Review of Meaning, Discourse and Society①

FENGCHAO ZHEN

(Shanghai Jiao Tong University, China)

InMeaning,DiscourseandSociety, Teubert explains how a reality can be constructed in discourse.The central stance is that the discourse-external reality is just shapeless and meaningless stuff, and what people talk about in both verbal communication and written texts are actually the objects they have created collectively within discourse.Discourse therefore becomes the only seat of meaning.By putting forward this argument, Teubert finds himself in conflict with the tenor of cognitive scientists who claim that meaning can be found in mental concepts or neural activities.For Teubert, language is a social rather than a mental phenomenon, on the grounds that language only comes about as long as people interact and communicate.Language is needed to assign meaning.Without language, people’s behavior is meaningless.Meaning is negotiated collaboratively by the members of discourse communities, and thus enables them to have knowledge of themselves and the world around them as well.Discourse, viewed as the collective mind of a discourse community, constructs the realities people are confronted with.This means that through their participation in discourse, people can change these realities, i.e.the world according to their perceived needs.

A significant feature of the book is the persuasiveness and comprehensiveness in presenting critical reviews about cognitive sciences.Being highly informative, it not only cites many schools of thought in the domains of philosophy, psychology, linguistics, science and so on, and establishes links among them, but also reviews the main work and central stance of the representative scholars.More importantly, Teubert has his own critical insights into the potential problems of those thoughts.Another outstanding feature of the book is its inclusion of both theory and practice of the construction of reality within discourse.From Chapter 1 to Chapter 14, Teubert mainly deals with the theoretical investigation, arguing that meaning cannot be found in mental concepts or neural activity as implied by the cognitive sciences but in people’s collaborative negotiation within discourse.The last two chapters introduce Teubert’s approaches to exploring meaning and to interpreting texts.Chapter 16 in particular takes an example of a haiku generated randomly by a computer program to show how to interpret meaning of text or text segments by means of looking for the paraphrastic content in the discourse and establishing its intertextual links with other texts.

The opening chapter introduces the turn in scientific paradigm in the 1950s from behaviorism and pragmatism to cognitivism.It signals a change in research focus from human interaction to the working of individual mind.There are actually two schools of cognitive linguistics.One school usually labeled with a lower-case ‘c’ and ‘l’ aims to model the language system as an innate computational mechanism linked to an equally innate language of thought, and that it is the task of linguistics to analyze this language faculty.The other school, labeled with an upper-case ‘Cognitive Linguistics’, complements the language mechanism with other mental modules that reconstitute the intended meaning of the conveyed message.Although sounding attractive, these two types of cognitive linguistics are not without problems.As for the first type, take Chomsky’s postulation of the existence of universal grammar as an example.In order to formulate inviolable rules explicitly, constant adjustments, adaptation and redefinition have to be made over time, due to recognized inconsistencies.The major problem of Cognitive Linguistics lies in its attempt to equate mind with computers.According to the computational theory of mind, the working of the mind can be understood and described by analogy with the working of the computer, which shows an obvious circularity.The actual failures of the attempts of artificial intelligence and machine translation can well explain this infecund cross-fertilization.Although it may sound attractive, cognitivism cannot solve the mystery of meaning.

The second chapter gives a brief retrospection to the history of mind linguistics, tracing cognitive linguists’ idea back to the ancient Greeks, namely that the mind is the seat of meaning, the arbiter of knowledge.According to the medieval language philosophy, the meaning of words is realized only by the mental images of the object they signify, having nothing to do with their concrete forms, and the mind is the place to discover the language system.These ideas prepare the theoretical ground for the cognitive science, but this ground has not been explored fully.

Chapter 3 proceeds to address the fundamental issue of cognitive linguistics, namely that word meanings are mental concepts.According to cognitive linguistics, these mental concepts are both primary and universal.By primary it is meant that humans were born with a language of thought which is now also called mentalese.By universal it is meant that all natural languages, albeit in different forms, share a common reservoir of mental concepts.But the question is what we know about these mental concepts.One possible answer given by cognitive linguists is to establish the correspondence between natural language and mental concepts.And cognitive linguists argue that each single word in natural languages corresponds to a single mental concept.However, once the fact of genetic variation among people and the cross-linguistic comparison and contrast have been taken into account, this argument is no longer convincing.Another school of theory assumes that there exist a rather limited number of semantic primes, or basic conceptual bricks, that are seen as the building blocks of all more complex concepts.And these semantic primes can be translated into natural languages.But the problem is that we are unable to control the appropriateness of this translation.

Chapter 4 introduces another formalist model for meaning which rests on the assumption that the meaning of words can be explained and described by the presence or absence of a (language-independent) seme, i.e.the minimal units making up the distinctive semantic features of lexemes.The problem of this model is that the seme is no more than a theoretical construct within a linguistic model and that no methodology can be provided to assign semes to a given mental concept or word.

Chapter 5 deals with the concepts employed in the fields of machine translation and artificial intelligence.Since natural languages are prone to constant changes and their meanings are context-dependent, universal concepts are the critical weak points in these fields.These concepts are not the mental concepts proposed in cognitive linguistics, but like those used in terminology, or the formulae in chemistry.They are defined as the units of knowledge which are language-independent and context-free.In other words, they are not subject to constant change regardless of where they occur.The standard method in machine translation is to put the fuzzy, ambiguous natural language into the orderly and formalized concepts of a consistent conceptual ontology.This approach, however, may work fairly well with processing a controlled language such as the maintenance manuals, but less well with the unpredictabilities of fuzzy and ambiguous natural languages.

Chapter 6 discusses the different ways of looking at meaning by cognitive linguists and corpus linguists.Cognitive linguists believe that meaning does not exist in the actual linguistic utterance but in people’s heads.In the camp of cognitive linguistics, there are two main schools of thought about meaning: one-level semantics and two-level semantics.The advocates of one-level semantics suggest that the mental representation of what is being uttered refers to the utterance itself.Those who subscribe to two-level semantics believe that the mental representations conveyed by what is being uttered are much richer than the natural linguistic utterances, which can be identified as the speaker’s intentions.In order to reconstruct the speaker’s intentions in full, the hearer in verbal communication has to employ a mental mechanism to read the speaker’s mind.This approach to looking for meaning in mind is criticized by Teubert on the grounds that 1) the mind is a blurred construct, 2) there is no scientific evidence for the mind, and 3) the only content that can be negotiated between people is symbolic or can be verbalized.Corpus linguists, on the contrary, look for the meaning of a lexical item in real language data, i.e.by including the context in which it is embedded.They hold that meaning does not exist in mental concepts but can be negotiated in what is actually said.In the process of negotiation, we learn about the meaning of lexical items from other people’s contributions to the discourse.And this is how we learn to use words when we grow up.

Chapter 7 summarizes four seats of meaning.The first one is the individual intentional mind.By intentionality or consciousness is meant that we experience and become aware of what is going on inside us, such as our moods and feelings and around us, and such as the behavior or facial features of other people.It is however impossible to look for meaning in individual’s intentionality on the grounds that no direct access to it is available except if a social perspective is used, which means that people negotiate their intentionality in discourse.The second option is the individual computational mind/brain.The computational model of the mind takes an analogy with computers and programs, viewing the brain as the hardware and the hard-wired circuits and the mind as the totality of all the programs running in the brain.However, no matter how advanced the computer becomes, it can only follow the pre-programmed instructions to process languages without interpreting signs or symbols or discussing their meanings, and therefore this model cannot be linked to intentionality.The third category falls into the brain where language becomes reality.Due to the advanced technology in the paradigm of neural sciences, it is possible to discover fascinating facts about the brain itself.Experiments have been conducted to demonstrate the neural activities in cerebral tissue triggered by utterance, and the neural sciences attempt to link these activities with the content of representations, arguing that the neural parameters involved in an action represent the internal structure of this action, and are mapped onto the internal structure of the corresponding concepts which can realize the meaning of an utterance.However, neurons are neurons are neurons.In other words, neurons themselves don’t stand for meaning.We have to resort to the use of language to express the neural representations.Inaccessibility is the unsolved problem for the above three options of looking for meaning.What is real and accessible is only the collective mind which is embodied in the discourse and this is the last option.We contribute to the discourse and get access to it, and in the discourse we share, interpret, inspect and investigate the meaning of lexical items.

Chapter 8 defines the concept of discourse.In its broad sense, discourse consists of all the spoken, written or signed utterances from the first time when people start communicating via language.The discourse, constituted by texts or utterances in context, is the only reality constructed and shared by the members of a discourse community.People use language to exchange and share content, and meaning is what people have said about a word, a phrase, a text segment or a text.It is the discourse that forms the seat of meaning.In other words, meaning can only be explored in discourse rather than mind or brain.The discourse in the broad sense is, however, not available to us, and what we can get access to for study is only a small part of it.This is why we have to define explicit criteria for selecting discourse data to study so that we can be sure for the scope, quality and representativeness of each text.This chapter also criticizes Critical Discourse Analysis for its conviction that discourse is just one of many social practices.As Teubert sees it, it is discourse that establishes all other social practices by assigning meaning to them.

Chapter 9 aims to link language to society, arguing that they presuppose each other.That society presupposes language implies that people’s verbal communication is a prerequisite of society.In communication, people assign symbolic content to what they encounter in the realities they share, and therefore their behavior becomes meaningful, which sets people’s society apart from a pack of wild animals.Moreover, people assign an interpretation to each other’s actions or to the meaning of texts, and this negotiation of meaning is not an individual’s intentionality but a collaborative activity which defines and constructs society.The reason for us to interpret other’s behavior or to discuss the meaning of a text is that we want to have people in a discourse community to accept our understanding.We do not interpret for ourselves.Therefore, any first-person experience remains meaningless until it is interpreted and made public, thus becoming an object of negation.This is Teubert’s social perspective towards language.

In order to differentiate spoken from written language, Chapter 10 takes a closer look at the primordial speech situation which refers principally to an informal gathering where all the parties share a given reality by engaging in collective interaction through ongoing negotiations, explanations and reflections.In this primordial speech situation, the shared environment is integrated into discourse and thus becomes discourse internal.Once this has happened, the realities encountered by people are no longer realities of a world outside but have become realities constructed through discourse.By means of definitions and paraphrases, all the participants negotiate and achieve consensus on the meaning of a lexical item and its reference to an object, concrete or abstract.And this process is never ceasing in the speech situations.Teubert also points out that once a discourse-internal reality has been established, narration becomes possible, in which no direct link exists between utterances and a shared reality.However, the primordial speech situations turn the shared reality into a discourse-internal reality.From then on, people’s first-person experiences are not any longer ‘raw’ experiences, but experiences informed by the categories that have been established in discourse.

Then Chapter 11 proceeds to distinguish oral from literate societies because the languages used in both situations differ.Spoken language contains not only text but also includes the setting where it occurs, its acoustic features, the deictic acts, and even the reality shared by the participants of the situation.Spoken language as a transient and impromptu acoustic phenomenon, is hardly realized as symbolic until recorded and transcribed.The discourse of oral societies mainly deals with concrete things with little venture into abstraction.In written language, it is only the meaning of text segments that matters.In other words, the content of a text is its meaning which can tell this text from all other texts.The members of a discourse community negotiate the meaning of a lexical item collaboratively.It is generally believed that written language gives a more dependable account of reality than speech, and thus has more profound effects on many fields of social life concerning law and justice, history, schools and knowledge, property, and so on.Moreover, literacy has enormous effects on societies’ modalities of thinking, such as people’s ability to think in and about hypothetical and counterfactual cases and thus to develop new solutions to old problems, and to recognize and reflect on symbolic content.

After differentiating spoken from written language, Chapter 12 claims that only the recorded language is the available data resource for empirical linguistic research.The problem of analyzing the spoken language in the primordial speech situation is that it seems impossible to separate what is said from the context in which the utterance takes place.Moreover the boundary between the linguistic and the non-linguistic symbolic interaction involved in the primordial speech situation is not clear-cut.Because the primordial speech situation is transient and impromptu, once it has taken place, it is irretrievably lost and never fulling recoverable in recordings or transcripts no matter how detailed the annotation of the non-linguistic factors is.Finally, the chapter draws three conclusions.The first one is that linguistic studies should stop attempting to describe the way in which people’s process language because peoples’ minds are not directly accessible.The second one is that any integrationist approach will have to be severely limited because it is almost inconceivable to record and transcribe all the linguistic and non-linguistic factors that are indispensable ingredients of symbolic interactions.The third one is that what we can do to deal with spoken language is to focus our investigation on the symbolic nature of the interaction.

Chapter 13 explores the relation between discourse and reality, knowledge and meaning.There are two types of reality: the external reality and the reality constructed in the discourse.What is meaningful is not the non-symbolic reality outside of discourse but the shared reality constructed and represented by the symbolic content of the discourse.Without the discourse, the external reality would be just meaningless and chaotic stuff.As for knowledge, a traditional view relates it to truth.In other words, the reason why we could know something is that they really exist.As a result, this tradition believes that it is reason, a faculty of the mind, that can recognize something true and naturally knowledge, as knowing what is true, has always to be private.However, if knowledge is just private, and all the other members of the discourse community might disagree, how could the true knowledge be shared or exchanged? In order to overcome this conundrum, two types of knowledge are proposed, namely private knowledge and public knowledge.First-person experiences are the basis for private knowledge, and once it is talked and negotiated via language in discourse, it turns into public knowledge.It is further argued that all our knowledge is firmly situated within the discourse and that therefore all our (public) knowledge does not refer to a discourse-external reality but to the realities we encounter in discourse.Thus the meaning of a lexical item is co-referential with the knowledge of the discourse object for which it stands.In other words, the meaning of lexical items, or the knowledge about the discourse objects they represent, are collaboratively constructed in the contributions made by the members of a discourse community.

Chapter 14 addresses the issue of the language in scientific reports.It is conventionally assumed that by conducting experiments, the scientific research aims to invent or discover something which has already existed in the real world.In other words, the philosophy of science used to presuppose the existence of a reality which is discourse-external and independent from the observers.Holding a different opinion, Teubert argues in this chapter that the language of scientific reports endeavors to re-create a virtual primordial speech situation, and the data reported in the research are no more than a set of meaningless stuff unless a consensus on the interpretation of the data has been reached.Undoubtedly, there is a reality out of discourse, but without discourse it remains meaningless and chaotic.What scientists actually attempt to illustrate is their interpretation of the reality which they construct and share in their discourses.

Chapter 15 mainly discusses hermeneutics, the art of interpreting texts.What modern hermeneutics is interested in is the meaning of texts and their components rather than the author’s intentions or the relationship between the text and reality out of discourse.In other words, discourse should be the only resource to undertaking the hermeneutic act.There are two instruments employed in a hermeneutic analysis, namely intertextuality and paraphrase.Intertexutality means that any new text or a text segment can be seen as a reaction to certain previous texts.In other words, the next text segments can be interpreted as referring to similar text segments in previous texts via intertexual links.New texts habitually explain, affirm, modify, reject or merely elaborate on earlier texts, for instance by redefining the meaning of lexical items or larger phrases.Such new definitions are what Teubert calls paraphrases.The marked linguistic form of paraphrase is the copula ‘is’.However, in most cases, paraphrase is not so direct and straightforward.Therefore, we have to interpret the context of given text segments to explore their paraphrastic content.By resorting to paraphrases and other paraphrastic material, we are able to explore the meaning of a lexical item in a specific discourse.Teubert also argues that unlike traditional linguistics including corpus linguistics as well, which still attempts to find out the regularities obtained for a given language by looking at large corpora from a synchronic perspective, hermeneutics is interested in the meaning of a given occurrence of a lexical item or text segment in a given text from a diachronic perspective, allowing us to explore not only the difference but also the similarity between its present occurrence and other occurrences in previous texts.In this sense, hermeneutics entails diachronicity as a crucial notion, and it is the impact of the diachronic dimension that will determine any new interpretation.

In order to examine the feasibility of interpreting meaning of text or text segments through looking for the paraphrastic content in the discourse and its intertexual links in other texts, Chapter 16 takes an example of haiku generated randomly by a computer program.Teubert’s aim is to demonstrate that meaning is never stable and interpretation is never a solitary exercise, which means that new interpretations will be constantly added to the previous ones and the meaning of a word keeps changing forever.It is the context that can solve the problem of ambiguity, vagueness and fuzziness in the meaning of isolated single word, and meaning only emerges within the discourse, namely in the way that it is negotiated and discussed in discourse without referring to a discourse-external reality.This chapter also argues for corpus linguistics to explore the meaning of lexical items based on the actual evidence of real language data, i.e.discourse data, and working with the notion of statistical significance can help to discover regularities and trends rather than inviolable rules, but it must be complemented by interpretation.

In concluding this book, Teubert raises three questions.First, can the quest for meaning dispense with the investigation of the solitary mind? He argues that searching for meaning would be futile if we attempt to explain how a monadic mind is endowed with a mechanism for processing mental states, because first of all there is no such object as the mind whose workings can be observed directly and secondly mental representations confined to the mind and not expressed in discourse are per se meaningless.Therefore, meaning is not a mental representation of what is said but everything said in the self-referential discourse and it can only be interpreted in collaboration.Secondly, how real are the realities we experience? It is argued in this book that discourse constructs the reality in which we experience ourselves and without discourse, the reality being nothing but stuff.This stance does not necessarily deny the existence of the discourse-external world, but shows that this shapeless stuff can be made comprehensible only through the mediation of discourse.Lastly, can we change the realities we are confronted with by reconstructing it in our contributions to discourse? Media discourses are expressions of interests of those who are in charge of these discourses, and if we do not like what we find there we must make our voices heard.Media discourse advertises, for instance, individual agency.The west media glorifies individual agency or personal responsibility, discouraging all forms of collective action although the top hierarchies exercise solidarity among themselves in order to suppress the interests of others.But this discourse of those in positions of power is largely hidden from us normal people.In other words, the reality they design for our consumption is the opposite of the reality they exercise among themselves.Teubert argues that we have to learn that there is an alternative.Collaborating in our direct involvement in discourse can make a difference.We should not believe in doing things alone.That is to say, we should give up on individual minds and start acting together, as a collective mind, and re-interpret the realities we find in discourse in such a way that this new interpretation suits our interests more than the one which comes from those exercising power over us.If reality is to be found only in discourse, it is within our power to change it.

Providing an account of the nature of meaning is one of the key issues inMeaning,DiscourseandSociety.As Jackendoff (2002: 267) put it, “meaning is the ‘holy grail’ not only of linguistics, but also of philosophy, psychology, and neuroscience...For most people, meaning is intuitively the central issue in the study of language-far more important than understanding details of word order or morphology”.Linguistics as well as a range of related disciplines in the humanities and the social and cognitive sciences attempt to propose theories from different perspectives to contribute to explaining the mystery of how word means.In modern linguistics, the dominant view of word meaning is literalism (Recanati 2004) which distinguishes between semantics which aims to nail down the context-independent aspects of meaning, and pragmatics which explains the context-dependent aspects of meaning.However, as many scholars notice, this distinction is illusory, and the direct consequence of this distinction is that “word meanings are assumed to be stable and relatively delimited atoms of meaning which are context-independent” (Evans 2009: 26).However, the evidence from the perspectives of social psychology, cognitive psychology, interactional sociolinguistics, cognitive linguistics, corpus linguistics, and computational linguistics shows that the meaning of word is not stable, but flexible, open-ended, and highly sensitive to context.In other words, the meaning of a word is realized when it is used in actual situation, or the meaning of a word is the function of context.However, even a cursory glance at the literature in this field reveals that there is clear lack of an agreed view on what types of context should be taken into account when explaining meaning.For Evans (2009: 26), “the notion of context is a complex and multifaceted phenomenon which includes linguistic as well as non-linguistic aspects of the communicative event”, and the encyclopedic knowledge which is “primarily non-linguistic or conceptual” (Evans 2009: 17) constitutes a word’s semantic potential and the context serves to narrow or constrain the precise part of this semantic potential which is relevant in any given utterance.Obviously, Evans employs a cognitive perspective to explain meaning.However, for Teubert, meaning as any conceptual representation is not accessible and therefore cannot be validated or falsified.What he trusts when explaining the meaning of a lexical item is the linguistic context, that is, the text or discourse.Undoubtedly Teubert’s views are influenced by Sinclair who firmly believes that a new understanding of the nature of language can be available only by means of examining extended naturally occurring texts (Sinclair 1991).However, different from corpus linguists, Teubert does not attempt to discover the patterns or regularities obtained for a given language by looking at large collections of text from a synchronic perspective, but on the contrary is interested in the inherent variation in the nature of meaning and adopts a diachronic perspective to explain the meaning of a given occurrence of a lexical item or text segment in a given text.The intense trust in text would make Teubert ignore the role of encyclopedic knowledge which includes background knowledge, common-sense knowledge, sociocultural knowledge and real-world knowledge (Evans 2009: 17) in understanding the meaning of a lexical item.Moreover, Teubert does not define precisely the scope of text which is adequate to interpret the meaning of a lexical item or a text segment.The evidence from corpus linguistics reveals that the meaning of a lexical item will be different if more extended texts are included.As for the process of interpretation proposed by Teubert, it is not so easy to find out the paraphrastic content because in spite of the marked linguistic form of paraphrase, that is the copula is, in most cases, the paraphrase is not direct or straightforward.These, however, cannot deny the value ofMeaning,DiscourseandSociety, in which Teubert provides strong grounds for arguing that discourse is the seat of meaning and human’s society is constructed in discourse.Teubert’s critical views against cognitive linguists who claim that meaning is conceptual representation or found in neural activities and his powerful and philosophical argument that language is a social phenomenon shed light on the nature of meaning, and the relationship among language, society and the world.

REFERENCES

Evans,V.2009.HowWordsMean:LexicalConcepts,CognitiveModels,andMeaningConstruction.Oxford: Oxford University Press.

Jackendoff,R.2002.FoundationsofLanguage:Brain,Meaning,Grammar,Evolution.Oxford and New York: Oxford University Press.

Recanati,F.2004.LiteralLanguage.Cambridge: Cambridge University Press.

Sinclair,J.1991.Corpus,Concordance,Collocation.Oxford: Oxford University Press.