LINGUIST List 14.3105

Thu Nov 13 2003

Review: Pragmatics: Panther & Thornburg (2003)

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. Geert Br�ne, Metonymy and Pragmatic Inferencing

Message 1: Metonymy and Pragmatic Inferencing

Date: Thu, 13 Nov 2003 01:41:31 -0500 (EST)
From: Geert Br�ne <>
Subject: Metonymy and Pragmatic Inferencing

Panther, Klaus-Uwe and Linda L. Thornburg, eds. (2003) Metonymy and
Pragmatic Inferencing, John Benjamins Publishing Company, Pragmatics
and Beyond New Series 113.

Announced at

Geert Br�ne, Department of Linguistics, University of Leuven (Belgium)


This volume presents a collection of ten papers, most of which were
first presented at a workshop on ''Metonymy and Pragmatic
Inferencing'' at the 7th International Pragmatics Conference in
2000. In the introduction to the volume, the editors describe the
general aim of the collected papers as ''contributions to pragmatics
from a cognitive linguistics perspective'' (1). Indeed, given the
radical redefinition in cognitive linguistics of metaphor and metonymy
from the traditional view as figures of speech to a cognitive approach
that treats them as central cognitive construal mechanisms pervasive
in language and thought, a genuine pragmatic perspective emerges. More
specifically in the case of metonymy, the accepted view in cognitive
linguistics that metonymy is a reference-point phenomenon (Langacker
1993) in which one element of a cognitive frame or ICM (Idealized
Cognitive Models, Lakoff 1987) serves as an access point to a
different element in the same frame or to the frame as a whole,
implies a pragmatic force for metonymic reasoning. As Gibbs (1999)
argues, such a frame-based definition naturally leads to the
hypothesis that metonymic reasoning or reference-point reasoning is
the driving force behind conversational implicature. Where exactly
this interaction between pragmatic inferencing and the metonymic
processing of language (Gibbs 1999: 69) is to be situated, is the
central question of the volume.

In the introduction to the book by the editors, Klaus-Uwe Panther and
Linda L. Thornburg, present a state of affairs in the cognitively
oriented research on metonymy, with specific focus on the interaction
with pragmatic inference. The central notions that are clustered in
this overview are the conceptual nature of metonymy, the kind of
relation that is exploited in metonymy (contingency relation), the
strength of that relation, the connection between different types of
metonymy (referential, predicational, illocutionary), and the main
research question of the volume: the relation between metonymy and
pragmatic inferencing in implicatures and explicatures.

SECTION I: The place of metonymy in cognition and pragmatics

The volume is divided into four major sections, each focussing on a
different aspect of the interaction between metonymy and pragmatic
inferencing. Section I opens with a paper on ''Cognitive operations
and pragmatic implication'' by FRANCISCO JOSE RUIZ DE MENDOZA IBA�EZ
and LORENA P�REZ HERN�NDEZ. The authors provide a detailed
theoretical account, in which the role of metaphor and metonymy as
cognitive operations in pragmatic implication is explored. More
specifically, they argue that metaphoric and metonymic mappings need
to be included as mechanisms in the derivation of what is called
'explicature' in Relevance Theory (in contrast to
implicature). Explicatures are derived when an incomplete logical
form, an assumption schema, is developed into a full proposition (the
explicated meaning) through a number of cognitive processes, including
disambiguation, saturation, loosening, but also, according to Ruiz de
Mendoza and P�rez Hern�ndez, metaphor and metonymyy" In the case
of metaphor, the authors argue that the number of potential
explicatures correlates with the complexity of the metaphorical
mapping structure, ranging from one-correspondence to
many-correspondence metaphors (Ruiz de Mendoza 1998), and that the
principle of relevance guides the inferential process towards the
contextually most relevant explicature. Metonymy, on the other hand,
per definition revolves around a one-correspondence mapping between a
domain and a subdomain based on salience differences (Croft & Cruse,
in press), yielding only one potential explicature. Rather than
arguing for a strict division between both mechanisms, the authors
illustrate that there is a metaphor-metonymy continuum, which is
''crucial to determining the communicative effects of the explicatures
derived by means of metaphoric and metonymic mappings'' (35). In the
last part of the paper, more complex patterns are analyzed, including
double metonymic mappings and cases of the interaction between
metaphor and metonymy in generating explicatures.

SEANA COULSON and TODD OAKLEY explore the role of metonymy in
conceptual integration networks or 'conceptual blends', the theory of
which has been developed by Fauconnier & Turner (1998, 2002), Coulson
(2000), among others. Blends, according to this framework, are complex
representations involving structure from multiple domains (or 'mental
spaces') connected via different relations, including (among others)
analogical, metaphorical and metonymical mappings. In order to
regulate the interaction between different mental spaces and their
inter- and inner-space connections, blending theory postulates a
number of optimality principles that guide and constrain the
integration process. One of the central principles that is often
satisfied through the use of metonymic reasoning potential, is that of
'integration', which dictates that representations in the blended
space should be manipulable as a single unit. Structure from different
mental spaces can be compressed into a single coherent, integrated
representation in the blend partly because ''conventional metonymies
help speakers to unpack mappings from the compressed element in a
blended space to its various counterparts in other spaces in the
network'' (77). Through the analysis of highly disparate integration
networks, including a newspaper headline, a metaphorical passage from
Ernest Hemingway, the English idiom 'Blowing your own Horn' and a
sculpture by Viktor Schreckengost, the authors illustrate that
metonymically structured blends often involve a trade-off between
different optimality principles. The analysis leads the authors to the
conclusion that metonymic connections in the blend play a crucial role
in providing access to the web of mental spaces that are activated in
the interpretation process.

In the last paper of the first section, ANTONIO BARCELONA explores the
metonymic basis of pragmatic inferencing through an analysis of jokes
and witty anecdotes. The central question that drives the study is how
complex inferential structures can be processed with apparent ease. A
locus that is particularly suited for this research question is humor,
one of the most elusive types of language use involving complex, often
ad-hoc inferences. Barcelona argues that it does not suffice to merely
uncover the inferences involved by means of the communicative
principles described in pragmatics (e.g. through the violation of the
Gricean maxims). Rather, the question needs to be pursued what guides
and motivates these inferences, so as to appear natural and
effortless. A detailed analysis of the material shows that
pre-existing metonymic connections provide essential clues for the
hearer/reader towards the most plausible inferential interpretation,
and in consequence ''provide the very 'skeleton' of pragmatic
inferencing'' (97). This conclusion is very much in line with the
insights from the other two papers discussed above. In contrast to
Ruiz de Mendoza and P�rez Hern�ndez, however, the cases discussed
by Barcelona belong to the realm of pragmatic implicature rather than
explicature. By showing that implicature derivation is
guided/facilitated by metonymic connections as well, this argument
extends the view of metonymy and metaphor to a broader status as
''pragmatic implication-deriving mechanisms''.

SECTION II: Metonymic inferencing and grammatical structure

In the first chapter of the section on metonymy in grammatical
structure, ANATOL STEFANOWITSCH provides a ''construction-based
approach to indirect speech acts''. More specifically, he treats
conventionalized indirect speech acts (ISAs) of the type in
'disguised' requests like ''Can you pass the salt?'' as cases of
constructions in the sense of Goldberg's Construction Grammar (CG). By
means of a number of formal arguments, Stefanowitsch illustrates the
(relatively) independent and unpredictable status of conventionalized
ISAs in comparison to the direct speech acts they derive from, and to
non-conventionalized ISAs with the same illocutionary force. This
leads him to the conclusion that they qualify as
constructions. However, CG does not treat the constructional lexicon
as an unstructured glossary, but rather as a network of interrelated
constructions, connected via so-called 'inheritance links'. In order
to cover the relationship between direct speech acts and their
conventionalized indirect counterparts, Stefanowitsch proposes to add
to the existing list of four types of inheritance links one additional
type, viz. the metonymy link. On the basis of Panther and Thornburg's
theory of speech act metonymies (see next paper to be discussed), it
is argued that the metonymic linking type ''allows us to capture the
(partial) motivation behind ISA constructions, while at the same time
acknowledging their independent status'' (108). The paper closes off
with some tentative neurolinguistic evidence (particularly the
behavior of people with right hemisphere damage confronted with
(in)direct requests) in support of the central claim of the study.

''Metonymies as natural inference and activation schemas: The case of
dependent clauses as independent speech acts'' analyzes the
exceptional status of independent speech acts that have the syntactic
form of dependent clauses, like in expressions of surprise ''Why, if
it isn't Susan!'' and indirect requests such as ''If you will all be
quiet now''. Focussing on independent if-clauses that lack the
expected explicit consequent clause, the authors pursue the question
of what motivates the particular pragmatic meaning this type of
constructions can conventionally get. By means of their own,
previously elaborated approach to speech acts as scenarios with
metonymic structure (Panther & Thornburg 1998), a classification of
conventionalized pragmatic functions of independent if-clauses
-constructions in the sense of Goldberg (1995)- is
introduced. Basically, three such metonymically motivated functions
can be distinguished: first, the deontic use as directives (requests
as in ''If you will all be quiet now''), offers, suggestions and
wishes; second, expressive use to communicate surprise (as in ''Why,
if it isn't Susan!''), disapproval, etc.; and third, an epistemic use
for reasoning to an unknown conclusion (as in counterfactual uses like
''If Sonia and I hadn't made love...''). Much in the line of
Stefanowitsch (previous paper), Panther and Thornburg conclude that
many of the 'independent dependent clauses' functioning as speech acts
have a degree of conventionalization (and independence) that grants
them the status of construction, without, however, losing their
metonymic motivation.

In the third paper of the section, ''Metonymic pathways to
neuter-gender human nominals in German'', KLAUS-MICHAEL K�pcke and
DAVID A. ZUBIN explore the systematic (and productive) character of
the use of neuter-gender classification for female human beings in
German. The study shows that the distribution of these
human-reference nouns with neut-gender is not random. Based on
Lakoff's (1987) model of metonymic ICMs, the authors argue that these
neut-gender nominals trigger a metonymically structured social
stereotype. More specifically, nine types of metonymic motivation,
each evoking different perspectival metonymic ICMs, are presented that
can account for 80% of the collected data. Most of these principles
have a clear downgrading character, exploiting negatively valued
cultural stereotypes. In the last part, the authors pursue the
question to what extent this play on metonymic ICMs for females is
exploited for semantic-pragmatic purposes in discourse processes, and
more specifically in lexical choice and pronominal anaphoric
reference. Two highly disparate text types, an excerpt from a short
story by Karl Waggerl and some passages of journalistic writing, are
analyzed to illustrate the marked semantic status of the neut-gender

SECTION III: Metonymic inferencing and linguistic change

The two contributions in section III explore the role of metonymy from
a diachronic perspective. DEBRA ZIEGELER's paper ''The development of
counterfactual implicatures in English. A case of metonymy or
M-inference?'' deals with the cognitive mechanisms that build the
foundation of counterfactual implicatures in the use of modal
expressions expressing ability, like 'was/were able to', 'could', and
'had the ability to'. According to Levinson (1995), a sentence like
''John could solve the problem'' implicates the actuality of the
complement clause (John did solve the problem), whereas in the
periphrastic alternative ''John had the ability to solve the
problem'', non-factuality of the event is inferred (John did not solve
the problem). The counterfactual implicature in the latter example, in
Levinson's view, is guided by what he calls M-inferences, an inference
that accompanies the marked expression in contrastive sets of marked
and unmarked alternate expressions. Ziegeler counters that it is
highly unlikely that only the relative markedness of an expression
leads to contrasting implicatures, and gives several arguments that
support that claim. Rather, she argues that counterfactual
implicatures developed diachronically as the result of metonymic
inferencing. In line with, and in addition to Panther and Thornburg's
scenario-approach to metonymy, this counterfactual metonymy can be
labelled POTENTIALITY FOR NON-ACTUALITY. By means of a survey of
diachronic texts, Ziegeler empirically investigates what (con)textual
factors contributed to the development of the counterfactual metonymic
implicature. It is concluded that ''counterfactual implicatures have
developed from metonymic extensions in past modal verbs, especially
with perfect auxiliaries, e.g. 'could have + V-ed', in which the
modal clause may stand alone for the entire construction containing a
modal clause and followed by an adversative or contrastive clause''
(199), and can hence be treated as a part-for-whole metonymy.

In the second paper of the section, SHIGEKO OKAMOTO explores the role
of ''Metonymy and pragmatic inference in the functional reanalysis of
grammatical morphemes in Japanese''. The specific topic of the paper
is the reanalysis of complementizers (COMPs) marking subordinate
clauses as sentence-final particles (SFPs) with a conventionalized
modal meaning. Okamoto focuses on one specific morpheme in Japanese
that can be used either as a complementizer or as a sentence-final
particle, viz. 'koto'. As a complementizer it marks a clause boundary
and indicates an abstract or indirectly perceived event/state. As SFP,
it has primarily a modal function, expressing either an exclamation or
an order, in what is labeled 'subordinate-clause-as-main-clause
constructions' (SCMCCs). The central question is how 'koto' has come
to be used as a SFP, and what motivates the development from a COMP to
the SFP function. It is argued that the reanalysis has developed from
a metonymically motivated conversational implicature (in Panther and
Thornburg's frame-based approach) to a conventional implicature, thus
yielding the grammaticalization effect. Possible motivations for the
reanalysis are situated in the social and rhetorical domain: the use
of a subordinate clause is argued ''(1) to foreground the information
in the ''original'' complement as the most important part of the
message, [and] (2) to bring about certain expressiveness, that is, to
perform a given speech act with particular stylistic nuances'' (215).

SECTION IV: Metonymic inferencing across languages

The last two chapters of the volume provide a more typological
perspective to the study of metonymy, in that they focus on
cross-linguistic differences in the application and force of metonymic
principles. G�NTER RADDEN and KEN-ICHI SETO's paper on ?Metonymic
construals of shopping requests in HAVE- and BE-languages'' starts
from the recent insight in cognitive linguistics that indirect speech
acts, and more specifically indirect requests, are structured
metonymically (Panther and Thornburg, 1998, this volume). The use of
indirect requests in different languages is investigated for one
specific setting, viz. requests in a shopping scenario. It is shown
that different languages have a different metonymic construal of
indirect requests, in that they refer to different subevents of the
commercial transaction scenario (availability, transfer, reception,
resulting possession). This distribution is argued to be related to
the typological differences between the languages. For example, there
is a correlation between the choice for a specific metonymic reference
point in the transaction scenario and the way a language construes
possession. Two language groups are distinguished on the basis of this
distinct construal: HAVE- and BE-languages. For instance,
HAVE-languages tend to metonymically access the availability aspect of
the transaction scenario by referring to possession, like in English
'Do you have 40-Watt light bulbs?'. BE-languages, in contrast,
typically have EXISTENCE rather than possession as metonymic reference
points to availability (e.g. Japanese, Chinese, Korean, Finnish,
Hungarian, etc.). Next to the different distribution in the metonymic
construal of the availability aspect of the commercial transaction
scenario, both groups also display a distinct metonymic use of the
other stages of the scenario (transfer, reception). The differences
in the request strategies are linked to different cultural models of
politeness (indirectness vs. deference) and a different construal of
events (action vs. process).

The last chapter of the volume, MARIO BRDAR and RITA BRDAR-SZAB�'s
contribution on ''Metonymic Coding of Linguistic Action in English,
Croatian and Hungarian'' deals with the English use of predicational
adjectives for reporting linguistic action (e.g. 'Arthur was brief
about his other teachers') and its counterparts in Croatian and
Hungarian. These constructions are argued to be based on a MANNER FOR
(LINGUISTIC) ACTION predicational metonymy, since they specify the
manner in which an action is conducted, but not the verbal action
itself. In a crosslinguistic comparison, it is revealed that Croatian
and Hungarian exhibit only some of the subtypes that are found in
English, and impose more specific constraints on the use of
predicational metonymies than in English. In the case of referential
metonymies, these constraints do not seem to hold. According to the
authors, these crosslinguistic regularities ''seem to indicate that
the distinction between referential, predicational and illocutionary
metonymies may be an important parameter in establishing a typology of
metonymies.'' (251). What is more, a tentative conclusion is proposed
that there may be an implicational relationship between the use of
referential and predicational metonymies, meaning that a language that
makes extensive use of predicational metonymy will also have a wider
use of referential metonymy. Other languages will restrict the scope
of metonymy primarily to the referential use.


The volume presents an excellent overview of the current debate in the
new trend of Cognitive Pragmatics. Given the observation in cognitive
linguistics as well as recent pragmatic approaches that a distinction
between semantics and pragmatics is no longer tenable, the
contributions in the volume together provide a much-needed attempt to
'reconcile' both research traditions. Despite the underlying
assumption in recent cognitive linguistic accounts that metonymy is a
driving force in inferencing, this interplay had not received
sufficient scrutiny until the publication of this volume. The most
essential conclusion that can be drawn from this endeavor is that a
cognitive linguistic approach in terms of metonymic reasoning does not
necessarily preclude a pragmatic approach in terms of conversational
implicature. Rather, both approaches can be combined to yield an
encompassing, cognitive-functional approach.

Still, the volume does not present a closed chapter in the research on
metonymic reasoning. On the contrary, linking research on cognitive
mechanisms and pragmatic inferencing raises the question what guides
and constraints the choice for a specific metonymic construal. One
possible solution to this problem, proposed by a number of
contributors to the volume, is adding a relevance theoretic
perspective to the account. This remains an insufficiently addressed
question that will have to be the focus of future research. Also,
there does not seem to be a consensus among the authors on the exact
role of metonymy in the inferential process. Whereas some seem to
suggest that all pragmatic inferencing is metonymically driven, others
reserve a more specific, well-defined place for metonymy. In my
opinion, there is still need for further discussion on the exact scope
of metonymy. Nevertheless, this volume promises to be one of the
pioneering works in the growing research on the pragmatics-cognition


Coulson, Seana (2000): Semantic Leaps. Frame-shifting and Conceptual
Blending in Meaning Construction. New York/Cambridge: Cambridge
University Press, 2000.

Croft, William & Alan Cruse (in press): Cognitive Linguistics
[Cambridge Textbooks in Linguistics]. Cambridge: Cambridge University

Fauconnier, Gilles & Mark Turner (1998): ''Conceptual integration
networks''. Cognitive Science 22:2, 283-304.

Fauconnier, Gilles & Mark Turner (2002): The Way We Think. Conceptual
Blending and the Mind's Hidden Complexities. New York: Basic Books.

Gibbs, Raymond W., Jr. (1999): ''Speaking and thinking with
metonymy''. K.-U. Panther & G. Radden (eds.), Metonymy in Language and
Thought [Human Cognitive Processing 4], 61-76.
Amsterdam/Philadelphia: John Benjamins.

Goldberg, Adele (1995): Constructions: A Construction Grammar Approach
to Argument Structure [Cognitive Theory of Language and Culture].
Chicago/London: University of Chicago Press.

Lakoff, George (1987): Women, Fire and Dangerous Things. What
Categories Reveal about the Mind. Chicago: University of Chicago

Langacker, Ronald W. (1993): ''Reference-point constructions''.
Cognitive Linguistics, 4, 1-38.

Levinson, Stephen C. (1995): ''Three levels of meaning''. F.R. Palmer
(ed.), Grammar and meaning: Essays in honor of Sir John Lyons'',
90-115. Cambridge: Cambridge University Press.

Panther, Klaus-Uwe & Linda L. Thornburg (1998): ''A cognitive approach
to inferencing in conversation''. Journal of Pragmatics, 30, 755-769.

Ruiz de Mendoza, Francisco J. (1998). ''On the nature of blending as a
cognitive phenomenon''. Journal of Pragmatics, 30, 259-374.


Geert Br�ne is a PhD student in linguistics at the University of
Leuven (Belgium). He is currently preparing a dissertation on a
cognitive linguistic approach to humor interpretation (supervised by
Kurt Feyaerts). His main research interests are cognitive semantics,
cognitive stylistics, (linguistic) humor theories and German
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue