have lifted up yourself against the Lord ofheaven;'

The parallelism between "I n'p rb^^T] N'7. 'you have not humbled yourheart', and nCDTinn, 'you have lifted up yourself, seems to be evidentat first sight, but it nmst be said that the exact opposite of the firstexpression is rather n3:ib Q"i, 'his heart was lifted up", which v as saidof king Nebuchadnezzar in verse 20. than PGDrinn. Therefore, though itis not at once apparent whether the alternative translation is superiorhere, it certainly is rather hkely. It would appear that the fact thatBelshazzar has not humbled his heart is presented in direct oppositionto the humility expressed by his ancestor Nebnchadnezzar, who afterhis former haughtiness "knew that the Most High God rules the king-dom of men, and sets over it whom he will' (5.21): 'But you his son,Belshazzar. have not humbled yonr heart. Though yon knew all this,you have lifted up yourself against the Lord of heaven;'

Dan. 6.4:

RSV: "Then this Dauiel became distinguished above all the other presi-dents and satraps, becansc an excellent spirit was in him; and the kingplanned to set him over tlie whole kingdom.'

Daniel's success is illustrated and emphasized by his possibly beingplaced 'over the whole kingdom' in the third sentence, and by his suc-cess over the satraps and other presidents, who were said in verses 1-3to have power over the entire kingdom, in the first. Here it is not atonce evident whether the second sentence should be taken with thefirst or the third, and I prtjpose the following alternative translationmainly because of the parallehsm with the other instances of '?5pr' 3^.. :i. Note, however, that HTP' mi, 'excellent spirit', is also mentionedas a reason for assigning Daniel to a high position in 5.12; this maybe an indication that it should indeed be taken with the last sentence:'Then this Daniel distinguished himself above all the other presidents


and satraps. Because an excellent spirit was in him, the king plannedto set him over the whole kingdom."Dan. 6.5:

a-b nrnoi -b^-h^-i Kin ]a'nD-'-i ^zp-"?:: nnDon* yb^-i^b nr-nci

RSV: 'hnt they could find no ground for complaint or any fanlt, becausehe wiis faithful, and no error or fault was found in him.'

The first and third sentences are evidently parallel to each other. It maybe felt to be slightly more appropriate to connect 'faithfulness' withthe following sentence, in which we find "b^, 'error', instead of with thepreceding, where ifhv, Aground for complaint', is not a good parallel,but otherwise it is not easy to decide with which sentence the middlesentence should be connected. The alternative translation makes indeedat least as good sense as the traditional one: 'but they could find noground for complaint or any fanlt. Because he was faithful, no error orfanlt was found in him.'

A comparable instance appears to be Dan. 6.23 (vs. 22 in the RSV):

rniv Kb r^b^2u SD' Q ["joipi jnip f]Ki ^bRSV: 'My God sent his angel and shut the lions' mouths, and they havenot hurt me, because 1 was found blameless hefore him; and also beforeyou. O king, I have done no wrong'.

Again the syntactical relations would seem to be not very (dear if weassume this translation; especially the last sentence follows in a ratherunexpected way, seemingly not connected with the preceding sentences,except for the use of the stem 'PDH in the first as well as in the thirdsentence. The third sentence is. however, much better integrated if wetranslate as: 'My God sent his angel and shut the lions' mouths, andthey have not hurt me; just as I was found blameless before him I havealso done no wrong before you, () king'.

The reason why this construction apparently escaped the attentionof most readers may be that it is usually possible to connect the secondsentence in such a construction with the first because of the rather widerange of meaning of n ^^^'b^ aud because it is indeed usually closelyrelated with it. The syndetic addition of a third sentence usually doesnot impede our ability to understand the passage as a whole, as wefeel free to translate the word 1, 'and', which introduces this sentence,in various ways. We saw, in fact, that in some cases we are unable to


choose between assigning the sentence with '1 'P pf' S to the precedingor to the following sentence, and it is only the cumulative evidenceconcerning this construction T.. .'~i ' ^p"' ? and its stylistic use in Daniel2 6 whif'h allows us to connect it with the following sentence withreasonable confidence.

Before continuing with a discussion of the background and functionof this phenomenon, we shall look at some more or less close parallelsin other Aramaic texts. As noted above, the construction ^.. .'l ' 3p"'73apparently appears also in Ezra 4.14 nr.ri itzn'?^ abz^n uba"! b^p'bD ]VDNDQ*? K]::nm xin'^s r]:i-b:) n.\nnb K:'7"-;nx ab s^':'^, which is usually, e.g.in the RSV, translated as 'Now because we eat the salt of the palaceand it is not fitting for us to witness the king's dishonor, therefore wesend and inform the king...', but which could also be rendered as 'Nowbecause we eat the salt of the palace, it is not fitting for us to witnessthe king's dishonor; therefore we send and inform the king...'. In thiscase we find the construction without the first sentence which alwaysprecedes it in Daniel. An interesting, though not a literal, parallel, with^K.. .'T b2pb instead of onr ".. .'i b2p~b'3 can be fonnd in one of theAramaic documents from Elephantine. In E. G. Kraeling, The BrooklynMuseum Aramaic Papyri (New Haven: Yale University Press, 1953),text 9, 16-18 we find a passage which exhibits a great stylistic likenessto the Daniel passages which we studied above:

nin n:T xnson pTD ':l^bn^ pTa nnnBDi -mQirrn -r N- ' I X]-b 20 yw n:xi 'Tb20 'T bnpb -annn T.ID3 rnn 17

b ran- 18 n:x ^N T

This passage is probably to be translated as:

'This honse whose boundaries and measurements are written and whosewords are written in this document I, Anaui, have given Jehoisma mydaughter gratis at my death. Because she supported me while I was oldof days—unable (to use) my hands, yet she supported me—I, in turn,have given (it) to her at my death.'' "

^^B. Porten followt l my proposal in his The Elephantine Papyri, m English: ThreeMillennia of CWs.s-Cultural Continuity and Change (Leiden: Brill. 1996). p. 240. Acomparable translation was already in his earlier Jews of Elephantine and Arameansof Syene (Jerusalem: The Hebrew University, Department of the History of theJewish People, 1980). p. 61. Ditfcrently in his Textbook of Aramaic Documents fromAncie.nt Egypt..., vol. 2: Contracts (Jerusalem: The Hebrew University, Departmentof the History of the Jewish People, 1989), pp. 8G-89.


Note that the structural resemblance is very striking. The first sen-tence would have been sufficient by itself, bnt a second sentence indi-cating the reason for the gift is inserted, after which the scribe returnsto the theme of the gift in the third.

In any case, it becomes evident that this idiom ) . . .'i '7?p,'''73 on theone hand is an additional instance of the anthor of Daniel giving theelements from other books which he derived, or rather alluded to, extraemphasis: against one case in the Aramaic chapters of Ezra he used itnine times in his own book (four times in ch. 2, once eacli in 4 and 5,and three times in ch. 6). On the other hand we see that this linguisticfeature, which onr author assigned to a very specific, probably newhterary function in his book, namely to slow down and evalute theaction at certain important places in the narrative, appears to connectthe Aramaic chapters of Daniel, as against the considerable differencesbetween them. In my 1988 article I interpreted this as meaning that theAramaic chapters of the book belong closely together, and that theyprobably had a pre-existence of their own before being inserted intothe present-day book of Daniel. It is now clear that their drawing theAramaic chapters together is part of the larger picture of the balanceof continuity and discontinuity within this unitary composition.

[AS 3.2 (2005) 285-87]DOI: 10.1177/1477835105059097

REVIEW

Hezy Mutzafi, The Jewish Neo-Aramaic Dialect of Koy Sanjaq (IraqiKurdistan) (Wiesbaden: Harrassowitz, 2004).

This volume is a description of a Neo-Aramaic dialect that is now onthe verge of extinction. Until the middle of the twentieth century it wasthe vernacular language of a thriving Jewish community in the townof Koy Sanjaq in Northern Iraq, but now is spoken only by a smalland ever dwindling number of elderly people who moved to Israel in1951. Until the exodus of all Jews from Iraq in the 1950s, numerousdialects of Aramaic were spoken by numerous Jewish communities inthe North of the country. These communities were located in the areawhere Kurdish is the predominant language of the Muslim population,i.e., broadly speaking, to the North of a diagonal line drawn on a mapacross the country between Mosul and Kirkuk. Christian communitiesin this region still speak a variety of dialects of Aramaic to this day,though the number of Christian Aramaic speakers has severely dimin-ished in the last few decades. Moreover, the dialects spoken by theJews were difi erent from those of the Christians. Even where Jews andChristians resided together in the same town, they spoke completelydifferent dialects. This applied to the town of Koy Sanjaq, where thereis a small Christian community. The Aramaic dialects of the region,therefore, constitute a remarkably diverse group, which is known asNorth Eastern Neo-Aramaic (NENA). The majority of the dialects ofthis group are today endangered. This applies to all the Jewish dialects.

Civen this background, the volume under review, which is a superbdescription of one of the endangered Jewish dialects, is particularlywelcome. Mutzafi has located some of the best surviving speakers ofthe dialect and, on the basis of extensive fieldwork, has produced adescription of the grammatical structure of the dialect, which is bothremarkably comprehensive and clear in its presentation.

In the introductory chapter, the position of the dialect within theNENA group is examined. For this Mutzafi draws on his wide knowl-edge of other dialects. He demonstrates the existence of a sub-group



of Jewish NENA dialects that comprise those that were spoken to theeast of the Great Zab river, including that of the Jewish community ofKoy Sanjaq. This group exhibits various distinctive features, such asthe shift of the interdental consonants *d and *t to / ] / , e.g. * 'eda > 'eia'festival', sata > sala 'fever'. The chapter on phonology is a very de-tailed account of the complex sound system in the dialect. Despite thiscomplexity, Mutzafi has devised an economical phonemic transcriptionsystem. The creation of a satisfactory type of transcription is alwaysone of the most difficult tasks for Neo-Aramaic dialectologists. Withgreat perception and a keen ear he has detected numerous interest-ing phonetic processes, including the neutralization of emphasis in theenvironment of pharyngal consonants, e.g. tem'a > tdui'a 'taste', andthe lengthening of short vowels before sonorants and sibilants, e.g. 'drba'four', 'ista 'six'.

The ensuing chapters describe the morphology, with sections de-voted to pronouns, verbs, nouns, adjectives, numerals and particles.As is the case with most NENA dialects, the verbal system of KoySanjaq exhibits numerous interesting features. There are, for example,some traces of an intransitive perfect that is conjugated with a non-ergative set of infiections, e.g. zil 'He has gone', zilex 'We have gone'.This is the only form of conjugation with intransitive past verbs in theJewish dialect of Sulemaniyya lying to the East whereas in the Jewishdialect of the Arbil region lying to the West only the ergative type ofinflection with prepositional suffixes is used with past verbs, e.g. zil-li,zil-lan. In Jewish Koy Sanjaq, both types of conjugation occur withpast verbs, but they have different functions. The zil type of conjuga-tion is, however, in the process of being replaced by a compound verbalconstruction qtila + copula, which is used with both transitive and in-transitive verbs. This compound construction exhibits an interestingstructural difference according to whether it expresses what Mutzafiterms a 'stative perfect' or a 'dynamic perfect'. The dynamic perfectis more 'verb-like' in meaning and, accordingly, the compound verbalconstruction loses some of its nominal properties, in that the qtila baseis not infiected with the plural nominal inflection when the subject isplural, e.g. dmixe-wex 'we are asleep' (stative perfect) but qtila-wex'we have killed' (dynamic perfect).

The morphology section is followed by samples of texts, which con-stitute only a small proportion of the large text corpus that Mutzafihas collected in his research. These texts include an oral translation ofGenesis chapter 1.

REVIEW 287

The volume closes with a glossary of all the words that occur inthe grammar and texts. Unlike most published grammars of NENA,this glossary contains etymologies of all words. Mutzafi has been aidedin establishing many of these etymologies by his wide knowledge ofother NENA dialects and indeed copious references are made to otherdialects in the notes to each lemma.

This is a model grammar of a Neo-Aramaic dialect, which makesan invaluable contribution to the field.

Geoffrey Khan, University of Cambridge

[AS 3.2 (2005) 288-91]10.1177/1477835105059098

BIBLIOGRAPHY OF THE ARAMAIC BIBLE

Compiled by the staff of the Peshitta Institute, Leiden, and of the Targum Institute,Kampen. Please send relevant bibliographical material to [email protected],and if possible also an offprint to the Semitic Institute, Kampen Theological Uni-versity, POB 5021, 8260 GA Kampen, The Netherlands.

AviSHUR, Yitzhak, 'The "Great Sin" and its Punishment: Remarks onthe Letter of the King of Sidon to the King of Ugarit, from the RecentPubhcation, in the Light of the Bible (OT), the "Mishna", and theII Targum of the Book of Esther', Zeitschrift filr Altorientalische undBiblische Rechtsgeschichte 10 (2004), pp. 207-12.

Keywords: Texts: Targum Sheni to Esther

CHILTON, Bruce D., 'From Aramaic Paraphrase to Greek Testament'in Craig A. Evans (ed.). From Prophecy to Testament: The Functionof the Old Testament in the New (Peabody, MA: Hendrickson, 2004),pp. 23-43.

Keywords: Aramaic background of the New Testament

CHILTON, Bruce D., 'Targum Jonathan of the Prophets I', in J. Neusnerand A. Avery-Peck (eds.). Encyclopedia of Midrash: Biblical Interpre-tation in Formative Judaism (2 vols.; Leiden: E.J. Brill, 2005), cols.889-908.

Keywords: Targum Jonathan to the Prophets

CHILTON, Bruce D., 'Targum Jonathan to the Prophets II: The EormerProphets', in J. Neusner and A. Avery-Peck (eds.). Encyclopedia ofMidrash: Biblical Interpretation in Formative Judaism (2 vols.; Leiden:E.J. Brill, 2005), cols. 908-27.

Keywords: Targum Jonathan to the Prophets

DiEZ MERINO, Luis, 'Los estudios targumicos en la actualidad', EstBib62 (2004), pp. 347-90.

Keywords: Targum (general)

BIBLIOGRAPHY 289

DuHAiME, Jean, review of David Shepherd, Targum and Translation:A Reconsideration of the Qumran Aramaic Version of Job (Assen: VanGorcum, 2004), JHS 5 (2004-2005). [onhne journal:http://www.jhsonline.org or http://purl.org/jhs]

Keywords: Targum Job from Qumran

EVANS, Craig A., 'The Aramaic Psalter and the New Testament: Prais-ing the Lord in History and Prophecy', in Craig A. Evans (ed.). FromProphecy to Testament: The Function of the Old Testament in the New(Peabody, MA: Hendrickson, 2004), pp. 44-91.

Keywords: Aramaic background of the New Testament

FLESHER, Paul, 'Pentateuchal Targums as Midrash', in J. Neusner andA. Avery-Peck (eds.). Encyclopedia of Midrash: Biblical Interpretationin Formative Judaism (2 vols.; Leiden: E.J. Brill, 2005), cols. 630-46.

Keywords: Pentateuchal targums, midrashic interpretations

HOUTMAN, Alberdina, and Johannes de Moor, Introduction, Additionsand Corrections, Indices (A Bilingual Concordance to the Targum ofthe Prophets, 21; Leiden: E.J. Brill, 2005) [ISBN 90-04-13107-8].

Keywords: Targum Jonathan to the Prophets, concordance, lexicography

HOUTMAN, Alberdina, 'The Role of Abraham in Targum Isaiah', AS3 (2005), pp. 3-14.

Keywords: Abraham, Targum Isaiah

KuNTZMANN, Raymond, 'Le Targum des Chroniques I, 1 (TC I, 1) etla geographie synagogale', RevScRel 79 (2005), pp. 147-61.

Kejrwords: Targum Chronicles, geography

LATTKE, Michael, 'Eine iibersehene Textvariante in den Oden Salomos(OdSal 36,1a)', Zeitschrift fiir Antikes Christentum 8 (2004), pp. 346-49.

Keywords: PeshittaTexts: Odes of Solomon

LEGRAND, Thierry, 'Les targums du Pentateuque: Leur rapport auxdocuments inscrits, aux livres et a la Torah', RevScRel 79 (2005), pp.127-46.

Keywords: Keywords : Pentateuchal targums

LEVINE, Etan, 'Targum, Conceptual Categories of, in J. Neusner andA. Avery-Peck (eds.). Encyclopedia of Midrash: Biblical Interpretationin Formative Judaism (2 vols.; Leiden: E.J. Brill, 2005), cols. 927-94.

Keywords: Targums (general)


LOREIN, Geert W., 'NniD'pa in the Targum of the Prophets', AS 3(2005), pp. 15-42.

Keywords: kingship, lexicography

MAIER, Johann, 'Beobachtungen zum Text von Gen 18,21', in MartinEbner and Bernhard Heininger (eds.), Paradigmen auf dem Priifstand.Exegese wider den Strich. Festschrift fiir Karlheinz Miiller zu seinerEmeritierung (Neutestamentliche Abhandlungen, 47; Miinster: Aschen-dorff Verlag, 2004), pp. 141-54.

Keywords: Pentateuchal targumsTexts: Gen. 18

MORRISON, Craig, 'The Relationship of the Peshitta Text of SecondSamuel with the Peshitta of First Chronicles', AS 3 (2005), pp. 59-81.

Keywords: Peshitta Samuel, Peshitta Chronicles

PETERSEN, Wilham L., review of Robert E. Shedinger, Tatian and theJewish Scriptures: A Textual and Philological Analysis of the Old Tes-tament Citations in Tatians Diatessaron (Corpus Scriptorum Chris-tianorum Orientahum 591; Subsidia 109 Leuven: Peeters, 2001), RBL6 (2005) [onhne journal: http://http://www.bookreviews.org/].

Keywords: Tatian's Diatessaron, Targum

RIBERA-ELORIT, Josep, 'Tecnicas de traduceion del Targum : Laversion interpretada de ntn en el Targum de Ezequiel', in A. GonzalezBlanco, J.P. Vita and J.A. Zamora (eds.), De la tablila a la inteligen-cia artificial (Zaragoza: Instituto de Estudios Islamicos y del OrienteProximo, 2003), pp. 641-48.

Keywords: Targum Ezekiel, lexicography, grammar

RIBERA-PLORIT, Josep, review of Eveline van Staalduine-Sulman, TheTargum of Samuel (SAIS, 1; Leiden: E.J. Brill, 2002), Biblica 85 (2004),pp. 443-46.

Keywords: Targum Samuel

RIBERA-ELORIT, Josep, 'Hagiographa, Targums to', in J. Neusner andA. Avery-Peck (eds.). Encyclopedia of Midrash: Biblical Interpretationin Formative Judaism, 2 vols. (Leiden: E.J. Brill, 2005), cols. 148-73.

Keywords: Targum to the Writings

Rizzi, Giovanni, 'Penomeni ermeneutici nella tradizione di Pesh Sapienza',in Giuseppe Bellia and Angelo Passaro (eds.), II Libro della Sapienza;tradizione, redazione, teologia (Roma: Citta Nuova, 2004), pp. 233-69.

Keywords: PeshittaTexts: Wisdom of Solomon

BIBLIOGRAPHY 291

SALVESEN, Alhson, 'Jacob of Edessa's Version of Exodus 1 and 28',Hugoye: Journal of Syriac Studies 8, 1 (January 2005). [online journal:http://syrcom.cua.edu/syrcom/Hugoye]

Keywords: Peshitta, Jacob of EdessaTexts: Exodus 1, Exodus 28

SMELIK, Willem P., 'Translation as Innovation in b. Meg. 3a', in L.Teugels and R. Ulmer (eds.). Recent Developments in Midrash Re-search: Proceedings of the 2002 and 2003 SBL Consultation on Midrash(Piscataway: Gorgias Press, 2005), pp. 25-49.

Keywords: Aquila, Jonathan, Onqelos, Theodotion, Talmud

SPERBER, Daniel, 'pi;Q nQ:in icnp bw 'nuQ-iuran:j'7"n nx 'prb -[-n in bsi'^onp^S'' [On the Need to Correct Krauss's Lehnworter: Pnqrysyn],Jewish Studies: An Internet Journal 3 (2004), PDF-file, five pages.[online journal: www.biu.ac.il/JS/JSIJ]

Keywords: Lexicography

SuOMALA, Karla R., Moses and God in Dialogue. Exodus 32-34 ™Postbiblical Literature (Studies in Bibhcal Literature, 61; New York:Lang, 2004).

Keywords: Pentateuchal TargumsTexts: Exodus

VAN ROOY, Harry p., 'The Headings of the Psalms in the Two SyriacVersions of the Commentary of Athanasius', Old Testament Essays 17(2004), pp. 659-77.

Keywords: Syriac headings of the Psalms

VAN ROOY, Harry P., 'The Syro-Hexaplaric Headings of the Psalmsin Manuscript 12t3', AS 3 (2005), pp. 109-26.

Keywords: Syro-Hexapla, Peshitta, Syriac headings of the Psalms

cursodearameo.com.ar - aramaic studies volume 3

Documents

american oriental

finnish oriental

universita

grammatik

la nuit pascale

van der woude

thousand oaks

hebrew union