LINGUIST List 14.2251

Tue Aug 26 2003

Review: Ling Theory/Pragmatics: MacKenzie (2002)

Editor for this issue: Naomi Ogasawara <>

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


  1. Francisco Yus, Paradigms of Reading; Relevance Theory and Deconstruction

Message 1: Paradigms of Reading; Relevance Theory and Deconstruction

Date: Tue, 26 Aug 2003 11:52:13 +0000
From: Francisco Yus <>
Subject: Paradigms of Reading; Relevance Theory and Deconstruction

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

Announced at

Francisco Yus, Department of English Studies, University of Alicante,

[For another review of this book, please see --Eds.]


In this book, Ian MacKenzie (henceforth I.M.) discusses and contrasts
two approaches to literary communication and reading, namely
deconstruction, as supported by Paul de Man (henceforth d.M.), and
relevance theory (henceforth RT), created by Sperber and Wilson
(1986/95). He also addresses the (now almost rhetorical) question of
where the meaning of a text is supposed to lie. As I pointed out in a
former review (Yus, 2002: 624f), ''literary theories [are] divided as
to where the meaning of a literary text lies: either that meaning is
only the author's intended interpretation (intentio auctoris), or the
'objective' meaning of the text itself (intentio operis), or the
reader's personal interpretation of it (intentio
lectoris)''. Relevance theory is clearly reader-centred, but
d.M. tries to deconstruct this emphasis by resorting to a series of
''deconstructivist tricks'' aiming at subverting this role of the
reader in the eventual meaning of a text. Clearly, this is a losing
battle, as I will try to show below.


Chapter 1 (''Pragmatic Banality and Honourable Bigotry'', p.1-15) is
an introduction to the general issues addressed in the book, together
with a main introduction to the basic tenets of relevance theory
(henceforth RT) and deconstruction. From the beginning, I.M. positions
himself clearly on the pragmatic side of this contrastive analysis,
favouring the inferential account of communication suggested by RT
instead of the deconstructionist account which insists on the
materiality of language, and disregard the enormous role played by
contextual factors in linguistic communication (p. 2). For d.M.,
language is invariably ironic, random, arbitrary, irresponsible,
mechanical and inhuman, very far away from the interpretive picture
within RT, according to which utterances (or texts) resemble the
thoughts that the speaker or writer intends to communicate, providing
clues and blanks to be filled inferentially by the addressee (listener
or reader), a far more interesting approach. Indeed, for I.M.,
''substituting the word inference, or more specifically the word
relevance, for d.M.'s term rhetoric, very often allows one to explain
language use more adequately'' (p. 3). Later in the chapter,
I.M. provides the basic aims of the book: he is ''specifically
concerned to contest d.M.'s notions of the random, mechanical,
arbitrarily or aberrantly referential, ironic and inhuman nature of
language'' and proposes ''a pragmatic alternative to this theory''
(p. 4).
In chapter 2 (''Relevance Theory and Spoken Communication'', p. 17-
28), I.M. provides a general outline of RT, with the central issue of
linguistic underdetermination (the fact that utterances or texts
rarely -if not never- code the thoughts that the speaker or writer
wants to communicate with them).
I.M. also introduces the notion of poetic effect, basically linked to
the notion of weak implicature, and which is the traditional RT
approach to figurative language. As I sketched in Yus (2002: 621-622),
underlying this notion is the claim that every utterance is used to
represent a thought of the speaker's (or writer's). Often (if not
always) communicators are not literal in the way their utterances
communicate their thoughts. Even at the explicit level, utterances are
often incomplete and the communicator expects the addressee to be able
to fill up the blanks, as it were, in the (context-bound) processing
of the utterance. The same applies to metaphors and related tropes. In
this case, there is an interpretive relation between the speaker's (or
writer's) thought and what it represents. In highly innovative
metaphors, the result is a wide array of contextual implications whose
extraction is the interlocutor's responsibility (weak implicatures).
Therefore, there is a whole continuum in metaphors between
communicator-backed strong implicatures and the weak implicatures
which the interlocutor is responsible for extracting. However, we
should not make a direct equation between implicature and weak
effect. Kenesei (2003), in a recent review of this book, seems to make
this link when stating that ''explicatures are strong assumptions of
the hearer, and implicatures, which produce the poetic effect, are the
weak ones''. Both explicatures and implicatures can be -and often are-
ostensively backed up by the speaker, but there can be a point in
which their extraction is increasingly the addressee's
responsibility. Indeed, readers of literature will normally be willing
to devote some additional mental resources in order to access a wider
array of weak implicatures because the sensations obtained offset the
cognitive effort required in exchange, regardless of whether these are
intended by the writer or not.
Personally, I wish I.M. had focused more on the current development of
this view of metaphor in terms of ''ad hoc concept formation'', within
which metaphoric concepts are viewed as making a contribution to the
proposition expressed by the utterance (and are no longer viewed as
implicatures), a development of RT which is extensively dealt with by
Pilkington (2000) and also by other authors within an RT perspective
(e.g. Carston, 2002).
Chapter 3 ('''Positive Hermeneutics': Relevance and Communication'',
p. 29-46) is about literature from an RT perspective. The relevance-
theoretic focus is, predictably, on how the reader mentally attempts
to find a relevant -i.e., intended- interpretation of the text, that
is, a close parallel to the writer's thoughts using the words on the
page as ''a guiding light''. Sometimes the reader will fall short of
reaching the author's thoughts or intended interpretations, but this
is not essential to enjoy literature. Even if there are easy-to-spot
intended meanings in the text, ''rather than attempt to infer
manifestly communicated informative intentions, readers can let
meaning proliferate, and manipulate, decompose and recompose fiction
for as long as the effort brings results'' (p. 33).
The chapter is filled with opinions by different authors on the task
of writers and readers, often within specific research areas such as
hermeneutics (e.g. Dilthey) or reader-response theories (e.g. Iser). I
found this section extremely useful, and I enjoyed reading about how
intuitive Bahktin was on the dialogic nature of literary
communication. Besides, when writing about the interpretive quality
of human interpretation, there is also some reference to the
epidemiological model of cultural spread envisaged by Sperber (1996).
Chapter 4 ('''Negative Hermeneutics': Themes, Figures, Codes and
Cognition'', p. 47-61) introduces the term ''negative hermeneutics'',
which falls outside RT in its attempt to access unconscious authorial
intentions. RT is instead ''concerned with conscious intentions and
does not consider the possibility that we are definitively cut off
from authors' communicative intentions because we are bound by
interpretive strategies emanating from our own unconscious''
(p. 48). There is again (as there is throughout the book) a contrast
between the RT view and the views of other authors (e.g. Holland).
Chapter 5 (''Words, Concepts and Tropes'', p. 62-83) is about tropes
and especially metaphor. Again, I wish I.M. had devoted more time to
current developments of RT in the account of figurative language as ad
hoc concept formation (see above).
Ad hoc concepts apply not only to figurative language. Wilson and
Sperber (2002), for instance, describe Peter's interpretation of
Mary's utterance in (1b) not as conveying the encoded concept FLAT (as
stabilised in dictionaries, for example) but the related ad hoc
concept FLAT*, with a more restricted encyclopedic entry and a
narrower denotation, constructed ad hoc for this particular occasion,
as paraphrased in (1c):

(1a) [Peter and Mary are discussing their next cycling trip. Peter has 
 just said that he feels rather unfit].
(1b) Mary: ''We could go to Holland. Holland is flat''. 
(1c) Holland is FLAT* (where FLAT* is the meaning indicated by 'flat', 
 and is such that Holland's being FLAT* is relevant-as-expected in the 

Mary's word ''flat'' is taken here to warrant only those effects which
make it worth processing in the specific context (1a). Since there is
no one-to-one correspondence between the dictionary entry ''flat'' and
the ad hoc concept FLAT*, its relationship is a matter of resemblance,
rather than pure description. FLAT* is an unglossed version of the
word 'flat' which retains only the attributes which are relevant for
the processing of the utterance. On a different context, Mary may well
use the word ''flat'' not to communicate the concept FLAT* but another
ad hoc concept FLAT** whose specific attributes will be relevant in
that context but not in context (1a). Obviously, the biologically
rooted relevance-seeking mental procedure should guide the hearer on
every occasion to the recovery of the intended CONCEPT* (metaphorical
or otherwise) based on the word encoding it.
A similar analysis has been provided for figurative language in
general and metaphors in particular, but it is not pursued in the book
as deeply as I expected. What is indeed pursued, and very well so, is
the account of all the aspects in which d.M. is wrong. I.M also
wittily uncovers d.M.'s own (typically deconstructivist) liking for
contradictions, for example when d.M. writes that ''it is impossible
to say whether denomination is literal or figural: from the moment
there is denomination, the conceptual metaphor of entity as difference
is implied, and whenever there is metaphor, the literal denomination
of a particular entity is inevitable'' (p. 148 of Allegories of
Reading, quoted on p. 67).
Chapter 6 (''Rhetoric as an Insurmountable Obstacle'', p. 84-106) also
focuses on d.M.'s insistence on the unpredictable and ambiguous nature
of language, the impossibility ''of making the actual expression
coincide with what has to be expressed, of making the actual sign
coincide with what is signifies. It is the distinctive privilege of
language to be able to hide meaning behind a misleading sign'' (p. 11
of Blindness and Insight, quoted on p. 84). As I pointed out above,
within RT it is claimed that utterances and written texts normally
underdetermine the thought(s) that the speaker or writer intends to
communicate, but the claim is made without such pessimistic
connotations. Speakers and writers do leave blanks to be filled
inferentially, but hearers and readers normally manage to reach an
adequate level of understanding so that communication can often be
labelled successful.
d.M. is clearly wrong when he argues that neither grammar nor intended
reference can contain the figural possibilities in language, so it is
impossible to decide between literal and figural readings. Needless
to say, utterances are often ambiguous and reference assignment is
hard due to many potential referents, but on every occasion the same
principle applies: the gap between the semantic representations of
sentences and the thoughts communicated by utterances is filled by the
inferential recognition of the communicator's intentions with the aid
of contextual information (p. 85). The same insufficient argument by
d.M. is found in his account of irony (''irony is something that
language does, rather than a conscious attitude on the part of a
language-user'', p. 100), clearly misguided if we contrast it to
current pragmatic theories such as the one suggested by RT: irony as
the ostensive communication of an echo and a parallel attitude of
dissociation towards the proposition expressed by the utterance.
In chapter 7 (''Words and the World: The Problem of Reference'', p.
107-130), I.M. continues with d.M.'s account of the ''slippery''
nature of language, especially when dealing with indexicals. Again,
''although... the relationship between words and things is
conventional or contractual rather than phenomenal or constitutive, it
does not follow that the referential function of language is always
unstable, or that we cannot express our thoughts'' (p. 107).
Chapter 8 (''Mechanical Performatives'', p. 131-151) is about speech
acts. The basic point of discussion is the codification of acts by the
grammar. In my opinion, some analysts of speech acts are often too
concerned with analysing what speakers do with language, instead of
guessing what intentions and attitudes underlie the public use of
language. Indeed, there are many linguistic devices to show what acts
are performed with the utterance (e.g. verbs such as warn, ask,
apologise...), but often these acts can only be grasped inferentially
if there are no linguistic cues. Inside an RT point of view, the
discussion is now centred upon which acts are ostensively communicated
and which are not (see, for instance, Nicolle, 2000). However, d.M.'s
concern is, rather, that ''language is capable of performing randomly,
mechanically, non-referentially and non-cognitively, entirely beyond
anybody's will or control'' (p. 132). The relevance-theoretic and
deconstructivist positions cannot be more different. The extreme case
of this differentiation is when d.M. treats texts as mechanical,
devoid of users' intentionality.
Chapter 9 (''The Madness of Words and the Enunciating Subject'', p.
153-175) is a continuation of the arguments against d.M.'s
deconstructivist tricks in his analysis of language. The mechanical
view of language places d.M. close to other analysts such as Saussure
and other French critics such as Foucault or Derrida. In all cases the
mechanic nature of texts dissociates them for their users. Culler's
quote on p. 155 is illustrative: ''we often think of the meaning of an
expression as what the subject or speaker 'has in mind'. But as
meaning is explained in terms of systems of signs -systems which the
subject does not control- the subject is deprived of his role as
source of meaning''. The result is a excessive emphasis on ''intentio
operis'' over ''intentio auctoris'' or ''intentio lectoris''.
Finally, chapter 10 ('''When Lucy ceas'd to be', p. 176-195) collects
all the interpretive hypotheses which analysis and critics have
suggested for Wordsworth's short lyric ''A slumber did my spirit
seal''. All the opinions about what the right interpretation of the
poem is confirm the unpredictable nature of literary communication, in
which ''there is a difference... between language directed at known
addressees and language directed at addressees most, if not all, of
whom will be unknown, which is the typical case with literature. Only
in the former case can communicators make ready assessments concerning
which contextual assumptions are easily accessible to their
addressees'' (Pilkington, 2000: 82). In literature, it is more
difficult (if not impossible) to make assumptions mutually manifest,
and thus a greater load of responsibility is laid upon the reader in
extracting the intended (or, alternatively, his/her own)
interpretation of the text plus whatever feeling and emotions are
associated with the comprehension of the text.


I.M. provides a good number of valid arguments against d.M.'s
''deconstructionist tricks'', as I call them. Trying to subvert, in a
''topsy-turvy'' kind of way, the analysis of where the meaning of text
is supposed to lie, is a losing battle in my opinion. Indeed, the
relevance-oriented cognition-centred arguments accounting for what
goes on in the reader's mind are far more scientific than the
philosophically-oriented ones provided by d.M., even if some of d.M.'s
ideas are, no doubt, thought-provoking and some are even illuminating.
It could even be argued that I.M. knew this battle was lost long
before he undertook this analysis of relevance theory and
deconstruction. And I must say this is not the only case. In a
previous publication of mine (Yus, 1998), I also compared Derrida's
deconstructionist approach to the oral-written interface (basically
centred upon the logocentrism/phonocentrism debate) to a pragmatic
account of the similarities and differences in the contextual support
available for these two types of discourse. I also knew this was going
to be a losing battle for deconstruction. Derrida's attempts to place
written communication in higher order compared to oral communication
was unsuccessful. He does cling to deconstructionist tricks, for
instance when he states that if writing is a supplement to oral
communication, it can only supplement speech if speech itself is
limited and not self- sufficient, only if there is a fault in speech
that allows writing to become its supplement. Again, the pragmatic
research on the richness of contextual support in face-to-face
situations provides good arguments to invalidate these
deconstructionist tricks.


Carston, R. (2002) Thoughts and Utterances. Oxford: Blackwell.

Kenesei, A. (2003) ''Review of I. MacKenzie's Paradigms of Reading.
Relevance Theory and Deconstruction.'' The Linguist List 14.2104, 8-8-
2003. Available at

Nicolle, S. (2000) ''Communicated and non-communicated acts in
relevance theory''. Pragmatics 10(2): 233-245.

Pilkington, A. (2000) Poetic Effects. A Relevance Theory Perspective.
Amsterdam: John Benjamins.

Sperber, D. (1996) Explaining Culture. A Naturalistic
Approach. Oxford: Blackwell.

Sperber, D. and D. Wilson (1986/95) Relevance: Communication and
Cognition. Oxford: Blackwell.

Wilson, D. and D. Sperber (2002) ''Truthfulness and relevance''. Mind
111, 443: 583-632.

Yus, F. (1998) La Preeminencia de la Voz. Alicante: University of
Alicante, Servicio de Publicaciones.

Yus, F. (2002) ''Review of A. Pilkington's Poetic Effects. A Relevance
Theory Perspective''. Journal of Pragmatics 34: 619-628.


Francisco Yus teaches pragmatics at the University of Alicante, Spain.
He has a PhD in linguistics and has specialised in the application of
pragmatics (especially relevance theory) to media discourses and
conversational issues. For instance, he has made two applications of
pragmatics to characters in alternative comics (Conversational
cooperation in alternative comics, 1995; El discurso femenino en el
c�mic alternativo ingl�s, 1998), proposed a pragmatic verbal-visual
model of communication in media discourses (La interpretaci�n y la
imagen de masas, 1997), studied the written-oral interface (La
preeminencia de la voz, 1998) and developed a pragmatic approach to
Internet-mediated communication (Ciberpragm�tica, 2001). Latest
research has to do with the application of relevance theory to the
analysis of misunderstandings and irony in conversation, as well as to
the production and interpretation of humorous discourses.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue