article (published version) - archive ouverte unige

Article

Reference

(Un)common denominators in research on emotion language: a

postscript

OGARKOVA, Anna, BORGEAUD, Philippe

Abstract

In this special issue we brought together a number of articles whose convergence may not be

immediately evident to our readers. Varying in almost all respects from their conceptual

underpinnings, through the methodologies used, and to the very objects of their inquiry the

articles may leave the reader wondering what was the ultimate motivation to bring them

together in one special issue of the journal. Indeed, at their face value, the foci of the papers,

namely, the conceptualization of the distinctions within the affective domain (Deonna &

Teroni) and, more specifically, within the emotion domain (Cochrane), the affective

(non-propositional and connotative) effects of language that go beyond its representational

capacity (Moeschler, Soriano & Valenzuela), and the pan-temporal and pan-cultural

translatability of lexical and narrative emotion (Nelis, Bouvier, Zufferey) may seem reminiscent

of the famous Jamesian ‘shapes of the rocks on a New Hampshire farm', all quite unique and

interesting, but not providing a unifying, central point of view (James, 1890: 448).

OGARKOVA, Anna, BORGEAUD, Philippe. (Un)common denominators in research on emotion

language: a postscript. Social Science Information, 2009, vol. 48, no. 3, p. 523-543

DOI : 10.1177/0539018409106204

Available at:

http://archive-ouverte.unige.ch/unige:99585

Disclaimer: layout of this document may differ from the published version.

1 / 1

http://archive-ouverte.unige.ch/unige:99585

Special issue: The language of emotion – conceptual and cultural issues

Numéro spécial: Le langage de l’émotion – questions conceptuelles et culturelles

Anna Ogarkova & Philippe Borgeaud

(Un)common denominators in research on emotion language: a postscript

Key words. Affect/emotion terms – Connotative structure – Dimensional models – Narrative emotion – Non-propositional effects

Mots-clés. Effets non-propositionnels – Emotion narrative – Modèles dimensionnels – Structure connotative – Terminologie des affects/émotions

In this special issue we brought together a number of articles whose conver-gence may not be immediately evident to our readers. Varying in almost all respects – from their conceptual underpinnings, through the methodologies used, and to the very objects of their inquiry – the articles may leave the reader wondering what was the ultimate motivation to bring them together in one special issue of the journal. Indeed, at their face value, the foci of the papers – namely, the conceptualization of the distinctions within the affec-tive domain (Deonna & Teroni) and, more specifically, within the emotion domain (Cochrane), the affective (non-propositional and connotative) effects of language that go beyond its representational capacity (Moeschler, Soriano & Valenzuela), and the pan-temporal and pan-cultural translatability of lexical and narrative emotion (Nelis, Bouvier, Zufferey) – may seem reminiscent of the famous Jamesian ‘shapes of the rocks on a New Hampshire farm’, all quite unique and interesting, but not providing a unify-ing, central point of view (James, 1890: 448).

© The Author(s), 2009. Reprints and permissions: http://www.sagepub.co.uk/journalsPermissions.navSocial Science Information, 0539-0184; Vol. 48(3): 523–543; 106204

DOI: 10.1177/0539018409106204 http://ssi.sagepub.com

524 Social Science Information Vol 48 – no 3

Given our interdisciplinary approach, it would have been preposterous to strive for such a view. Instead, our aim has been to illustrate the heterogene-ity of the affective sciences domain through a narrower focus on language and culture – two relevant venues from which the diversity of the field becomes apparent. As we hope to have shown in the introduction, the role of language in emotion, the ways in which the labeling process can be con-ceived of by major emotion theorists, as well as universalism vs cultural relativism in emotion are among several long-standing debates where the either–or stance has rarely been particularly productive. And while it is com-mon sense that ‘… what emotions are is very much a matter of how they are conceived and supposed to function’ (Beatty, 2005: 18), it is the loose affinities which the articles share that fit our purpose better, for they indicate more the variety of directions of present and future research, rather than privilege a particular avenue or a specific approach.

Conceptualizations of the affective domain based on language analysis

The first two articles, ‘Taking affective explanations to heart’ by Julien Deonna and Fabrice Teroni and ‘Eight dimensions for the emotions’ by Tom Cochrane, are philosophical attempts to elaborate on broader and narrower distinctions within the affective domain. In Deonna & Teroni’s article, the thrust is to differentiate between several categories of affect-related phe-nomena, such as emotions, moods, sentiments, character traits and tempera-ments. In Cochrane’s article, an alternative model for the dimensionality of emotion concepts is presented, where the dimensions are derived from very general conditions of human ‘being-in-the-world’.

What unites both papers is their heading towards a perspective on emo-tion language where the latter is taken up as both the departure point and the important source of evidence substantiating the authors’ claims. Whereas in Deonna & Teroni’s article the focus is mostly on third-person, present-tense psychological ascriptions of mental states (e.g. John is angry with Maria, John is sanguine, etc.), Cochrane’s unfolds at the lexical level of emotion representation in language, taking emotion labels as the departure point for the analysis.

Both articles refrain from overstating the role of language in carving up ‘the affective space’. However, they vary in the assertiveness of their claims. Deonna & Teroni repeatedly assert that the affective language in the tokens within the affective categories they consider is far from being systematic. On the contrary, the affective terms usually employed to explain human

Ogarkova & Borgeaud (Un)common denominators in research 525

behavior are ambiguous, interchangeable, fuzzy, or misleading. This asser-tion echoes similar observations made by many researchers working in different paradigms (cf. Ryle, 1961; Ortony, Clore & Foss, 1987; Sabini & Silver, 2005; Scherer, 2005). Despite this lack of systematicity, the ‘language of moods/temperaments/character traits’ used in commonsense psychological explanation is readily used as supporting evidence in argu-mentation. For instance the claim that emotions are rational, reason- responsive states as opposed to moods, which are said to be a-rational (and irrational), is tightly linked to the ways we explain behavior in our conven-tional language. Similarly, whereas the possibility to use ordinary language for excuse/mitigation rather than justification serves as a complementary benchmark to distinguish emotions from moods or temperaments, the lack of such a possibility distinguishes character traits from both moods and temperaments.

In a more language-sensitive fashion, the initial position in Cochrane’s article sides with the view that the conceptual level is not completely deter-mined by the semantics of language items, as would be the case in what is frequently misleadingly labeled as ‘the strong Whorfian’ position (cf. Hill & Mannheim, 1992).1 However, a concession is made that, should the object language not be English, another set of dimensions could have emerged in the analysis. As in any dimensional model, emotion labels are taken as default language representations of emotion concepts, from where either a conceptual or a semantic route is possible (cf. Pavlenko, 2005: 83). Cochrane’s analysis operates at a fairly abstract level, aspiring to overstep the boundaries of language, and his focus is explicitly on the ‘encyclopedic’ meaning of emotion terms – a view where the semantic representation verges on the conceptual one, and both are posited to be governed largely by the same principles.2

Both papers differ in the sensitivity of their proposals to the social dimen-sion of affective phenomena. Deonna & Teroni’s analysis of the use of affec-tive adjectives to denote moods/temperaments/sentiments/character traits in ‘naive’ psychological explanations of behavior operates at the level of indi-viduals (e.g. John, Mary, etc.), where the relation between the agents is not specified. Yet recent evidence (Fiedler, 2008: 41) seems to suggest that peo-ple show consistent differences in explaining behavior conditional on whose behavior they are explaining. In the ascription of behavior to members of a different social group, people regularly use more abstract terms to describe negative behaviors (e.g. adjectives such as ‘disrespectful’) and more con-crete terms (e.g. verbs such as ‘help’) to describe positive ones. This sug-gests more internal (stable, consistent, context-independent) attributions for negative behaviors and more external (less stable, situational) attributions


for positive behaviors. On the contrary, in the ascription of behavior to members of the same social group, more abstract terms are used to describe positive behaviors, and more concrete terms are used to describe negative behaviors, suggesting internal, global attributions for positive behavior and external, local attributions of negative behaviors.

Cochrane’s proposal offers more in terms of capturing an inherently social nature of emotions by proposing the connected/disconnected dimen-sion. Yet, praiseworthy in its departure from the conventional overlooking of the social in major dimensional models, this is just the tip of the iceberg of the social component of meaning in emotion lexis. For instance in ordi-nary language use, emotion labels signal superior/inferior and distanced/close distinctions, among others, all of which are relevant to the ways the self is construed in a culture.3

Both articles point to problematic areas in contemporary emotion theory and research. Deonna & Teroni’s acknowledgment of the largely intuitive character of their introspections as to what lexical items (even in English) would be central to represent moods, temperaments or character traits clearly points to the lack of the large-scale enquiries into the structure of referential structures of relevant lexical domains, from which (ideally, cross-culturally valid) extrapolations to the conceptual level could be made. Despite the richness of taxonomical studies seeking to tabulate – at least the very core of – ‘emotion lexicons’ on the basis of a variety of languages including (among others) English (e.g. Storm & Storm, 1987), Dutch (Hoekstra, 1986), Taiwanese Chinese (Boucher, 1979), Indonesian (Heider, 1991) and Filipino (Church et al., 1998), systematic studies venturing out of the semantic fields of emotion labels into the wider area of other affect- related states are rare. With several exceptions – as, for instance, the research on English affect terms that suggests a number of broader classes referenc-ing, in various combinations, the affective, cognitive and behavioral aspects of states (Ortony, Clore & Foss, 1987), several studies on ‘personality descriptors’ in Ifaluk (Lutz, 1982), A’ara and Oriya (White, 1978, 1980), or affective disposition terms in Samoa (Mageo, 1991) – most of the research – ethnographic, linguistic and psychological alike – has been concerned with the emotion terms proper. Therefore, future large-scale research is needed on affective lexicons that would incorporate labels for moods, tempera-ments, affective dispositions and character traits.

The lexical domain of the ‘emotion terms proper’, which is the focus of Cochrane’s article on dimensions, is in itself a fairly treacherous area, fraught with different problems not fully accounted for in mainstream emo-tion theories and research methodologies. One of these problems is the absence of the category of emotion in several societies (e.g. Russell, 1991;


Pavlenko, 2005, for overviews), as well as its ‘fuzziness’ (Fehr & Russell, 1984), exposed by a relatively recent cross-disciplinary liberalization in the views on categorization as such (e.g. Rosch, 1978; Hacking 2001; Aarts, 2006). This has important implications for the semantic and conceptual structure of emotions, also affected by the absence of a finite set of criteria to differentiate emotion terms from other lexical domains4 (even in English) and the controversy over the grammatical form of the items to be used in the analyses, where the bias is towards considering only emotion nouns (but see Galati et al., 2008). These difficulties have also added fuel to the natural-kind debate, pioneered by the psychologist Elizabeth Duffy (e.g. 1941), the philosopher Amelie Rorty (1978),5 and the linguist Anna Wierzbicka (e.g. 1972), a debate which is now expanding further across disciplines (e.g. see Griffiths, 2004a,b; Charland, 2002, 2005; Barrett, 2006; Zinck & Newen, 2008; for both ‘pro’ and ‘contra’ views).

Another worry is the commensurability of the emotion terms as concerns their reference (e.g. Lutz, 1982; Russell, 1991): in many societies emotions are culturally construed as statements about the relationship between a per-son and an event, which is the case for Samoans (Gerber, 1975), Pintupi Aborigines (Myers, 1979), Ifalukians (Lutz, 1982) and A’ara speakers of the Solomon Islands (White, 1978), rather than references to internal states.6 In Chinese contexts, as shown by Nicolas Zufferey’s article in this issue, emo-tions are conceptualized as attitudes and behaviors in relation to others rather than individual subjective states in any absolute sense.

All these problems pose certain challenges to any large-scale comparative analyses of emotion lexicons, including proposals for dimensional models. In addition, the dimensions, given their general character, are able to capture only most abstracted differences between the terms. Furthermore, since the major rationale of dimensional models is frequently that of economy and testability, a bias towards discrete, prototypical, semantically non-overlapping items to be scaled on dimensions is almost inevitable in most cases.

Beyond the referential in language: connotative structure and non-propositional effects

As becomes evident from the preceding discussion, much of the research on affective language – in philosophy, anthropology, psychology and linguis-tics alike – has been focused on the most immediate, lexical level of affect representation. Yet it is obvious that the complex interaction between affect and language is by no means exhausted by the denotative capacity of lan-guage. The motivation to get at other levels has determined our disciplinary


choices on psycholinguistics and pragmatics, illustrated by the next two articles presented in the issue, ‘Emotion and colour across languages: implicit associations in Spanish colour terms’, by Cristina Soriano and Javier Valenzuela, and ‘Pragmatics, propositional and non-propositional effects’, by Jacques Moeschler.

Both papers, although stemming from very different branches of linguis-tics, converge in their focus on those language and cognition effects that go beyond the referential and therefore are only tangentially concerned with affective concepts per se. Both depart from the purely semantic view of meaning by emphasizing its cognitive component, be it by investigating the implicit affective associations of non-affective lexis (Soriano & Valenzuela), or by highlighting the role of inferential activity based on contextual assumptions (Moeschler).

Soriano & Valenzuela’s article focuses on the connotative structure of basic color terms, whose pairing with the connotative structure of emotion labels is hypothesized by the authors as one out of at least four alternative (although not mutually exclusive) explanations of why color names are fre-quently associated with the emotions in a given language. The article sug-gests the use of the Implicit Association Test, a methodology that, in contrast to the explicit/conscious semantic differential technique (e.g. Osgood, Suci & Tannenbaum, 1957) and other scaling methodologies, aims to measure automatic affect, or implicit attitudes, underlying evaluation (Greenwald, McGhee & Schwartz, 1998: 1464). The article’s particular focus on Peninsular Spanish contributes to discussions on the anthropology of color and emotion, a research line of extraordinary richness and variety (see e.g. Maclaury, Paramei & Dedrick, 2007).

In contrast to Cochrane’s proposal on alternative dimensionality, Soriano & Valenzuela’s case study is based on the classical Osgoodian dimensions of evaluation, potency and activity, and their results seem to corroborate earlier findings on two (potency and activity) out of three dimensions in color semantics and color association. Yet the divergence in Soriano & Valenzuela’s data on the evaluation dimension and the authors’ interpreta-tion of this finding in terms of a higher contextual dependence, vagueness and (hypothesized) cross-cultural (and inter-personal) variability of valence in color perception align with similar observations in Cochrane’s paper, albeit raised on conceptual grounds and in a very different context.

Diverging from Soriano & Valenzuela’s empirical approach, Moeschler’s contribution is a theoretical account of how the field of post-Gricean cog-nitive pragmatics and, more specifically, Relevance Theory (Sperber & Wilson, 1995; Carston, 2002), can be applicable to the study of emotions, which are – most generally – conceptualized here as effects triggered by


utterance perception and interpretation. In a relevance-theoretic fashion, the article overviews main changes in the way cognitive pragmatics describes verbal communication. The author argues for the inferential view, which then allows him to delineate propositional against non-propositional (non-representational, non-truth-conditional and, most specifically in this case, affective) effects that accompany and/or are produced by the utterance inter-pretation. These effects, Moeschler argues, surface most prominently in speech blunders, creative language use – including novel metaphor, syntac-tic parallelism and other tropes – all of which are treated by mainstream research on language emotiveness as ‘poetic effects’. The interaction between propositional and non-propositional effects is then considered as twofold: the emergence of non-propositional effects can either block the inferential process, as in erroneous referencing or paradoxical witticisms, or reinforce propositional effects, as in positive expressive speech acts or puns in the news which lead to supposedly shared references.

In contrast to Soriano & Valenzuela’s culture-sensitive approach, which strives to account for implicit associations that are formed under the influ-ence of the participants’ cultural background, Moeschler’s article unfolds at a largely culture-free level, where the inferential capacity involved in utter-ance interpretation is posited as a basic cognitive ability that evolved in humans prior to the development of the language faculty, and is thus immune to language and cultural differences. Non-propositional effects are tacitly assumed to be triggered equally well by utterances in any language – a position that could be justified by the fact that all languages have presup-positions. But it is equally clear that the way those presuppositions behave may differ from language to language and that the cultural differences may influence the content of the conversational implicatures which arise (cf. von Fintel & Matthewson, 2008).

Translatability of emotions

Three concluding articles, by David Bouvier, Damien Nelis and Nicolas Zufferey, shift the focus to narrative structures where the depiction or expression of emotion plays a crucial role. Diverging in the types of narra-tives they explore (Greek amorous or Latin epic poetry, or both ancient and contemporary Chinese prose), their convergent focus is on the pan-temporal and pan-cultural translatability of emotions.

David Bouvier’s ‘Peut-on traduire une émotion érotique? L’exemple des traductions françaises de l’ode 31 de Sappho’ is a case-study discussion of Sappho’s ode 31, where, on the basis of an extensive corpus of the ode’s


translations into French, the metamorphoses of the original text in transla-tion and imitation are traced from the 16th to the 20th century. The article exposes the limitations of translation, including the commentators’ uncer-tainties as to the nature of the emotion depicted (love, friendship or jeal-ousy?), its locus (heart, soul?), the gender of the three protagonists in the poem and the polymorphism of the ‘emoting body’, for which each of the translations provides an alternative interpretation by adding and amplifying the original depiction with references to bodily expressions that were not explicitly mentioned in the original text. Importantly, these successive meta-morphoses of the Greek original in French translations, as Bouvier shows, reflect historical period-sensitive construals of the emotion. For instance the emphasis – through reference to the soul or the heart – on either intellectual-izing or sensualizing the emotion reflects the commentators’ and translators’ (epoch-sensitive) preferences for either a hetero- or a homosexual Sappho.

Bouvier’s analysis of how the ode progresses from the depiction of an amorous sensation to the sensation of death – from Eros to Thanatos – shows the diachronic variability in the translators’ perceptions on this (lit-eral? metaphoric?) dying of the protagonist. Whereas in the 18th- and the 19th-century imitations and translations the verb phainomai, as shown by Bouvier, was more typically rendered with allusion to literal dying or an impression of it (although varying, through the use of grammatical nuances, in either the muting or strengthening of this death sensation, whose grada-tion Bouvier alludes to as ‘crescendo’), from the beginning of the 20th century and on, the translators prefer to use this verb in its primary sense of ‘to appear’, ‘to seem’. Therefore, an image of the dying poetess is succeeded by an image of the purported self-dramatized emotional display, where the internal subjective feeling is externalized and, as Bouvier argues, leads to overcoming the emotion.

In contrast to Bouvier’s analysis of the narrative where the emotion depicted is never explicitly named and gets verbalized exclusively through the allusion to emotional (and bodily) symptoms (showing interesting paral-lels with those of fear conveyed in Homeric texts), Damien Nelis’s paper, ‘Translating the emotions: some uses of animus in Vergil’s Aeneid’, is focused on the analysis of how a single Latin word, animus – a key word in the depiction of emotional states in Vergil’s Aeneid – has been reacted to by the English translators of the poem.

Animus, as shown by the author, is a polysemous word, whose meanings range from ‘the mind as the seat of consciousness/desire/volition/the emo-tions’ and ‘anger’ to ‘the moral and mental constitution of a person, their disposition or character’. In Vergil’s use, echoing that of Greek anemos (‘wind’ or ‘tempest’), it denotes either ‘the mind as the seat of feelings and


emotions’ or, more simply, ‘the emotions or feelings themselves’. The word’s extensive usage throughout the poem – 163 occurrences in total, and five uses in the opening 156 lines – creates a dense network of thematic associa-tions, involving the emotion of anger based on resentment at perceived wrongs and a desire for revenge. This repeated usage thematizes anger within the poem’s narrative structure and also alludes to the Aeneid’s subtext – the then-recent history marked by post-civil-war traumas of the early 20s BC.

Similarly to Bouvier’s conclusion that, through different historical peri-ods, the transformations of Sappho’s poem in translation have gained cul-tural references fundamentally different from those implied in the original, Nelis’s analysis shows that the linguistic and cultural specificity of the vocabulary of the emotions renders the study of literary texts extremely complex. This becomes apparent in view of the evidence that the translation of Latin animus in the Aeneid has never been consistent either within any single translation or across a number of English translations of the poem.

Nicolas Zufferey’s philological and historical approach in ‘Quelques réflexions sur la représentation des émotions en Chine et en chinois’ is close to those demonstrated by Bouvier and Nelis in its carefully historicized perspective on emotions. Similarly to Nelis in his analysis of Latin animus which, to the educated Roman readers, was an allusion to, or a ‘bilingual play’ on Greek anemos ‘wind’, Zufferey’s article offers several observations on the metaphoric underpinnings of Chinese generic terms to denote emo-tions that allude to resonance and music. In addition, Zufferey’s work con-verges with both Bouvier’s and Nelis’s in asserting the specificity of the cultural codes of emotional expression, based in this case on the analysis of sincerity in Chinese communicative behavior, which is hardly penetrable to most Europeans.

Yet the most prominent feature of Zufferey’s contribution, which distin-guishes it from other papers in the issue, is the persistent emphasis on the social aspect of emotions in Chinese culture. This emphasis is explicated by the author through the semantic analysis of exemplary cases of timidity and shame words in Chinese. The Chinese emotion vocabulary, as shown by Zufferey, is largely dependent on the social contexts in which the emotion terms are used; that is, it is an emotion’s antecedent event that gives rise to a distinct label, not the emotion’s qualitatively distinct and internal ‘raw feel’. Such is the specification of timidity in qiechang (lit. ‘to be afraid of the scene’), pasheng (lit. ‘to be afraid of the unknown’), or rensheng (lit. ‘to recognize as unknown’), all of which signify important aspects of the situa-tions where these emotions are experienced.

The social aspect of the emotions is furthered by Zufferey’s analysis of the concept of ‘correlative thinking’, central for both ancient and modern


Chinese thought, and the notion of ‘face’, which emphasizes among other things the external, social dimension of shame and other emotions in Chinese society. This general tendency is shown by Zufferey to impact tra-ditional Chinese ancient and modern novels so that the preference is given to the external rather than the internal ‘focalization’ of emotions; that is, the emotions are interwoven into the narrative structure through the depiction of their overt and visible manifestations. Moreover the protagonists’ inner speech – a potential locus for genuine reflection on inner sensations and for subjectivity in general – gets deliberately theatricalized. At least in part, as Zufferey remarks, this tendency is explicable by the specificity of the con-strual of the self in Chinese society, where an individual is defined first of all through his/her relations within a group, such as a family or a social network. This particular construal of the self in Chinese culture leads to an alternative conceptualization of emotions as attitudes and behaviors in rela-tion to others, rather than as individual subjective states in any absolute sense. This then lets Zufferey voice certain concerns in relation to the self-report methods – widely adopted in the social sciences – which he claims to be ill-suited to the Chinese contexts.

Further considerations: challenges and future directions

Apart from providing a multi-faceted perspective on how language and cul-ture issues are relevant for the analyses of emotions in different disciplines, the articles in this issue imply, or directly point to, problematic areas in contemporary emotion theory and research at large. They also indicate the variety of directions of future investigations.

The contributions by Deonna & Teroni and by Cochrane both show that much would be gained were affective scientists of different research orienta-tions able to rely on a broader set of rigorously defined affect terms for their analyses. Yet, although extensive research on working emotion lexicons (e.g. Storm & Storm, 1987; Frijda, Markam & Sato, 1995) or on emotion labels derived from the literature and thesauri (e.g. Ortony, Clore & Foss, 1987; Scherer, 2005: 713–14) are available, these studies emphasize the similarities among states represented by different emotion labels, which could be grouped around a few emotion prototypes.7 Thus far it is the basic-level categories that have been the major focus of categorical approaches; much less attention has been paid to subtler subordinate-level, within- category variation.

This tendency, which is part and parcel of the mainstream philosophy of ‘exploring the affective’ and the quintessential feature of the academic


language on emotions, requires a more detailed commentary. It reflects a more general thrust to achieve homogeneity in (emotion) research at large, which in very general terms lies at the very core of Western science, where the urge for the elegant simplicity of few-element ontologies can be traced from modern times back through the European Enlightenment to Antiquity (Wierzbicka, 2001; Solomon, 2003). More recently this tendency has mate-rialized in a host of studies seeking to establish cultural, language and emo-tion universals.

In the language and culture domain, this orientation has resulted in diverse theories and research programs that emphasize the ‘sameness’ of human culture and language at large, as well as specify the universal principles of their functioning. Such are a modular view of language (e.g. Fodor, 1975), the ‘ratchet effect’ (Tomasello, 1999) and other adaptive mechanisms that ensure the cumulativeness of human culture (e.g. Carruthers, 2009), linguis-tic nativism (Chomsky, 1965; Pinker, 1994), semantic primitives and lingua mentalis (Wierzbicka, 1972, 1980), as well as, more narrowly, the search for phonological, morphological, semantic and syntactic universals in human language (see e.g. Bobaljik, 2008; Hyman, 2008; Newmeyer, 2008; van der Hulst, 2008; von Fintel & Matthewson, 2008, for recent overviews).

In the emotion domain, the tendency has manifested itself in the focus on broad (presumably universal) emotion categories named in language (e.g. Hupka, Lenton & Hutchinson, 1999; Wierzbicka, 1999),8 affect programs (Izard, 1993), basic emotions and ‘emotional plots’ (Ekman, 1992, 1999), and core relational and ‘universal themes’ of emotions (Lazarus, 1991; Ekman, 2004). Most importantly, it has also largely shaped the academic discourse on emotion and resulted in its reliance on a short list of ‘safe’, ‘undisputable’ (English) emotion labels.

Unlike ordinary people, who, despite the givens of the systematic lan-guage constraints and individual expressive powers,9 are able to verbally name or describe a multitude of qualitatively different feelings, academic language repertoires to refer to those subtleties – even where universalist assumptions are not shared – rarely go beyond the common set of ‘anger’, ‘fear’, ‘happiness’, ‘sadness’ and ‘the like’. As a careful overview of the literature suggests, most emotion researchers are prepared to acknowledge between about five and twenty-five meaningful emotion categories (cf. Shiota & Keltner, 2005: 33, for a similar observation), and concerns about the standardization of the list are frequently voiced. This is at least in part a consequence of emotion researchers’ endorsing what can be called an ‘expressive’ perspective on emotion, which maintains that emotion, over and above anything else, results in a set of recognizable behavioral and physio-logical outcomes or symptoms – all seen as relatively involuntary indices, or


‘leakages’, of the person’s internal states.10 The observed clusters of expres-sive symptoms are then named with overarching emotion labels that are – at most – crude approximations of a variety of subjective feelings.

It is not that emotion theorists and empirical researchers are not aware of this ‘one-name-suits-all’ inclusiveness. Ekman (1999: 53) remarks that each of the distinct emotions he proposes is not a single affective state but a ‘fam-ily’ of related states, delimited from other ‘families’ by a set of common characteristics. Panksepp (1982) concedes that rather than arguing for dis-crete emotions (acceptance, fear, panic and rage), what he means are four systems, which he refers to as ‘exploration–curiosity–foraging–expectation–desire’, ‘flight–caution–anxiety–fear–horror’, ‘offense–irritability–anger–rage–fury’, and ‘crying–sadness–sorrow–grief–panic’. Albeit for a very different reason, Griffiths (2004a: 901) argues that specific emotion terms (‘anger’) and the very term ‘emotion’ are examples of what can be called ‘partial reference’, by analogy with the term ‘jade’, which partially refers to both jadeite and nephrite, two minerals with distinct properties and geologi-cal origins. That is, in Rosch’s terms (e.g. 1978), words such as ‘anger’ and ‘fear’ can be names of both basic-level and subordinate-level categories.

Yet adopting Frege’s distinction (1960) between the two meanings a word can have – the ‘sense’, i.e. the thought the word expresses, and the ‘refer-ence’, contained in objects and events to which a word points11 – one would agree with Jerome Kagan that ‘scientists often treat changes in reference as having minimal implications for the meaning of a construct’ (1988: 616). For instance, various states, ranging from the state experienced by a rodent in response to a simple tone paired with shock to the state experienced by a person as s/he enters a doctor’s office to learn the outcome of a breast biopsy (Davidson & van Reekum, 2005: 16), are all labeled as ‘fear’. The same label is ascribed to the state that people are assumed to unconsciously undergo when exposed to phylogenetic or ontogenetic stimuli, such as pic-tures of spiders, snakes, syringes or guns (e.g. Öhman, Flykt & Esteves, 2001; Öhman, Lundqvist & Esteves, 2001; Blanchette, 2006). Without denying that the states above (can) have a convergent set of characteristics such that it is possible to place them in a single category, it is important to stress that they all refer to different situations. But the change in reference entails neither a shift from the prototypical (e.g. ‘fear’) to the particular label (e.g. ‘shock’, ‘apprehension’, ‘dread’, etc.) nor a dramatic change in the ‘sense’ of the label used,12 implying an assumption that those differences are not important.

This is a disputable point, and arguments against such a ‘loose usage’ of emotion terms in scientific discourse have been raised by many researchers working on emotion language (cf. Wierzbicka, 2009). However, the major


line of their argumentation has been concerned with the idea that such a use promotes an ethnocentric bias as well as disregards the cultural connotations of the emotion lexis. A frequent retort to this criticism has been that what psychologists are interested in is not emotion words but emotions as such (Frijda, Markham & Sato, 1995: 121; Ekman, 2004: 13). Leaving the intri-cacies of this debate aside, it is clear that the controversy can be reconciled at least partially by the appeal to the ‘terminologization’ of a limited set of English basic-level emotion terms in academic discourse on emotion, which then leads to fear, anger, happiness (and ‘the like’) becoming scientific rather than ordinary language concepts. Yet it could be argued that the choice of some English emotion terms that are ‘terminologized’ in academic discourse on emotion may not be determined only by the exactness of refer-ence. For instance, Ortony & Turner (1990: 325–6) discuss at length how ‘distress’ can be re-interpreted as a more suitable term to refer to prototypi-cal ‘fear’ scenarios, or show that ‘embarrassment’, rather than ‘shame’, complies with most of the criteria posited for psychologically and physio-logically decomposable emotions. Yet it is the prototypical labels ‘anger’, ‘fear’ and ‘shame’ that retain their central status in academic language on emotion. This might lead us to think that what really counts here is not the (extralinguistically valid) exactness of their reference, but these terms’ pro-totypicality in (the English) language derived from their broad semantics and their frequency of everyday use.

One more important reason that has contributed to the reliance of the academic language about emotions on a short list of ‘safe’, ‘undisputable’ labels to index the affective is the widely assumed skepticism of emotion researchers about the capacity of language to adequately represent emo-tional experience. This skepticism is not limited to emotion psychologists: in philosophy, arguments were raised against ‘the linguistic thesis’ (one-to-one correspondence between linguistic forms that people use to ascribe mental phenomena and the ways in which such phenomena ought to be clas-sified) (Wollheim, 1999: 20). Similarly, but from a very different angle, the ‘theory of resonance’ in anthropology (e.g. Wikan, 1992) advocates going ‘beyond words’ in pursuit of understanding indigenous peoples’ ways of feeling. In social sciences, radical claims are made to the effect that lan-guage ‘is not suited to represent emotional experiences at all’ (Halberstadt, 2005: 19) or that the diverse vocabularies of emotion developed in many societies are necessarily wrong about the feelings they purport to describe (Needham, 1981: 23, quoted in Lutz, 1986: 295).

The contrast between the ordinary and the scientific in conceptualizing the emotional is implied in both Soriano & Valenzuela and Moeschler’s contributions. Their objects of enquiry are the non-representational effects


arising from ordinary language use, which is assumed to be associated to equally ‘naive’, language- and culture-specific mental representation. Both authors theoretically and experimentally seek to account for these affective language effects by scientific structures – a universally validated (Osgood, Suci & Tannenbaum, 1957) dimensional structure of the connotation, or the inferential view of linguistic communication where emotions are seen as products of cognitive processing of utterances.

Yet it is clear that the link between the non-representational affective effects and specific emotion concepts (both ‘lay’ and scientific) needs fur-ther elaboration in future studies of the non-representational. Soriano & Valenzuela explicitly allude to their future research on the connotative struc-ture of Peninsular Spanish emotion terms, which is expected to show how it coheres with that of the Spanish color lexis. In Moeschler’s contribution, the link between specific types of non-propositional effects and the specific types of emotion concepts (e.g. AMAZEMENT, AWE, DELIGHT, REVERENCE, etc.) is fairly loose. This results thus far in rather crude dis-tinctions between affective effects, which are for the most part classified by the author as either pleasant or unpleasant. In part, this is explainable by commonly admitted difficulties in assessing exactly which emotion is at stake in expressive language (cf. Foolen, 1997). Secondly, given that much of the non-propositional arises from creative language usage – implying aesthetic judgment on the part of the hearer – the difficulties in explicitly naming those emotional effects can be linked to similar problems that arise when lexicalizing esthetic emotions.

Another important implication for emotion research at large can be inferred from the articles by Bouvier, Nelis and Zufferey. All offer a perspec-tive on emotion words and concepts that is sensitive to the historical aspects of feeling and provides substantial evidence on the successions in the percep-tions of narrative emotion, both cross-lingually and cross-culturally. This perspective is particularly valuable given the dominant focus of the main-stream affective sciences on the synchronic, rather than diachronic, level of emotion representation, and the tendency to privilege the lay and the naive over the artistic in the conceptualization and verbalization of emotions.

In a broader perspective, the problems addressed by Bouvier, Nelis and Zufferey border on three distinct research orientations in literary criticism, semantics and (broadly conceived) cross-cultural research that promise to be extremely rewarding to pursue in further studies. The first is a research line on emotion in literature which posits emotion as inherently narrative in char-acter and aims to account not only for the ways in which expressed emotions may become narratives, and thus bear on the materiality and structure of their expression, but also for how the readership’s affective response is related to


the narrative structures conveying the emotions (e.g. Frijda & Schram, 1995; Pence, 2004). Secondly, the translation uncertainties and limitations, as dem-onstrated by Greek/French, Latin/French and Chinese/French contrasts, contribute to the semanticists’ debate on the strong-translatability (cf. Katz, 1976) vs the ineffability hypothesis as concerns the possibility to adequately render meanings from one language to another (see also von Fintel & Matthewson, 2008). This then brings the articles closer to the research program on lacunal and untranslatable emotion concepts, a research line with a considerable history in anthropology, linguistics and psychology (e.g. Levontina & Zalizniak, 2001; Wierzbicka, 2001; Ye, 2004; Farell, 2006; Shweder, 2008, among many others). All these research lines are highly rel-evant for the affective sciences domain at large, and the attention called to them by Bouvier, Nelis and Zufferey is hoped to sensitize the social scientists to these issues.

To sum up, the contributions to this special issue show that affect perme-ates language at multiple levels – both explicit and implicit, within and outside of the referential, synchronically and diachronically, lexically and discursively, in a universal and in a culture-specific way – and that the emer-gent interaction between language and affect is complex and bi-directional. At their intersections the articles generate insights into several problematic areas in mainstream emotion research where further research is needed. These include the quantitative and qualitative aspects of the use of the affect terms in scientific discourse on emotion, the interrelationship between ‘ordi-nary’ and ‘scientific’ emotion concepts and the affective, in general, and the necessity to better balance the synchronic and the diachronic, as well as the ordinary and the artistic, in the affective scientists’ endeavors.

All papers also highlight the instrumentality of language analyses for multidisciplinary research on the affective. This allows us to conclude by rephrasing Searle’s (1969) famous tenet that one can have language without money, property, government or marriage, but not all of these without lan-guage. As suggested by the collection of articles we present, one can plausi-bly have emotion without language, but one is much less sure to have language without emotion.

Anna Ogarkova is a postdoctoral researcher at the Swiss Centre for Affective Sciences of the University of Geneva. Her major research interests focus on cognitive linguistics (espe-cially Conceptual Metaphor Theory), (cognitive) lexical semantics and critical discourse analysis applied to the study of emotional expression in natural languages. Dr Ogarkova is one of the organizers for the NCCR ‘Language and culture’ interdisciplinary focus. In addi-tion, she collaborates on the GRID study as the coordinator of the Slavic section and is a member of the ELIN Project on emotional language in international communication. Author’s address: CISA, 7 rue des Battoirs, 1205 Geneva, Switzerland. [email: [email protected]]


Philippe Borgeaud studied history of religions at the University of Geneva and then in Chicago with Mircea Eliade and Arnaldo Momigliano, and regularly attended Jean-Pierre Vernant and Marcel Detienne’s seminars and courses in Paris. His doctoral dissertation (Geneva, 1979) was published under the title The cult of Pan in Ancient Greece (Chicago: University of Chicago Press, 1988). He is presently Professor of History of Religions at the University of Geneva. His recent publications include Aux origines de l’histoire des reli-gions (Paris: Le Seuil, 2004) and Des exercices de mythologie (Paris: Labor et Fides, 2004). He also edited, in 2008, a special issue of Revue de l’histoire des religions (t. 225) on ‘La mort et l’émotion: attitudes antiques’, and has devoted a number of articles to the emotions in cults and myth. Author’s address: 2 rue de la Candolle, 1205 Geneva, Switzerland. [email: [email protected]]

Notes

1. Like several other commentators on linguistic relativity, Hill & Mannheim insist on historical contextualization of Boas, Sapir and Whorf’s claims in the general intellectual background of behaviorism, the Zeitgeist (Carruthers, 2008) of the second quarter of the 20th century, as well as what they (politically incorrectly) term ‘a naive and racist universalism in grammar and equally vulgar evolutionism in anthropology and history’ (Hill & Mannheim, 1992: 384). See also e.g. Lucy (1997) for a detailed overview.

2. It should be noted, however, that this conflation has been extensively challenged cross-disciplinarily (cf. Ortony, 1988; Levinson, 1997; Pavlenko, 2005). See also e.g. Ariel (2002) and Berg (2002), on further differentiation of what is traditionally assumed to be ‘semantic meaning’. In a stricter sense any definitive claim on the emotions’ conceptuality should entail comprehensive analyses of how various aspects of emotions – such as causal antecedents, categorization, consequences, regulation, and so on – configure our mental representations.

3. For example in Russian the two ‘anger’ words, gnev and zlost’, capture the superior/inferior distinction: in an unequal dyadic communicative situation, the former can be attributed to someone with a higher social status but not to the person with a lower one. Likewise the ‘compassion’ words sostradanie (‘compassion’)/sochuvstvie (‘co-suffering’) instantiate the distanced/close distinction (cf. Apresjan, 1997).

4. Methodological diversity in the way the issue has been approached includes semantic analysis (e.g. Johnson-Laird & Oatley, 1989) and similarity judgments (e.g. Shaver et al., 1987) among several other methodologies and their combinations.

5. Duffy’s main concern was that the criteria that distinguish emotions from non-emotions differentiate them in degree rather than in kind (1941: 283, original emphasis); Rorty’s claim (1978: 156) was that the physiological and the intentional in the emotions (e.g. ‘malicious gos-sip in departmental politics; the terror of waking after a nightmare whose drama one has already forgotten; nostalgia-for-the-lilacs-of-yesteryear; and fear in the face of a powerful danger’) do not simply mix in the emotions in a variety of ways but, entering each emotion in a peculiar way, result in the ‘in kind’ differences.

6. In the study on hierarchical clustering of Ifalukian emotion terms, Lutz (1982: 116–20) proposes labeling the basic-level categories not with overarching emotion names, as is always done in the prototype approach (e.g. Shaver et al., 1987; Shaver, Murdaya & Frailey, 2001), but in accordance with the type of situation(s) where the emotions in the cluster have been reported to occur. For Ifalukians, these are the emotions whose antecedents are good fortune, danger, human error, connection and loss, or inability. In a way Lutz’s proposal here is similar to the


proposal from Lazarus on core relational themes (1991: 122); yet it differs from it by being more sensitive to culture-specific information on the relevant types of events that generate emotions in a specific population.

7. The focus on broad emotion categories has given rise to studies seeking to establish if the emerging broad categories would show a coherent sequential pattern of development in natural languages (e.g. Hupka, Lenton & Hutchinson, 1999) or if the frequency of use of central cate-gory members would be consistent diachronically and across varieties of the same language as spoken in different countries (e.g. Delgado, 2007).

8. Despite Wierzbicka’s enduring emphasis on the cultural specificity of any emotion term in any language, she also hypothesizes cross-lingual universality of ‘anger-like’, ‘fear-like’ and ‘shame-like’ words (1999: 286–94). This hypothesis corroborates Hupka and colleagues’ (1999) finding that ANGER, GUILT and ALARM are the first emotion categories to evolve cross-lingually.

9. One of the psychological concepts to account for (very general) individual differences in the tendency to represent emotional experiences is that of ‘emotional granularity’, proposed by Barrett (e.g. 2004).

10. The ‘expressive’ position has been a convenient locus of departure not only because empirical emotion researchers are, as Robert Solomon (2003: 119) puts it, ‘desperate for the observables and something to measure’. Another reason is what Lutz (1986: 294–5) discusses as ‘emotion as physicality’, and ‘emotion as natural fact’, Western conceptions of emotion – the two most pervasive views on emotion in 20th-century academic psychology. The ‘expressive’ stance on emotion has also been influenced by Darwinian evolutionary theory, further pro-moted by the emotion universalists, by the dominance of behaviorism in the second quarter of the 20th century and beyond, by philosophical perceptions of emotion as directly perceivable in the expression (e.g. Austin, 1979) and, for that matter, an earlier famous Jamesian view on emotion (conscious feeling) as caused by its bodily (including expressive bodily) component.

11. Frege’s conception of meaning roughly captures the common tripartite distinction between words, concepts and ‘reality’, a common currency in many disciplines. It should be noted, however, that the conceptions of meaning proposed within linguistics are much more versatile. To give just one example, back in 1969 Komlev (1969: 10) overviewed seven major conceptions where the meaning is construed as: (1) the named object; (2) idea of the object/ ‘ideal object’; (3) notion; (4) relation between: (a) the linguistic sign and the object; (b) the linguistic sign and the idea of the object/‘ideal object’; (c) linguistic sign and the notion; (d) linguistic sign and extralinguistic activity; (e) the signs; (5) function of the word-sign; (6) invariant set of information; (7) reflection of reality.

12. I leave out the question of whether different types of fear between these involve different underlying neurobiological systems. Davidson & van Reekum (2005) and Davidson & Irwin (1999) argue that they do.

References

Aarts, B. (2006) ‘Conceptions of categorization in the history of linguistics’, Language sci-ences 28: 361–85.

Apresjan, V. (1997) ‘Emotion metaphors and cross-linguistic conceptualization of emotions’, Cuadernos de filologia inglesa 6(2): 179–96.

Ariel, M. (2002) ‘Introduction’, Journal of pragmatics 34: 345–8. (Special issue on ‘Literal, minimal, and salient meanings’ edited by M. Ariel)


Austin, J. L. (1979) ‘Other minds’, Philosophical papers, pp. 76–116. Oxford: Oxford University Press.

Barrett, L. F. (2004) ‘Feelings or words? Understanding the content in self-report ratings of emotional experience’, Journal of personality and social psychology 87: 266–81.

Barrett, L. F. (2006) ‘Solving the emotion paradox: categorization and the experience of emo-tion’, Personality and social psychology review 10(1): 20–46.

Beatty, A. (2005) ‘Emotions in the field: what are we talking about?’, Journal of the Royal Anthropological Institute (New series) 11: 17–37.

Berg, J. (2002) ‘Is semantics still possible?’, Journal of pragmatics 34: 349–59.Blanchette, I. (2006) ‘Snakes, spiders, guns, and syringes: how specific are evolutionary con-

straints on the detection of threatening stimuli?’, Quarterly journal of experimental psychol-ogy 59(8): 1484–504.

Bobaljik. J. D. (2008) ‘Missing persons: a case study in morphological universals’, The linguis-tic review 25: 203–30.

Boucher, J. D. (1979) ‘Culture and emotion’, in A. J. Marsella, R. G. Tharp & T. V. Ciborowski (eds) Perspectives on cross-cultural psychology, pp. 159–78. San Diego, CA: Academic Press.

Carruthers, P. (2008) ‘Language in cognition’, in E. Margolis, R. Samuels & S. Stich (eds) The Oxford handbook of philosophy of cognitive science. Oxford: Oxford University Press.

Carruthers, P. (2009) ‘The distinctively human mind: the many pillars of cumulative culture’, in G. Hatfield (ed.) The evolution of mind, brain, and culture. Pittsburgh, PA: Penn Museum Press.

Carston, R. (2002) Thoughts and utterances. Oxford: Blackwell.Charland, L. C. (2002) ‘The natural kind status of emotion’, British journal of philosophical

science 53(4): 511–37.Charland, L. C. (2005) ‘The heat of emotion: valence and the demarcation problem’, Journal

of consciousness studies 12(8–10): 82–102.Chomsky, N. (1965) Aspects of the theory of syntax. Cambridge, MA: MIT Press.Church, A. T., Katigbak, M. S., Reyes, J. A. S. & Jensen, S. M. (1998) ‘Language and organisa-

tion of Filipino emotion concepts: comparing emotion concepts and dimensions across cul-tures’, Cognition and emotion 12(1): 63–92.

Davidson, R. J. & Irwin, W. (1999) ‘The functional neuroanatomy of emotion and affective style’, Trends in cognitive science 3:11–21.

Davidson R. J. & van Reekum, K. M. (2005) ‘Emotion is not one thing’, Psychological inquiry 16(1): 16–18.

Delgado, A. (2007) ‘Spanish basic emotion words are consistently ordered’, Quality and quan-tity, available at http://www.springerlink.com/content/m9306h75nm0083w5/?p=1abdf20c64a043c9b0ff2874e0b9ed48&pi=83

Duffy, E. (1941) ‘An explanation of emotional phenomena without the use of the concept emo-tion’, Journal of general psychology 25: 283–93.

Ekman, P. (1972) ‘Universals and cultural differences in facial expressions of emotion’, in J. K. Cole (ed.) Nebraska symposium on motivation, vol. 19. Lincoln: University of Nebraska Press.

Ekman P. (1992) ‘Are there basic emotions?’, Psychological review 99: 550–3.Ekman, P. (1999) ‘Basic emotions’, in T. Dalgleish & M. Power (eds) A handbook of cognition

and emotion, pp. 45–60. Sussex: John Wiley & Sons.Ekman, P. (2004) Emotions revealed. New York: Henry Holt.Farrell, P. (2006) ‘Portuguese SAUDADE and other emotions of absence and longing’, in

B. Peeters (ed.) Semantic primes and universal grammar, pp. 235–58. Amsterdam: John Benjamins.


Fehr, B. & Russell, J. A. (1984) ‘Concept of emotion viewed from a prototype perspective’, Journal of experimental psychology: General 113: 464–86.

Fiedler, K. (2008) ‘Language, a toolbox for sharing and influencing social reality’, Perspectives on psychological science 3(1): 38–47.

Fodor, J. A. (1975) The language of thought. Cambridge. MA: Harvard University Press.Foolen, A. (1997) ‘The expressive function of language: towards a cognitive semantic

approach’, in S. Niemeier & R. Dirven (eds) The language of emotions: conceptualization, expression, and theoretical foundation, pp. 15–32. Amsterdam: John Benjamins.

Frege, G. (1960) ‘Uber Sinn und Bedeutung’ [On sense and reference], in P. Geach & M. Black (eds & trans.) Translations from the philosophical writings of Gottlob Frege, pp. 56–78. Oxford: Basil Blackwell. (Reprinted from Zeitschrift für Philosophie und philosophische Kritik, 1892, 100: 25–56)

Frijda, N. & Schram, D., eds (1995) Emotions and cultural products, Poetics 23. (Special issue)Frijda, N., Markam, S. & Sato, M. (1995) ‘Emotions and emotion words’, in J. A. Russell,

J.-M. Fernandes-Dols, A. S. Manstead & J. C. Wellenkamp (eds) Everyday conceptions of emotions: behavioural and social sciences 81: 121–45. (NATO ASI Series, Series D)

Galati, D., Sini, B., Tinti, C. & Testa, S. (2008) ‘The lexicon of emotion in the neo-Latin languages’, Social science information sur les sciences sociales 47(2): 205–20.

Gerber, E. (1975) ‘The cultural patterning of emotions in Samoa’, doctoral dissertation, Anthropology Department, University of California, San Diego.

Greenwald, A. G., McGhee, D. E. & Schwartz, J. L. K. (1998) ‘Measuring individual differ-ences in implicit cognition: the Implicit Association Test’, Journal of personality and social psychology 74: 1464–80.

Griffiths, P. E. (2004a) ‘Emotions as natural kinds and normative kinds’, Philosophy of science 71: 901–11.

Griffiths, P. E. (2004b) ‘Is emotion a natural kind?’, in R. C. Solomon (ed.) Philosophers on emotion, pp. 233–49. Oxford; New York: Oxford University Press.

Hacking, I. (2001) ‘Aristotelian categories and cognitive domains’, Synthese 126: 473–515.Halberstadt, J. (2005) ‘Language, emotion attribution, and emotional experience’, Psychological

inquiry 16(1): 18–21.Heider, K. G. (1991) Landscapes of emotion: mapping three cultures of emotion in Indonesia.

Cambridge: Cambridge University Press.Hill, J. H. & Mannheim, B. (1992) ‘Language and worldview’, Annual review of anthropology

21: 381–406.Hoekstra, H. A. (1986) Cognition and affect in the appraisal of events. Groningen:

Rijksuniversiteit Groningen.Hupka, R. B., Lenton, A. P. & Hutchison, K. A. (1999) ‘Universal development of emotion

categories in natural language’, Journal of personality and social psychology 77(2): 247–78.Hyman, L. M. (2008) ‘Universals in phonology’, The linguistic review 25: 83–137.Izard, C. E. (1993) ‘Four systems for emotion activation: cognitive and noncognitive pro-

cesses’, Psychological review 100: 68–90.James, W. (1890) The principles of psychology. New York: Holt.Johnson-Laird, P. N. & Oatley, K. (1989) ‘The language of emotions: an analysis of a semantic

field’, Cognition and emotion 3: 81–123.Kagan, J. (1988) ‘The meanings of personality predicates’, American psychologist 43(8):

614–20.Katz, J. (1976) ‘A hypothesis about the uniqueness of natural language’, Annals of the New

York Academy of Sciences 28: 33–41. (Special issue on ‘Origins and evolution of language and speech’, edited by S. R. Harnad, H. D. Steklis & J. Lancaster)


Komlev, M. G. (1969) Komponenty soderzhatel’noj struktury slova [The components of the content structure of a word]. Moscow: Moscow State University Press.

Lazarus, R. S. (1991). Emotion and adaptation. New York: Oxford University Press.Levinson, S. C. (1997) ‘From outer to inner space: linguistic categories and non-linguistic think-

ing’, in E. Pederson & J. Nuyts (eds) With language in mind: the relationship between lin-guistic and conceptual representation, pp. 13–45. Cambridge: Cambridge University Press.

Levontina, I. B. & Zalizniak, A. A. (2001) ‘Human emotions viewed through the Russian language’, in J. Harkins & A. Wierzbicka (eds) Emotions in crosslinguistic perspective,pp. 291–336. Berlin; New York: Mouton de Gruyter.

Lucy, J. A. (1997) ‘Linguistic relativity’, Annual review of anthropology 26: 291–312.Lutz, C. (1982) ‘The domain of emotion words on Ifaluk’, American ethnologist 9: 113–28.Lutz, C. (1986) ‘Emotion, thought, and estrangement: emotion as a cultural category’, Cultural

anthropology 1(3): 287–309.Lyons, J. (1977) Semantics, vol. 1. Cambridge: Cambridge University Press.MacLaury, R., Paramei, G. & Dedrick, D., eds (2007) Anthropology of colour. Amsterdam:

John Benjamins.Mageo, J. M. (1991) ‘Samoan moral discourse and the Loto’, American anthropologist (New

series) 93(2): 405–20.Myers, F. (1979) ‘Emotions and the self: a theory of personhood and political order among

Pintupi Aborigines’, Ethos 7: 343–70.Needham, R. (1981) Circumstantial deliveries. Berkeley: University of California Press.Newmeyer, F. J. (2008) ‘Universals in syntax’, The linguistic review 25: 35–82.Öhman, A., Flykt, A. & Esteves, F. (2001) ‘Emotion drives attention: detecting the snake in the

grass’, Journal of experimental psychology: general 130(3): 466–78.Öhman, A., Lundqvist, D. & Esteves, F. (2001) ‘The face in the crowd revisited: a threat advan-

tage with schematic stimuli’, Journal of personality and social psychology 80(3): 381–96.Ortony, A. (1988) ‘Are emotion metaphors conceptual or lexical?’, Cognition & emotion

2: 95–103.Ortony, A. & Turner, T. (1990) ‘What’s basic about basic emotions?’, Psychological review

97: 315–31.Ortony, A., Clore, G. & Foss, M. (1987) ‘The referential structure of the affective lexicon’,

Cognitive science 11: 341–64.Osgood, C. E., Suci, G. J. & Tannenbaum, P. H. (1957) The measurement of meaning. Urbana;

Chicago; London: University of Illinois Press.Panksepp, J. (1982) ‘Toward a general psychobiological theory of emotions’, The behavioral

and brain sciences 5: 407–67.Pavlenko, A. (2005) Emotions and multilingualism. Cambridge: Cambridge University Press.Pence, J., ed. (2004) ‘Introduction’, Journal of narrative theory 34(3). (Special issue on

‘Narrative emotion: feeling, form and function’)Pinker, S. (1994) The language instinct. London: Penguin Books.Rorty, A. (1978) ‘Explaining emotions’, The journal of philosophy 75(3): 139–61.Rosch, E. (1978) ‘Principles of categorization’, in E. Rosch & B. Lloyd (eds) Cognition and

categorization, pp. 27–48. Hillsdale, NJ: Lawrence Erlbaum Associates.Russell, J. A. (1991) ‘Culture and the categorization of emotions’, Psychological bulletin

110: 426–50.Ryle, G. (1961) The concept of mind. New York: Barnes & Noble.Sabini, J. & Silver, M. (2005) ‘Ekman’s basic emotions: why not love and jealousy?’, Cognition

and emotion 19(5): 693–712.


Scherer, K. R. (2005) ‘What are emotions? And how can they be measured?’, Social science information sur les sciences sociales 44(4): 695–729.

Searle, J. R. (1969) Speech acts. Cambridge: Cambridge University Press.Shaver, P. R., Murdaya, U. & Frailey, R. C. (2001) ‘Structure of the Indonesian emotion lexi-

con’, Asian journal of social psychology 4: 201–24.Shaver, P. R., Schwartz, J., Kirson, D. & O’Connor, C. (1987) ‘Emotion knowledge: further

explorations of a prototype approach’, Journal of personality and social psychology 52: 1061–86.

Shiota, M. N. & Keltner, D. (2005) ‘What do emotion words represent?’, Psychological inquiry 16(1): 32–7.

Shweder, R. (2008) ‘The cultural psychology of suffering: the many meanings of health in Orissa, India (and elsewhere)’, Ethos 36(1): 60–77.

Solomon, R. C. (2003) Not passion’s slave. Oxford: Oxford University Press.Sperber, D. & Wilson, D. (1995) Relevance. Oxford: Blackwell.Storm, Ch. & Storm, T. (1987) ‘A taxonomic study of the vocabulary of emotions’, Journal of

personality and social psychology 53(4): 805–16.Tomasello, M. (1999) ‘The human adaptation for culture’, Annual review of anthropology

28: 509–29.Van der Hulst, H. (2008) ‘On the question of linguistic universals’, The linguistic review

25: 1–34.Von Fintel, K. & Matthewson, L. (2008) ‘Universals in semantics’, The linguistic review

25: 139–201.White, G. (1978) ‘Ambiguity and ambivalence in A’ara personality descriptors’, American

ethnologist 5: 334–60.White, G. (1980) ‘Conceptual universals in interpersonal language’, American anthropologist

82: 759–81.Wierzbicka, A. (1972) Semantic primitives. Frankfurt: Athenaeum.Wierzbicka, A. (1980) Lingua mentalis. Sydney: Academic.Wierzbicka, A. (1999) ‘Emotional universals’, Language design 2: 23–69.Wierzbicka, A. (2001) Sopostavleniye kul’tur cherez posredstvo leksiki i pragmatiki [Comparison

of cultures through lexis and pragmatics]. Moscow: Yazyki slavyanskoj kul’tury.Wierzbicka, A. (2009) ‘Language and metalanguage: key issues in emotion research’, Emotion

review 1(1): 3–14.Wikan, U. (1992) ‘Beyond the words: the power of resonance’, American ethnologist 19(3):

460–82.Wollheim, R. (1999) On the emotions. New Haven, CT; London: Yale University Press.Ye, Z. (2004) ‘The Chinese folk model of facial expressions: a linguistics perspective’, Culture

and psychology 10: 195–222.Zinck, A. & Newen, A. (2008) ‘Classifying emotion: a developmental account’, Synthese

161(1): 1–25.

article (published version) - archive ouverte unige

Documents