What People Say They Do With Words

ByJozef Francisca Verschueren

Grad. (University of Antwerpen, Belgium) 1974 M.A. (University of California) 1976

DISSERTATION

Submitted in partial satisfaction of the requirements for the degree ofDOCTOR OF PHILOSOPHY

inLinguistics

in the GRADUATE DIVISION

OF THEUNIVERSITY OF CALIFORNIA, BERKELEY

Approved:,. ...........7kwv. & l . 13 7

hairman

DOCTORAL DEGREE CONFERRED JUNE 14,1980

ACKNOWLEDGMENTS

After writing a d is ser ta t ion i t i s not easy to take one's pen again

to express some acknowlegments — however s in cere ly they are meant —

in view of the fa c t that a l i s t of the people who contributed d ir ec t ly

or in d ir ec t ly to the creative process i s bound to remain as hopelessly

incomplete and unfinished as the d is ser ta t ion i t s e l f . Let me, therefore ,

r e s t r i c t myself to the absolute e s s e n t i a l s .

F irs t of a l l I should thank Charles Fillmore, Louis Goossens, George

Lakoff and John Searle who have a l l witnessed and guided, each in his

own way, the growth o f my ideas. I am convinced that the shortcomings

of t h i s work can a l l be traced to fa i lu res to follow th e ir advice; ever

so often when I thought I knew what I was doing I chose a direction

d i f feren t from the one they indicated only to find out much la ter that

they were right to begin with; such experiences make i t quite c lear to

me that i f I had not followed {or had) any o f the ir advice the work for

th is t h e s i s , from conception to b ir th , would have taken a decade instead

of ju s t two years — or, more probably, I would not have written i t at

a l l .

(i)

Among the people who have read e a r l i e r drafts of the t e x t I have to

thank Geoffrey Leech who took the trouble to write down his comments.

Of course I have a lso benefited from discussions with numerous people,

e sp e c ia l ly some participants at the Twelfth International Congress of

Linguists (Vienna 1977) and the 13. Linguistisches (Colloquium (Gent

1978) who were patient enough to s i t through my half -bo i led programmatic

statements, as well as a number o f participants at the Conference on

P o s s i b i l i t i e s and Limitations of Pragmatics (Urbino 1979); I refrain

from naming people out of fear to forget some.

Thanks are a lso due to the Commonwealth Fund o f New York for granting

me the Harkness Fellowship which enabled me to s ta r t studying l in g u i s t i c s

at the University of California , Berkeley, in 1975, and to the Belgian

National Science Foundation which has been my source of support ever since

the end o f my Harkness tenure.

F ina l ly , I want to express my gratitude to Ann who succeeded in keeping

me away from the d is ser ta t ion j u s t about enough for me to be able to keep

writing i t .

( i i )

TABLE OF CONTENTS

I. THE COMPARATIVE LEXICAL APPROACH TO LINGUISTIC ACTION 1A. Introduction 1B. Speech act theory at the breaking point ? 2C. Reflections on the study of social action 15D. Advocating unorthodox scholarship 26

II . RESEARCH PRINCIPLES AND TERMINOLOGY 34A. Introduction 34B. The object of investigation 35C. Principles of lex ical semantics 46D. Circularity and how to avoid i t 69E. Problems of representation 76F. Concluding remark 81

II I . THE SEMANTICS OF SILENCE 83A. Introduction 83B. The frame of s i lence 84C. S i lence , meaning, context and communication type 86D. The codes of s i lence 97E. The sound of s i lence 107F. Silence and i t s causes 119G. The tacens and his motives 127

(111)

H. Silence and i t s interlocutors 139I. The topics of s i lence 143J. Silence and non-propositional meaning 150K, Silence and context 151L. The 11locutionary force of s i lence 155M. Conclusions 155

IV. THE SEMANTICS OF LYING 158A. Introduction 158B. The frame of lying 160C. Lying and truth 162D. The illocutionary force of lying 176E. Lying and perlocutionary intent 183F. The textual dimension of lying 188G. Lying and value judgments 191H. Note on human imagination 193I. Conclusions 195

V. THE SEMANTICS OF DIRECTING 198A. Introduction 198B. The frame of directing 206C. The d irect iv i ty of directing 206D. Directing and i t s social sett ings 210E. Directing and i t s goals 226F. The direct ionali ty of directing 236G. Directing and authority 252H. Miscellaneous dimensions 258I. Conclusions 259

VI. THE SEMANTICS OF FORGOTTEN ROUTINES 262A. Introduction 262

(1v)

B. The expression o f emotions and a t t i tudes 265C. Negative and pos i t ive responses 278D. Fixed routine responses 287E. Conclusions 292

VII. CONCLUDING REMARKS 294

REFERENCES 296

( v)

1

I. THE COMPARATIVE LEXICAL APPROACH TO LINGUISTIC ACTION

A. INTRODUCTION

1. The best known theory o f l i n g u i s t i c action i s no doubt speech act

theory which entered i t s incubation period with the la te J.L. Austin's

How t o Do Things w i th Words and f i n a l l y hatched when J.R. Sear le 's

Speech A c t s was published. Since then dozens o f scholars have been

engaged in ref ining the theory, applying i t to n on - l i tera l and non-serious

uses o f language and indirect speech a c t s , extending i t to speech act

sequences and incorporating i t in a wider speech a c t i v i t y theory (what the

Germans ca ll S p r e c h t a t i g k e i t s t h e o r i e as opposed to S p r e ch h a n d lu n g s th eo r ie )

or in the theory o f action in general. In th i s chapter I shall argue that

in sp i te of the bulk and d iv e rs i ty o f the result ing 1iterature^ nearly a l l

speech act scholars, whether l in g u i s t s or philosophers, approach l i n g u i s t i c

action from an e s s e n t i a l l y theoret ica l point of view. Empirical studies are

rare. I t i s not my intention to repudiate th i s approach. However, I am about

to present one o f the possible a l t e r n a t iv e s , the comparative lex ica l

approach which I w i l l present as an example o f what I ca l l the empirical-

1 To get an idea of the voluminous and diverse literature on the subject, the reader should have a look at Verschueren (1978a).

2

conceptual method. This a lternat ive should be regarded as a necessary

supplement rather than as a replacement.

2. The point of th is introductory chapter i s to elucidate the foregoingof

obscurit ies . F irst I shall point out some the current threats to speech

act theory. Then the ir origins will be discussed with reference to some

ref lec t ions on the study of social action in general; in th is context the

contrast between theoretical and empirical-conceptual methods wil l be

explained. Finally , the comparative lex ica l approach wil l be proposed to

divert the threats and to give a new impulse to the study of l in g u i s t i c

action. As said in the previous paragraph, the proposal in question is

not intended as a s e l f - s u f f i c i e n t way of describing l in g u i s t i c action:

i t i s a one-sided approach meant to counterbalance the diagonally opposed

methodology embodied in almost all inquiries into speech act phenomena —

with a couple of rare exceptions to be mentioned la ter .

B. SPEECH ACT THEORY AT THE BREAKING POINT ?

3. For the sake of the uninit iated reader I pass the basic concepts of

speech act theory, as they are habitually used, in review. A speech a c t

i s an act performed in or by using language; the intentional utterance

of any sentence const i tutes a speech act . Its main components are a

p r o p o s i t i o n a l c o n te n t (consist ing of a reference and a predication) and

an i l l o c u t i o n a r y f o r c e ( i . e . i t s role as a statement, a request, an

order, a promise, a question, e t c . ) . The e f f e c t produced in a hearer by

performing a speech act i s i t s p e r lo c u t io n a r y e f f e c t (e .g . convincing,

3

persuading, pleasing, annoying, e t c , ) - A speech act viewed under the

aspect of i t s i l locu t ionary force i s ca l led an ' i l l o c u t io n a r y a c t ; viewed

under the aspect of the perlocutionary e f f e c t i t produces i t i s a p e r l o -

c u t io n a r y a c t .

I l locut ionary acts are analyzed by means of formulating the const i tu t ive

2w l e s for th e ir f e l i c i t o u s performance: propositional content rules ( sp ec i ­

fying r e s tr ic t io n s with respect to reference and predication; e . g . when

making a promise the speaker predicates a future action to be carried out

by himself); preparatory rules (expressing what the act implies or the

conditions without which the act has no point; e .g . a promise only makes

sense i f the speaker be l ieves that the hearer would prefer his doing the

promised act over his not doing i t ) ; s in ce r i ty rules (formulating the

psychological s ta te of the speaker; e .g . a promise expresses the speaker's

intention to carry out the action in question); e s sen t ia l rules (specifying

the essence o f the act; e .g . a promise counts as the speaker's undertaking

of an obl igation to do something). Often such rules are offered in the form

of necessary and s u f f i c i e n t conditions or f e l i c i t y c o n d i t i o n s .

Most — i f not a l l — languages possess a number of verbs such as TO

PROMISE and TO ORDER which describe types of i l locut ionary acts but which can

also be used to perform the act in question as in "I order you to leave

the room" and "I promise not to forget the book". These are c a l led p erfo rm ­

a t i v e v e r b s , and the two utterances presented are e x p l i c i t p e r f o r m a t i v e s .

Verbs which cannot be used performatively but which can also describe types

of i l locu t ionary acts form, together with the performative verbs, the c lass

of sp eech a c t v e r b s .

Some speech acts such as "Can you pass the s a l t ?" have a double i l l o -

cutionary layering: the ult imately intended force ( i . e . the primary i l l o -

4

cutionary force) i s that of a request to pass the s a l t ; but th i s i s

achieved in d irec t ly by means o f a secondary force (which i s the force

ty p ic a l ly associated with the grammatical form of the u t terance) , namely

that of a question. Hence these are ca l led i n d i r e c t sp eech a c t s .

Another example i s "The door i s s t i l l open" which i s a plain statement

but which, in certain s i tua t ion s ( e .g . i f the hearer i s the speaker's

servant) , can take the function of a request or even an order to c lose

the door.

4. In le s s than a decade, speech act theory has become a prodigious

excrescence from the body of language studies — and i t i s s t i l l growing

every day. The continued attract ion i t exerts does not mean that i t i s

in good health. On the contrary, I am about to present some problems

which could soon prove to be fa ta l i f no cure i s found. I discuss them

under the following captions: the meaning/force controversy, the perlo-

cution controversy, the se r io u sn e s s / l i t e r a ln e s s controversy, the sequencing

controversy, the c la s s i f i c a t io n controversy and the indirectness contro­

versy. Of course the controversies themselves do not const i tu te threats

to the theory o f speech a c ts . The trouble i s in one of the solutions pro­

posed in each case, which menace some of the basic in s igh ts which made

speech act theory appealing in the f i r s t place.

5. The m ea n in g /fo rce c o n t r o v e r s y i s the l e a s t harmful o f th e l o t . I t

caused an early childhood ailment from which speech act theory seems to

have recovered. But unlike with measles there i s no guarantee that i t w i l l

not recur.

5

Austin paraphrased the periocutionary act as an act performed by saying

something and the i l locut ionary act as an act performed i n saying some­

thing. The act o f saying something he dubs the 1ocutionary act . Apart

from i t s phonetic-phonological and i t s grammatical properties a locution-

ary act i s a lso characterized by a certain meaning (which Austin def ines ,

in Frege's terms, as i t s sense and reference) . In that way the term meaning

acquired a technical sense (even though Austin did not intend i t as such)

in opposition to the f o r c e of an utterance which i s i t s status as an

i l locut ionary act o f some type.

Quite early in the development of speech act theory some scholars pointed

out correct ly that in the ordinary language sense of the word 'meaning'

i l locut ionary force i s part o f meaning and cannot be placed in contrast

with i t . Though th i s argument, as i t stands, cannot be exploited to refute

Austin who, as any other scholar, i s free to iposit his own terminological

conventions, the remark is not f u t i l e . I ts j u s t i f i c a t i o n i s the underlying

question of whether i l locu t ionary force can be distinguished from other

aspects o f meaning in any c lear and consistent way. Usually such a d i s ­

t in c t io n i s taken for granted. Hence the controversy between those who

give a negative reply to the question and those who answer i t in the po s i t ive

i s v ir tu a l ly absent from the l i t e r a tu r e . At f i r s t s igh t th i s absence o f an

overt controversy conceals i t s p o te n t ia l ly dangerous implicat ions. But at a

deeper level i t reveals the extreme acuteness o f the problem. The few who

believed that there was no c lear d i s t in c t io n between i l locut ionary force and

other aspects of meaning have concluded that therefore i l locu t ionary forces

do not e x i s t or at l e a s t that the concept i s u t te r ly use less ( e .g . L.J. Cohen

1964). Such a conclusion i s completely logical i f one adheres to the pr inc ip le

6

that terminological c la r i ty i s an absolute prerequis ite for f ru i t fu l

thinking. And ye t i t deprives us of a way o f ta lking about basic se ­

mantic phenomena such as the difference between "He opened the door"

and "Did he open the door ?". In other words, i t deprives us o f a notion

which carried the promise of opening completely new horizons of under­

standing when i t was f i r s t introduced and which has gone a long way by

now in trying to f u l f i l that promise.

With the acceptance or reject ion of the notion of i l locut ionary force

speech act theory stands or f a l l s . Therefore, the threat radiating from

t h is controversy is not f e l t immediately: those who rejec t the notion

disappear from the speech act camp without leaving a trace. The real

trouble s tar ts when speech act th eor is ts themselves begin to doubt the

v a l id i ty of the notion. I t i s hard to predict when th i s wil l happen.

But sooner or la ter i t may, because i l locu t ionary force i s n o t c lear ly

dist inguishable from other aspects of meaning. E xp lic i t performatives

should su f f ic e to demonstrate t h i s non-discreteness: in such utterances force

the i l locu t ionary i s a lso indicated in the propositional content and

i t i s mainly due to th i s presence in the proposition that the utterances

get th e ir force; in other words, part o f the propositional content

functions as an i l locu t ionary force indicating device.

6. Next in l in e i s the p e r l o c u t i o n c o n t r o v e r s y . Remember that periocution­

ary acts were defined as the production, by saying something, o f any con­

sequential e f f e c t in the hearer. By ordering somebody to leave the room

you can make him leave the room; but you can a lso make him laugh, annoy

him, anger him, make him commit su ic id e , e t c . All o f these are perlocution-

7

ary e f fec ts - - enough to drive any law-and-order-loving mind to despair.

The contingent and capricious nature of periocutionary e f fe c t s has

prompted most speech act theorists to apply the time-honored practice

of avoiding trouble; in other words, the periocutionary aspect of l in ­

g u is t ic acts was omitted from most accounts. Some intended e f fe c t s such

as the intention to make the hearer do something, which i s inherent in

the act of commanding, could not be shunned (but got relabeled as the

i l l o c u t i o n a r y p o i n t , which i s not to say that there i s a one-to-one

correspondence between intended perlocutionary e f f e c t s and i llocutionary

points) . But no systematic invest igation has ever been made of a l l the

e f fec ts that are typical ly associated with particular types of i l lo c u ­

tionary acts — l e t alone the less typical ones.

Whereas the meaning/force controversy only threatens speech act theory

from without, the perlocution controversy is a menace from within. Not

al l perlocutionary e f fe c t s (even those typical ly associated with certain

types of i llocutionary acts) can be given an equally prominent place.

Consider the subtle pragmatic difference between "I t e l l you that P but

I'm not trying to make you believe that P" and "I order you to do P but

I'm not trying to get you to do P": both are deviations from a standard

s i tuation , but th is i s more so for the order than for the statement. Yet

the absence of a systematic investigation i s a gap that no bridge can span.

This i s where I disagree with Searle (personal communication) who would

not deny that a l l i llocutionary act types have an intended e f f e c t typically

associated with them, but who regards these as n ecessa ry by d e f i n i t i o n for

some i llocutionary acts and as optional for others; according to him only

the necessary ones are worth considering in a theory of speech acts .

8

Though the peHocution-phobia i s founded on the b a s ica l ly sound urge for

contro l lable data, i t v e i l s fundamental aspects of l in g u i s t i c action such

as the speaker's intention to make something known to the hearer when

making a statement.

7. The f e l i c i t y conditions formulated for the performance o f i l locu t ionary

acts usually bear on th e ir performance in l i t e r a l and serious discourse.

In the so -ca l led p a ras i t ic forms of discourse, i . e . in the non- l i tera l

ones ( irony, metaphor, e t c . ) and in the non-serious ones (play ac t ing ,

language teaching, e t c . ) , these condit ions are usually v io la ted . A common

and acceptable procedure i s to describe such v io la t io n s in terms o f de­

v ia t ions from a standard. Recently, however, voices have been raised

against t h i s l in e of thought. I t i s argued that i f a promise that the

speaker does not intend to keep s t i l l counts as a promise, that then i t

i s wrong to regard the intention to carry out the promised action as a

condition on the f e l i c i t o u s performance of the act of promising. I f th i s

conclusion i s correct — which I think i t i s not — then i t i s no longer

possible to formulate f e l i c i t y conditions at a l l : I am not aware of the

existence o f any that are immune against s imilar c r i t ic i sm . Therefore the

s e r i o u s n e s s / l i t e r a l n e s s c o n tr o v e r s y i s capable of undermining the

apparatus to describe speech acts .

8. A s im ilar worry ar ises from the sequ en cin g c o n t r o v e r s y . I t i s argued,

quite co rrec t ly , that speech acts rarely occur in i so la t io n and that the

sequence to which they belong inf luences the ir semantic properties . Again,

trying to incorporate such influences in the description o f speech act

9

types instead of departing from a standard analysis which i s allowed to

change depending on the context i s a maneuver which menaces the very

p o s s i b i l i t y o f speech act ana lys is . Therefore some scholars have recently

proposed to supplant speech act theory with a t o t a l l y new speech a c t iv i t y

theory rather than to supplement i t .

The attacks on speech act theory presented in th i s and the previous

paragraphs are based on the practice of cr ic iz in g a theory for what i t

does not do; such c r i t ic i sm i s usually misguided unless the theory c r i t i ­

cized i s in ten ded to do what i t does not do. F e l i c i t y conditions on speech

acts are not meant (at l e a s t not in Sear le 's proposal) to cover cases of

non- l i tera l and non-serious language, nor to explain the interact ion of

individual speech acts with the ir context. The conscious — and valid —

policy i s to describe standard forms. By f a i l i n g to r e a l iz e t h i s , those

who rejec t speech act theory on those grounds make the age-old mistake

o f throwing out the child with the bath-water: the very fa c t that speech

act theory concentrates on standard forms enables i t to shed l i g h t on a l l

'dev ia t ions ' . I t should be kept in mind that the term 'deviation' i s not

meant to imply ' in f e r io r i t y ' nor ' r a r i t y ' .

9. Both Austin and Searle proposed a f i v e - f o ld c la s s i f i c a t io n of i l l o -

cutionary ac ts . Searle (1976) comes up with the following c la s se s : repre­

sentat ives (the point of which i s to represent a s ta te of a f f a i r s ; which

have a word-to-world direct ion of f i t , i . e . the intention i s to make the

words f i t the world; in which a b e l i e f i s expressed; and in which any

proposition can occur) , e .g . statements; d irec t ive s (the point of which

i s to d ir ec t the hearer towards doing something; which have a world-to-

10

word direction of f i t ; in which a wish i s expressed; in which the propo­

s i t i o n i s a future act done by the hearer), e .g . orders; commissives (the

point of which i s that the speaker commits himself to doing something;

which have a world-to-word direct ion of f i t ; in which an intention i s

expressed; in which the proposition i s a future act done by the speaker),

e .g . promises; expressives (the point o f which i s that a certain psycho­

log ica l s ta te i s expressed; which have no direct ion of f i t ; in which a

wide range of psychological s ta te s can be expressed; in which the propo­

s i t i o n ascribes a property or act to the speaker or the hearer), e .g . con­

gratu lat ions; declarations (the point of which i s to bring something about

in the world; which have both a word-to-world and a world-to-word direct ion

of f i t ; in which no psychological s ta te i s expressed; in which any propo­

s i t i o n can occur), e .g . an excommunication.

Apart from those who present s l i g h t l y modified versions o f t h i s c l a s s i ­

f i c a t io n there are l in g u is t s who maintain ser ious ly that the number of

d i s t i n c t c la sses i s somewhere between f iv e hundred and s ix hundred, and

others who contend that there are exactly three basic speech acts from which

al l the others can be derived. Let us ca l l them the s p l i t t e r s and the lumpers,

respect ive ly . The s p l i t t e r s ' a t t i tu de i s based on some postulated pr inc ip les

of s c i e n t i f i c taxoncmies such as absolute discreteness ( i . e . the unaccepta­

b i l i t y of overlap between categories) and on the use of the speech act verb

vocabulary of a natural language as one's point of reference. What the

lumpers are doing i s not r ea l ly c la s s i fy in g speech acts but trying to de­

termine the basic function o f a couple of grammatical forms in such a way

that the other uses can be derived from i t . The i l locut ionary character of

such functions i s doubtful. Therefore i t i s not surprising that they use

11

terms such as ' in d ic a t iv e ' , ' in terrogat ive ' and 'imperative' as names for

the categories they propose.

Both for s p l i t t i n g and for lumping theoret ical j u s t i f i c a t i o n s can be

adduced. Yet these two poles of the c l a s s i f i c a t i o n c o n t r o v e r s y , but es ­

p e c ia l ly the second one, seem to move away from the central tenet of

speech act theory, namely an understanding o f the phenomenon of i l l o c u -

tionary force.

10. The lumper a t t i tu d e , i . e . a strong preoccupation with grammatical

forms, inev itab ly leads us to the i n d i r e c t n e s s c o n t r o v e r s y . The in t u i t iv e

relevance of the notion o f an ind irec t speech act i s the following: in

some cases people mean more than what they say; t h i s i s not only true

with respect to the propositional content o f utterances, but a lso with

respect to th e ir i l locu t ionary force; such indirectness needs to be

accounted f o r , which can be done in connection with the propositional

content in terms of irony, metaphor, presupposition, some types o f impli-

cature, e tc . and in connection with the i l locut ionary force in terms of

ind irec t speech acts (which does not mean that ind irec t speech acts do not

often involve 'unsaid' propositional meaning as w e l l ) . However, i f one

adopts the grammar-based lumper pos i t ion one i s not only forced to des­

cribe the request "Can you pass me the s a l t ?" as an ind irect "eech act

because i t i s an imperative subtype disguised as an in te r r o g a t iv e , but a lso

the e x p l i c i t performative "I order you to leave the room" because i t

functions as an imperative though i t s structure i s ind ica t iv e .

Leaving aside the issue of whether one should not regard certain pro-

posit ional features {such as 'I order') as l in g u i s t i c indicators o f i l l o -

cutionary force on a par with syntactic features {such as grammatical

mood), we can regard the conclusion as lo g ic a l ly correct. However, the

claim of the indirectness (or, for that matter, the non-directness) of

e x p l i c i t performatives i s in t u i t iv e ly vacuous since the speaker does not

mean more than what he says. I f you say "I promise to bring the book"

but you do not bring i t , you cannot reply to my accusation of your break­

ing your promise by saying "I did not promise, I only stated that I pro­

mised to bring the book". However troublesome vagueness may be, i t i s les s

harmful to scholarship than vacuity: though one can f i l l an empty b o t t le ,

no one has ever succeeded as yet in substantiating a vacuous claim.

The main argument in favor of the lumper-view of e x p l i c i t performatives

i s based on the notion of l i t e r a l meaning: the l i t e r a l meaning of "I order

you to leave the room" i s said to be a statement-meaning. The argument can

e a s i ly be turned against th i s claim. The idea of l i t e r a l meaning as a

context-free meaning has long been abandoned as a f i c t io n and replaced by

the meaning a word or an utterance has in some kind of neutral , unmarked or

minimal context. Neutrality and markedness in th is sense are basica l ly

cognit ive notions. A simple t e s t to discover the typ ica l ly associated

minimal context for "I order you to leave the room" would be to present

the sentence to a number o f informants and to ask them to construct a

typical context for i t . I am convinced that practica l ly a l l o f them would

come up with a context in which "I order you to leave the room" i s an

order instead o f a context in which i t i s a reply to a question such as

"What do you do i f I make too much noise ?". Hence, the psychologically

basic force of the utterance, i . e . the force i t has in a maximally neutral

context and which can thus be regarded as (one of) i t s l i t e r a l f o r c e ( s ) , i s

13

that of an order. Therefore no indirectness i s involved. I am not advo­

cating the promotion of such t e s t s to the s in g le (or even the main) decis ion

procedure to decide whether an i l locut ionary force i s d ir e c t ly or in d ir e c t ­

ly conveyed, because i t i s quite c lear that the statement-meaning o f "I or­

der you to leave the room" as a reply to "What do you do i f I make too much

noise ?" i s a lso quite l i t e r a l and d irec t . Moreover, a s t r i c t adherence to

the t e s t might prove "Can you pass me the s a l t ?" to be l i t e r a l l y and d i ­

r ec t ly a request the pass the s a l t rather than an ind irec t speech act .

Trying to solve these problems would lead us to a discussion o f the d i f f e r ­

ent types o f conventions involved in language and language use. All I am

in terested in here i s demonstrating that the notion o f l i t e r a l meaning

cannot be re l ied upon to support the lumper view of indirectness in lan­

guage.

11. We are confronted with a ver i tab ly paradoxical s i tu a t io n . With respect

to each o f the controversies discussed I have drawn the attention to one

or two poss ib le solutions which have been proposed at some point in the

philosophical or l in g u i s t i c l i t e r a t u r e . The paradox i s the following:

though each o f these so lutions i s l o g ic a l ly sound (at l e a s t from some

theoret ical point of view), they are not compatible with the fasc inat ing

and i n t u i t i v e l y correct premises of speech act theory. Let me recapitulate:

( i ) I l locutionary force , the fundamental concept o f speech act theory,

can be rejected as a valid notion i f one be l ieves that i t would only

be valid i f i t could be c le a r ly distinguished from other types of

meaning. (See paragraph 5 . )

14

( i i ) The need for studying the perlocutionary aspect of l in g u i s t i c action

can be opposed by those who want to r e s t r i c t scholarship to the in v e s t i ­

gation o f more contro l lable data. (See paragraph 6 . )

( i i i ) F e l i c i t y conditions can be discarded i f one adheres to the pr inc ip le

that such conditions are only acceptable i f they describe a l l occurrences

o f speech acts o f a certain type (no matter whether they are serious and

l i t e r a l or not, and no matter what sequence they occur in ) . (See para­

graphs 7 and 8 . )

( iv ) Sear le 's in s igh tfu l c la s s i f i c a t io n of i l lo cu t io n ary acts can be re­

futed i f one be l ieves (as do the s p l i t t e r s ) in the absolute d i s c r e t e ­

ness o f c la s se s in a taxonomy, or i f one takes grammatical forms

rather than meaning categories as one's s tar t in g -p o in t (as do the

lumpers). (See paragraph 9 . )

(v) The lumpers' strong preoccupation with grammatical forms also leads

to a rebuttal of the contention that e x p l i c i t performatives are d irect

rather than ind irec t or non-direct representatives of the speech act

type they belong to . (See paragraph 10.)

I f they a t t i tudes l i s t e d would ge t too many advocates, speech act theory

would collapse l ik e a house o f cards, though not all of them would play

an equally important role in the process: leaving perlocutionary e f f e c t s

untouched only makes the theory incomplete; but abstaining from a bas ic ­

a l ly semantic or pragmatic approach to the c la s s i f i c a t i o n o f speech acts

and the problem o f indirectness makes the theory l e s s relevant; and the

theory i s not possib le a t a l lwithout the notion o f i l lo cu t ion a ry force

and the formulation of f e l i c i t y conditions.

15

In the fol lowing paragraphs I shall inquire into the origins o f these

threats to speech act theory, with d irect reference to some r e f l e c t io n s on

the study of soc ia l action in general (o f which l i n g u i s t i c action i s one

i n s t a n t ia t io n ) .

C. REFLECTIONS ON THE STUDY OF SOCIAL ACTION

12. Most — i f not a l l — soc ia l practices show both universal and par­

t i c u la r aspects . Therefore there i s a need for generalizat ion and for

dealing with the here-and-now in studies o f social behavior. General­

iza t ions can be made soon i f a researcher acts as an observer operating

with a preconceived framework o f abstract and general concepts; the here-

and-now can be grasped i f an attempt i s made to capture the culture-

s p e c i f i c understanding of an action that the partic ipants themselves

have, in other words, i f one t r i e s to penetrate the conceptual space

associated with an action by the partic ipants themselves. Ideal ly these

two methodological components should be blended into one coherent approach.

In pract ice , they always are blended; thus the framework o f abstract and

general concepts with which the object of invest iga t ion can be approached

i s , consciously o f unconsciously, based on some reseracher's previous ob­

servations of concrete and part icu lar phenomena; conversely, i t i s hard

for a scholar to g ive up his habits of general izat ion and probably he

wil l never succeed completely in abstracting from the abstrcat and general

concepts he has been confronted with. But rarely are the two components in

balance. Hence the methodological spectrum of the soc ia l sciences i s a

continuum ranging from studies with a high degree o f general i ty to i n v e s t i ­

16

gations with a high degree of part icu lar ity , but with absolutely nothing

at the two extremes of the continuum and hardly anything in the middle.

I t i s usually agreed upon that the ultimate aim of scholarship i s to

be able to make generalizations. Therefore those studies tending more

towards the use of a preconceived framework of abstract and general con­

cepts somehow seem to go more s traight to the point; hence i t i s not

surprising that th is approach seems to be favored in the social sciences.

Of these invest igations I shall say that they use the t h e o r e t i c a l approach.

I f the balance leans more towards the cu lture-sp ec i f ic understanding of

some kind of behavior, I shall say that an e m p ir ic a l - c o n c e p tu a l method is

used. I rea l ize that my usage of the term 'empirical' deviates from common

practice; many so-called empirical invest igations (e .g . most s t a t i s t i c ,

data-oriented studies) apply a purely theoretical method in my terminology.

That i s why I prefer to introduce the more complex term 'empirical-conceptual'.

13. Earlier (in paragraph 1) I claimed that nearly all speech act scholars,

whether l in g u is t s or philosophers, approach l in g u i s t i c action from an

es se n t ia l ly theoretical point of view. In other words, they are rarely in­

terested in the speech acts performed in a particular language - - not to

mention the way in which speakers of a particular natural language conceptual­

ize and understand the ir own acts — but mainly in speech act universals ,

i . e . t r a i t s of l in g u i s t i c action which are expected to be traceable in all

languages. Even when universals are not d irec t ly envisaged the approach i s

usually th eo re t ica l , i . e . based on a preconceived framework of abstract and

general concepts.

17

In order to prevent misunderstandings I want to repeat my point about the

im poss ib i l i ty of absolute general i ty of u n iv er sa l i ty , with d ir ec t reference

to speech act theory. The point i s e sp e c ia l ly c lear with respect to e x i s t ­

ing c la s s i f i c a t io n s of speech act types or, in the professional jargon,

of i l locu t ionary acts or i l locu t ionary forces . Though both Austin (1962)

and Searle (1976), to c i t e the two best known examples, are e x p l i c i t l y

in terested in c la s s i fy in g the types of speech acts that can be performed in

natural language as such — not in one part icu lar language - - they cannot

avoid taking as the ir points of reference the English verbs used to per­

form those acts or to describe them. Thus, though an attempt i s made to

detach the universal claims from concrete data, they are at l e a s t partly

based on them.

14. The uniformly theoretical and generaliz ing approach i s the underlying

cause of the problems of speech act theory. A natural outcome of such a

lopsided approach i s the r e le n t le s s growth o f mutually independent and

often contradictory ' t h e o r ie s ' . There i s nothing wrong with the ex is tence

o f those theories as such: most o f them can be su ccess fu l ly defended from

some theoret ica l point o f view. But because of the ir mutual independence,

which resu l t s from differences in the theore t ica l points o f departure

( i . e . d if ferences in the preconceived frameworks of abstract and general

concepts) most i ssues remain unresolved.

I bel ieve that the i s sues can be s e t t l e d only by introducing a counter­

balancing empirical-conceptual approach, i . e . by examining the way in which

l in g u i s t i c action i s conceptualized by speakers of d i f feren t natural

languages. For instance , what i s th e ir notion of indirectness in language,

18

and how do they c la s s i f y (in a ' fo lk taxonomy' sense) speech acts ? The

s im i la r i t i e s and differences with respect to such notions discovered

between natural languages can const i tu te the basis for a reappraisal of

the universals proposed by t h e o r i s t s . Inc identa l ly , the p a r t icu la r i ty -

oriented approach which I am about to propose should be thought of as

a p o s s i b l e e m p ir ic a l b a s i s f o r th e s tu d y o f u n i v e r s a l s o f l i n g u i s t i c

a c t io n .

The crucial question i s now: Why should we not s e t t l e the unresolved

i ssues by theorizing even more — and more care fu l ly ? Why do we have to

cataract down into the pre-theoretical world of fa c t s ? Does t h i s not mean

an in to lerab le regression of the s c i e n t i f i c enterprise ? In the next

paragraphs the need for supplementing the theoret ica l method w i l l be explained.

15. There i s no denying that some universals o f behavior must e x i s t ;

therefore the theoret ica l approach i s i n t r i n s i c a l l y j u s t i f i e d and necessary.

It seems that the vast majority o f soc ia l s tudies have been conducted from

th i s point of view. There are, however, some problems with i t as pointed

out by Winch (1958) and re-emphasized by McDaniel (1978).

One obvious r e s tr ic t io n i s the fa c t that abstract and general concepts

can never capture the fu l l meaning of soc ia l phenomena which are themselves

concrete , particular and constantly subject to change. Deductive th e o r is t s

would not rea l ly regard th is as a problem. I t i s not so obvious why anyone

would want to capture the p art icu lar i ty o f soc ia l phenomena. The dispute i s

not necessar i ly an epistemological one. The main issue i s : What kind of

knowledge i s most appropriate ? The answer depends very much on what type

of behavior one i s trying to understand. The empirical-conceptual approach

19

works on the assumption that what people say and think should be taken

ser iously , which i s not always the case, as with some types of schizophrenic

behavior; in such cases the theoretical approach i s particularly appropriate.

Moreover, a theory-based method i s needed to unmask motives of behavior

which may be hidden to the agents themselves. But even so, an epistemo-

logical aspect comes in. Though ultimately the search for universals may

be our task, universals are intolerable i f they are based on poorly under­

stood particular facts — or i f they do not have any correspondence with

p a r t icu la r i t ie s at a l l .

16. A stronger plea for supplementing theoretical approaches with the

empirical-conceptual method i s rooted in the meaningful ness of social

action. Social action is by def in i t ion meaninhful in the sense that the

participants in the action perceive i t in terms of a background of shared

meanings. A quote from Schuetz (1967:468) may c la r i fy th is idea of shared

meanings.

I posit meaningful acts in the expectation that Others will interpret them meaningfully, and my schema of positing is oriented with respect to the Other's schema of interpretation. On the other hand, I can examine everything which, as a product of Others, presents itself to me for meaningful interpretation as to the meaning which the Other who has produced it may have connected with it. Thus, in these reci­procal acts of positing meaning, and of interpretation of meaning, my social world of mundane intersubjectivity is built; it is also the social world of Others, and all other social and cultural phenomena are founded upon it.

In that way there i s a fundamental unity between the actor's concepts and

the practices they serve to conceptualize. (Notice that th is formulation

implies the non-unity of the e n t i t i e s of which I claim that they are one;

20

I have not found a way to avoid th i s problem in English; therefore I

would l ik e the reader to condone further manifestations o f the same l i n ­

g u i s t i c f la b b in es s . ) Action i s always interpreted action. I t was Gotthold

Ephraim Lessing who remarked that i t i s unjust to give any action a

d i f fe re n t name from that which i t used to bear in i t s own time and

amongst i t s own people, even though i t may be true that the actions

are always the same in themselves however d i f f e r e n t the times and the2

s o c i e t i e s in which they occur. This i s because a l l actions have an 'under­

stood context ' . I f there i s such a strong unity between actions and con­

cep ts , then a fu l l understanding of soc ia l phenomena can only be reached

i f studies based on a preconceived framework of abstract and general

concepts are supplemented with an attempt to penetrate the conceptual

world of the participants in the action to be described.

17. At th i s point i t i s cer ta in ly useful to i l l u s t r a t e the unity between

concepts and pract ices . My examples, some of which are based on i l l u s t r a t i o n s

given by Winch and McDaniel, show at the same time the f a i lu r e of the theo­

r e t i c a l approach to capture the par t icu lar i ty of socia l phenomena.

Consider, for example, the f ixed rules o f in teract ion between members of

d i f feren t castes in India. Employing his framework o f comparative concepts,

a theory-oriented researcher would characterize a 'caste ' as a 'c lass without

2 Actually, Winch (1958) starts his book with the following quote from Lessing (1778:423): "Denn wenn es schon wahr ist, dass moralische Handlungen, sie mogen zu noch so verschiednen Zeiten, bei noch so ver- schiednen Volkern vorkommen, in sich betrachtet immer die namlichen bleiben, so haben doch darum die namlichen Handlungen nicht immer die namlichen Benennungen, und es ist ungerecht, irgendeiner eine andere Benennung zu geben als die, welche sie zu ihren Zeiten und bei ihrem Volk zu haben pflegte."

21

mobil ity' . There i s nothing wrong with this description as such. But in that

way the real cultural meaning which the caste system has for people in India

would get lo s t : i t i s a hierarchical ranking of men and groups in terms of

religious cr i ter ia of purity of which the hierarchy, the s t r i c t separation

and the division of labor are overt manifestations. (McDaniel quotes

Dumont (1970) as an excellent attempt to penetrate this meaning of the

caste system.)

To take an easier example, consider the practice of bathing. Bathing i s

not just a matter of cleaning oneself with water. In the United States i t

i s associated with an obsession with hygiene and sex appeal. In other

cultures i t may be a means of reaching the purification of the soul.

Or take the conceptual apparatus that Freud developed to explain neuroses.

I f a psychiatrist wants to reveal the origin of neuroses among, say, the

Trobrianders i t would be unwise for him to apply Freud's ideas without

further re f lec t ion . First he would have to grasp many of their own culture-

sp ec i f ic concepts, such as their concept of fatherhood.

Similarly, i t i s wrong for an anthropologist operating with Western

standards of rationali ty and observing the frequent recourse to magic by

a primitive culture, to conclude that these people's behavior i s basical ly

irrational . The fact that the Azande, as pointed out by Winch (1964), blame

all instances of bad fortune on witchcraft does not mean that they do not

perceive rea l i ty clearly: they see as well as we do that the woman was not

ki l led by a witch but by the roof of her hut which collapsed because of a

tree which f e l l on top of i t because of the strong wind. Or again, the

constant use of magic by the Melanesian tribes studied by Malinowski (1954)

does not prevent their minds from working with s c i e n t i f i c accuracy: they

22

know that rich crops depend mainly on precise and hard work and that safe

sa i l ing i s in the f i r s t place a function of the application of certain

principles of s t a b i l i t y and hydrodynamics. So, why magic ? Here i s Mali­

nowski's reply (1954:29):

Thus there is a clear-cut division: there is first the well-known set of conditions, the natural course of growth, as well as the ordinary pests and dangers to be warded off by fencing and weeding. On the other hand there is the domain of the unaccountable and adverse in­fluences, as well as the great unearned increment of fortunate coin­cidence. The first conditions are coped with by knowledge and work, the second by magic.

If s c i e n t i f i c experiments do not work the experimenter's conclusion is

usually that there was something wrong with the setup. Likewise with

magic. The fact that i t often does not work does not destroy the b e l ie f

in i t : i f a ritual does not have the expected e f f e c t , i t s performance

must have been improper somehow. Witherspoon (1977) reports that every time

when he witnessed the Navajos' rain dance i t started raining within

twenty-four hours but only once was the precipitat ion enough to be of any

help. The Navajos' reaction was to smile and say "How feeble-minded we've

become".

I t should be c lear from such examples that detaching social behavior from

the participants' own interpretation of i t , though useful for certain

purposes, i s in general an unwarranted reduction.

18. I t may be interest ing to point out, in passing, that the idea of the

unity between concept and practice i s in keeping with one of the fundamental

principles of phenomenology. Instead of departing from one s ing le basic

princ ip le , such as Descartes' cogito , phenomenology takes as i t s starting

23

point an entire f i e ld of original experiences, basic acts of consciousness,

such acts are always intentional: consciousness i s always consciousness

of something.

If,then, an act of a certain structure is present, then by that very fact a certain object is also present; moreover, the character of this object is Co-determined by the character of the act in which the object appears. The character, therefore, of the known object depends on the character of the act by which it is grasped. (Kockelmans 1967:32-33.)

Applied to the topic of this paper this can be translated as follows: i f

social actors conceptualize their own practices , as they certainly do, the

character of those practices must be partly determined by their own way of 3

perceiving them.

Another f i e ld of scholarship which brings out how in tr ica te ly concepts

and actions are related i s ethnomethodology. Probably i t su ff ices to refer

to some student experiments reported by Garfinkel (1972). Garfinkel asked

his students to go home and try to give up some of the ir basic assumptions,

for instance by imagining that they were paying boarders instead of part

of the family. In that way even the most common everyday actions can be shown

to get a radically different meaning — i . e . to be radically different - -

i f the background assumptions, the ideas of the participants , are changed.

3 The reader should resist the temptation to conclude from this digression that what I am advocating is a phenomenological approach to language. In spite of some similarity with respect to basic principles any phenomeno­logy of language will look substantially different from what I have in mind. Indeed, Husserl seems to have defended the idea of a general and a priori grammar, which would be an extremely theoretical approach (see Jakobson 1970:13).

24

19. Once we have accepted as a necess ity the attempt to understand the

social agent’s own views of what he i s doing, we are confronted with a

technical question: How can we achieve such understanding ?

An apparent impediment i s the im poss ib i l i ty for a scholar of abstracting

from the general conceptual framework he has acquired. Following Gadamer's

(1960) hermeneutical l ine o f thought we can say that such a ' l imitation'

i s no impediment at a l l , but the very prerequisite for understanding.

Understanding results from a dialogue between one’s b e l ie f s and the ob­

servation of new phenomena.

There are, however, a couple of real problems. For one thing, as far as I

can t e l l , there i s no d irec t way of studying a person's conceptualization

of his own actions. But i t can be studied indirect ly i f one accepts that

there i s some kind of relationship between words and concepts. The e x i s t ­

ence of such a relationship i s taken for granted by many scholars. Winch

(1958:123) for instance, a f ter discussing the link between the actor's

concepts and his actions, changes his topic to the link between words and

social behavior:

To give an account of the meaning of a word is to describe how it is used; and to describe how it is used is to describe the social inter­course into which it enters.

Winch i l l u s t r a t e s th is contention by quoting a book on social services

in which i t i s claimed that i t i s the duty of a social worker to establ ish

a relationship of friendship with her c l i en ts but that she must never forget

that her f i r s t duty i s to the policy of the agency by which she i s employed.

The notion of friendship used to exclude th is sort of divided loyalty .

According to Winch, th is change in the notion of friendship r e f l e c t s a

25

change in soc ia l re lat ionsh ips . Indeed, a soc ie ty in which one no longer

makes a d i s t in c t io n between a friend and an acquaintance i s probably a

so c ie ty which, because of i t s extreme m obil ity , i s in need of instant

friendship: make friends f a s t , otherwise the people you meet w i l l be gone

before you get around to knowing them.

I f , then, there i s a relat ionship between words an concepts, i t should

be possible to learn something about the conceptual space associated with

certain practices by examining the words {and expressions) that the p a r t i ­

cipants in the action have at the ir disposal to ta lk about those practices .

The hermeneutic c i r c l e i s playing us tr icks : i t i s not rea l ly possible to

understand the one without understanding the other, and v ice versa. But

i f we do not take th is c i r c l e as an e v i l s p i r i t haunting our in te l l e c tu a l

enterprise but as one o f i t s inev itab le structural features , then the

word-oriented procedure proposed provides us with an ind irect way of

approaching the practices we want to comprehend in an empirical-conceptual

way.

20. I f the empirical-conceptual method i s l imited in the sense that the

soc ia l agent's conceptual world can only be approached in d ir e c t ly , we are

now confronted with a second problem. My claim that an agent's conceptual­

iza t ion o f his own behavior can be in d ir ec t ly invest igated by scrut in iz ing

the words and expressions he has at his disposal to ta lk about the actions

in question was predicated on the assumption that there i s "some kind of

relat ionship between words and concepts". But what kind of a relationship

i s i t ? jo come back to Winch's example, i t i s c lear that a person's be­

longing to a soc ie ty with a changed notion of friendship does not mean

26

that he i s not capable of o ld - s t y le friendship or of conceiving i t . We

would not want to make the Whorfian mistake — which Alford (1978) has

convincingly shown not to have been Whorf's error but the brainchild of

his c r i t i c s — of pos it ing a one-to-one correspondence between words and

concepts and claiming that language completely determines thought.

In order to c la r i fy the relat ionship between language and thought, one

could compare language with Kuhn's notion of a s c i e n t i f i c paradigm.^

Though a language, in particular i t s lex icon , by no means determines the

capacity of i t s speakers to draw d i s t in c t io n s , i t does r e f l e c t the d i s t i n c t ­

ions they h a b i tu a l ly draw. The r e f l e c t io n ^ moreover, i s not perfect: i t

i s l ike the re f lec t io n on the surface o f a pond rather than in a mirror.

Though there wil l be some d i s to r t io n s , the 'Gestalt ' i s recognizable.

Reformulating the end o f the previous paragraph in those terms we can

say: Since the lexicon o f a language (conceived here as a c o l l e c t io n o f

words and expressions) r e f l e c t s the conceptual d i s t in c t io n s that i t s

speakers habitually draw, i t becomes possib le to penetrate the conceptual

space that the participants in an action assoc ia te with i t by scrut in iz ing

the words and expressions that th e ir language provides for talking about

the action in question. When doing th is type of research, however, one

should never lose s igh t o f the l im ita t ion s we sp e c i f ie d .

D. ADVOCATING UNORTHODOX SCHOLARSHIP

21. The aspiration of most scholars i s to embody the higher stages o f

truth, namely syntheses. The formulation o f strong theses and antitheses

i s generally shunned which, in p rac t ice , often resu l ts in the disruption

4 The comparison was first made by Van Valin (1976).

27

of the normal development of thought and hence in the very rare emergence

of real syntheses. The proposal I am about to make clashes with th is common

cautious practice. Therefore most readers will regard i t as an instance

of unorthodox scholarship though I propose i t for the sake of scholarship

i t s e l f , that i s , with the intention of pushing the l imits of our knowledge

jus t a l i t t l e b it further.

22. Verbalizations of many aspects of social l i f e have been investigated

by cultural, especial ly l in g u is t i c , anthropologists. Linguistic action,

however, has mainly been theorized about. It i s true that verbs used in

the performance and description of speech acts have been studied, but

such studies were usually undertaken in view of the advancement of the

theory (which is also one of my ultimate aims) but also with strong theo­

retical prejudices; remember the attempts to c la s s i fy speech act types.

To counterbalance the exist ing theoretical approaches a systematic study

of the words and expressions used to describe l in g u is t ic action i s needed

in order to l i f t the veil that i s covering the language user’s culture-

sp ec i f ic and language-specific experience or conceptualization of his own

speech ac t iv i ty . In other words, I am advocating one type of empirical-

conceptual approach to l in g u is t ic action, namely a le x ic a l approach. By

undertaking this type of research, a s ign if icant though not to ta l ly new

contribution to the study of l in g u is t ic action will be made. My hope is

that thrusting the methodological balance to the opposite extreme may

provide us with revealing data on the basis of which the many unresolved

theoretical issues can be tackled.

28

B rie f ly , whereas th e o r is ts o f l in g u i s t i c action have been trying to ex­

plain how p e o p le do th in g s w ith w ords , I propose to undertake a supple­

mentary study o f w hat p e o p le sag th e y do w ith w o rd s . Notice, however, that

we are le s s interested in what people a c tu a l l y say than in what they can

say about th e ir verbal behavior; trying to reveal the concepts they have

at th e ir disposal i s no doubt the more basic question of the two in view

of the fa c t that people do not always say what they mean.

Let me formulate the enterprise in one more d i f fe re n t way. Trying to

leave speech act theory behind us — an attempt in which we can at best be

partly successful - - we can take a guided tour through l in g u i s t i c action

with language i t s e l f as our guide.

23. Some additional remarks are ca l led for . F ir s t , l in g u i s t i c action

provides us with an area in which the unity between action and concept i s

part icu lar ly c le a r , so c lear in fact that i t i s almost tempting to postu­

la te a complete overlap between words, concepts and act ions. The area I

am thinking of i s that part icu lar subf ie ld of the s e t of l ex ica l items

used to describe l in g u i s t i c action which contains the truly performative

verbs, i . e . the verbs which, i f used in the f i r s t person singular simple

present ind icat ive a c t iv e , are devices for the performance of the very same

acts as the ones they describe when used d i f f e r e n t ly . Since, in most con­

t e x t s , saying ser ious ly and l i t e r a l l y "I promise that P" const i tu te s an

act o f promising, the verb TO PROMISE i s more than j u s t a r e f le c t io n of

a conceptual category that happens to be associated with some type of

verbal behavior. Somehow the word, the concept and the action seem to be

co-extensive; no doubt most speech act th e o r is ts would hotly deny th is

29

because i t shows the weakness of the ir universal claims based on concrete

mono-linguistic data. This strong link can probably be generalized to

non-performative speech act verbs ( i . e . a l l other verbs describing speech

act types) s ince , as I have shown elsewhere (see the f i r s t chapter of

Verschueren 1977), the two c lasses are not s t r i c t l y separable; rather,

there i s a performativity continuum running through the complete se t of

speech act verbs. The further a verb i s removed from the truly perform­

ative type, the weaker the word-concept-action overlap may be expected

to become. Without pressing the point too strongly, I'd l ike to suggest

that th is i s an area in which the l imitations of the lexical approach are

themselves limited.

Second, the invest igation I propose wil l lead us not only into the domain

of l in g u i s t i c a c t iv i ty covered by speech act theory. Obviously, words and

expressions describing l in g u i s t i c action can focus on any of a large number

of i t s aspects, not only on the speech act type.

Third, the relevance of the invest igation can be expected to increase

dramatically as soon as a comparative perspective i s brought i s . The enter­

prise i s e s se n t ia l ly open-ended because of the i n f i n i t e p o s s i b i l i t i e s of

h is to r ic a l , cross-cultural and c r o s s - l in g u is t ic comparison, but also because

of the p o s s ib i l i t y of s ig n i f ica n t d i f ferent ia t ion within one culture and one

language at one particular point in time. What I am advocating i s a

com para tive l e x i c a l approach to l in g u is t i c action: the comparative study of

the se t of words (not only verbs but also nouns and maybe even adjectives

and adverbs), idioms and other fixed formulae and expressions used to

describe l in g u is t i c action.

30

24. I do not want to create the impression that s im ilar invest igat ions have

never been advocated or attempted before. After a l l , Austin i s said to have

defended a strongly verb-oriented approach himself though he never got

around to applying i t ex tens ive ly . Here are some more examples to the

contrary.

After arguing that speech act descriptions have no place in l in g u i s t i c s

— a pos it ion which I would cer ta in ly not want to support — L.J. Cohen

(1974), for instance, contends that l in g u is t s could shed l ig h t on the

problem o f speech acts by discussing the terms used in describing speech

acts and by surveying the empirically detectable v a r ie t ie s of ways in

which these terms function and the empirically detectable v a r ie t ie s of

conditions for applying them. He even l i s t s a number of relevant semantic

dimensions o f the words and phrases used in describing what people do with

words. Unfortunately, his proposal has gone large ly unnoticed so that

further r e f l e c t io n s on i t s relevance are hard to f ind .

Nevertheless , some attempts have been made, for instance by D. Lehmann

(1976) and the Berliner Gruppe which prepared a lengthy — s t i l l un­

published — manuscript on German speech act verbs. Other scholars such

as J. Allwood (see 1976 and 1978) have a d e f in i t e in te r e s t in th is kind

of research but do not want to embark on i t unti l they have constructed

a sa t i s fa c to ry theory o f l in g u i s t i c action which can be taken as the ir

point of departure. As far as I can t e l l none o f these researchers derive

the ir i n t e r e s t from a methodological commitment s im ilar to mine.

One scholar whose a t t i tu de i s much c loser to mine i s R. Meyer-Hermann.

His recent work (e .g . 1978) shows an in t e r e s t in what people say about

l in g u i s t i c a c t iv i t y for the sake of a better understanding o f what l i n -

31

g u is t i c action i s . However, he concentrates on what they a c tu a lly say

rather than on the lex ical expression of the concepts they have at their

disposal to talk about verbal behavior. Though I bel ieve that the la t ter

i s the more basic of the two enterprises , they are both necessary and,

in practice , mutually dependent.

Inquiries in which an inference i s drawn from the mere presence of

certain lex ical items to the way in which verbal behavior i s conceptual­

ized or experienced in a particular language community are rare. A

marginal example i s Kernan, Sodergren and French's (1977) br ie f account

of Belizian City Creole in which i t i s claimed that "the proliferation

of metal inguistic terms for types of speech acts i s ind icative of the

cultural emphasis on speech and cultural appreciation of talented and

elegant speech behavior" (p .36).

More to the point i s M. Reddy's (1977) insightful paper called "The

conduit metaphor — A case of frame c o n f l ic t in our language about

language". Here i s how he announces the theme of his a r t ic le :

In this paper, I am going to present evidence that the stories English speakers tell about communication are largely determined by semantic structures of the language itself. This evidence suggests that English has a preferred framework for conceptualizing about communication, and can bias thought process towards this framework, even though nothing more than common sense is necessary to devise a different, more accurate framework, (p.3)

Reddy claims that the basic view of l in g u is t i c communication that emerges

from the English words and expressions used to describe i t could be called

the 'conduit metaphor': language i s a channel through which thoughts are

conveyed; the speaker puts his thoughts in at one end and at the other

the hearer catches them; the thoughts are rea l ly in the words. He gives

32

dozens of i l lu s tr a t io n s such as "Try to GET your THOUGHTS ACROSS better",

"None of Mary's FEELINGS CAME THROUGH TO ME with any c lar ity" , "You s t i l l

haven't GIVEN ME any IDEA of what you mean", "Try to PACK more THOUGHT

INTO fewer words" and "That THOUGHT IS IN pract ica l ly every other WORD".

Reddy claims that th is metaphor d is torts rea l i ty : a hearer never simply

receives thoughts; instead, understanding results from an active process

of interpretation, i . e . from hard work. In terms of the conduit metaphor

fa i lure to communicate seems impossible as long as the speaker uses the

'right' words; success appears to be automatic. Reddy also elaborates on

the dangerous impact such a distorted view can have. I bel ieve that he

overstates his point a l i t t l e b i t in that he loses s ight of the partial

truth on which the conduit metaphor i s based. I f i t were not true that

somehow language i s responsible for thought processes ( in other words,

that there are ideas in the words) then there would be no basis for him to

contend that the metaphor can be perilous. But th is does not undermine the

importance of Reddy's claims.

The research project that I am advocating i s more ambitious than Reddy's

because of i t s wider scope, but at the same time i t i s l e s s ambitious in

that I am not immediately looking for comparable sweeping statements.

25. The se t of words and phrases to describe l in g u i s t i c action i s such a

vast expanse that any attempt to give a complete description even for a

s ing le language within the scope of a s ingle essay would be doomed to

fa i lu re . Therefore the topic of th is text w i l l be narrowed to a small

number of r e la t iv e ly small subfields each of which wil l be representative

of a particular level of l in g u i s t i c action. A f i r s t chapter wil l deal with

33

the most marginal aspect of verbal behavior imaginable, namely the se ­

mantics of s i len ce , the absence of speech. The semantics of lying will

be next, because of i t s relation to propositional content. Illocutionary

force will be represented in a chapter of directing. Then there i s a

chapter on conversational routines, concentrating on those for which at

l ea s t some languages do not provide a descriptive term.

Because of the suspected relevance of a com parative lexical approach

these subfields will be studied contrastively in English and in Dutch. A

comparison between two less related languages might have been more re­

vealing, but I believe that i f interesting conclusions can be drawn from

a comparison between two languages which are so c losely related, the same

must a fort ior i be true for all other confrontations.

Before starting the actual work, some methodological and terminological

considerations are needed. They wil l be provided in the next chapter. Be­

fore moving on I want to repeat my point that this study, though evidently

particularity-oriented, should be regarded as an exercise preparing a

potential ly in f in i t e series of wide-ranging comparative investigations

which could be the e m p ir ica l b a s is for ref lect ions on u n iv e r s a ls o f l in ­

g u i s t i c a c t io n .

34

II . RESEARCH PRINCIPLES AND TERMINOLOGY

A. INTRODUCTION

1. In the previous chapter I issued a manifesto professing my b e l i e f

that the e x is t in g theories of l i n g u i s t i c action su f fer from some kind of

anemia due to the lack of an acceptable decis ion procedure to s e t t l e un­

resolved i s s u e s , and that invest igat ions of what people say they do with

words could furnish us with such a procedure and hence increase the

v i t a l i t y and v i a b i l i t y o f the theory. I a lso announced that th i s essay

would be res tr ic te d to the comparative study o f some r e la t iv e ly small

subfie lds o f the English and Dutch words and phrases used to describe

l in g u i s t i c action. In t h i s chapter I shall f i r s t present a further sp ec i ­

f i c a t io n of my object of invest igat ion and introduce some terminological

conventions. Then some pr incip les o f lex ica l semantics w i l l be proposed.

Further, a methodology w i l l be propounded to elude the impending danger

of c ir c u la r i ty in the semantic analysis of l ex ica l items. Finally I shall

bring into focus the problem of how to represent the findings of th i s type

of research; in th is context the notion of a semantic dimension {as opposed

to a semantic c la s s ) w i l l be brought in .

35

B. THE OBJECT OF INVESTIGATION

2. What I advocated in the f i r s t chapter was a comparative study of the

se t of words, idioms and other f ixed formulae or expressions used to

describe l in g u i s t i c action . Groups o f words such as TO BEAT AROUND THE

BUSH ( i . e . ' talking without coming to the p o in t1) or the Dutch IEMAND

IETS IN DE SCHOENEN SCHUIVEN ( l i t e r a l l y ' to shove something into somebody's

sh oes ' , meaning 'to accuse somebody unjustly o f something'), the meaning

of which cannot e a s i ly be derived from the sense of the component elements

we ca l l id io m s . The o th e r f i x e d fo rm u la e are expressions with a more

transparent meaning such as TO PRONOUNCE X+Y HUSBAND AND WIFE ( i . e . to

marry X+Y). In that way idioms and other f ixed formulae are by no means

two completely separate c la s se s o f l in g u i s t i c e n t i t i e s . Together they

c onst i tu te what I shall henceforth refer to as com plex l e x i c a l i z a t i o n s .

They are contrasted with w ords ( i . e . monomorphemic words, derived words

and many compounds) which I ca l l s im p le l e x i c a l i z a t i o n s . Both simple and

complex l e x i c a l i z a t io n s are l e x i c a l i te m s .

The coining of new terms may seem to be superfluous since we have the

p er fec t ly c lear English term c o l lo c a t io n at our disposal to refer to

most cases of what I call complex l e x i c a l i z a t i o n s . Indeed, but groups of

words such as TO BE SILENT (which i s the most commonly used lex ica l item

to describe the absence o f speech in English; i t s status as a lex ica l

item wil l be c la r i f i e d la ter ) can hardly be ca l led co l lo ca t io n s ; y e t they

are no words e i ther; in my c l a s s i f i c a t i o n they are complex l e x i c a l i z a t i o n s .

36

3. Again, there i s no c lear-cu t boundary between simple and complex l e x i ­

c a l i z a t io n s . Most compounds are borderline cases . TO BAD-MOUTH i s often

characterized as an idiom; y e t I am inc l ined to include i t in the c la s s

of simple l e x i c a l i z a t i o n s . On the other hand, a DEAR JOHN LETTER i s prone

to be described as a word because i t forms such a strong unity that even

men can be said to write a Dear John l e t t e r to a woman, and because i t s

formation seems to be gett ing productive in the sense that i t i s no longer

unusual to hear a woman say, rather s a r c a s t i c a l l y , to a man "Thank you very

much for that nice Dear John phone ca l l o f yours"; y e t I am inclined to

regard a DEAR JOHN LETTER as a complex l e x i c a l i z a t io n . To j u s t i f y these

in c l in a t ion s I propose a rather arbitrary rule: l e x ic a l i z a t io n s are simple

i f they form an orthographical unity, i f not , then they are complex. Note,

however, that th i s rule wil l not be applied in the case o f preposit ional

verbs such as TO TALK ABOUT (SOMETHING) and phrasal verbs such as TO ORDER

(SOMEONE) ABOUT which are simple l e x i c a l i z a t io n s .

Apart from the fa c t that what I want to ta lk about cannot be dealt with

in terms of the contrast between words and c o l lo c a t io n s , the new termino­

logy has the advantage of that a lex ica l item i s not j u s t a l in g u i s t i c

e n t i t y but a l e x i c a l i z e d ( i . e . l i n g u i s t i c a l l y processed) form of something

e l s e , namely of part o f the conceptual space constructed by a particular

language community.

4. However vague the d i s t in c t io n may be between c las ses o f l ex ica l items,

the boundary between lex ica l items and groups o f words which do not form

a lex ica l unity i s even fu z z ie r . The decis ion to treat a word group as a

l ex ica l item (a complex l e x i c a l i z a t io n ) w i l l always be based on a judgment

of how strong a unity the group forms. But 'strength of unity' can be

37

viewed from two angles: the encoder's and the decoder's. The cr i ter ion

I generally use i s the following: i f someone encoding his thoughts would

be able to produce the group of words in question without having learnt

i t as a unity, then we have to do with a simple word group; i f previous

learning can be expected to be required, then the combination of words

i s a l ex ica l item even i f someone decoding the message would have no

trouble at a l l to interpret i t without having come across i t before.

There are only a couple o f cases in which i t could be claimed that I

deviate from th is cr i ter ion . For instance, I often tre a t the combination

of an adject ive with an auxi l iary ( e .g . TO BE SILENT), or the combination

of a verb with the negative adverb 'not' { e .g . NOT TALK) as lex ica l

items. This procedure seems to be j u s t i f i a b l e (in view o f our search

for the lex ica l re f lec t io n of conceptualization habits) on the ground that

these combinations are such conventional ways of verbalizing an adject ive

and o f turning a pos i t ive verb into a negative one that they often make

the creation o f a separate verb superfluous to express the s ta te or action

referred to . I f we refrained from treat ing them as l ex ica l items we would

be confronted with a whole l o t of l ex ica l gaps which are not rea l ly gaps

at a l 1.

5. The question o f why we want to study complex as well as simple l e x i c a l ­

iza t ions cannot be avoided. Bolinger (1975) couches the problem in the

following terms:

Since the lexicon purports to record all the pre-set meaning-bearing units of a language, ideally it would have to include every collo­cation as well as every word. In practice this is impossible - and probably in theory too. Practically, there would be no room. Theo­

38

retically, one would not know where to stop, because collocations shade off into more or less freely formed constructions and fluctuate too much from place to place and from individual to individual. Further­more, there is no reliable way to test them, at least at the borders between collocations and constructions, (p.105)

To counter Bol inger's objection I contend that i t would be suspicious

to find only categories with c lear boundaries in the description of a

phenomenon which i s i t s e l f in constant f luc tuat ion . An immediate corollary

of th is stance i s the commitment to undertaking an i n t r in s i c a l l y endless

job. Note, however, that the apparent f in i t e n e s s of d i f feren t approaches

i s rarely more than an i l l u s i o n .

The necess i ty to take complex l e x i c a l i z a t io n s into account i s quite

clear i f we do not j u s t think about our enterprise in terms of the study

of part of the lexicon as such but as a r e f le c t io n of the underlying con­

ceptua l iza t ions . Every l e x i c a l i z a t io n of r e a l i t y or experience, whether

simple or complex, i s a unique r e f le c t io n of some conceptualization

habit. We have shown th is in connection with a DEAR JOHN LETTER. Further

i l lu s t r a t io n s are not hard to f ind.

6. While further demonstrating the unique conceptualization habits re­

f lec ted in the complex (as well as simple) l e x i c a l i z a t i o n s , we can at the

same time point out some d i f fe re n t types o f relat ionships between the

lex ica l forms and the object or event they are applicable to .

F ir s t , there i s the e x e m p lify in g re la t ion . An i l l u s t r a t i o n i s , once

more, our DEAR JOHN LETTER. A l e t t e r in which a wife asks for a divorce

or a g ir l f r ie n d breaks o f f an engagement i s not l ik e ly to s ta r t with

"Darling" or "Honey", but rather a d is tant mode o f address w i l l be chosen,

such as "Dear John" or, more general "Dear X" where X stands for the

39

hearer's Christian name. The term i s exemplifying in two ways: i t repre­

sents a particular mode of address (which i s not the only one possib le)

and i t picks out the name 'John' as an example.

Second, the relat ionship can be a n a ly t i c a l as when the expression

TO RAISE ONE'S BLOOD SUGAR LEVEL i s used instead of the verb TO EAT. The

a c t i v i t y of eating i s analyzed in terms o f what i t does for the body. Thus,

as a matter of course, the expression i s mainly used in c i r c l e s where

there i s a strong (though often simply fashionable) preoccupation or

obsession with health. There i s a comparable formula in Dutch, namely

CALORIEEN SLIKKEN ( l i t . 'to swallow c a l o r i e s ’ ) . Here the eating substance

i s analyzed. Saying "I think i t i s about time for me to go and raise my

blood sugar level" or "Ik denk dat het t i j d wordt nog eens wat calorieen

te gaan slikken" ( l i t . "I think i t i s about time again to go and swallow

some ca lor ies") i s somehow a euphemistic denial of the fa c t that what one

i s about to do i s sa t i s fy in g some primary bodily need or indulging in some

earthly de l igh ts ; i t almost sounds l ik e "I am sorry, I simply have to ,

otherwise I wouldn't do such a thing".

Third, there are m e ta p h o r ic a l re la t io n s . A c lear example of a metaphor

i s TO CASH IN ONE'S CHIPS meaning TO DIE. Even dead metaphors such as

TO KICK THE BUCKET seem to preserve the ir metaphorical character ( i . e . th e ir

relat ionship to l i t e r a l meaning) to some extent . Though "Harry's been

dying for weeks now" ia an acceptable sentence, "*Harry's been kicking

the bucket for weeks now" i s not, because o f the associat ion of abruptness

carried by the verb TO KICK.

This l i s t i s no doubt incomplete. There are a lso complex l e x i c a l i z a t io n s

in which there seems to be no re lat ionship whatsoever between the ir lex ica l

40

forms and the objects or events they describe. As far as I can t e l l there

i s nothing in TO DO A COLD TURKEY which makes i t inherently su itab le to

express someone's withdrawing, without medical help, from an addictive

habit such as drugs, alcohol or c igaret te smoking.

7. After se t t in g up a terminological framework and explaining why we

want to in v es t iga te complex as well as simple l e x i c a l i z a t i o n s , another

question crops up: What c la s se s o f lex ica l items {in the sense of the

tradit ional word c la s se s ) are used to describe l in g u i s t i c action and are

thus relevant to the type of invest iga t ion proposed ?

Not only v e r b s are used to describe l in g u i s t i c ac t ion , but a lso nouns.

Though for many o f the verbs in question (e .g . TO PROMISE) there i s a

corresponding noun ( e .g . PROMISE), th i s i s not always the case. For the

verb TO URGE there i s no corresponding noun URGE which describes the same

l in g u i s t i c act (though that gap would be at l e a s t p a r t ia l ly covered by

the noun EXHORTATION). Conversely, the Dutch noun ORDER has no one-to-one

correspondence to any verb, but i t describes l i n g u i s t i c acts which form

a subclass of the acts described by the verb BEVELEN ( ' t o order ' ) . There­

fore both nouns and verbs are relevant to the study of what people say they

do with words.

A d je c t i v e s provide us with pertinent information about the nature of

certain l in g u i s t i c actions through the ir combination p o s s i b i l i t i e s with

the nouns describing those act ions . Note, for instance , that one can ta lk

about A TRUE STATEMENT but not * A TRUE ORDER (unless TRUE i s used in a

d i f fe re n t sense); A SOUND ARGUMENT but not *A SOUND PROMISE; A.SOLID OFFER

but not *A SOLID QUESTION; A BASIC QUESTION but not *A BASIC OFFER; or in

41

Dutch EEN GAVE BEKENTENIS ('an unconditional confession') or EEN GAVE

TOESTEMMING ('an unconditional permission') but not *EEN GAVE VRAAG ('an

unconditional question'); EEN VAST AANBOD ('a sol id o f fer ' ) or EEN VASTE

BELOFTE (' a firm promise') but not *EEN VAST BEVEL ( l i t . 'a solid order').

Adjectives always reveal a semantic dimension of the nouns they modify:

the acceptabil ity of the phrase THE TALL TREE exposes the fact that there

i s a ta l lness dimension to trees . Therefore, investigating such combination

p o s s ib i l i t i e s could y ie ld some insights into the nature of certain types of

verbal behavior.

Similarly, adverbs may prove useful in disclosing the most important

dimensions of the acts described by the verbs or verb-like expressions they

can be used to qualify.

8. Though verbs, nouns, adjectives and adverbs al l deserve a place in the

examination of what people say they do with words, there i s no denying that

when i t comes to describing action (in this case l in g u is t ic action) verbs

and verb-like expressions occupy a key position. This essay is not more

than a p i lo t study on which severe l imitations have to be imposed. The

discussion in the subsequent chapters will therefore be confined to verbs

and verb-like expressions. Nouns, adjectives and adverbs will only occasion­

a l ly be treated when they are part of a complex verb-l ike lex ica l iza t ion .

Another terminological remark is needed. In order to avoid the s t y l i s t ­

i c a l l y tiresome recurrence of the phrase 'verbs and verb-l ike expressions'

or 'verbs or verb-like expressions' I create the term v e r b ia ls on the ana­

logy of the word 'adverbial' which covers both adverbs and adverb-like

expressions. I f a need arises to distinguish the two, I shall talk about

42

s im p le v e r b ia l s (or simply v e r b s ) and com plex v e r b i a l s on the basis of my

e a r l ie r d i s t in c t io n between simple and complex l e x i c a l i z a t i o n s . (Notice

that the term 'verbal' would not have been su itab le because i t indicates

words combining properties o f verbs with those o f adject ives and nouns,

such as gerunds, i n f i n i t i v e s and p a r t i c i p l e s . )

The verbials used to describe l in g u i s t i c action under one aspect or

another, i . e . the objects for invest igat ion in th i s essay, I c a l l l i n ­

g u i s t i c a c t io n v e r b i a l s .

9. Now that we have confined our task to the Dutch-English comparative

study of the verbials covering the small subfie lds o f l in g u i s t i c action

sp ec i f ied at the end of the f i r s t chapter, we ought to look for a cr iter ion

to decide on when a verbial can be r ig h t fu l ly included in the se t of

lex ica l items used to describe the action in question. As with most of

the d i s t in c t io n s drawn so f a r , th is c r i ter io n i s going to be a hazy one

too.

Some verbs such as TO ASK have a type of verbal behavior as th e ir pr i­

mary meaning (though the meaning can be metaphorically extended as in "He

i s asking for trouble"). Evidently, these have to be accounted for .

Other verbs such as TO THREATEN denote actions which are not necessari ly

l i n g u i s t i c . S t i l l the l i n g u i s t i c part o f the ir meaning is prominent enough

to warrant unqualified inc lus ion .

Yet another c la s s o f verbs such as TO PUT FORWARD have an o r ig in a l ly

n o n - l in gu is t ic sense. But the metaphorical extension of th e ir use to kinds

of l in g u i s t i c action has become so widespread and conventional that we can

43

talk about dead or l e x ic a l i z e d metaphors. Hence they should be dea lt with.

Finally we have verbs such as TO KILL which can only be used to describe

verbal behavior i f th e ir meaning i s very widely and a c t iv e ly extended

metaphorically as in the following sentence from a song: "Killing me

s o f t l y with his words". Since the meaning of nearly a l l words can be ex­

tended in a s im ilar fashion, the invest igat ion would get t o t a l l y out of

hand i f these were incorporated into the s e t o f lex ica l items used in saying

what one does with words.

Fuzziness w i l l be inev itab le when i t comes to making decis ions about

l ex ica l items which are s i tuated on the borderline between the third and

l a s t categories l i s t e d . Notice that the d i s t in c t io n s we made here with re­

ference to verbs are also applicable to complex verb ia ls .

10. A f ina l remark about the object of invest igat ion i s needed. I announced

a comparison between English and Dutch. I t happens that the v a r ie t ie s of

these two languages that I have been confronted with most are American

English and the Flemish variety o f Standard Dutch {spoken in the northern

provinces of Belgium). Consequently the subsequent discussion w i l l contain

some lex ica l items and judgments of usage peculiar to these v a r i e t i e s .

But they are so c lose to Brit ish English and the Standard Dutch spoken in

the Netherlands that there i s no harm in using the br ie f names English and

Dutch, as long as the above remark i s kept in mind whenever necessary.

There i s , however, a minor problem involved in any lex ica l comparison

between American English and Flemish Standard Dutch, which has to do with

the history of the two languages and the s o c i e t i e s in which they are spoken.

Mobility i s maybe the basic structuring force in American so c i e t y , re­

44

suit ing in a kind o f cultural uniformity which, given the vast geogra­

phica l , racial and occupational d i f feren ces , would be hard to imagine

in the l e s s mobile European cultures . Apart from the language used by

previously segregated groups such as the American Blacks, there i s also

a quite impressive l in g u i s t i c uniformity. Though the English spoken in

the deep South sound quite d i f feren t from New York Jewish English — they

are, a f te r a l l , more than a thousand miles apart — they are s t i l l mutual­

ly i n t e l l i g i b l e , whereas people in Limburg (the easternmost province of

Flanders) do not understand the people in West Vlaanderen (the western­

most province of Flanders) i f they speak the ir local d i a l e c t , and vice

versa, though the two places are l e s s than a hundred miles apart (that i s ,

l e s s than the distance between New York and Philadelphia or between Los

Angeles and San Diego). Because o f the l in g u i s t i c uniformity (and support­

ed by a vague but strong principle of l ib er ty ) Americans can speak the same

variety o f English on a l l occasions: at home, at work, addressing a fr iend ,

or addressing the whole nation. Their language i s not j u s t a means of

communication scrupulously adapted to the circumstances, but a means of

expression with which they have become fused. As a r e s u l t , though an

American w i l l not f a i l to respond when asked to make judgments about the

formality or informality of l in g u i s t i c expressions, in actual usage th i s

contrast i s gett ing l e s s and l e s s relevant. Thus, the same Americans who

euphemize about a basic a c t i v i t y such as eating by talking about RAISING

THEIR BLOOD SUGAR LEVEL, are l ik e ly to use a nice c o l le c t io n of f o u r - l e t t e r

words in the same conversation. Or again, the use o f an expression such as

TO PUT THE LID ON (which some d ic t io n a r ie s , even recent ones, characterize

45

as slang) will cause no surprise at al l in the discussion of the gross

national product in a te levised presidential campaign debate.

In Flanders, on the other hand, there was until recently a complete

lack of l in gu is t ic uniformity. Even after Belgium became an independent

country in 1830 the spontaneous growth of a standard language was prevent­

ed by the fact that for all purposes for which a standard language is

needed, such as government, the press, science and higher education,

French was used (or, indeed, French was the only lega l ly accepted lan­

guage). Though a standard Flemish variety of Dutch has been used for all

o f f i c ia l purposes for several decades now, and though everybody under­

stands i t , most people have considerable d i f f i c u l ty speaking i t : their

language at home, and often at work, remains a regional dialect (which is

maintained because of the s t i l l l imited mobility of the soc iety) . The

consequence is l in gu is t ic insecurity and an over-anxiousness to use the

right register and the right degree of formality in the right circum­

stances. Moreover, many speakers find i t absolutely impossible to reach

any degree of informality when using the standard language.

In the subsequent treatment of the English and Dutch expressions used

to describe l in gu is t ic action also informal lexical items and slang ex­

pressions are dealt with. Since the patterns of usage of formal versus

informal lexical items in American English and Flemish Standard Dutch

d if fer considerably, the overall comparison (in which labels such as

'formal1, 'informal' and 'slang' wil l be used) i s in tr in s ic a l ly distorted.

This fa c t , which could be a fascinating topic for further soc io l ingu is t ic

research, should be kept in mind throughout. A comparison between British

46

English and the Standard Dutch spoken in the Netherlands would not present

such a problem.

C. PRINCIPLES OF LEXICAL SEMANTICS

11. How should the meaning of verbia ls be described ? The question can be

broken up in two parts. F i r s t , how can the internal semantic structure

of verbia ls be brought to l ig h t ? Second, how should the semantic re­

lat ionships between the verbials covering a part icu lar conceptual f i e l d

be id e n t i f ie d ?

In the fol lowing paragraphs lex ica l decomposition, which i s one o f the

possib le reponses to the f i r s t question, and the s t r u c t u r a l i s t study of

lex ica l f i e l d s , which was for a long time the favored so lution to the

second problem, w i l l be discussed b r i e f ly . Their l im ita t ions w il l be con­

sidered and an a l ternat ive approach wil l be proposed.

12, Saussure's d i s t in c t io n between longue and p a r o le induced l in g u i s t s

such as Trubetzkoy to postulating abstract 'phonemes' which can be rea l­

ized in d i f feren t ways and which can be characterized in terms of a

universal se t of d i s t i n c t i v e features such as + vo ice , + dental , e tc .

On the analogy o f these s t r u c t u r a l i s t i c phonological notions L. Hjelmslev

and R. Jakobson proposed to analyze the meaning o f words in terms o f a

se t o f universal m eaning com ponents or sem a n tic f e a tu r e s . In that way the

meaning o f MAN could be analyzed as + male, + adult , + human, and the

meaning o f WOMAN as - male, + adult and + human. This semantic procedure

i s ca l led co m p o n en tia l a n a ly s i s .

47

Componential analysis labors under a number of more or l e s s serious

d i f f i c u l t i e s . F ir s t , i t i s not too c lear which pole of a binary contrast

such as male/female should be taken as the descript ive point of depart­

ure. This problem i s not too serious because the proponents of component­

ia l analysis can e a s i ly get away with a decis ion by declaring that the

semantic features used should be regarded as a to m ic c o n c e p ts which can

but need not be l ex ica l ize d in individual languages; hence a semantic

feature such as + adult must not be id e n t i f i e d with the word ADULT. None­

th e le s s componential analysts try to motivate th e ir decis ions by relying

on the notion of m a rk ed n ess: s ince one normally asks "How t a l l i s John"

and not "How short i s John" when making inquiries about John's s i z e , ' t a l i '

i s regarded as the unmarked pole of the t a l l / s h o r t contrast and therefore

i t i s taken as the point of departure i f the contrast i s needed in seman­

t i c analys is .

But how can the meaning of semantic features be grasped i f i t i s so ob­

jec t ionab le to id en t i fy them with the corresponding words in natural lan­

guages ? Especially i f the components are atomic in the sense that they

r e s i s t further decomposition i t i s hard to see how th e ir meaning can be

explicated other than by means o f such an id e n t i f i c a t io n . Obviously th is

second problem cannor be smothered under terminological tr ickery .

A third , equally ser iou s , problem i s the dubi.table ex is tence o f a uni­

versal s e t o f semantic features belonging to some universal mental lan­

guage from which speaking i s always a transla t ion to a part icu lar natural

language. Consider the rather random l i s t of th ir teen fundamental con­

cepts proposed by A. Wierzbicka (1977): I , you, someone, something, world,

t h i s , want, don't want, think o f , say, imagine, be part o f , become. I t i s

48

hard to detect any log ic in th is s e r i e s any any scholar w i l l be able to

think of concepts which are not readily analyzable in terms of Wierzbicka's

Fundamental Thirteen. One concept which most scholars would spontaneously

add to the l i s t i s TO BE. But then we remember Witherspoon's (1977) claim

that in Navajo TO GO appears to be a much more basic concept than TO BE.

Maybe we ought to a t tr ibute the fa i lu re to construct an acceptable se t of

universal semantic primit ives to the fa c t th a t , as Fortescue (1979) points

out, the language of thought, i f conceived in terms of the view underlying

the search for universal meaning components, i s no language at a l l .

13. When attempts were made to apply componential analysis to verbs, i t

was soon realized that the meaning components (which are then ca l led

a to m ic p r e d ic a te s ) should be viewed as const i tu t ing a hierarchic semantic

structure instead of a simple sum. A verb such as TO KILL, for instance ,

was thought to be analyzable as

k i l l ( x , y ) : CAUSE(x, BEC0ME(y, NOT ALIVE))

(where x i s the agent o f TO KILL and y i s the p a t ie n t ) . This further ex­

tension of componential analysis i s ca l led l e x i c a l d e c o m p o s itio n .

One common cr i t ic ism i s th a t , for instance , TO KILL and CAUSE TO DIE

cannot be substituted for each other: one can say "Yesterday John caused

Jim to die" even i f Jim did not die unti l today; t h i s i s impossible with

"Yesterday John k i l l e d Jim". Such c r i t ic i sm may reveal the need for time

indexes in some lex ica l decomposition formulae, but scholars who maintain

that atomic predicates , j u s t l ik e other semantic fea tu res , should not be

confused with the corresponding words are of course immune to i t .

49

If we abstract for a moment from the opacity and the dubious universality

of the semantic primitives, which i s a problem for lexical decomposition as

well as for other types of componential analysis , then we may ask the

question whether the verbials used to describe l in gu is t ic action are also

susceptible to the type of analysis under discussion.

14. The semantic complexity of notions such as promising, ordering, e tc .

i s such that i t cannot be captured in a lexical decomposition formula

for the corresponding verbs (which i s not to say that analyzing into

components i s impossible, but simply that the formalism does not work).

Elsewhere (see Verschueren 1977) I used the decomposition paradigm to

i l lu s tr a te this complexity. Compare the following two formulae; the f i r s t

one i s a variation of the one given earl ier for the verb TO KILL.

k i l l ( x ,y ) : CAUSE (x, COME ABOUT (DIE ( y ) ))

order(x,y,P): CAUSE (x, COME ABOUT (DO (y ,P)))

(In the second formula x i s the speaker, y i s the hearer, and P stands for

a proposition which, in this case, spec if ies a future act on the part of

the hearer.) Obviously the second formula is not complete as a description

of the meaning of TO ORDER. At lea s t the following elements need to be

added: unlike the dying associated with k i l l in g , the doing associated

with ordering is not a necessary oitcome of the action but merely an in­

tended one; the intention i s carried out by mean of uttering a sentence;

i t i s the speaker's intention that his uttering that sentence will make the

hearer do a certain action. Inserting all this additional information

results in the following formula (in which Se stands for a sentence).

50

ord er(x ,y ,P ) : SAY ( x ,y ,S e ) a INTEND (x , CAUSE ([SAY ( x ,y ,S e Q , COME

ABOUT (DO ( y ,P ) ) ) )

Put in plain words th is i s : 'x orders y to do P' means that ‘x utters

a sentence Se to y and x intends his uttering Se to y to cause i t to come

about that y does P'. However complex the formula i s , i t i s not y e t ex­

p l i c i t with respect to several points: f i r s t , the two conjoined parts of

the formula are part of the same act; second, the two occurrences of

SAY (x ,y ,Se) are id en t ica l ; th ird , the Se may be the surface expression

of ord er(x ,y ,P ) ; fourth, P i s part o f what Se expresses . But l e t us for a

while expect a cooperative a t t i tude from everybody so that making these

points e x p l i c i t in the formula would be redundant.

So far so good. But what happens i f we try to construct comparable

formulae for d i f feren t speech act verbs ? Here i s an attempt for TO REQUEST,

TO STATE, and TO ARGUE.

request(x ,y ,P): SAY (x ,y ,Se ) a INTEND (x, CAUSE ([SAY ( x ,y , S e f j ,

COME ABOUT (DO ( y ,P ) ) ) )

s t a t e ( x ,y ,P ) : SAY (x ,y ,S e ) A INTEND (x , CAUSE {[SAY ( x . y . S e ) ] , COME

ABOUT (KNOW ( y ,P ) ) ) )

argue(x ,y ,P ): SAY (x ,y ,S e ) A INTEND (x , CAUSE ([SAY ( x , y , S e ) j , COME

ABOUT (KNOW ( y ,P ) ) ) )

A serious problem emerges immediately. The formulae do not provide for any

d i f fe re n t ia t io n in meaning between TO ORDER and TO REQUEST and between

TO STATE and TO ARGUE; y e t TO ORDER i s not a synonym of TO REQUEST, and

TO STATE and TO ARGUE are not synonyms e i th er . There i s no obvious way

51

in which the authority implied by TO ORDER and the expectation of the

hearer's res is tance against bel ieveing P in the case o f TO ARGUE can be

formalized in such a way as to obtain s u f f i c i e n t ly d i f fe re n t ia ted de­

composition formulae.

Lexical decomposition od speech act verbs {by means of the formalism

i l lu s t r a t e d ) does not lead us any further than the construction of a more

or l e s s general formula of the following kind, in which some basic se ­

mantic components of a l l speech act verbs are brought together.

SAV(x,y,(P)): SAY (x ,y ,S e ) INTEND (x , CAUSE ( SAY (x ,y ,Se ) , COME

ABOUT (ACCEPT (y ,SA ') ) ) )

Note that SAV stands for speech act verb and SA1 for a l l the aspects of

the speech act in question, except ( i ) the e f f e c t s x t y p ic a l ly intends

to bring about, for which ACCEPT (y,SA') stands (which should be read as

'y accepts or appropriately reacts to the speech a c t ' ) , and ( i i ) the fac t

that x utters a sentence Se by means of which he intends to bring about

those e f f e c t s , for which the res t of the formula stands. I t should be

clear that such formulae cannot make e x p l i c i t the fu l l meaning of ind i­

vidual speech act verbs. F ir s t , in most cases i t w i l l not be poss ib le to

replace ACCEPT (y,SA1) by a s in g le word such as KNOW, DO or TELL; though

there i s cer ta in ly an e f f e c t ty p ic a l ly associated with acts such as pro­

mising (e .g . grat i fy ing the hearer in one way or another), i t i s not

always easy to describe, l e t alone to formalize. Second, a complete re­

presentation of SA' requires the formulation of most of the f e l i c i t y

condit ions on the act in question, which i s hardly f e a s ib l e within the

52

scope of a lex ica l decomposition formula.

The conclusion is simple. We must look for a d if ferent way of describing

the internal semantic structure of the verbials used to describe l in g u i s t i c

action {of which the speech act verbs referred to in this paragraph con­

s t i t u t e a subset.

15. How should the semantic relationships between the l in g u i s t i c action

verbials be ident i f ied ? An obvious place to s tart looking for an answer

i s in the s tr u c tu r a l i s t theory of lex ica l f i e l d s .

Saussurean structuralism regards a language as a unique system or net­

work of functionally related elements within which each separate element

derives i t s essence from i t s functional relation with the other elements.

With respect to the vocabulary of a language th is means that the meaning

of each word depends on the existence of other words. This posit ion in­

evitably leads to the conclusion that i t does not make sense to try to

define the meaning of one word separately. The meaning of each word

covers a r e la t iv e ly small co n cep tu a l a rea which i s part of a wider con­

c e p tu a l f i e ld .T h e co l lec t ion of words which together cover a complete

conceptual f i e l d i s called a sem a n tic f i e l d s a l e x i c a l f i e l d or l e x ic a l

domain. Within such a lex ica l f i e l d the s i z e of the conceptual area

associated with a particular word i s determined by the s i z e of the con­

ceptual areas of the surrounding words.

Since conceptual f i e l d s can be le x ic a l i z e d , i . e . s p l i t up into con­

ceptual areas, in d if ferent ways, the structure of the lex ica l f i e ld s

to which certain words belong has to be taken into account when semantic­

a l ly comparing those words with the ir equivalents in d if ferent languages.

53

Figure 1 v i su a l ize s th i s s ta te of a f f a i r s .

Language 1 word 1 word 2 word 3 word 4j ; ; ; j 11Conceptual f i e l d ----------- >Ll~n------ *-*— rr-------------------i * 1 • *

: :: .Language 2 word 1 word 2 word 3 ’

F igu re 1

There are three d i f feren t types of lex ica l f i e l d s (see Vassilyev 1974).

F ir s t , f i e l d s consist ing o f words belonging to the same part of speech,

such as TOWER, STEEPLE, TURRET, PINNACLE, e t c . , are ca l led p a ra d ig m a tic

f i e l d s . Second, we are confronted with a sy n ta g m a tic f i e l d i f i t cons is ts

of syntagmatically related words from d i f feren t parts o f speech; the

syntagmatic relat ionships are based on essen t ia l semantic connections

such as between BARK and DOG, HEAR and EAR, BITE and TOOTH, BLOND and

HAIR. A com plex f i e l d , f i n a l l y , includes words from d i f feren t parts of

speech ( i . e . the syntagmatic dimension) along with the ir paradigmatic

corre la tes . I t i s exemplified in Figure 2.

teacher teach pupil

paradigmatic dimension instructor ins truct student

professor

1f :

English

syntagmatic dimension

F igu re 2

54

Semantic f i e l d th e o r is ts have a lso drawn the attention to the ex istence

of le x ic a l gaps. We can talk about a lex ica l gap whenever a particular

language lacks a word to cover a certa in conceptual area within a con­

ceptual f i e l d which belongs to the experientia l world o f the language

community involved. An example i s to be found in Figure 3.

Eng. sleet'

Eng. snow Du. sneeuw

EngV Tiai 1 Du. hagel

Eng. rain Du. regen

Eng. prec ip i ta t ion Du. neerslag_____

F igu re 3

Though a Dutch equivalent to SLEET may e x i s t , i t i s absent from the com­

petence of most speakers and in most cases as paraphrase such as SMELTENDE

SNEEUW ('melting snow'), which does not f i t our cr i ter ion for l ex ica l

unity (see paragraph 4 ) , w i l l be used. Figure 3 a lso reveals the h ierar­

chical structure of the lexicon.

16. The problemsof the s t r u c tu r a l i s t approach to lex ica l f i e l d s are

manifold. The simple fac t that semantic f i e l d th e o r is t s usually exclude

complex l e x i c a l i z a t io n s and even compounds from the ir accounts should

already arouse our suspicion. Let us pass some o f the trouble spots in

review.

55

F irst , i f the conclusion that i t does not make sense to try to define

the meaning of one word separately i s so inevitable , how i s i t that people

do i t al l the time ? Second, how can the theory of lexical f i e ld s account

for the semantic difference between two words covering an identical con­

ceptual area, such as ANIMAL and BEAST ? Third, does the absolute arbi­

trariness of the way in which conceptual f i e ld s can be lex ical ized or

s p l i t up into conceptual areas, which is implied in the theory of lexical

f i e l d s , r e f l e c t rea l i ty ? Fourth, the practicabi l i ty of semantic f i e ld

theory depends on the p o s s ib i l i ty of defining the boundaries of word

meanings; whether such s t r i c t boundaries e x i s t i s an extremely doubtful

matter.

The four problems mentioned can a l l be reduced to an inconsistency

characterizing structuralism. On the one hand i t i s postulated that every

element within a l in g u is t ic system derives i t s essence from i t s function­

al relationships with other elements, and that every individual element

can only be described as a junction of relationships. On the other hand,

orthodox structuralis ts bel ieve that the language system is an independent

being which can be described as such without making reference to extra-

l in gu is t ic rea l i ty . The inconsistency should be clear: the structuralis t

mode of thinking is not applied to language as a phenomenon among other

phenomena. How the four trouble spots mentioned result from this lack of

consistency wil l be easy to understand after reading through the rest of

this chapter.

To put my crit ic ism in one sentence, as far as lexical semantics is

concerned, structuralism ignores the internal structure of ex tra- l inguis t ic

56

r e a l i t y , human cognitive processes, the expression of social r e a l i t y , and

l in g u i s t i c context.

17. Insofar as componential analysis i s an outgrowth of structuralism

i t shares the l a t t e r ' s problems. One att itude which the two usually have in

common i s the b e l i e f that an adequate description of the meaning of a

word must be such that i t covers the meaning that word has in a l l of i t s

l i t e r a l occurrences. In the case of semantic f i e l d theory th is att itude

was ident i f ied in terms of the requirement to define the boundaries of

word meanings. Fillmore (1975) dubs i t the c h e c k l i s t approach to meaning:

a word meaning i s represented as a l i s t of semantic features; i f an ob­

j e c t or event s a t i s f i e s a l l the features in the l i s t , the word can be used

to refer to i t ; otherwise i t cannot. The absurdity of such an approach

wil l be demonstrated soon.

18. So far i t was claimed that lex ica l decomposition was inadequate to

account for the internal structure of the verbials we want to describe,

and that the s tru c tu ra l i s t approach to lex ica l f i e ld s cannot be expected

to y ie ld s u f f i c i e n t insight into the semantic relationships among those

verbials .

Psychologists such as E. Rosch, so c io lo g i s t s such as E. Goffman, l i n ­

g u i s t i c anthropologists such as B. Berlin and P. Kay, computational l i n ­

gu is ts such as M. Minsky, R. Schank and R. Abelson, and semanticists such

as C. Fillmore have created numerous concepts by means of which structu­

r a l i s t reductionism can be avoided: sce n e 3 fram e3 schema3 d e s c r ip t io n 3

te m p la te 3 s c e n a r io 3 p r o to ty p e 3 module3 models p la n 3 s c r i p t 3 b a s ic l e v e l

57

term . As in most innovative areas of research there i s considerable termi­

nological confusion and c o n f l ic t . In my search for an adequate methodology

to analyze l in g u i s t i c action verbials I shall only handle the terms p r o to -

ty p e 3 b a s ic l e v e l term 3 scen e3 fram e and l i n g u i s t i c fram e .

19. The lex ica l iza t io n of the potent ia l ly i n f in i t e d ivers ity of rea l i ty

looks l ike a mission impossible for language. How then, can we explain

i t s re la t ive success ?

Though the boundaries between categories of objects are mostly d i f fu se ,

y e t r ea l i ty i s no undifferntiated continuum of phenomena. Flying animals

are nearly always feathered and only rarely furred. An object which looks

l ike a chair i s more frequently used to s i t on than an object which looks

l ike a pig. Such considerations led E. Rosch (1977a, 1977b) to believe

that ' in t r in s ic a l ly separate things' e x i s t in the world. If the conclusion

i s correct then this fac t may grease the wheels of the lex ica l iza t io n pro­

cess considerably.

Not only i s rea l i ty structured, but moreover human perception operates

according to some s t r i c t mechanisms. After an age-old b e l i e f in the arbitra­

riness of the color terminology, for instance, B. Berlin and P. Kay (1969)

have demonstrated quite convincingly that d if ferent languages show a strong

conformity with respect to the focal point of the area of the color spec­

trum covered by a particular basic color term (whereas the boundaries of

the area are always vague), and that there i s a universal evolutionary

pattern governing the acquisit ion of basic color terns by a language.

Though the original study called for several modifications, i t remains an

example of the interaction between the internal structure of rea l i ty and

58

the laws determining the human perceptual apparatus, in this case the un­

even color se n s i t iv i ty of the eye (which is discussed at length in P. Kay

and C.K. McDaniel 1978).

20. Both phenomena, i . e . the internal structure of rea l i ty and the laws

governing the perceptual apparatus, lead us to the notion of a p r o to ty p e .

The formation of prototypes, ' id e a l ‘instances of a particular category,

i s one of the cognitive processes enabling man to grasp the diversity of

the world with his mind. This process i s partly determined by the laws of

perception and the structure of rea l i ty . An example of the f i r s t influence

was mentioned with reference to the color terminology. The second influence

can be i l lu s tra ted by saying that a pigeon or a sparrow are more typical

or 'better' examples of a bird than a chicken or an ostr ich , and that a

dog or a cow is a more typical example of a mammal than a whale. Brief ly ,

for each lex ica l ized category human cognition contains a prototypical

example (which does not have to be an actual member of the category i t s e l f ,

but which is a kind of 'mental image1) which i s used as a yardstick to decide

whether a particular object can or cannot be referred to by means of the

lexical item in question. I f the deviation from the standard is too big,

the word cannot be used.

This brief exposition may create the fa lse impression that the prototype

idea i s only valid for nouns. But remember that color adjectives were adduced

as examples already; the 'focal points' referred to are the prototypical

examples of the colors denoted by the terms. To show that a similar ana­

ly s i s i s feas ib le for verbs I adopt an example given by C. Fillmore (1978):

the verb TO CLIMB typical ly describes an ascen d in g motion in a clam bering

59

fashion. I quote:

A monkey climbing up a flagpole satisfies both of these conditions.The monkey climbing down the flagpole satisfies the clambering compo­nent only, but is nevertheless engaged in an action that can be pro­perly called climbing. A snail climbing up the flagpole satisfies the ascending condition and can still be said to be climbing. But the snail is not privileged to olimb down the flagpole, since that acti­vity would involve neither clambering nor ascending, (p.153)

So there i s a c lear prototype, deviations from which are poss ib le . But i f

the meaning strays too far away from the standard, the word can no longer

be used appropriately. Also more abstract verbials can be approached in a

s imilar fashion. Consider the speech act verb TO CONGRATULATE. A typical

congratulation i s an expression of the speaker's being pleased about the

hearer's success in doing or obtaining something important. The f i r s t as­

pect of t h i s prototypical meaning i s completely absent from many formal

acts of congratulating. The second aspect i s being tampered with in the

fol lowing headline from the International Herald Tribune: "Begin congratu­

la te s Sadat on th e ir Nobel".

21. Such a prototype approach to meaning, which takes the internal s tru c t ­

ure of r e a l i t y , the laws o f perception and human cognit ive s tr a te g ie s into

account, immediately solves three of the problems pointed out in connection

with the s t r u c t u r a l i s t i c semantic f i e l d theory.

F ir s t , i t demonstrates that there i s no absolute arbitrariness in the

l e x i c a l i z a t io n of conceptual f i e l d s . Second, i t explains the age-old habit

o f defining the meaning o f many words without d irec t reference to the

surrounding lex ica l items in the same f i e l d : when doing so , the definer

r e s t r i c t s himself to the prototypical meaning. Third, the non-existence

60

of sharp boundaries between the conceptual areas covered by related words

in a particular lex ica l f i e ld i s no longer a problem for lex ica l semantics.

The main preoccupation of s tru c tu ra l i s t semantic f i e l d theoris ts was to

determine those boundaries, which turn out to be subject to strong in ter ­

personal variation {within one language community as well as between

different languages). With respect to the prototypical examples there i s

a much greater consistency in the word use of a speaker, a much greater

correspondence between the uses made of a lex ica l item by d if ferent speakers

of the same language, and a str iking s im i lar i ty in the use of equivalent

words in d if ferent languages. Evidently we have encountered a fundamental

property of word meaning, namely a psychological structure which i s im­

posed on i t (which might be represented as in Figure 4) .

word 1conceptual f i e l d <-

word 2 1 word 3I

F igure 4

This figure i s not meant to be usable for the representation of the over­

al l structure of complete lex ica l f i e l d s . As we shall see la t e r , i t can

only be used to depict the patterning of lex ica l f i e l d s along a s ingle

{or at best a few) 'semantic dimensions' — a term to be explained la ter .

22. The prototype idea also reveals the absurdity of the check lis t view

which i s s t i l l held by many l in g u is t s (most of those practicing lex ica l

f i e l d theory and/or componential ana lys is ) . To repeat, the check lis t

approach i s based on the view that an adequate description of the meaning

of a word must present the complete l i s t of conditions under which the

61

word can be appropriately used. The alternative i s to give only the 'ty­

pical' conditions, i . e . the prototype approach.

In trying to determine the meaning of the word BOAT one could come up

with a def init ion such as "a man-made object which can be used for trave l ­

ing on water". A defender of the checklist approach, coming across a boat

with a hole in i t and deciding that he s t i l l wants to call i t a BOAT

(though i t cannot be used for traveling on water any more), would have to

revise his definit ion: " a man-made object which can normally be used for

traveling on water, but in which there can also be a hole". Further, he

would have to determine how big the hole can be before the object in

question is no BOAT any more, but simply a WRECK. The impracticabil ity of

the checklist approach appears from the fact that not even i t s proponents

would want to be guilty of the absurdities mentioned. A defender of the

alternative theory could simply st ick to his def init ion and describe a

boat with a hole in terms of deviations from the prototypical boat. It

should be kept in mind that such 'deviations' are not necessaily 'defects ' ;

for that reason the example i s a b i t misleading. Probably the best argu­

ment for the prototype approach i s that , in practice, most analyses of

word meanings are based on i t , though often unconsciously.

23. It i s th is prototype approach to meaning that I shall make use of.

Though the standard or typical uses of semantic units are more stable than

the borders of their applicabi l i ty which can be expected to change very

much from speaker to speaker, even the prototypes are subject to some

interpersonal var iab i l i ty . This should not be surprising at a l l given that

prototypes are products of cognitive processes which serve to make perceived

62

r e a l i t y mental 1- manageable and given the fa c t that the world surrounding

a member o f a particular l in g u i s t i c community i s not necessar i ly ident ica l

to that of other members. Such v a r ia b i l i ty does not make the sp ec i f ic a t io n

of prototypes impossible. My judgments about prototypical meanings, how­

ever , must be regarded as h y p o th e se s because I have not had any opportunity

to ver i fy them by means of extensive psychological te s t in g . Not everybody

w il l agree with my judgments, but the important thing i s that such judgments

can be made.

Since prototypes are basic too ls in human categorizat ion , t h i s approach

to the meaning o f l in g u i s t i c action verbials w i l l lead us c lo ser to what

we are ult imately interested in , namely people's conceptualization of l i n ­

g u i s t i c behavior.

24. Just as there i s a psychological structure imposed on the semantic

structure o f lex ica l items (as v isua l ized in Figure 4 ) , there i s a lso a

psychological structure imposed on lex ica l f i e l d s . Not a l l c la s s i f i c a t o r y

l e v e l s in the hierarchic structure of the lexicon play an equally import­

ant role in the cognitive processing of r e a l i ty . The ex is tence of a c o g n i­

t i v e l y h a s ie l e v e l emerges from simple t e s t s of the following kind: i f

you show a guitar to someone and ask him "What i s th i s ?", the reply i s

more l ik e ly to be 'a guitar' than 'a musical instrument' or 'a fo lk g u i t a r 1;

i f you present an apple to someone, the answer to the same question i s more

l ik e l y to be 'an apple' than ' f r u i t ' or 'a golden d e l i c i o u s ' . Therefore

APPLE and GUITAR are ca l led b a s ic l e v e l te rm s : the others are superordi­

nate or subordinate terms. More examples are given in Table 1 (part of

which i s adopted from Rosch 1977a).

63

s u p e ro r d in a te

musical instrument

f r u i t

ai rcraft

bird

b a s io l e v e l

gu i ta r

■piano

apple

peach

airplane

he l icopter

sparrow

su b o rd in a te

folk guitar

■classical guitar

grand piano

upright piano

golden de l ic ious

mackintosh

freestone peach

-cl ing peach

DC 10

song sparrow

f i e l d sparrow bald eagle

golden eagle

T ab le 1

25. Though i t may be possib le to discover the basic l e v e l s for the average

member of a language community, they are, j u s t as prototypes, subject to

interpersonal d i f ferences . The v a r ia b i l i t y , however, i s not arbitrary: i t

depends on the degree of sp ec ia l i za t io n or the knowledge someone obtains

in a part icu lar f i e l d . For instance , an a irc ra f t engineer i s l ik e l y to have

DC 10 and 747 as th e ir basic leve l terms rather than AIRPLANE. On the other

64

hand, my limited knowledge of that f i e l d does not even allow me to f i l l in

subordinates for HELICOPTER except for HUEY which I acc identa l ly got to

know by reading about the Vietnam war. Conversely, a l i f e - lo n g c ity -dw el ler

i s not unl ikely to have BIRD as a basic level term rather than SPARROW

and ROBIN. Differences may even be more subtle: inhabitants o f some pigeon-

in fes ted European c i t i e s may be expected to have PIGEON as a basic level

term for pigeons and BIRD for a l l other birds (see Table 2 ) . In th i s case

there i s no simple correspondence between cognitive structure and b io lo­

g ical hierarchy.

animal bird robin California robin

animal bi rd sparrow f i e l d sparrow

animal bi rd pigeon fancy pigeon

su p e ro r d in a te b a s ic l e v e l s u b o rd in a te

T ab le 2

26. The a p p l ic a b i l i ty of the notion o f a basic leve l term i s not r e s t r i c t ­

ed to nouns. I ts applicat ion to adject ives i s im p l ic i t in Berlin and Kay's

color term studies: they r e s tr ic ted th e ir invest iga t ion s to basic color

terms which, though they were not defined on a purely cognit ive ba s is ,

might be regarded as the basic level adject ives in the color lex icon .

As for verbs, there i s probably a high degree o f correspondence between

basic leve l terms and what R.M.W. Dixon (1971) c a l l s n u c le a r v e r b s , i . e .

65

verbs which cannot be defined in terms of other verbs; non-nuclear verbs are

those that can be defined in terms of nuclear ones. Though th is def in i t iona l

cr i ter io n d i f f e r s considerably from the cognit ive c r i t e r ia adduced to s in g le

out basic level terms, there seems to be a correlat ion . The cognitive

sa l ien ce of nuclear verbs appears from the fac t that the mother-in-law

variety o f Dyirbal ( i . e . Dyal<juy, which i s only spoken in the presence of

certain tabo o -re la t iv es , and which i s characterized by the r e s tr ic t io n of

the number of l ex ica l items to an absolute minimum) contains mainly nuclear

verbs and hardly any non-nuclear ones.

But what about the relevance o f basic l e v e l s to l i n g u i s t i c action verb­

i a l s ? No doubt some s imilar judgments can be made, but extreme caution i s

ca l led for. Since the hierarchic structure of the lexicon i s l e s s c lear for

abstract concepts than for concrete ob jects , a lso the psychological

structure imposed on the hierarchy wil l be more d i f f i c u l t to detect .

Whereas my judgments about prototypes w i l l be merely hypotheses due to the

lack o f psychological t e s t in g , the judgments that I w i l l occasionally make

about basic level terms have to be regarded as extremely ten ta t iv e hypo­

theses . Such judgments wil l be made, however, whenever i t seems appropriate

and relevant. But because of th e ir ten ta t ive character I shall sometimes

ta lk about the most n e i tr a l terms instead of basic level terms (which i s

too much associated with Rosch's experimental methods).

27. Words are not i so la ted e n t i t i e s . Not only are they related to other

words, but a lso to e x tr a - l in g u i s t i c r e a l i t y . The verb TO SELL, for instance ,

immediately evokes a commercial scene consist ing of a buyer, a s e l l e r , money

and goods, between a l l of which there are relat ionships which are changed

66

throughout the interact ion . {For a detai led analysis see Fillmore 1977.)

Just as the human mind i s forced to create prototypes to process the in ­

f in i tude o f stimuli presented by r e a l i t y , so man i s a lso obliged to

construct prototypical instances of scenes in order to be able to handle

socia l r e a l i ty cogn it ive ly . Such a prototypical instance o f a scene I

call a frame. Hence, to be prec ise , we would say that the verb TO SELL

evokes a commercial frame, i f we are interested in describing i t s pro­

totypical meaning.

28. Lexical meanings are always partly determined by an associat ion with

such frames and/or non-prototypical scenes. Here i s an example adapted

from Fillmore (1977). We can say about John "He spent three hours on

land" and about Dick "He spent three hours on the ground". Though these

sentences may be used to describe the nearly ident ical behavior of John

and Dick in Antwerp on a rainy afternoon, ye t they do not mean the same.

The only reason i s that ON LAND evokes the frame of navigation whereas

ON THE GROUND i s associated with the frame of av iat ion .

Not only words with a c lear lex ica l meaning such as LAND, GROUND and

TO SELL are connected with frames. Also l e x i c a l l y empty items such as

WHAT carry such assoc ia t ions . Consider the question "What are you ?".

The interrogative pronoun in th i s sentence used to be associated with

the world o f profess ions , the appropriate answer being 'a doctor ' ,

'a teacher' , 'a student' e tc . But nowadays i t i s ambiguous in some soc ia l

c ir c l e s since i t may e i th er evoke the professional world or the practice

of astrology ( in which case the appropriate answer i s 'a Taurus', 'an

67

Aquarius', e t c . )

The notion of a frame provides us with a solution to the problem of how

to account for the semantic difference between two words covering an iden­

t ica l conceptual area, such as ANIMAL and BEAST. In many contexts the two

nouns simply evoke d if ferent frames.

Needless to add that also scenes and frames are subject to interpersonal

variation,

29. The se t of l in g u i s t i c choices - - textual as well as grammmatical and

lex ical — which i s associated with a particular frame I call a l i n g u i s t i c

fram e. Its lex ical part i s a l e x i c a l fram e. I t wil l be c lear that the

la t t e r i s not independent of the rest of the l in g u i s t i c frame. To give

a simple i l lu s tr a t io n (again borrowed from Fillmore 1977), even the choice

of a determiner may influence the frame with which a noun i s associated.

Compare "He did not give me a tip" with "He did not give me my t ip" . The

indef in i te a r t i c le in the f i r s t sentence places TIP in the frame of gene­

ros i ty , whereas the possessive pronoun in the second sentence transposes

i t into a frame of obligatory payment. Hence "He did not give me a tip"

i s an accusation of bad manners, whereas "He did not give me my tip" i s

an accusation of doing an i l l e g a l thing. By connecting lex ica l meaning in

that way with other l in g u i s t i c e n t i t i e s (on every level of structure)

lex ical semantics transcends yet another l im itat ion of the s tr u c tu r a l i s t i c

approach to semantic f i e l d s .

One might be tempted to ident i fy lex ical frames with what the structural­

i s t s ca l led semantic or lex ica l f i e l d s . Such an ide n t i f ic a t io n would be

wrong. Lexical frames are even more complex than complex lex ica l f i e l d s .

68

The set of lexical choices associated with the commercial frame, for in­

stance, does not only include TO BUY, TO PURCHASE, TO PROCURE, e t c . , but

also TO SELL, TO PAY, TO CHARGE, etc . (As a complex f i e l d , i t does, of

course, not only include verbs.) It i s impossible to force a l l - o f these

into a paradigmatic-syntagmatic scheme such as Figure 2.

30. Summarizing, I am about to undertake a comparative study of a number

of subsets of the verbials belonging to the lexical frame associated with

l in g u is t ic action. When i t comes to semantic analyses the notions of

frame, basic level and prototype will be made use of .

In the foregoing discussion of these concepts I repeatedly drew the

attention to interpersonal var iab i l i ty . Such var iab i l i ty i s bound to make

the job endless. An approach which obliges you to keep working forever

i s l ik e ly to be regarded as problematic. But frankly, I have never seen

a phenomenon described completely, even with the most reductionistic

approach imaginable. So where i s the advantage ?

Furthermore, the fact that success in communication is not automatic

even i f the speaker uses the ‘r ig h t ’ words to express his thoughts, but

that i t depends crucially upon a laborious interpretation process, i s not

only veiled by the conduit metaphor (remember M. Reddy 1977) but also by

all l in g u is t ic invest igations based on the view that i t i s j u s t i f i e d to

abstract from the human (cognitive and social) aspects of a basically

human phenomenon such as language. Such invest igations act ive ly support

the conduit metaphor and may thus impede e f f i c i e n t communication. Not

only are they out of touch with the actual character of the object of

investigation (which makes them s c i e n t i f i c a l ly irrelevant) but they are

69

also misleading in an acutely dangerous way (which could almost be said

to make them s o c ia l ly irresp o n s ib le ) . The approach sketched above cannot

be blamed on e i th er of these two counts. I bel ieve that we should not

try to construct troubleproof theories; rather, we should i n s t i t u t i o n a l i z e

he troubles which r e f l e c t actual interpretation processes — for the sake

of both the theory and the people who are confronted with i t .

D. CIRCULARITY AND HOW TO AVOID IT

31. There i s always a danger of c ir c u la r i ty in the attempt to describe

the meaning of l in g u i s t i c units: one can hardly avoid using language for

i t , i . e . using paraphrases every part of which i s as much in need o f an

explanation as the element described. But t h i s invest igat ion runs the risk

of double c ir cu la r i ty : language w i l l have to be used to describe the way

in which language is described by means of language. I b e l ie v e , however,

that the risk of being c ircu lar can be minimized.

F ir s t , there i s no c ir c u la r i ty in ostensive d e f in i t ion s such as "That

i s a tower" pronounced while pointing at a tower. Since we are concerned

with lex ica l items used to describe l in g u i s t i c actions we can usually pre­

sent some ostensive d e f in i t io n s simply by reproducing examples o f the

verbal acts described by means of the words and expressions in question.

Of course t h i s cannot work with respect to those lex ica l items focusing

on an aspect o f verbal behavior which cannot be reproduced on paper. In

those cases a circumscription wil l have to replace the i l l u s t r a t i o n .

Needless to say that ostens ive d e f in i t io n s do not provide real c la r i t y .

I t s u f f ic e s to be reminded of the story about Captain Cook who, when he

70

landed in Austral ia , saw kangaroos jumping around, asked some aborigines

"What i s that ?" while pointing at them, and got as a reply something l ik e

"Kangaroo", which (again according to the story) i s supposed to mean "What

do you say ?".

32. Though adducing i l l u s t r a t i o n s or ostensive d e f in i t ion s can be regarded

as a non-circular way o f c la r i fy in g the meaning of the verbia ls we are

in terested in , i t i s by no means a fu l l - f le d g e d semantic procedure.

The main component o f the approach to meaning which I shal l take i s

im p l ic i t in writings as divergent as Wittgenstein (1953) and Lei si (1973).

The meaning of the semantic units referring to (an aspect of) a l in g u i s t i c

action can be presented by formulating the conditions to be s a t i s f i e d in

order for the description to be appropriate. Again there can be no c i r ­

cu la r i ty involved since no paraphrases are needed. This approach i s c le a r ­

ly pragmatic. Linguist ic pragmatics i s , a f ter a l l , the study of the appro­

priateness conditions on the use of natural language.

We are talking about the appropriateness conditions on the use o f the

verbials we are inves t ig a t in g . But remembering the . double.; !inguistic layering

of the enterprise (analyzing le x ic a l items which are used to describe l i n ­

g u i s t i c behavior) most — though not all — of them can be formulated in

terms of conditions on or chara cter is t ic s of the acts described by means■ s i - .i

of the verbia ls in question. The general form of the analys is w i l l be th is :

a verbial V can be appropriately used in a description D o f a l i n g u i s t i c

action A i f A s a t i s f i e s condition C (or possesses c h a r a cter is t ic C). The

symbols D ( for the describing a c t ) , A (for the act described) and C (for

71

the conditions on or charateris t ics of A necessary for D to be an appro­

priate description of A) will be used throughout. Because we shall have

to talk about many describing ac ts , acts described and conditions, they

wil l be numbered consecutively (Dl, D2, e t c . ) per section (where a 'sect ion'

i s any part of a chapter with a t i t l e preceded by a capital l e t t e r ) . In

the few cases in which an appropriateness condition on the use of a

l in g u i s t i c action verbial cannot be formulated in terms of conditions on

the act described, the symbol Cv wil l be used for i t i f i t can be formulated

in terms of conditions on the use of the verbial as such or Cd i f a formu­

lation in terms of conditions on or properties of the describing act i s

needed. A few more symbols: S stands for speaker and H for hearer; the

speaher and the hearer of A are abbreviated as Sa and Ha, and those of D

as Sd and Hd.

In the following paragraph a br ief overview of appropriateness conditions

will be given which I use as a heuris t ic procedure. For a further j u s t i ­

f ica t ion of the d is t inct ions I make, see Verschueren (1978).

33. Appropriateness conditions are not ju s t attached to utterances as such,

but to d if ferent leve ls of l in g u i s t i c structure. At the highest level of

generality we find a number of conditions attached to the use of language

as such: i f a speaker S t r i e s to communicate with a hearer H by means of

a language L, he must assume that H understands L, that H is not deaf,

or that he can read, e tc . This i s part of what Searle (1969) includes in

his 'normal input and output conditions'.''' As with many other conditions

1 Searle's formulation of his 'normal input and output conditions': "To-

th e ir importance i s only f e l t in case they do not obtain.

The next level i s that of the com m unication t y p e . Different types of

communication are, for instance , conversation, l i t e ra r y w rit ing , scho­

la r ly writ ing , advertis ing , e tc . The maxims of quantity, q u a l i ty , re lat ion2

and manner sp ec i f ied by Grice (1975) were intended as conditions attach­

ed to a particular communication type, namely conversation (though they

can be generalized to some other types o f communication as w e l l , and

though they are not equally val id for a l l types o f conversation).

Next i s the s t y l e or c o d e . A language i s not an absolutely homogeneous

instrument. Rather, c lo s e ly related intentions can get very d i f feren t

expressions as in "Hi, Joe" versus "Good morning, Professor Carruthers"

both of which can be addressed to the same person. Different appropriate­

ness conditions are attached to these d i f feren t s t y l e s .

gether they include such things as that the speaker and hearer both know how to speak the language; both are conscious of what they are doing; they have no physical impediments to communication, such as deafness, aphasia, or laryngitis; and they are not acting in a play or telling jokes, etc. It should be noted that this condition excludes b o th impediments to communication such as deafness and also parasitic forms of communication such as telling jokes or acting in a play." (1969:57) The problem of seriousness and literalness (which is referred to in this quote by means of the term 'parasitic forms of discourse' as opposed to serious and literal types of communication) is not to be situated at the level of language as such.

2 For the sake of convenience I list Grice's (1975) maxims of conver­sation: C o o p e ra tiv e P r in c ip le : "Make your conversational contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged"; Maxims o f Q u a n ti ty:"Make your contribution as informative as is re­quired (for the current purposes of the exchange)" and "Do not make your contribution more informative than is required"; Maxims o f Q u a lity supermaxim: "Try to make your contribution one that is true", plus "Do not say what you believe to be false" and "Do not say that for which you lack adequate evidence"; Maxim o f R e la t io n : "Be relevant"; Maxims o f Manner: supermaxim: "Be perspicuous", plus "Avoid obscurity of expression", "Avoid ambiguity", "Be brief (avoid unnecessary pro­lixity)", and "Be orderly".

73

With the level of the t e x t there i s , for instance, a condition of co­

herence associated.

At the level of the speech a c t ty p e we have to s i tu a te the whole

apparatus of conditions on i l locutionary acts worked out by Searle (1969).

Many of these are sp ec i f ic applications of Grice's maxims of conversation.

Then comes the occas ion-spec i f ic speech a c t . This level i s separate

from the previous one for the following reason. Whereas, for instance,

the speaker's intention to do a future act i s a condition on the perform­

ance of a promise in general, the speaker's intention to come tomorrow

i s a condition on the occasion-specif ic utterance of the promise "I pro­

mise to come tomorrow".

In order for certain sen ten ce s tr u c tu r e s to be appropriate, certain facts

have to obtain. The wh-question "When did George come back ?" presupposes

that George came back. Similarly, the c l e f t "It was John who murdered the

grocer" presupposes that someone murdered the grocer, and the pseudo-cleft

"What Henry did was to burn down the garage" presupposes that Henry did

something.

Appropriateness conditions are also associated with smaller l in g u i s t i c

units. For some noun p h ra se s such as de f in i te descriptions to be appro­

priate the object referred to has to be id e n t i f ia b le . Many w ords carry

presuppositions (which const itute one type of appropriateness conditions):

a verb of transit ion such as TO AWAKE is only appropriate i f the person

who i s said to awake was asleep before; a fact ive predicate such as TO RE­

GRET presupposes that the regretted s ta te of a f fa ir s obtains; s im ilar ly ,

many adverbials such as AGAIN, e tc . presuppose something. Even sound

f e a tu r e s such as a particular intonation pattern are only appropriate

74

under certain circumstances.

A word of caution i s ca l led for. The foregoing i l lu s tr a t io n s may be

misleading in that a l l the conditions presented were somehow conditions

on standard forms of language. Needless to say that l in g u i s t i c behavior

deviates from th is standard a l l the time. But there i s a serious danger

of confusion here. I claimed before that what I was mainly interested in

was the prototypical or standard meaning of l in g u i s t i c action verbials

because I believe that the prototype approach to lex ica l meaning i s the

only tenable one. But what should be kept in mind i s that a deviation

from the standard conditions, spec i f ied in this paragraph, attached to

some level of the l in g u i s t i c structure of A may be a standard condition

on the appropriate use of a verbial V in a description D of act A.

34. The reader should be warned against interpreting the attempt to

formulate 'appropriateness conditions' as a return to l in g u i s t i c s as a

prescriptive rather than a descriptive enterprise . Such an interpretation

completely misses the point. I hope th i s statement i s not in need of

further c la r i f i c a t io n .

35. The conditions we are talking about are a matter of contextual

appropriateness. Thus a condition attached to a particular level of l i n ­

g u is t ic structure may be related to features of a d i f ferent level of

the same l in g u i s t i c ac t , but a lso to features of the e x tr a - l in g u is t ic

context including the speaker’s intentions and the e f f e c t s produced in

the hearer. All these aspects of l in g u i s t i c action are represented

in Table 3.

75

ex tra -1 in g u is t ic r ea l i ty

speakerINTENTION

Linguist ic action "* language

communication type

s ty le

tex t

speech act type

speech act

propositional con­tent

sentence s t r u c t ­ure

phrases

words

hearer ' EFFECT

T able 3

As I have said before, the previous overview of appropriateness con­

dit ions on the basis of the d i f feren t l e v e l s o f l in g u i s t i c structure to

which they are attached mainly serves as a h e u r is t ic procedure and w i l l

not be immediately r e f lec ted in the ensuing exposi t ions . Thus i t may

happen that , when specifying a condition C on the appropriate use o f a

verbial V i s a description D of an act A, I shall be talking only about

an aspect o f the e x t r a - l in g u i s t i c context o f A or of the communication

type to which A belongs though in fa c t C i s attached to the speech act

type or the propositional content of A. For instance , consider the con­

dit ion that a speaker S must have authority over the hearer H in order

for the verb TO COMMAND to be appropriately used in the description "S

commanded H to leave" o f the act o f uttering “Leave I". This condition i s

completely formulated in terms of an e x t r a - l in g u i s t i c re lat ionship between

76

the speaker and the hearer, though i t i s a condition attached to the speech

act type.

In most cases conditions w i l l be attached to several of the l e v e l s of

l i n g u i s t i c structure of the act to be described. But usually not a l l of

those conditions w i l l be equally s a l i e n t . These degrees of sa l ience provide

us with one cr i ter ion for ident ify ing subfie lds within the vast lex ica l

frame under invest iga t ion . Actually, th i s cr i ter ion was used in my se lec t io n

of the subfie lds I am going to concentrate on in t h i s essay.

PROBLEMS OF REPRESENTATION

36. In the f i r s t chapter of t h i s essay I have advanced my view o f th i s

enterprise as an attempt to map the conceptual space associated with

l in g u i s t i c action as re f lec ted in the words and expressions used in d i f f e r -

ent languages to describe i t . But how can such a 'geography of the mind1

be undertaken ? Consider Figures 5 and 6 (both of which I borrowed from

Bolinger 1975). Such mappings are quite straightforward. But notice that

in Figure 6 there i s only a dimension of moistness at work. In Figure 5

there are three such dimensions, namely humanness, age and sex; ye t the

diagram retains a high degree of s im p l ic i ty . But what can we do with words

which are much more complex and cannot be adequately accounted for in

terms o f a couple of fea tures , such as our l in g u i s t i c action verbials ?

Clearly, a comparable graphic representation i s impracticable i f i t s

purpose i s to be a sa t i s fa c to ry account of the d if ferences in meaning

between those verb ia l s , and, in that way, to s p l i t them up into semantic

c la s se s .

Female

Non-humanHuman

Non-adul

BOY

Male

F igure 5

F igure 6

78

37. As mentioned before, attempts have been made to distinguish sem an tic

c la s s e s of l in g u i s t i c action verbials . Remember the attempts to c la s s i f y

speech act Verbs or the i l locutionary forces they describe. Though the

number of semantic dimensions that have to be taken into account makes

graphic representations of the type i l lu s tra te d impossible, such c l a s s i ­

f ica t io n s are fea s ib le . There i s , however, one condition: a decision has

to be made as to the re lat ive importance of the dimensions involved.

Several d if ferent decisions can be made, depending on one's theoretical

point of departure. Since the purpose of th is invest igation i s to aban­

don theoretical biases as much as poss ib le , I wil l not engage in the con­

struction of c la s s i f i c a t io n s as a l l . I am only interested in dicovering

the sem a n tic d im ensions along which l in g u is t i c action verbials can be

compared. Not only does th is enable us to remain fa i l th fu l to the o r i ­

ginal in tent ions , i t also makes graphic representations possible again.

Most verbials under invest igation will be presented in diagrams which I

call sem an tic d im ension com parison ta b le s (henceforth SDC-Tables).

38. The SDC-Tables will have the following general form. At the center

the semantic dimension will be represented by a l i s t of actsfnumbered Al,

A2, e t c . ; those numbers wil l be substantiated in the text ) and the r e le ­

vant conditions (numbered Cl, C2,etc .; again those numbers occur in the

tex t ) under which they are performed, i . e . those conditions relevant to

the semantic dimension in question. To the l e f t the Dutch verbials wil l

be l i s t e d in such a way that i t wil l be clear which acts performed under

which conditions can be appropriately described by them; whenever ne­

cessary also the description D (a lso numbered as in the tex t ) in which

DUTCH SEMANTIC DIMENSION

- A1 f - Cl

- - r? -------

Word 1

Word 2

A6

F igure ?

ENGLISH

Word 1

Word 3

Word 2

80

they are so used may be indicated; moreover, large dots wil l often indi­

cate (as in Figure 4) the prototypical uses of the verbials. To the right

the same will be done for English. As a resu lt we get a visualized com­

parison of the relevant Dutch and English verbials with respect to a

particular semantic dimension. The general form of these diagrams i s de­

picted in Figure 7. Such diagrams will be presented for most dimensions

along which appropriateness conditions are formulated and compared. Need­

less to say that often there wil l be several layers of overlapping words

on each side of a single SDC-Table, and that the area covered by a parti­

cular word does not have to be a continuous part of the dimension.

39. Occasionally i t wil l happen that a particular semantic dimension is

relevant to the meaning of so many verbials that including them in the

SDC-Table in question is physically impossible. In such cases the d i f f e r ­

ent sem an tic p a t te r n s (abbreviated SP) to which the verbials conform

will often be numbered in the SDC-Table and a complete or representative

l i s t of the verbs an phrases belonging to each of the patterns d is t in ­

guished will be given in the tex t . Thus a sem an tic p a t te r n i s defined

as the way in which a word (or group of words) relates to a particular

dimension of i t s meaning. Figure 8 depicts the general form of such sim­

p l i f ie d SDC-Tables.

Another way out when the number of relevant verbials i s too large to

squeeze into the tables i s to simply give one or two examples. It wil l

always be clear from the text which other verbials show the same semantic

pattern as the examples given in the SDC-Table.

81

DUTCH SEMANTIC DIMENSION ENGLISH

SP2

SP3

A1

- C2

- C3 SP1

A2

- C4

A3

- C5

V- C6

A4*

SP2

F-iguve 8

F. CONCLUDING REMARK

40. In th i s chapter I hope to have succeeded in defining the object of

invest igat ion more p r e c i se ly , in explaining the pr inc ip les o f l ex ica l s e ­

mantics underlying my approach, and in c lar i fy in g the way in which the

analyses wil l be formulated and the resu lts w i l l be presented. Thus we

should be ready now to take language as our guide for the tour through

l in g u i s t i c action. In view of the inev itab le l im ita t ion s o f the i n v e s t i ­

gation some readers might be prompted to abandon the adventure altogether .

But a f ter a l l , curta i led ins ights are better than no in s igh ts at a l l .

There i s one aspect o f the inves t iga t ion that I have neglected to mention

so far. One of the ways to pursue my ultimate in te r e s t in the conceptual­

82

ization of l in g u is t ic action would be to study the metaphors revealed in

i t s Dutch and English lex ica l iza t ions . However, considerations to that

e f f e c t will only sporadically be found in the following chapters.

83

III. THE SEMANTICS OF SILENCE

A. INTRODUCTION

1. In the introductory chapter we announced that we are going to make

a guided tour through l in g u is t ic action with language, in particular the

verbials i t furnishes to describe what one does with words, as our guide.

A good guide does not only take you to the obvious places which can be

seen and recognized from miles away. Language f i t s that definit ion of a

good guide. Immediately i t draws our attention to an aspect of l in g u is t ic

action which a l l theoreticians have ignored: the many ways of being s i l e n t .

There is more to si lence than the absence of speech. Silence can be golden,

deathlike, tomblike, solemn, and even pregnant; but i t is rarely neutral.

That i t , when s i lence is neutral i t i s rarely talked about or even no­

t iced .

One can legit imately wonder whether s i lence is a l in g u is t i c a l ly s i g n i f i ­

cant category. Let us make a comparison. Though for a geographer the com­

plete set of holes in the surface of the earth does not form a natural

category, i t consists of a number of subsets (e .g . vulcanic craters , holes

caused by the impact of meteorites, those caused by earthquakes, e t c . )

84

which do. Similarly, though al l instances of l in gu is t ic s i l en ce , taken

together, do not constitute a natural l in gu is t ic category, i t i s l in g u is t ­

i c a l ly s ign if icant to get an overview of the complete se t because i t con­

s i s t s of a number of subsets the 'natural category' status of which would

be hard to deny.

2. For the sake of brevity I baptize all the verbials describing a person's

being s i l e n t verba ta een d i (Latin TACERE means 'to be s i l e n t ' ) . And instead

of having recourse to the quite informal OYSTER or CLAM, the s i l e n t person

will be dubbed the ta een s.

Due to the marginal position of the verba taeendi among the l in g u is t ic

action verbials, this chapter d if fers from the others in two important

respects. F irst , some negations of non-silence l in g u is t ic action verbials

such as NOT SPEAK and NOT TALK (the lexical status of which was discussed

in paragraph 4 of the preceding chapter) will be regarded as complex

l ex ica l izat ions because they are conventional means for talking about

the absence of speech. Almost no comparable lex ica l iza t ions wil l occur

in the following chapters. Second, ostensive def init ions will be impossible

here: acts of being s i l e n t cannot be reproduced on paper. Therefore the

acts referred to by means of the lexical items under investigation can only

be described, whereas i l lu s tra t io n s can be adduced when we come to the

study of the other lexical subfields singled out in this essay.

B. THE FRAME OF SILENCE

3. Silence is no doubt the most marginal aspect of l in g u is t ic action

imaginable. Yet in our talking about l in gu is t ic action i t i s certainly

85

part o f i t . One would never make an assert ion about someone's being s i l e n t

unless he could have been expected to speak. Whereas the marginal character

of s i l en ce within the frame of l in g u i s t i c action i s the reason why theo­

r i s t s o f communication have usually neglected to ta lk about i t , i t pro­

vides me with two arguments to treat the verba taeendi in th i s essay which,

as announced, i s intended to concentrate only on some r e p r e s e n ta t i v e sub­

f i e l d s of the l in g u i s t i c action verb ia ls .

F ir s t , no matter what phenomenon one i s truing to account fo r , the

account i s hopelessly inadequate i f only i t s 'spectacular' aspects are

dealt with and nothing i s said about the unspectacular ones. Let me try

to c la r i f y the relevance o f th i s general stetement for the problem at

hand by carrying the metaphors o f th is undertaking as a kind of 'geogra­

phy o f the mind' and as a 'guided tour' a b i t further. Someone who has

v i s i t e d Yellowstone National Park and the Grand Canyon but has no idea

of the l e s s impressive Midwestern landscape, or someone who has only seen

New York City or San Francisco but has no idea of the kind of l i f e going

on in a l e s s exc i t ing small Midwestern town, cannot be said to know the

United States . Si lence i s the usually unnoticed and rarely appreciated

Midwest o f l in g u i s t i c action. The area has to be included in any sample

which purports to be representative .

Second, because of i t s marginality , s i l en ce i s a r e f lec t io n of a l l the

major aspects of l in g u i s t i c action. This i s e sp e c ia l ly c lear a f te r a sur­

vey of the verba taeendi. S i lence i s not j u s t the absence of l i n g u i s t i c

sounds. There is a lso a meaning to i t . I t shows contextual l in ks . There

are social value judgments attached to s i l e n c e . There i s a speaker and a

hearer involved - - however strange t h i s may sound. And so on. B r ie f ly ,

86

the frame of si lence i s the frame of l in g u is t ic action i t s e l f . Whereas all

other subfields of l in g u is t ic action verbials studied in this essay are

centred around a few heavily emphasized aspects of verbal behavior, no

such emphasis i s clear with the verba taeendi. But l e t us now start

l is tening to what the words have to say; in other words, l e t us s tart

invest igating what the lexical frame of s i lence has to teach us.

C. SILENCE3 MEANING, CONTEXT AND COMMUNICATION TYPE

4. Though I am more interested in the semantic dimensions along which

l in g u is t ic action verbials can be compared than in the semantic c lasses

into which i t i s possible to group them, i t may be helpful to observe

that the se t of verba taeendi seems to f a l l apart in three quite natural

categories (which can themselves be regarded as points of a semantic

dimension). The contrast between the f i r s t two c lasses re f lec ts the d is ­

t inct ion between the form and the content of utterances or between the

act of speaking as such and what one says. Let us call the lexical tools

to describe the absence of acts as speaking as such verba s i l e n d i . The

basic ones in English are probably TO BE SILENT, NOT TALK, NOT SPEAK, and

TO SAY NOTHING. Many other verba taeendi mean that the tacens refrained

fromtalking about a particular subject (in which case he may e ither be

completely s i l e n t or talking about something e lse ) : these I call verba

r e tv e e n d i (Latin RETICERE i s synonymous with the transit ive form of

TACERE); the object in question will be referred to as the ta cen d im , pi.

tacen da . The basic English examples are TO BE SILENT ABOUT, NOT TALK ABOUT,

NOT SPEAK ABOUT and TO SAY NOTHING ABOUT. In addition to the verba si lendi

87

and the verba reticendi there i s a third category of words and expressions

by means of which a person's being s i l e n t can be described. With the verba

s i lendi they share the property o f designating the absence of speech as

such; a spec i f ica t ion of what i t i s that i s not being talked about i s

not necessary, though many of the verbials in question can get or even

require a complement. I am talking about what could be called the v erb a

e e s sa n d is verbials denoting a person's s i lence a f ter he has been speaking

(or writing) and sometimes implying that he wil l resume his l in g u i s t i c

a c t iv i ty a f ter a br ief interval . The complements they can get may specify

the nature of the preceding discourse.

Observe that there i s another se t of verbials which describe l in g u i s t i c

actions related to s i len ce . I am thinking of TO SILENCE, TO PUT TO SILENCE,

e tc . These, however, are not verba taeendi: they describe (usually l i n ­

g u i s t i c ) acts performed in order to cause somebody e ls e to be s i l e n t .

Therefore they do not belong to my object of invest igation in th i s chapter.

Remember that the purpose of th i s paragraph i s not to se t up a taxonomy

of verba taeendi, which would be in contradiction with the overall aim

of th is work. Rather, I present the d is t inct ions in order to demonstrate

the kinds of verbials that I am about to analyze, and in order to s e t up

a terminology which wil l make the discussion eas ier . Moreover, and th is i s

the main reason, the d is t in ct ions provide us with a kind of semantic dimension

along which the basic verba taeendi can be usefu lly compared, as wil l soon

become clear.

With b a s ic verb a ta e e n d i I mean those which are somehow neutral, un­

marked; that i s , those verba taeendi the app l icab i l i ty of which i s governed

by the le a s t sp ec i f ic appropriateness conditions. I f my l in g u i s t i c in t u i t -

88

ions had been backed by elaborate psychological t e s t in g , I would probably

have dared to call them basic level terms. Notice that I have not given

an example of a basic verbum cessandi; i t seems to me that there i s no

such thing in English; TO STOP TALKING i s not a lex ica l item (according

to the criter ion put forward in paragraph 4 of the previous chapter) and

TO FALL SILENT is not neutral as wil l be shown la ter .

5. From the few examples adduced so far two facts about English verba

taeendi emerge. F ir s t , the basic verba taeendi are always com plex l e x i c a l -

i z a t i o n s . Many simple lex ica l iza t io n s wil l be encountered in the fol low­

ing discussion, but the ir meaning i s always more spec ia l ized .

Second, English lacks a g e n e ra l term for the absence of speech. TO BE

SILENT (which will only be studied in i t s absence-of-an-act-of-speaking

sense, not in the sense of being habitually reluctant to speak) does not

necessari ly mean that no words were uttered at a l l , though th i s i s no

doubt i t s central meaning; i t i s a more re la t ive notion. I f Mr. Smith

refrained from speaking on a certain occasion, the absolute absence of

his speech a c t iv i ty would be accurately described by saying

(Dl) Mr. Smith was s i l e n t

But the same sentence could be used i f Mr. Smith had indeed been talking

but i f he had been more taciturn than he used to be or than was espected

of him on such an occasion. In other words, TO BE SILENT i s applicable to

(Al) and (A2).

(Al) Sa utters no words at a l l

(A2) Sa utters few words (or fewer than usual under comparable c i r ­

cumstances

89

Consequently, TO BE SILENT i s not only applicable to s i tua t ion s in which

a person abstains from using language.

Moreover, not a l l s i l en ce acts can be described with i t . In order to

show th is we dist inguish two more poss ib le types of such ac ts .

(A3) Sa utters no words about a part icu lar top ic

(A4) Sa disconstinues his uttering o f words

The a t ten t ive reader w i l l have noticed that (Al) sp e c i f i e s the domain

of the verba s i l e n d i , (A3) that of the verba reticendi and (A4) the domain

of the verba cessandi. TO BE SILENT, as i t stands, can only be used in

some marked descriptions of (A3) and (A4). I regard a description of

(A3) as marked when (Cd2) instead of (Cdl) obta ins , and a description o f

(A4) when (Cd4) instead of (Cd3) obtains.

(Cdl) The tacendum i s expressed in the propositional content o f D

(Cd2) The tacendum i s not expressed in the propositional content o f D

(Cd3) The discontinued discourse i s not overt ly indicated in D

(Cd4) The discontinued discourse i s overt ly indicated in D

Since the tacendum of a s i l e n c e act can be anything, (Dl) cannot be con­

sidered an adequate account of (A3) except , f o r instance , i f (Dl) i s

uttered in reply to a question such as "Did Mr. Smith say anything about

his forthcoming divorce ?". Because such a specia l context i s required

to avoid an expression o f the tacendum in the propositional content o f a

description of (A3), any D describing (A3) and conforming to (Cd2) i s

marked whereas i t would be unmarked or neutral when conforming to (Cdl).

TO BE SILENT can only be used in such a marked description o f (A3). How-

90

ever, only a small adaptation i s needed to make TO BE SILENT applicable

in D describing (A3) and conforming to (Cdl), namely the addition of the

preposition ABOUT. But what about (A4) ? It seems that verba cessandi

such as TO FALL SILENT do not require an overt sp ec i f ic a t io n of the d i s ­

continued discourse in the D in which they are used to describe (A4):

the fa c t that the s i l e n c e act fol lows speech i s c lear from the verbial

i t s e l f . Hence D is unmarked i f i t conforms to (Cd3) but marked i f (Cd4)

i s s a t i s f i e d . Again, TO BE SILENT can only be used to describe (A4) in a

marked D: (Dl) does not mean that the utterance o f words was discontinued.

But in (D2), which s a t i s f i e d (Cd4), TO BE SILENT i s a verbum cessandi.

(D2) He was s i l e n t for a couple of seconds (before he resumed the

thread of his discourse)

Notice that even the addition of ' for a couple o f seconds' may be s u f f i c i e n t

as an overt indication o f preceding discourse. In th is case a small adapt­

ation cannot make TO BE SILENT applicable in D describing (A4) and con­

forming to (Cd3). The verbial TO FALL SILENT can be used in an unmarked D

of (A4), but i t w i l l be shown la t e r that for other reasons i t cannot be

regarded as a basic verbum cessandi.

6. Another verb ia l , TO SAY NOTHING (or NOT SAY ANYTHING), has two ad­

vantages over TO BE SILENT when i t comes to deciding which term for the

absence of speech i s the most general . F ir s t , i t excludes acts o f type

(A2) except in strongly hyperbolic speech; hence i t i s a more absolute

notion than TO BE SILENT. The r e la t iv e character of TO BE SILENT i s easy

to understand i f we r ea l iz e that in one of i t s senses SILENT i s synonymous

91

with TACITURN, which i s predicated of a person who i s temperamentally d i s ­

inclined to ta lk (though, o f course, he talks whenever talking i s inev i ­

tab le) .

Second, TO SAY NOTHING covers more acts of type (A3) than TO BE SILENT.

The res tr ic t ion to descriptions conforming to (Cd2), as well as the poss i ­

b i l i t y of extending i t s use to descriptions sa t is fy ing (Cdl) by adding the

preposition ABOUT, are ident ica l . But imagine two ruffians torturing Mr.

Smith to force some secrets out of him. I f one of those brutes returns

a fter having gone out for a while and asks the other one "Did he say any­

thing ?", the answer could be (D3).

(D3) No, he didn't (say anything); he was ju s t begging for mercy

In (D3) "he didn't (say anything)" cannot possibly be replaced by "he was

silent"because i t would be incompatible with "he was ju s t begging for

mercy". Thus TO BE SILENT can only be used in D describing (A3) and con­

forming to (Cd2) i f condition (Cl) i s s a t i s f i e d by (A3). TO SAY NOTHING

on the other hand, i s applicable no matter whether (Cl) or (C2) obtains.

(Cl) Sa utters no words at al l

(C2) Sa utters only words unrelated to the tacendum in question

(Notice the ident i ty between (Cl) as a property of (A3) and my previous

formulation of (A l) . ) Here TO BE SILENT turns out to be a more absolute

notion than TO SAY NOTHING. Is th is not in contradiction with the d i f f e r ­

ence in app l icab i l i ty of the two verbials to (A2) ? The p o s s ib i l i t y of

using TO SAY NOTHING in D describing (A3) and conforming to (Cd2) no

matter whether (Cl) or (C2) i s s a t i s f i e d by (A3) i s due to the ambiguity

92

of TO SAY SOMETHING between 'making an utterance1 and 'stat ing something'.

Its meaning can be even more spec i f ic : 'stating something about a sp e c i f ic

subject ' . Thus, for the ruff ian's purposes

(D4) He did not say anything

would s t i l l be accurate i f Mr. Smith had been making a l l kinds of s ta te ­

ments such as

(A5) The weather i s gorgeous today

or

(A6) I love peanut butter and j e l l y sandwiches

Sentence (Dl) could not possibly replace {D4) with reference to the same

acts . However, i t might be possible to use (D5).

{D5) He kept s i l e n t

But ju s t as TO FALL SILENT, the verbial TO KEEP SILENT is no basic verbum

taeendi. Once more, th is w i l l be c la r i f i e d la ter .

7. The negations of TO SPEAK and TO TALK cover (Al) , and also (A3) in

conformity with (Cd2) whether (A3) s a t i s f i e s (Cl) or (C2). Notice that ,

unlike TO SAY SOMETHING, the verbs TO SPEAK and TO TALK are not ambiguous

between 'making an utterance' and 'stat ing something'. Yet they ore am­

biguous between 'making an utterance' and 's tat ing something about a

sp e c i f i c subject ' . That i s why

93

{D6) He did not speak (ta lk)

can replace (D4) in the tor tu re -s i tu a t ion sketched.

NOT SPEAK and NOT TALK pose an additional problem because o f the am­

biguity of TO SPEAK and TO TALK between 'to ut ter words', 'to conduct

a conversation', and ' to de l iver a speech, give a t a l k ' . The second

and third meanings explain why i t i s possib le to say (D7) and (D8), res­

p ec t ive ly .

(D7) During the l a s t f i v e years o f his l i f e he never spoke (talked)

to anyone; only occasionally did he shout a greeting from in­

side his car

(D8) Because he did not speak ( ta lk ) at the conference he had some

time for a nice private conversation

Thus NOT SPEAK and NOT TALK cover (A7) and (A8), neither of which can be

included in any o f the major types of s i l en ce acts distinguished so far:

here the tacens abstains from engaging in a particular communication type.

(A7) Sa does not conduct a conversation

(A8) Sa does not d e l iver a speech (give a ta lk)

None of the other verba tacendi discussed so far can be used in any way

sim ilar to {D7) and (D8).

Observe, f i n a l l y , that adding ABOUT to NOT SPEAK and NOT TALK r e s t r i c t s

the ir use to descriptions o f (A3) conforming to (Cdl) , In th i s respect they

are s imilar to both TO BE SILENT and TO SAY NOTHING.

94

8. Whereas English only possesses complex l e x ic a l i z a t io n s to express some­

body's being s i l e n t in a more or l e s s neutral way, many languages have

s im p le l e x i c a l i z a t i o n s at th e ir disposal: Dutch ZWIJGEN, German SCHWEIGEN,

French SE TAIRE, Latin TACERE and SILERE, Greek tfiwiWOand e t c .

As announced, I only concentrate on the contrast between Dutch and

English. ZWIJGEN i s not only a simple l e x i c a l i z a t i o n , i t s meaning i s also

move g e n e ra l than any of the English equivalents: i t covers the three

major areas o f the conceptual domain o f the verba tacend i , i . e . the ab­

sence o f speech ac such, the absence o f speech about a certain topic as in

(D9) Hij kon het n ie t 1 anger ZWIJGEN

(He couldn't KEEP SILENT ABOUT i t any longer)

and the act o f discontinuing one's discourse. English simply lacks such

a general term. Moreover, unlike TO BE SILENT, ZWIJGEN does not cover

(A2), which I regard as outside the proper scope of the verba tacendi.

NIET SPREKEN {'not speak ') , NIET PRATEN ( 'not t a l k 1) and NIETS ZEGGEN

( ' t o say nothing') have the same scope as th e ir English counterparts.

The meaning o f these three v e rb ia l s , as well as the meaning o f ZWIJGEN,

can be r e s tr ic t e d , as with the English v e r b ia l s , to descriptions of (A3)

conforming to (Cdl) by adding the preposit ion OVER. For the description

of (A3) conforming to (Cdl) Dutch provides another simple l e x i c a l i z a t i o n ,

namely VERZWIJGEN as in (DIO).

(DIO) Hij VERZWEEG z i jn communistisch verleden

(He KEPT SILENT ABOUT his communist past)

9. The d if ferences between the basic verba tacendi in English and in

Dutch are represented in SDC-Table 1. A str ik ing fa c t which emerges from

DUTCH SEMANTIC DIMENSION ENGLISH

zwijgen

niet spre- ken, niet praten

niet spreken niet praten

m etszeggen

to be s i len tto say

zwijgen oververzwijgenetc .

to be s i l e n t about, etc .

A4 f-Cd3

....................................................................... / -C d 4 jJJ to be s i l e n t

notspeak,nottalk

not speak not talk

SDC-Table 1

96

the table i s that for all the verbials which have (Al) as part of their

meaning, (Al) i s at the same time the focus of their meaning. The asymme­

tr i e s between Dutch and English are no doubt clear.

How can Al-3-4-7-8 be regarded as a semantic dimension of the verba ta­

cendi ? I said earl ier that due to the marginality of si lence in the frame

of l in g u is t ic action, the lexical frame associated with i t i s bound to

re f lec t the whole spectrum of l in gu is t ic action i t s e l f . Thus there i s an

important language dim ension to the verba tacendi. Since the verbials dis­

cussed so far were all said to be somehow basic in the lexical frame of

s i l en ce , (Al), (A3) and (A4) — and maybe also (A7) and (A8) — can be

regarded as focal points in that language dimension of the verba tacendi:

the simple absence of language, the absence of speech about a particular

topic (which i s related to the meaning aspect of l in g u is t ic act ion) , the

absence of speech in the c o n te x t of surrounding utterances, and the absence

of utterances belonging to a particular communication ty p e .

Though the importance of (Al) , (A3) and (A4) might have been realized

by any theoretician studying acts of being s i l e n t , the importance of (A7)

and (A8) would probably have been overlooked. Indeed, i t appears arbitrary

to single out the communication type within the realm of the different

aspects of l in g u i s t i c action, and within the realm of the different commu­

nication types i t seems even more arbitrary to s ingle out the conversation

type and the 'speech'. Yet (A7) and (A8) occupy a sa l i en t place in our

western experience of l in g u i s t i c action: they are our prototypes of dialogue

and monologue, respectively. I t v/ould not cause too much surprise to find

a highly ritual culture or subculture in which the equivalent of NOT SPEAK

indicated the non-performance of a particular r itual .

97

10. The previous paragraphs provide us with an additional argument for not

trying to make a taxonomy of verba tacendi or, for that matter, of l i n ­

g u i s t i c action verbials in general. No doubt the boundaries between the

verba s i l e n d i , verba reticendi and verba cessandi are fuzzy. A high per­

centage of the verbials to be invest igated belong to more than one of

these categories. Indeed, the meaning of the few basic verba tacendi

discussed so far was always spread out over more than one of the three

types and often extended beyond the scope of all three. In some cases one

could claim that the verbials in question have several d i f ferent meanings.

But there are obvious cases of mixed meanings. A typical example is the

Dutch AFSTAPPEN VAN { l i t . ' to step down from fa t o p ic ] ' ) which indicates

at once s i len ce about a particular topic and the discontinuation o f speech

about that topic (as well as the taking up of another to p ic ) .

Moreover, a lso the bounadry between the verba tacendi and the verbials

describing acts of putting someone to s i lence i s vague; consider in th is

respect the verbs TO CENSOR and TO HUSH UP. Furthermore, there are probably

borderline cases between the verba tacendi and the verbials of lying to be

discussed in the next chapter: there may be occasions on which acts of

keeping back information ( i . e . the domain of the verba reticendi) should

be considered acts of lying.

D. THE CODES OF SILENCE

11. The reader's attention should be drawn to the fact that all the verbials

in the foregoing paragraphs e x p l i c i t l y concentrate on the absence of speech

as an oral manifestation of language, and not on the absence of l in g u i s t i c

98

behavior as such. With respect to oral versus written communication we

can distinguish (Al) and (A2).

(Al) Sa abstains from speaking

(A2) Sa abstains from writing

Though al l the basic verba tacendi represented in SDC-Table 1 have (Al)

as the focus of their meaning, some — i f not most — of them cam be

applied metaphorically to (A2) as in (Dl).

(Dl) The novelist kept s i l e n t for almost a decade before publishing

But this application i s metaphorical. The only basic verba tacendi center­

ing on and restricted to (A2) are the negations of English TO WRITE and

Dutch SCHRIJVEN. With respect to the codes of s i lence we get the quite

symmetrical situation depicted in SDC-Table 2.

his la s t masterpiece

DUTCH SEMANTIC DIMENSION ENGLISH

• zwijgen to be s i l e n t #

zwijgen to be s i l e n t

vrniet schrijven £ A2J not write

SDC-Table 2

12. With respect to the dist inct ion between the absence of spoken versus

written communication, the verba tacendi might be expected to show f ive

different semantic patterns.

99

(SP1) The meaning i s res tr ic te d to (Al)

(SP2) The meaning i s res tr ic te d to (A2)

(SP3) The focus of the meaning i s (A l) , but i t can be extended to (A2)

(SP4) The focus of the meaning i s (A2), but i t can be extended to (Al)

(SP5) The meaning includes (Al) and (A2) on an equal basis

These patterns are depicted in SDC-Table 3.

SP5

DUTCH

SP4

SEMANTIC DIMENSION

SP1 £ Al J SP1 f

V SP2 A2 SP2 Y

ENGLISH

SP4

SDC-Table 3

How are the verba tacendi d is tr ibuted across these patterns ?

13. An overview o f the verba tacendi shows that the numerically best re­

presented pattern is (SP5). The group inc ludes, amongst many other v erb ia ls ,

TO BURY, TO CEASE, TO COME TO AN END, TO CONCEAL, TO END, NOT GIVE AWAY,

TO KEEP BACK, TO KEEP UNDER ONE'S HAT, TO KEEP SECRET, TO MAKE NO SIGN,

TO SECRETE, TO STOP, TO VEIL. Notice that in th i s group there are only

verba reticendi and verba cessand i , with the poss ib le exception o f TO MAKE

NO SIGN.

Also on Dutch the majority o f verba tacendi show pattern (SP5). Some

examples: AFSTAPPEN VAN ( l i t . ' to step down from'), EINDIGEN ( ' t o e nd ') ,

GEHEIMHOUDEN { ' t o keep s e c r e t ' ) , ACHTERHOUDEN ( ' t o keep back1}, e tc .

100

Also in Dutch the l i s t only includes c lear cases of verba reticendi and

verba cessandi. I called TO MAKE NO SIGN a p o ss ib le exception to th i s re­

g u lar i ty . Consider i t s use in (D2) and i t s Dutch equivalent NIET VAN ZICH

LATEN HOREN ( l i t . 'not l e t [[somebody ̂ hear about o n e se l f ' ) in (D3).

(D2) I don't know his whereabouts; he's made no sign during the past

three years

(D3) Hij heeft al drie jaar n ie t meer van zich laten horen

( l i t . ’he has not l e t Qus] hear about himself for three years

already')

The context in (D2), as well as the r e f l e x iv e VAN ZICH {'about o n e se l f ' )

in the Dutch verb ia l , show that i t maybe n e c e s s a r y to interpret the phrases

as indicating the absence o f (oral or written) communication about onese l f

rather than the absence of communication as such. This hypothesis may be

further supported by the rough equivalence o f (D2) with (D4) and o f (D3)

with (D5).

(D4) I don't know his whereabouts; he's given no sign of l i f e during

the past three years

(D5) Hij heeft al drie jaar geen teken van leven meer gegeven

(He has given no sign of l i f e for three years already)

But the rough equivalence o f TO MAKE NO SIGN and NIET VAN ZICH LATEN HOREN

with TO GIVE NO SIGN OF LIFE and GEEN TEKEN VAN LEVEN GEVEN could a lso be

adduced as evidence to disprove the hypothesis: one might regard an act

of 'giving no sign of l i f e ' as an act of refraining from communicating as

such so that'hearer'cannot in fer the ta ce n s1 being al ive* rather than as

101

an act o f not communicating about onese l f (or one's being a l i v e ) . Hence

the four verbials in question are to be situated somewhere along the border­

l in e between verba s i lend i and verba re t icend i . That i s , we have one more

argument for not y ie ld ing to the temptation to make hasty taxonomic

judgments.

But why the nearly absolute r e s tr ic t io n o f pattern ( SP5) to verba r e t i ­

cendi and verba cessandi in both languages ? Why are the verba s i lend i not

normally ne i tra l with respect to the oral versus the written code of s i l e n c e ?

We are not y e t in a pos it ion to venture upon a generaliz ing explanation

— and we won't be within the scope of t h i s essay. More comparative work

i s needed f i r s t in order to discover the u n iversa l i ty or non-universal ity

of the phenomenon. Only then w i l l i t be poss ib le to decide where an ex­

planation has to be found: in the nature o f language as such, or in the

cultural t r a i t s o f part icu lar l i n g u i s t i c communities.

14. Next on the sca le of the number of representatives in the lexicon

comes pattern (SP1). Some examples: NOT BREATHE A WORD, TO BUTTON ONE'S

LIP, TO CLOSE ONE'S MOUTH, TO HOLD ONE'S TONGUE, TO KEEP QUIET, NOT SAY

'BOO', TO KNOCK IT OFF, TO SHUT ONE'S BAZOO, NOT UTTER A WORD. Dutch:

BOE NOCH BA ZEGGEN ( l i t . 'to say neither boo nor b a ' ) , GEEN BEK/MOND

0PEND0EN ( l i t . 'not open a beak/mouth'), ZIJN BAKKES HOUDEN ( l i t . 'to

hold one's t r a p ' ) , e tc . Two remarks ought to be made.

F ir s t , there i s a c lear correlat ion between the formation of most (SP1)

verbials and the r e s tr ic t io n o f the ir meaning to (Al) . Most of them con­

ta in an e x p l i c i t reference to part of the oral communication process ( e .g .

breathing, opening one's mouth, uttering sounds) or am instrument of oral

102

communication (e .g . l i p s , tongue, mouth, e t c . ) .

Second, the (SP1) verbia ls form the only s e t o f verba tacendi which con­

ta ins many slang expressions such as TO KNOCK IT OFF, TO SHUT ONE'S BAZOO,

e tc . Slang a lso occurs in verbials the meaning o f which i s patterned

d i f f e r e n t ly , but much l e s s frequently. Hence there i s an in terest ing

correlat ion between the oral code o f the acts o f s i l en ce to be described

and the informal, slangy code o f the appropriate describing ac ts . But

again, an explanation w i l l have to be based on further comparative data

and on the observation of the ex is tence or non-existence o f a comparable

correlat ion with respect to non-si lence aspects of l in g u i s t i c action.

15. Pattern (SP3) comes next. I ts representatives in the lexicon c o n st i ­

tute the only remaining large group of verba tacendi. An example i s the

basic verbum tacendi TO BE SILENT and i t s Dutch equivalent ZWIJGEN, as

wil l have appeared from SDC-Table 2. I shall d iscuss some other examples

in contrast with c lo s e ly related (SP1) v erb ia ls . Consider the following

two columns of verb ia ls .

to f a l l s i l e n t to be s i l e n t

to stand mute to be mute

to button up

not u t ter a word

to keep buttoned up

not say a word

not say a wordnot say ‘boo

to shut one's face/mouth/head— to shut up

to hem and haw — to beat around the bush

If I am not mistaken, the l e f t column contains (SP1) verbia ls whereas

those in the r ight column a l l conform to (SP3). The meaning o f TO BE SILENT

has already been shown to be extendable to (A2). No such extension seems

103

to be possible for TO FALL SILENT the typical use of which i s i l lu s tr a te d

in (D6).

(D6) The p o l i t i c ia n made a desparate attempt to answer the press

agent's question, but he soon f e l l s i l e n t

TO BE MUTE usually refers to (A l ) , but in (D7), uttered in response to

"I know you wrote to him about the i r r e g u la r i t i e s you discovered, but

did you get a reply ?“ , the act described may be of the (A2) type.

(D7) No, he i s s t i l l mute on the subject

An (A2) interpretation of TO STAND MUTE as used in (D8) i s not poss ib le .

(D8) He stood mute upon hearing about his superior's resignation

The next verb ia l , TO KEEP BUTTONED UP can be used with reference to the

absence of written communication, as in (D9) which i s a reply to "Did he

answer your l e t t e r ?".

(D9) Yes, but he keeps buttoned up about the i r r e g u la r i t i e s I d i s ­

covered

TO BUTTON UP, on the other hand, i s r es tr ic ted to (Al) . The (SP3) pattern

of NOT SAY A WORD and the (SP1) pattern of the corresponding verbials NOT

UTTER A WORD and NOT SAY 'BOO' w i l l be c lear from the fa c t that the l a t t e r

two cannot replace the former in (DIO).

(DIO) In his l e t t e r he did not say a word about the i r r e g u la r i t i e s

I discovered

Also TO SHUT UP can be extended to (A2) as in (D l l ) .

104

(D ll) I want th i s l e t t e r to be mailed today, so I must shut up in

a minute

Similar extensions are not possib le for TO SHUT ONE'S FACE/MOUTH/HEAD.

Final ly , one can BEAT AROUND THE BUSH both ora l ly and in writ ing , but

one can only hem and haw o r a l ly , though both verbia ls designate the same

kind of s i l en ce about a particular topic brought about by talking about

vaguely related but irrelevant things.

The interaction of three quite d i s t i n c t parameters seems to be res­

ponsible for the dif ferences in semantic pattern shown in the above pairs

of verb ia ls . F ir s t , there i s the degree o f im m ediacy of the l i n g u i s t i c

a c t io n to be described. The immediacy o f oral communication as opposed

to the delayed interact ion mediated by writing might explain why TO FALL

SILENT, which implies the abrupt character o f the s i l en ce described, i s

not normally used to describe ' s i le n c e in i t s written form'; there are ex­

ceptional cases such as "After writing one novel a f te r the other for more

than th ir ty years he suddenly f e l l s i l e n t , in which the abruptness i s pre­

served. A second factor involved i s the fo rm a tio n of the v e r b ia l s (see

also the previous paragraph). NOT UTTER A WORD and NOT SAY 'BOO' carry an

e x p l i c i t reference to the production of sounds, j u s t as the im ita t ive TO

HEM AND HAW, whereas TO SHUT ONE'S FACE/MOUTH/HEAD and TO STAND MUTE focus

on the instruments o f oral communication. Therefore, th e ir being res tr ic te d

to (Al) i s not surprising. The same factor may explain why TO BE SILENT,

TO BE MUTE and NOT SAY A WORD are in the f i r s t place concerned with the

absence of speech instead of being neutral as to the (Al) or (A2) type of

s i l en ce . Third, the degree of f o r m a l i ty of the d e s c r ip t io n i s involved.

105

TO BUTTON UP, NOT SAY 'BOO', TO SHUT ONE'S FACE/MOUTH/HEAD and TO HEM AND

HAW are a l l slangy. Also the relation between slangy verba tacendi and the

oral code of acts of s i lence has already been observed in the previous para­

graph. The restr ic t ion of the four verbials enumerated to (Al) i s ent ire ly

in keeping with i t . Moreover, the same correlation may explain why the

pretty informal expressions TO KEEP BUTTONED UP, TO SHUT UP and TO BEAT

AROUND THE BUSH are in the f i r s t place concerned with the absence of speech

rather tahn being neutral as to the (Al) or (A2) type of s i l e n c e . This

follow-up remark, as well as the one attached to the comments on verbial

formation, indicates that none of the parameters discussed i s s e l f - s u f f i ­

c ient to assign a particular semantic pattern to a verbial; rather,

an in tr ica te intercation of these seems to be responsible. Again, a spe­

c i f i c a t io n of this interaction will require a l o t of further research. Quite

similar arguments could be made for the equivalent Dutch verbials discussed

in this paragraph.

16. Pattern (SP2) i s extremely sparsely populated. We have already men­

tioned NOT WRITE, which i s the only basic verbial belonging here. I had to

stretch my imagination to find a second and, admittedly, marginal example,

namely TO DISCONTINUE in the sense of 'stopping the publication of' as in

(D12). (But in pract ica l ly the same sense one can talk about discontinuing

a ser ie s of radio ta lk s . )

(D12) The publishing company discontinued the unprofitable journal

One might be tempted to include verbials such as TO CLASSIFY and TO FILE

AND FORGET in the l i s t of (SP2) verba tacendi, but they do not belong here

106

at a l l : they do not indicate the absence o f written communication, but

the use of writing as a medium to secrete information.

In Dutch there i s only NIET SCHRIJVEN ( 'not w r i t e 1) . Including the

Dutch equivalent to TO DISCONTINUE, namely DE PUBLIKATIE STOPZETTEN VAN

( ' s top the publication o f ) requires an even stronger stretching of the

amigination because i t s emantic unity i s so weak that I would not be

inclined to regard i t as a lex ica l item (however complex).

17. F in a l ly , pattern (SP4) i s , as far as I can se e , completely empty,

both in English and in Dutch. In th is respect another in terest ing question

a r ise s . How i s i t that there i s such an asymmetrical relat ionship between

(Al) and (A2) that basic (Al) verbials such as TO BE SILENT can be ex­

tended to (A2), whereas the reverse i s not true ? I am not going to

answer the question. Some readers may be l ieve that i t would be absurd

even to assume that there might be a language in which the meaning of

NOT WRITE would get extended to (Al) . Yet there i s no log ica l necess i ty

in the r e s tr i c t io n o f NOT WRITE to (A2), as appears from the fa c t that

words describing the reception and reproduction poles of written communi­

cat ion , such as TO READ and TO COPY, do get extended to the reception of

oral language (transmitted by radio) as in "Do you read me ?" and "I copy

you".

18. In connection with (SP2) we hinted at a t o t a l l y d i f fe re n t aspect of

the code of some acts of being s i l e n t . Some verba reticendi describe acts

of withholding information or of being s i l e n t about a part icu lar topic

107

which may be e s s e n t i a l ly brought about by l in g u i s t i c means. I am thinking

of verbials such as TO CONCEAL, TO COVER UP, TO HIDE, TO KEEP UNDER WRAPS,

TO PUT THE LID ON, TO VEIL, and two mentioned already, namely TO CLASSIFY

and TO FILE AND FORGET. I f language i s used to hide information, that

language has to appear e i th er or in a written form as w e l l . Most o f the

verbials in question are neutral as to the code o f the secreting language.

Only TO CLASSIFY and TO FILE AND FORGET focus o f the written code, though

they may be metaphorically extended. Surprising enough, I have not been

able to find any which focus s p e c i f i c a l l y on the oral code o f communica­

t io n . So here the relat ionship between the two codes i s reversed. Again

the s i tuat ion in Dutch seems to be completely analogous.

E. THE SOUND OF SILENCE

19. Absolute concepts are not suscept ib le to comparison or gradation.

Thus, i f two people are absent from a meeting i t makes no sense to say

that the one i s more absent than the other or /that both are very absent.

Yet the absence of speech, ju s t as speech i t s e l f , can be perceived as

more or l e s s intense or ' loud' . Moreover, as to speaking, there i s an

e s s e n t i a l ly temporal dimension to not speaking. Therefore i t should not

surprise us that most o f the expressions used to describe i t contain some

indication of duration. I n t e n s i t y and d u r a tio n are what I ca l l the two

sound-related dimesions of s i l e n c e . Their relevance to an understanding

of the verba tacendi w i l l be discussed in th i s s ec t ion .

108

20. In theory, i t should be possible to distinguish between acts of being

s i l e n t with a high, average and low in tens i ty , and between long, average

and short ones. The following semantic patterns of the verba tacendi might

be expected to correspond to these degrees. They are represented in SDC-

Tables 4 and 5.

(SP1) The verbial describes a highly intense s i lence act

(SP2) The verbial describes a si lence act of average intensity

(SP3) The verbial describes a s i lence act of low intensity

(SP4) The verbial indicates a long si lence

(SP5) The verbial indicates a s i lence act with average length

(SP6) The verbial indicates a short si lence

However, only one of these degrees of comparison, the superlative, i s

reflected in the verba tacendi. In other words, patterns (SP2), (SP3),

(SP5) and (SP6) in SDC-Tables 4 and 5 characterize empty c lasses . All

verbials which do not indicate highly intense s i lence acts are completely

neutral with respect to in tens i ty , i . e . they conform to (SP7).

(SP7) The verbial i s applicable to a s i lence act with any degree

of intensity

And those which do not indicate a long duration are neutral as to the

temporal dimension, i . e . they conform to (SP8).

(SP8) The verbial i s applicable to a s i lence act with any duration

S tr ic t ly speaking, the neutral (SP7) and (SP8) verbials are outside the

scope of the two sound-realted dimensions of s i lence . Therefore, only the

109

(SP1) and (SP4) verb ia ls , i . e . those verba tacendi emphasizing the in­

t e n s i ty and duration of the acts of being s i l e n t , wil l be reviewed in

the following paragraphs.

DUTCH

SPI [

SPP [

SP3 [

SEMANTIC DIMENSION ----------

high in tens i ty SPI

average in tens i ty j SP2

low in tens i ty j SP3

ENGLISH

SDC-Table 4

DUTCH SEMANTIC DIMENSION ENGLISH

Sp4 [ long ]SP 4

SP8 SP5 £ average J SP5 SP8

SP6 [ short ] SP6

SDC-Table 5

21. Some of the most common verbials for describing extreme cases o f being

s i l e n t contain a reference to death and the symbols o f 'c losedness1. In

English the ve feven a e to d ea th i s present in TO MAINTAIN A DEATHLIKE/TOMB­

LIKE SILENCE and the the metaphorical verbum reticendi TO BURY. Dutch,

however, does not only possess the d irect equivalents of these two (EEN

D00DSE STILTE BEWAREN and BEGRAVEN), but in addition there are Z0 GESLOTEN

ZIJN ALS EEN GRAF ( l i t . 'to be as closed a a grave ') , ZWIJGEN ALS VERM00RD

( l i t . 'to be s i l e n t as i f [one had been]murdered') , EEN GEHEIM MEE ONDER

110

DE AARDE/IN HET GRAF NEMEN ( l i t . 'to take a secret under the earth/into

the grave1) , and IETS DOODZWIJGEN ( l i t . 'to be s i l e n t about something

until i t i s dead ' ) . In a l l of these the death referred to i s purely

metaphorical, except in EEN GEHEIM MEE IN HET GRAF NEMEN. One can only

say

(Dl) Hij neemt heel wat geheimen mee in het graf

(He takes quite a few secrets with him into the grave)

i f the person talked about has died. This again shows the marginality of the

verba tacendi: here we are confronted with what i s probably the only example

of a l in g u i s t i c action verbial which can be predicated of a dceased per­

son. Notice that the l i t e r a l equivalents of ZWIJGEN ALS HET GRAF, namely

TO BE SILENT AS THE GRAVE and TO BE SILENT AS THE TOMB, as predicated o f a

conversational tacens are not acceptable to many American speakers of

English. Whereas the Dutch sentence

(D2) Hij zwijgt als het graf

( l i t . 'he i s s i l e n t as the grave')

i s not at a l l uncommon, the English l i t e r a l equivalents of the Dutch verbial

are mainly used to describe s i lence as such as in "It 's as s i l e n t as a

tomb in here", though I remember reading the following sentence in a British

novel : "During the taxi drive she remained as s i l e n t as the tomb".

22. Also with respect to the sym bols o f 'c lo s e d n e s s r there are a couple

of differences between English and Dutch. The most important English verb­

i a l s in question are TO BE/BECOME MUM AS AN OYSTER, TO CLOSE UP LIKE A

I l l

CLAM/AN OYSTER, and TO CLAM UP. In Dutch there i s the comparable expression

ZO GESLOTEN ZIJN ALS EEN OESTER ( l i t . 'to be as closed as an o y s t e r ' ) , but

clams do not enter the picture and they are not even replaced by mussels

which are more common in the Dutch-speaking region o f the world (maybe be­

cause the use of the word for mussel as a symbol for a weakling — for

which one uses a j e l l y f i s h in English — i s too strong a convention). How­

ever, Dutch extends i t s symbols well beyond the world of bivalve mollusks:

ZO DICHT ZIJN ALS EEN POT/BRIEF ( l i t . 1 to be as closed as a p o t / l e t t e r ' ) ,

POTDICHT ZIJN ( l i t . 'to be p o t -c lo se d ' ) , ZO GESLOTEN ZIJN ALS EEN POT/PE-

PERDOOS/BRANDKAST ( l i t . ' to be as closed as a pot/pepper-box/safe1).

F inally there i s the l e s s transparent image ZWIJGEN ALS EEN MOF ( l i t . 'to

be a s i l e n t as a German'); MOF i s a quite contemptuous term for 'German'

(comparable to the American KRAUT), but i t i s hard to see an essent ia l

re la t ionsh ip between ta c i tu rn i ty and a German (whether he i s a contemptible

one or n o t ' ).

23. TO BE SILENT AS A POST and TO BE SILENT AS A STONE, in which the

in te n s i ty of s i l en ce i s emphasized by the refrence to t y p ic a l ly inanimate

and therefore s i l e n t objects do not f ind , contrary to what one could

have expected, any counterparts in Dutch.

24. So far we have presented an overview of those verba tacendi which

derive the expression o f in te n s i ty from the symbolic re lat ionship be­

tween s i l en ce and death, 'c losed' and inanimate objects . A second im­

portant se r ie s expresses highly intense s i l en ce by referring to the

112

absence of basic speaking processes or the inact ivat ion o f certain organs

of speech. The absen ce o f b a s ic sp ea k in g p r o c e s s e s wil l be c lear in most

of the examples given in Table 1. The processes referred to are the pro­

duction of words, as in NOT BREATHE A WORD, and the production of sounds

as in MOT LET OUT A PEEP, with somewhere in between NOT SAY 'BOO'. In th i s

respect there i s a complete paralle l ism between English and Dutch. The

underlying rationale i s no doubt that i f h igher-level processes would be

referred to the in te n s i ty o f the s i l e n c e could not possibly be highlighted:

descriptions such as “During the cocktail party, Smith did not ut ter a

s in g le f u l l sentence" or "He was not able to susta in a prolonged conver­

sation" imply that there was not a complete absence of speech on Smith's

part.

Closely related verbia ls which were not entered into the tab le are the

following: TO FORSWEAR SPEAKING/SPEECH, NOT HAVE A WORD TO SAY, GEEN

WOORDEN VOOR IETS HEBBEN ( l i t . ' n o t have any words for something1) , HET

EERSTE WOORD OVER IETS NOG MOETEN ZEGGEN ( l i t . ' s t i l l have to say the

f i r s t word about something') and the quite colorful DE WOORDEN BLEVEN HEM

IN DE KEEL STEKEN ( l i t . 'the words remained stuck in his throat ' ) and

HET WOORD STIERF OP ZIJN LIPPEN ( l i t . 'the word died on his l i p s ' ) .

113

------------------- ENGLISH ---------------not breathe a word

not l e t a word escape one

not u t ter /sa y a word

DUTCHgeen woord over z ijn lippen laten komen

( l i t . 'not l e t a word pass through one's 1i p s ')

geen woord lossen(over i e t s )( l i t . 'not l e t loose a word (about some­th in g’ )

geen (stom) woord (meer) zeggen ( l i t . 1 not say a (dumb) word (any more)

not say 'boo' boe noch ba zeggen ( l i t . 'to say neither boo nor ba')

not l e t out a peep geen geluid uitbrengen ( l i t . 'not produce a sound')

geen piep (meer) geven/laten ( l i t . ' not g i v e / l e t out a peep (any more)

geen kik geven/laten ( l i t . 'not g i v e / l e t out a sound'; KIK i s a word for a very f a in t no ise , which i s only used in expressions such as these)

kik noch mik geven ( l i t . ' to give neither KIK nor MIK'; MIK i s semantically completely empty; i t s only function i s to in te n s i fy the a l ­ready intense GEEN KIK GEVEN)

n ie t kikken (van i e t s )( l i t . 'not produce any sounds (about some­thing ')

T able 1

25. On the other hand, the in a c t iv a t- to n o f sp eech o rgan s (the fa v or i te

ones being the l i p s , the tongue and the mouth as a whole) i s involved

i s a l l the examples given in Table 2. Again the paral le l ism between Eng­

l i s h and Dutch (except for TO MUZZLE ONESELF), which has no d irec t

114

counterpart) i s s tr ik ing , though not surprising. About the only difference

i s that the button-and-zip-symbolism i s missing i s Dutch.

------------- ENGLISH ----------------------

to seal one's l ips

to button one's l ip

to zip one's l ip

to bite one's l ip s

DUTCH

de lippen op elkaar drukken/klemmen ( l i t . ' to press the l ip s together')

zich op de lippen bijten ( l i t . ' to b ite one's l i p s ' )

to bite one's tongue

to put a bridle on one'stongue

to keep one's tongue incheck

op z i jn tong bijten ( ' t o b i te one's tongue')

z ijn tong in bedwang/toom houden ( ' t o keep one's tongue in check/ bridled')

not open one's mouth

to keep one's trap/yap shut

to shut one's bazoo/face/head

geen mond/bek opendoen ( l i t . 'not open a mouth/trap')

zijn mond/bek/bakkes niet opendoen ( 'not open one's mouth/trap/yap')

to muzzl.e oneself

T able 2

In contrast to the verbials in Table 2 , there are also some which in­

volve an unintentional inactivation of some organ of speech or the organs

of speech altogether. Surprisingly, however, one o f these , TO HAVE LOST

ONE'S TONGUE and the Dutch translat ion ZIJN TONG VERLOREN HEBBEN, i s used

ty p ica l ly in description such as (D3) and (D4) which are necessari ly ironical

115

and imply that the s i l en ce i s quite voluntary.

(D3) He must have l o s t his tongue

(D4) Hij heeft zeker z ijn tong verloren

The others , which do not carry such implications at a l l , are: TO STAND

MUTE, MET VERSTIJFDE TONG STAAN ( l i t . ' to stand with a s t i f f e n e d tongue') ,

VERSTOMD STAAN VAN ( l i t . ' t o stand struck dumb because o f ' ) , SPRAKELOOS

STAAN ( l i t . ' to be struck with dumbness'). A picturesque Dutch verbial

which i s in the same l in e o f business i s MET DE MOND VOL TANDEN STAAN

( l i t . 'to stand with the mouth fu l l of t e e t h ' ) : having teeth everywhere

in one's mouth i s a rather farfetched but unquestionable e f f e c t i v e image

for the inact ivat ion of the organs of speech and thus for being s i l e n t .

26. We are s t i l l l e f t with a miscellaneous s e t of verbials describing

intense acts of being s i l e n t , which do not f i t the schemes presented in

the previous f iv e paragraphs. F ir s t , there are some verba cessandi: TO

BUTTON UP, TO CLOSE UP, TO DRY UP, TO DUMMY UP, TO FALL SILENT and the

Dutch DICHTKLAPPEN ( l i t . ' to smack c lo sed ') and STILVALLEN ( ' t o f a l l

s i l e n t ' ) . In the case o f verba cessandi in te n s i ty i s often associated

with the sudden character o f the discontinuation o f speech. I t i s be­

cause o f the in te n s i ty of the s i len ce act described by TO FALL SILENT

that we did not want to regard the verbial as a basic verbum tacendi.

Further, there are several verba reticendi such as NEVER LET ON, TO

REFUSE COMMENT, TO REPRESS, TO SMOTHER, TO STIFLE, TO SUPPRESS and the

Dutch NIETS LOSLATEN ( l i t . 'not l e t anything l o o s e ' ) , OPKROPPEN ( l i t .

116

to p i l e up in one's g izz a rd ' ) , and EEN DIEP/GROOT GEHEIM VAN IETS MAKEN

( l i t . 'to make something into a deep/big s e c r e t ' ) .

F in a l ly , there are a couple o f additional Dutch verba s i l en d i : IN ALLE/

ZEVEN TALEN ZWIJGEN ( l i t . 'to be s i l e n t in a l l / s e v en languages') , and the

l e s s common GEEN SLAG AAN DE BAK KRIJGEN (the BAK referred to i s a wooden

container which was used to get hot food for the sa i lo r s from the ship's

kitchen; the expression means 'not be able to reach the BAK' because of

the pushing of the others so that one does not obtain food; i t s f igurat ive

meaning i s 'not be able to say anything because of the excess ive talking

o f o th e rs1).

27. The in te n s i ty expressed by the verbials that we have ju s t discussed

i s not merely a matter of in te n s i ty . I t can be related to d i f feren t aspects

o f being s i l e n t : the duration of the s i l e n c e , the obstinacy of the person

who keeps s i l e n t , the reason for being s i l e n t , and the meaning with which

the s i l en ce i s imbued. E xp lic i t l inks wil l not always be made when we come

to the invest igat ion of these other dimensions, but l e t us not forge t that

such in terre la t ionsh ips e x i s t .

28. The second sound-related dimension to be dealt with i s the duration

of the s i l e n c e . I t should be noted at once that acts of being s i l e n t

always have some duration. But whereas one can say "He was s i l e n t for

a second", i t would be pretty weird to t e l l "He did not utter a word for

a second", "He did not open his mouth for a minute'1 or even "He sealed

his l ip s for f i v e minutes'' unless one wanted to joke about an excess ive ­

ly ta lk a t iv e person. The longer duration associated with NOT UTTER A WORD,

117

NOT OPEN ONE'S MOUTH and TO SEAL ONE'S LIPS becomes apparent through the

incompatibi l ity of the expressions with adverbials indicating a short

period o f time. Notice that these three verbials were included among those

which express intense acts of being s i l e n t . As a matter of f a c t , s i l en ce

cannot be intense unless i t i s r e la t iv e ly long. Therefore a l l the verbials

discussed with refrence to the in te n s i ty dimension belong here as w e l l .

However, i t i s not true that a l l long acts o f being s i l e n t are a lso in ­

tense: TO KEEP ONE'S MOUTH SHUT i s , but TO KEEP QUIET i s not.

In the fol lowing paragraphs we shall only concentrate on verba tacendi

the form of which draws e x p l i c i t attention to a longer duration. As a

r e s u l t , most of those discussedb with refrence to the in te n s i ty dimension

w il l be l e f t out. The same princip le would a lso exclude those verbials

which derive the ir assoc ia t ion with a longer duration only from the ir

incompatibi l ity with adverbials indicating a br ie f period of time. However,

I have not been able to discover any verba tacendi which describe long

s i l en ce acts and which neither carry an e x p l i c i t reference to duration

nor belong to those emphasizing the in t e n s i t y of s i l e n c e .

29. The most common indication o f duration in English i s the presence of

the verb TO KEEP as in : TO KEEP BACK, TO KEEP BETWEEN US, TO KEEP BUTTON­

ED UP, TO KEEP CLOSE, TO KEEP DARK, TO KEEP FROM, TO KEEP IN, TO KEEP IN

IGNORANCE, TO KEEP IN PETTO, TO KEEP IN THE DARK, TO KEEP IT A DEEP DARK

SECRET, TO KEEP IT UNDER ONE'S HAT, TO KEEP MUM, TO KEEP ONESELF TO ONE­

SELF, TO KEEP ONE'S MOUTH SHUT,TO KEEP ONE'S OWN COUNSEL, TO KEEP ONE'S

TONGUE BETWEEN ONES CHEEK, TO KEEP ONE'S TRAP/YAP SHUT, TO KEEP QUIET,

TO KEEP SECRET, TO KEEP SILENCE, TO KEEP SILENT (the long duration of

118

which was our reason not to regard i t as a basic verbum tacend i) , TO KEEP

SOMEBODY OUT OF SOMETHING, TO KEEP STILL, TO KEEP TO ONESELF, TO KEEP

UNDER WRAPS, TO KEEP WITHIN THE BOSOM OF THE LODGE, TO KEEP WITHIN THESE

WALLS.

Some other indicators are TO MAINTAIN, TO LEAVE, TO REMAIN and TO SIT:

TO MAINTAIN A SECRET, TO MAINTAIN (A DEATHLIKE/TOMBLIKE) SILENCE, TO LEAVE

IN THE DARK (ABOUT), TO REMAIN SILENT (ABOUT/AS TO), TO SIT MUM, TO SIT

ON. And f i n a l l y there i s the morpheme HOLD in TO WITHHOLD, and the verb

TO HOLD in: TO HOLD ONE'S PEACE, TO HOLD ONE'S TONGUE, TO HOLD OUT ON.

30. In Dutch there are three comparable indicators of duration, namely

HOUDEN ( ' t o keep, h o ld ' ) , LATEN { ' l e a v e ' ) and BEWAREN ( ' t o maintain,

keep') as in: DE HANDEN IN DE MOUW HOUDEN ( l i t . ' to keep one's hands in

one's s l e e v e ' ) , IEMAND ERGENS BUITEN HOUDEN/LATEN ( l i t . 'to keep/leave

someone out of something') , IETS ACHTER DE ELLEBOOG HOUDEN ( l i t . 'to

keep something behind one's e lbow') , IETS ACHTER DE HAND HOUDEN ( l i t . 'to

keep something behind one's hand'), IETS ACHTER HOUDEN ( ' t o keep some­

thing back'), IETS IN PETTO HOUDEN ( ' t o keep something in p e t t o ’ ) , IETS

TUSSEN DE TANDEN HOUDEN ( l i t . 'to keep something between one's t e e t h ' ) ,

IETS VOOR ZICH HOUDEN ( ' t o keep something to o n e s e l f ) , IETS IN DE MOUW

HOUDEN ( l i t . ' to keep something in one's s l e e v e 1) , STILHOUDEN ( ' t o keep

s t i l l ' ) , DE TONG VOOR DE TANDEN HOUDEN ( l i t . ' to keep the tongue in front

o f one's t e e t h ' ) , IETS BEDEKT HOUDEN ( ' t o keep something covered ') , ZICH

ERGENS BUITEN HOUDEN ( ' t o keep.out o f something') , ZICH KOETHOUDEN ( ' t o

keep mum'), ZIJN MOND/BEK/BAKKES/BABBEL/RAMMEL/SNATER/SNOET/SNUIT/TOET

HOUDEN ( l i t . 'to hold one's mouth/ trap/yap e t c . ' ) , ZIJN TONG IN BEDWANG/

TOOM HOUDEN ( l i t . 'to keep one's tongue in check /b r id led ' ) , ZIJN PIJPEN

119

IN DE ZAK HOUDEN ( l i t . 'to keep one's pipes in one's pocket'); IEMAND VAN

IETS ONKUNDIG LATEN ( l i t . 'to leave somebody ignaorant about something'),

IETS BUITEN BESCHOUWING LATEN ( ’to leave something out of account') , IETS

MAAR BLAUWBLAUW LATEN ( l i t . ' to leave something blueblue') , TERZIJDE

LATEN ( l i t . 'to leave something a s id e ' ) , IETS IN DE D00FP0T LATEN ( l i t .

'to leave something in the extinguish ing-pot'); DE STILTE BEWAREN ( l i t .

to maintain the s i l e n c e ' ) , EEN DOODSE STILTE BEWAREN ( ' to maintain a

deathlike s i l e n c e ' ) , HET STILZWIJGEN BEWAREN ( l i t . 'to maintain the [act

of} being s i l e n t ' ) , EEN GEHEIM BEWAREN ( ' t o keep a s e c r e t ' ) .

F. SILENCE AND ITS CAUSES

31. In the vast majority of cases the act of being s i l e n t i s del iberate.

But the se t of l in g u i s t i c action verbials in English and in Dutch leaves

room for instnaces of s i lence which lack th is w i l l fu l character. Thus

s i lence acts can be distinguished into those corresponding to (Al) and

those describable in terms of (A2).

(Al) Sa i s s i l e n t , but not del iberately

(A2) Sa i s s i l e n t del iberately

What are the cau ses of the s i lence in case that (Al) obtains ? I t i s th is

question which will occupy me in the next couple of paragraphs. Silence

can only be involuntary i f an i n a b i l i t y to speak i s involved. But th is

general cause needs further analysis: the in a b i l i ty i t s e l f can be caused

by many d if ferent circumstances.

120

32. When we move from deliberate to non-deli berate s i len ce , we find just

across the border the acts of being s i l e n t which sa t i s fy (Cl), i . e . which

are due to a general, temperamentally determined, d isp o s itio n not to

talk.

(Cl) Sa is temperamentally dis inclined to talk

There are no verba tacendi which draw exclusive attention to th is type

of act. The verbials which can be used to describe i t are ambiguous between

an (Al)-(Cl) act and an (A2) act of some kind. Consider, for instance,

(Dl).

(Dl) Sally 's pride was wounded, but she suppressed her feel ings

The suppression of Sa l ly 's feeings can either be a conscious act perform­

ed for a particular reason or a temperamentally determined ref lex . In the

f i r s t case (Dl) describes an act of the (A2) type; in the second case

an (Al)-(Cl) act i s involved. The association of TO SUPPRESS with (Cl)

i s probably due to the fact that the verb does not only mean 'to refrain

from expressing1 but also 'to exclude from consciousness' and excluding

something from consciousness i s , according to our heavily freudianized

twentieth-century world view, mainly an unconscious ac t iv i ty (in spite

of Webster's definit ion of SUPPRESSION as "the conscious intentional ex­

clusion from consciousness of a thought or fee l ing"). The association

with (Cl) i s even stronger in (D2) so that the ambiguity between (Al) and

(A2) i s almost — though not completely — resolved.

(D2) Sal ly 's pride was wounded, but, as always, she suppressed her

feelings

121

On the other hand, in (D3) the balance t ips completely towards {A2).

(D3) S a l ly ' s pride was wounded, but somehow she managed to suppress

her f e e l in g s

Notice that not only l in g u i s t i c elements in the description can influence

the del iberateness-balance, Also the nature of the fee l in g to be expressed

i s important. Consider (D4).

(D4) Sa l ly was on the verge o f gett ing as cross as two s t i c k s , but

she suppressed her f e e l in g s

In th is case the suppression i s almost necessar i ly d e l ib era te , which

implies that anger i s regarded as a fee l in g which i s e a s ier to express

or more d i f f i c u l t to control than, say, being hurt. In f a c t , anger and

i t s expression are so firmly united in our western mind that i t would

be hard to say (at l e a s t for most o f the speakers o f English I consulted)

"Sally was r ea l ly p i s se d -o f f , but she didn't show her f e e l in g s" , whereas

"Sally was r ea l ly hurt, but she d id n ' t show her f ee l in g s i s quite l ik e l y

to occur.

Similar verbs on the border between (Al) and (A2) are: TO REPRESS, TO

SMOTHER, TO STIFLE, and in Dutch ONDERDRUKKEN, VERSMOREN and OPKROPPEN.

I have the fee l in g that OPKROPPEN ( l i t . 'to p i l e up in one's gizzard')

i s more strongly associated with (Cl) than any o f the others.

33. Another source of involuntary s i l en ce i s the ta cens1 in d e c is io n , as

described in (C2).

(C2) Sa i s unable to decide what to say (next)

122

Again the most common verbial belonging here i s ambiguous between (Al) and

(A2), but (Al) i s i t s central meaning: TO HESITATE, as used in (D5)

(D5) Mr. Smith hesitated before continuing his talk

typica l ly describes a si lence due to indecision (which i s no voluntary

state of mind); i f the act described is w i l l f u l , there i s pretense in­

volved. The same holds for the Dutch equivalent AARZELEN, but the meaning

of a second equivalent, WEIFELEN, i s completely restricted to (Al).

34. The inab i l i ty to talk may also result from aston ish m en t, g r i e f , shook,

or other s tro n g em otion.

(C3) Sa is unable to talk because of some strong emotion

The verbials TO BE MUTE and TO STAND MUTE belong here, as used in (D6)

and (D7), respectively.

(D6) After hearing the news of his w ife 's death, he was mute for

several minutes

(D7) The news was so unexpected that he stood mute for several mi­

nutes

These two verbials , however, can also be used to describe deliberate acts

of being s i l e n t . Consider (D8).

(D8) The mayor i s s t i l l mute on the subject

And TO STAND MUTE may be predicated of a person arraigned by law who

makes no answer, maintains s i l en ce , or refuses to plead direct ly or stand

t r i a l .

123

The Dutch expressions which use muteness as a metaphor exclude (A2)

completely from their meaning: VERSTOMMEN { l i t . 'to become mute'), VER-

STOMD STAAN VAN ( l i t . 'to stand mute because o f ' ) , MET STOMHEID GESLAGEN

ZIJN { l i t . 'to be struck with muteness') , MET VERSTIJFDE TONG STAAN

( l i t . 'to stand with a s t i f fen ed tongue') , SPRAKELOOS STAAN { l i t . 'to

stand sp eech les s ' ).

Some more verbials s a t i s f y (C3): TO DRY UP, which again i s ambiguous

between (Al) and (A2) as appears from the command "Dry up !", TO FALL

SILENT and i t s Dutch equivalent STILVALLEN, both o f which are restr ic ted

to (Al), and the expression EEN KIKKER/ROGGESTAART IN DE KEEL HEBBEN

( l i t . 'to have a frog /ray-ta i l in one's th r o a t ' ) , which i s ambiguous

between the in a b i l i ty to speak because of hoarseness or becuase of some

strong emotion, DE WOORDEN BLEVEN HEM IN DE KEEL STEKEN ( l i t . 'the words

remained stuck in his throat') and HET WOORD STIERF HEM OP DE LIPPEN ( l i t .

'the word died on his l i p s ' ) .

An interest ing question i s what kinds of emotions are involved in pro­

ducing the tacens' in a b i l i ty to speak. For one thing, a l l the (A1)-(C3)

verbials imply that the emotion was caused by something unexpected; in other

words, there is always s u r p r is e involved. Moreover, the surprise i s a l ­

most necessarily of an unpleasant kind. I f someone 'stood mute' or ' f e l l

s i l e n t ' upon hearing a piece of good news, bystanders would soon infer

that the person in question was not glad at a l l ; he or she i s expected

to say at lea s t "I can't bel ieve it" (and to make some additional funny

noises ) . Jumping to a conclusion, one might say that the vision underlying

the Dutch and English lex ica l systems does not allow for the existence

of joy unless i t i s expressed whereas g r ie f and sorrow are not in need of

124

expression in order to e x i s t . Joy i s not the only emotion with which TO

STAND MUTE, TO FALL SILENT, e tc . are not generally associated . Another

one i s anger, which i s a lso firmly connected with i t s expression because

of what we referred to before as the re la t iv e ease with which i t can be

expressed or the e f fo r t i t takes to control (see paragraph 32). A further

explanation may be that unlike g r ie f or shock, which may ar ise suddenly

due to an unexpected piece of unpleasant news(remember the element of

su rp r ise ) , intense f ee l in g s o f anger or wrath are considered to need more

time to generate.

35. A further cause of the in a b i l i t y to speak may be a la c k o f a n y th in g to

s a y .

( C4) Sa i s unable to speak because he does not know what to say

This cause may be involved in an act described by means of TO DRY UP,

TO FALL SILENT and STILVALLEN. More spec ia l ized (A1)-(C4) verbia ls are

NOT HAVE A WORD TO SAY, TO HAVE LITTLE TO SAY, and the Dutch GEEN WOORDEN

VOOR IET HEBBEN ( l i t . ' n o t have words for something1) and MET DE MOND VOL

TANDEN STAAN ( l i t . ' to stand with one's mouth f u l l o f t e e t h 1) . The l a s t

one of these indicates the tacnes' defeat; metaphorically speaking, his

conversational partners have checkmated him so that he i s at a loss for

things to say; o f course, th i s does not y e t make the image transparent.

36. A special case o f not knowing what to say i s the occasin on which

one has f o r g o t te n vskat one was g o in g to s a y .

(C5) Sa i s unable to speak because he has forgotten what he was going

to say

125

The verbials which can be used to express th is quite specialized but not

uncommon sta te of a f fa irs are the same as for (C4): TO DRY UP, TO FALL

SILENT and STILVALLEN. This type o f occasion i s exemplified in (D9).

(D9) The play could have been good. I t ' s a pity that one of the

debuting actresses dried up in the second act and lacked the

necessary presence of mind to improvise

In the example TO DRY UP simply means 'to forget one's l in es ' (which

Webster even l i s t s separately as one of the meanings of TO DRY UP).

37. The next cause of s i lence to which e x p l i c i t attention i s drawn by

some verbials i s the ta lk in g o f o th e r s .

(C6) Sa is unable to talk because o f the excessive talking o f others

Examples are NOT GET A WORD IN EDGEWAYS, Dutch ER GEEN WOORD TUSSEN

KRIJGEN and GEEN SLAG AAN DE BAK KRIJGEN {the formation of which was

explained before; see paragraph 26).

38. The f inal not . to say terminal — cause of s i lence which i s mani­

fested in the verba tacendi i s d e a th .

(C7) Sa i s unable to talk because he i s dead

This cause of s i len ce i s c lear in TO TAKE A SECRET INTO THE GRAVE and i t s

Dutch equivalent EEN GEHEIM MEE IN HET GRAF NEMEN. The same cause i s often

implied by GEEN PIEP/KIK (MEER) GEVEN/LATEN ( l i t . 'not g i v e / l e t out a

peep/sound (any more)').

The causal dimension of s i lence i s represented in SDC-Table 6. In the

DUTCH

s t i lv a l l e n

(opkroppen, e tc . )

(aarzelen), weifelen

verstommen, e tc .

geen woorden voor ie t s hebben, met de mond vol tanden staan

0

geen slag aan de bak krijgen

een geheim mee in het graf nemen

SEMANTIC DIMENSION--------------- ENGLISH-

£ (AX) -{Cl) ^ (to suppress, e tc . )

£ (A1)-(C2) 1 {to hesitate)

£ (A1)-(C3) ( to be/stand mute)

f (Al)—(C4) not have a word to 1 (to dry up)say, to have l i t t l e j to fa l l s i l e n tto say

[ (Al)—(C5) 0

£ (Al)-(C6) not get a word in edgeways

[ (A1)-(C7) J to take a secret into the grave

SDC-Table 6

127

table round brackets are used for the verbials which do not only show

an (Al) meaning, but an (A2) meaning as wel l .

G. THE TACENS AND HIS MOTIVES

39. This section deals with the second pole in the contrast between acts

describable in terms of (Al) and those representing (A2).

(Al) Sa i s s i l e n t , but not del iberately

(A2) Sa i s s i l e n t del iberately

As said before, in the cast majority of cases the acts of being s i l e n t

are del iberate. Thus the question to ask i s th is : What are the tacens'

motives or reasons for being s i l e n t ? Whereas in a b i l i t y to speak was the

one basic cause of a l l involuntary acts of being s i l e n t , i t i s hard to

f ind a basic reason or motive for the w i l l fu l ones. Not only i s there

a greater d i f fere n t ia t io n , often several reasons are intertwined.

Before giving an overview of the motives singled out by the verba ta-

cendi i t i s important to notice that the border between (Al) and (A2)

i s not a clear one. A troublesome case i s TO FAIL TO MENTION: the pre­

suppositions of TO FAILand those often carried by the expression as a

whole are contradictory. TO FAIL denotes an involuntary non-performance

of a certain a c t iv i ty because i t presupposes a del iberate attempt to

perform i t . However, TO FAIL TO MENTION, as used in (Dl) and (D2)

(Dl) In his speech about American foreign pol icy he fa i led to men­

tion the CIA's involvement in Chile

128

(D2) He to ld the pol ice that he had been home at the moment o f the

crime. But he f a i l e d to mention that he had been at the scene

of the crime f i f t e e n minutes e a r l i e r and that he had seen the

suspect

usually carries the implicat ion that the s i l en ce act was conscious

or even planned though i t was performed as i f through forget fu lness .

40. A second preliminary remark i s that there are quite a lo t o f verba

tacendi belonging to the (A2) category which do not s in g le out any par­

t i c u la r reason for the act of s i l e n c e they describe: TO BECOME MUM, TO

CLOSE, TO COME TO AN END, TO SAY NOTHING, TO MAINTAIN

SILENCE, and the Dutch AFSTAPPEN VAN ( l i t . 'to step down from'), EINDI-

GEN ( ‘to end ') , NIETS ZEGGEN ( ' t o say nothing1) , HET STILZWIJGEN BEWAREN

( ' t o maintain s i l e n c e ' ) and many others are used to describe acts of

being s i l e n t which can be performed for any r e a s o n . On the other hand,

I have not been able to discover any (Al) verba tacendi which do not

s in g le out a particular cause for the involuntary s i l en ce described. The

obvious rationale underlying the contrast i s th i s : as any other a c t i v i t y ,

being s i l e n t i s ty p ic a l ly a w i l l fu l act; therefore , i f the act i s not

voluntary, an explanation i s always needed.

Along the l in es o f th is contrast we could d is t ingu ish the following

properties of the verba tacendi.

(Cvl) The verbial does not s in g le out a part icu lar cause of the

s i l en ce

129

{Cv2) The verbial s ingles out a particular cause of the s i lence

(Cv3) The verbial does not s ingle out a particular motive for the

{Cv4) The verbial s ingles out a particular motive for the s i lence

Correspondingly, (Al) verbials could theore t ica l ly show patterns (SP1)

or (SP2).

(SP1) The verbial s a t i s f i e s (Cvl)

(SP2) The verbial s a t i s f i e s (Cv2)

And (A2) verbials can show either (SP3) or (SP4).

(SP3) The verbial s a t i s f i e s (Cv3)

(SP4) The verbial s a t i s f i e s (Cv4)

These four patterns are represented in SDC-Table 7. According to what

we said above (SP1) i s a completely empty pattern. (SP2) verbials were

discussed in the previous section on the causal nature of s i len ce . This

section on the tacens' motives wil l be mainly concerned with (SP4) verb­

i a l s .

s i lence

DUTCH ■SEMANTIC DIMENSION •ENGLISH

0 SP1 SP1 0

see par. 32-38 SP2

see par. 40-41 SP3

see par. 42-49 SP4

[ (Al)-(Cv2) ] SP2 see par. 32-38

SP3 see par. 40-41

SP4 see par. 42-49

SDC-Table 7

130

41. A few more words need to be said about (SP3) verbials . Though they

do not indicate any particular reason for the tacens to be or become

s i l e n t , many of them refer to a s tro n g d e te rm in a tio n , a refusal to speak

or, in the case of verba silendi and reticendi, even obstinacy. (Proba­

bly the reader does not have to be reminded of the possible interrelat ion­

ships between different semantic dimensions; in this case i t i s clear

that verbials implying a strong determination will often indicate a

long duration and a high intensity as w e l l . ) Some examples are TO BE MUM

AS AN OYSTER, NOT BREATHE A WORD, GEEN STOM W00RD ZEGGEN ( l i t . 'not say

a dumb word'), IN ALLE TALEN ZWIJGEN ( l i t . 'to be s i l e n t in al l languages'),

etc . It i s clear that often some strong emotion will be responsible for

the generation of the determination or obstinacy. The nature of the emo­

tion is never specif ied. But one may assume that , contrary to what we

discovered for the emotions involved in some of the involuntary acts of

being s i l e n t , anger is one of the favorites here. On the other hand, joy

i s even more unlikely to underlie an (A2) act than an (Al) act.

Some (SP3) verbials do not draw the attention to a strong determination

as such, but to a determination not to speak in spite of one's inclination

to the opposite. I am thinking of TO BITE ONE'S LIP/TONGUE, TO MUZZLE

ONESELF, TO PUT A BRIDLE ON ONE'S TONGUE, TO BUTTON/ZIP ONE'S LIP and in

Dutch OP ZIJN TONG BIJTEN ( ' to bite one's tongue1) , ZICH OP DE LIPPEN

BIJTEN ( ' to bite one's l i p s ' ) , ZIJN TONG IN BEDWANG/T00M HOUDEN ( ' to

keep one's tongue in check/bridled'). Another example i s ZIJN TONG GEWELD

AANDOEN ( l i t . 'to do violence to one's tongue'). I present th is one se­

parately not only because i t i s an extremely clear case in point but also

because i t manifests very plainly the talkative character of western

131

culture: moving from non-action to action certa in ly takes more e f f o r t

than remaining in a s ta te of res t ; therefore one could an expression

which means l i t e r a l l y l to do v io lence to one's tongue' to describe the

action o f moving, against one's in c l in a t io n , from not speaking to speak­

ing; but quite the opposite i s true: ZIJN TONG GEWELD AANDOEN is f ight ing

one's inc l in a t ion to talk in order to remain in a s ta te of s i l e n c e .

42. One of the most common reasons for being s i l e n t , which i s manifested

by many verba r e t ice n d i , i s co n cea lm en t.

(Cl) Sa wants to conceal something

Some obvious examples: TO KEEP SECRET, TO CONCEAL, TO VEIL; GEHEIMHOUDEN

( ' t o keep s e c r e t ' ) , VERZWIJGEN ( ' t o be s i l e n t about') , VERBERGEN ( ' t o

concea l ' ) ; e tc . Concealment i s often associated with subsidiary reasons

or motives such as deception, s o l i d a r i t y , e t c . However, the vast majority

of (A2)-(C1) verbia ls do not r e s t r i c t the ir meaning to a particular one

of these . But a few of them do. Let us pass those in review.

The following four subsidiary motives (which often in terac t) play a role

in at l e a s t some verba tacendi:

(Cla) Sa wants to conceal something in order to deceive somebody

(Clb) Sa wants to conceal something out o f s o l id a r i ty

(Clc) Sa wants to conceal something out o f d iscret ion

(Cld) Sa wants to conceal something out o f mercy

Though the - in te n tio n to d e c e iv e can be supposed to be a very common

reason for being s i l e n t about something, I have not been able to d i s ­

132

cover more than one verbial with which i t i s very strongly (though not

even necessari ly) associated: TO HOLD OUT ON as used in (D3), Dutch IETS

ACHTERHOUDEN.

(D3) Ever since we married she's been holding out on me

The association with deception resu lts from the fa c t that TO HOLD OUT ON

means 'to withhold information to which someone has a r igh t '; but even

so, other reasons, such as fear , may be at work. How i s i t that for a

common action such as withholding information with the intention to

deceive, there are almost no verbials ? Two phenomena may be responsible

for th is s i tuat ion . F ir s t , our rules of politeness may curb our i n c l i ­

nation to openly accuse a conversational tacens of deception. Second, and

more importantly, our generally accepted rules of conversation dictate us

not to make any claims for which there i s no s u f f i c i e n t evidence; since

s i len ce i s a completely intangible phenomenon i t i s not possible to adduce

i t as evidence of any kind; in other words, the tacens can always get away

with i t by taking refuge to arguments based on forgetfulness and the l ik e .

Therefore one usually does not have the courage to go any further than

making insinuations as with TO FAIL TO MENTION which, as said before,

l i t e r a l l y describes the involuntary non-performance of certain acts of

speaking though i t i s often used in such a way that i t implies the conscious

and purposeful performance of an act o f not mentioning something.

43. The three remaining subsidiary reasons for concealment which are

manifested in some verba tacendi might be regarded as 'good reasons':

they transform the acts described into cases of 'white s i len ce ' ( i f we

133

are allowed to use the formation o f a 'white l i e ' productively) .

S o l i d a r i t y i s the underlying reason in NOT GIVE AWAY as used in (D4).

(D4) She did not give John away

Some other examples are NEVER LET ON, TO KEEP BETWEEN US/THEM, TO KEEP

WITHIN THE BOSOM OF THE LODGE, TO KEEP WITHIN THESE WALLS, NOT LET IT

GO FURTHER, and the Dutch NIET VERKLAPPEN and NIET VERRADEN (both of

which are more or l e s s equivalent to NOT GIVE AWAY).

D is c r e t io n i s apparent as a motive only in TO BE THE SOULD OF DISCRE­

TION and DE DISCRETIE IN PERSOON ZIJN ( l i t . 'to be d iscret ion in person') .

These v erb ia ls , however, are mainly used to describe a general d isp o s i ­

t ion . I f applied to a s p e c i f i c occasion as in (D5)

(D5) He was the soul o f d iscret ion la s t night

the implication i s that the tacens talked about i s not normally the soul

of d iscret ion .

F in a l ly , there i s one Dutch verb ia l , IETS MET DE MANTEL DER LIEFDE BE-

DEKKEN ( l i t . 'to coversomething with the coat of lo v e ' ) which reveals

m e r c ifu ln e s s as the tacens' reason to be s i l e n t about something; the

tacens conceals some fa c t or event because the person(s) i t i s related

to might be hurt i f i t were brought to l ig h t .

We have seen that bad intent ions underlying acts of conceil ing are

raely focussed on because of the d i f f i c u l t y of obtaining evidence for

them and the generally accepted princ ip le that one should not make accu­

sat ions unless there i s s u f f i c i e n t evidence. But when one praises someone,

evidence i s rarely asked for unless there i s a strong suspicion that the

person in question does not deserve the praise at a l l . Therefore, no such

134

princip les can be re l ied on to explain why there are so few verba tacendi

focusing on good intentions underlying concealments. On the contrary, the

fol lowing argument could be made. Since there are no such pr inc ip les res ­

t r i c t in g the occasions on which one i s allowed to ascribe 'good1 proper­

t i e s or acts to someone, one could conclude from the small number of

l ex ica l too ls at our disposal to describe cases of benevolent s i l e n c e ,

e i th er that in our soc ie ty 'white s i l e n c e ' i s considered to occur once

in a blue moon, or that being s i l e n t because of s o l i d a r i t y , d iscret ion

or mercy i s not regarded as a rea l ly praiseworthy a c t iv i t y ; in e i th er

case the importance attached to speech i s emphasized once more.

44. Next to concealment there are two types of avoidance in the spectrum

o f motives for being s i l e n t .

( C2) Sa wants to avoid a topic

(C3) Sa wants to avoid a confrontation

The most important (A2)-(C2) verbials are: TO BEAT AROUND THE BUSH, TO

HEM AND HAW; ROND DE POT DRAAIEN ( l i t . 'to turn around the p o t ' ) , and

EEN SNAAR NIET AANROEREN ( l i t . 'not touch a [^particular] s t r i n g ' ) . Close­

ly related are TO SAY NEITHER YES NOR NO and ABRAHAMMETJE SPELEN ( l i t .

' to play l i t t l e Abraham', which means 'to try to escape by t e l l i n g half

the tr u th 1). The attempt to avoid a confrontation i s implied by TO CHECK

ONE'S SPEECH, TO HOLD ONE'S PEACE as in (D6) and (D7).

(D6) From the moment the supervisor entered, a l l employees checked

the ir speech

(D7) I did not agree with the teacher but held my peace as he was

rather angry

135

The Dutch verbials OP ZIJN MOND PASSEN ( l i t . ' to guard one's mouth'), OP

ZIJN WOORDEN PASSEN ( l i t . 'to guard one's words') and ZIJN WOORDEN AF-

METEN ( l i t . 'to measure one's words') are comparable to TO CHECK ONE'S

SPEECH; on the other hand, ZICH KOEST HOUDEN ( l i t . 'to keep quiet'*, the

adject ive KOEST i s a contracted form of the French "Couche-toi !" and

apart from i t s use in ZICH KOEST HOUDEN i t i s mainly used as an imperative

addressed to a dog and meaning ''Lie down !" or "Stop barking I") i s

c lo ser to TO HOLD ONE'S PEACE.

45. Another reason for being s i l e n t or for being s i l e n t about something

i s the ta ce n s ' in d i f f e r e n c e .

(C4) Sa i s s i l e n t (about something) because he regards the matter

at hand as too unimportant

Both TO SAVE ONE'S BREATH and TO WASTE NO WORDS draw e x p l i c i t attention

to such indifference: the tacens regards ta lk ing as a waste o f energy.

TO SAVE ONE'S BREATH may suggest that the deeper reason for not talking

about unimportant matters i s to build up energy for la te r ta lk ing . Com­

parable Dutch verbials are NIET VEEL WOORDEN VERSPILLEN ( 'not waste many

words') and NIET VEEL WOORDEN AAN IETS VUILMAKEN ( l i t . 'not dirty many

words at something').

46. The expectancy of disappointment may induce a speaker to be l ieve that

a part icu lar matter i s not worth wasting any words on. D isa p p o in tm en t

i t s e l f can a lso be the reason for w i l l fu l s i l e n c e .

(C5) Sa i s s i l e n t because he f e e l s disappointed

136

A fee l ing of disappointment presupposes a prior attempt to achieve some­

thing. Therefore, in the context of language i t presupposes prior speech.

As a r e su l t , the few verbials which f i t (C5) are al l verba cessandi. The

verbials I am thinking of are TO ABANDON and TO QUIT as used in (D8) to

(DIO).

(D8) The reporter abandoned asking questions after real iz ing that

he would never get a reply

(D9) He abandoned trying to convince his friend

( DIO) The senator quit speaking after the audience's rude reactions

The Dutch equivalent, OPGEVEN, may be more exc lus ive ly associated with

disappointment, but i t s use i s more res tr ic ted .

47. TO BREAK OFF, as used in ( D l l ) and ( D 1 2 ) , generally reveals a sudden

im pu lse as the tacens’ reason for discontinuing his speech.

(C6) Sa stops talking because of a sudden impulse to do so

( D l l ) The senator broke o f f in the middle of a sentence

( D12) The speaker was interrupted so often that he broke o f f and

sat down

The impulse in question may be generated by any kind of emotion, such as

surprise , anger, fear , and others. Similar verbials in Dutch are AFBREKEN

( ' to break o f f ' ) and BOT STILZWIJGEN ( ' t o stop talking abruptly1).

48. The postpon em en t of divulging a certain piece of information i s y e t

another motive for keeping s i l e n t .

137

(C7) Sa wants to postpone divulging some information

I t i s revealed in TO KEEP IN PETTO, Dutch IETS IN PETTO HOUDEN. The

tacens keeps something in petto i f he wants to reserve the tacendum

in order to convey i t at a more appropriate time or, in the case of an

argument, to be able to use i t more fo r c e fu l ly la te r on. (Originally

' in petto' was predicated o f a cardinal appointed by the pope but not

named in cons is tory . )

49. The f i n a l , and probably most t r i v i a l , reason for being s i l e n t re­

vealed by the verba tacendi i s the tacens' wish to c a tc h h is b r e a th .

(C8) Sa wants some breathing-time

This motive i s manifested in TO TAKE A BREAK and i t s Dutch equivalent

PAUZEREN (which does not mean that th i s i s the only reason for taking a

break). The implication i s that the speaking wil l be resumed after a

b r ie f in terval .

Both TO TAKE A BREAK and PAUZEREN are verba cessandi describing a

break in a speech a c t iv i t y belonging to a part icu lar communication type;

in an everyday conversation, for instance, one does not 'take a break'.

There i s an additional (A2)-(C8) verbial in Dutch which i s also a verbum

cessandi but at the same time a verbum ret icend i; moreover, i t i s not

r e s tr ic t e d with respect to communication type , but i t implies that the

suspension of the speech a c t i v i t y i s o f a considerable longer duration.

The verbial in question i s ERGENS EEN SPELDJE BIJ STEKEN; a SPELD(JE) i s

a ( l i t t l e ) pin; the pin referred to here i s an object attached to a par-

-DUTCH-

geheimhoudenetc .

0niet verklappen, e tc .

(de d iscret ie in per- soon zijn}

ie t s met de mantel der l ie fde bedekken

-SEMANTIC DIMENSION-

(AZ)-(Cl)

[ (Cla) ]

[ ( C l b ) l

[ (C lc ) ]

[ (C ld ) J

f ( A 2 ) - ( C 2 ) ]

[ (A2)-(C3) 3

rond de pot draaien, etc .

op zijn woorden passen, e tc .

niet veel woorden aan i e t s vuilmaken £ (A2)-(C4) 1

opgeven £ (A2)-(C5)

afbreken, e tc .

in petto houden

pauzeren, etc .

[ (A2)-(C6) 1 C (A2)-(C7) 1

C ( A 2 H C 8 ) 1

SDC-Table 8

-ENGLISH-

to hold out on

not give away, etc.

(to be the soul of discretion)

0

to keep secret etc .

to beat around the bush, etc .

to check one's speech

to waste no words, e tc .

to abandon, etc .

to break off

to keep in petto

to take a break

139

t icu lar page of a book to indicate how far one has gotten; the expression

means l i t e r a l l y 'to attach a pin to something'; within the domain o f the

verba tacendi i t means 'to suspend talking about a particular to p ic ' .

The motivational dimension of s i len ce discussed in th is section is re­

presented in SDC-Table 8.

H. SILENCE AND ITS INTERLOCUTORS

50. Not only do the verba tacendi inform us about the codes of s i l e n c e ,

i t s sonound properties, i t s causes and the tacens' motives. Some of them

even give away the nature of the tacens and his audience. There are several

verba tacendi which describe acts of being s i l e n t directed at the public

in general. I am thinking of TO COVER UP, TO SIT ON, TO BLACK OUT, TO

CLASSIFY and TO FILE AND FORGET as used in (Dl) to (D6).

(Dl) The banker tr ied to cover up his s teal ing some of the bank's

money

(D2) The senator kept s i t t in g on the plans until he knew for certain

which way winds were blowing

(D3) The wartime government blacked out a l l news

(D4) Dictators usually black out a l l cr i t ic ism of the government

(D5) CIA o f f i c i a l s c la s s i f i e d all the information about their in ­

volvement in Chile

(D6) All the information about the CIA's involvement in Chile was

f i l e d and forgotten

All of these examples conform to (Cl).

140

(Cl) The audience i s the public in general

(Cl) appears to be the only audience-oriented condition attached to any

verbum tacendi .

51. There are two c oro l la r ie s following from (Cl) . F ir s t , there would be

no need to spec i fy that the audience envisaged in the acts described in

(Dl) to (D6) i s the public in general unless the tacendum i s o f general

public in te r e s t and hence the tacens holds some kind o f o f f i c i a l function.

Thus the (Cl) verbia ls also furnish information about the tacens. Depend­

ing on the character of the tacens they can be divided into those s a t i s ­

fying (Cla) and those conforming to (Clb).

(Cla) The tacens i s usually an impersonal o f f i c i a l body

(Clb) The tacens may e i th er be an individual holding an o f f i c i a l

function or an impersonal o f f i c i a l body

TO COVER UP and TO SIT ON are of the (Clb) type, whereas TO BLACK OUT,

TO CLASSIFY and TO FILE AND FORGET s a t i s f y (Cla) . The anonymity o f the

tacens i s e sp e c ia l ly clear in the case of TO FILE AND FORGET which cannot

even be used in the act ive voice as a verbum tacendi.

A second corollary of the special re lat ionship between tacens and au­

dience implied by (Cl) verbials i s that a certain motive, which was not

y e t mentioned in the previous s e c t io n , seems to be involved: the tacens

withholds information from the general public because he be l ieves that

divulging i t might weaken his own pos it ion in public l i f e .

141

Sometimes (Clb) verbials may also be used to describe acts o f being

s i l e n t performed by an individual without the kind of o f f i c i a l function

referred to before, as in (D7) which conforms to (Clc).

{D7) The journa l i s t kept s i t t in g on some information that could

have produced a second Watergate

(Clc) The tacens i s an individual without o f f i c i i function

In th is case also the motivational structure of the act i s transformed:

the tacens withholds information from the general public because he be­

l ie v es that divulging i t might e ither harm himself or weaken someone

e l s e ' s posit ion in public l i f e .

52. As far as I can t e l l Dutch only possesses a s ing le verbial which

indicates that the audience i s the public in general. The verbial in

question i s IETS IN DE D00FP0T STOPPEN ( l i t . 'to put something into

the ext inguish ing-pot ' ) which means TO HUSH UP. These two verba tacendi

presuppose previous discourse about the tacendum and neither of them gives

any further information about the character of the tacens. (Note that

an act of hushing up i s not necessari ly directed at the public in gene­

ral , though usually i t i s . )

53. There i s only one more verbial relevant to the interlocutor-dimension

of the verba tacendi. TO DISCONTINUE, as used in ( DS) and (D9) necessari­

ly conforms to (C2).

(D8) The publishing company discontinued the unprofitable journal

142

(D9) The BBC discontinued the s e r ie s o f p o l i t i c a l debates

(C2) The tacens i s a publisher or a broadcasting service

I ts Dutch translat ions DE PUBLICATIE STOPZETTEN VAN { l i t . 'to stop

the publication o f ' ) and DE UITZENDING STOPZETTEN VAN ( l i t . 'to stop

the broadcasting o f ' ) , as said before, cannot be regarded as l ex ica l

items.

54. SDC-Table 9 represents the interlocutor-dimension, which i s c lea r ly

marginal, and even more so for Dutch than for English. I t i s noteworthy

that the only important manifestat ion of the dimension emphasizes public

acts o f being s i l e n t .

I t should be noted, before going on to a next semantic dimension, that

a number o f verba tacendi require a complement spec ify ing the person from

whom a tacendum i s withheld. They are: TO KEEP FROM, TO KEEP IN IGNORANCE

(ABOUT), TO KEEP IN THE DARK (ABOUT), TO LEAVE IN THE DARK (ABOUT), and

a few others. But nothing about the person in question i s predictable

from the verbials themselves.

DUTCH- SEMANTIC DIMENSION- ENGLISH-

i e t s in de doofpot stoppen

(Cl)

[ ( C l a ) ] to black out

( C l b ) fto s i t on, e tc ,

(Clc)*

[ (C2) to discontinue

(to hush up)

SDC-Table 9

143

I . THE TOPICS OF SILENCE

55. The next semantic dimension to be discussed takes us into the realm

of the propositional content of acts of being s i l e n t . In other words, we

shall be concerned with the tacenda, i . e . that which i s not being talked

about. As a resu l t , only verba reticendi wil l be studied in th is section.

According to whether a tacendum, which may be e i th er predictable of

unpredistable from a given verbum ret icend i , i s to be specif ied as a

complement to the verbial or not, we can distinguish four se ts o f verba

ret icen d i .

(Cdl) The tacendum has to be sp ec i f ied ; i t i s not predictable from

the verbial

(Cd2) The tacendum has to be sp ec i f ied though i t i s more or le s s

predictable from the verbial

(Cd3) The tacendum i s not further spec if ied because i t i s c lear

from the verbial

(Cd4) The tacendum cannot be sp ec i f ied though i t i s not predictable

from the verbial

TO KEEP SECRET, TO BE SILENT ABOUT, TO WITHHOLD, NOT GIVE AWAY and many

others belong to the (Cdl) category which const i tutes the largest part

of the verba ret icend i , both in English and in Dutch. The (Cd4) category

i s probably, and quite predictably, empty in both Dutch and English.

Since only (Cd2) and (Cd3) verbials provide us with information about

the tacendum, they are the only ones to be investigated in th is sect ion.

144

The variety of topics that can be talked about i s without l im i t . And

so i s the variety of topics that one can choose not to talk about. The

choice re f lec ted in the l e x ic a l i z a t io n of the absence o f speech, i . e .

in the verba tacendi , can be expected to further our understanding of

the nature of l in g u i s t i c s i l e n c e , as perceived in our l in g u i s t i c commu­

n i t i e s , considerably.

56. The topics to which (Cd2) verbials draw our attention are a l l s i ­

tuated a t e i th er one o f the two extremes of what could be ca l led the

'privacy continuum' o f ta lking matter. On the one hand there are

matters of general public in te re s t and on the other there are complete­

ly private ones.

(Al) Sa i s s i l e n t about a private matter

(A2) Sa i s s i l e n t about a matter o f general public in t e r e s t

The membership in the (Cd2)-(A2) category provides us with another exam­

ple o f the re lat ionships between d i f fe re n t semantic dimensions. In the pre­

vious sect ion i t was claimed that TO COVER UP, TO SIT ON, TO BLACK OUT,

TO CLASSIFY, TO FILE AND FORGET, TO HUSH UP and the l a t t e r ' s Dutch equi­

valent IETS IN DE D00FP0T STOPPEN a l l describe acts of being s i l e n t d i ­

rected at the public in general. There i s no need for being so sp e c i f i c

about the audience unless the tacendum i t s e l f i s o f public in te r e s t .

Therefore the same verbials belong to the (Cd2)-(A2) category.

The phrase 'matters of general public in teres t ' covers a wide range of

possible topics of speech; th is i s no doubt the reason why the verbials

enumerated require a complement. Most ( Cd2)—(Al) verbials are s l i g h t l y

145

more s p e c i f i c . They do not ju s t indicate that the topic i s a 'private

matter' , but most of them require some kind of fee l in g or emotion as a

tacendum. The verbials TO REPRESS, TO STIFLE, TO MUFFLE, TO SUPPRESS and

TO SMOTHER, as well as the ir Dutch counterparts OPKROPPEN and VERSMOREN

al l indicate that the tacens i s s i l e n t about some fee l in g or emotion.

Of course, a further sp ec i f ic a t io n i s s t i l l necessary in the description

because of the many sorts o f emotion. Thus one can smother one's rage,

repress one's worries, s t i f l e one's anger e tc . Notice that the range of

tacenda i s limited to unpleasant f e e l in g s . Apart from these emotional

v erb ia ls , which describe acts of the (Ala) type, there i s probably only

one ve rb ia l , namely the Dutch IETS NIET AAN IEMANDS NEUS HANGEN { l i t .

'not hang something on someone's n o s e ' ) , which includes both (Ala) and

(Alb) in i t s meaning, and there are none at a l l with a sense res tr ic ted

to (Alb).

(Ala) Sa i s s i l e n t about some (unpleasant)emotion or fee l ing

(Alb) Sa i s s i l e n t about a non-emotional private matter

IETS NIET AAN IEMANDS NEUS HANGEN can be used for any act o f not commu­

nicating something which the tacens regards as his own private business .

Thus, "Dat ga ik n i e t aan je neus hangen" ( l i t . ‘ I'm not going to hang

that on your nose") i s synonymous with "That's none of your business".

Notice that the ( Cd2) - ( Ala) verbials also get non-emotional comple­

ments as in 'to smother a secret ' and 'to muffle g o ss ip ' . I t seems that

in these cases the tacendum has to be a matter o f general public in te r e s t

as w e l l . Thus TO STIFLE, TO SMOTHER, e tc . seem to incorporate both ex­

tremes o f the 'privacy continuum'. It i s hard to t e l l whether the ir

146

(Ala) and (A2) meanings are on a par with each other or whether one i s a

derivat ive of the other.

57. Though the private-public contrast i s s t i l l detectable in the (Cd3)

verba tacendi, i t only plays a role in some modified forms. For one thing,

s ince (Cd3) verbials do not allow further sp ec i f ic a t io n s of the tacendum

in the form of a complement, the verbial i t s e l f should give more precise

information: i t should be possib le to ca lcu late the nature of the tacen­

dum on the basis of the verbial i t s e l f , combined with some contextual

information. The most important topics o f s i len ce revealed by (Cd3) verba

tacendi are presented in (A3) through (A7b).

(A3) Sa i s s i l e n t about everuthing that he regards as personal

(A4) Sa i s s i l e n t about other people's private matters (which may or

may not be of public in te r e s t )

(A5) Sa i s s i l e n t about his ideas or plans (which are of potential

in t e r e s t to the hearer)

(A6) Sa refrains from voicing his disagreement or disapproval

(A7a)Sa i s s i l e n t about a matter which i s o f in t e r e s t to the hearer

because i t i s the topic of conversation

(A7b) Sa i s s i l e n t about a matter which i s o f in t e r e s t to the hearer

because the l a t t e r has a r ight to know about i t

A b r ie f overview of the verb ia ls in question fo l low s.

58. Act (A3) d i f f e r s from (Al) in that the former i s an act o f being

s i l e n t about everything that the tacens regards as personal or pr ivate ,

147

whereas the la t ter i s an act of being s i l e n t about one particular private

matter. This fact can be reconciled with our previous statement that ( Cd3)

verbials should give more precise information about the topic of s i lence

than (Cd2) verbials because they do not allow further specif icat ions of

the tacendum in the form of a complement. The tacendum of a (Cd3)-(A3)

verbial , being'every personal matter1, i s l in g u is t ic a l ly more sp ec i f ic

than 'a personal matter1, though at the same time i t i s more general.

The se t of (Cd3)-(A3) verbials includes: TO KEEP ONE'S DISTANCE, TO KEEP

AT A DISTANCE, TO KEEP ONESELF TO ONESELF, TO STAND ALOOF, TO HOLD ONE­

SELF ALOOF, and the Dutch OP EEN AFSTAND BLIJVEN ( ' to keep at a d is tance ' ) ,

ZICH OP EEN AFSTAND HOUDEN (idem), GERESERVEERD BLIJVEN ( ' t o remain re­

served') . One of the ways in which the tacens may succeed in keeping his

distance is by talking about non-personal things; this observation shows

one of the many bridges between s i lence and non-silence.

Observe that TO KEEP AT A DISTANCEand all three Dutch verbials may also

describe acts of type (A8).

(A8) Sa i s s i l e n t about a certain topic of conversation which he

regards as the other speaker's private business

Again i t i s hard to decide whether the (A3) or (A8) meaning of these verba

tacendi i s derived from the other, or whether they are on a par.

59. Silence acts of the (A4) type are described by one verbial only,

namely TO BE THE SOUL OF DISCRETION, Dutch DE DISCRETIE IN PERSOON ZIJN

( l i t . 'to be discretion in person'). However, these two are marginal

in the se t of verba tacendi because they are not normally used to describe

148

a particular act of being s i l e n t , but rather to indicate a general a t t i ­

tude or inc l in a t ion .

A (Cd3)-(A5) verb ia l , reporting the ta ce n s1 s i l en ce about plans or

ideas which are of potential in te r e s t to the hearer, i s TO KEEP ONE'S

OWN COUNSEL. Its Dutch equivalent i s DE KAARTEN DUIKEN (meaning more

or l e s s 'not show one's cards') or NIET IN ZIJN KAARTEN LATEN KIJKEN

( l i t . 'not show one's cards ' ) .

Both (Cd3)-(A6) and (Cd3)-(A7b) are empty c la s se s in Dutch. In English

they are represented by TO HOLD ONE'S PEACE and TO HOLD OUT ON, respect­

iv e ly .

F in a l ly , (A7a) i s described in English by means o f TO BEAT AROUND THE

BUSH and TO HEM AND HAW, and in Dutch by means of ROND DE POT DRAAIEN

( l i t . ' to turn around the p o t ' ) .

60. Two additional verbials which s a t i s f y (Cd3) marginally are TO REFUSE

COMMENT and TO SAY NEITHER YES NOR NO. Since the contextual information

that they carry i s more important (and cer ta in ly l e s s vague) than the

propositional information, we shall come back to them in the sect ion

on s i l en ce and context.

The propositional dimension o f the verba tacendi i s v isua l ized in

SDC-Table 10. Again, apart from some minor d i f fere n c es , The English and

Dutch lexicon are quite symmetrical.

■DUTCH -SEMANTIC DIMENSION

/ C .ie-t-s n-jet aan iemands opkroppen, t neus hangen

etc.'*

L i e t s in de doofpot s

'(Cd2)T- (Ala)

stoppen

op een afstand blijven, etc .0

0

[(Cd3)

L(de d iscret ie in persoon z i j n T

de kaarten duiken

0 Irond de pot draaien V

0 [

(Alb) J[ I’ (A2) ]

(A3) 1

(AS) ]

(A4) 1

(AS) ]

(AS) T

(A7a) 3

(A7b) 1

SDC-Table 10

-ENGLISH

1\

.to s t i f l e , etc .

to black out, etc . to hush up

to keep oneself to oneself etc .

0to keep at a distance

(to be the soul of discretion)

to keep one's own counsel

to hold one's peace

to beat around the bush

to hold out on

150

J . SILENCE AND NON-PROPOSITIONAL MEANING

61. A person's s i len ce may be imbued with a kind of meaning other than

the topics or the propositions that the tacens i s being s i l e n t about.

This non-propositional meaningfulness i s generally indicated by means

of adject ives as in TO MAINTAIN A DEATHLIKE/TOMBLIKE/GOLDEN/SOLEMN/

PREGNANT SILENCE. DEATHLIKE and TOMBLIKE do not only mark in ten s i ty ;

in addition they s tres s the s i n i s t e r or ominous character o f the s i l e n c e .

The absence of speech can be SOLEMN only when something serious happened,

i s happening or i s going to happen, usually as part of a { re l ig io u s or

secular) ceremony. Just as the 'calm before the storm' s i l en ce i s PREG­

NANT when the s i l e n t people r ea l ize that a l o t of (usually unpleasant)

things could be said and are probably going to be said before long. When

the s i l en ce i s highly desirable to all p a r t i e s , i t may be said to be

GOLDEN.

The same types o f adject ives can be combined with STILTE { ' s i l e n c e ' )

in Dutch: DOODS { 'd ea th l ik e ) , PLECHTIG ( ' so lem n ') , GELADEN ( l i t . ' loaded',

equivalent to 'pregnant' in th i s context) , e t c . It i s s ig n i f i c a n t that

there i s almost no l im i t to the negatively-colored adject ives which can

be combined with ' s i l e n c e ' in both languages. Here are some more: OMINOUS

and i t s equivalent ONHEILSPELLEND; THREATENING and i t s Dutch counterpart

DREIGEND; e tc . On the other hand, i t i s hard to find any p o s i t iv e adjec­

t i v e s apart from GOLDEN and SOLEMN (which i s only semi-posit ive since

s i l en ce at a funeral i s even more l ik e ly to be solemn than s i len ce at a

wedding ceremony). Moreover, the meaning o f an inherently po s i t ive ad­

151

j e c t i v e such as PREGNANT {pregnancy being associated with 'joyful ex­

pectation' which i s the l i t e r a l translat ion of one o f i t s Dutch l e x i c a l -

i z a t io n s ) gets perverted in i t s combination with ' s i l e n c e ' . These facts

emphasize once again our repeated conjecture that for Anglo-Americans

and speakers of Dutch there i s no such thing as joyful or pleasant

s i l en ce — at l e a s t in the sense of s i l e n c e as the absence o f speech

K. SILENCE AND CONTEXT

62. The contextual dimension o f the verba tacendi was f i r s t introduced

when the verba cessandi were defined (see paragraph 4 ) , All verba

cessandi presuppose preceding discourse. However, the nature of that

discourse i s rarely sp ec i f ie d . Observe (Dl) to (D9).

(Dl) He abandoned s tu t t e r in g , rea l iz ing i t provoked r id icu le

rather than pity

(D2) He abandoned asking questions a f ter rea l iz ing that he would

never get a reply

(D3) He abandoned speaking a f ter the audience's rude reactions

(D4) He abandoned his impolite speech a f ter his father punished him

(D5) He abandoned writing poetry

(D6) He abandoned speech for several days

(D7) He abandoned his native speech and adopted the French tongue

(D8) He abandoned trying to convince his friends

(D9) He abandoned the subject when he did not get any response

The examples show that TO ABANDON, which denotes the discontinuation

of previous speech, may focus on part icu lar sound fea tu res , a certain

152

speech act type, a certain type of t e x t , a particular s t y l e , a particular

communication type, language in general , one individual language, a cer­

tain intention and a particular intended e f f e c t , and a particular sub­

j e c t . Thus the verb i s maxinally general and does not carry information

about the nature of the discourse the discontinuation of which i t des­

cribes. Most verba cessandi are s im ilar in this respect. The number of

exceptions i s extremely small.

63. An example of a verbum cessandi which carries sp e c i f ic information

about the discontinued discourse i s TO KNOCK IT OFF as used in (DIO).

(DIO) John f in a l ly knocked i t o f f

The use of TO KNOCK IT OFF presupposes that i t s user (e .g . the person

uttering (DIO) regards the preceding discourse e i ther as nonsensical

or as inappropriate. In other words, condition (Cdl) obtains.

(Cdl) Sd regards the discontinued discourse as nonsensical or

inappropriate

Consider (Dll) and (D12).

(Dll) After talking nonsense for hours, John f in a l ly knocked i t o f f

(D12) John had been rude a l l night , but he f in a l ly knocked i t o f f

In these two descriptions the p o s s i b i l i t i e s are exemplified e x p l i c i t l y .

A Dutch verbum cessandi which provides s p e c i f ic information about the

nature of the discontinued discourse i s STAKEN ( l i t . 'to s t r ik e ' ) which

can only be used to describe the discontinuation of conversations or

other dia logic forms o f communication.

153

The information carried by TO KNOCK IT OFF i s mainly contained in i t s

presuppositions. In the case of STAKEN the information about the nature

of the preceding discourse is supported by the valence of the verb: the

su bject1 i s necessarily plural and the object has to be a substantive

denoting a dialogic form of discourse.

64. Following discourse i s as much a matter of context as preceding

discourse. These two aspects of context combined y ie ld three subclasses

of contextualized acts of being s i l e n t .

(Al) Sa discontinues discourse

(A2) Sa discontinues discourse which i s to be continued

(A3) Sa i s s i l e n t before starting discourse

These three acts create several possible semantic patterns for verba ta ­

cendi. Not only are there three patterns corresponding to (Al), (A2) and

(A3), respectively, but also the following ones: (Al) or (A2); (Al) or

(A3); (A2) or (A3); (Al) or (A2) or (A3). But surprisingly only (Al),

(A2) and ((A2) or (A3)) are lex ica l ized . An obvious (Al) verb is TO FI­

NISH, Dutch OPHOUDEN (MET SPREKEN). Most verba cessandi belong to this

type. The (A2) pattern is only represented by TO PAUSE, Dutch PAUZEREN.

The most striking gap in the lexical frame i s the absence of (A3) verbials ,

which would form the logical counterpart to the verba cessandi. (A3) is

only represented as one possible meaning of the ((A2) or (A3)) verb TO

HESITATE, Dutch AARZELEN. The absence of (A3) verbials re f lec ts the fact

that pre-discourse s i lence i s not noticed unless the hearer i s waiting

154

for the conversational tacens to ta lk , whereas post-discourse s i lence

i s always noticed, whether more discourse i s expected to follow or not.

The lex ica l frame under discussion i s represented in SDC-Table 11.

65. Not only verba cessandi provide us with information about preceding

discourse. For instance, both TO REFUSE COMMENT and TO SAY NEITHER YES

NOR NO presuppose preceding discourse. One cannot 'refuse comment'

unless somebody asked to comment on some statement of s ta te of a f fa ir s .

And one cannot 'say neither yes nor no' unless somebody asked a question

or made a proposal. Note, however, that in contrast with the verba

cessandi, these two verba reticendi do not presuppose discourse on the

part of the conversational tacens himself , but on the part o f somebody

e l s e .

More examples are hard to find. Thus i t seems that there are no verba

si lendi which carry contextual information.

-j----- DUTCH SEMANTIC DIMf:nsion ENGLISHr 1- rr / ophouden | ( a i ) 1 to f in ish

/ L 0- 3 \ •

i10 01 pauzeren (A2) to pause 0

I0

\ aarzelen - - to h es i ta te Ii

T 0 (A3) 0/

L L - -

SDC-Table 11

155

L. THE ILLOCUTIONARY FORCE OF SILENCE

66. A l a s t , quite marginal, dimension of the verba tacendi i s the inform­

ation they carry about the i l locu t ionary force of the absent speech. Both

TO FAIL TO MENTION and TO REFUSE COMMENT are c lear ly concerned with

statement- l ike ac ts . The verbials TO REFUSE COMMENT and TO SAY NEITHER

YES NOR NO also provide information about the i l locu t ionary force o f the

preceding discourse. In the case of TO REFUSE COMMENT, a comment must

have been asked for. In the case of TO SAY NEITHER YES NOR NO the i l l o ­

cutionary force o f the absent speech described even depends on the force

of the preceding discourse: i f a question precedes, then the absent speech

i s s tatem ent- l ike; i f a proposal precedes, then acceptance and non-

acceptance i s referred to by means of the 'yes' and the 'no'.

Once again, more examples are rea l ly hard to f in d , which j u s t i f i e s our

claim about the marginality o f th i s dimension o f the verba tacendi.

M. CONCLUSIONS

67. The discussions in th is chapter should enable us to make detai led

semantic comparisons between 'equivalent' verba tacendi in English and

in Dutch (and probably in many other languages as w e l l ) by comparing

th e ir pos i t ions with respect to each o f the semantic dimensions in ­

vest iga ted . In most cases additional dimensions of meaning w i l l have

to be taken into account. But I tru s t that the most important ones for

the characterization of the contrasts and s i m i la r i t i e s between the se ts

156

of verba tacendi in English and in Dutch have been dealt with.

68. I t i s high time we wondered about the overall s ign if ican ce of th is

chapter on the semantics of s i l e n c e . Undoubtedly i t has revealed the

importance o f s i l en ce as an aspect o f l in g u i s t i c action (and hence I

hope i t may function as an incentive for further s i l e n c e -r e se a r c h ) :

the whole structure o f language and l i n g u i s t i c action , from i t s sound

properties via i t s meaning to i t s contextual architecture , i s re f lec ted

in l in g u i s t i c s i l e n c e . In the process of uncovering the l in g u i s t i c

structure o f the absence of speech we have made numerous noteworthy

observations, some o f which were merely confirmations o f pre-theoret ica l

in t u i t i o n s , others being quite surprising , and s t i l l others being

downright mysterious. Here i s a partial l i s t .

( i ) The language dimension of the basic verba tacendi reveals the con­

ducting of a conversation and the del ivering of a speech as the two

fundamental communication types in the experience of l i n g u i s t i c

action. (Paragraph 7)

( i i ) The absence o f speech as such ( i . e . the domain o f the verba s i l e n d i )

occupies the central pos it ion in our conceptualization o f l in g u i s t i c

s i l en ce (as appears from SDC-Table 1) . (Paragraph 9)

( i i i ) The verba tacendi reveal the primacy o f the oral mode of communi­

cation. (Paragraphs 11 to 18)

( iv ) The verba tacendi which are neutral between oral and written communi­

cation are never verba s i l e n d i . (Paragraph 13)

(v) The metaphorical extension of the meaning o f verba tacendi from the

oral to the written code i s ir r e v e r s ib le . (Paragraph 17)

157

(v i ) There i s a correlat ion between the code of the acts to be described

and the code of the describing ac ts . (Paragraph 14)

( v i i ) The formation of verbials can determine the ir a p p l ic a b i l i ty . (Para­

graph 15)

( v i i i ) The absence o f speech i s gradable along the two sound-related

dimensions of in te n s i ty and duration, but the superlat ive i s the

only point o f the gradation sc a le that i s rea l ly emphasized. (Para­

graph 20)

( ix ) The verba tacendi reveal a t t i tu d es towards the expression o f pleasant

versus unpleasant f ee l in g s or emotions. (Paragraphs 32 and 34)

(x) The ta lkat ive character of western culture i s demonstrated repeated­

ly and in various ways.

The l i s t i s completely random and most a t ten t iv e readers wil l be able to

supplement i t .

69. Apart from the fa c t that English does not whereas Dutch does possess

simple l e x ic a l i z a t io n s as basic verba tacendi and a quite general basic

term to cover the domain of l i n g u i s t i c s i l e n c e , most of the d if ferences

between the English and the Dutch se ts o f verba tacendi are o f minor

importance, though not uninterest ing . The strong paral le l ism was, of

course, predictable . I t i s necessary to carry out further comparative

invest igat ions in order to find out which ones o f our conclusions are

language-specif ic and which ones stand a good chance of being universal .

158

IV. THE SEMANTICS OF LYING

A. INTRODUCTION

1. After surveying the marginal area of s i l e n c e , we shall now attempt

to penetrate one of the cores of l i n g u i s t i c ac t ion , namely the domain

of the propositional content. We shall do so by way of analyzing the

verbials describing l in g u i s t i c actions performed by a speaker who pre­

sents the propositional content o f his speech act as r e f le c t in g a true

sta te of a f f a ir s but who knows or be l ieves that i t deviates from the

truth.

Though no further arguments are needed for our decis ion not to engage

in an endeavor to c l a s s i f y l in g u i s t i c actions or l in g u i s t i c action verb­

i a l s , we are now confronted with a part icu lar ly compelling one which we

should not neglect to mention. Even between the extremely marginal do­

main of s i l e n c e and the quite central area o f propositional content there

i s no sharp boundary. Not only can acts of being s i l e n t about something

be occasionally regarded as ly in g , but there are a lso a number of l i n ­

g u i s t i c action verbials which c le a r ly cover portions of both domains

without being ambiguous between two completely d i s t in c t senses . TO DIS-

159

GUISE, for instance, i s not only a verbum reticendi describing an act of

being s i l e n t about some particular topic or f a c t , but disguising can be

done by lying. Suppose there i s a p o l i t ic o -r e l ig io u s movement called

Communists for Christ, which people are so suspicious of that i t i s forced

to s tar t operating through cover organizations, one of them being the

Creative Thought Project. I f a communist for Christ introduces himself

as a member of the Creative Thought Project , then he i s not ly ing, but

he i s simply disguising his ident ity by being s i l e n t about i t (which he

can keep doing by assuming a d if ferent ident i ty which he hopes wil l not

be associated with the 'real' one). Suppose, furthermore, that i t i s a

well-known fact that Communists for Christ do not accept the authority

of the Pope. I f , then, our Communist for Christ claims that he accepts

the authority of the Pope, then he i s disguising his real convictions by

ly ing. And i f he makes the same claim after being asked whether he i s a

Communist for Christ, then he i s disguising his ident i ty by being s i l e n t

about i t and by t e l l in g a l i e . Similarly, TO COVER UP describes acts

which often combine s i lence about a particular tacendum (e .g . a crime)

with a l i e (e .g . a fa l se a l i b i ) . Such marginal cases will be l e f t out of

the subsequent account.

2. Not a l l l in g u i s t i c actions in which the truth i s distorted would be

called l i e s . Within the scope of th is chapter we shall not only discuss

lying proper, but also exaggerating, understating, d is tor t ing , blackening,

coloring, etc . Given this scope i t would be misleading to talk about

'verbials of ly in g ' . Therefore we introduce the term verb a m en tien d i

(from Latin MENTIRI meaning 'to l i e ' ) to cover the complete domain under

160

invest igation. The ' l ia r ' wil l be cal led the m en tio n s. And what i s being

' l ied ' about wil l be referred to as the mentiendum , plural m entienda .

Notice that in a case such as "Yesterday I went to the moon" the men­

tiendum includes, s t r i c t l y speaking, both the tr ip to the moon and y e s ter ­

day's a c t i v i t i e s . Which one of the two aspects i s focused on wil l depend

on the context.

Remember that the set of verba tacendi included some negations of

non-silence l in g u is t i c action verbials such as NOT SPEAK and NOT TALK.

One similar formation wil l be encountered in th is chapter, namely NOT

TELL THE TRUTH, which wil l be regarded as a complex lex ica l iza t io n

because i t i s a conventional means of talking about untruthful l i n ­

g u i s t i c acts . No comparable lex ica l iza t io n s will occur in the following

chapters.

Whereas ostensive def in i t ions were impossible in the previous chapter

because acts o f being s i l e n t cannot be reproduced on paper, they wil l be

used from now onwards.

B. THE FRAME OF LUNG

3. The l in g u i s t i c actions described by the verba mentiendi involve a con­

scious distort ion of the truth. The notion of truth i s ambiguous between

truth as such ( i f i t e x i s t s ) and what the speaker knows, bel ieves or

thinks to be true. Given our lex ica l approach, a more tangible contrast

i s that between what i s true for the user of a verbum mentiendi and what

i s true for the performer of the act described. I f John bel ieves that the

earth i s round and Jim believes that i t i s f l a t , and John i s aware of

161

Jim's b e l i e f , then John wil l not describe Jim's claim that the earth is

round as a l i e (even though the propositional content deviates from Jim's

b e l ie f s ) ; instead, John will have to say that Jim told the truth (even

though John realizes that Jim was trying to convey misleading information

about his own world of b e l i e f s ) . Such attempts at making an untrue s ta te ­

ment are not describable in terms of a verbum mentiendi. Thus, a verbum

mentiendi i s appropriately used only i f the describing act D s a t i s f i e s

conditions (Cdl) and (Cd2), and i f A, the act described, s a t i s f i e s (Cl).

(Cl) Sa believes that what he presents as a true state of a ffa irs

i s a distortion of the truth

(Cdl) Sd believes that what i s presented in A as a true state of

a f fa ir s , i s a distortion of the truth

(Cd2) Sd believes that (Cl) obtains

These three conditions define the lexical frame of lying. As a resu lt ,

the verba mentiendi do not include verbs such as TO ERR which denote

untruthful ness for which the speaker is not (completely) reponsible, nor

verbs such as TO FLATTER and TO BOAST which may but need not imply de­

viations from the truth.

4. In the previous chapter we claimed that the verba tacendi, because of

the marginality of s i lence as an aspect of l in g u is t ic action, reflected al l

the najor aspects of l in g u is t ic action. In other words, the frame of

s i lence was said to be the frame of l in g u is t ic action i t s e l f . This e-

merged very c learly from the sa l ien t role of what was called the language

dimension of s i lence . Thus studying the semantics of s i lence proved to

162

be relevant as a way of gaining insight into our experience of language.

In a similar way, studying the semantics of lying may y ie ld ins ight

into truth as an aspect of language (or as an aspect of the propositional

content of utterances) . The most important semantic dimension of the

verba mentiendi will be shown to be the truth dimension. In other words,

the frame of lying i s the frame of truth i t s e l f .

C. LYING AND TRUTH

5. A natural expectation of what a truth dimension would look l ike could

be in terms of the gradabil i ty of the notion of truth. One could expect

there to be a continuum of verba mentiendi ranging from those indicating

a f l a t falsehood to those indicating a minor deviation from the truth.

Of course such gradabil ity i s ref lected in the se t of verba mentiendi

as appears from the contrast between TO LIE and TO COLOR (and the corres­

ponding Dutch verbs LIEGEN and KLEUREN). However, the contrast i s rarely

th is clear since in most cases t o t a l ly d i f ferent semantic features play

an important role . For instance, the fact that TO FIB appears to be

weaker than TO LIE i s not due to the gradability of the notion of truth;

rather, i t resu lts from the type of falsehood involved: in the case of

TO FIB, a t r iv ia l falsehood i s referred to . On the other hand, TO LIE

LIKE A TROOPER appears to be stronger than TO LIE, but th is strength usu­

a l ly derives from the frequency implied rather than from the kind of

gradabil ity under discussion. Consequently, i f there were only the grada­

b i l i t y of truth, there would be very l i t t l e to say about the truth d i­

mension of lying.

163

Nevertheless, i t i s worth comparing, in passing, the se ts of verbials

available in English and in Dutch to describe highly intense acts of

ly ing. Whereas the English repertoire does not go much further than

TO LIE FLATLY and TO LIE LIKE A TROOPER, in Dutch we find considerable

variation: LIEGEN ALS EEN KETTER ( l i t . ' t o l i e l ike a h e r e t i c ' ) , LIEGEN

ALS EEN ALMANAK ( l i t . 'to l i e l ik e an almanac'), TEGEN DE KLIPPEN/STERREN

OP LIEGEN ( l i t . 'to l i e up against the c l i f f s / s t a r s ' ) , LIEGEN ALSOF HET

GEDRUKT STAAT ( l i t . ' to l i e as i f i t were pr in ted ') , LIEGEN DAT MEN HET

ZELF GELOOFT ( l i t . ‘to l i e so that one believs i t o n e s e l f ) , LIEGEN DAT

MEN ZWART/SCHEEL ZIET ( l i t . 'to l i e so that one looks black/cross-eyed'} .

6. Apart from verbials such as TO LIE, TO COLOR, TO FIB, TO LIE LIKE A

TROOPER and many others, a l l of which simply indicate that the truth i s

being d is torted , there are verba mentiendi which give information about

how the truth i s being distorted. The la t t e r can be placed at the extremes

of two s c a l e s , the quantity scale and the quality scale .

The se t of verba mentiendi r e f lec t s four ways of distort ing the truth.

F irst , i t i s possible to make something look 'smaller' than i t i s . Second,

one can make something look 'bigger' than i t i s . The basic verba mentiendi

situated at these two poles of what I call the q u a n ti ty s c a le o f tr u th

are TO UNDERSTATE (which does not always describe the figure of speech 'un­

derstatement') and TO EXAGGERATE. We shall ca l l the f i r s t one a q u a n ti ty

d im in ish in g v e r b , the second one a q u a n ti ty in c r e a s in g v e r b . It i s in te r ­

esting to note that Dutch lacks a basic equivalent to TO UNDERSTATE though

there i s a quantity diminishing verb such as VERKLEINEN (the adjective

KLEIN means 'sm all ' ) ; however, we refrain from regarding th is one as a

164

quantity diminishing verbum mentiendi because i t i s often qual ity

diminishing. Consequently, we seem to be confronted with a quite sur­

pris ing l ex ica l gap in the Dutch se t of basic verba mentiendi. The Dutch

equivalent of TO EXAGGERATE i s OVERDRIJVEN.

The two remaining ways o f d is tor t ing the truth co n s is t in making something

look 'better* than i t i s and making something look 'worse' than i t i s .

The basic verba mentiendi s i tuated at these two poles o f the q u a l i t y

s c a le o f t r u th are TO WHITEWASH and TO SLANDER (Dutch GOEDPRATEN and

(BE)LASTEREN); We shall ca l l the f i r s t one a q u a l i t y in c r e a s in g v e rb

and the second one a q u a l i t y d im in ish in g v e r b .

The f iv e types of verba mentiendi discussed so far are presented in

Figure 1. TO LIE, TO COLOR, e tc . are s i tuated in the middle, with arrows

directed at the two poles o f the qual ity and quantity sca les because

some instances of d is tor t ing the truth along these sca les can be described

in terms of ly in g , co lor ing, e tc .

7. The qu a l t i ty and qual i ty sca les o f truth seem to show a curious one-

direct ional interdependence relat ionship ( indicated by means o f the

broken arrow in Figure 1) . I t i s possible to whitewash somebody's 'bad'

behavior by making an understatement about i t or even by exaggerating i t s

p o s i t iv e aspects . S im ilar ly , one can slander someone by exaggerating

his v ices (though not as e a s i l y by making understatements about his

v ir tu e s ) . But one cannot make an understatement by whitewashing, nor

exaggerate by slandering. Why i s the qual i ty sca le dependent on the

quantity sca le and not v ice versa ? Probably because i t i s not possible

Quantity sca le o f tru th < -

Eng. to understate

Eng. to l i e

to color etc .

Du. liegen kleuren etc .

Eng. to exaggerate

Du. overdrijven

+

F igure 1

Quality sca le o f tru th

+

V

Eng. to slander

Du. (be)lasteren

Eng. to whitewash

Du. goedpraten

166

to take qual ity as a cr i ter ion for quantity, whereas in our culture quan­

t i t y i s continuously taken as a cr i ter ion for qual ity . This explanation

would be e n t ir e ly in keeping with the MORE IS BETTER value judgment

which Lakoff and Johnson (1979) discover in some chains o f metaphors

permeating the English language.

8. Once we leave the domain o f the b a s ic verba mentiendi we discover a

relat ionship between the l ik e poles of the sca les which the opposite

poles do not have. Indeed, the s e t s of qual ity diminishing and quantity

diminishing verbs share one member in Dutch, VERKLEINEN, though none in

English. S imilarly , TO DRESS UP and TO EMBROIDER, as well as the Dutch

BORDUREN ( ' t o embroider') belong to the quantity increasing and qual ity

increasing verbs.

Just as the d ir e c t io n a l i ty o f the dependence re la t ionsh ip discussed

in the previous paragraph, the verba mentiendi enumerated here show that

quantity i s regularly taken as a cr i ter ion for qual ity . This s ta te of

a f fa ir s i s r e f lec ted even more c lear ly in the verbs TO BELITTLE and

KLEINEREN which mean l i t e r a l l y 'to make small' (the adject ive KLEIN

meaning 'sm all ' ) but which can o n ly be used as q u a l i t y diminishing verbs.

9. After indicating the main l in es of force of the truth dimension of

ly in g , our next task i s to make a more deta i led study o f the English and

Dutch verba mentiendi along th i s dimension. There are a number o f gene­

ral conditions on the standard appropriate use o f the verba mentiendi

which wil l be not be repeated for each verbial in the res t of th i s chapter.

167

The conditions in question are (C l) , (Cdl) and (Cd2), which have already

been mentioned in the previous sect ion to define the lex ica l frame of

ly ing .

(Cl) Sa be l ieves that what he presents as a true s ta te o f a f fa ir s

i s a d is tor t ion o f the truth

(Cdl) Sd believes that what i s presented in A as a true s t a t e of

a f f a i r s , i s a d is to r t ion o f the truth

(Cd2) Sd bel ieves that (Cl) obtains

Consequently, when we ta lk about an act (A) conforming to (C) and des­

cribed by means o f (D), we mean that a lso (C1‘ ) , (Cdl')and (Cd2‘ ) obtain.

(A) The earth i s f l a t

(C) The earth i s round

(D) Sa l ied

(C l1) Sa be l ieves that the earth i s round

(Cdl1) Sd bel ieves that the earth i s round

(Cd2‘ ) Sd be l ieves that (C l1) obtains

In that way redundant repet i t ions can be avoided.

One may be inclined to bel ieve that (Cl) and (Cd2) do not have to obtain

in order for a d is tor t ion o f the truth to be appropriately described by

means o f certain verba mentiendi such as TO EXAGGERATE. However, the v a l i d i ­

ty of my general claim i s probably s u f f i c i e n t ly demonstrated by pointing

out that an adequate response to the accusation "You are exaggerating" is

"No, I r ea l ly bel ieve what I am saying".

168

10. A simple case of deviating from the truth i s {A1) i f uttered in c i r ­

cumstances in which (C2) obtains.

(Al) John was home at the time of the robbery

{C2) In f a c t , John was not home at the time of the robbery

In any D describing (Al) conforming to (C2) one could use the verba men-

t iendi TO LIE, TO LIE FLATLY, TO TELL A LIE, NOT TELL THE TRUTH, TO SPEAK

FALSELY, and in Dutch LIEGEN ( ' t o l i e ' ) , LEUGENS VERKOPEN ( l i t . 'to s e l l

l i e s ' ) , LEUGENS VERTELLEN { ' t o t e l l l i e s ' ) , DE WAARHEID GEWELD AANDOEN

( l i t . ' to v io la te the t r u t h ' ) , DE WAARHEID NIET ZEGGEN ( 'not t e l l the

t r u t h ' ) . Let us ca l l these v e r b i a l s o f l y in g .

All the (A1)-(C2) verbia ls mentioned, except TO LIE FLATLY, are also

applicable to (Al) when i t conforms to (C3).

(C3) John was almost home at the time of the robbery ( in fa c t c lose

enough to be sure that he could not have been at the scene of

the crime when the crime was committed)

In addition to the (A1)-(C2) v erb ia ls , a large subset of the verba men-

t i e n d i , which cannot be applied to (Al) when (C2) obtains, are per fect ly

f i t to describe (Al) when (C3) app l ie s , or at l e a s t to acts o f a s im ilar

type: TO DISTORT, TO COLOR, TO BEND THE TRUTH, TO DEVIATE FROM THE TRUTH,

TO GIVE A COLOR TO, TO GIVE A FALSE COLORING, TO MISREPRESENT, TO PREVA­

RICATE, TO PUT A FALSE APPEARANCE UPON, TO PUT IN A FALSE LIGHT, TO SLANT,

TO STRAIN THE TRUTH, TO STRETCH THE TRUTH, TO TWIST, TO WARP; and in

Dutch KLEUREN ( ' t o c o l o r ' ) , IN EEN VALS DAGLICHT STELLEN ( l i t . 'to put

in a f a l s e d a y l ig h t ' ) , VERDRAAIEN ( ' t o t w i s t ' ) , VERKEERD VOORSTELLEN

( ' t o m isrepresent ') , VERWRINGEN ( ' t o t w i s t ' ) , EEN VERWRONGEN VOORSTELLING

169

GEVEN VAN ( l i t . 'to give a distorted representation o f ) , VAN DE WAARHEID

AFWIJKEN ( ' to deviate from the tru th ' ) , DE WAARHEID TE KORT DOEN ( l i t .

'to wrong the tru th ' ) , DE WAARHEID VERDRAAIEN/VERWRINGEN/VERKRACHTEN

( l i t . 'to twist/warp/rape the tru th ') . These are called the v e r b ia ls o f

d i s t o r t i n g .

In addition to the verbials of lying and dis tort ing , there are a l l the

quality and quantity diminishing and increasing verbials which I shall

now go on to l i s t and i l lu s tr a te .

11. To start with the quality diminishing verba mentiendi, consider (A2)

uttered in a context which s a t i s f i e s (C4).

(A2) Mr. Kohl was a member of the Nazi party

(C4) Mr. Kohl was never a member of the Nazi party (though, perhaps,

he had some Nazi friends)

This act can be adequately described by means of TO SLANDER as in (Dl)

or (D2).

(Dl) Sa i s slandering

(D2) Sa i s trying to slander Mr. Kohl's good name

Similar quality diminishing verbials (not all of which necessarily in­

volve a distortion of the truth though they are usually associated with

i t ) are TO ASPERSE, TO BACKBITE, TO BAD-MOUTH, TO BESMEAR, TO BESMIRCH,

TO BLACKEN, TO CALUMNIATE, TO CAST A SLUR ON, TO CAST ASPERSIONS ON, TO

DEFAME, TO DEFILE, TO DENIGRATE, TO DISPARAGE, TO DRAG THROUGH THE MUD,

TO GIVE A BAD NAME, TO LIBEL, TO MALIGN, TO RUN DOWN, TO SLUR, TO SMIRCH,

170

TO SPEAK ILL OF, TO SPEAK SLIGHTINGLY OF, TO SULLY, TO TARNISH, TO TRA­

DUCE, TO VILIFY; and in Dutch ACHTERKLAPPEN {'to backbite' ) , BEKLADDEN

{ ' to besmirch1) , (BE)LASTEREN ( ' t o s lan der ') , BEZWALKEN ( ' t o besmear1) ,

DENIGREREN { ' to denigrate ') , IEMAND IN ZIJN EER/GOEDE NAAM AANTASTEN

( l i t . 'to injure a person in his honor/good name'), IEMAND OVER DE HEKEL

HALEN ( l i t . ' to pull someone across the hackle'; a hackle i s a board

with long metal teeth for dressing f lax or hemp; the expression means

'to talk about and judge a person in a very sharp, merciless and male­

volent manner' ) , KWAADSPREKEN ( l i t . 'to speak i l l ' ) , VAN IEMAND KWAAD

STOKEN ( l i t . 'to brew i l l about someone'), IEMAND LELIJK/SLECHT MAKEN

( l i t . 'to make someone ugly/bad') , DOOR DE MODDER SLEUREN ( ' to drag

through the mud'), DOOR HET SLIJK HALEN ( ' t o drag thorugh the mud'),

IEMAND ZWART MAKEN ( ' t o balcken someone').

There are two aspects of the lex ica l frame associated with acts of the

(A2)-(C4) type which r e f l e c t one of the profoundest, though often un­

conscious, value judgments inherent in our l in g u is t i c culture. F ir s t ,

apart from TO CRITICIZE (Dutch BEKRITISEREN) I cannot think of any l i n ­

g u i s t i c action verbial , neither in English nor in Dutch, which contains

the meaning 'to say something bad or unfavorable about someone or some­

thing' , and which i s not, in a l l or most o f i t s uses, associated with

untruthful ness. This i s in sharp contrast with the large s e t of verbials

meaning 'to say something good of favorable about someone or something1

without being associated with a lack of truthfulness (such as TO PRAISE,

TO LAUD, TO GLORIFY, and many others) . Second, whereas the verbials of

distort ing can be used to cover the domains of the quality and quantity

diminishing and increasing verbials (as w i l l be clear from what fo l low s) ,

171

the verbials o f lying can be e a s i ly extended to the conceptual domain of

the quality diminishing verbials ( i . e . the acts o f the (A2)-(C4) type)

but not so e a s i ly — and in most cases not at a l l — to the area of the

qual ity increasing and the quantity diminishing and increasing l in g u i s t i c

action verbia ls . Both o f these facts reveal a strong associat ion of

'saying bad things about' with 'saying u n tru e things about'. Thus the

lexicon r e f l e c t s a usually unconscious value judgment which i s en t ire ly

in keeping with Freud's (1917:48) observation that "society makes what

i s disagreeable into what i s untrue” . How e l s e can we explain that an

expression such as TO DISH THE DIRT has come to assume the meaning of

goss ip ing, which i s strongly associated with untruthful ness ? Dishing

the d i r t , even i f the d ir t i s very r e a l , i s rarely appreciated, and

therefore the truth i s branded as untrue.

12. The quantity diminishing verb TO UNDERSTATE can be used to describe

such acts as (A3) conforming to (C5).

(A3) Once or twice a year Jim sends some money to the World Health

Organization

(C5) In f a c t , at l e a s t twice a year Jim sends a complete monthly

allowance to the World Health Organization

Another quantity diminishing verb, with a quite s p e c i f i c meaning, i t TO

UNDERREPORT in the sense o f reporting ( e .g . to the Internal Revenue Ser­

vice ) an amount ( e .g . as one's income) l e s s than the actual one. In

Dutch there seem to be no purely quantity diminishing l in g u i s t i c action

172

verbials . There i s , however, the verb VERKLEINEN, which i s quantity d i ­

minishing in form but which can be used e ither as a quantity diminishing

verb as in (D3) describing (A3)-(C5) or as a quality diminishing verb

as in (D4) describing (A2)-(C4).

(D3) Sa tracht Jims bijdrage te verkleinen

(Sa i s trying to make Jim's contribution jjlook} smaller')

(D4) Sa tracht Mr. Kohls verdiensten a ls na-oorlogs s o c i a l i s t te

verkleinen

('Sa i s trying to b e l i t t l e Mr. Kohl's merits as a post-war

s o c i a l i s t ' )

Apparently English lacks a verbial which can e i th er be quantity ov

quality diminishing. TO BELITTLE can be used to describe some acts of the

(A3)-(C5) type, as in (D5).

(D5) Sa i s trying to b e l i t t l e Jim's contribution

But in such cases i t i s a quantity and quality diminishing verb at the

same time.

13. Quantity increasing verbs such as TO EXAGGERATE and i t s Dutch equi­

valent OVERDRIJVEN can be applied to acts such as (A4) conforming to

(C6).

(A4) Jim donates ha lf his income to UNESCO

(C6) In fa c t , Jim sends a small contribution to UNESCO twice a year

Other quantity increasing verbials are TO AGGRANDIZE, TO BLOW UP, TO

DRAW THE LONGBOW, TO HYPERBOLIZE, TO MAGNIFY, TO TALK IN SUPERLATIVES,

173

TO OVERLAUD, TO OVERPRAISE, TO OVERSTATE, TO OVERSTRESS; and in Dutch

AANDIKKEN ( l i t . 'to thicken') , CHARGEREN ( ' to exaggerate in order to

r id ic u le ' ) , EXAGEREREN { 'to exaggerate'; rarely used), OPBLAZEN {'to

blow up'), OPSNIJDEN ( l i t . 'to cut up') , OPSCHROEVEN ( l i t . 'to screw

up') , ER EEN SCHEPJE OP DOEN ( l i t . 'to add a l i t t l e scoop1) , MET SPEK

SCHIETEN ( l i t . 'to shoot with bacon'), VERGROTEN ( l i t . 'to enlarge') ,

VAN EEN SCHEET EEN DONDERSLAG MAKEN ( l i t . 'to make a fart into a thun­

derbolt1) , VAN EEN VLIEG/MUG/MUIS EEN OLIFANT MAKEN ( l i t . 'to make a

fly/mosquito/mouse into an elephant') .

Notice that i t is possible to 'make something look bigger' , i . e . to

exaggerate something, by making i t look n u m erica lly smaller. An example

would be the utterance "In Belgium the sun shines once or twice a year".

14. There are two different types of quality increasing verba mentiendi.

The f i r s t type, represented by TO EMBELLISH in English and i t s Dutch

counterparts VERFRAAIEN and AANKLEDEN ( l i t . 'to dress ' ) , i s used to

describe acts such as (A5)-(C7).

(A5) We arrived in Los Angeles in the morning. The same day we

v is i ted Hollywood where we happened to see Peter Falk and

Jane Fonda walking around

(C7) Sa v is i ted Hollywood, but he did not manage to see any well-

known moviestar

About Sa, who is inventing deta i ls to make his story nicer , one can say

174

(D6) As usual, 5a i s embellishing the story of his vacation

TO EMBELLISH, VERFRAAIEN and AANKLEDEN describe acts of making something

look nicer or better which may already have been good or nice to begin

with.

A second type of quality increasing verba mentiendi i s used to describe

acts of making something look better which was not good at a l l to s tart

with. Consider (A6)-(C8).

(A6) Jack's steal ing i s ju s t a natural outcome o f the deprivation

he suffered during his childhood

(C8) Jack i s a t h i e f (who may or may not have suffered deprivation

during his childhood)

This type of act i s described by means of TO WHITEWASH, TO DEODORIZE,

TO GILD, TO GLOSS (OVER), TO MAKE SMELL LIKE ROSES, TO VARNISH and the

Dutch GOEDPRATEN ( l i t . 'to good-ta lk ' ) , BEWIMPELEN ( l i t . 'to cover with

a f l a g ’ ) , MOOIPRATEN ( l i t . ' to n i c e - t a l k ' ) , VERGULDEN ( ' t o g i l d ' ) ,

VERBLOEMEN ( l i t . 'to turn into a f lo w er ' ) , VERGOELIJKEN ( l i t . 'to make

good') , and WIT WASSEN ( l i t . 'to wash white'; not to be confused with

'to whitewash').

We have mentioned before that TO DRESS UP, TO EMBROIDER and the Dutch

BORDUREN ( ' t o embroider1) are quantity and quality increasing verba men­

t ien d i . Notice, however, that they do not cover the complete domain of

the quality increasing verbials but only the area o f the (A5)-(C7) sub-

type. Also notice the curious fact that the Dutch verb AANKLEDEN ( l i t .

'to dress ' ) i s mainly quality increasing and n o t both quality and

kleuren

-DUTCH- -SEMANTIC DIMENSION- ENGLISH-

• liegen

\f_ ver­kleinen

borduren

(be) las- teren .

0

overdrijven

verfraaienaankleden

goedpraten

(A1)-(C2

(A1)-(C3

[(A2)-(C4

[(A3)-(C5

[(A4)-(C6

[(A5)-(C7

[(A6)-(C8

J to slander

J to understate

J to exaggerate

J to embellish

j to whitewash

to l i e

to dress up to embroider

to d i s to r t

SDC-Table 1

176

quantity increasing as i t s l i t e r a l English equivalent TO DRESS UP.

The fa c t that a d is t inct ion had to be made between (A5)-(C7) and

(A6) - (C8) whereas no similar s p l i t was to be found in the se t of qua­

l i t y diminishing verbials (though i t i s quite conceivable to distinguish .

between making good things look bad and making bad things look worse)

i s another l in g u i s t i c ref lec t ion of the habit to regard unpleasant

things as untrue.

The preceding observations on the truth dimension of lying are summar­

ized in SDC-Table 1.

D. THE ILLOCUTIONARY FORCE OF LYING

15. The t i t l e of th is section i s misleading in the sense that lying

cannot be said to possess an i l locutionary force of i t s own, Lying i s

simply a matter of performing speech acts the propositional content

of which deviates from the truth in any of several possible ways. How­

ever, a l l examples of verba mentiendi mentioned so far described acts

of the i l locutionary type (Al).

(Al) Sa makes a statement

Therefore, though they cannot be said to have an i l locutionary force

of the ir own, they are not neutral with respect to i l locutionary force:

in Searle 's terminology, the point o f the acts described i s always to

represent a certain s ta te of a f fa ir s . This r es tr ic t io n to statement-like

speech acts i s not surprising. Though s incer i ty plays an important part

with respect to other types of speech acts such as d irec t iv e s , commissives

177

and express ives , questions o f truth, i . e . questions of the correspondence

between ' r e a l i t y ' and propositional content, are l e s s relevant.

Yet there i s another s e t o f l in g u i s t i c action verbials which are c lo s e ly

related to the verba mentiendi mentioned so far but which describe acts

of the type (A2), i . e . there i s no i l locu t ionary type sp ec i f ie d by the

verbials themselves.

(A2) Sa performs (a) speech a c t ( s ) with any i l locut ionary force

They can be regarded as verba mentiendi in so far as they can be used

to describe statement- l ike utterances deviating from the truth; when they

refer to acts of a d i f f e r n t i l locu t ionary type they cease to be verba

mentiendi because no deviation from the truth i s involved though s in ­

cer i ty i s t r i f l e d with in some other way. Here i s a partial l i s t : TO

PUT ON AN ACT, TO PUT UP A FRONT, TO PUT ON A ( FALSE)FRONT, TO ACT, TO

ACT/PLAY A PART, TO ACT/PLAY THE HYPOCRITE, TO CANT, TO AFFECT, TI ASSUME,

TO SHAM, TO SIMULATE, TO DISSEMBLE, TO DISSIMULATE, TO FAKE, TO FEIGN,

TO GAMMON, TO LET ON, TO MAKE A PRETENSE, TO MAKE A SHOW OF, TO MAKE FALSE

PRETENSES, TO POSSUM, TO PRETEND; and in Dutch AFFECTEREN ( ' t o a f f e c t ' ) ,

DOEN ALSOF ( ' t o act as i f ' ) , ZICH VAN DE DOMME HOUDEN ( l i t . ' t o hold one­

s e l f s tu p id ' , meaning ' to act as i f one does not know about something') ,

GEBAREN ( l i t . 'to g e s tu r e ' , meaning 'to l e t o n ' ) , HUICHELEN { ' t o f e i g n ' ,

often 'to d issem ble') , ONTVEINZEN { ’to d issem ble ' ) , IN ZIJN ROL BLIJVEN

( l i t . 'to s tay in one's r o l e ' ) , EEN ROL SPELEN ( ' t o act a p a r t ' ) , SIMU-

LEREN ( ' t o s im ula te ') , VEINZEN ( ' t o f e i g n ' ) , VOORGEVEN ( ' t o pretend') ,

VOORWENDEN ( ' t o pretend') , FEMELEN ( ' t o c a n t ' ) , KWEZELEN { ' t o c a n t ' ) ,

DE HYPOCRIET UITHANGEN ( ' t o play the h y p o cr i te ' ) , e tc . Whenever these

178

verbials are discussed further on, only their (Al) meaning w i l l be referred

to.

16. A number of verba mentiendi are applicable to certain sp e c i f ic sub-

types of statement-like acts . One o f these subtypes i s the act of a ccu sin g

which i s of the type (Al) conforming to (Cl)

(Cl) the propositional content of Sa's utterance ascribes an act

or property, which i s generally regarded as 'bad1 or ’wrong1,

to a person

An (Al)-(Cl) verbial would be TO IMPUTE (something to a person); in Dutch

there are the verbials AANWRIJVEN, AANTIJGEN and IEMAND IETS IN DE SCHOENEN

SCHUIVEN ( l i t . 'to shove something into somebody's shoes' , meaning 'to

accuse u n just ly ' ) . About a l l quality diminishing verbia ls , though their

meaning i s wider, can be used in the sense o f accusing unjustly.

17. The opposite o f accusing i s p ra is in g . There are two types of praising

which are covered by sp e c i f ic verba mentiendi. The f i r s t type i s (Al)

conforming to (C2)

(C2) the propositional content of Sa's utterance ascribes an act

or property, which i s generally regarded as 'good' or ' r ig h t 1,

to a person or object

Some (A1)-(C2) verba mentiendi are TO OVERLAUD, TO OVERCOMMEND, TO OVER­

PRAISE and the Dutch OVERROEPEN ( l i t . 'to overshout') . The second type

of praising unjustly i s (Al) conforming to (C2) plus (C3).

179

{C3) the person about whom something 'good' i s said or who has an

immediate in te r e s t in the s ta te of a f fa ir s talked about, i s

(one of) Sa's hearer(s)

Relevant verbials here are TO PLAY UP TO and the Dutch IEMAND NAAR DE

MOND PRATEN ( l i t . 'to ta lk to someone's mouth' meaning 'to say what he

l ik es to hear, which ,akes you wonder why the expression i s not 'to talk

to someone's e a r ' ) . Verba mentiendi which are neutral between (A1)-(C2)

and {Al) - { C2+C3) are TO GIVE MOUTH HONOR TO, TO RENDER/GIVE/PAY LIP

SERVICE and the Dutch EEN LIPPENDIENST BEWIJZEN ( ' t o render l ip s e r v i c e 1).

I t i s worth noting that whereas most qual ity diminishing verbials such

as TO CAST ASPERSIONS ON can be extended from a neutral (Al) to ( A l ) - (C l ) ,

i t i s not possib le to extend the basic (Al) meaning o f qu a l i ty increasing

verbials such as TO WHITEWASH into the domain o f praising unjustly. On

the other hand, the qua l i ty increasing verb TO EMBELLISH as well as most

quantity increasing verbs such as TO EXAGGERATE can be extended from (Al)

to (A1)-(C2) and even (A1)-(C2+C3). But the quantity diminishing verb

TO UNDERSTATE, ju s t as TO WHITEWASH, s t ic k s to the neutral (Al) meaning.

The interaction between the truth dimension and the i l locut ionary dimension

in Dutch i s completely analogous.

18. The third statement-l ike i l lo cu t ion a ry force tupe which some verba

mentiendi draw e x p l i c i t attention to i s sw e a r in g . In other words, (Al)

conforms to (C4) or (C5).

(C4) Sa i s taking an oath

(C5) Sa i s under oath

180

The Dutch EEN MEINEED AFLEGGEN ( ' t o commit perjury') i s r e s tr ic ted to

(A1)-(C4). TO BEAR FALSE WITNESS and i t s equivalent EEN VALSE GETUIGENIS

AFLEGGEN seem to be confined to (A1)-(C5); in some cases they do not

belong to the domain o f swearing at a l l . The English verbials TO PERJURE

ONESELF and TO COMMIT PERJURY are neutral between (Al)-(C4) and (A1)-{C5):

they can e i th er be used to describe an act o f swearing as such or an act

of making an untruthful statement under oath ( i . e . a statement made a f ter

swearing that what was to fo l low would be true) .

19. The i l lo cu t ion a ry dimension of the verba mentiendi, which i s re­

presented in SDC-Table 2, shows that the question of truth i s most r e l e ­

vant with respect to statement- l ike speech acts or representatives . Among

the representat ives , acts o f accusing, praising and swearing are focused

on ind iv idua l ly , which shows th e ir sa l ien ce . Praising untruthfully i s

sa l i e n t because because o f i t s frequent occurrence in soc ia l in teract ion .

Accusing and swearing are s a l i e n t because the question of truth i s absolute­

ly central to these two a c ts . This i s why the verb TO LIE and the other

verbials of lying can be e a s i ly applied to acts o f accusing unjustly and

swearing f a l s e l y , but not to overlauding or rendering l ip serv ice .

20. The c e n tr a l i ty of the question o f truth to representative speech acts

also emerges from the fa c t that the basic Dutch verb in th is conceptual

area, BEWEREN, i s often associated with untruthful ness . BEWEREN means

'to s ta te without proof or to maintain in sp i te of evidence to the con­

trary ' . In that sense the Dutch verb i s much c loser to TO CLAIM than to

TO STATE. There seems to be no Dutch representative the meaning o f which

-DUTCH- -SEMANTIC DIMENSION- -ENGLISK

CLOro3

(Uwo-h

fDC“SfD3

<fD O "S < -h fD^ s cu djCU -s■ - I t — I

fD3 < fD 3

1i ppen-dienstbewijzen

overroepen

iemand naar de mond praten

liegen

(be)-las-ter-en

goedpraten

£(A1)-(C2) J to overlaud

[(A1)-(C2+3)|

[ ™ ]to play up to

to give mouth honor to

to understate to whitewash

iemand i e t s r "Iin de schoe- (Al)-(Cl) to imputenen schuivenueen meineed afleggenvalse ge-tuigenisafleggen

[ ( A l ) - (CA)J

[ (A1)- (C5) ]

0to bearfa lsewitness.

to perjure oneself

(A2)

rU </)

cn CD

CU CD

SD C -Table 2

to pu

t on

an ac

t

182

i s as general and as neutral as the meaning of TO STATE and which i s not

associated with untruthful ness. Though the claim that th is shows something

about the c re d ib i l i ty of Dutch statements would be hardly credible i t s e l f ,

we are at l ea s t confronted with a curious lex ical gap. In practice , the

gap i s usually f i l l e d by the statement-specif ic use of the more general

l in g u i s t i c action verb ZEGGEN { ' to s a y ' ) , a use to which also the English

TO SAY can be put; another common way out i s to use VERKLAREN (English

'to d ec lare ' ) , but this i s c learly not as basic a word as TO STATE.

21. Lying i s a domain without e x p l i c i t performatives. This does not

only resu lt from the fact that lying has no i l locutionary force of i t s

own (though i t s domain i s restr ic ted to the representative i l locutionary

act type), but also from some simple pragmatic constraints on the use

of the verba mentiendi. Elsewhere (Verschueren 1977:13-14) I have shown

that an act of boasting would destroy i t s e l f i f the performative formula

'I boast' were to be used, and that a moral judgment would be passed on

an act of threatening by using the performative formula 'I threaten'.

Similarly, Grice (1978:125) draws our attention to the fa c t that prefixing

the phrase "to speak ironically" to an ironical statement would destroy

the irony (whereas "to speak metaphorically" can be prefixed to metaphorical

utterances) . In the same way, adding 'I l i e ' to an untrue statement

would turn i t into a true statement, or at l e a s t i t would destroy the

p o s s ib i l i t y o f deceiving the hearer. This f inal remark leads us toward

the perlocutionary intent of lying.

183

E. LYING AND PEELOCUTIONARY INTENT

22. The intended perlocutionary e f f e c t which i s typically associated with

statements i s to make the hearer believe something or to inform him about

something. Thus, i f one l i e s , one typically intends to bring i t about

that the hearer believes something which i s untrue. In other words, there

i s an in te n t io n to d e ce iv e involved. The verb TO DECEIVE was i t s e l f

excluded from the realm of the verba mentiendi because i t focuses completely

on a particular perlocutionary e f f e c t which i s , moreover, necessarily

achieved. I f one were to describe (Al) by means of (Dl) in circumstances

in which (Cl) obtains, also (C2) has to be f u l f i l l e d in order for TO DE­

CEIVE to be used appropriately in (Dl) as a description of (Al).

(Al) John was home at the time of the robbery

(Dl) Sa deceived the police

(Cl) In fa c t , John was not home at the time of the robbery

(C2) The police came to believe that John was home at the time of

the robbery

Similarly, TO FLATTER was also excluded from the set of verba mentiendi

though i t may be used to describe untruthful statements, because ju s t as

TO DECEIVE i t real ly focuses on a particular perlocutionary e f fe c t which

in this case may or may not ne achieved. About the verba mentiendi such

as TO LIE we can only say that they have a strong association with a

typical ly intended perlocutionary e f fe c t which could generally be called

deception or the creation of fa l se b e l ie f s .

184

Though there i s th is one basic e f f e c t which i s ty p ic a l ly intended, there

i s considerable variation among the verba mentiendi along the perlocution­

ary dimension. There are mainly two parameters for the d i f f e r e n t ia t io n .

F ir s t , the type of fa l s e b e l i e f may be further sp ec i f ie d . Second, d i f feren t

types of further e f f e c t s may be aimed at via the fa l se b e l i e f induced.

These two parameters wil l now be considered.

23. Most verba mentiendi g ive no further sp ec i f ica t ion of the f a l s e b e l i e f

that the speaker i s trying to induce in the hearer. Consequently, most of

them are neutral with respect to th is part of the perlocutionary dimension.

Probably there i s only one c lass of exceptions const ituted by verbials such

as TO CANT, TO ACT/PLAY THE HYPOCRITE and the Dutch KWEZELEN ( ' t o c a n t ' ) ,

FEMELEN ( ' t o cant ') and DE HYPOCRIET UITHANGEN {'tomplay the hypocr i te ' ) .

As an i l l u s t r a t i o n of th e ir use, consider (D2) and (D3) as descriptions

of (A2)-(C3) and (D4) and (D5) as descriptions of (A3)-(C4).

(A2) I t i s hard to understand why so many young people should d ie ,

but I think we should not try to judge God's decis ions

(D2) Sa i s canting again

(D3) Sa i s playing the hypocrite again

(C3) Sa does not be l ieve that God e x i s t s , or at l e a s t he thinks that

i f God e x is t s he must be a super-cruel being in order to l e t

so many young people die

The second example:

(A3) Let t h i s incident not be a reason for declaring war; they are,

a f t e r a l l , our brothers

185

(D4) Sa was canting about brotherly love

(D5) Sa was playing the hypocrite again

{C4) Sa wants vengeance, he i s only waiting for a better moment

These two examples show that the kind of b e l i e f which i s typ ica l ly in­

tended to be generated by means of canting and playing the hypocrite i s the

hearer's b e l i e f that the speaker i s a virtuous or pious person. In other

words (C5) should be a character is t ic of (A2) in order for (D2) and (D3)

to be appropriate descriptions of A2), and (C6) should be a character ist ic

of (A3) in order for (D4) and (D5) to be appropriate descriptions of (A3).

(C5) Sa wants Ha to bel ieve that he i s a pious, God-fearing person

(C6) Sa wants Ha to bel ieve that he i s a virtuous, peace-loving person

I have not found other verba mentiendi which gave equally s p e c i f ic inform­

ation on the perlocutionary intent at the level of the b e l ie f s which the

speaker wants to arouse.

24. There i s a perlocutionary level beyond the b e l ie f s which the speaker

wants to arouse when making an untruthful statement. The intention may be

to produce a further e f f e c t by means of generating a certain fa l se b e l i e f .

Also at that level many verba mentiendi do not provide any sp e c i f ic in­

formation. But th is parameter y ie lds a somewhat stronger variation than

the previous one.

For one thing, the four types of verba mentiendi specifying the way in

which the truth i s being distorted (quality increasing, quality diminish­

ing, e t c . ) also s ingle out, as a direct corollary, d if ferent intended

186

e f f e c t s . Thus, the ty p ic a l ly intended e f f e c t of an act describable in

terms o f a qual ity increasing verbial such as TO WHITEWASH i s to create

a favorable a t t i tude towards something or someone by making the hearer

bel ieve that i t or he i s 'b e t t e r 1 than i t or he i s in r e a l i t y . The

ty p ic a l ly intended e f f e c t of an act describable in terms o f a qual ity

diminishing verbial such as TO SLANDER i s to create an unfavorable a t t i ­

tude towards something or someone by means of making the hearer be l ieve

i t or him to be 'worse' than i t or he i s in r e a l i t y . S im ilar ly , the

t y p ic a l ly intended e f f e c t o f an act describable in terms of a quantity

increasing verbial such as TO EXAGGERATE i s to impress the hearer by

making him believe that the blown up representation given matches r e a l i t y .

Conversely, acts describably by means o f quantity diminishing verbials

such as TO UNDERSTATE are t y p ic a l ly intended to y i e ld the reverse e f f e c t

of impressing.

I t should be noted that understating, in actual d iscourse , i s very

often not intended to make the hearer be l ieve the propositional content.

Thus "He i s fu l l of understatement" may mean that the person in question

i s fu l l of humor and that he achieves his humoristic e f f e c t s by means of

understatements (as f igures of speech). Does th i s inval idate our analys is ?

By no means. This observation simply points to a more general problem.

Making a statement the propositional content of which does not match rea­

l i t y or the 'truth' i s not necessar i ly ly ing. Consider (A4) uttered in

circumstances to which (C7) obtains.

(A4) John i s a genius

(C7) In f a c t , Johncan hardly produce two coherent sentences

187

(A4) can e i th er be an instance of lying or an instance of irony. What

dist inguishes the two i s the speaker's perlocutionary in tent . In the case

of irony i t i s not the speaker's intention to make the hearer bel ieve

that John i s a genius; quite the opposite i s true. Humoristic under­

statements d i f f e r from l i e - l i k e understatements in the same way. This

observation shows the necess i ty of formulating the ty p ic a l ly intended

perlocutionary e f f e c t of acts within the domain of the verba mentiendi.

25. There are a couple o f quantity increasing verbs the intended e f f e c t

of which deviates from the general intention of 'impressing' formulated

in the previous paragraph. Not only are t h e ir e f f e c t s d i f feren t in type,

they are a lso much more s p e c i f i c .

F ir s t , the Dutch verb CHARGEREN means ' to give an exaggerated repre­

sentation o f something in order to make i t r id ic u lo u s ' . I have not

found an exact equivalent in English, though TO CARICATURE i s c lo se .

Second, the verb TO OVERCHARGE denotes acts which are intended to make

someone pay more for something than i t i s r e a l ly worth, by making him

bel ieve that the charged price r e f l e c t s i t s real value. In Dutch only

circumscriptions such as TEVEEL AANREKENEN ( ' t o charge too much'), which

can hardly be regarded as l ex ica l items, can be used to describe such

a cts .

26. There are a lso some verbials which do not belong to any of the four

s e t s of verba mentiendi s in g l ing out the way in shich the truth i s being

dis to r ted , which spec ify an intended e f f e c t to be achieved via the arousing

o f f a l s e b e l i e f s in the hearer.

188

F irst , verbials such as TO PLAY UP TO and i t s equivalent IEMAND NAAR DE

MOND PRATEN ( l i t . ‘to talk to someone's mouth') describe acts typical ly

intended to gratify or f l a t t e r the hearer by making him believe that the

speaker holds b e l ie f s which the hearer would want him to hold.

Second, TO BEAR FALSE WITNESS and i t s equivalent VALSE GETUIGENIS AF­

LEGGEN describe acts typical ly intended to influence judgments (in or

out of court) by inducing fa l se be l ie f s in the hearer.

Finally , the Dutch ABRAHAMMETJE SPELEN ( l i t . 'to play l i t t l e Abraham')

i s used to talk about acts of t e l l in g part of the truth (and thus holding

back the other part or lying about i t ) in order to escape from further

questioning, which may be achieved i f the hearer comes to bel ieve that

what the speaker says i s the whole truth.

These remarks conclude our overview of the sa l ien t points on the perlo­

cutionary dimension of acts of lying.

F. THE TEXTUAL DIMENSION OF LYING

27. There are several ways in which the verba mentiendi r e f l e c t the

textual dimension of lying. For one thing, there are quite a few verba

mentiendi which denote l in g u is t ic actions which almost necessarily con­

s i s t of more than one speech act.

Many of the verbials which were said not to give any information about

the i llocutionary type of the acts described belong here. Acts described

by means of TO PUT ON AN ACT, TO PUT UP A FRONT, TO MAKE FALSE PRETENSES,

and the Dutch DOEN ALSOF ( ' t o act as i f ' ) , EEN ROL SPELEN ( ' to act a par t ' ) ,

189

e tc . w i l l rarely be r e s tr ic ted to one s in g le speech act . Usually a whole

t e x t , o f any kind, i s involved.

Similarly TO DRESS UP, TO EMBELLISH and TO EMBROIDER often require

a suprasentential segment o f l in g u i s t i c action , as wil l be c lear from

the ir usage in (Dl) to (D3).

(Dl) He dressed up the story of what he did on vacation

(D2) He kept inventing d e ta i l s to embellish his story

(D3) He embroidered the story o f his adventures in Hollywood

The Dutch equivalents BORDUREN ( ' t o embroider'), VERFRAAIEN ( ' t o em­

b e l l i s h ’ ) and AANKLEDEN ( ' t o dress up') behave in a s im ilar fashion.

The verb TO STORY can a lso be used as a verbum mentiendi. But i t i s

quite remarkable that th i s verb can be used to refer to one s in g le un­

true statement in sp i te of the fac t that a story normally cons is t s of

a sequence of speech acts as appears from (Dl) to (D3). Thus, TO STORY

i s used quite appropriately in (D4) as a description o f (Al) given that

(Cl) obtains.

(Al) I am nearly 30 years old

(Cl) Sa i s over 40

(D4) Sa stor ied about his age

A comparable Dutch verb ia l , SPROOKJES VERTELLEN ( l i t . 'to t e l l f a i r y - t a l e s ' ) ,

can a lso be used as a verbum mentiendi. But unlike TO STORY i t rarely re­

fers to a s in g le speech act .

190

28. A second group of t e x t - s e n s i t iv e verba mentiendi does not only require

that the l in g u i s t i c action described cons is t s o f more than one speech a c t ,

but also that these speech acts could be separately characterized as l i e s .

Most of the in te n s i f i e d verba mentiendi such as TO LIE LIKE A TROOPER

and the Dutch TEGEN DE STERREN OP LIEGEN ( l i t . 'to l i e up against the

s t a r s ' ) belomg here. I t would be hard to predicate of a certa in speaker

that he l i e d l ik e a trooper i f he only uttered one untrue statement.

There are a couple of Dutch verbials which bas ica l ly ind icate a person's

general inc l in a t ion towards ly ing. They are WAT AFLIEGEN ( ' t o l i e a l o t ' ;

we are rea l ly concerned with a l ex ica l item here since AFLIEGEN has no

meaning of i t s own when in i s o la t io n ) and VAN ALLE KATTEN KWAAD WETEN

( l i t . ' to know something bad about a l l c a t s ' ) . But ju s t as TO LIE LIKE

A TROOPER and other in te n s i f i e d verba mentiendi they can a lso be used

to describe multiple acts o f ly ing . This wil l not be hard to understand

for WAT AFLIEGEN, but the use of the second verbial in i t s ac t-sense

instead of i t s inc l in a t ion -sen se requires an i l l u s t r a t i o n ; Consider (D5).

(D5) Hij wist weer van a l l e katten kwaad

( l i t . 'Again he knew something bad about a l l c a t s ' )

(D5) i s an adequate description of a speaker's l in g u i s t i c a c t i v i t y

during a tea-time conversation in which he or she claimed, without proof

or evidence, that Jane was about to get her third divorce, that Jim passed

his exam because his father paid a large sum o f money to the professor

in question, that Kate had an a f f a i r with an o f f i c i a l o f the Internal

Revenue Service so that she did not have to pay any tax es , e t c .

191

29. Some verba mentiendi may presuppose discourse preceding the untruthful

statement described. For instance, the use of ZICH VAN DE DOMME HOUDEN

( l i t . 'to hold oneself dumb') which means 'pretending not to know any­

thing about a certain matter' , often implies previous questioning about

that matter by the hearer. In th is case even the type of preceding speech

act i s made e x p l i c i t .

Others may presuppose e ither preceding or following discourse. One can

only describe the taking of an oath by means of EEN MEINEED AFLEGGEN

( l i t . 'to take a fa l se oath') i f the oath in question bears on a l i e

uttered before or after the oath was taken. Again the nature of the pre­

ceding discourse or the following discourse, which in this case must

contain a l i e , i s made e x p l i c i t .

30. A f inal s e t of verba mentiendi which re late acts of dis tort ing the

truth to other l in g u i s t i c action includes TO MISQUOTE, TO FALSIFY, TO

TWIST ONE'S WORDS and the Dutch VERVALSEN ( ' t o f a l s i f y ' ) and IEMANDS

W00RDEN VERDRAAIEN /VERWRINGEN/VERVALSEN ( ' t o twist someone's words').

In th is case previous l in g u is t i c utterances (usually by someone e ls e )

are misrepresented. Needless to say that TO MISQUOTE i s only a verbum

mentiendi in case i t refers to a voluntary action with the intention to

deceive.

G. LYING AND VALUE JUDGMENTS

31. Though there are supposed to be such things as WHITE LIES (in Dutch

LEUGENS 0M BESTWIL, l i t . ' l i e s for the best ' ) negative value judgments

192

are attached to pract ica l ly all verba mentiendi. That i s to say, neither

Dutch nor English provides us with instruments for talking favorably about

distort ing the truth. The only verbials which present d is tort ions o f the

truth as excusable acts include TO FIB, TO KID, TO PULL SOMEONE'S LEG

and the Dutch JOKKEN ( ' to f i b ' ) and IEMAND BIJ DE NEUS NEMEN ( l i t . 'to

grab someone's nose' , meaning 'to pull someone's l e g ' ) . TO FIB and

JOKKEN describe acts which are excusable because of the unimportance or

t r i v i a l i t y of the facts l ied about. The acts referred to by means o f TO KID,

TO PULL SOMEONE'S LEG and IEMAND BIJ DE NEUS NEMEN are excusable be­

cause of the humor involved. But the very fact that these few exceptions

merely present some acts of lying as ex cu sa b le s tresses the general nega­

t iv e att itude towards distort ions of the truth.

32. Two more indicators of these value judgments may be adduced. F ir s t ,

there are a couple of fixed expressions used to describe acts of saying

something in favor of a d is tort ion of the truth, namely TO COLOR A LIE,

and TO GILD A LIE. Now, TO COLOR and TO GILD imply d is tort ions o f the

truth themselves. Since I know of no other fixed verbials describing

acts of saying something favorable about a l i e and not carrying the same

implicat ion, we might risk to formulate the conclusion that whenever

something favorable i s said about a l i e th is i s generally regarded as

a dis tort ion of the truth and therefore that l i e s are generally regarded

as inherently 'bad' and lacking praiseworthy features.

Second, the lex ica l items available for describing the end of a ly ing-

sequence ( i . e . admitting that what was said before were a l l l i e s ) imply

that lying is an a c t iv i ty that one usually does not stop de l iberate ly .

193

(Here we disregard the neutral phrase TO STOP LYING because i t i s not a

l e x ic a l item, and because i t does not necessar i ly imply that the truth

was revealed at the end of the ly ing-sequence .) I am thinking of verbials

such as UIT ZION ROL VALLEN ( l i t . 'to f a l l out of one's r o le ' ) and DOOR

DE MAND VALLEN ( l i t . ' to f a l l through the basket') both of which refer

to a fa i lu re to uphold one's l i e s or pretenses. Again the explanation i s

to be found in a negative a t t i tude towards ly ing . A mentiens t r i e s to

maintain his f a l s e representation of the truth in order to avoid being

judged and losing face.

33. Together with the importance of questions o f truth for statements,

which was revealed in the sec t ion on the i l lo cu t io n ary force o f ly ing,

the overwhelming prominence o f negative value judgments attached to

d is tor t io n s o f the truth supports the relevance of Grice's (1975) con­

versational maxim of qual i ty "Try to make your contribution one that i s

true". Though under innumerable circumstances the question of truth be­

comes irre lev an t , as several authors have shown and as Grice certa in ly

r e a l i z e s , the verba mentiendi show beyond any doubt that truth i s one

o f the basic parameters in terms of which statement- l ike utterances

are conceptualized.

H. NOTE ON HUMAN IMAGINATION

34. The propositional content o f l in g u i s t i c actions describable in terms

of verba mentiendi i s always a product of the human mind or human imagi­

nation since the speaker knows or thinks i t does not match r e a l i t y . There

194

i s , however, a se t o f lexical items which strongly emphasize th is imagi­

nation aspect of lying; in other words, they e x p l i c i t l y present the men-

tiendum as a product of the human mind. They are: TO CONCOCT, TO COOK UP,

TO FABRICATE, TO FANTASIZE (ABOUT), TO INVENT, TO MAKE UP, TO MANUFACTURE,

and TO TRUMP UP. Here are some i l l u s t r a t i v e sentences showing the ir use

as verba mentiendi.

(Dl) They concocted an a l ib i for the ir friend

(D2) Whenever he arrived late he cooked up an elegant excuse

(D3) He fabricated a complicated explanation for his being la te

(D4) At tea-time she always keeps fantasiz ing about her neighbours

(D5) What he did not know, he invented

(D6) He made up most o f the story

(D7) The historian was evidently manufacturing a l l the evidence he

adduced

(D8) Whenever he was la te he trumped up a new excuse

Notice that these verbs are often ambiguous between a purely mental a c t i ­

v i ty and the l in g u is t i c a c t iv i ty of putting creations of the mind into

words. Though imagination or mental a c t iv i ty i s much appreciated in

our soc ie ty , a pejorative connotation gets attached to the verbs mention­

ed as soon as they are used as verba mentiendi. This phenomenon provides

us with additional support for the thes is that strong negative value

judgments are attached to lying in our culture and that Grice's maxim

of quality shows a strong factual sa l ience in our experience of l in g u i s t i c

a c t iv i ty .

Similar verbials are to be found in Dutch: FABRICEREN ( ' t o f a b r ic a t e ' ) ,

FANTASEREN ( ' t o fa n t a s iz e ' ) , FINGEREN ( l i t . ' to produce f i c t i o n ' , meaning

195

•to make up ') , VERZINNEN { ' t o make u p ' ) , UITVINDEN { ' t o in v e n t ' ) , VER-

DICHTEN ( l i t . 'to turn into poetry ' , meaning 'to invent1) and the picturesque

expression UIT ZIJN DUIM ZUIGEN ( l i t . 'to suck from one's thumb', meaning

' to invent’ ).

Notice that most o f these imagination-oriented verbs belong to the s e t

of verba mentiendi which usually refer to more than one s in g le speech act.

On the truth dimension they are to be s i tuated among the verbia ls of lying

proper.

I . CONCLUSIONS

35. In th i s chapter we have touched upon the most important semantic

dimensions along which English and Dutch verba mentiendi seem to be

comparable: the truth dimension (which i s no doubt the basic one),

the i l locut ionary dimesnion, the perlocutionary dimension, the textual

dimension, and f i n a l l y the dimension of value judgments and of the role

o f imagination. This i s not to say that no additional semantic dimensions

w il l have to be taken into account when further comparing the meaning

o f individual verbials within the domain of ly ing .

36. As I did at the end of the previous chapter, I shall now give a

r e la t iv e ly small and random l i s t o f noteworthy observations which forced

themselves upon us in the process o f s cru t in iz ing the le x ic a l f i e l d asso­

ciated with l i n g u i s t i c acts of d is tor t ing the truth.

( i ) The gradabil i ty of the notion o f truth i s a r e la t iv e ly unimportant

parameter of the truth dimension o f the verba mentiendi. (Para graphs

5 and 6)

196

( i i ) From the set of verba mentiendi i t appears that the different ways

in which the truth can be deviated from are conceived in terms of a

fourfold contrast: quality increasing, quantity increasing, quality

diminishing, quantity diminishing. (Paragraph 6}

( i i i ) There i s a one-directional dependence relationship between the

quantity scale of truth and the quality scale. (Paragraph 7)

{i v ) The lexicon re f lec ts the general habit of taking quantity as a

criterion for quality. (Paragraph 7)

(v) Only the l ike poles of the too truth scales share descriptive verbials .

(Paragraph 8)

(vi) The verba mentiendi re f lec t the following attitude prevailing in

our culture: what is disagreeable i s regarded as untrue. (Paragraph

11).

( v i i ) Questions of truth are most relevant in connection with repre­

sentative speech acts . (Paragraphs 15 f f . )

( v i i i ) The acts of praising, accusing and swearing are sa l ien t points in

the area of the representative speech acts when i t comes to the

question of truth. (Paragraphs 16 to 19)

(ix) Deceiving i s an e f fec t typical ly intended by utterers of untrue

statements; th is reinforces the independent claim that there i s a

typica l ly intended e f fe c t associated with all representatives, namely

making the hearer know or believe something. (Paragraphs 22 f f . )

(x) I t i s shown that usually negative value judgments are attached to

distortions of the truth; together with ( v i i ) th is strengthens the

l in g u is t ic relevance of Grice's conversational maxim of quality with

respect to statement-like utterances. (Paragraphs 31 f f . )

197

37. Apart from the quite surprising gap for a basic quantity diminishing

verb in Dutch, differences between the sets of Dutch and English verba

mentiendi are minimal. Of course a l l the generalizations summed up in the

previous paragraph should be subjected to further comparative research.

An e sp ec ia l ly promising topic for further invest igation would be the

universal ity or non-universality of the two scales of truth (the quanti­

ty sca le and the quality scale) in terms of which dif ferent types of

deviations from the truth are conceptualized, and of their in terre la t ion­

ships.

198

V. THE SEMANTICS OF DIRECTING

A. INTRODUCTION

1. We are now entering a second central area of l in g u i s t i c meaning, the

realm of i l locut ionary force. Within th is vast area we s in g le out speech

acts of a d irec t ive kind. What types o f acts are encompassed by the

general term ' d i r e c t i v e 1 w il l be e lucidated in the fol lowing paragraphs.

The verbs and expressions used to describe l in g u i s t i c actions of a d i ­

rect ive kind wil l be referred to as v e r b ia l s o f d i r e c t i n g .

One aspect o f the subsequent treatment o f the verbia ls o f d irecting

wil l no doubt arrest the reader's thoughts. Though th i s chapter i s supposed

to explore a subfie ld o f the vast area o f i l locu t ionary fo rce , we shall

not e x p l i c i t l y discuss an i l locut ionary dimension of the verbia ls of d i ­

recting . The reason i s that i l locu t ionary force i t s e l f i s the cr i ter ion

for including a verbial in the subject matter of th is chapter. The d i f f e r ­

ent shades in force which wil l present themselves depend on a large

variety o f factors which have to be dealt with in terms o f d i f fe re n t s e ­

mantic dimensions. The overview of d irec t iv e speech act types in the next

few paragraphs could be regarded as an outl ine of the i l locu t ionary

199

dimension in question. The individual discussion o f other dimensions a f t e r ­

wards wil l c la r i f y the d i f feren t shades o f i l locut ionary force pointed

out.

Though primarily a matter of i l locu t ionary force i s at s take, some of

the l in g u i s t i c action verbia ls to be invest igated do not describe what

would tra d i t io n a l ly have been ca l led an ' i l lo cu t io n a ry a c t ' . Instead, many

o f them refer to larger configurations o f l in g u i s t i c acts such as t e x t -

leve l structures . These configurations o f speech ac ts , however, can be

sa id to possess a force s imilar to individual d ir ec t ive speech ac ts . I t

should be kept in mind that when I use 'a d ir ec t iv e act' in th i s chapter,

t h i s phrase can frequently be replaced by 'a se t of d ir ec t iv e acts ' or

'a ser ie s of acts with a d irec t iv e fo r c e ' . I t s permeating d i f feren t l e v e l s

o f l in g u i s t i c structure shows how fundamental d irect ing i s as a function

o f language.

2. According to Sear le 's (1976) c la s s i c a l d e f in i t io n of d ir ec t ive speech

acts the point of a d ir ec t ive i s to get the hearer to do something; there

i s a world-to-word d irect ion of f i t ; the speaker expresses a wish; and

the proposition sp e c i f i e s a future act to be done by the hearer. This

d e f in i t io n i s , not su rp r is in g ly , geared to the central instances of d i ­

rect ing , namely commands and requests. This means that i t would be wrong

to adopt i t as a r ig id cr i ter ion for including or excluding a verb or

expression in or from the set of verb ia ls of d irec t in g . The r e su l t would

be the exclusion of many verbials about which our in tu i t io n maintains

that they re fer to l in g u i s t i c actions which possess a d ir ec t iv e aspect of

some kind; we would be l e f t with TO COMMAND, TO REQUEST, TO BEG and th e ir

200

synonyms. I t would be equally wrong, a f ter using our in tu it ion to round

up the verbials of d ir ec t in g , to force th e ir meaning in to the mold of

Sear le 's d e f in i t io n . For instance , i t w i l l not be poss ib le to describe

every d irec t iv e speech act as an attempt to get the hearer to do some­

thing. But how can a l in g u i s t i c action which i s not, s t r i c t l y speaking,

an attempt to get the hearer to do something, be regarded as an act of

directing ? The answer w i l l , hopefully , reveal i t s e l f in the course of

the following introductory and sketchy overview of d irec t iv es which de­

v ia te somehow from the central instances of commanding and requesting.

The overview i s meant to shape the reader's expectations as to the

l ex ica l domain covered by th is chapter.

3. A f i r s t type of d irec t iv e which departs from the central cases out­

l ined by Sear le 's d e f in i t io n i s commonly described by means o f TO PROHIBIT

and i t s synonyms. Prohibitions d ir ec t the hearer to not doing something

instead of doing something. In t h i s case the deviation i s minimal. The

act i s c lear ly d irec t ive .

A second deviant type: TO ASK, when used to describe the asking of a

question ( instead of as an equivalent o f TO REQUEST), refers to acts the

propositional content o f which does not spec i fy the future act to be done

by the hearer. What i s expected from the hearer i s a response to the

question. This i s c lear from the verb i t s e l f .

TO DARE, in the sense of to challenge to do something requiring bold­

ness and s k i l l , stands apart from the central d ir ec t iv e s in that i t does

not usually imply the speaker's b e l i e f that the hearer i s able (or bold

enough) to do what he i s trying to get him to do. Such an implication i s

201

tr a d i t io n a l ly assigned (as a 'preparatory condit ion') to acts o f command­

ing and requesting.

4 . Also acts of threatening can be d ir ec t iv e . I am not thinking about un­

conditional threats such as " I ' l l k i l l you" but about conditional ones

such as “If I ever see you with my s i s t e r again, I ' l l k i l l you". The

l a t t e r i s c lea r ly an attempt to keep the hearer away from the speaker's

s i s t e r , and the act as a whole can be described by means of TO THREATEN

as in "S threatened H into staying away from S's s is ter" or "S threatened

to k i l l H i f he ever saw H with S's s i s t e r again".

Also in th i s case the future act to be performed by the hearer i s not

sp ec i f ied in the propositional content. Moreover, threatening i s never

purely d ir ec t iv e . In i t s unconditional form i t i s simply commissive and

lacks d ir ec t iv e aspects. In i t s conditional form i t i s at the same time

commissive (though not as strongly as a promise since the 'obligation'

i t creates to i n f l i c t harm i s unlikely to lead towards the speaker's

being reproached in case he does not act in accordance with his commit­

ment.) .

Notice that a description such as "S threatened H" focuses completely

and exc lus ive ly on the commissive aspect o f threatening. This means that

a complement i s needed to bring out the d ir ec t iv e character o f the act .

This i s a lso true for the Dutch equivalent DREIGEN or BEDREIGEN. The

form BEDREIGEN can only be used in a structure such as "S theatened H".

Dreigen i s used in descriptions of the type "S threatened to k i l l H i f he

ever saw H'with S's s i s t e r again". Neither of them can be used in a

structure such as "S theatened H into staying away from his s i s t e r . Thus

TO THREATEN and DREIGEN can focus both on the d ir ec t iv e and the commissive

202

aspects of conditional threats whereas BEDREIGEN does not allow for a

direct ive aspect at a l l . Therefore BEDREIGEN will be l e f t out of account

in this chapter. There is another c lose ly related verb in Dutch, namely

AFDREIGEN which i s one of the Dutch equivalents of TO BLACKMAIL: i t means

'trying to get the hearer to give something by means of threatening'. In

this verb both aspects of the conditional threat are united.

There are two more types of directives which are in fact mixtures between

directives and some other type of l in g u is t ic action. Both of them derive

their lack of central ity in the area of directives from their being

mixed classes and from the fact that they cannot really be paraphrased

as 'attempts to get the hearer to do something*.

5. The f i r s t of these mixed c lasses i s represented by TO ADVISE and TO

WARN. Usually acts of advising and warning are partly - - i f not primari­

ly — representative. In what sense can they be said to be directive ?

A piece of advice i s an indication of a preferable course of action.

Its directive force does not necessarily derive from a wish on the part

of the speaker — though in some circumstances i t may — but rather from

the fact that a particular course of action is presented as preferable

because i t i s to the hearer's benefit. On the other hand, warning is

an indication of an event or course of action to be avoided because i t could

be detrimental to the hearer.

6 . A second mixed class i s more troublesome. Not because i t i s less c learly

directive than advising and warning, but simply because of recent attempts

to declare i t a monolithic commissive. The acts in question can be des­

cribed by means of TO PERMIT and i t s equivalents. The argument goes l ike

203

th i s : when giving his permission to do something the speaker commits him­

s e l f to not obstructing a particular action on the part of the hearer;

hence a permission i s a kind o f promise; therefore i t i s a commissive

rather than a d ir ec t iv e . This i s a decent argument in favor o f the claim

that a permission is commissive. But i t presents no evidence a g a in s t

c la s s i fy in g permissions as d ir ec t iv e s . Unless one accepts that every

type of speech act can only belong to one c la s s . But that i s one o f the

main errors t h i s essay w i l l hopefully help to abolish.

Yet permitting i s not a central type of d irect ing . I t cannot be des­

cribed as an 'attempt to get the hearer to do something'. What, then,

makes i t d ir ec t iv e at a l l ?

Just as advis ing , a permission i s an indication o f a certain course of

action. Unlike advising, permitting does not imply that the speaker pre­

sents that course of action as preferable to the hearer. But permitting

implies that the hearer himself prefers his doing the action over his not

doing i t . Moreover, the speaker does not only ■ indicate a course o f action

which the hearer regards as des irab le , but he a lso uses his authority

over the hearer to open th e way to that course of action . The l a t t e r aspect

emerges very c lear ly from some equivalents of TO PERMIT such as TO GIVE

THE GO-AHEAD (which can be used to describe a speaker's permitting to

s ta r t with the execution of a plan). Thus, in a very real sense, a person

giving the permission to do something i s r e g u la t in g another person's

behavior and using his authority to do so. In other words, permitting

might even be said to be more strongly d ir ec t iv e than advising or warning.

What about the wish on the part of the speaker, which i s c h aracter is t ic

of the central instances o f directing ? I t was not necessar i ly present in

204

advising and warning. In contrast , a weak form of wishing i s always present

in permissions: i f S did not want H to do P at a l l , he would simply forbid

him to , which he can do by virtue o f the same authority which he applies

when permitting. This type of wanting i s , admittedly, quite weak because

i t does not even require the speaker's b e l i e f that the course o f action

in question i s advantageous to anyone. But what i t bo i l s down to i s that

the speaker, ju s t as when he is commanding (with authority ) , i s w i l l in g

to take the r e s p o n s i b i l i t y for an act to be performed by the hearer. Since

S has power over H's doing of not doing P, he can be held responsible i f

H does P. This w il l ingness to take r e p o n s ib i l i ty , or th is weak form of

wanting H to do P, makes permitting d ir ec t iv e . But at the same time i t i s

the very basis of the commissive aspect involved. The same kind of commit­

ment characterizes a l l authoritat ive acts of d irect ing . Recalling Nurnberg

where o f f i c e r s were convicted of ordering so ld ie r s to shoot people, should

convince anyone of the truth o f th is claim. And i f i t i s true that both

commanding and permitting share th i s type of commitment, i t s presence in

permissions i s cer ta in ly no adequate ground for excluding acts of permitting

from the realm of the d ir ec t iv e s . This paragraph probably contains the

strongest indication given so far o f the dangers of theoret ica l c l a s s i f i ­

cations of speech ac ts .

Just as with commanding and requesting, the propositional content of a

l in g u i s t i c act of permitting s p e c i f i e s a future act to be done by the

hearer. This fac t pul ls permissions away from other commissives which

usually contain a proposition specifying an act to be done by the speaker.

Thus one could say that the d irec t ive aspect o f permissions i s quite ex­

p l i c i t whereas the commissive aspect remains largely im p l ic i t . One could

object that an utterance such as "You may go to the movies tonight" i s

205

quite e x p l i c i t l y and overtly commissive because i t means — due to the

presence of 'may' - - "I commit myself to not obstructing your going to

the movies tonight". But such a claim would already be based on a semantic

analysis which uses as a premise the b e l i e f that permissions are primari­

ly commissive.

7. The discussion in the foregoing paragraph should not be misinterpreted

as an attempt to decide whether acts of permitting are primarily commissive

or primarily d irec t ive . In t h i s , as in many cases , I regard such attempts

as irrelevant. The important thing to real ize i s that permissions have both

a direct ive and a commissive aspect. The classes of speech acts proposed

in the l i terature are rarely mutually exclusive since they are based on

c r i t e r ia of c la s s i f i c a t io n which are always arbitrary to a certain extent

and since there i s no logical impediment to the occurrence of several

dif ferent functions of language. The reader wil l recall that this r e a l i ­

zation was the very incentive to the invest igations reported in this

essay and i t vindicates our decision to study semantic dimensions rather

than semantic c la s se s . There are two reasons, one theoretical and one

pract ica l , why in sp ite of th is decision some class-terminology i s s t i l l

being used. F ir s t , i t links th is essay with e a r l ie r research. Second,

an uncompromising inquiry directed at semantic dimensions would have to

take as i t s point of departure the complete se t of l in g u i s t i c action verb­

i a l s ; practical considerations d ictate me to l im it the object of in v e s t i ­

gation; in th is chapter a subfie ld i s s ingled out on the basis of what

has been regarded up to now as a major c lass on i l locutionary acts .

206

B. THE FRAME OF DIRECTING

8 . The introductory overview of some d irect ive speech act types leads

us towards the following d e f in i t io n of verbials of direct ing: a l l verbs

and verb-l ike expressions which describe l in g u i s t i c action by means of

which a speaker ty p ica l ly d irects or influeneces a hearer's subsequent

behavior. Directing i s not only one of the fundamental functions of

language, i t i s also one o f the v i ta l roles in socia l l i f e ; for the per­

formance of t h i s role the d ir ec t ive function of language i s usually put

in to action.

Whereas s i l e n c e was to be situated in the frame of language and lying

in the frame o f truth, direct ing belongs in one of the most basic frames

of social interact ion in general. I t involves parameters such as authority

and wishing, which wil l be re f lec ted as semantic dimensions in the verbials

of d irect ing . The verbials can also be expected to reveal the sp e c i f i c

areas of soc ia l interaction in which directing i s a major function. Let

us go ahead and find out what the words have to say.

C. THE DIRECTIVITY OF DIRECTING

9. I t w i l l be c lear from the introduction that not a l l acts denoted by

verbials of direct ing are equally strong as d ir e c t iv e s . I would l ik e to

ca l l these d if ferences in strength d e g re e s o f d i r e c t i v i t y . Though there

i s some interaction with other dimensions, such as the authority in ­

volved, these degrees o f d i r e c t iv i t y depend mainly on the strength of

the w ish or w an tin g expressed by the speaker. As with a l l other dimensions

involving s im i lar types o f gradabil i ty i t i s not poss ible to mark o f f

207

rea l ly discrete points . What we can do, however, i s to indicate the two

extremes of the scale and try to arrange some basic verbials of d irec t ­

ing with respect to these. The extremes are, of course, {A1) and (A2).

(Al) Sa expresses a strong wish for Ha to do or not do P

(A2) Sa expresses a weak wish for Ha to do or not do P

Thoigh i t i s c lear that TO COMMAND, TO ORDER and TO PROHIBIT are quite

high up on the sca le , there are d irect ives expressing even stronger

wishes, namely TO BEG and TO PLEAD. At the other end of the scale we

approach a zero wish with TO ACQUIESCE (in the sense of permitting) as

used in (Dl) describing (A3) conforming to (Cl) , (C2) and (C3).

(A3) O.K., you can go to medical school

(Cl) Sa i s Ha's father

(C2) Sa has been trying to get his son to take over his business

(C3) Ha has been trying for months to get his father's permission

to go to medical school instead of taking over his business

(Dl) Sa f in a l ly acquiesced in his son's going to medical school

In this instance, the speaker's wish for the hearer to do P i s even weaker

than with regular permissions. But though i t i s c lear from the context

that his son's going to medical school was or ig ina l ly a g a in s t Sa's wishes,

for (A3) the argument holds that i f Sa did not want his son to go at a l l ,

he would keep res i s t in g the decision (which he could do on the basis of

some kind of parental authority which Ha accepts because otherwise he

would not have asked his fa ther 's permission in the f i r s t p lace .)

208

10. Before discussing the degree of d irec t iv i ty of any further individual

cases br ie f ly , I submit a tentative picture of the scale in question,

namely SDC-Table 1. Needless to say that the elusive character inherent

to matters of gradability i s responsible for the tentativeness of the

picture. There are no claims attached to the actual sequence of verbs on

the scale; I do not even want to contend that the sequence i s an accurate

representation of my own judgment and intu it ion . All that counts i s the

approximate placement of the verbs on the scale.

In the following paragraphs I shall draw the attention to some of the

more striking aspects of the SDC-Table.

11. What may strike the reader f i r s t i s the position of TO ASK in the

sense of TO ASK A QUESTION, Dutch VRAGEN or EEN VRAAG STELLEN. Asking a

question i s high up in the scale . But i s i t s d irec t iv i ty so strong that

i t warrants i t s placement as high up as TO DEMAND and considerably

higher up than TO INSIST and TO REQUEST ? I believe that asking a question

always involves a strong wish to get a reply. This i s even true in the

course of cocktail conversations in which both the topic of conversation

and the truth of the statements made are irrelevant to a large extent.

Not answering a question i s always a rupture in the social interaction

between two interlocutors; usually i t wil l be taken as a lack of considera­

tion or even as an insu l t . In this sense asking a question i s comparable

to demanding. Both imply the speaker's b e l ie f that he has a right to what

he wants the hearer to do (which i s an unspecified kind of act in the case

of TO DEMAND and a l in g u is t ic act in the case of TO ASK A QUESTION).

TO INSIST is les s strongly directive and so i s TO REQUEST: the wish im-

209

-DUTCH- -SEMANTIC DIMENSION- -ENGLISH-

smeken, bepleiten

bevelen, gebieden, verbieden

oproepen

vragen, een vraag s t e l l e n

e isen

dreigen

aandringen

0uitdagen

verzoeken

vragen

uitnodigen

aansporen

aanmoedigen

aanbevelen

v o ors te l len , suggereren

aanraden, waarschuwen

toe laten

instemmen met

(Al)A

(A2)

to beg, to plead

to command, to order, tohi b i t

to summon

to ask (a question)

to demand

to threaten

to i n s i s t

to t e l l

to dare

to request

to ask

to inv i te

to exhort

to encourage

to recommend

to propose, to suggest

to advise , to warn

to permit

to acquiesce

SDC-Table 1

pro-

210

plied i s s t i l l strong but not becessar i ly based on a r ight and not strong

enough to cause a disruption of the social interaction i f the acts of

in s i s t in g and requesting are not acted upon.

12. TO SUMMON and TO INVITE both describe acts of a speaker trying to get

a hearer to go or come to some place. In sp i te of the near-synonymy they

are quite far apart in SDC-Table 1. The difference in the degree of d irec t ­

i v i t y i s probably the main dif ference between them: summoning i s 'order­

ing to come' while inv it ing i s 'asking to come'.

13. There are many ways to tran s la te TO TELL, as used in (D2) describing

(A4) conforming to (C4).

(A4) I f you wait a minute, I ' l l find you a table

(C4) Sa and Ha are in a restaurant; Sa i s the waiter , Ha i s the

customer

(D2) The waiter to ld the customer to wait unti l he'd find a table

for him

But there i s no Dutch verbial available with exactly the same degree of

d i r e c t iv i t y , which i s somewhere between requesting or asking and order­

ing.

D. DIRECTING AND ITS SOCIAL SETTINGS

14. Directing was said to be one of the basic functions in soc ia l in te r ­

action. I t i s not surprising, therefore , to find that the majority of

211

verbials of directing contain strong clues as to the social se tt ing or

se tt ings in which the acts they describe occur. Such clues are absent

from many of the basic verbs such as TO COMMAND, TO ASK, TO TELL, TO IN­

VITE, TO SUGGEST, TO ADVISE, e t c . ; a f ter a l l , i t i s because of the ir not

being sp e c i f ic or of the ir being 'neutral' with respect to a number of

semantic dimensions that they can be said to be more 'basic' than the

others. I t i s certainly not too bold to assume that the ensuing over­

view of the less neutral verbials of directing wil l s ingle out some areas

of social interaction in which directing other people's behavior i s a

prominent a c t iv i ty . The semantic dimension under invest igation in th is

section deals with the d if ferent frames of action with which l in g u i s t i c

acts of directing are associated.

15. A large number of verbials s i tuate the act described in a vaguely

' o f f i c i a l ' s e t t in g , without further sp ec i f ica t io n s . In other words, (Cl)

applies to the act described.

(Cl) The social se t t ing i s of an unspecified ' o f f i c i a l ' nature

Some examples are: TO ACCREDIT (in the sense of 'to authorize ') , TO APPLY

FOR, TO APPOINT, TO ASSIGN, TO AUTHORIZE, TO BLACKLIST, TO CERTIFY (in the

sense of 'to l i c e n s e 1, which i s a type of permission), TO CHARGE (in the

sense of imposing a certain duty or task ) , TO COMMISSION, TO CONVENE (in

the sense of causing to come together) , TO CONVOKE, TO GIVE CARTE BLANCHE,

TO GIVE FULL POWER, TO GIVE AN OPEN MANDATE, TO GIVE OFFICIAL SANCTION,

TO LICENSE, TO MEMORIALIZE (in the sense of 'to p e t i t i o n ' ) , TO PETITION,

212

TO POST (in the sense of assigning to a sp ec i f ic posit ion or s t a t io n ) ,

TO PRESENT A PETITION, TO QUALIFY, TO SUSPEND, TO VOUCHSAFE (in the sense

of granting as a previlege, allowing, permitting), TO WARRANT (in the sense

of 'to a thor ize ') ; Dutch: AANSTELLEN ( ' t o appoint), EEN AANVRAAG INDIENEN

( ' t o make an app l ica t ion ' ) , AANVRAGEN ( ' t o apply f o r ' ) , AANZEGGEN ( ' to

announce as an o f f i c ia l order') , AUTORISEREN ( ' t o authorize ') , BEKRACHTI-

GEN ( ' to give o f f i c i a l sa n ct io n ' ) , BELASTEN MET ( ' t o charge') , BENOEMEN

TOT ( ' to appoint') , BIOEENROEPEN ( ' t o convene'), CARTE BLANCHE GEVEN ( ' to

give carte blanche') , CONVOCEREN ( ' to convene, to convoke'), GELASTEN

( ' t o order o f f i c i a l l y and solemnly') , MACHTIGEN ( ' t o authorize ') , ONT-

BIEDEN ( ' t o inv i te o f f i c i a l l y and a u th o r i ta t iv e ly ' ) , OPDRAGEN ( ' t o charge'},

ORDONNEREN (equivalent to GELASTEN but rarely used), PETITIONEREN ( ' to

p e t i t i o n ' ) , SCHORSEN ( ' to suspend from membership or from du ty ') , SOLLI-

CITEREN NAAR ( ' t o aplly f o r ' ) , VEROORLOVEN { ' to allow o f f i c i a l l y ' ) , EEN

VERZOEKSCHRIFT INDIENEN ( l i t . 'to hand in a written request' meaning 'to

pet it ion ' or 'to memorialize'), VOLMACHT GEVEN ( ' t o authorize ' ) , EEN

VOORSTEL INDIENEN ( l i t . 'to hand in a proposal') .

The Dutch and the English l i s t s presented are quite p a ra l le l . One of

the f iner d i f ferences , however, i s the following. GELASTEN and ONTBIEDEN

describe orders and in v i ta t io n s , respect ively , of an unspecified o f f i c ia l

nature. No comparable English verbials seem to e x i s t : a l l the English

verbials of directing which describe o f f i c i a l acts of ordering and in­

v i t ing s i tua te the act in a more sp ec i f ic o f f i c i a l context; examples of

these wil l be adduced in the next few paragraphs.

213

16. Many verbials of direct ing s i tu a te the act described in a s p e c i f i c

o f f i c i a l s e t t in g . The most frequently spec i f ied o f f i c i a l contsxts are

l i s t e d in (C2) through (C6).

(C2) The act i s performed in a ' lega l ' se t t ing

(C3) The act i s performed in a ' r e l ig io u s ' s e t t in g

(C4) The act i s performed in a 'commercial' se t t in g

(C5) The act i s performed in a 'mil i tary' se t t in g

(C6) The act i s performed in a ' p o l i t i c a l ' s e t t in g

We shall supply an overview of the verbials focusing on l in g u i s t i c

actions of a d irec t ive kind conforming to each of these f iv e conditions. <

The reader is advised to keep in mind that c la s s i fy in g a verb or verb-l ike

expression as, for instance , a (C2) verbial does not mean that i t can

o n ly be applied to a d irec t iv e l in g u i s t i c action performed in a legal

context. I t only means that the verbial in question i s strongly associated

with that context.

17. By far the largest group i s the s e t o f (C2) verbials which s i tu a te

the acts described in a le g a l s e t t in g . The prominence of th is frame of

action in the domain of direct ing i s not surprising. After a l l , regu­

lat ing people's behavior i s what the law i s a l l about.

Here is a l i s t o f verbia ls focusing on l e g a l ly d irec t ive l in g u i s t i c

acts: TO APPEAL, TO ASSERT/VINDICATE A CLAIM/RIGHT/TITLE TO, TO BAN, TO

CITE (in the sense o f ca l l in g upon to appear before a court) , TO CLAIM,

TO CROSS-EXAMINE, TO CROSS-INTERROGATE, TO CROSS-QUESTION, TO DECREE,

214

TO ENJOIN, TO ECAMINE (in the sense of interrogating c l o s e l y ) , TO INTER­

DICT, TO ISSUE A WRIT/INJUNCTION, TO LAY CLAIM TO, TO LAY DOWN THE LAW,

TO LEGALIZE, TO LEVY, TO MAKE MANDATORY, TO MAKE REQUISITION, TO MANDATE,

TO NEGATIVE, TO OBTEST (in the sense of invoking as a w i tn e ss ) , TO ORDAIN,

TO OUTLAW, TO PROSCRIBE, TO PUT ONE'S VETO UPON, TO PUT UNDER AN IN­

JUNCTION, TO PUT UNDER AN INTERDICT, TO PUT UNDER THE BAN, TO READ THE

RIOT ACT (in the sense of ordering a mob to d isp erse ) , TO RECLAIM, TO

REGULATE, TO. REQUISITION, TO RULE, TO RULE AGAINST, TO SANCTION, TO STIPU­

LATE, TO SUBPOENA, TO SUMMON, TO TRY (in the sense of interrogating an

accused person) and TO VETO.

18. Some mebers o f the foregoing l i s t ca l l for further comments. TO SUMMON

can be used to i l l u s t r a t e my e a r l i e r observation that inc lusion o f a verb­

ia l in the l i s t does not imply that i t cannot be used to account for acts

performed outside the legal frame o f action. The verb refers to an o f f i c i a l

and authoritat ive in v i ta t ion which can be e i th e r issued by a court or not.

Thus, i t s uses in (Dl) as a description of (Al) and in (D2) as a descr ip t ­

ion o f (A2) are both acceptable.

(Al) You are requested to appear in court as a witness in the Cutler

murder t r ia l

(Dl) Sa summoned Ha (as a witness)

(A2) You are requested to appear before Mr. Allworthy at 9:00 A.M.

(D2) Sa summoned Ha to appear before Mr. Allworthy

But since summoning i s strongly associated with the court-room s i tu a t io n

one can hardly regard TO SUMMON as neutral with respect to the type of

215

o f f i c ia l sett ing in which the acts i t can be used to describe are performed.

In Dutch DAGVAARDEN would be used in (Dl) and ONTBIEDEN in (D2); hence

the former requires an act sat isfy ing (C2) whereas the la t ter only speci­

f i e s that (Cl) obtains. Though ONTBIEDEN is quite neutral with respect

to the type of o f f i c ia l se t t in g , i t may be interesting to point out that,

most probably because of the existence of DAGVAARDEN, ONTBIEDEN is hard

to apply to acts of the (Al) type. In English there i s , of course, also

the verb TO SUBPOENA which is completely restr icted to acts of the (Al)

type. The minute portion of the so c ia l - se t t in g dimension of directing

which i s under consideration here, i s visualized in SDC-Table 2.

DUTCB-

ontbieden

dagvaarden

-SEMANTIC DIMENSION-

( C l )

(C2)

-ENGLISH-

to subpoena#to summon

SDC-Table 2

19. The reader may have wondered why TO CLAIM, and TO REQUISITION were

included among the (C2) verbials . Consider (D3) and (D4).

(D3) He claimed his bags at the station

(D4)During the war the army requisitioned most hotels

The acts of demanding denoted by these two verbs are based on the speaker's

b e l ie f that he has a right to what he asks. Such b e l ie f can only be based

on laws, written or unwritten. That i s how a legal sett ing comes in.

216

TO READ THE RIOT ACT, in i t s original sense of ordering (a mob) to d is ­

perse, i s only obliquely related to a legal frame of action: the speaker

must be a person invested with the authority to execute the law. Needless

to say that i t s extended meaning of giving a strong warning, which i s more

prominent now, i s not related to a legal se t t in g .

F inally , the direct ive verbial i s most e x p l i c i t l y related to a legal

context, TO LAY DOWN THE LAW, i s more commonly used in i t s metaphorically

extended sense of giving s t r i c t orders (as a teacher, a father, e t c . ) than

as a description of a d irect ive act to which (C2) obtains.

20. The main (C2) verbials in Dutch are: AANSPRAAK MAKEN OP ( ' to c la im ') ,

IN DE BAN DOEN ( ' t o ban'}, BUITEN DE WET STELLEN ( ' t o outlaw'), DAGVAARDEN

( ' t o subpoena'), DECRETEREN ( ' t o decree') , EEN EIS INSTELLEN ( ' to demand'),

HEFFEN { 'to l e v y ' ) , INVORDEREN ( ' t o c la im ') , AAN EEN KRUISVERHOOR ONDER-

WERPEN ( ' to cross-examine'), LEGALISEREN ( ' t o l e g a l i z e ' ) , MANDATEREN

( ‘to mandate'), 5NACTI0NEREN ( ' t o san ct ion ' ) , SOMMEREN ( ' t o admonish'),

STIPULEREN { ' to s t i p u l a t e ' ) , TERUGVORDEREN ( ' t o rec la im ') , TOEWIJZEN ( ' to

a s s ig n ' ) , VERBANNEN { ' to ban') , VERHOREN ( ' t o t r y ' ) , VERORDENEN ( ' t o

ordain, to decree ' ) , ZIJN VETO UITSPREKEN OVER ( ' t o v e to ' , l i t . ' to pro­

nounce one's veto upon'), VORDEREN ( ' t o c la im ') , and WETTIGEN (*to l e g a l ­

ize' ).

21. When comparing the Dutch l i s t with the English one a number of con­

spicuous facts emerge. For instance, why are EEN EIS INSTELLEN, SOMMEREN

and TOEWIJZEN l i s t e d as (C2) verbials whereas the ir translat ions , namely,

TO DEMAND, TO ADMONISH and TO ASSIGN are not considered to focus on a

217

legal s e t t in g for the acts they describe ? The answer i s simply that they

refer to acts of demanding, admonishing and assigning in a legal context,

for which no d irec t English equivalents can be given. The Dutch equi­

valent of TO DEMAND i s EISEN, which only d i f fe r s from EEN EIS INSTELLEN

{ l i t . ' to i n s t i t u t e a demand1) in that the l a t t e r refers to an o f f i c i a l

jud ic ia l way o f demanding.

SOMMEREN, as used in (D5) and (D6), i s an act of admonishing to f u l f i l l a

certain duty or to comply with a request; the act i s based on jud ic ia l

authority on the part of the speaker or on a legal r ight .

{D5) De p o l i t i e sommeerde de menigte u i t elkaar te gaan

(The pol ice adminished the mob to disperse)

(D6) De winkelier sommeerde z i jn klanten to t betaling

(The shopkeeper admonished his customers to pay)

I f the more common equivalent to TO ADMONISH, namely AANMANEN TOT, had been

used, there would not have been d irec t legal implications.

TOEWIJZEN i s used to refer to jud ic ia l acts of assigning as appears from

description (D7).

(D7) Na de scheiding wees het gerecht haar de kinderen toe

( l i t . "After the divorce the court assigned the children to her")

This sentence means that "After the divorce the court granted her the

custody o f the children". TOEWIJZEN draws the attent ion to both the r ights

and the duties involved, whereas 'granting custody1 emphasizes the r igh ts .

218

22. The verbials of directing focusing on the legal se t t in g o f the act

described are abundant. This would probably not be the case in languages

spoken in s o c i e t i e s which lack an elaborate legal system. However, there

are no s o c i e t i e s without laws. But in more 'primitive' s o c i e t i e s be­

havior i s regulated by r e l ig io u s -e th ic a l codes rather than by what we would

ca l l ' c i v i l law'. Indeed, in many cases i t may be impossible to separate

these two types of regulating i n s t i t u t i o n s . The traces of such a s i tuat ion

are s t i l l to be found in the English and Dutch verbials o f d irect ing: a

considerable number of them are ambiguous between an emphasis on the frame

of c i v i l law or the frame of e c c l e s i a s t i c a l law. Nevertheless c i v i l law

c le a r ly dominates, which i s no doubt in keeping with regular observations

of the directing forces in our western s o c i e t i e s . My hunch i s that in the

d ir ec t iv e verbial sect ion of the lexicon o f every language a prominent

place w il l be occupied by verbia ls s i tuat ing l in g u i s t i c acts of directing

in a legal or r e l ig iou s s e t t in g , depending on the nature of the s o c ie ty .

This i s a hypothesis to be tes ted by further cross-cu ltural and cross-

l in g u i s t i c research on l in g u i s t i c action v erb ia ls .

23. These observations lead us to the verbials of d irect ing implying a

r e l i g i o u s context; in other words, (C3) obtains.

(C3) The act i s performed in a r e l ig io u s se t t in g

Among those which are c le a r ly ambiguous between (C2) and (C3) we find

TO ENJOIN (which i s e sp e c ia l ly strongly associated with the legal and

e c c l e s i a s t i c a l context in i t s prohibit ion se n se ) , TO INTERDICT (which

219

describes e c c le s ia t ic a l or judicia l prohibit ions) , TO ISSUE AN INJUNCTION

(which refers to an e c c le s ia s t i c a l or judicial order or prohibit ion) , and

TO ORDAIN (which denotes an order or appointment based on a court decision

or on the authority of the church). For al l of these , except for TI IN­

TERDICT (Du. EEN INTERDICT UITSPREKEN OVER, l i t . 'to pronounce an in ter ­

d ic t on1) , i t i s hard to find Dutch equivalents carrying the asame impli­

cations with respect to the social se t t ing .

Notice that we have only l i s t e d c learly ambiguous cases in th is para­

graph. There are, in addition, many (C2) verbials the use of which can be

extended without d i f f i c u l t i e s to (C3) contexts.-

24. Some verbials focusing on (C3) contexts are TO CATECHIZE ( i . e . giving

systematic re l ig ious instruction by means of asking questions, receiving

answers and offering explanations or corrections) , TO CENSURE (the meaning

of which can be e a s i ly extended to non-rel igious and non-ethical s e t t in g s ) ,

TO EXCOMMUNICATE ( i . e . to shut o f f by an e c c le s ia t ic a l sentence from commu­

nion with the church), TO PRECONIZE ( i . e . to approve a high e c c le s ia s t i c a l

appointment publicly by papal proclamation in consis tory) , TO PUT ON THE

INDEX and TO TABOO. Dutch: CATECHISEREN ( ' to c a te ch ize ' ) , CENSUREREN

{ ' t o censure') , EXCOMMUNICEREN { ' to excommunicate'), PRECONISEREN ( ' t o

preconize') , OP DE INDEX PLAATSEN { ' to put on the index') and TABOE VER-

KLAREN ( l i t . 'to declare taboo', i . e . 'to taboo') . Here the parallelism

between Dutch and English i s complete.

Apart from TO ORDAIN, TO CATECHIZE and TO PRECONIZE all the verbials

l i s t e d in th is and the previous paragraph refer to acts which are or can

be prohibit ive in nature. The predominance of negative d irect ives i s

220

certa in ly not unrelated to the prohibit ive nature o f Christian e th i c s .

After a l l , e igh t of the Ten Commandments are prohibit ions .

25. The (C3) verbials discussed so far describe acts of d irect ing the

behavior of people. Within a re l ig iou s or semi-rel ihious frame of action

people a lso try (or used to try) to d irec t the behavior o f supernatural

beings, i . e . gods, devi ls or s p i r i t s . Thus acts of directing s i tuated in

a re l ig iou s context can e i th er conform to (C3a) or (C3b).

(C3a) Ha i s /a r e (a) human being(s)

(C3b) Ha i s /a r e (a) supernatural being(s)

Whereas {C3) - ( C3a) defines a re l ig iou s or r e l ig io u s -e th ic a l context,

(C3)-{C3b) points at a relig ious-magical s e t t in g . Some (C3)-(C3b) verbs

are TO CONJURE (in the sense of summoning a supernatural being to appear

or to disappear; Dutch BEZWEREN), TO CONJURE UP, TO EXORCISE (Dutch UIT-

DRIJVEN). Also TO INVOKE (Dutch 0PR0EPEN) is often used in a (C3)-(C3b)

sense but can hardly be said t i imply a religious-magical se t t in g as part

of i t s central meaning.

26. Next in l in e i s the world of commevce. (C4) verbials describe attempts

to get money, goods or s e r v ic e s , to get someone to use or buy something,

or to make someone accept a particular price or condition for a commercial

transaction.

(C4) The act i s performed in a 'commercial' se t t in g

Some relevant verbials are: TO ADVERTIZE, TO BARGAIN, TO BOOK, TO CANVASS

221

(in the sense of seeking orders, contributions or subscriptions), TO

CHARGE (in the sense of asking as a fee or payment), TO COMMISSION (as used

in (08)) , TO HAGGLE, TO HIGGLE, TO INDENT, TO MAKE RESERVATIONS, TO ORDER

(as used in (D9)), TO PROMOTE, TO PUBLICIZE, TO PUT IN AN ORDER FOR,

TO RESERVE.

(D8) Most of his prominent contemporaries commissioned Rubens to

paint the ir portraits

(D9) He ordered a cheeseburger and a coke

The Dutch set of (C4) verbials i s quite analogous: ADVERTEREN ( ' to

advert ise ' ) , AFDINGEN { ' to bargain'), AFPINGELEN { ' to haggle1) , BE-

STELLEN ( ' to order') , BESPREKEN ( ' to book, reserve'; the original sense

of this verb, in which i t i s s t i l l currently used, i s 'to talk about';

another one of i t s derived but prominent meanings i s 'to review'),

RECLAME MAKEN VOOR ( ' to a dvert i se ' ) , RESERVEREN ( ' to reserve ' ) , WERVEN

( ' to canvass'). A (C4) verb without a parallel in English i s AANPRATEN

which means 'to talk into buying or using' .

27. When asked to sketch a prototypical directing or commanding s i tuat ion ,

informants can be expected to call attention to a m ilita ry frame of action.

Yet there are almost no (C5) verbials of directing.

(C5) The act i s performed in a 'military' sett ing

Indeed, the Dutch verb COMMANDEREN which must or ig ina l ly have been asso­

ciated with mili tary commands, has come to be used almost exclusively in

a semi-ironical sense to describe directing behavior based on assumed

222

— as opposed to real — authority; in many cases i t can be translated

by means of ' to order about1.

There seem to be only a couple of verbs focusing on (C5) contexts , but

even the use of these few can be e a s i ly extended outside the m il i tary

world: TO CONSCRIPT ( i . e . to enroll into service by compulsion) and

TO DRAFT; in Dutch: INLIJVEN ( ' t o conscript , d r a f t ' ) . TO CALL UP (Dutch

OPROEPEN) i s frequently used in the same sense but i t i s l e s s strongly

associated with the m il i tary .

Curiously enough, these ty p ic a l ly mil i tary commands are directed at

non-soldiers . The (C5) s i tuat ion could be subdivided into (C5)-(C5a)

and (C5)-(C5b).

(C5a) Ha i s not a so ld ier

(C5b) Ha i s a so ld ier

In Dutch I have not been able to discover verbia ls describing m il i tary

commands directed at so ld ie r s . As far as English i s concerned I am only

aware of the ex is tence o f one, namely TO ORDER UP (which also has the non­

m il i tary meaning of ordering someone to go or come u p s ta ir s ) . I t s m i l i ­

tary sense i s i l l u s t r a t e d in (DIO).

(DIO) The general ordered two batta l ions up to strengthen a weak

point in the l in e

I t means ' to order from a pos i t ion in the rear to the front l i n e ' .

Is there any explanation for the discrepancy between the conceptual­

iza t ion of the m il i tary world as a prototypical d irect ing s i tua t ion and

the lack of verbials of d irect ing focusing on a m il i tary se t t in g ? The

223

solution i s probably to be found in the very fa c t that the s i tuat ion i s

so prototypical: i f a mil i tary command i s a command p a r e x c e l l e n c e , there

i s no need to ca l l i t d i f f e r e n t ly .

28. Some verbials focus on p o l i t i c a l acts of d irec t in g , i . e . (C6) obtains.

{C6) The act i s performed in a ' p o l i t i c a l ' se t t ing

I t i s not surprising that the {C4) and (C6) s e t s share a member, namely

TO CANVASS (Dutch WERVEN) which means, in the (C6) s e t t in g , 'to s o l i c i t

votes or seek p o l i t i c a l support in an e le c t io n campaign'.

A typical (C6) verb i s TO INTERPELLATE (Dutch INTERPELLEREN), meaning

' to question formally aboit a governmental po l icy or d e c i s io n 1. A couple

of other verbials of dirceting s i t u a t e the act described in even more

precise soc ia l contexts , namely those sp ec i f ied in (C6a) and (C6b), the

s p e c i f i c i t y o f which increases progress ively .

(C6a) The act i s performed in a 'diplomatic' se t t in g

(C6b) The act i s performed in a frame o f ' inetrnational p o l i t i c a l -

economical diplomacy'

A (C6)-(C6a) verb i s TO ACCREDIT in the sense o f appointing as ambassador;

there i s no Dutch counterpart. The (C6)-(C6b) verbials I am thinking of

are TO EMBARGO and TO LAY AN EMBARGO ON (Dutch EEN EMBARGO LEGGEN OP)

in the sense of a governmental prohibit ion for the departure or entry of

foreign ships or for importing or exporting.

224

29. We are s t i l l l e f t with a wastebasket o f verbials o f d irect ing sp ec i ­

fying d i f feren t soc ia l contexts . Another more or l e s s o f f i c i a l se t t in g

i s mentioned in {C7).

(C7) The act i s performed in an 'educational' se t t in g

The Dutch verb EXAMINEREN ( ' t o examine') i s r e s tr ic ted to acts sa t i s fy in g

(C7). The English equivalent TO EXAMINE i s ambiguous between a legal and

an educational context. TO INTERROGATE (as well as i t s Dutch equivalent

ONDERVRAGEN) i s even wider in usage. Remember that the exc lus iv e ly legal

counterpart to EXAMINEREN was VERHOREN (English TO TRY).

A number of verbs s i tu a te the d irec t ive act described in the world of

crime.

(C8) The act i s performed in a 'criminal' se t t ing

The examples are: TO BLACKMAIL, TO EXTORT and TO SCREW (in the sense of

practis ing extortion upon). Dutch: AFDREIGEN ( ' t o blackmail ' ) , AFPERSEN

( ' t o e x t o r t ' ) , CHANTAGE PLEGEN and CHANTEREN (both 'to blackmail' ) .

A few iso la ted cases remain: TO INTERVIEW (Dutch INTERVIEWEN) points

at the world of the m e d ia ; a m e d ie a l se t t in g i s often focused upon by

TO PRESCRIBE (Dutch VOORSCHRIJVEN); and TO CALL OUT as used in (D ll ) i s

to be s i tuated in the world of i n d u s t r i a l r e l a t i o n s .

(D ll ) The union ca l led out a l l bus drivers

In such a context TO CALL OUT means 'to order or authorize workers to go

on s t r i k e ' ; there i s no obvious Dutch counterpart (unless one would re­

gard TOT STAKING OPROEPEN as a l ex ica l unity , which I am not inc l ined to

-DUTCH- -SEMANTIC DIMENSION-

een inter­dict u i t - spreken over

opdragen

dagvaardenverhoren

censureren

werven

[

bezweren

/ £ bestellen

in l i jven

0

interpen eren

0

{Cl)

[ (C2)

(C3)-(C3a)

(C3)-(C3b)

(C4)

{C5)-(C5a)

(C5)-(C5b)

[ (C6)

£ (C6)-(C6a)

een embargo leggen op £ (C6)-(C6b)

examineren I {C7)

chanteren (C8)

]]11

]]11J

SDC-Table 3

ENGLISH

to assign

to summon to try

to censure

to conjure

to order

to draft

to interdict

)■.\

1 to canvassto order up J

/

]>

/

to accredit

to embargo

• 1to blackmail

to examine

225

226

do). No doubt some additional iso lated cases could be found in the world

of sports (e .g . TO CALL has a very specialized direct ive meaning in card-

games).

The results of the discussion in this section are summarized in SDC-

Table 3.

E. DIRECTING AND IT S GOALS

30. Directing can be conceived as orienting somebody to some course of

action. Only in cases with a high degree of d ir e c t iv i ty can such a c t i ­

v ity be paraphrased as an attempt to get the hearer to do something.

The course of action towards which the hearer i s oriented, i . e . the goal

of the directing a c t iv i t y , i s specif ied in many verbials of directing.

There i s a strong correlation between the verbials s ituating the act

described in a particular social se t t ing and those specifying i t s goal.

In other words, there will be a s ign if ican t overlap between the sets of

verbials discussed in th is section and those in the previous one.

31. The most commonly emphasized goals of directing are specif ied in (Al)

through (A3).

(Al) Sa directs Ha towards an act of responding ( i . e . of performing

a l in g u s i t i c act)

(A2) Sa directs Ha towards an act of coming or going somewhere

(A3) Sa directs Ha towards an act of giving or granting (Sa) some­

thing

Ill

The s e t of (Al) verbials includes: TO ANGLE FOR, TO ASK (ABOUT), TO ASK

A QUESTION, TO CATECHIZE, TO CROSS-EXAMINE, TO CROSS-INTERROGATE, TO CROSS­

QUESTION, TO ELICIT, TO EXTRACT INFORMATION, TO EXAMINE, TO FISH FOR,

TO GRILL, TO INQUIRE, TO INQUISITION, TO INTERPELLATE, TO INTERROGATE,

TO INTERVIEW, TO MAKE INQUIRY, TO MAKE INQUISITION, TO PICK THE BRAINS OF,

TO POSE A QUESTION, TO PRY/PRIZE OUT, TO PUMP, TO PUMP FOR INFORMATION,

TO PUT A QUESTION TO, TO PUT ON THE GRILL, TO PUT QUERIES, TO QUERY,

TO QUESTION, TO QUESTIONNAIRE, TO QUIZ, TO ROAST, TO SHOOT QUESTIONS AT,

TO TAKE UP/CARRY ON AN INQUIRY, TO TRY, TO WORM OUT OF. Dutch: EXAMINEREN

( ' t o examine1, in the educational s e t t i n g ) , INFORMEREN ( ' t o ask for in ­

formation') , INTERPELLEREN ( ' t o i n t e r p e l l a t e ' ) , INTERVIEWEN ( ' t o in te r ­

v iew ') , AAN EEN KRUISVERHOOR ONDERWERPEN ( l i t . 'to subject to a cross-

examination') , ONDERVRAGEN ( ' t o in te rro ga te ' ) , ONTLOKKEN { ' t o e l i c i t ' ) ,

ONTWRINGEN ( l i t . 'towfrench out o f 1) , OP DE ROOSTER LEGGEN ( ‘to put on the

g r i l l ' ) , UITHOREN ( l i t . ' to hear o u t ' , i . e . ' to pump'), UITVRAGEN ( l i t .

' to ask o u t ' , i . e . to in te rro g a te ) , VERHOREN ( ' t o examine1, in the legal

s e t t in g ) , VISSEN NAAR ( ’to f i sh f o r ' ) , EEN VRAAG STELLEN ( ' t o ask a

q u es t io n ' ) , VRAGEN ( ' t o a s k ' ) , VRAGEN AFVUREN OP ( ' t o shoot questions a t ' ) .

This whole chapter deals with l i n g u i s t i c acts o f d irect ing . The foregoing

l i s t of (Al) verbials represents a subclass o f such acts aimed a t the pro­

duction of other l i n g u i s t i c a c ts . They d i f f e r from most of the others in

that the ir propositional content does not spec ify or even hint at the

actions to be carried out by the hearer. Such sp e c i f ic a t io n i s not necessary

because of the interrogat ive mood and other markers which signal the type

of d irec t iv e behavior under invest iga t ion . I f th i s type of d ir ec t iv e behavior

i s important enough to j u s t i f y the emergence o f a special grammatical mood,

228

i t should not surprise us that (Al) verbia ls occupy an important place

among the verbials o f d irect ing . Moreover, ethnomethodologists have

shown us c lear ly enough that e l i c i t i n g responses i s the most s a l i e n t

type of control the speaker usually t r i e s to exert over the course of

l in g u i s t i c in teract ion .

32. The (A2) v e rb ia l s , describing acts of trying to get the hearer to

come or go somewhere, include: TO BAN, TO BEG LEAVE, TO BID {in the sense

of i n v i t in g ) , TO BID COME, TO CALL, TO CALL AWAY, TO CALL BACK, TO CALL

IN, TO CALL OUT, TO CALL TOGETHER, TO CALL UP, TO CITE (in the sense of

summoning), TO CONJURE, TO CONJURE UP, TO CONVENE, TO COMVOKE, TO DEMAND

( in the sense of asking to see as in "The crowd demanded the star" , or

summoning to court) , TO DRAFT, TO EVOKE, TO EXORCIZE, TO EXTEND AN IN­

VITATION, TO GIVE LEAVE, TO INVITE, TO INVOKE, TO MUSTER (as in "They

did not muster much of a crowd"), TO OBTEST (in the sense of invoking as

a w i tn e ss ) , TO ORDER UP, TO RECALL, TO REQUEST THE PLEASURE OF SOMEONE'S

COMPANY, TO REQUEST THE PRESENCE OF, TO SEND AFTER, TO SEND FOR, TO SUB­

POENA, TO SUMMON, TO WARN OFF. Dutch: BEZWEREN ( ' t o e x o r c iz e ' ) , BIJEEN-

ROEPEN ( ' t o call t o g e t h e r ' ) , BINNENROEPEN { ' t o ca l l i n ' ) , CONVOCEREN

( ‘to convoke') , DAGVAARDEN ( ' t o summon'}, INROEPEN ( ' t o ca l l i n ' ) ,

INVITEREN { ' t o i n v i t e ' ) , LATEN HALEN ( ' t o send f o r ' ) , MEELOKKEN ( ' t o en­

t i c e someone to come a lo n g ' ) , MEETRONEN ( ' t o en t ice someone to come a lo n g ' ) ,

ONTBIEDEN ( ' t o i n v i t e ' ) , OPROEPEN ( ' t o d r a f t 1 or 'to ca l l u p ' ) , TERUG-

ROEPEN { ' t o call back ') , TE VOORSCHIJN ROEPEN ( l i t . ' to ca l l /o rd er to

appear') , TRONEN ( ' t o ent ice someone to go or come somewhere1) , UITDRIJVEN

( ' t o e x o r c iz e ' ) , UITNODIGEN ( ' t o i n v i t e ' ) , VERBANNEN ( ' t o ban ') , VERZOEKEN

229

( ' t o i n v i t e ' ) , WEGTRONEN ( ' t o ent ice to go away1).

The prominence of the (A2) verbia ls among the verbials o f d irect ing may

be surprising. What makes coming and going into such an important a c t i v i t y

that i t i s the focus of so many verbials of direct ing ? The fa c t i t s e l f

i s in terest ing to observe even though the answer to the question may be

r e la t iv e ly simple. Not only i s movement a basic category o f voluntary

action (voluntary action being the only reasonable object for acts of

d ir e c t in g ) , i t i s such a s a l i e n t cognit ive category in general that

for some languages such as Navajo (according to G. Witherspoon 1977) the

principal verb i s 'to g o 1 rather than ' to be' (which i s commonly re­

garded as the principal verb o f so many other languages).

33. The third prominent area in the goal dimension of the verbials of

direct ing i s the s e t of (A3) verbials describing acts of trying to get

the hearer to give or grant (the speaker) something. The s e t includes:

TO APPEAL TO ONE FOR, TO APPLY FOR, TO ASK FOR, TO BEG, TO BLACKMAIL,

TO CLAIM, TO DEMAND, TO EXACT, TO EXTORT, TO LAY CLAIM TO, TO MAKE DE­

MANDS/A DEMAND, TO MAKE REQUISITION, TO PRAY FOR, TO PUT IN A CLAIM,

TO PUT THE BITE ON (which means 'to ask for money, favors , e t c . ' as in

"Johnput the b i te on his friend for several t i c k e t s to the dance"), TO

REQUEST, TO REQUIRE, TO REQUISITION, TO SCREW, TO SOLICIT, TO STIPULATE

FOR, TO SUE FOR, TO SUPPLICATE, TO WHEEDLE SOMETHING OUT OF A PERSON.

Dutch: AANSPRAAK NAKEN OP { ' t o lay claim t o ' ) , EEN AANVRAAG INDIENEN

( l i t . ' to submit an a p p l ic a t io n ' ) , AANVRAGEN ( ' t o apply f o r ' ) , AFBEDELEN

( l i t . ' to beg o f f ' ) , AFDREIGEN ( ' t o extort by means o f threaten in g ') ,

AFDWINGEN ( ' t o e x t o r t ' ) , AFPERSEN ( ' t o e x t o r t ' ) , AFSMEKEN ( l i t . 'to beg

230

o f f ) , AFTROGGELEN { ' to wheedle something out of a person') , BEDELEN

{ ' to beg1) , BEDINGEN ( ' to st ipulate f o r ' ) , BIDDEN 0M ( ' t o pray f o r ' ) ,

CHANTAGE PLEGEN ( l i t . 'to commit blackmail'), CHANTEREN ( ' to blackmail'),

EISEN ( ' to demand'), EEN EIS INSTELLEN ( ' to put in a c la im ') , OPEISEN

( ' to claim, demand'), OPVRAGEN ( ' to claim') , RECLAMEREN ( ' to claim1) ,

SCH00IEN ( ' to beg', but a derogatory term), SOLLICITEREN NAAR { 'to apply

f o r 1) , TERUGEISEN ( l i t . 'to demand the return o f ) , TERUGVORDEREN ( ' to

demand the return o f ) , VORDEREN ( ' t o demand'), VRAGEN 0M ( ' to ask f o r ' ) .

What the (A3) verbials describe i s probably central in the area of

directing other people's behavior. This cannot be concluded from a com­

parison of the three sets of goal-spec if ic verbials l i s t e d so far. However,

a close look at the remaining goal-spec if ic verbials of directing reveals

that about half of them refer to acts of the (A3) type with an even further

specif ied goal. They wil l be presented in the next paragraph.

34. (A3a) thorugh (A3g) can all be paraphrased as attempts to get Ha to

undertake an action from which Sa will profit in some tangible material

way, i . e . to engage in an ac t iv i ty of 'giving' or 'granting' in a broad

sense.

(A3a) Sa directs Ha towards buying Sa's product

(A3b) Sa directs Ha towards paying something

(A3c) Sa directs Ha towards making or keeping something for Sa

(A3d) Sa directs Ha towards voting for Sa

(A3e) Sa directs Ha towards mailing something to Sa

(A3f) Sa directs Ha towards marrying Sa

231

(A3g) Sa d irects Ha towards having sexual intercourse with Sa

The principal (A3a) verbials are: TO ADVERTISE, TO CANVASS, TO PROMOTE,

TO PUBLICIZE and TO RECOMMEND; Dutch: AANBEVELEN { ' t o recommend'), AAN-

PRATEN ( l i t . 'to ta lk a person into buying something') , AANPRIJZEN (to

recommend'), ADVERTEREN ( ' t o a d v e r t i s e ' ) , RECLAME MAKEN VOOR ( ' t o adver­

t i s e ' ) , WERVEN ( ' t o canvass ') .

(A3b) verbials are: TO CHARGE, TO DUN and TO LEVY; Dutch HEFFEN ( ' t o

l evy ' ) and AANREKENEN ( ' t o charge') . There seems to be no obvious Dutch

equivalent for TO DUN in the sense of making p e r s i s ten t demands for

money as used in "The grocer dunned the customer monthly by mail for

payment o f his b i l l" or in "dunning for contributions".

The (A3c) verbials include: TO BESPEAK, TO BOOK, TO COMMISSION, TO

INDENT ( in the c h ie f ly Brit ish sense o f 'to order by an indent, i . e .

an o f f i c i a l requis i t ion as for s u p p l i e s ' ) , TO MAKE RESERVATIONS, TO

ORDER, TO RESERVE, TO PUT IN AN ORDER FOR; Dutch: BESPREKEN ( ' t o book,

re se r v e ' ) , BESTELLEN ( ' t o o rd er ' ) , RESERVEREN ( ' t o r e serv e ' ) . The actual

'g iv in g 1 towards which Sa's utterance i s directed i s of course delayed

in acts of the (A3c) type.

Attempts to get someone's vote can be described by means of the (A3d)

verb TO CANVASS, Dutch WERVEN. Note that the same word applies to seeking

p o l i t i c a l support and trying to a t t r a c t buyers.

Whereas (A3c) verbials refer to acts of 'giving' which are delayed,

the 'g iv in g 1 indicated by (A3e) verbia ls i s in d ir e c t , namely through the

mail. The relevant l ex ica l items are TO SEND FOR and TO ORDER, Dutch

AANVRAGEN and BESTELLEN.

232

In the remaining two se ts the object o f the 'giv ing' i s quite s p e c i f i c .

In the case of (A3g) verbials such as TO IMPORTUNE (in the sense of making

advances towards), TO SEDUCE and TO SOLICIT, Dutch EEN AANZOEK DOEN ( ' t o

s o l i c i t ' ) and VERLEIDEN ( ' t o seduce ') , Sa i s a f ter Ha's body. In the case

of (A3f) verbials such as TO POP THE QUESTION and TO PROPOSE, Dutch EEN

AANZOEK DOEN and IEMANDS HAND VRAGEN ( l i t . 'to ask someone's hand') Ha's

whole l i f e i s at stake.

35. The remaining s e t s of g o a l - s p e c i f i c verbials of d irect ing are a l l

r e la t iv e ly small. The acts they describe are l i s t e d in (A4) through (A16).

(A4) Sa d irects Ha towards not buying, using, e t c . something

(A5) Sa d irects Ha towards taking on a certa in duty

(A6) Sa d irects Ha towards going on s tr ike

(A7) Sa d irec ts Ha towards making war

(A8) Sa d irec ts Ha towards writing something down

(A9) Sa d irec ts Ha towards keeping something where i t i s

(A10) Sa d irects Ha (a mob) towards dispersing

(All) Sa d irec ts Ha towards attacking somebody

(A12) Sa d irec ts Ha towards accepting a lower price than he wanted

o r ig in a l ly

(A13) Sa d irec ts Ha towards changing his party or opinions

(A14) Sa d irects Ha towards hurrying up

(A15) Sa d irec ts Ha towards stopping his (profess ional) a c t i v i t i e s

temporarily

(A16) Sa d irec ts Ha towards a rebel l ion of some kind

233

We can b r ie f ly l i s t the most important lex ica l representatives .

(A4) verbia ls: TO BAN, TO BAR, TO BLACKLIST, TO CENSURE, TO OUTLAW,

TO PROSCRIBE, TO PUT ON THE INDEX, TO TABOO; Dutch: BUITEN DE WET STELLEN

( ' t o outlaw', l i t . 'to place outside the law1) , CENSUREREN ( ' t o cen sure ' ) ,

OP DE INDEX ZETTEN ( ' t o put on the in d ex ' ) , TABOE VERKLAREN ( l i t . 'to

declare taboo') . Notice that in acts of the (A4) type Ha i s never an

indiv idual , but a large group of people; the objects the f e r b ia l s take

refer to what the group of people in question i s not supposed to buy,

use, e tc .

The (A5) verbials: TO ACCREDIT (in the sense of 'to order to proceed

on an o f f i c i a l m is s io n ') , TO APPOINT, TO ASSIGN, TO CHARGE, TO COMMISSION,

TO MANDATE, TO ORDAIN, TO POST, TO PRECONIZE, TO SET and TO STATION;

Dutch: AANSTELLEN ( ' t o appoint ' ) , BELASTEN MET { ' t o charge') , BENOEMEN

TOT ( ' t o appoint ' ) , MANDATEREN ( ' t o mandate'), OPDRAGEN ( ' t o charge') ,

OPLEGGEN (*to charge1) , TOEWIJZEN ( ' t o a s s ig n ' ) .

I only know of one (A6) verb, namely TO CALL OUT as in ''Bus-drivers

may be ca l led out in support of the ir recent wage-claim". The Dutch

equivalent TOT STAKING OPROEPEN cannot be regarded as a lex ica l item.

Another one without an obvious Dutch counterpart i s the (A7) verb

TO DEFY in i t s quite s p e c i f i c meaning o f 'to challenge to combat'.

The main (A8) verb i s TO DICTATE, Dutch DICTEREN, in the sense o f 'to

speak or read for a person to write down or tran scr ib e ' .

TO EMBARGO and TO PUT AN EMBARGO ON, Dutch EEN EMBARGO LEGGEN OP, are

(A9) verbia ls .

TO READ THE RIOT ACT in the sense o f ordering Ha (which cannot be an

individual but which i s generally a mob) to disperse i s an (A10) verb ia l .

234

Apparently, a Dutch counterpart i s lacking.

The set of (All) verbs i s again a one-member c lass . The verb TO SIC,

Dutch AANHITSEN, means 'to inc ite to an attack'. The Ha of an (All) act

i s typically Sa's dog.

(A12) verbs: TO BARGAIN, TO HAGGLE and TO HIGGLE; Dutch: AFBIEDEN ( ' to

bargain'), AFDINGEN ( ' to bargain'), AFPINGELEN ( ' to haggle, h igg le ' ) .

TO TALK OVER as in "Fred is trying to talk Bill over to our side" i s

clearly an (A13) verbial; i t s Dutch equivalent in OMPRATEN.

OPJAGEN in the sense of 'to inc ite to excessive haste' i s an (A14)

verb. The same verb i s used to describe acts of starting or rousing game.

In i t s (A14) meaning i t i s hard to find a good English equivalent.

An (A15) verb: TO SUSPEND as in "He was suspended from the army for a

year", Dutch SCHORSEN.

Finally, a Dutch (A16) verb without a clear English equivalent i s

OPRUIEN which always describes an act of incit ing Ha to a rebellion of some

kind. Similar, at l ea s t in one of i t s senses, i s the verb OPZETTEN.

36. The goal dimension of l in g u is t ic acts of directing i s represented

in SDC-Table 4. In the table some comparisons are indicated which have

not been mentioned so far. By far the most important one i s the following.

Though in English acts of the (Al) and (A3) types are joined by the

common verb TO ASK, the lexical link i s not as strong as in Dutch.

TO ASK in i t s (A3) sense requires the preposition FOR whenever the object

of the asking i s indicated; as a resu lt , the (A3) verbial i s TO ASK FOR

rather than TO ASK. Though in Dutch there i s also a prepositional (A3)

verb VRAGEN OM, the verb VRAGEN can i t s e l f be used in the (A3) sense with-

235

---------------- Din CH--------------------- SEMANTIC D MENSION------------- ENGLISH-

Avragen |

\

een vraag s t e l l e n

uitnodigen

(Al

(A2

J to ask (a question)"

2 to i nvi te\

• 0 /vragen om (A3 ^ to ask for K

4 adverteren (A3a J to advertise

1werven t

aanrekenen

reserveren

(A3b

(A3c

'j to charge

to reserve

ii\•to canvassU4 t*

bestenerTI 0 (A3d 1 0 :'Ito order

kaanvragen (A3e *1 to send for

/4

u

een aan-j zoek d o e n ^

iemands hand vragen

verleiden

(A3f

(A3g

J to propose

-f to seduce J to sol i c i t r

taboe verklaren (A4 J to taboo

benoemen t o t (A5 ^ to appoint

0 (A6 | to call out

0 (A7 2 to defy

di cteren (A8 J to d ic ta te

een embargo leggen op (A9 J to embargo

0 (A10 J to read the r i o t act

aanhitsen (All 2 to s i c

afdingen (A12 2 to bargain

ompraten (A13 2 to talk over

opjagen (A14 1 »schorsen (A15 2 to suspend

opruien (A16 1 »SDC-Table 4

2 36

out adding a presposit ion , even when the object o f the asking needs to be

made e x p l i c i t .

F. THE DIRECTIONALITY OF DIRECTING

37. In the introduction to th is chapter we remarked that i t would not

be possib le to regard every d ir ec t ive speech act describable in terms

of the verbials of d irect ing as an attempt to get the hearer to do

something. Indeed, in some cases the speaker even t r i e s to make the

hearer n o t do something. Such cases can be accounted for by means of

verbs such as TO PROHIBIT. In th i s sect ion speech acts d irect ing the

hearer towards n o t doing something wil l be said to display a n e g a t i v e

d i r e c t i o n a l i t y as opposed to the p o s i t i v e d i r e c t i o n a l i t y of attempts

to make the hearer do something. Thus d ir ec t iv e speech acts can be divided

in to two types, (Al) and (A2).

(Al) Sa t r i e s to get Ha to do something

(A2) Sa t r i e s to get Ha not to do something

At f i r s t s igh t one could expect the verbia ls of d irect ing to f a l l into

two corresponding categories with TO COMMAND in the (Al) category and

TO PROHIBIT in the (A2) category. But an atgempt to place a l l our verb­

i a l s in one of the categories soon turned out to be in vain. The lexicon

i s patterned in a much more in tr ic a te way which I am about to sketch.

38. With respect to the d ir e c t io n a l i ty dimension a l l verb ia ls o f d ir e c t ­

ing can be placed on a nine-point sca le . At the two extremes there are

237

the e x c l u s i v e l y p o s i t i v e and e x c l u s i v e l y n e g a tiveo x \es . Next are those

which are i n t r i n s i c a l l y p o s i t i v e an those which are i n t r i n s i c a l l y n e g a t iv e .

I regard a verbial of directing as in t r in s i c a l ly pos it ive i f i t shows

pos i t ive d irec t io n a l i ty whenever combined with a pos it ive propositional

content, and negative d ir ec t ion a l i ty whenever the propositional content

of the description in which i t i s used i s negative. Conversely, a verbial

of directing i s i n t r in s i c a l ly negative i f i t s ignals negative direct ion­

a l i t y when combined with a pos i t ive propositional content and posit ive

direc t io n a l i ty when the propositional content of the description in which

i t i s used i s negative. As a r e su l t , the f i r s t se t nearly always indicates

a pos i t ive d irect ive whereas the second se t nearly always describes a

negative one. Examples wil l be given soon.

Further towards the center of the scale we should find verbials tending

towards the pos i t ive side and some tending towards the negative s ide . The

remaining se ts can be expected to be: those that can be e i ther pos i t ive

or negative, with equal probabili ty; those that can be pos i t ive and

negative simultaneously; those that are neither negative not p os i t ive .

The nine-point scale of the dimension under discussion should be for­

mulated d irect ly in terms of conditions on the use of the verbials (since

an a c t of directing can i t s e l f only be po s i t iv e or negative):

(Cvl) The verbial describes (Al) exclus ive ly

(Cv2) The verbial describes nearly always (Al)

(Cv3) The verbial describes usually (Al)

(Cv4) The verbial describes e i ther (Al) or (A2)

(Cv5) The verbial describes (Al) and (A2) simultaneously

(Cv6) The verbial describes neither (Al) nor (A2)

238

(Cv7) The verbial describes usually (A2)

(Cv8) The verbial describes nearly always (A2)

(Cv9) The verbial describes (A2) exclusively

The relevant verbials can now be l i s t e d systematically. But before doing

so, some preliminary remarks will be made.

39. An in tu it iv e ly plausible expectation would be to find TO COMMAND

at the positive extreme of the scale and TO PROHIBIT at the negative one.

However, these verbials are not exclusively positive and negative, res­

pectively. Consider (Dl) through (D4).

(Dl) He commanded the sold ier to scrub the f loor

(D2) He commanded the soldier not to leave the building

(D3) He prohibited the soldier to leave the building

(D4) He prohibited the soldier not to scrub the f loor

Of these descriptions, (D2) and (D4) are less l ik e ly to occur than the

others; indeed, (D4) wil l probably only be found in a logic handbook.

But the point i s that both are possible: (D2), with TO COMMAND, describes

a negative directive and (D4), with TO PROHIBIT, a pos it ive one. Thus

TO COMMAND and TO PROHIBIT f i r our definit ion (in the previous paragraph)

of i n t r i n s i c a l l y p o s i t i v e and i n t r i n s i c a l l y n e g a t iv e verbs of directing.

At this point one question crops up immediately: i f TO COMMAND and

TO PROHIBIT are not exclusively positive and negative, respectively, are

there any (Cvl) and (Cv9) verbials at all ? In f a c t , there are as many

exclusively positive and negative verbials of directing as there are

in tr in s ica l ly posit ive and negative ones. But the acts describable in

239

terms of the exc lus iv e ly p o s i t iv e and negative ones have rarely been regard­

ed as central cases of l in g u i s t i c d irect ing . An exc lus ive ly p o s i t iv e verb

i s , for instance, TO INVITE in i t s original sense o f requesting Ha's pre­

sence or part ic ipat ion . Someone's being present or his partic ipating

i s always a p o s i t iv e act. In i t s derivative sense of requesting formally

or p o l i t e l y , TO INVITE i s of course a (Cv2) verb. Another (Cvl) verb i s

TO ORDER as in "He ordered a meal"; in i t s other, non-commercial, senses

also TO ORDER belongs to the (Cv2) s e t . A (Cv9) verb would be TO BAN in

the sense of prohibit ing (by legal means or social pressure) the per­

formance of some a c t i v i t i e s or the dissemination or use of something; an

act o f banning in t h i s meaning can only be directed at a non-action.

40. By far the most numerous s e t s are the (Cvl) and (Cv2) verb ia ls . This

means that the lex icon r e f l e c t s a d e f in i t e tendency towards focusing on

the posutive aspect of d ir ec t ive behavior. The main English (Cvl) verb­

i a l s are: TO ADVERTISE, TO ANGLE FOR, TO APPLY FOR, TO APPOINT, TO APPROVE,

TO ASK (A QUESTION), TO ASK FOR, TO ASSIGN, TO BESPEAK, TO BID (COME),

TO BLACKMAIL, TO CALL (in the sense of commanding or requesting to come),

TO CALL AWAY, TO CALL BACK, TO CALL FOR, TO CALL FORTH, TO CALL IN,

TO CALL ON, TO CALL OUT, TO CALL TOGETHER, TO CALL UP, TO CANVASS, TO

CATECHIZE, TO CERTIFY (in the sense of l icens ing to do something as in

"to c e r t i f y a teacher"), TO CHARGE, TO CITE ( in the sense o f ca l l in g

upon o f f i c i a l l y or au thor ita t ive ly to appear before a court) , TO CLAIM,

TO CLAMOR FOR, TO COMMISION, TO CONJURE (UP), TO CONVENE, TO CONVOKE,

TO CROSS-EXAMINE, TO CROSS-INTERROGATE, TO CROSS-QUESTION, TO DEMAND

( in the sense of cla iming), TO DICTATE (in the sense of speaking or read­

240

ing for someone to write down), TO DRAFT, TO ELICIT, TO EXTRACT INFORMATION,

TO EVOKE, TO EXACT, TO EXAMINE, TO EXTEND AN INVIATTION, TO EXTORT, TO

FISH FOR, TO FOMENT (in the sense of incit ing as in "to foment r io t s , re­

volutions"), TO GIVE A FREE HAND, TO GIVE THE GO-AHEAD, TO GIVE THE GREEN

LIGHT, TO GRILL, TO IMPORTUNE (in the sense of making advances towards),

TOP IMPOSE, TO IMPOSE ON ONE FOR, TO INQUIRE (AFTER), TO INQUISITION,

TO INTERPELLATE, TO INTERRROGATE, TO INTERVIEW, TO INVITE (in the sense

of asking to come over), TO INVOKE, TO LAY CLAIM TO, TO LEVY, TO MAKE

INQUIRY, TO MAKE INQUISITION, TO MAKE REQUISITION, TO MAKE RESERVATIONS,

TO MANDATE, TO MUSTER, TO OBTEST (in the sense of invoking as a witness) ,

TO ORDER (as in "to order a meal"), TO ORDER UP, TO PICK THE BRAINS OF,

TO POP THE QUESTION, TO POSE A QUESTION, TO POST, TO PRECONIZE, TO PROMOTE,

TO PROPOSE (in the sense of asking to marry), TO PRY OUT, TO PUBLICIZE,

TO PUMP (FOR INFORMATION), TO PUT A QUESTION TO, TO PUT IN AN ORDER FOR,

TO PUT ON THE GRILL, TO PUT QUERIES, TO QUERY, TO QUESTION, TO QUESTIONNAIRE,

TO QUIZ, TO READ THE RIOT ACT (in the sense of ordering to disperse),

TO RECALL, TO REQUEST THE PLEASURE OF SOMEONE'S COMPANY, TO REQUEST THE

PRESENCE OF, TO REQUISITION, TO ROAST, TO SCREW, TO SEDUCE (in the sense

of enticing into partnership in sexual intercourse), TO SEND AFTER,

TO SEND FOR, TO SET (in the sense of appointing or assigning to an o f f ice

or duty as in "to set pickets around the camp"), TO SHOOT QUESTIONS AT,

TO SIC, TO SOLICIT (in the sense of offering sexual intercourse), TO STATION,

TO SUBMIT, YO SUBPOENA, TO SUMMON, TO TAKE UP AN INQUIRY, and TO WORM OUT OF.

Notice that the (Cvl) verbials include all those describing acts directed

at Ha's responding verbally, such as TO ASK A QUESTION. No need to explain

why these are exclusively pos i t ive . Another goa l-spec if ic category which

i s completely represented at the posit ive extreme of the d irect ionali ty

241

scale i s the one encompassing the lexical items used to account for acts

of directing Ha towards coming or going somewhere. The same is true for

the category of verbials describing attempts to make Ha give or grant

(Sa) something.

41. The Dutch set of (Cvl) verbials: AANPRATEN ( ' to talk a person into buying,

doing soething') , AANSPRAAK MAKEN OP ( ' t o c la im') , AANSTELLEN TOT ( ' to

appoint') , AANVRAGEN ( ' to apply for' or 'to send f o r ' ) , EEN AANZOEK DOEN

(in the sense of 'to pop the question' of trying to seduce), ADVERTEREN ( ' to

advertise1) , AFBEDELEN ( l i t . 'to beg o f f ' ) , AFDREIGEN ( ' to extort by means of

threatening1) , ADDWINGEN ( ' to e x to r t ' ) , AFERSEN ( ’to e x tor t ’ ) , AFPINGELEN

{ ' to haggle, g ig g le ' ) , AFSMEKEN ( l i t . to beg o f f ' ) , AFTROGGELEN { 'to wheedle

out of a person1) , BEDELEN ( ' to beg') , BELASTEN MET { 'to charge w ith ') ,

BEMOEMEN ( ' to appoint'), EEN BEROEP DOEN OP ( ' to call on1) , BESPREKEN

( ' to book'), BESTELLEN ( ' to order'as in "to order a meal"), BEZWEREN (in

the sense of 'to conjure up1) , BIJEENROEPEN ( ' to call together') , BINNEN-

ROEPEN ( ' to call i n ' ) , CHANTAGE PLEGEN ( ' to blackmail'), CHANTEREN ( ' to

blackmail'), CONCOCEREN ( ' to convoke'), DAGVAARDEN ( ' to subpoena'),

DICTEREN { 'to dictate' in the sense of reading or speaking for someone to

write down), EISEN (in the sense of ‘to c la im ') , EEN EIS INSTELLEN { 'to

ut in a c la im') , EXAMINEREN ( ' to examine' in the sense of 'to subject to

an examination', in the educational s e t t in g ' ) , HEFFEN ( ' to l e v y ' ) , INFOR-

MEREN ( ' to ask for information'), INROEPEN (both in the sense of 'to call in'

and 'to call on') , INTERPELLEREN { ' to in te r p e l la te ' ) , INTERVIEWEN ( ' to in­

terview1) , INVITEREN ( ‘to i n v i t e 1 in the sense of asking to come over') ,

INVORDEREN ( ' to c la im') , AAN EEN KRUISVERHOOR ONDERWERPEN ( ' to subject to

242

a cross-examination'} , LATEN HALEN { ' t o send f o r 1) , MANDATEREN ( ' t o mandate'),

MEELOKKEN { ' t o ent ice to come a lo n g ' ) , MEETRONEN ( ' t o ent ice to come a lo n g ' ) ,

ONDERVRAGEN ( ' t o in terrogate1) , ONTBIEDEN ( ' t o in v i t e o f f i c i a l l y ' ) , ONT-

LOKKEN ( ' t o e l i c i t ' ) , ONTWRINGEN ( l i t . ’to wrench out o f ' ) , OPEISEN ( ' t o

c la im ') , OPGEVEN { ' t o a s s ig n ' ) , OPJAGEN { ' t o ent ice to hurry up ') ,

OPROEPEN (in the sense of 'to d r a f t ' ) , OPRUIEN ( ' t o in c i t e to r e b e l l i o n ' ) ,

OPVRAGEN ( ' t o c la im ') , PRECONISEREN ( ' t o p recon ize ' ) , RECLAME MAKEN VOOR

( ' t o a d v e r t i s e ' ) , RECLAMEREN ( ' t o c la im ') , RESERVEREN ( ' t o r e se r v e ' ) ,

SCHOOIEN (pejorative for ' to b e g ' ) , SOLLICITEREN NAAR ( ' t o apply f o r ' ) ,

TERUGEISEN ( ' t o demand the return o f ' ) , TERUGROEPEN ( ' t o ca l l back') ,

TERUGVORDEREN ( ' t o demand the return o f ' ) , TE VOORSCHIJN ROEPEN ( l i t .

'to c a l l , order to appear') , TOEWIJZEN ( ' t o a s s ig n ' ) , TRONEN ( ' t o entice

to come or go somewhere'), UITDRIJVEN ( ' t o e x o r c i s e ' ) , UITHOREN ( l i t .

' to hear o u t ' , i . e . 'to pump for information') , UITNODIGEN (in the sense

of asking to come over, 'to i n v i t e ' ) , UITVRAGEN ( l i t . 'to ask o u t ' , i . e .

' to in te r r o g a te ' ) , VERHOREN ( ' t o examine', in the legal s e t t i n g ) , VERLEIDEN

( ‘to seduce ' ) , VERLOKKEN ( ' t o e n t i c e , sed u ce ' ) , IN VERZOEKING BRENGEN

( to seduce ') , VISSEN NAAR { ' t o angle f o r ' ) , VORDEREN ( ' t o demand' in the

sense o f asking to g i v e ) , EEN VRAAG STELLEN ( ' t o ask a q u es t io n ' ) , VRAGEN

( in the sense of asking a quest ion) , VRAGEN AFVUREN OP ( ' t o shoot questions

a t ' ) , WEGTRONEN ( ‘to entice to go away1) , DE WEG WIJZEN ( l i t . 'to show the

way', i . e . ' to give d i r e c t io n s ' ) , WERVEN ( ' t o canvass') .

42. The i n t r in s i c a l l y p o s i t iv e (Cv2) verbia ls are equally numerous, but i t

i s harder to re la te them to categories sp ec i f ie d previously in the d is -

243

cussion of different semantic dimensions. In th is class all the traditional

central examples of directive l in g u is t ic action verbs will be encountered

(but not only those). The English set: TO ACCEDE, TO ACCREDIT, TO BLACK­

MAIL, TO BRIEF, TO ACQUIESCE, TO ADJURE, TO ADMINISH, TO ADVISE, TO ADVO­

CATE, TO AGREE, TO ALLOW, TO ALLURE, TO APPEAL TO ONE FOR, TO APPLY

PRESSURE, TO APPROVE, TO ASK (in the sense of requesting), TO ASSENT,

TO AUTHORIZE, TO BEG, TO BESEECH, TO BULLY (SOMEONE INTO DOING SOMETHING),

TO CAJOLE, TO CALL (in the sense of requesting as in "He called for an

invest igation of the facts") , TO CALL ON (or UPON), TO CAHLLENGE, TO COACH,

TO COAX, TO COMMAND, TO CONJURE (in the sense of entreating earnestly or

solemnly as in "I conjure you to weigh my case wel l") , TO CONSENT, TO

COUNSEL, TO CRAVE (in the sense of asking earnestly as in "He craved his

superior's pardon"), TO DARE, TO DECREE, TO DEFY, TO DELIVER AN ULTIMATUM,

TO DEMAND, TO DICTATE (in the general sense of speaking commandingly

and imposing orders), TO DIRECT, TO DROP A HINT, TO EGG (ON) (in the

sense of provoking to action as in "They egged the ir governments on to

spend hundreds of mill ions"), TO ENCOURAGE, TO ENDORSE (in the sense of

expressing def in i te approval o f ) , TO ENTICE, TO ENTREAT, TO EXHORT,

TO GIVE CARTE BLANCHE, TO GIVE FULL POWER, TO GIVE AN OPEN MANDATE, TO

GIVE AN ORDER, TO GIVE A PIECE OF ADVICE, TO GIVE LEAVE, TO GIVE OFFICIAL

SANCTION, TO GIVE PERMISSION, TO GIVE THEWORD (OF COMMAND), TO GOAD,

TO GRANT, TO HINT, TO IMPETRATE, TO IMPLORE, TO IMPORTUNE (in i t s general

sense of making frequent and troublesome requests or begging), TO INCITE,

TO INDUCE, TO INSIST ON, TO INSTIGATE, TO INSTRUCT, TO INVEIGLE (as in

"He used the most subtle means to inveigle the author into the o f f i ce " ) ,

244

TO INVITE ( in the general sense of requesting p o l i t e l y ) , TO ISSUE A

COMMAND, TO LAY DOWN (THE LAW), TO LEGALIZE, TO LICENSE, TO LURE, TO MAKE

A DEMAND, TO MAKE A REQUEST, TO MAKE OBLIGATORY (MANDATORY), TO MANI­

PULATE, TO OBTEST (in the sense of beseeching), TO OKAY, TO ORDAIN,

TO ORDER, TO PERMIT, TO PERSIST, TO PERSUADE, TO PETITION, TO PLEAD,

TO PRAY, TO PRESCRIBE, TO PRESENT A PETITION, TO PRESS, TO PRICK, TO

PROD, TO PROMPT, TO PROPOSE, TO PUT A FLEA (BUG) IN SOMEONE'S EAR ( i . e .

to give a hint or secre t information to make someone act as in "He saw

Mary at the jeweler 's admiring the diamond pin; so he was able to put

a bug in Henry's ear"), TO PUT THE PRESSURE ON, TO PUT THE SCREWS TO,

TO QUALIFY (in the sense o f l i c e n s in g ) , TO RECOMMEND, TO REGULATE,

TO REQUEST, TO REQUIRE ( in the sense o f asking a u th o r i ta t iv e ly ) , TO RULE,

TO SANCTION, TO SAY THE WORD, TO SEDUCE (in the sense of persuading into

disobedience, d i s lo y a l ty , e t c . ) * TO SET (in the sense of 'to de c re e ' ) ,

TO SET CONDITIONS, TO SOLICIT (in the sense of in c i t in g or strongly

urging), TO SPUR, TO STAND ON, TO STIPULATE, TO SUBSCRIBE, TO SUE, TO

SUGGEST, TO SUPPLICATE, TO TELL, TO TEMPT, TO TROUBLE SOMEONE FOR, TO

URGE, TO VOUCHSAFE, and TO WHEEDLE.

43. No doubt there i s enough to surprise the reader when going through

these l i s t s . For instance , how can we explain that TO GIVE A FREE HAND,

TO GIVE THE GO-AHEAD and TO GIVE THE GREEN LIGHT seem to be ex c lu s iv e ly

applicable to pos i t ive d ir e c t iv e s , whereas a very s im ilar verbial such

as TO GIVE CARTE BLANCHE, though i n t r in s i c a l l y p o s i t i v e , seems to allow

for negative d ir e c t io n a l i ty as well ? And, to take another example, why

245

i s the Dutch EEN BEROEP DOEN OP d e f in i t e ly a (Cvl) verbial whereas the

English equivalent most commonly used to translate i t , namely TO CALL ON,

had to be l i s t e d in the (Cv2) s e t ? We have hardly started to make any

systematic observation o f th i s kind. I t i s quite natural, therefore , that

explanations are s t i l l far o f f in the future.

44. The Dutch (Cv2) verbia ls include: AANBEVELEN ( ' t o recommend1) , AAN-

DRINGEN { ' t o urge ') , AANHITSEN ( ' t o egg on ' ) , AANMANEN ( ' t o ur g e ' ) ,

AANMOEDIGEN ( ' t o encourage') , AANPORREN ( ' t o i n c i t e ' ) , AANPREKEN { ' to

recommend s t r o n g ly ' ) , AANPRIJZEN ( ’to recommend'), AANRADEN ( ' t o a d v i s e ' ) ,

AANSPOREN ( ' t o i n c i t e ' ) , AANSTICHTEN ( ’to i n s t i g a t e ' ) , AANSTOKEN ( ' t o

foment, i n s t i g a t e ' ) , EEN AANVRAAG DOEN/INDIENEN ( ’to make an a p l l i c a t i o n ' ) ,

AANVUREN ( ' t o i n c i t e ' ) , AANWAKKEREN ( ' t o foment') , AANWIJZINGEN GEVEN

( ' t o give d i r e c t io n s ' ) , AANZEGGEN ( ' t o order o f f i c i a l l y ' , a sense in

which the verb i s rarely used), AANZETTEN ( ' t o i n c i t e 1) , EEN AANZOEK DOEN

(OM) (in the sense o f making a solemn, serious request) , VAN ADVIES DIE-

NEN ( ' t o co u n se l ' ) , ADVISEREN ( ' t o a d v i s e ' ) , AUTORISEREN ( ’to a u th or ize ' ) ,

BEKRACHTIGEN ( ' t o sa n c t io n ' ) , BEPALEN ( ' t o r u l e ' ) , BEPLEITEN ( ' t o plead

f o r ’ ) , BEPRATEN ( ' t o persuade') , EEN BEVEL GEVEN { ' t o give an order ' ) ,

BEVELEN ( ' t o order’ ) , BEWEGEN TOT { ' t o i n c i t e ' ) , BEZWEREN ( ' t o adjure ' ) ,

BIDDEN (OM) ( ' t o pray f o r ' ) , BILLIJKEN ( ' t o approve o f ' ) , BRENGEN TOT

( ' t o i n c i t e ' ) , CARTE BLANCHE GEVEN ( ' t o give carte b lanche') , COMMANDEREN

( ' t o command'), DECRETEREN ( ' t o d e c re e ' ) , DICTEREN ( ' t o d ic ta te in i t s

general sense of speaking commandingly or imposing orders) , DRUK UITOE-

FENEN OP ( ' t o put pressure o n ' ) , EISEN ( ' t o demand'), EEN EIS STELLEN

246

( ' t o make a demand1) , GEBIEDEN ( ' t o order ' ) , GOEDKEUREN ( ' t o approve') ,

IEMAND HET MES OP DE KEEL ZETTEN ( l i t . 'to put the knofe on someone's

th ro a t ' , meaning 'to put the screws t o ' ) , IEMAND IETS IN HET HOOFD PRATEN

( l i t . ' to talk something into someone's head') , IEMAND IETS OP HET HART

DRUKKEN ( l i t . ' to press something on to someone's hear t ' , i . e . to urge

a person t o ) , IEMAND LASTIG VALLEN MET ( ' t o importune' in the sense of

urging with frequent or unreasonable requests) , IEMAND VOOR IETS WINNEN

( l i t . ' to win someone for something') , INBLAZEN ( ' t o su g g es t ' , l i t . ' to

blow i n ' ) , INGEVEN ( ' t o s u g g e s t ' ) , INSTEMMEN MET ( ' t o accede ') , INSTRU-

EREN ( ' t o i n s t r u c t ' ) , INWILLIGEN ( ' t o accede t o ' ) , LEGALISEREN ( ' t o

l e g a l i z e ' ) , MACHTIGEN ( ' t o a u th o r iz e ' ) , MACHTIGING GEVEN TOT ( ' t o author­

i z e ' ) , MANIPULEREN ( ' t o manipulate') , MISRADEN ( ' t o advise wrongly') ,

OPDRAGEN ( ' t o order ' ) , OPHITSEN ( ' t o i n s t i g a t e ' ) , OPLEGGEN ( ' t o charge

w ith ' ) , OPPEREN ( ' t o sugges t1) , OPROEPEN (in the sense of ca l l in g on) ,

OPSTOKEN ( ' t o i n s t i g a t e ' ) , OPZWEPEN ( ' t o i n c i t e ' , l i t . ' to whip up ' ) ,

ORDONNEREN ( ' t o order' , rarely used) , OVERHALEN TOT ( ' t o persuade') ,

OVERREDEN ( ' t o persuade') , PETITIONEREN ( ' t o p e t i t i o n ' ) , PLEITEN VOOR

( ' t o plead f o r ' ) , RAAD GEVEN ( ' t o a d v i s e ' ) , OP ZIJN RECHTEN STAAN ('to

assert one's r i g h t s ' ) , REGLEMENTEREN ( ' t o r e g u la t e ' ) , SANCTIONEREN ( ' t o

sa n c t io n ' ) , SMEKEN ( ' t o b eg ' ) , SOMMEREN ( ' t o demand a u t h o r i t a t i v e l y ' ) ,

STAAN OP ( ' t o i n s i s t ’ , l i t . ' to stand o n ' ) , STIPULEREN { ' t o s t i p u l a t e ' ) ,

SUGGEREREN ( ' t o s u g g e s t ' ) , TARTEN ( ' t o d a r e ' ) , TOEGEVEN ( ' t o accede ') ,

TOELATEN ( ' t o perm it ' ) , TOELATING GEVEN ( ’to give permiss ion') , TOESTAAN

( ' t o perm it ' ) , TOESTEMMEN IN ( ' t o accede ') , TOESTEMMING GEVEN ( ' t o give

permission') , UITDAGEN { ’to c h a l le n g e ' ) , EEN ULTIMATUM STELLEN { ' t o de­

247

l iv e r an ultimatum'), VASTLEGGEN { ' t o s e t ' ) , VERGUNNEN ( ' t o permit') ,

VERGUNNING GEVEN/VERLENEN ( ' t o give permission') , VEROORLOVEN ( ' t o per­

m i t ' ) , VERORDENEN ( ' to decree, ordain') , VERZOEKEN ( ' t o request ' ) ,

EEN VERZOEKSCHRIFT INDIENEN ( ' to request formally and in w r i t in g ' ) ,

VOET BIJ STUK ZETTEN ( ' t o i n s i s t ' ) , VOLMACHT GEVEN ( ' t o authorize ') ,

VOORSCHRIJVEN ( ' to p rescr ibe ' ) , VOORSLAAN ( ' t o suggest, propose'), EEN

VOORSTEL INDIENEN ( ' to make a propos i t ion ') , VOORSTELLEN ( ' to suggest,

propose') , VOORWAARDEN STELLEN ( ' to make cond it ions ' ) , VRAGEN ( ' t o ask ' ) ,

ZICH WENDEN TOT { 'to ca l l o n ' ) , WETTIGEN ( ' to l e g a l i z e ' ) .

45. Again we can ask a number of questions to which we do not know any

answer. For instance, why i s OPLEGGEN a (Cv2) verb whereas i t s most common

English equivalent TO CHARGE is exclus ive ly pos i t ive ? The t r iv ia l answer

i s , of course, that the ir meanings do not overlap completely. But i s th is

purely accidental , or i s there a deeper rationale ? Though I am unable to

answer the question, the in tu i t iv e s im ilar i ty in meaning makes me re­

luctant to accept the accident-hypothesis without further scrutiny.

Another potent ia l ly interest ing fa c t i s the difference in semantic

patterning between TO ASK and Dutch VRAGEN. When discussing the goal

dimension of the verbials of directing we were forced to distinguish be­

tween two types of asking: TO ASK in the sense of ASKING A QUESTION as in

(D5), and TO ASK FOR in the sense of requesting Ha to give or grant (Sa)

something, as in (D6). The d irec t io n a l i ty dimension o f directing imposes

a third type on us: TO ASK simply in the sense of requesting, as in (D7).

(D5) He asked where the stat ion was

248

( D6) He asked for money

(D7) He asked me to go home

In a l l three cases Dutch VRAGEN could be usedthough in the case o f (D6)

the preposition OM may be added. Both TO ASK in the (D5) sense and TO ASK

FOR are {Cvl) verb ia ls , whereas TO ASK, as well as VRAGEN, in the (D7)

sense i s a (Cv2) verb.

46. Let us jump immediately to the opposite pole of the s c a le , i . e .

the ex c lu s iv e ly and i n t r i n s i c a l l y negative verbials o f d irec t ing . Both

s e t s are much smaller than the corresponding p o s i t iv e ones. Some English

(Cv9) verbials are: TO BAN {as in "to ban a p o l i t i c a l party, a book", im­

plying a categorical c i v i l or e c c l e s i a t i c a l prohib it ion) , TO BAR {as in

"barring the use of poison gas in war"), TO BLACKLIST, TO CENSURE, TO EM­

BARGO, TO EXPOSTULATE {in the sense of reasoning earnest ly with a person

for purposes of d is suas ion) , TO INTERDICT, TO LAY AN EMBARGO ON, TO RE­

MONSTRATE (in the sense of expostu la t ing) , TO NEGATIVE (in the sense of

refusing assent t o ) , TO PROSCRIBE, TO PUT ONE'S VETO UPON, TO PUT ON THE

INDEX, TO PU UNDER AN INTERDICT, TO PUT UNDER THE BAN, TO READ THE RIOT

ACT (in the sense of issueing a peremptory warning to cease doing some­

thing or not to do something again, as in "Three boys were la te to c lass

and the teacher read the r io t act to them"), TO SUSPEND, TO TABOO, TO VETO.

The reader may have noticed that almost a l l the verbia ls of d irect ing

specifying the act (not) to be performed belong e i th er to the (Cvl) or the

(Cv9) category. This i s hardly surprising since the behavior towards which

Ha i s directed i s e i th er a p o s i t iv e or an negative act; as a r e s u l t , i f the

249

act in question i s sp ec i f ie d also i t s being p os i t ive or negative i s l ik e ly

to be sp ec i f ied and hence the d ir e c t io n a l i ty of the verb i s f ixed .

Some Dutch (Cv9) verbia ls are: CENSUREREN { ' t o censure' ) , OP DE INDEX

PLAATSEN ( ' t o put on the ind ex ' ) , EEN INTERDICT UITSPREKEN OVER ( l i t . 'to

pronounce an in terd ic t about') , SCHORSEN ( ' t o suspend'), TABOE VERKLAREN

( l i t . 'to declare taboo' ) , UITSLUITEN (in the sense of not admitting),

ZIJN VETO UITSPREKEN OVER ( ' to v e t o ' ) .

47. The l i s t of i n t r i n s i c a l l y negative (Cv8) verbials can be equally

short: TO DISALLOW, TO DISCOURAGE, TO FORBID, TO INHIBIT ( in the sense of

prohibiting from doing something), TO OUTLAW, TO PROHIBIT, TO RULE

AGAINST, TO RULE OUT (not ice that TO RULE was l i s t e d as a (Cv2) verb),

TO DISSUADE; Dutch: AFBRENGEN VAN ( ' t o d is su ad e ' ) , AFRADEN ( ' t o advise

a g a in s t ' ) , BUITEN DE WET STELLEN ( ' t o out law') , IEMAND IETS UIT HET

HOOFD PRATEN ( l i t . ' to ta lk something out of a person's head', i . e . to

dissuade), ONTMOEDIGEN ( ' t o discourage1) , ONTRADEN ( ' t o advise a g a in s t ' ) ,

VERBIEDEN ( ' t o p r o h ib i t ' ) .

Whereas the central instances o f po s i t ive d irec t ive s such as TO ORDER

and TO COMMAND were found to be (Cv2) verbs, The central negative example,

TO PROHIBIT, turns out to be a (Cv8) verb. So f a r , the symmetry i s com­

ple te .

48. Moving in the d irect ion of the center of the d ir e c t io n a l i ty sca le

we come across the (Cv3) and (Cv7) s l o t s . In those s l o t s we should f ind

verbials which are not i n t r i n s i c a l l y negative or p o s i t i v e , but which, in

a vague probablity-of-occurrence sense , tend towards the negative or

250

posit ive poles. At the negative side we f ind {Cv7) verbials such as TO

ADDRESS A WARNING TO, TO CAUTION, TO GIVE FAIR WARNING, TO ISSUE A CAVEAT,

TO UTTER A CAVEAT, and TO WARN; Dutch: WAARSCHUWEN { ' to warn') and EEN

WAARSCHUWING GEVEN ( ' t o give a warning'). No doubt a description such as

(D8) i s more l ik e ly to occur than (D9) though both are acceptable.

(D8) I warned him not to stay there

(D9) I warned him to go home

Therefore warning, as a d ir ec t ive , tends towards the negative pole of

the d irec t ion a l i ty sca le . But TO WARN i s not in tr in s ic a l ly negative because

i t s d irec t io n a l i ty i s not negative , but p o s i t iv e , when i t i s followed

by a pos it ive complement, and negative when a negative propositional

content signals the act towards which Ha i s directed.

To preserve the symmetry o f the scale one would expect TO ADVISE to f i l l

the (Cv3) s l o t . But TO ADVISE was, quite correct ly , l i s t e d as an in tr in ­

s i c a l l y pos i t ive direct ive . Thus the (Cv3) s l o t remains empty. This shows

that there must be more to warning than (usually) negative advising; the

Dutch lexicon shows, moreover, that negative advising (re f lec ted in the

verb AFRADEN as opposed to AANRADEN) belongs in the (Cv8) rather than the

(Cv7) se t . How, then, can we explain that TO WARN tends towards the ne­

gative pole of the scale though i t does not behave l ike in tr in s ic a l ly

negative d irect ive verbials but rather l ik e in t r in s i c a l ly pos it ive ones ?

The question wil l remain unanswered in th i s essay.

49. Every act of directing pushes Ha to e i th er doing or not doing something.

Thus i t comes as no surprise to find that there are no (Cv6) verbials e ither .

251

This leaves us with (Cv4) and (Cv5). There are some English verbia ls which

can be used to describe e i th er pos i t ive or neagtive acts o f d irec t in g , with

equal probabili ty: TO ENJOIN, TO ISSUE A WRIT/AN INJUNCTION, TO MENACE,

TO PUT ONE'S FOOT DOWN, TO PUT UNDER AN INJUNCTION, TO THREATEN, and

TO WARRANT (in the sense o f giving authority or power for doing or for ­

bearing to do something). Dutch : DREIGEN ( ' t o threaten1) and OMPRATEN

(which can be e i th er 'to persuade1 or ' to d issuade') . Notice that apart

from the pair TO THREATEN-DREIGEN, the English and Dutch se t s of (Cv4)

verbials do not contain equivalents .

50. I have not been able to discover any verbials o f d irect ing which

are necessari ly of the (Cv5) type. However, there are quite a few which

can e i th er s a t i s f y (Cv4) or (Cv5), depending on the context. All of them

ind icate s e r ie s of d irect ive acts (which, in the case of the (Cv5) in te r ­

pretat ion , are a mixture of p o s i t iv e and negative ones) rather than a

s in g le one. The English examples: TO CALL THE SHOTS, TO CALL THE SIGNALS,

TO PLAY/CALL THE TUNE, TO MAKE THE RULES and TO ORDER ABOUT/AROUND. Dutch:

COMMANDEREN (which i s often used more or l e s s in the sense of 'to order

about' ) , IEMAND DE WET STELLEN (more or l e s s equivalent to ' to make the

r u l e s ' ) , and RINGELOREN (s im ilar to 'to order about') .

The d ir e c t io n a l i ty dimension o f d irect ing i s summarized in SDC-Table

5. Keep in mind th a t , in the case of the (Cv4) s l o t , the verbia ls on both

sides of the dimension are not necessar i ly each other's equivalents .

252

vragen

-DUTCH-

vragenuitnodigen

vragenverzoekenbevelen

0

ompraten

0

0

waarschuwen

verbieden

een in te rd ic t uitsprekenover

-SEMANT

commanderen

[

C DIMENSION- -ENGLISH-

Cvl

Cv2

Cv3

Cv4

Cv5

Cv6

Cv7

Cv8

Cv9

]]

to ask (a question) to ask for to in v i te

to ask to request to order

0

to ask

to order aboutto enjoin

] 0

] '"2 to warn

J to prohibit

to put under an in terd ic t

SDC—T a b le 5

G. DIRECTING AND AUTHORITY

51; Though authority can be expected to be one o f the major phenomena in ­

volved in acts of d irect ing and to understand in order to grasp the nature

of such behavior, i t s discussion in the l i t e r a tu r e on speech acts i s res­

t r i c t e d to some cursory statements such as J.R. Sear le 's claim that the

253

d is t in c t io n between a request and a command i s simply that the l a t t e r im­

p l i e s authority on the part of the speaker whereas the former lacks such

an implication. The purpose o f th i s sect ion i s not to provide support

for these cursory statements, nor to refute them. My intention i s simply

to show that authority rea l ly const i tu te s a complex semantic dimension

in the l ex ica l frame of d irect ing . Authority i s not ju s t e i th e r absent

or present. Nor i s the authority dimension simply a matter o f gradabil i ty .

The verbials o f directing force us to d is t ingu ish a whole range of

d i f feren t typ es o f authority .

52. Very c lear cases of d irec t iv e verbials implying authority of some

kind are TO ORDER (Dutch BEVELEN) and TO COMMAND (Dutch BEVELEN). They

can be used to describe l in g u i s t i c acts of type (Al) conforming to

condition (Cl).

(Al) Sa directs Ha towards some action

(Cl) Sa has authority over Ha

A l e s s s a l i e n t but equally c lear example i s TO TELL as used in (Dl) .

(Dl) The waiter told us to wait

The main difference between TO TELL and TO ORDER or TO COMMAND seems to

be that the former implies authority of a weaker kind, though e s s e n t ia l ly

of the same type. Also TO PERMIT (Dutch TOELATEN) belongs in the (Cl) c a te ­

gory.

The reader may have noticed that both TO ORDER and TO COMMAND have BE­

VELEN as a Dutch equivalent. Also TO TELL can be translated by means of

254

BEVELEN under certain conditions, but in the case of (Dl) i t i s hard to

find any good Dutch equivalent at a l l . The most l ike ly translation for

(Dl) would be (D2)

(D2) De kelner deed ons wachten

hich means l i t e r a l ly "The waiter made us wait"; a good lexical equivalent

seems to be lacking.

53. The number of verbials applicable to directive acts conforming to (Cl)

i s extremely large. But what about all the others ? Do they all imply a

simple negation of (Cl) ? Take, for instance, TO ADVISE (Dutch AANRADEN)

and TO WARN (Dutch WAARSCHUWEN). I t i s clear that a piece of advice and

a warning also involve a type of authority, which could be called knowledge

a u th o r i ty as opposed to the power a u th o r i ty underlying orders and commands.

In other words, in the case of advising and warning Sa has authority over

Ha as a result of knowing more. Therefore (Cl) has to be made more spe­

c i f i c .

(Cla) Sa has power authority over Ha

(Clb) Sa has knowledge authority over Ha

( I t i s by virtue of the concept of 'knowledge authority1 that an expert

in a particular f i e ld can be called an 'authority' in that f i e l d . ) Direct­

ive acts conforming to (Cla) are describable in terms of TO ORDER e t c . ,

whereas those satisfying (Clb) can be accounted for by means of TO ADVISE

etc .

255

So far so good. But are there also mixed categories of acts , conform­

ing both to (Cla) and (Clb), or are these two mutually exclusive ? And

i f they are not mutually exclusive, are these mixtures also reflected

in the lexicon ? In other words, are there any (Clc) verbials ?

(Clc) Sa has both power and knowledge authority over Ha

A clear case seems to be TO INSTRUCT (Dutch INSTRUEREN).

54. A next question to ask i s the following: Should a l l directive acts

which are not describable in terms of (Cla), (Clb) or (Clc) verbials be

regarded as conforming to (C2) ?

(C2) The act involves no authority

There are no doubt a large number of (C2) verbials including TO HINT,

TO SUGGEST, TO INSIST, TO ASK (in i t s requesting sense) , and maybe even

TO REQUEST. There i s no need to l i s t them all here.

But we have not ye t reached the end of the authority story. Not only

are there different kinds of authority which the speaker may have over

the hearer. The lexicon imposes a fact on us the formulation of which

might have been regarded as contradictory in a purely theoretical approach

to directive l in g u is t ic behavior. I t turns out that in some cases of di­

recting the hearer has authority over the speaker. How can a speaker,

who holds a subordinate pos it ion , direct the hearer, his 'superior' ?

In other words, how can there be any acts of directing sat isfying (C3) ?

(C3) Ha has authority over (Sa)

256

The mystery disappears when looking at some obvious (C3) verbs such as

TO PETITION {Dutch EEN VERZOEKSCHRIFT INDIENEN) and even TO BEG (Dutch

SMEKEN). Both verbs are used to describe acts of trying to get the

hearer to do (or allow) something, and in both cases the decis ion to

perform the act or not i s completely in the hearer's hands. There is

a clear r e v e r s a l o f a u th o r i ty and s t i l l the speaker's behavior remains

direct ive .

55. Is therea also a d is t in ct ion between power and knowledge authority

when the authority i s on the hearer's s ide ? In other words, i s there a

d is t in ct ion between (C3a) and (C3b) verbials ?

(C3a) Ha has power authority over Sa

(C3b) Ha has knowledge authority over Sa

The examples given so far , TO PETITION and TO BEG, are no doubt of the

(C3a) type. But there are, indeed, (C3b) verbials as wel l . Consider TO

ASK (in the sense of asking a question, Dutch VRAGEN), for instance.

When asking a question the speaker's attempt to get an answer (at leas t

a helpful or correct answer) f a i l s unless the hearer has what we call

knowledge authority over the speaker: with respect to the subject matter

of the question he i s supposed to know more than the speaker. The same

holds for TO CONSULT (in the sense of asking for advice, Dutch RAADPLEGEN)

and many others.

There seems to be only one point where there i s no symmetry between

speaker and hearer authority, namely (C3c).

257

(C3c Ha has both power and knowledge authority over Sa

I have not been able to find any (C3c) verbia ls , neither in English nor

in Dutch.

In th is section we have refrained from giving extensive l i s t s of examples.

But no doubt the reader wil l be able to s e le c t lo ts of additional examples

for the s ix points on the authority dimension from the long l i s t s

provided in previous sect ions. The authority dimension i s v isualized in

SDC-Table 6.

-DUTCH- -SEMANTIC DIMENSION- -ENGLISH-

bevelenverbieden

aanradenwaarschuwen

instrueren

voorstellenaandringenverzoeken

een verzoekschrift indie- nen, smeken

vragenraadplegen

0

[[[LtL[

(Cla)

(Clb)

(Clc)

(C2)

(C3a)

(C3b)

(C3c)

to order to prohibit

to advise to warn

to instruct

to suggest to i n s i s t to request

to pet i t ion to beg

to ask (a question) to consult

]SDC-Table 6

258

H. MISCELLANEOUS DIMENSIONS

56. The f i v e semantic dimensions discussed so far are cer ta in ly not the

only ones we need to understand the patterning o f d irec t ive verbia ls in

the English and Dutch lex icons. But no doubt they are among the most

important ones. In th i s sect ion I want to draw the attention to a couple

of additional dimensions which deserve further research.

A quite relevant dimension for further invest igat ion could be the spe­

c i f i c a t i o n o f -he hearer of the d ir ec t iv e utterances {or, i f you wish,

the d i r e c t e e ) . The hearer can not only be an indiv idual , but a lso a mob

(as implied by TO READ THE RIOT ACT in one of i t s senses) or the public

in general {as in TO DECREE), or even s p i r i t s (as in TO CONJURE UP).

Of course such hearer-spec i f ica t ions are often connected with, for in ­

stance, the social se t t in g in which the d ir ec t ive act i s s i tuated . Along

the hearer dimension a part icu lar ly in terest ing case i s TO SEND FOR as

used in (Dl) as a description o f (Al).

(Al) Go and get the doctor, quickly I

(Dl) Sa sent for the doctor

In th is case there i s , in addition to the immediate hearer Ha who i s

told to go and get the doctor, a second or d e la y e d h e a r e r » namely the

doctor who i s requested to come. Unlike in other cases where there i s

more than one hearer, the d i f feren t status o f the two directees o f (Al)

turns the utterance into a d o u b le d i r e c t i v e .

259

57. A second neglacted but potent ia l ly relevant l ine of research i s the

study of the value judgment dimension. For one thing, many verbials of

directing pass value judgments on the act towards which Ha i s directed.

For instance, both AANSTOKEN and AANVUREN mean TO INCITE, but the former

implies that the act towards which Ha i s directed i s questionable, where­

as the second implies that i t i s praiseworthy. In this respect Dutch seems

to show sharper contrasts than English; the questionable character of the

act i s often implied by TO INSTIGATE but not always; and I have not been

able to find an English d irect ive verbial implying the unquestionably

praiseworthy character of the act.

Value judgments are also passed on the l in g u i s t i c act of directing

i t s e l f . Verbs such as AFTROGGELEN ( i . e . TO WHEEDLE SOMETHING OUT OF A

PERSON), AFDWINGEN ( i . e . TO EXTORT), AFDREIGEN ( l i t . 'to extort by

threatening1, i . e . TO BLACKMAIL) and TO SCREW (in the sense of practis ing

extortion upon)all imply disapproval, with judgments ranging from

'blameworthy' to 'criminal' .

I . CONCLUSION

58. Nothing in th is chapter can be regarded as conclusive; i t merely

scratches the surface of the semantics of d irect ing . Let us recapitulate ,

in the same random and cryptic way as in the previous chapters, SOme

of the main points made

{i ) This chapter, as a whole, reveals how fundamental directing i s as a

function of language.

260

( i i ) The dangers of purely theoretical c la s s i f i c a t io n s were demonstrated

even more c learly than before; direct ive meaning i s shown to s p i l l

over into the domains of the representatives and commissives. (Para­

graphs 2 to 7)

( i i i ) Commanding and ordering turn out not to be the strongest types of

direc t ives . (Paragraph 9)

( i v ) Asking a question i s shown to be more strongly direct ive than in­

s i s t in g and requesting. (Paragraph 11)

(v) The verbials of directing s ingle out a number of areas of social in­

teraction in which direct ive behavior i s prominent. (Paragraphs 14 f f . )

(v i ) A discrepancy i s noted between the conceptualization of the mili tary

world as a prototypical directing s i tuat ion and a lack of verbials

focusing on the military se t t in g . (Paragraph 27)

( v i i ) The verbials indicate that ' responding1, 'moving to or from a p la c e 1

and 'granting or giving something' are the basic types of acts towards

which people direct each other l in g u i s t i c a l l y . (Paragraphs 30 f f . )

( v i i i ) Though the act towards which a hearer i s directed can only be po­

s i t i v e or negative ( i . e . doing something or not doing something)this

two-sided d irec t ion a l i ty i s ref lec ted in a s ix fo ld d is t in ct ion which

i s required to understand the semantic patterning of the verbials of

direct ing. (Paragraphs 37 f f . )

( ix ) The lexicon shows that warnings are not ju s t negative pieces of ad­

v ice . (Paragraph 48)

(x) The verbials of directing shed l ig h t on the nature of the authority

which i s often involved in d irect ive behavior; for one thing, they

show that a d is t inct ion i s necessary between d i f ferent types of

261

authority. (Paragraphs 51 f f . )

(x i ) Contrary to what one would expect, i t turns out that even cases in

which the hearer has authority over the speaker with respect to the

action to be performed, can be genuine cases of d irect ing . (Paragraph

54)

Each of these observations can be taken as the point of departure for

extensive further inves t iga t io n s .

262

VI. THE SEMANTICS OF FORGOTTEN ROUTINES

A. INTRODUCTION

1. So far we have already used the comparative lex ica l approach to further

our understanding of the l i n g u i s t i c a l l y marginal area o f conversational

s i l e n c e . We also tr ied to penetrate the experience of truth , a central

feature of propositional meaning, fay way of studying the verbia ls used to

describe representative speech acts in which the propositional content

d is to r t s the truth in one way or another. In the previous chapter we

scratched the surface of the vast area of i l locu t ionary meaning by can­

vassing the verbia ls describing speech acts o f the d ir ec t iv e type.

This f ina l sample study enters an area which i s marginal in some res­

p ec t , but very central in another. The topic of inves t iga t ion i s the area

of conversational routines. I am not only referring to formulaic ex­

pressions (as described by Tannen and 6ztek 1977) and po l i ten ess formulas

(Ferguson 1976), but a lso to routine utterances of a l e s s f ixed nature

and with d i f f e r e n t social functions . Thus I shall not only (not even mainly)

be talking about expressions such as "God b less you"(uttered when someone

sneezes) which i s formulaic in nature and which const i tu tes a form

263

of l in g u i s t i c po l i ten ess , but I shall also pay attention to a number of

quite f l e x i b l e , non-formulaic responses to acts of in s i s t in g , giving per­

mission, requesting, advising, e tc . which are not necessarily induced

by pol i teness .

2. The centra l i ty of conversational routines in l in g u is t i c action i s beyond

doubt. Without them, conversation would cease to e x i s t . Further, the

occurrence of formulaic expressions and politeness formulas appears to

be a universal phenomenon, though the ir actual shapes vary cross-1in-

g u i s t i c a l ly and cross -cu ltural ly . Moreover, the ir importance emerges

from the fact that omitting them (e .g . by neglecting to greet a person

one knows) or fa i l in g to acknowledge them (e .g . by not responding to

a greeting) inevitably creates tensions in interpersonal relationships.

These observations lead naturally towards the following question: In what

sense could we say that conversational routines (inclusing formulaic

expressions and pol i teness formulas) are marginal ?

3. The rpoduction of conversational routines involves a high degree of

a u to m a tio i ty . Whereas lying requires a conscious messing around with

the propositional content of a statement-like utterance, and whereas

commanding involves a conscious act of the w i l l , replying "You're welcome"

to "Thank you" i s largely automatic. Therefore, i f we agree that a

human action typ ica l ly resu lts from a conscious impulse o f the w i l l ,

we have to conclude that routine utterances are less central instances

of l in g u i s t i c action than, say, l i e s or commands or even many acts of

being s i l e n t .

264

This observation i s not meant to detract from the importance, empha­

sized in the previous paragraph, of routine utterances as aspects of

conversational interaction. But i t may help us understand why, as Fer­

guson (1976:137) puts i t , "this universal phenomenon has been very l i t t l e

studied by l in g u is t s or anthorpolagists or other students of human beha­

vior". It may also explain why, for a number of these routines, many

languages lack descriptive verbia ls . I f , for instance, John says "God

bless you" when Jane sneezes, there i s no easy way for speakers of English

to describe what John d id or, in other words, what type of l in g u i s t i c act

he performed. A description such as "John wished that God would bless

Jane"would be hopelessly inadequate; a native speaker of English would

not take i t seriously . The problem remains the same for the a lternative

and equivalent formula "Gesundheit" which could not be accounted for by

claiming that "John wished Jane health". One would have to take refuge

in the phrase "John said 'God bless you1" or "John said 'Gesundheit'11

which does not y ie ld any ins ight into the type of act performed because

the l in g u i s t i c action verb TO SAY is so general that i t can be prefixed

to any utterance (in d irect ly or indirect ly reported speech). The auto-

maticity with which routine acts are performed may explain why there are

many gaps in the lexical frame associated with them: the ir being per­

formed more or l e s s unthinkingly diminishes their cognitive sa l ience ,

which i s ref lected in the absence of a l e x ica l iza t io n .

4. I t i s the s e t of non-lexical ized conversational routines that I call

fo rgo tten rou tin es . This chapter i s an overview of some major lex ical

gaps in th is area in English and in Dutch.

265

The foregoing ten ta t iv e explanation (in terms o f the automaticity in ­

volved) of why some types o f routine utterances are e a s i l y 1forg o t ten ' ,

however a t tra c t iv e i t may be, s t i l l leaves us with mysteries . Indeed,

i t might even make some fa c ts harder to comprehend. For instance, how

i s i t that certain types o f routines ( e .g . greetings) are so s a l i e n t that

they get the ir own descriptive verb ia ls , whereas others can hardly be

described at a l l in a given natural language ? Sometimes i t w i l l be

poss ib le to present a hypothetical answer to th i s question; sometimes

i t w i l l not. But in a l l cases the question w i l l be relevant.

There i s something peculiar about a semantic discussion of lex ica l

gaps. What I hope th i s chapter w i l l lead to i s a partial understanding

of the type of conversational routines which are e a s i ly 'forgotten'

in the sense mentioned and which, therefore , require specia l study.

We w i l l be skimming through a semi-conscious area of l in g u i s t i c action

in which some of i t s most fundamental peoperties may be hidden.

B. THE EXPRESSION OF EMOTIONS AND ATTITUDES

5. A large proportion of l i n g u i s t i c routine acts can be situated in the

domain o f what Searle (1976) ca l led e x p r e s s i v e s , i . e . the c lass of speech

acts the i l locut ionary point o f which i s that a certain psychological

s t a t e , emotion or att i tude i s expressed. Of course t h i s area shades o f f

into the other tradit ional c la s se s because the speech acts o f which the

expression of a psychological s ta te i s not a prominent aspect ( e .g .

Sear le 's declarations) are rare.

266

In English and in Dutch, as in many other languages, there are sp e c ia l ­

ized l in g u i s t i c action verbia ls to describe the expression of some emo­

t io n s . For others , there are no such lex ica l items. This f i r s t contrast

between l e x ic a l i z e d and non-lex ica l ized l in g u i s t i c ac t ions , which i s

amply i l lu s t r a t e d in the following paragraphs, w i l l lead us to some

in terest ing conclusions as to why expressives were made into a separate

c lass of speech acts .

6. Let us f i r s t consider some of the ways in which sorrow {or regret)

and pleasure can be expressed.

(Al) I am sorry that you could not come over for a v i s i t

(A2) I regret that you could not come over for a v i s i t

(A3) I am sorry for being so rude to you

(A4) I regret that I have been so rude to you

(A5) I am sorry that your father died

(A6) I am glad that I have gotten the fel lowship

(A7) I am glad you've gotten the fel lowship

(A8) (I am) glad/pleased to meet you

Descriptions (Dl) to (D8) correspond to acts (Al) to (A8), in the same

order.

(Dl) Sa sa id /stated/c la im ed that he was sorry t h a t . . .

or: Sa expressed his regret t h a t . . .

(D2) Sa sa id /stated/c la imed that he regretted t h a t . . .

(D3) Sa apologized for being rude

267

(D4) Sa apologized for being rude

{D5) Sa commiserated/condoled Ha's fa ther 's death

{D6) Sa s a id /s ta te d /c l aimed that he was glad that h e . . .

(D7) Sa congratulated Ha on g e t t i n g . . .

(D8) Sa greeted Ha

It wil l be c lear that the phrase "Sa sa id /c l aimed/stated . . . " could have

been employed in a l l of the above descrip t ive a c ts . Notice, however, that

th i s phrase, which can introduce any type o f reported statement- l ike

utterance, would not normally be used by a speaker of English who was

asked to t e l l what Sa did when uttering (A3), (A4), (A5), {A7) and (A8).

Instead, the spec ia l ized expressive l i n g u i s t i c action verbia ls TO APOLO­

GIZE, TO COMMISERATE, TO CONDOLE, TO CONGRATULATE and TO GREET would be

used. Notice, moreover, that English lacks such spec ia l ized verbia ls for

the description of (A l) , (A2) and (A6). In other words, there i s no

verbial available for describing (Al) , (A2) and (A6) a s e x p r e s s io n s o f

e m o tio n s , i . e . verbials which focus on the emotion-expressive aspect

of these ac ts . Thus, some expressions o f psychological s ta te s are

l ex ica l ize d whereas others are not. The discrepancy i s s tr iking because

sometimes ident ical formulas are used for the l e x ic a l i z e d and the non-

l e x ic a l i z e d expression o f odentical emotions. A case in point i s the

formula "I am sorry" which always expresses a fee l in g o f regret but

which introduces what would be described as a simple statement in (Al)

whereas i t tr iggers the descript ive verbials TO APOLOGIZE and TO COMMI­

SERATE in (D3) and (D5). Similar things happen with "I regret" and

"I am glad".

268

7. A completely analogous picture emerges in Dutch. The formula "I am

sorry" can be replaced by the impersonal phrase "Het s p i j t me" ( l i t . ' i t

sorrows me') in (A l) , (A3) and (A5); "I regret" can be translated as

"ik betreur" in (A2) and (A4); and "I am glad" i s equivalent to "Ik ben

b l i j " . In (D l) , (D2) and (D6) the phrase "Sa zei /beweerde. . ." ('Sa sa id /

c l a i m e d . . . ' ) , which can introduce any type o f reported statement- l ike

utterance, would be used, whereas the verbia ls ZICH VERONTSCHULDIGEN

( ' t o apo log ize1) , ZIJN DEELNEMING BETUIGEN ( ' t o commiserate1) , GELUK-

WENSEN ( ' t o congratulate') and GROETEN ( ' t o g r e e t ' ) would be used in

the remaining describing acts .

This complete paralle l ism shows that probably we are not dealing

with one of the whimsical features o f natural language. But before

attempting to formulate an explanation we shall take a look at some

more comparable data.

8. Consider the following acts in which l ik ing and d i s l ik in g i s expressed.

(A9) I (don't) l ik e working la te

(A10) I l ik e your plan very much

(All) I don't l ik e the way you dress

Act (A9) can be described as in (D9).

(D9) Sa sa id /s tated/c la im ed that he l iked (did not l ik e ) working la te

Though the same statement-formula "Sa sa id /s ta te d /c la im ed . . ." could be

used to account for (A10) and (A ll) in a l l circumstances, there i s another

269

p o s s ib i l i ty as well . Imagine that Sa i s Ha's employer. Thus condition

(Cl) applies to (A10) and (A l l ) .

(Cl) Sa has authority over Ha (with respect to the topic of Sa's

utterance)

In such circumstances we are l ik e ly to get (DIO) and (Dll) as descriptions.

(DIO) Sa approved of Ha's plan

(Dll) Sa disapproved o f the way Ha dressed

Again we are confronted with l in g u i s t i c actions in which identical psy­

chological s ta te s are expressed by means of identical formulas, though

for only some of them English provides us with descriptive verbials .

Once more the parallelism with Dutch i s s tr ik ing . The verbs GOEDKEUREN

( l i t . 'to judge good', i . e . ' to approve') and AFKEUREN ( ' t o disapprove1)

would be used in (DIO) and (D l l ) . Moreover, the d is t inct ion between (A9)

on the one hand and (A10) and (All) on the other, i s reinforced by the

fac t that the verb HOUDEN VAN { ' to l i k e ' ) can be replaced by GOEDVINDEN

( l i t . 'to find good') , which i s ambiguous between l ikeing and approving,

in (A10) and ( A l l ) , but not in (A9).

A minor difference between English and Dutch i s the following: whereas

TO APPROVE and TO DISAPPROVE are not necessari ly l in g u i s t i c action verbials

( i . e . they can be used to express the att i tudes rather than the ir ex­

pression) , GOEDKEUREN and AFKEUREN can o n ly be used to describe the ex­

p r e s s io n of approval and disapproval.

9. The divergence of human emotions and att i tudes i s l imited . So i s

the number of the ir l in g u i s t i c forms of expression. In th i s paragraph

270

I shall offer a random se lec t ion of the remaining ones. Unlike those d i s ­

cussed so far , the following ones will not const itute -pairs of l e x ic a l -

ized and non-lexicalized forms of expression. Yet they wil l present

exactly the same problem in the sense that , for no obvious reason, some

of them are lex ica l ized whereas others are not. The expression of a wish,

and the expression of gratitude are two examples of lex ica l ized ex-

pressives. Consider (A12) and (A13).

(A12) I wish I could go

(A13) I'm truly grateful for your hosp i ta l i ty

These two can be described by means of the l in g u i s t i c action verbs

TO WISH and TO THANK as in (D12) and (D13).

{D12) Sa wished he could go

(D13) Sa thanked Ha for his hosp i ta l i ty

In contrast, there are no emotion or att itude-oriented verbials asso­

ciated with the expression of a hope as in (A14), the anticipation of

pleasure as in (A15), disappointment as in (AI6), surprise as in (A17)

and concern or worry as in (A18).

(A14) I hope he arrives on time

(A15) I am looking forward to receiving your reply

(A16) I am disappointed that he did not come after all

{A17) I am surprised that he came

(A18) I am concerned/worried th at /abou t . . .

271

Both in English and in Dutch these can only be described as statements.

(A12) and (A13) are also lex ica l ized in Dutch: the verbs WENSEN { ' to

wish') and (BE)DANKEN ( ' to thank') would be used in (D12) and (D13).

Notice that TO WISH (Dutch WENSEN) is a borderline case in the sense

that i t s use as a l in gu is t ic action verbial i s extremely limited.

An act such as (A12) i s l ik e ly to be described by means of "Sa expressed

the wish that he could go" rather than by means of (D12). But in the case

of (A14) through (A18) similar paraphrases are the only way out.

The data presented so far are summarized in SDC-Table 1 (with TO WISH

and WENSEN between brackets because of their marginal character). We

should now start looking for explanations.

10. What needs explaining i s the fact that some l in g u is t ic expressions

of emotions or attitudes can only be described as statements whereas

others are lex ical ized in such a way that they seem to take on a to ta l ly

different speech act status. In other words, the problem is that some of

these expressions, as ex p re ss io n s o f em otions o r a t t i t u d e s , are simply

forgotten in the conceptualization underlying the sets of l in g u is t ic

action verbials in English and in Dutch. The best place to look for an

an explanation i s in the pairs of identical formulas expressing identical

emotions, which are sometimes lex ica l ized and sometimes not, i . e . in

(Al) through (All) .

Let us take, for instance, the formula "I am sorry". What d i s t in ­

guishes i t s use in (Al) from (A3) and (A5) ?

(Al) I am sorry that you could not come over for a v i s i t

■DUTCH­ 'S EMANTIC DIMENSION-------- ENGLISH-

0(Al)

(A2)0

zich verontschuldigen(A3)

(A4)to apologize

zijn deelneming betuigen - <A5) ] to commiserate, to condole

0 I (A6) ; 15

gelukwensen ] (A7) ‘ to congratulate

groeten I (A8) ^ to greet

0 . (A9) 0goedkeuren JjfA10) - ( C l t o approve

afkeuren JA11)-{C1)1 to disapprove

(wensen) - (Ai2) : (to wish)

(be)danken _ (A13) "

(A14) '

(A15)

to thank

0 (A16)

(A17)

(A18)

0

SDC-Table 1

273

(A3) I am sorry for being so rude to you

(A5) I am sorry that your father died

I be l ieve that the major difference i s the fol lowing. Whereas iri.(Al)

Sa i s simply conveying information about himself which may or mey not

in te r e s t Ha, in both (A3) and (A5) the speaker's psychological s ta te

expressed i s of crucial importance to the hearer. (A3) presupposes that

Sa has been rude to Ha; therefore , a l l further interaction between the

two interlocutors dpends cru c ia l ly on Ha's coming to know whether Sa

f e e l s sorry about his previous behavior or whether he can expect s imilar

behavior in the future. S im ilar ly , (A5) presupposes that Ha's father

died; whether Sa f e e l s sorry ( for Ha) about th i s unfortunate event

partly defines Sa's a tt i tudes towards Ha, and therefore i t i s important

for Ha to know Sa's f e e l in g s . My hypothesis i s that the importance of

the emotions expressed in (A3) and (A5) for the re lat ionship between Sa

and Ha and for th e ir soc ia l interact ion increases the cognit ive sa l ience

of these expressions within the domain of l in g u i s t i c a c t iv i t y to such

an extent that th e ir conceptualization and l e x i c a l i z a t io n as e x p r e s s io n s

o f em o tio n s become inev itab le . An act such as (Al) lacks th is cognitive

sa l ience and i s therefore ' forgotten' in the l e x i c a l i z a t io n process , though

both the emotion expressed and the formula used are ident ica l with (A3)

and (A5).

In Searle 's typology of speech acts (Al) would simply be described as a

statement (a representative) whereas (A3) and (A5) would be said to be

expressive. Thus, i f my hypothesis i s correc t , the d is t inguishing t r a i t

of expressives i s not the expression of a psychological s ta te as such

274

{because also the point of (Al) i s to express a psychological s tate ) but

the expression of a psychological s tate important to the hearer. Our

lex ical data support the val id ity of the dist inct ion between represent-

atives and express!ves while enabling us to formulate the dist inct ion

more accurately.

11. We s t i l l have to check the va l id i ty of our explanation for the rest

of our data. The contrast between (A2) and (A4) can be put in the same

terms as that between (Al) and (A3). Also the congratulation (A7) and

the greeting (A8) d i f fer from the statement (A6) in that the attitudes

expressed are important to the hearer. Exactly the same holds for the

difference between the approval (A10)-(C1), the disapproval (A l l ) - (C l) ,

and the simple statement (A9).

But what about the non-paired examples (A12) through (A18) ? The

importance to Ha of the gratitude expressed in an act of thanking as in

(A13) i s evident. On the other hand, the hope in (A14), the disappoint­

ment in (A16), the surprise in (A17) and the concern in (A18) do not

show the same kind of relevance to the hearer. This does not mean that

sentences starting "I h o p e . . .1' e tc . are always ju s t stements. Consider

(A19).

(A19) I ’m giving a party tomorrow. I hope you can come

In this case "I hope you can come" i s c learly an act of inviting Ha to

come to the party; an invitat ion i s a direct ive speech act. The only

points I want to make are that sentences starting with "I hope . . . " e t c . ,

though they clearly constitute the expression of a psychological s t a t e ,

cannot be described by means of a l in g u is t ic action verbial focusing on

275

the psychological s ta te expressed, because no such verbials are ava i la b le ,

and that the lex ica l gaps in question resu l t from the fa c t that the

expression of the f e e l in g s expressed lacks an inherent importance for the

relat ionship between speaker and hearer.

So far the hypothesis seems to work. There are , however, two trouble­

some cases , namely (A12) and (A15).

12. The f i r s t problematic case i s (A15), which i s a widely used phrase

to end a l e t t e r .

(A15) I am looking forward to receiving your reply

One might want to argue that the ant ic ipat ion of pleasure expressed in

(A15) i s important to the reader of the l e t t e r . Of course i t i s . But

upon c lo ser inves t iga t ion the problem turns out to be imaginary. The

antic ipat ion of pleasure i s only important to Ha in the same way as the

hope expressed in "I hope you can come )as in (A19J). That i s , the

verbalized psychological s ta te does not matter as such. What rea l ly

counts for Ha i s i t s implication. The expression o f Sa's hope that Ha

would be able to come implies that Sa wants Ha to come, and therefore

the sentence "I hope you can come” counts as an in v i ta t io n . S im ilar ly ,

Sa's looking forward to receiving Ha's reply implies that Sa wants Ha

to reply , and therefore (A15) counts as a p o l i t e request to reply. The

d ir ec t iv e overtone of {A15) and (A19) dominate so strongly that i t

would be surprising to get a description emphasizing the expressive

aspect o f these acts .

276

13. The only real problem i s (A12), which can be described by means of

TO WISH used as a l in g u is t i c action verbial (though, as said before, such

usage i s extremely marginal).

(A12) I wish I could go

What makes wishing so d if ferent from hoping that (D12), as a description

of (A12), can be interpreted as an account of the l i n g u i s t i c a c tio n

performed, whereas (D14), as a description of (A20), can only be regarded

as an account of the p s y c h o lo g ic a l s t a t e expressed in the act ?

(A20) I hope I can go

(D12) Sa wished he could go

(D14) Sa hoped he could go

In other words, what makes TO WISH into a l in g u i s t i c action verbial ( in one

of i t s marginal senses ) , whereas TO HOPE can only refer to a s ta te of

mind ? Obviously, there i s no difference in importance to the hearer

between the expression of a wish and the expression of a hope. I t i s even

hard to see any difference between a wish and a hope as s ta tes of mind.

I want to suggest that in th is case the hearer-oriented principle I for­

mulated i s overcome by a second principle which i s idiosyncratic to acts

of wishing. What made TO WISH into a l in g u is t i c action verb may have

been the b e l i e f that a s ta te of a f fa ir s can be created by putting one's

wishes into a l in g u is t i c form (e .g . a magical formula, a prayer, e t c . ) .

As a r e su l t , the meaning of the verb TO WISH got extended from a s tate

of mind to the l in g u i s t i c expression of that s ta te of mind.

277

14. The b e l ie f that verbalizing one's wishes may bring about the descired

sate of a f fa ir s , and the result ing strong association of wishes (as

mental acts) with the ir expression (as l in g u i s t i c acts) and with the

desired state of a ffa irs i t s e l f (brought about by the formulation of

the wish), might explain another pecul iar ity of the verb TO WISH. Wishing

normally concerns a s ta te of a f fa irs which does not yet obtain, as in

"wishing someone a happy New Year", which i s done at the beginning of a

year. However,"wishing someone a happy birthday", though not ent ire ly

synonymous with "congratulating someone on his birthday", has as i t s

object a s ta te of a ffa irs which in most cases does already obtain.

This extension of the meaning of TO WISH could be explained as fol lows.

Take t l to be the time at which a wish i s uttered and t2 as the time

at which the s ta te of a f fa irs which i s the object of the wish obtains.

In the paradigm case of "wishing someone a happy New Year" t2 i s la ter

in time than t l . Now, imagine someone wishing that i t would stop raining.

I f one believes that uttering a wish can bring about the desired s ta te

of a f f a i r s , then t l and t2 could nearly coincide:immediately after some­

one's wish that i t would stop raining, i t might indeed stop raining.

This quasi - ident i f ica t ion of the act with i t s object makes i t possible

to understand how wishing got extended to circumstances in which i t s

object i s already present. This extension i s very nicely ref lected in the

fact that the Dutch verb for TO CONGRATULATE, in a l l i t s occurrences,

i s GELUKWENSEN ( l i t . 'to wish happiness') .

But whatever the explanation for i t s behavior, TO WISH remains marginal

as a l in g u is t i c action verb. Due to i t s marginal character i t does not

present a real threat to our hypothesis about the l ex ica l iza t io n of

278

expressions o f emotion.

C. NEGATIVE AND POSITIVE RESPONSES

15. The subject matter of the previous sect ion could be called the

e x p r e s s iv e d im en sion of forgotten routines. We discussed the types

of psychological s ta tes for the l in g u i s t i c expression of which languages

such as English and Dutch do not provide descr ipt ive v erb ia ls . Now we

are going into the t e x tu a l d im en sion of forgotten routines. We shal l

present an overview of the gaps in the lex ica l apparatus to describe

r e s p o n s e s to diverse types of speech ac ts . In th i s sect ion the over­

view w i l l be l imited to responses to speech acts o f a d irec t iv e nature.

All of the responses can be negative or p o s i t iv e . Thus, we are concerned

with two types o f a c ts , (Al) and (A2).

(Al) Sa responds p o s i t iv e ly to a speech act performed by Ha

(A2) Sa responds negatively to a speech act performed by Ha

The nature o f the preceding discourse can be sp ec i f ied in the form of

conditions on these acts . Consider (Cl).

(Cl) The act performed by Ha i s a suggestion such as "I suggest

we take the night train"

Examples of responses can be given as fo l lows: (A3) i s a response of the

(Al)-(Cl) type; (A4) i s a response o f the (A2)-(C1) type.

(A3) Let's do that

(A4) I don't fee l l ik e i t

279

These two responses could be described as in {D1) and (D2).

(Dl) Sa agreed to take the night train

(D2) Sa disagreed about taking the night train

Thus, the verb TO AGREE can refer to pos it ive reponses to suggestions,

whereas TO DISAGREE denotes, amongst other th ings, negative reponses

to suggestions. Conseuently, i f (Cl) obtains, both types of reponses

are lex ica l ized in English. Dutch provides the verbial AKKOORD GAAN

( ' t o agree') for (A3); however, i t s negation NIET AKKOORD GAAN ( ' to

disagree') would be rarely used for (A4); instead, the verb AFWIJZEN

(more or le s s equivqlent to 'to dec l ine ' ) would be more l ik e ly to

occur as in (D3).

(D3) Sa wees het voorstel af

( l i t . 'Sa declined the suggestion')

16. Let us go on to a second type of preceding discourse.

(C2) The act performed by Ha i s an act of in s i s t in g such as "I

i n s i s t that you come home with me"

A possible (A1)-(C2) act i s (A5). , whereas (A6) represents the (A2)-(C2)

type.

(A5) O.K. Fine with me

(A6) No, I won't

A quite natural way of describing (A5) would be to use the verb TO GIVE IN

(Dutch TOEGEVEN, which i s ambiguous between 'to give in' and 'to admit)

as in (D4). For (A6) one might use TO REFUSE (Dutch WEIGEREN) as in (D5).

280

(D4) Sa gave in (and went home with Ha)

(D5) Sa refused (to go home with Sa)

As with suggesting, there are no lex ica l gaps here.

17. In the case of advis ing , an additional complication a r i s e s .

(C3) The act performed by Ha i s an act of advising such as "You'd

better take your medicine"

To a piece o f advice two types of p o s i t iv e and negative reponses can be

given: one can e ither concentrate on the representative aspect of

advice as in the responses (A7) and (A9), or on i t s d ir ec t ive aspect

as in (AS) and (A10).

(A7) I think you're right

(A8) O.K. I w i l l

(A9) I don't think so; i t ' s s t i l l too early

(A10) No, I won't

The verbs TO AGREE and TO DISAGREE can be used to account for (A7) and

(A9), as in (D6) and (D7).

(D6) Sa agreed (with Hs on taking his medicine)

(D7) Sa disagreed (with Ha on taking his medicine)

In Dutch the comparable verbia ls AKKOORD GAAN ( ' t o agree') and NIET

AKKOORD GAAN ( ' t o d isagree') could be used. The d irec t ive -or iented

negative reply (A10) can be described by means of TO REFUSE (Dutch

WEIGEREN) as in (D8).

281

(D8) Sa refused {to take the medicine)

A problem arises with the directive-oriented positive response: I did

not succeed in finding a l in g u is t ic action verbial (apart from the

passe-partout verb TO SAY) suited for the description of (A8). Here

we seem to be confronted with a f i r s t gap.

18. Next in l ine i s the speech act of invit ing.

{C4) The act performed by Ha i s an act of inviting such as "Would

you l ike to come to my party tomorrow night ?"

The posit ive (A1)-(C4) response could simply be (A l l ) , and (A12) could

be the negative (A2)-(C4) reply.

(All) Yes, I would

(A12) No, I can't

To describe these two acts the verbs TO ACCEPT {Dutch AANNEMEN) and

TO DECLINE {Dutch AFSLAAN) might be used as in (D9) and (DIO).

(D9) Sa accepted (Ha's invitat ion to come to the party)

(D19) Sa declined (Ha's invitat ion to come to the party)

Again, a l l lexical s lo t s are f i l l e d . (Notice that "No, I wouldn't"

cannot be regarded as an appropriate negative reponse to an invitation;

i t would count as an insu lt rather than as an act of declining the

in v i ta t io n . )

282

19. What about responses to requests ?

(C5) The act performed by Ha i s an act of requesting such as

"Can you possibly give me a l i f t ?"

Acts (A13) and (A14) are adequate replies of the (A1)-(C5) and the (A2)-

(C5) types, respectively .

(A13) Sure

(A14) No, I'm afraid not

The negative response (A14) can be described as a refusal (using the

verb TO REFUSE, Dutch WEIGEREN) as in (D l l ) .

(Dll) Sa refused (to give Ha a l i f t )

For the pos i t ive response, however, no adequate l in g u is t i c action verb­

ial presents i t s e l f . Thus, (A1)-(C5) i s our second lexical gap. Notice

that the gap cannot be f i l l e d by means of such phrases as "to comply

with the request"; th is phrase, ju s t as "to follow someone's advice"

or "to obey someone's order", indicates a pos i t ive behavioral reponse,

not a pos i t ive l in g u is t i c response.

20. Similar to requests are orders and commands. According to the tra ­

ditional accounts they mainly d i f f e r from requests in that they require

authority on the part of the speaker over the hearer. As far as the s e t

of descriptive l in g u is t i c action verbials i s concerned, the responses

to both types of acts show an identical pattern. Consider (C6).

283

(C6) The act performed by Ha i s an act of ordering such as

"Come here, Michael !"

Acts (A15) and (A16) are poss ib le responses of the (A1)-(C6) and the

{A2)—{C6) types , respec t ive ly .

(A15) O.K.

(A16) No, I won't

(A16) could be described by means o f the verb TO REFUSE, Dutch WEIGEREN.

Once more, an adequate l in g u i s t i c action verbial does not present i t s e l f

for the p o s i t iv e reply (A15).

21. At l e a s t two more types of d irec t iv es deserve our attention: per­

missions and prohibit ions.

(C7) The act performed by Ha i s an act of giving permission such as

"You can go home now"

(C8) The act performed by Ha i s an act of prohibit ing such as

"You mustn't leave th i s early"

(A17) i s an appropriate p o s i t iv e response to the permission; (A18) i s

a possible negative one.

(A17) Thank you

(A18) I'm not going now

The prohibit ion can be replied to p o s i t i v e ly as in (A19) and negatively

as in (A20).

284

(A19) Fine with me

(A20) I ' l l do i t anyway

(A17) i s simply an act of thanking for the description of which we have

the verb TO THANK, Dutch( BE)DANKEN, a va i lab le . As a r e s u l t , (D12) i s an

adequate account of (A17).

(D12) Sa thanked Ha ( for the permission to go home)

But for (A18) through (A20) no descrip t ive verbia ls seem to be avai lab le .

22. SDC-Table 2 summarizes the verbia ls avai lab le for the description

of p o s i t iv e and negative responses to a number of d irec t ive speech acts .

How can we explain the gaps ? How can we explain the fa c t that some

response types were 'forgotten' in the l e x i c a l i z a t io n process ? I bel ieve

that a vague princip le o f harmony o f i n t e r a c t i o n can show us why some

responses to part icu lar types o f d irec t ive s are cogn it ive ly l e s s s a l i e n t

and, therefore , do not need to be l e x i c a l i z e d .

Acts of i n v i t in g , suggesting and even in s i s t in g leave the hearer complete­

ly free to do as he chooses. These acts c le a r ly ind icate the speaker's

preference with respect to the hearer's ac t ions . But no matter what

course o f action the hearer takes there i s nothing disharmonious about

the in teract ion . Therefore, both p o s i t iv e and negative responses can be

expected with equal probabil i ty . As a r e s u l t , both options got l e x i c a l ­

ized .

The same reasoning applies to advising when i t s representative aspect

i s focused upon. But what happens i f i t s d irec t ive component i s envisaged ?

I t seems to me that advising may be stronger as a d ir ec t iv e than i n s i s t - .

285

-DUTCB- -S EM ANTIC DIMENSION- -ENGLISH-

akkoord gaan

afwijzen

toegeven

weigeren

akkoord gaan

0n ie t akkoord gaan

weigeren

aannemen

afslaan

0weigeren

0weigeren

(be)danken

1 (Al) -(C2

[ (A2) -(Cl

[ (Al) — ( C2

[ (A2) -(C2

' / ’(Al) - ( C3

[ W - yr ( a i o )

rA2

Al

A2

Al

A2

Al

A2

Al

A2

Al

A2

-(C3

"(C4

— (C4

— ( C5

— ( C5

— ( C6

-{C6

~{C7

“ (C7

-(C8

-(C8

1 1 1 ]/ ( A 7 ) J

(AS) ]

. (A 9 ) ]

(AIO)]

]]

to agree

to disagree

to give in

to refuse

to agree

0

to disagree

to refuse

accept

decline

1]]

to

to

0

to refuse

0

to

to

refuse

thank

0

S D C -T a b le 5

ing because i t implies some kind of authority (the kind that we ca l led

'knowledge authority1 in the previous chap ter ) . I t i s assumed that

the speaker giving the advice knows what the best course of action i s .

286

As a r e s u l t , the hearer i s expected to respond p o s i t iv e ly and not to

disregard the advice. Since harmony of interaction can only be obtained

i f the response i s p o s i t iv e , the pos i t ive reponse i t s e l f i s taken for

granted to such an extent that i t loses i t s cognit ive sa l i en ce . This

explains the absence of a verbial to describe (A8). On the other hand,

the negative reponse i s cognit ive ly s a l i e n t because i t disrupts the

harmony of in teract ion .

Completely analogous accounts can be given for the gaps in connection

with requests and orders. Complying with a request and obeying an order

are the harmonious reactions expected. In the case of orders the expect­

ation resu l t s from the 'power authority' involved. Requests are quite

compelling in sp i te of the absence of such authority.

What about permissions and prohibitions ? Why are the negative responses

to these types of acts not l e x i c a l i z e d , though they no doubt disrupt

interactional harmony ? A poss ib le explanation i s that in the case of

prohibit ions negative reponses are not only unexpected and disruptive

but a lso rare because of the high degree of authority involved: a speaker

who prohibits something i s often in a posit ion to a c t iv e ly prevent the

hearer from doing what he prohibits him to do (whereas in the case of

an order such ac t ive intervention i s usually harder s ince i t i s much more

d i f f i c u l t to make somebody do something than to prevent him from doing

i t ) . On the other hand, permissions imply that the hearer wants to do

what he i s allowed to do; therefore , negative responses to permissions

are rare as w e l l . The scarc i ty o f negative responses to acts of permitting

and prohibiting deprives them of the cognitive sa l ience which they gained

287

by being disruptive. Therefore, they are 'forgotten' in the l e x i c a l i z a t io n

process.

A f ina l question has to be answered. How i s i t that there i s a verbial

to describe the p o s i t ive reponse to acts of permitting, even though that

response i s cer ta in ly the expected and harmonious one ? Since permissions

imply that the hearer wants to do what he i s allowed to do, and since

the speaker has authority over the hearer with respect to the act in

quest ion , the p o s i t iv e response i s , not surpris ing ly , an expression of

gratitude. And since expressions of gratitude a lso occur in other contexts

and are generally ca l led acts of thanking, i t i s only log ica l that the

same name i s given to the expression of gratitude fol lowing a permission.

( I t i s quite possib le that in an imaginary language in which gratitude

would only be expressed a f ter obtaining a permission, there would be

no word equivalent to TO THANK.)

Z ) . FIXED ROUTINE RESPONSES

23. The responses studied in the previous sec t ion could be e i th er p o s i t iv e

or negative. There are a lso routine responses of a f ixed formulaic

nature which lack the pos i t ive -negat ive option. Some of these are easy

to describe in English and in Dutch. Consider the reponse act (Al)

uttered in the contexts sp ec i f ied by (Cl) through (C3).

(Al) Thank you

(Cl) The act performed by Ha i s an act o f congratulating such as

"Congratulations on your birthday"

288

(C2) The act performed by Ha i s an act of condoling such as

"I'm terr ib ly sorry that your father died"

(C3) The act performed by Ha i s an act of welcoming such as

"Welcome home"

In a l l of these contexts (Al) can be described by means of TO THANK,

Dutch (BE)DANKEN, as in (Dl).

(Dl) Sa thanked Ha

Sa expresses his gratitude for Ha's kind att itude towards him. Hence

(A l) , in response to acts of congratulating, condoling and welcoming,

i s not an obscure or empty routine. I ts conversational meaning coincides

with the lex ical meaning of the words used.

24. A b i t more complicated are the responses to greetings. In the context

spec if ied by (C4), several rep lies are poss ib le , some of which are l i s t e d

as (A2) to (A4).

(C4) The act performed by Ha i s an act of greeting such as

"How are you ?"

(A2) Very w e l l , thank you

(A3) Fine. How are YOU ?

(A4) Very w e l l , thank you. And you ?

Though in (A2) the same formula, "thank you", i s used as in (Al) , and

though (A2) could be said to express appreciation ( i f not gratitude)

for Ha's in terest in Sa or for his p o l i t e display of recognition (which

i s probably the essence of every greet ing) , (A2) can hardly be described

as an act of thanking. Nor do we have any other l in g u i s t i c action verbial

289

at our disposal to describe the reponse act . Acts (A3) and (A4) are

return greet ings , describable by means of TO GREET, Dutch (BE)GROETEN.

25. F ina l ly , there are a number of f ixed routine responses for which no

descriptive, verbia ls e x i s t at a l l . F ir s t , consider (A5) through (A9),

uttered when (C5) obtains.

(C5) The act performed by Ha i s an act o f apologizing such as

"I am sorry for being so rude to you"

(A5) No harm done

(A6) Never mind

(A7) That's quite a l l right

(A8) Please don't worry

(A9) Forget i t

For none of these acts do we have descriptive verb ia ls .

Second, consider (AIO) through (A14) uttered in a context in which

(C6) ontains.

(C6) The act performed by Ha i s an act o f thanking such as

"Thank you very much"

(AIO) You're welcome

(All) That's quite a l l right

(A12) Not a t a l l

(A13) Forget i t

(A14) Don't mention i t

Again, no descr ip t ive verbia ls seem to be avai lab le in English and in

290

Dutch. Just as "thank you" as a response to a certain type o f greeting ,

"you're welcome" as a response to an act of thanking i s a forgotten

routine, though TO THANK and TO WELCOME are l in g u i s t i c action v erb ia ls .

The reason i s that the conversational meaning of the.formulas does not

coincide with the lex ica l meaning of the words used. S im ilar ly , in

many languages there are f ixed routine responses to acts o f thanking

which include an equivalent of the verbs TO PRAY, TO BEG, TO ASK: German

BITTE, French JE VOUS EN PRIE, I ta l ian PREGO, Hungarian KEREM {BITTEN,

PRIER and PREGARE mean ' to pray, to berg, to ask; KERNI means ' to ask

( for something)') . Yet these acts would never be described as acts of

praying, begging or asking.

Notice that (A7) i s identical with (A ll ) and that (A9) i s ident ical

with (A13). This i s possib le by v irtue of the fa c t that both apologizing

and thanking express the speaker's f e e l in g that he owes something to the

hearer.

26. The material presented in t h i s sect ion i s v isua l ized in SDC-Table 3

which i s , so to speak, f i l l e d with gaps. The acts surveyed are a l l res­

ponses to expressive speech acts . The high proportion o f l ex ica l gaps

should not surprise us given the f ix ed , formulaic nature of these acts .

The au to m atic ity involved in t j e i r performance i s extremely high, and

therefore th e ir cognit ive sa l ience can be expected to be low. I t i s only

when the routine reponses are not made with the expected accuracy and

speed (as in the speech of a foreigner) that they come to mind.

The presence of a verb,TO THANK, to describe responses to acts of con­

gratu la t ing , condoling and welcoming i s due to th e ir formal and semantic

291

-DUTCH-

(be)danken

0

(be)groeten

0

0

-SEMANTIC D

(Al

(Al

(Al

(A2

(A3

(A4

(A5

(A6

(A7

(A8

(A9

(AIO)

(All )

(A12)

(A13)

(A14)

MENSION-

Cl)C2)

C3)

C4)

C4)

C4)

C5)

C5)

C5)

C5)

C5)

(C6)

(C6)

(06)

(06)

(06)

-ENGLISH-

to thank

0

to greet

0

SDC-Table 3

s im i la r i ty to independent ( i . e . non-response) expressions of grati tude .

Responses to greetings are no l e s s automatic than rep l ies to apologies

and thanks. But they can often be described by means of TO GREET because

of the frequent rec iprocity of acts o f greeting.

292

In the previous sect ion i t was suggested that the normal, expected,

harmonious responses to speech acts are cogn it ive ly l e s s s a l i e n t and are,

therefore , l ik e ly to escape l e x i c a l i z a t io n . This hypothesis may explain

the many gaps in the area covered by th is s ec t io n . The routine responses

under inv es t ig a t io n , lacking the pos i t ive -negat ive option, are so

compelling that they form part of a strong expectation pattern of

harmonious interact ion . One could almost say that acts of apologizing,

thanking, and the l ik e are not complete unless an appropriate routine

response fo l lows. The responses are part of the act and hence, as s e ­

parate speech ac ts , they tend to escape our attent ion .

E. CONCLUSIONS

27. This chapter i s cer ta in ly not a complete overview of the gaps in the

l ex ica l frame associated with l i n g u i s t i c act ion . I t s purpose was to

draw the attention to certain types of acts which, because of th e ir

routine character, are e a s i ly overlooked and which, therefore , require

special study.

The complete paralle l ism between Dutch and English, which w i l l be hard

to f ind in other areas o f l i n g u i s t i c act ion , shows that language, in

particular the lex icon , does not even make forgett ing into a t o t a l l y

whimsical a c t iv i t y . Such adherence to a kind o f c r o s s - l in g u i s t i c system-

a t i c i t y lends our hypothetical explanations some strength. This i s not

to say that there i s any u n iversa l i ty to the f indings o f the extremely

l imited invest igat ion presented. However, the apparent strength o f the

l e x i c a l i z a t io n pr inc ip les involved, re f lec ted in the complete paralle l ism

293

between English and Dutch, makes the search for universals in th i s domain

of the l in g u i s t i c action verbials part icu lar ly tempting and compelling.

28. All the hypothetical explanations presented for the phenomena under

invest igat ion boil down to one basic pr inciple: the gaps in the lex ica l

frame associated with l in g u i s t i c action are due to the low cognit ive

sa l ience of the corresponding ac ts . This p r in c ip le , i f i t i s correct ,

proves that there i s a relat ionship between the lex icon , or the l e x i c a l -

i za t ion process, and habits of conceptualization. Hence, th i s chapter

confirms our b e l i e f that studying and comparing the verbia ls available

in d i f feren t natural languages for the description of l in g u i s t i c action

may y ie ld ins igh ts into people's language-specif ic and c u l tu r e - sp ec i f ic

conceptualization and experience of l in g u i s t i c behavior and, therefore ,

into the nature of l in g u i s t i c behavior i t s e l f .

294

VII. CONCLUDING REMARKS

Not a single portion of this essay can be regarded as conclusive. The

comparative lexical approach advocated in the f i r s t chapter i s a l ine of

research which will require investigations on a large scale before i t s

f e a s i b i l i t y and fru itfu lness can be shown to be beyond doubt. Even the

methodology proposed in the second chapter i s merely tentative and will

have to be refined. The remainder of th is essay i s no more than a bunch of

pi lo t studies. But I believe that they already hint at the f e a s ib i l i t y

and fruitfu lness of a comparative lexical approach to l in g u is t ic action.

Its comparative aspect has remained extremely limited: there was only

a comparison between two c losely related languages. Yet we have reached

a number of more or less important ins ights . Silence was shown to be an

integral part of l in g u is t ic interaction; we learnt about the nature of

truth and the ways in which i t can be deviated from; the verbials of

directing have helped us to understand the direct ive function of language

better; and even the absence of lexical items to describe certain types

of routine utterances has turned out to be revealing in several ways

(for instance, by laying bare the reasons why the members of the tra­

ditional class of expressive speech acts can be regarded as different

295

from representative expressions of emotions or att itudes.

To make another comparison, the approach i s l ike a psychoanalysis of

language: put language on a sofa and l e t i t talk to get i t s unconscious

out into the open. A b e l ie f in the s e n s ib i l i ty of the same kind of

approach underlies the work on metaphor by Lakoff and Johnson (1979):

metaphorical lex ica l izat ions are taken as evidence for conceptualization

habits.

Most detailed l in g u is t ic invest igations display some tediousness. There­

fore I hope the monotonous nature of th is essay can be forgiven. My

sincerest wish is that, however c lose ly form and content may be related,

the s t y l i s t i c impediments have not made the reader stumble over the

meaning I was trying to convey.

296

REFERENCES

Alford, Danny K.H.

1978 "The demise of the Whorf hypothesis (A major revision in the

history of l in g u i s t i c s ) . Proceedings of the Fourth Annual Meeting

of the Berkeley Linguistics Society 485-99.

Allwood, Jens

1976 Linguist ic Communication as Action and Cooperation: A Study in

Pragmatics. (= Gothenburg Monographs in Linguistics 2) Department

of L inguist ics , University of Goteborg.

1978 "On the analysis of communicative action". Gothenburg Papers in

Theoretical Linguist ics 38.

Austin, John Langshaw

1962 How to Do Things With Words. Oxford: Oxford University Press.

(Second edit ion , 1975, Cambridge, Mass.: Harvard University Press .)

Berl in, Brent and Paul Kay

1969 Basic Color Terms: Their Universality and Evolution. Berkeley/

Los Angeles: University of California Press.

297

Bolinger, Dwight

1975 Aspects of Language. New York: Harcourt Brace Jovanovich.

Cohen, L. Jonathan

1964 "Do mocutionary forces ex is t ?" Philosophical Quarterly 14:55.

118-37.

1974 "Speech acts". Current Trends in Linguistics 12.173-208.

Dixon, R.M.W.

1971 "A method of semantic description". In D.D. Steinberg and L.A.

Jakobovits eds. : Semantics. An Interdisciplinary Reader in

Philosophy, Linguistics and Psychology. Cambridge: Cambridge

University Press, pp. 436-71.

Dumont, Louis

1970 Homo Hierarchicus. Chicago: University of Chicago Press.

Ferguson, Charles A.

1976 "The structure and use of politeness formulas". Language in

Society 5:2.137-51.

Fillmore, Charles J.

1975 "An alternative to checklist theories of meaning". Proceedings

of the First Annual Meeting of the Berkeley Linguistics Society,

123'31.

/1977 "Topics in lexical semantics". In R.W. Cole ed.: Current Issues

in Linguistic Theory. Bloomington: Indiana University Press, pp.

76-138.

298

1978 "On the organization o f semantic information in the lexicon".

Papers from the Parasession on the Lexicon, Chicago Linguist ic

Society , 148-73.

Fortescue, Michael

1979 "Why the 'language of thought1 i s mot a language: Some incon­

s i s t e n c i e s of the computational analogy of thought". Journal

of Pragmatics 3:1 .67-80.

Freud, Sigmund

1917 Introductory Lectures on Psychoanalysis (= 1978, The Pelican

Freud Library 1) . Harmondsworth: Penguin Books.

Gadamer, H.G.

19 75 Truth and Method. New York: Seabury.

Garfinkel, Harold

1972 "Studies of the routine grounds of everyday a c t i v i t i e s " . In D.

Sudnow ed. : Studies in Social Interaction. New York: The Free

Press, pp. 1-30.

Grice, H. Paul

1975 “Logic and conversation". In P. Cole and J.L. Morgan eds.:

Syntax and Semantics 3: Speech Acts. New York: Academic Press,

pp. 41-58.

1978 "Further notes on log ic and conversation". In P. Cole ed.:

Syntax and Semantics 9: Pragmatics. New York: Academic Press,

pp. 113-27.

299

Jakobson, Roman

1970 Main Trends in the Science of Language. New York: Harper and Row.

Kay, Paul and Chad K. McDaniel

1978 "The l in g u is t i c s ignif icance of the meanings of basic color

terms". Language 54:3.610-46.

Kernan, K.T., J. Sodergren and R. French

1977 "Speech and social prestige in the Belizian speech community".

In B. Blount and M. Sanches eds.: Sociocultural Dimensions of

Language Change. New York: Academic Press, pp. 35-50.

Kockelmans, Joseph J.

1967 "Some fundamental themes of Husserl's phenomenology". In J .J .

Kockelmans ed.: Phenomenology. The Philosophy of Edmund Husserl

and Its Interpretation. Garden City, N.Y.: Doubleday and Company,

pp. 24-36.

Lakoff, George and Mark Johnson

1979 Metaphors We Live By. Ms.

Lehmann, Dorothea

1976 Untersuchungen zur Bezeichnung der Sprechaktreferenz im Englischen.

{= Forum Linguisticum 8) Frankfurt am Main: P.Lang.

L e is i , Ernst

1973 Praxis der Englischen Semantik. Heidelberg: Carl Winter Univer-

sitHtsverlag.

300

Lessing, Gotthold Ephraim

1778 "Anti-Goeze". In G.E. Lessing, 1965, Werke in Sechs Banden 6:

Philosophie/Theologie. Zurich: Stauffacher-Verlag, pp. 398-458.

Malinowski, Bronislaw

1954 Magic, Science and Religion, and Other Essays. Garden City, N.Y.:

Doubleday and Company.

McDaniel, Timothy

1978 "Meaning and comparative concepts". Theory and Society 6.75-92.

Meyer-Hermann, Reinhard

1978 "Aspekte der Analyse metakommunikativer Interaktionen". In R.

Meyer-Hermann ed. : Sprechen - Handeln - Interaktion. Tu’bingen:

Max Niemeyer, pp. 103-42.

Reddy, Michael

1977 "The conduit metaphor - - A case of frame c o n f l ic t in our language

about language". In A. Ortony ed.: Metaphor and Thought.

Cambridge: Cambridge University Press, (in press)

Rosch, Eleanor

1977a "Human categorization" In N. Warren ed.: Studies in Cross-Cultural

Psychology. New York: Academic Press, pp. 1-49.

1977b "Classif icat ion of real-world objects: Origins and representations

in cognition". In P.N. Johnson-Laird and P.C. Wason eds.: Thinking.

Readings in Cognitive Science. Cambridge: Cambridge University

Press, pp. 212-22.

301

Schuetz, Alfred

1967 "Phenomenology and the social sciences". In J .J . Kockelmans ed.:

Phenomenology. The Philosophy of Edmund Husserl and Its Inter­

pretation. Garden City, N.Y.: Doubleday and Company, pp. 450-72.

Searle , John R.

1969 Speech Acts: An Essay in the Philosophy of Language. Cambridge:

Cambridge University Press.

1976 "A c la s s i f i c a t io n of i l locutionary acts". Language in Society

5:1.137-51.

Tannen, Deborah and Piyale COmert Oztek

1977 "Health to our mouths: Formulaic expressions in Turkish and

Greek". Proceedings of the Third Annual Meeting of the Berkeley

Linguistics Society , 516-34.

Van Valin, Robert D.

1976 "Meaning and the Sapir-Whorf hypothesis". Ms.

Vassi lyev, L.M.

1974 "The theory of semantic f i e l d s : A survey". Linguist ics 137.79-93.

Verschueren, Jef

1977 "The analysis of speech act verbs:Theoretical preliminaries".

Distributed by Indiana University Linguistics Club. To be re­

printed in Pragmatics and Beyond {Amsterdam: JohnBenjamins)

1978a Pragmatics: An Annotated Bibliography. Amsterdam: John Benjamins.

302

1978b "Reflections on presupposition fa i lu re : A contribution to an

integrated theory of pragmatics". Journal of Pragmatics 2:2.107-51.

Wierzbicka, Anna

1977 "Mental language and semantic primitives". Communication and

Cognition 10:3/4.155-79.

Winch, Peter

1958 The Idea of a Social Science and I ts Relation to Philosophy.

London: Routledge and Kegan Paul.

1964 "Understanding a primit ive soc iety" . American Philosophical

Quarterly" 1.307-24.

Witherspoon, Gary

1977 Language and Art in the Navajo Universe. Ann Arbor, Michigan:

University of Michigan Press.

Wittgenstein, Ludwig

1953 Philosophische Untersuchungen. Oxford: Basil Blackwell.