From: Kerstin Fischer <kerstinfuni-bremen.de>
Subject: Context as Other Minds
AUTHOR: Givón, Talmy TITLE: Context as Other Minds SUBTITLE: The Pragmatics of Sociality, Cognition and Communication PUBLISHER: John Benjamins YEAR: 2005 Announced at http://linguistlist.org/issues/16/16-2353.html
Kerstin Fischer, University of Bremen
The main point of the book, as I understand it, is to show that speakers constantly take into account their interlocutors' supposed mental models. Givón argues from an evolutionary point of view that it makes sense to constantly monitor the other's mental states because in this way the behaviours of others can be predicted, which has an adaptive value. He suggests that we know other minds because we know our own minds, and by feature association we can transfer knowledge of ourselves to the others, which makes sense in a society of intimates, in which we are taken to have lived until about 8000 years ago. Givón holds grammar to exhibit the function to anticipate or influence the others' minds in many respects. Grammar is thus proposed to have evolved as a perfect adaptation to our need to induce others to comprehend what is in our mind.
Chapter 1 discusses the notion of context and context-dependency on the basis of ''recurrent themes'', such as relevance, analogy and metaphor, and from a historical perspective, illuminating the relationship between categorisation and context in works by Plato, Aristotle, Kant, Peirce, and Wittgenstein, among others.
In chapter 2, Givón lays the foundations for the rest of the book by establishing particularly three important concepts:
Firstly, he argues that organisms in order to survive need to represent most tokens of the same type in the same way and to recognize deviant tokens as exceptional (p. 39). That is, an organism needs to develop context-sensitive adaptive responses. He argues that the tension between typical examples and exceptions is mirrored in the core vs. periphery distinction, the discrete vs. gradual distinction associated with Plato and Wittgenstein respectively, generative grammar vs. emergent grammar, and logical AI vs. semantic networks. Moreover, it corresponds to the rapid, repetitive, robust processing of the bulk versus the slow and error-prone contextual discrimination, which however is taken to have a high adaptive value.
Secondly, he introduces the mechanism of feature association as a form of abductive reasoning (p. 59).
Thirdly, he argues that until recently (from an evolutionary point of view) we have mostly been living in societies of intimates in which we shared the cultural, situational and individual background with our interlocutors. By feature association we thus construct the others' minds as like our own, as a prototype-based category (p. 62).
In chapter 3, Givón outlines his model of communication and discusses metaphor in some detail. The cognitive representation system is divided into the lexicon, propositions, and discourse. Communicative codes are divided into sensory-motor and grammatical codes. Sensory-motor codes comprise phonetics, phonology, and neurology, where the phonetic/phonological codes encode the lexicon. Grammatical codes (which should in principle also be encoded phonologically) encode discourse coherence, which is equated with communicative intention (p. 69).
The chapter continues with a discussion of metaphor, especially of Lakoff's notion of conceptual metaphor. Givón criticises that conceptual metaphors are identified out of context, whereas the felicity of metaphors crucially depends on the serendipity of the context (p. 75). In contrast, he shows how conceptual metaphors can be activated in discourse (p. 81).
Chapter 4 addresses in more detail the notion of the other mind. On p. 91, Givón shows how other minds are constructed by feature association from self. He continues by exploring the three cognitive representation systems introduced in the previous chapter: semantics/lexicon, grammar, and discourse. The lexicon in this view corresponds to what is generically, culturally shared. Grammar represents the particular interlocutors' mental models at particular times that constitute conventionalized, i.e. grammaticalized, common, recurrent, and adaptively relevant types of contexts (p. 92).
These three kinds of representation are taken to correspond to three kinds of memory (p.101):
shared generic network (lexicon) = permanent semantic memory shared speech situation = working memory/attention shared current text = early episodic memory
Givón then discusses the issue of consciousness, which brain areas are most likely involved and what kind of consciousness is necessary for a representation of other minds.
He argues that the ability to forecast the behaviour of others is the most important adaptive capacity for a social cooperating species (p. 120) and that ''the systematic on-line construction of mental models of the current epistemic and deontic states of one's interlocutor is the central adaptive motivation for the evolution of grammar'' (p. 121).
Chapter 5 is concerned with referential coherence, i.e. with the grounding (strangely without any reference to Clark's work) of referents in the universe of discourse, which is equated with working/episodic memory. He shows how anaphoric and cataphoric linguistic devices are used to indicate continuity versus discontinuity with previously grounded referents (pp. 136-143) and ''to anticipate the epistemic mental states of the interlocutor'' (p. 133) respectively.
Chapter 6 addresses epistemic and deontic modality, tense, aspect, and evidentiality. ('deontic' is taken to mean ''matters of desirability, preference, intent, ability, obligation, manipulation or power'' (p. 149)) Givón concludes that these ''propositional modalities'' display a ''fine- tuned sensitivity on the part of the speaker to the informational and social reality around them, most conspicuously to the constantly shifting epistemic and deontic states of their interlocutors'' (p. 177).
Chapter 7 is concerned with discourse. According to Givón, discourse coherence is established by the grammatical cues discussed in chapters 5 and 6, which together constitute ''an elaborate system of cues that speakers give hearers about highly specific mental structures and operations'' (p. 193).
Chapter 8 is an essay in the philosophy of science. It is related to the rest of the book by means of the question: who is the scientist's relevant interlocutor? The answer is of course the community of scholars (p. 196). First Givón discusses the two positions of deductivism and inductivism. Against this background, he establishes ''the pragmatics of empirical science'', by which he understands contextual, abductive reasoning (p. 205). Abductive reasoning can be illustrated by the following example (p. 207):
a. Puzzling facts F are incompatible with theory T, b. But facts F are fully compatible with Hypothesis H, c. whose truth value is yet to be determined. d. That is, if Hypothesis H were the case, e. then facts F would be explained as a matter of course. --------------------------------------------------------- f. Therefore Hypothesis H must be the case.
Abductive reasoning is taken to be contextual because it means to place ''erstwhile disparate facts in a wider context'' (p. 209). In his model the community of scholars plays important roles, for instance, regarding their ''deontic commitment to the status quo'' or as ''deontic resistance against the new hypothesis'' (p. 218), or regarding criteria of falsification (p. 219).
Chapter 9 concerns the concept of self. Givón argues that until recently (from an evolutionary point of view), an essentialist theory of self was ''highly predictive and adaptive'' (p. 223). He discusses different perspectives on the self and argues that schizophrenia and autism are caused by problems with the controlling self (p. 234).
Chapter 10 addresses the role of 'other minds' in martial arts, which is peculiar since the 'interlocutor' here is an adversary. The chapter can be read as an introduction to Confucian thinking and as a defence of Tai Chi Chuan as a martial art which ''as a discipline founded upon the presence of a real, on-line adversary-interlocutor, whose constantly- shifting epistemic and deontic states, and concomitant actions, must be anticipated'' (p. 254) should be practised as a martial art (and not as a physical exercise).
Givón's 'Context as Other Minds' is certainly an interesting book. However, I would think that in order to answer the question whether context consists of representations of other minds, we need to invoke evidence on whether speakers really build up such representations, what these representations consist of (especially if speakers cannot infer the others' mental models on the basis of their own), and under which conditions speakers make use of these representations.
Recently, much work has been devoted to the question whether speakers build up models of their interlocutors, what these models consist of, whether speakers display attention to this kind of information and how interlocutors contribute to it. Thus, many sources of information are available to the researcher interested in the questions raised by Givón. In particular, there is currently a lively discussion in psycholinguistics on whether speakers rely on knowledge about their hearers and under which conditions (see Schober & Brennan 2003 or Horton & Gerrig 2005 for overviews). Moreover, there is conversation analytic work showing participants' orientation to subtle cues with respect to information processing and topic development (e.g. Gardner 2001). Furthermore, there are numerous linguistic studies on how people talk to communication partners who are in some ways different from the speaker, such as children, retarded adults, or even computers. Finally, work in pragmatics has shown that speakers do not only orient to epistemic and deontic modality, as Givón suggests, but that they also model their interlocutors' attitudes and evaluations in their utterances (for discussion, see Nemo to appear).
Givón does not discuss any of the above. Of the more than 500 references, only 48 were published within the last five years, 30 of which are neurolinguistic or psychological papers and seven the author's own recent work. Only about five references concern recent work in linguistics or pragmatics.
However, even though Givón does not enter a detailed discussion with 'the community of scholars', there are many interesting things the book does do. In fact, Givón himself presents us with a framework in which to locate his scientific work (ch. 8).
The evidence presented in the book rests on arguments from evolutionary linguistics and on detailed analyses of grammatical features (especially ch. 4-7), which show that grammar can be interpreted as having the function to anticipate and influence the interlocutors' mental states, or, in Givón's words:
''a dedicated signaling system whose purpose is to induce others to comprehend what is in one's mind. Not only to comprehend, but hopefully also to spring into relevant action. Such behavior is inconceivable without a running on-line mental model, however subconscious, of the interlocutor's rapidly shifting intentional and epistemic states.'' (p.120)
Givón's argument can thus be understood as abductive hypothesis formation:
a. there are puzzling facts F on the function of grammar and on the tasks involved in social behaviour, b. facts F are fully compatible with Hypothesis H, that speakers subconsciously model their communication partners' mental states, c. whose truth value is yet to be determined. d. That is, if Hypothesis H were the case, e. then facts F would be explained as a matter of course. --------------------------------------------------------- f. Therefore Hypothesis H, that speakers subconsciously model their communication partners' mental states, must be the case.
Givón not only succeeds in supporting this line of argumentation very well, he also demonstrates the ubiquity of abductive reasoning and contributes thus to clarifying the kinds of evidence linguistic theorizing may rely on.
The book furthermore comprises many shorter discussions to which I cannot do justice here. One seems to be to clear Aristotle's name of the suspicion of being objectivist (p. 80), another one is to argue for the adaptive value of prototype-based reasoning (ch.2), and a third and fourth ones concern psychiatric diagnoses of mental disturbance (ch. 9) and Confucian philosophy (ch. 10) respectively. These discussions are not always as thorough as they could be; for instance, during Givón's invoking of Freud's model of self, only id and ego are discussed, super-ego has been dropped, although the super-ego would be a plausible candidate for the representation of others in our mind. Also the discussion of prototype theory lacks the depth of, e.g. the one by Kleiber (1998). Moreover, I would like to cast doubt on the usefulness of a discussion of ''relevant themes'' in pragmatics that, for instance, devotes five lines to the notion of relevance (p. 8). However, the discussions are certainly thought-provoking.
Regarding formal matters, the volume is a treasure for researchers working on garden path sentences, and it is sloppy with respect to references (e.g. Firth instead of Frith (p. 232), Fussell & Kreuz are cited as Fussel & Kreutz (p.108), Sweetser 1990 is cited as 1990 and 1991 on the same page (p. 73), Scherer becomes Schere (p. 257) etc.) and typos, the most amusing of which are ''sort-term working memory'' (p. 107) and ''metal representation in the brain'' (p. 92).
Besides its shortcomings, the book is interesting and inspiring and a useful source for everyone working on the role of the partner in communication.
Gardner, R. (2001) When Listeners Talk. Amsterdam/Philadelphia: John Benjamins.
Horton, W. S. & Gerrig, R. J. (2005) The impact of memory demands on audience design during language production. Cognition 96, 127- 142.
Kleiber, G. (1998) Prototypensemantik: eine Einführung. 2. Auflage Tübingen (= La sémantique du prototype. Categories et sens lexical. Paris 1990).
Nemo, F. (to appear) The Pragmatics of Common Ground. In Fetzer, A. & Fischer, K. (Eds.), Lexical Markers of Common Ground. Amsterdam: Elsevier.
Schober, M. F. & Brennan, S. E. (2003) Processes of interactive spoken discourse: The role of the partner. In A. C. Graesser, M. A. Gernsbacher, & S. R. Goldman (Eds.), Handbook of discourse processes (pp. 123-164). Hillsdale, NJ: Lawrence Erlbaum.
ABOUT THE REVIEWER
Kerstin Fischer is assistant professor at the University of Bremen, Germany. She has been working on recipient design and common ground for several years, especially between unequal interlocutors, for instance, in communication with computers, robots, foreigners, or children. She is co-editor of 'Lexical Markers of Common Grounds', to appear with Elsevier this year. In the framework of the SFB/TR8 'Spatial Cognition', she elicits and analyses corpora of human-robot interaction that differ only with respect to single variables in order to identify the factors that influence speakers' choices for their artificial interlocutors.