ISSA Proceedings 2002 – A Non-Propositional Approach To Emotions In Argument
In this paper we try to show why it is inadequate to approach emotions within a pragma-dialectic or propositional approach to argument. We confine ourselves to the arguments related to language and discourse. We do not claim that pragma-dialecticians must disappear, what we claim is that the constant movement they try to make in favor of an expansion of their approach is illegitimate. Pragma-dialectics is only a theory of written and highly critical discourse, of computational logical analysis.
Van Eemeren and Houtlosser accept the importance of three levels of rhetorical strategies:
1. the selection of the material,
2. its adaptation to the audience and
3. its presentation.
But the way they analyze them is valid only in their framework. Their claim that we need to start “from the assumption that rhetoric may be considered to operate within a dialectical framework” (van Eemeren and Houtlosser 2000, 2) is not valid beyond the critical, propositional and artificial intelligence approach to argument.
I propose five interrelated lines of thought: the three cited by van Eemeren and Houtlosser and two more:
4. the presence of emotions in speech acts as a nuclear component furnishing motives for action; and
5. the necessity of studying emotions as enculturated phenomena. The difference is that for me the selection operation with which any argumentative discourse begins (the old inventio) is necessarily a relative cultural phenomenon. The relevance of the disposition of the arguments challenges the propositional approach. And the presentation of arguments shows us the inevitability of emotions in language. We cannot consider emotions as a minor component or as a mere accompaniment of the logical-dialectical component. Reason is not divisible; it encompasses the whole of logic, emotion, belief, value and intuition.
1. Emotions and the selection operation
Van Eemeren and Grootendorst (van Eemeren and Grootendorst, 1996) consider the topoï in their logical sense. They perceive that two operations occur in regard to topoï: the selection operation and the warrant operation. They do not care about selection and focus on the warrant role.
Van Eemeren and Houtlosser change the original pragma-dialectical view. They consider the inventio in order to achieve the optimal rhetorical result. The selected moves must be an effective choice made from the available potential moves. The moves must be in such a way adapted to the audience that they comply with auditorial demands.
We cannot discuss this option in the pragma-dialectician terms. What we want to argue is that if there is a warrant function of topoï it is because there was before or at the same time a selection operation. And this selection is rhetorical and is crucial to the argument in a sense that goes beyond the pragma-dialectical interest. If the selection is not well done, then the very possibility of convincing is ruled out, and we will not be able to work on the discourse. This selection still has a logic component but is deeply rooted in intuition and emotion, in what we feel is right and adequate. The conjecture of the neuro-physiologist Antonio Damasio (Damasio, 1994) is that making a decision is an activity guided emotionally by “somatic markers” that filter decisions. And, moreover, decisions are also a cultural and ideological election. And the emotional appeal of a selection can have dramatic consequences even in the most critical discourse, which is scientific discourse.
In argumentative discourse, in any discourse indeed, emotion is inevitable. Language and affect are correlates. We select topics and notions to talk or write about. They are affected by an emotional valence and have cultural-and language-specific resonances. Topics are enculturated and bring with them an emotional world. Let us imagine, for example, a declaration of war. We cannot imagine it without an emotional appeal to people. And this is not a fallacy, it is just a condition of a war discourse. The sole word “people” evokes emotional responses and they are in accordance with the language, the culture and the specific historical moment. In English, the word “people” refers only to human beings and is plural. In Spanish we differentiate between “pueblo” and “gente”. “Gente” has very low emotional appeal, while “pueblo” has a strong one. Moreover, “pueblo” is singular and can be contrasted with the plural “pueblos”. An appeal to “pueblos” has a greater emotional impact. It may be a reference to humankind. “Pueblo” and “pueblos” may refer to people, but they can also refer to the town. The amalgamation of people and town has an Indian, rural and communal resonance, with a different emotional appeal. And we can continue, contrasting this with an Indian culture with no specific term to refer to the people and talking about themselves as the “true human beings”.
Emotions are evoked differently according to the recipient. A right wing audience will reject in many cases an emotional appeal to the people. A left wing audience will be prone to reject an emotional appeal to people coming from a bourgeois discourse; it will be considered a matter of populism. The same audience will probably empathize immediately with an appeal to people made by left wing speakers. History matters in these respects. Let us imagine now an appeal to people by a functionary in a communist country after the soviet occupation of Prague. Let us imagine in contrast the appeal to people made by Charles de Gaulle during the “résistance” in France. The emotional appeal is opposite.
What the example of the people shows us is that there is no such thing as an abstract proposition in the arguments in the real world. Propositions are useful. They are a human construct to assess logical aspects of a discourse. Nonetheless discourse – even the logical component of a discourse – is not only propositional. Discourse also refers to schematizations of discursive objects. And both schematizations and propositions are tinted with emotion in the deepest level of their construction.
The Natural Logic of Grize and Vignaux tries to depict the cultural (and for us also emotional) aspects of argumentation schematizations. They call the very first instance of argumentation, the act of selecting a topic, a word, the anchorage operation (opération d’ancrage, Grize, 1973, 1974, 1982). To their work we must add the consideration of the pervasiveness of metaphor and metonymy, often appealing to emotions, also in an enculturated fashion. Of course, the pragma-dialectical approach tries to show us that there is a need to distinguish the “argumentation” component of emotions and the “elocutionary, decorative, belletristic Burkian rhetoric” (van Eemeren and Houtlosser, 2000, 3). Nonetheless, others are right now working in the exact opposite direction, trying to establish the continuity between the two branches of rhetoric and emotion in discourse: the rhetoric of inventio and dispositio, and the rhetoric of elocutio (Klinkenberg, 1996). Nietzsche (2000), as it is known, considers the eradication of emotion and metaphor as impossible, because they form part of the very essence of language.
2. Other affective components of language and discourse
The discursive appeal to emotions does not end with the anchorage operation. Any premise of a judgment that includes an evaluation may easily convey an emotional appeal. The borders between evaluation and emotion are quite tiny. Not all evaluations are emotional, but all emotions are evaluations. These problematic aspects of argumentation are studied by the Natural Logic analysis of schematizations. They are also analyzed by the French School of Argumentation (L’Argumentation dans la Langue). This school of thought (Ducrot and Anscombre, 1980) do not analyze emotions directly but their approach to “argumentative scales” is a good instrument to analyze the emotional components of a proposition. In the emotional linguistic components of a proposition or a schematization we have three operations taking place at the same time.
The first operation is a matter of ethos. If Michael Leff tells the argumentation theory community that “the pragma-dialectics treatment of emotions is absurd”, then the emotional appeal will be very strong in favor of or in opposition to Leff. If the same thing is said by Pedro Reygadas then it would surely generate a reaction against the speaker who has no recognition as an expert in the argumentation theory community. These words have an emotional appeal in the latter sense because they are recommendations to act or consider things in a certain way and this recommendation is affected by ethos.
The second operation is a matter of pathos and, again, of selection. If we say, “Pragma-dialectics treatment of emotions is wrong”, this is one thing. Quite another thing is to affirm, “Pragma-dialectics treatment of emotions is absurd”. Many terms imply a selection in a scale and this selection connotes logical-emotional aspects. For Ducrot (still within an ideal and propositional analysis of argumentation) the sense of an utterance cannot be taken seriously without the description of certain intentions. The logical-emotional aspects of a word selection allow us to follow or not to follow a certain conclusion. Certain utterances are characterized, within a given society, by the power we recognize in them to orient us towards certain conclusions (Ducrot 1980, 11). This orientation is logical and emotional. The linguistic dressing of propositions and the bias they impose on interpretation is logical and emotional.
We cannot talk about emotions in a proposition without considering modality in a broad sense. Modality has been studied as a matter of modal logic. This useful approach tells us the scope of the term and the conclusion considered. We can study modal terms in a mathematical perspective of the operators of necessity (□) and possibility (◊). We can study this in an argumentative way within the Toulmin model. Indeed, Toulmin’s interest in modality was crucial. He considered modals one of his important contributions to argumentation analysis, but his followers, so to speak, did not follow him in this respect.
Modal logic is useful, but does not exhaust modality. Modality pervades discourse as demonstrated by Charles Bally, a disciple of Ferdinand de Saussure. And modal terms are often emotional, not only intellectual (evaluative). Why do pragma-dialecticians not consider the modality approach to language and argument? We can guess that it is because it drives us toward subjectivity and the agents of argumentation.
Modality appears tinted with emotion in different dimensions. In many languages the emotion permits us to distinguish the modality of enunciation, whether we are in front of an exclamation or an imperative or declarative assertion. The proposition also has a certain modality, completely logical sometimes (probability, certainty, verisimilitude) but also appreciative and emotional (sad, happy, is a pity that). And the discourse (the message) has an inevitable presence of modality: of intonation, lexical and syntactical (“he is kind of an asshole”). Intonation and lexical modality are frequently emotional.
Intonation is pervaded by affect. We can read intonation affect (rising and falling tone, segment gemination, aspiration, voice quality, volume, speed, pitch) in context.
Even names, the apparently neutral class of words, can be emotional. If I say as a conclusion: “Well, he is a professor”, emotion is minimized. If I say “Well, he is a stupid” or “Well, he is a genius”, that may be highly emotional. The sign shows an ideological and emotional competence.
The use of adjectives in discourse is rooted in subjectivity. To proffer an adjective is to reduce multidimensionality of names to one dimension (Wierzbicka). Even so, adjectives may be relatively objective (male/female, single/married, white/yellow/black). Subjective adjectives may be evaluative, rooted in axiology (good/bad) or not (small/big). Evaluative adjectives easily become emotional in the appropriate context. Finally, adjectives may be directly affective: funny, pathetic, etc.
Verbs may also be modal, and some of them are emotional, like the ones of feeling: love, appreciate, wish, hope, want. There are adverbs of affection also, like badly or unfortunately. Evaluative adverbs like obviously, allegedly, plainly may be easily taken as emotional.
As we can see, to talk about emotions and language is not only to talk about the field of emotion words. The description of connotative emotional aspects of languages is an almost infinite task. In some cultures kinship terms are affect terms. The same happens in other languages with diminutives and other linguistic processes such as reduplication. Even negation may be subject to emotional manipulation. Emotional aspects are part of the description of exclamations, expletives, interjections, curses, insults, imprecations, onomatopoeias and ideophones (words whose phonological structure itself encodes meanings).
Of course the classification of words is relative. There is only a tendency of the referred words to be evaluative or emotional. What makes these words really evaluative or emotional is the whole proposition, the whole discourse, the whole context and the agents. This is what makes propositionalization possible. Terms are not fully emotional in themselves. It is the agents that are emotional when they produce an argumentative judgment or when they interpret it. When propositional theories exclude agents, they are allowed to exclude emotions. Nevertheless, do they study argumentative discourse or an obstinate fiction of the Western thought?
Not only words and intonation are affective. Code switching from English to Latin in academic discourse may be affective and formulaic discourse like proverbs and idioms may frequently express emotional components. Frames of conversation, for example in politeness, have emotional components and even laughter and weeping may be normative, ritualized and be a part of conversation in a given culture.
An additional problem for propositionalism is that emotion and schematizations are not isolated. Within an argumentative discourse there is a whole web of resonance of terms, both emotional and non-emotional. There is also a resonance of propositions. The impact of a judgement is also emotional and this impact rests not only in what is said by a set of premises and conclusion but by how the appearance of the conclusion is prepared. How many times does the word “people” appear in the declaration of war? How is it modalized? How many senses does it have? How do we discuss, avoid or attack what undermines our argument? These questions frequently raise emotional appeals.
3. The disposition of discourse
Rhetoric has shown for centuries that the opening and closing movements in a discourse are highly emotive. They are not only so, but they are crucial for convincing (we do not accept a rigid distinction between conviction and persuasion). Pragma-dialectics and all propositional dialectical theories of argument were not interested in discourse order. They just extracted the proposition out of the totality and analyzed it logically. Of course this can be done. This is always done, even in natural processes of arguing. Nonetheless, what we do when propositionalizing arguments is not to analyze the real arguments; we analyze logical phantoms. Pragma-dialecticians now realize this and they consider that the presentation of the moves must be discursively and stylistically appropriate. They also consider that invention (selection of the material), adaptation to the audience and presentation of the discourse must converge to be optimally successful. But, to accept these undermines the foundations of pragma-dialectics. The consideration of the rhetorical shift drives us towards the necessity of considering not just the judgment but the whole of a discourse, abandoning the restricted propositional approach to argument. The pragma-dialectical rules of reconstruction are severely questioned, especially permutation and deletion.
We place the arguments of a discourse one after another in different ways: according to cool numbers; presenting first the strongest argument; trying to begin in a calm way and then making stronger and stronger arguments to knock out the enemy. We repeat arguments according to different nuances, insisting on them. These strategies are rhetorical and are emotional. When we analyze propositions, all this is of no interest. But can we really affirm that the value or even the logical validity of an argument, resides in itself? Is there such a thing as an isolated judgment pending in the sky of logic?
The disposition of a discourse, as Plantin has showed us, is not only a matter of logic. An argumentative interaction is also structured by the emotional exchange development: the construction of confidence, the appearance of conflict, the feeling of treason, the rage, the apology and the reconstruction of the relation, for example. Emotions are the glue and the organizers of interactive arguments.
4. Emotions as motives for action
Here we are back to speech acts, in a new level of our discussion. Pragma-dialectics and all propositional approaches to argumentation write about speech acts privileging Searle instead of Austin. This seems to be an innocent decision. The idea for selecting him is simple: Austin proposed a theory of speech acts, Searle developed it and gave us instruments to analyze them in argumentative discourse. But this simple assumption is completely misleading.
Austin cannot be understood until a relation with Wittgenstein has been established nor without considering the Austinian project of attacking apriorism, determinism and the epistemological foundation of philosophy which he wanted to substitute with a logical-linguistic foundation.
1. For Wittgenstein (1999), the fact that language is action and refers to language games is universal. However, the games in themselves are relative and refer to the totality of the language-culture network. For Austin (Austin, 1970, 1975), language seems to be an addition of little systems and we cannot generalize anything but the kinds of acts, the types of acts (locutionary, illocutionary and perlocutionary) and the “felicity conditions” of them, always in a provisional way. For Austin the object of study is the agent in a context. Searle (1990) is a universalist. His project is the “logistization” of Austin, abstracting speech acts rules. He relegates the agents and creates metaphysics, as Habermas does. The question is, is it legitimate to pass from the types of acts to the rules?
Austin considers that every language has certain types of acts, and may have certain classes. What Searle considers is that if that is the case, then there must be something they share. He proposes that what is shared are rules. We think that this is simultaneously right and wrong. Right, because there must be a form underlying Speech Acts universal functioning, allowing us to know when the felicity conditions are fulfilled and at which level. Wrong, because Searle focuses only on the logical dimension, losing sight of the idea of language as a complex and enculturated phenomenon. Maybe we can use Searle’s rules if we complement them with other institutional and cultural facts. I am skeptical about the utility of Searle’s project, because it implies the loss of the agency of speech acts.
Another aspect to be clarified is that for Austin there is no limit to actions. For Searle there are universal restrictions. Austin talks about the limit of the actions we perform, whereas Searle talks about the underlying principles of actions. Perhaps we can reconcile the two positions in such a way that restricted rules can generate unlimited actions in context. Is this possible and adequate?
2. The theory of meaning in Wittgenstein and Austin is very different from that of Searle. For Austin the meaning is not a thing, it is metalinguistic; meaning is not meaning because it refers to something. The meaning, as Wittgenstein has shown, operates within the calculus of language. For Searle, predication becomes somehow independent. We lose the linguistic turn of philosophy.
Maybe in this case we need an intermediate and more complex position. First of all, we know by means of non-linguistic and non-cognitive ways. Wittgenstein accepts that we need to include emotions in our reflection. Our “lived body” as phenomenologists (Merleau Ponty, 1962) investigated, knows by means of motility, perception, emotion, sexuality, and maybe also by other inner states. The first move then is to change from the logical-cognitive-linguistic paradigm of knowing to a wider scope of knowledge. Within the linguistic and cognitive level, unlike Searle, we have to accept that we know everything through language and prejudice (in the Gadamer sense of it). But at the same time, we have to accept that there is something outside of language, even if we know it through language.
3. For Searle, to “understand” is a mental state, while for Austin understanding refers to the comprehension of the adequate actions to be performed. “To understand” is for Austin an ability (the nature of an action satisfies such and such criteria). It is not a matter of communication as with Searle but of behavior. For Wittgenstein to understand is related to being able to do something, to being empowered and to knowing. To understand is for him to know how to use the word, how to correct the improper uses based on known criteria.
Searles’ approach to speech acts drives us towards propositionalization and the logical components of argument (to what is true or false). Austin’s and Wittgenstein’s approach to language as action study the agents of argumentation. And from this point to considering the role of emotions for agents there is no gap. The study of agents of the speech acts allows us to study their emotions. This is not at odds with the Wittgensteinean approach. The German philosopher clearly stated (Wittgenstein, 1985, 182):
Let us consider the voice intonation, the inflexion, the gestures, as an essential part of our experience, not as non-essential accompaniment elements or as mere communication means.
And in all his latest work Wittgenstein defended the inseparability of reason and emotion. The agents are rational and emotional. Reason is emotional. Emotion is rational. And this pervasiveness of emotionality in rationality and argument cannot be well described in the framework of any propositional attitude towards argumentation theory. To really study emotions in argument we have to study the emotional agents and their paralinguistics: intonation, gaze, gesture, position, proximity, etc. Otherwise we will commit the same crime pragma-dialecticians committed when constructing their framework. We will go from a theory of speech acts and actions without agents to an emotional analysis of emotions without emotional agents. Actions without agents are not actions. Emotions without emotional agents are not emotions. It is time now to abandon the propositional approach to argument and definitely analyze argumentation as an enculturated phenomenon where agents and paralinguistics are placed at the very center of the discussion.
The dominant trend of pragma-dialectics left aside the real pragmatic approaches of Natural Logic, the French school of discourse analysis and the English and American traditions of communication, speech theory and interaction. In the American tradition of interaction analysis, for example, there is an analytical hierarchy that really drives us to the assessment of speech acts in argumentative discourse. This hierarchy varies from author to author and also the categories may vary but usually have certain core elements (Kerbrat-Orecchioni, 1998):
– The communicative situation, which is the whole of the interaction
– The sequence, which has a semantic unity
– The exchange, that has (in dialogue) two turns, one of each speaker
– The turn and, sometimes, the move or moves within a turn
– The speech act, the minimum unity of interaction
Pragma-dialectics may use this scheme, but will always leave aside the real agents of the speech act. The speech act is a unity of interaction. The agent makes logical and emotional sense of the discursive exchange. Emotions in the pragma-dialectical approach lose part of their very essence: being a motive for action.
Emotion may be analyzed by an Informal Logic model (e.g. Walton), by a Gricean model or even in Speech Acts’ theory. Gilbert has studied speech acts (for us “discourse acts”) as emotional. In this conception, the illocutive force may be an emotional one. In an analogy with linguistics, we may say that we do things by expressing emotions: frighten, accuse, accept, condescend, blame, etc.
Emotions not only transcend arguments and propositions but they also transcend language. In oral argumentation we only express what we feel in an approximate way. It is not a matter of propositions or even modality, it is the whole agent acting. In linguistic arguments we just can try again and again to express what we feel, approximating us more and more to what we feel throughout interaction in a way that has absolutely no relation to any possible modification of a pragma-dialectical or artificial intelligence model.
5. Emotions and languages
To admit that emotions have a role in argument is to admit – at least to a certain extent- the necessary relativistic approach to argumentation insistently and sharply rejected by pragma-dialecticians.
It seems inevitable to consider the cultural and social construction of emotions. There are basic emotional expressions that are related to the maintenance of a certain social position. There are emotions exclusive of a certain culture. Even similar emotions in Western cultures cannot be exactly mapped. The Spanish “nostalgia” has not the religious resonance of English “Sunday melancholy” or the meaning of the Portuguese “saudade” (considered almost as an emotional identity trait of Brazilians).
There are different rhetorics of emotion. Chinese complaint is somatic while in Western societies it is a speech practice. Emotions are expressed overtly and covertly. They are expressed verbally and in non-verbal ways.
To understand the other’s emotion requires sharing the basic experiences of a “form of life” that evoke a given feeling. We essay to comprehend each other, but it takes long time really to do it. It is within each culture and its emotion that we learn to define and negotiate social relations of ourselves according to cultural and practical ethical principles.
Cultures “hypocognyze” or “hypercognize” certain emotions (Levy, 1984). They give certain emotions a predominant role and decide to exclude some others.
Emotions cannot be adequately studied if we do not consider the social position of the agents. Equally important are the global ideological structures of the person, the gender and the whole story of interactions between people in politics or in an interpersonal argument. Emotional meaning systems reflect social relations. Emotions constitute social behavior. Emotions may support the caste structure of India or may be a way of presenting the self in non-agonistic ways in Java. Emotions may even be the equivalent of law, regulating interactions and the moral order, as occurs in the Ifaluk society (Lutz, 1998). Each class has different emotional register. A culture may repress an emotion like as in the case of sadness in Tahiti or may have an emotion unknown to other cultures as the Spanish historical “pundonor” and “gracia”.
The expression of emotions varies from culture to culture. Not all cultures conceive anger as a metaphor of a hot fluid in a container (Lakoff and Kövecses, 1987). This idea of anger fits with the hydraulic metaphor in Western theory of emotions. But, what about the pintupi aborigines and their idea of anger? For them, anger is dialectically related to compassion. Anger “leads beyond its particular construction to existencial and evolutionary issues of attachment and dependency”. For pintupi aborigines it is relevant to consider in anger both human evolution and the significance of hierarchy in a cultural semiotics that is far from American anger (Myers, 1988).
The way out for the necessary consideration of relative and specific patterns of emotion and emotional expression is to consider emotions as propositional attitudes. These emotional propositions are even susceptible to computational operations. Nonetheless, once again, is that really a reflection about the emotions in argument? Or is it just the leaking phantom of the emotions? Emotions in propositional analysis are thoughts seeped out of the essential fact of having an emotion, the understanding that I’m involved in something.
In our view, what we need is an approach to emotions outside the propositional attitude. The only thing that has human interest about emotions in argument is how agents conceptualize emotion. How they constitute emotion through their behaviors. How they make sense of emotions in argument in a “languaculture”.
Austin, John (1970). Philosophical Papers, 2nd ed. Oxford: Clarendon Press.
Austin, John (1975). How to Do Things with Words. Oxford: Clarendon Press.
Berlin, Isaiah et al. (1973). Essays on J.L. Austin. Oxford: Clarendon Press.
Burleson, Brant B. & Susan L. Kline (1979). Habermas Theory of Communication. Quarterly Journal. of Speech, V 65, 0033-5630.
Damasio, Antonio (1994). Descartes’ Error. New York: Putnam.
Ducrot, Oswald (1980). Les échelles argumentatives. Paris: Minuit.
Eemeren, Frans van & R. Grootendortst (1996). La nouvelle dialectique. Paris: Kimé.
Eemeren, Frans van, & Grootendorst, R. & Jackson, S., & Jacobs, S. (1994). In: Eemeren, Frans van y R. Grootendortst (Eds.), Studies in Pragma-Dialectics, International Center for the Study of Argumentation. Amsterdam: Sic-Sat Collection.
Eemeren Frans van et al (1996). Fundamentals of Argumentation Theory – a handbook of historical backgrounds and contemporary developments. Mahwah, New Jersey: Lawrence Erlbaum Associates Publishers.
Eemeren, Frans van & Peter Houtlosser. Rhetoric in pragma-dialectics, http://www.argumentation.spb,ru/2000_1/papers/I_2000pl.htm.
Eekman, P. & W.V. Friessen (1971). Constants across cultures in the face and emotion. Journal of Personality and Social Psychology 17 (2): 124-29.
Eibl-Eibesfeldt (1980). Strategies of social interaction. In: R. Plutchik and H. Kellerman (Eds.), Emotion: Theory, Research and Experience (pp. 57-80). New York: Academic Press.
Grize, Jean Blaise (1973). Logique et discours pratique. Revue Communications, 20. Paris: Seuil.
Grize, Jean Blaise (1974). Argumentation, schématisation et logique naturelle. Revue Européene des sciences sociales, 32, XII. Geneva: Librairie Oroz.
Grize, Jean Blaise (1982). De la logique a l’argumentation. Geneva: Librairie Oroz.
Habermas, Jürgen 81976). Some Distinctions in Universal Pragmatics: A Working Paper. Theory and Society 3.
Habermas, Jürgen. Theories of Truth, trans. Richard Grabau (unpublished manuscript, Purdue University). Cited in Burleson and Kline (1979).
Habermas, Jürgen (1989). Teoría de la Acción Comunicativa, Buenos Aires: Taurus.
Kerbrat-Orecchioni, Catherine (1998). Les interactions verbales, 3 vols. Paris: Armand Colin.
Klinkenberg, Jean Marie (1996). Sept leçons de sémiotique et de rhétorique. Tornoto: Du Gref.
Lakoff, G. and Z. Kövecses (1987). The conceptualization of anger in American English. In: N. Quinn and D. Holland (Eds.), Cultural models in language and thought. Cambridge: Cambridge University Press.
Levy, R.I. (1984). Emotion, knowing and culture. In: R.A. Schweder and R.A. Levine (Eds.), Culture Theory, Essays on Mind, Self and Emotion, Cambridge University Press, Cambridge.
Lutz, Catherine (1998). Unnatural Emotions –Everyday Sentiments on a Micronesian Atoll & Their Challenge to Western Theory. Chicago and London: University of Chicago Press.
Merleau-Ponty, Maurice (1962). Phenomenology of perception, trad. Colin Smith. Londres: Routledge.
Meyer, Michel (1986). De la problématologie –philosophie, science et langage. Bruxelles: Pierre Mardaga.
Myers, Fred R (1988). The logic and meaning of anger among pintupi aborigines.
Nietzsche, Friederich (2000). Escritos sobre retórica. Madrid: Trotta.
Plantin, Christian (1998). Les raisons des émotions. In: Forms of argumentative discourse, M. Bondi (Ed.). Bologna, University of Bologna.
Searle, John (1990). Actos de habla – ensayo de filosofía del lenguaje. Madrid: Cátedra.
Wittgenstein, Ludwig (1985). Preliminary Studies for the Philosophical Investigations as the Blue and Brown Books. New York: Harper and Row.
Wittgenstein, Ludwig (1999). Investigaciones filosóficas. Barcelona: Altaya.