A Pragmática costuma ser definida numa perspectiva cognitiva como o estudo do uso da linguagem. As interpretações não são feitas mecanicamente de acordo com regras cognitivas, mas a partir de princípios contingentes que implicam também escolhas indeterminadas quando o contexto não permite qualquer tipo de desambigüização.


From a cognitive perspective pragmatics may be briefly defined as the study of language use. It is not surprising then that interdisciplinary studies are fundamental in any attempt at ordering mechanisms and motivations behind conscious or unconscious choices and effects of utterances. Psychology and cognitive perspectives have been involved all along. In relation to psychiatric involvement the name Bateson stands for pragmatics as a general program aimed at explaining pathological human behaviors including verbal communication. In Pragmatics of human communication: a study of interaction patterns, pathologies, and paradoxes (1967)[1], Watzlawick, Bavelas and Jackson – belonging to the Batesonean school – , state some definite conclusions about argumentative utterances in schizophrenic language. According to them, double binding or paradoxical injunctions are pathogenic verbal behaviors often observed in schizophrenics, and also generally found in other types of psychoses.

Paradoxical injunctions are certainly not contradictions. In the case of contradictions, a choice between two possible interpretations may be made, whereas double binding makes such choice impossible. Consider, for instance, the following statement made by a young schizophrenic woman when she bounced into a psychiatrist’s office for her first interview (in Watzlawick et al., 1967)[1]. The woman cheerfully announced:

[1] My mother had to get married and now I am here.

According to the authors, there would be four possibilities to paraphrase this statement:

  • she was the result of an illegitimate pregnancy;
  • the fact caused her illness;
  • her mother cannot be blamed for being forced into the marriage due to social pressure;
  • such facts had driven her crazy as she will be eternally indebted to her mother for giving birth to her.
  • The impossibility of choosing one interpretation among the four would confirm a double binding phenomenon. This seemingly unquestionable hypothesis fails first when considering that arguments in general can be transparently interpreted. A single utterance does not bring all necessary elements for judging it as true or false or convincing the interlocutor. Secondly, it disregards the nature of the subjects included in the communication context. The psychotic does not recognize contradictions and double binding, while the other, being non-psychotic, judges the statement as a paradox despite the fact that the young woman may be effectively explaining why she had to go to a psychiatric interview. Basically the difference cannot simply arise from the existence (or not) of mental disease itself, or from singular life histories and idiosyncrasies; it clearly arises from appropriation patterns of language which make argumentation possible. The phenomenon remains intriguing as language is shared, language meaning a set of rules and significant forms, and not, as we shall see later, the intrinsic possibility of an individual to be a subject and establish pragmatic ties.

    Considering argumentation in psychosis as double binding without statements including the relations between subjects may lead us to false conclusions such as the ones drawn by the cited authors:

    (i) psychotics behave as if they would avoid commitment by not communicating

    (ii) psychotics behave as if they tried to deny that they are communicating;

    (iii) psychotics want to communicate without accepting the commitment inherent in all communication.

    Could we say that the patient was not committed to her utterance? How can we deny the fact that the argument to explain her visit to the psychiatrist was a real attempt to communicate her truth and to manifest the reason why she believes she was taken to the psychiatrist? To what extent can we be certain that we can choose a precise interpretation for a given argument?

    Elsewhere (Novaes 2000[2], 1999a[3], 1999b[4], 1999c[5], 1998[6], 1997a[7], 1997b[8], 1997c[9], 1996a[10], 1996b[11], 1995[12]) I have presented the results of my research on speech acts in psychosis, which confirm that psychotics take turns in verbal communication appropriately as if they were affected by interpretative constraints. Thus they behave as if they were committed to communicating. It seems too that they try to fit arguments to these constraints. To a certain limited extent their arguments are constrained by general principles of pragmatics and they are cooperative and polite. Hence, they do not deny that they are engaged in communication. They try to answer according to the interlocutor’s question despite the fact that they do not change their points of view. They act as if their arguments were unquestionably true and not dependent upon paradoxical effects. Therefore, if we consider such arguments as double binds derived from non-choice of interpretation, we will be eliminating the whole process of interacting, which is the basis for argumentation.

    According the Watzlawick et al. (1967)[1], a phenomenon remains unexplainable as long as the range of observation is not wide enough to include the context in which the phenomenon occurs. Part of the context includes two subjects engaging in language by different links to each other and to the world and this seems to be ignored when considering double binding as a result of a judgement of only one of the participants. A central feature of most psychotics is precisely their commitment to the truth, which definitely anchors their speech. There is nothing in the young lady’s arguments about being ill or being there to induce us to affirm she has no commitment to communication. She advances an answer already implicit in being there in the psychiatrist’s office proving she is not rejecting communication.

    By its own definition, argumentation implies commitment to a truth. Because the psychiatrist cannot give an interpretation, her argument may as well not be considered paradoxical. Her argumentation cannot be conceived simply as a product of double binding. The implied circularity and autonomy of psychotic argumentation lead us to positing that any cognitive approach undertaken to deal with psychotic speech is based on questionable premises that disregard differences between psychic structural patterns for psychotics and non-psychotics.

    The attempt to use structural patterns to explain what is going on in psychotic language is not uncommon. Gernsbacher, Tallent and Bolliger (1999)[13] argue that the Structure Building Framework is a model of the general cognitive process and of the mechanism underlying discourse that can be used to account for these phenomena. According to the Structure Building Framework, the goal of comprehension is to build coherent mental representations or structures. Psychotics in general exhibit verbose disordered discourse as they have inefficient suppression mechanisms to lay a foundation for mapping subsequent information. When producing discourse, one would expect hyper-activation of semantically related information. As a consequence, irrelevant information may remain activated contributing to the production of disordered discourse features. When asked why he thinks people believe in God, one psychotic answered:

    [2] Uh, let´s see, I don’t know why, let’s see, balloon travel. He holds it up for you, the balloon. He don’t let you fall out, your little legs sticking out down through the clouds. He’s down to the smoke stack, looking through the smoke trying to get the balloon gassed up you know. Way they’re flying on top that way, legs sticking out, I don’t know, looking down on the ground, heck, that’d make you so dizzy you just stay and sleep you know, hold down and sleep there. The balloon’s His home you know up there. I used to sleep outdoors, you know, sleep outdoors instead of going home. (Chapman and Chapman, 1973[?] in Gernsbacher, Tallent and Bolliger, 1999)[13]

    Such cognitive model of normative discourse to explain cognitive processes in psychoses appears to be inefficient. For somebody to exhibit generalized cognitive deficits it is suggested that a variety of general cognitive processes may be dysfunctional in the disorder. So the only conclusion is that they have speech disorders; such statement, however, had already been reached without a cognitive model. Thus there is no theory to explain the difference, since dysfunctional disorders do not always occur in every psychotic and some non-psychotics may also present such impairment.

    The crucial point seems to be the nature of the subject who utters any argumentation. The pragmatic links to the others seem to depend on a structural means of representing objects. Pragmatics has no theory of the subject that can be implied from its approaches to explain argumentation as a cognitive product. Either a normative and cognitive approach is used to explain deficit or it is admitted that something is pathological because it is judged like that by one participant in the communication. The basic question to answer is whether someone convinces someone of something. Who convinces whom? That is the pertinent aspect to explain argumentation. For that reason I would like to explore some theoretical formulations found in the psychoanalytical theory to explain differences in the subject’s relation to speech and truth.

    In a sense, I partially agree with Watzlawick, Bavelas and Jackson (1967)[1] on considering psychosis in communication. It is precisely in the characterization of psychosis that we may encounter speech strictly related to the subject. Freud located the difference between neurosis and psychosis in the nature of the representation of things and words. One relevant point in such formulation would be the need to represent states and selves in speech. In Lacan’s proposal such representation demand goes further and he provides us with an explanation of psychosis as a specific psychic structure and as a speech phenomenon occurring between two subjects. Since communication is an interchange of meanings between subjects, I think that it is theoretically relevant to define differences and similarities between both psychic structures. If we disregard this aspect we may be caught in a tautological conclusion that something is pathological because it is double binding or the argumentation is a double bind because the subject is psychotic.

    The role of psychic structures in producing meanings during verbal interaction is decisive to define argumentation as a pragmatic act between two participants committed to convincing each other of a truth. Evidently, a description of psychotic structure in itself is not sufficient to explain paradoxes found in communication. The crucial theoretical point may be to develop a hypothesis pointing out why psychotics cannot judge their arguments as false since the effects caused in the others do not trigger any change in the argumentation itself. Requests for restating arguments considered contradictions or double bind are void. They neither doubt their certainties nor recognize that the other may have different meaning patterns for interpretation. As a consequence, I understand that different psychic structures are intrinsically related to the processes of argumentation in production as well as in persuasion.

    As the raison d´être for drawing on psychoanalytical theory in a pragmatic approach to argumentation in psychoses, I would like to argue that psychotic language cannot be explained without a theoretical formulation of its underlying specific psychic structure and, consequently, of its own signification mechanisms and motivations. It is my contention that the three psychic functions that link subjects to others and to the world, proposed by Lacan (the Symbolic order, the Imaginary order and the Real), directly determine argumentative statements. The function of the Imaginary order is specifically that of constructing phantasies during various stages of psychic development. Lacan´s view of the self as an alienated self corresponds to the internalization of the other through identification. The child’s release from his/her alienating image will occur through his/her appropriation of language from the Other, which is his/her means of entry into the Symbolic order. Since the Symbolic, the Imaginary, and the Real co-exist and intersect in the subject – the Real is not synonymous with external reality, but rather with what is real for the subject – any change in one order will have repercussions on the others.

    The Symbolic is the primary order, since it represents and structures the other orders. A symbol in the traditional sense is not necessarily part of the symbolic order nor is an image necessarily part of the imaginary, since these terms define functions rather than the elements in these functions. The emphasis of the structural view is upon relationships rather than upon objects. What is incompletely symbolized must logically appear in the Real (the domain outside symbolization). Psychotics indeed lack symbolization of objects they could not deal with in the past. According to Freud (1894), psychosis involves a means of defense against incompatible ideas that cannot be symbolized. The whole truth is internally abolished and projected outwards as another truth that constructs the argumentation to convince others and themselves. Thus, as there is incomplete symbolization, the proliferation of objects in the Imaginary order will prevail in utterances. Evidently, that proliferation of objects will determine the impossibility of making pragmatic ties.

    The symbolic function seems to fail in causing a means of anchoring psychotics’ appropriation of language in order to be engaged in interpretative restrictions imposed by the other in dialogue. To attribute meanings is to put forward a judgement of paradox coming from the symbolic order, from its structuring effect on the imaginary. Therefore, interlocutors of psychotics cannot make responsible evaluation of argumentative points. Argumentation may contain unexpressed premises underlying certain ways of interpreting, for example, the reason why they are taken for treatment, and this standpoint appears normally connected to other points of argumentation.

    To further develop my arguments, I will analyse some fragments of an interview between a psychiatrist and a young woman named Oleana. The fragments were chosen in order to illustrate how the arguments are maintained in several sequences despite the effects on the psychiatrist’s speech. Basically, I shall demonstrate that to make some arguments Oleana names herself differently and denies her mother and father, the figures who normally name children. Her illness is also denied, but it is explained as a consequence of brainwashing. As she does not recognize herself as Oleana, but as someone else named Nea, she states that she was not born in the family some attribute to her. She says she was found in a medical laboratory in Sweden.

    I have here attempted a close transcription and translation of her speech (D. is the psychiatrist and P. is Oleana):


    D: What is your full name?

    P: My name is Oleana Silva Gomes, my brainwashing name, because I have been brainwashed four times

    D: So tell me how was that?

    P: It´s-it´s-it´s-my real name is Nea I came from Stockholm Sweden [pause] to Rio de Janeiro

    [como é o seu nome todo?]

    [meu nome é Oleana Silva Gomes o meu nome de lavagem cerebral porque eu sofri quatro lavagens cerebrais]

    [então conta prá mim como foram as lavagens?]

    [é é é meu nome verdadeiro é Nea eu vim de Estocolmo, Suécia [pausa] pro Rio de Janeiro.


    D: So tell me something that … I would like to know, well, about your family uh who are the people in your family that are your friends?

    P: Well look my family, there in Stockholm, I do not even have a family I was born … no … medical lab and it was – and - and- it was there, yeah I found myself [pause] they found me there in a medical lab in Stockholm

    D: Did they?

    P: Yes

    [e me diga uma coisa que f-, eu queria saber assim da sua família, é quem são as pessoas da sua família quem são seus amigos]

    [é olha a minha família lá de Estocolmo, eu não tenho nem família eu nasci … não … laboratório de medicina e foi e e foi lá é me encontrei, [pausa] me encontraram lá não laboratório de medicina de Estocolmo]




    D: And who else lives with you?

    P: Who else lives with me?

    D: Right

    P: It is … the … the mo - my mother who says that she is my mother Dalva, Alfredo, he is Silva Gomes already dead, brainless, Maria harassed the guy so much that he died brainless

    D: Then who is this … this Alfredo?

    P: It was the-the-the father it was was Dalva’s husband, there is also Fabiana

    D: But why? But …why does she say that she is your mother? Isn’t she your mother? You told me this – she says she is my mother – so is she not your real mother?

    P: No, she is not my mother I came from Sweden from Stockholm they are lying the story to me I am blond with blue eyes I can see myself with-with hair yeah yeah …

    D: Brown

    P: Brown and-and-and with-with a-a and brown eyes but I am blond with blue eyes

    [e quem mais mora com vocês?]

    [quem mais mora comigo?]


    [é… a … a mã- a minha mãe que diz que é a minha mãe Dalva o Alfredo é Silva Gomes já morreu descerebrado a Maria desgraçou tanto o cara que ele morreu descerebrado]

    [mas esse quem era, esse Alfredo?]

    [era o a o pai, era o o o marido da da dona Dalva, tem a Fabiana]

    [mas por quê? mas por que que ela diz que é sua mãe? ela não é sua mãe? você disse prá mim assim ela diz que é minha mãe, mas ela não é sua mãe de verdade?]

    [não, não é minha mãe eu vim da Suécia de Estocolmo eles tão mentindo a história prá mim eu sou loira dos olhos azuis eu estou me vendo é é com com os cabelos é é é é é]


    [castanhos e e e com com um um e os olhos castanhos mas eu sou loira dos olhos azuis]


    Throughout the interview, it may be observed that Oleana is coherent in constituting herself according to some imaginary ties as:

  • Nea is her real name
  • She is blond with blue eyes
  • She was found in a medical laboratory in Sweden
  • She came from Stockholm to Rio de Janeiro
  • She underwent brainwashing
  • Alfredo is not her father
  • Dalva is not her mother
  • She has no family.
  • These imaginary ties do not have pragmatic effects, even though the psychiatrist attempts to destroy her arguments when he insists and asks about her family. By doing that she might be forced to face the fact that she only has the one name given by her parents. However, all his attempts fail. If she changes one argument all argumentation may be destroyed. Possibly what is avoided is the symbolic representation of the family and her relation to each of its members. Thus she does not belong to the Silva Gomes family; they are lying to her; they chase her having her changed her name after brainwashing. What Oleana wants to communicate with those arguments nobody can risk interpreting. Any attempt at her remaking her arguments is void. She has no other argumentation to convince us that she is not ill as we can not convince her that she is ill.

    I have been observing her in a psychiatric institution since 1991. In 1996, I met her again in another institution and she was repeating the same argumentative points of view. The question for pragmatics is not to propose an interpretation and in this sense double binding is a false judgement to any argumentation. Double binding presupposes two symmetrical subjects with the same communication competence to judge on equal grounds what is convincing or not. To choose one interpretation for Oleana’s argumentation about her identity and her illness would be that of a communicative illusion, which directs the search for univocal nature in communication. The question for pragmatics is how to consider indeterminacy in judgments of psychotics’ arguments in its theory since only one of the participants can recognize double bind. The status of mental pathologies cannot be decisive to define argumentation in pragmatics.

    Lacan’s view of the loss of reality in psychosis is therefore that of a loss of symbolic reality. In the widest sense, psychotics’ difficulties in relating to people around them would be the crucial reason why we may be unconvinced by their arguments. If the first young woman (example [1]) states that she is there in the psychiatrist’s office because her mother had to get married, this may be judged as a real truth to which she is committed even if it seems to us complete non-sense. She also cannot be said to be avoiding communication. Similarly, if the man in the second example gives other reasons to justify why people believe in God, his argument cannot be judged as double bind or disordered.

    To conclude, an investigation of argumentation simply as a product submitted to one of the participant’s judgments in communication does not come to grips with the complexity of pragmatic phenomena. Cognitive approaches are sustained by the assumption that participants in communication can equally judge utterances and remake their arguments in order to convince one another of the true choice. When viewing pragmatics as a general social and human perspective on language use, an additional dialogue may be established with theories that consider differences in the way subjects get related to others and to the world. Such a perspective implies that the choices to interpret utterances are not made mechanically or according to cognitive rules, but on the basis of highly flexible principles that imply also indeterminacy of choices when context does not allow for restatements of any kind.


    1. Disordered discourse in schizophrenia described by the Structure Building Framework Gernsbacher M A, Tallent K A, Bolliger C M. Discourse Studies.1999;:355-372.
    2. Os discursos paralelos nas entrevistas psiquiátricas: a violação do princípio de cooperação entre médico e paciente Novaes M. (a sair nos Anais do VII GELNE).2000.
    3. Clinical interview in psychiatry: a specific genre in pragmatics Novaes M. Actas del Simposio del VI Simposio Internacional de Comunicación Social.1999a.
    4. Novaes M. CD-rom da 51a. reunião anual da SBPC.1999b.
    5. Terapias de valorização da palavra: identidade e linguagem Novaes M. Conferência apresentada na Jornada de Saúde Mental do Norte Fluminense..1999c.
    6. A heterogeneidade constitutiva no discurso psicótico Novaes M. Caderno de Letras da UFF.1998;1:57-73.
    7. A linguagem como fator de diagnóstico nas esquizofrenias Novaes M. Cadernos do IPUB.1997a;5:127-148.
    8. O mito da comunicação: os efeitos de estranhamento dos dizeres nas psicoses Novaes M. Revista do Instituto de Letras da PUCCAMP.1997b;16:75-85.
    9. Sobre a homonímia: a linguagem nas estruturas neurótica e psicótica Novaes M. Revista do Departamento de Psicologia da UFF.1997c;9:73-82.
    10. Os dizeres nas esquizofrenias: uma cartola sem fundo Novaes M. São Paulo: Escuta; 1996a.
    11. O equívoco: a senha de entrada do sujeito do inconsciente no dizer Novaes M. Anais do XLIV Seminário do GEL.1996b;:372-378.
    12. O dizer coisa com coisa: o efeito de coerência do texto produzido pelo outro Novaes M. Anais do IV Congresso da ASSEL-Rio.1995;:100-105.
    13. Communication: the social matrix of psychiatry Ruesch J, Bateson G. W.W. Norton & Company; [1987(1951)].
    14. Pragmatics of human communication: a study of interactional patterns, pathologies, and paradoxes Watzlawick P, Bavelas J. B, Jackson D. W.W. Norton & Company.; 1967.
    15. The language of change: elements of therapeutic communication Watzlawick P. W.W. Norton & Company.; 1978.