Michael Bamberg

Department of Psychology, Clark University

1. Introduction

The findings I am about to report in this paper were not anticipated when I began this line of research a few years ago. Originally, I did not intend to investigate and compare children's accounts of emotion situations, in particular their accounts of anger and sadness situations. Rather, I had started with the aim to investigate how children orient themselves toward - so to speak - 'the same situation', but from different genre-perspectives. More specifically, I was interested in children's accounts of situations in which they used the personal pronoun I (in order to refer to past events, or "personal experiences") in contrast to accounts of events in which a third person (s/he) went through the same "experience". In addition, I attempted to also compare those two genres to one in which a generalized person (one, or the generalized you) acted or was acted upon. In short, my original investigation aimed at a genre comparison of (a) personal narrative, (b) third-person narrative, and (c) explanatory discourse. The idea to employ emotion situations such as 'being angry' or 'being sad' came up in the attempt to find a situation that was ecologically meaningful for both younger and older children, and was 'of the same kind', so I could compare the linguistic devices used according to the age of the children and according to the genre that was targeted by the children.

This in mind we [See footnote 1] asked 80 (American-English-speaking) children ranging from preschool to 3rd-grade (4-10 years of age) to tell us about "one time when you were angry / sad / scared / happy" - prompting for the account of a personal experience; in addition, we asked them to imagine a little boy or girl and to give us an account for "one time, when s/he was angry / sad / scared / happy"; and last, we simply asked to explain, "what it means to be angry / sad / scared / happy". This resulted in 12 interview questions, which were randomly assigned, all revolving around different perspectives on those particular emotional situations.

    In the course of interviewing the children, we stumbled across something that should turn around the way we conceptualized children's abilities to use linguistic constructions for discourse purposes. We repeatedly heard some of the children maintaining that one or the other question had already been answered, which initially did not surprise us, because the battery of questions could easily confuse interviewers, particulary if they did not fully concentrate. However, when scrutinizing the data more thoroughly, we realized that these kinds of confusions occurred solely when we had asked to give anger or sadness accounts, but never with any of the others. When we asked in these situations to give us an answer anyway, the accounts were most often word-for-word repetitions. In addition, we noticed that these confusions were more typical for the younger than for the older children.

    This preliminary evidence seemed to hold some water, although it is not in agreement with what one would expect based of research reports of children's emotion knowledge (cf. Stein & Trabasso, 1992; Stein, Trabasso & Liwag, 1992). According to those reports, children as young as 3 years of age are able to perform at a high level of proficiency in figuring out the different components that lead to emotional states such as 'angry' and 'sad' and what typically follows from them. The only encouragement to probe deeper into this observation came from anthropological reports about a number of African languages, which, at least at the lexical level, do not seem to differentiate between what is divided according to the English lexicon into 'anger' and 'sadness' (cf. Davitz, 1969; Heelas, 1986; Leff, 1973; Matsumoto, 1994). Thus, this original accidental stumbling across some children's confusions of angry and sad launched us into a closer look at how the accounts of sadness and anger were linguistically constructed by younger (and older) American-English-speaking children, and what these accounts were actually used for when it came to a comparison between the different genres.

    For the purpose to clarify how young children actually come to confuse two so-called basic emotion concepts - at least in the genre of narrative accounts (although this genre is highly relevant for self accounts and identity presentations) - I will first discuss some general tenets of the relationship between language, thought and emotion, and their relationship for developmental studies. Then, I will show in more detail, how my study of emotion talk led to the differentiation of two different 'grammars', that of 'anger' and that of 'sadness', and to how young children's confusion between these two grammars can be accounted for. In my concluding section, I will take up on the relationship between narrating and emotion talk with the somewhat 'radical' argument that talk is more foundational than traditionally credited _ not only for the way we make sense of emotions, but also - at least to a degree - for how we actually feel.

2. Emotion Concepts and Emotion Words versus Emotion Talk as a Practice

Generally speaking, talk about emotions, i.e., talk in which emotions are thematized, seems to imply that emotions are objects or entities that have an existence outside of talk and apart from language in general. In this, they are very similar to our folk notions of events that have their existence outside of talk, but can be referred to - just like emotions - in and through talk. Events and emotions could have taken place in the past or they can be imagined; they can be of a 'personal nature', i.e., the teller could be centrally (or peripherally) involved, or they can be of a completely detached, impersonal nature, where the teller is not simultaneously thematizing him-/herself as in accounts of emotions of others, past or imagined, or as in explanations, definitions or other more detached situations such as in card sorting tasks (cf. Lutz, 1988).

    The question, however, that immediately comes to mind, is how do we know what emotions are and what they mean, and more specifically, how do children learn the mesnings? _ In order to answer this question, we may be thrown back onto language and emotion talk as the sources and possibly even resources that tell us what we know about emotions and how they are dealt with in the social, communicative realm.

    A way to avoid the issue of dealing with language and emotion talk as somewhat foundational to our understanding of emotion, would be by way of borrowing from a theory of "natural perception". In this theory, emotions are not really learnt. They are bodily experiences that are directly sensed and differentiated into a limited number of emotion categories. What is learnt are the language appropriate labels for these categories. And although much of our everyday talk about emotions and feelings seems to rely on this theory, anyone who has struggled with a foreign language knows that the emotion categories we learnt with our first language are not the same as in any other language: Natural perception cannot automatically read off from bodily sensations the categories that are considered meaningful for the speakers of particular languages. Thus, we seem to be thrown back onto language as one of the sense giving foundations when it comes to emotion categories.

    A second route to avoid taking talk in any way as a foundational factor for the constitution of emotions as meaningful entities, though by far more sophisticated than the theory of natural perception, relies on the intuition that all humans have emotions, and that the particular language that we learn as our first, just carves up the "emotion spectrum" different from any other language, leaving us with the impression that our (first) language does it somewhat more naturally, while other languages are somewhat derived. This theory is actually quite similar to the one that has been developed for color categories, and been proposed in its most sophisticated version by Wierzbiecka (1992, 1994, 1995), and recently also Goddard (1997). The basic tenets on which this theory rests are cognitive universals. In short, resting on the assumption that human cognition (the mind) can differentiate between the different emotion categories and translate emotion terms from one language into another by use of a limited set of (cognitive/semantic) universals, the foundational capacity for making sense is attributed to cognition, not to language.

    A similar theoretical advance on the relationship between emotion, language, and cognition, although not from a cognitive/semantically universalist point of view has been made by Stein and her associates (Stein & Levine, 1990; Stein & Trabasso, 1992; Trabasso & Stein, 1993). In her approach, emotions are viewed as tied into relations between people, although they are approached as a representational system of the goal-plan-outcome knowledge that is held to regulate and coordinate the relations between people. Thus, knowledge of goals and plans is assumed to form the core prerequisite for making sense of others, and it figures foundationally in explaining and accounting for one's own actions, i.e., in the process in making sense of one's own self. According to Stein and her associates, this type of knowledge is acquired relatively early, at around three years of age. At this point, children are assumed to be apt to successfully differentiate between the components of actions and goal-plans that lead to (English) anger, sadness, fear, or happiness (Stein, Liwag, & Wade, 1997).

    In contrast to Wierzbicka and her colleagues, who view emotions as a semantic domain that governs the patterns of discourse, Stein uses narratives of real life emotion situations and subjects them to on-line questions for on-line reasoning. However, similar to Wierzbicka, Stein and her colleagues use discourse data to analyze language in its ideational, representative function, i.e., as a more or less transparent window into the conceptual underpinnings of what their talk is about. The aboutness of talk (or what is 'behind' the talk) is taken as basic, irrespective whether the speaker wants to be understood as blaming someone else or saving face, i.e., the directive force of language (the interpersonal function) is not considered as of immediate relevance to the meaning of the emotion account, nor to the meaning of the situatedness of 'the emotion', nor to an 'emotion' in general. Thus, what both theories, that of Wierzbicka and that of Stein, share is a theoretical proceeding from the abstract to the concrete: The meaning of 'emotion' is a foundational concern for its application in situated expressions or displays, and those are foundational for situated verbal accounts.

    While there has been an abundance of theorizing over the last decade on the relationship between language, cognition and emotion, most of it nevertheless has centered on the more narrow relationship between (emotion) concepts and (emotion) words. And although ethnomethodological approaches to emotion talk in other cultures/languages (Basso, 1992; Lutz, 1988; Ochs, 1988, 1996; Schieffelin, 1990; Schieffelin & Ochs, 1986) have repeatedly underscored the situatedness and cultural contextualization of emotion talk for the way emotions make sense and can enter as meaningful entities the interactions of participants, their contributions nevertheless have mostly gone unheard or they have been misconstrued as dealing predominantly with concepts and words (REFERENCES TO reviews). Developmental studies, with only few exceptions ( Clancy, ....), have predominantly targeted emotion concepts and in this sense are very much in line with mainstream developmental attempts to contribute to a debate regarding by what age children have emotion concepts (or at least the basic ones), have 'a theory of mind', the basic narrative components, or other psychological 'objects'. It is also interesting to note that not much of this debate has been incorporated into standard language acquisition volumes ranging from general psycholinguistic introductions to the acquisition of the lexicon (cf. Gleason & Ratner, 1993; Clark, 1993; Fletcher & MacWhinney, 1995; Hoff-Ginsberg, 1997), which may be due to the fact that the grounds of what actually develops and what it is that facilitates development, are rather murky: Is it concepts that develop and at one point or another become mapped onto the appropriate linguistic forms, or is it linguistic forms that develop, channeling thoughts and cognitions toward socially appropriate ways of making sense? And further, is making sense primarily a more reflective, conceptual activity, or can it be also described in mor practical, participatory terms?

    In this chapter, the attempt is made to break out of the cycle of describing language use as principally based on cognitive terms or conceptual entities. Starting from the assumption that emotion displays are deeply embedded in our human way of displaying ourselves as situated selves in situations with others, we do not deny that emotion displays have physiological ('bodily caused reactions') correlations, or that situations can be conceptually structured and talked about. However, in order to determine how emotion displays gain their meaning as meaningful events, we cannot solely rely on physical reactions to stimuli or to conceptual structures in the mind of private individuals as foundations. Rather, body and mind occupy (jointly) a social space in unfolding episodes, communication the (relational) position of a self vis-á-vis others. This is where joy, anger, shame, surprise, and the like "show" as meaningful position -- taken up by a person purposely. Thus, these positions are displayed as actions that are purposely taken up to signal and signify a self-other relationship [See footnote 2] . In this way, we can study the kinds of judgements, aesthetics, morality, and prudence that are expressed in emotion displays _ we can determine what interactionally has led up to an emotion display, and what the display has accomplished. In this sense then, emotions are like conversations (Harré & Gillett, 1994) and accounts of emotion situations typically work up the aesthetics, judgements, morality involved in such situations.

    In sum, the present chapter does not take narrative accounts of emotion situations as windows into some <underlying> conceptual (mind) or physical (body) foundations of human meaning making, but as windows to the positionings that are being performed in the form of narrative actions. In these positionings narrators provide the audience with an order so that they can convince, blame, or save face, i.e., practically orient the audience to an order within which judgements, aesthetics and morality are purposely arranged. And since these narratives are constructed for an audience, the window metaphor should have better been replaced by the image of a signpost or pointer: The actual performance of the narrating act orients the audience to attend to the order constructed. And since the researcher is participant in this construction process, the construed order may also be considered as part of the research situation itself. It should be noted that this account of the relationship between language, emotion, self and other, mind and body is centrally dialogical.. Since language is always an embodied act and it always is centrally dialogical, it orients selfhood and otherness to one another in a foundational way (see Bamberg, in preparation, for further discussion on the centrality of language).

    It should also be noted that the approach presented in this chapter bears heavily on the notion of development: In contrast to mapping out changes over time of uses of words (semantic structures) or linguistic applications of conceptual structures and claiming that this is what develops, we see the issue of development much more closely tied to the issue of participating in (linguistic) practices. And since these practices are no longer conceived of as structures that have their existence apart from the person (or within the person as internalized or matured mental structures), but rather as embodied discourse activities, the trajectory of language development (here: emotion talk) is no longer a single domain - such as lexical development, but closely intervowen with the development of the person as a whole. We will have to return yo this issue in the concluding section.

3. The Grammars of Anger and Sadness in American English-Speaking Children

In the following I will extrapolate the linguistic devices that are typically employed in the construction of two types of situations, being-angry and being-sad. I will not detail the findings for each single age group, but for contrasting purposes compare the older children (the third-graders - mean age: 9;1) with the younger age groups (preschoolers <mean age: 5;2> and kindergartners <mean age: 6;1>). In addition, I will only draw on the comparison between the first-person and the explanatory genre, neglecting here a more detailed discussion of the third-person genre (see Özçaliskan, 1997, for a report of the genre findings across all four emotions).

The Grammar of Anger in the First-Person Genre. The linguistic devices employed by older and younger children to construct angry situations typically consist of (i) a highly individuated agent (my sister - see example [1] below), and a highly individuated undergoer (me); (ii) a marking of the action as highly transitive; (iii) a positioning of the I as the recipient and target of the action in the direct object slot; and (iv) a positioning of the other (the agent) in subject slot. See examples (1) and (2) as illustrations for how anger is typically constructed in terms of the above five linguistic construction types:

(1)    I was in the room
    and my sister kicked me
    and it went right into the rib bone

(2)    when my sister slapped me across the face
    just because she didn't let me in her room
    and I wanted to play a game
    but she didn't let me
    and slapped me across the face

In terms of the discursive purposes for which these lexico-syntactic devices are employed, we can tentatively point into two directions: On the one hand, the construction of a highly individuated target of others' actions may orient the audience toward empathy or sympathy, particularly if the action is not sufficiently motivated or justified. On the other hand, introducing the other as the topical focus in the position of the syntactic subject, opens her/him to become subjected to blame, again, particularly when the action was unmotivated or unjust. In the anger accounts of children across the age-ranges, this topical focus on the perpetrator (for the purpose of attributing blame) overshadows, so to speak, the discursive purpose of eliciting empathy for the victim. Or, in other words, the construction of anger in American-English consists (developmentally from very early on) of two discursive purposes: blame and empathy; with the latter subordinated to the former.

The Grammar of Sadness in the First-Person Genre. Typical for the accounts of older American children (as well as adults) are two different construction types: (i) positioning the other in subject position, as in example (3), or (ii) positioning the I in subject slot, as in example (4):

(3)    it was when I was about 5- or 4-years old
    my biggest sister got into a car accident
    so she died
    because of a car accident
    and I was really sad for a few weeks

(4)    I was in Charlton
    and I moved to Worcester
    and I couldn't see my neighbors and their dogs

While construction type (i) holds up the possibility to make the other (here my sister) the potential topical focus, and as such orients the discourse purpose toward blame, this option is ruled out by two additional linguistic devices; (a) the avoidance and downplay of the other's agentivity, and (b) the absence of a target of the activity referred to (dying is atelic). These two devices are similarly employed in construction type (ii) (cf. example 4), denying the I to achieve the status of a topical focus - which could only have resulted in self-blame.

Thus, it can be maintained that the grammar typically employed for the construction of sadness differs from the grammar of anger in degree of complexity: While anger consists of one (formal) construction type, but comprises the two discursive orientations empathy and blame (and as such requires a delicate balance between these two orientations), sadness is less complex in terms of discursive purposes, because it is geared toward only one discourse purpose, namely empathy, but more complex in terms of the existence of two constructive options. In addition, taking the prototypical English construction type of the transitive scene, the construction types employed to orient toward empathy are also more complex, since they are deviations from the prototype. They require an additional downplay of the topical focus. In other words, after the subject (which typically is the topical focus) has been established, this focus has to be "defocused" in subsequent clauses. The clause "because of a car accident" (in example 3) illustrates this function, removing "my sister" from potentially becoming the topical focus, implying that 'she didn't really do anything' - 'this is not really about her'. Thus, although the construction of anger - in English - is more complex when it comes to its discursive purposes, the construction of sadness is more complex in terms of its actual linguistic construction types.

Turning next to the description of how sadness was linguistically constructed in the first-person genre by the younger children, we find their accounts structurally equivalent with the anger accounts of all American English-speaking subjects (including their own). Examples (5) and (6) illustrate this point:

(5)    when Nikki hit me in the eye
    I was really really sad
    I cried for a whole half an hour

(6)    my Mommy hit me
    she hit me in the eye
    and I was sad
    and cried

These accounts typically consist of two components: The first part topicalizes the perpetrator by constructing a highly transitive event, which is likely to be taken to orient the audience toward an attribution of blame to the agent. However, in the second part, the topic shifts from the other to the I, orienting the audience toward empathy as the discursive purpose of the two components as a whole.

Within the frame of these structural characteristics, the younger children's constructions of sadness (in the first-person genre) are different from the older children's in that they consist of two units which co-exist next to one another. While the older children either topicalized a non-agentive other in order to shift to a sad I as in example (3), or topicalized the I all the way as in example (4), the younger subjects seemed to be glued to a highly agentive other in subject position as the topic of what the described situation 'is all about'. Thus, it seems to be the fixation with the English construction type typical for the construction of a transitive scene that is difficult for the younger subjects to overcome. They must either de-agentivize the other and construct the topic of an helpless I, or construct a non-agentive I in subject position as illustrated in example (4). Consequently, for our younger American children, the attempt to construct a sadness scenario ends up very much like the construction of an anger scenario.

The Grammar of Anger in the Explanatory Genre. The construction of anger in explanatory discourses is achieved by a number of (mostly running in concert) construction types: (i) an unspecified agent in subject slot, most often plural they; (ii) the unspecified target of the activity described in direct object slot, most typically you (where it remains unclear whether you refers to an unspecified hypothetical person or to the interviewer); (iii) an active verb which nevertheless is much less specific (and therefore less transitive) when compared with verbs used in the first-person genre (e.g. doing something or hurting in the explanatory genre versus hitting and kicking in the first person genre); (iv) the clause modus is most often marked by if or when, in conjunction with (v) the present tense, taking the situation into the realm of the timeless and possible world. Examples (7) and (8) illustrate these types:

(7)    if someone hurts you
    and you get really really really mad
    then you are angry

(8)    you are angry at someone
    because they did something to you
    and you didn't like it
    what they did

In more general terms, the construction types result in the overall construct of an anger scenario that is much less of a bounded event, less vivid, and presented from a much more detached perspective than anger was constructed in the first-person perspective. Seeking empathy from the audience or blaming the other for any transgression clearly do not matter. If there is a particular discursive orientation, it lies in 'describing' or making 'what usually occurs' explicit, though clearly from a detached vantage point. The audience is led out into a world of usual occurrences, distanced from the realm of the special occurrence of subjective experience that made examples (1) through (6) tellable narratives.

As already mentioned, the six construction types are used in concert by the older children. Younger children have difficulties in using all five simultaneously, and they also often slip after having given an account in the explanatory genre into the first-person genre, telling how this once happened to them. However, across the board, they all are able to employ at least a few of the above mentioned construction types. And in terms of younger children's general competence to construct sadness scenarios in the explanatory genre, their accounts are well in place.

The Grammar of Sadness in the Explanatory Genre. Similar to the accounts given in the first- person genre, a sad situation is constructed by our older subjects as well as by the younger children in terms of either 'something [bad] is happening to you' (as in example 9), or in terms of 'you want something, but you can't have it' (example 10). Agentive others, who could be held responsible, very rarely figure in these accounts, and if they do, they are always defocused as potential targets of blame. Example (9) illustrates the 'something bad is happening to you' scenario by transforming in line 4 the potential agents into an impotent it of a mere happening that leaves you helpless (because there is no target for revenge):

(9)    when like someone calls you four-eyed
    if you have glasses
    and you get not mad at them
    but it hurts your feelings
    and you're sad
(10)    like your favorite blanket was up high
    where you couldn't get it
The only difference between the younger and the older children in their construction of sad scenarios in the explanatory genre was that the younger children at times seemed to consider a description of the behavioral display of being sad a sufficient explanation of "what it means to be sad" as in crying or if somebody cried .

Thus, in spite of some difficulties in sorting out the linguistic complexities of the explanatory genre at an early age, all children clearly demarcated what it means to be angry from what it means to be sad, construing the former as a bounded unit, which was generalized and presented from a detached discourse orientation; and construing the latter as a non-agentive happening, with no particular other as a target to be blamed. The discursive orientations of attributing blame or eliciting empathy were backgrounded, while the discursive orientation to describe general occurrences from a detached perspective became the essential interactive goal.

Summary. In this paper I have described, thus far, the construction types children between four and ten years use in their anger and sadness accounts in the two genres (a) personal experience narratives, and (b) explanations. The descriptions of the construction types revealed that young children construed their personal experience of anger and sadness by use of the same devices, while older children constructed personal experiences of anger and sadness distinctly. Further, as documented in the descriptions of the construction types used for explanations, younger children clearly are able to differentiate between anger and sadness situations. From this it was concluded that the reasons for the confusion between anger and sadness in our younger population had to do with the particular affordances of the first-person genre. More specifically, a differentiation between the discursive purposes of personal experience accounts of anger and sadness situations revealed different degrees of complexities in how these accounts are constructed. It was concluded that this was the source of confusion for young American English-speakers, due to an early underdifferentiation of how the linguistic devices necessary to construct personal experiences of anger and sadness situations are to be put to work.

4. Conclusion

In conclusion, there is more to these data (and their analysis) than the finding that young children conflate anger and sadness in the genre of personal narratives. Since we were able to rule out the possibility that these findings are due to some general lapse in concentration, and since children at this age seem to be well equipped to differentiate between sadness and anger in other genres, a reconsideration of the relationship between emotion, cognition, and the role of language in their configuration seems to be at place. In the remainder of this paper, I will pursue two lines of reasoning.

    First, it appears to be problematic to maintain the (folk) assumption that the development of emotion talk is governed by, and only possible, if a conceptual structure of emotions is already in place. In contrast to this "cognition- first-hypothesis" for the development of emotion talk, children's ability to engage in emotion talk appears to depend primarily on their abilities to coordinate lexico-syntactic constructions with the discursive purposes of such talk. Thus, the cognitive-developmental assumption that holds that conceptual structures of emotion knowledge develop independent of language is in serious need of revision.

In addition, the findings seem to suggest that emotions are - at least to a degree - linguistically constructed. As we could show, learning to make sense what it means to be angry or sad is closely tied to how we organize the relationship between agents and undergoers. And this is primarily a linguistic achievement. _ This is not meant to imply that the English language as a formal system constrains speakers' sense making of emotions. However, participating in emotion talk, where we lay out what happened in terms of who did what to whom, and with what kind of underlying motivations - in other words, where we take a stance in who was right and who was wrong -, socializes children (and not only children) into conventional ways of using construction types for discourse purposes. These kinds of practices not only shape an understanding of what emotions are, but even more radically, they enable the person to learn how to feel.

I am aware that this construal of the relationship between language practices, cognition, and emotions (and ultimately also [subjective] feelings) poses somewhat of a challenge to the reader, who customarily has been prepared to view emotion talk as a reflection of underlying emotions and the ways these are subsequently "cognized". Nevertheless, what has come to the forefront in this paper is the suggestion to reconsider the constructive function of language in the formation of social constructs (such as emotions) which are being achieved (constructed) in development and interaction.


Bamberg, M. (1996a). Emotion talk(s): The role of perspective in the construction of emotions. In S. Niemeier & R. Dirven (Eds.), The language of emotions (pp. 209-225). Amsterdam, The Netherlands: John Benjamins.
Bamberg, M. (1996b). Perspective and agency in the construal of narrative events. In D. Stringfellow, E. Cahana-Amitay, E. Hughes, & A. Zukowski (Eds.), Proceedings of the 20th Annual Boston Conference on Language Development, Vol. 1 (pp. 30-39). Somerville, MA: Cascadilla Press.
Bamberg, M. (1997). Positioning between structure and performance. Journal of Narrative and Life History, 7, 335-342.
Bamberg, M. in preparation. Narrative identity.
Berko, J.B., & Ratner, N.B. (1993). Psycholinguistics. Fort Worth, TX: Harcourt Brace Jovanovich.
Basso, E.B. (1992). Contextualization in Kalapo narratives. In A. Duranti & C. Goodwin (Eds.), Rethinking context (pp. 253-269). Cambridge, England: Cambridge University Press.
Clark, E. (1993). The lexicon in acquisition. Cambridge, England: Cambridge University Press.
Davitz, J. (1969). The language of emotions. London: Academic Press.
Fletcher, P., & MacWhinney (Eds.) (1995). The handbook of child language. Cambridge, MA: Basil Blackwell.
Goddard, C. (1997). Contrastive semantics and cultural psychology: 'Surprise' in Malay and English. Culture and Psychology, 3, 153-181.
Harré, R. & Gillett, G. (1994). The discursive mind. Thousand Oaks, CA: Sage Publications.
Heelas, P. (1986). Emotion talk across cultures. In R. Harré (Ed.), The social construction of emotions. Oxford: Basil Blackwell.
Hoff-Ginsberg, E. (1997). Language development. Pacific Grove, CA: Brooks Cole Publishing.
Leff, J. (1973). Culture and the differentiation of emotional states. British Journal of Psychiatry, 123, 299-306.
Lutz, C. A. (1988). Unnatural emotions: Everyday sentiments on a Micronesian Atoll and their challenges to Western theory. Chicago, IL: University of Chicago Press.
Matsamuto, D. (1994). People. Psychology from a cultural perspective. Pacific Grove, CA: Brooks/Cole Publishing
Ochs, E. (1988). Culture and language development. Cambridge, England: Cambridge University Press.
Ochs, E. (1996). Linguistic resources for socializing humanity. In J.J. Gumperz & S.C. Levinson (Eds.), Rethinking linguistic relativity (pp. 406-437). Cambridge, England: Cambridge University Press.
Özçaliskan, S. (1997). Genre construction in emotional discourse. Unpublished Master of Arts Thesis. Clark University, Department of Psychology.
Schieffelin, B.B. (1990). The give and take of everyday life. Cambridge, England: Cambridge University Press.
Schieffelin, B.B., & Ochs, E. (1986). Language socialization. Annual Review of Anthropology, 15, 163-191.
Stein, N.L., & Levine, L.J. (1990). Making sense out of emotional experience: The representation and use of goal-directed knowledge. In N.L. Stein, B. Leventhal, & T. Trabasso (Eds.), Psychological and biological approaches to emotion (pp. 45-73). Hillsdale, NJ: Lawrence Erlbaum Associates.
Stein, N.L., Liwag, M.D., & Wade, E. (1997). A goal-based approach to memory for emotional events: Implications for theories of understanding and socialization. In R.D. Kavanaugh, B.Z. Glick, & S. Fein (Eds.), Emotions: The G. Stanley Hall Symposium. Mahwah, NJ: Lawrence Erlbaum Associates.
Stein, N.L., & Trabasso, T. (1992). The organization of emotional experience: Creating links among emotion, thinking and intentional action. In N. Stein & K. Oatley (Eds.), Cognition and Emotion (special issue), 6, 225-244.
Stein, N.L., Trabasso, T., & Liwag, M. (1992). The Rashomon phenomenon: The role of framing and future-orientation in memories for emotional events. In M.M. Haith, J.B. Benson, R.J. Roberts, Jr., & B.F. Pennington (Eds.), The development of future oriented processes (pp. 409- 436). Hillsdale, NJ: Lawrence Erlbaum Associates.
Trabasso, T., & Stein, N.L. (1993). How do we represent both emotional experience and meaning? A review of Richard S. Lazarus' 'Emotions and adaptation'. Psychological Inquiry, 4, 326-333.
Wierzbicka, A. (1992). Defining emotion concepts. Cognitive Science, 16, 539-581.
Wierzbicka, A. (1994). Emotion, language, and cultural scripts. In S. Kitayama & H.R. Markus (Eds.), Emotion and culture (pp. 133-196). Washington, DC: American Psychological Association.
Wierzbicka, A. (1995). Emotion and facial expression: A semantic perspective. Culture and Psychology. 1, 227-258.

Footnote: 1 1 Thanks go to the students who participated in this project, including: Andrea Berger, Sunil Bhatia, Ayden Reynolds, and Michelle Sicard.
Footnote: 2     2     And although we also have emotions as private individuals, with no one else around, self and other remain the unit around which 'selfhood' and 'otherness' are constructed.