LINGUIST List 14.2104

Fri Aug 8 2003

Review: Linguistic Theories: MacKenzie (2002)

Editor for this issue: Tomoko Okuno <>

What follows is a review or discussion note contributed to our Book Discussion Forum. We expect discussions to be informal and interactive; and the author of the book discussed is cordially invited to join in.

If you are interested in leading a book discussion, look for books announced on LINGUIST as "available for review." Then contact Simin Karimi at


  • Kenesei Andrea, MacKenzie (2002), Paradigms of Reading.

    Message 1: MacKenzie (2002), Paradigms of Reading.

    Date: Mon, 4 Aug 2003 18:34:43 +0200 (CEST)
    From: Kenesei Andrea <>
    Subject: MacKenzie (2002), Paradigms of Reading.

    MacKenzie, Ian (2002) Paradigms of Reading. Relevance Theory and Deconstruction, Palgrave/Macmillan

    Announced at

    Andrea Kenesei, Department of English and American Studies, Veszprem University, Hungary

    MacKenzie (M.) discusses Relevance Theory (RT) versus the deconstructionist approach to reading through the works of Paul de Man (d. M.). The author very systematically and convincingly confronts pragmatics with deconstruction.

    Chapter 1 introduces Sperber and Wilson's RT of reading as opposed to the deconstructionist approaches to interpretation and understanding. On one end of the scale there is d. M.'s and Derrida's negative and pessimistic account of language, and on the other end there is the positive pragmatic perspective. The author contests d. M.'s claim that language is ironic and inhuman by proposing a pragmatic alternative. The active human agency, which pragmatics takes for granted, is obviously a more beneficial basis to analyse the reading process than the impersonal approach of the poststructuralist, deconstructionist and Lacaian theories. RT opposes the Saussurian claim that there is no correct reading because meaning can never be fixed, and d. M.'s claim that the tropes and ironies in language undermine the intended meaning. According to RT, the signifiers provide the starting point for an inferential process that leads to the discovery of the intended meaning. The author describes the different views on reader interpretation of literature-the theory of Schleiermacher and Dilthey, Poulet and Richards, Gadamer and Sperber and Wilson (S&W). He points out that he does not intend to provoke ''Damascene conversions'' among deconstructionists. Nevertheless, he emphasises that from a pragmatic viewpoint d. M.'s account is inadequate.

    Chapter 2 is devoted to the discussion of spoken communication in view of RT, since S&W concentrated on this aspect of communication. M. interprets S&W's theory on how utterances express thoughts, what additional attitudes and performances are added to utterances. Even pragmatics admits the fact that due to semantic and referential ambiguities utterances can never fully express thoughts. No comparison is made at this point with the deconstructionist view, but the reader infers that the point M. makes is that pragmatics focuses on the possibility of filling the gaps between what goes on in the mind and what is verbally expressed. Inference is given priority to recognition. Besides inferences there is ostension that the speaker has, i.e., the effort to make the audience understand that information is being passed on. The contextual effects are in direct proportion to relevance, whereas the processing efforts are in indirect proportion-the greater the first is the greater the relevance; the greater the second is the lower the relevance. M. briefly describes what is well-known about assumptions and how the memory works, however, the already known is treated in view of the novelty-RT. And he does this very convincingly, systematically and in a readable fashion. The use of ''she'' and ''he'' for speaker and hearer, respectively, taken from S&W, gives the reader a familiar feeling about communication. A comparison is made between Grice's principles and S&W's model; the former must be consciously applied, the latter need not. Explicatures are strong assumptions of the hearer, and implicatures, which produce the poetic effect, are the weak ones. At the end of the chapter there is a shift from spoken communication to poetic language.

    Chapter 3 is about how the understanding of literary language is governed by relevance. M. outlines S&W's assumptions about how the theory of communication and the theory of cognition determine literary interpretation. The conclusion is that the optimal relevance of communication must shift towards the maximal relevance of cognition in order to fulfil the claims of the understanding of literature. Mention is made how Hirsch, Barthes and Focault see the role of the author in reading literature. The history of hermeneutics is discussed briefly through Schleiermacher, Dilthey, Parry (romantics) with reference to the opposition represented by Nietzsche, Husserl, Heidegger, and Gadamer. As for the shared cultural background of people, S&W assume that this is the basis of the similar interpretations of public representations. From the description of the reader-response theories (Iser, Mailloux, Eco, Gutt, Poulet) we learn about the different views concerning the relationship between author and reader. As for the poetic effects, d. M. and S&W agree that the weak implicatures are dominant and cognition is primary to communication. Then comes the depiction of the opposition between S&W and Dawkins regarding culture. Positive hermeneutics regards the reader as one who reconstructs the context which is shared with the author, invents the author and her intentions, and projects into the author's creative process.

    In Chapter 4 M. goes on to introduce the notion of negative hermeneutics, which, unlike positive hermeneutics, considers the literary work a resource rather than a source, and which explains the unconscious authorial motivations, and which, for this reason, is not the scope of RT. The unconscious identity of the reader is revealed through Holland's theory, which says that reading is a process where the reader only ''replicates'' himself, thus never gets any novel experience. M. argues that Holland's wrong in attributing too much to the reader's identity but we may object that in interpretation the reader's world schemata are of primary importance. The question is, to what extent identity and knowledge mutually determine each other. This is proved by the ''anticipatory hypotheses'', which are based on these two factors. RT incorporates the positive elements of other models-while reading literature the hearer/reader takes his time to go over the tremendously many processing processes, in conversation the same processes take place in the nick of time. The intentions are conscious, the processes are unconscious. I do not agree that Fish's ''anticipatory hypotheses'' are ''premature'', as M. claims, because if nothing but the title of a text, which is a macroproposition, has the force to bring about preliminary assumptions. He talks about the reader's expectations-if there are expectations, there are hypotheses as well. The fundamental difference between RT and deconstruction is that they treat the same thing only from two different angles-the former accepts that texts are more than often ambiguous; the latter rejoices in hunting obscurities. Is that such a big difference? It is true, however, that RT respects the reader, and deconstruction does not, as it seems to disrespect everything.

    Chapter 5 starts with McCloskey's definition of rhetoric and metaphor, and refers to d. M.'s description of (tropical, mechanical and indecipherable) literary language. Relying on Rousseau's views on discourse, d. M. seems to have an erroneous starting point of ''metaphor'' instead of ''common noun''. In this part of the book, like elsewhere, there are cross/ and backward references to authors citing somebody else. All agree in one thing-that language is by nature poorer than the world around us. Neither actions nor objects and concepts can be fully described and referred to with words. This is the basis of the apparently intricate, yet fundamentally bipolar debates about meaning. M.'s technique of arguing and reasoning is smooth and easy to follow. He shifts to d. M.'s approach clearly and convincingly. d. M.'s views are not novel; he builds his theory upon Rousseau's and Nietzsche's ideas. Even a naive layman would draw the conclusion that provided d. M. were right, nobody including himself would understand what he said. If figural meaning is self-destructive, there is not much meaning left in ordinary language, which operates with more figural meanings than one would ever think of. d. M.'s logic is not logical at al-for him figures represent language, figures do not represent things but then what represents things? There is language and there are things, the two are not connected in his view. Then what is language for? d. M., by questioning the role of a signifier in number, questions communication as such, but gives no other solution, as M. vividly shows. The discussion of metaphor, truth, lie etc. is more about Rousseau, Friedman, Nietzsche and Marx than d. M.. However, the conclusion d. M. draws about the tropological nature of numbers is as illogical as almost everything he is convinced of. d. M. does not seem to notice that he has fallen into his own trap-if things become real in the mind, and if language is created by the mind, then the two must meet in the mind, that is, they cannot do without each other, which contradicts what he stated earlier. While discussing catachreses, we must bear in mind that if every dead metaphor were substituted by a non-metaphorical word, every language would contain an insurmountable number of words, which would work against the common sense of people. Man invented metaphors to simplify language and to keep the linguistic tools within treatable boundaries. This common sense is best proved by the simple fact that the same metaphors and dead metaphors can be found in many languages, taking the foot of a mountain or the legs of a piano as very good examples.

    In Chapter 6 the counter-argument M. gives to ''Unmediated expression is a philosophical impossibility'' could have been ''but a linguistic possibility'' because thoughts are indeed more complex than language. d. M. insists that even everyday language is exposed to uninterpretability and ambiguity. If so, he should not have written a single sentence because to interpret it is a task with no end or progress. We are relieved though because M. puts it in words what we have in mind as a counter-argument to d. M.. In ''What's the difference'' d. M. is right after all-in everyday conversations there are many cases of unsuccessful communication of the Bunker type-it is true, however, that they often are the result of people's bluntness and language itself. d. M. seems to neglect the fact that language is man's invention and it is used by human beings. In connection with literature d. M. makes a distinction where he should not-between aesthetic values and linguistic structures. The opposition between rhetoric and aesthetics also lacks ground. If, according to d. M., the literary reading is rhetorical and rhetoric contains no aesthetic component, then what text is aesthetic? The question is left unanswered. And this is not a logical continuation of Nietzsche's equating rhetoric and language. (M.'s grouping of the literary devices is very useful.) That grammar cannot be considered to be a trope is a question that deserves discussion in a separate book. But there are views according to which literature, or one literary work, is one macro-metaphor; this represents a somewhat similar logic as d. M.'s collapsing of grammar. d. M.'s account of irony is again something startling-irony is an inherent feature of language and not the speaker's attitude, he says. d. M.'s account of reading is totally sceptic-for him readability as such is non-existent because the readings of a text are incompatible. d. M. questions the rightfulness of the trivium and believes that the rhetorical function undermines grammar and logic. M. argues that d. M.'s main shortage is the ignorance of inferences a reader makes. Also, M. attacks d. M. for making apodiptic statements, for being too self-confident setting up his theory, which is not too difficult to find fault with.

    In Chapter 7 the refutation of d. M.'s claim that the word-thing link is conventional starts with the use of the deictic elements. We could argue though that it is just these conventions that allow for a consensus in understanding. Claiming that language constitutes experience rather than reflects it, d. M. shares the Whorfian view that language is prior to thought, language determines culture, or as d. M. says, the ''self''. However, the binary oppositions d. M. makes include the separation of the self and the world-the question is whether culture is made up of selves or it is the reflection of the world. The conclusion of ''Being and becoming'' might be that all those binary oppositions are rather for uniting than separating the self and the world, language and cognition and representation and thought. Reading ''Concepts, metaphors, catachreses and reference'' I had a feeling that d. M. writes about blunt people who are indeed blind to metaphorical meaning-I guess that the basis of the contradiction between the deconstructionist and the pragmatic approach is that the former is disillusioned by the bluntness and blindness of so many of us, which is a reality, and the latter has an idealistic and optimistic opinion of people's mental abilities. In ''Reference and ideology'' M. can but repeat that man has not been able to invent any better system to convey thoughts than the system of signs, therefore no matter how much d. M. is right that language is not the best model for cognition, we can but surrender to it. In ''Ich kann nicht'' M. discusses symbol and sign as seen by the Man through Hegel's account of the subject. d. M. is criticised for not paraphrasing but misinterpreting Hegel. Part of the problem might be the translation of ''Phenomenology''. Individuation or generality is the main question discussed here. d. M. dislikes the idea that his opinion is consequently others' opinion as well. Hegel is a starting point for d. M. and the pragmaticians alike concerning the deictic ''I'' but they take different routes-language is not possessed by anybody, says he and language belongs to all, they say. The problems of reference cluster around names and grammar. The question is whether grammar suspends referential meaning or not. Again, d. M. claims that it does, unlike the pragmaticians.

    >From Chapter 8 we learn that d. M. subverts Austin's notion of performative and says that performative utterances-like all other linguistic performances-are independent of the user, they are only bound to language. ''The purloined ribbon'' is about d. M.'s interpretation of Rousseau, which, according to M., is a distorted translation and thus a misinterpretation. Also, d. M. favours a rhetorical instead of a psychoanalytic reading of ''Confessions'', and, as M.'s greatest accusation, d. M. understands a metaphorical meaning where there is none. In ''Excuses, fictions and machines'' it becomes clear that Rousseau's confession must be seen as a case of truth-falsity opposition rather than d. M.'s fictional referential reading, which results in an over-euphemistic explanation of Rousseau's story. d. M. is constantly criticised for ''interpreting [many philosophers] extravagantly'' for his own purposes. The exclusion of the speaker's intentions is again visible in treating language and texts in a mechanical fashion. (From d. M.'s articles written during WWII we learn that he took sides with fascism.)

    In Chapter 9 d. M. is again accused of not having too many novel ideas but only paraphrasing others. Saussure's deconstructionist followers distort his signification by implying a passive participation of the hearer as well as the speaker. (The distortion was not that hard though.) M. comes up with a good solution to the problem of man and language interaction: we should not apply exclusion, as Holland suggests, but inclusion. Rhetoric is revisited from the point of view of the self, which is the result of language for d. M. He attributes cognition to language rather than the individual. d. M. deprives especially literary language of every human feature; the text is autonomous for him. The conclusions he draws from Benjamin's speculations on translation are subversive, far-reaching and again free from human intention. M. clearly demonstrates how RT can cope with all these nonsensical approaches. I can argue with Frost's claim that a poem is lost when translated with my own research where I proved the contrary (Fordítástudomány 2003). (Mention must be made though that I worked mainly with free verse.) Currie is called for to support M's criticism concerning d. M.'s nearly exclusive wearing borrowed plumes.

    Chapter 10 is devoted to interpretations of a Wordsworth poem. Bateson, then Hirsch operate with contextual implication. There is disagreement concerning how many readings are allowed and which is the optimally relevant. Davies' interpretation is based on the personal deixis of ''spirit'' and the reference to the female character. Caraher's reading is based on the punctuation and a different denotation of the feminine deixis, and in general supplements other readings. His reading barely suits RT. Holland gives a psychoanalytic explanation making sexual inferences, and also concludes that this kind of analysis is very much like the formalist approach. He says that the number of the readings of a text is the same as the number of the readers. Matlak stresses relying on the biographical background when interpreting texts. d. M. relates the message to temporal allusions-both futuristic and eternal. Miller's opinion resembles Hollands in that any reading is possible. M. makes it clear at the beginning of the chapter that ''All [RT] aims to do is to offer explanations of existing readings in cognitive pragmatic terms''. Well, to choose from a basket is easier than weaving the basket... What is more, the reader is curious to know which interpretation RT or M. finds the most relevant-there is no choice made and the final paragraph does not seem sufficient to solve the problem. If it is true that a contextual basis and inferential processes are required to decipher any piece of information, why does RT not accept that any interpretation is viable or that there are as many interpretations as readers/hearers?

    Chapter 11 ends with an overall definition of RT: ''[RT] focuses on the potential richness of intended or interpreted meanings, including poetic effects.''


    MacKenzie aims to prove that RT, being a novel approach to communication, is capable of overriding the hitherto existing theories of understanding and interpretation. According to RT, hearers are able to choose for themselves the most relevant information with the smallest processing effort and the greatest possible cognitive effect; let it be literary or non-literary language, literal or non- literal meaning. The book is not narrowed down to the description of RT but gives a comprehensive outline of the deconstructionist views opposing the pragmatic approach of RT. The author's knowledge cannot be more comprehensive and his arguing technique cannot be more convincing. However, the penultimate chapter seems to leave something behind-without an example of how RT sees for example the poem interpreted the reader is not fully convinced. If we choose the most relevant interpretation then there are as many of them as there are readers. And then who decides which is the most relevant? If the reader does, then the above statement must be true.

    About the reviewer

    Andrea Kenesei is a senior lecturer of linguistics. Her interests include pragmatics, psycholinguistics, discourse and text analysis, linguistic analysis of literature, translation and reader-response theories. She is working on her PhD dissertation on "Reader-response of translated verse".