Pragniatics

Cognition, Context & Culture

Laura Alba-Juez
J. Lachlan Mackenzie

'This book is more than a university course on pragmatics. With exq,uisite

balance, it offers a fascinating journey into the often complex relationship
between linguistic form and what it can communicate in relation to cultural
contexts and the mind. It further combines cutting-edge erudition with
reader-friendliness and provides students -and any reader- with an abundance
of insights for critical reflection on topics that are still undergoing exploration
in linguistic theory. I strongly recommend making extensive and continued use
of this great work.'
Ricardo Mairal, Full Professor of Linguistics,
UNED, Madrid, Spain

•••
'A very readable and didactically strong introduction to Pragmatics. Besides
offering a complete overview of the field, this book explores connections to
cognitive, cultural, and grammatical approaches to language, thus contributing
to an integrated theory of human communication.'
Ad Foolen, Associate Professor of Linguistics,
Radboud University, The Netherlands

·A carefully-crafted account of pragmatics that will guide readers through the

discipline's many intricacies.'
Angela Downing, Full Professor of Linguistics,
Universidad Complutense de Madrid, Spain
IIIIII IIIII IIIIIIIIII II IIII II IIIIIIIIIIIII IIIIIIII III Table of contents IX

TABLE OF CONTENTS

Acknowledgements ...................... :.......................................................................... IX

Introduction .............................................................................................................. XI

1. Pragmatics: definition and scope ············:······················································ 1

• Main objectives of the chapter .............................................................................. 1
• Main topics of the chapter ...... ........ ........ ..................... ........... ..... ......... ... .............. 1
1. 1. What is pragmatics and why should we develop pragmatic awareness? ..... 2
1.2. Context: a dynamic concept ............................................................................... 4
1.3. The underdeterminacy of language and its relation to context ................... 9
1.4. The scope of pragmatics: Main schools and central topics ......................... 12
1.4.1. Micro and macropragmatics ....................................................... :......... 14
1.5. Pragmatics and cognition ................................................................................. 15
1.6. Indirect meanings: why they are always pragmatic in nature ............... ..... 18
1.6.1. Conventionalized and grammaticalized in~rect meanings .............. 19
1.6.1.1 Conventionalization .................................................................. 19
1.6.1.2 Grammaticalization .................................................................. 21
1.7. Pragmatics and culture·······················.······························································ 23
1.8. Conclusion .......................................................................................................... 27
• Summary ................................................................................................................ 27
• Self-evaluation q_uestions .................................................................... ................. 30
• Developing our pragmatic competence: exercises for reflection
and amplification .................................................................................................. 32
2. What we do with language: Speech acts .................................................... 35
• Main objectives of the chapter ........ ..................... ...... ........... .. ........ .. ........ ...... ... 35
• Main topics of the chapter ............... ........... ..... ...... .......... ....................... ......... ... 35
2.1. On saying and doing .......................................................................................... 36
2.2. Austin's reaction to logical positivism and truth-conditional semantics .... 37
2.3. Performative acts .............................................................................................. 38
2.3.1. Types of performatives .......................................................................... 39
2.3.2. Felicity conditions .................................................................................. 42
2.3.3. Explicit and implicit performatives ...................................................... 43
2.4. Types of speech acts (Austin 1962): Locution, Illocution
and Perlocution .................................................................................................. 45
2.5. Form and function: Do they always coincide? .............................................. 47
X Table of contents /// ///// // //////////// ///// /// //// // /// ////// ////// //

2.6. Types of speech acts (Searle 1969) ................................................................. 51

2.6.1. Indirect speech acts ................................................................................ 55
2.7. The impact of Austin and Searle's work on linguistics ................................. 58
2.8. Conclusion ........................................................................................... :.............. 63
• Summary ................................................................................................................. 63
• Self-evaluation q,uestions ..................................................................................... 65
• Developing our pragmatic competence: exercises for reflection
and amplification ................................................................................................... 68

3. On referring, saying, and implicating: Reference and inference ...... 73

• Main objectives of the chapter ............................................................................ 73
• Main topics of the chapter .. .... .. .......... ...... ..... ... .. ............ .. ... ... .. ... ..... .. .. ....... ...... .. 73
3.1. Reference .. .......... .... ....... .. .............. ... ....... ............ ..... ....... ........ ............... .... ..... ... 74
3.1.1. Why is reference an essentially pragmatic mechanism? .................. 74
3.1.2. Types of reference .................................................................................. 76
3.2. Deixis: indexicals and context ......................................................................... 80
3.2.1. Types of deixis ···························································:···························· 82
3.2.2. Grammaticalized deictic meanings ...................................................... 86
3.3. Inference and implicit meanings ..................................................................... 91
3.3.1. Types of Inference .................................................................................. 91
3.3.2. Presupposition: shared assumptions and background
knowledge ................................................................................................ 92
3.3.2.1 Types of presuppositions .......................................................... 93
3.3.2.2 Grammar and presupposition .................................................. 96
3.3.3. Implicature ............................................................................................... 99
3.3.3.1 Conversational Implicature .................................................... 101
3.3.3.2 Other ways of not observing the maxims ............................ 103
3.3.3.3 Conventional Implicature ....-................................................... 104
3.3.3.4 How do conventional and conversational
implicatures differ? ................................................................. 105
3.3.3.5 Generalized and particularized conversational
implicature ............................................................................... 107
3.3.3.6 Neo-Gricean theories of implicature ..................................... 109
3.4. Conclusion ........................................................................................................ 114
• Summary ................ ........................ ....... ..................... .................. ........................ 115
• Self-evaluation q,uestions ................................................................................... 118
• Developing our pragmatic competence: exercises for reflection
and amplification (Chapter 3) ..................... ....................................................... 121
II IIII IIIIIIII IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII IIIIIII Table of contents XI

4. Politeness and Impoliteness: How to verbally reconcile

our interlocutors' wants with our own (or not) ..... ......... ...................... . 129
• Main objectives of the chapter .......................................................................... 129
• Main topics of the chapter ...... ....... .... ..... ........ ... ..... ......... ....... ........ ...... .... ......... 129
4.1. Politeness as an important element for the development
of pragmatic competence .... ... .. ..... ...... .. .... ...... ... ... .. ... ..... .... .... .. ..... .. .... .... .. .. ... ...... ... ..... .. 130
4.2. Approaches to politeness ................................................................................ 131
4.2.1. The conversational maxim view: Lakoff, Leech and Gu ................. 135
4.2.1.1 Robin Lakoff ............................................................................. 136
4.2.1.2 Geoffrey Leech's Politeness Principle (PP) ........................... 137
4.2.1.3 Yueguo Gu ................................................................................. 143
4.2.2. The face-saving view: Brown & Levinson
·, . ......................................... 143
4.2.2.1 The concept of face ... ............ ..... ........... .. .............. ........ ..... ..... 145
4.2.2.2 Politeness strategies .. ....... ... .. ...... .......... .................. ......... ....... 148
4.2.2.3 Sociological variables: D, P & R ............................................ 151
4.2.3. Criticisms leveled at both Leech's and B&L's approaches ............... 154
4.3. Theories of impoliteness ........................................ :.............. .......................... 157
4.3.1. Culpeper's approach ............................................. ................................ 158
4.3.2. Kaul de Marlangeon's approach .......................................................... 160
4.3.3. Further work on (im)politeness: the rise of the relational view .... 168
4.4. Conclusion ........................................................................................................ 169
• Summary ... ..... ... ... ..... ................ ........ ........ ....... ....................... .. ...... ..... ........ ........ 170
• Self-evaluation q_uestions ................................................... ................................ 173
• Developing our pragmatic competence: exercises for reflection
and amplification ...................................,............................................................. 176
5. Relevance: Processing effort and contextual effects of
communication ................................................................................................. 181
• Main objectives of the chapter .................... ..... ................................................. 181
• Main topics of the chapter ........................................................ .- ........................ 181
5.1. Relevance Theory ............................................................................................ 182
5.2. The Principle of Relevance .. ....... ........ ............ .. ... ... .... ........ ............. .. ............ 183
5.3. Relevance and cognition ................................................................................ 185
5.4. The notions of communication and context within.RT ................................ 187
5.5. The comprehension process ........................................................................... 190
5.5.1. Implicature and explicature ................................................................ 190
5.5.1.1 Explicature ............................................................................... 190
5.5.1.2 Implicature ... ...... ...................................................................... 194
XII Table of contents I I I I I///// I I I I// I I/ I I I I I/ I I I I I I/ I I I I I I I I I I// I/ I I I I I I I

5.6. Conceptual vs. procedural meaning/encoding .......................................... 196

5. 7. Relevance and grammar ............ ... ......... ... .... .. ... .... .. ........... ... .... .. .. ...... .. .... .... 199
5.8. Salience and inference .................................................................................... 202
5.9. Images and RT ................................................................................................. 204
5.10. Conclusion ...................................................................................................... 206
• Summary ..... ......................................................................................................... 207
• Self-evaluation q_uestions ................................................................................... 209
• Developing our pragmatic competence: exercises for reflection
and amplification ................................................................................................ 211
6. Other topics of interest in pragmatics ..................................................... 215
• Main objectives of the chapter .......................................................................... 215
• Main topics of the chapter ................................................................................. 215
6.1. Introduction ..................................................................................................... 216
6.2. Pragmatics as an important component of grammar ................................ 216
6.2.1. Pragmatics and syntax: how they interact ....................................... 216
6.2.2. Pragmatics and the lexicon ................................................................. 221
6.2.3. Pragmatics and phonology: prosody ................................................. 226
6.3. Stance and Evaluation as a dynamical, pragmatic system ....................... 230
6.4. Emotion: Intersubjective pragmatic meanings ........................................... 236
6.5. Historical pragmatics ..................................................................................... 242
6.6. Pragmatics and computational linguistics .................................................. 245
6. 7. Experimental pragmatics ............................................................................... 24 7
6.8. Cy berpragmatics ............................................................................................. 249
6.9. Conclusion .....................................................................- -. .................................. 251
• Summary .............................................................................................................. 252
• Self-evaluation q_uestions ................................................................................... 254
• Developing our pragmatic competence: exercises for reflection
and amplification ................................................................................................ 257
References .............................................................................................................. 259

Key to self-evaluation questions and exercises ......................................... 283

Index ........................................................................................................................ 300

I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I Acknowledgements XIII

ACKNOWLEDGEMENTS
As Stephen Levinson points out in his well-known book on pragmatics, writers
"must be intellectual sponges and spongers" (1983: xiv), and our case as
writers of this book has been no exception. We are grateful to all the authors
from whose works we have drawn, without whom our work would have been
impossible.
In particular, we want to thank our colleagues and friends Silvia Kaul de
Marlangeon, Victoria Escandell-Vidal, Francisco Yus and Ad Foolen, for having
sent and shared with us not only their published books and articles, but also some
of their unpublished work. Sil':ia and Victoria were also kind enough to read the
first manuscript of two different chapters, and their comments on their respective
topics of expertise were of invaluable help.
We also owe a debt of gratitude to Joaq,uin Armijo, for his hard work and
artistry in designing the cover and the illustrations in the different chapters,
and to Cristina Sanchez, of McGraw Hill, for her guidance through the process of
preparing the volume.
Finally, due acknowledgement has to be given to the funding received from
the Spanish Ministry of Economy and Competitiveness through the EMO-
FUNDETT project (FFI2013-47792-C2-1-P PI Laura Alba-Juez) and the project
The construction of discourse as social interaction: Contrastive implications
and applications (FFI2013-40517-P, PI Maria de los Angeles Gomez Gonzalez).
I I I I I I I I I/ I I I I I I I I I I/ I I I I I/ I I I I I I I!// I I I I I I I I I I I I I I I I I I I I Introduction XV

INTRODUCTION
Pragmatics is a relatively new field of research in the history of linguistic studies,
but its findings and scope are so relevant to the understanding of human language
that it would nowadays be unthinkable to make a serious analysis of any
language without taking pragmatic considerations into account. Pragmatics is
about meaning, but as we shall see throughout this book, it goes beyond the
traditional, truth-conditional approaches to semantics. It deals with a broader
body of information that draws not only on the text but also on the discursive
context, and depending on the perspective taken by its practitioners, its scope
may be limited to a micropragmatic analysis or may be macropragmatic in
nature, in which case - as we shall point out - the field is difficult to distinguish
from that of discourse analysis. This book attempts to explain these perspectives
and to bring them together, as well as presenting the major topics and issues that
have been explored within the field.
The early pragmatic approaches to language came from the field of
philosophy, and therefore many of their main concepts and tenets are
philosophical in nature and may be difficult to grasp·on a first encounter, but this
is also one of the reasons why the study of pragmatics provides us with
fundamental information not only about language but also about human nature
in general. In this respect, it can be said that anyone who embarks on the study
of pragmatics will simultaneously embark on an inner voyage that will lead to a
better understanding of their own motives, intentions and reactions when
interacting with other people.
The main aim of this book is, therefore, to present the reader/student with
the main concepts, tenets and perspectives of pragmatics as they have been
expounded by the main authors in the field and at the same time to contribute our
own perspective to the different issues discussed. In addition, the volume has
been conceived as a university course that can be used in both traditional and
distance education environments. It is precisely because the possibility of
autonomous learning was contemplated that the book contains self-evaluation
q,uestions and a body of exercises with their corresponding keys in its final pages.

Chapter overview and organization

This book has six chapters. The first presents the two main perspectives on
pragmatics and seeks to define and delimit the field, bearing in mind that this is
no easy task, considering the different approaches taken by the scholars working
on the subject. In addition, we discuss the three crucial concepts profoundly
intertwined with pragmatics which form the subtitle of this book: cognition,
XVI Introduction /////////////////////////////////////////////////////////

contex t and culture. Chapters 2 and 3 present the reader with other concepts and
topics that are fundamental for the comprehension of the way human language
works, namely speech acts, reference a;nd the different types of infere_nce.
Chapters 4 and 5 deal with two of the most influential pragmatic theories
developed to date, Politeness Theory and Relevance Theory. Finally, Chapter 6
looks into both the micropragmatic and the macropragmatic approaches, first
dealing with pragmatics as an important component of the grammar of any
language and its relationship to other levels of linguistic analysis (syntax,
lexicon, phonology), going on to explore a) two functions of language that have
gained considerable attention from pragmaticists in the 2ist century, the
evaluative and the expressive/emotive functions, and b) four relevant subfields
that have emerged from the original, philosophical-linguistic conception of the
discipline, namely historical, computational, experimental and cy berpragmatics.
All chapters have the same overall organization, with the following fixed
sections:
• Main objectives of the chapter
• Main topics of the chapter
• Main body of the chapter, divided into sections
• Conclusion
• Summary
• Self-evaluation q,uestions
• Developing our pragmatic competence: Exercises for reflection and
amplification

At the end of the book there are keys to both the self-evaluation q,uestions
and the exercises from each chapter. These have been especially designed for
those readers who want to approach the study of the subject within distance
education environments or who simply want to approach the study of pragmatics
in an autonomous manner.
 PRAGMATICS:
DEFINITION AND SCOPE
//////////////////////////////////////
MAIN OBJECTIVE OF THE CHAPTER: To define and delimit the field of
linguistic pragmatics

MAIN TOPICS OF THE CHAPTER:

• Pragmatics as a sub-discipline of linguistics: origins and scope
• Context as a dynamic concept
• Language underdeterminacy
• Main approaches to and central topics in pragmatics
• The relationship of pragmatics to cognition
• Indirect meanings
• Conventionalization and grammaticalization
• Pragmatics and culture
2 Pragmatics: definition and scope /!l/!/ll!///l//ll/l//ll!I/III/I///////I/

I.I. WHAT IS PRAGMATICS AND WHY SHOULD

WE DEVELOP PRAGMATIC AWARENESS?
When defining pragmatics, as when defining any other field of knowledge, the
approach or perspective taken for its study will undoubtedly shape and influence
the definition finally adopted. As will be made apparent in this chapter, pragmatics
is a complex, dynamic discipline whose boundaries are far from clear, and which
is therefore not easy to define in static, absolute terms.
Let us start our discussion by saying that pragmatics is a subdiscipline of
linguistics that can be traced back to the classical traditions of rhetoric and
stylistics, but whose status as a modern, independent field of study was not
acq,uired until the first half of the 20 th century, when Morris (1938: 21) wrote
about the "pragmatic dimension of semiosis" as "the relation of signs to
interpreters". Indeed, pragmatics does not deal with abstract idealizations of how
language is structured, or prescriptive rules about how language should work,
but rather with how, when, why, etc. speakers/writers (Morris's 'interpreters')
actually use language for different purposes. One thing we know for sure:
Pragmatics deals with meaning. But the concept of meaning in pragmatics is
much broader than in traditional, truth-conditional semantics. 1 Meaning in
pragmatics goes further than the word or the sentence, to be regarded as a
crucial aspect of the whole social, cultural and even cognitive context. We
should then distinguish between a) the basic meaning of an expression
resulting from its syntactic structure (i.e., Grice's 1967 timeless meaning), and
b) what someone means by using such an expression in a given situation.
Semantics is mainly concerned with a), while pragmatics is mainly concerned
with b ). Thomas (1995) refers to the former as 'abstract meaning' and to the
latter as 'utterance' or 'contextual' meaning. Leech (1983: 6) explains this
difference very clearly and in a nutshell, when pointing out that the difference
between semantics and pragmatics lies in two different uses of the verb to
mean. Semantics will try to answer the q,uestion What does X mean?, while
pragmatics will try to find an answer to What did you mean by X? But in spite
of these differences, the boundaries between the two subdisciplines of
linguistics are not clear-cut: they share a great deal of common ground and
thus it can be said that, far from being opposed, they not only complement
each other but even overlap in some respects.

1 Truth-conditional semantics is an approach to semantics that accounts for the meaning

of propositions in terms of the conditions under which they are logically true. The first
attempts to account for meaning on the basis of truth are attributed to Donald Davidson.
See Davidson (1967).
//I/I/I//////I///III/III////////I/I///// Pragmatics: definition and scope 3

Therefore, when doing pragmatics we are taking a theoretical and

methodological approach to language that focuses more on context and actual
linguistic behavior than on some abstract, idealized system. This view is
inevitably linked to a functional perspective on language, as opposed to a formal
one. Taking a functional view entails giving paramount importance to the
purposes and functions of language, including all of its uses. Notwithstanding, a
pragmatic approach to the study of language may (and, in fact, should) take into
account many of its formal aspects, because they also form part of and contribute
to a satisfactory determination of the meaning of utterances.
Mackenzie (2016: 470) points out that all functional views of language share
the assumption that languages in human societies "have the primordial function
of permitting sophisticated communication of ideas and feelings" and this is
precisely what pragmatics allows us to do: to understand communication by
trying to work out not only the literal meaning of words and sentences, but also
the hidden, indirect or implied meanings of utterances and/or whole texts. Indeed,
one interesting aspect of the study of the pragmatic dimension of language is the
fact that it can help us understand how human beings can "say what they mean
without meaning what they say" ( as Cutler 1974 put it in the title of her famous
article on ironic meanings).
And even if pragmatics as a field is not easy to define, it is undisputed that
all ( competent) users of any given language share a vast array of both general
and specific pragmatic knowledge without which it would be impossible for them
to interact. This kind of knowledge allows us, among other things, to interpret the
same utterance in completely opposite ways if said in different situations, as is
the case of How clever of her in (la) and (lb), taking into account that in (la)
being able to work out the solution to a math problem is in general considered as
something positive, while in (lb) both Susan and Peter have a negative opinion
of John, and thus, in this particular case, they both think that Ann has made the
wrong choice. Thus, their background knowledge of the situation and their
shared pragmatic knowledge 2 are both key factors to interpreting Peter's
utterance in a literal way in (la) and as an ironic remark in (lb):
(1) a. Susan: Ann has finally worked out the solution to that difficult math
problem.
Peter: How clever of her!
b. Susan: Ann has finally decided to marry John.
Peter: How clever of her!

2 Such as, for instance, the knowledge of the fact that in English (as in all languages, it would
appear) one can say something and mean the opposite (or something very different from
the literal meaning), given the appropriate discursive conditions.
4 Pragmatics : definition and scope /!llllll/l!ll!llllllllllll!lllll/ll!IIII

But pragmatic knowledge is necessary not only for understanding complex

and intricate ironic meanings, but also for such simple things as disambiguating
the real person or thing behind a referring term or a deictic 3 pronoun. For
instance, knowing the syntactic properties of pronouns in English is not enough
to interpret you or him in (2):
(2) You and you, but not him, will come to London with me.
We know that both pronouns could be used to identify different referents in
different contexts, and therefore we need contextual information about the
situation in which they are used, as well as some knowledge about the speaker's
intention.
Normally, the speakers of a language use their pragmatic knowledge (and
thus work out the meaning of utterances and whole discourses) in an unconscious
way. However, a conscious management of pragmatic resources and strategies
can be of great help not only in the obvious case of a learner of a foreign language
and culture, but also in cases of intracultural communication, where a skillful use
of these resources will most surely bring about payoffs of various kinds. When it
comes to achieving successful communication, developing pragmatic awareness
is a crucial factor: Not only is it a sign of high communicative competence, but
also of social and emotional intelligence. 4
Pragmatic knowledge and skills, therefore, are indispensable tools for any
communicatively competent speaker of a language. But before discussing the
possible definitions and our own view of pragmatics, it is worth looking into
some key points and aspects related to the field, such as the concept of context,
the underdeterminacy of language, or the relation between pragmatics and
cognition. We now turn to these matters.

1.2. CONTEXT: A DYNAMIC CONCEPT

Everything people do, whether they are speaking or not, occurs in a context.
Context has a powerful effect on discourse, not only for the hearer as he 5
interprets a message but also for the speaker as she formulates it. Contexts are
far from static: the contextual elements of any given situation may change once
or several times in the course of a single conversation or any other type of
discourse intera;ction. Even within a single utterance, the context is changed by

3 See Chapter 3.
4
For further treatment of the concept of emotional intelligence, see Goleman (1995).
5 For organizational purposes, we refer to the speaker as "she" and to the hearer as "he".
l/ll/ll!l/l/l/l/ll/ll!/II/I/II/I//I//II/ Pragmatics: definition and scope 5

what the speaker says, affecting both how she can complete her utterance and
the ongoing interpretation by the hearer. By way of example, although (3a)
and (3b) contain the same propositional message, they are (and have to be)
encoded in different ways due to the particular contextual circumstances in each
case:
(3) Tom is very happy because he has won the lottery and delivers the good
news to a) his wife and later to b) his boss.
a) To his wife, at home: Honey, you won't believe this: We're rich!!! We won
the first prize in the lottery!! Yoohooo!! We won't need to work anymore
for the rest of our lives!!
b) To his boss, at the company's headquarters: Mr. Williams, I'm very happy
to inform you that I have won the first prize in the national lottery, and
as a conseq_uence I'm giving you a month's notice of my resignation from
my position in this company. As you will understand, it will not be
necessary for· me to continue working here or indeed in any other
company.
But what do we exactly mean by contex t, and what types of context can
there be? We shall refer to context as both the linguistic and extra-linguistic
information affecting the meaning that is finally settled on by both speaker and
hearer. Fetzer (2004: 4-12) views context as a multifaceted phenomenon which
influences the connection between language and its use, and identifies several
types of context, namely the linguistic, the social, the socio-cultural and the
cognitive context. But apart from these, we believe there is another, very relevant
type that should be considered, namely, the emotional context. All of them we
shall explain in the fallowing paragraphs, going on to show how each type of
context is interrelated with the other types.
The linguistic contex t covers the preceding and upcoming words and
sentences that condition the ongoing utterance (sometimes called the co-text)
but much more besides. The time and location of utterance are also included in
the linguistic context since they affect the understanding of words such as now
and here; so is the identification of the speaker and hearer(s), which determines
the interpretation of pronouns like I and you. These factors together make up the
speech situation. Whereas it is uncontroversial that current utterances are
affected by what went before, it needs to be stressed that predicted follow-up
contributions also belong to the linguistic context: for example the nature of a
desired response will determine the form of a q,uestion. In other words, utterances
are both constrained by contexts and themselves build new contexts. Linguistic
context also encompasses more global settings: for example, the genre 'job
interview' provides a context for (and limits) the exchange that takes place
between the candidate and the committee.
.

6 Pragmatics: definition and scope /ll///l/l/!I/I//I//I////I/////I////I//I/

The social context is broader, including such q_uestions as the institutional

context in which the interaction is occurring (a school, a doctor's office, a
courtroom, etc.). These wider contexts entail certain roles and statuses (known
as footings 6), which for instance determine who leads a conversation, who asks
the q_uestions, etc. Another aspect of social context concerns how participants in
an interaction conceive their own identities in terms of seniority, gender, sexual
orientation, ethnicity, and so on: clearly, any one person's identity is a complex
matrix, different aspects of which will be important in different situations. Yet
another aspect of the social context affects such fundamental choices as that
between spoken and written communication (as well as intermediate forms like
chatting, 'whatsapping', etc.) or the possible use of more than one language
(code-switching) and/or two varieties of the same language (diglossia). It is
immediately obvious how social pressures impact upon communicative choices
but also how changes in society are reflected in those choices: it nowadays
happens that bosses fire employees with a mere text message, something
unthinkable even ten years ago. Seen from this social perspective, context is
something negotiable, arising from the interaction of human beings who are
aware of their multiple roles in society but also are continually mc1:neuvering
towards the achievement of their personal goals.

The cultural context reflects how the variables of the social context are
interpreted from the perspective of a particular culture. A culture may alfgn with
a specific language community, as when people talk of 'English-speaking culture'
or with a group of such communities, as in 'Latin culture'; alternatively, different
cultures can be recognized within one and the same group speaking (more or
less) the same language, for example distinguishing Brazilian and European
Portuguese cultures. Cultural context can influence such fundamental aspects of
language as the understanding of time (for example as linear or not) or of space
(for instance as independent of the human body or not). At a more micro level, it
can determine what associations people have with words: in many cultures, the
word normal is evaluated negatively and is associated with notions like
'ordinary', 'run-of-the-mill', 'unexceptional'; on the contrary, in Dutch culture,
normaal tends to be evaluated positively in the sense of 'agreeable', 'not
irritating'. Cultural context is so important for analyzing language use that
researchers studying undocumented languages have to immerse themselves in
the speakers' culture before they can grasp how those languages are used to
communicate. Cultural context is also key to understanding how a community
uses language in serious vein for government and religious rituals and in lighter
vein for humor and mockery. Cultures are very far from being static or monolithic:
just consider how British culture has changed and internationalized under the

6 E.g. Goffman (1981).

///////l!/I//I///II///I/J/I////II/I////I Pragmatics: definition and scope 7

influence of immigration, EU membership and globalization. The dynamism of

modern cultures means that the impact of cultural context on language use is
sometimes hard to trace; but no one doubts its powerful influence.
The cognitive context refers to the mental processes that allow language
users to perform within their own social and cultural contexts. This context
contains a set of beliefs and assumptions, some of which have the status of facts
for particular speakers. These can be expressed as propositions. Within any given
culture, there will be atheists, agnostics and believers (in very different
percentages, of course, depending on the culture under analysis). For the atheists,
the proposition 'God exists' has the status 'untrue'; for the agnostics, it has the
status 'may or may not be true'; for the believers, it has the status 'true' -in other
words, it is a fact. The particular setting of their cognitive context will
subseq,uently affect how they form and interpret utterances about faith, religion
and theology. One belief that is shared by everyone from the age of three
upwards, with the partial exception of sufferers from autism and related
conditions, is that part of the cognitive eq_uipment of other people ( and of oneself)
is intentionality: the desire to inform others, coupled with the desire to have this
desire recognized as such by those others.
To the four types of context proposed by Fetzer (2004) we wish to add a fifth
that has had increasing attention in recent years, the emotional-attitudinal
context. Long before we can communicate with language, we are very effective
at communicating our emotions (as any childcarer will confirm) as well as our
attitudes of acceptance and refusal. The impact of emotion on our communication
continues through childhood and adult life, providing the engine for our thinking,
inspiring our cultures, motivating our sociality and impacting our language in
countless ways (Foolen 2012). Everything we do is dependent upon our emotional
state and our attitudes. Imagine you have a week to write an essay on a poem.
The way you go about this linguistic task is ultimately determined by your
feelings: How do you feel about the message of the poem? How do you feel about
having to write about it? How do you feel about the teacher who gave you the
assignment? It is common experience that all these feelings and many more have
a decisive impact on how your mind sets to work. Even if your essay is expressed
in the 'cool' prose of scholarly discourse, there is a profounder context for the
entire experience that is deeply rooted in the processes of your body. This
emotional-attitudinal context also affects how readers and hearers react: voters
may be bombarded with written manifestoes and political broadcasts but
ultimately their decision at the ballot box will be determined by their feelings.
Although the details of the relation between language and emotion are still not
well understood, the increasing interest among researchers in this elusive but
fascinating area justifies our recognizing emotional-attitudinal context as a
crucial type of context, which affects and is affected by all the other types.
8 Pragmatics : definition and scope l!lll//l!/ll////l/l///l//l!//l/l!//I////

In effect, our memories for the linguistic context, our sensitivity to the social
context and our exposure to the cultural context are all mediated by the thinking
brain as well as by our emotions and attitudes, but at the same time it can also be
said that the language, the society and the culture we are immersed in also shape
our cognitive system and our emotions in some way or another. Figure 1 shows
our attempt to represent this phenomenon.

Figure I: Types of con+exf

.:Emotional/ . ·
- attitudinal
. :_ eontext

Cognitive
I \ Cultural'. -
cont~xt . context

\ I
· Social
,:_.inguistic . ',
. · context context

Hence, it is apparent that utterances cannot be treated in isolation, but as a

function of their possible contexts. We have conseq,uently tried to explain here
that pragmatic studies include phenomena which reflect the interaction between
elements of the purely linguistic context and elements of the more general
cognitive, attitudinal and socio-cultural context (e.g. perception, personal and/or
cultural beliefs, inference, etc.). And in relation to this, it should be pointed out
that these phenomena cannot be properly investigated without drawing on the
findings of other disciplines, such as cognitive science or sociology, which makes
pragmatic studies interdisciplinary. It is clear, thus, that an approach to
pragmatics that takes into consideration both the 'purely linguistic' and the non-
linguistic factors will be far more enriching and revealing for the analyst than an
approach that considers only the 'purely linguistic' ones.
Our main intention in this section has been to explain that the concept of
context is a multifarious and dynamic one, which changes continually and always
influences the linguistic product of a discourse interaction. Conseq,uently, it can
contain various kinds of interrelated information that is crucial for the
characterization of a wide range of pragmatic phenomena such as deixis,
reference or implicature (to name a few), all of which will be presented and
analyzed in this book in due course.
//l////l/////l////////l/////!I///////I// Pragmatics : definition and scope 9

1.3. THE UNDERDETERMINACY OF LANGUAGE

AND ITS RELATION TO CONTEXT
If we analyze the utterances we hear or read every day, we shall see that there
could always be some way in which they are unclear. This is what in pragmatics
we call the 'underdeterminacy' of language, which refers to the fact that any
given utterance could have more than one possible meaning, depending on the
conditions surrounding the speech event. The cognitive work we carry out as
users of a given language - based on all the contextual information we have - will
determine which of all the possible meanings is the one intended by the speaker
in each particular situation. Thus, it can be said that most (if not all) linguistic
expressions are underdetermined in some way or another and that their
interactive meanings can only be assessed and worked out within a given
context.
One of the authors of this book always remembers how much at a loss she
felt the first time she had to pay for her groceries at an American supermarket,
when the cashier asked the underdetermined q,uestion Paper or plastic? The
exchange went as transcribed in (4): ·

(4) Cashier: Good morning. Have you found everything you needed Ma'am?
Laura: Yes, thank you. Everything's fine.
Cashier: Great. Paper or plastic?
Laura: Erm .... Excuse me? (Thinking: what in the world could she mean???)
Cashier: P-a-p-e-r o-r p-1-a-s-t-i-c?
Laura: Sorry, but ... Do you mean "Am I going to pay with bills or credit
card??"
Cashier: No, ma'am, I mean, "do you want paper or plastic bags for your
groceries?"

The q,uestion put by the cashier (Paper or plastic?) is a prototypical case of

an underdetermined utterance which is normally understood by everyone in a
certain context or culture and therefore does not generally need to be explained.
In this particular case, since Laura was visiting the USA for the first time in her
life, she did not have the cultural background to automatically know that the
cashier was referring to the type of bag she wanted to have her groceries put
- o and therefore it was necessary for her to ask for the explicature7 that would
clarify the intended meaning of the cashier's q,uestion. Without it, Laura was
chinking of a more far-fetched explanation, for as can be seen in (4), she had

We shall deal with the concept of explicature in Chapter 5.

10 Pragmatics: definition and scope /////////////////////////// /////// II

speculated about paper and plastic being metaphors for cash and credit
respectively, considering that bills are made of paper and credit cards of p
(but of course, she was wrong!).
In this particular case the meaning encoded in the linguistic expres: ·
used underdetermines the proposition expressed (i.e. what is said). Some autncxs
( e.g. Carston 2002) argue that this is the only type of underdeterminacy
should be covered in pragmatics. However, other authors (e.g. Grundy 20
Huang 2016) include other types within its scope: they also include instanc ·
which a) the linguistic meaning underdetermines what is meant, orb) what ·
said underdetermines what is meant. An example of a) would be the sign "
Psychiatrist" (Grundy 2008: 9) on the door of an office in a clinic, where the m
retrievable8 and appropriate interpretation would be that it refers to a psychia ·
who treats children and not to a psychiatrist who happens to be a child, but if
only considered the possibilities given by the logic and grammar of the Eng ·
language, both meanings are eq,ually possible. In the following examples (5b and
6) we shall deal with cases of b) above, i.e. examples in which what is said
underdetermines what is meant. And here is where we could include instances o
irony, understatement and the like. Consider, then, (5a) and (5b):

(5a) Welcome to fabulous Las Vegas ( the famous sign at the entrance of the city
of Las Vegas, Nevada, USA)

i 'f
I

(5b) A: I didn't know there was so much crime and vice in this city.
B: Welcome to fabulous Las Vegas.

8 To use a term from Relevance Theory (See Chapter 5).

//ll/ll//////l/!/ll/lll/llll!ll/!//I!!// Pragmatics : definition and scope II

As the reader will surely have realized by now, the meaning of (5a) is rather
different from that found in (5b ), even if the linguistic expression used is exactly
the same one. And the reason why they are interpreted differently is precisely
becaus~ the context of appearance of the expression is not the same: In (5a) we
will normally interpret it as a friendly welcome message to all people visiting
Las Vegas, whereas the message in (5b) has little to do with the expressive
illocutionary act of welcoming. In fact, it is a warning message, which intends to
make the listener aware of the fact that the city of Las Vegas can be a dangerous
place, and therefore not so 'fabulous'. Conseq,uently, if the expression Welcome to
fa bulous Las Vegas were devoid of context, its meaning would be q,uite
underdetermined and we would not be able to define the specific function fulfilled
by it in each case.
Another interesting example, in which it is important to take the variable of
time as part of the context, is found in the following exchange of What's App
messages between one of the authors and her younger son:
(6) (Julian is helping out at the Tech Department of his school during the
summer holidays)
Laura: Julian, will you come for lunch today? i"l:08
Julian: I'll let you know soon. 11:09
Laura: What does "soon" meanfor you? 13:53
Julian: Sorry! I won't be long. I'll be back later. 13:54
It is obvious here that Laura is not really interested in knowing what the
word soon means for Julian, but that she is being reproachful towards her son
because almost three hours have elapsed since the time he said he would "let her
know soon" whether he was coming for lunch or not. Thus what she says
nnderdetermines what she means, a fact that can also be clearly observed in
Julian's reaction to his mother's (rhetorical) q,uestion, which turns out to be an
apology instead of an explanation of what the word "soon" means for him.
The function and the meaning of the q,uestion What does "soon" mean for
you? in (6) are different from, for instance, its function and meaning in (7), simply
because the context is different and, among other things, the variable of time is
not affecting the q,uestion in the same way as in (6):
(A is going on an errand)
..,,,
A: I'll be back soon
B: What does "soon" mean for you?
A: Ermmm ... .In an hour or so.

This is a normal conversation in which each turn occurs immediately after

the previous one, and where B only wants to have a more precise idea of the time
12 Pra9mafics: definition and scope /////////////// // / / / //////I/// I/////////

at which B will be back, no reproach being intended. Thus, contrary to Julian's

response in (6), the response given by A in (7) is not an apology but a simple act
of providing B with the piece of information previously req,uested.
As the reader may have noticed, underdeterminacy is related to ambiguity.
Both are properties of everyday language that are important and dealt with in
pragmatics. Huang (2016: 206), for instance, provides interesting examples of
lexical and syntactic ambiguity which also present underdeterminacy if taken
out of their contexts, such as (8) and (9):
(8) John is looking for his glasses.
(9) They are cooking apples.
(8) is an example of an ambiguous, underdetermined utterance in that - if
the context is not provided - there are at least two possible interpretations:
a) John is looking for his spectacles, orb) John is looking for his drinking vessels.
And the utterance in (9) could be the answer to at least two different q,uestions:
a) What are they doing in the kitchen? Or, b) What kind of apples are those?
Possessive constructions in English constitute another case of context-
dependent, underdetermined and ambiguous expressions. For instance, if someone
is speaking of Lachlan's book, there could be at least four immediate possible
interpretations: a) the book that Lachlan wrote or is writing, b) the book that he
possesses, c) the book that he is now reading, or d) the book that he is holding in
his hand. Only if we have access to the context(s) in which the expression is used
can we choose one of these four (or any possible other) alternatives.
For all the reasons explained above, it is understood that the feature of
language called underdeterminacy is of key importance within pragmatic studies.
Other properties of everyday lan9'uage related to underdeterminacy (and
ambiguity), such as relevance, appropriateness, non-literal/indirect meanings, etc.
are also part of the scope of pragmatics. In the next sections ( and chapters) we
shall try to delimit its scope by looking into these, as well as other central topics.

1.4. THE SCOPE OF PRAGMATICS: MAIN SCHOOLS

AND CENTRAL TOPICS
As we pointed out at the beginning of this chapter, pragmatics is a complex and
dynamic sub-discipline of linguistics whose boundaries are not easy to define.
One of the reasons why it is not easy to draw clear lines regarding the scope of
pragmatics is because, as we also pointed out above, pragmatic studies step out
of the frame of 'pure linguistics', and thus also draw on the findings of other
/////I//I///II///II////////////I//II//// Pragmatics: definition and scope 13

disciplines such as psychology, sociology or neurology. And it is precisely due to

this widening of scope that, within pragmatic studies, we may find different
schools of thought, the two principal ones being a) the Anglo-American School,
and b) the European Continental School.

The Anglo-American School holds what has been called the component view
of pragmatics, because within this conception, pragmatics is considered to be
one more of the core components of a linguistic theory, along with phonetics,
phonology, morphology, syntax and semantics. This tradition is represented by
authors such as Levinson (1983), Leech (1983), Yule (1996) or Cummings (2005).
As Huang (2016: 208) remarks, "the component view of pragmatics is to some
extent a reflection of the modular conception of the human mind", and thus
within this framework of thought, pragmatics may be defined as "the systematic
dy of meaning by virtue of, or dependent on, the use of language. The central
opics of inq,uiry include implicature, presupposition, speech acts, deixis and
reference" (2016: 208). Indeed, the topics alluded to in Huang's definition are
central and unavoidable topics for any researcher in the field, but a great number
of scholars believe that this constitutes too narrow a view of pragmatics.

The European Continental conception of linguistics is embodied in the work

of linguists such as Jef Verschueren, who defines pragmatics as "the cognitive,
social, and cultural science of language and communication" (2009: 1), and who
conseq,uently advocates a broader view, presenting a functional perspective on
levels of linguistic description and beyond. As Bublitz & Norrick (2011: 4) put
· within this broader point of view, pragmatics is seen as "the scientific study of
aspects of linguistic behavior" and as an area of research that is "fundamentally
concerned with communicative action in any kind of context". Conseq,uently,
ccording to the Continental view, pragmatics is not confined within the
undaries of just one linguistic component, but is 'omnipresent' in all
mmp onents and every aspect of linguistic behavior. This broader view has been
re inclined than the component view to consider pragmatic phenomena as
~:::!'a1m·...,tic data that necessarily have to be studied in relation to other domains,
· · g rise to new sub-areas of research such as computational, historical,
ropological or experimental pragmatics.
But the above dichotomy of views does not in fact present the real current
·;:~n.rre: As Huang (2016) points out, there has been some convergence between
Anglo-American and the Continental traditions, for there has been important
rk done on topics such as implicature or presupposition from a Continental
lftol:li1""'"""'ective, and at the same time the Anglo-American School has delved into
er research on the relation of pragmatics to areas such as cognitive
~JCflolinguistics, computational linguistics or neuro-linguistics, just to name a
. . As we see it, this convergence was a necessary and fruitful one, for it unites
14 Pragmatics: definition and scope ////////////////////////////////////////

the efforts, knowledge and strengths of both schools, which obviously

complement each other. Other authors, such as Ariel (2010) or Mey (2008 [1993]),
also adhere to the reconciliation of the component and the perspective views.
Mey makes the following concluding remark, which is very much in-sync with the
approach taken in this book:
Thus, we could have a pragmatic component, understood as the set of whatever
pragmatic functions can be assigned to language, along with a pragmatic
perspective, i.e. , the way these functions operate. We could either ask how users
'mean what they say', that is, how they communicate, using language, or how
they 'say what they mean', employing the linguistic devices at their disposal to
express themselves. (Mey 2008 [1993): 9)

1.4.1. MICRO AND MACROPRAGMATICS

The scope, interests and methodology of the Anglo-American School could be
identified with what has been called micropragmatics and those of the
Continental European School with so-called macropragmatics. So, when doing
micropragmatics we would be concerned with specific theor~tical discussions
about topics such as implicature, reference or deix is , while when doing
macropragmatics the emphasis would be placed more on the perspective taken
than on any theoretical point in particular, which would simply entail taking a
functional perspective (in a very broad sense) on the study of language in all its
aspects.
But, again, this dichotomization of the field of pragmatics presents its
problems and raises some important q,uestions, such as:
a) Can a researcher do micropragmatics and relate her findings on the topics of
implicature, presupposition, etc. to other non-linguistic social, cultural or
clinicalphenomena?,or
b) If macropragmatics is about 'the study of language in all aspects', how does it
differ from discourse analysis, and why would it be necessary in the first
place?
To the first q,uestion we would feel inclined to respond that the answer is
'yes', and that we in fact should try to find the relationship between these
theoretical topic~ and their contexts of whatever type, as we illustrated in 1.2. So,
again, the boundaries between one view and the other remain blurry and
undefined, and ·it would be limiting to stick to either one of them in a
fundamentalist way. As for q,uestion b ) , it is very likely that, if we define
pragmatics as a discipline that 'studies language in all its aspects', we shall
immediately start to wonder if there is any use in distinguishing it from discourse
analysis or even linguistics (if taken in a broader sense). Thus, while a
lllll/llll//lll/!ll//!l!/I//II/I//I//II/ Pra9mafics : definifion and scope 15

convergence of both approaches seems desirable, the all-embracing character of

macropragmatics seems too broad to make pragmatics a necessary sub-field of
linguistics, distinct from any other.
Along this line of thought, then, we feel more prone to think that the findings
and objectives of both micro and macropragmatics are of great use for linguists
and discourse analysts, but that, whereas pragmatics is an indispensable source
for the analysis of discourse (Alba-Juez 2009: 46), it is, however, not the same
thing. Pragmatics provides discourse analysts with important tools and basic
concepts without which their work would be impossible to carry out, but
discourse analysis is a broader, more empirically-oriented discipline that is not
particularly interested in theorizing about, for instance, inferences or deixis
(pragmatics already does that!). Among other things, discourse analysis includes
many schools and traditions ( e.g. critical discourse analysis, ethnography of
communication, etc.) which in our view do not form part of pragmatics, although
they do draw on its findings.

Based on the above discussion, in this book we shall not cover all approaches
to the study of language or discourse or, for obvious reasons, 'all aspects of
language', and shall thus concentrate more on the topics that have traditionally
been studied within micropragmatics, without disregarding the findings of
macropragmatics. We shall mainly explore, therefore, topics such as politeness,
speech acts, implicature or reference, placing our emphasis on the fact that they
always occur in cognitive, social, emotional or cultural contexts (Chapters 1 to
5), and at the same time also devote some space and time to the discussion of
ome topics from the areas of, for instance, historical or computational
pragmatics, which could be placed more within the field of what so far has been
called macropragmatics (mainly in Chapter 6).

1. 5. PRAGMATICS AND COGNITION

_.\s we mentioned in 1.2 above, our ability to function in social and cultural
contexts is dependent on our mental capacities. In the approach to the mind
favored by generative linguistics, q,uestions of communicative context were ruled
out as belonging to performance rather than competence. Scholars interested in
the cognitive context of communication therefore turned to other traditions,
notably functionalism and in particular pragmatics. As a result, these traditions
began to thrive, spawning such new fields as Cognitive Linguistics and Discourse
_.\nalysis in which the relation of utterances to their social, cultural and cognitive
settings is of central importance.
16 Pragmatics : definition and scope l/ll//ll//ll/l!/ll/!/ll/l!ll/ll!///!I/I/

An important contributor to the study of the cognitive context of

communication is Bara (2011). He defines Cognitive Pragmatics as the study
of the mental states of people who are engaged in communication. Some of these
mental states are shared, such as those that concern the nature of the ongoing
communication: a teacher and her student, for example, will have a shared
representation of the class activities that are taking place. Other mental states
are not shared, of course: the teacher may well be assessing the student during
the interaction without communicating that assessment; and the student may
be secretly daydreaming about his favorite football team.
When the brain is damaged, as in traumatic brain injury, the patient's
linguistic abilities may be affected (luckily, this does not happen very often since
only about 2% of cases suffer persistent language disorders, Granacher 2008:
275). The disorders can, as one might expect, also affect patients' communicative
abilities. McDonald & Pearce (1996) tested brain-damaged patients for their
ability to recognize sarcasm, as in their example (10):

(10) Mark: What a great football game.

Wayne: Sorry I made you come.

Here Wayne's reaction shows that he has understood (probably on the basis
of the evidence of his own eyes, too) that Mark was being sarcastic. The
researchers found that a proportion of their subjects were unable to detect the
sarcasm, whereas normal subjects generally could. The brain-damaged subjects
typically could process the literal meaning but failed to infer what Mark really
intended to communicate.
Whereas generative linguistics holds that our language faculty is
autonomous, i.e. distinct from our other cognitive abilities, cognitive pragmatics
emphasizes how language is inseparably interwoven with the totality of our mind
and body. This position has been characterized by Lakoff (1987) as
experientialism, a philosophical stance that sees reason as not being objective
but rather as arising from our genetic and environmental involvedness with the
world around us. For some scholars in pragmatics, this entails (cf. Cuenca 2003:
2) conceiving "the system ... in a dynamic way , interrelating linguistic structure
(syntax), meaning (semantics), language use (pragmatics) and conceptual
structure (cognition)". As a conseq,uence of this view, the advocates of this
approach tend not to believe in clear-cut dichotomies (e.g. competence/
performance, syntax/semantics, semantics/pragmatics) and rather call for an
integration of all linguistic levels along a dynamic continuum.
One of the trickiest issues in the study of language and cognition concerns
the extent to which our conceptualization, i.e. the way we categorize our
experience of the world, is determined by the language that we happen to speak.
//////////////////III//// / .r, Pragmatics: definition and scope 17
* 8 t8LIO~~o't
Let us take as an example one of the objective properties of a human being: his/
her age. For every person I know, including myself, I have in my 'mental database'
a value for the variable age: "27", "in her sixties", "q,uite a bit younger than me",
etc. When it comes to communicating my own age, an extract from my database,
I express it in English by means of the construction I am n years old and in
Spanish using the construction Yo tengo n anos (lit. 'I have n years'); other
languages have yet other constructions. The q,uestion that arises is whether I
conceptualize my age differently according as I communicate in different
languages: Does the Spanish-speaker regard herself as 'possessing' years ( as she
might possess consumer goods)? Does the English-speaker see himself as to a
certain extent old? In any case, it is clear that the two ways of putting it are q,uite
conventionalized - in English, after all, we could say "I have n years", but we
happen not to.
Where our concepts are more abstract, things get more difficult. In the area
of emotions, for example, we talk about our feelings differently according as we
are speaking one language or another. Bosq,ue (2010) discusses how the Spanish
lexicon does not make the distinction between shame and embarrassment, using
only the word vergiienza for both, and how (in our translation) "the best
dictionaries tell us that Spanish carino is love, affection, fondness and tenderness,
eq,uivalents that leave any native speaker of Spanish unsatisfied." Indeed, the
English word love, both as a noun and a verb, aligns with both the nouns amor
and carifio and the verbs amar and querer in Spanish; yet no Spanish-speaker
would fail to make a sharp distinction between two emotions here. Bosq,ue goes
on to point out that conceptual distinctions in such areas are so language-specific
that no theory of emotions can be formulated in an individual natural language.
He praises Wierzbicka's (1999) attempt to analyze them in terms of a neutral
metalanguage, and emphasizes how each culture's emotional life has been shaped
by the social history of the speech community.
Yet we must not be trapped into thinking that our language's classification of
our emotions determines how we feel as individuals. A bilingual speaker
of English and Spanish will have the same feelings towards her lover, her children
and her dog, no matter whether she is speaking one language or the other; and
he will feel the same about her own bad behavior, whether she classifies her
feelings as shame or embarrassment when she is speaking English. That said, we
cannot exclude the likelihood of some feedback from linguistic distinctions to
our emotional life. In a culture in which open discussion of emotions is favored,
the individual will be exposed to more instances of refinements in vocabulary
and thereby stimulated to reflect on the nature of her feelings.

The relationship between language use and cognition is far from having
been settled and continues to be actively debated. We have seen that distinctions
18 Pragmatics: definition and scope ////////////////////////////////////////

in vocabulary can also affect our cognition and there is little doubt that patterns
of interactive behavior also come to be entrenched in our cognition as schemas
and scripts. If, as is claimed by the cognitive linguist Langacker (1987: 57), a
grammar is "a set of cognitive routines, which are constituted, maintained and
modified by language use", a view strengthened by the 'usage-based' approach of
Bybee (2010), then the most deep-seated and subconscious component of our
linguistic abilities, the g~ammar of our language, is itself the result of language
in action.

1.6. INDIRECT MEANINGS: WHY THEY ARE ALWAYS

PRAGMATIC IN NATURE
One of the most fascinating features of human language is the fact that a single
utterance may have many different meanings in different contexts. As Grundy
(2008: 178) puts it, "When you think about it, it's really odd that what we mean by
our words is more important than what our words mean". And in order to
understand what speakers mean by their words, we usually have to put into
motion a series of more or less complex inferential processes. This occurs in all
natural languages, and it is the one feature that makes language evolution
possible: if all meanings were fixed and direct, and there were no possibility of
inferring indirect meanings, communication as we know it today would be
practically impossible.
When dealing with indirectness in pragmatics, we normally refer to the kind
of indirectness that is consciously chosen by the speaker, assuming that she is
behaving in a rational manner9 and wants to cause or obtain some special effect
by doing so. Example ( 11) is an instance of one of the ways in which Laura uses
indirectness in some conversational exchanges with her husband, because she
does not want to impose on him but at the same time knows that he is always
willing to help. As the reader will surely notice, the utterance We're running out
of peanuts and I have to make peanut butter constitutes an indirect way of
asking her husband to go and buy peanuts at the usualfrutos secos ('dried fruit
and nuts') shop in Majadahonda (a town close to their house where they always
buy nuts). This is a clear example of an indirect speech act 10 (in this case, a
statement that is in fact functioning as a req,uest):

9 As Thomas (1995: 121) points out, the reader may think that assuming that speakers
behave rationally is a pretty big assumption to make, but it is true that, in spite of the
fact that people freq,uently behave irrationally, we tend to assume that people are
rational until we have evidence to the contrary.
10 We shall examine the topic of speech acts in detail in Chapter 2.
 ////I///////I////I//I/I//I//II/I//I//I/I Pragmatics : defin ition and scope 19

(11) Laura: We're running out of peanuts and I have to make peanut butter.
Gus: Don't worry I'll stop by at Majadahonda to buy some this evening. How
much do you want me to get?

Indirectness can also be unintentional, however, as for instance when a

speaker forgets a word and has to give an indirect explanation or definition of it
in order to be understood, but this is not the kind of indirectness we are most
concerned with in pragmatics. The type of indirect meanings that pragmaticians
look into is the intentional one, which is normally produced at the speaker's 'own
risk and expense', considering that the indirect utterances produced are 'costly'
- in the sense that it takes longer for them to be produced and processed - and
risky', because the hearer may not understand the meaning(s) the speaker is
trying to convey (Dascal 1983).
Thus a great number of the meanings we want to communicate are not "said"
lit.erally , and therefore have to be reached by the hearer(s) through some kind of
cognitive process, namely an inference of some sort. What makes it possible for
the hearer to make these inferences - apart from his cognitive capacity to process
information - is precisely his pragmatic knowledge of the whole discursive
ituation, i.e., his interlocutor's (complex) identity, the place and time of
occurrence of the utterance, their shared background knowledge of the possible
environments in which a given utterance may occur, etc.
Since all these inferential processes are so relevant from the pragmatic point
of view, we shall explore their occurrence and nature at different points along
d:ris book. Thus we shall be looking into such phenomena as conventional and
conversational implicatu re, presupposition, entailment, etc. These inferential
meanings may sometimes become conventionalized or even grammaticalized,
which constitutes a very relevant aspect of language, to which we now turn .

.6.1. CONVENTIONALIZED AND GRAMMATICALIZED

INDIRECT MEANINGS
. .I. I Conventionalization

me constructions which were originally indirect - and the refore in the past
:req_uired some processing cognitive effort on the part of the hearer to be
derstood - do not req,uire such effort any longer because the indirect meaning
· ended has been conventionalized, i.e. , it has become an accepted and
con entional way of expressing the indirect meaning in q,uestion. For instance,
ery competent speaker of English knows that if someone asks the q,uestion
20 Pragmatics: definition and scope //////// ////////////////////////////////

Could you open the door, please?, the speaker is not asking about the ability of
the hearer to open the door, but just req,uesting that he open the door as a favor
to the speaker, be it because she is carrying a heavy load or because of any other
reason that makes her unable ( or unwilling) to open it h~rself. Other instances of
this phenomenon can be found in some generally rude ironic utterances of the
type shown in italics in examples (12), (13) and (14), all of which have become
conventionalized ways 11 of either telling an interlocutor that his q,uestion or
remark was a stupid/unnecessary one (example 12) or of expressing disbelief
or disagreement (examples 13 and 14, respectively). The fact that the expressions
and/or strategies used to convey these indirect meanings are now conventionalized
in the English language entails a simplification of the inferential process, in the
sense that the hearer no longer has to work out the conversational implicature
that leads to the indirect interpretation, because it has been 'short-circuited'
(Morgan 1978) due to its freq,uent use among English speakers.

(12) A: Do you think our government is corrupt?

B: Is the Pope Catholic? ( = The answer to your q,uestion is obvious: YES)
(13) A: I have won the "Mr. Universe" contest.
B: Yes, and I'm Marie the Queen of Romania ( = I don't believe a word of
what you say)
(14) A: Peter is a very good guy, don't you think?
B: If Peter is a good guy, I'm the king of France ( = I don't agree with you or
with the proposition expressed by your utterance= NO)

In all three examples we are confronted not only with conventionalized

expressions, but also with conventionalized pragmatic strategies which can be
described as follows:
a) Answer your interlocutor's ridiculous q,uestion with another, even more
ridiculous and superfluous q,uestion, in order to show her that the answer to
her q,uestion was self-evident (example 12). Other possible replies could have
been Does a bear shit in the woods? (another, even ruder conventionalized
expression), Am I Marilyn Monroe? (not conventionalized as an expression,
but ridiculous enough to fulfill the req,uirements of the strategy), or any other
q,uestion whose answer is obvious.
b) Reply to your interlocutor's lie (or what you believe to be a lie) with an even
more exaggerated and ridiculous lie in order to show that you have not been
cheated by her (example 13). Another possible reply could have been: Yes, and

11 For a detailed account of this phenomenon, see Alba-Juez (1998).

 I //////I////I/I//////////II///I/I/I//I Pragmatics: definition and scope 21

I am Pope Francis, or any other statement flouting Grice's Quality Maxim 12

in a similar manner.
c) Reply to your interlocutor's statement with the formula "If p, then q, = not p"
to show your disagreement with her statement (example 14). The condition
for this strategy to cause its desired effect is that the main clause (p) has to
carry an absurd proposition, and the refore there are other possible alternative
expressions, different from/'m the King of France (such as I'm a Martian,
or I'm Elvis Presley, among a million others), which could be used in the
formula.

In all these examples we are confronted with instances of utterances which

originally triggered conversational implicatures but no longer do so in present-
day English, because the expression or the strategy ( or both) have been
conventionalized, and therefore the implicature now worked out by the listener
- of the conventional type. 13 But this process can even go further, to the point
chat the original indirect meaning becomes deeply ingrained in the grammar of
the language, as we shall see in the next section.

1.6.1.2 Grammaticalization

It sometimes happens that the indirect or inferred meaning of an item comes to

co-exist with or even replace the direct meaning. As we mentioned in 1.6.1.1,
Could you open the door? has become the accepted and conventional way of
making the relevant req,uest. Many pragmatic markers, too, have arisen in this
way. A pragmatic marker is a fixed expression used by a speaker to indicate how
the propositional content of what she is saying links to the context. One good
example is the pragmatic marker 14 well, which is often used before the answer to
a q,uestion in order to indicate that the answer is not a direct response:
(15) A: What's your excuse for coming late?
B: Well, I'm not sure what to say.

B does not answer A's q,uestion and signals this by prefacing what he has to say
with well. In this usage, this is clearly felt to be the direct meaning (or function)
of well and not a contextual understanding of the manner adverb well 'in a good
way', as found in/ meant that well. Nevertheless, in historical times, well must
have been used to indicate that although the answer does not address the

12 See Chapter 3 for Grice's Cooperative Principle.

13 We shall return to the topic of implicatures in Chapter 3.
14 We are here following Fraser's (2006b) distinction between pragmatic and discourse
markers.
22 Pragmatics: definition and scope / / // / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

q,uestion, it is still meant well. Other pragmatic markers take the form of mini-
clauses, like you know or I mean, as found in (16):
(16) a. You know, I think we should stop right here.
b. I mean, there's no point in carrying on.

These expressions are q,uite fixed (You mean, I know and You knew, for
example, cannot function as pragmatic markers) and they may fulfill different
discourse functions depending on the context of use. In (16a) the speaker is using
the discourse marker to claim control of the ongoing situation and in (16b) she is
using I mean to introduce a justification for her previous statement. These are
nowadays strongly felt to be the direct meanings of those units. Not only are the
two units fixed in form, but' they are also incomplete: the verbs know and mean
usually req,uire an object (as in You know the answer and I mean the other shoe).
Invariability and incompleteness ( or 'reduction') are typical signs of what is
called grammaticalizatio n, the historical process in which freely associating
linguistic forms come to be not just conventionalized but actually incorporated
into the grammatical resources of the language.
Most pragmatic and discourse markers are usually fairly separate from the
units that they relate to: well almost always precedes, and you know and I mean
typically precede or follow, and they each have their own intonation unit. But
there are some pragmatic markers which are different in being integrated into
the clause in which they function, but they too have usually arisen from
conventionalization. In Dutch, for example, the pragmatic marker wel (which has
the same historical origin as English well) is found in the middle of a clause,
stressed, to indicate that the ongoing proposition is a denial of something
negative that someone else ( often the conversation partner) believes:
(17) A: Je hebt die e-mail nooit gestuurd. (You have that e-mail never sent) 'You
never sent that e-mail.'
B: Ik heb 'm wel gestuurd. (I have it DENIAL sent) 'I DID send it.'

In English, the grammatical techniq,ue of introducing the dummy operator

do is used to capture the stressed denial; the fact that Dutch uses a stressed
particle rather than an empty verb shows that wel in this usage has become part
of Dutch grammar. So this is another case of grammaticalization.
In English, please is used as a basic pragmatic marker to preface or follow a
polite req,uest (18a & b ), but can also occur inside a clause, even a subordinate
clause (18c & d):
(18) a. Please tidy up this mess.
b. Tidy up this mess, please.
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I/ I I I I I I Pragmatics: definition and scope 23

c. Will you please tidy up this mess?

d. He asked me to please tidy up the mess.
What is clear is that please is the result of grammaticalization. An entire
adverbial clause such as If it please you (with please in the present subjunctive
form; we find it in this form in Shakespeare's Two Gentlemen of Verona, for
example) has been worn down to the single word please. It is instructive to
compare eq,uivalent expressions from three other languages (19):
(19) a. French s'il te plaft (if it you pleases)
b. Catalan sisplau (from si us plau = if you it-pleases)
c. Dutch alsjeblieft (from als hetje belieft = if it you pleases)

In French, the expression is only used as a pragmatic marker and has not
been reduced in form. In Catalan, univerbation (the creation of a single word) has
taken place and one phoneme has been lost. In Dutch, we again see univerbation
involving loss of one original word and one syllable from belieft); alsjeblieft
occurs both as a separate and an integrated pragmatic marker, in the eq,uivalent
o (18c) but not (18d). As we see, grammaticalization te~ds to run along pathways
across languages, but the degree to which a form is grammaticalized differs from
one to the other.
Grammaticalization affects many different areas of the language, not just
pragmatic markers. Some examples of grammaticalized units that are not
markers are How about .. . ? for suggestions, Let's ... for proposals or You('d)
better for recommendations. In all cases, grammaticalization involves going
beyond conventionalization in the sense that inference is no longer involved:
lrn.at was historically an indirect meaning has become the new direct meaning.

1.7. PRAGMATICS AND CULTURE

the reader may have concluded by now, pragmatic studies cannot be
· dependent from cultural studies, given the fact that pragmatic meanings are
ays sensitive to the context, and therefore culturally situated. Human beings
can interact in different ways, and most of these modes of interaction are culture-
.specific. When people interact with other members of their own culture, there
ds to be less ground for misunderstanding or pragmatic misfires, considering
t all participants generally share the same language and the same modes of
· eraction. This is what has been called intracultural communication.
But communication does not always occur among members of the same
culture, which makes it inevitable for the researcher to pose q,uestions about how
24 Pragmatics: definition and scope I I/ I/ I I I I I// I I I I I I I//// I I I/// I/// I// I I I I

communication works when individuals pertaining to different cultures have to

interact. Grundy (2008: 233) invented the term trans-cultural to name any kind of
communication that is not intracultural, and he identifies two kinds of trans-
cultural communication: a) cross-cultural, which refers to the "communication
where a non-native member interacts with a native member of a particular
culture", and b) intercultural, which refers to the communication occurring
"when the interactants do not share a common culture, and therefore have to use
a lingua franca". The distinction between cross-cultural and intercultural
communication is not always easy to draw, but a prototypical case of the former
occurs when, for instance, a Spanish businessperson travels to France and has to
interact with his/her French colleagues using French as the language of
communication. And a prototypical example of intercultural communication can
be found in the context of international conferences, where the participants come
from different countries and cultures, and therefore use a linguafranca (English
in most cases) to communicate with one another. It is interesting to note here
that, even if some of the participants have English as their native language, the
rules of interaction are not going to be the same for them as in an intracultural
communication situation.

In both intracultural and trans-cultural communication it is crucial for the

speakers to have the necessary pragmatic competence to handle the
corresponding rules of politeness. Anyone who has traveled out of her country
will have noticed that the politeness rules of her native culture do not always
apply in cross-cultural or intercultural situations, and therefore new politeness
strategies have to be learnt in order to survive. A more detailed account of the
phenomenon of linguistic politeness will be made in Chapter 4, but we shall
extend our discussion for this section only by making reference to some instances
of pragmatic mismatches which have occurred or might occur due to so-called
"cross-cultural illiteracy" (Luce & Smith 1987).

Wierzbicka (2003: 49) for instance, explains how in the Polish cultural
tradition it is valued and desirable to hold firmly to one's beliefs (and not to make
concessions to those of others), while in what she calls the Anglo-Saxon tradition
similar attitudes "would be regarded as dogmatic and inflexible". She herself (as
a Polish woman living in Australia) had to learn new modes of interaction, such
as new rules for turn-taking (e.g.: interrupting one's interlocutor as he is speaking
is not perceived as rude behavior in Polish, as opposed to what she perceives to
be the case be in English), or different rules for the use of the imperative form
(e.g.: telling someone 'Do this!' in the same situations as she did in Polish would
be considered too imposing or rude by an English speaker). In general, and as
she herself puts it, she had to learn to be "less 'emotional' and more 'dispassionate'
(at least in public speaking and academic writing)" (1997: 120). It must be added
 //I///// I// I I I I/ I I// I I I I//////// I I I I I I I I Pragmatics : definition and scope 25

here, as we shall also try to clarify after the next example, that the tendency to
reify cultures and to operate as though there were an identifiable "Polish culture"
or an "Anglo-Saxon culture" may be dangerous and misleading, for there are
many more variables, other than the country or region of origin, that intervene in
communication and therefore affect the resulting discourse.
Pragmatic mismatches may obviously occur between speakers of different
varieties of the same language as well. For instance, the use of the imperative in
Peninsular Spanish, as in example (20) - where there is no rudeness implied or
intended, because A & B know each other and therefore A knows that B will
recognize her voice and she does not need to "beat around the bush" to make her
sister open the door - may be interpreted as rude by a speaker of (Latin-American
Spanish, who would tend to use the formula shown in (21):

(20) (A is ringing the bell of her sister's entry phone at the main door.)
A: (rings bell)
B: 6Sf? (Yes?)
A: jAbre! (Open up!)
(21) (A is ringing the bell of her sister's entry phone at the main door.)
A: (rings bell)
B: 6Quien es? (Who's there?)
A: Soy yo. 6Me puedes abrir? (It's me. Can you open the door?)

Needless to say, and as was pointed out above, these are very general
comparisons, and the refore they may apply or not depending on many variables,
such as the intonation used, the age of the interlocutors, the relationship between
them, the sub-culture within the general culture in which they are interacting, etc.
For instance, the rules may vary within the Peninsular Spanish situation
depending on whether the interactants are from Madrid or Andalucia, or whether
they are Romani or white, immigrants or natives, white collar or construction
workers, etc.
We would thus like to point out that comparing pragmatic systems is no easy
ask. Many attempts have been made to find universal patterns or a tertium
comparationis to be used as the basis for comparison. Two honorable and very
well-known examples of such attempts are Grice's (1975) Maxims of the
Cooperative Principle, or Brown and Levinson's (1987) politeness strategies. As
we shall see in Chapter 4, Grice advocates for the existence of four main maxims
of conversation (Quality, Quantity, Manner and Relation) that are universally
followed by the speaker of any language, and Brown & Levinson argue in favor
of the universality of their model of politeness, which is based on the abstract
n otion of face. But contrary to Grice or Brown & Levinson, there are other
26 Pragmatics: definition and scope /II I I I I I I I II I I I I I I I I I I I I I I I I I I I I I I I I I I I I

authors (e.g. Kramsch 1993 or Wierzbicka 2003) who advocate for cultural
relativism (i.e. the view that the norms are specific to each culture) and have
persuasive arguments against the existence of pragmatic universals. And even
more relativistic are the views of some authors, such as Scollon, Wong Scollon &
Jones (2012), who argue against the very concept of 'culture', on the grounds that
it is a fuzzy and tricky term to define and conseq,uently cannot be used as the
basis for any kind of comparison.
In spite of these differences of approach, all researchers into trans-cultural
phenomena have to find a way to obtain reliable resulfs from the comparison of
languages and cultures, and they are therefore always in search of the most
appropriate units of analysis. There is no agreement as to which those units
should be and how many there are of them, but it is reasonable to state that the
choice of unit will depend on the purposes of the study and the research q,uestions
posed. Conversational routines (Coulmas 1981) and discourse systems (Scollon,
Wong Scollon & Jones 2012) could be considered two such units.
Coulmas suggests that most conversational language is predictable, in the
sense that our ability to determine what to say and how to s_a y it in a given
situation is the product of what we have heard others say in previous interactions.
These choices then turn into routines which give a sense of authenticity to our
speech and make conversation fluent; for this reason, a lack of knowledge or use
of such routines would make a participant 'culturally illiterate'.
Scollon, Wong Scollon & Jones point to the fact that what people refer to as
'cultures' (be it Spanish, Dutch, English or any other culture) may differ
significantly from one another and that there is hardly any dimension on which
cultures could be compared "and with which one culture could be clearly and
unambiguously distinguished from another" (2012: 277). For this reason, and in
order to overcome this analytical obstacle, they have introduced the notion of
discourse systems, which "are defined in terms of four interrelated components:
ideology, face systems, forms of discourse, and socialization" (2012: 277).
Discourse systems are complex systems which interact with one another and in
which each speaker participates throughout her life. All of them adopt a given
discursive form of expression, and they include the social practices and the
values of the group in q,uestion, but they will differ for every speaker, even within
the limits of the same 'culture': it is interesting to observe and wonder how it is
possible for a speaker to have a sense of stable identity when she is navigating
across and within all these identity sources (e.g. the discourse system
corresponding to her gender, her sexual identity, her country, the historical
period she is living in, her work, her hobbies, etc.). This is the reason why Scollon
et al. try to avoid the term intercultural communication, favoring interdiscourse
system communication (2012: 278).
I I I I I I I I I I I I I I I I I I I I I I I/ I I I I I I I I I I I I I I I I Pragmatics: definition and scope 27

Much more could be said about conversational routines, discourse systems,

or any of the other possible units of analysis that could help us understand the
phenomena of transcultural and/or intracultural communication. Likewise, much
more could be added about many other aspects of these phenomena, but for the
purposes of this chapter we shall simply remark that they cannot be studied
without being looked at through a pragmatic prism, which allows us not only to
see the purely linguistic facts but also their relationship to speakers, hearers,
place, time, etc., among numerous other variables.

1. 8. CONCLUSION
The purpose of this chapter has been to offer a first impression of pragmatics. We
have seen how this branch of linguistics can be approached from both a broad
and a narrow perspective. Its focus is always on understanding the fact that we
regularly communicate much more than we actually say. The standard dictionary
definitions of our words do not cover all the meanings that are conveyed in
everyday verbal interaction, which is why we sag that these meanings are
nnderdetermined. We have seen the fundamental role of context, and it is here
t the breadth of pragmatics shows itself: so many different levels and types of
con text conspire in complex ways to invest our words with a richer meaning than
ey themselves can carry. This has led to the realization that meanings are
· erwoven with the nature of human cognition and with the social groups and/
cultures in which we live and have our being. Pragmatics thus cannot but
rlap with the cognitive and social sciences.
In the next five chapters, we will delve into the major theories of pragmatics,
owing how each has contributed to providing a deeper understanding of the
estions that have been raised in this chapter. We will start with work that arose
a philosophical tradition but which q,uickly entered linguistics and became an
.Dl[lSJ)ensable component of almost all pragmatic studies: the Theory of Speech

MMARY
Pragmatics is the branch of linguistics that deals with what people mean by
the things they say. It thus overlaps partially with semantics, the study of
iimeless meanings', but it focuses mainly on the meanings that have to do
lrith the relationship of language to its interpreters and the goal of their
erances within a given spatial, temporal and situational context. Language
28' Pragmatics : definition and scope ////////////////////////////////////////

users have pragmatic knowledge, built up over a lifetime, which allows them
to use and understand language appropriately and thus to know when to
interpret an utterance as ironic, how to relate pronouns to context, how to
adapt to another culture, etc.

2. A central notion in pragmatics is context. Every verbal interaction takes

place in a context, which develops and changes as the interaction
continues. Different kinds of context are distinguished: linguistic, social,
socio-cultural, cognitive and emotional-attitudinal; these all enter into
complex interplay with one another in any communicative event. The
linguistic context is a combination of the speech situation, predictions
about upcoming utterances and the overall genre setting. The social
context covers the broader context in which the speaker and addressee
have roles and identities which are in counterpoint with their communicative
goals. The socio-cultural context is broader still, reflecting the impact of
community and even national culture on understandings. The cognitive
context takes us back to communicating individuals, referring to the
mental processes that allow them to perform appropriately within their
own social and cultural contexts. This is complemented by the emotional-
attitudinal context, which is the full range of feelings and motivations that
impinge on our verbal communication.

3. One of the main insights of pragmatics is that language forms underdetermine

how they are understood. The implication of this standpoint is that
interpreting an utterance involves using the context (in any or all of the five
senses distinguished in 2) to figure out ( or 'infer') what is intended.
Underdeterminacy applies to many phenomena, ranging from incomplete
utterances, through nominal compounds, to irony, vagueness and ambiguity
and beyond.

4. The study of pragmatics divides into two major schools, known as Anglo-
American and European Continental respectively. The former sees pragmatics
as a module of linguistic theory, alongside semantics, syntax, etc., whereas
the latter takes a broader view, seeing every aspect of linguistic behavior as
having a pragmatic component; the distinction is related to that between
micropragmatics and macropragmatics respectively. In recent years, there
has been a meeting of minds between the two views, and this book reflects
that convergent tendency, although it is careful to avoid the major pitfall for
the macropragmatic view, the loss of any distinction between pragmatics and
discourse analysis.

5. Pragmatic knowledge is a phenomenon that has attracted the interest of the

cognitive sciences, giving rise to the new (inter)discipline of Cognitive
 I I I I I I I I I I I/ I I I I I/ I I I I I I I I I I I I I I I I I I I I I Pragmatics : definition and scope 29

Pragmatics. A central tenet of this approach is that language is inseparably

interwoven with the totality of the human mind and body. The q_uestion then
arises whether our thinking is so bound up with the distinctions made in our
language that we actually think differently according as we are speaking one
language or another.

- An important focus of work in pragmatics is on indirect meanings, whether

these come about intentionally or not. Indirect meanings can only be
understood by carrying out inferences, i.e. using a combination of reason
and contextual knowledge to work out what is intended. This is not a rare
aspect of language use; on the contrary, our communication is full of indirect
meanings. Being so common, some of these indirect meanings have come to
be conventionalized. In other words, they have become the direct way of
saying something, so that inference is no longer needed. Pragmatics can thus
c ontribute to our understanding of language change. Where a
conventionalized item becomes part of the grammar, we may speak of
grammaticalization.

- Even within one and the same culture, social. differences can lead to
misunderstandings, where the wrong inferences are drawn and/or where
background knowledge is missing. These 'misfires' tend to be even more
common and troubling in trans-cultural communication, where, for example,
issues can arise concerning the norms for politeness. There is ongoing
discussion about the extent to which the conversational maxims developed
by Grice and the Politeness Theory developed by Brown & Levinson are
universal in their application. Rather, integrating into an alien culture may be
a matter of learning to follow the 'routines' that apply there or to navigate the
local multi-dimensional 'discourse systems'.
30 Pragmatics : definition and scope ////////////////////////////////////////

SELF-EVALUATION QUESTIONS
Choose the most appropriate answer for each q,uestion (unless it is
specified that more than one answer is possible):
1. Linguistic pragmatics, as an independent field of study, originated in
a) Classical Antiq,uity
b) The 20th century
c) The beginning of the 21st century

2. By timeless meaning Grice (1967) referred to ...

a) the meaning of an expression that has nothing to do with time.
b) what someone means by using an expression.
c) the meaning of an expression as a composite of the dictionary meanings
of its parts.

3. Pragmatics is mainly concerned with ...

a) utterance or contextual meaning.
b) abstract meaning.
c) timeless meaning.

4. The context of an utterance contains ...

a) only linguistic information.
b) only extra-linguistic information.
c) both linguistic and extra-linguistic information.

5. The cognitive context refers to ...

a) the preceding and upcoming words uttered by the speaker.
b) the information about the speaker's physical environment.
c) the mental processes involved in the process of communication.

6. All types of context ...

a) both affect and are affected by one another.
b) are independent of one another.
c) are not found in all utterances.
! //l///!//ll!l/!l//ll!I/I//I///I/I////I Pragmatics : definition and scope 31

7. The 'underdeterminacy' of language refers to the fact that ...

a) one can never determine the meaning of an utterance.
b) any utterance can have more than one meaning.
c) there is always a hidden meaning behind every utterance .

. What is the most probable meaning of the following underdetermined

utterance, if said by someone who's just visited her friend Paul and has had
tea in the garden with him?
I saw the trees in Paul's garden.
a) I perceived with my eyes the trees that were in Paul's garden.
b) I normally cut the trees in Paul's garden with a saw.
c) Neither of the above.

. Specify the possible meanings of the utterance "The Scotsman blew through
his pipe to clear it" (More than one answer can Qe chosen here)
a) He removed dirt from the plumbing tube.
b) He removed spittle from his bagpipe chanter.
c) He removed tobacco from his favorite smoking device.

. The European Continental School views pragmatics as ...

a) a core component of a linguistic theory.
b) the study of language and communication.
c) the study of pure linguistic data.

1. Micro-pragmatics is concerned with ...

a) a perspective more than with particular discussions.
b) discussions about specific topics, such as deixis or inferences.
c) the dichotomization of the field.

Cognitive Pragmatics focuses on ...

a) the relationship between language, mind and body.
b) the generation of ironic meanings.
c) the language of brain-damaged individuals.
32 Pragmatics: definition and scope I I I I I I I I I I I/ I I/ I/ I I/ I I///// I I// I/ I I I I I I I

13. The utterance He's poor but honest constitutes ...

a) a direct way of saying that poor people are not honest.
b) a direct way of saying that poor people are honest.
c) an indirect way of saying that poor people are not honest.

14. B's utterance in the following exchange shows ...

A: Isn't Lily wonderful?
B: If she is wonderful, I'm Marilyn Monroe
a) an indirect conventionalized strategy to mean B does not agree with A.
b) an indirect grammaticalized utterance to mean B does not agree with A.
c) a direct expression to mean that B does not agree with A.

15. Cross-cultural communication differs from intercultural communication in

that ...
a) speakers within the former belong to more than two languages and
cultures.
b) speakers within the latter always have to use a lingua franca.
c) speakers within the former always have to use a lingua franca.

DEVELOPING OUR PRAGMATIC COMPETENCE:

EXERCISES FOR REFLECTION AND AMPLIFICATION
A) Search the web for one or two definitions of linguistic pragmatics. Do
they coincide with the definition and scope of the field presented in
this chapter? In what senses do they coincide or not?

B) The word pragmatics was inspired by the Latin pragmatic us, an

adjective taken from the Greek npo. 1 µo.TLt.,6, (pragmatikos), which
means 'fit for action', which in turn comes from npo. 1 µo. (pragma),
'deed, act', and from npo.crcrw (prasso), 'to pass over, to practise, to
achieve'. What do you think is the connection between its origins
and the sense given to it within the discipline of linguistic
pragmatics? Write a paragraph with your conclusions and ideas
about it.
 1 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I/// I I Pragmatics: definition and scope 33

C) Think of at least two possible contexts in which the following

utterances could occur. Does the meaning change in each case?
a. The manager told the forwards to use their heads.
b. The singer totally failed to move me.
c. Wall Street was badly shaken.
d. As I rushed down the stafrs, I ran into my friend.

D) Suppose you are a woman (if you are not) and you've just learnt that
you're pregnant. Considering the different contexts implied, how
would you tell the news to: a) your friend from childhood? b) your
boss? c) your mother? d) an orthopedic surgeon who prescribes you
an X-ray? e) your husband/your significant other f) the baby's father
(if he is not your husband or your significant other)?

Think of the ways in which the foil owing utterances are underdeter-
mined and ambiguous, and discuss the possible meanings they could
have depending on the context of occurrence:
a. Mary's right
b. The President's right hand is in the corner
c. He saw a woman on a hill with flowers
d We saw her run

Discuss the possible indirect meanings that the utterances in italics

( containing the expression Thank you very much) might have in the
following three examples:
I) A: I'm having a party tonight and I was wondering if you could lend me
your beautiful black dress for the occasion.
B: I'm sorry, I never lend my clothes to anyone.
A: Well, thank you very much!!
2) A: I was thinking that you could give the inaugural lecture this year.
B: Me??? Inaugural lecture??? No way, thank you very much.
3) Mother: That's q_uite enough noise now, children; thank you very much.

Think of possible situations in which you could be involved in:

a Intracultural communication
b. Intercultural communication
c. Cross-cultural communication
 WHAT WE DO
WITH LANGUAGE:
SPEECH ACTS
/I I///////////////////////////////////
N OBJECTIVE OF THE CHAPTER: To define and explain the
pragmatic concept of speech act and identify its different types

TOPICS OF THE CHAPTER:

Austin's (1962) Theory of Speech Acts and its contrast with logical
positivism
P erformative acts ( definition and types)
Types of speech acts: Locution, Illocution and Perlocution
Relationship between the grammatical form of utterances and their
function
Searle's (1969) taxonomy of speech acts
Direct and indirect speech acts
P ost-Searle studies and work on speech acts
36 What we do with lan9ua9e : Speech ads // // //////// // /////// ///////// ////

2.1. ON SAYING AND DOING

Words are not just words. As speakers of a human language, we all know that
our words can praise, hurt, convince or disappoint, among numerous other
things. By way of illustration, if a friend told you that you are stupid, you would
most likely report the event by saying that your friend insulted and therefore
offended you, thereby referring more to the action(s) performed than to the
actual words uttered by your friend. And indeed, by doing so you would be
somehow expressing that what really affected you was what your friend DID to
you with his words. Another example can be found in (1), a simple exchange
to which we think everyone can relate, where B is not so much concerned about
the meaning of the words "You're different" but about what B is trying to do by
saying them:

(1) A: You're different.

B: Is that a compliment?
As we know, 'being different' could mean that B is special, beautiful,
intelligent, etc. (in which case A would be praising B), but it could also mean
q_uite the contrary, so it is clear that what matters to B here is what A intends to
do with her words, rather than what the dictionary definition or timeless
meaning 15 of "different" is.
One might conclude that it should be q,uite clear to any body's eyes that we
not only say but also DO things with language. However, this idea of relating
actions to words is a relatively new one, and is considered to be one of the
milestones in the emergence and development of pragmatics as a sub-discipline
of linguistics. The observation that concrete actions can be carried out by means
of words lead John Austin to develop a theory of speech acts in his well-known
monograph entitled How to do things with words (1962), which is the published
version of his William James Lectures delivered at Harvard in 1955.
As Sadock (2007: 53) observes, the theory of speech acts "is especially
concerned with those acts that are not completely covered under one or more of
the major divisions of grammar - phonetics, phonology, morphology, syntax,
semantics or under some general theory of actions". This means that, when we
speak of speech acts in pragmatics, we do not refer to acts such as reducing the
q_uality of a vowel or constructing a complex clause. We refer to what Austin
called illocutionary acts, as we shall explain right away in 2.2.

15 See 1.1 for Grice's (1967) definition of timeless meaning.

1I IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII What we do with language : Speech ads 37

2. 2. AUSTIN'S REACTION TO LOGICAL POSITIVISM

AND TRUTH-CONDITIONAL SEMANTICS
The reader may find it interesting to know that Austin was not a linguist, but a
philosopher who worked at Oxford University in the 1940s and 1950s. He was
Paul Grice's teacher, and they both - together with a group of like-minded
philosophers -were later known as 'ordinary language philosophers', for they set
the basis for what may be regarded as a change of paradigm in the study of
language, based on a new conception of meaning.
Previous to Austin, other philosophers belonging to the logical positivism
earn of thought (and working at the University of Cambridge), such as
Bertrand Russell and George E. Moore , had also been concerned with the
relationship between philosophy and language, but from a different perspective.
Russell, for instance, took the perspective of truth-conditional semantics, thereby
,resenting a view of everyday language as a deficient vehicle which is full of
biguities and contradictions and which therefore should be refined in search
for an ideal language. Logical positivists maintained th,at the only meaningful
tements are those that can be tested empirically, and they were conseq_uently
d mainly interested in the properties of sentences which could be evaluated in
of truth or falsity. Within this frame of mind, the only meaningful sentences
ould be those that can be assessed as true or false, such as (2), and other
tences such as (3) would be meaningless because, strictly and logically
speaking, it is not true that the speaker's heart was broken into pieces .

. . :) London is the capital of the United Kingdom.

My heart was broken (when I saw a child suffering).

In opposition to this view, Austin claimed that a sentence like (3) IS

mearu·mgful, and that people are normally able to communicate without major
blems by using language just as it is, thereby calling the attention of scholars
e idea that language is not imperfect or defective, but on the contrary, is a
gnificent tool whose functioning should be studied in order to understand how
- that humans manage to communicate so well by using it. Indeed, Austin's
?.) first observation was that most utterances have no truth conditions, an
rvation that emerged from his analysis of the behavior of what he termed
ormative acts (see 2.3).
Austin was therefore convinced that there is much more to language than
'raw' meaning of its words and phrases, and this is what led him to develop
- theory of illocutionary acts. In developing this theory , he explored the
afferent things that we do when we speak and how we do them. He also analyzed
38 What we do with language: Speech ads IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

what determined the probabilities of success or failure in performing such

actions. This led him to develop the notion that illocutionary acts have felicity
conditions.
We shall now examine Austin's theory in more detail (2.3 and 2.4).

2.3. PERFORMATIVE ACTS

In his famous Harvard lectures (delivered in 1955 and posthumously published in
1962), John L. Austin (1911-1960) pointed to the fact that certain types of
utterances, such as (4) and (5), are designed to change the state of affairs and
actually do something (here to marry a couple and declare war, respectively)
rather than to say something. He called these performatives, as opposed to
constatives, the latter being the descriptive sentences that had been the main
concern of the positivist philosophers, i.e. those that could be described under the
conditions of truth or falsity.

(4) I pronounce you man and wife (uttered by a priest at a wedding ceremony).
(5) We declare war upon the Empire of Japan (uttered by U.S. President
Roosevelt, on December 8, 1941).

In order to identify a performative act, it is useful to test whether the adverb

lwreby can be inserted between the subject and the verb. If it can, then it is most
likely that the utterance in q,uestion is an example of a performative act. Compare
(6) and (7), where the test would tell us that certify is a performative verb (since
it shows an appropriate use of the adverb hereby in English), while describe is
not, considering that inserting lwreby before this verb would be ungrammatical
and inappropriate; this is shown by the asterisk (*):

(6) I hereby certify that this document is an authentic copy of the original
diploma.
(7) *I hereby describe a landscape.

It is important to point out that,.in spite of his initial differentiation between

performatives and constatives, Austin later argued that this distinction was not
defensible, for the facts showed that all utterances have both a descriptive and an
effective aspect, namely, that saying something always implies doing something.
Thus, utterances like those in (4), (5) and (6) constitute a sub-type (declarations,
according to Searle's 1975a later classification) of what he called illocutionary
acts.
 I I I I/ I I I I I I I I I I I I/ I I I I I I I I I I I I I What we do with language: Speech ads 39

However, because of the great influence and relevance that the theory of
ormative acts has had within pragmatic studies, we find it pertinent to devote
me time and space to describing these acts in more detail.

.3.1. TYPES OF PERFORMATIVES

performatives have common characteristics. They are self-verifying
containing their own truth conditions 16), self-referential (the verb refers to the
tion the speaker is doing) and non-falsifiable (they can never be untrue).
owever, Thomas (1995) observes that not all performatives are the same in
ture, and she classifies them into four types: 1) Metalinguistic, 2) Ritual,
Collaborative, and 4) Group performatives.
Metalinguistic performatives are the most direct and clear kind of
ormatives, because the verb in itself is very explicit as to what type of action
e speaker is performing by means of her utterance. Examples (8) to (10)
trate this type:

) I apologize for having been disrespectful to you . .

(9) I sag that Kevin is innocent.
10) I object to his contract being renewed.

Notice that (11), (12) and (1~) are their non-performative counterparts,
cause they do not contain the corresponding performative verb, and are
erefore not self-verifying, but subject to truth conditions:

11) I'm sorry for having been disrespectful to you.

12) Kevin is innocent.
13) I do not want his contract to be renewed.

Thomas (1995: 34-35) explains that saying "I am sorry" as in (11) is a direct
tement about one's feelings which is perceived by native speakers of English
o be more sincere than saying "I apologize" as in (8), the latter often being
experienced as something the speaker has said for form's sake. Similarly, (12) is
different from (9) in that (12) has truth conditions: if in the real world it can be
own that Kevin is innocent, then (12) is true. However, (9) is self-verifying
because, strictly speaking, the speaker is only making a statement about what
e is saying, not about Kevin's innocence, and therefore an utterance starting

16 This means that what can be self-verified is that the action named b9 the performative
is the action being uttered, e.g., in (9) above, what is self-verif9ing is that the speaker
saus it, NOT that Kevin is innocent. This last statement ("Kevin is innocent"), however,
is subject to truth conditions and thus not self-verif9ing.
40 What we do with language: Speech ads I I I I I I I I I I I I I I I I I I Ii I I I I I I I I/ I I/ I I

"I say that ... " will always be true in strictly logical terms. In the case of the pair
(10) and (13), (13) expresses the speaker's desires, while in (10) she is making her
opposition explicit or possibly registering a formal demurral.
As the reader may have noticed, the difference between the pairs of
utterances above is a very subtle and even tricky one, and this is one of the
reasons why Austin concludes in his final lectures that the theory of performatives
is unsustainable. However, these subtle differences from the initial lectures are
useful for understanding and interpreting discourse situations like the following:
(14) a. Jane (to her boyfriend): I'd like to go to the movies tomorrow.
Leonard: 0.K. Let's go to see the latest Woody Allen film tomorrow
evening. I'll pick you up at 7:00.
(The following day Leonard calls Jane to tell her he has other commitments
and cannot pick her up to go to the movies, to which Jane replies):
b. Jane: But you invited me and promised we'd go to the movies today!
Leonard: I didn't PROMISE you anything! I just expressed my intention of
going with you; that's all!

Example (14) shows that, even if from the illocutionary point of view the
utterance O.K. Let's go to see the latest Woody Allen film tomorrow evening is
normally taken as an invitation, and I'll pick you up at 7:00 as a promise, Leonard
takes advantage of the fact that he did not use the performative verbs invite and
promise in order to justify himself. This is one more instance of how speakers
make use of pragmatic resources and nuances in order to save face 17 or try to get
out of delicate discourse situations 'unscathed'.
Ritual performatives are the kind of performatives that are associated with
certain rituals or very formal events, and which are therefore highly dependent
upon the particular culture in which they occur. Example (4) above, as well as
(15) to (17), are prototypical examples of ritual performatives:

(15) I name this ship The Conq,ueror.

(16) I baptize you in the name of the Father, Son, and Holy Ghost.
(17) I sentence you to 20 years in prison.

Ritual performatives are more dependent onfelicity conditions than any

other type of performative. We shall deal with the felicity conditions of speech
acts in 2.3.2, but for the time being, suffice it to say that one felicity condition for
the performance of the acts in (4), (15), (16) and (17) is that the person uttering

17 See Chapter 4 for the concept offace.

 I / I/ I//// I I I I I I//// I I I I/ I I I////// What we do with language: Speech ads 41

· has to be an authority who is allowed to, respectively, marry couples, name

ships, baptize children or sentence people to death. Otherwise, the act would not
be 'happily' performed, and it would have no effect whatsoever. This is the reason
why even when, as is the case with all performatives, nobody can attribute truth
conditions to any of these utterances by saying That is not true!, Austin observed
that in these cases it would be perfectly valid to say You have no right to do that!.
Collaborative performatives are the kind of performatives that req_uire the
collaboration of another person for its success to be guaranteed. So in examples
1 ) and (19) both the bet and the challenge will only be successful if the other
person accepts the bet and the challenge, respectively.

l ) I bet you £1,000 that Real Madrid will win the game tonight.
19) I challenge you to a battle of wits.

Group performatives are the type of performative speech acts that are
o rmally uttered by more than one person, and they only succeed if they
are performed on behalf of all the people involved. Thomas explains that group
performatives may also belong to any of the three oth~r categories (metalinguistic,
rimal or collaborative), as shown in (20), (21) and (22):

L20) Combination of group and metalinguistic performative:

(from the Constitution of the United States of America): We, the People of
the United States, in Order to form a more perfect Union, establish Justice,
insure domestic Tranq_uility, provide for the common defence, promote the
general Welfare, and secure the Blessings of Liberty to ourselves and our
Posterity, do ordain and establish this Constitution for the United States of
America.
:.1) Combination of group and ritual performative:
(A verdict from a jury): We find you guilty of first degree murder.
~) Combination of group and collaborative performatives
(A statement on a joint will made by a married couple):
We bequeath all our properties to our two children.

As can be deduced from these examples, and as Thomas (1995: 42) herself
acknowledges, there is an overlap of categories, for it could be argued, for
:=15",t.".knce, that many ritual performatives are also collaborative in nature or vice-
a. And in some cases it can even be argued that the performative belongs to
ee or even all four of the categories, such as the act of beq_ueathing in (22),
wmch is collaborative because in order for it to be successful in legal terms the
drildren will have to accept the goods that are beq,ueathed to them by their
42 What we do with language: Speech ads //////////////////////////////////

parents, but at the same time it also constitutes a group and a ritual performative
given that the parents have joint ownership of all their goods and therefore have
a common will, and that the act of beq_ueathing implies some rituals without
which it could not be 'happily' performed.

2.3.2 . FELICITY CONDITIONS

As anticipated above, there are some ways in which performatives can go wrong,
or be infelicitous. A very simple and obvious example of this would be for
someone who is neither a public officer nor a priest to say to her two friends "I
hereby pronounce you man and wife", in which case it would most likely be
interpreted as a joke or (perhaps) as a sign that the speaker is mentally insane,
considering that she has no authority or public status to marry anyone. Thus the
act would not meet the necessary felicity conditions and would not be
accomplished. Austin (1962: 14-15) distinguished three main categories of felicity
conditions:
(A. 1) There must exist an accepted conventional procedure having a certain
conventional effect, that procedure to include· the uttering of certain
words by certain persons in certain circumstances, and further,
(A. 2) the particular persons and circumstances in a given case must be
appropriate for the invocation of the particular procedure invoked.
(B. 1) The procedure must be executed by all participants both correctly and
(B. 2) completely.
(C. 1) Where, as often, the procedure is designed for use by persons having
certain thoughts or feelings, or for the inauguration of certain
conseq_uential conduct on the part of any participant, then a person
participating in and so invoking the procedure must in fact have those
thoughts or feelings, and the participants must intend so to conduct
themselves, and further,
(C. 2) must actually so conduct themselves subseq,uently.

When one or more of these conditions are not met, Austin observes that the
cause may incur one or more of the following three infelicities:
A) Misinvocations: These occur when any of the A conditions above are not met,
and the refore the purported act is disallowed. The example of the
unauthorized person saying the words of the marriage ceremony clearly falls
within this category.
B) Misexecutions: These occur when any of the B conditions above are not met,
and therefore the purported act is vitiated by errors or omissions. This
 IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII What we do with lan9ua9e: Speech ads 43

category includes examples in which, for instance, an appropriate authority

baptizes a child, or pronounces a couple man and wife, but uses the wrong
names or formulas, or does not sign the corresponding documents, in which
case the purported act does not take place.
C) Abuses: These refer to those situations in which the act succeeds but is
"hollow" and therefore does not meet the C conditions, because the
participants do not have the expected commitment or feelings associated
with the felicitous performance of the act in q,uestion. Prototypical examples
of abuses would be insincere promises, fake apologies or congratulations,
unfulfilled invitations, etc.
In spite of the great influence that the concept of infelicity had on subseq,uent
ork by linguists and philosophers, Austin came to the conclusion that his theory
ed to make a substantial distinction between performatives and constatives in
· respect, because both types of acts req,uire some correspondence with the
"real' facts which has to do with their felicity conditions. That is, constatives can
o be unhappy in the manner of performatives. For instance, if we take
ples (8) and (11) above, where (8) is a performa_tive (I apologize ... ) and (11)
constative (I'm sorry ... ), (11) can also turn out to be an infelicitous act (of the
e kind) if the speaker does not 'really' (in her heart of hearts) feel sorry.

As the reader may have concluded from this last example, there are some
es in pragmatics which are hard to determine (how do I know if someone is
y sorry??), but this has not discouraged pragmaticians in the least. On the
ntrary, it has generated further research and work into realms which had
viously been thought of as unconq,uerable. One of these realms, in which
gmaticians specialize, is the realm of implicit meanings. To conclude this
·on, we shall try to make the distinction between explicit and implicit
ormatives more precise .

.3. EXPLICIT AND IMPLICIT PERFORMATIVES

his lectures, Austin warns his audience that there is an internal evolution to
- argument, so that what is proposed at the beginning will be rejected by the
On the one hand, what is viewed at the beginning as a special class of
ences 18 with uniq,ue syntactic and pragmatic properties (performatives) is
11[1~nted at the end as a general class of act that includes both explicit (the
explained in 2.3 and 2.3.1 above) and implicit performatives (practically

should be noticed that Austin writes about perjormative sentences and perjorma-
·ve utterances rather interchangeably, although at some point he states that not all
ranees are sentences (1962: 6).
44 What we do with language : Speech ads ////////////// ////// /////// ///// //

any other kind of utterance). On the other hand, a shift is observed from the
opposition between performative and constatives to a general theory of
illocutionary acts, in which performatives and constatives are just different
subtypes of the same phenomenon.
As for the difference between explicit and implicit performatives, it could be
said that when using explicit performatives the speaker is being very specific and
unambiguous about the act being performed, whereas when using implicit
performatives the speaker is less specific and, instead of using a performative
verb, uses other devices, such as discourse markers, adverbs, mood or even
intonation to indicate what she intends. Compare the pairs of examples ( a & b) in
(23), (24) and (25), where a) is the explicit performative and b) is a possible
implicit counterpart:

(23) a. I saw him cheating at the exam. I conclude that he is a dishonest person.
b. I saw him cheating at the exam. Therefore, he is a dishonest person.
(Device used to indicate it is a concluding remark: Discourse marker
Therefore)
(24) a. I order you to leave this room immediately.
b. Leave this room immediately. (Device used to indicate it is an order:
Imperative mood)
(25) a. I assure you I will send you the document tomorrow morning.
b. I will send you the document tomorrow morning for sure. (Device used
to show it is an assurance: adverbial for sure)

The reader should note, however, that, as pointed out above, in some cases
the force of the intended act may be given by means of the intonation used (in
conjunction with other discourse variables surrounding the act), in such a way
that, for instance, the utterance in (26) could be a threat, a promise or a complaint
( among other things):

(26) I will not come to your house again.

And even if the performative verb is used (i.e. I hereby promise that I will
not come to your house again), the circumstances of the particular discourse
situation could be such that they oblige the hearer not to interpret the utterance
as a promise, but as a joke, for instance. This points to the fact that, in some
cases, not even the presence of a performative verb can be taken as infallible
proof that the action intended is the one expressed by the verb.
To conclude this section, we find it relevant to remark that, as Austin himself
admitted at the end of his lectures, there is no substantial distinction in meaning
 , /II I I/// I I I I/ I I I/ I I I I I I I I/ I I I I What we do with language: Speech ads 45

tween explicit and implicit performatives. Both may carry the same kind of
utionary force. However, and as Thomas (1995) observes, some situations
--~--· e that the performative verb be used, and these are generally more formal,
emn or forceful in nature. For instance, saying I swear to God that I didn't do
- sounds more forceful than just saying I didn't do it, or saying I'll pick you up
rrow to go to the movies, as in example (14) above, is probably perceived as
eaker and less formal promise than saying "I hereby promise that I will take
to the movies tomorrow" (although it should be said that the latter is q,uite
· ely to occur in normal everyday exchange).
As already mentioned, the development of Austin's argument led him to the
n clusion that there is no substantial distinction between constatives and
ormatives of any kind, be they explicit or implicit. He therefore concluded
they should all be included within the broader category of illocutionary acts,
hich we now turn.

TYPES OF SPEECH ACTS (AUSTIN 1962):

LOCUTION, ILLOCUTION AND PERLOCUTION
relevant observation within Speech Act Theory is that utterances not only have
Se!IJSe and reference, but also force. For instance, in (27) Lucy is not only referring
er dislikes by saying she does not like rollercoasters but also rejecting Peter's
_-n-.,,n· on to get on one. This rejection constitutes the force of the utterance and

at Austin called illocutionary force . Likewise, Peter's words (Come on ...

· tn.ke a ride ... ) are not limited to their sense or to what they refer to; they
have the illocutionary force of an invitation.
_ (Lucy and Peter are at an amusement park)
Peter. Come on, Lucy. Let's take a ride on the rollercoaster. It's fantastic!
Lucy: I don't like rollercoasters, sorry.

The act containing the force of the utterance is called the illocutionary act.
this is not the only kind of act that is performed when language is put to use.
&zsc;·m noticed that there are three main kinds of acts that are characteristic of all
erances (which include both his original performatives and constatives):
Locutionary acts, 2) Illocutionary acts, and 3) Perlocutionary acts.
Locutionary acts are the acts involved in the construction of the utterance,
as choosing certain words and pronouncing them, as well as arranging
llll:5e words according to the grammatical rules of, and with the sense and
m en~ce determined by, a particular language. Thus, in (28), the locutionary act
46 What we do with language: Speech ads IIIII III IIII IIIIIIIII IIIIIIIII IIII

consists of saying and pronouncing the utterance Would you help me with the
dishes?, meaning by help 'help', and by dishes 'dishes' (i.e. their semantic
(timeless) meaning) and referring by you to Susan and by me to John:
(28) (John and Susan have just finished dinner)
John: Would you help me with the dishes?
Susan: Sure! No problem.

fllocutionary acts are the acts done or performed in speaking (hence the
name 'illocutionary'). They convey the force or intention behind the words, and
for that reason they are the central type of acts in Austin's theory. As has already
been noted, illocutionary acts include both performatives and constatives, given
that both of these can be said to contain an illocutionary force of some kind. In
example (28), for instance, the illocutionary act performed by John is clearly a
req,uest, while that performed by Susan is an acceptance of the req,uest.
Perlocutionary acts are the by-product of the illocutionary act, having to do
with the effect that the illocutionary act causes on the hearer or other more or
less indirect participants, an effect which may or may not be in accordance with
the speaker's intention. It is often said that a perlocutionary act is not performed
in saying something, but by saying it. For instance, in (28) the perlocutionary
effect intended by John seems to have been to convince or persuade Susan to
help him, and to judge by Susan's answer, he succeeded in his intention. But if
Susan's answer had been different, and she had replied by saying something like
How could you be so inconsiderate? I've been working all day and you still want
me to do the dishes?, then the perlocutionary act produced by John's utterance
would have been one of angering or offending Susan, even if that was not the
effect intended by John. Other prototypical examples of perlocutionary acts
could be comforting, inspiring, inciting or humiliating. A way of testing, in
principle, whether an act is illocutionary or perlocutionary is to use the following
frame:

(29) Speaker PERLOC Hearer by ILLOC.

Applying this to (28) and its alternative, we get John convinced/angered

Susan by requesting that she help him with the dishes; this confirms that
convincing and angering are perlocutions while req,uesting is an illocution.
Even though the distinction among these three types of acts has been very
useful in understanding different pragmatic phenomena, in practice it may be
difficult to draw clear lines between them. Austin himself acknowledged that in
some cases it might be troublesome to separate illocutions from perlocutions. His
suggested solution to this problem was to consider that illocutions are
conventional, in the sense that the act can be made explicit by the performative
 IIIII III II IIIIIII I IIIIII II IIII III What we do with language : Speech ads 47

formula, while perlocutions are not; i.e. in (28), for instance, John could make his
illocution explicit by saying I hereby request you to help me with the dishes, but
could not do so by saying I hereby convince you by asking you to help me.
y authors (e.g. Strawson 1971, Warnock 1973, Urmson 1977 and Sadock 2007)
ter produced works which weigh up the validity and determination of the
erences among these three types of act, but we shall not discuss them in this
k because they go beyond its main objectives and scope .

.5. FORM AND FUNCTION: DO THEY ALWAYS

COINCIDE?
language is a tool that is adapted for communication, we will expect the
aticalform of clauses to reflect the communicative functions of the major
s of illocution. As Sadock (2007: 53) says, the "formal features of the
erance used in carrying out the act might be q,uite directly tied to its
omplishment". To a considerable extent this· is true. English grammar
-n·rnguishes a number of moods of the verb, each of which roughly aligns with
erent broad group of illocutions. The moods of English are three in number.
imperative mood is associated with a range of illocutions that involve
ae1l::i·n,g the addressee to do something ( or stop doing something). The illocutions
range from a recommendation (as in (30a)) through a req,uest (as in (30b))
order (as in (30c) ):

a. Change your password regularly.

b. Pass me the salad.

c. Get out of this room.

The subjunctive mood is not applied very much anymore but when it is used
illocutionary force involves the expression of a wish, a hope or a prayer, as
. I):

) a. God save the Queen.

b. Thy will be done. 19

This sentence is taken from the Lord's Prayer, a prayer which according to Christians
taught by Jesus Christ to his disciples. The prayer is in Early Modem English. In
11taS
~lb) the subject is Thy will ( = 'Your wishes') and the verb phrase in the subjunctive
mood is be done ( = 'may be carried out'), so that the meaning of the whole is 'I pray
your(= God's) wishes are carried out'.
48 What we do with language: Speech ads // / / / / /// / / / // / / / / / / / / //// / //// / / /

All other finite forms of the verb belong to the indicative mood, which is
associated with all other types of illocution. English has a rather impoverished
system of moods (some other languages are much richer in this regard) but it
does use different types of indicative clauses to distinguish further groups of
illocutions. Thus, illocutions of assertion involve Subject-Verb-Object order in an
indicative clause, as in (32a); illocutions that serve to elicit information involve
initial placement of the finite verb or of a q_uestion word in an indicative clause
as in (32b) and (32c) respectively; and illocutions of exclamation can involve
Object-Subject-Verb order in an indicative clause, as in (32d):
(32) a. Peter kicked the football away. (Declarative clause type)
b. Did you see the lightning? (Interrogative clause type)
c. Where did I put my keys? (Interrogative clause type)
d. What a sour face you are making! (Exclamative clause type)

Table I: Moods, clause +ypes and associated illocutions in English

.. . . ;
·,
;

.-
.
MOOD
'
'
CLAUSE ·
.
'T YPE · ,., .
.
· ASSOCIATED,
ILLOCUTION
..
EXAMPLE .
•.

-
'
"'
declarative asserting, stating, ... John has written a fine novel.

indicative interrogative eliciting information Which novel did he write?

exclamative expressing emotion What a fine novel John wrote!

subjunctive hoping, wishing, ... Long live free downloads.

imperative ordering, req,uesting, ... Please come home soon.

As Table 1 makes plain, the indicative is the most general mood and, of the
three, is the one most freq_uently applied. This mood makes no distinction between
performative and constative utterances. Both (33a) and (33b) have the same
basic grammatical characteristics, namely indicative mood, Subject-Verb word
order, 1st person singular pronoun, simple present verb:
(33) a. I apologize for my error. (Performative)
b. I go for a long walk every day. (Constative)

As Austin (1962: 78) points out, not only the grammar but even the choice of
verb may be identical: (34a) is performative (cf. I bid you welcome) while (34b),
meaning that the speaker is happy about the reduction, is constative:
 I I I I I I I I I I I I I I I I I I I I I I I/ I I I I I I What we do with language : Speech ads 49

..;, a. I welcome you to this conference.

b. I welcome the reduction in taxes.

_rotice that only the constative can also appear in an exclamative clause
. while only the performative can be paraphrased as I bid ... welcome:

a. *How I welcome you to this conference!

b. How I welcome the reduction in taxes!
a. I bid you welcome to this conference.
b. *I bid the reduction in taxes welcome.

Form does not always map so neatly onto function. Many of the moods and
-e types discussed above can at times be used to convey an illocution that is
cally associated with another mood or clause type. The indicative mood, for
- -----uple, can be used to express the illocutions normally associated with the
rative mood, as in the declarative (37) or the interrogative (38), which are
_ polite ways of saying 'Tie your shoelaces' and 'Pass me the salad', in the
rative mood:

our shoe laces need tying.

Can you pass me the salad?

1th a rising intonation, a declarative clause type can be interpreted as an

-- ogative, albeit one in which the speaker not only asks a q,uestion but also
-.-.. _.,...., res some skepticism, as in (39):

John has written a novel?

_ ere the speaker is probably indicating she doubts whether John has it in
-o write a novel. The interrogative clause type can, at least in some people's
.....,.,,,'""To. be used to express what would otherwise be an exclamative, as in (40),
• · g 'How happy I am to see you!', in which the q,uestion would have a falling
- ·on, instead of the normal rising intonation for yes/no q,uestions:
_ 1 I happy to see you!

_ e imperative mood can sometimes be used as an eq,uivalent of the

m::~.:1.n·ye mood in a clause beginning with if, as in (41):

_ ess the delete button and your entire book will be wiped from the
computer.

1early, the speaker is not telling the hearer to press that button but means
press the delete button, your entire book will be wiped from the computer'.
50 What we do with language : Speech ads //// //////// /////////////// /// ////

Examples (37-41) have shown that although grammatical form is a guide to the
intended function, there are cases of mismatch between form and function.
Especially in conversation, the relation between the literal meaning of an
utterance and its illocutionary force is often q,uite indirect. The words that are
heard do not give direct access to the speaker's intentions. Rather, the hearer
infers what the speaker will presumably have meant by considering the context
in which the words were said. In (42):
(42) Andy: I've decided I won't be using your services this time.
Bill: Well, I'm not going anywhere.
Andy informs Bill that he has chosen not to employ him on this occasion;
Bill's reply may seem strange or nonsensical, since the conversation is not about
his movements. So, given the context, Andy works out - and Chapter 3 will show
how - that Bill means that he remains available to Andy, should he change his
mind or have some other work for Bill. The literal meaning of Bill's utterance is a
statement but the illocutionary force is that of an offer (to undertake future
work). Another example is (43):
(43) Charlene: Michelle has let us down again.
Dolly: Well, what did you expect, taking on someone with so little experience?

Here Dolly's response to Charlene's statement has the form of an

interrogative, but in terms of illocutionary force Dolly is not asking a q,uestion at
all but reproving Charlene for employing an inexperienced person. If Charlene
took Dolly's q,uestion literally, she would have to answer it and concede that she
had expected Michelle not to disappoint them. Since context and experience
indicate that Dolly is not likely to be forcing the issue here, Charlene will interpret
her colleague's comment as a rebuke that doesn't req,uire an answer. Notice that
Bill's utterance in (42) and Dolly's in (43) both begin with the pragmatic marker
well, which often signals to the other partner in the conversation that the answer
is going to be indirect (see 1.6.1.2).
It is not just answers that can show a complex relation between literal
meaning and illocutionary force. Speakers can also initiate conversations with
utterances of this kind. (44) will often have the force of an exhortation (eq,uivalent
to 'Let's look in Elaine's fridge'); and (45) is not literally the issuance of a wager
but has the force of a committed statement meaning 'It will snow tonight for
sure':
(44) I wonder what Elaine has in her fridge.
(45) I bet it's going to snow tonight.
Both these examples start with the first person singular pronoun and the
simple present. (44) is a constative (since you cannot wonder by saying 'I wonder';
 I I I I I/ I I I I I I I I I I I I I I I I I I I I I I I I I I/ 'N'nat we cto w"it'n \ansuase: S?eec'n acts S\

cf. (7) above, with describe). (45) admittedly has the appearance of a performative
(since you make a bet by saying I bet (you) (SUM-OF-MONEY) (that) ... ), but it,
too, is here a constative since it states the speaker's confident expectations about
the weather.
Examples (42-45) were designed to show how utterances can be interpreted
as having an illocutionary force that is different from the literal meaning of the
words uttered. However, under different circumstances they could have a direct
fo rm-function correspondence. I'm not going anywhere could be a refusal to
move, just as What did you ex pect? could be a q,uestion about someone's
expectations. It all depends on the context. In a court of law, for example,
witnesses are supposed to attend to the literal meaning of the q_uestions put to
them and to answer accordingly. What we may conclude is that every utterance
we hear is interpreted against the background of the context shared by the
speaker and hearer. Where there is a disparity between the literal meaning and
the context, the hearer infers that the illocutionary force, which reflects the
speaker's communicative intentions, is different from what the words suggest.
This observation was the basis for the development of Austin's work by the
philosopher John Searle, who is still contributing aetively to the philosophy of
language and mind, and for Searle's distinction between direct and indirect
speech acts. These matters will be dealt with in the next section.

2.6. TYPES OF SPEECH ACTS (SEARLE 1969)

Alongside Austin's foundational insights, it is Searle's work (1969) on speech act
theory that has been most influential on contemporary linguistics. When
reviewing Austin's locutionary acts, Searle identifies the propositional act as
particularly important. The propositional act ( eq_uivalent to Austin's locutionary
act) consists in turn of two parts: 1) an act of reference, by means of which the
speaker pinpoints a particular object or being through the use of a given noun
phrase, and 2) a predication (also a kind of act).
Searle pointed out that very often the form of an utterance also displays a
bipartite structure: one part expresses the propositional act, and the other the
illocutionary act. He symbolized the former asp, and the latter (i.e. those features
that determine the literal illocutionary force, which he called fllocutionary Force
I ndicating Devices - IFIDs) as F. This he represented with the formula F(p), and
it is prototypically illustrated by the following example:
(46) I promise I will come F(p)

where I promise is F and I will come is (p).

52 What we do with language: Speech ads ///////////////////// ////// /// // //

One of Searle's central ideas was that we perform acts according to certain
rules which he called constitutive rules, given that these are conventional rules
that are constitutive of each different kind of act. Thus, in order to discover the
rules, and in line with Austin's work, Searle proposed to examine the felicity
conditions that must be met for the different types of acts to be performed. For
instance, the rules he postulates for the illocutionary act of promising are the
following (Searle 1969: 63):

1 Pr (the IFID for promising) is to be uttered only in the context of a sentence

( or larger stretch of discourse) T the utterance of which predicates some
future act of A of S.
2 Pr is to be uttered only if the hearer H would prefer S's doing A to his not doing
A, and S believes hearer H would prefer S's doing A to his not doing A.
3 Pr is to be uttered only if it is not obvious to both S and H that S will do A in
the normal course of events.
4 Pr is to be uttered only if S intends to do A.
5 The utterance of Pr counts as the undertaking of an obligation to do A.

According to Searle, rule 1 is the Propositional Content Rule, rules 2 and 3

are Preparatory Rules, rule 4 is a Sincerity Rule, and rule 5 is the Essential Rule.
It is interesting to note that a violation of the Preparatory Rules would produce
an infelicity of Austin's type A (misinvocation), and a violation of the Sincerity
Rule would be more or less eq,uivalent to Austin's third class of infelicities
(abuses); but neither the Propositional Content Rule nor the Essential Rule
seems to have been contemplated in Austin's taxonomy of infelicities (see 2.3.2
above).
One important concern within Speech Act Theory is the classification of
speech acts, a problem that is related to what are called 'speech act verbs', i.e.
those verbs that are the natural way of expressing the different kinds of speech
acts. Searle (1975a) offered a taxonomy of the five types of act he considered to
be basic, which is commonly understood to have been a refinement of Austin's
taxonomy and which to date seems to have been the classification most widely
used by linguists and other scholars. Verschueren (1979: 10) notes that there are
linguists who lump together speech acts into a few, large categories (whom he
therefore calls 'lumpers'), and those who split up speech acts into a great number
of classes (whom he therefore calls 'splitters'), a number that, according to Mey
(2008 [1993): 105), "may go up to the tens of thousands". Within this framework,
Searle would be a lumper, for his taxonomy contains only five main classes. His
classification was based on essentially pragmatic parameters, the most important
of which Searle (1975a) called direction offit, i.e. whether the words fit the facts
of the world or the world comes to fit the words. In the former, the speech act
 IIIII IIIII IIIIIIIIIIIIIIIIIIII III What we do with language: Speech ads 53

reacts to a pre-existing state of affairs, while in the latter, the point of the speech
act is to change the current or future state of affairs. Examples of the former are
assertions, and examples of the latter, promises and req_uests. 20
The five types proposed by Searle are explained and illustrated in Table 2.

able 2: Summary of Searle' s (1975a) taxonomy and explanation

of speech acts

Acts which commit the speaker

Assert, conclude, boast,
to the truth of the expressed
1. REPRESENTATIVES complain, deduce,
proposition
recite
(Words-to-world direction of fit)
Ask, order, command,
Acts which reflect the attempt
request, beg, forbid,
by the speaker to get the hearer
instruct, urge, warn,
2. DIRECTNES to do something (in varying ·
pl,ead, pray, entreat,
degrees).
invite, permit, advise,
(World-to-words direction of fit)
dare, defy, challenge

(Table 2 continues)
Searle (1975a: 3-4) further explains: "The best illustration of this distinction I know of
is provided by Miss Anscombe (1957). Suppose a man goes to the supermarket with
a shopping list given him by his wife on which are written the words 'beans, butter,
bacon, and bread'. Suppose as he goes around with his shopping cart selecting these
items, he is followed by a detective who writes down everything he takes. As they
emerge from the store both shopper and detective will have identical lists. But the
function of the two lists will be q,uite different. In the case of the shopper's list, the
purpose of the list is, so to speak, to get the world to match the words; the man is
supposed to make his actions fit the list. In the case of the detective, the purpose of
the list is to make the words match the world; the man is supposed to make the list
fit the actions of the shopper. This can be further demonstrated by observing the role
of 'mistake' in the two cases. If the detective gets home and suddenly realizes that
the man bought pork chops instead of bacon, he can simply erase the word 'bacon'
and write 'pork chops'. But if the shopper gets home and his wife points out he has
bought pork chops when he should have bought bacon he cannot correct the mistake
by erasing 'bacon' from the list and writing 'pork chops'.
In these examples the list provides the propositional content of the illocution and the
illocutionary force determines how that content is supposed to relate to the world.
I propose to call this difference a difference in direction of fit. The detective's list
has the word-to-world direction of fit (as do statements, descriptions, assertions, and
explanations); the shopper's list has the world-to-word direction of fit (as do req,uests,
commands, vows, promises)."
54 What we do with language: Speech ads I I I I I/// I I I I I I/ I I I I I I I/ I I I I/ I I I I I I

Acts whose point is to commit Promise, threaten,

the speaker (in varying offer, guarantee,
3. COMMISSIVES degrees) to some future course pledge, vow, undertake,
of action. warrant, swear,
(World-to-words direction of fit) volunteer
Acts which express a
Thank, congratulaw,
psychological state specified in
deplore, apologize,
the sincerity condition about a
4. EXPRESSIVES condole, detest,
state of affairs specified in the
welcome, appreciate,
propositional content.
regret
(No direction of fit)
Acts which bring about
Appoint, nominate,
immediate changes in the
christen, declare (e.g.
institutional state of affairs
war), excommunicate,
and thus tend to rely on extra-
5. DECLARATIONS se1'_1,tence (e.g. someone
linguistic institutions.
to death), pronounce
(Both words-to-world and
(e.g. a couple husband
world-to-words direction of
fit)21 and wife), resign

The reader should bear in mind that some of these categories may overlap,
and in some cases it is not so easy to assign an act to a single category. For
instance, there may be a certain overlap between cornrnissives and directives: the
act of inviting in (47) is a directive in the sense that it reflects Mary's attempt to
get Torn to come to her house for dinner, but it can also be seen as a cornrnissive
in the sense that Mary commits herself to preparing some kind of dinner for Torn
and receiving him in her house on the agreed date and at the agreed time.

(47) Mary: It'd be nice if we could spend some time together, so come on, Tom, I
invite you for dinner at my house tomorrow evening at 8.

Searle (1975a: 15) also noted that "some members of the class of declarations
overlap with members of the class of representatives". For instance, even though
( 48) lines up more with the class of representatives and ( 49) with the class of

21 Searle (1975a: 15) explains that this double direction is due to the peculiar character of
declarations (having no sincerity conditions), and that "declarations do attempt to get
language to match the world. But they do not attempt to do it either by describing an
existing state of affairs (as do representatives) nor by trying to get someone to bring
about a future state of affairs (as do directives and commissives.)"
 I I I I I I I I I I I I I I I I I I I I I I I I I I!/ I I I I What we do with language: Speech acts 55

larations, when a judge makes this factual claim in either of these two ways,
th claims will be representatives in the sense that they can be assessed with
ard to word-to-world fit (i.e. as true or false) , but at the same time they both
e the force of a declaration because for legal purposes the accused will be
ty, irrespective of the truth or falsity of the statement in ( 48). Searle noted
t acts like these could be called Representative declarations, because both
are the sincerity condition, a condition that is not met by other types of
declaratives. 22
) Judge: On charge 1 of murder in the first degree, the accused is guilty.
(49) Judge: On charge 1 of murder in the first degree, the court finds the accused
guilty .

In its third column, Table 2 above shows many of the prototypical verbs
(normally called Speech Act Verbs - SAVs) that are used to perform the different
classes of acts, but as is clear from ( 48), it is not strictly necessary to use any of
these specific verbs in order to perform an act, for, as Austin ultimately concluded,
performativity is a characteristic of all verbs, and as we shall see throughout this
book, of all utterances too. Similarly, examples (50) and (51) have no need of the
v erbs thank and request to perform the act of thanking or req,uesting
(respectively), which leads us to the next section in this chapter, on direct and
:indirect speech acts.
(50) A student to her teacher: I would like to express my gratitude for your
kindness in helping me with my homework.
(51) Leo: Henry, could you open the door? (in a situation where Leo has his arms
loaded with boxes and Henry is near the door)

2.6.1. INDIRECT SPEECH ACTS

Imagine Henry's answer to the q,uestion in (51) were:
(52) Yes, I could.

Now imagine he did not respond verbally but simply by opening the door.
Which of the two responses would you consider more appropriate? Clearly, the
latter is the expected and more appropriate response, and in fact, merely to give
an answer like (52) without any accompanying action would be considered
highly inappropriate (and even rude in some contexts). This shows that- within
the above-specified conditions - a competent speaker of English will understand

22 This is a delicate issue that shows some differences between Austin's initial division
between constatives and performatives and which could be disputed, but we are here
only presenting Searle's (1975) point of view.
56 What we do with language: Speech ads / / / / / / / / / / / / / / / // / / / / / / / / / / / / / / / / /

(51) as an indirect req,uest, and not as a q,uestion about the hearer's ability to open
the door. Similarly, given the proper circumstances, the utterances in (53) and
(54) could also be regarded as indirect attempts by Leo to make Henry open the
door:
(53) Henry!!!!
(54) The door!!!

In both (53) and (54) Leo may be indirectly asking Henry to open the door,
given that they are friends, that Leo's boxes are too heavy and he is in a hurry to
dump them somewhere out in the garden, and that the context is rich enough for
Henry to realize what Leo wants.
It is clear, then, that in order to account for indirect speech acts, an analysis
is needed of both the mutually shared background information about the
conversation and the relevant linguistic conventions, since in spite of the
indirectness of (53) and (54), all competent speakers of a language can perfectly
understand the intentions behind these acts. Indirect acts are very freq,uent in
normal everyday conversation, especially in req,uests.
In his analysis of indirect speech acts, Searle introduced the notions of
primary and secondary speech (illocutionary) acts. In his view, the primary act
of utterances such as (51), (53) or (54) above is the indirect one (the req,uest), and
the secondary act is the direct one (performed in the literal utterance of the
sentence). By way of a further example, consider (55):
(55) Lily: We should pay the gardener today.
William: We have no cash in the house.

Here the primary illocutionary act of William's utterance is his rejection of

Lily's suggestion, and the secondary act is his statement that they have no cash
in the house. But the primary act is derived from the secondary act, for in order
to interpret the indirect meaning of William's statement, we first have to
understand the literal meaning of the statement (assessing it within the given
context) and then by way of inference conclude which of the two is the correct
meaning to respond to. This co-occurrence of acts allows us to understand how
it is possible for someone to say something and mean it but at the same time
mean something else. In (55) it is clear that William means both things: a) that
they have no cash in the house, and b) that therefore they won't be able to pay the
gardener for his services today (with the further implication that it will have to
be some other day, after they withdraw money from the bank or ATM).
Searle (1975b) argues that in order to derive primary illocutions from literal
ones, we go through an unconscious process of several steps. In the case of (55)
the steps could look as follows:
 // II I I I I I I I I I I I I I I I I I I I I I I I I I I I I What we do with language : Speech ads 57

ep 1: A suggestion is made by Lily, and William responds by means of an

utionary act. (Facts about the conversation)
ep 2: Lily assumes that William is cooperating in the conversation and that he
made a statement that is relevant. (Principle of conversational cooperation)
p 3: The literal meaning of William's utterance is not relevant to the

p 4: Since Lily assumes that William is cooperating, there must be another

meaning to his utterance.

5: Based on mutually shared background information, Lily knows that, apart

from not having cash in the house, they cannot go to the bank now because it is
and the bank is closed ( and also neither of them feels like going out and
searching for an ATM). She also knows that the gardener has no problem about
waiting until tomorrow or some other day to collect his money. Therefore,
illiam has rejected Lily's suggestion about paying him today. (Mutual
kground information)
p 6: Lily knows that William has intended something other than the literal
eaning, and the primary illocutionary act must have been the rejection of her
proposal. (Theory of speech acts)

Thus, what superficially looks like a statement about the availability of cash
to William and Lily is primarily a rejection of Lily's previous suggestion. As we
c:an see, the apparatus used in order to explain indirect speech acts mainly
includes "mutual background information, a theory of speech acts, and certain
general principles of conversation" (Searle 1975b: 64). And in spite of the detailed
description of the process, the example above is still underdescribed, for we
could break down each part of the process into many more little unconscious
eps.
Indirectness may be said to be universal in the sense that it occurs to some
degree in all languages, but speakers of different languages and cultures do not
always use indirectness in the same way. We are going to study this phenomenon
in more depth in Chapter 4, but for the purposes of this section, it is worth
mentioning that some authors (e.g. Brown & Levinson 1987, or Thomas 1995)
have noted that the variables affecting the use of indirect speech acts are mainly
the following: a) the social distance between the speaker and the addressee,
b) the relative power of the speaker over the hearer, c) the degree to which a
given speech act is rated as an imposition in the given culture, and d) the relative
rights and obligations between the speaker and the hearer. The assessment of
these variables will help the speaker decide how direct or indirect her speech acts
should be for the purposes of the ongoing conversation and discourse situation.
58 What we do with lan9ua9e: Speech ads /////// ///// ///////// //////// /////

Since the first three of these variables will be dealt with in more detail in
Chapter 4, for the sake of illustration we shall here explain the variable related to
the rights and obligations of the speaker. For instance, if a male gynecologist is
in his surgery with a new patient hitherto unknown to him, he will have the right
to ask her to take off her clothes for an examination of her breasts (which also
forms part of his obligations as a doctor if the patient's ailment so req,uires ), so
that he would be entitled to be direct, as in (56):
(56) Take off your blouse and bra, and lie down on the bed.

This same utterance would not be considered appropriate if the same doctor
said it to an unknown woman outside his surgery because in that case he would
not be judged as a gynecologist but as an ordinary man. Now, suppose he were
somewhere else outside his surgery with one or more women friends, in which
case he again would have neither the right nor the obligation to examine them. If
one of the women told him that she was worried about a lump that she had found
in her breast, he might show his intention to help her by being more indirect, as
in (57), for in that situation he would not be taken to be acting as a gynecologist,
but as her friend.

(57) Well, if you don't mind my examining your lump to give you my opinion, I
would need you to take off your blouse and bra.
Thus the type of act would also change: in (56) the doctor is entitled to give
a direct order, while in (57) the doctor-friend has no right to give such an order
and therefore his utterance is just an indirect offer to help her, with no imposition
whatsoever on his part. Notice that the aforesaid applies eq,ually, mutatis
mutandis, in the case of a female gynecologist.
Although this chapter is coming to an end, the topic of indirectness is not.
We will continue treating it throughout the book, since it is key to understanding
many pragmatic phenomena such as implicatures, relevance or politeness.

2.7. THE IMPACT OF AUSTIN AND SEARLE'S WORK

ON LINGUISTICS
The impact of Austin and Searle's work on linguistics has been strong and lasting.
It seems unlikely that Austin's work, which was concerned with language as
action (and thus was thoroughly pragmatic), would have had much effect on the
major streams of linguistic thought at the time, had it not been for Searle's
systematization of speech act theory. Rajagopalan (2000: 365) goes as far as to
accuse Searle of a timely "bowdlerisation" of Austin's work, i.e. the removal of
 I I I I I I/ I I I//// I I I I I I I I I I I I I I I I I I I I What we do with language : Speech ads 59

any aspects of Austin's work that would be unpalatable to generative linguists. Be

that as it may, Searle's approach shares with generative grammar a concern with
an idealized speaker taken out of context; the orientation is to the speaker and
her world (cf. his treatment of words-to-world relations) rather than to the
speaker and her hearer; and, as in the generative tradition, the data are derived
from introspection rather than, say, from the inspection of a corpus. Small
wonder, then, that some representatives of generative grammar sought to
incorporate Searle's insights into their theories.
Among the most famous of these attempts was Ross (1970), who interestingly
does not mention Searle at all but only Austin, and only in the first sentence of the
article. Ross (1970: 224) argues that the insights of speech act theory req,uire that
the syntactic analysis (the 'deep structure') of (58) be (59), represented as a tree
diagram in (60):
(58) Prices slumped.
(5.9) I [Verb, +performative, +communication, +linguistic, +declarative] 23 you
that prices slumped.
(60) s
NP VP

I V NP NP
I

+V
I I
you ---------s
+ performative
+ communication
+ linguistic
-----------
NP
I I
VP

prices V
+ declarative
I
slumped

This analysis of course entails that in the generation of (58) a rule of

"performative deletion" will have to remove the abstract illocutionary clause that
precedes prices slumped from (59). Ross presents fourteen arguments. There is
no room here to present them all, but we will summarize one to give a flavor.
Whereas sentence (61a) is grammatical, sentence (61b) is not:
(61) a. As for myself, I won't be invited.
b. *As for herself, she won't be invited.

Z1 Ross could have simply used "I tell" here, but for technical reasons concerned with the
formalism, he preferred this abstract set of features.
60 What we do with language: Speech acts IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

It's not clear why this should be. Ross explains it by claiming that myself in
(61a) refers back (as a reflexive pronoun) to the first person pronoun in the
abstract illocutionary clause, while herself is impossible because it has no
antecedent there. Ross was writing in the framework of Generative Semantics, a
version of generative grammar that reached its apogee around 1970 and then
disappeared. Nevertheless, Ross's analysis has remained familiar, and more
recent work in generative grammar, especially in the so-called cartographic
approach of Luigi Rizzi, has worked with a level of grammatical analysis known
as the Force Phrase (cf. Rizzi 2013), which is a syntactic unit corresponding to the
illocutionary force of the sentence.
Functionally oriented work, such as Hengeveld & Mackenzie (2008), has
also recognized illocution in grammar, but rejecting the notion that it
has anything to do with syntax. Their grammar situates the illocution at the
'interpersonal level', which handles the impact of pragmatics on linguistic form,
making it a core element of every 'discourse act'. They observe (2008: 81-82)
that certain uses of adverbials relate to the illocution rather than to the
proposition, as in (62) and (63):

(62) Sincerely, this is not a trick.

(63) Please tidy your sister's room, although why am I asking you?
In both these examples, the material in bold modifies the speech act, not the
propositional content. (62), for instance, is eq,uivalent to "I am telling you
sincerely that this is not a trick". The adverb hereby mentioned in 2.3 above is
also a modifier of this kind.
In 2.5, we saw how the English language has a number of moods and clause
types that indicate different illocutionary forces in direct speech acts. Working
in the framework of Functional Discourse Grammar, Hengeveld et al. (2007)
studied the range of clause types available in a middle-sized sample of indigenous
languages of Brazil and found a great deal of variety. Nevertheless, they
succeeded in reducing that variety to an implicational hierarchy, i.e. they showed
that certain clause types are impossible if others are absent from a language's
repertoire. This is a welcome contribution to what is still an open q,uestion in the
linguistics of speech acts: what illocutions are distinguished in the grammars of
the world's languages? Among the riches already found by descriptive linguists
are laudatives (used for praising), prohibitives (used for forbidding) and
imprecatives (used for cursing). Declaratives, interrogatives and imperatives
appear, q,uite understandably, to be very widespread distinctions and may even
be universal.
The remarks in the previous paragraph concern grammatical distinctions.
There are also massive cultural distinctions in the use that is made of speech acts.
 IIIIIIIII IIIIIIIIIIIIIIIIIIIIIII What we do with language : Speech ads 61

In Muslim cultures, although there are differences here between Shia and Sunni
Islam, for a man to announce "I divorce thee" three times (the triple t,aliiq,;
ffinchcliffe 2000) suffices to bring about the end of his marriage (women do not
ve this right, though there is a distinct procedure for them); clearly this has
no eq_uivalent in other cultures in which divorce is not a performative verb.
Huang, in a useful brief survey (2006: 8600), reports on cultures that lack
promising or thanking as possible illocutions. He also points out that the same
-tuation may trigger different speech acts in different cultures: in English-
speaking cultures, one will apologize after stepping on someone's toes; in the
t-African Akan culture, which does recognize apologies, such a situation will
lead to an expression of sympathy rather than an assumption of guilt. In Japan,
writes, one does not say thanks for a meal but apologizes for intruding on the
dinner party; similarly , one responds to a compliment not with thanks but with
self-denigration (e.g. "No, I really don't know how to cook well"). These matters
ere studied in the 1980s in the Cross-Cultural Speech Act Realization Patterns
Project (Blum-Kulka et al. 1989), which revealed huge differences in the degree
to which indirect speech acts are used, especially with req_uests. Argentinian
anish speakers were found to be the most direc~, and Australian English
51>Cakers the most indirect!
The impact of speech act theory has gone far beyond core linguistics. In
applied linguistics, there is no doubt that work on indirect speech acts weakened
the traditional assumption of a correlation between form and meaning and thus
]ped to inspire the communicative or functional-notional approach to language
teaching that arose in the 1970s. 24 This approach stimulates the use of authentic
Janguage in identifiable contexts of communication with particular functions in
human interaction and with specific notions in play. In psycholinguistics, efforts
ere made to determine whether indirect speech acts, as their name suggests,
take longer to process than direct ones, notably in young children. They
discovered ( e.g. Bucciarelli et al. 2003) that conventional indirect speech acts
· e Can you open the door?) are not processed more slowly than direct ones
but that unconventional ones are. If a father declines to read to a child by
saying (64):
(64) I'm trying to watch this program on TV.

the child will take longer to understand that its req_uest has been refused than if
he had been direct, as in (65), or conventionally indirect, as in (66):
(6.5) No, not now.
(66) Sorry.

See, for instance, Widdowson (1978), Wilkins (1976, 1978) or Canale & Swain (1980).
62 What we do with lan9ua9e : Speech ads I I I I I I I I I/ I I I I I I I I I/ I I I I I I I I I I I I I I

It seems that it is more important for the child to grasp the overall context
than to work out mechanically that the direct meaning does not make sense. This
offered experimental confirmation of Morgan (1978)'s notion of a "short-circuited
implicature" (see 1.6.1 for more examples of this process and 3.2 for details of
implicatures). Something very similar applies to adults' understanding of irony:
if they know from past experience that the 'mood' of the ongoing conversation is
sarcastic, it will not take longer for them to understand an indirect ironic
meaning. And even if the mood is not sarcastic, the freq,uent use of a particular
kind of irony within a discourse community will normally lead to short-circuiting:
the ironic intention will be understood immediately and effortlessly.
Working out what is meant is also of crucial importance for computational
linguistics and artificial intelligence, where machines are being trained to
understand language used by humans. When a customer asks (67):
(67) Do you know when the next train to Canterbury leaves?
she is not inq,uiring whether the computer has this information in its database;
in fact she assumes it does. Rather, the indirect meaning is 'Tell me when the
train leaves'. Work on this problem started in the 1970s (Perrault, Cohen & Allen
1980), first using mechanisms rather similar to those suggested by Searle's
work, with the computer program assessing felicity conditions until the most
plausible interpretation was reached. In more recent work, highly conventional
indirect formulations can be stored with their indirect meanings, much as with
the language processing of conventional indirection by children mentioned
above.
Turning finally to discourse analysis, we find that speech act theory has
been integrated, but with changes. In particular, attempts have been made to
break away from Searle's 'armchair philosophy' by confronting his claims with
empirical data taken from transcriptions of conversational interactions. A
pioneer here was Geis (1995), who observes that the kind of example used by
Searle is vanishingly rare in actual conversation, which is more fragmentary,
more dynamic and more interactive than the standard examples suggest. He also
points out that several indirect meanings may be in play at any one time,
distinguishing between two types of perlocutionary effect, namely 'transactional
effects' ( concerning the ostensible goal of the interaction) and 'interactional
effects' ( concerning more personal goals). If a girl, having been asked out on a
date, answers that she is planning to wash her hair that evening, the transactional
effect on the hearer is 'She disappointed me' while the interactional effect may be
'She is giving me hope': her answer may show that she's not the kind of girl who
accepts every invitation but ( encouragingly) that she might accept another. The
girl had countless options for achieving the transactional effect; which option she
chooses has much more to do with the desired interactional effects. Geis shows
 I I I I I I I I I I I I II I I I I I I I I/ I I I I I I I I What we do with language: Speech ads 63

t these interactional effects are decisive in understanding the course of

versations.
The preceding paragraphs have given a brief impression of just some of the
gs in which Austin's and Searle's contributions have been followed up and
plified. We have noted some limitations of classical speech act theory: it
m ncentrates too much on IFIDs rather than on the speech act in its entirety; it
·es too strongly on native-speaker intuitions; and it has no developed theory of
text. Nevertheless, these are all issues that have been addressed in pragmatics
and they do nothing to undermine the central place that speech act theory
occupies in the discipline.

2.8. CONCLUSION
have seen in this chapter how the concept of a speech act first arose and
blossomed in 'ordinary language philosophy'. Above all, thanks to Searle's
sgstematization of speech act theory, that concept was to have a profound impact
on how linguists now understand language in context, with expressions such as
fflocutionary force', 'implicit performatives' and the like having become part of
their stock in trade. The greatest effect has been on pragmatics itself, which has
gained a set of tools with which to tackle the fundamental problem of indirect
speech acts, where the form of words differs from what would have been a
direct expression of the speaker's communicative intention. The result has been
to open up everyday language use, where indirect speech acts abound, to
scholarly investigation and to offer an alternative to the traditional study of
decontextualized sentences.
The pragmatic take on language is, then, all about the things speakers do,
their activities. The following chapter will focus on three of the most important
of these, referring, saying and implicating. We will consider the work of various
scholars, giving pride of place to another philosopher of language, H. Paul Grice.

SUMMARY
1. The insight that the use of language is a matter of not only saying something
but also doing something, which arose in the mid-twentieth century in
reaction to logical positivism, is the essential idea underlying modern
pragmatics. The work of the Oxford philosopher John Austin on speech acts,
64 What we do with language : Speech ads //// //////// ///////// // /////// ////

as he called them, triggered an explosion in the study of language in use.

Language was no longer seen as largely flawed and 'illogical' as it had been
by the logical positivists.
2. Austin initially introduced the term 'performative' to describe those utterances
that are explicitly used to perform an action, calling all other utterances
'constatives'. However, he came to see that all utterances, including
constatives, have a performative aspect. Later work by Thomas (1995) argued
for a subdivision of performatives into metalinguistic, ritual, collaborative
and group performatives; she also showed that one and the same example
can belong to several of these classes.
3. Austin proposed that speech acts are subject to 'felicity conditions', so called
because they specify when an utterance is felicitous in the sense of
'appropriate for a particular occasion'. Where these fail to apply, the result is
a 'misinvocation', 'misexecution' or an 'abuse'. He also introduced a distinction
that continues to resonate in pragmatics, that between implicit and explicit
performatives: I'll pay you back the $100 you lent me, for example, would
count as an implicit promise, while an explicit one.would need the words I
promise to have been uttered.

4. A further insight of Austin's is that each utterance involves the simultaneous

performance of three acts. The production of a grammatically and
phonetically correct utterance is a locutionary act; the act of conveying a
communicative intention (e.g. to promise) is an illocutionary act, which is
then said to have the force of (in this case) a promise; and the act that has to
do with the effect caused by the illocution on the thought and action of the
hearer is a perlocutionary act. A promise may be intended to reassure the
hearer, and this is the intended perlocutionary effect.
5. English has a number of grammatical moods and clause types (imperative,
subjunctive, indicative; declarative, interrogative, exclamative) that are
associated with illocutionary distinctions: commands are usually expressed
with the imperative mood, for instance. However, in conversation the
correlation between form ( = grammar) and function ( = illocutionary force) is
often q,uite indirect. The context plays an important role in helping the hearer
to work out if the form of words corresponds directly or indirectly to the
speaker's intention.
6. John Searle, also a philosopher, is responsible for systematizing speech act
theory. Without changing Austin's fundamental perceptions, Searle made
some terminological reforms and introduced a few new concepts, such as the
Illocutionary Force Indicating Device ( or IFID) to refer to explicit
performative verbs and their subject ( e.g. I promise). He proposed a five-way
 IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII What we do with lan9ua9e : Speech ads 65

classification of speech acts, focusing on the performative verbs they can

contain. Other authors (so-called 'lumpers') have since proposed fewer classes,
while others (so-called 'splitters') have suggested many more subdivisions.
Searle picked up on the idea that we are very often not explicit about the
illocutionary force of our utterances, proposing the notion of an 'indirect
speech act' and laying out the logical steps by which the addressee of an
indirect speech act figures out the speaker's intended meaning. Indirectness
became established as one of the core concepts of pragmatics.
More recent work has built on Searle's proposals (and through him Austin's)
in many related areas of linguistics. In theoretical linguistics, certain branches
of generative grammar and most branches of functional grammar have
embraced illocutionary force as having a significant effect on the form of
languages. There has also been work in linguistic anthropology, in
psycholinguistics, in computational linguistics, and in discourse analysis that
both owes a profound debt to speech act theory and has contributed to
refining it further.

SELF-EVALUATION QUESTIONS
Choose the most appropriate answer for each q,uestion.
1. An example of a speech (illocutionary) act would be ...
a) reducing the q,uality of a vowel sound.
b) declaring a couple husband and wife.
c) constructing a compound clause.
2. According to the logical positivists, the only meaningful sentences are those
that ...
a) are true.
b) do things with words.
c) can be assessed as true or false.
3. The sentence I burst into tears would be considered as ...
a) meaningful by Austin.
b) meaningful by the logical positivists.
c) meaningless by Austin.
66 What we do with lan9ua9e: Speech acts I I I I I I I I I I I I I I I I I I I I I I I/ I I I I I I I I I I

4. According to Austin's initial formulation (1962), the utterance I promise

that I will be obedient next time is an example of a ...
a) constative.
b) performative.
c) both a and b.
5. What type of performative does the utterance I name this ship The Titanic
prototypically exemplify?
a) Ritual
b) Metalinguistic
c) Collaborative
6. If an authority named a ship by using the wrong name, what type of infelicity
would that be?
a) A misinvocation
b) An abuse
c) A misexecution
7. The utterance Get out of my house! is an example of ...
a) an explicit performative.
b) a non-explicit performative.
c) an explicit constative.
8. Austin developed his argument in such a way that he finally concluded
that ...
a) there is a substantial distinction between performatives and constatives.
b) both performatives and constatives blend into the category of illocutionary
acts.
c) implicit performatives are not illocutionary acts.
9. What would be a very possible perlocutionary act produced by the following
utterance (said by a woman to her boyfriend, in.the middle of an argument)?
You're the stupidest person I've ever met!
a) Insulting
b) Humiliating
c) Uttering a sentence
IIIIII IIIII III IIIIIIIIIIIIIIIIIIII What we do with language : Speech ads 67

10. Regarding the relationship between form and function, it can be said that ...
a) there is a one-to-one relationship between the grammatical form of
utterances and the illocutionary acts they perform.
b) each mood and clause type expresses only one, distinctive illocutionary
act.
c) the same illocution may.be expressed by different forms and vice-versa.

11. What mood and clause type is the illocutionary act of expressing emotion
normally associated with?
a) Indicative exclamative
b) Imperative
c) Indicative interrogative

12. According to Searle's classification of speech acts, the utterance I swear that
I will not cheat on you ever again (said, for instance, by a man to his wife)
would be an example of a ...
a) commissive.
b) directive.
c) expressive.

13. The direction of fit of the utterance in 12 (J swear that I will not cheat on you
ever again) is ...
a) world to words.
b) words to world.
c) both a and b.

14. In the exchange between A and B below, the primary, indirect act contained
in B's words (where B is a very responsible person who puts duty ahead of
pleasure) is ...
A: How about going for a walk?
B: I have to finish mowing the lawn

a) stating what B's duties are at the moment of talk.

b) rejecting A's invitation.
c) offending A.
68 What we do with language: Speech ads / / / / / / / / / / / / / / / / / / / / // / / / / / / / / / / / /

15. Hengeveld & Mackenzie's (2008) Functional Discourse Grammar situates

illocutions at ...
a) the morphosyntactic level.
b) the representational level.
c) the interpersonal level.

DEVELOPING OUR PRAGMATIC COMPETENCE:

EXERCISES FOR REFLECTION AND AMPLIFICATION
A. Given the following utterances, and taking into account Austin's initial
distinction between performatives and constatives, identify which of
them belong in a more prototypical way to one category or the other.
Justify your answers.
a) I sing.
b) I certify that the above is an exact and accurate translation of the original.
c) I swear to God that I didn't do it!
d) I am sewing this new dress.
e) I'm sorry for your loss.
f) I apologize for having offended you.
g) I name this ship "The Wanderer".
h) I can swim 2 km in 25 minutes.
i) I bet you €100 that she's not coming.

B. Classify the utterances you identified as performatives in A) according

to Thomas's (1995) taxonomy of performatives.

C. What kind of infelicities may situations a) to d) incur?

a) A friend says that she will invite you to dinner at her house, which you
accept gladly , but then she never calls you to confirm the invitation and
give you an exact day and time for you to go to her place.
b) A priest goes through all the rituals to baptize a baby, but on that day he is
baptizing three babies in a row, and when it is time to name the first he
uses the name of the second baby instead.
c) Someone who is not a Notary Public (pretends that s/he) certifies the
validity of a diploma.
d) Your friend promises you that she will lend you €500 but then she never
gives you the money.
 II I I I I I I I I I I I I I I I I I I I I I I I I I I I I I What we do with language : Speech ads 69

D. Say whether the following utterances are examples of explicit or

implicit performatives, and in the case of implicit ones, specify the
device used to indicate the intention of the speaker.
a) I order you to stop smoking.
b) Would you be so kind as to lend me your Ferrari?
c) Stop talking!
d) Tomorrow I will take you to the theatre.
e) I promise to buy you a new bicycle for your birthday.
f) I warn you that the food in Mexico is very spicy and hot.

Applying Searle's classification of speech acts, identify the main type

and subtype of act performed in each of the following images, as
exemplified in (a):
a) Main type: Representative II Subtype: Assertion

( YESWECAN!

b)

I swear to uphold the

Constitution of the
United States against
all enemies, foreign
and domestic.
70 What we do with language : Speech acts IIIIIIIIIIIIIIIIIIIII IIIIII IIIIIII

c) d)
1
I WA~Nl~G 1
HIGH
.VOLTAGE ·'T I

I now
e)
pronounce
you husband
and wife.

g)
I HAVE A
DREAM!
 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I/ I What we do with language : Speech ads 71

. Consider again the acts in E) above and identify the locution, the
illocution and one possible perlocution in each case (from a) tog)).

G. Indicate the mood of the structures in bold and specify their function.
Do form and function coincide in all of them? Show the results of your
analysis in the table below.
a) And this will we do, if God permit. (Hebrews 6:3)
b) (At the dinner table): Can you pass the pepper?
c) A teacher to his student: I believe in you.
d) The British people: Long live the Queen!
e) Please fasten your seatbelts. We're flying through an area of turbulence.
f) A mother to her child: You need to take a shower.
g) A woman to her chubby, lazy friend: You have run the mile in 3 minutes
50???
h) A girl to the boy who is in love with her: i) Do you want me to accompany
you?
Boy: ii) Do I want you to accompany me!!
i) A man to a friend who is being disrespectful of him: Raise your voice
again and I'm out of here!

a)
b)
c)
d)
e)
f)
g)
h)
i)

H. Now return to the utterances analyzed in G above and say whether

they are examples of direct or indirect speech acts.
72 What we do with language: Speech ads / /// / // / / / / / ///// / / // / / / // / /// /// /

I) Taking into account functionally oriented work such as that of

Hengeveld & Mackenzie (2008), indicate the function of the adverbials
in italics in a) and b ):
a) Seriously, I don't want to carry on with this relationship anymore.
b) I might not be able to be here tomorrow, although why am I giving any
you explanations at all?

J) Examine all the speech acts in the images below (two each) and classify
them according to their type (Representative, Commissive, etc.). Also,
specify if they are performatives or constatives, direct or indirect.

a)
I hereby declare
three days of official
mourning in Brazil
after losing the World
Cup Semifinal.

b)

K) Pick a cartoon or any text that interests you and analyze the speech
acts in it. Specify the mood of the structures used and their function,
and therefore whether the acts performed are direct or indirect. Also,
classify all the acts according to Searle's taxonomy of speech
(illocutionary) acts.
 ON REFERRING, SAYING,
AND IMPLICATING:
REFERENCE AND INFERENCE
/ /////////////////////////////////////
AIN OBJECTIVE OF THE CHAPTER: To define and explain the
pragmatic concepts of reference and inference, identifying their different
types and focusing on Grice's theory of implicature.

MAIN TOPICS OF THE CHAPTER:

• Reference
• Deixis
• The relation between deixis and grammar
• The different types of inference
• Presuppositions and their impact on grammar
• Conversational and conventional implicatures in the work of H.P. Grice
• Neo-Gricean proposals for general principles of communication
74 On referring, saying , and implicating: Reference and inference // / / // / /// // ///

3.1. REFERENCE
As we build up a discourse, we mention and often re-mention various people and
things. The pragmatic mechanism that allows us to introduce these entities into
discourse and to keep track of them is called reference. This resource makes
crucial use of the speaker's and the hearer's shared knowledge, and this justifies
seeing reference as a pragmatic phenomenon. In addition, reference is often
achieved through the use of proper names, and these have a pragmatically
determined conventional association with particular people, places and objects
known to a speech community.

3.1.1. WHY IS REFERENCE AN ESSENTIALLY PRAGMATIC

MECHANISM?
We all know - at least intuitively or unconsciously - that context has a powerful
impact on how referring expressions are to be interpreted. This entails that the
semantic meaning or denotation of a word or expression may or may not coincide
with the meaning intended by the speaker or interpreted by the addressee(s). By
denotation linguists normally designate the relation between words and classes
of elements in the world. This alludes to the conceptual prototype or 'dictionary
meaning' of a word or expression. But reference is not the same thing as
denotation. Reference is an action carried out by speakers and therefore alludes
to the relationship created by a speaker between words and specific entities. By
way of illustration, consider utterances (1) and (2), both said by a hotel
receptionist to a workmate:
(1) Room 25 has to be cleaned by 12:00 noon.
(2) Room 25 has complained about the noise at night.
The denotation or dictionary meaning of the word room would be something
like "a portion of space within a building or other structure, separated by walls
from other parts". In both (1) and (2) the reference of the word room is different
from its denotation, for in (1) the speaker does not refer to the prototype, but to
a room (#25) in a particular hotel (not a dining-room or any other type of room
in a house, for instance), and in (2) the speaker does not refer to a room at all. It
is clear, though, that in (1) the meaning intended by the use of the word room is
closer to its denotation than in (2), where this same word is used to make
reference to a person or persons, more specifically, to the guest who happens to
be occupying Room 25 at the moment of talk. (2) and (4) below are examples of
what Huang (2016) calls deferred reference, defined by Ward (2004: 262) as the
"use of an expression to refer to an entity related to, but not denoted by,
I/ /// I I//////// On referring , saying , and implicating : Reference and inference 75

the conventional meaning of that expression".25 This shows that it is very difficult
o understand any message without the pragmatic information given by the
context. And by context we refer here both to the linguistic context (or co-text)
and the extra-linguistic context, taking into account its different types, as shown
in Chapter 1 (1.2.). If we consider (1) and (2) again, we shall see that the physical
context of talk is the same for both (a hotel reception desk), but in this case what
helps us identify the different referents (an actual room or a person) is the
co-text: it is evident that rooms cannot literally "complain", which therefore leads
us to infer that by using the word room the speaker in (2) has to be referring to
the guest in Room 25. A similar process of inference is necessary to distinguish
the referent of H uxley in (4) from that in (3), where the context of talk is in both
cases an interaction between two students of literature studying together for an
exam, but the co-text helps the hearer identify the Huxley in (3) as a man (Aldous
Huxley) and that in (4) as a book:
(3) Huxley is the author of Brave New World.
(4) Can you pass me your Huxley?

In some cases, such as (5) below, the exact referent may be difficult to
determine, for the expression used (Mar Adentro) may allude to a number of
possible referents in the cognitive and emotional world of the audience, namely
the movie alone, its producers, its director (Alejandro Amenabar), or the country
where it was filmed and produced (Spain), and all of these are valid within the
positively charged emotional context surrounding the glory of getting such an
important award:
(5) (Context: 2005 Academy awards)
Presenter: And the winner of the Oscar award for best foreign language film
is ... MAR, ADENTRO!!!!

Referential expressions are normally noun phrases (as shown in (1) to (5)
above), but verbs, whether nominalized or not, can be also used as referents of
particular actions or activities, such as driv ing and to watch TV in (6) and (7)
respectively:
(6) Driving is not my favorite thing.
(7) She doesn't like to watch TV.

In the case of verbs, just as in that of nouns, the identification of the person,
thing or activity cannot be properly made if there is not a conventional agreement

25 In fact, most cases of deferred reference display instances of metonymy, a figure of

speech in which a person, thing or concept is called not by its own name but by the
name of something associated in meaning with that person, thing or concept.
76 On referring, saying, and implicating: Reference and inference //// // / / / ///// /

within the given community as to what the referent of that person, thing or
activity is. For instance, an extra-terrestrial being coming from a different planet
where there are no cars, even in the hypothetical case that s/he knew how to
speak English, would not be able to assign a referent to the word driving, in the
same way as a terrestrial four-year-old would most likely not be able to assign a
referent to Huxley in (4), because s/he does not belong in the context of educated
adults or literature students and the refore lacks the background and necessary
pragmatic competence to do this.
In sum, the main point of this introductory section has been to show that the
process of referring is a pragmatic one, for the referents introduced and used in
any type of discourse could not be identified without taking into account both the
speaker's and the hearer's intentions, actions and knowledge.

3.1.2. TYPES OF REFERENCE

There are different types of reference depending on different variables, such as
the definiteness or indefiniteness of the expressions used, the type of thing or
person being referred to, or the direction in which it refers within the text
(backwards or forwards) or out of the text.
Speakers normally introduce referents into discourse by using terms which
are indefinite and explicit ( e.g. A dog that was astray in the street), and later in
the discourse refer to them by means of definite and inexplicit expressions (e.g.
the dogl(s)helit). Definite reference is realized by expressions containing definite
determiners (e.g. the cat), certain pronouns (e.g. they), noun phrases and proper
names (e.g. motorbikes, Kenneth), locative adverbials such as here or in the box,
time adverbials such as yesterday or the day after tomorrow, etc. Indefinite
reference is found in expressions with indefinite determiners ( e.g. a man, many
people), certain pronouns (e.g. anyone), such locative adverbials as anywhere,
and time adverbials such as some time or any time. One and the same referent
may be evoked by using either a definite or an indefinite expression, either
explicit (e.g. the guy who was looking at you) or inexplicit (e.g. he), depending
on the shared background knowledge, the relationship between the interlocutors,
and many other variables affecting the speech situation. For instance, we could
refer to our friend and colleague Elena in many ways, such as those in (8),
depending on the context of talk and on who our interlocutors are:
(8) a. Elena
b. Dr. Martinez Caro
c. Professor Elena Martinez Caro
d. a friend of ours
 I / / I I I I I I/ I I/ On referring, saying , and implicating : Reference and inference 77

e. a colleague of ours
f. our common Spanish friend
g. the friend that introduced us to each other

In this case we are referring to a particular individual, so all these referring

ressions are instances of specific reference. There are cases, however, in
·eh we do not refer to anything or anyone in particular, but to an object or
on as pertaining to a group or class, as shown by the noun phrases in bold
- (9). These are instances of what we call generic reference:

a. A washing machine is a very useful appliance.

b. Computers can save your life.
c. Men are very primitive creatures.
d. The Italians are all very passionate.

Notice that both specific and generic referring expressions can subsume
e-itnu·te or indefinite reference: (8 a, b c, f, g) and (9 b, c, d) illustrate the former,
· e (8 d, e) and (9 a) illustrate the latter.
Martin & Rose (2003: 169) explain that "whenever the identity of a participant
- presumed, that identity has to be recovered", and this recovery can be achieved
- various ways, depending on where the relevant information is. Sometimes the
- onnation is found by looking backward in the text or discourse, as in (10). This
- what we call anaphoric reference:

0) That's my friend Suzanne. She is an actress.

There is also another kind of anaphoric reference which is indirect and is

called bridging reference. It refers backwards indirectly, as can be seen in (11),
re the food refers indirectly backwards to the restaurant:

11) We went to a new Italian restaurant yesterday and the food was delicious.

Some other times we may have to look forward in the text in order to recover
participant or thing being referred to, as in (12), in which case we have an
instance of cataphoric reference:

12) Immediately after she saw him, Camilla turned round and left the room.

In some cases, we may find the referent within the same noun phrase, as in
13), where the speaker makes clear that the house she is referring to is the one
78 On referring, saying , and implicating : Reference and inference ///////////////

that is "in the woods", thereby preventing the hearer from having to track the
referent somewhere else in the text or context:
(13) The house in the woods was her hiding place.

This type of reference is called esphora, and it allows us to identify

participants without having to look elsewhere in the discourse. Martin & Rose
(2003: 172) explain that this happens when one element modifies another one and
it answers the q,uestion "Which one?" (in (13): Which house?).
Speakers may also make reference to people and objects outside the text. If
the reference looks out to shared or cultural knowledge, we speak of homophora,
a kind of reference that is normally realized through names or definite noun
phrases whose reference is obvious within the speech community, such as The
Pope, God, Shakespeare, the Senate, etc., as in (14):
(14) In his lifetime, Michelangelo was considered the greatest living artist.

In some cases, the reference looks out of the text but not to famous people
or obvious, shared cultural knowledge, as in (15), where the red motorbike refers
to a real motorbike that is outside the text and can be perceived with the senses.
This is a case of so-called exophoric reference:
(15) Look! The red motorbike has crashed into a tree!!

Exophora normally refers to something outside the text that can be heard,
seen, touched, tasted or felt by the interlocutors. In (15) the red motorbike is
introduced for the first time, and the refore it is not mentioned anywhere else in
the text, but the interlocutors can see the referent in their physical environment.
There is still another kind of reference (not included by Martin & Rose in
their taxonomy) that looks out of the text but that is neither homophoric (because
the referents involved are not obvious or do not form part of shared knowledge)
nor exophoric (because the referent cannot be seen, touched, etc. in the physical
environment of the interlocutors). This is the case of, for instance, the first time
someone or something is introduced in a narration, as in (16) and (17), where
Mrs. Dalloway and Dick Gibson are presented by the authors in the opening line
of their respective novels. We shall call this type of reference ideophoric, because
it creates the first idea of a referent that is not in the immediate physical
environment but in the interlocutors' mind.
(16) Mrs. Dalloway said she would buy the flowers herself. - Virginia Woolf,
Mrs. Dalloway (1925)
(17) When Dick Gibson was a little boy he was not Dick Gibson. - Stanley
Elkin, The Dick Gibson Show (1971)
 II /////////// On referring, saying, and implicating: Reference and inference 79

It is worthy of mention that, in the case of both (16) and (17), once the
_ _ ,..,,cters in q,uestion have been introduced, the successive reference made to
becomes homophoric; for they are now part of the authors' and readers'
ed knowledge. This illustrates the fact that the phenomenon of reference is
a static one, and that the way of referring to the same participant or entity
y change according to the conditions and circumstances of the ongoing

Table 1 summarizes the seven types of reference that help us track the
·cipants and entities in discourse:

Backward in the text I love water-skiing. It's fun!

He sat the exam last week. The

BRIDGING Indirectly backward
CATAPHORIC Forward in the text
ESPHORIC Forward within the same NP
Out of the text, to shared
HOMOPHORIC
knowledge
Out of the text, to the
EXOPHORIC situation or immediate
physical environment of talk
Out of the text, to some new
imaginary person or idea
IDEOPHORIC
introduced by the speaker/
writer
mark was higher than he had
expected.
Much to her surprise, Lily
received an anonymous bunch
of flowers on St. Valentine's Day
The kid you met yesterday is
looking for you.
Pope Francis marked a day
of prayer against human
trafficking.
That noise is killing me!
Miss Brooke had that kind
of beauty which seems to be
thrown into relief by poor dress.
- George Eliot, opening line
of Middlemarch (1872)

As noted in Alba-Juez (2009: 60), when reference is viewed as a discourse

ocess, referring seq_uences (not just isolated referring terms or expressions)
may become crucial for the analysis. Thus, when analyzing a text such as a
ronversation, a letter, an e-mail message or a recipe, it is important to look at
80 On referring, saying , and implicating: Reference and inference / / /// / // / //// / /

how the participants and entities are introduced and later tracked throughout the
discourse, because the development of the referential process may reveal
important aspects of the general meaning of the particular discursive event.
One way of tracking participants or things in discourse is by means of the
phenomenon of deixis, a type of reference to which we now turn.

3.2. DEIXIS: INDEXICALS AND CONTEXT

The word deixis comes from Ancient Greek, where it means "pointing to" or
"picking out". Some words or expressions (such as here, the day before yesterday,
there, now, this, that, come or go) req_uire that we pick out a person, place, time
or situation to determine how the expression refers; i.e. these expressions allow
us to identify or "point to" referents that are particular to the context, and for that
reason they are called deictics or also indexicals. The use of indexicals in
discourse is thus one of the processes that clearly show the relationship between
language and context, because utterances like (18) or (19). cannot be fully
understood or interpreted if there is no further indication of who the speaker and
addressee(s) are, or when and where they were uttered.
(18) I'll phone you next week.
(19) You and you, come here now!

Now consider (20) and (21), where, as is the case with the deictic expressions
in (18) and (19), YOU, ME, I, you, and NOW are indexicals or deictic expressions
because they are indexes that point to some person or aspect in the context of
talk.
(20) Little Tommy to his friend: This ice-cream is not for YOU. It's for ME.
(21) Susan to her daughter. I cannot pay attention to you NOW. I gotta work.

Another characteristic of deictic expressions is that their referents shift

depending on the contextual conditions. Thus YOU, ME, I, or NOW do not always
have the same referent. In (20) YOU refers to Tommy's friend and ME to little
Tommy, but this YOU is not the same one as the you in (21), which refers to
Susan's daughter, and the ME in (20), little Tommy, is not the same person as the
Jin (21), which refers to Susan. The same happens with the word NOW in (21),
where it refers to the particular moment of Susan's utterance, but if the same
sentence were uttered by a different - or even the same - person before or after
that moment, then the word NOW would refer to a different moment from that in
(21). In fact, each time these words are used their referent will shift, and for this
reason they have also been called shifters.
 /II I I I I I I I I I I I On referring , saying, and implicating : Reference and inference 81

The role of context is crucial in helping to determine reference, for this

determination very much depends on who the speaker is, where she is located,
hen she is speaking or writing, and so on. The specific context where the
speaker is located is what has been called the deictic center, for in this particular
context the "I" will refer to the speaker, the "now" to the moment when the
speaker is talking, the "here" will refer to the place where the speaker is located,
etc. Therefore, all the referring expressions in a particular discourse situation
will be interpreted in terms of the speaker's deictic center, represented by the
oval in Figure 1.

f igure I: Deictic center

Last (year, etc.)

Past~~ Then (Future)

Tomorrow, later, etc.

You /he/ she/ it/ they
,,....,,,.,~...,..,..,..........,.. There
That
Left right

come/bring

Figure 1 serves to illustrate the fact that deixis is a feature of language

which is generally organized in an egocentric manner in the sense that the
speaker is the center of it all. Another interesting thing to notice about the deictic
center (also called origo by Buhler 1934) is that sometimes the criterion for
choosing one deictic expression or another may be psychological rather than
physical. A good example in English is given by some uses of the verb to come,
as seen in Jack's reply in (22):

(22) Mother. Jack! Dinner's ready!

Jack: Corning!

Strictly speaking, and taking into account Jack's physical deictic center, the
response to his mother's call should be "Going!", because Jack will move from
where he is physically located (his deictic center) to the dining room. However,
82 On referring, saying, and implicating: Reference and inference / / / / / / // / / / / / / /

more often than not in a situation like this English speakers will use the verb to
come and not to go. A possible explanation for this is that in (22) the speaker
visualizes himself as being where the mother and the food are and imagines he is
watching himself moving towards the place where he has been psychologically
self-located. A similar phenomenon occurs when in narratives the deictic
expressions are transposed or relativized (to use Fillmore's 1997a term) to the
deictic center of the protagonist, so that the narrator or writer removes himself
from the deictic center, as in (23), the opening line of the novelMoby Dick (1851),
where me refers to one of the protagonists (Ishmael) and not to Herman Melville
(the author of the novel):
(23) Call me Ishmael.

As the reader will have noticed by looking at Figure 1 and examples (18) to
(23), deixis is a complex phenomenon that has to do with many aspects of the
context, such as speakers and hearers, time, place, direction of movement, etc.
We shall examine these in more detail in 3.2.1.

3.2.1. TYPES OF DEIXIS

As Levinson (2007: 97) points out, deixis "is a much more pervasive feature of
language than normally recognized" and there is still a lot to study and understand
about it. The phenomenon of deixis also shows the complex relationship between
semantics, cognition and intention, i.e. between what the words mean, the
thoughts that they express, and what human beings mean when saying these
words. In this section we are going to examine the main types of deixis and
indexical expressions, but it is important to remark from the very beginning that
practically every referring expression can be used deictically. This is the reason
why there is neither a clearly delimited list nor a complete or adeq,uate cross-
linguistic typology of deictic expressions.
Most analysts agree that there are three main types of deixis: 1) Person, 2)
Time, and 3) Spatial or Place Deixis. To these three basic types two more have
been added by some authors (e.g. Levinson 2007), namely 4) Discourse and 5)
Social deixis.
1) Person deixis is reflected in the grammatical category of person. Thus
personal pronouns are the prototypical indexical expressions for this type.
The first person normally points to the speaker and the second person to the
addressee, while the third person points to a third party that may or may not
form part of the particular discourse situation. These roles shift according to
conditions such as conversational turn-taking, and therefore the origo or
deictic center shifts along with them.
/ II I I I I I I I I I I I I On referring , saying , and implicating : Reference and inference 83

2) Time deixis is prototypically encoded in the grammatical category of tense,

but it is also found in adverbials such as now, then, today, yesterday, ten
years ago, next year, etc. As noted in 3.2, the temporal deictic center regarding
time deixis is the particular moment at which the utterance is issued.
3) Spatial or Place deixis is prototypically encoded in adverbials such as here,
there, in this place or in that room, which point to places related to the
context of talk. This and here are examples of prox imal (i.e. close to the
speaker) place deixis, while that and there constitute instances of distal (i.e.
non-proximal to speaker) place deixis. There are languages, such as Spanish,
that have a third sub-type, namely medial place deixis, encoded in the
demonstrative ese, the proximal and distal ones being este and aq,uel,
respectively. In both English and Spanish, verbs such as come and go (Sp.
venir and ir) or bring and take (Sp. traer and Uevar) encode spatial motion to
or from the deictic center and thus are prototypical examples of place-
indexical expressions.
) Discourse deix is is observed when certain expressions (e.g. this (situation),
as pointed out before, in the previous chapter, in the nex t section, etc.) are
used to refer to some portion of the preceding or·f orthcoming discourse. A
distinction should be made here between discourse or textual deixis and
anaphora or cataphora. Discourse deixis refers to portions of the discourse
itself, whereas anaphoric or cataphoric expressions refer to people or entities
outside the text by connection to a prior or later referring expression
(Levinson 2007: 119). Thus, the demonstrative That in (24) is discourse-deictic
(because it refers back to the whole of David's utterance), while the one in
(25) is just an example of anaphora, which is co-referential with the house and
the one:
(24) David: Get ready, Mary. We're off to the Maldives!
Mary: That's what I've been waiting to hear for ages!!
(25) I like the house and that is the one I want to buy.

Discourse markers of the connecting type, such as to summarize, in

conclusion, however, or in addition are also considered discourse-deictic due to
the fact that they relate to a prior utterance or segment of discourse. An example
is found in (26), where however points back to everything the mother had said
before (the two previous sentences).
(26) Mother to child: Both Dad and I are very upset with the grades in your report
card, as well as with your behavior. Besides, all that back-talking sounds
very disrespectful of us. However, and in hope of radical changes in your
conduct, we're going to give you one more opportunity and will allow you to
go to the mountains to ski with your classmates.
84 On referring , saying, and implicating: Reference and inference / //// / / / / //// /

A curious instance of this type of deixis is given by the relative pronoun

which when used as deictic discourse marker, as shown in the following extract
from The Washington Post, where the occurrence of which in bold print poin:
back to the complete previous paragraph:

(27) ... This case is about both - Obama's personal feelings, and how those
feelings would eventually be reflected in policy decisions. But it isn't that he
was secretly plotting all along to create marriage eq,uality for every
American (the Supreme Court is the one that will do that). The policy steps
his administration has taken on gay rights issues - ending "Don't Ask, Dont
Tell," opposing DOMA- were things he said in 2008 he was going to do, and
he ultimately did them. All of which could reasonably lead us to the
conclusion that although it was certainly wrong for Obama to hide his true
feelings on this issue, it may not be among the worst sins presidential
candidates have committed.
Which brings us to today, and the presidential campaign that's now
beginning. Most of the Republican candidates are taking a position that ...
(From: How bad was Obama's 2008 deception on ·same-sex marriage? The
Washington Post, 10 February 2015.)

5) Social deixis is the type of deixis that points to aspects of the social
relationship between interlocutors such as power, distance, social status or
role of the participants in the speech event. This is the case of honorific
expressions found in different languages, such as the so-called T/V
pronouns ( e.g. tulvous; tu!usted; du/Sie in French, Spanish and German,
respectively), which encode social aspects such as respect, different social
class, or different age of the participants. Levinson (1983) distinguishes
four axes on which the relations among participants are defined: 1) Speaker
to referent, encoded in the use of referent honorifics such as titles
(e.g. Dr. Sigmund Freud, Count Bismarck); 2) Speaker-to-addressee
honorifics, an example of which are the T/V pronouns named above;
3) Bystander honorifics, which signal respect to non-addressed but present
participants (for example, Keating 1998 writes about a kind of suppletive
verbs and nouns which are used in the indigenous languages of Pohnpei ( of
the Federal States of Micronesia) in the presence of a chief); and 4) speaker-
to-setting honorifics, which have to do with the levels of formality used
depending on the setting or event. An example of this type of honorific is
found in the distinction made in English between words of Germanic origin
and those of Romance origin ( e.g. house/residence, land/territory, mean!
signify, etc.), the latter being generally used in more formal or technical
contexts.
I I I I I I I!/ I I I I I I On referring , saying , and implicating : Reference and inference 8'5

Deictics are semantically deficient, and therefore many times, with all five
kinds of deixis, gestures are necessary to help us identify the referent, as in (28).
This is what we call gestural deixis:

(28) Teacher (pointing to Daniel and Karla): You and you will come with me to
the Headmaster's Office!!

But sometimes no gesture is req,uired because the knowledge of the basic

parameters of the speech event is sufficient for the hearer to identify the referent.
In these cases we speak of a symbolic usage of deixis, exemplified in (29), where
you obviously refers to Martha's husband because there is no one else in the
room but him, and conseq,uently there's no need for Martha to point to him:

(29) (Martha enters the dining room to find her husband standing on the table)
Martha: What are you doing???

Also, it is interesting to note that deictic expressions can have non-deictic

usages, as exemplified by (30), where you is used in a generic, non-specific way:
(30) With people like these, you never know how to react.

However, sometimes it is difficult to tell whether the use of a pronoun is

deictic or not, as in (31), where the pronoun you could be used in a general way,
t-0 refer to every person, or in a particular way, referring to the addressee:

(31) You're not supposed to eat with your hands!

If we try to relate the types of deixis to the types of reference seen in 3.1.2
above, at first sight it might seem that that deictic reference is always exophoric
and non-deictic reference is endophoric or intratextual, but things are not that
simple. Suffice it to say that, for instance, discourse deixis is endophoric but
deictic, while some uses of demonstratives are recognitional or empathetic
(Levinson 2007: 108), such as those exemplified in (32) and (33), and are non-
deictic:

(32) Do you remember that blond woman we met the other day in the park?
(Recognitional use: that is used to help the interlocutor recognize the person
or thing in q,uestion)
(33) I hate him - that son of a bitch. (Empathetic use: that is charged with some
kind of emotional meaning which has nothing to do with pointing to a distal
person or object)

The following additional examples can further help the reader comprehend
the above distinctions:
86 On referring , saying , and implicating : Reference and inference / / / / // / ///// / //

(34) I want this chocolate. (Exophoric gestural - req,uires pointing to the

chocolate)
(35) This is my favorite park. (Exophoric symbolic - does not req,uire gesture)
(36) A duck sounds like this: Quack, Quack! (discourse deictic)
(37) Her boss allowed her to go on vacation. This man was really kind and
understanding. (Anaphoric use, non-deictic)
(38) She suddenly stood up and slapped that charlatan. (Empathetic use, non-
deictic)
(39) I'm trying to find that book we bought on our last trip to the U.S.A.
(Recognitional use, non-deictic)

In examining all the above examples, the reader may have noticed that the
phenomenon of deixis is closely connected to certain grammatical aspects of
language. This is what we shall try to explain and illustrate in more detail in the
next section.

3.2.2. GRAMMATICALIZED DEICTIC MEANINGS

The forms used to express various types of deixis distinguished in the preceding
section are interconnected in various ways. In English, the same words this and
that are used for spatial deixis (e.g. this cat), for time deixis (e.g. this century),
for discourse deixis (This was what I was trying to say), sometimes for person
and social deixis ( e.g. This author disagrees, a rather formal way of saying "I
disagree" in writing), and for non-deictic uses like the recognitional ( e.g. this
trend of using social mediafor business communication) or the empathetic use
(this lovely girl I met last night). All these various uses are linked within the
system of English grammar but they are also related to one another historically
through the process of grammaticalization. Grammaticalization is a diachronic
process in which constructions or expressions dealing with the concrete world
come to develop meanings that are more abstract (cf. 1.6.1.2).
In the case of deixis, the historical evidence from many languages indicates
that spatial deixis is the most basic of all types. This is because the distinction
between this and that (or between este, ese and aq,uel in Spanish) has to do with
immediate perceptions of the speaker's environment, specifically concerning the
q,uestion whether objects are relatively close to or distant from the deictic center,
with English distinguishing two degrees of distance (proximal and distal) and
Spanish three. Other languages, e.g. Hausa and Tlingit (Huang 2013: 197), make
four distinctions; and yet others have additional refinements, like whether the
object is visible or not. Person deixis in many languages, especially in the third
person, is historically derived from spatial deixis: Spanish el and ella, for example,
 I ////// I I I I I I On referring , saying , and implicating : Reference and inference 87

derive historically from Latin illum and illam, meaning 'that one (masc.)' and
tone (fem.)' respectively; English he can ultimately be traced back to the
oto-Indo-European (PIE) *ki- 26 ('here') and she to *so- ('this, that'). As for
origins of the first person pronoun I, these are lost in the mists of time, but
1K>tice how the deictic here is still associated with the speaker:

(41) might be said in response to someone else saying something about

mselves like "I'm fed up with this". In Japanese, the second person pronoun
,mata 'you' comes from a spatial deictic pronoun meaning 'over there' (Heine &
ng 2011: 605, 607-608). In English, too, we could say (42), meaning 'How are
things with you?', for example when speaking to a friend in another country by
telephone:

42) How are things over there?

Time deixis is also derived from spatial deixis. T~e present time is correlated
with the proximal deictic this and the past time with the distal deictic that; this is
d early because the present time coincides with the deictic center and thus is near,
while the past is remote:
(43) These days far too many young people are out of work.
{44) In those times, the power of religion was much stronger in society.

The deictic adverb then can be used for either past or future time, as in (45)
and (46) respectively. Both the past and the future are remote from the present,
so it is not surprising to discover that historically then comes from the same PIE
root as that:
(45) Then I was still fit enough to stay up all night dancing.
(46) I'm going out for dinner and then I'll go dancing all night.

Discourse deixis can also derive historically from spatial deixis. This is a
good example of how grammaticalization leads to more abstract expressions. In
the following mini-conversations, we see the proximal deictics this and here and
the distal deictics that and there take on discourse roles:
(47) A: We need a new government immediately!
B: This is what I've been telling you for weeks.

26 We have no docurnenta11J evidence of PIE, so all the forms have been reconstructed by
linguists. Reconstructed forms are always prefixed by an asterisk (*).
88 On referring, saying , and implicating: Reference and inference ///////////////

(48) A: The company needs to avoid lay-offs.

B: Here I agree with you completely.
(49) A: We need a new government immediately!
B: That's typical of you - always blaming the government.
(50) A: The company needs to avoid lay-offs.
B: There I disagree - you're being too soft-hearted, as usual.

As you will have noticed, the proximals are associated with agreement and
the distals with disagreement. Other expressions that are used as 'signposts' in
texts are above-mentioned andfollowing; these are also examples of what were
originally spatial deictics being used in discourse deixis.
A freq,uent marker of discourse deixis in English is so, as in (51B):
(51) A: She's failed her exam again.
B: I thought so!

What B means here is that she thought that the event A describes would
happen. So is therefore a discourse deictic in this use. Histopcally, so meant 'in
this/that way', as it still does in German. But gradually it developed away from
that meaning - now usually expressed as like this or like that - and took on
various discourse-deictic roles. One clear case is so in (52), which first looks back
in the discourse to what has preceded and then introduces the conseq,uence: 27
(52) The loopholes in the law are a symptom of a broken tax system. Tinkering
clearly hasn't worked. So we now need to treat the cause. (Adapted from
Daily Mail, 23/7/2012)

Social deixis, finally, is different from the other types in not deriving
historically from spatial or other types of deixis. Rather, honorifics work as a
kind of overlay on the basis person-deictic system. In many of the languages of
Asia, the first and second person pronouns derive from words for 'slave, servant'
and 'lord, master' respectively (Heine & Song 2011). Here we can clearly see the
grammaticalization of extremely deferential language use, with "Your slave
wishes to invite her master for dinner" meaning 'I would like to invite you to
dinner'. We see traces of this in European languages, too: in Dutch the V-form in
the T/V pair is u, derived from 'your nobility'; in Portuguese formal letters the
addressee is often referred to as "Vossa Excelencia" ('your excellence'), and in
Spanish the V-form usted is historically 'your mercy'. There is not much like this
in modern English, except in very obseq,uious usage, as in:
(53) Did Sir enjoy his lunch?

27 This is a prototypical use of so as a discourse marker.

 / /II I I I I I I I I/ On referrin9, sayin9, and implicafin9 : Reference and inference g9

ere Sir is historically 'lord' ( cf. Spanish Sefior) and the third person is used
d'. his lunch) to avoid the possible unwanted intimacy of your.

Spatial deixis has infiltrated other areas of the grammar, too. As mentioned
ve, the verbs come and go are thoroughly deictic. They, too, are easily
;ended to the domain of time deixis (in days gone by, in the coming months).
all wonder, then, that in many languages these verbs have adopted the
mmatical role of indicating aspectual distinctions, i.e. refinements of
temporal meaning of lexical verbs, sometimes developing into fully fledged
e markers. In English the verb go is used to indicate the prospective aspect,
m vering intention, immediacy and high likelihood:
54) I'm going to marry the girl next door.
55) It looks like it's going to rain.

Here, clearly, no spatial motion is involved. The constructions shown in (54)

d (55) result from a centuries-long process of grammaticalization and be going
ID is now a recognized feature of English grammar. Many other languages -
Spanish, Portuguese, French, Dutch - have followed a similar course so that the
rb 'go' has developed into a new way of forming the future tense. Catalan,
- t~restingly, has had a different history, with particular forms of anar 'go'
ving developed into a marker of the past tense, as in Es va menjar el peix 'He
e the fish'. In both cases, the spatial meaning of 'go' is present, namely 'move
away from the deictic center', but the movement in the first-named languages
- away into the future and in Catalan away into the past! The verb come in
English has developed a so-called inchoative sense, i.e. the notion of starting
mething, as in (56):
56) She came to see that he had been lying to her.
In (56) the subject enters into awareness, much as in spatial deixis 'to come'
is 'to enter into the speaker's domain'. In both these examples of
grammaticalization, the original spatial-deictic sense has disappeared. Spanish
w rbal periphrases (Olbertz 1998) offer a rich array of grammaticalizations of the
originally spatial-deictic ir 'go' and venir 'come'.

Time deixis involves identifying points and stretches of time as preceding,

coinciding with, or following the deictic center: think, for example, of the adverbs
yesterday, today and tomorrow respectively. Yesterday occurs in clauses in the
past tense and is incompatible with the present tense (including the present
perfect) and with any way of marking the future in English (with will, is going
to, etc.):

(57) Yesterday I bought/*have bought/*am buying/*will buy/*am going to buy a

new coat.
90 On referring , saying , and implicating : Reference and inference ///////////////

Tomorrow is incompatible with the past tense, occurring either with a future-
time marker or with the present tense (which is then understood as identifying a
future time that has already been identified):
(58) Tomorrow she will graduate/is going to graduate/graduates/*graduated
from college.

Today is compatible with all three tenses, but with the past the meaning is
'during this day but before the moment of speaking' and with a future marker
it is 'during this day but after the moment of speaking':
(59) Today the new president is being sworn in/was sworn in/will be sworn in/is
going to be sworn in.

It follows from this that the past and present tenses as well as the markers
of future time are also time-deictic. These tenses are known technically as
absolute tenses, since they demarcate absolute stretches on the time line, before,
during and after the deictic center. English also has a relative tense, the perfect
tense, 28 which is non-deictic since it does not involve the deictic center but simply
relates events to each other. Consider the following exam:ples:
(60) Martha has now finished her new novel.
(61) Martha had finished her new novel before she started on a new one.
(62) Martha will have finished her novel by the end of the year.

In (60), the absolute tense is the present tense (as is shown by the form lias
and the adverb now) but the relative perfect tense indicates that the moment of
finishing preceded the time identified as 'now'. In (61), the absolute tense in both
clauses is the past, but the first clause also contains the relative perfect tense,
indicating the relation of precedence (in the past) between the two events. In
(62), the same applies, but in the future.
This section has dealt with deixis. Deixis is thoroughly pragmatic in crucially
involving the deictic center, i.e. the moment of speaking and everything that goes
with it. It is also pragmatic in having expanded from the indexical domains of
person, space and time into those of discourse organization (discourse deixis) and
the social relations between interlocutors (social deixis). We have also seen, too,
that deictics can develop non-deictic uses, which substantially extend the range of
pragmatic options for language users. Finally, we considered how deixis impinges
upon grammar through processes of grammaticalization and how absolute tense,
a central feature of English grammar, is a fundamentally deictic phenomenon.

28 We use the term 'tense' here in line with authors such as Comrie (1985), but admit that
relative tenses could also be - and very often are - analyzed as 'aspect' rather than
'tense'. This debate has no bearing on the matter discussed here.
 On referring, saying, and imp li cating: Reference and inference 91

_3_ INFERENCE AND IMPLICIT MEANINGS

both speaking and writing, our discourse contains many implicit meanings that
interlocutors apprehend through the mechanism of inference. In fact,
• IOl[l]·stic communication could not exist without the parallel cognitive process
making inferences, which may be based on different types of meaning, as we
_.l&ll now explain .

.I. TYPES OF INFERENCE

• w· e inferences derive chiefly from contextual assumptions, while others are

[lly built into the structure of the linguistic units that trigger them. The former
said to be pragmatic in nature, while the latter are considered to be primarily
111111:alli:u tic. As we shall see later in this chapter, conversational implicatures are a
- d of pragmatic inference (related to what Grice 1967 called 'non-natural'
. .ieairn·mg or meaning-nn), while the entailments found in syllogisms have a more
·ca1 and semantic basis (related to what Grice called the 'natural' meaning of
erances), and other inferences such as presuppositions normally have both a
~ucu. tic and a pragmatic basis.

A prototypical and famous example of inference with a logical-semantic

· ( defined by the ancient Greek philosophers) is found in bold type in the
ogism in (63):

) All men are mortal.

Socrates is a man.
Therefore, Socrates is mortal.

syllogism has a valid form of the type:

All A are B.
C is A.
Therefore C is B.

The validity of an inference depends on the form of the inference, not on the
th of its premises. But if the form is valid and its premises are true, the
conclusion will always be true as well, as in (63). An invalid form would be:

All A are B.
C is a B.
Therefore, C is A.
92 On referring, saying, and implicating : Reference and inference / / / / / // // / / / ///

in which the true premises lead to a false conclusion, as shown in (64):

(64) All dogs are animals. (Correct)

Lions are animals. (Correct)
Therefore, lions are dogs. (Wrong)

But the fact that the conclusion is wrong does not mean that it is not an
inference. It is still an inference, only a wrong one with an invalid form.
(63) and (64) are not the types of inference that we shall be most concerned
with in this book, for they mainly describe the truth-conditions of the propositions
involved. For obvious reasons, the inferences that mainly occupy the minds of
pragmaticians are those which originate from both the form and the context
of the utterance rather than from its form or truth conditions alone. In fact, when
it comes to analyzing actual language in use, all inferences (including those that
have been labeled as 'purely semantic') have to be tested and evaluated by taking
into account pragmatic aspects such as common knowledge, mutual assumptions,
etc. One such case is that of presupposition, a kind of inference to which we now
turn.

3.3.2. PRESUPPOSITION: SHARED ASSUMPTIONS

AND BACKGROUND KNOWLEDGE
Presupposition is the name given to a variety of linguistic inferences related to
propositions whose truth (or assumption of truth) is taken for granted. These
phenomena have been widely discussed by both philosophers and linguists;
however, they have not yet been totally covered or comprehended. Logicians
(e.g. Strawson 1950) originally described presuppositions in semantic terms,
defining them as propositions entailed by a sentence and its negation, as shown
in (65):

(65) I regret not having accepted the invitation

I do not regret not having accepted the invitation
Presupposition I did not accept the invitation

However, this definition presents some problems, because it fails to describe

the presuppositions of some sentences such as that in (66), whose apparent
negation seems to presuppose just the opposite:

(66) Why did you buy that car? You bought that car)
Why didn't you buy that car? You did not buy that car)
 // II I I I///// I I On referring , saying , and implicating : Reference and inference 93

This and other issues (which are not relevant for the purposes of our
discussion here) have led scholars to reformulate the definition in pragmatic
tenns, the simplest reformulation containing the notion that what matters is not
whether the presupposed is true or not, but that both the speaker and hearer
assume that the proposition said is presupposed as true. Thus, presuppositions
deal with information that is assumed by both speaker and hearer to be necessary
background information for the utterance to make sense and/or to be appropriate.
e inference is normally triggered by a particular expression or grammatical
cture, but there are additional beliefs that have to be presupposed for the
erance to be considered appropriate, which do not necessarily have to be
objectively evaluable as true or false. This is why presuppositions are currently
treated by linguists as properties of utterances (not of sentences) and therefore
primarily pragmatic phenomena. The first author to speak of 'pragmatic
suppositions' (in contrast with semantic ones) was Robert Stalnaker (1977).
He pointed out that even to decide on matters of truth or falsity , a context is
needed in order to correctly interpret the presuppositions triggered by any
ranee. As Mey (2008: 186-187) remarks:
A semantic notion of presupposition [... ] merely links.sentences together on the
basis of what is true or false. A serious theory of pragmatic presupposition goes
beyond this, and inq,uires metapragmatically into the ways an utterance is
understood in the context of the language users' 'common ground' .

.3.2.1 Types of presuppositions

ere are three main kinds of presuppositions that are considered to be
representative of the phenomenon: 1) Existential, 2) Factive, and 3) Connotative
uppositions.
Ex istential presuppositions are inferences made in relation to the existence
of people or things normally described in definite terms. Thus (67) presupposes
that Peter exists, and that he has a brand new Mercedes Benz car, and (68) that
e speaker has a next door neighbor, and that this neighbor has bagpipes. The
expressions triggering these presuppositions are, respectively, Peter, Peter's
nd new Mercedes, Our next door neighbor, and his bagpipes. For this reason
se definite noun phrases are considered to be the 'presupposition triggers':
67) I went for a ride in Peter's brand new Mercedes.
(68) Our next door neighbor is playing his bagpipes.

Factive presuppositions are those triggered by the complements of

epistemic and emotive factive verbs such as know, realize, be aware, regret, be
glad/sorry/surprised/amaz ed that, etc., or the subject complements of m ean,
94 On referring , saying, and implicating : Reference and inference /!/I I I I/ i /// i //

prove, be obvious/ forturiate, etc., as shown in (69), where the presupposition for
all cases (a to f) is the same, namely, that it is a fact (and therefore assumed to
be true by the interlocutors) that Jim is dating Catherine:

(69) a. Did you know that Jim is dating Catherine?

b. I'm glad that Jim is dating Catherine.
c. That Jim is dating Catherine is wonderful news.
d. She is amazed that Jim is dating Catherine.
e. He wasn't aware that Jim is dating Catherine.
f. I was stupid not to realize that Jim is dating Catherine.

lnchoative verbs29 as in (70), cleft constructions as in (71), wh- q,uestions as

in (72), and adverbial and relative clauses introduced by the corresponding
subordinators as in (73) are also associated with factive presuppositions (FP):
(70) Fred discovered that his friend had lied to him. (FP: Fred's friend lied to him)
(71) It was in London that we went to see Tom Jones. (FP: we went to see Tom
Jones)
(72) Why did you call so late at night? (FP: You called late at night)
(73) When he recited the poem, everyone started to cry. (FP: He recited a poem)

It is interesting to note that presuppositions can also be associated with

counterfactive verbs such as pretend or wish in (74) or counterfactual
conditionals, as shown in (75), in which case the complement of the verb and the
proposition expressed by the if-clause (respectively) are presupposed NOT to be
true:
(74) a. He pretended that he was German. (FP: He is NOT German)
b. I wish I had a house in Miami. (FP: I do NOT have a house in Miami)
(75) If I had obeyed my mother, I would be happier now. (FP: I did NOT obey my
mother)

Connotative presuppositions (or connotations) are associated with specific

lexical items that are used in a restricted number of situations in which they
apply. An example of this kind of presupposition can be observed in verbs of
judging, such as accuse, blame or criticize, which involve the propositions 'A did
B' and 'Bis bad', both of which may or may not be presupposed, depending on the
verb, as shown in (76):

29 Inchoative or 'inceptive' verbs are those verbs which show a process of beginning or
becoming.
/ /II I I I I I I I I I I I On referring , saying , and implicating: Reference and inference 95

{76) a. She was accused of stealing the money. (CP: Stealing money is bad)
b. She was blamed for stealing the money. (CP: Stealing money is bad)
c. She was criticized for stealing the money. (CP: She stole the money, and
stealing money is bad)

This shows the normal situation with the verbs in (76) but it is possible for a
speaker to use a verb of judging without herself sharing the judgment, as in (77):

(77) The visitor to Saudi Arabia was accused of showing her hair in public.

In (77), the speaker, if she is an average Westerner, is unlikely to believe

that showing one's hair in public is a 'bad' thing. Nevertheless, by saying (77) she
can indicate that in Saudi Arabian law it is an offense while not committing
herself to any judgment; she is adopting the presuppositions of the accusers. She
could make her personal presuppositions clear, of course, by adding an adverb
such as ridiculously or justifiably, depending on her viewpoint.
Consider also the use of claim and admit in cases like (78):
8) a. The terrorist organization claimed responsibility for the attack.
b. The terrorist organization admitted responsibility for the attack.

In sentences like (78a), the use of claim traditionally went with the
connotative presupposition 'good thing', as in He claimed responsibility for his
daughter's success; admit, seen in (78b), clearly is like blame in having the
connotative presupposition 'bad thing', as in He admitted responsibility for
the plagiarism in the book. However, in recent usage as shown in (78a), claim
bas lost its traditional connotative presupposition and is used without any
presupposition about the goodness/badness of what it claimed. There is thus
a certain flexibility in connotative presuppositions, which can be suspended in
certain cases or can disappear over time.
Other examples of connotations include verbs such as assassinate or
murder, whose use presupposes that the killing was intended, and in the case of
assassinate, that the victim was someone with political power. A great number
of other words, such as those in bold in (79) are also the triggers of certain
connotative presuppositions (CP):
(79) a. My neighbor is playing his bagpipes again. (CP: My neighbor has played
his bagpipes before, and depending on the context and how it is said, it
can also carry the connotation that this repetition is something negative
and annoying)
b. When did you stop singing in bars? (CP: The addressee used to sing in
bars before).
96 On referring, saying, and implicating: Reference and inference //// ///// // // / /

c. We were walking in the jungle when suddenly we heard some roaring

(CP: The roaring, together with the fact of being in the jungle, presupposes
the presence of some animal, probably a tiger or a lion).
d. I started playing tennis when I was ten years old. (CP: I had not played
tennis before I was 10).
e. All my friends are fantastic. (CP: the set of the speaker's friends has at
least three members). 30
f. The regime in that country has to come to an end. (CP: The leader of that
country is a dictator)

In (79a), we again see that connotative presupposition is variable: after all,

to a lover of bagpipe music, the presupposition could be q,uite different. And in
(79c ), we see that they are defeasible: it could be that the walkers in the jungle
were deluded or in fact heard a waterfall or even overheard a film set with
computer-generated animal noises!
Notice that in some cases the same expression can be the trigger for different
kinds of presupposition at the same time. For instance, in (79f), the noun regime
triggers the connotation that the leader of the country is a dictator, and the noun
phrase the regime triggers the existential presupposition that there is a regime in
that country.
As the reader may have noticed by now, there are constructions that
invariably trigger certain presuppositions, a fact that exhibits the close
connection between grammar and this kind of inference, something we shall now
proceed to explain in more detail.

3.3.2.2 Grammar and presupposition

Presuppositions are sparked off by the use of various grammatical

constructions. As noted above, the specific linguistic items that have this effect
are called presupposition triggers. Every time you use a definite noun phrase,
for example, you presuppose the existence of what you are referring to. This is
why some logicians originally argued that you cannot say 'the King of
Switzerland' since Switzerland is a republic and therefore does not have a king.
(This gave rise to many disputes among philosophers about the status of
sentences like The King of Switzerland is bald - this sentence is clearly not
true, but is it false?)

30 Notice that all is not used when the set has only two members. It would be ungrammatical
to say something like "All my ears are big".
 ////////////// On referring , saying , and implicating : Reference and inference 97

As we saw in 3.3.2.1, one freq,uently mentioned kind of grammatical

construction serving as a presupposition trigger is the factive predic ate,
i e. a verb or adjective that is followed by a complement clause, the content of
which must be true for the speaker. Someone who says (80), using the factive
predicate of cognition know, must presuppose the truth of the proposition
'Smoking can kill people':
) He knows that smoking can kill people.

The speaker can only say this if the speaker herself believes that smoking
an kill people. Other factive predicates are realize, learn, discover, remember
and be aware. Certain (but not all) predicates that denote emotions, such as
regret and like, have the same property.
Yet another kind of presupposition trigger is constituted by the 'implicative
predicates' manage andfail. These both presuppose effort on the part of the
bject, differing only in whether success was achieved or not:
1) Henry managed to get the car to start.
) Henry failed to get the car to start.

In both (81) and (82), Henry tried to get the car to start, so this attempt is the
iresupposed content.
Cleft sentences, finally , are also grammatical constructions involving
presupposition. These have the form It+ be+ focus+ presupposition, as in (83):

83) It was Cecilia who persuaded Donald to join us.

This construction is used if it is presupposed that 'Someone persuaded

Donald to join us', with the information marked as 'focus' supplying the identity
of the 'someone'.
Returning to the emotion predicates mentioned above, note that certain
emotion predicates that are followed by a complement clause do not involve a
pr esupposition. Consider (84):

(84) Elizabeth feared that Fritz was too clever for her.

The speaker of (84) does not presuppose that Fritz was too clever for
Elizabeth; in fact, she takes no position on the q,uestion. We therefore say that
f ear is a nonfactive predicate. Not only emotion predicates can be nonfactive in
this sense. Verbs of cognition and communication like think, believe, say, state,
da im and many more share this property. In (84), for example, the speaker
reports Guy's belief without presupposing that it is true:
(85) Guy believed that Hyacinth was a vampire.
98 On referring, saying, and implicating: Reference and inference I I I// I I////////

As noted in the previous section, there are also predicates that are
counterfactive. These involve the speaker presupposing that the content of the
complement clause is untrue. A good example is imagine, as in (86):
(86) The little girl imagined that her bedroom was a castle.
Here the speaker presupposes that the little girl's bedroom was in fact not a
castle. Other counterfactive predicates are wish and prevent, as in (87):
(87) Jonathan prevented Saul from killing David.
Notice that the 'negation test' for factive predicates does not apply to
counterfactives. In (87), the presupposition is that Saul didn't kill David; if you
negate prevented, as in (88), the presupposition is that Saul did kill David:
(88) Jonathan didn't prevent Saul from killing David.
One pragmatic use that is made of presuppositions is to make a req,uest
more polite, by presupposing that it has already been complied with. Thank you
for is a factive presupposition trigger, as in (89), where the speaker presupposes
that the addressee believed in her. However, in (90), the same trigger is used in a
situation where the content cannot be presupposed; rather, that content is -
politely - being imposed on the addressee, giving the meaning 'Don't smoke; and
we're grateful that you are obeying this injunction':
(89) Thank you for believing in me.
(90) Thank you for not smoking.
Another pragmatic use is found in academic writing. Freq,uently it is
necessary to introduce someone else's idea in a complement clause. If the idea is
one they agree with, authors will often use a factive predicate, subtly
presupposing that the contents are true; and if they disagree, a nonfactive
predicate may be applied. Consider (91) and (92):
(91) Jones (2008: 123) observes/shows/proves that Binding Theory has a fatal
flaw.
(92) Jones (2008: 123) claims/states/argues/assumes that Binding Theory has a
fatal flaw.
In any of the variants of (91 ), the speaker presupposes that Jones is correct;
in any of the variants of (92), she makes no presupposition about the correctness
of Jones's position. Only after writing (92) can she go on to contest what Jones
has written.
Barristers tend to be skilled at exploiting presuppositions, as in the famous
example (93):
(93) So, tell the court, when did you stop beating your wife?
I I I I I I I I I I I I I I/ On referring, saying , and implicating: Reference and inference 99

The poor accused cannot answer this sentence without condemning himself,
since, as we have already seen, stop is a factive predicate. The only possible
answer - if the accused is innocent or wishes to appear so - is to reject the
presupposition of the q,uestion, for example by saying But I never started beating
her or, more formally, I cannot accept the presupposition of your question.
ormally, of course, the judge will disallow such a q,uestion, known in legal
circles as a 'loaded q,uestion'.
In this section, we have seen that various aspects of the grammar of
English can only be understood if we take the pragmatics of presupposition
into account. The interconnectedness of grammar and pragmatics is one of the
phenomena dealt with in functional grammars, which aim to understand not
just what the structures of the language are but also how they are deployed in
communication.

3.3.3. IMPLICATURE
As is shown all throughout this book, what a speak~r intends to communicate
is normally far richer than what she literally expresses, and thus she has to
exploit pragmatic principles that the hearer can consciously or unconsciously
invoke in order to bridge the gap between the literal content ( determined by its
grammatical structure) of the 'said' and her intended meaning.
In Chapter 2 we saw that one of the basic concepts in Gricean pragmatics
is speaker meaning. Grice (1967) makes a distinction bet ween natural
m eaning, which is devoid of human intentionality, and non-natural meaning
(meaning- nn), which refers to the intended message (i.e. the meant-but-unsaid).
The bridge from what is said to what is meant is very often built through a kind
of inference that Grice called implicature, a notion that is one of the single most
important ideas in pragmatics, and which is rooted in the fact that messages are
radically underdetermined if only the natural meaning is taken into account for
their production and interpretation.
Grice's explanation of meaning-nn is sketched by Levinson (2000: 13) as
follows:
S means-nn p by "uttering" U to A iff31 S intends:
a. A to think p
b. A to recognize that S intends (a)
c. A's recognition of S's intending (a) to be the prime reason for A thinking p.

31 If and only if.

100 On referring , saying, and implicating: Reference and inference //// ///////// //

Thus the semiotic picture of the total signification of an utterance is

composed of both what is said (the natural meaning) and what is implicated (the
meaning-nn). But meaning-nn is not homogeneous: Grice wrote about two main
kinds of inference that generate such meaning, which he called 1) Conversational
lmplicature, and 2) Conventional Implicature. He also contemplated other
subtypes of intended meaning, such as those generated by presupposition or
nonconversational and nonconventional implicatures, but these did not constitute
the focus of his research and work. Within Grice's theoretical framework, the
said and the conventionally implicated are coded by the linguistic system.
However, those meanings that are conversationally implicated are not; they are
only inferred on the basis of some basic rational assumptions stated in the
Cooperative Principle and its maxims of conversation, which all rational speakers
are assumed to respect and follow (Grice 1975: 4H6):

The Cooperative Principle:

Make your contribution such as is req,uired, at the stage at which it occurs, by the
accepted purpose or direction of the talk exchange in which you are engaged.
The maxim of Quantity
1. Make your contribution as informative as is req,uired (for the current purposes
of the exchange).
2. Do not make your contribution more informative than is req,uired.
The maxim of Quality
Try to make your contribution one that is true, specifically:
1. Do not say what you believe to be false.
2. Do not say that for which you lack adeq,uate evidence.
The maxim of Relation
1. Be relevant.

The maxim of Manner

Be perspicuous, and specifically:
1. Avoid obscurity of expression.
2. Avoid ambiguity.
3. Be brief (avoid unnecessary prolixity).
4. Be orderly.

But speakers do not always follow the maxims to the letter, which may
generate conversational implicatures, as we shall now explain.
 / /II I I I I I I I I I On referring, saying, and implicating : Reference and inference IOI

3.3.3.1 Conversational lmplicafure

pointed out above, it is normally expected that rational speakers will follow
the maxims of the Cooperative Principle. Nevertheless, very freq,uently these
maxims are not observed, and speakers intentionally 'flout' one, two, three or all
of the maxims at the same time. In these cases, the hearer still assumes that the
speaker is trying to be cooperative, and therefore looks for meaning at a deeper
level than the literal one. In doing so, he makes an inference of the type called
m nversational implicature which is not coded in the language and has its basis
in the rational shared assumptions of the interlocutors. The utterances in bold in
examples (94) to (97) show how the four maxims can be flouted.
) Flouting the maxim of Quality:
Charles: Look! It's raining! Why don't we rush to the beach and bathe in the
middle of the storm? It would be very exciting, don't you think?
Lucy (rolling her eyes in disagreement): Yes, VERY exciting!!

Here there may be many variables that will make Charles realize that Lucy
is not being cooperative: Two obvious signs are 1) the fact that at the same time
she is saying "Yes, VERY exciting" she is rolling her eyes, and 2) the exaggerated
stress (and intonation - probably falling) given to the adverb VERY. But there
may be other features of the context or shared knowledge, such as the fact that
Lucy is afraid of storms, or that swimming in the sea is prohibited when there are
storms, etc. that will make any hearer realize that Lucy is flouting the Quality
maxim intentionally, thereby being ironic. This is a prototypical case of irony, in
which the speaker implies the opposite of what is literally said; 32 thus the
conversational implicature triggered by flouting the maxim of Quality is that she
does not think that bathing in the sea in the middle of a storm is exciting, but on
the contrary, that it is a crazy idea.

(95) Flouting the maxim of Quantity:

A: I think Robert and Eunice are very honest people.
B: Well, I think Robert is honest, ges.

In this example B is being 'less informative than req,uired' by not including

Eunice in her statement, and therefore not totally cooperative by flouting the
maxim of Quantity. This will most probably make A work out the implicature

But ironic utterances can also flout any of the other three maxims, for irony is not
always realized by not saying that which is true. For instance, a statement like "I love
people who are polite" said after someone has insulted the speaker, expresses a true
proposition but is ironic anyway, and the maxim being flouted is not the Quality but
the Manner maxim.
102 On referring, saying, and implicating : Reference and inference / / / / // / // / / / / / /

that B thinks that Eunice is NOT honest. And again, there may be other aspects
of A and B's shared knowledge that will make A infer this, such as his knowledge
that B and Eunice don't like each other, or that they both had an argument in the
past, etc.
(96) Flouting the maxim of Relation:
C: Would you like to come to my home town with me this weekend?
D: I have to keep on working on my project during the weekend or
else I'll never finish.
In (96), D's response to C's q,uestion does not seem to be relevant, for she is
not directly answering it by saying whether she would like to go to C's home town
with him or not. However, C will most probably infer (by means of conversational
implicature) that D is declining his invitation because she either cannot or does
not want to spend the weekend with him.
(97) Flouting the maxim of Manner:
E: Have you ever met Richard?
F: No, what's he like?
E: Well, he's not what one would call 'handsome'. ·
E's statement is not a direct one. She could have been briefer and less
obscure by simply saying "He's ugly", but for reasons of politeness, she has
chosen not to obey the Manner maxim by using an obscure circumlocution and
thus minimizing or softening the implicit negative judgment regarding Richard's
appearance.
In all the above examples the blatant non-observance of the maxims on the
part of the speaker is what triggers the search for a conversational implicature.
In effect, Grice (1989 [1975]: 30-31) explains that a speaker flouts a maxim when
he blatantly fails to fulfill it and thus characterizes the notion of conversational
implicature in the fallowing manner:
A man who, by (in, when) saying (or making as if to say) thatp has implicated
that q,, may be said to have conversationally implicated that q,, provided that (1)
he is to be presupposed to be observing the conversational maxims, or at least
the Cooperative Principle; (2) the supposition that he is aware that, or thinks
that, q, is req,uired in order to make his saying or making as if to say p (or doing
so in those terms) consistent with this presumption; and (3) the speaker thinks
(and would expect the hearer to think that the speaker thinks) that it is within
the competence of the hearer to work out, or grasp intuitively, that the
supposition mentioned in (2) is req,uired.
But the reader should know that this is not the only way in which a maxim
may be 'not observed'. There are other forms of non-observance of the maxims,
as we shall now see.
I////// I/ I/ I I I/ On referring , saying , and implicating: Reference and inference 103

.3.3.2 Other ways of not observing the maxims

·ce also pointed out that there are many occasions on which speakers fail to
serve the maxims even though they have no intention of generating an
• plicature. Grice writes about three other such types of non-observance:
Violating a maxim, 2) Opting out of observing a maxim, and 3) Infringing a
- AJ·u.• . Thomas (1995: 72) points out that several writers since Grice have argued
need for another category, namely 4) Suspending a maxim. We shall briefly
lain each of these categories here in order to make it clear that the only type
non-observance of the maxims that generates conversational implicatures is
flouting of a maxim. The other types may also generate inferences, but of a
erent kind.
Violating a maxim: Some authors have used the term 'violate' for all cases in
which the maxims are not observed, but Grice clearly defines 'violation' as the
u nostentatious non-observance of a maxim, which means that the speaker
who violates a maxim does not expect the hearer to know or realize that she
is doing so. This category would then include cases of lying, in which the
speaker violates the maxim of Quality but with -DO intention whatsoever of
making the hearer realize that she is not telling the truth (in fact, this is the
essence of lying), and therefore without generating any conversational
implicature. This is very different fromflouting this maxim and being ironic
(as in (94) above), in which case the speaker DOES want the non-observance
of the maxim to be noticed, and this is precisely what triggers the
conversational implicature. Also, this category includes cases in which
the speaker is not lying but is 'liable to mislead', as Grice (1989 [1975]: 30) puts
it Thomas (1995: 73) gives an example from a fictional interaction in which a
woman is cheating on her husband with a woman, and the husband suspects
she may be having an affair with another man, so he asks her if there is
another man', to which she replies 'No, there isn't another man'. With such an
answer, the woman is not lying but (by violating the Quantity maxim and not
t.elling the 'complete' truth) she is certainly misleading her husband, who could
never suspect she was having a homosexual relationship with another woman.
Opting out of a maxim: A speaker opts out of a maxim when she indicates
that she does not want to cooperate in the way the maxim req,uires but does
not wish to generate a false implicature or appear uncooperative. Examples
of this type of non-observance are freq,uently found in public life when, for
instance, a priest who is called to declare in a court refuses to reveal what the
accused told him in confession.
Infringing a maxim: When a speaker has neither the intention of generating
an implicature nor of deceiving anyone, she is said to 'infringe' a maxim. This
may occur because the speaker is nervous, excited or drunk, or has some
104 On referring , saying , and implicating : Reference and inference /////// ////////

cognitive impairment, or because she does not have full command of the
language (e.g. if she is a foreigner or a little child) and thus is incapable of
making a clear point, etc.
4) Suspending a ma.xim: This category applies in cases in which there is no need
to opt out of a maxim because there is no expectation that they will be
followed , and therefore their non-observance does not generate any
implicatures. A clear example of the suspension of the maxim of Quantity can
be found in the telegrams of the 20th century, or in SMS messages nowadays,
where everyone expects that the text will be short, and therefore the cut in the
q,uantity of words does not generate any conversational implicatures.

Now the rationale for conversational implicature has been sketched, it is

time for us to turn to the other type of implicature that Grice included within his
description of 'meaning-nn', namely conventional implicature, which, as we
shall see, differs from conversational implicature in essential ways.

3.3.3.3 Conventional lmplicature

Grice also characterized another type of implicature which he called conventional

because it is generated by the conventional meaning of the words or expressions
used. Consider (98) and (99):
(98) He is poor but honest.
(99) Even teachers can afford one of these laptops.

By using the connector but, the speaker in (98) is committing herself to

supporting the notion that poor people are normally not honest, and she cannot
say she did not mean this because of the adversative meaning which is inherent
to and conventionally implicated by but. Likewise, in (99) the meaning
conventionally implicated by even tells us that the speaker commits herself to the
idea that teachers cannot afford to buy expensive computers.
Grice describes conventional implicatures by explaining that "in some cases
the conventional meaning of the words used will determine what is implicated,
besides helping to determine what is said" (1989: 25). Thus, conventional
implicatures, being more attached to the linguistic form 33 of the utterance than

33 Conventional implicatures are attached to the form of utterances, but not to their
semantic content. That is why, in contrast with conversational implicatures - which
are non-detachable from semantic meaning- conventional implicatures are said to be
detachable from it. This means that, for instance, the word but will always carry the
same conventional implicature, irrespective of the general semantic content of the
utterance in which it is placed. See 3.3.3.4.
/ /////I//////// On referring, saying , and implicating: Reference and inference 105

conversational implicatures, deal with non-cancelable aspects of meaning and

therefore, like presuppositions, are triggered by features of the natural meaning
of utterances. In fact, conventional implicatures are akin to pragmatic
presuppositions, and many authors consider them to be the same phenomenon.
K.arttunen & Peters (1975, 1979), for instance, suggest that the prototypical
examples of presupposition should be best treated as conventional implicatures.
Grundy (2008: 106), too, observes that "conventional implicatures and implicative
erbs share identical properties, although traditionally (and perhaps
inappropriately) have been treated as presuppositions" . Our view is that
ronventional implicatures are indeed the Gricean eq,uivalent of presuppositions;
1his is one instance of where the various traditions within pragmatics meet.

Presuppositions aside, what matters most for the purposes of this section is
to draw the difference between conventional and conversational implicatures.
The property of non-cancelability for conventional implicatures is one important
aspect in which they differ, but there are other properties that play a part, as we
all now try to clarify.

3 .3.3. 4 How do conventional and conversational implicatures differ?

While conventional implicatures are mainly derived from the conventional

eanings of the words and expressions used, conversational implicatures deal
with non-natural, non-conventional meaning, and therefore the latter are said to
display the following distinctive properties (Grice 1975: 57-58):
a) Cancelabilityldefeasibility: This means that the inference derived from the
flouting of the maxims can be canceled or defeated by adding some premises.
Thus, for instance, in (95) above (A: I think Robert and Eunice are very honest
people/ B: Well, I think Robert is honest, yes) the speaker could always add,
for instance: "Well, I don't know Eunice very well, so I cannot judge her",
thereby canceling the implicature that Eunice is dishonest. In contrast with
conversational implicatures, deductive or logical inferences of the type
illustrated in (63) and (64) above (syllogisms) are not defeasible or cancelable,
and neither are presuppositions or conventional implicatures.
b) Nondetachability: By saying that conversational implicatures are
no ndetachable, Grice means that the implicature is not attached to the
linguistic form, but to the content of what is said, i.e., they are nondetachable
from the semantic content of the utterance. Thus, conversational implicatures
cannot be detached from an utterance by simply replacing the words by
synonyms. For instance, if we change the form of (95) to "I think Robert is a
sincere and reliable person" or "Yes, Robert is extremely frank and genuine" it /
106 On referring, saying, and implicating: Reference and inference ///////////////

will still trigger the implicature (provided the conditions of the context are the
same) that the speaker does not think that Eunice is honest. lmplicatures
deriving from the Manner maxim, however, are considered to be exceptions
to this, because they are not totally detachable from form (i.e. they are based
on forms which are synonymous in context but vary in length or lexical
freq,uency). 34 In contrast to conversational implicatures, conventional
implicatures are always detachable from the content of the utterance
(See 3.3.3.3).

c) Calculability: Conversational implicatures are calculable because, according

to Grice, for every occurrence of the phenomenon, and based on the
assumption of cooperation, it should be possible to show how an addressee
can make the desired inference from the literal meaning or sense of the
utterance in q,uestion. In other words, based on the Cooperative Principle, all
implicatures can be 'calculated'. Conventional implicatures, in contrast, are
not calculable by using pragmatic principles and contextual knowledge; they
are given by convention.

d) Nonconventionality: This property simply alludes to the fact that

conversational implicatures are not conventional, and therefore are not
generated from the conventional meaning of linguistic expressions, as
conventional implicatures are.

Levinson (2000: 15) adds the following features to Grice's list, also taking
into consideration the works of Sadock (1978) and Horn (1991):

e) Reinforceability: This property refers to the fact that it is always possible to

reinforce what is implicated by going on to say it in an explicit, overt way,
which is clearly easier on the listener even if there are risks about being so
direct. So for instance, in (95) above, the speaker could add (after a silent
moment) "But Eunice is NOT" as a reinforcement of the original utterance
"Well, Robert is an honest person, yes"; this is unsubtle but it is clear.

f) Universality: Since this type of inference is not arbitrary but motivated, and
it is assumed to derive from basic considerations of rationality, it is expected
to be universal, in contrast with coded meanings. This means that in any
language into which a given utterance is directly translated, the eq,uivalent
form should carry the same standard implicatures. So, for instance, the
utterance "I have three nieces" carries the universal generalized implicature

34 Grice's distinction here, as well as his general treatment of these properties, can be
problematic, for one could find examples, for instance, of Manner implicatures that
are nondetachable from content as well.
 / /II I I I I I I I I I I On referring , saying , and implicating : Reference and inference 107

that 'I have no more than three nieces' in any language into which this
utterance can be directly translated.
Some authors, such as Sadock (1978) or Levinson (1983, 2000), have pointed
out that - when it comes to testing implicatures and distinguishing them from
other kinds of pragmatic inferences - all the above properties can be problematic,
for several counter-arguments can be marshaled to show that they do not apply
to all possible occurrences of the phenomenon.

3 .3.3.5 Generalized and particularized conversational implica+ure

In spite of the shared and distinctive properties of conversational implicatures

set out in 3.3.3.3, not all of them are 'born eq_ual': Grice drew a distinction between
what he termed generalized and particularized conversational implicatures.

Generalized conversational implicatures (GCI) are those implicatures that

are triggered without the need for a special context or scenario. Thus if a speaker
otters (100), irrespective of the scenario where she got into a car, the hearer will
be taken to infer that the car is not the speaker's, for the expression an X seems
to trigger the generalized conversational implicature that X is not closely related
o the speaker. Similarly , in any possible context, an utterance like (101) will
always tend to generate the implicature that the speaker cannot run the mile in
less than 4 minutes. Another example is the one already given to illustrate the
property of universality in 3.3.3.4 (I have three nieces) which will tend to trigger
the implicature that the speaker has no more than three nieces, no matter what
the context or scenario.

(100) I got into a car.

(101) I can run the mile in 4 minutes.
Levinson also notes that GCis are very important for linguistic theory, for
they are hard to distinguish from the semantic content of linguistic expressions,
given the fact that "they will be routinely associated with the relevant expressions
in all ordinary contexts" (1983: 127).

In contrast to GCI, there are other implicatures that will only be triggered if
certain conditions are met and thus are classified as particularized conversational
implicatures (PCI). Consider (102), where B's flouting of the Relation maxim
could only implicate that Lily is with Jack in the particular sort of setting that A
is Lily's brother (or relative) and has come to Jack's (and B's) house to ask about
her whereabouts, that Lily had previously come to visit Jack, and that the
participants share the knowledge that Jack likes Lily very much and therefore is
happy when in her company:
108 On referring, saying, and implicating: Reference and inference // / // ///// / / / / /

(102) A: Hi, I'm looking for Lily. Have you by any chance seen her?
B: Jack is very happy (PCI: Lily is now with Jack)

Examples (94)-(97) above display instances of utterances generating PCis as

well.
Grice does not devote too much space to the treatment of the distinction
between GCis and PCis; his formulation lacks detail. Levinson (2000: 16) thus
extends this treatment and proposes the following formulation:
The distinction between PCis and GCis
a. An implicature I from utterance U is particularized iff U implicates I only by
virtue of specific contextual assumptions that would not invariably or even
normally obtain
b. An implicature i is generalized iff U implicates I unless there are unusual
specific contextual assumptions that defeat it.

Example (103) illustrates and clarifies this difference by showing that a

given utterance can generate both a GCI and a PCI at the same time. This is an
interesting point, because it shows how the assessment of the weightiness of the
two implicatures can cause a certain balance in the decision later taken by
the speakers, given that, on the one hand, they should probably consider buying
a water regulator because some of their neighbors already have one (PCI), but on
the other, perhaps they don't need one, for not all their neighbors have installed
one (GCI):
(103) A: Do we really need to install a water regulator?
B: Some of our neighbors already have one.
PCI: Perhaps we also need to install one, yes.
GCI: Not all of our neighbors have installed a water regulator.

As to the differences between GCis and conventional implicatures

(something the reader might be wondering about), Grice himself admitted that
"non-controversial examples are perhaps hard to find, since it is all too easy to
treat a generalized conversational implicature as if it were a conventional
implicature" (1989: 37), but the basic criteria for distinguishing them would be, as
explained in 3.3.3.3, that while GCis are cancelable, nondetachable, reinforceable
and universal, conventional implicatures have none of these properties.
Grice's distinctions regarding species of meaning are summarized in Figure 2,
although the picture is partial, for he himself allowed that there are other
subtypes of signification (e.g. presupposition, other types of implicatures, etc.)
that are not included in his model.
 I/ //// I///// On referring, saying , and implicating : Reference and inference 109

- ure 2: Grice' s classification of meaning

Furthermore, this model does not show that there are other kinds of
- erence, such as those based on interactional politeness or conversational
anization that were later to be treated by other authors ( e.g. Leech 1983,
wn & Levinson 1987), nor does it include the neo-Gricean interpretation of the
ry of implicature, which - among other things - points to the difficulties
din making the distinction between what is said 3:nd what is implicated, and
within which not all of the Gricean maxims receive eq,ual attention, as we shall
·efly explain in 3.3.3.6.

.3.3.6 Neo-Gricean theories of implicature

Hom (1972, 1984, 2007) and Levinson (1995, 2000) have devoted much of their
ork to a refinement of Grice's theory of implicature, above all by exploring the
m ncepts of scalar implicature and generalized conversational implicature.
Scalar (or Quantity) implicatures derive from utterances in which a scalar
rm is used to suggest that the speaker had a reason for not using a more
informative or stronger term on the same scale. As the reader may have noticed
· some of the examples above (e.g. (103)), many utterances display scalar
operties. Consider now the utterances in (104):
(104) a. Some students do not like Professor James.
b. Sam should be coming back from Paris now.
c. I can buy the red dress or the blue one.

In these three utterances, some, should and or are members of different

scales. The scale to which some belongs includes <all, many, some, few>, and
llffillg some in this particular utterance will imply that the members of the scale \
to its left are not applicable. Likewise, should is a member of a scale that includes
other modal verbs <will, should, may, might>, with the verb to the left of should
110 On referring, saying, and implicating: Reference and inference // / / // / / / / / /// /

(but not to its right) being excluded from the meaning of (104b) by scalar
implicature. In the case of or in (104c), there is an indeterminacy that could be
solved by adding "but it has to be one or the other, because I can't afford to buy
both", or in contrast, "I need two dresses for a special event, so I think these two
colors are the most appropriate". In the first case, or would mean "either the red
dress or the blue one, but not both", while in the second, it would mean "either the
red one or the blue one, or both". Oris in a scale <or, and> with and, and therefore
a speaker using or implies "not and". The scalar view of or has helped linguists
account for the two meanings of this connector, the inclusive and the exclusive
one. They have resolved it by explaining that in cases like (104c) "either the red
dress or the blue one, or both" is an entailment and "either the red dress or the
blue one but not both" is an implicature.
Horn's (1972) work on lexical gaps is an early example of neo-Gricean
pragmatics. He noticed there was a gap in the group of lexical items containing
the negative particle. For instance, he pointed out that in the English language we
have none (not one), but not *nall (not all), and he argued that since not all is
implied by the use of some and therefore is already a Ql 35 implicature, its
meaning does not need to be lexicalized. 36 The important contribution of Horn's
work in this respect was to show that the Ql Maxim ("Make your contribution as
informative as is req,uired") is crucial to determining which concepts can be
lexicalized. This leads to the conclusion that the foundation behind this kind of
lexical gap is pragmatic and favors the idea that other phenomena might also be
explained by examining such pragmatic conversational principles.
In later works, Horn (1984, 2007) reinterpreted Grice's Quantity, Manner and
Relation Maxims and argued that the Maxims of the Cooperative Principle can be
reduced to just two principles: the Q Principle and the R Principle. The Q
Principle is a reinterpretation of the Ql maxim, and the R Principle is a
reformulation of the Q2 maxim and the Relation and Manner maxims, as
reproduced here (Horn 1984: 13):
The Q Principle: Make your contribution sufficient. Say as much as you can.
The R Principle: Make your contribution necessary. Say no more than you
must.

35 A Ql implicature is an implicature derived from the flouting of the first part of the
Maxim of Quantity, i.e. "Make your contribution as informative as is req,uired". A Q2
implicature is one derived from the non-observance of second part of Grice's formulation
of the maxim: "Do not make your contribution more informative than is req,uired".
36 The same happens with a) sometimes, which implies not always, and thus there is no
such term as *nalways; b) or, implying not and, and thus there is no "'nand; c) one,
implying not both, and thus there is no *noth, etc.
 I I I// I I I I I I// I On referring , saying, and implicating : Reference and inference Ill

Example (105) shows an utterance from which the addressee will most
probably recover the Q inference that the woman in question was a stranger, or
least not a person close to John, since he didn't say 'my wife', 'my mother', etc.
because it is assumed that he has said 'as much as he can':
(105) John: I met a woman at the supermarket yesterday.

In (106) the hearer will recover the R inference that there will not be more
than seven people for lunch, because it will be assumed that she is following the
Principle of 'not saying more than you must':
106) There will be seven of us for lunch today.

Thus, according to these Principles, a speaker will only be as informative as

is required, and will not use any stronger terms unless required. The hearer will
· erefore assume that the stronger term does not apply; so for instance, the
onger term "more" will not apply in (106), and that is why the speaker did not
say "There will be more than seven of us for lunch today".

Another neo-Gricean author is Stephen Levinson who, within his theory of

preferred interpretation (1995, 2000), defends the notion of generalized
m nversational implicature as an essential explanatory notion for a great
ariety of linguistic facts, such as those exhibiting the lexicalization constraints
studied by Horn and exemplified above in this section. This author argues in
favor of the GCI as the default or preferred inference, i.e. "one that captures our
intuitions about a preferred or normal interpretation" (2000: 11). To explore
default inferences, Levinson focuses on the maxims of Quantity and Manner,
arguing that this kind of inference can neither be reduced to the level of
entence-meaning nor to that of speaker-meaning, for they are midway
phenomena, influencing both grammar and speaker-meaning.
As we saw in 3.3.3, Grice distinguished between what is said or entailed (i.e.
what our words mean) and what is implicated (i.e. what we mean with our
words). Levinson identifies not two, but three types of meaning:

l) Entailment (which is context-free and non-inferential),

2) Utterance-type meaning (which is context-free an9 inferential), and
3) Utterance-token meaning (which is context-sens/ ive and inferential).
!
The Theory of Preferred Interpretation concentrate~ upon the second type,
and is thus a theory of utterance-type meaning. An utterance type is a type of
utterance with a 'predictable' GCI (Levinson 2000: 176), i.e. its inferred
interpretation is regular across a range of contexts. Thus, GCis belong to the
realm of utterance types. Because utterance type meaning is expectable and
112 On referring , saying , and implicating : Reference and inference ///////////////

conventional to a certain extent, it resembles sentence meaning; however, it is

different from sentence meaning in that it is inferred and cancelable, as in (107),
a reformulation of (105) above, where the GCI that the woman is unknown or not
close to the speaker is defeated by the addition made after the pause:
(107) I met a woman at the supermarket yesterday .. . and that woman was my
mother!
In contrast, an utterance token is an utterance that has a single inferred
interpretation which depends upon its context. Thus, particularized
conversational implicatures (see 3.3.3.5 above) belong to the realm of utterance
tokens.
Levinson bases his tripartite theory of utterance-type meaning on three
basic heuristics or insights that in his view govern the process of inferential
enrichment and that enable him to formulate the three following principles,
derived from Grice's Quantity (divided into Q1 and Q2) and Manner maxims:
1) Q-principle (Ql)
Speaker's m axim: Do not provide a statement that is info.rmationally weaker
than your knowledge of the world allows, unless providing an informationally
stronger statement would contravene the I-Principle. Specifically, select the
informationally strongest paradigmatic alternate that is consistent with the
facts.
Recipient's corollary: Take it that the speaker made the strongest statement
consistent with what he knows. (Levinson 2000: 76)

The recipient's corollary will induce scalar and clausal37 Q-implicatures, as

illustrated in (108):
(108) a. Some of my classmates are lawyers.
(Scalar implicature Not all of my classmates are lawyers)
b. Peter believes his illness was caused by a virus
(Clausal implicature Peter does not know whether his illness was caused
by a virus or not) /

2) I-Principle (or Principle of Informativeness, based o1 G.~ce's Q2 maxim)

Speaker's maxim: The maxim of Minimization. "Say as little as necessary";
that is, produce the minimal linguistic information sufficient to achieve your
communicational ends (Bearing Q in mind).

37 Clausal irnplicatures are those derived "from contrasts between one expression that
entails its embedded sentence(s) and another that does not" (Levinson 2000: 76-77).
 IIIIIIIIIIIIII On referring, saying, and implicating: Reference and inference 113

Recipient's corollary: the Enrichment Rule. Amplify the informational content

of the speaker's utterance by finding the most specific interpretation, up to
what you judge to be the speaker's rn-intended point, 38 unless the speaker has
broken the maxim of Minimization by using a marked or prolix expression.
(Levinson 2000: 114)

Levinson explains that the I-Principle collects a range of inferences that

appear to go in the reverse direction to that in which Q-irnplicatures tend:
'1--irnplicatures are inferences from the lack offurther specification to the lack of
lleedfor it, whereas Q-implicatures are inferences from the lack of informational
richness to the speaker's inability to provide it" (2000: 116; emphasis in the
"ginal).
Some examples exploiting the I-principle are found in (109), which show that
1-implicatures are based on principles of economy:
109) a. Alba-Juez and Mackenzie have written a book on pragmatics.
(I-implicature Alba-Juez and Mackenzie wrote the book together)
b. Charles left the building and a bomb exploded.
(I-implicature Charles first left the building and then the bomb
exploded)
c. Possessive interpretations:
Ken's house (I-implicature the house he lives in)
Ken's children (I-implicature the children to whom he is a father)
Ken's proposal (I-implicature the proposal he created or made)
Ken's duty (I-implicature the duty falling on him)

3) M-Principle (Manner)
Speaker's maxim: Indicate an abnormal, nonstereotypical situation by using
marked expressions that contrast with those you would use to describe the
corresponding normal, stereotypical situation.
Recipient's corollary: What is said in an abnormal way indicates an abnormal
situation, or marked messages indicate marked situations. (Levinson 2000:
136)
Levinson observes that M-implicatures seem to be essentially parasitic on
corresponding 1-implicatures. Thus compare the M-implicatures in (110) with
their corresponding 1-lmplicatures in (109a & b):

M-intention, according to Grice (1989: 105), is the complex reflexive intention involved
in speaker's meaning, i.e. her intention to cause an effect in the recipient just by getting
the recipient to recognize that that was her intention.
114 On referring, saying, and implicating: Reference and inference ///////////////

(110) a. Alba-Juez and Mackenzie both have written a book on pragmatics.

(M-implicature Alba-Juez and Mackenzie wrote two separate books
on pragmatics).
b. Charles left the building and almost immediately thereafter a bomb
exploded.
(M-implicature Charles leaving the building and the bomb's explosion
may have been simultaneous.) .

As the reader may have noticed by now, Levinson's principles were inspired
and based on Horn's. Levinson (2000: 137) himself acknowledges that Horn's Q
Principle is "pretty much eq,uivalent" to both his Q and his M Principles, while
Horn's R principle "is roughly coextensive" with Levinson's I-Principle.
Levinson's theory of presumptive (i.e. preferred or presumed) meanings is
much more complex and extensive than we are able to show here, but its essence
lies in being a set of generalizations about the interrelations of linguistic
expressions and the default inferences generated by them. Finally, it should be
added that Levinson claims his theory is predictive in the sense that it can make
predictions about presumptive meanings at different levels of generalization
(morphological, lexical, etc.).

3.4. CONCLUSION
This chapter has shown how thoroughly pragmatics penetrates our everyday use
of language by examining two notions with rather similar names, reference and
inference, but with very different effects. Reference is concerned with how we
identify and talk about the entities that form part of our discourse (persons,
things, events, etc.). Inference is involved in how we interpret what is being said
to us, and this is especially important when the words uttered are not a direct
reflection of the speaker's communicative intention.
What is common to both notions is that speakers, whether referring or
inviting an inference, regularly leave it to the hearer to figure out what is meant.
When someone leaves the office building and says to a co-worker, "Now, where
am I parked?", she uses the first-person pronoun I; the denotation of that pronoun
is the speaker but the reference is to her car. The colleague has no trouble
understanding this because (a) you don't park people and (b) the context of
looking for your car in the company's parking lot is a familiar part of the day's
routine. Reference in this case, as in so many instances, is a co-operative activity
involving input from the speaker and inferential work from the hearer. With
conversational implicature, similar processes are in play. When Romeo sighs that
Juliet is the sun, he doesn't mean she is a fiery ball of gas 150 million kilometers
/ II I/ I I I I I I I I I/ On referr ing, saying , and implicating : Reference and inference 115

away; rather, as with all poetic metaphors, the listener in the theatre understands,
using his own experience of love (and his knowledge of Renaissance cosmology),
what it is like to feel that someone is the center of your universe, warming and
nurturing you. That understanding is based, then, on processes of inference that
can involve any aspect of a hearer's cognitive capacities (emotions, experience,
education).
One major reason (but certainly not the only one) for a speaker to use
indirect reference or a conversational implicature is to avoid hurting her
interlocutor's feelings. Rather than using critical words, for example, a speaker
can employ a formulation from which the hearer can deduce that (a) the
speaker is not pleased but that (b) she still respects the hearer. Questions like
these have been studied under the heading of 'politeness', and it is to the various
theories of linguistic politeness that the next chapter turns.

SUMMARY
1. The first major subject of this chapter is reference, covered by points 1 to 5 of
this summary. This involves expressions that either introduce a participant
(a 'referent') into discourse or, once it has been introduced, maintain its
presence there. Referring expressions appeal not only to the denotation of the
words (their dictionary meaning) but also - and ~!Ucially - to the context, a
fact that justifies their treatment in pragmatics.
2. The initial introduction of a referent is usually associated with indefinite
reference, while subseq,uent mentions involve definite reference. Reference
may also be ~ither specific (identifying a particular referent) or generic
(identifying a class of referents). In addition, reference may be anaphoric,
bridging, cataphoric, esphoric, homophoric, ex ophoric or ideophoric. The
various references in a text create tracks through the discourse.
3. Where reference is made to an aspect of the ambient context, we speak of
deictic reference or deix is. Deictic expressions like today, over there,
Tuesday, etc. cannot be interpreted without access to details of the
circumstances under which they have been uttered, known collectively as
the deictic center. Individuals in close proximity share the deictic center but
if they are in different spaces, a speaker may adopt the deictic center of the
hearer.
4. Five types of deixis are recognized: person deix is, time deix is, spatial or
place deixis, discourse deixis and social deixis. Cross-cutting this taxonomy,
especially with person and place deixis, is the distinction between gestural
deixis (which involves actual pointing) and symbolic deixis (which does not).
Certain deictic expressions can develop non-deictic uses.
116 On referri ng, saying, and impl icating : Reference and inference / /// // / / / / /////

5. Historically, spatial deixis is the most basic of all types, and its traces can be
found in the grammar of all the other types of deixis. Whereas English spatial
deixis only differentiates prox imal and distal (this vs. that), other languages
have more distinctions. Verbs like come and go are also deictic and in many
languages have penetrated the tense system. Since the opposition between
present, past and future time relates to the deictic center (identifying events
coincident with, preceding and following it respectively), absolute verb tenses
are also regarded as deictic in nature. Relative verb tenses, by contrast, are
non-deictic.
6. Inference is a cognitive process that accompanies all communication.
Inferences come in different types: entailments are relations between
propositions such that one of them is strictly implied by the other-
presuppositions are traditionally defined by logicians as background
assumptions that are entailed by a proposition and its negation;
and implicatures are meanings that are not 'there in the words' but are
derived from interpreting an utterance in its context. The second and third
types have been much studied in pragmatics.
7. Presuppositions were initially analyzed by logicians but later came to be
identified by pragmaticians as conveying part of the background information
that is necessary for any utterance to make sense in context. Three basic
kinds are distinguished: Ex istential presuppositions are normally associated
with definite referring expressions, factive presuppositions with the
complements of certain epistemic and emotive verbs and other grammatical
phenomena, and connotative presuppositions with individual words or
expressions. Words with the same or very similar denotation can have q,uite
different connotations although these may vary from one speaker to another,
e.g. conservative will have different connotative presuppositions depending
on your political leanings.
8. Various grammatical phenomena, known as presupposition triggers , are
known to spark off particular presuppositions. On this basis, complement-
taking verbs can be classified asfactive, nonfactive or counterfactive. Other
grammatical phenomena of interest are inchoative verbs, implicative
predicates and cleft constructions.
9. Implicatures were first identified and studied by H. Paul Grice. He observed
a distinction between natural meaning (reflecting denotation) and non-
natural meaning ('meaning-nn', reflecting the speaker's communicative
intentions). It is by figuring out the implicatures of an utterance that we can
access the meaning-nn. Implicatures are inferences that may be either
conventional or conversational: in the former case, they are coded in the
language system; in the latter, they have to be worked out through a
 I //////// I/// On referring , saying , and implicating : Reference and inference 117

combination of inferential processes and consideration of the context of

utterance.

0. Conversational implicatures are processed against the background

assumption that speakers and hearers observe the Cooperative Principle and
its associated four maxims of Quantity , Quality, Relation and Manner. A
conversational implicature arises when a speaker flouts at least one of those
maxims: the hearer notes that the speaker is apparently being blatantly
uncooperative and the refore infers that she must mean something different.
In the case of irony, she may even mean the very opposite of what she has
said. Not all deviations from the Cooperative Principle and the maxims
involve flouting in this technical sense. Other forms of non-observance
involve violating, opting out, infringing or suspending the maxims, but
none of these gives rise to a conversational implicature.

11. Conventional implicatures, in being tied into the language system, are like
presuppositions and are ultimately indistinguishable from them.
Conversational implicatures differ from them in six ways: they are cancelable
(or defeasible) - the speaker can deny the indirect meaning was intended);
they are nondetachable in being associated with the meaning rather than the
specific words chosen; they are calculable (the hearer can work out what's
really meant); they are nonconventional (obviously); they are reinforceable
(the speaker can go on to make the intended meaning explicit); and they are
universal (they apply across languages).

12. A further distinction is drawn between generalized and particulariz ed

conversational implicatures (GCis and PCis). GCis do not need a special
context or 'scenario' to apply, and for this reason can sometimes be difficult
to distinguish from conventional implicatures. The criteria listed in point 11
can be used as tests, however.

13. Recent treatments of GCis and related phenomena, notably by Lawrence R.

Horn and Stephen C. Levinson, are described as neo-Gricean. These deal
with implicational scales and lexical gaps as well as putting forward
proposals to replace the maxims with more general principles of
communication. Horn differentiates between a Q principle and an R principle
which, in finding a balance between maximizing and minimizing what you
say , together clarify the interpretation of scalar expressions. Levinson
makes a three-way distinction: his Q and M principles line up q,uite well with
Horn's Q principle (with the M-principle reviving Grice's maxim of manner)
while his I-principle aligns with Horn's R principle. Levinson focuses on GCis,
which leads him to propose a treatment of utterance types, thereby providing
a theory of preferred interpretation.
118 On referring , saying , and implicating : Reference and inference / / / / ///////// / /

SELF-EVALUATION QUESTIONS
Choose the most appropriate answer for each q,uestion.
1. The phenomenon of reference alludes to ...
a) the denotation of words or expressions.
b) the relationship between words and specific entities created and mentioned
by speakers.
c) the conceptual prototype of words or expressions.

2. The fact that in the utterance "The pneumonia in Ward 25 needs his antibiotics"
(said by a nurse in a hospital) the pneumonia refers to the patient in Ward 25,
shows that ...
a) the process of referring is a pragmatic one.
b) nurses sometimes use incorrect language.
c) referential expressions are always noun phrases.

3. What type of referent is the first his (in bold) in the following sentence?
"Even with the backing of his coalition partners in the New Centre
Right (NCD) [. .. J Mr Renzi could not have passed his electoral reform
without Mr Berlusconi's support" (From The Economist, January 24-30,
2015, p. 21).
a) Anaphoric
b) Cataphoric
c) Bridging

4. What type of referent is Mr Berlusconi in the same sentence (in 3 above)?

a) Esphoric
b) Ideophoric
c) Homophoric

5. Deixis is a feature of language that ...

a) is organized in an egocentric manner.
b) is always accompanied by physical pointing or gestures.
c) is only used in spoken language.
/ /I/ I I I/// I I I/ On referring, saying, and implicating: Reference and inference 119

6. In the well- and world-known utterance "This is CNN", used by the reporters
working on the CNN TV channel, the demonstrative This is an instance of ...
a) person deixis.
b) discourse deixis.
c) place deixis.
7. As we have seen, in English (and also in Spanish and other languages) some
personal pronouns, as well as the present or past tenses, derive from, or are
historically correlated with spatial deictics. This is an instance of the
phenomenon called ...
a) grammaticalization.
b) deictic center.
c) spatial correlation.
8. Inference is ...
a) an indirect linguistic utterance.
b) a cognitive phenomenon that allows us to work out implicit meanings.
c) an expression that triggers different and various indirect meanings.
9. The utterance "I wasn't aware of the big mistake I was making" presupposes
that the speaker made a mistake some time in the past. What type of
presupposition is this?
a) Factive
b) Existential
c) Connotation
10. Counterfactive predicates such as the one found in "She imagined that she
had married the Prince" trigger the presupposition that ...
a) the complement clause is true.
b) the Prince does not exist.
c) the complement clause is untrue.
11. Within Grice's theoretical framework, the meanings triggered by
conversational implicatures are ...
a) coded by the linguistic system.
b) not related to the Maxims of the Cooperative Principle.
c) not coded by the linguistic system.
120 On referring, saying , and implicating: Reference and inference / /// / / / // / / / ///

12. What maxim of the Cooperative Principle is B flouting in the following

exchange?
A: How about going to the beach this afternoon?

B: The Weather Channel said it's going to rain this afternoon.

a) Relevance
b) Quantity
c) Quality

13. What type of non-observance of the Quantity maxim of the Cooperative

Principle can be seen in B's response, (where he is the accused and is called
to testify in his own trial in the U.S.)?
A: Mr. B, where were you the day the victim was murdered?
B: I take the Fifth Amendment and so will not give an answer that might
incriminate me.
a) Violation
b) Opting out
c) Infringement

14. What kind of implicature is the following, generated by Rick's utterance?

Rick: I didn't know where the museurn was, so I asked a man who was at the
corner of the street for directions.
Implicature Rick does not know the man who was at the corner of the
street (Or: The man at the corner was a stranger for Rick)
a) Generalized conversational implicature
b) Particularized conversational implicature
c) Conventional implicature

15. Levinson's (2000, 2007) neo-Gricean approach to meaning is mainly a theory

of ...
a) utterance-token meaning.
b) entailment.
c) GCis as preferred interpretation.
/ II// I I I I I/ I I l / On referring , saying, and implicating: Reference and inference 121

DEVELOPING OUR PRAGMATIC COMPETENCE:

EXERCISES FOR REFLECTION AND AMPLIFICATION
(CHAPTER 3)

A) Indicate whether the referents in bold are specific or generic, definite

or indefinite, explicit or inexplicit. Where applicable, specify if it is
also a case of deferred reference.
a) The woman who called yesterday is here now.
b) She is a tough cookie.
c) At the corner there was a man in a blue raincoat.
d) May I borrow your Shakespeare?
e) Women are not allowed to drive in this country.
f) The ham sandwich wants the check (said by a waiter to another waiter
in a restaurant).

B) Indicate to which type of reference ( taking into account the direction

in which they refer in the text) the referents in bold belong in the
following texts (First lines of novels):
a) He was an old man who fished alone in a skiff in the Gulf Stream and he
had gone eighty-four days now without taking a fish. In the first forty days
a boy had been with him. - Ernest Hemingway, The Old Man and the
Sea (1952)
b) Having placed in my mouth sufficient bread for three minutes' chewing,
I withdrew my powers of sensual perception and retired into the privacy
of my mind, my eyes and face assuming a vacant and preoccupied
expression. - Flann O'Brien, At Swim-Two-Birds (1939).
c) I was born in the Year 1632, in the City of York, of a good Family, tho' not
of that Country, my Father being a Foreigner of Bremen, who settled
first at Hull; He got a good Estate by Merchandise, and leaving off his
Trade, lived afterward at York, from whence he had married my Mother,
whose Relations were named Robinson, a very good Family in that
Country, and from whom I was called Robinson Kreutznaer; but by the
usual Corruption of Words in England, we are now called, nay we call
ourselves, and write our Name Crusoe, and so my Companions always
call'd me. - Daniel Defoe, Robinson Crusoe (1719).
122 On referring, saying, and implicating: Reference and inference ///// // / //// /,

C) Analyze the deictic expressions found in the following texts. Say what
type of deixis they exhibit (Person, Place, Time, Social or Discourse.
// Gestural, symbolic or non-deictic?)
a)
I have decided to
go on a diet
Umm ...
goodluck
with that ...

b)

c)
I don't like that tie,
Mr. Medvedev...
 / ////I//////// On referring , saying , and implicating: Reference and inference 123

d)
Checkmate, Your
Majesty!

D) What kind of inferences are those marked in bold in the following

utterances?
a) All dolphins are mammals. Flipper is a dolphin, therefore Flipper is a
mammal
b) She only learnt about her friend's illness only a week before he died.
( Her friend was ill and then died)
c) She saw a boy in the street who called her attention. ( She didn't know
the boy)
d) Even though he is a man, he can express his emotions. ( Men cannot
express their emotions)
e) A: Would you like me to take you to the movies?
B: I'm a grown-up girl now, Dad. I don't want you to take me to
the movies, because I can go alone and don't need to be taken
anywhere)

E) What are the presuppositions triggered by the expressions in bold,

and what type are they?
a) I'm so glad that you finished your PhD Thesis!!
b) My son was born at the Holy Cross Hospital in Maryland.
c) Yussef pretended he was a rich man.
d) I stopped smoking three years ago.
e) When I met her, I immediately fell in love with her.
f) He was accused of arson.
g) All her children are very intelligent
h) Linda was pregnant with Cynthia in 2005.
i) Jane is now at the school swimming-pool.
124 On referring , saying, and implicating : Reference and inference //// / / /// ///// /

F) Analyze the utterances in bold by saying which Maxim(s) of Grice's

Cooperative Principle they are flouting, and which conversational
implicature they trigger.

1) A: Do you like this country?

B: I like its people very much.
2) A: Shall we go out for a walk?
B: It's very windy today.
3) (B didn't want her rrwther to know
abou t the surprise party they
were preparing for her)
A: I put my foot in it and told
Mom about the party
B: My goodness!! How clever of
you!!!
4) A: So you finally met your in-laws.
What are they like?
B: Well, they are not of the
nicest variety .. .
5)Mother: Have you done your
homework?
Daughter: Nicholas brought
candy to class today.

G) Indicate what type of non-observance of a maxim is found in tile

following exchanges/situations:
a) A: Will you tell me the secret Julia told you?
B: Of course not. She is my friend and I would never do such a thing.
b) A: Will you tell me the secret Julia told you?
B: It's a gorgeous day today, isn't it?

c) (Johnny was punished at school for misbehaving during the break (not in
class) this morning, and has not told his Mom about it)
Johnny's mother: Johnny, did you behave well in class today?
Johnny: Yes, Mom, I did behave well in class.

d) Situation: When a writer or oral narrator tells fictional stories and

therefore the Quality maxim is not observed, because everyone knows and
accepts that the stories told are not real.
 / ////I/////// On referring, saying , and implicating : Reference and inference 125

e) This is a real situation that Laura experienced in Mexico when once talking
with an Austrian Professor who did not have a good command of the
Spanish language:
Laura: 6Que tal Herta? Veo q,ue disfrutas de tu estadia en Mexico (Hi
Herta. I see you're enjoying your stay in Mexico)
Herta: Si, me gustan mucho las frutas (Yes, I like fruit very much. (With
a heavy German accent, and obviously mistaking the verb
'disfrutar' for the noun 'frutas')).

8) Search the web for texts in which both conventional and conversational
implicatures are generated, and explain the distinctive properties of
each one of them (cancelabilitg, nondetachabilitg, etc.) for the
particular examples you find.

Search the web for any kind of interview, and find instances in which
both particularized and generalized conversational implicatures are
triggered. Compare them and justify your _a ssigning them to one
category or the other.

J) Taking into account Horn's neo-Gricean approach, sag whether the

inferences in bold are Q inferences or R inferences:
a. (At a restaurant)
Waiter: Good morning. (A table for) how many people?
Customer: Oh, yes; it's five of us, please. ( In the customer's group
there are no more than five people)
b. I got on a boat that was full of dead fish. ( The boat did not belong to
the speaker)
c. That's my mother's husband. mother's husband is not mg father)

K) Complete the table with the type of meaning that, according to

Levinson's neo-Gricean approach, would be associated with each type
of inference:

Syllogism

Generalized conversational implicature

Particularized conversational implicature

126 On referring, saying, and implicating: Reference and inference /////////I I II I/

L) Specify the kind of implicature (I, Mor Q) in each case, taking into
account Levinson's approach:
a) She invited some of her friends to the Not all of her friends
were invited to the party)
b) Peter opened the door and kissed his wife ( Peter first opened the
door and then kissed his wife)
c) Sam's car ( The car Sam owns and drives.)
d) Peter and his wife both bought a new car ( Each one of them bought
a different car).

M) Analyze the following fragment from the first act of The Importance
of Being Earnest by Oscar Wilde, by looking at how the author handles
reference, deixis, and the different kinds of inference to create
different effects and fulfill different functions, such as amusing the
audience.

FIRST ACT
SCENE: Morning-room in Algernon's flat in Half-Moon Street. The room is
luxuriously and artistically furnished. The sound of a piano is heard in the
adjoining room. LANE is arranging afternoon tea on the table, and after the
music has ceased, ALGERNON enters.
ALGERNON: Did you hear what I was playing, Lane?
LANE: I didn't think it polite to listen, sir.
ALGERNON: I'm sorry for that, for your sake. I don't play accurately - any one
can play accurately - but I play with wonderful expression. As far as the
piano is concerned, sentiment is my forte . I keep science for Llfe.
LANE: Yes, sir.
ALGERNON: And, speaking of the science of Life, have you got the cucumber
sandwiches cut for Lady Bracknell?
LANE: Yes, sir. (Hands them on a salver)
ALGERNON: (inspects them, takes two, and sits down on the sofa) Oh! ... by the
way, Lane, I see from your book that on Thursday night, when Lord
Shoreman and Mr. Worthing were dining with me, eight bottles of
champagne are entered as having been consumed.
LANE: Yes, sir; eight bottles and a pint.
ALGERNON: Why is it that at a bachelor's establishment the servants invariably
drink the champagne? I ask merely for information.
I I I I/// / I I I/ On referring , saying, and implicating : Reference and inference 127

LANE: I attribute it to the superior q,uality of the wine, sir. I have often
observed that in married households the champagne is rarely of a first-
rate brand.
ALGERNON: Good heavens! Is marriage so demoralising as that?

LANE: I believe it IS a very pleasant state, sir. I have had very little experience
of it myself up to the present. I have only been married once. That was in
conseq,uence of a misunderstanding between myself and a young person.
ALGERNON: (languidly) I don't know that I am much interested in your family
life, Lane.
LANE: No, sir; it is not a very interesting subject. I never think of it myself.
ALGERNON: Very natural, I am sure. That will do, Lane, thank you.
LANE: Thank you, sir.
Lane goes out.
 POLITENESS
AND IMPOLITENESS:
HOW TO VERBALLY
RECONCILE OUR
INTERLOCUTORS' WANTS
WITH OUR OWN (OR NOT)
/////////////////////////////////////

MAIN OBJECTIVE OF THE CHAPTER: To define and explain the concepts

of politeness and impoliteness in pragmatics by presenting the major
theories developed in recent decades, summarizing the criticisms they
:ve provoked and outlining the issues currently under research.

MAIN TOPICS OF THE CHAPTER:

• Politeness
• The conversational-maxim view of politeness
• Leech's Politeness Principle
• Negative and positive face
• Brown & Levinson's Politeness Theory
• Culpeper's approach to impoliteness
• Kaul de Marlangeon's approach to impoliteness
130 Politeness and Impoliteness / / / / / / / / / / / ///////I I I I I I I I/ I I I I I I I I I I I I I I I I I

4.1. POLITENESS AS AN IMPORTANT ELEMENT FOR THE

DEVELOPMENT OF PRAGMATIC COMPETENCE
I normally expect that my interlocutor is not going to oblige me to do something I
do not want to do, or that s/he is not going to insult me. Moreover, I will expect him
or her to behave in a rational way and to be tactful, and I may also assume that s/he
is going to be kind to me. All of this has to do with rules and strategies of social
and personal interaction that linguists have studied under the name of Politeness.
However, these expectations are sometimes not fulfilled, for people may opt
to be impolite, simply because they feel unbalanced, annoyed or attacked, or
because the particular discourse situation req,uires the speakers to disregard
their interlocutors' wishes and purposefully take them out of their 'comfort zone .
Examples of this latter option are found in the discourse of some political debates
or some reality shows on T.V., where one of the main functions of the ongoing
communication is to discredit the opposition or to exalt rivalry and 'bad manners'
for purposes of entertainment. In these cases, the other side of the politeness coin
is shown, but it cannot be said that the interactants are not following any rules of
social interaction. They are following certain rules and using special strategies.
namely rules and strategies of impoliteness, which constitutes the other pole of
a continuum showing interactive behavior ranging from 'most polite' to 'most
impolite'. In this chapter we are going to examine both politeness and impolitene
strategies as illumined by the main theoretical approaches to the description of
both phenomena.
Politeness has a lot to do with putting ourselves into our interlocutor's shoe .
i.e., with being empathetic and trying to find the most appropriate words that will
show generosity, tact and concern regarding his wants and feelings. Therefore, it
is clear that not only the social and cultural contexts of utterances are very
important for the determination of the strategies to use, but also, and crucially.
their emotional context. 39 Compare (1) with (2), where in (1) the father is in a
good mood, and in (2) the father is, on the contrary, in a bad temper, having just
scolded Tim's sister for misspending some money he had given her:
(1) Tim: Dad, give me some money for the bus, please.
(2) Tim: Dad, I don't want to bother you, but I have to go to the library now and
I need some money for the bus, but I can also walk there, no problem.
if necessary.
It is evident that Tim is much more careful with his words in (2) than in (1).
because while in (1) Tim is making a normal 'default' req,uest (because his father

39
See Chapter 1 for the different types of context.
I I I I I I I I I I I I I I I I/ I I I I I I I I I I I I I I I I I I I/ I I I I I I I Politeness and Impoliteness 131

is in a positive emotional state), in (2) he is taking into account his father's

feelings and negative emotional state, and therefore his discourse is shaped
accordingly, by using hedges(/ don't want to bother you but ... ) and by trying to
minimize the imposition of the req,uest as much as possible ( ... but I can also
walk there if necessary ... ).
As we shall see throughout this chapter, the phenomenon of (im)politeness
is very much related to the Cooperative Principle and its maxims, 40 for it has to
do with what we human beings expect and consider to be rational or 'civilized'
behavior. And in spite of the fact that (im)politeness as a general phenomenon
has been claimed to be universal by many linguists, these expectations will vary
ccording to the different cultures, communities, social groups or discourse
stems involved in the interaction: The strategies of politeness used by a
sian speaker to make a req,uest may be different from those used by an
glish speaker within the same situation, but even within the same culture or
guage, different groups or even different families may have different rules of
liteness to function within their inner circles.
From all the above considerations it is easily deduced that the phenomenon
(im)politeness is essentially a pragmatic phenomenon, for it would be
ssible to even conceive of it without taking into account the people and all
variables of the context around the language used. This is the reason why the
pie occupies a whole chapter in this book, as well as being why it is one of
key elements to be included in any foreign language syllabus: it is unthinkable
achieve a high command of a language without learning to handle its
)politeness strategies and rules.
Before turning to an examination of the main and most influential linguistic
roaches to the phenomenon of politeness, we want to make clear that when
speak of politeness in linguistics we are not speaking of just social 'good
ers': Linguistic politeness involves the management and handling of many
d various discourse-pragmatic variables, which may make the difference
een what is considered appropriate and coherent language and what is not.
shall now look into as many of these variables as possible, in order to
eate and illustrate this very important aspect of human language .

.2. APPROACHES TO POLITENESS

e topic of politeness in linguistics started to gain considerable importance in
the 1970s. Since then, there have been many attempts to define and describe the

-m See Chapter 3 for Grice's Cooperative Principle and the maxims of Qualitl), Quantitl),
Relation and Manner.
132 Politeness and Impoliteness / / / / / / / / / / / / / / / / / / / / / / / / // / / / / / / / / / / / / / / / / / /

phenomenon, the most prominent and influential one being Penelope Brown and
Stephen Levinson's ([1978] 1987), which we shall examine in more detail in 4.2.2.
Green (1989: 141) defines politeness as "strategies for maintaining or
changing interpersonal relations". Grundy (2008: 187) explains the phenomenon
in the following way:
We each have expectations as to how we should be addressed by the various
people we meet in the various contexts in which we meet them. The study of
politeness is the study of the ways in which these expectations are met (or not,
as the case may be).
Leech (1983: 104) introduces the Principle of Politeness as "an important
missing link between the Cooperative Principle and the problem of how to relate
sense to force" , and proposes a number of maxims dealing with polite behavior
that we shall examine in 4.2.1.
Other authors, such as Watts (1989, 2003), distinguish betweenfirst-order
and second-order politeness, the former referring to the everyday sense of the
term and the ways in which it is evaluated and commented on by lay members of
society, and the latter referring to the abstract characterizatio~ or scientific
concept which transcends the everyday notion of (im)polite behavior. This
distinction was later relabeled by Eelen (1999, 2001) as politeness 1 andpoliteness1
respectively. Watts rejects politeness2 on the grounds that we should investigate
politeness as an instance of real social interaction and behavior and not as a
theoretical abstract concept. Thus - according to Watts - research on politeness
should deal with "the reproduction and reconstruction of the values of socially
acceptable and socially unacceptable behaviour" (i.e. with politeness 1).
The main four approaches to (im)politeness identified at the end of the 20m
century are outlined by Fraser (1990) as follows:
1) the social-norm view,
2) the conversational-contract view,
3) the conversational-maxim view,
4) theface-saving view.
To these we should add other approaches such as:
5) the emotive perspective taken by Arndt & Janney (1979, 1983) and also
Janney & Arndt (1992),
6) the relational perspective developed in works such as Locher (2004
2006), Locher & Watts (2005), or Culpeper (2008), which emanates from
Watts's (2003) postmodern approach,
7) the frame approach supported by authors like Aijmer (1996), Terkourafi
(1999, 2001a, 2005) or Deutschmann (2003),
8) the rapport management view (Spencer-Oatey 2000).
IIIIII II III III IIIII II IIIIIIIIII IIIIIII IIIIII Politeness and Impoliteness 133

The social-norm view is associated with the understanding of politeness as

good manners". Authors such as Jespersen (1965) or Quirk et al. (1985) can be said
t-0 hold this view, as when Quirk et al. ( 1985: 188) state that it is more polite to use
the past tense in I wondered if you could help us or that Could I see your driving
Licence? is politer than Can I ... (1985: 233). More recently the concept of politeness
has evolved into a much more complex construct, and so nowadays this perspective
cannot be said to be supported by the majority of researchers and scholars.
The conversational-contract view is the perspective adopted by Fraser
(1990) and Fraser & Nolen (1981). These authors acknowledge the importance of
Grice's Cooperative Principle and Goffman's (1967) notion of face, but in contrast
with other theories of politeness recognizing these notions, they put forward the
idea that all the participants in an interaction enter into a conversation and
maintain it in agreement with a current conversational contract at every turn.
Fraser (1990: 233) explains that "being polite constitutes operating within the
then-current terms and conditions of the Conversational Contract", the contract
thus having a dynamic q,uality, which has to be adjusted in accordance with the
unfolding conditions of the ongoing conversation.
The conversational-maxim view is the one based on Grice's Cooperative
Principle and its maxims, and among the most important authors subscribing to
this approach we can name Robin Lakoff (1973), Geoffrey Leech (1983, 2014) and
Yueguo Gu (1990, 1997).
The face-saving view is the most renowned of the perspectives on politeness,
and it was presented by Brown & Levinson in their book Politeness: Some
Universals in Language Use ([ 1978] 1987), which is the most widely cited work
on the subject.
Both the conversational-maxim and the face-saving views have been very
influential, and therefore we shall deal with them in more detail in 4.2.1, 4.2.2
and4.2.3.
Arndt & Janney's (1979, 1983, 1985) perspective presents a general model
of emotive interactional behavior. Instead of focusing on the linguistic channel of
communication between individuals, they put emphasis on their holistic behavior
and emotions, thereby developing the notion of interpersonal supportiveness,
which not only affects language, but also paralinguistic and kinesic features such
as gesture, intonation, laughter, etc.
The relational perspective is a postmodernist, discoursal approach to
(im)politeness. Its founding father is Richard Watts, who bases his views on
human interaction on Bourdieu's work (e.g. 1977) and his concepts of practice
theory and habitus. Within this perspective, politeness is viewed as relational
work, a term which Watts (2003: 277) defines as "Efforts made by the participants
in verbal interaction to be as considerate towards one another as possible".
134 Politeness and Impoliteness I I I I I I I I/ I I/ I I I I I! I I I I I I I I I I I I I I I I I I I I I I I I I I

Since this is a perspective that became prominent in the first years of the 2ist
century and is thus now in vogue in certain linguistic circles, we are going to
devote some more space to it later in this chapter (4.3.3.), but for the time being
let us summarize it by saying that the postmodern perspective does not believe in
theoretical conceptions or in any attempts to objectify the phenomenon of
politeness. Leech (2014: 43) remarks that, in contrast to approaches such as
Brown & Levinson's or his own, Watts's standpoint does not view politeness as a
means for maintaining face and social concord, but as "a tool for maintaining
hegemony in the hands of the powerful".
Theframe approach defines politeness in terms of "culture-specific ready-
made patterns" (Terkourafi 1999: 107). The concept of fram e had already been
researched within and across different disciplines (e.g. Minsky 1975 and Schank
& Abelson 1977 in Artificial Intelligence, or Fillmore 1976, 1977b, Goffman 1974
and Tannen 1993 in linguistics), and even though the concept is not treated in the
same way in all of them, there is agreement as to the existence of stereotypical
pieces of knowledge (stored in human beings' memory and easily retrievable)
which are acq,uired as the result of their interactional experience with other
people and with the environment. An analysis of politeness within frame analysis
will then look at conventionalized forms of polite behavior and its stereotyped
features, as well as the formulaic patterning of language associated with them.
Other authors supporting this approach are Aijmer (1996), who studies frames
with reference to certain speech acts, such as thanks, req,uests or apologies in her
book about conversational routines in English, and Deutschmann (2003), who
studies the prototypical apology as occurring within a frame which contains four
explicit or implicit elements, namely an Offender, an Offended, an Offense and a
Remedy. As Leech (2014: 39) notes, the concept of frame is particularly useful for
routinized, conventionalized polite behavior, but it disregards the fact that "being
polite means using reasoning and imagination, not just memory".
The rapport management view was put forward by Spencer-Oatey (2000).
This is a sociocultural perspective that places politeness within the broader
concept of rapport management, alluding to the management of interpersonal
relations, which has two main components: management offace and management
of sociality rights. Spencer-Oatey takes her definition of face from Goffman
(1972), 41 and although she adopts Brown & Levinson's distinction between
positive and negative face (see 4.2.2), she views these notions in a different light
for her, the former concerns face, while the latter is not concerned with face but
with sociality rights or a desire for autonomy. In Spencer-Oatey's first model
(2000: 15),face management includes two kinds of face, namely 1) quality f ace

41
Goffman (1972: 5) defines face as "the positive social value a person effectively claims
for himself by the line others assume he has taken during a particular contact".
' I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I/ Pol iteness and Impoliteness 135

(i.e. the value we claim for ourselves in terms of personal q,ualities, which is
closely related to self-esteem) and 2) identity face (i.e. the value we claim for
ourselves in terms of social roles, such as parent, boss, teacher, etc.). Within the
management of sociality rights this author includes two kinds of rights: 1) equity
rights (i.e. the entitlements we feel we should have in terms of personal
consideration and fair treatment from others) and 2) association rights (i.e. the
entitlements we feel we should have in terms of relations with other people, such
as friendship or respect). In a later publication (2008), Spencer-Oatey no longer
makes the distinction between quality face and identity face but characterizes
f ace as involving three levels of representation: 1) the personal level, 2) the
relational level, and 3) the group level. It is interesting to note here that, as in
Arndt and Janney's model, the emotive context is taken into account, in this case
as a crucial factor for the distinction of these levels, for each is associated with
particular sensitivities. This means that it is possible to threaten a person's face
by underrating or degrading their person, their circle of friends, or their
:nationality, just to give one example of each of the three 'sensitive' levels.

In this section we have included the main approaches to politeness that have
emerged to date and the key points that they make, but the authors included here
are only some of the many scholars that have written and theorized about this
pragmatic phenomenon, and therefore our account of it here is far from complete.
Some of these authors and models of politeness deserve special attention because
they have had the greatest impact within linguistics and across other disciplines,
and this is the reason why in the following sections we shall give a more
exhaustive description and account of them.

A final reflection that we would like the reader to consider in this section is
that, whatever the approach taken, there are two things that we know for sure
about politeness. One is that politeness is not an innate phenomenon, i.e. we have
o acq,uire it by learning different interpersonal strategies and patterns of
behavior, and for this reason it is an element of language and human behavior
without the consideration and inclusion of which the design of a language
sgllabus would be totally incomplete, hence its importance in language teaching/
learning. The other thing we know for sure is that, as we anticipated in 4.1 ,
politeness phenomena constitute an archetypical example of pragmatic usage,
and cannot be disregarded in any study of the nature of human language .

.2.1. THE CONVERSATIONAL MAXIM VIEW : LAKOFF ,

LEECH AND GU
.\8 was noted above, the conversational-maxim view of politeness has been one
of the two most influential perspectives on politeness. The authors supporting
136 Politeness and Impoliteness //////////// //////// ///// ///// //// ////// ////

this view take Grice's Cooperative Principle (CP) and its maxims 42 as the
foundation for an account of politeness phenomena in assuming that speakers
normally observe the CP and its maxims when being polite, but even if they do
not observe them, they can still be following certain strategies of politeness. As
we saw in Chapter 3, sometimes people do not observe the maxims of the CP and
flout them purposefully in order to signal the working of conversational
implicatures, and this non-observance of the maxims may often be generated by
a desire to be (im)polite in some way or another.

4.2.1.1 Robin Lakoff

Robin Lakoff is one of the authors supporting the conversational maxim/rule
view of politeness, and as early as in 1973 she proposed the following two Rules
of Pragmatic Competence (1973: 296):
1) Be clear
2) Be polite

These rules may reinforce each other, but at times may come into conflict
with each other: According to Lakoff, if a speaker has to choose between being
clear and being polite, politeness will supersede, because it is more important to
avoid conflict than to achieve clarity. She divides the second rule into three Rules
of Politeness (1973: 300):
1) Don't Impose
2) Give Options
3) Make A43 feel Good - Be friendly

Lakoff explains that each of these rules req,uires a different type of

politeness. Thus Formal Politeness is req,uired if we want to follow Rule 1
Informal Politeness is req,uired for Rule 2, and Intimate Politeness for Rule 3.
Examples (3), (4) and (5) illustrate these three options, respectively , for the
speech act of requesting:

(3) Excuse me for bothering you, but would you mind if I used your telephone?
(4) May I use your telephone, please?
(5) Hey, I need to use your phone. Pass it to me, darling.

42
See Chapter 3.
43
By "A" here Lakoff refers to the hearer.
II I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I Politeness and Impoliteness 137

As is obvious, (3) req,uires a more formal formulation of the req,uest because

it is obeying the rule of non-imposition, which the speaker may deem necessary
for different reasons, such as distance between her and her interlocutor, the
special mood the hearer is in, or the situation and context surrounding both. In
(4) the speaker is less formal than in (3), but more than in (5), so it exemplifies an
intermediate case, where the speaker deems the situation to req,uire Informal
Politeness while at the same time giving her interlocutor the option of choosing
hether he wants to comply with the req,uest or not. In (5), in contrast with the
other two examples, the speaker shows that she has an intimate relationship with
e hearer, and therefore does not need to resort to any formalities. On the
contrary, the direct req,uest is expected to be taken by the hearer as a sign of
friendliness, closeness and even love on the part of the speaker. As Lakoff puts
- the speaker's intention is to make the hearer "feel good" by showing him that
eir relationship is one of intimacy. This type of politeness would fall within
,,..-hat Brown & Levinson (B&L) later called positive politeness, while the other
o types would fall within B&L's negative politeness (See 4.2.2.).

In sum, Lakoff's model of politeness presents a rational system of rules that

· relative to the different levels of formality that hold in human relationships .
.And even if at the time it was influential within the field, her approach has been
·ticized, as Fraser (1990) notes, for not being clear about, for instance, how the
aker or hearer is to assess the level of politeness that is req,uired for each
ticular situation .

.2.1.2 Geoffrey Leech's Politeness Principle (PP)

Geoffrey Leech has been one of the pioneering authors in the field of politeness.
He views the phenomenon as dependent on Grice's Cooperative Principle (CP)
d therefore proposed a Tact Maxim (1977, 1983) and a Politeness Principle
P) as complementary to the CP. The PP presents an idea of politeness in which,
other things being eq,ual, people prefer to be polite rather than impolite. In the
same fashion as the CP, and unlike grammar rules, 44 the PP can be observed,
ached, suspended or flouted, and can also be classified into maxims, which in
ech's model are the following (1983: 132):

According to Leech (2014: 34) pragmatics is principle-governed (unlike grammar,

which is rule-governed) because he assumes that rules are constitutive in force, while
principles are regulative. He takes this distinction from Searle (1969: 33), for whom
regulative rules "regulate antecedently or independently existing forms of behaviour"
(e.g. rules of etiq,uette), and constitutive rules "do not merely regulate, they create or
define new forms of behaviour" (e.g. rules of chess or football).
138 Politeness and Impoliteness ///// /////// //////// //////////////////// ////

(I) TACT MAXIM (in impositives and commissives ):

(a) Minimize cost to other [(b) Maximize benefit to other]
(II) GENEROSITY MAXIM (in impositives and commissives):
(a) Minimize benefit to self [(b) Maximize cost to self]
(III) APPROBATION MAXIM (in expressives and assertives):
(a) Minimize dispraise of other [(b) Maximize praise of other]
(IV) MODESTY MAXIM (in expressives and assertives ):
(a) Minimize praise of self [(b) Maximize dispraise of self]

(V) AGREEMENT MAXIM (in assertives):

(a) Minimize disagreement between self and other
[(b) Maximize agreement between self and other]
(VI) SYMPATHY MAXIM (in assertives ):
(a) Minimize antipathy between self and other
[(b) Maximize sympathy between self and other]

Leech makes it clear that, in spite of their being formulated in the imperative
mood, his intention is not to present the maxims as prescriptive; on the contrary
he claims they are "descriptive of what happens in communication by default"
(2014: 35).
One aspect that has been controversial in Leech's (1983) model is that it
distinguishes between Relative Politeness and Absolute Politeness, the former
referring to the politeness which varies according to specific situations and the
latter to the degree of politeness which in his view is normally associated with
specific actions. Thus for Leech some illocutionary acts, such as offers, are
inherently polite and others, such as orders, are inherently impolite. This has
been criticized by some authors, such as Fraser (1990), who points out, for
instance, that an order does not necessarily have to be impolite, for one can
always find situations in which an order would be a very positive and polite
speech act (e.g. if a teacher ordered a student to present his excellent results on
a test or a project to the class, this would be interpreted as polite on the part of
the teacher, as an attempt to make her student 'feel good'). So taking into account
this and other criticisms leveled at his model ( e.g. that it has an unnecessarily
large number of maxims, or the different status of the PP as compared with the
CP), Leech presented a revised version of the model in what - sadly - turned ou
to be the last book he published before his unexpected death (in August 2014):
The Pragmatics of Politeness (2014).
 IIIII IIIII IIIIIII III IIIIIII III II III II IIIIII Politeness and Impoliteness 139

In the revised version of his model of politeness, Leech responds to all the
criticisms directed at his thirty-one-year-old model of 1983, and presents a more
detailed and refined account of his PP, which he now reformulates as the General
ategy of Politeness (GSP), in response to the criticism (by Brown & Levinson
1987 and others) about the "expansionist" approach of his PP, which in their
opinion introduced too many maxims (in contrast with, for instance, the
reductionist approach of Relevance Theory 45 ). The GSP is a "single
erconstraint" (2014: 90) which includes all the maxims previously formulated
· the 1983 model, which Leech defines in the following way (where O refers to
e hearer or addressee, but may also apply to a third person):

General Strategy of Politeness: In order to be polite, S expresses or implies

meanings that associate a favourable value with what pertains to O or
associates an unfavourable value with what pertains to S (S= self, speaker).
(2014: 90)

Before presenting his rearrangement of the maxims, it is important to note

re that Leech distinguishes between pas-politeness and neg-politeness, which,
cording to Leech (2014: 89) and as we shall see in 4.2.2, "are not q,uite the same
what B&L understand by positive politeness and negative politeness". For
Leech, pas-politeness is the kind of politeness involved in, for instance, praising
meone, whose purpose is to place a high value on O's q,ualities, and neg-
liteness is the kind of politeness used for instance in making a req,uest, which
a negative purpose because it is intended to avoid offense by mitigating the
· position on the hearer. Both kinds of politeness are illustrated in (6), where A's
req,uest is an example of neg-politeness, and it is therefore minimized by the
ormula Would you mind, and B's reply is an example of pas-politeness which
also illustrates the Generosity Maxim:

6) A: Would you mind if I ate one of your cookies?

B: Of course I wouldn't. Eat as many as you like.

The rearrangement of the maxims within the GSP is shown in Table 1, which
reproduces Leech's (2014: 91) own table, but in which we have added a
prototypical example (in italics) in each case, to illustrate each of the maxims.
e maxims oriented towards the hearer (0) illustrate pas-politeness, and those
·ented towards the speaker (S) illustrate neg-politeness. Here Leech has added
o constraints not present in Leech (1983): two pas-politeness Maxims of
Obligation and two neg-politeness maxims ( opinion reticence and feeling
reticence):

See Chapter 5 for Relevance Theory.

140 Politeness and Impoliteness // // // / // //// // // // / / // // / // // / / // // / / // ////

Table I: The component maxims of the General Strategy of Politeness

(with examples) ·
,. -.

E.~MPLE
,. '

A woman to her
(Ml) give a
friend: Let me pay
high value to Generosity Commissives
for this meal. It's
O'swants
my treat!
Generosity, A woman to her
Tact friend: If it would
(M2) give a
not bother you
low value to Tact Directives
too much, could I
S's wants
borrow your dress
for the party?
Son to his mother:
(M3) give a Your tiramisu
high value to Approbation Compliments is the best I have
O's q_ualities- tried in my whole
life, Mom;
Approbation,
Modesty A student to her
teacher: I'm so
(M4) give a
stupid I don't even
low value to Modesty Self-devaluation
understand how
S's q_ualities
to calculate square
roots!
An employee to his
boss: I'm terribly
sorry, but I will
(M5) give a
have to ask you to
high value to Obligation Apologizing,
leave early today
S's obligation (of Sto 0) thanking
because my wife
too
is sick and I have
Obligation to take her to the
doctor's.
A I'm very sorry
(M6) give a to have bothered
Responses to
low value to Obligation you.
thanks and
O's obligation (of O toS)
apologies B: No problem at
toS
all. It's all right.

(It continues)
//////III/IIII/I/IIII//I/II/IIII/IIIIII/II/I Politeness and Impolite ness 141

A:. Jane is a
(M7) give a wonderful teacher,
Agreeing, don't you think?
high value to Agreement
disagreeing
O's opinions B: Yeah, she's
excellent!
A woman to her
Opinion
friend: My humble
opinion is that
(MS) give a Opinion you should end
low value to Giving opinions
reticence that relationship,
S's opinions
but of course I
may be completely
wrong.
A woman to her
friend, who has
(M9) give a won a scholarship
Congratulating;
high value to Sympath11 to go abroad:
commiserating
O's feelings Congratulations!!!
I'm sooo happy for
you!!!!
Adam to his
Feeling
girlfriend, after
she bumped into a
(MIO) give a rock and fell down:
Feeling Suppressing Are you O.K.?
low value to
reticence feelings
S's feelings Eve: Yes, my ankle
hurts a little bit
but it's nothing;
don't worry.

Leech includes numerous examples and many more considerations about

this new formulation of his theory, which we encourage the reader to examine in
his 2014 book, but which we cannot include here for space reasons. However,
there are two important aspects of the restatement of his theory that we would
like to remark on by way of conclusion to this section. One is that, in order to
avoid misunderstandings, he now writes about a pragmalinguistic politeness
scale and a sociopragmatic politeness scale, in lieu of what he formerly had
called absolute and relative politeness, respectively. The former concept refers to
the fact that utterances can be ordered on a scale of politeness while keeping
142 Politeness and Impoliteness ////// ///////// ///// /// /// //////// //////////

context invariant; this scale is unidirectional and has to do with the

lexicogrammatical form and semantic interpretation of the utterance. So for
instance, on a pragmalinguistic scale, we can say that, other things being eq,ual,
(7) is less polite than (8):
(7) Give me $10.
(8) Would you be so kind as to give me $10?
The latter concept, namely the sociopragmatic scale, is a bidirectional scale
that refers to the kind of politeness that is relative to the particular norms of a
given group, society or situation, and is sensitive to context. This means that
a form that is considered very polite in a certain context may be not seen as such
in a different environment or social group. Hence, and to illustrate, the otherwise
polite form Would you please ... , as used in (9), will most surely be interpreted as
sarcastic in the context of a family conversation:
(9) Would you please let me speak for a while?
The second aspect we would like to emphasize about Leech's revision of his
theory is that it presents a very reasonable, coherent and consistent approach to
the phenomenon, which not only accounts for the English language, but for other
languages as geographically and culturally distant as Korean, Chinese and
Japanese. For that reason, he responds to the attacks on the alleged ethnocentrism
of both Brown & Levinson's and his own model (e.g. by Wierzbicka 2003), by
giving authentic examples from these and other foreign languages, and as to the
related q,uestion of whether his model is universal or not, he very sensibly replies
in the following manner:
My own position is not very different from this [Brown & Levinson's]: although
I would not press for "universal principles" [... ], I would still argue that a model
of politeness should be generalizable to various cultures, and should provide the
basis for studying (im)politeness in different languages and societies. [... )
Consider [... ] the distinction between "universalism" (the position espoused by
Brown & Levinson) and "relativism" (the position adopted by Wierzbicka). An
absolute universalist position is clearly untenable: it is obvious, from common
experience as well as from studies over the past twenty or thirty years, that
politeness manifests itself in numerous guises according to the language and
culture. On the other hand, a completely relativist position is eq,ually untenable.
If there were not a common pattern shared by many languages or cultures, it
would be meaningless to apply a word like "politeness" or "face" to those
cultures, and there would be no point, for example, in having an Internatio nal.
Journal of Politeness Research, such as now exists. (2014: 83).

Yueguo Gu, the author we will deal with in the following section, is precisely
one of the authors who has criticized Leech for not having taken Oriental
languages and cultures into consideration in his 1983 model.
 / II I I I I I/ I I I I/ I I I! I I I I I I I I I I I I I I I I I I I I I I I I I Politeness and Impoliteness 143

.2.1.3 Yueguo Gu
Yueguo Gu is another of the linguists who take a Gricean approach to politeness,
but in response to Leech's (1983) proposal of the PP he argues that it should be
modified to account for the traditions and practices of politeness in Chinese,
mething that, as we have just explained, Leech did when he restated his model
in 2014. 46 Gu (1990) proposes four politeness maxims: 1) Self-denigration ,
2) Address, 3) Tact, and 4) Generosity, as well as a Balance Principle, whose
function is to maintain social eq,uilibrium and through which a given speaker
balances the favors done to her by her interlocutor B by herself doing favors to
B. In a later publication, Gu (1997) proposes three categories for the analysis of
goal-oriented linguistic activity: a) goal development analysis, b) talk exchange
evelopment analysis, and c) interpersonal management analysis, where
politeness, as well as other aspects of rapport management (as defined by
encer-Oatey 2000, 2008), are included in the third category.
A feature that distinguishes Gu's approach from other perspectives (such as
B&L's) is that the concept of 'face' is regarded as a normative social phenomenon
(m the Confucian tradition) and not as a psychological characteristic of the
individual. Thus, Gu's work is one of several works (e.g. Ide 1982, 1989, 1993) on
politeness that have emphasized the contrast between the individualist ethos of
estern civilization and the collectivist ethos of the Eastern world .

.2.2. THE FACE-SAVING VIEW: BROWN & LEVINSON

Brown & Levinson's ([1978] 1987) approach to politeness seeks to unveil the
rational principles by which participants in interaction deploy strategies to
achieve their communicative goals in a social context. These principles do not
derive from individual cultures but are grounded in human thinking and are
therefore cross-cultural or potentially universal in nature; Brown & Levinson's
own data is taken from English, Tzeltal 47 and South Indian Tamil. They assume
that the individuals who participate in interaction, alongside their ability to use
the local language, have two major properties. The first is rationality, their
ability to reason out how to achieve their goals; the second is face, which
combines two desires ( or 'wants'): the desire to be unimpeded and the desire to
be approved of. The former is known as negative face, since it concerns the
absence of barriers to the achievement of the individual's goals, and the latter as
positive face.

In fact, Leech had presented some results on the topic some years earlier. See Leech
(2005, 2007) .
.r. A Mayan language spoken in the Mexican state of Chiapas.
144 Politeness and Impoliteness // // ///////////// ///////// //// //// //////////

Whereas Lakoff and Leech see politeness as implying rules, principles or

maxims that are additional to Grice's Cooperative Principle, Brown & Levinson
(henceforth B&L) argue that it is attention to face that motivates the appeal to
conversational implicatures and the use of politeness strategies in general. In a
dialogue with two participants A and B, A and B know of each other that they are
being cooperative, that they are rational agents and that they each have face
wants. Now, if A says something that is in conflict with the maxims of
conversation and yet is being cooperative, then B has to use reasoning to work
out what A means. That reasoning is steered by B's knowledge that A is rational,
that A has face wants and that B himself has face wants. B will then conclude that
A has appealed to a conversational implicature in order to be polite. Politeness
now emerges as a rational strategy for achieving one's desires without offending
the face of the other.

Let us consider an example:

(10) Norma: Bill, sweetie, shouldn't you put that book down now?

Here Norma produces a speech act with three components._Firstly, she

pronounces her husband's name (let's assume they are married). Rationally
Norma has no need to do this, since there's no one else in the room and Bill
knows that she knows his name. So Bill, again rationally, draws the conclusion
that this needless linguistic activity is motivated by the reason that she is
attending to his positive face: by pronouncing his name, she gives him
recognition and indicates a positive attitude to him. Secondly, she explicitly
attends to his face by using the term of endearment sweetie. Finally, she asks
him a q_uestion about his duties using the modal verb should; in addition, she
negates the verb with n't, which, if we look at it rationally, indicates that she is
not sure about what Bill's duty is in this case. In processing this q_uestion, Bill
uses his reasoning abilities to draw the conversational implicature that Norma
wants him to stop reading but also to observe that Norma has refrained from
actually saying "Stop reading!" and instead has used a combination of
interrogative syntax and a negated modal of duty. From all this it is reasonable
for him to conclude that Norma has done this to respect his negative face: she
has shown that she respects his desire for freedom of action, in this case to carry
on reading. The combination of the respect for positive face evident in the first
and second parts of Norma's speech act and the respect for negative face in the
second may, she hopes, lead Bill to stop reading and (as a perlocutionary effect)
to pay attention to her, which may have been the goal that motivated Norma
speech act in the first place.

Let us now develop some of the concepts that have been mentioned here to
give a fuller picture of B&L's approach.
IIIII IIIIIII II IIII IIII IIIIIII IIII III IIIIII II Politeness and Impoliteness 145

4.2.1.1 The concept of face

In B&L's writings,face is used as a technical term. Its sense, which derives from
the work of Goffman (1967), is close to its use in the expression to lose face,
which means to undergo damage to one's self-esteem. In our social life, our face
in this sense is very important to us and we invest a lot of emotional effort in
protecting it and trying to improve it; if we lose face, we may feel embarrassed
or even humiliated and will suffer as a conseq,uence. Since we know that face is
an important issue for everyone, we can benefit our own chances of feeling good
about ourselves by showing respect for the face of others. B&L (1987: 70) refer to
this as 'anointing' the face of the other. In general, it is in everyone's interests
to respect the face of other people; although there are cultural or group variations
here, some communities investing more effort in 'facework', as it is called, than
others, rationally speaking no society could survive without its members
attending to one another's faces.
The combination of emotional investment and rational self-interest that
underlies individuals' concern with face is translated in social interaction into
commitments or desires. These allow Brown & Levinson to define face as 'wants',
using the verb want as a noun. As mentioned above these are either negative or
positive, as shown in the following definitions, where 'member' means 'member
of a society':
Negative face: the want of every 'competent adult member' that his actions be
unimpeded by others.
Positive face: the want of every member that his wants be desirable to at least
some others. (B&L 1987: 62)

Notice that positive face is defined in terms of 'at least some others'. If a
woman wears a new dress that her lover has given her, she will want to be
admired (to have her positive face anointed) by her lover and perhaps some close
friends, but probably not by everyone.
Now, problems arise when one's face wants come into conflict with other
people's desires. Those other desires and the speech acts tµat follow from them
can be seen as threats to the maintenance of face, and such acts are therefore
known as Face-Threatening Acts (or FT.As for short). FTAs can endanger either
positive or negative face.
Any speech act that indicates that the speaker (S) is indifferent to the face
wants of the hearer (H) is a threat to H's positive face. Among many examples of
such threats to H's positive face are those in (11):
(11) a. criticisms, ridicule and insult You've burned the toast again, you
blockhead!
146 Pol iteness and Impoliteness / /// //////////////// ////////// //// /////////

b. challenge and disagreement You're totally wrong about that.

c. violent displays of emotion I'mfk****g FURIOUS with you.
d. mention of sensitive subjects Did you ever consider gastric band
surgery?
e. bad news You've failed the exam.
f. boasting I've always succeeded in everything I do.
g. bringing up divisive subjects Parked two feet from the curb, typical
woman driver!
h. interruption, inattention What was that you said just now?
i. wrong address form Hi prof, what's the assignment for next
week?
All of the examples in (11) are threats to H's face because they show a lack
of respect, of praise, or of sensitivity. This is emphatically not to say that it is
never appropriate to criticize, challenge, mention sensitive subjects, impart bad
news, etc., but that all of the FTAs call for what Brown & Levinson call redressive
action, i.e. additional linguistic material that will show Sis aware of the harm
being done to H's positive face but has to go through with the FTAanyway. This
will be dealt with in detail in 4.2.2.2 below. Suffice it to give one simple example
of redressive action here. Rather than saying (lle), an examiner could say You've
been a good student all semester, but you've failed the ex am. The examiner is
duty-bound to convey the result, but by also giving praise she anoints the
candidate's face, showing that she does not disq,ualify him as a person.
Possible threats to the hearer's negative face - given the appropriate
circumstances - are exemplified in the following set of examples:
(12) a. orders, req,uests Give up your seat on the bus to elderly
passengers.
b. suggestions, advice You should not wear that color of lipstick.
c. reminders I told you the boss needs the report by 4
o'clock, remember?
d. threats, warnings, dares I dare you to go into that field with the
bull in it.
e. offers I'd like to make you a loan of $5000.
f. promises We promise to visit you next time we're
in town.
g. compliments I really liked that article you published
last summer.
h. strong emotions about H You know, I could just kiss you all over
right now.
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I/ I I I/ I I I Politeness and Impoliteness 147

Notice that offers and promises, as in (12e) and (12f), although well meant,
may constitute face-threatening acts towards H's negative face because of the
reciprocity that is built into our social relations. If A offers or promises B
something, B usually has to make a similarly valued offer or promise back to A,
and this restricts B's freedom of action. Compliments, as in (12g), similarly call
for reactions, which even if the original compliment was meant sincerely, might
not themselves be sincere. It is also possible for an FTA to threaten H's positive
and negative face at the same time. An expression of strong affection like (12h)
is threatening to H's negative face in restricting his freedom of action but also to
his positive face in presenting him with a more emotionally charged situation
than he is comfortable with.
In all the cases of FTAs we have examined up to this point we have mainly
focused on threats to H's face. However, a speech act may also be threatening to
S's own face. The speech acts that are a threat to S's negative face usually occur
in reaction to something that H has said or done, since S has no choice but to
utter them. Here are some examples.
(13) a. thanks I am very grateful for the box of socks
you sent me. ·
b. acceptance of thanks You're welcome, please don't mention it.
c. excuses I'm really sorry I parked in your spot; I
didn't realize.
d. acceptance of offers I'm very glad to be getting this ride home.
e. responses to afaux pas Don't worry about it, you weren't to know
I had cancer.
f. unwilling promises Of course I'll do your taxes for you next
year too.
These examples are all instances of where S would normally not consider
the option of remaining silent. In (13a), for example, the - probably
\
unenthusiastic
- recipient of the socks would most probably be judged as impolite if s/he left the
present unacknowledged.
As for threats to S's positive face, these all involve cases in which she has to
admit something that is damaging to her (self-)image. See the examples in (14).
{14) a. apologies I'm so sorry I spilled wine over your new
fitted carpet.
b. acceptance of a compliment Really, I just threw that cake together,
nothing to it.
c. loss of physical control I'd like to introduce you to my [HICCUP]
parents.
148 Politeness and Impoliteness // // ///////////// /// ////////// //// /////////

d. self-humiliation, acting stupid My talk will focus on the human orgasm

e. confessions
f. emotional 'leakage' 48
... err... organism.
I haven't done the homework.
I haven't got the foggiest
[hahahahahahahahahahaha]

In all these cases, the speaker loses face by humiliating herself or by being
untrue to herself. In the case of (14b), the speaker possibly spent hours preparing
and baking the cake in the most careful manner but in accepting a compliment
feels constrained to belittle her own efforts. In a case of misspeaking (especially
in a 'Freudian slip' like in (14d)), the speaker has to take redressive action by
providing the correct word. Further examples of redressive action will come up
in the next section.

4.2.1.2 Politeness strategies

Politeness in B&L's view is intimately connected with compensating for the
threats to face that are inherent in the various types of FTA. The speaker takes
redressive action by using a form of words (and intonation can be vitally
important here, too) from which H can infer that S is sensitive to the damage that
she is doing to H's face and to her own face. However, the use of redressive action
is not obligatory and in fact is but one kind of strategy available to the speaker.
B&L identify a range of strategies that are open to a speaker when confronted
with the challenge of performing an FTA. They (1987: 60) represent the speaker's
options as a choice tree as follows:

Figure I: B&L' s strategies of politeness

Circumstances determining choice of strategu
Lesser 1. without redressive

<
action, baldly
onrecord
2. positive politeness
Do the FTA with redressive

4. off record

5. Don't do the FTA

Greater

48
This is B&L's term for when the expression of emotions, e.g. tears or laughter, imposes
itself on speech.
 IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII Politeness and Impoliteness 149

B&L (1987: 60) explain that the more an act threatens S's or H's face, the
more Swill want to choose a higher-numbered strategy, because in their view,
and as the Figure shows, the higher the strategy, the lower the risk of face loss.
Thus if a speaker has to say something that the hearer might deem offensive or
that could attack his face, she will tend to use strategies 4 or 5, rather than 1, 2
or 3, and if she had to choose among these first three, she would go for 3 and 2
before 1. This, however, is one of the aspects of B&L's theory that has been
disputed, for as some other scholars49 have shown, the silence coming out of not
doing an FTA may be more offensive than using an on-record, verbal strategy, or
the sarcasm used in an off-record strategy may be more pungent and offensive
than using a direct, on-record strategy.

The strategies shown in Figure 1 will now be discussed one for one. If the
speaker chooses to go on record without redressive action, she will simply
perform the FTA without more ado; this is what B&L call going bald on record,
here 'bald' is used in the sense of 'directly'. Going bald on record is the right
ategy where considerations of face are less important than getting the message
across. In situations where you're told "You don't need to beat about the bush,
·ve it to me straight", you're being invited to go bald on record. In 4.2.2.3 we
all see how the circumstances for choosing the right strategy are determined.
or example in urgent situations, it would be obstructive to use a higher-
numbered strategy: in the immediate aftermath of an accident, you can shout
Call an ambulance! to a perfect stranger without using any 'good manners' or
redress for your action.
The next two strategies involve going on record with redressive action. B&L
1987: 69-71), in explaining this, focus on redressive action with respect to the
rer's face, both positive and negative. In positive politeness, S shows that she
ants H's wants, and therefore polite formulation is used as a strategy to show
rapport and camaraderie. It is also useful in situations in which H's self-esteem
could be damaged, in order to minimize the threat, as we illustrate in (15), where
e speaker wants to show that in some way she values him as a fellow human
being and there is no offense intended. Let us reconsider how some of the bald on
record examples given in (11) above could be made more positively polite:
15) a. You've burned the toast again, you blockhead! ( = lla) » Silly old you,
you've burned the toast again, you sweet blockhead!
b. You're totally wrong about that.(= llb) » I'm sure you've got your
reasons for saying that, but my honest opinion is that you're totally
wrong about it.

See, for instance, Akrnan (1994), Alba-Juez (2007), or Kaul de Marlangeon's over-
whelming silence impoliteness strategy in this chapter (4.3.2).
150 Politeness and Impoliteness /////////////// ///// /////////// ///////// ////

c. You've failed the exam. ( = 1lc) » I know you worked hard and prepared
well, but unfortunately you've failed the exam.

In all these cases, S 'sweetens the pill' by anointing H's face (with the words
in bold) as well as imparting the message proper.
With negative politeness, the speaker reassures H that she does not want to
get in H's way. This involves various forms of self-effacement that persuade H
that his freedom of action is, after all, not impeded. Negative politeness is
essentially face-saving. Let us consider how some of the examples in (12) could
be modified to display more consideration for H's negative face, and thus be
made negatively polite:

(16) a. Give up your seat on the bus to elderly passengers. ( = 12a) » Why not
give up your seat on the bus to elderly passengers?
b. I told you the boss needs the report by 4 o'clock, remember? ( = 12c) » The
report is needed by 4 o'clock, remember?
c. You know, I could just kiss you all over right now. ( = 12h) » Let me know
if I'm coming on too strong but, you know, I could just kiss you all over
right now.

The aim of negative politeness used in (16) is to persuade the hearer that he
is expected to do something but is not being coerced into it. In (16a), the FTA is
made more polite by replacing the imperative mood (cf. 2.5) with a conventional
formula (Why not ... ) that contains the negative not and thus suggests the
possibility of non-compliance. In (16b ), the use of the passive voice removes any
need to mention the controlling agent or to start the utterance with the somewhat
reproachful I told you. And in (16c), the speaker gives the hearer an 'out' with a
lengthy preliminary.
The option of going off record arises when we say something that could be
but is not necessarily, interpreted as face-threatening. For example, asking
someone to feed you dinner is a major imposition that threatens your own
positive face and your hearer's negative face. However, if you're in someones
home and gradually making to leave, you might say something like (17):

(17) Oh dear, I didn't realize it was so late, the shops will be closed and I still need
to get something for dinner.

You have not explicitly begged to be invited for dinner. You have gone off
record, which allows your hosts to act as if their negative face has not been
affected (Oh, don't worry, you'll find a night shop still open on the way home)
and your own positive face, too, is saved if you are not invited (I'm not all that
hungry anyway). In off-record utterances, the conversational implicature ('Feed
 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I//// I I I I I I I I I I Politeness and Impoliteness 151

me!') may be drawn (and usually is), but both parties can act as if there was no
intention of launching the implicature.
The option of simply abstaining from performing the FTA (Don't do the FTA)
arises when the circumstances are so threatening that it is better to forget about
it altogether. We all know that every day we 'bite our tongue' and refrain from
saying what we might have said for fear of damaging our interlocutor's face or
our own. However, and in spite of what B&L argue here, as was pointed out
above, the strategy of remaining silent and not doing the FTA can at times have the
opposite effect of being less polite than doing the FTA by using any of the other,
lower numbered strategies. A simple example would be given by the impolite
silence involved in not thanking someone when they have done you a favor.

.2.2.3 Sociolo9ical variables: D, P & R

The seriousness of an FTA, as we have mentioned before, varies with the context
in which it arises. B&L re cognize three contextual ( or as they call them,
ociological) variables that impinge upon communication in ways that are
relevant for how speakers and hearers deal with threats to their face. They
assume that co-participants in a verbal interaction will have the same perception
of these variables - for this reason they are known as 'actors' assumptions' (B&L
1987: 76). In politeness theory, the three variables are known by their initials (D,
P, R). D stands for the social (relative) Distance between Sand H (or between H
and S, which is the same thing). People differ in how they assess social distance,
but freq,uent interaction and the q,uality of the interaction (for example with
regard to respect for one another's face) are important indicators of closeness.
The value for D will, for example, be generally low between family members or
lovers, while it will be high between strangers, especially if they come from
different cultures.
P stands for 'the relative Power of Sand H' (1987: 74), which, in contrast with
D (which is symmetrical) represents an asymmetrical type of relationship.Sor H
can exert P as a result of material or metaphysical control. By material power is
meant either physical force or financial capacity: for this reason an invading
army has P over a civilian population, an armed police force has P over peaceful
demonstrators, employers have P over their employees, parents have P over their
dependent children, etc. Metaphysical power refers to the control attributed by
belief systems: thus a priest will have P over his flock, the leader of a cult will
have P over his followers, a beloved pop singer will have P over his admirers, and
so on. Doctors have this kind of P over their patients (since they are seen as
having an impact on the patients' well-being); this is a good example of a
relationship where the D may be low but the P is high.
152 Politeness and Impoliteness / // // / // / // // / / /// / / /// // / /// / / // / // // / / / ///

R stands for Ranking and reflects how the culture in which S and H are
operating defines the 'imposition', i.e. the threat to the negative and positive face
wants of the two interactants. Impositions come into two varieties: those that are
concerned with the services His being asked to deliver and those that have to do
with the material goods (for example money) that H is being asked to part with.
An example of the latter comes from the area of begging: in certain cultures
begging for, say, a dollar would be ranked as a very high imposition, while in
others that might be the lowest amount a beggar would ask for. B&L recognize
that R may also differ from one situation to the next: begging outside a church,
which is associated with giving alms to the poor, may lower the ranking of the
imposition. People also vary individually: some rank the req,uest to fill out a
q_uestionnaire ( an example of delivering a service) as a very high imposition
while others do not.
B&L's claim is that S's choice of strategy results from an interaction among
the three factors D, P and R. They present an eq,uation for calculating the
weightiness of an FTA:
Wx = D(S,H) + P(H,S) + Rx

This says that the weightiness (W) of x, an FTA, is the sum of the D between
Sand H, the P of Hover Sand the cultural/situational R of x . Since none of the
variables can be given numerical values, the eq,uation is merely indicative.
Nevertheless, it does show that an FTA which involves high D between Sand H,
high P of H over S, and a high R of that FTA will be extremely weighty and thus
will call for an exceptionally high investment in politeness. Imagine wanting to
ask a heavily armed, total stranger for a gift of $50,000! Conversely, where D, P
and R are all very low, for example when asking a beloved sibling to buy you a
coffee, Scan safely go bold on record and say Buy m e a coffee!.
It gets interesting where one or two of the values are high while the others
are low. Consider the case of borrowing someone's cell phone in order to call
home. Here the R of the imposition is for most people moderate, neither low nor
high. The R is not low since a cell phone is a personal and valuable possession,
often containing information, pictures and other data that are important to its
owner. The R is also not high, since the user will soon get the phone back and the
expense of a single call is not much. The moderate R, then, will already add to the
weightiness of the FTA and predispose S to take some redressive action,
explaining that she will only make one call, possibly offering to pay for i
promising to stay within sight of the owner, etc. As for the D, the weightiness of
the FTA will be affected by the previous relationship between borrower and
lender. If they have helped each other out on earlier occasions, that will have
entailed 'exchanging face' and thereby lowering the D; similarly , if they are
 II III I I II II III I I II I II I II I I II I I III III II II I I II Politeness and Impoliteness 153

related by family or some other emotional bond, D will be lower. And P is relevant
in the sense that in a hierarchical work environment, asking a favor of one's
superior increases the weight of the FTA. Similarly, a patient cannot easily ask a
doctor for a loan of his cell phone. However, if the patient and the doctor regularly
see each other at the gym, that will lower the D factor and make the req,uest less
weighty.
Let us now imagine possible formulations for different settings of the
weightiness of the FTA:
(18) D low, P low, R moderate (e.g. to S's brother): Gimme your cell for a couple
of minutes!
(19) D moderate, P low, R moderate ( e.g. to a work colleague): Can you lend me
your cell to make a single phone call home?
(20) D moderate, P moderate, R moderate (e.g. to the office manager): Excuse
me, I've forgotten my cell phone, could I possibly borrow yours for a minute
to make a q,uick call home?
(21) D moderate, P high, R moderate (e.g. to a member of senior management):
I'm very sorry to trouble you, but this is a bit of an emergency: I've stupidly
left my cell phone at home and I really need to call my husband about a
very important matter, and I just wondered if there was any way I could
possibly borrow your phone for the briefest of calls?
(22) D high, P moderate, R high (e.g. to a stranger on the street): Excuse me,
madam, I'm terribly sorry to trouble you but the thing is that I have a real
emergency going on. I've come out without my cell phone and I desperately
need to make an important call home. I know I'm a complete stranger to you
and you're probably anxious about me approaching you like this, but I'd like
to know if there's any way, any way at all, I might borrow your cell phone
for two seconds?

Over the five examples, the weightiness (Ylx) of the FTA grows. Notice how
this is paralleled by an increase in the length of the examples: generally speaking,
the greater Wi , the more words S needs to put into redressive action. The details
of the examples will repay careful examination, but suffice it to point out here
(a) the use of verbs (give - lend - borrow) , stressing the impermanence of the
transfer of the phone in the weightier examples; (b) the increasing insistence on
the brevity of the transaction (for a couple of minutes through to for two
seconds); (c) the use of increasingly tentative modal verbs (none - can - could-
m ight); (d) the increasing explicitness about the motivation for the req_uest; (e) in
the most extreme examples, the speaker's willingness to damage her own face
in the interests of persuading the hearer (stupidly -I'm a complete stranger to
y ou), and her explicit recognition of the hearer's reluctance (you're probably
154 Politeness and Impoliteness ////////////////////////////////////////////

anxious ... ). As B&L (1987: 81) put it, S's redressive strategy 'encodes' how she
estimates the danger of the FTA. It is therefore also important for her not to
exaggerate, since that may put H on his guard: Is the req,uest perhaps more
threatening than I imagine if it is prefaced by so much verbal effort?
From the perspective of politeness theory, finding a balance between
communicating effectively and avoiding loss of face involves a kind of calculation
on the speaker's part. She will be prepared to invest more energy in politeness if
she reckons that the 'payoff' (B&L 1987: 71-74) will be rewarding. Going bald on
record, for example, can be risky since it may be taken as impolite but has
payoffs, too: someone who 'speaks her mind' may, especially in certain cultures
or social groups, be praised for her honesty and reliability. Going off record has
the pay off of being less confronting but may give the speaker a reputation for
being devious. Positive politeness has its payoffs, too: for example, by showing
that she shares H's wants, S can make criticisms of H without offending him; on
the other hand, she may come across as being insincere or ingratiating. The
payoff of negative politeness, finally, is that H grants S's req,uest almost as if it
were by his own free will; a risk for S, however, is that by giving Han 'out', H
feels free not to comply.
A remarkable point made by B&L about their theory is that they claim it to
be universal, i.e., that it can be applied to all human languages and cultures.
Indeed, their approach is very revealing and has given ground for discussions of
politeness in all cultures around the world, but this claim for universality is one
of the aspects of their theory that have been most widely criticized. We shall
examine some of these criticisms in the next section.

4.2.3. CRITICISMS LEVELED AT BOTH LEECH'S AND B&L'S

APPROACHES
Since both Leech's and B&L's models present an essentially Gricean treatment of
politeness, they were criticized in similar ways regarding certain aspects of their
respective theories. As was anticipated above, B&L's claim for universality was
not well received by several linguists and scholars (e.g.: Gu 1990, Lee Wong 1999
Mao 1994, Ide 1993, Matsumoto 1988, Nwoye 1992, Werkhofer 1992), in particular
by those specialized in Eastern languages and cultures, such as those of China,
Korea or Japan. Wierzbicka (2003) is another of the authors who took a relativistic
approach to politeness as opposed to the universalistic view of B&L, and who
also accused Leech (1983) of being ethnocentric in claiming that his maxims are
universal. Leech (2014) has responded to this accusation by implying that
Wierzbicka most probably did not read his work properly, for in 1983 he never
made any claim for the universality of his model of politeness, although he did
mention in his conclusion that it was his hope that his model could be used in
 II I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I Politeness and Impoliteness 155

contrastive studies. Leech complains of having been caricatured in some way, as

B&L have been too, because even if these authors did claim universality, they
also pointed out that "the application of the principles differs systematically
across cultures, and within cultures across subcultures, categories and groups"
(1987: 283). Leech (1983) was also criticized for having too many maxims in his
model, a criticism that as he later explained (in his 2014 book) could have been
avoided if he had made clear that all the maxims represent variants of the same
constraint, namely the General Strategy of Politeness (see 4.2.1.2 above).
In general, both models were accused of having a Western bias which comes
across as, in Schmidt's (1980: 104) expression, "rather paranoid" because, among
other things, they emphasize individual wants while they disregard the group-
oriented politeness of the Eastern World (e.g., see Matsumoto 1989, Wierzbicka
1991, Koutlaki 2002), where people tend to give high importance to their rights
and obligations in relation to in-group and out-group members, as well as to
superiors and inferiors.
Some authors ( e.g. Spencer-Oatey 1993) also criticized B&L's concepts of
Power and Distance as sociological variables affect~ng the strength of the FTA,
on the grounds that interpretations of what she calls "vertical or horizontal
distance" differ from one culture to another, and that, especially in the Eastern
cultures, Power is not such an appropriate term. B&L's formula for calculating the
strength of an FTA was also criticized for treating the three factors (P, D and R)
as cumulative ("The greater the power of h over s, the greater the imposition
upon h" (B&L 1987: 76), etc.), which makes the formulation too simplistic.
In the Spanish-speaking world, authors such as Bravo (1999), Boretti (2001)
and Hernandez Flores (1999) argued that face does not always show an individual
orientation; for instance, Boretti pointed out that in Argentina req,uests and offers
using a direct, bald on-record strategy (in the imperative mood) are not perceived
as face-threatening by hearers. In general, according to these and other authors,
Spanish speakers seem to be mainly oriented towards positive politeness because
for them the expression of individuality is less important than the expression of
solidarity. For this reason, Bravo (1999) q,uestioned the universality of the
concepts of positive and negative face, and proposed the categories of autonomy
and affiliation instead, on the grounds that her categories are culturally-oriented,
and simultaneously universal and empty (because they vary from one culture to
another, so for each culture this category will be filled with different elements).
According to Bravo, autonomy alludes to the speaker's need to perceive herself
and be perceived by others as different from the group, and affiliation stands for
the need to belong to and be perceived as part of the group. These categories
may at first sight seem to be the same as B&L's negative and positive face
respectively, but Bravo explains that while in what (in the Spanish-speaking
156 Politeness and Impoliteness I I I I I I I I I I I I I/ I I I I I I I I I I I I I I I I I I I I I I I I I I I I I

world) is commonly known as 'Anglo-Saxon cultures' autonomy has to do with

negative face and affiliation with positive face, this correspondence does not
apply to other cultures such as the Spanish one, where, for instance and in
contrast, autonomy is characterized by many of the features of what B&L term
positive face.
We find it necessary to note here that, in spite of all this discussion about the
differences and similarities among different 'cultures' initiated by authors at the
end of the 20th century (not only regarding politeness, but also in cultural studies
in general), in the 21 st century many authors, including the writers of this book,
find it inappropriate to speak about "Spanish culture" or "Anglo-Saxon culture
due to the evident fact that it is very difficult and indeed undesirable to make
generalizations about any culture in the world, considering the large number of
sub-cultures and sub-groups that are found within these macro-cultures.
Furthermore, even within the tiniest of sub-groups there may be differences
affecting its linguistic-pragmatic features, having to do with numerous variabl
other than the social group, such as the gender or the personal history of
its members. This is the reason why - as was anticipated in Chapter 150 - authors
like Scollon, Wong Scollon & Jones (2012) have rejected the use .of the term
culture in favor of the concept of discourse system.
Lavandera (1988), another Spanish-speaking author, was one of the fir
scholars to notice some inconsistencies in B&L's model. She observes that the
phenomena politeness/impoliteness should not be seen as a dichotomy but as a
continuum, and she proposes the notion of politeness force as supplementary to
that of illocutionary force , placing emphasis on the 'obligatory nature' (1988:
1196) of the former force in all utterances. Lavandera adheres to Fraser & Nolen
(1981) view that it is the conditions under which the expressions are used, and
not the expressions themselves, that make them more or less (im)polite, for (im)
politeness is a property of utterances, not of sentences. Moreover, she points to
the following weaknesses in B&L's model:
a) ascribing the degree of politeness to a strategy - as B&L do - and not to the
entire speech act within which the strategy occurs makes it impossible t-0
account for the fact that the same speech act may show different ( or at least
more than one) strategies being used at the same time;
b) B&L do not make the distinction between strategies like "Be ironic", which
are purely pragmatic, and other strategies which contain specific linguistic
descriptions, such as "Employ a diminutive";
c) B&L do not recognize the fact that politeness is a permanent component of all
speech acts, and they do not present any strategies aimed at impoliteness.

50
See 1. 7. in this book.
 IIII IIIIII IIIIIII III IIIIIIIIII IIIIIIIII II II Politeness and Impoliteness 157

And it was precisely because impoliteness was q,uite disregarded in B&L's

theory that some authors started investigating the phenomenon in the 1990s. Two
of the pioneers were Silvia Kaul de Marlangeon (1995) in the Spanish-speaking
orld and Jonathan Culpeper (1996) in the English-speaking world. Even though
Kaul de Marlangeon's publication was previous to Culpeper's, and although both
approaches to impoliteness are very sensible and have many points in common,
the one that became more widely known around the world was Culpeper's,
probably because Kaul de Marlangeon's publication was written in Spanish and
Culpeper's in English. Because of the impact that these two approaches had in
lheir respective macro-cultures, we consider it important to include them in the
following section of this chapter.

.3. THEORIES OF IMPOLITENESS

fu spite of the undeniable impact and influence of B&L's Theory of Politeness,
Tery soon after its publication it became clear to many scholars that what people
do when they talk or interact is not limited to avoiding face-threat. On the
contrary, sometimes they purposefully threaten their interlocutor's face , or
additionally, they may sometimes threaten someone's face accidentally ,
consciously, or gratuitously. It was therefore necessary to reflect a broader
range of phenomena, and this is the reason why the term ( im)politeness was
coined. The wider scope of this term allowed researchers to avoid dichotomizing
e phenomenon in terms of two poles (politeness vs. impoliteness) in favor of
considering a continuum along which numerous types of behavior can be found,
or which different terms have been used (e.g. Watts's 1989 politic behavior,
erkourafi's 2001b u nmarked politeness, or Kerbrat-Orecchioni's 2011 non-
liteness).
We find it important and necessary to point out here that there is a very
· timate association between impoliteness and emotion. It is in fact a matter of
common knowledge that impoliteness may be caused by or may arouse feelings
of anger or humiliation, and thus negative emotional reactions are a common
eature of most instances of impoliteness. As Terkourafi (2016) explains, the
d ose link between impoliteness and emotion is a final point on which politeness
and impoliteness differ: while impoliteness is almost universally associated with
egative feelings, politeness can arise from or trigger a varied gamut of emotions,
and may even pass unnoticed.
Even though some authors from the 1980s and early 1990s (e.g. Craig et al.
1986, Lakoff 1989, Tracy 1990, Penman 1990) had pointed out that politeness
eory could be extended to include antagonistic or confrontational
158 Politeness and Impoliteness I I I I I/ I I I I I I I I I I I I/ I I I I I I I I I I I I I I I I I I I I I I I /

communication, they did not develop a theory of impoliteness, as Kaul de

Marlangeon and Culpeper did, respectively, in 1995 and 1996. In effect, Jonathan
Culpeper (e.g. 1996, 2005, 2008, 2010, 2011, 2012; Culpeper at al. 2014) and Silvia
Kaul de Marlangeon (e.g. 1995, 2005 a & b, 2008 a & b, 2010, 2014) have been
pioneers in noticing that the phenomenon of politeness includes a much wider
spectrum of strategies than those initially envisaged by B&L. Thus both Culpeper
(for English) and Kaul de Marlangeon (for Spanish) have investigated the use of
strategies designed to have the effect of social disruption and oriented towards
attacking face. Both authors take as a point of departure B&L's strategies of
politeness in order to propose a framework of impoliteness strategies, and both
of them initially resort to specific types of discourse (the former the discourse of
an army training camp in the film Sold ier Girls, and the latter the Argentinian
Tango discourse of the 1920s) to illustrate the fact that there are discourse
communities within which the practice of impoliteness, NOT of politeness, is
taken as the norm. We now turn to these two authors.

4.3.1. CULPEPER'S APPROACH

Culpeper (1996) points out that some areas of politeness are not well represented
in B&L's politeness model, and he the refore proposes an impoliteness framework
where each of B&L's politeness superstrategies has its opposite impolitene
superstrategy which attacks face instead of enhancing or supporting it. Thus.
his taxonomy includes the following strategies ( adapted from Culpeper 1996:
356-357, including some examples of our own):
1) Bald on record impoliteness: The use of strategies with a clear intention of
attacking face in a direct way. A prototypical example of this strategy is found
in B's utterance in (23), where the emotional context is that of a heated
argument, and where B is very angry at her interlocutor:
(23) A: I don't understand why you're so stubborn and don't wanna listen
tome!
B: Shut up, you idiot!
2) Positive impoliteness: The use of strategies designed to damage the
addressee's positive face wants. Examples of this strategy could be to exclude
someone from the conversation by using obscure or secretive language, to be
unsympathetic by denying association or common ground with the other, or
to simply ignore someone.
3) Negative impoliteness: The use of strategies designed to damage the
addressee's negative face wants. Examples of this strategy could be to ask
someone about his private life with the clear intention of invading his privat.e
space and making him feel ashamed, or to scorn or ridicule an interlocutor.
IIIIII III IIIIIII IIIIIIIIIIIIII IIIIIIII II II II Politeness and Impoliteness 159

4) Sarcasm or mock politeness: the use of politeness strategies that are obviously
insincere. 51 These strategies refer to the use of mock politeness with the aim
of provoking social disharmony. (24) exemplifies this strategy by means of
the sarcastic use of honorifics:

(24) Fiona and Tim are having an argument, and after Fiona rejects Tim's
invitation to make it up and instead decides not to talk to him anymore,
Tim says:
O.K., it seems Her Majesty is not in the mood to talk anymore today!

5) Withhold politeness: the absence of politeness work where it would be

expected. A prototypical example of this impoliteness strategy is found in the
situation where someone does not thank somebody for a present and
deliberately remains silent in order to hurt and offend him.

Culpeper also points out that B&L's politeness model does not take into
account some aspects such as paralinguistic or non-verbal politeness. The bow
that Japanese people make when saluting others, for instance, can be considered
a non-verbal strategy of politeness. Culpeper provides counter-examples of
impoliteness in these areas, such as shouting and avoiding eye contact. Other
non-verbal strategies of impoliteness could be rolling one's eyes at someone's
stupid comment, looking at one's watch to show that one has better things to do
than being with one's interlocutor, or covering one's ears with one's hands to show
unwillingness to listen to one's interlocutor.
In general, polite behavior entails a recognition that one's interlocutor is a
person like oneself, whereas impolite behavior does not involve that
recognition. That is why impoliteness occurs in discursive situations and
contexts where there is great power ineq,uality. 52 One of these is the context of
the army, where impoliteness is a common and systematic practice deployed by
sergeants, as Culpeper (1996) illustrates with examples taken from the
documentary film Soldier Girls (1981). He shows how three sergeants attack
Private Alves' mental stability, psychological make-up, and even her genetic
make-up, as seen in (25):

51
For Culpeper sarcasm is close to Leech's (1983) conception of irony, i.e. negative irony
(in Alba-Juez's 1995, 1996 [2000] or Alba-Juez & Attardo's 2014 terms), leaving positive
irony or banter (in Leech's terms) aside.
52
It is normally assumed that the powerful are those who are entitled to be impolite, but
it may also be the case that it is the powerless who, in an act of rebelliousness, choose
to be impolite. Consider, for instance, a cheeky boy who answers his teacher back in
a disrespectful manner, and does not care or is not conscious about the conseq,uences
his behavior may bring about.
160 Politeness and Impoliteness /// ////// //// //// // //// /// //// //// /// /// ////

(25) Sl: Do me a favor don't have any children ... because unfortunately there is
such a thing as heredi hereditary genes that I would hate to think that
anybody would even closely come out like you. (1996: 362)

This early work of Culpeper's, as well as that of Kaul de Marlangeon (which

we shall discuss in 4.3.2), raised various issues and further research q_uestions
such as How do we really know if the speaker's intention is to be impolite? What
counts morefor the analysis of impoliteness, the intention of the speaker, or the
interpretation of the addressee? What aspects of a speaker's behavior lead w
the interpretation that she is being impolite? When is impoliteness used i n
eq,ual relationships? Are there different degrees of impoliteness? These two
authors, as well as a number of other researchers (e.g. Marina Terkourafi, Sarah
Mills, Miriam Locher, Derek Bousfield) have tried to find an answer to these and
other crucial q_uestions.
In a later publication, Culpeper clarifies some aspects of the phenomenon of
impoliteness and he provides the following definition:
Impoliteness comes about when: (1) the speaker communicates face-attack
intentionally, or (2) the hearer perceives and/or constructs behavior
as intentionally face-attacking, or a combination of (1) and (2.) (2005: 38)

In his later work, as well as in the work of many researchers in the 21

century, the study of (im)politeness has mainly been approached from the
relational perspective (as we shall see in 4.3.3), taking into account Halliday
( 1978) interpersonal metafunction.
We shall now examine the work of Silvia Kaul de Marlangeon, who, as was
anticipated above, with her 1995 article on the impoliteness of tango song
became a pioneer of impoliteness studies, in particular within the Spanish-
speaking world.

4.3.2. KAUL DE MARLANGEON'S APPROACH

Taking as data a corpus of tango discourse (found in the lyrics of tango songs)
of the 1920s in Spanish, Kaul de Marlangeon (1995) points out, in the same wag
Culpeper did a year later in English, that there are discourse strategies which are
not used with a mitigating or redressive intention, but, on the contrary, with the
express aim of attacking or threatening face. Thus she argues that B&L
taxonomy of politeness strategies may be the base for a taxonomy of impoliteness
strategies, both taxonomies complementing each other. In consonance with
Lavandera (1988), she views the phenomenon of politeness as a continuum,
within which she contends there is a natural grading for impoliteness. She
illustrates, using Spanish examples, that a speaker may choose to be overtly
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I/ I I I I I I I Politeness and Impoliteness 161

impolite by performing an act on record (baldly, or with some degree of redress,

attacking her interlocutor's positive or negative face), or she may choose to do
so in a covert way, i.e. off record, by means of irony and/or sarcasm, thereby
reducing - at least in its formal appearance - the face threat.
She also points out that power (P) is the main variable determining the use
of impoliteness, and that distance (D) cannot be regarded as a symmetrical
relationship (as B&L indicate for politeness) in cases of impoliteness. Likewise,
she argues that within non-institutional contexts the P and D variables respond
t-0 personal factors which serve the purposes of attitudes in which referential
social patterns are absent. With respect to the R variable, the author observes
that when the values of friendship, faithfulness (on the part of a woman in a love
relationship), love for one's mother, etc. are not shared by one of the participants
or characters in a tango song, then impoliteness is triggered. 53 Within the tango
discourse world of the 1920s in the Rio de la Plata area, the model person (in the
sense given to it by B&L), then, is a rational agent who chooses to attack and
threaten face with little or no redressive action as the best means for attaining his
communicative goals. And we use the masculine 'his' here because normally the
tango singer or speaker is a man who often addresses a woman by showing a
negative evaluation of her positive face through the use of criticisms, reproaches,
insults, complaints, etc., as illustrated in (26):

(26) 6No manyas, che pelandrunal q,ue te vende de muy lejos/ esa piel de zorro
viejo/ mas pelada q,ue Alvear? (Don't you realize, you, poor stupid woman,
that your old fo x skin -balder than Alvear54 - gives you away? [Our
translation] 55) (1995: 29).

Silvia Kaul de Marlangeon has devoted a great part of her life to the study
of impoliteness and has refined her model of impoliteness in successive
publications, such as those of 2005a and b, 2008a and b, 2010, or 2014. Her (2008a)
definition of impoliteness considers many factors and differs from Culpeper's
(2005), for instance, in that she does not make the distinction between the hearer's
perception and construction of the speaker's intentionally impolite act. Instead,
she prefers to replace both with the concept of interpretation, thus preventing the
analyst from having to make the choice between these two options, since

a.1 The author here analyzes the tango poetics as an aspect of the etlws of the Rio de
la Plata culture, represented in tango songs and expressing a devalued view of
the world and in particular of women, who are treated with impoliteness.
The reference here is to Marcelo T. de Alvear, an Argentinian politician who was bald.
5o
It should be noted here that the variety of Spanish used in the tangos of the 1920s is
the so-called Lunfardo of the Rio de la Plata, which has a very peculiar vocabulary for
which it can be very difficult to find accurate translations into English.
162 Politeness and Impoliteness /// ///////////////// /////////// ////// ///////

differentiating what the hearer perceives from what s/he constructs as impolite
behavior is a complicated, if not impossible task. Another way in which her
definition differs from Culpeper's is that she considers that impoliteness can
sometimes be involuntary, whereas Culpeper has advocated the idea that
impoliteness 'cannot be unintended' (2005: 37).
Thus, according to Kaul de Marlangeon, impoliteness occurs in any of the
following eleven instances (2008a: 258-259 [our translation]):
A) When the speaker (S):
1) tries to be polite to the hearer (H), but for H, S's manner of expression is
reminiscent of improper, indecorous or disrespectful language. (1)
2) involuntarily offends H by:
2.1.) committing a gaffe or faux pas, or (2)
2.2.) stinting on the politeness expected by H, or (3)
2.3.) ignoring politeness norms (4)
3) deliberately uses offensive language toward him/herself with different
motivations (5)
4) is very polite or excessively polite to the hearer, in order to hurt or mock
him/her. (6)
5) voluntarily stints on the politeness expected by H (7)
6) deliberately offends H with a purpose that may:
6.1.) damage H's face (8)
6.2.) defend S's face (9)
B) WhenH:
1) interprets S's behavior as an intentional face attack that induces him/her to
accept the attack or reject it through defense or counter-attack. (10)
2) remains silent intentionally, in order to indicate disagreement/
discontentment with S's utterance. (11)

Based on this "endecatomic"56 definition, she proposes a scale of impoliteness

acts, according to the growing intensity of the impoliteness force transmitted by
the acts in q,uestion. For the intended acts, the higher the number, the stronger
the impoliteness force, i.e., the higher the degree of impoliteness (Figure 2):

56
This is indeed what she calls it (Sp. endecat6mica).
 III IIIIIIII IIIII IIIII IIII IIIIIIIIIIIIIIIIII I Politeness and Impoliteness 163

Fi9. 2: The politeness-impoliteness force continuum (as in Kaul

de Marlan9eon & Alba-J uez 2012: 74)

§
;:::l
.s
§
u
Face flattering acts

t(J'J
rn
(l.)

B
;.:::l
0
0..
0)
.s
(J'J a,)

.....
B
(J'J Bald on record face threatening acts
C.)

0
0.. Neutral Degree
t Acts which are neither polite nor impolite
of Politeness
Formally impolite acts with a polite purpose S'
0 g
P.. ro
1Jl
s·
co
§'
"d
......
0

::::i
ro
1Jl
1Jl

We shall now briefly describe each category, using some of the examples
presented in Kaul de Marlangeon & Alba-Juez (2012) (henceforth KdL & AJ),
where all the categories are illustrated with real or fictional examples (taken
from movies or some TV series), both in Spanish and English:
l) Formally impolite acts with a polite purpose: The ludicrous use of irony
(i.e. mock impoliteness), which may have no impolite intention, but on some
occasions may include such an intention or at least come across as mildly
impolite. A prototypical example of this type is the famous utterance uViva
Mexico, Cabrones!! (whose more or less literal translation would be "Long live
Mexico, you bastards!!", but which is pragmatically used by Mexicans to
164 Politeness and Impoliteness ///// //// ////////// //// // ////// ///// //// ////

express love for their country, irrespective of what anyone might think, and
showing contempt for those who don't love Mexico or have evil intentions or
thoughts about that country).
2) Involuntary impolite acts: These are by definition devoid of impolite
intentions, a fact that does not exempt them from having an impolite effect in
some contexts or situations. There are three types:
a) Gaffes: (27) shows an embarrassing example of a real-life gaffe, when the
American Vice-President Joe Biden, unaware of the fact that Senator
Graham was confined to a wheel chair, asks him to stand up:
(27) Eiden: Ah ... Chuck Graham, State Senator is here. Chuck, stand up,
Chuck! Let me get to see you ... (Biden then notices Graham is sitting in
a wheel chair). Oh, God love you, what am I talking about! I'll tell you
what: You're making everybody else stand up though, pal. Thank
you very, very much I'll tell you what: stand up for Chuck! (the audience
cheers) Thank you pal ... (Transcribed from YouTube at https://www.
youtube.com/watch?v=C2mzbuRgn14 September 9, 2008, in KdL & AJ
2012: 79-80)
b) S's involuntary stint on the politeness expected by H: This type reflects
those cases in which the speaker unintentionally falls short of politeness
as illustrated by (28), in a scene from the movie Bridget Jones's Diary
when Mark Darcy, instead of responding to Bridget's thanks with a polite
"You're welcome", gives a very sincere response, thereby unintentionally
offending Bridget, who by now is notably in love with Mark and would
have preferred him (not his parents) to have invited her to the party:
(28) Bridget: Thank you for inviting me.
Mark: I didn't. It must've been my parents.
(Bridget Jones's Diary, in KdL & AJ 2012: 80)
c) Involuntary omission of politeness: A prototypical example of involuntary
impoliteness occurs when children or teenagers forget to thank adults,
which brings about the inevitable subseq,uent remark (in retrospective
awareness) by their mothers/fathers/caregivers: (Johnny) what do you
say (to Mrs. Xfor having invited you to her home)? Everyday experience
tells us that it is possible that the adult in q,uestion will be heard saying
afterwards that 'Johnny's manners left much to be desired', which shows
that s/he has taken the teenager's omission as an instance of mild
impoliteness or rudeness.
3) Self-impoliteness acts: Acts by means of which people use impolite or rude
language toward themselves. Self-impoliteness may be a) authentic orb)
IIIIII IIIII III IIIIIII IIIIIIII IIIIIII IIIIII II Politeness and Impoliteness 165

feigned. Examples of these two kinds are given in (29) and (30) respectively.
While in (29) the speaker is sincere about his feelings, in (30) he strategically
manipulates his message with the aim of eliciting a Face-Flattering Act
(Kerbrat-Orecchioni 2004) from the hearer.
(29) Charles realizes he has made a big mistake in his Math exam, and
sincerely expresses:
Charles: Damn! What an idiot I am!!
(30) Vladimir (whose native language is Russian) has made a silly mistake
when speaking in English, and in order to save face with his teacher (but
not because he really thinks he is a fathead), he says:
Vladimir. How stupid of me, what a fathead I am!
Teacher: Oh, no! How can you say such a thing! You are a very smart
student.

4) Formally polite acts with an impolite purpose: Here politeness forms are
paradoxically used as a means to achieve impoliteness. All ironic uses of
polite forms of address fall into this category, as well as some manifestations
of cynicism, as shown in (31), an example taken from The Shawshank
Redemption, 57 where both the linguistic and non-linguistic context help the
viewer interpret Hadley's last utterance (in bold) as not precisely a
"welcoming" one, in spite of its literal meaning:
(31) Norton: This is Mr. Hadley, Captain of the Guard. I am Mr. Norton, the
warden. You are convicted felons; that's why they've sent you to me. Rule
number one: no blasphemy. I will not have the Lord's name taken in vain
in my prison. The other rules you'll figure out as you go along. Any
q,uestions?
Convict: When do we eat?
Hadley: (Cued by Norton's glance, steps up to the con and screams right
in his face): YOU EAT WHEN WE SAY YOU EAT! YOU SHIT
WHEN WE SAY YOU SHIT! AND YOU PISS WHEN WE SAY YOU
PISS! YOU GOT THAT YOU MAGGOT-DICK MOTHERFUCKER?
(Hadley rams the tip of his club into his belly. The man falls to his knees,
gasping and clutching himself. Hadley takes his place at Norton's side
again, and softly, says):

57
A film in which the protagonist, Andy Dufresne, is accused of having killed his wife
and is therefore sent to prison. On the first day in Shawshank prison, Warden Norton
introduces the new prisoners to the Captain of the Guard, Mr. Hadley, a corrupt,
cynical and cold character who abuses his power to a cruel extent.
166 Politeness and Impoliteness / / / / // / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

Norton: Any more q,uestions? (Silence) ... I believe in two things:

Discipline and the Bible. Here, you'll receive both (holds up a
Bible). Put your trust in the Lord. Your ass belongs to me.
Welcome to Shawshank.
(From The Shawshank Redemption, in KdL & AJ 2012: 82)

5) S's voluntary stint on the politeness expected by H: The impolite behavior is

here interpreted as such based on the speaker's deliberate omission, rather
than action. However, the omission is not complete, for she participates in the
exchange, but withholds politeness to a certain extent, thus flouting Grice's
Quantity Maxim by avoiding the upper points of a compliment or admission,
as illustrated in (32), where Humphrey's impoliteness is brought about by his
reluctance to tell Bernard his secret:
(32) Bernard: What are we supposed to do about it?
Humphrey: Can you keep a secret?
Bernard: Of course!
Humphrey: So can I.
(From Jay & Lynn's (1994) "Yes, Minister" Video Series Episode: 'Open
Government', in KdL & AJ 2012: 83)

6) Overwhelming silence acts: Overwhelming silence is the silence that is used

to show disagreement or a certain degree of contempt for, or disapproval of,
the interlocutor's previous comment or behavior. A prototypical example
of this kind of impoliteness is found in (33), where the speaker makes a
flattering comment about a third person, to which the interlocutor responds
with this type of silence, which is interpreted as a form of disagreement,
eq,uivalent to other occasions when the answer is verbal but derogatory,
namely when someone responds by saying "No comment", implying that it is
better not to say anything because if she did, it would include a very negative
criticism. Overwhelming silence in these cases is, then, the extreme
manifestation of stint on the politeness expected by H (Type 5 above).
(33) A: I think Mr. White is a wonderful teacher.
B: (Overwhelming silence)

7) Fustigation 58 impoliteness acts: this type of impoliteness consists of verbal

aggression in a confronting or challenging situation, where the main and

58
Kaul de Marlangeon (2005a) draws on the meaning of the termfustigation (in the
sense of "whipping somebody or something") and uses it in a metaphorical way.
IIIIII IIIII IIIIIIII IIIIIII IIIIIII IIIII II IIII Politeness and Impoliteness 167

express aim is to damage the interlocutor's face. It may be enacted through

direct or indirect strategies, as shown in the utterances in bold in (34) and (35)
respectively:
(34) Hadley: What's gour malfunction gou fat fuckin' barrel of monkeg-
spunk?
Fat ass: PLEASE! I AIN'T SUPPOSED TO BE HERE! NOT ME!
Hadley: I ain't gonna count to three! Not even to one! Now shut the
fuck up or I'll sing gou a lullaby!
(From The Shawshank Redemption, in KdL & AJ 2012: 86)

(35) Bridget, who is in a relationship with Daniel, finds a woman naked and
hiding in Daniel's bathroom, and therefore looks at her in astonishment,
while Daniel clumsily tries to introduce them:
Daniel: This is Lara, from the New York office. Lara, this is Bridget.
Lara: Hey there. (To Daniel) I thought gou said she was thin.
(From Bridget Jones's Diary, in KdL & AJ 2012: 87)

In (34) we find an instance of direct verbal aggression within the context of

a prison, and in (35) Lara's sarcastic allusion to Bridget's overweight condition,
together with the fact that she had been having sex with Daniel shortly before
Bridget's arrival, constitutes an instance of indirect but utterly humiliating
fustigation impoliteness towards Bridget.
As we have seen in this section, Kaul de Marlangeon's (2008a) taxonomy
aims to include the whole gamut of impolite acts, from the mildest to the
strongest. She adopts a socio-cultural and pragmatic perspective by pointing out
that (im)polite verbal behavior is idiosyncratic to each culture or community of
practice, 59 and although her work is free from any universalistic assumptions, she
aligns with Leech's (1983, 2014) belief that there exist common shared patterns or
basic universal principles without which it would be impossible to speak of
(im)politeness in the first place. This author (2010) also adopts a topological
perspective for the delimitation of both polite and impolite community practices,
which constitute her units of analysis of (im)polite discourse, and which are
characterized by their social, extra-linguistic and multidimensional nature within
a sociocultural pragmatic approach. In her many publications about impoliteness,
she has looked into different contexts: theatre plays, ethnic jokes, horoscopes,
graffiti, football discourse, the discourse of celebrities, reviews of scientific
publications, and social network discourse.

59
See also Kaul de Marlangeon (2010).
168 Politeness and Impoliteness ////I I I I/ I! I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I

Having presented the earliest theoretical approaches to the phenomenon of

impoliteness, in order to round off the chapter we shall now comment briefly on
some further work on the topic that has been done in the 21 st century so far.

4.3.3. FURTHER WORK ON (IM)POLITENESS: THE RISE

OF THE RELATIONAL VIEW
As the reader may have noticed, most of the authors (e.g. Leech, Culpeper, Kaul
de Marlangeon) who began writing about (im)politeness at the end of the 20 th
century have continued to do so in the 2l8t century, showing an evolution in their
work. In particular there has been a tendency towards the consideration of
(im)politeness as an all-pervasive phenomenon that is by no means one-
dimensional or easy to grasp (Locher & Bousfield 2008: 4), its main characteristic
being that it involves relational work (i.e. the work done by social actors who are
constantly negotiating their positions, cf. 4.2) and therefore forms part of a wider
communicative experience.
Part of the discussion has also turned to whether impoliteness and rudeness
are the same thing or not. Both Bousfield (2008) and Culpeper (2008) consider
that the key for impoliteness is the hearer's understanding of the speaker's
intentions, while in contrast Terkourafi (2008) argues that the recognition of
intentions constitutes rudeness rather than impoliteness, but in any case, it is
clear that in the relational work spectrum, the two terms occupy a similar
conceptual space.
Another important issue is what kind of (im)polite behavior is marked or
unmarked. In Locher and Watts's (2005) perspective, relational work includes
various kinds of behavior: appropriate polite behavior is positively marked,
whereas over-polite behavior is negatively marked. Inappropriate impolite
behavior is also negatively marked, rudeness being its most extreme
representation (which shows that for these authors, rudeness is the worst or most
aggressive kind of impoliteness). The unmarked behavior would be what Watts
(2003) calls politic behavior, 60 something in-between the two marked poles,
which, like polite behavior, is also appropriate. Culpeper (2008: 23) notes that this
raises the issue of how 'markedness' differs from 'appropriacy', and suggests that
the distinguishing characteristic of markedness is its relationship with affect.
Once more when dealing with pragmatic topics, we see how the emotional
context is crucial for the interpretation of a phenomenon: the unmarked relational

60
Politic behavior is defined by Watts (2003: 19) as "linguistic behavior which is per-
ceived to be appropriate to the social constraints of the ongoing interaction, i.e. as
non-salient".
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I/ I I I I I I I I I I I Politeness and Impoliteness 169

options, which tend to display a neutral emotion, are contrasted with the
positively and the negatively marked, which involve a negative or a positive
emotion. This is precisely the main topic touched on in Culpeper et al. (2014), as
seen from a cross-cultural perspective.
In summary and to conclude, it can be said that, in the view of many
specialists on the topic, the relational framework offers an all-embracing
perspective, covering not only politeness and impoliteness, but also other kinds
of relational practice (such as politic behavior). Holmes & Schnurr (2005: 124)
indicate that one of the advantages of treating the phenomenon as relational
work is that it avoids "the definitional traps, referential slipperiness, and
emotional baggage of the term 'politeness"'. But whatever the approach taken, it
is clear that (im )politeness constitutes one of the key topics for understanding
the pragmatics of any given language, culture, or social group.

4.4. CONCLUSION
The body of work presented in this chapter constitutes one of the major
achievements of pragmatics to date. Half a century ago, the notion of politeness
was only associated with good manners. If linguists or dictionary-writers
mentioned it at all, they did so in order to warn their readers how to avoid rude
or inappropriate language. It was in the wake of Grice's insights into the principles
underlying conversation that linguists gained a radically new understanding of
how politeness is ruled by principles and maxims analogous to those identified
for conversation in general. Using the methodology and techniq,ues of pragmatics,
linguists came to see that politeness was central to the strategies deployed in
interaction, and various theories were developed to account for those aspects of
linguistic behavior that follow from such notions as face, rationality, strategic
options, self-effacement and respect. It was only a matter of time before linguists
developed an eq,ually strong interest in what had been mistakenly taken to be
merely absence of politeness, namely impoliteness. Authors from various cultures
have now looked into different aspects of impolite behavior, but for the purposes
of this book we have placed special emphasis on the pioneering work of the
English-speaking and Spanish-speaking worlds, where ways have been found to
classify and understand impolite language behavior as a much more complex
phenomenon, from a sociocultural viewpoint which involves different shades and
types of relational work.
The next chapter will turn to another of the great achievements of
pragmatics, the development of Relevance Theory.
170 Politeness and Impoliteness //// /////////////// /////////// ////////// ////

SUMMARY (CHAPTER 4)
1. The study of linguistic politeness and impoliteness is a prominent aspect of
current research in pragmatics. Since what counts as (im)polite is strongly
determined by the wishes, communicative goals and emotional states of the
interlocutors and by the context in which they are interacting, pragmatics
constitutes an ideal framework for the analysis of politeness phenomena in
language.

2. Politeness has been studied within linguistic pragmatics since the 1970s. Eight
main views can be distinguished in the literature: the social-norm view, the
conversational-contract view, the conversational-maxim view, the face-
saving view, the emotive perspective, the relational perspective, the frame
approach, and the rapport management view. The most influential have been
the conversational-maxim and the face-saving views.

3. The conversational-maxim view takes as its major inspiration Grice's

Cooperative Principle (CP) and the associated Maxims. The first prominent
advocate of this view was Lakoff (1973), who proposes Rules·of Pragmatic
Competence, the second of which ('Be polite') is divided into three Rules of
Politeness, covering a scale Formal-Informal-Intimate.

4. The major exponent of the conversational-maxim view is Leech (1977, 1983)

who formulates a Politeness Principle that is complementary to the CP and
embraces six Maxims of Politeness. For Leech, a distinction needs to be
drawn between relative and absolute politeness. As examples of the latter, he
argues that certain illocutionary acts are inherently polite, while others are
inherently impolite and therefore need to be tailored for politeness. In his
2014 book he revises these concepts by including them within what he calls a
pragmalinguistic politeness scale and a sociopragmatic politeness scale
respectively.

5. In his last publication, Leech (2014) returns to politeness, putting forward a

General Strategy of Politeness which embraces the various Maxims. This
book emphasizes the sociopragmatic nature of politeness, examining the way
in which societal norms and communicative context impinge upon politeness
strategies.

6. To counterbalance criticisms of Leech's work as being too Anglocentric

Yueguo Gu (1990, 1997) puts forward an alternative Balance Principle
associated with four Maxims, which together reflect traditions and practices
of politeness in Chinese. His work also sets great store by the concept of
'face' that is central to the work of Brown & Levinson (B&L).
///////I// I I I I I I I I// I I I! I!// I I//// I I////// I/ Politeness and Impoliteness 171

7. The face-saving view, represented above all by B&L ((1978] 1987), takes face
and rationality as the basis for an understanding of politeness. Politeness is
seen as a rational strategy for achieving one's desires without offending the
other's or one's own face. Face is defined as incorporating two 'wants':
the desire to be unimpeded (negative face) and the desire to be approved of
(positive face).
8. Both negative and positive face can be threatened by a whole range of
illocutionary acts. These are known as Face-Threatening Acts (FTAs), which
are classified as threatening to the hearer's negative face, the hearer's
positive face, the speaker's negative face or the speaker's positive face.
Politeness in this view involves using language to make an FTA less
threatening, either through negative politeness and/or through positive
politeness.
9. B&L distinguish five strategies, ordered according to the speaker's estimate
of the loss of face involved: FTA without redressive action ('bald on record'),
FTA with positive politeness, FTA with negative politeness, FTA off record,
and abstaining from doing the FTA. In addition, they recognize three
contextual variables that together set the background for the operation of
these strategies: D(istance), P(ower) and R(anking) (of imposition in the
particular culture). They propose an eq,uation for assessing the weightiness
of an FTA as a function of the three variables.
10. Both Leech's and B&L's work, despite claims especially by the latter to have
proposed a universally valid system, have been criticized for a failure
to account for politeness in non-Western cultures. In addition, it has been
argued (Bravo 1999) that in Spanish-speaking cultures values such as
autonomy and affiliation play the role taken by considerations of face in the
data considered by B&L. Finally, Leech and B&L have both encountered
criticism for excluding impoliteness from their proposals.
11. In pragmatics, impoliteness is now not seen as the absence of politeness but
as a use of language sparked off by a range of emotions. It has been studied
since the nineties, notably for English by Culpeper and for Spanish by Kaul
de Marlangeon, in a series of works by each from the mid-nineties to the
present. Culpeper reacts to B&L's Politeness Theory by distinguishing five
kinds of impoliteness: bald on record impoliteness, positive impoliteness,
negative impoliteness, sarcasm or mock politeness, and the withholding of
politeness.
12. Kaul de Marlangeon provides a classification of impolite acts as formally
impolite with a polite purpose, involuntarily impolite (gaffes, involuntary
stints, involuntary omissions), self-impolite, formally polite but impolite,
172 Politeness and Impoliteness / / / ///////I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I

voluntary stints, overwhelming silences, or involving fustigation

impoliteness.
13. Issues that have occupied impoliteness researchers in recent times have
been the q,uestion whether rudeness is to be distinguished fro m
impoliteness and the possibility that excessive politeness can be
experienced as impolite, giving rise to the notion of 'politic behavior
(Watts 2003) lying between extremes of politeness and impoliteness.
Within this framework, politeness and impoliteness are found within the
spectrum of relational work, where appropriate behavior is unmarked,
and inappropriate behavior is marked.
IIIIIIIIIII III IIIII II IIIIIIII IIII III IIIIIIII Politeness and Impoliteness 173

SELF-EVALUATION QUESTIONS (CHAPTER 4)

Choose the most appropriate answer for each q,uestion.
1. Politeness in linguistics is mainly concerned with ...
a) social manners.
b) appropriateness.
c) correct grammar.
2. There is/are ... to the phenomenon of politeness.
a) only one approach
b) four approaches
c) various approaches
3. The most influential approaches to politeness have been ...
a) the social-norm view and the conversational-contract view.
b) the frame approach and the emotive approach.
c) the face-saving view and the conversational-maxim view.
4. According to Lakoff ( 1973), if a speaker has to choose between politeness
and clarity, she will choose ...
a) politeness.
b) clarity.
c) neither of the two.
5. With the concept of absolute politeness, Leech (1983) refers to his view that ...
a) some illocutionary acts are inherently polite or impolite.
b) politeness varies according to specific situations.
c) there is a fixed number of politeness maxims.
6. Leech (2014) is in favor of a view of politeness that is ...
a) completely relativistic.
b) completely universalistic.
c) in between the relativistic and the universalistic views.
7. Gu (1990, 1997) and Ide (e.g. 1989, 1993) are authors who have emphasized ...
a) the similarities between Eastern and Western cultures.
b) the contrast between the politeness behavior in the Eastern and the
Western worlds.
c) their opposition to the Grice an approach to Politeness.
174 Politeness and Impoliteness I I I I I I I I I I I I I I I/ I I I I I I I I I I I I I I I I I I I I I I I I I I I I

8. In Brown & Levinson's (1987) view, negative face has to do with ...
a) the hearer 'wanting the wants' of the speaker.
b) the want of every human being that their actions be unimpeded.
c) the desire of the speaker to please her interlocutor.

9. What type of politeness strategy (following B&L's taxonomy) is used by

Anthony in the following situation?
Anthony is having a conversation and celebrating his birthday with his
friends in his house when all of a sudden he hears a strange noise, so he says:
"Shhhh!! Stop talking!"

a) Off record
b) Bald on record
c) On record with negative politeness

10. What type of politeness strategy (following B&L's taxonomy) is used by

Carmen in the following situation?
Carmen is a bit annoyed because her husband never takes the initiative to go
out or travel to different places, so when her husband comes back from work
one day and asks her if dinner is ready, she replies: "I wish I had an
adventurous friend to enjoy life with!"
a) Off record
b) On record with positive politeness
c) On record with negative politeness

11. Which of the following combinations of the sociological variables (P, D & R)
would be the most probable in the following exchange, in a situation in which
an English Engineer (A) is talking to his friend (B), who nevertheless happens
to be his boss:
A: Will you be coming to the office tomorrow morning?
B: Yes, why?
A: I just wanted to invite you for lunch.

a) Low D, low P, and high R

b) Low D, high P, and high R
c) Low D, high P, and low R
III II IIIIII IIIIIIIIII IIIIIII III II IIIIIIIII II Politeness and Impoliteness 175

12. B&L's politeness theory was criticized for its ...

a) relativistic view.
b) claim to uniq,ueness.
c) claim to universality.

13. Following Culpeper's (1996) taxonomy of impolite acts, say what type of
impoliteness is found in B's reply, in a situation in which she knows that A is
lying to her:
A: Believe me, I have never cheated on you.
B: Yeah, right!!!

a) Sarcasm or mock politeness

b) Bald on record impoliteness
c) Positive impoliteness

14. Following Kaul de Marlangeon's (2008) taxonomy .of impolite acts, say what
type of impoliteness is found in B's reply, in a situation in which A is very
angry at her friend:
A: What's the matter with you today???
B: Shut up and get the hell out of here, NOW!!!

a) Fustigation impoliteness
b) Overwhelming silence
c) Formally polite act with an impolite purpose

15. Within the relational work perspective on (im)polite behavior, over-polite

behavior is ...
a) positively marked.
b) negatively marked.
c) neutral.
176 Politeness and Impoliteness / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / //

DEVELOPING OUR PRAGMATIC COMPETENCE:

EXERCISES FOR REFLECTION AND AMPLIFICATION
(CHAPTER 4)
A) Include in the second column at least one author representing each
one of the politeness views in the first column of this table:

Social-norm
Conversational-contract
Conversational-maxim
Face-saving
Emotive
Relational
Frame
Rapport-management

B) Considering Lakoff's (1973) two Rules of Pragmatic Competence, i.e.

Be clear and Be polite, which of the two do you think the speakers are
giving preference to in each of these two situations?
a) Sarah has bought a new dress and shows it to her friend, who does not like
it at all.
Sarah: I've put on my new dress for you to see ... Do you like it?
Cristina: Yeah .. .It looks nice on you.
b) Tim has been singing in a very loud voice all morning, while his father has
been trying to concentrate on a paper he is writing:

Tim: lalalaaaaaaala .$t~-..::;,--~

"~ lalaaa
---
Tim's dad: Cut it out! No more singing! I have to work

C) Taking into account Leech's (2014) General Strategy of Politeness,

specify which of its component maxims are illustrated by the following
examples:
a) A woman to her partner: How can I be stupid enough to have forgotte n
that today is your birthday???
IIIIII IIIII III IIIIIIIIIIIIIII I III II III II II II Politeness and Impoliteness 177

b) A young woman to her friend: If I were you, I would break up with John,
but of course, it's completely up to you
c) A man to his neighbor: I'm very sorry about the noise my son is making
with his guitar. He is rehearsing for a concert, but we'll try not to bother
you too much.
Neighbor: It's O.K. Don't worry.
d) A woman to her boyfriend on his birthday: Today I won't let you do any
chores at all in the house. I'll treat you by cooking your favorite dish, and
will give you one of my world-famous massages. Steel yourself!
e) A woman to her friend, after receiving the news that her friend's father has
died: Oh, dear, I'm very sorry about this great loss in your life; you know
you always have a friend here, and if you need company or just someone
to talk to, just call me or come to my house.

D) Following Leech's view of politeness, and on a pragmalinguistic scale,

arrange the following utterances in order, from most polite to least
polite:
a) Shut the door, you blockhead!
b) If it is not too much of a bother, would you be so kind as to shut the door,
please?
c) Would you mind shutting the door?
d) Shut the door, will you?

E) Given the following situations and utterances (in bold), specify the
politeness strategy being used by the corresponding speakers, taking
into account Brown & Levinson's taxonomy of politeness strategies.
In the case of off record strategies, specify the Maxim being flouted
and the implicature being triggered.
a) A woman to her husband: Could you please see what's wrong with the
printer? It won't print and I need to finish this project.
b) John to his friend Sammy, who has misbehaved in class, when he sees the
Headmaster approaching: Run, Sammy, run!!
c) Lucy to her friend Susan: You're such a talented person, Susan. I
admire you so much!
d) Cecily to her friend Jane, showing annoyance at Jane's self-destructive
attitude: There you go again! I love it when you are so masochistic!!
178 Politeness and Impoliteness / / / / / / / / / / / / / / / // / / / / / / / / / / / / / / / / / / / / / / / / / / /

e) A: Will you tell me the truth about your relationship with Mary?
B: Nice weather today, don't you think?
f) A woman to her neighbor: Sorry to bother you with this, but would it
be too much if I asked you to lend me your mixer? Mine has broken
and I have to finish the dessert I'm preparing for dinner.

F) Provide 5 examples of utterances that could threaten the face of your

interlocutor in some way, and propose an alternative that minimizes
the face threat, specifying the strategy of politeness being used
(following B&L's approach).
E.g.:

I don't like that dress. Perhaps you should try Off record (Give Hints)
Put on something else and wear a different
for the party. dress to the party,
sweetie.

G) Given the utterances in the first column, provide a possible formulation

of the weightiness of the FTA by assigning a value to the three
sociolinguistic variables P, D, and R:

E.g.: Tom to his sister: Pass me that Low D, low P, low R

book!

1) A woman to a stranger in the street:

Excuse me, but could you possibly
change me this $10 bill? I need
coins for the parking meter and I
don't have any change.

2) A woman to her boy-friend: C'mon,

tell me about your previous
girlfriend. "What was she like?

3) Jonathan to his colleague at work:

Look dude, I'm having some
financial problems at the moment.
Do you think you could lend me
$20,000?
I I I I I I I I I I I I I I I I I I I I I I I I I I I I! I I I I! I I I I I I I I I I Politeness and Impoliteness 179

H) Given the following impolite utterances or situations (in bold),

indicate the strategy that the speaker is likely to be using in each
case, according to Culpeper's (1996) taxonomy of impolite acts:

1) In the elementary school

playground:
Martha (to a group of
classmates): May I play
hopscotch with you?
Lorna: No, we don't wanna
play with you!

2) An exchange between
workmates:
Richard: Don't do that, George,
it's not an honest thing to do!
George: You of all people are
saying this to me ... YOU ... who
were convicted in a famous
corruption case some years
ago!!

3) Julia is angry with her friend

Cynthia, who ignores this
circumstance and when Julia's
birthday comes, she goes
to visit her and gives her a
present. Julia does not thank
or even talk to her friend.

4) A couple are having an argument:

She: How could you do that to
me?
He: What are you talking about?
I haven't done anything to you!
She: It's incredible how
irresponsible and cheeky you
can be. Get the hell out of my
sight!!
180 Politeness and Impoliteness //////////// ////////////////// //////////////

I) Try to find instances of impolite acts in the following extract from the
movie Saltwater (U.K., 2000), and classify them using Kaul de
Marlangeon's categories of impoliteness.

Academics and students are gathered in a room drinking wine and eating
sandwiches. Ray sees Trish Meehan coming over to him. He turns away and
lowers a glass of wine, picks up another one.
TRISH: That's it. Don't mind me. A little drink in moderation never hurt anyone.
Rtw: Why the hell do you always look at me like I just gingerly slipped a wasp
up your crack?
TRISH: Because I know what you're like, maybe.
RAY: Look, I know, it's a terrible thing. You've gone through your whole life, no
one'll come near you with a ten-foot pole. But it's nobody's fault. It's not the
fault of mankind, Trish. Let it go.
TRISH: I'm going to be a little bit more specific. I had a student visit me today.
Very concerned about the relationship between you and a third year
Deborah McCeever, being inappropriate.
Ray sees Deborah
TRISH: Tell me about you and Deborah McCeever. Oh, that got your attention.
What are you playing at?
RAY: You have no idea what you're talking about
TRISH: Oh really? Then why has one of her classmates reported you?
RAY (helpless): What do you want me to do? What would make you happy in
this situation, Trish?
TRISH: Sort it out. Ray. Don't make me tell Tony.
Ray regards Trish, then he lowers his glass of wine.
RAY: I'm going to do it for you. Do you know why?
TRISH: Why?
RAY: Because you may not have a nice appearance or demeanour, or .. .
anything, you have a fantastic little arse thing going on there. And it just
about saves you.
TRISH: There's nothing you can say to offend me, Ray. Because I think your
opinion is worthless.

J) Search the web or look for any text (literary or of any kind) where
people or characters display different kinds of impolite behavior and
classify the impolite acts identified according to either Culpeper's
(1996) or Kaul de Marlangeon's (2008) taxonomies.
 RELEVANCE:
PROCESSING EFFORT
AND CONTEXTUAL EFFECTS
OF COMMUNICATION
I/////////////////////////////////////
MAIN OBJECTIVE OF THE CHAPTER: To introduce Relevance Theory
(RT) as a prominent approach to communication, to explain the principal
notions used in RT and to determine its contribution to the understanding
of the roles of context and cognition in human interaction.

MAIN TOPICS OF THE CHAPTER:

• The Principle of Relevance ·
• The presumption of optimal relevance
• Ostensive, inferential and ostensive-inferential communication
• Processing effort and contextual effects
• Accidental, covert and overt communication ...
• Coding, encoding and decoding
• Implicature and explicature
• Salience
• Images and RT
182 Relevance: Processing effort and contextual effects of communication /////////

5.1. RELEVANCE THEORY

This chapter will focus on the argument set out by Dan Sperber and Deirdre
Wilson in their book Relevance: Communication and Cognition (1986) 61 that the
cognitive notion of relevance is crucial for communication. As Wilson & Sperber
(2007: 607) explain in a later article, their theoretical account may be seen as an
attempt to work out in detail two of Grice's (1989) central claims:
l) The claim that an essential feature of most human communication is the
expression and recognition of intentions. In so claiming, Grice laid the
foundations for an inferential model of communication according to which a
speaker provides evidence of her intention to convey a certain meaning
which is inferred by her interlocutor(s) on the basis of the evidence provided
2) The claim that utterances automatically create expectations which guide the
hearer toward the speaker's meaning. As we saw in Chapters 3 and 4, Grice
described these expectations in terms of the Cooperative Principle and its
Maxims (Quality, Quantity, Relation and Manner).

Sperber & Wilson fully agree with the first claim, but only partially with the
second one, because they contend that the Principle of Relevance is sufficient for
the process of interpreting and understanding utterances and therefore that there
is no need to bring any other of the Grice an Maxims into the picture for the
comprehension of that process. Relevance is the only principle that matters when
it comes to interpreting a message, and so they assume that speakers have
intuitions of relevance, which means that they can distinguish relevant from non-
relevant information, and that these intuitions are relative to contexts.
By way of illustration, when Johnny says (1) to his mother, she will take into
account that he already knows she is going to work and how long she normally
stays there, which will allow the mother to work out that the most relevant way
to understand his q_uestion is that he expects her to be back by lunchtime so she
can cook for and have lunch with him.
(1) Mom, will you be back at midday?
This example is just one of the many instances in which speakers do not
have any difficulty working out the most relevant interpretation of
underdetermined utterances, because there is obviously a close relationship
between Johnny and his mother, which allows her to know all the conditions
under which the utterance is made, and therefore to handle the context. Thus we
can say that Johnny's utterance is relevant to his morn for that matter. If, however,

61 This original version was updated in Sperber & Wilson (1995, 1998, 2002) and Wilson &
Sperber (2002).
IIIIII III Relevance : Processing effort and contextual effects of communication 183

the same q,uestion were asked his mother by a complete stranger, she would most
probably lack the contextual information to interpret the q,uestion with accuracy,
and therefore the q,uestion would tum irrelevant - or at least, much less relevant
than in ( 1) - because she would need more such information to understand why
this complete stranger is asking such a q,uestion. In fact, according to Sperber &
Wilson, the greater the difficulty in working out what someone means, the less
relevant the speaker's utterance will be for the hearer.
RT not only accounts for the process of recovering implicatures of the kind
Johnny's mother recovered from her son's q,uestion, but also for the kind of lexical
enrichment needed in utterances such as the one found in example ( 4) in
Chapter 1, partially reproduced here in (2), where the necessary contextual effect
is for the customer to realize or know that the cashier is asking about the kind of
bag that he wants for his groceries, and not, for instance, about the kind of money
he is expected to pay with.
(2) Cashier: Paper or plastic?
In effect, Sperber & Wilson explain that an assumption is relevant in a
context if and only if it has some contextual effect in that context (1986: 122).
Relevance is used by these authors as a technical term to describe the degree of
cognitive effort req,uired for an addressee to achieve an understanding of what
a speaker means by what she says. In the next section we shall examine these
concepts (contex tual effects, assumptions or cognitive effort) in more detail, as
fundamental notions for the comprehension of the Principle of Relevance.

5.2. THE PRINCIPLE OF RELEVANCE

Sperber & Wilson (hereinafter S&W, or W&S, depending on the publication) base
their theory on the common-sense assumption that human cognition tends to
focus attention on certain aspects of the world, and so by drawing our audience's
attention to some kind of stimulus, we intend to be relevant in all our exchanges.
In this line of thought, these authors argue that relevance is a much more
pervasive principle than Grice's Cooperative Principle. Hence the Principle of
Relevance basically tells us that an act of expressing something carries a
presumption of relevance, and this fact makes manifest the intention behind the
utterance.
In order to sort out the incompleteness or indeterminacy of all linguistically
coded messages, contextual information is needed, and because of the
presumption of relevance, the hearer will choose the context involving
the greatest contextual effects and the least processing effort. More than one
1g4 Relevance: Processing effort and contextual effects of communication /////////

mental process is involved in this act, and therefore W&S (2007) present the
Principle of Relevance in two different formulations:
1) Cognitive Principle of Relevance: "Human cognition tends to be geared to the
maximization of relevance" (2007: 610). Inferential communication takes
place against this cognitive background.
2) Communicative Principle of Relevance: "Every ostensive stimulus conveys a
presumption of its own optimal relevance" (2007: 612).
An act of ostensive communication is an act of deliberate, overt
communication by means of which the speaker not only intends to convey a
particular message, but also actively helps the hearer to recognize this. In order
to understand ostensive messages or stimuli, hearers engage in acts of inferential
communication, which constitute the other side of the same process. Hence the
speaker is involved in ostension, while her audience is involved in inferencing.
The use of an ostensive stimulus creates a presumption of optimal relevance, a
concept that W&S (2007: 612) explain in the following manner:
Presumption of optimal relevance
a. The ostensive stimulus is relevant enough to be worth the audience's
processing effort.
b. It is the most relevant one compatible with communicator's abilities and
preferences.
From these two clauses it is derived that, since the communicator ~ ants to
be understood, it is in her interest to make her ostensive message as easy as
possible for her audience to understand, as well as to provide evidence for the
cognitive effects she aims to achieve. For instance, in (3) the teacher's goal is to
inform his students that they are having a History exam the following week, and
he finds that the most effective way of achieving this goal is by offering more
information about the specific topic to be covered, which given clause (a) will
hold his class's attention, and will also make them reasonably infer that no topics
other than WWII will be included in the exam, because if other topics were
included, the teacher should have said so, given clause (b) of the presumption of
optimal relevance: 62
(3) Mr. Watson: Next week I will be giving you an exam covering WWII.
Thus, there are two main factors that govern the degree of relevance of an
ostensive act: contextual effects and processing effort. S&W (1986: 118) explain
that the notion of a contextual effect is essential to a description of the

62 See also the connection of this observation with the concept of scalar implicature in
Chapter 3.
I I I I I/ I I I Relevance : Processing effort and contextual effects of communication 185

comprehension process, and it is explained by the fact that the hearer processes
a number of assumptions as the discourse proceeds, assumptions which form a
gradually changing background against which new information is processed.
"Interpreting an utterance involves working -out the conseq,uences of adding
this assumption to a set of assumptions that have themselves already been
processed" (1986: 118).
Cruse (2004: 383) points out that there are four kinds of contextual effects:
a) adding new information, b) strengthening old information, c) weakening old
information, and d) canceling old information. A new fact which triggers many
new inferences, if taken together with information already known, will probably
be worth processing. Contrariwise, a new fact which has no connection with
anything already known is probably not worth processing. Processing effort,
then, is not only a function of the linguistic complexity of the utterance itself, but
also of the cost of accessing and using contextual assumptions in the derivation
of contextual effects. The less effort it takes to recover a fact the greater its
relevance. Consider (4), an utterance said by an officer on a train, where - other
things being eq,ual - the most relevant interpretation, and therefore the one
involving the least processing effort given our background knowledge and
experience of what normally happens on trains, will be that the officer wants to
see the passengers' tickets to check that there's no one in the carriage that has
not paid for their trip. A more far-fetched interpretation for this utterance, and
therefore a much less relevant one, would be that the officer wants to buy
a ticket, because this would have no connection with anything a ready known,
and therefore would not be worth processing.
(4) Tickets, please!
As the above considerations and examples indicate, the Principle of
Relevance has not only to do with linguistic communication, but also with
complex cognitive processes; that is why we find it necessary to examine this
relationship in the coming section.

5.3. RELEVANCE AND COGNITION

We presented the Cognitive Principle of Relevance in 5.2 in anticipation of the
assumption of RT that utterances are expected to be relevant because the search
for relevance is a basic feature of human cognition which communicators
normally exploit. Any type of input (e.g. an utterance, a sound, a memory) is
relevant to an individual when its processing yields a positive cognitive effect.
For instance, the confirmation of a contextual implication as a true conclusion
186 Relevance : Processing effort and contextual effects of communication /////////

constitutes a positive cognitive effect. False conclusions are also considered to

be cognitive effects, but not positive ones. According to W&S, the most important
type of cognitive effect is a contextual implication, i.e. "a conclusion deducible
from input and context together, but from neither input nor context alone" (2007:
608). An example of contextual implication can be found in Ann's final utterance
in (5), which shows that Ann comes to the conclusion that Lucy will not be coming
for dinner, on the basis of the input she has, as well as of the context (it's already
7 o'clock and she hasn't called):
(5) Ann: Is Lucy coming for dinner?
Ben: She said that we shouldn't wait for her if she didn't call before 6:00 p.rn.
Ann: It's 7:00 p.m. and she hasn't called. I don't think we should wait any
longer, then.

But there are other types of cognitive effects different from contextual
implications, such as the strengthening, revision or abandonment of available
assumptions. For example, in (5) the fact that Lucy is not coming for dinner may
confirm Ann's impression that Lucy is not interested in meeting them, or make
Ann change her previous plan of laying the table with her finest china. These are
positive cognitive effects that constitute the necessary condition to make Ann's
input relevant.
An important aspect to remark is that there are infinite potentially relevant
inputs, but humans cannot attend to them all. There is a mass ·of competing
stimuli, and what makes one stimulus worth picking is not just that it is relevant,
but that it is more relevant than any other alternative. Thus, the greater the
positive cognitive effects obtained after processing an input, the more relevant
the input will be. And, as we anticipated in 5.2., the greater the processing effort,
the less relevant the input will be. W&S (2007: 609) formulate the relevance of an
input to an individual in the following manner:
a. Other things being eq,ual, the greater the positive cognitive effects achieved
by processing an input, the greater the relevance of the input to the individual
at that time.
b. Other things being eq,ual, the greater the processing effort expended, the
lower the relevance of the input to the individual at that time.
To illustrate this, consider the three utterances in (6), as possible responses
by Myriam to her friend, who has called to ask her if she will be free on Friday
(because if so, her intention is to invite her to the movies):
(6) a. I'm free on Friday, yes.
b. I don't have to work on Friday.
c. The answer is positive, unless (2 - 3)(2- 2) = (-1) x Ois not 0.
/II I I I I I/ Relevance: Processing effort and contextual effects of communication 187

It is obvious here that (6a) is the most relevant response because it is the one
with the greatest positive cognitive effects and req,uires the least processing
effort, while (6b) is less relevant than (6a) because Myriam's friend will have to
work out an implicature in order to conclude that Myriam is free on Friday, and
therefore, it is less direct than (6a), but more relevant than (6c). (6c) is the least
relevant, because in order to deduce a conclusion from the input, Myriam's friend
would have not only to work out an implicature, but also first to solve an eq,uation,
and therefore it would req,uire the greatest processing effort. The three utterances
will derive the same cognitive effects, but these are easier to derive from (6a)
than from (6b), and more difficult to derive from (6c) than from (6b).
The conclusion to be drawn from this section is that speakers and hearers
maximize the relevance of the inputs they process simply because they make the
most efficient use of the available processing resources, and this is the reason
why RT claims that we have an automatic tendency to maximize relevance
because of the way our cognitive systems have evolved, and not because we
have any choice in the matter. This cognitive tendency 63 to maximize relevance
is thought to be universal by S&W, and it makes it possible to predict and
manipulate the mental states of others to a certain extent when engaged in
communication.

5.4. THE NOTIONS OF COMMUNICATION AND

CONTEXT WITHIN RT
Within the framework of RT, understanding is achieved when the communicative
intention is fulfilled, i.e. when the hearer or audience recognizes the speaker's
informative intention. A crucial element for this recognition is context. Thus the
concepts of communication and context are fundamental to the development of
this theory.
Wilson (1994) distinguishes three types of communication: 1) Accidental,
2) Covert, and 3) Overt. An example of accidental communication may be a case
in which a speaker is overwhelmed by some kind of negative emotion (like anger,
sadness or mourning) and she fails to conceal it while, for instance, she is giving
a professional talk, engendering a feeling of discomfort in the audience. Covert
communication is the kind of communication where intended intentions are
backgrounded, and overt communication is that in which they are foregrounded.

63 See also Escandell-Vidal (2009) or Scott-Phillips (2010) on the biological evidence for
this allegedly universal cognitive tendency.
188 Relevance: Process ing effort and contextual effects of communication / / / / / / / //

We have seen many examples of covert communication in the preceding chapters

of this book, when we discussed indirect speech acts or different types of
inferences. A further example for the reader to consider here could be that of a
history teacher who, when teaching about a given period in the history of her
country, does her best to show her position towards the historical facts without
actually saying what her political ideas are, because she will be more successful
in her attempt to express her intentions (which may be convincing her students
that she is right) only if she hides them. In contrast, and as its name implies, in
overt communication the intended intentions are made explicit, and to all
purposes , understanding will involve recovering the overtly intended
interpretation.
Thus, speakers always intend to make some particular assumptions manifest
to their hearers. This is what S&W (1986: 49) call ostensive communication or
ostension. The act of simply showing something to someone is a case of
ostension, and this is true, too , of intentional communication. Ostensive
communication is, therefore, communicative behavior which is planned to draw
someone's attention. But this communicative behavior does not necessarily have
to be linguistic. Sometimes communicators transmit their intentions by means
of non-linguistic noises, gestures or movements which make their audience
infer what they are thinking or feeling. Taking this fact into account, as well as
the knowledge gained in previous chapters from other approaches ( e.g. Grice
1967), a distinction could be made among three main types of communicative
behavior:
1) Coded behavior. the most straightforward form of communication. If I want
you to come to my house for dinner, I simply and directly say to you: "I invite
you to my house for dinner tonight".
2) Coded behavior that needs to be assisted by inference: This is the case of any
of the uses of indirect language we have studied in previous chapters. For
instance, if my interlocutor replies to the above invitation as follows: "I
promised my aunt that I would visit her this evening", the non-acceptance of
the invitation has to be deduced by inference, which in this case involves a
conversational implicature.
3) Pure inference: This kind of communicative behavior does not resort to
language. Consider a situation in which Jack is on the phone and is at the
office with Jill, his workmate. A gust of strong wind blows into the room and
so Jack (who cannot interrupt his telephone conversation) visually calls Jill's
attention by pointing to the window to make her infer that he wants her to
close it.
Obviously, communicative types 2 and 3 differ from type 1. The distinction
between them is that between linguistically decoded meaning and pragmatically
//////II I Relevance : Processing effort and contextual effects of communication 189

inferred meaning. Hence in the first case we would be talking of 'purely' linguistic
behavior (if this is at all possible) and in the other two of pragmatic behavior.
However, in RT the identification of explicit content is seen as eq,ually inferential
and eq,ually guided by the Principle of Relevance as that of non-explicit content:
"The relevance-theoretic comprehension procedure applies in the same way to
resolving linguistic underdeterminacies at both explicit and implicit levels"
(W&S 2007: 625).
Thus, S&W (1986) apply the term communication to all cases of ostension,
and therefore they treat ostensive communication, inferential communication,
and ostensive-inferential communication as fundamentally the same thing.
Inferential communication and ostension are one and the same process, only
seen from different points of view: the communicator is involved in ostension,
while the audience is involved in inference. Ostensive-inferential communication
comprises, then, two layers of information to be picked up: 1) the informative
intention and, 2) the communicative intention, i.e. the intention to inform the
audience of one's informative intention. This may sound like a tongue-twister, but
it simply points to the fact that understanding is achieved not only when the
information is given, but also and necessarily when the communicative intention
is fulfilled, i.e. "when the audience recognises the informative intention" (W&S
2007: 611). And a crucial notion for the achievement of understanding, which is
fundamental to RT, is - as was noted at the beginning of this section - the notion
of contex t. Whereas many previous researchers (e.g. Brown & Yule 1983, Hymes
1964) viewed context in a more traditional way, as a less dynamic background
against which utterances could be understood, S& W say that the inferential
process is a matter of creating a context in which the communicative act achieves
relevance. The role of linguistic form in this creation of context is to limit ( or
constrain') the possible contexts that the addressee considers.
We shall try to illustrate all these concepts by means of the situation in
example (7):
(7) Larry comes across his friend Sally on the street. Sally is with her beautiful
baby daughter, Trisha, whom Larry has never seen before, and after she is
presented to Larry, he says (winking): "My goodness! This girl is as ugly as
her mother!"
Sally's challenge is to correctly identify Larry's communicative intention.
The purely linguistic information suggests that he is insulting both the baby and
her mother, but Sally also bears in mind (i.e., entertains the assumptions) that
Larry intentionally winked his eye, that he and Sally are old friends who feel
affection for each other, that Sally knows Larry is a very good-natured man who
could never intend to insult, that 'positive irony' is a conventional, harmless and
amusing form of communication among people who know each other well, etc.
190 Relevance: Processing effort and contextual effects of communication ////////

On this basis, the most relevant interpretation is that Larry's communicative

intention is actually to praise both of them (meaning that they are both beautiful).
In such a case, it can be said that Sally creates a context in which Larry
informative intentions can be recognized, and therefore that his intention (of
praising mother and baby) was fulfilled.

5.5. THE COMPREHENSION PROCESS

In his attempt to achieve comprehension, the hearer's goal is to construct a
hypothesis about the speaker's communicative intention that satisfies the
presumption of relevance conveyed by her utterance. W&S (2007) explain that
this general task can be broken down into a number of subtasks which should not
be seen as seq,uentially ordered, but as carried out in parallel, each involving an
inference process which is embedded in the overall process of constructing a
hypothesis about the speaker's meaning.
Subtasks in the overall comprehension process
a. Constructing an appropriate hypothesis about explicit content
(EXPLICATURES) via decoding, disambiguation, reference resolution, and
other pragmatic enrichment processes.
b. Constructing an appropriate hypothesis about the intended contextual
assumptions (IMPLICATED PREMISES).
c. Constructing an appropriate hypothesis about the intended contextual
implications (IMPLICATED CONCLUSIONS). (2007: 615)
According to this model, then, the comprehension process comprises two
main kinds_of constructed hypotheses: those having to do with explicatures, and
those related to implicatures. We shall now explain these concepts.

5.5.1. IMPLICATURE AND EXPLICATURE

S&W make a distinction between the assumptions that are explicated or explicitly
communicated, and those which are implicated or implicitly communicated.

5.5.1.1 Explicafure
An explicature is an enrichment of the information that is explicitly encoded in
an utterance to a fully elaborated propositional form. Consider the explicatures
I I I/// I// Relevance : Processing effort and contextual effects of communication 191

for the different utterances in (8), which obviously help in the construction of the
hearer's hypothesis, which contains a combination of the processed assumptions,
including both the explicated and the implicated meaning:

(8)
Utterance Explicature (Simplified) possible
hearer's hypothesis
(including both explicature
and implicature)

a. (On the train): Could you show The ticket inspector wants to
Inspector: Tickets, me your tickets, see my ticket as proof
please! please? of payment.

b. (My neighbor sees Are you having My neighbor wants to know

I'm struggling with any trouble if I am having trouble in
the lock of my front opening your opening the door because he
door): door? wants to help me.
Neighbor: Any trouble
with the door?

c. (Fred introduces This is Jessica. Fred is introducing Jessica,

Jessica to his college She has been who has been his friend for
friends, expecting my friend for many years now, and for
them to accept her in many years now that reason he wants us to
the group): (rather than "My consider her as one of us.
Fred: This is Jessica. friend is old")
She is an oldfriend
of mine.

Because explicatures have to be complete from a logical point of view, a

relevance-driven process of elaboration is req_uired in order to get them from
a given utterance. Cruse (2004) illustrates four aspects of explicature in which
the Principle of Relevance guides the recourse to inference. These four aspects
are the following:

1) Disambiguation: We normally do not notice ambiguities in language because

the context and the Relevance Principle guide the process of disambiguation.
Consider the utterances in (9):

(9) a. I feel like eating fruit today. Give me a date, please!

b. Tim took Lisa out on a date.
c. What's the date of your exam?
192 Relevance: Processing effort and contextual effects of communication / // / / / ///

In these three examples, the information contained in the utterances, together

with the Relevance Principle, play an important role in disambiguating the
possible ambiguity found in the word date. In (a) the hearer will be
predisposed to select the reading "a type of fruit" because of the presence in
the linguistic context of the verb "eating" and the noun "fruit" (among other
aspects of the particular situation). Choosing this meaning of the word date
makes more sense and involves less processing effort than choosing the
meaning "an appointment with a person to whom one is romantically or
sexually attached", which is in fact the most relevant choice for the utterance
in (b ), where the interpretation of date as a fruit would be totally irrelevant
As for (c), neither the selection of the fruit meaning, nor that of the romantic
meeting one would be relevant, considering that the speaker is obviously
asking about an exam, and therefore it is easier to create a scenario where
the meaning "a specific day of the month" for date is involved, rather than
either of the other two meanings.
2) Reference assignment: The identification of the referents in discourse also
plays a very important role in the construction of an explicature. Context is
crucial in this process too, as is made patent in (10), where it would be
impossible to assign a referent to the pronoun "it" and hence to construct an
explicature for B's utterance without a proper maximization of the contextual
effects:
(10) Sam comes back home after a long day at school, on which he had a
very difficult Math exam, so when he opens the door, his mother, who
has been worried all day and waiting to hear how he has done, says:
Sam's mother. So, how was it??!! Was it too difficult?
In order to properly construct the explicature of Sam's mother's utterance,
Sam here would have to integrate it with previous discourse in which Sam
told his mother about this exam and about his difficulty in understanding all
the concepts covered in it, as well as resorting to knowledge stored in our
memory about how difficult math exams can be, etc.

3) Enrichment: This phenomenon consists in fleshing out skeletal propositions

in order to make an utterance complete. Two main kinds of enrichment are a)
the recovery of missing elements in cases of ellipsis, and b) the resolution of
semantic incompleteness. In the examples in (8) above, we have two cases in
which it is necessary to recover the missing elements because they have been
elided, namely Tickets, please! and Any trouble with the door? in which, as
we have seen, we need to reconstruct the whole syntactic structure of the
utterance (by means of grammatical rules) in order to get its explicature.
Other examples can be seen in (11), where the explicature would provide
answers to the bracketed q,uestions. This completion process of linguistically
/ /// //I// Relevance : Process ing effort and contextual effects of communication 193

mandated completion is called saturation by some authors (e.g. Carston

2007):
(11) a. Coca-Cola is worse. [than what?]
b. She is too thin. [for what?]
c. I'm not good enough. [for what?]
d. That would be the same. [as what?]

As for the resolution of semantic incompleteness, the examples in (12)

illustrate utterances that would req,uire this kind of enrichment:
(12) a. We were chatting for some time after lunch.
b. Modern humans evolved in Africa possibly from Homo
heidelbergensis, Homo rhodesiensis or Homo antecessor and
migrated out of the continent some 50,000 to 100,000 years ago,
gradually replacing local populations of Homo erectus. (from
Wikipedia http://en.wikipedia.org/wiki/Human_evolution)

The meaning of some we have to recover is different in (a) and (b).While in

(a) it refers to a relatively short period of time (minutes or one or two hours
at the most), in (b) it refers to thousands of years.

4) Higher order explicatures: this kind of explicature has to do with the speech
act of the utterance in q,uestion. In RT there exist two kinds of (implicit)
speech act: 1) communicated speech acts, and 2) non-communicated speech
acts. Cruse (2004: 387) identifies thanking and promising as examples of the
first kind, because they are defined as institutional speech acts and could not
function without the existence of appropriate constitutive rules and social
structures. On the contrary, non-communicated speech acts need not be
identified as such by the audience to be successfully performed. Higher-
order explicatures reveal the non-communicated propositional attitude of
the speaker to her utterance. Let us consider (13):
(13) A teacher (to her student John, when she sees that he is upset about
something his classmate Tony has said to him and looks like he wants to
hit him): Think before you act, John!

Depending on the conditions of the context, the student here could make the
higher-order explicature that the teacher is reprimanding him or, perhaps,
that she is just giving him some advice, thereby perceiving the teacher's
communicative intention.

It is worth noticing, by way of a conclusion to the treatment of explicatures,

that Carston (2007) points out that some of the so-called explicatures within the
194 Relevance : Processing effort and contextual effects of communication II II II II/

framework of RT would be considered generalized conversational implicatures 64

according to various neo-Gricean accounts. Carston (2007: 645), however,
observes that there are substantive differences between the two conceptions, a
topic we are not going to elaborate on here because it goes beyond the scope and
objectives of this book.

5.5.1.2 lmplicature
At the same time as the hearer fleshes out the semantic representation of an
utterance by constructing the explicature in some of the above-mentioned ways,
he derives the implicatures from it. Unlike explicatures, the propositional form of
an implicature is different from that of the original utterance, as illustrated in
(14), where Jeremy wants to tease his girlfriend Paula, and therefore replies to
her q,uestion in an ironical way:
(14) Paula: Do you love me?
Jeremy (smiling and hugging her): No, I don't!

The propositional form of the explicature for Jeremy's utterance will be the
same as that of the utterance: No, I do not love you (reconstructed by means of
enrichment), whereas that of the implicature, given the conditions of the context
(Paula knows that Jeremy is teasing her, Jeremy is smiling and hugging her, etc.)
will be different from that of the utterance, and could be approximately
formulated as: Yes, of course I love you. How could you even ask such a thing?
It must be pointed out here that the relevance-theoretic approach to
implicatures does not exactly coincide with Grice's (1967) view of the
phenomenon. One main aspect in which they differ is in the already mentioned
fact that for RT only one of Grice's Maxims, that of Relation, is to be taken into
account. Besides, as S&W (1986: 162) explain, in their view:
[C]ommunicators do not 'follow' the principle of relevance; and they could not
violate it even if they wanted to. The principle of relevance applies without
exception: every act of ostensive communication communicates a presumption
of relevance.
Thus, S&W do not adhere to Grice's view that implicatures are triggered by
the flouting of the Maxims of the Cooperative Principle. For them, the Principle
of Relevance explains both explicit and implicit ostensive communication as a
whole. Implicatures are recovered by taking into account "the speaker's
manifest expectations about how her utterance should achieve optimal
relevance" (1986: 194).

64 See 3.3.3.4 and 3.3.3.5.

III/ I I III Relevance : Processing effort and contextual effects of communication 195

As anticipated in 5.5, from the relevance-theoretic viewpoint, implicatures

may be of two kinds: 1) implicated premises, and 2) implicated conclusions. (15a)
presents an example of the former, and (15b) of the latter:
(15) Jenny: Have you read Fahrenheit 451?
Robert: I don't like science fiction books.
a. Implicated premise: Fahrenheit 451 is a science fiction book.
b. Implicated conclusion: Robert wouldn't read a science fiction book, and
therefore he has not read Fahrenheit 451.
Implicated premises must be retrieved or constructed by developing
assumption schemas from memory, while implicated conclusions are deduced
from the explicatures of the utterance and the context.
By way of conclusion, we shall analyze one more example, integrating all
these aspects of meaning we have discussed in this section. Let us then consider
Linda's utterance in (16):
(16) Brenda: I'm sick and tired of this. I'm going to q,uit my job!
Linda: Your terribly well-paid job? Are you sure?
Linda's reply here is an economical linguistic representation of an
underlying logical form which needs to be enriched by means of explicature to
a full propositional form in which both of her q,uestions are fully reconstructed
by recovering the referent of 'you' and 'your' and the syntactic structures that
are missing, as well as the senses of the nouns and adjectives used (job, sure,
well-paid). Thus the explicature would be something like: Are you (Brenda) sure
that you (Brenda) want to quit your (Brenda's) terribly well-paid job? (by
choosing the most relevant meanings for job, sure and well-paid, according to
the context).
Also, depending on aspects of the context that have to do with Linda's
personality and/or the relationship between these two friends, Brenda might
take Linda's reply as a warning (Mind you, if you quit this job, you're never
going to find a betterpaidjob) or as a reprimand (How could you even think of
leaving such a good job???!!!), which would constitute the higher-order
explicature.
On the other hand, a hearer will most surely construct the implicated premise
that Brenda has a very high salary, and will deduce the implicated conclusion
that Linda would never q,uit a job like Brenda's if she had one.
We see in all these examples that communicators are dealing with different
types of meaning. One important distinction in RT is the one between conce'J)tual
and procedural meaning. We now turn to this issue.
196 Relevance: Process ing effort and contextual effects of communication / / ///////

5.6. CONCEPTUAL VS. PROCEDURAL

MEANING/ENCODING
The distinction between conceptual and procedural meaning is one of the most
influential notions to have emerged from RT. It was developed extensively by
Blakemore (1987, 2002) and continued in the works of other linguists, such as
those found in the volume edited by Escandell-Vidal, Leonetti & Ahern (2011).
Blakemore explains that there are important differences between regular
'content' words such as house, green, or play on the one hand, and discourse
connectives such as but, thus, or so. According to her view, then, there is more
than a single type of encoded meaning: Some words encode concepts and others
encode procedures.
Wilson (2011) points out that lexicalized expressions such as nouns, verbs
and adverbs encode concepts which together constitute a conceptual
representation system or "language of thought" (2011: 5). 65 These contribute to
the proposition expressed by an utterance and affect the conditions under which
it is true. On the other hand, grammaticalized discourse markers or connectives
encode procedural meaning which contributes "to the computational rather than
the representational side of comprehension" (2011: 5). These connectives do not
affect the truth conditions of utterances and therefore do not encode concepts.66
However, discourse connectives are a good example of elements existing and
working at the semantics/ pragmatics interface, for they can also be said to
contain arbitrary linguistic-semantic meanings. Indeed, authors like Fraser
(2006a), Escandell-Vidal & Leonetti (2011) or Wilson herself point out that a single
word or expression can encode both conceptual and procedural meaning at the
same time, which logically entails that these two types of meaning are not
mutually exclusive.
But procedural meaning is not only found in connectives. Further research
on the topic ( e.g. Wilson & Sperber 1993; Escandell-Vidal 1998, 2002; Fretheim
1998; Andersen & Fretheim 2000; Wilson 2000; Wharton 2003, 2009; Wilson &
Wharton 2006) has shown that linguistic elements such as pronouns, mood
indicators, interjections or intonation encode procedural meaning and contribute
to the construction of higher-order explicatures which carry speech-act,
propositional-attitude or affective-attitude information. Wilson & Sperber (1993),
for instance, propose that pronouns, like discourse connectives, not only encode
conceptual meaning, but also procedural constraints on the inferential phase of
the comprehension (e.g. on the phase of identifying who "I" and "you" are in an

65 See also Fodor (1975, 1998) or Sperber & Wilson (1986: Chapter 2).
66 However, see Bach (1999), or Potts (2005) for a contrary view.
/I// I I I I I Relevance: Processing effort and contextual eff, 197

utterance like I like you). Thus, on the one hand, the pronoun in (17a) constrains
the set of potential referents to those picked out by a masculine singular pronoun
(conceptual meaning), and on the other, the hearer will have to identify from all
of those referents which of them is the intended referent (Peter Brown in ( 17b))
based on contextual information and expectations of relevance (procedural
meaning):
(17) a. He is my father's best friend.
b. Peter Brown is my father's best friend.
Another example, this time of prosodic procedural meaning, is found in (18)
where the type of affective intonation added may trigger higher-order
explicatures of the kind expressed in (18a) or (18b):
(18) I told her the truth.
a. I regret having told her the truth.
b. I'm very proud and happy to have told her the truth.

While the notion of conceptual meaning is relatively easy to understand, the

notion of procedural meaning still remains somewhat complex and mysterious:
very little is known about the cognitive mechanisms involved in it or about how
it is acq,uired. Clark (2013) points out that it is hard to say exactly what procedural
expressions encode. If we ask a native speaker what house means, she will
promptly answer something like "a building where a family lives", but if you ask
her what but means, she will most likely answer along the lines of "It's like
however" or "Sort of a contrast or something" (2013: 324). Furthermore, according
to Blakemore (1987) , procedural meanings are more difficult to translate than
conceptual meanings. For instance Spanishpero seems easily translatable in (19),
but much less easily so in (20):
(19) Julia es muy buena persona pero es algo tacafia. (Julia is a very good
person but she's a bit stingy)
(20) (A has repeatedly been asking B to accompany him on his jungle adventure)
A: Bueno, al final, 62,te atreves a venir a la jungla conmigo o no?? (Well then,
when all's said and done, do you dare come to the jungle with m e or
not?)
B: jPero si! jCuenta conmigo! (*But yes! You can count on me!)
Another important aspect of procedural meanings is that they normally do
not have synonymous conceptual counterparts. It cannot be said, for instance,
that the use and meaning of well in (21) is synonymous with its use in (22):

(21) She sings very well. (Well = adverb form of good , encoding conceptual
meaning)
198 Relevance : Processing effort and contextual effects of communication /////////

(22) A: How was the trip?

B: Well, I wouldn't say it was exciting. (Well = parallel pragmatic marker, in
Fraser's 2006b classification, encoding procedural meaning)

However, as we anticipated above, conceptual and procedural meaning are

not mutually exclusive, and so we find cases like (23b ), in which the conceptual
meaning of indeed is similar to that of (23a), but in (23b) it has an additional
procedural component that makes the marker deictic and exhibits some
connection with previous chunks of discourse. Thus we can say that a given
expression, depending on how it is used, can encode either a conceptual or a
procedural meaning, or both at the same time. In those cases in which both types
of meaning are present, as in (23b ), the procedural component contributes an
additional meaning to the expression that will make it different from the meaning
of the same expression when used only in its conceptual meaning, even though
the conceptual meaning encoded in both may be similar.

(23) a. She was an excellent writer indeed.

b. Indeed, she was an excellent writer.

While conceptual expressions form part of any communicated proposition,

procedural expressions do not normally contribute to propositional meaning, but
they rather indicate the ways in which the utterance might be relevant. For
instance, in (24) Wow is a pragmatic marker which shows the surprise and
admiration of the speaker, but it does not contribute to the propositional meaning
or to the truth conditions of the proposition expressed:

(24) Wow, Andrew's new car is a Ferrari Maranello!

All the above characteristics and distinctions between conceptual and

procedural meaning may logically lead the reader to conclude that there is a
strong connection between conceptual and truth-conditional meaning, as well as
between procedural and pragmatic/non-truth-conditional meaning. This
conclusion is correct, but W&S (1993) explain that while in linguistics the truth-
conditional/non-truth-conditional distinction normally applies to linguistic
expressions, the conceptual/procedural distinction in RT is cognitively motivated
(i.e. it is a contrast between two kinds of processing, not two kinds of
expression). Wilson (2011) remarks, then, that because of its cognitive
commitments, the conceptual/procedural distinction may be useful when
explaining the contrast between many traditional semantic and pragmatic
distinctions, an important one being that between grammaticalized and
lexicalized meaning. This is one aspect that we will touch on in the next section,
although much research still needs to be done to clarify this issue.
IIIIIIIII Relevance : Processing effort and contextual effects of communication 199

5.7. RELEVANCE AND GRAMMAR

RT has not only had a major impact on pragmatics and on how we understand the
interpretation of utterances but has also developed its own distinctive view on
grammar. There are various areas of grammar where one and the same form is
generally regarded as having different meanings. Modal verbs, for example, are
typically analyzed as being polysemous. Consider the three uses of the verb can
in (25):
(25) a. My three-year-old can already write her own name.
b. If you have a ticket, you can now enter the auditorium.
c. It surely can't rain on my birthday!

In (25a), the meaning of can is said to be dynamic, indicating 'ability'; in

(25b), it is deontic, indicating 'permission'; and in (25c ), it is epistemic, indicating
'possibility'. For the speaker, this is not a problem: she knows that the three
notions of ability, permission and possibility are expressible by one and the
same form can. It is for the hearer that there is a ptoblem: how does he know
which of the meanings is intended by the speaker? The answer given by
RT is that the interpretation of can arises from his understanding the
propositional meaning (in which can has an underdetermined meaning) against
the background of relevant contextual assumptions. In (25a), for example, the
background knowledge comes into play that the great majority of children (in
our Western societies) are encouraged to learn to write, as well as the
assumption that this skill usually first manifests itself around the fourth or fifth
year of life, coupled with some linguistic evidence, namely the presence of the
adverb already. All this combines with the propositional meaning - as yet
another example of explicature - to yield the interpretation 'is (at an unusually
early age) able to'. In (25b), can again has a vague meaning which is rendered
more precise by the contextual assumptions: buying a ticket is usually done in
order to be allowed to do something, for example enter an auditorium to hear a
concert or a lecture; unlike a wheelchair for an invalid attendee, for example,
the ticket does not give you the ability but the permission to enter. The
explicatures brought along by the interpretation process thus establish the
'permission' sense of can as appropriate here. In (25c), finally, the presence of
the impersonal subject it excludes the ability and permission senses, thus
enforcing the possibility interpretation: 'It surely isn't possible for it to rain on
my birthday'.
Another clear case of underdeterminacy is that of the three moods of
English. In 2.5, we discussed the moods of English in the context of the
observation that form does not always align with function. From the Relevance
200 Relevance : Processing effort and contextual effects of communication /II I I I I/

Theory viewpoint, we can say that the mood of a clause underdetermines how it
will be interpreted. Consider cases like the following, which all involve clauses
linked by and or or, with the first clause in the imperative mood, and the second
in the indicative mood (clause type: declarative), future tense:

(26) a. Airport official: Join this line and show your passport to the officer.
b. Police officer. Put down that knife or I'll have to shoot you.
c. Restaurant owner. Serve another undercooked steak and you will be
fired.

In (26a), the hearer, using contextual assumptions about the speaker

authority to tell passengers how to behave, finds no clues in the grammar or in
the context that would disbar interpreting the two clauses as orders; this case is
thus straightforward. In (26b), by contrast, the presence of the conjunction or
suggests that the speaker is offering the hearer a choice between putting down
the knife and being shot; however, since background assumptions tell the hearer
that, to put it mildly, the second option is not a desirable choice, he is forced into
understanding (26b) as a negative-conditional threat: 'If you don't put down the
knife, I'll have to shoot you'. In (26c), which has the same grammatical structure
as (26a), the hearer (an incompetent chef) is apparently being ordered by
the speaker (the restaurateur) to serve another undercooked steak. Background
assumptions about what bosses expect of their employees, however, militate
against this interpretation of the imperative, and the relevant inferences are
reinforced by the presence of what appears to be a prediction of dismissal in
the second clause. The hearer will thus understand (26c) as a positive-conditional
threat: 'If you serve another undercooked steak, you will be fired'.
What we have seen so far is that certain grammatical forms are
underdetermined but also that others serve to guide the hearer's interpretation.
As examples of the latter, consider that in (25a), the adverb already predisposes
the hearer to understand can as 'ability'; in (25c) the impersonal it rules out the
'ability' and 'permission' interpretations; and in (26b) and (26c) the conjunctions
or and and contribute to the interpretation of a threat as being subject to a
negative or positive condition respectively. The speaker, then, in formulating her
ostensive act, is using a mix of underdetermined forms and of 'compensatory'
forms which together, in the context created by the utterance, will guide the
hearer towards the intended explicatures.
Discourse connectives are seen in RT as having this very function of guiding
the hearer towards understanding the speaker's intention. We have already
considered some uses of the connectives and and or to constrain the search for
relevance; in the terminology introduced in 5.6, these convey procedural
information. Procedural discourse connectives (like and, or, but, so) are seen by
/ /II I I I I I Relevance : Processing effort and contextual effects of communication 201

the Relevance theorist Blakemore (2002: 82-84) as having three properties:

(a) they are 'elusive', i.e. it is hard to pin down their function; (b) they do not have
synonymous conceptual com1terparts; (c) they cannot be semantically complex.
Let us consider what she might mean by each of these properties. Firstly,
procedural discourse connectives are elusive in the sense that it would make little
sense to look them up in a dictionary, for it would be difficult to write an entry
for all the possible contextual effects that could affect them in any possible
situation in which they might be used. Secondly, we can use certain adverbs
(these are usually words derived from adjectives by adding the suffix -ly) as
connectives, e.g. additionally, alternatively, conversely, etc. Blakemore's point is
that in some cases, as in (21)-(23) above and (27) below, the same word may
appear to mean the same when it functions as an adjunct as when its function is
that of pragmatic marker. However, this is not so, considering that in (27a) the
meaning of additionally is part of the conceptual meaning of the main clause,
and in (27b) the meaning of this connective is mainly procedural, and therefore it
does not affect the basic meaning of the clause.
(27) a. Student to another student: You may have to additionally send the
Professor a report.
b. Tutor: Students will have to hand in their complete portfolio to the
Professor. Additionally, they will have to upload their reports onto
the virtual course.

But Blakemore's point here is better appreciated in cases like (28a) and
(28b ), where it is more obvious that the two uses of well are not synonymous.
Whereas in (28a) well is just an adjunct within the main clause, in (28b) it
functions as a pragmatic/discourse marker which has nothing to do with
whether Julia recites poems well or not (in fact, to judge from what B says, she
does not!).
(28) a. She recites poems very well.
b. A: Did you like Julia's performance?
B: Well, I liked what she said, but not how she recited the poem.
Finally , in saying that procedural connectives cannot be "semantically
complex", Blakemore means that they do not contain or affect the conceptual
meaning of utterances; rather, they are like the plus, minus, multiplication and
division signs in eq_uations, showing the relations among the values but not the
values themselves.
One example freq_uently used in the RT literature is the following:
(29) Peter's not stupid. He can find his own way home.
202 Relevance: Processing effort and contextual effects of communication / / / / / / / /

If a speaker were to say (29), the hearer will assume there is some link
between the two statements but he cannot be certain what that link is.
However, if the speaker had added a discourse connective to the second
statement, the hearer would have been guided towards making the intended
inference, cf. (30):

(30) a. Peter's not stupid. So he can find his own way home.
b. Peter's not stupid. After all, he can find his own way home.

In (30a), the discourse connective so indicates that the speaker presupposes

Peter is not stupid and is drawing the conclusion that he can find his own way
home; in (30b), the speaker is asserting that Peter is not stupid and the presence
of after all indicating that she is taking his ability to find his own way home as
evidence for that assertion. By using a discourse connective, the speaker of (30a)
or (30b) is thus orienting herself to the hearer's inferential processes and easing
his task of finding the intended interpretation. Notice that (29) is fine as it stands·
adding a discourse connective, as in (30), is an option that speakers can use to
minimize the danger of being misunderstood.
This section has shown some ways in which RT interacts with grammar. In
particular, we have seen how certain grammatical forms like modal verbs or the
imperative mood are semantically underdetermined and therefore have to be
interpreted against the background of contextual assumptions. We have also seen
how one and the same form can have either procedural or conceptual meaning
and how this interacts with the grammatical distinction between discourse
markers and adjuncts. Let us now turn to the concept of salience.

5.8. SALIENCE AND INFERENCE

Throughout this chapter we have been dealing with the concept of salience
without explicitly pointing to it, but we have noticed that we always make
inferences of some kind to interpret what other speakers say and that the most
relevant interpretation of an utterance - which is normally the most salient -
may sometimes be an explicature, sometimes an implicature, and sometimes a
higher-order explicature. And when we consider more than one possible
explicature or implicature it is the principle of salience that helps us choose the
most relevant one. Thus, in (31), the most salient stimulus and explicature will
be that the oil the speaker needs to buy is the kind of oil that is made from
olives, and not an oilfor olives (because, among other things, our knowledge of
the world tells us that olives cannot eat or consume oil). However, in (32), the
I I I I I I I I I Relevance: Processing effort and contextual effects of communication 203

most salient explicature is that what the speaker needs to buy is oilfor babies,
NOT an oil made from babies (in fact this would not be allowed to be sold as a
product in our society!).
(31) I need to buy some olive oil at the supermarket.
(32) I need to buy some baby oil at the supermarket.

(31) and (32) illustrate cases in which the salient meaning is easy for any
competent speaker of a language (English in this case) to recognize, but this is
not always the case; it can sometimes be difficult to determine how a speaker
intends a contribution to be relevant, especially when one has to decide which is
the most salient implicature or higher-order explicature to result from a whole
utterance. This is in fact a point made by the postmodern theorists, who consider
that, for instance, a novel and the different stories told in it can have as many
relevant interpretations as the number of people who read them. But without
analyzing a whole novel, let us simply discuss the exchange in (33), where William
(a university professor) has arrived late for his date with Lorna:
(33) William: Hi Lorna. I'm terribly sorry about this delay. I had one of these
hectic days and on top of it all I had to give a speech to the doctoral students.
Lorna (with a serious expression on her face): Yes, sure ... one of your
world-famous profound speeches ...

Here William, depending on his knowledge of Lorna's personality, might

consider that the most salient interpretation of Lorna's utterance is the higher-
order explicature that Lorna is reprimanding him. This would imply that he has
identified her utterance as being ironic, but that he has not given much importance
to the irony because he knows that Lorna does not really mean it and that if she
is ever sharp-tongued with him, it is when she is angry. For him, the salient and
most important aspect of the situation is that Loma is upset with him; he will try
and make up for his lapse by being nice to her, indulging her and talking her out
of her mood. However, another possibility would be that William found the ironic
meaning of Lorna's utterance to be the most salient, in which case he would be
focusing on the implicature, i.e. that Lorna thinks that his talks are not world-
famous or profound at all. This could lead to William feeling offended and, instead
of trying to fix the situation, he could react by ending the conversation abruptly
or storming out of the room, which in turn could have undesirable conseq,uences
for their relationship. This example, though apparently very simple, is highly
typical of the complexity of human interaction, and shows that even though the
notion of salience usually allows us to interpret people's utterances, it is not
always simple to decide which of the possible interpretations is to be taken as the
most salient and relevant, this difficulty being the source of most
misunderstandings and pragmatic misfires.
204 Re levance : Processing effort and contextual effeds of communication / / / / / / / / /

In short, and as the reader may have concluded by now, the concept of
salience is a useful concept when speaking about relevance, precisely because
the most salient interpretations are the most relevant ones, i.e. the ones with most
contextual effects and involving least processing effort.

5.9. IMAGES AND RT

At the beginning of this chapter, we made it clear that RT is a theory of human
communication. Like the great majority of theorists in this area, we took all our
examples of communication from the use of language in interaction, with some
mention of the supportive role of gesture. However, as we all know, it freq,uentlg
happens that we communicate by means of images, both moving pictures like
movies and videos and still pictures like photographs. The q,uestion arises
whether RT is eq,ually applicable to pictorial communication. At a time when
images are playing an ever more important role in our daily lives, this section will
give brief consideration to the role of RT in understanding how we communicate
via the pictorial mode.
A first observation is that many examples of pictorial communication
involve a combination of language and pictures. For example most pictorial
advertisements published on billboards or in magazines ensure that the name of
the product being advertised is prominently visible. In public communication,
signs like that in (34) combine the word 'exit' with a graphic representation of a
human being exiting; this co-presence may indeed be essential for readers who
are illiterate or who do not know the English language:

(34) - -

Such communication is known as multimodal, since it involves more than

one mode of ostensive communication. We can conclude that the reader of this
sign can process either the linguistic stimulus or the pictorial one (or both) and
that the processing is essentially identical: the word 'exit' and the image of the
person exiting are relevant and spark off positive cognitive effects in exactly
the same way.
If pictorial communication indeed runs parallel to verbal communication, we
would expect the presentation of a picture to spark off implicatures in the same
way. Let us now consider the pictorial part of (34) further, as shown in (35):

(35)
I I I I I I/ I I Relevance : Processing effort and contextual effects of communication 205

In ordinary verbal communication, we know who the communicators are. In

the case of a sign like (35), that is not immediately obvious, so we have to infer
the identity of whoever had the picture affixed to the wall. We achieve sufficient
relevance when we conclude that the authority in q_uestion is responsible for the
safety of people in the space where the sign is visible. As for the receiver of the
intended message, the most relevant inference is that the human figure in the
picture represents the recipient himself, i.e. the observer of the picture; any
observer therefore draws the strong implicature that the picture is relevant to
him. Further implicatures are that the sign is not addressed only to people whose
heads float free from their shoulders (since such people would have little ability
to perceive the sign!) and is not addressed to males only, despite the fact that the
human figure has a physical structure and clothing (pants) that are closer to
the stereotype for adult men, but to anyone, male or female, adult or child, single
or in a group, looking for the way out. All of these are implicatures, reflecting the
artist's manifest expectations about how her picture should achieve optimal
relevance. Notice that the image also involves a higher-order explicature, since
the illocutionary status of the picture is not that of a statement "Near this picture
there is an exit" (since that in itself is of little relevance) but is eq_uivalent to an
imperative which will be highly relevant in an emergency: "Use the exit near this
picture". The fact that the figure is depicted as running encourages this
interpretation, although it is a weak implicature, since the exit could also be used
by someone who is not in a hurry.

The q_uestion has arisen in the RT literature (see Forceville & Clark 2014 for
a balanced discussion) whether pictures can trigger explicatures. Forceville &
Clark argue that some nonverbal behaviors can be understood as having coded
meanings, which would allow for the possibility of non-linguistic explicatures.
Some gestures ( e.g. thumbs up) carry non-verbal codes that can however be
fleshed out to arrive at an interpretation by means of an explicature. The same
happens with some pictorial codes, such as the one in (35), where the meaning is
so precise that its interpretation is not a matter of only inferencing, but includes
an element of coding as well. Thus, the most relevant explicature for (35) will be
something like: "This is the (way to the nearest) exit", depending on whether the
sign is placed on top of the actual exit or is found on the way to it, pointing in that
direction. The same holds true for other pictorial elements such as the icons
that aid navigation on the internet, or the logos of some famous and very well-
known brand names. And apart from the most relevant explicatures found in
these signs, there are other elements that appear to be encoded and could
the refore be fleshed out and explicated if we wanted to pay attention to detail.
Consider for instance the diagonal line under the figure in (35). This line achieves
relevance if interpreted as a shadow of the figure's right leg, since that engenders
the interpretation of the white area as representing the sunlit exterior of the
206 Relevance : Processing effort and contextual effects of commun ication / / / / / / / / /

space in which the sign is visible. The interpretation of the diagonal line as a
shadow is arguably an example of the 'fleshing out' that is characteristic of
explicatures. Another example is the fact that the figure's right arm appears to
have already exited while his left foot is clearly still inside. It is by making these
matters explicit that we have a basis for understanding the implicatures of the
picture, for example the implicature that the sun may not be shining in
the exterior; in fact, the exit may simply lead to another internal room.

5.10 . CONCLUSION
RT has been applied prolifically as a way of understanding the many forms of
'overt and covert communication' (see 5.4) that surround us in our daily lives (see
Yus n.d. for a complete and constantly updated bibliography). Jokes, for example,
often involve fooling the hearer into interpreting a story as a matter of overt
communication, only for the covert interpretation to emerge later as the intended
one. Advertising often plays with language in ways that puzzle.us. Not until we
put in the cognitive work described in RT do we understand what product we are
being encouraged to buy; that very cognitive effort reinforces our memory for
and approval of the item being advertised. In literary studies, too, RT has given
scholars a rigorous framework in which to think about the multiple interpretability
of poetry and other multi-layered texts. In translation studies, the debate about
what it means to be 'faithful' to the original has been impacted by the RT
distinction between decoding and interpreting. Yet there have also been criticisms
of the strongly cognitive orientation of the theory, in which the social embedding
of language behavior is downplayed. In addition, critics have pointed out that
there has been little experimental verification of the claims of RT: for example,
the confident claim that it is always the most relevant interpretation that hearers
reach first is ripe for testing. Questions of the general applicability of pragmatics
and of its wider context will be treated in the final chapter, in which we will
consider how the various theories we have reviewed interact with other aspects
of our linguistic abilities ( syntax, lexis, intonation, etc.) as well as how pragmatic
theories can contribute to our understanding of the human being in general,
especially with regard to our propensity to evaluate everyone and everything
we come across and the powerful role of emotions in our communication and
interaction.
IIIIIIIII Relevance : Processing effort and contextual effects of communication 207

SUMMARY (CHAPTER 5)
1. RT has been developed by Dan Sperber and Deirdre Wilson and their
colleagues, establishing itself as one of the major theories of pragmatics to
have arisen in the aftermath of Grice's work. It sees the hearer's task as being
to uncover the speaker's communicative intention on the basis of her words
and the context in which they were uttered. RT focuses on how hearers infer
intentions by explicating underdetermined utterances and identifying
implicatures.

2. On the permanently applicable assumption that the speaker is using the

optimally relevant formulation of her 'act of ostensive communication' (an
assumption that is spelled out as the Cognitive and Communicative Principles
of Relevance), the hearer will select as the most relevant interpretation the
one that req,uires minimum processing effort from him. The relevance of an
utterance is reflected in the degree to which it has contextual effects, i.e. the
extent of its impact on the information currently available to the hearer. The
optimal formulation is thus the one that has the.best cost-benefit ratio, i.e.
maximum contextual effect for minimum processing effort.

3. The search for relevance is seen as an innate efficiency constraint on human

cognition and one that is vital to our survival. It is essential that our cognition
is flexible and changeable, and accordingly the cognitive effect of an
utterance can be to strengthen existing assumptions, to modify them or even
to abandon them altogether. These are all called positive cognitive effects,
since the hearer is better informed as a result and better eq,uipped to deal with
the world.

4. RT distinguishes three types of communication: accidental, covert and overt.

The first occurs when a speaker unintentionally reveals some information
that she would have preferred to conceal; the second is found when a speaker
intends to convey certain ideas but just hints at them; the third is where the
speaker's intentions are foregrounded and openly reflected in the chosen
formulation. RT clearly focuses on the second and third.

5. The speaker's effort to make assumptions manifest to her hearer is

achieved by using language to draw her hearer's attention. She can also
use a combination of language and gesture or, in certain circumstances,
simply gesture. In all these cases, the speaker indulges in ostensive-
inferential communication, which involves getting the hearer to recognize
her intention to inform and also her intention to inform him of her
informative intention.
208 Relevance: Processing effort and contextual effects of communication / / / / / / / / /

6. The hearer's task is a composite of various subtasks: (a) to develop a

hypothesis about the formulation used by the speaker by decoding its
meaning and where necessary enriching the content with explicatures; (b) to
create a context (or a set of 'contextual assumptions' or 'implicated
premises') in which the utterance can be understood as maximally relevant;
(c) to work out what the implications of (a) and (b) are for the context (the
so-called 'implicated conclusions').

7. Explicatures involve enriching the content offered by the speaker, which

often underdetermines the process of discovering the intended meaning. This
may be a matter of deciding which meaning of an ambiguous word is
intended, of assigning reference to pronouns or of fleshing out truncated
utterances. Higher-order explicatures are also needed: these involve deciding
what kind of speech act is intended where there is more than one possibility.

8. RT sees the speaker's ostensive act of formulating an utterance as 'encoding'

and the hearer's analysis of the incoming utterance as 'decoding'. Some of
the words in that utterance are used to encode concepts, while others encode
procedures. Discourse markers are prominent examples of expressions with
procedural meaning, since they guide the hearer through the procedure of
determining the optimally relevant interpretation. However, many discourse
markers also have conceptual meaning and research has shown that
conceptual and procedural meanings are thoroughly interwoven in every
utterance.

9. RT has also had some impact on grammar in emphasizing how conceptual

meanings are regularly underdetermined by the forms in which they are
expressed. This is, for example, true of the modal verbs of English or the
moods of the clause: in both cases there are many fewer distinctions than
possible meanings.

10. Although the number of implicatures that a hearer could in principle draw
from an utterance is enormous, some are more accessible than others, and
it is the most salient implicature which is the most relevant. It can happen
that a hearer is unsure which of two or more eq_ually salient implicatures is
the intended one: if he chooses the wrong one, this can lead to
miscommunication.

11. RT has been applied above all to communication via language, but there are
of course other modes of communication, notably by means of images. It
can be shown that pictures spark off implicatures, but it has also been
argued that their interpretation involves explicatures, including higher-
order ones.
I I I I I I I I I Relevance : Processing effort and contextual effects of com_m unication 209

SELF-EVALUATION QUESTIONS (CHAPTER 5)

Choose the most appropriate answer for each q,uestion.

1. Sperber & Wilson fully agree with Grice that ...
a. the four Maxims of the CP are necessary for interpreting utterances.
b. the expression and recognition of intentions is essential for human
communication.
c. only the Maxim of Manner is necessary for interpreting utterances.

2. It is an assumption of Relevance Theory that the lesser the difficulty in

working out what someone means,
a. the less relevant the speaker's utterance will be for the hearer.
b. the more difficult it will be for the speaker to make herself understood.
c. the more relevant the speaker's utterance will be for the hearer.

3. An assumption is relevant in a context ...

a. only when it has some contextual effect in that context.
b. when it has no contextual effects in that context.
c. when both speaker and hearer belong to the same cultural context.

4. By means of an act of ostensive communication the speaker intends to ...

a. both convey a particular message and help the hearer to recognize this.
b. hide the real meaning of her message, if possible.
c. make it obvious that her interlocutor is not interpreting her message.

5. Which would be the most relevant answer given by Billy in response to his
teacher's q,uestion?
Teacher: Have you done your homework for today, Billy?
Billy:
a. If the Sun revolves around the Earth, the answer to your q,uestion is
positive.
b. No, I haven't. I'm sorry, Sir.
c. I had to help my mother with the house chores yesterday, and had no time
for anything else.
210 Relevance : Processing effort and contextual effects of communication /////////

6. The kind of communication sent out by a woman who comes back from work
after a very stressful day, and when she greets her children tries to be happy
but cannot hide the fact that she is tired and overwhelmed would be an
example of ...
a. overt communication.
b. covert communication.
c. accidental communication.

7. For Relevance theorists, inferential and ostensive communication are ...

a. different things.
b. one and the same thing.
c. similar in some respects but different in others.

8. In a conversation where two men are discussing the beauty of two women,
Lara and Angel, the utterance "Lara is more beautiful" will be interpreted as
"Lara is more beautiful than Angel" by means of the process of .. ·.
a. enrichment.
b. reference assignment.
c. higher-order explicature.

9. The propositional form of an implicature is likely to be ...

a. the same as that of the original utterance.
b. different from that of the original utterance.
c. the same as that of the corresponding explicature.

10. Which would be the implicated premise of Jeremy's utterance?

Richard: Have you seen The Last Rescue?
Jeremy: I hate war movies.

a. Jeremy does not like going to the cinema.

b. Jeremy has not seen The Last Rescue.
c. The Last Rescue is a war movie.

11. Which of the options in 10 above would be the implicated conclusion of

Jeremy's utterance?
IIIIIIIII Relevance: Processing effort and contextual effects of communication 211

12. Procedural meaning ...

a. is found only in discourse connectives.
b. does not encode concepts.
c. is the same as truth-conditional meaning.

13. From the Relevance Theory viewpoint, we can say that the mood of a
clause ...
a. underdetermines how it will be interpreted.
b. is all we need to interpret it appropriately.
c. is the only grammatical clue for understanding an utterance.

14. The meaning of luckily in the utterance Luckily, I could get a ticketfor the
Eric Clapton concert at the Royal Albert Hall next month, is mainly ...
a. conceptual.
b. figurative.
c. procedural.

15. The most salient interpretation of an utterance ...

a. is always given by its explicature.
b. may be its explicature, its implicature or its higher-order explicature.
c. is always given by its implicature.

DEVELOPING OUR PRAGMATIC COMPETENCE:

EXERCISES FOR REFLECTION AND AMPLIFICATION
(CHAPTER 5)
A) Explain the hurnor found in the following jokes based on the RT
concept of relevant interpretation:
a) Did you hear about the guy whose whole left side was cut off? He's all
right now.
b) I wondered why the baseball was getting bigger. Then it hit me.
c) I can't believe I got fired from the calendar factory. All I did was take a
day off.
212 Relevance: Processing effort and contextual effects of communication / / // / / / / /

B) Which of the two options given in each situation would be the more
relevant interpretation (involving less processing effort and having
more contextual effects) of the following utterances? Once you
answer this q,uestion, sag what kind of communication is found in the
two utterances ( overt, covert or accidental).
1) An officer at immigration, when you're entering a foreign country: Your
passport, please.
a) The officer wants you to give him your passport as a present.
b) The officer wants you to show him your passport for a moment.
2) Oscar's wife Eve comes home and tells him how sweet and generous her
workmate Jenny has been to her, for she has helped her with a very
difficult task she had to do urgently. Without Jenny's help, Eve's boss
would have gotten very angry at her because she wouldn't have finished
the job on time. So when Oscar meets Jenny the following day, he says to
her: They tell me you're a very selfish person!
a) Oscar thinks Jenny is a selfish person.
b) Oscar thinks Jenny is a very generous person and is grateful for that.

C) Formulate the most relevant explicature for the utterances in italics

in their corresponding situations, specifying which process is req,uired
to get to such explicatures ( disambiguation, reference assignment,
enrichment or higher-order explicature?).
,- \ ··1 - • • , •_ .' •;:..-.-.... >, ,· ' . I

' MOST RELEVANT .

'• ~B>LICATURE
Jerry is a cook and is now in the
kitchen of his restaurant. His
colleague asks him:
When will youfinish the souffle?
Lindsay gives her daughter a present
for her birthday, but her daughter
does not thank her for it, so Lindsay
sags out loud:
You're welcome!!
Sandra comes back home from a blind
date with Ernest. Her sister, who
knew about her blind date, anxiously
asks her: So, did you like him?
JoAnne knows that what her friend is
telling her did not really happen, so
she sags: You're lying!
//I/ I I I I I Relevance : Processing effort and contextual effects of communication 213

D) What are the implicatures (implicated premise and implicated

conclusion) of the utterances in bold?

Torn: Will you invite Henry to

your party?
Abby: I never invite jerks to
mg parties.

A: Look! There are Wilma and

David.
B: Ah, I have never seen David
before, but once more we have
proof of the fact that Wilma
always dates handsome men.

Anne: Is there any dulce de leche

in the fridge?
Beatrice: If I had dulce de
leche, I would be fat.

E) According to Blakemore, what do the words/ expressions in bold

mainly encode, concepts or procedures?
a) There's a spider on the table.
b) Don't cry, please.
c) I don't like him. Conseq,uently, I won't accept his invitation.
d) She would never do such a thing.
e) To conclude, I will q,uote Shakespeare once more: Let me not to the
marriage of true minds admit impediments. Love is not love which alters
when it alterationfinds.

F) Indicate which is the kind of meaning ( conceptual or procedural)

mainly encoded by O.K. in each of the following utterances:
a) A: How are you?
B: I'm 0.K.

b) O.K., are you coming with me or not?

c) Is that 0.K. with you?
d) O.K. Stop it! I won't put up with this anymore.
214 Relevance : Processing effort and contextual effects of communication /////////

G) Which is the more salient ( and therefore more relevant) interpretation

of the compound words in bold?
a) I need to get an ironing board ( a board for ironing clothes on or a board
of people who iron clothes?)
b) Gloria is Peggy's half-sister (Has Gloria been cut in half, or does she have
the same father or mother as Peggy?)
c) Be careful! There's a honeycomb there. (A comb made of honey, or a wax
structure made by bees to store their honey?)

H) The conjunction or can have two interpretations. Imagine I have rented my

flat to Peter and Richard for a month. When the month is over, I inspect the
flat and find that it smells badly of tobacco. If I say "Well, now I know Peter
smokes or Richard smokes", I mean three possibilities: Peter smokes; Richard
smokes; they both smoke. This is called the inclusive interpretation of or. In
a different context, or will have an exclusive reading, where the possibility of
interpreting as relevant both the elements linked by or is excluded, as in
Students may sit a three-hour exam or write a 5,000-word essay; here the
students must choose one of two options. The exclusive reading can be
signalled by the word either: Students may either sit a three-hour exam or
write a 5,000-word essay. Now consider the following old advertisement.
Taking into account the contextual assumptions needed to interpret
it, is or here inclusive or exclusive? Rephrase the main message of the
advertisement as a negative conditional.
 OTHER TOPICS OF INT-EREST
IN PRAGMATICS
//////////////////////////////////////
MAIN OBJECTIVE OF THE CHAPTER: To introduce the reader to
prominent connections between pragmatics and other areas of academic
work from both the micropragmatic and macropragmatic perspectives.

MAIN TOPICS OF THE CHAPTER:

• Pragmatics and syntax, the lexicon and prosody
• Evaluation and Appraisal Theory
• The relation between cognition, emotion and language
• The impact of pragmatics on historical linguistics
• The impact of pragmatics on computational linguistics
• Experimentation in pragmatics
• Cy berpragmatics
216 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

6.1. INTRODUCTION
The reader will remember that in Chapter 1 we discussed the two main general
approaches to pragmatics: a) the Anglo-American view (micropragmatics)
which sees pragmatics as one more component of the grammar of a language, on
a par with the phonological, morphological, lexical, syntactic and semantic
components, and b) the broader, European Continental view (macropragmatics)
which sees pragmatics as the cognitive, social, and cultural science of
communication (Verschueren 2009). In this last chapter of our book we shall
explore some additional topics from the former view, namely the relationship
between pragmatics and the other components of grammar (syntax, lexicon,
etc.), moving on to explore other topics from the latter view, such as how our
systems of values and our emotions interact with language and the perspective
offered by four interdisciplinary approaches to pragmatics: historical
computational, experimental and cy berpragmatics.

6.2. PRAGMATICS AS AN IMPORTANT

COMPONENT OF GRAMMAR

6.2.1. PRAGMATICS AND SYNTAX: HOW THEY INTERACT

The scientific treatment of language traditionally focused on phonology and
morphology, the study of the organization of phonemes and of the componen
parts of words respectively. Syntax, the study of how words combine to make
phrases, clauses and sentences, was poorly developed until the sixties of last
century, when Chomsky (1965) set out his well-known transformational-
generative theory of syntax, which unleashed an enormous spate of publications
on the syntax of English and, later, other languages. Syntax moved from being a
marginal area of interest to occupying the very center of linguistics. The claim
was made that syntax has properties that are uniq,ue to human language, that our
syntactic capacity is inborn (i.e. genetically pre-programmed) and that syntax is
autonomous, i.e. independent of meaning. It was pointed out, for example, that
nonsense sentences, like the first stanza of Lewis Carroll's poem 'Jabberwocky
can still be parsed for their syntax:

(1) 'Twas brillig, and the slithy toves

Did gyre and gimble in the wabe;
All mimsy were the borogoves,
And the mome raths outgrabe.
I I I I I I I I I I I I I I I I/ I I I I I I// I I I I I I I I I I I Other topics of interest in pragmatics 217

Syntacticians were skeptical that meaning could be analyzed with the same
rig or as syntactic structure and the refore concentrated on establishing the
syntactic ('transformational') relations between sentences without attempting to
specify their meaning.
To some extent this is still the approach of Noam Chomsky and his followers.
However, in his (1965) book he had claimed that there is an abstract structure (or
'deep structure') that underlies the 'surface structure' of the actual sentences we
use. Whereas he denied that deep structure was a representation of meaning, a
number of scholars (such as George Lakoff 1971 or Charles Fillmore 1968) argued
that deep structure was in fact a semantic structure and that transformational
rules could be used to derive syntactic structure from that deep semantic
structure. The result was a schism within linguistics between those who remained
true to Chomsky's autonomous syntax and those who pursued the alternative line
(Harris 1993). It was the latter group who confronted the problem of representing
semantic meaning within linguistics (which actually also inspired some of the
former group to involve themselves with semantics, e.g. Jackendoff 1972).
Nevertheless, for all their differences and the animosity between them, both
groups shared a number of theoretical and methodological presuppositions,
notably the idea that clauses and sentences could and should be analyzed
independently of any context. When a problem arose in their work that had
pragmatic aspects to it,. there was a tendency to dismiss it as 'a matter of
pragmatics', without entering into any detailed treatment of how that problem
would in fact be treated. As a result, pragmatics came to be seen by syntacticians
and semanticists as a 'wastebasket', a term first used by Bar-Hillel (1971). The
idea was that pragmatics was a rather mysterious collection of odds and ends 1eft
over from the supposedly neat and tidy study of syntax and semantics. As this
book has shown, this is far from being a true picture of pragmatics, but it was one
that dominated linguistics for a while and discouraged many linguists from
contributing to it.
In the seventies, the idea began to take hold that pragmatics contained
insights that were directly relevant for understanding the syntactic structure of
sentences. For example, it could be shown that alternative formulations of the
same message were not only connected by grammatical rules but also that
the reasons for a speaker to choose one rather than another of the options could
be formulated in pragmatic terms. Consider the following examples:

(2) a. ?A bus driver contacted the police.

b. The police were contacted by a bus driver.
c. It was a bus driver who contacted the police.
d. The one who contacted the police was a bus driver.
218 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

There are many more variants than these four, and they all can be linked by
grammatical rules; syntacticians who limit themselves to doing this are called
formalists. However, you can also specify the circumstances under which a
speaker will use each structure: if you concern yourself with this q,uestion, you
are a functionalist. Functionalists cannot do their work without invoking the
context in which the options are appropriate, and this means that their work
involves doing pragmatics. The circumstances for the use of each of the examples
in (2) are q,uite different.
Example (2b) could occur at the beginning of a text. The contextual
knowledge that makes it appropriate to begin the sentence with 'the police' is that
police stations are open 24 hours a day and are manned by operators who are
waiting for calls from the public. The information at the beginning of the clause
is thus said to be 'given', or contextually determined. Every clause will also
contain 'new' information, information that is not (yet) shared by the speaker and
the addressee and which is therefore not part of the context. In (2b) this comes
as an identification of the source of the contact, a bus driver. The natural order
in English, and in many other languages, is for given information to precede new
information, as in (2b). In addition, given information is typically grammatically
definite (the) while new information is typically indefinite (a) . There is a tendency
stronger in English than in several other languages, for the given information to
be the subject of the clause, again as in (2b ). If the subject of the clause is not the
active participant (the 'agent'), as is the case in (2b) - the bus driver is active, not
the police - then the passive voice is used (as in were contacted). The context
of the clause thus determines (a) the word order, (b) the use of determiners, (c) the
choice of grammatical subject and ( d) the voice of the verb phrase.
Context can also help explain why certain formulations are less 'acceptable
than others, a major concern of workers in syntax. (2a) is preceded by a q,uestion
mark, which is syntacticians' way of indicating that an example is grammatical
but not fully acceptable. From a pragmatic viewpoint, we may say that it is
difficult to find a context in which (2a) would be usable. The problem is that the
sentence begins with new information and ends with given information; this is
less than ideal and the problem can be easily avoided, namely by using the
formulation in (2b ).
(2c), which involves a cleft construction (cf. 3.3.2.1, 3.3.2.2) needs a more
elaborate context to be appropriate. The speaker of (2c) has already shared the
information with the addressee that 'someone contacted the police'. What she
adds in (2c) is the identity of that 'someone'. Notice that if she is directly asked to
provide the information, she could omit the shared 'who contacted the police':
(3) A: Who contacted the police?
B: It was a bus driver.
IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII O+her +opics of inferes+ in pragmafics 219

In (2d), the context is different again: both speaker and hearer know that
someone contacted the police and have other information about him/her as well.
Now the speaker is informing the hearer that this person has bus driving as a
profession. In this case, there is greater shared knowledge, and the speaker - for
whatever reason - is increasing that stock of knowledge. All these examples
show that although it would be possible to do syntax without understanding
speakers' motivations for choosing one or another variant, it is also possible, and
in the functionalist's view essential, to have pragmatics and syntax interact in
fruitful ways. Another advantage is that we come to understand how different
syntactic structures function in discourse: only (2b) could start a narrative, for
example, while (2c) is most natural as an answer and (2d) is best suited in the
middle of a narrative.
One of the main problems of syntax is ambiguity. Consider the two sentences
in (4):
(4) a. John talked to Bill about himself.
b. Mary talked to Bill about himself.

The most natural understanding of (4a) is that the reflexive pronoun himself
refers back to John. In (4b), the only possible interpretation is that himself refers
back to Bill, since Mary is female and cannot be an antecedent for himself (only
herself). However, the interpretation of (4b) makes one think that (4a) may in
fact be ambiguous between two meanings, 'John talked to Bill about John' or
'John talked to Bill about Bill'. Rather than enter into the complex syntactic
arguments about this kind of q,uestion, suffice it to say that the problem arises
solely as a result of 'decontextualization', the treatment of stretches of language
as 'autonomous', as independent of the contexts in which they would be
appropriate.
It is not just the pragmatics of context that interacts with syntax. The
pragmatics of speech acts ( cf. Chapter 2) also has an impact on how syntactic
constructions are understood (for a similar case, see Comrie 1984). Consider the
following pair:
(5) a. Since I have a train to catch, I am asking you [to be allowed to leave
early].
b. Since you are annoying me, I am asking you [to leave early].

In syntactic theory, the elements in sq,uare brackets are both 'embedded

non-finite clauses with an understood subject'. In (5a), the understood subject is
the speaker (I am asking you that I be allowed to leave early) while in (5b), the
understood subject is the hearer (I am asking that you leave early). If the matter
were purely syntactic, the determination of the understood subject should be
220 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

rule-governed, but that is clearly not the case. Rather it is the nature of the speech
acts expressed as I am asking you ... that is the determining factor: in (5a) it
means 'I am req,uesting' while in (5b) it means 'I am (politely) ordering'; a req,uest
is clearly self-oriented while an order is oriented to the actions of the hearer.

The way in which pragmatics can overrule syntactic regularities and

determine syntactic form inspired the functionalist Simon C. Dik to develop a
'functional grammar' (Dik 1978, 1997), in which "pragmatics is seen as the all-
encompassing framework within which semantics and syntax must be studied.
Semantics is regarded as instrumental with respect to pragmatics, and syntax as
instrumental with respect to semantics" (Dik 1997: 8). This has more recently led
to Functional Discourse Grammar (FDG; Hengeveld & Mackenzie 2008), in
which the form taken by an utterance results from the joint operation of an
'Interpersonal Level' of analysis which covers all the pragmatic factors that
co-determine form and a 'Representational Level' which deals with semantics.
Importantly, the Representational Level lies within the scope of the Interpersonal
Level, perpetuating Dik's conception of pragmatics as all-encompassing. Among
the pragmatic aspects of language covered by FDG's Interpersonal Level are the
recognition of the actional nature of language use, the grouping of discourse
acts into conversational moves, the notion of illocutionary force, the
characteristics of the speaker and the hearer, the observation that the
communicated content of an utterance is distinct from its propositional content
and the contention that reference and predication are thoroughly actional
pragmatic phenomena. In addition, the grammatical component of FDG interacts
with a contextual component in such a way that relevant contextual factors
impinge on the workings of the grammar. For all its orientation to the pragmatics
of language use, however, FDG remains a grammar, centrally interested in
q,uestions of linguistic form. As two of its proponents have argued, "FDG [... ]
uses pragmatic notions only to the extent that languages encode the distinctions
corresponding to these notions, be it morphosyntactically or phonologically"
(Hannay & Hengeveld 2009: 114).

The various functionally and cognitively oriented models of grammar

currently practiced in linguistics vary in the extent to which they explicitly ( or
implicitly) involve pragmatics in their approaches to syntax. Among the most
interesting is Giv6n's (2009) 'adaptive approach'. He points to the functional
advantages of a grammatical system and draws a sharp contrast between
pragmatic communication, where interlocutors do not share a system, and
grammatical communication, where they do. In a situation of purely pragmatic
communication, as for example when foreign tourists encounter monolingual
inhabitants of the country they are visiting, morphology will be heavily reduced,
constructions will be simple and the word order will flow directly fro m
IIIIIIIIIIII II IIIII II IIIIIIII IIII III Other topics of interest in pragmatics 221

communicative needs; processing will be slow, laborious and conscious, and

speakers will be highly dependent on context. In grammatical communication, by
contrast, morphology is rich, constructions are varied and hierarchical, and word
order is often dependent on syntactic relations (such as subject and object), while
processing is q,uick, effortless and unconscious. Context still plays a vital role but
interacts with a developed morphosyntax.
This section has covered some aspects of the interaction between pragmatics
and syntax in contemporary grammar. The next section will turn to the impact of
pragmatics on another aspect of our linguistic competence, the lexicon, our store
of words.

6.2.2. PRAGMATICS AND THE LEXICON

Complementing our knowledge of the syntactic rules in the language we speak is
our knowledge of the words and idiomatic phrases that are combined by those
rules. This knowledge is referred to as our 'lexicon', which contains a large list of
entries, rather like a dictionary. Each entry includes various parts: (a) a
specification of the sound of the word in phonemes, e.g. lknfi:/ and possibly of its
spelling (coffee); (b) an indication of its word class (or 'part of speech'), e.g. noun,
plus other features such as [-countable]; (c) a statement of the word's meaning,
possibly also in feature format [+drink, +coffee]. Lexicology is the study of the
structure of the lexicon. Lexicologists have observed that, on the basis of their
meaning, many words group into patterns that can be described with the use of
features. Consider the words man, woman, boy and girl. Two features (or
'meaning components') are enough to distinguish the four words, as follows:

Table I: Example of componential analysis

+adult -,adult
'
'"
1,
+male · man bO!J

-male woman girl

Woman can thus be defined as [+human, -male, +adult], and so on for the
other words. This approach to word meanings and the lexicon is known as
componential analysis. Although it was popular for a while, especially in the
circles around early generative grammar described in 6.2.1, it has been largely
abandoned as being too simple and mechanical (we are now better aware of
subtler gender distinctions, for example) and as only being applicable to part
of the vocabulary ( coffee, for example, cannot be broken down into meaning
components). It should also be added that, for instance, many women are
222 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

uncomfortable about being defined negatively as [-male]; and many objectively

adult females still consider themselves to be girls!
Another aspect of componential analysis that has run into difficulties is the
assumption that each word has one meaning. Consider the adjective red. It would
be possible to define red as corresponding to a wavelength of 670 nanometers on
the visible spectrum, but the actual hue that is associated with red differs
enormously according to which item it is associated with syntactically. Consider
the following example:
(6) A girl with red hair was sitting at a red table, eating red meat accompanied
by a glass of red wine. For dessert, she ordered a red apple. She was reading
a red book.
Clearly, red identifies a different color in each case: the girl's hair is ginger-
colored; the table top is painted red on the outside (and the legs may be of a
different color); the cooked meat is liable to be brown with possibly some pink
inside; the wine is of a dark color; the apple is red on the outside but white inside,
and may have been peeled; the red book has a red cover but white pages (perhaps
it was this book she was reading!). The meaning of red, then, is very much
dependent upon what it combines with. One solution to this problem is to specify
that the lexicon contains a large number of entries with the phonemic form /red/
and the word class Adjective: red 1, red2, red3, ... redn; each entry would then
have its own meaning, with red 1, for example, have the same meaning as ginger.
This multiple-entry approach, which would apply in greater or lesser measure to
all words, was invented to rescue the idea that all meaning could be stored in the
lexicon. A radical alternative solution would be to deny that words have any
fixed meaning and to declare that word meaning cannot be analyzed scientifically.
Luckily, it is not necessary to follow the latter counsel of despair, since we can
have recourse to a notion that was mentioned in Chapters 1 and 5 above,
underdeterminacy.
The insight that lies behind the concept of underdeterminacy is that our
knowledge of the world is much richer than the lexicon of words we use to
describe it. We know from experience, for example, what colors human hair can
have (without dye), the colors of the wine we drink and the types of the meat we
eat; we know that a red apple is outwardly red but inwardly white, while a red
grapefruit is outwardly yellow and inwardly red. We also know that books
typically have white pages but that the cover can have any color. And so on. Just
as all the members of a family are similar, even though third cousins may look
almost totally different, so all the concrete phenomena we describe as red are
related in their hue (Wittgenstein 1953 called this 'family resemblance'). We can
say that all red things are nearer or further from a prototype (Rosch 1978, Lakoff
1987): the prototype for red is probably the color of blood, just as the prototype
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I Other topics of interest in pragmatics 223

for green is grass or foliage. If the meanings of words (lexical entries) are defined
by a prototype, then it becomes possible to include a meaning in the lexicon for
each word. If red is defined in the lexicon as 'having the color of blood', however,
then this clearly underdetermines how the word is used, as we have seen in our
discussion of (6). In order to deal with actual use, we need the concept of
pragmatic enrichment.
Let us consider how the notion of pragmatic enrichment works. Consider the
following examples:
(7) a. We took the bus back to the zoo.
b. We took the escaped camel back to the zoo.
Syntactically, (7 a) and (7b) are completely parallel. They both also use the
verb take. Our natural inclination is to understand (7 a) as meaning that we went
back to the zoo by bus and (7b) as meaning that we captured the camel and
returned it to its enclosure in the zoo. However, these understandings do not lie
in the verb take. Our interpretations are based upon a combination of the
meanings of the words, the syntactic structure of the clauses and essentially -
and this is where pragmatic enrichment comes in - our experience as human
beings. We normally use buses as means of transport and normally believe that
in our everyday environment it is safer for wild animals to be in a zoo. However,
this enrichment is called pragmatic because it is based on context. In other
contexts, it would be possible to understand (7 a) and (7b) differently: if the zoo's
bus had been stolen and abandoned at another location, we could have taken (i.e.
driven) the bus back to its rightful location, the zoo; and if we found an escaped
camel that was tame enough, we could have mounted it and returned to the zoo
on its back. The conclusion is that the two understandings of take in (7 a) and (7b)
are not represented in the lexicon but are dependent upon the context for
enrichment.
Linguists hold that the lexicon is dynamic in the sense that it is not only a
storehouse of words but also contains operations on words. For example, we can
derive new words by adding prefixes or suffixes ( e.g. friendly from friend,
unfriendly fromfriendly, and unfriendliness from urifriendly). We can also join
words together to form compound words, for example chocolate and cake to
make chocolate cake. In the lexicon, the meaning specification for a compound
has to be very general, in this case 'cake that is associated with chocolate'. This
is because cake can also link up to form a compound with many other words, e.g.:
(8) birthday cake
oil cake
fish cake
marble cake
224 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

Clearly, if chocolate cake were shown in the lexicon to mean 'cake one of
whose major ingredients is chocolate', this definition would extend (replacing
'chocolate' with 'fish') to fish cake but not to birthday cake or to marble cake (cake
which looks like marble) and certainly not to oil cake (food for animals, since this
is food from which the oil has been removed). This does not only apply to
compounds with cake but to thousands of other words that enter into compounds.
The moral is that the meaning of compounds involves a combination of a very
general specification in the lexicon and processes of pragmatic enrichment
Our final example of pragmatic enrichment comes from the area of
metonymy. Very often, instead of using an exact description, we cut corners and
mention something that is merely connected with what we mean, leaving it up to
pragmatic enrichment to ensure successful communication. Here are some
examples of this procedure:
(9) a. I love to listen to Baroq,ue composers.
b. The entire school came out to greet the mayor.
c. The Caesar salad left without paying.

In (9a), the Baroq,ue composers are long dead, so the speaker can only mean
that she is fond of their music; in (9b ), the school cannot move, so the speaker
must mean all the teachers and pupils came out; and in (9c ), a dish of food also
cannot move, so the speaker must mean the person who ordered and consumed
a Caesar salad in a restaurant. Notice how the hearer needs to assume that the
speaker is being cooperative and is not speaking nonsense: he uses pragmatic
enrichment to 'fill in the gaps' in what the speaker has said. In any case, the
lexical meaning of the words composer, school and salad has not changed. It
should be pointed out, however, that the regular metonymic use of a word can on
occasion lead to its lexical meaning changing. Drinking vessels were traditionally
made of metal, but when it became possible to make them from glass, the first
uses of glass must have been metonymic since the lexical entry for glass had the
feature [-countable] (and in one of its meanings still does). As a result of
the freq,uent metonymic use of glass, there arose in the language alongside the
original glass 1 (the uncountable substance) a countable noun glass 2, meaning a
drinking vessel made of glass.
In 3.3.3.5 above, we mentioned Horn's (1984: 13) Q Principle and R Principle.
These are relevant to an aspect of lexical pragmatics known as narrowing (see
also Huang 2009). This applies where understanding a word involves narrowing
down its interpretation. Let us talk of Q-narrowing where the Q Principle ("Make
your contribution sufficient. Say as much as you can.") is involved. Consider the
following example:
(10) Margaret broke her finger.
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I Other topics of interest in pragmatics 225

On the assumption that Margaret has ten fingers, the hearer understands
that Margaret broke one of them. If one of the fingers had been a thumb, then
according to the Q Principle the speaker would have specified that it was a
thumb; since she did not do so, the hearer understands that it was one of the
remaining eight fingers. In all cases of Q-narrowing, the speaker is at liberty to
cancel the inference, as in (11):
(11) Margaret broke her finger, in fact it was her thumb.

R-narrowing involves an appeal to the R Principle ("Make your contribution

necessary. Say no more than you must."). Consider examples like the following:
(12) a. Could I have some milk?
b. Could I have a drink?

The lexical entry for milk covers the lactic fluid of all mammals, but here the
speaker is understood to mean cow's milk. Similarly, the lexical entry for drink
covers all potable liq,uids, but in most layers of western culture, the hearer will
understand (12b) as a req,uest for an alcoholic drink. This is R-narrowing because
the speaker is not saying more than is necessary to have the desired effect. A
speaker will only tend to say some cow's milk or an alcoholic drink if their hearer
is - let's say - a drinker of soy milk or a teetotaller. R-narrowing often exposes
the stereotyped way in which some people think. Consider the following cases:
(13) My neighbor is a doctor.
(14) My cousin worked as a prostitute.

There is nothing in the lexical entries neighbor, doctor, cousin and prostitute
to indicate the sex of the referent. Nevertheless, responses like Oh really, where
does he work? and Oh, how awful for her! respectively show that the speaker's
conversational partner assumes that doctors are male and prostitutes female and
has R-narrowed the interpretation in line with those (possibly erroneous)
assumptions. Finally, R-narrowing explains how we can sometimes communicate
by means of apparently vacuous statements such as those in (15):
(15) a. My child has a temperature.
b. My daughter has a brain.
c. My son has muscles.
d. Now, that's a salad!

In (15a), the child has a fever, a higher body temperature than is healthy;
(15b) is understood to mean that the daughter is unusually intelligent and (15c)
to mean that the son has strong, well-defined muscles. (15d), finally, if pronounced
with an appropriately enthusiastic intonation, will be understood to mean that the
226 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

speaker is referring to a fine example of a salad, with many different appetizing

components.

The purpose of this section has been to show that, from the point of view of
language use, the lexicon covers only part of the meaning associated with words.
In order to understand how words are used in context we need such pragmatic
concepts as enrichment and narrowing. In discussing (15d) we mentioned the
pragmatic role of intonation. That will be the topic of the next subsection.

6.2.3. PRAGMATICS AND PHONOLOGY: PROSODY

The examples we provide in this book are all given in the form of written
language, with a few images. However, this means that a very important, some
would say vital, part of language behavior has been missing from the eq,uation,
the phonology of speech. In all the preceding chapters we have mentioned
intonation (e.g. in 1.6.1.2, 2.3.3, 3.3.3.1, 4.2 and 5.6). We have been expecting you,
as readers, to reconstitute the intended sound of the utterances on the basis of a
combination of words and punctuation marks, sometimes with the additional use
of capital letters to indicate emphasis. Nevertheless, the written word can never
be more than a poor approximation to the full richness of connected speech, with
its subtle distinctions of melody, pitch, rhythm, tempo, voice q,uality and
amplitude ( = loudness). In order to enrich the study of language, greater
attention is now being paid (e.g. Wilson & Wharton 2006, Barth-Weingarten
et al. 2009, Prieto & Rigau 2012) to the relation between pragmatics and
prosody, the latter being defined as the phonology of syllables and longer
stretches of speech.

It is fair to say that many domains of linguistics have tended to ignore

prosody. Not only have many linguists been content to work with decontextualized
items (see 6.2.1 above) but they also have tended to exclude prosodic q_uestions
from their work. One reason for the disregard of prosody has been the sense that
there is enormous individual variation (in sex, age, dialect, emotional state, etc.)
making it hard to discern patterns. Another has been a feeling that prosody
cannot be reduced to the neat oppositions like [± animate], [± plural] or [± voiced)
that linguistics discern in semantics, syntax and segmental phonology
respectively. A third reason has been the suspicion that although there is some
connection between pragmatics and prosody, it is unclear whether there are ang
regular mappings between the two domains that would allow a researcher to
reliably recognize, say, an ironic intonation. A fourth reason has been that non-
pragmatic factors can have an influence on prosody, for example the two
syntactic structures of old men and women, namely [old [men and women] and
[ [old men] and women]. In addition, in tone languages such as Chinese
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I Other topics of interest in pragmatics 227

Vietnamese or (to some extent) Swedish the difference between words is signalled
by prosodic contrasts: in Swedish tanken means either 'the tank' or 'the thought',
depending on the tones applied.
There have been q_uite some disagreements about the best way to notate
prosodic contours. The early work in this area by Halliday (1967) and Brazil
(1997) was based on the authors' personal observations and introspections and
thus - although brilliant - was inevitably primarily applicable only to the variety
of English they spoke and was rather holistic in approach, not revealing detailed
properties of utterances. More recently, techniq_ues (e.g. Boersma 2001) have
become available for the very precise measurement of the acoustic properties of
recorded speech. This in turn has led to the development and general acceptance
of the ToBI (Tones and Break Indices) framework (Beckman et al. 2005), a set of
conventions for labeling the distinctive properties of the sound spectrographs
of utterances, no matter what language, or what variety of a language, is being
spoken. The use of these techniq_ues req,uires extensive training, but there is no
doubt that the study of prosody has matured in recent years, allowing a fresh
consideration of its connection to pragmatics.
In the model of Functional Discourse Grammar (Hengeveld & Mackenzie
2008) mentioned in 6.2.1, there is a direct connection between the Interpersonal
Level of the grammar (at which the grammatical conseq_uences of pragmatic
strategies are represented) and the Phonological Level (at which prosody is
treated). This allows the grammar, for example, to show how distinct illocutions
(Declarative, Interrogative, Imperative, etc.) are reflected in different intonation
contours. Sometimes, an interrogative may have the syntactic form of a
declarative, as in (16):
(16) You said the priest had arrived?

Here, a rising intonation (L*H in ToBI) can be used to help indicate that the
speaker is just checking rather than asking a new q_uestion. Where there is a
mismatch between syntax and phonology, as in this case, we will expect the
speaker to signal the rise in her voice more clearly than otherwise, since it is
the prosody that alone carries the burden of indicating the nature of the speech
act. Intonation is often used to signal contrast, as in (17), where capitalization is
used to represent prosodic prominence:
(17) Henry wore a conservative suit and tie, but his WIFE was in HIGH HEELS
and a LOW-CUT DRESS.

If the speaker- for whatever reason (approval, disapproval, ... ) -wants to

bring out the implications of the contrast between the fashion styles of Henry and
his spouse, she can use marked intonation and a slower speech rate to emphasize
228 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

wife, high heels and low-cut dress. This kind of effect carries over to cases where
a speaker wishes to trigger a Grice an implicature (cf. 3.3.3.1 ):
(18) Marcia: Did Bruce contribute to the fund?
Noreen: Well, CHARLIE did.
By using a contrastive intonation, Noreen can further ensure that her words
will trigger the desired implicature - we say 'further ensure', since the flouting
of the Maxim of Relation and the pragmatic marker well will already have
this effect. At times, greater reliance is placed on the prosody, as in an example
like (19):
(19) a. I have had some (Ism/) nice experiences here.
b. I have had SOME nice experiences here.

If the speaker does not give prominence to some, as in (19a), then the
message will be purely positive, with no implicatures being sparked off. However
if the speaker emphasizes some, as in (19b ), the implicature will be that not all her
experiences were pleasant.
Even when there is no intention to trigger implicatures, the placement of
what is called nuclear stress depends on the context. Nuclear stress involves not
only prosodic prominence but also a change of pitch, either upwards or
downwards or some combination of those movements, on the affected syllable(s).
Consider the following examples:
(20) a. Tom never gets up EARLY.
b. Tom NEVER gets up early.
c. Tom never gets UP early either.
d. It is TOM who never gets up early.

The meaning of all four examples is the same, but they fit into q,uite differen
contexts. (20a) is appropriate in a neutral context, where Tom is the topic and the
rest of the clause is a comment about him; so it could occur in response to, for
instance, Tell me something about Tom. (20b), with nuclear stress on the negati e
word never, fits into other contexts, for example one in which it is 10 a.m. and
Tom has still not appeared for breakfast. (20c) is most natural where the
conversation has been about Tom's tardiness in everything: it could occur after.
say , Tom is always late for meals and forgets his appointments .... (20d).
finally, shows the special syntactic structure known as the cleft construction; this
construction (cf. our discussion of (2c) above) is often used to correct an
erroneous assumption that has arisen in the context. In (20d) it is used to poin:
out that Tom, not someone else, is the late riser. In cleft constructions it is the unit
that follows It + be that carries the nuclear stress.
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I Other topics of interest in pragmatics 229

Various hypotheses have been put forward about the role of prosody in
politeness and these are now being examined. So far, the results have suggested
that some of the generalizations proposed in earlier work were too hasty: for
example Brown & Levinson's speculation that positive politeness is associated
with a low, creaky voice and negative politeness with a high pitch (1987: 267-268)
has been shown not to apply generally across languages (Grawunder & Winter
2010). The general conclusion among researchers (e.g. Prieto & Rigau 2012) is
that there are no inherently polite intonation contours. Prosodic effects can
certainly contribute to the achievement of politeness, but can do so only as one
of part of an ensemble of lexical, grammatical, facial, bodily, discursive and
contextual factors. In work on prosody and Relevance Theory, too, the current
position (Wilson & Wharton 2006) is that prosodic contrasts do not carry
conceptual meaning but rather procedural meaning, suggesting that, much like
discourse markers and facial expressions, they serve to guide the hearer during
the inferential process.
Hirschberg (2007) notes that some studies show a relationship between for
instance, turn-taking phenomena and prosodic elements such as contour, pausal
duration and final lowering (Sacks et al. 1974, Auer 1996, Selting 1996). Other
studies ( e.g. Ladd et al. 1985, Cahn 1989, Mozziconacci 1998) have scrutinized the
role of intonation in conveying affect or emotional state, a topic that we shall
develop in more detail in 6.4.
In addition, according to some scholars ( e.g. Lehiste 1979, Hirschberg &
Nakatani 1996, Swerts 1997, Van Donzel 1999) prosodic phenomena such as pitch
range, rate, loudness, and duration of inter-phrase pause can have discourse-
pragmatic effects related to the topic structure of a text. In this respect,
Hirschberg (2007: 531) observes that:
... phrases beginning new topics are begun in a wider pitch range, are preceded
by a longer pause, are louder and are slower, than other phrases; narrower
range, longer subseq,uent pause, and faster rate characterize topic-final phrases.
Subseq,uent variation in these features then tends to be associated with a topic
shift.
Another feature that has also been identified as a signal of topic shift is
amplitude. Brown et al. (1980), for instance, have found that amplitude seems to
rise at the start of a new topic and fall at the end. Regarding variation in overall
pitch contour, several studies have characterized it as conveying not only the
syntactic mood of the utterance, but also the speech act intended and
the speaker's attitude, belief or emotion (O'Connor & Arnold 1961, Bolinger 1989,
Ladd 1996).
Even though the consensus among researchers is that prosody is intimately
linked with pragmatics, many hypotheses built on impressions and introspections
230 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

have been shown to be ill-founded. It is therefore wise to be cautious about

proposing that prosody is a direct reflection of pragmatic phenomena, the main
reason being that prosody does not work alone. In fact most of these studies have
suggested that prosody interacts with various other phonological variables, as
well as other aspects of human communicative behavior, like gesture, gaze and
body movement. Thus from a more general sociological perspective it has been
shown, for instance, how discourse can be manipulated at the phonological level
in order to achieve certain social or political objectives ( e.g. Margaret Thatcher
was said to change her accent purposefully during her campaign in order to
obtain more votes; Wilson 2001). Also, Wilson & Gunn (1983) observe that
speakers may adopt certain alternative phonological forms in order to sound
more Catholic/ Irish/ Unionist, etc., depending on their persuasive purposes.
In these first sections of the chapter we have seen how speakers resort to
the different linguistic levels and systems in order to fulfill their pragmatic
objectives, and how complex and intertwined the weave of all these elements can
become. One thing we know for sure: there is no uniq,ue or fixed way for all these
levels and systems to interrelate because language is dynamic and the contexts
in which utterances occur are also dynamic and ever-changing. Therefore we
cannot speak of strict one-to-one mappings when analyzing these variables. In
language and discourse nothing can be said to be static, as we shall also try to
show in the next section, dealing with another of the discourse-pragmatic
functions of language, namely that of positioning oneself towards the discourse
situation as well as evaluating people, things and events.

6.3. STANCE AND EVALUATION AS A DYNAMICAL,

PRAG MATIC SYSTEM
Language is very often used to reflect our systems of values and therefore to
express our evaluations of people, things and situations, as well as our emotions.
Scholarly reflection on this fact has stimulated the study of the evaluative
function of language, which has gradually grown in strength within the linguistic
research of the late 20th century and first two decades of the 21 st century. Bally
(1913), Labov & Waletsky (1967), Labov (1972, 1997a and b) and Aarts & Calbert
(1979) were among the authors whose early approaches paved the way for the
more recent approaches to the phenomenon. Bally, a Swiss linguist who
integrated axiology, the study of values and value judgments, into linguistics in
the early 20th century, noted that the axiological parameter permeates most
linguistic expressions, pointing to the close relationship between value
judgements and the expression of linguistic emotion. Labov examined evaluation
I I I I I I I I I I/ I I I I I I I I I I I I I I I I I I I I I I I I I Other topics of interest in pragmatics 231

as one of the elements of narrative structure, and Aarts & Calbert contributed
semantic taxonomies of values for the English language.
Adding a new dimension to the true/false axis that has always been given
great prominence in linguistics, many researchers in the field of Axiological
Semantics (e.g. Osgood et al. 1957, Osgood 1980, Krzeszowski 1990, Felices-Lago
1997, 2003) started to argue that the most important axis is in fact the good/bad
one. They claimed that this positive/negative axis is a much more basic parameter
in linguistics, being founded on very deep emotions and thus is the earliest
distinction (in terms of values) that human babies make. This basic categorization
manifests itself in language in various different ways, reflecting the more or less
complex systems of values adhered to by different individuals and social groups.
A new wider perspective was arising, then, which conceived of linguistic
evaluation as essentially discourse-pragmatic, interpersonal, context-dependent
and negotiable. This broader view has been developed in works such as Hunston
& Thompson (2000), Martin (2000), Macken-Horarik & Martin (2003), Halliday &
Matthiessen (2004), Hunston (2008), Martin & White (2005), Englebretson (2007),
Bednarek (e.g. 2006, 2008a & b, 2009a & b) and Thompson & Alba-Juez (2014).
These authors have noted, for instance, that the values reflected in language are
not always the speaker's, for they may reflect the 'social mind' or the values of
the corresponding culture or group within which the speaker is interacting. As
Bednarek (2009: 405) explains, they may be strategically related to notions such
as self-presentation or politeness, thus becoming the mirror of the 'social mind'
or culture.
The Appraisal Model developed by Martin & White (2005) stands out as by
far the most elaborate and fully developed model of evaluation in the literature
to date. They define Appraisal as "an interpersonal system located at the level of
discourse semantics" (2005: 33), which is in turn divided into three main
subsystems or domains, briefly described and illustrated here as follows:
l. Attitude: the subsystem concerned with our feelings, including emotional
reactions, judgements of behavior, and evaluation of things. It is divided into
three further subsystems: Affect, Judgement and Appreciation. Affect is
concerned with registering positive and negative emotions or feelings;
Judgement deals with attitudes towards people's behavior which we praise or
criticize; and Appreciation involves evaluations of things (semiotic and
natural phenomena).
2. Engagement: the subsystem dealing with the source of attitudes and the play
of voices in discourse.
3. Graduation: the subsystem concerned with adjusting the gradability of an
evaluation.
232 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

Prototypical examples of evaluation within the major subsystem of Attitude

for each of its three subsystems can be found in (21), (22) and (23), where an
emotion, a judgement of a person, and an appraisal of a movie are expressed,
respectively:

(21) I fell in love with your blue eyes the moment I met you. (Affect)
(22) It was silly of her to say such a thing to you. (Judgement)
(23) The movie was boring and totally simplistic in its approach to life.
(Appreciation)

The subsystem of Engagement is based on the assumption that everything

speakers say encodes a certain point of view or stance, and that all verbal
communication is dialogic. 67 Thus all utterances show how their speakers present
themselves (i.e. as standing with, against, or neutral with respect to other
speakers and other value positions). But even if all communication is dialogic,
speakers on certain occasions do not make reference to other voices and
viewpoints in their utterances, in which case their Engagement is categorized by
Martin & White as 'Monoglossic'. On the contrary, when the utterances do invoke
or allow dialogistic alternatives, they are categorized as 'Heteroglossic' (2005:
100). In (24) and (25) we find examples of Monoglossic and Heteroglossic
Engagement found in the headlines of a broadsheet and a tabloid, respectively:
(24) Jordan hangs 11 men after eight-year halt to death penalty (The
Guardian, December 21 st, 2014).
(25) I've called time on boozing - after 24 bottles in two weeks, says LIZ JONES
(The Daily Mail, January 3rd, 2015).

In (24) the journalist who wrote the headline does not recognize any
dialogistic alternative: she just states as a fact that 11 men have been hanged in
Jordan. On the contrary, in (25) the journalist does not take responsibility fo r
what is stated and explicitly states that this is what Liz Jones said, thereby
acknowledging the dialogistic alternative by introducing another person's
voice.
However, reporting what other people say is not the only way a speaker can
recognize other voices. Interrogative utterances, for instance, are alway s
dialogistic and heteroglossic, because they recognize an alternative to what
would be their declarative counterpart. The same can be said of other utterances
that contain some clue as to the consideration of other people's stances or
opinions. Compare the monoglossic utterance in (26) to some of its possible
heteroglossic alternatives in (27), (28), (29) and (30):

67 See Bakhtin (1981).

I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I/ I Other topics of interest in pragmatics 233

(26) Robert has been rude.

(27) Has Robert been rude?
(28) In my view, Robert has been rude.
(29) Everyone thinks Robert has been rude.
(30) Nobody can deny Robert has been rude.

As for the third major subsystem of Graduation, Martin & White point out
that it is a general property of all values of Affect, Judgement and Appreciation
that they construe greater or lesser degrees of positivity or negativity, and that
gradability is also generally a feature of the Engagement system (2005: 135). The
system of Graduation allows speakers to grade according to intensity or amount
(Force), as well as by reference to the degree to which the graduation phenomena
match some supposed core or instance of a semantic category (Focus). The
utterance in (31) presents an example of Graduation of the Force type, and that
in (32) an instance of the Focus type:

(31) Kate is an extremely generous person. (Graduation: Force)

(32) Leo has been acting kind of weird lately. (Graduation: Focus)

All the systems and subsystems of Appraisal interact and interrelate

in various and intricate ways, which is evidence of the fact that linguistic
evaluation is a complex, context-dependent phenomenon, as other authors have
also argued. In a more recent publication, Alba-Juez & Thompson (2014: 13)
define evaluation as:

... a dynamical sub-system of language, permeating all linguistic levels and

involving the expression of the speaker's or writer's attitude or stance towards,
viewpoint on, or feelings about the entities or propositions that s/he is talking
about, which entails relational work including the (possible and prototypically
expected and subseq,uent) response of the hearer or (potential) audience. This
relational work is generally related to the speaker's and/or the hearer's personal,
group, or cultural set of values.

These authors also point out that in their view, stance and evaluation are not
exactly the same thing. Stance is a more abstract concept, a position a speaker
may hold for herself without expressing it, whereas evaluation is the actual
verbal realization or manifestation (through linguistic means) of the stance.
Evaluation is thus characterized as a complex, pragmatic, dynamical phenomenon
that has many faces, for it may: a) be overt or covert, b) be placed at different
points along the continuum on the polar evaluative positive/negative axis,
c) express one, two or more different values or parameters at the same time, and
d) permeate all the linguistic levels. Examples (33) and (34) illustrate overt as
234 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

opposed to covert evaluation (inscribed vs. invoked evaluation in Martin &

White's 2005 terms), respectively:
(33) Jonathan is a liar.
(34) Jonathan is very fond of making up stories about other people.

Whereas the criticism of Jonathan is made clear and is inscribed in the

semantics of the word liar in (33), in (34) it is made in an indirect way, invoking
the judgement by describing what he is fond of doing, thus leaving the ultimate
assessment about Jonathan up to the hearer's own interpretation and deduction.
With respect to polarity, evaluations can be positive, as in examples (21) and
(31), or negative as in (22), (23), (26) or (33). However, evaluations are not always
so clearly positioned at one pole or the other: they may be placed on a continuum
whose midpoint is a neutral kind of evaluation, examples of which are found in
(35), where the speaker commits neither to one extreme nor the other, and
in (36), where a simple, rather objective appreciation68 of the River Nile is given
(also involving Monoglossic Engagement), without showing or implying any
positive or negative attribute:
(35) I neither love nor hate her.
(36) The Nile is a long river.

Regarding point c) above, the fact that different kinds of evaluation can be
found in a single utterance also constitutes proof of its complexity. In (23) above
for instance, there is a combination of the subsystem of Appreciation within
Attitude and the subsystem of Graduation (found in totally). An even more
complex interplay of systems and subsystems can be seen in (37), an analysis 69
of a headline in The Guardian (18 December 2014) where a combination can be
observed of Heteroglossic Engagement (the journalist is reporting what the
Court said), two types of Attitude (an Appreciation of the way in which the father
shook the baby, and an invoked Judgment of the father as a murderer), and
Graduation (the father's loss of self-control is assessed as "momentary", perhaps
in an attempt to minimize his murderous intention, in the sense that it was not
premeditated but only the result of a moment of loss of self-control) (the analysis
and classification is within sq,uare brackets):
(37) Father fatally [Attitude (negative, inscribed Appreciation)] shook babg
[Attitude (invoked negative Judgment)] in momentary [Graduatio n
(downscaling)} loss of self-control [Attitude (invoked negative
Judgement)], court told. [Whole headline: Engagement (Heteroglossic)].

68 This is called "descriptive appraisal" by philosophers like Richard Joyce (Price 2014).
69 Reproduced here from Alba-Juez (forthcoming).
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I Other topics of interest in pragmatics 235

Our final remark about evaluation will concern d) above, namely the fact
that evaluation permeates all the levels of linguistic description. At the
phonological level, it is well-known that certain prosodic features ( e.g. a given
intonation or high pitch) may be used, for instance, to indicate an ironic criticism
or to infuse utterances with different types of evaluative/emotive meanings. The
example in (38) has been taken from Escandell-Vidal, Marrero Aguiar & Perez
Ocon (2014: 163), who show how the lengthening of the nuclear stress that
Spanish uses to indicate verum focus 70 (in this particular example, on the vowel
/i/ in the second syllable of abrigo in A's last utterance) may be linked to an
evaluative message of insistence and/or impatience:
(38) A: Abrfgate bien, q,ue dicen q,ue va a hacer mucho frio.
(It will be very cold outside, so wrap up warm)

B: jPero si hace un sol esplendido!

(But the sun is shining bright!)
A: Llevate el abri:go
(Do take your coat)

At the morphological level, evaluation can be encoded in some prefixes and

suffixes, such as the suffix-ie in sweetie (showing affection), or the prefixpro-
in pro-life (showing a positive stance towards any measure in favor of life, and
in some contexts, a simultaneous negative attitude towards, for instance,
abortion).
Evaluation at the lexical level is normally the easiest to recognize, given that
many words or expressions (e.g. good, bad, pretty, ugly) include positive or
negative evaluative content ingrained within their semantic features.
At the syntactic level, evaluation can be made manifest, for instance by
resorting to the lexico-grammatical systems of modality (Halliday 1994, Eggins
& Slade 1997, Bybee & Fleischman 1995, Biber & Finegan 1989) or by changing
the expected order of constituents in a clause. Compare (39), where the order of
constituents is the expected one, to (40), where the adjunct has been fronted,
adding an element of Graduation (Intensification) to Jennifer's invoked negative
evaluation of Ernest (because by fronting the negative adjunct she puts emphasis
on the fact that she does not want to visit him, making her interlocutors infer,
given the proper context, that she does not like Ernest). Jennifer's utterance has
to be placed in a context where her friends are planning on visiting Ernest and
giving him a surprise, but where Jennifer, who is Ernest's ex-girlfriend and had a

70 Verumfocus is also called 'positive polarity focus', and is defined by Fery as a "special case
of narrow focus, namely on the affirmative part of a declarative sentence" (2006: 167).
236 Other topics of interest in pragmatics / / / / / i I I I I I I I I! I I I Ii I I I I I I I I I I I I I I I I

very bad experience in her relationship with him, now hates him and wants to
avoid any possible contact with him:
(39) I will not visit Ernest on anu account.
(40) On no account will I visit Ernest.
At the 'purelu' semantic level, we find, as in the lexicon, the evaluative
meanings that are inherent to words and expressions and are therefore a part of
their semantic features. At the broader semantic/pragmatic level we find a kind
of evaluation that is context-dependent, ranging from the local co-text up to
personal, social, emotional or cultural contexts. Thus, as Alba-Juez & Thompson
(2014: 11) note, "'a fat woman' was considered positive, beautiful and healthu at
the beginning of the 20th centuru, but not so much so in our present Western
culture". However, fatness mau become desirable when talking about pigs that
are being raised for food. Another interesting example of pragmatic evaluation
is to be spotted in the utterance He wears sandals with socks which, according
to Thompson (2014: 51), is recognized bu manu British readers/hearers as an
indirect (invoked) negative judgement of the sandal-wearer's taste. Indeed, in
order for the hearer to understand pragmatic-level evaluation, he has to share or
at least have some knowledge of the speaker's personal, social and cultural
values, or else the invoked meaning will be lost or misinterpreted. It is important
to note here that an utterance that has a positive valence within a certain group
or culture mau have a negative meaning in another or vice-versa and that
evaluative language also depends on and varies according to the text tupe in
which it is used.
As the reader mau have noticed, a crucial aspect of evaluation is that it is
closelu connected to emotions and the expressive function of language. We now
turn to these.

6.4. EMOTION: INTERSUBJECTIVE PRAGMATIC

MEANINGS
20 th-centuru linguistics was mainly centered on the referential function of
language. As was noted in 6.2, it was onlu at the end of the 20th and beginning
of the 21 st centuries that other functions of language started to be explored in
depth, such as the evaluative and the expressive functions. We have seen that the
expression of evaluation is veru closely connected to the expression of emotion:
human beings normallu have opinions and feelings about other people, things or
situations, which are expressed by both verbal and non-verbal means (e.g. gestures
posture). Verbal communication, as we now know, plaus a very important part in
/II I I/ I I I/ I I/ I I I I I I I I I I I I I/ I I I I I I I I I Other topics of interest in pragmatics 237

the expression of stance, attitude and emotions. In fact, there is very little
language (if any at all) that can be said to be devoid of stance or emotions.
As we saw in the previous section, theories of evaluation have incorporated
emotion into their conceptual schemes. Appraisal Theory includes the subsystem
of Affect, involving emotions, within the major subsystem of Attitude. But the
researchers who are now exploring emotion in detail (e.g. Foolen 2012, 2016,
Dewaele 2013, or the EMO-Fundett group 71 ) have seen that there is more to
emotion than just the subsystem of Affect: Thompson (2014) has shown how both
Emotion and Emotional talk (using Bednarek's 2008b terms) are not only found
in the Affect subsystem but also overlay Judgement and Appreciation, as well as
the main areas of Engagement and Graduation. The reflections developed within
our research group (EMO-Fundett) have led us to hypothesize that all emotive
language is evaluative but not all evaluative language is emotive. Testing this
hypothesis will depend upon how we define emotion (a difficult task indeed!),
which will involve considerations from not only linguistics but also psychology
and sociology. This matter, as with all issues in the Humanities, is complex, and
we cannot yet say that all the variables that underpin the phenomenon have been
established. Above all, there is still no consensus as to which and how many
emotions there are. There have been many attempts to categorize and represent
the human emotional space (see for instance the psychological theories of Russell
1980, James 1884, or Ekman 2007) but there is a great diversity of opinions as to
how this should be done. Within the field of linguistics, Schnoebelen (2012) points
out that the basic categories which are mentioned in the majority of studies are ten
in number, namely: 1) Anger/rage/hostility, 2) Fear/fright/terror, 3) Joy/happiness/
elation/enjoyment, 4) Sadness/sorrow/distress/dejection, 5) Disgust, 6) Shame,
7) Love/tender emotion, 8) Anxiety/worry, 9) Surprise, and 10) Guilt. But this is by
no means the standard, for some authors only work with two main types (positive
vs. negative emotions), while others speak of an infinity of categories.
We shall try, however, to present here a succinct picture of what has been
done so far in the field of linguistics regarding this important aspect of human
communication, taking into account that emotions are crucial for the
comprehension of human nature and interaction, and that the expression of
emotion is therefore essentially a pragmatic phenomenon. For that reason, in
Cognitive Linguistics the conceptualization and expression of emotion is
considered to be a natural function of language.
Dewaele (2013: 6) points out that, when learning a language, knowing the
affective valence and degree of emotionality of a word or expression is as
important as learning its grammatical class or its gender, because an inaccurate

71 http://www.uned.es/proyectofundett/.
238 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

understanding of this aspect may result in unwanted illocutionary effects "which

might be more embarrassing than phonological, morphological or syntactical
errors". Dewaele also remarks that emotion can be a very strong motivating
force behind the acq_uisition of a foreign language. Furthermore, other authors
( e.g. Ar bib 2005 or Tomasello 2008) support the thesis that emotion was an
important basis for the emergence of human language in the first place, given
human beings' obvious need and desire to share their emotions and/or attitudes
with others. Within Cognitive Linguistics, Langacker (2012: 100) has confirmed
this general perception by stating that every instance of language use "has
conceptual import involving four dimensions: descriptive, expressive/emotive
interactive, and discursive".
All these considerations make it rather difficult to conceive of communicating
in any natural language without expressing emotion. Foolen (2016) points out
that emotions are relevant for language for two main reasons: a) because they
are expressed in language, and b) because they are part and parcel of the
grammar and lexicon of languages. The difference between these two aspects is
illustrated in (41), where Lucy's pain immediately after having hammered her
finger is directly conveyed and expressed but not described, as opposed to (42)
where Lucy shows that she has conceptualized the feeling of pain and is
describing it by making use of the grammar and vocabulary of English:
(41) Lucy: Ouch!!!!!
(42) Lucy: My finger hurts! It's a terrible, acute pain.

These two options, i.e. the direct expression and the conceptualization of
emotion, are in fact two of the ways in which the relationship among cognition,
language and emotion can be seen. Foolen (2012) explains that there are four
ways to see this relationship:
1) We might hypothesize that there is no direct connection between language
and emotion, and that cognition always stands as an intermediary between
them: emotion is conceptualized in cognition, and cognition is in turn reflected
in language, as seems to be the case with the lexical differentiation between
emotions (I'm happy, sad, angry, etc.). This possibility is illustrated in
Figure 1:

Figure I: Hypothesis I: There is no direct connection between

lan9ua9e and emotion

EMOTION · COGNITION LANGUAGE ·'

I I I I I I I I I I I I I! I I I I I I I I I I I I I I I I I I I I I I Other topics of interest in pragmatics 239

2) We might also think that language always has a direct connection to emotion,
as seems to be the case with verbal utterances such as Ouch! or Yummy!, a
possibility that is illustrated in Figure 2:

Figure 2: Hypothesis 2: Language and emotion are directly

connected

EMOTION .

LANGUAGE .

3) A more reasonable hypothesis, in our opinion (which is in line with Foolen's),

would be that language has both a direct and an indirect link to emotion,
i.e. language both reflects the conceptualization of emotion and expresses it,
as represented in Figure 3:

Figure 3: Hypothesis 3: Language has both a direct and

an indirect link to emotion

1 + 2
'

•. EMOTION
EMO'flON LA;NGUAGE

LANGUAGE

4) A final hypothesis would state that the relation between language and
emotion varies, depending on the types of emotion. Thus, Foolen explains
that it could be claimed that for instance, surprise is typically expressed,
while anger or fear are only conceptualized in language but expressed in non-
verbal ways. There are no conclusive studies or results that could support this
hypothesis.

As noted in 3), our view of this issue tends towards conceiving of this
relationship as 'mixed', given the observable fact that people not only have the
ability to conceptualize their emotions and those of others (e.g. I love ice-cream,
but my boyfriend hates it), but also to express their own emotions directly via
240 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

language, resulting in expressive language ( e.g. Argh! Eek! Pew!) and that at
times these two possibilities may be conflated in the same expression (e.g. That
news is soooo good!!!!!).
Potts (2007) writes about the q,ualities of expressives, the most salient of
which are the following: a) they have an immediate and powerful impact on the
context; b) they reveal the perspective from which the utterance is made and may
have a dramatic impact on the way current and future utterances are perceived
by the hearer or audience; c) their meaning cannot be easily articulated; and d)
they are indispensable to language.
Let us discuss some of the ways in which emotion, as we saw was also the
case for evaluation, is encoded in language at different levels.
Some words or morphemes have special emotional meanings: The -k that is
appended to Cantonese particles acts as an emotion intensifier (Sy besma & Li
2007). Schnoebelen (2012: 13) points out that Klima (1964) writes of negative
polarity items like any as affective elements that are normally attached to
(negative) emotion, which, for instance, allows "It is (not possible/impossible}
for him to do any more without allowing *It is possible for him to do any more".
Olmos & Ahern (2009) explain that the connectors p ero 'but' and aunq,ue
'although' in Spanish suggest the speaker's positive emotional stance in utterances
like La casa estd lejos pero hay una estaci6 n de metro (The house is far but there
is a metro station).
a) Emotions are taken into account in the categorization of verbs in some
languages like the East Caucasian languages, in which emotion verbs are
split from perception verbs and mark experiencers with a different case.
Schnoebelen (2012: 10) notes that in Godoberi and Tsakhu, for instance, the
dative case is used for the emotional experiencer, while the affective case is
used for the experiencer of a perception.
b) We saw in the previous section that diminutives can also have an evaluative
dimension which interacts with the emotional dimension in obvious ways, as
is the case in Spanish. For instance, if someone is called mi amorcito
(literally: 'my little love') by their partner/lover, the meaning added by the
diminutive -(c)ito to the noun amor (love) has nothing to do with its 'small
size' (it is not the case that the love s/he feels for her/him is little, or that the
addressee is a small person), but much to the contrary, with an intensification
of the closeness, love and affection felt for the addressee.
c) Changing the syntactic order of the constituents of a clause can alter not only
the evaluative load of an utterance (as we saw in 6.3, but also its emotional
load. Compare the following two possible ways of starting an e-mail message
to a friend: Dear John, vs. John, dear, or their eq,uivalents in Spanish: Querido
IIIIIIIIIII IIIIIII III IIIIIIIII III III Other topics of interest in pragmatics 241

Juan: vs. Juan q,uerido:. In both languages the second option, which is the
marked 72 one, will be most surely understood or interpreted as showing a
closer, more affectionate relationship between the writer and John.
d) There is an undeniable relationship, as is also the case with evaluation,
between phonetics/phonology and emotion. Pitch, intonation, high stress or
the lengthening of a syllable or sound is connected in many languages with
strong ( or at least some kind of) emotional involvement, displaying a great
variety of affective meanings which may go from ironic, mild negative
criticism to the dramatic and direct verbal expression of anger, pain, joy or
any other emotion (as for instance, when screaming out of fear).
e) In Computer Mediated Communication, or communication involving any of
the new technologies (mobile phones, tablets, etc.), emoticons have become
a normal feature of messages which aid in the representation of the writer's
emotional states and reactions. As Wharton (2009: 14) remarks, emoticons
encourage the reader "to construct a higher level explicature", 73 and therefore
add to the pragmatic meaning of the message, making up for the absence of
real face gestures. Indeed, there is a connection between emotions and
procedural meaning, 74 as we shall briefly explaining) below.
f) Blakemore (2009) writes about interjections such as Ah, Oh, and Good
heavens! in fiction as a useful way of capturing emotions that are new to a
character or mixtures of emotions like excitement and regret. These
expressions encode "a procedure for activating a range of emotional
attitudes" (2009: 22), a process that the reader will do on his own, given the
context and his own experiences in real life, and that may give more emotional
immediacy as well as increase involvement.
g) Taboo words are typically loaded with emotional content, as we all know, and
this is why Dewaele (2013) points out that they play an important part in the
learning of a foreign language. Furthermore, he writes about his own
experience with the "double standards" of native speakers who do not always
interpret foreign learners' use of expletives in the same way as they would
judge or interpret other native speakers using the same words in similar
situations: a swearword or risq,ue joke uttered by someone who sounds
foreign may be judged as inappropriate or even gross (thereby raising
negative emotions in the natives), while if the same word or joke were uttered
by a native speaker, it would be judged by her peers as totally appropriate

72 Foolen (2012) also remarks that there seems to be a tendency for markedness in language
to be associated with emotion.
73 See Chapter 5 for the concept of higher level explicature.
74 See also Chapter 5 for the concept of procedural meaning.
242 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

(and would therefore be received with positive emotion). In relation to this

topic, Schnoebelen (2012: 17) notes that taboo words "seem to carry with
them the imprint of power, authority, and rebellion".

h) The abstract character of emotions makes them good candidates for

conceptualization by means of metaphors. It is in fact very common to use
figurative language to describe emotions, as in She exploded or He was
boiling with anger, where anger is seen as a fluid in a container. Many of
these metaphors use body parts and organs to refer to emotions, and are
therefore metonymical, as in My heart sank into my boots, He was head
over heels in love with her, He had coldfeet, or His knees trembled. Some
others are orientational, as in She has her ups and downs or We sank into
despair. Crawford (2009) observes that metaphors are used in all kinds of
discourse but they are especially freq,uent when the discourse topic is
emotional, and even more so when dealing with intense emotions. In short, it
appears to be a fact that the use of metaphors and figurative speech in general
is fundamental to expressive language.

To conclude this section, we would like to stress that language is used by

human beings to position themselves with respect to their opinions, thoughts
feelings and emotions, thereby also positioning their audiences and topics. When
expressing emotions, people not only make their internal states visible but also
perform linguistic actions which are interpersonal in nature and have particular
conseq,uences. There is perhaps nothing more human than the verbal expression
of emotion (animals may express certain basic emotions in non-verbal ways, but
cannot talk about them!) and by doing so human beings reveal, and at the same
time affect, certain aspects of the cognitive and social systems they form part of.
It is high time the humanities, and linguistics in particular, paid more attention to
them. We are now in the process of what might be labeled 'a happy start'.

6.5. HISTORICAL PRAGMATICS

It is a fact that pragmatics has played a very important role in much of the recent
historical work on semantic change. Present-day historical linguistics has
branched into several subfields, one of them being historical pragmatics, which
became institutionalized as a field of study in the mid-1990s, mainly thanks to the
work done by Andreas Jucker and the body of research presented in the Journal
of Historical Pragmatics, whose founding editors were Andreas Jucker and Irma
Taavitsainen. Jucker (2006: 329) defines historical pragmatics as "a field of study
that lies at the intersection of historical linguistics and pragmatics". Thus, as
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I Other topics of interest in pragmatics 243

Traugott (2007: 538) has put it, "Historical Pragmatics is a usage-based approach
to language change". The field is also known as Historical Discourse Analysis 75
or Historical Dialogue Analysis, and although the choice of one or the other two
names may imply a particular perspective, the three terms are used largely
interchangeably.
According to Jacobs & Jucker (1995), the field of historical pragmatics is
subdivided into two main types, reflecting the distinction between external and
internal language change respectively: 1) pragmaphilology (a macro-approach
whose focus is on the pragmatic aspects of linguistic change as found in historical
texts within their sociocultural context of communication) and 2) diachronic
pragmatics ( a micro-approach whose focus is on the interface of linguistic
structure and use, and on the historical development of pragmatic elements, such
as speech acts or discourse markers). The data for the former are written or
spoken text types, such as conversations, monologues, etc. viewed in the context
of their legal, pedagogical, etc. norms of production and reproduction. The data
for the latter come from textual evidence of the development of the pragmatic
elements under analysis.
Historical pragmatics can be considered a kind of contrastive analysis,
insofar as different stages in the development of pragmatic elements or issues
down the history of a language are compared. Brinton's (1996) study, for instance,
compares a variety of pragmatic and discourse markers in Old English and in
Middle English, exploring a diachronic form-to-function mapping in which some
markers whose forms stayed relatively stable are taken as a starting point, and
tracing their various functions. Arnovick's (1999) study, by contrast, carries out
a diachronic function-to-form mapping by retracing the histories of a number of
speech acts and speech events. Here the starting point is the function, and what
is explored are the different forms used to realize the function at different stages
in history. The distinction between these two perspectives is, however, sometimes
not easy to make, because certain linguistic elements may change both in form
and in function in the course of time or because the object of research may in
some cases be a whole system of interconnected entities (such as, for instance,
evaluative subsystems or politeness strategies), and not a particular form or
function.
Taavitsainen & Fitzmaurice (2007: 15) note that the methods of research
used in historical pragmatics "rely upon the adaptation and application of current
theories of language to earlier periods in language history", i.e., it has been
generally assumed that the linguistic forces that operate today are the same as

75 This is one more instance in which researchers ( especially those who support the wider
view of pragmatics (macropragmatics) do not seem to find or establish a clear distinction
between discourse analysis and pragmatics, as was explained in Chapter 1.
244 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

operated in the past. However, these authors point out that this assumption may
bring along methodological problems and therefore they commit themselves to
accepting a weaker statement (in line with Romaine 1982: 123) that there is "no
reason for claiming that language did not vary in the same patterned ways in the
past as it has been observed to do today".
Some historical pragmatic studies have used dictionaries as data but much
of the current work is based on large electronic corpora ( e.g. of literary and
biblical texts) or on new corpora which have been compiled for specific research
projects. As for the language scrutinized, the bulk of research in historical
pragmatics has been devoted to the history of the English language, but some
important studies have also been carried out on other languages (e.g. Xing 2004
for Mandarin Chinese, Lunde 2004 for Medieval East Slavic, Bax 2003 on various
rituals (for instance in the Vedic Hymns or in modern Dutch) or Fanego 2010
on Spanish). Other studies have focused on a certain genre: Ungerer (2000)
and Herring (2003), for example, have studied the language of mass media
communication and have shown that historical pragmatic analysis can also be
applied to changes currently in progress (Jucker 2006: 331).
In order to understand a discipline, it is important to know what research
q,uestions are being asked in that discipline. Traugott (2007: 540) states that there
are two central q,uestions that have been asked within the Gricean pragmatic
tradition since the mid-1970s, and a third one that has been posed primarily
within historical linguistics, giving the following three:
1) Do different conversational maxims motivate different types of semantic
change?
2) Does Grice's distinction between particularized and generalized
conversational implicatures help account for how semantic change occurs?
3) Are there additional important factors that need to be considered in
accounting for freq,uently observed types of semantic change?

We do not intend to answer these q,uestions in this section; rather we will

conclude by showing how some scholars have attempted to find the answer to
some fundamental issues by taking a glimpse at how historical pragmatic
research is carried out in the study of discourse markers (DMs). Fanego (2010)
tracks the evolution of the DM de hecho in Spanish and finds that it is analogous
to that of in fact, indeed or actually in English, the essential property of these
markers being that they have primarily procedural meanings (Blakemore 1987,
2002) in present-day Spanish and English and hence are to be distinguished from
sentence adverbials, which have conceptual meaning. She thus shows how these
expressions started out at an earlier period of time as clause-internal adverbials,
then came to function as epistemic sentence adverbials and ended up as discourse
I I I I I I I I I I I I I I I I I I I I I I I I I I I/ I I I I I I I I Other topics of interest in pragmatics 245

markers. Traugott & Dasher (2002: 40) assume the following correlated paths of
directionality in semantic change, viewing pragmatic implicatures as playing a
crucial bridging role in semantic change:

Truth-conditional---- non-truth conditional

Content----- content/ procedural---- procedural
Scope within scope over scope over discourse
Non-subjective---- subjective---- intersubjective

The diagram illustrates the fact that some expressions with a given
conceptual meaning gradually develop a semantic or pragmatic meaning
primarily indexing speaker attitude or viewpoint (subjectivity) and ultimately
speaker's attention to the addressee's self-image (intersubjectivity). This position
is also supported in Traugott's (2007) study of the DM after all, or Alba-Juez &
Martinez Caro's (2015) study of the pragmatic/discourse marker no wonder.
However, as we saw in Chapter 5 and as Fanego (2010) explains, discourse
markers evince different layers of meaning at the same time and therefore cannot
be said to be purely procedural.
Another q,uestion that these and other researchers have posed is whether
this ultimate procedural meaning becomes grammaticalized or conventionalized
in the corresponding language at some point of its historical development. Aijmer
& Simon-Vandenbergen (2004: 1788) note that "some contextual meanings are
more freq,uent and conventionalized than others" so that the answer for DMs has
to be explored case by case. But there is no room for doubt that the processes of
grammaticalization and lexicalization are indeed, and will continue to be, two
of the main concerns of historical pragmatics.
We now turn to another subfield of pragmatics that has developed and grown
in recent years: computational pragmatics.

6.6. PRAGMATICS AND COMPUTATIONAL

LINGUISTICS
The main concern of computational pragmatics, as is also the case with general
pragmatics, is the phenomenon of inference. Jurafsky (2007: 578) defines
computational pragmatics as "the computational study of the relation between
utterances and context", which means that it is concerned with the relationship
between utterances and action, or with the place, time and environment in which
they are produced.
246 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

The basic aim of computational modeling in linguistics has been to provide

mathematical foundations, the formal underpinnings of the models, such as
rules or trees. Another purpose of computational models is to do language
engineering, a major goal within this line of research being to build artificial
agents that are able to carry out a conversation with human beings and therefore
to perform speech acts such as answering a q,uestion, giving directions or even
reacting to people's emotions in an expressive way. This leads us to the third goal
of these models, to develop computational psycholinguistics, defined by
Jurafsky (2007: 579) as "the use of computational techniq,ues to build processing
models of human psycholinguistic performance".

Thus the problem for computational pragmaticians is how to compute the

relationship between language and context, i.e. how to decode those aspects of
linguistic expressions that encode contextual information. This involves, as Bunt
& Black (2000) explain, the difficult task of building effective and explicit
representations at both ends of the relationship. The linguistic side of the
relationship does not pose as many problems as the contextual side, which is
much less well understood by computational linguists, and for that reason they
have not yet been able to generate accurate formal representations of it. It is
assumed that any natural language processing system will need to be able to
understand the intentions behind utterances relative to an evolving context,
which means that it will be capable of making inferences of a non-deductive, or
better, an abductive kind. There are different kinds of abductions, but their
common characteristic is that "the inference mechanism is permitted to assume
additional premises in order to reach a conclusion deductively" (Bunt & Black
2000: 9). As the reader may have understood by now, the pragmatic inferences we
studied in Chapter 3 of this book, such as conversational implicatures, are
abductions. Abduction is formally related to Default Logic, a type of logic
proposed by Reiter (1980) which is concerned with drawing conclusions from
incomplete evidence. Default logic draws inferences of the kind If p, then q,,
unless there is evidence that not q, (Bunt & Black 2000:10). Some authors
(e.g. Errault 1989, Beun 1989, Jurafsky 2007) have used Default Logic to explain,
for instance, how a declarative utterance such as He's dating our friend Marta
may be interpreted as a q,uestion rather than an assertion. But as we know
abductive reasoning may produce non-valid conclusions, which is one of the
reasons why implicatures are cancelable (as explained in Chapter 3).

Another area that preoccupies computational pragmaticians is the

distinction between knowing and believing and, for that matter, the reasoning
that human beings make about their beliefs. A complete computational pragmatic
system should therefore have an inference engine that is able to reason about
beliefs, which poses another very serious challenge to researchers in the field.
I I I I I I I I I// I I I I I I I I I I I/ I I I/ I I I I I I I I I Other topics of interest in pragmatics 247

Some attempts have been made in this respect (e.g. Ballim, Wilks & Barnden
1991, Lee & Wilks 1996), but these are restricted to simple forms of reasoning
about belief. A more recent model is the Rational Speech Act model, developed
by Frank & Goodman (2012), which presents an attempt to model cognitive
aspects of the interpretation of sentence fragments and implicatures.

To conclude and summarize, computational pragmatics is a rapidly growing

field which is to be found at the interface between linguistics, cognitive science
and artificial intelligence and which is very relevant for linguistic pragmatics,
considering that it can enable advanced experimental testing of pragmatic
theories. And it is precisely this subfield, experimental pragmatics, which we
shall deal with in the following section of this chapter.

6.7. EXPERIMENTAL PRAGMATICS

Experimental pragmatics is a relatively new sub-field of pragmatic research that
draws on pragmatics, psycholinguistics and the psychology of reasoning
(Sperber & Noveck 2004).

Within the field of pragmatics alone, there has been very little experimental
evidence that could clearly confirm (or disconfirm) any of the theories we have
discussed in this book (e.g. Gricean, Neo-Gricean or Politeness theories). Not
being familiar with or interested in experimental psychology, pragmaticists
initially had their own ways to account for their findings, and were in general
guided by their intuitions and observational data. It has only been in the 21 st
century that they have started to make use of Experimental Psychology to
obtain experimental evidence that could allow them to evaluate and compare
their pragmatic claims.

As Sperber & Noveck (2004: 8) argue, pragmatic research is not to be censured

or discarded on the grounds that it is mostly based on intuition and observational
data. However, the incorporation into pragmatic studies of the more established
and developed field of experimental psycholinguistics has opened up the scope
of research by adding experimental evidence to researchers' intuitions and
recordings of discursive data, thereby offering them better tools to understand
a variety of topics, from the perception and decoding of the acoustic or visual
signal to the interpretation of discourse in general. The three kinds of evidence
put together (intuitions, observations and experiments) can therefore give high
reliability to the results of research. There are, however, some disadvantages
in including experimental data within pragmatic studies, the main ones being
248 Other +opics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

that they are harder to collect and that they are generally more artificial than
observational data of the type found in corpora, for instance.
Ervas & Gola (2013: 8) point out that "the tension between philosophical and
experimental pragmatics seems to be the dialectic motor of the evolution of
pragmatics itself", and indeed it can be said that in the last decade or so there has
been a considerable body of research in the field supporting this assertion, much
of it coming from Relevance Theorists. Breheny et al. (n.d.), for instance, suggest
that an important place to start research in the field is the source of relevance for
an utterance, by exploring the more general mechanisms that anticipate its
completion. Some research on children's pragmatic performance (e.g. Farroni et
al. 2003, Guasti et al. 2005, Liebal et al. 2009, Southgate et al. 2009) has shown that
babies as young as 4 months of age are already employing sophisticated
pragmatic abilities, which suggests that human pragmatic competence may
involve specific neural structures and thus have an innate basis (Grossman et al.
2008). In Guasti et al.'s (2005) experiment, children had to carry out the felicity
judgement task of obligatorily choosing between two images, one in which all the
characters performed a given action and another in which some did perform it
but some others did not: the 5-year-old children consistently chose the scenario
which respected Grice's expectations of informativeness. Southgate et al.'s (2009)
experiments show that infants seem to understand communicative acts in terms
of what their interlocutor would deem relevant and not simply in terms of what
is relevant to them.
Some other authors have devoted their research to testing the generation of
scalar implicatures, such as Breheny et al. (2006) or Cummins et al. (2012), who
provide experimental data that examines the apparently generalized assumption
that categories of q,uantifying expressions such as more than n, at least n, and
fewer than n, fail to give rise to scalar implicatures in unembedded declarative
contexts. In a later study, Cummins (2015) also investigates, by means of two
experiments, the interpretation of modified fractions.
From a different perspective, Gibbs & Colston (2012) present a new,
dynamical approach to the interpretation of figurative language (e.g. metaphor,
irony) and propose some theoretical alternatives, based on empirical work from
psycholinguistics and neurolinguistics.
All in all, it can be said that psycholinguistic experimental research has shed
and will continue to shed light on the cognitive mechanisms that underpin
pragmatic phenomena. However, as Sperber & Noveck remark, "fruitful
interactions between pragmatics and experimental psychology are not limited to
psycholinguistics" (2004: 12), for ALL experiments that involve human verbal
communication present results concerned with how the participants understand
language and discourse and should conseq,uently involve considerations of the
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I Other topics of interest in pragmatics 249

pragmatic interpretations of, for instance, the instructions given for the correct
execution of the experiments. This opens an enormous window of interdisciplinary
research for pragmatics, where only the sky seems to be the limit.

6.8. CYBERPRAGMATICS
Cyberpragmatics is the term coined by Francisco Yus (2001) to refer to a strand
of pragmatic research devoted to the study of internet-mediated communication
and interaction from a cognitive perspective, and more specifically, from the
viewpoint of Relevance Theory. Yus (2011: xi) argues that this approach can
explain not only face-to-face communication but also asynchronous
communication and the multiplicity of multimodal exchanges on the internet. He
explains that most of these exchanges have an oral connotation, and this is the
reason why most texts in internet-mediated communication are hybrid in nature:
on the one hand they have the rigidity of typed text, but on the other they present
the spontaneity of speech. This feature has also been discussed by other authors
(e.g. Murray 1990 or Herring 2001) for computer-mediated communication in
general. There is a blending of speaking and writing that may cause the
impression that this type of discourse is 'less correct' than standard written
language, due to the fact that interlocutors on the web generally try to mimic the
spoken language as well as to economize on typing effort, among other features
such as the use of symbols or emoticons to compensate for the lack of facial
expressions, sound or body language and the freq,uent use of the so-called
electronic utterance (Sotillo, 2000), a single clause with complements and
adjuncts.
One interesting aspect of internet-mediated communication - Yus notes - is
the fact that it is very common for participants to engage in multi-tasking,
especially in the case of adolescents, "who usually chat with other users while
answering messages from Messenger and searching for information in a portal"
(2011: 12). While doing all this (although it is obvious that more mental effort is
involved76), these users are permanently searching for relevance and dismissing
potentially irrelevant information at the same time. But the social character of
much of internet discourse may bring along some payoffs, such as the
strengthening of community-related assumptions, in order to compensate for this
extra mental effort. However, and at the same time, this social orientation of
internet communication may be detrimental to the estimation of relevance in

76 See Chapter 5 in this book, where the relationship between cognitive effort and contextual
effects within Relevance Theory is explained.
250 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

terms of a reduction of both the credibility of the source and willingness to

process it. Yus even argues that it is very possible that multi-tasking on the web
is altering the way human beings process and store information more generally,
which eventually will affect the organization and functioning of their brains.
They may, for instance, feel reluctant to invest cognitive resources in stimuli that
involve deferred relevance (e.g. reading a novel).
Yus points out that the main objective of cy berpragmatics is to explain the
role of context "in the eventual q,uality of interpretations" (2011: 3), a task for
which Sperber & Wilson's (1986) Relevance Theory has proved useful. In this
respect, he explains that the different forms of internet communication exhibit
variations related to accessibility to context, which has conseq,uences for the
q,uantity and q,uality of information communicated and interpreted. The following
are, then, the hypotheses that "make up the foundations of cyberpragmatics"
(2011: 14):
1) On the Internet, the "addresser users" have communicative intentions and
devise their utterances with the expectation that these intentions will end up
being relevant to the other users and that their utterances will be interpreted
correctly. Since users are aware that, in principle, there are multiple ways in
which their utterances can be coded, they type ( or talk in voice-enabled
Internet communication) with the expectation that these coded utterances
will be adeq,uate evidence to lead "addressee users" effectively towards the
intended interpretation.
2) Internet users use inferential strategies when they interpret messages on the
Net, and these do not differ from the ones used for the comprehension of
utterances in oral conversations shaped by physical co-presence. We are
eq,uipped with a biologically evolved tendency to maximize the relevance of
the utterances that we process, but we do not apply different inferential
procedures for our interpretation of stimuli (verbal and nonverbal) in physical
or virtual contexts.
3) Internet users expect their interlocutors to be able to access the necessary
amount of contextual information that will allow them to arrive at a correct
interpretation of their utterances. In the same way, their interlocutors will
invariably access contextual information as a necessary stage in a relevance-
oriented interpretation of these utterances.
4) The attributes of the different cyber-media (chat rooms, e-mail, messenger,
web pages, social networking sites ... ) influence the q,uality of the user's
access to contextual information, the amount of information obtained, the
interpretation selected, the cognitive effects derived and the mental effort
involved in obtaining these effects. (2011: 14)
/II I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I Other topics of interest in pragmatics 251

Thus, the different cy ber-media are placed on a scale of contextualization

ranging from highly context-saturated media (videoconferencing, Internet-
enabled phone calls, etc.) to highly cues-filtered media (e-mail, instant messaging,
etc.), and so they may have a greater or smaller information gap to be filled in
inferentially by the interactants.
As we know, the Internet world is constantly changing and evolving, and its
impact on our lives has been huge. Some initial forms of internet communication
have become obsolete, but others have emerged and others will emerge in the
future for sure. The challenge for cyberpragmatics will be to answer the new
cognitive and analytical q,uestions that will come along with these new forms.
But cyberpragmatics is not isolated from other strands of pragmatic study.
In fact, all the different avenues of research within pragmatics are interconnected,
and it may very well be the case that we find studies in which, for instance,
theoretic-philosophical, experimental, computational and cyberpragmatics are
all combined and used to attain the objectives of a particular piece of research.

6.9. CONCLUSION
In this chapter we have seen, as was anticipated in Chapter 1, that pragmatic
analysis may be carried out both at a micro level (in relation with the syntax,
lexicon or prosody of utterances) and a macro level - which involves a more
holistic view of the field and takes into account its interaction with other fields
of research ( e.g. historical or computational linguistics), as well as with different
subsystems of language, such as those related to the expression of stance or
emotion.
Our overall intention throughout the book, as also in this chapter, has been
to present this linguistic subfield of research in all its vigor and complexity with
an open mind and heart, in the hope that the reader will make a proper assessment
of how important and necessary it is to look at language from a pragmatic
perspective. Exploring a language without taking into account its pragmatic
aspects is like contemplating a flower without being able to smell its perfume: a
great and fundamental part of the information that we should perceive is lost.
Pragmatics is an extremely rich, healthy, and prosperous field of knowledge. We
hope to have succeeded in making the readers enjoy the trip through all this
richness, health, and prosperity.
252 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

SUMMARY (CHAPTER 6)
1. This chapter deals with the relations between pragmatics and other areas of
academic work, first considering the interaction of pragmatics with syntax,
the lexicon and phonology from a micropragmatic perspective and then, from
a macropragmatic perspective, exploring how language use is intertwined
with value systems and emotions. The chapter ends with a presentation of
historical, computational, experimental and cy berpragmatics.

2. Whereas certain formal theories of syntax held that the structure of clauses
can and should be studied without reference to pragmatics, functional
theories have shown the importance of context for understanding the
relationships among syntactic structures and for resolving ambiguities as
well as exploring the impact of speech acts on syntax.
3. The meanings of words are subtle and complex. Since the componential
meaning representations to be found in the lexicon underdetermine this
complexity, pragmatic enrichment is req,uired to account for many actual
uses of words. In addition, Q-narrowing and R-narrowing offer pragmatically
inspired explanations of how words are understood in context.
4. There is little doubt among scholars that prosody is intimately linked to
pragmatics. The relative prominence of syllables, for example, has been
linked to contextual circumstances, and various aspects of politeness and
relevance have also been connected to prosodic phenomena. The social
embedding of language also has effects on the prosodic characteristics of
speech.
5. One salient use of language is its evaluative function, first studied in
axiological semantics. More recently, a body of work has arisen, most
prominently given form in Martin & White's (2005) Appraisal Model, which
distinguishes the subsystems of Attitude, Engagement and Graduation.
Evaluation permeates all linguistic levels, including semantics, syntax and
phonology, and serves to express the speaker's stance on the matter under
discussion.

6. Important distinctions within Attitude are Affect, Judgement and

Appreciation. Within Engagement, we distinguish Monoglossic
and Heteroglossic Engagement, and within Graduation, Force and Focus.
7. The expression of evaluation is very closely connected to the expression of
emotion. Emotion is more all-embracing than Martin and White's notion
of Affect, impinging on all kinds of evaluation and also playing a vital role in
language acq,uisition.
I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I Other topics of interest in pragmatics 253

8. The relation between cognition, emotion and language is a complex one.

Language can be used both to conceptualize emotion (when we classify
feelings) but also to express emotion in a direct way. There are countless
ways in which language forms and structures can be used to express the full
gamut of emotions, from diminutives to word order, prosody, interjections,
taboo words, metaphor, or the insertion of emoticons.
9. The rise of pragmatics has enriched the study of language change, spawning
such fields as pragmaphilology and diachronic pragmatics. It has emerged
that semantic change follows certain pathways, notably from non-subjective
to subjective and intersubjective uses, opening new opportunities for
expressing pragmatic meanings, for example by means of discourse markers.
10. Computational linguistics has recently also incorporated insights from
pragmatics, giving rise to computational pragmatics. A pragmatically
sensitive computational system for use in artificial intelligence needs to have
(abductive) inferential capacities so that machines can deal with the
fragmentary nature of much human communication; this is the focus of a
major research effort in this field.
11. Although the bulk of research in pragmatics takes place in the philosopher's
study, out in the field or at the computer (with corpora), it has recently also
entered the laboratory in the form of experimental pragmatics. The aim is to
use psycho- and neurolinguistic techniq_ues to determine whether and, if so,
how the inferential mechanisms postulated by pragmaticians operate.
12. Finally, the new field of cyberpragmatics is concerned with applying
Relevance Theory to internet-mediated communication. Given the
prominence of social media in the current world, pragmatics has important
insights to bring to understanding the degrees of context-dependence of the
various forms of internet communication.
254 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

SELF-EVALUATION QUESTIONS (CHAPTER 6)

Choose the most appropriate answer for each q,uestion.
1. What contextual knowledge makes it appropriate to begin the following
sentence the way it begins (i.e. with "The firefighters")?
The firefighters were called to save a little cat which was standing near the
high voltage power lines.
a. The knowledge that power lines are dangerous.
b. The knowledge that firefighters are on call 24 hours a day, seven days a
week.
c. The knowledge that little cats are cute.

2. In English, new information is typically .. .

a. definite, and precedes given information.
b. definite, and neither precedes nor follows given information.
c. indefinite, and follows given information.

3. Within Functional Discourse Grammar (Hengeveld & Mackenzie 2008),

the Interpersonal Level of analysis covers pragmatic aspects of language
such as ...
a. the grouping of discourse acts into conversational moves.
b. the order of constituents in the sentence.
c. the formation of words and sentences.

4. In situations of purely pragmatic communication ( e.g. when a foreigner has

to interact with a monolingual inhabitant of the country s/he is visiting) ...
a. morphology is reduced, the constructions used are simple, and the
interpretation of the message will highly depend on context.
b. morphology is rich, the constructions used are varied, and there is little
dependency on context.
c. morphology is rich, the constructions are complex, and context interacts
with a developed morphosyntax.

5. Behind the concept of underdeterminacy lies the insight speakers (un)

consciously have that ...
a. each word in the language has a uniq,ue meaning.
I I I I I I I I/ I I I I I I I/ I I/ I l I l I I I I I I I I I I I I Other topics of interest in pragmatics 255

b. our knowledge of the world is much richer than the lexicon of words we
use to describe it.
c. there is a different word for every single thing or thought we want to
describe or name.

6. In order to understand the meaning of compound words we need ...

a. only some general specifications about the lexicon.
b. a thorough knowledge of the process of metonymy.
c. a very general specification in the lexicon complemented by processes of
pragmatic enrichment.

7. The utterance You said you're not coming to the party tomorrow? with a
rising intonation (L*H in ToBI) can be used to help indicate that the
speaker ...
a. is asking a new q_uestion.
b. is just checking what her interlocutor said.
c. is inviting the hearer to a party tomorrow.

8. The nuclear stress placed on SUSAN in the utterance It was SUSAN who told
me her family secret will most likely be used to point out that ...
a. Susan, not someone else, told the speaker about her family secret.
b. the speaker knows what Susan's family secret is.
c. Susan should not have told the speaker her family secret.

9. In order to interpret the pragmatic meaning of utterances, we should take

into account that prosody ...
a. works alone.
b. interacts with various other phonological and paralinguistic variables.
c. has no connection with any sociolinguistic factors.

10. Which subsystems of appraisal (according to Martin & White's 2005 Appraisal
Theory) are reflected in the expressions in bold in the utterance Emma
i) told me that John was ii) head over heels iii) in love with her.
a. i) Engagement (monoglossic), ii) Graduation (focus), iii) Appreciation.
b. i) Engagement (heteroglossic), ii) Graduation (force), iii) Affect.
c. i) Engagement (monoglossic), ii) Appreciation, iii) Judgement.
256 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

11. The evaluation made of Cecil as ugly (or at least 'not pretty') in the utterance
If Cecil is pretty, I'm the Archbishop of Canterbury is an instance of ...
a) inscribed, positive evaluation.
b) invoked, positive evaluation.
c) invoked, negative evaluation.

12. By means of the utterance I'm so happy!! in the following exchange, speaker
B ...
A: Hi Lauren, how's it going?
B: Great! I'm so happy!! (with a smile on her face) My son has graduated
from Medical School!!

a. expresses her emotion in a direct, 'non-conceptualized' way.

b. both expresses (by means of intonation and gestures) and conceptualizes
her emotion.
c. does not show any linguistic signs of conceptualization of her emotion.

13. A major concern of historical pragmatics ...

a) is grammaticalization as well as lexicalization.
b) is the narration of personal experience.
c) is the computational study of utterances.

14. Experimental pragmatics is a relatively new sub-field of pragmatic research

that ...
a. focuses on computer-mediated communication.
b. draws mainly on politeness theories.
c. draws mainly on psycholinguistics and the psychology of reasoning.

15. Cyberpragmatics focuses on ...

a. relevant aspects of discourse in multimodal exchanges.
b. spontaneous, hybrid texts and discourse.
c. internet-mediated interaction from the Relevance-theoretical perspective.
/I.II I I I I I//// I/ I/ I//// I I I I I I I/ I///// Other topics of interest in pragmatics 257

DEVELOPING OUR PRAGMATIC COMPETENCE:

EXERCISES FOR REFLECTION AND AMPLIFICATION
(CHAPTER 6)
A) Explain, in no more than two paragraphs and in your own words, how
the scientific study of language has evolved from the mere treatment
of phonology and morphology, to the pragmatic analysis of text/
discourse as occurring within different contexts.
B) Explain the meaning and reference of the compound words and
expressions (in bold) in the table by resorting to pragmatic
enrichment.
. ;.,

.' EXPLANATION BY
COMPOUND WORD/EXPRESSION
,,, C
PRAGMATIC ENRICHMENT
He wrote two words on the
chalkboard.
Little Bill9 wanted a skateboard for
his birthda9.
This is the chessboard Fisher used to
pla!J against Kasparov.
M!J laptop has a touch keyboard.
The whole course went on an
excursion to the Highlands.
I love reading Shakespeare.

C) Choose any video from YouTube where there is spoken interaction in

English and identify at least two instances in which a prosodic variable
( tone, nuclear stress, amplitude, etc.) serves pragmatic meanings and
purposes. Justify your choice of examples.
D) Taking into account the direct connection between the Interpersonal
and the Phonological Levels of the grammar in Functional Discourse
Grammar (Hengeveld & Mackenzie 2008), explain the relationship
between syntax and phonology in the following examples:
a. You're going out in that pink jumper??
b. Ann: Are you guys coming to the beach with me?
Ben: Well, LESLIE is.
c. I will ALWAYS remember you.
258 Other topics of interest in pragmatics / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

E) Examine the following headlines on British online newspapers and

classify the subsystems of Appraisal (Martin & White, 2005) at work
in the expressions in bold, including the type of Engagement reflected
by the whole utterance. Also, specify if the evaluation is inscribed or
invoked, and if it is positive, negative or neutral.
a) Nigel Slater: Cooking is a joyous, lifelong pleasure. (The Guard i an,
20 September 2015).
b) Greek conservatives admit defeat. (BBC Online, 20 September 2015).
c) Canadian leaders gone wild in three-way debate battle. (BBC Online,
15 September 2015).

F) Consider the emotive expressions in bold in the following exchanges

or utterances and say whether the emotion is mainly conceptualized
or directly expressed:
a) (Andy sees Jack coming with an expression of anger on his face)
Andy: Hi Jack, what's the matter with you?
Jack: GRRRRRR!
b) The sudden, strange noise gave me a terrible fright.
c) It makes me so sad to hear such news!
d) Ewwww! There's a worm in your apple!

G) Provide examples of emoticons used to express at least six different

emotions in computer-mediated discourse.

H) Considering the areas of research of historical pragmatics, computational

pragmatics, experimental pragmatics and cyberpragmatics, suggest at
least one topic of research ( with its corresponding research q,uestion)
for each of these four sub-fields of pragmatic study. In other wor~
think of at least one pragmatic aspect for each of these branches of
pragmatics that you think would fit their scope of research.
 REFERENCES
//////////////////////////////////////
Aarts, Jan M.G. & Joseph P. Calbert. 1979. Metaphor and non-metaphor. The
semantics of adjective-noun combinations. Tiibingen: Niemeyer.
Aijmer, Karin. 1996. Conversational routines in English: Convention and
creativity. London: Longman.
Aijmer, Karin & Anne-Marie Simon-Vandenbergen. 2004. A model and
methodology for the study of pragmatic markers: The semantic field of
expectation. Journal of Pragmatics 36. 1781-1805.
Akman, Varol. 1994. When silence may mean derision. Journal of Pragmatics
22. 211-218.
Alba-Juez, Laura. 1998. On the conventionalization of verbal irony in English: A
corpus-based study. In Ignacio Vazq,uez Orta & Ignacio Guillen Galve (eds.),
Perspectivaspragmaticasenlinguisticaaplicada, 11-17. Zaragoza:Anubar.
Alba-Juez, Laura. 1995. Irony and politeness. Revista Espafiola de Linguistica
Aplicada 10. 9-16.
Alba-Juez, Laura. 2000 [1996]. The functions and strategies of ironic
discourse: An analysis. Ph.D Thesis. Madrid: Servicio de Publicaciones
de la Universidad Complutense.
Alba-Juez, Laura. 2007. On the impoliteness of some politeness strategies: A study
and comparison of the use of some pragmatic markers of impoliteness in
British English and American English, Peninsular Spanish and Argentine
Spanish. In Pilar Garces-Conejos, Manuel Padilla Cruz, Reyes Gomez Moron
& Lucia Fernandez Amaya (eds.), Studies in intercultural, cognitive and
social pragmatics, 37-56. Newcastle-upon-Tyne: Cambridge Scholars.
Alba-Juez, Laura. 2009. Perspectives on discourse analysis: Theory and practice.
Newcastle-upon-Tyne: Cambridge Scholars.
Alba-Juez, Laura. Forthcoming. Evaluation in the headlines of broadsheets and
tabloids: A comparative study. In Ruth Breeze (ed.), Evaluation in media
discourse: European perspectives. Berlin: Peter Lang.
Alba-Juez, Laura & Salvatore Attardo. 2014. The evaluative palette of verbal
irony. In Geoff Thompson & Laura Alba-Juez (eds.), Evaluation in context,
93-115. Amsterdam & Philadelphia, PA: John Benjamins.
260 References / / / / / / / / / / / / / / / / / / // / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

Alba-Juez, Laura & Elena Martinez Caro. 2015. 'No wonder' as a marker of
epistemic modality and affective evaluation. In Juan Rafael Zamorano-
Mansilla, Carmen Maiz, Elena Dominguez & Maria Victoria Martin de
la Rosa (eds.), Thinking modally: English and contrastive studies on
modality, 231-248. Newcastle-upon-Tyne: Cambridge Scholars.
Alba-Juez, Laura & Geoff Thompson. 2014. The many faces and phases
of evaluation. In Geoff Thompson & Laura Alba-Juez (eds.) 231-248,
Evaluation in context, 3-23. Amsterdam & Philadelphia, PA: John Benjamins.
Andersen, Gisle & Thorstein Fretheim (eds.). 2000. Pragmatic markers and
propositional attitude. Amsterdam & Philadelphia, PA: John Benjamins.
Arbib, Michael. 2005. The mirror system hypothesis: How did protolanguage
evolve? In Maggie Tallerman (ed.) , Language origins: Perspectives on
evolution, 21-47. Oxford: Oxford University Press.
Ariel, Mira. 2010. Defining pragmatics. Cambridge: Cambridge University Press.
Arndt, Horst & Richard W. Janney. 1979. Interactional and linguistic models
for the analysis of speech data: An integrative approach. Sociolog ia
Internationalis 17(1-2). 3-45.
Arndt, Horst & Richard W. Janney. 1983. Towards an interactional grammar of
spoken English. In John Morreal (ed.), The NinthLACUS Forum 1982, 367-
381. Columbia, SC: Hornbeam.
Arndt, Horst & Richard W. Janney. 1985. Improving emotive communication:
Verbal, prosodic, and kinesic conflict-avoidance techniq,ues. Per Linguam
1(1). 21-33.
Arnovick, Leslie K. 1999. Diachronic pragmatics: Seven case studies i n
English illocutionary development. Amsterdam & Philadelphia, PA:
John Benjamins.
Auer, Peter. 1996. On the prosody and syntax of tum-continuations. In Elizabeth
Couper-Kuhlen & Margret Selting (eds.), Prosody in conversation, 57-100.
Cambridge: Cambridge University Press.
Austin, John L. 1962. How to do things with words. Oxford: Clarendon Press.
Bach, Kent. 1999. The myth of conventional implicature. Linguistics a nd
Philosophy 22. 327-366.
Bakhtin, Mikhail. 1981. Discourse in the novel. In Michael Holq,uist (ed.), Th
dialogic imagination, 259-422. Austin: University of Texas Press.
Ballim, Afzal, Yorick Wilks & John Barnden. 1991. Belief ascription, metaphor.
and intensional identification. Cognitive Science 15(1). 133-171.
Bally, Charles. 1913. Le langage et la v ie. Geneva: Atar. (Spanish edition: El
lenguaje y la vida. Buenos Aires: Editorial Losada, 1941).
//II/II////I/I//II//////////I//II////////II/I/I//II//I//II References 261

Bar-Hillel, Yehoshua. 1971. Out of the pragmatic wastebasket. Linguistic

Inquiry 2. 401-407.
Bara, Bruno G. 2011. Cognitive pragmatics: The mental processes of
communication. Intercultural Pragmatics 8(3). 443-485.
Barth-Weingarten, Dagmar, Nicole Dehe & Anne Wichmann (eds.). 2009. Where
prosody meets pragmatics. Leiden: Brill.
Bax, Marcel (ed.). 2003. Ritual language behaviour. Special issue of Journal of
Historical Pragmatics 4:2.
Beckman, Mary E. , Julia Hirschberg & Stefanie Shattuck-Hufnagel. 2005. The
original ToBI system and the evolution of the ToBI framework. In Sun-Ah
Jun (ed.), Prosodic typology: The phonology of intonation and phrasing,
9-54. Oxford: Oxford University Press.
Bednarek, Monika. 2006. Evaluation and cognition: Inscribing, evoking and
provoking opinion. In Hanna Pishwa (ed.), Language and memory: Aspects
of knowledge representation, 187-221. Berlin: Mouton de Gruyter.
Bednarek, Monika (ed.). 2008a. Evaluation in tex t types. Special Issue of
Functions of Language 15(1).
Bednarek, Monika. 2008b. Emotion talk across corpora. New York: Palgrave
Macmillan.
Bednarek, Monika. 2009a. Dimensions of evaluation: Cognitive and linguistic
perspectives. Pragmatics and Cognition 17(1). 146-175.
Bednarek, Monika. 2009b. Emotion talk and emotional talk: Cognitive and
discursive perspectives. In Hanna Pishwa (ed.), Language and social
cognition: Expression of the social mind, 395-431. Berlin: Mouton de
Gruyter.
Beun, Robbert-Jan. 1989. The recognition of declarative questions in information
dialogues. PhD Thesis, Tilburg University. Available at <https://pure.uvt.nl/
portal/files/1206907/3955384.pdf >.
Biber, Douglas & Edward Finegan. 1989. Styles of stance in English: Lexical and
grammatical marking of evidentiality and affect. Text 9. 93-124.
Blakemore, Diane. 1987. Semantic constraints on relevance. Oxford: Blackwell.
Blakemore, Diane. 2002. Relevance and linguistic meaning: The semantics and
pragmatics of discourse markers. Cambridge: Cambridge University Press.
Blakemore, Diane. 2009. Communication and the representation of thought: The
use of audience-directed expressions in free indirect thought representations.
Journal of Linguistics 46. 575-599.
Blum-Kulka, Shoshana, Juliane House & Gabriele Kasper (eds.). 1989.
Crosscultural pragmatics: Requests and apologies. Norwood: Ablex.
262 References / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / / /

Boersma, Paul. 2001. Praat: A system for doing phonetics by computer. Glot
International 5(9/10). 341-345.
Bolinger, Dwight. 1989. Intonation and its uses: Melody in grammar and
discourse. London: Edward Arnold.
Boretti, Silvia. 2001. Aspectos de la cortesfa lingtiistica en el espafiol coloq,uial
de la Argentina. Oralia 4. 75-102.
Bosq,ue, Ignacio. 2010. Aspectos individuales y sociales de las emociones: Sobre
la noci6n de "vergtienza" y sus variantes. P<iginas de Guarda 10. 13-27.
Bourdieu, Pierre. 1977. Outline of a theory of practice. Cambridge: Cambridge
University Press.
Bousfield, Derek. 2008. Impoliteness in the struggle for power. In Derek Bousfield
& Miriam A. Locher (eds.), Impoliteness in language: Studies on its
interplay with power in theory and practice, 127-154. Berlin & New York:
Mouton de Gruyter.
Bravo, Diana. 1999. 6Imagen "positiva" vs. imagen "negativa"? Pragmatica
sociocultural y componentes deface. Oralia 2. 155-184.
Brazil, David. 1997. The communicative value of intonation in English (2nd
edition). Cambridge: Cambridge University Press.
Breheny, Richard, Napoleon Katsos & John Williams. 2006. Are scalar implicatur
generated by default? Cognition 100. 434-463.
Breheny, Richard, Header Ferguson & Napoleon Katsos. No date. How q,uantity
implicatures are accessed in incremental utterance interpretation: Evidence
for full-blown Gricean inferencing. Found at http://www.phon.ucl.ac.
home/richardb/project/[ 1]%20Breheny%20Ferguson%20Katsos.pdf (August
27, 2015).
Brinton, Laurel J. 1996. Pragmatic markers in English: Grammaticalization a
discoursefunctions. Berlin: Mouton de Gruyter.
Brown, Gillian, Karen Currie &Joanne Kenworthy. 1980. Questions ofintonatio
Baltimore, MD: University Park Press.
Brown, Gillian & George Yule. 1983. Discourse analysis. Cambridge: Cambrid
University Press.
Brown, Penelope & Stephen C. Levinson. 1987. Politeness: Some universals i
language usage. Cambridge: Cambridge University Press.
Bublitz, Wolfram & Neal R. Norrick. 2011. The burgeoning field of pragmati
In Wolfram Bublitz & Neal R. Norrick (eds.), Foundations of pragmati
1-20. Berlin: De Gruyter Mouton.
Bucciarelli, Monica, Livia Colle & Bruno G. Bara. 2003. How children compreh
speech acts and communicative gestures. Journal ofPragmatics 35. 207- 2
//l!/llll///ll!ll/lll/l/ll/!l/ll!lll////ll!ll//ll/!I/II/I/ Refurences 263

Buhler, Karl. 1934. The deictic field of language and deictic words. Reprinted
in Robert J. Jarvella & Wolfgang Klein (eds.), Speech, place and action:
Studies of deix is and related topics (1982), 9-30. New York: John Wiley.
Bunt, Harry & William Black. 2000. The ABC of computational pragmatics.
In Harry Bunt & William Black (eds.), Abduction, belief and contex t in
dialogue: Studies in computational pragmatics, 1-46. Amsterdam &
Philadelphia, PA: John Benjamins.
Bybee, Joan. 2010. Language, usage and cognition. Cambridge: Cambridge
University Press.
Bybee, Joan & Suzanne Fleischman. 1995. Modality in grammar and discourse.
Amsterdam & Philadelphia, PA: John Benjamins.
Cahn, Janet E. 1989. Generating expression in synthesized speech. Master's thesis,
MIT. Available at www.media.mit.edu/~cahn/theses/masters.pdf.gz.
Canale, Michael & Merryl Swaine. 1980. Theoretical bases of communicative
approaches to second language teaching and testing. Applied Linguistics
1(1). 1-47.
Carston, Robyn. 2002. Thoughts and utterances: The pragmatics of explicit
communication. Malden, MA, Oxford, Carlton & Berlin: Blackwell.
Carston, Robyn. 2007. Relevance Theory and the saying/implicating distinction.
In Laurence R. Horn & Gregory Ward (eds.), The handbook of pragmatics,
633-656. Oxford: Blackwell.
Chomsky, Noam. 1965. Aspects of the theory of synta:x. Cambridge MA: MIT Press.
Clark, Billy. 2013. Relevance Theory. Cambridge: Cambridge University Press.
Comrie, Bernard. 1984. Subject and object control: Syntax, semantics, pragmatics.
Proceedings of the Tenth Annual Meeting of the Berkeley Linguistics Society,
450-464.
Comrie, Bernard. 1985. Tense. Cambridge: Cambridge University Press.
Coulmas, Florian. 1981. Introduction. In Florian Coulmas (ed.), Conversational
routine: Explorations in standardized communication situations and
prepatterned speech, 1-17. The Hague: Mouton.
Craig, Robert, Karen Tracy & Frances Spisak. 1986. The discourse of req,uests:
Assessment of a politeness approach. Human Communication Research
12. 437-468.
Crawford, L. Elizabeth. 2009. Conceptual metaphors of affect. Emotion Review
1. 129-139.
Cruse, Alan D. 2004. Meaning in language: An introduction to semantics and
pragmatics. Oxford: Oxford University Press.
264 References / ///////I I I I I I I I I I I I I I I I I I I I I I I I I II I I I I I I I I I/ I I I I I I I I I I I I I

Cuenca, Maria Josep. 2003. Cognition, pragmatics and grammar. Naves SL.
Revista de SociolingiUstica. Available at http://www.gencat.cat/llengua/
noves/noves/hm03hivern/docs/a_cuenca.pdf
Culpeper, Jonathan. 1996. Towards an anatomy of impoliteness. Journal of
Pragmatics 25. 349-367.
Culpeper, Jonathan. 2005. Impoliteness and entertainment in the television q,uiz
show: 'The Weakest Link'. Journal of Politeness Research 1(1). 35-72.
Culpeper, Jonathan. 2008. Reflections on impoliteness, relational work and
power. In Derek Bousfield and Miriam A. Locher (eds.), Impoliteness in
language: Studies on its interplay w ith power in theory and pract ice
17-44. Berlin & New York: Mouton de Gruyter.
Culpeper, Jonathan. 2010. Historical pragmatics. In Louise Cummings (ed.).
The pragmatics encyclopedia, 188-192. London: Routledge.
Culpeper, Jonathan. 2011. "It's not what you said. It's how you said it". In
Linguistic Politeness Research Group (eds.), Discursive approaches to
politeness, 57-83. Berlin: Mouton de Gruyter.
Culpeper, Jonathan. 2012. (lm)politeness: Three issues. Journal of Pragmatics
44(9). 1128-1133.
Culpeper, Jonathan, Gila Schauer, Leyla Marti, Meilian Mei & Minna Nevala.
2014. Impoliteness and emotions in a cross-cultural perspective. SPELL:
Swiss Papers in English Language and Literature 30. 67-88.
Cummings, Louise. 2005. Pragmatics: A multidisciplinary perspecti e.
Edinburgh: Edinburgh University Press.
Cummins, Chris. 2015. Pragmatic enrichment of modified fractions. Paper
presented at the Experimental Pragmatics Conference. Chicago, July l&-1
2015.
Cummins, Chris, Uli Sauerland & Stephanie Solt. 2012. Granularity and scalar
implicature in numerical expressions. Linguistics and Philosophy 35.
135-169.
Cutler, Anne. 1974. On saying what you mean without meaning what you sag.
In Michael W. La Galy, Robert Fox & Anthony Bruck (eds.), Papers fr
the Tenth Regional Meeting of the Chicago Linguistic Society, 117-12- .
Chicago, IL: CLS.
Dascal, Marcelo. 1983. Pragmatics and the philosophy of mind I: Thought ·
language. Amsterdam & Philadelphia, PA: John Benjamins.
Davidson, Donald. 1967. Truth and meaning. Synthese 17. 304-323.
Deutschmann,Mats.2003.ApologisinginBritishEnglish.Umea:UmeaUniversiteL
//II///////I/I/II////////I////II///I/I/IIII////II/I///I/I/ References 265

Dewaele, Jean Marc. 2013. Emotions in multiple languages. London: Palgrave

Macmillan.
Dik, Simon C. 1978. Functional Grammar. Amsterdam: North-Holland.
Dik, Simon C. 1997. The theory of Functional Grammar. 2 parts. Berlin & New
York: Mouton de Gruyter.
Eelen, Gino. 1999. Ideology in politeness: A critical analysis. PhD thesis,
Universitaire Instelling Antwerpen, Antwerp.
Eelen, Gino. 2001. A critique of politeness theories. Manchester: St. Jerome.
Eggins, Suzanne & Diana Slade. 1997. Analysing casual conversation. London: ·
Cassell.
Ekman, Paul. 2007 (2003). Emotions revealed: Recognizing faces andfeelings
to improve communication and emotional life. New York: St Martin's
Griffin.
Englebretson, Robert. 2007. Stancetaking in discourse. Amsterdam &
Philadelphia, PA: John Benjamins.
Ervas, Francesca & Elisabetta Gola. 2013. A philosophical guide to metaphor
use. In Elisabetta Gola & Franscesca Ervas (eds.), Metaphor in focus:
Philosophical perspectives on metaphor use, 1-8. Newcastle-upon-Tyne:
Cambridge Scholars.
Escandell-Vidal, Victoria. 1998. Intonation and procedural encoding: The case
of Spanish interrogatives. In Villy Rouchota & Andreas H. Jucker (eds.),
Current issues in Relevance Theory, 169-203. Amsterdam & Philadelphia,
PA: John Benjamins.
Escandell-Vidal, Victoria. 2002. Echo-syntax and metarepresentations. Lingua
112. 871-900.
Escandell-Vidal, Victoria. 2009. Social cognition and intercultural
communication. In Victoria Guillen Nieto, Carmen Marim6n Llorca &
Chelo Vargas Sierra (eds.), Intercultural business communication and
simulations and gaming methodology, 65-96. Berne: Peter Lang.
Escandell-Vidal, Victoria & Manuel Leonetti. 2011. On the rigidity of procedural
meaning. In Victoria Escandell-Vidal, Manuel Leonetti & Aoife Ahern (eds.),
Procedural meaning: Problems and perspectives, 81-102. Bingley, UK:
Emerald.
Escandell-Vidal, Victoria, Manuel Leonetti & Aoife Ahern (eds.). 2011. Procedural
meaning: Problems and perspectives (Current Research in the Semantics/
Pragmatics Interface, Vol. 25). Bingley, UK: Emerald.
Escandell-Vidal,Victoria, VictoriaMarreroAguiar&Pilar PerezOc6n.2014.Prosody,
information structure and evaluation. In Geoff Thompson & Laura Alba-Juez
(eds.), Evaluation in context, 153-177. Amsterdam & Philadelphia, PA: John
Benjamins.
266 Refurences l!!l//ll/ll/l!/l///lll/ / l/ll!ll!!lll/ll/llll!/ll!/III/II!!

Fanego, Teresa. 2010. Paths in the development of elaborative discourse

markers: Evidence from Spanish. In Kristin Davidse, Lieven Vandelanotte
& Hubert Cuyckens (eds.), Subjectification, intersubjectification, and
grammaticalization (Topics in English Linguistics 66), 197-237. Berlin:
Mouton de Gruyter.
Farroni, Teresa, Eileen M. Mansfield, Carlo Lai & Mark H. Johnson. 2003. Infants
perceiving and acting on the eyes: Tests of an evolutionary hypothesis.
Journal of Experimental Child Psychology 85. 199-212.
Felices-Lago, Angel. 1997. La compilaci6n de un diccionario de terminos
valorativos desde la perspectiva lexematico-funcional. In Gerd Wotjak (ed.),
Toward afunctional lexicologyl Hacia una lexicologiafuncional, 159-184.
Frankfurt am Main: Peter Lang.
Felices-Lago, Angel. 2003. Reformulation of the domain-level semantic pattern of
axiological evaluation in the lexicon of English verbs. Hermes 30. 179-198.
Fery, Caroline. 2006. Information structural notions and the fallacy of invariant
correlates. In Caroline Fery, Gisbert Fanselow & Manfred Krifka (eds.), The
notions of information structure, 161-184. Potsdam: Universitatsverlag
Potsdam.
Fetzer, Anita. 2004. Recontex tualizing contex t: Grammaticality meets
appropriateness. Amsterdam & Philadelphia, PA: John Benjamins.
Fillmore, Charles. 1968. The case for case. In Emmon Bach & Robert T. Harms
(eds.), Universals in linguistic theory, 1-90. New York: Holt, Rinehart and
Winston.
Fillmore, Charles. 1996. Frame semantics and the nature of language. In Stevan
R. Harnad, Horst D. Steklis & Jane Lancaster (eds.), Origins and evolutio n
of language and speech, 20-32. Annals of the New York Academy of
Sciences 280.
Fillmore, Charles. 1997a. Lectures on deix is. Stanford, CA: CSLI Publications.
Fillmore, Charles. 1997b. Scenes-and-frames semantics. In A. Zampolli (ed.)
Linguistic structures processing (Fundamental studies in computer science
59), 55-81. Amsterdam: North Holland.
Fitzmaurice, Susan & Irma Taavitsainen. 2007. Methods in historical pragmatics.
Berlin: Mouton de Gruyter.
Fodor, Jerry A. 1975. The language of thought. New York: Crowell.
Fodor, Jerry A. 1998. Concepts: Where cognitive science went wrong. Oxford:
Oxford University Press.
Foolen, Ad. 2012. The relevance of emotion for language and linguistics. In
Ad Foolen, Ulrike M. Ludtke, Timothy P. Racine & Jordan Zlatev (eds.)
//l//lll///l/l//l///lll!!l/ll///l///ll///l!/I/II/I//I///// Refurences 267

Moving ourselves, moving others: Motion and emotion in intersubjectivity,

consciousness and language, 347-368. Amsterdam & Philadelphia, PA: John
Benjamins.
Foolen, Ad. 2016. Expressives. In Nick Riemer (ed.), Handbook of semantics,
473-490. London & New York: Routledge.
Forceville, Charles & Billy Clark. 2014. Can pictures have explicatures?
Linguagem em (Dis)curso 14(3). 451-472.
Frank, Michael & Noah Goodman. 2012. Predicting pragmatic reasoning in
language games. Science 336. 998.
Fraser, Bruce. 1990. Perspectives on politeness. Journal of Pragmatics 14.
219-236.
Fraser, Bruce. 2006a. On the conceptual-procedural distinction. Style, Spring-
Summer 2006.
Fraser, Bruce. 2006b. Towards a theory of discourse markers. In Kerstin Fischer
(ed.), Approaches to discourse particles, 189-204. Bremen: Elsevier.
Fraser, Bruce & William Nolen. 1981. The association of deference with linguistic
form. International Journal of the Sociology of Language 27. 93-109.
Fretheim, Thorstein. 1998. Intonation and the procedural encoding of attributed
thoughts: The case of Norwegian negative interrogatives. In Villy Rouchota
& Andreas H. Jucker (eds.), Current issues in Relevance Theory, 205-236.
Amsterdam & Philadelphia, PA: John Benjamins.
Geis, Michael. 1995. Speech acts and conversational interaction. Cambridge:
Cambridge University Press.
Gibbs, Raymond & Herbert L. Colston. 2012. Interpreting figurative meaning.
Cambridge: Cambridge University Press.
Giv6n, Talmy. 2009. The adaptive approach to grammar. In Bernd Heine & Heiko
Narrog (eds.), The Oxford handbook of linguistic analysis, 27-49. Oxford:
Oxford University Press.
Goffman, Ervin. 1967. On face work. In Ervin Goffman (ed.), Interaction ritual,
5-46. New York: Anchor Books.
Goffman, Ervin. 1972. Interaction ritual: Essays on face-to-face behavior.
Harmondsworth: Penguin.
Goffman, Ervin. 1974. Frame analysis. New York: Harper and Row.
Goffman,Ervin.1981. Footing. In Ervin Goffman,Formsoftalk, 124-159.Philadelphia,
PA: University of Pennsylvania Press.
Goleman, Daniel. 1995. Emotional Intelligence: Why it can matter more than
I. Q. London: Bloomsbury.
268 References //!l////l/ll//l/ll/////l!/l/!l!ill//////l/l/////l//ll/ll//

Granacher, Robert P., Jr. 2008. Traumatic brain injury: Methods for clinical and
forensic neuropsychiatric assessment. 2nd edition. Boca Raton, London &
New York: CRC Press.
Grawunder, Sven & Bodo Winter. 2010. Acoustic correlates of politeness: Prosodic
and voice q,uality measures in polite and informal speech of Korean and
German speakers. Proceedings of the International Conference on Speech
Prosody 2010. 100316: 1-4. Chicago.
Green, Georgia. 1989. Pragmatics and natural language understanding.
Hillsdale, NJ: Lawrence Erlbaum.
Grice, H. Paul. 1967. Logic and conversation. Further notes on logic and
conversation. The William James lectures. Published as Part 1 of Grice (1989).
Grice, H. Paul. 1975. Logic and conversation. In Peter Cole & Jerry Morgan (eds.),
Synta.x and semantics 3: Speech acts, 41-58. New York: Academic Press.
Grice, H. Paul. 1989. Studies in the way of words. Cambridge, MA: Harvard
University Press.
Grossmann, Tobias, Mark H. Johnson, Sarah Lloyd-Fox, Anna Blasi, Fanni
Deligianni, Clare Elwell & Gergely Csibra. 2008. Early cortical specialization
for face-to-face communication in human infants. Proceedings of the Royal
Society, London B, 275. 2803-2811.
Grundy, Peter. 2008. Doing pragmatics. London: Hodder Education.
Gu, Yueguo. 1990. Politeness phenomena in modern Chinese. Journal of
Pragmatics 3. 237-257.
Gu, Yueguo. 1997. Five ways of handling a bedpan: A tripartite approach to
workplace discourse. Text 17. 457-475.
Guasti, Maria Teresa, Gennaro Chierchia, Stephen Crain, Francesca Foppolo
Andrea Gualmini & Luisa Meroni. 2005. Why children and adults sometimes
(but not always) compute implicatures. Language and Cognitive Processes
20(5). 667-696.
Halliday, M.A.K. 1967. Intonation and grammar in British English. The Hague:
Mouton.
Halliday, M.A.K. 1978. Language as social semiotic: The social interpretation of
language and meaning. London: Edward Arnold.
Halliday, M.A.K. 1994. An introduction to functional grammar. 2nd edition.
London: Edward Arnold.
Halliday, M.A.K. & Christian M.I.M. Matthiessen. 2004. An introduction to
Functional Grammar (Wd edition). London: Edward Arnold.
Hannay, Mike & Kees Hengeveld. 2009. Pragmatics in Functional Discourse
Grammar: Pragmatic aspects. In Frank Brisard, Jan-Ola Ostman & Jef
////////////////////l//ll///l//l//ll///ll/!I///II/////I/II Refurences 269

Verschueren (eds.), Grammar, meaning and pragmatics, 92-116. Amsterdam

and Philadelphia, PA: John Benjamins.
Harris, Randy Allen. 1993. The linguistics wars. New York & Oxford: Oxford
University Press.
Haugh, Michael & Derek Bousfield. 2012. Mock impoliteness, jocular mockery
and jocular abuse in Australian and British English. Journal of Pragmatics
44. 1099-1114.
Heine, Bernd & Kyung-An Song. 2011. On the grammaticalization of personal
pronouns. Journal of Linguistics 47. 587-630.
Hengeveld, Kees & J. Lachlan Mackenzie. 2008. Functional Discourse Grammar:
A typologically-based theory of language structure. Oxford: Oxford
University Press.
Hengeveld, Kees, Eli Nazareth Bechara, Roberto Gomes Camacho, Alessandra
Regina Guerra, Tafsa Peres de Oliveira, Eduardo Penhavel, Erotilde Goreti
Pezatti, Liliane Santana, Edson Rosa Francisco de Souza & Maria Luiza de
Sousa Teixeira. 2007. Basic illocutions in the native languages of Brazil. Alfa
- Revista de LingiUstica 51(2). 73-90.
Hernandez Flores, Nieves. 1999. Politeness ideology in Spanish colloq,uial
conversation: The case of advice. Pragmatics 9(1). 137-149.
Herring, Susan C. 2001. Computer-mediated discourse. In Deborah Schiffrin,
Deborah Tannen & Heidi Hamilton (eds.), Handbook of discourse analysis,
612-634. Malden, MA & Oxford: Blackwell.
Herring, Susan C. (ed.). 2003. Media and language change. Special issue of
Journal of Historical Pragmatics 4(1).
Hinchcliffe, Doreen. 2000. Divorce in the Muslim world. International Family
Law, May 2000. 63-65.
Hirschberg, Julia. 2007. Pragmatics and intonation. In Laurence R. Horn &
Gregory Ward (eds.), The handbook of pragmatics, 515-537. Oxford:
Blackwell.
Hirschberg, Julia & Christine Nakatani. 1996. A prosodic analysis of discourse
segments in direction-giving monologues. Association for Computational
Linguistics 34. 286-293. Santa Cruz, CA.
Holmes, Janet & Stephanie Schnurr. 2005. Politeness, humor and gender in the
workplace: Negotiating norms and identifying contestation. Journal of
Politeness Research 1(1). 121-149.
Horn, Laurence R. 1972. On the semantic properties of logical operators in
English. Mimeo, Indiana University Linguistics Club. Bloomington, IN.
 270 References //////II// I I I I I/ I I I I I/ I I// I I I// I I///// I I ///l I l ///II/ I I// I/

Horn, Laurence R. 1984. Toward a new taxonomy for pragmatic inference:

Q-based and R-based implicature. In Deborah Schiffrin (ed.), Meaning,
form, and use in context: Linguistic applications, 11-42. Washington, DC:
Georgetown University Press.
Hom, Laurence R. 1991. Given as new: When redundant affirmation isn't. Journal
ofPragmatics 15. 305-328.
Hom, Laurence R. 2007. Implicature: Some basic oppositions. In Laurence R. Hom
& Gregory Ward (eds.), The handbook of pragmatics. Oxford: Blackwell.
Huang, Yan. 2006. Speech acts. In Keith Brown (ed.), Encyclopedia of language
and linguistics, 2nd edition. 8593-8602.
Huang, Yan. 2009. Neo-Gricean pragmatics and the lexicon. International Review
ofPragmatics 1. 118-153.
Huang, Yan. 2013. Micro- and macro-pragmatics: Remapping their terrains.
International Review of Pragmatics 5. 129-162.
Huang, Yan. 2016. Pragmatics: Language use in context. In Keith Allan (ed.),
Routledge handbook oflinguistics, 205-220. London & New York: Routledge.
Hunston, Susan. 2008. The evaluation of status in multi-modal texts. Functions
of Language 15(1). 64-83.
Hunston, Susan & Geoff Thompson (eds.). 2000. Evaluation in text: Authorial
stance and the construction of discourse. Oxford: Oxford University Press.
Hymes, Dell. 1964. Language in culture and society. New York: Harper & Row.
Ide, Sachiko. 1982. Japanese sociolinguistics: Politeness and women's language.
Lingua 57. 49-89.
Ide, Sachiko. 1989. Formal forms and discernment: Two neglected aspects of
universals of linguistic politeness. Multilingua 8(2-3). 223-248.
Ide, Sachiko. 1993. Preface: The search for integrated universals of linguistic
politeness. Multilingua 12(1). 7-11.
Jackendoff, Ray. 1972. Semantic interpretation in generative grammar.
Cambridge MA: MIT Press.
Jacobs, Andreas & Andreas H. Jucker. 1995. The historical perspective in
pragmatics. In Andreas H. Jucker (ed.), Historical pragmatics: Pragmatic
developments in the history of English. Amsterdam & Philadelphia, PA
John Benjamins.
James, William 1884. What is emotion? Mind 9(34). 188-205.
Janney, Richard W. & Horst Arndt. 1992. Intracultural tact versus intercultural
tact. In Richard J. Watts, Sachiko Ide & Konrad Ehlich (eds.), Politeness in
language: Studies in its history, theory and practice, 21-41. Berlin: Mouton
de Gruyter.
////////////////////////////////////////////////////////// References 271

Jespersen, Otto. 1965. A Modern English grammar. Part IV. London: Allen &
Unwin.
Jucker, Andreas H. (ed.). 1995. Historical pragmatics: Pragmatic developments
in the history of English. Amsterdam & Philadelphia, PA: John Benjamins.
Jucker, Andreas H. 2006. Historical pragmatics. In Keith Brown (ed.), Encydopedia
of language and linguistics, 329-332. 2nd edition. Oxford: Elsevier.
Jurafsky, Daniel. 2007. Pragmatics and computational linguistics. In Laurence
R. Horn & Gregory Ward (eds.), The handbook of pragmatics. Oxford:
Blackwell.
Karttunen, Lauri & Stanley Peters. 1975. Conventional implicature in Montague
grammar. Proceedings of the First Annual Meeting of the Berkeley
Linguistic Society, 266-278.
Karttunen, Lauri & Stanley Peters. 1979. Conventional implicature. In Choon-Kyu
Oh & David A. Dinneen (eds.), Syntax and semantics II: Presupposition,
1-56. New York: Academic Press.
Kaul de Marlangeon, Silvia. (1993) 1995, 2003. La fuerza de cortesia-descortesia
y sus estrategias en el discurso tanguero de la decada del '20. Revista de la
Sociedad Argentina de LingiUstica 3. 7-38. Available at http://www.edice.
org/Documentos/SKaul. pdf.
Kaul de Marlangeon, Silvia. (1993) 1999. Uso estrategico de la ironia. Anales del
Instituto de Linguistica, Vols. XVIII, XIX, XX y XXI, 141-150. Universidad
Nacional de Cuyo, Mendoza.
Kaul de Marlangeon, Silvia. 2005a. Descortesia de fustigaci6n por afiliaci6n
exacerbada o refractariedad. In Diana Bravo (ed.), Estudios de la (des)
cortesia en espaiiol: Categorias conceptuales y aplicaciones a corpora
orales y escritos, 299-318. Buenos Aires: Dunken.
Kaul de Marlangeon, Silvia. 2005b. Descortesia intragrupal-cr6nica en la
interacci6n coloq,uial de clase media baja del espafi.ol rioplatense. L6dz
Papers in Pragmatics 2005/1. 121-138.
Kaul de Marlangeon, Silvia. 2008a. Tipologia del comportamiento verbal
descortes en espafi.ol. Actas del III Coloq,uio del Programa EDICE: Cortesia
y conversaci6n: de lo escrito a lo oral, 254-266. Valencia/Stockholm:
Departamento de Filologia Espafi.ola, Universidad de Valencia.
Kaul de Marlangeon, Silvia. 2008b. Impoliteness in institutional and non-
institutional contexts. Pragmatics 18(4). 729-749.
Kaul de Marlangeon, Silvia. 2010. Perspectiva topol6gica de la descortesia verbal.
In Franca Orletti & Laura Mariottini (eds.), La descortesia en espaiiol:
Ambitos te6ricos y metodol6gicos de estudio, 71-86. Roma: Universita degli
Studi Roma Tre.
272 References /II I I I I I I I I I I I I I I I// I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I/ I I I

Kaul de Marlangeon, Silvia. 2014. Delimitaci6n de unidades extralingiiisticas de

analisis del discurso de (des)cortesia. Signo y Sena: Revista del Instituto
de Lingi.Ustica 26. 7-22.
Kaul de Marlangeon, Silvia & Laura Alba-Juez. 2012. A typology of verbal
impoliteness behaviour for the English and Spanish cultures. Revista
Espanola de Linguistica Aplicada 25. 69-92.
Keating, Elizabeth. 1998. Honor and stratification in Pohnpei, Micronesia.
American Ethnologist 25. 399-411.
Kerbrat-Orecchioni, Catherine. 2011. From good manners to facework: Politeness
variations and constants in France, from the classic age to today. Journal
of Historical Pragmatics 12. 133-155.
Klima, Edward S. 1964. Negation in English. In Jerry A. Fodor & Jerrold J.
Katz (eds.), The structure of language, 246-323. Engelwood Cliffs, NJ:
Prentice-Hall.
Koutlaki, Sofia A. 2002. Offers and expressions of thanks as face enhancing acts:
tre'arof in Persian. Journal of Pragmatics 34(12). 1733-1756.
Kramsch, Claire. 1993. Contex t and culture in language teaching. Oxford:
Oxford University Press.
Krzeszowski, Tomasz P. 1990. The axiological aspect of idealized cognitive models.
In Jerzy Tomaszczyk & Barbara Lewandowska-Tomaszczyk (eds.), Meaning
and lexicography, 135-165. Amsterdam & Philadelphia, PA: John Benjarnins.
Labov, William. 1972. The transformation of experience in narrative syntax.
In William Labov, Language in the inner city, 354-405. Philadelphia, Pk
University of Pennsylvania Press.
Labov, William. 1997a. Narrative analysis: Oral versions of personal experience.
Journal of Narrative and Life History 7(1-4). 3-38.
Labov, William. 1997b. Some further steps in narrative analysis. Journal of
Narrative and Life History 7(1-4). 395-415.
Labov, William & Joshua Waletzky. 1967. Narrative analysis. In June Helm (ed.)
Essays on the verbal and visual arts, 12-44. Seattle, WA: University of
Washington Press.
Ladd, D. Robert. 1996. Intonational phonology. Cambridge: Cambridge
University Press.
Ladd, D. Robert, Kim E.A. Silverman, Frank J. Tolkmitt, Guenther Bergmann &
Klaus R. Scherer. 1985. Evidence for the independent function of intonation
contour type, voice q,uality, and f0 range in signaling speaker affect.
Journal of the Acoustical Society of America 78. 435-444.
//I//II////I/I/I/I/I//II////I/II//I/I/IIII//I//II//I/I//// References 273

Lakoff, George. 1971. On generative semantics. In Danny D. Steinberg & Leon A.

Jakobovits (eds.), Semantics: An interdisciplinary reader in philosophy,
linguistics and psychology, 232-296. Cambridge: Cambridge University
Press.
Lakoff, George. 1987. Women, fire, and dangerous things. Chicago: University
of Chicago Press.
Lakoff, Robin. 1973. The logic of politeness: Or, minding your p's and (lS. In
Claudia Corum, T. Cedric Smith-Stark & Ann Weiser (eds.), Papers from
the Ninth Regional Meeting of the Chicago Linguistic Society, 292-305.
Chicago: Chicago Linguistic Society.
Lakoff, Robin. 1989. The limits of politeness. Multilingua 8. 101-129.
Langacker, Ronald W. 1987. Foundations of Cognitive Grammar. Vol. I: Theoretical
prerequisites. Stanford, CA: Stanford University Press.
Langacker, Ronald W. 2008. Cognitive Grammar: A basic introduction. New
York: Oxford University Press.
Langacker, Ronald W. 2012. Interactive cognition: Toward a unified account of
structures, processing, and discourse. International Journal of Cognitive
Linguistics 3. 95-124.
Lavandera, Beatriz. 1988. The social pragmatics of politeness forms . In Ulrich
Ammon & Norbert Dittmar (eds.), Sociolinguistics: An international
handbook of the science of language and society, Vol. 2, 1196-1204. Berlin
& New York: Walter de Gruyter.
Lee, Mark & Yorick Wilks. 1996. An ascription-based approach to speech acts.
In Proceedings of 1996 International Conference on Computational
Linguistics COLING-96. Copenhagen.
Lee Wong, Song-Mei. 1999. Politeness and face in Chinese culture. Frankfurt:
Peter Lang.
Leech, Geoffrey N. 1977. Language and tact. Trier: Linguistic Agency University of
Trier, Series A, 46. Reprinted with revisions in Geoffrey N. Leech,Explorations
in semantics and pragmatics (1980), 79-117. Amsterdam & Philadelphia,
PA: John Benjamins.
Leech, Geoffrey N. 1983. Principles of pragmatics. London: Longman.
Leech, Geoffrey N. 2005. Politeness: Is there an East-West divide? Wai Guo Yu:
Journal of Foreign Languages (Shanghai) 6. 3-31.
Leech, Geoffrey N. 2007. Politeness: Is there an East-West divide? Journal of
Politeness Research 3(2). 167-206. (Revised version of Leech 2005).
Leech, Geoffrey N. 2014. The pragmatics ofpoliteness. Oxford: Oxford University
Press.
274 References I I I I I I I I I I I I I I I I I I I I I I I I I I I I! I I I I I I I I I I I I I I I I I I I I I I I I I I I I I

Lehiste, Ilse. 1979. Perception of sentence and paragraph boundaries. In Bjorn

Lindblom & Sven E.G. Ohman (eds.), Frontiers of speech research, 191-201.
London: Academic Press.
Levinson, Stephen C. 1983. Pragmatics. Cambridge: Cambridge University Press.
Levinson, Stephen C. 1995. Three levels of meaning. In Frank Palmer (ed.),
Grammar and Meaning - Essays in Honour of Sir John Lyons, 90-119.
Cambridge: Cambridge University Press.
Levinson, Stephen C. 2000. Presumptive meanings: The theory of generalized
conversational implicature. Cambridge, MA & London: MIT Press.
Levinson, Stephen C. 2007. Deixis. In Laurence R. Horn & Gregory Ward (eds.),
The handbook of pragmatics, 97-121. Oxford: Blackwell.
Liebal, Kristin, Tanya Behne, Malinda Carpenter & Michael Tomasello. 2009.
Infants use shared experience to interpret pointing gestures. Developmental
Science 12(2). 264-271.
Locher, Miriam A. 2004. Power and politeness in action: Disagreements in oral
communication. Berlin: Mouton de Gruyter.
Locher, Miriam A. 2006. Polite behaviour within relational work: The discursive
approach to politeness. Multilingua 25 (3), 249-267.
Locher, Miriam A. & Derek Bousfield. 2008. Introduction: Impoliteness and power
in language. In Derek Bousfield & Miriam A. Locher (eds.), Impoliteness
in language: Studies on its interplay with power in theory and practice,
1-13. Berlin & New York: Mouton de Gruyter.
Locher, Miriam A. & Richard J. Watts. 2005. Politeness theory and relational
work. Journal of Politeness Research 1(1). 9-33.
Luce, Louise Fiber & Elice C. Smith. 1987. Cross-cultural literacy: A
national priority. In Louise Fiber Luce & Elice C. Smith (eds.), Toward
internationalism, 3-10. Cambridge: Newbury House.
Lunde, Ingunn. 2004. Rhetorical enargeia and linguistic pragmatics: On speech-
reporting strategies in East Slavic medieval hagiography and homiletics.
Journal of Historical Pragmatics 5(1). 49-80.
Macken-Horarik, Mary & James R. Martin (eds.) 2003. Negotiating heteroglossia:
Social perspectives on evaluation. Special Issue of Text: An Interdisciplinary
Journal of Language, Discourse & Communication Studies 23(2).
Mackenzie, J. Lachlan. 2016. Functional linguistics. In Keith Allan (ed.), Routledge
handbook of linguistics, 470-484. London: Routledge.
Mao, LuMing Robert. 1994. Beyond politeness theory: "Face" revisited and
renewed. Journal of Pragmatics 21. 451-486.
////////////////////////////////////////////////////////// References 275

Martin, James R. 2000. Beyond exchange: Appraisal systems in English. In Susan

Hunston & Geoff Thompson (eds.), Evaluation in tex t: Authorial stance
and the construction of discourse, 142-175. Oxford: Oxford University Press.
Martin, James R. & David Rose. 2003. Working with discourse: Meaning beyond
the clause. London: Continuum.
Martin, James R. & Peter R.R. White. 2005. The Language ofevaluation: Appraisal
in English. New York: Palgrave Macmillan.
Matsumoto, Yoshiko. 1988. Reexamination of the universality of face: Politeness
phenomena in Japanese. Journal of Pragmatics 12. 403-426.
Matsumoto, Yoshiko. 1989. Politeness and conversational universals:
Observations from Japanese. Multilingua 8. 207-221.
McDonald, Skye & Samantha Pearce. 1996. Clinical insights into pragmatic
theory: Frontal lobe deficits and sarcasm. Brain and Language 53. 81-104.
Mey, Jacob L. (1993) 2008. Pragmatics: An introduction. 2nd edition. Oxford:
Blackwell.
Minsky, Marvin. 1975. A framework for representing knowledge. In Patrick H.
Winston (ed.), The psychology of computer vision. New York: McGraw Hill.
211-277.
Mozziconacci, Silvie J.L. 1998. Speech variability and emotion: Production and
perception. PhD thesis, University of Leiden.
Morgan, Jerry L. 1978. Two types of convention in indirect speech acts. In Peter
Cole (ed.), Syntax and Semantics 9: Pragmatics.New York: Academic Press.
261-280.
Morris, Charles. 1938. Foundations of the theory of signs. In Otto Neurath,
Rudolph Carnap & Charles Morris (eds.), International encyclopedia of
unified science, 77-138. Chicago, IL: University of Chicago Press.
Murray, Denise E. 1990. CmC. English Today 23. 42-46.
Noveck, Ira A. & Dan Sperber (eds.). 2004. Experimental pragmatics. London &
New York: Palgrave Macmillan.
Nwoye, Onuigbo G. 1992. Linguistic politeness and sociocultural variations of
the notion of face. Journal of Pragmatics 25. 1-32.
O'Connor, Joseph D. & Gordon F. Arnold. 1961. Intonation of colloquial English.
London: Longman.
Olbertz, Rella. 1998. Verbal periphrases in a Functional Grammar of Spanish.
Berlin: Mouton de Gruyter.
Olmos, Susana & Aoife Ahern. 2009. Contrast and propositional attitude: A
relevance theoretic analysis of contrast connectives in Spanish and English.
Lingua 119(1). 51-66.
276 References /////////////////////// / //////////// / //////////// // ///////

Osgood, Charles E. 1980. Lectures on language performance. New York: Springer.

Osgood, CharlesE., GeorgeJ. Suci,&Percy H. Tannenbaum.1957. Themeasurement
of meaning. Urbana, IL: University of Illinois Press.
Penman, Robyn. 1990. Facework and politeness: Multiple goals in courtroom
discourse. Journal of Language and Social Psychology 9. 15-38.
Perrault, Raymond. 1989. An application of default logic to speech act theory. In
Philip R. Cohen, Jerry Morgan & Martha E. Pollack (eds.), Intentions and
Communication, 161-186. Cambridge, MA: MIT Press.
Perrault, Raymond, Philip R. Cohen & James Allen. 1980. A plan-based analysis
of indirect speech acts. Computational Linguistics 6(3-4). 167-182.
Potts, Christopher. 2005. The logic of conventional implicatures. Oxford: Oxford
University Press.
Potts, Christopher. 2007. The expressive dimension. Theoretical Linguistics 33.
165-198.
Price, Michael (2014). Profiles in evolutionary moral psychology: Richard Joyce.
Available at https://evolution-institute.org/article/profiles-in-evolutionary-
moral-psychology-richard-joy ce/?source=.
Prieto, Pilar & Gemma Rigau.2012. Pragmatics and prosody. InJosep Lluis Payrat6
& Maria Cots (eds.), The pragmatics of Catalan, 17-47. Berlin: Mouton de
Gruyter.
Quirk, Randolph, Sidney Geenbaum, Geoffrey N. Leech & Jan Svartvik. 1985. A
comprehensive grammar of the English Language. London: Longman.
Rajagopalan, Kanovillil. 2000. On Searle [on Austin] on language. Language &
Communication 20. 347-391.
Reiter, Raymond. 1980. A logic for default reasoning. Artificial Intelligence 13.
81-132.
Rizzi, Luigi. 2013. Notes on cartography and further explanation. Probus 25.
197-226.
Romaine, Suzanne. 1982. Socio-Historical linguistics: Its status and m ethodology.
New York: Cambridge University Press.
Rosch, Eleanor. 1978. Principles of categorization. In Eleanor Rosch & Barbara B.
Lloyd (eds.), Cognition and categorization, 27-48. Hillsdale, NJ: Lawrence
Erlbaum.
Ross, John R. 1970. On declarative sentences. In Roderick A. Jacobs & Peter
S. Rosenbaum (eds.), Readings in English Transformational Grammar.
222-272. Waltham, MA: Ginn & Co.
Russell, James A. 1980. A circumplex model of affect. Journal ofPersonality and
Social Psychology 39(6). 1161-1178.
lllllll///l/l!l///l/l/ll///l!/ll//l/!I//II//I//I/II////I// Refurences 277

Sacks, Harvey, Emanuel A. Schegloff & Gail Jefferson. 1974. A simplest

systematics for the organization of turn-takingfor conversation. Language
50. 696-735.
Sadock, Jerrold. 1978. On testing for conversational implicature. In Peter Cole
(ed.),SyntaxandSemantics:Vol.9,Pragmatics,281-297.NewYork:Academic
Press.
Sadock, Jerrold. 2007. Speech acts. In Lawrence R. Horn & Gregory Ward (eds.),
The handbook of pragmatics. Oxford: Blackwell.
Schank, Roger C. & Robert P. Abelson. 1977. Scripts, goals, plans and
understanding: An inquiry into human knowledge structures. Hillsdale,
NJ: Erlbaum.
Schmidt, Richard. 1980. Review of Esther Goody (ed.), Questions of politeness:
Strategies in social interaction.Regional English Language Centre Journal
11. 100-114.
Schnoebelen, Tyler. 2012. Emotions are relational: Positioning and the use of
affective linguistic resources. PhD Dissertation, Stanford University.
Scollon, Ron, Suzanne Wong Scollon & Rodney H. Jones. 2012. Intercultural
communication: A discourse approach. Oxford: Wiley-Blackwell.
Scott-Phillips, Thom C. 2010. The evolution of relevance. Cognitive Science 34.
583-601.
Searle, John R. 1969. Speech acts. Cambridge: Cambridge University Press.
Searle, John R. 1975a. A taxonomy of speech acts. In Keith Gunderson (ed.),
Minnesota Studies in the Philosophy of Science. Vol. 9: Language, mind
and knowledge, 344-369. Minneapolis, MN: University of Minnesota Press.
Reprinted in Searle (1979), 1-29.
Searle, John R. 1975b. Indirect speech acts. In Peter Cole & Jerry L. Morgan (eds.),
Syntax and semantics 3: Speech acts, 59-82. New York: Academic Press.
Searle,JohnR.1979. Expressionandmeaning. Cambridge: Cambridge University
Press.
Selting, Margret. 1996. On the interplay of syntax and prosody in the constitution
of turn-constructional units and turns in conversation. Pragmatics 6. 371-389.
Sotillo, Susana M. 2000. Discourse functions and syntactic complexity in
synchronous and asynchronous communication. Language Learning &
Technology 4(1). 82-119.
Southgate, Victoria, Coralie Chevallier & Gergely Csibra. 2009. Sensitivity to
communicative relevance tells young children what to imitate.Developmental
Science 12(6). 1013-1019.
278 References I I I/!/ I I I I I I I I I I I I I I I I I! I I/ I! I I I/ I I I I I I I I I I I I I I I I I I I I I I I I I

Spencer-Oatey, Helen. 1993. Conceptions of social relations and pragmatics

research. Journal of Pragmatics 20. 27-47.
Spencer-Oatey, Helen. 2000. Rapport management: A framework for analysis. In
Helen Spencer-Oatey (ed.), Culturally speaking: Managing rapport through
talk across cultures, 11-46. London & New York: Continuum.
Spencer-Oatey, Helen (ed.). 2008. Culturally speaking: Culture, communication
and politeness theory. London & New York: Continuum. (Revised edition of
Spencer-Oatey 2000).
Sperber, Dan & Ira A. Noveck. 2004. Introduction. In Ira A. Noveck & Dan Sperber
(eds.), Experimental pragmatics, 1-22. London & New York: Palgrave
Macmillan.
Sperber, Dan & Deirdre Wilson. 1986. Relevance: Communication and cognition.
Oxford: Blackwell.
Sperber, Dan & Deirdre Wilson. 1995. Preface to the second edition of Relevance:
Communication and cognition. Oxford: Blackwell.
Sperber, Dan & Deirdre Wilson. 1998. The mapping between the mental and the
publiclexicon. InPeterCarruthers&JillBoucher(eds.) ,Languageandthought:
Interdisciplinary themes, 184-200. Cambridge: Cambridge University Press.
Sperber, Dan & Deirdre Wilson. 2002. Pragmatics, modularity and mind-reading.
Mind & Language 17: 3-23.
Stalnaker, Robert C. 1977. Pragmatic presuppositions. In John P. Murphy, Robert
Eugene Wall & Andy Rogers (eds.), Proceedings of the Texas Conference on
Performatives, Presuppositions and Implicatures, 135-147. Washington
DC: Center for Applied Linguistics.
Strawson, Peter F. 1950. On referring. Mind 59. 320-344.
Strawson, Peter F. 1971. Logico-Linguistic Papers. London: Methuen.
Swerts, Marc. 1997. Prosodic features at discourse boundaries of different strength.
Journal of the Acourstical Society of America 22. 25-41.
Sybesma, Rint & Boya Li. 2007. The dissection and structural mapping of Cantonese
sentence final particles. Lingua 117(10). 1739-1783.
Taavitsainen, Irma & Susan Fitzmaurice. 2007. Historical pragmatics: What it is
and how to do it. In Susan Fitzmaurice & Irma Taavitsainen (eds.), Methods
in historical pragmatics, 11-36. Berlin: Mouton de Gruyter.
Tannen, Deborah. 1993. Introduction. In Deborah Tannen (ed.), Framing i n
discourse, 3-13. New York: Oxford University Press.
Terkourafi, Marina. 1999. Frames for politeness: A case study. Pragmatics 9(1).
97-117.
/ll//!/I///I/I///II///I/////I/II//////I/II////II/I/////I// References 279

Terkourafi, Marina. 200 la. The distinction between generalised and particularised
implicatures and linguistic politeness. In Peter Kuhnlein, Hannes Rieser & Henk
Zeevat (eds.), Proceedings of the Fifth Workshop on the Formal Semantics
and Pragmatics of Dialogue. Bielefeld: ZiF. 174-188.
Terkourafi, Marina. 2001 b. Politeness in Cypriot Greek: A frame-based approach.
Unpublished PhD Dissertation, University of Cambridge.
Terkourafi, Marina. 2005. An argument for a frame-based approach to politeness:
Evidence from the use of the imperative in Cypriot Greek. In Robin Lakoff &
Sachiko Ide (eds.),Broadening the horizon of linguistic politeness, 99-116.
Amsterdam & Philadelphia, PA: John Benjamins.
Terkourafi, Marina. 2008. Toward a unified theory of politeness, impoliteness,
and rudeness. In Derek Bousfield & Miriam A. Locher (eds.), Impoliteness
in language: Studies on its interplay with power in theory and practice,
45-76. Berlin & New York: Mouton de Gruyter.
Terkourafi, Marina. 2016. The linguistics of politeness and social relations. In
Keith Allan (ed.), The Routledge handbook of linguistics, 221-235. Oxford:
Routledge.
Thomas, Jenny. 1995. Meaning in interaction: An introduction to pragmatics.
London: Longman.
Thompson, Geoff. 2014. AFFECT and emotion, target-value mismatches, and
Russian dolls: Refining the APPRAISAL model. In Geoff Thompson & Laura
Alba-Juez (eds.), Evaluation in context, 47-66. Amsterdam & Philadelphia,
PA: John Benjamins.
Thompson, Geoff & LauraAlba-Juez (eds.) 2014. Evaluation in context. Amsterdam
& Philadelphia, PA: John Benjamins.
Tomasello, Michael. 2008. Origins of human communication. Cambridge, MA:
MIT Press.
Tracy, Karen. 1990. The many faces of facework. In Howard Giles & W. Peter
Robinson (eds.), Handbook of language and social psychology, 209-226.
Chichester: Wiley.
Traugott, Elizabeth C. 2007. Historical pragmatics. In Laurence R. Horn & Gregory
Ward (eds.), The handbook of pragmatics, 538-561. London: Blackwell.
Traugott, Elizabeth C. & Richard B. Dasher. 2002. Regularity in semantic change.
Cambridge: Cambridge University Press.
Ungerer,Friedrich(ed.).2000.Englishmediatexts-Pastandpresent.Amsterdam
& Philadelphia, PA: John Benjamins.
Urmson, James 0. 1977. Performative utterances. Midwest Studies in Philosophy
2. 120-127.
280 References / ///////I I I I I I I I I I I I I I I I I I I/// I I I I I l I I I I/ I I I I I I I I /II/ I I I I I

Van Donzel, Moniq,ue E. 1999. Prosodic aspects of information structure in

discourse. PhD Thesis, University of Amsterdam.
Verschueren, Jef. 1979. What people say they do with words. Berkeley, CA:
University of California. PhD dissertation.
Verschueren, Jef. 2009. Introduction: The pragmatic perspective. In Jef
Verschueren & Jan-Ola bstman (eds.), Key notions for pragmatics, 1-27.
Amsterdam & Philadelphia, PA: John Benjamins.
Ward, Geoffrey. 2004. Eq,uatives and deferred reference. Language 80. 262-289.
Warnock, Geoffrey J. 1973. Some types of performative utterance, 69-89. In Isaiah
Berlin (ed.), Essays on J.L. Austin. Oxford: Oxford University Press.
Watts, Richard J. 1989. Relevance and relational work: Linguistic politeness as
politic behaviour. Multilingua 8(2-3). 131-166.
Watts, Richard J. 2003. Politeness. Cambridge: Cambridge University Press.
Werkhofer, Konrad T. 1992. Traditional and modern views: The social constitution
and the power of politeness. In Richard J. Watts, Sachiko Ide & Konrad Ehlich
(eds.), Politeness in language: Studies in its history, theory and practice,
155-199. Berlin: Mouton de Gruyter.
Wharton, Tim. 2003. Interjections, language and the 'showing-saying' continuum.
Pragmatics and Cognition ll. 39-91.
Wharton, Tim. 2009. Pragmatics and non-verbal communication. Cambridge:
Cambridge University Press.
Widdowson, Henry G. 1978. Teaching of language as communication. London:
Oxford University Press.
Wierzbicka, Anna. 1997. The double life of a bilingual: A cross-cultural perspective.
In Michael Bond (ed.), Working at the interface of cultures: Eighteen lives
in social science, 113-125. London: Routledge.
Wierzbicka, Anna. 1999. Emotions across languages and cultures: Diversity and
universals. Cambridge: Cambridge University Press ..
Wierzbicka, Anna (1991) 2003. Cross-cultural pragmatics: The semantics of
human interaction. Berlin & New York: Mouton de Gruyter. 2nd edition, with
a new Introduction.
Wilkins, David A. 1976. Notional syllabuses. London: Oxford University Press.
Wtlkins, David A. 1978. Approaches to syllabus design: Communicative, functional
or notional. In Keith Johnson & Keith Morrow (eds.), Functional materials
and the classroom teacher: Some Background Issues, 7-11. Reading: Centre
for Applied Language Studies, University of Reading.
IIfIIIIIIIIIIIfIIIIffIIfIIIIIIIIIIfIfIffIIIfIIIfIIIIffIfII References 1'6\

Wilson, Deirdre. 1994. Relevance and understanding. In Gill Brown, Kirsten

Malmkjrer, Alastair Pollit & John Williams (eds.), Language and
understanding, 35-58. Oxford: Oxford University Press.
Wilson, Deirdre. 2000. Metarepresentation in linguistic communication. In Dan
Sperber (ed.), Metarepresentations: A multidisciplinary perspective, 411-
433. New York: Oxford University Press.
Wilson, Deirdre. 2011. The conceptual-procedural distinction: Past, present and
future. In Victoria Escandell-Vidal, Manuel Leonetti & Aoife Ahern (eds.),
Procedural meaning: Problems and perspectives (Current Research in the
Semantics/Pragmatics Interface 25), 3-31. Bingley, UK: Emerald.
Wilson, Deirdre & Dan Sperber. 1993. Linguistic form and relevance. Lingua 90.
1-25.
Wilson, Deirdre & Dan Sperber. 2002. Truthfulness and relevance. Mind 111.
583-632.
Wilson, Deirdre & Dan Sperber. 2007. Relevance Theory. In Laurence R. Horn
& Gregory Ward (eds.), The handbook of pragmatics, 607-632. London:
Blackwell.
Wilson, Deirdre & Tim Wharton. 2006. Relevance and prosody. Journal of
Pragmatics 38. 1559-1579.
Wilson, John. 2001. Political discourse. In Deborah Schiffrin, Deborah Tannen
& Heidi Hamilton (eds.), Handbook of discourse analysis, 398-415. Malden,
MA & Oxford: Blackwell.
Wilson, John & B. Gunn. 1983. Political affiliation. Paper presented at the
Sociolinguistics Symposium, University of Liverpool.
Wittgenstein, Ludwig. 1953. Philosophical investigations. New York: MacMillan.
Xing, Janet Zhiq,un. 2004. Grammaticalization of the scalar focus particle lian if
in Mandarin Chinese. Journal of Historical Pragmatics 5(1). 81-106.
Yule, George. 1996. Pragmatics. Oxford: Oxford University Press.
Yus, Francisco.2001. Ciberpragm<itica. El uso del lenguaje en internet. Barcelona:
Ariel.
Yus, Francisco. 2011. Cyberpragmatics. Internet-mediated communication in
context. Amsterdam & Philadelphia, PA: John Benjamins.
Yus, Francisco. No date. Relevance Theory Online Bibliographic Service. http://
personal. ua.es/francisco. yus/rt.html.
 KEYS TO SELF-EVALUATION
QUESTIONS AND EXERCISES
I////////////////////////II//////////I

CHAPTER I
KEY TO SELF-EVALUATION QUESTIONS
1. b) 6. a) 11. b)
2. b) 7. b) 12. a)
3. a) 8. a) 13. c)
4. c) 9. a), b) & c) 14. a)
5. c) 10. b) 15. b)

KEY TO DEVELOPING OUR PRAGMATIC COMPETENCE:

EXERCISES FOR REFLECTION AND AMPLIFICATION (Ch. I)
A) OPEN EXERCISE
B) OPEN EXERCISE
C) Suggested possible meanings. Can you find more?
a. 1) The manager told the forwards to use their brains
2) The manager told the forwards to do more headers
b. 1) The singer totally failed to make me feel emotional
2) The singer totally failed to budge me from the spot
c. 1) Stocks fell badly at Wall Street Stock Exchange
2) There was an earthq,uake in New York, and it affected the buildings on
Wall Street badly
d. 1) As I rushed down the stairs, I encountered him/her
2) As I rushed down the stairs, I crashed into him/her
D) OPEN EXERCISE
284 Key ll/lll/llllllll/ll/lll/l/lllllll//l//lll!l/ll//l//!/ll//l/l!/I/I

E) Suggested possible meanings. Can you find more?

a. Mary's right
1) A legal right that Mary has (e.g. when discussing her rights as a citizen,
for instance)
2) Mary's right side (e.g. when trying to locate and identify an object that
is near Mary's right side)
3) Mary is correct ('Mary is right'. Here the ambiguity is caused by
phonetic reduction, and could occur in any context in which a speaker
is trying to agree with Mary)
4) Mary's punch, as delivered by her right hand (e.g. when narrating how
Mary gave someone a blow with her fist)
b. The President's right hand is in the corner
1) The president's most valued collaborator is in the corner (e.g. when
someone is watching TV and wants to indicate who the president's best
assistant is)
2) The president had an accident; his right hand was cut off, and it is now
in the corner (e.g., when someone is looking at a picture of the moment
after the accident, is describing what happened, and shows his/her
interlocutor that the hand is in the corner of the photo)
c. He saw a woman on a hill with flowers
1) He saw a woman, and the woman was on a hill covered with flowers
2) He was on a hill covered with flowers when he saw a woman
· 3) He saw a woman that was on a hill, who had flowers in her hand
d. We saw her run
1) We saw her when she was running
2) We saw the run that she had in her stocking

F) Suggested indirect meanings.

1) This is most likely an ironic "Thank you", therefore the meaning lacks
the gratitude component and on the contrary, it has the flavor of a
reproach.
2) Here B is most probably not actually thanking A, but emphasizing the
fact that she does not consider she is the right person for the task of
giving the inaugural lecture, and that therefore A's idea is not a good
one.
//////////////////////////////////////////////////////////////// Key 285

3) Here the mother is emphasizing her command to the children to make less
( or no) noise. She is not thanking them for anything.
In all cases, the intonation given to the utterances is of great importance to
understanding their indirect meanings.

G) OPEN EXERCISE

CHAPTER 2
KEY TO SELF-EVALUATION QUESTIONS
1. b) 6. c) 11. a)
2. c) 7. b) 12. a)
3. a) 8. b) 13. a)
4. b) 9. b) 14. b)
5. a) 10. c) 15. c)

KEY TO DEVELOPING OUR PRAGMATIC COMPETENCE:

EXERCISES FOR REFLECTION AND AMPLIFICATION (Ch. 2)
A) a) constative
b) performative
c) performative
d) constative
e) constative
f) performative
g) performative
h) constative
i) performative

Justification: The ones classified as performatives are those utterances in which

hereby can be inserted before the main verb (e.g.: b) I hereby certify that ... ).
286 Key l//ll/l/////ll/ll//ll/l///!///ll//l//ll!///////!/l!l/l//ll/!/I//

B) I certify that the above is an exact and accurate translation of the original
Metalinguistic perfonnative
I swear to God that I didn't do it! Metalinguistic perfonnative
I apologize for having offended Metalinguistic perfonnative
I name this ship "The Wanderer" Ritual perfonnative
I bet you $100 that she's not Collaborative perfonnative

C) a) Abuse
b) Misexecution
c) Misinvocation
d) Abuse

D) a) Explicit performative (order).

b) (Conventionalized) Implicit req,uest. Device used: conventionalized
formula for requests "would you be so kind as ... ".
c) Implicit order. Device used: Imperat_ive mood
d) Implicit promise. Device used: Indication of commitment to carrying out
a future action by means of "will".
e) Explicit performative (promise)
f) Explicit performative (warning).

,;
E) '
Act -
.•;
Main tgpe ·_ . ' ·subtype
.,
..

a) Representative Asserting

b) Cornmissive Swearing

c) Directive Warning

d) Expressive Congratulating

e) Declaration Pronouncing (a couple husband and wife)

f) Representative Asserting

g) Directive Forbidding
//lll/l l /l///!l//ll///l//l/l///l///l/////ll/!l///////!////II//// Key 287

F) , .<
Act '-, IUoc..;_tJonarg /_,' ir · · ,.· · (P()ssible) .. , ·
Lotutionary act · '. . .
•-:-1'; ~ -
• . c.'.,, . ac_t "
..
,. '
•r "'•
perlocutionar9 act
a) Saying and pronouncing Asserting Persuading (the
the utterance "Yes, we American people to
can", meaning by "we" the vote for him)
American people (and here
it could be taken as jokingly
meaning "Obama and his
dog" as well).
b) Saying and pronouncing Swearing Convincing (the
the utterance "I swear ... ", American people
meaning by "I" the speaker, that the speaker will
and by the rest of the words, be faithful to their
their literal, timeless meaning Constitution)
c) Writing the utterance Warning Keeping people away
"Warning, high voltage" using from such a dangerous
the timeless meaning of these place
three words.
d) Saying the utterance Congratulating Praising the
"Congratulations", meaning addressees (the Navy
"congratulations" (its graduates in the
timeless meaning) picture) for having
graduated.
e) Saying and pronouncing the Pronouncing (a Changing the marital
ritual words for marrying a couple husband status of the couple in
couple, meaning by "you", and wife) q,uestion
the couple, by "husband",
the man, and by "wife",
the woman who is getting
married.
f) Saying "I have a dream" Asserting Convincing the
meaning by "I", Martin Luther African-American
King, and by "a dream", population of the
a dream (i.e. its timeless U.S. that things could
meaning). change for them.
g) Writing the utterance Forbidding Showing strangers
"No trespassing. Private that they are not
property. Keep out", meaning wanted at that
by "private" 'private', by particular private
"property" 'property', place.
by "keep" 'keep' and by
"out" 'out' (i.e. their timeless
meanings)
288 Key ///!l/l/l/l//l/ll///l//ll/ll!!ll!/!ll!//!l/l/l/l/l!lll//!l/!/I!!

.,..
G) . •..
•·
. '
. .
Do form-and '
,J:;;, ' ';
FUNCTION/
' .,.

' ..
'/

STRUCTURE ", function MOOD ILLOCUTIONARY .

. coincide?
.· "
FORCE -
..
'
a) YES Subjunctive Hoping, wishing

b) NO Indicative Req,uesting
(interrogative)

c) YES Indicative Asserting

(declarative)

d) YES Subjunctive Hoping, wishing

e) YES Imperative Ordering

f) NO Indicative Ordering
(declarative)
g) NO Indicative Asking a q,uestion/
(declarative) Doubting
h) i) YES i and ii) Indicative i) Asking a q,uestion
ii) NO (interrogative) ii) Exclaiming/
expressing joy
i) NO Indicative Warning/threatening
(declarative)

H) a) Direct
b) (Conventionally) indirect
c) Direct
d) Direct
e) Direct
f) Indirect
g) Indirect
h) i) Direct// ii) Indirect
i) Indirect

I) In both a) and b) the adverbials modify the illocution rather than the
proposition. In a), the adverb seriously functions as a pragmatic marker
which is eq,uivalent to saying "I'm being serious when I tell you that I don't
want to carry on with this relationship anymore". In b) the adverbial although
why am I giving you any explanations at all? again modifies the illocution:
after an assertion whose main aim is to inform the addressee about the
speaker's future actions, the speaker adds a rhetorical q,uestion that indicates
//////ll/l//l//lill////l/l//////ll/l!//lll!III////II/II//I///I/J Key 289

the speaker's doubts about the very felicity conditions of the speech act s/he
has just performed, with the perlocutionary effect of expressing contempt for
the interlocutor.

J) a) I hereby ... Declaration, performative, direct

b) Welcome home Expressive, performative, direct

K) OPEN EXERCISE

CHAPTER 3
KEY TO SELF-EVALUATION QUESTIONS
1. b 6. b 11. C

2. a 7. a 12. a
3. b 8. b 13. b
4. C 9. a 14. a
5. a 10. C 15. C

KEY TO DEVELOPING OUR PRAGMATIC COMPETENCE:

EXERCISES FOR REFLECTION AND AMPLIFICATION (Ch. 3)
A) a) Specific, explicit, definite.
b) Specific, inexplicit, definite.
c) Specific, explicit, indefinite.
d) Specific, explicit, definite, deferred.
e) Generic, inexplicit, indefinite.
f) Specific, inexplicit, definite, deferred.

B) a) He Cataphoric
The Gulf Homophoric
A boy ldeophoric
Him Anaphoric
Ernest Hemingway Homophoric
290 Key llllilllll//ll//ll/lllll/lll/ll/ilil//l!ill/l/ll!ili/l//lll!//I/

b) my Exophoric
( three minutes') Bridging and anaphoric (it refers back to
'my mouth')
The privacy of my Esphoric
(a vacant and preoccupied) Bridging and anaphoric (it
refers back to 'my eyes and face')

c) The city of York Homophoric and Esphoric

That country Bridging and anaphoric (it refers back to 'the city of
York', though not exactly to the city, but to the country where the city is,
i.e. England).
My Ideophoric
Homophoric
He Anaphoric
Robinson Anaphoric
We Anaphoric
Me Anaphoric
Robinson Crusoe Homophoric

C) a) Person deixis, symbolic

That Discourse deixis, symbolic
b) I Person deixis, symbolic
You Person deixis, gestural
Place deixis, symbolic
c) I Person deixis, symbolic
that Place deixis, gestural
Mr. Medvedev Social deixis, symbolic
d) Your Majesty Social deixis, symbolic

D) a) Entailment, syllogism
b) Presupposition
c) (Generalized) conversational implicature
d) Conventional irnplicature
e) (Particularized) conversational implicature

E) a) Factive presupposition The addressee finished his/her PhD Thesis.

b) Existence presupposition There is a hospital called 'Holy Cross' in
Maryland.
////!l/ll///l///l//////l//l!//////II/I//II/I///I/I////////I//I// Key 291

c) (Counter)factive Yussef was not a rich man.

d) Connotation The speaker used to smoke in the past.
e) Factive presupposition I met her.
f) Arson is a bad thing.
g) Connotation She has more than two children.
h) Linda has a daughter called Cynthia.
i) Existence presupposition There is a swimming-pool at the school
(probably Jane's).
..
F)
UTTERANCE MAXIM(S) .
If•
IMPLJCATURE
..
··, ½ FtPDTEn
. '
·' ..
,;_ • ..

. '"
.' •·

1) A: Do you like this country? QUANTITY B does not like the

B: I like its people very much. country very much.
2) A: Shall we go out for a walk? RELEVANCE B does not feel like
B: It's very windy today. going/want to go out for
a walk.
3) (B didn't want her mother to QUALITY B does not think what
know about the surprise party her sister did is clever.
they were preparing for her)
A: I put my foot in it and told
Mom about the party
B: My goodness!! How clever of
you!!!
4) A: So you finally met your in- MANNER B's in-laws are not very
laws. What are they like? nice people.
B: Well, theg are not of the
nicest variety ...
5) Mother: Have you done your RELEVANCE Jill has not done her
homework? homework.
Jill: Nicholas brought candy
to class today.

G) a) Opting out (of the Quality and Quantity maxims).

b) Flouting (of the Relevance maxim).
c) Violation (of the Quantity maxim by not telling the complete truth).
d) Suspension (of the maxim of Quality).
e) Infringement (of the Relevance maxim).
 292 Key ////ll//l//ll//l/ll/l/ll///////l//l////l/l//l//!ll!I//I////I////

H) OPEN EXERCISE

I) OPEN EXERCISE

J) a) R inference
b) Q inference
c) Q inference

K) ' ''· .\.

.... .f ..
Type of meaning
.;
-

Syllogism Entailment

Generalized Utterance-type
conversational implicature

Particularized Utterance-token
conversational implicature

L) a) Q-implicature

b) I-implicature
c) I-implicature
d) M-implicature

M) OPEN EXERCISE

CHAPTER 4
KEY TO SELF-EVALUATION QUESTIONS
1. b 6. C 11. C

2. C 7. b 12. C

3. C 8. b 13. a

4. a 9. b 14. a

5. a 10. a 15. b
//////l//l/l/lll//l!l//l//l/////l//ll///ll/ll!/l//l/ll//!II//I/I Key 293

KEY TO DEVELOPING OUR PRAGMATIC COMPETENCE:

EXERCISES FOR REFLECTION AND AMPLIFICATION (Ch. 4)
A)

Social-norm Jespersen (1965), Quirk (1985)

Conversational-contract Fraser & Nolen (1981), Fraser (1990)
Conversational-maxim R. Lakoff (1973), Leech (1983, 2014),
Gu (1990, 1997)
Face-saving Brown & Levinson ([1978] 1987)
Emotive Arndt & Janney (1979, 1983, 1985)
Relational Watts (2003)
Frame Terkourafi (1999), Aijmer (1996),
Deutschmann (2003)
Rapport-management Spencer-Oatey (2000)

B) a) Be polite.
b. Be clear.

C) a) Modesty Maxim (self-devaluation)

b) Opinion Maxim (reticence)
c) Obligation Maxim (apologizing)
d) Generosity Maxim ( commissive)
e) Feeling Maxim (sympathy) and Generosity Maxim (commissive)

D) The order would be: b, c, d, a.

E) a) On record with negative politeness.

b) Bald on record
c) On record with positive politeness
d) Off record (She flouts the Quality Maxim, and the implicature is that she
does NOT love it at all when her friend becomes so masochistic)
e) Off record (He flouts the Relation Maxim, and the implicature is that he
does not want to talk about his relationship with Mary)
f) On record with negative politeness
294 Key l!//l/l///l/ll/lll//ll////ll///////ll///////l//l/l/////lll/l!II!

F) OPEN EXERCISE

G) ~. .. .. ,

Utterance
·a
.. ,. .,,
Possible f ormulatioil
'"'
E.g.: Tom to his sister: Pass me that book! Low D, low P, low R
A woman to a stranger in the street: High D, low P, moderate R
Excuse me, but could you possibly
change me this $10 bill? I need coins for
the parking meter and I don't have any
change.
A woman to her boy-friend: C'mon, tell me Low D, low P, moderate R
about your previous girlfriend. What was
she like?
Jonathan to his colleague at work: Low D, low P, high R
Look dude, I'm having some financial
problems at the moment. Do you think
you could lend me $20,000?

H) ' i.'~ - .~.. :

Utterance/situation Imp~liteness st~ategy

In the elementary school playground: Positive impoliteness
Martha (to a group of classmates):
May I play hopscotch w ith you?
Lorna: No, we don't wanna play with you!
An exchange between workmates: Negative impoliteness
Richard: Don't do that, George, it's not an
honest thing to do!
George: You of all people are saying this
to me ... YOU... who were convicted in a
famous corruption case some years ago!!
Julia is angry with her friend Cynthia, Withhold politeness
who ignores this circumstance and when
Julia's birthday comes, she goes to visit
her and gives her a present. Julia does not
thank or even talk to her friend.
A couple are having an argument: Bald on record impoliteness
She: How could you do that to me?
He: What are you talking about? I haven't
done anything to you!
She: It's incredible how irresponsible and
cheeky you can be. Get the hell out of my
sight!!
//l///l//l/l/l/l/////////l/l///!l/!l//l//!/l!l!II////////I//II// Key 295

I) In this passage from Saltwater there is constantfustigation impoliteness

on the part of both characters. There's only one occasion where Ray uses a
different type of impolite act, namely a formally polite act with an impolite
purpose, and that is when he says, almost at the end of the conversation,
"I'm going to do it for you", an act which on the surface appears to be polite
because Ray pretends that he is finally going to comply with Trisha's wish,
but as can be seen later on, this is not so.

K) OPEN EXERCISE

CHAPTER 5

KEY TO SELF-EVALUATION QUESTIONS

1. b 6. C 11. b
2. C 7. b 12. b
3. a 8. a 13. a

4. a 9. b 14. C

5. b 10. C 15. b

KEY TO DEVELOPING OUR PRAGMATIC COMPETENCE:

EXERCISES FOR REFLECTION AND AMPLIFICATION {Ch. 5)

A) In all three cases we find a pun on words. What at first sight seems to be
the most logical and relevant interpretation clashes with another possible
interpretation which in fact turns out to be the most relevant for this context
(that of someone telling a joke), thus becoming the punchline and causing a
clash of meanings that is precisely what makes the joke humorous:
a) Clash between the two possible interpretations of the word right (good,
correct vs. the side of the body opposite to the left one).
b) Clash between the literal and the figurative meaning of the verb to hit
(in the first case, it is the ball that hits the speaker, and in the second, it
is just an idea that comes into her mind). Because at the beginning of the
296 Key ////lll/////ll//l/////!//l/////ll!II/I//I//III//II/I//I///II////

utterance the speaker uses the verb wonder the hearer is led to think that
the most relevant interpretation of the verb hit will be the figurative one,
only to discover immediately after that the literal interpretation is the one
that will make the joke relevant.
c) Clash between the two possible interpretations of the expression to take
a day off (to take a free day from work, or literally to erase one day from
the calendar).

B) 1) b - overt communication
2) b - covert communication

C) ).f
"
'
... -. '
·., .
., ,,
.
, .
SITU_t\TIQN/ · MOST RELEV.t\NT .
· -PROCESS· · ·,
UTTERANCE EXPLICATURE
LC "·
'
Jerry is a cook and When will you finish Enrichment
is now in the kitchen cooking the souffle? (in
of his restaurant. His contrast to eating it)
colleague asks him:
When will youfinish
the souj'fl,e?

Lindsay gives her Lindsay is reproaching her Higher-order

daughter a present daughter for not having explicature
for her birthday, but thanked her.
her daughter does not
thank her for it, so
Lindsay says out loud:
You're welcome!!

Sandra comes back So, did you like Ernest Reference assignment
home from a blind date (i.e. your blind date)?
with Ernest. Her sister,
who knew about her
blind date, anxiously
asks her: So, did you
like him?

JoAnne knows that You are not telling the Disambiguation

what her friend is truth! (as opposed to
telling her did not "you're resting in a
really happen, so she horizontal position")
says: You're lying!
/ll/ll//llilll//lllllill//lll/illll//l/llll/ll!l/!ll/lll/ll/l!II Key 297

,, .,,
D) . '
.. IMPLICATED . IMPLICATED
.UTTERANCE . :. .,
CONCLUSION. .
·. · PREMISE
,, . .
•.
..
.

Tom: Will you invite Henry to Henry is a jerk (or so Abby will not invite
your party? Abby thinks) Henry to her party
Abby: I never invite jerks to
mg parties.
A: Look! There are Wilma and Wilma is dating David David is handsome.
David.
B: Ah, I have never seen
David before, but once
more we have proof of the
fact that Wilma always
dates handsome men.
Anne: Is there any du lce de Beatrice is not fat Beatrice never has or
leche in the fridge? eats dulce de leche.
Beatrice: If I had dulce de
lee he, I would be fat.

E) a) Concept
b) Concept
c) Procedure
d) Concept
e) Procedure

F) a) Conceptual
b) Procedural
c) Conceptual
d) Procedural

G) a) A board for ironing clothes on.

b) She has the same mother or father as Peggy.
c) A wax structure made by bees to store their honey.

H) Here or is exclusive, since the third interpretation (the hat will grow your hair
at no cost to you) is contextually impossible. The reader of the advertisement
has the strong contextual assumption that the company promoting its product
will not allow customers to use it for free. The exclusive interpretation of or is
therefore necessarily the relevant one. Reformulation as a negative conditional
is "If our new kind of hat, worn 10 minutes a day, does not grow your hair in
30 days, there will be no cost to you/you will get your money back".
298 Key /////l//l/l/ll/l//l/l/ll//l//l////l//!///l///l//ll!/I/II/II/I///

CHAPTER 6
KEY TO SELF-EVALUATION QUESTIONS
1) b 6) C 11) C

2) C 7) b 12) b
3) a 8) a 13) a
4) a 9) b 14) C

5) b 10) b 15) C

KEY TO DEVELOPING OUR PRAGMATIC COMPETENCE:

EXERCISES FOR REFLECTION AND AMPLIFICATION (Ch . 6)

A) OPEN EXERCISE

B) " . . . -~ . ;, .

EXPLANATION I;JYl>ilAGMATIC
q9MPOUND WORD/ EXPRESSION ENRICHMENT

He wrote two words on the A (normally wooden or slate) surface

chalkboard. to write on with chalk

Little Billy wanted a skateboard for A wooden, plastic or metallic oblong

his birthday. device on roller skate wheels used to
skate on smooth surfaces

This is the chessboard Fisher used to A checkered board for playing chess
play against Kasparov.

My laptop has a touch keyboard. Set of keys in a computer

The whole course went on an Metonymy: the course cannot go on

excursion to the Highlands. excursions. It refers to the people
attending the course.

I love reading Shakespeare. Metonymy: One cannot read

"Shakespeare" (the person), but a
book he has written. So it refers to
Shakespeare's books or works.

C) OPEN EXERCISE
llll//llll///l//ll/ll//ll///l//ll!ll!I/III/II/IIIIII/II/I/I//I// Key 299

D) a. You're going out in that pink Here there is a mismatch

between syntax and phonology, because a declarative form is used with
the intonation of an interrogative clause, in which case we speak of an
indirect speech act, and so normally in these cases the speaker raises her
voice more clearly in order to get across the right speech act (a q,uestion
loaded with -very possibly- some invoked negative evaluation of the pink
jumper).
b. Ann: Are you guys coming to the beach with me?
Ben: Well, LESLIE is. The contrastive intonation used on the word
LESLIE will trigger the implicature that Ben has no intention of going
to the beach with Ann.
c. I will ALWAYS remember you. The nuclear stress (prosodic prominence +
change of pitch) on ALWAYS distinguishes the utterance from its neutral
counterpart (I will always re'member you), and therefore could occur in
response to an utterance like You will not remember me any more in a
year or two.

E) a) Nigel Slater: Cooking is a joyous, lifelong pleasure [Attitude: positive,

inscribed Appreciation]; [Whole headline: Heteroglossic Engagement
(because it is Nigel Slater who says this, not the journalist)/
b) Greek conservatives admit defeat [Attitude: inscribed, negative
Judgement (of Greek conservatives)] [Whole headline: Monoglossic
engagement (it is the journalist who commits herself to the truth of the
assertion expressed) J
c) Canadian leaders gone wild [Attitude: inscribed, negative Affect
(referring to the leaders' feelings) with overlapping invoked negative
Judgement (of their behavior)J in three-way debate battle [inscribed,
negative Appreciation (of the debate)] [whole headline: Monoglossic
EngagementJ

F) a) Directly expressed emotion (anger).

b) Conceptualized emotion (fear).

c) Conceptualized emotion (sadness).
d) Directly expressed emotion (disgust).

G) OPEN EXERCISE

H) OPEN EXERCISE

//////////////////////////////////////
A Anointing 145, 150
/II///II/I/II/I//III/I/I//II///I/II
Abduction 246
Act
-Face-Threatening (FTA) 145-154, 171
- fustigation impoliteness 163, 166-167
- illocutionary 38, 44-4 7
- impoliteness 162-168
- involuntary impolite 163, 164
- locutionary 45-46, 51, 64
- of predication 51
- of reference 51
- ostensive 184, 200, 208
- overwhelming silence 163, 166, 172
- perlocutionary 45, 46, 64
-propositional 51
- self-impoliteness 163, 164-165
- with a polite purpose, formally
impolite 163-164
- with an impolite purpose, formally
polite 163, 165
Action, redressive 146, 148, 149, 152,
153,171
Affect (Appraisal Theory) 231, 232, 237,
252
Affiliation 155-156, 171
Ambiguity 12, 28, 191-192, 219
Analysis
- componential 221-222
- goal development 143
- interpersonal management 143
- talk exchange 143
INDEX
Appraisal Theory/ Model 231-233, 237,
252
Appreciation (Appraisal Theory) 231,
232,233,234,237,252
Approach
- adaptive 220
- to politeness, frame 132, 134, 170
Assumption 7, 92, 100, 101, 116, 151
- contextual 91, 108, 183, 185-186, 188,
190-191, 199,200,207,208
Attitude 7, 8, 196-197, 233,245
- (Appraisal Theory) 231, 232, 234, 237,
241,252
Autonomy 155-156, 171
Axiology 230
B
l/l///l/1////I/III////I/II/III/I//I
Behavior
- coded 188
- coded assisted by inference 188
- politic 157, 168, 172
C
/II//I/ / IIII///I/II//I/II//////II/I
Calculability ( of conversational
implicatures) 106
Cancelability ( of conversational
implicatures) 105-106, 108, 112, 117,246
Clause, types of and associated
l/l/ll!/l//l/!/ll//l///l/l//l/l/l//l/l///l/l!l/!l/l/l/l!/II/I
illocutions 48-49, 64
Coding 181, 196-198, 205
Cognition 15-18, 27, 183, 184, 185-187,
207, 238-239, 253
Cognitive Linguistics 15, 237-238
Communication
- accidental 187
- computer-mediated 249
- covert 187-188, 206
- cross-cultural 24
- inferential 182, 184, 189, 196-197, 207
- intercultural 24, 26
- interdiscourse-system 26
- intracultural 4, 23, 24
- ostensive or ostension 184, 188-189,
194,204,207
- ostensive-inferential 189
- overt 184, 187-188, 206, 207
- pictorial 204-206
- trans-cultural 24, 26, 29
Compound 28, 223-224
Comprehension 183-185, 189, 190-196
Conceptualization 17,237, 238-239, 242
Conclusion, implicated 190, 195, 208
Connective 196, 200-202
Connotation 94-96, 116
Constative 38, 43-45, 46, 48-51, 64
Construction, cleft 94, 97, 116, 218-219,
228
Context VIII, 2, 3, 4-8, 9-12, 27, 28, 63,
64, 74-75, 80-82, 93, 111-112, 115, 116,
117; 130,142,170, 182-185, 187-194,206,
207,208, 218-219, 228,236, 250-251, 252,
253
- cognitive 2, 5, 7, 8, 15-16, 28
- cultural 6-7, 8, 15, 28, 130,236
- dynamical 4-8
Index 301
- emotional/attitudinal or emotive 5, 7,
8,15,28, 75, 130,135,158, 168-169,236
- linguistic 5, 8, 28, 75
- social 6, 8, 28, 143
Context, types of 5, 7-8, 27
Conventionalization 19-21, 29, 245
Culture 4, 6-7, 8, 23-27, 28, 57, 61, 131,
142,152, 154-157, 171
Cyberpragmatics 249-251, 253
D
/IIIII//II//II/I/IIIIII/III/II/I/II
Decoding 188, 190, 206, 208, 246
Decontextualization 63, 219, 226
Deictic center 81-83, 86, 87, 89, 90, 115,
116
Deixis 80-90, 115-116
- and grammar 86-90
- discourse 83, 85, 86, 87, 88, 90, 115
-person 82, 86-87, 115
- social 84, 86, 88, 90, 115
- spatial or place 82, 83, 86-87, 89, 115,
116
- time 82, 83, 86, 87, 89, 115
- types of 82-86
Demonstratives, deictic/non-deictic use
of 85-86
Diminutive 235, 240, 253
Direction of fit (of speech acts) 52-54
Disambiguation 191
Discourse
- analysis VIII, 14-15, 28, 62,243
- system 26-27, 29, 156
Distance (D) 57, 151, 155, 161
302 Index !lll//l//lllll/l!//l//l/l////l//l/l//l//ll//llll///ll/ll/ll!I

E - identity 135
I///III///III//I//I//I/II/II//III// - management of or face management
Effect 134-135
- contextual 183-186, 192, 201, 204, 207 - negative 135, 143, 144, 145, 146-147,
-perlocutionary 46, 62, 64, 144 150, 155-156, 158, 171
- positive cognitive 185-187, 204, 207 -positive 143, 144, 145-146, 14 7-148,
150, 155-156, 158, 161, 171
Effort
- q,uality 134-135
- cognitive 19, 183, 206
Facework 145
- processing 133-188, 204, 207
Felicity conditions ( of speech acts) 38,
EMO-Fundett 237
40, 42-43, 52, 62, 64
Emotion 7, 8, 17, 97, 133, 158, 169, 171,
Focus 97, 233, 253
206, 230, 231-232, 236-242
-verum 235
Encoding 196-198, 208
Force
Engagement (Appraisal Theory) 231,
232-233, 237,252 - illocutionary 47, 50, 51-53, 60, 64, 65
- heteroglossic 232, 234, 252 - (im)politeness 156, 162-163
- monoglossic 232, 234, 252 Form and function 47-51
Enrichment 112-113, 183, 190, 192-193, Formalism 218
194 Frame 132, 134, 170
- pragmatic 223-224, 226, 252 Functional Discourse Grammar 60 220 ) )

Entailment 91, 110, 111, 116 227-228

Ethnocentrism 142
Evaluation 230-236, 252
- inscribed 234
- invoked 234, 235, 236
Experientialism 16
Experimentation ( experimental
pragmatics) 24 7-249
Explicature 9, 190-194, 199, 200, 202,
203, 205,206, 208
- higher order/level 193, 195, 196, 197,
202, 203, 205, 208, 241
Functionalism 15, 218, 219, 220
G
I/I/I//I////I///I//I/II//III/I/////
General Strategy of Politeness (GSP)
139-142, 155, 170
Gesture 85, 86, 115, 133, 188, 204, 205,
207, 230,236,241
Graduation (Appraisal Theory) 231 ,
233, 234, 235, 237,252
Grammar, usage-based approach to 18
Grammaticalization 21-23, 29 86 87
) ) )

F 88, 89, 90,245

II/III/I/I/I///I/I/////II/I///I//II

Face (Politeness Theory) 25, 26, 40,

132, 133, 134-135, 142, 143-148, 149, 150, H
II/I//II/II////I/I//III//II//III///
151, 153, 155-156, 157. 158, 160-161, 163,
165, 170-171 Habitus 133
///////////////////////////////////////////////////////////// Index 303

I 185, 188-189, 190, 191,200,202-204,205,

/IIIIII//II//III/IIII//II///I//I/I/
IFID (Illocutionary Force Indicating
Device) 51-52, 63, 64
I-implicature 113-114
lllocution 37-38, 39, 40, 44, 45--47, 48--49,
50-51,53-54,56,57,60,61,63,64,227
Image or Picture 204-206, 208
Implicature 20, 21, 62, 91, 99-114, 116-
117, 136,144, 150-151, 183,187,188,190,
194-195,202,203,204,205, 206,207,208,
228,244,245,246,247,248
- conventional 100, 104-107, 108, 117,
- conversational 20-21, 91, 100-102,103,
104, 105-115, 117, 136, 144, 150-151,
188,244,246
- generalized conversational 107-108,
109, 111-112, 117,244
- Neo-Gricean theories of 109-114, 117,
194
- particularized conversational 107-
108, 112, 117
- scalar 109-112, 117, 248
Impoliteness 130, 156-157, 157-169, 170,
171,172
225, 245-247,
-pure 188
-types of 91-114
Information
-given 218
-new 218
Intelligence, emotional 4
Intention/intentionality VIII, 4, 7, 19,
29,46, 50, 51, 56, 58, 62, 63, 64, 76, 82,
99,101,103,113,137,114,158,160,162,
163,164,168,182,183,187, 188,189-190,
200,207,246, 250
Interjection 196, 241, 253
Intersubjectivity 245
Intonation 22, 25, 44, 49, 101, 133, 148,
196, 197,206,225-226,226-230,235,241
I-Principle 112-113, 114, 117
Irony 3, 10, 20-21, 28, 62, 101, 103, 117,
159, 161,164-165, 189,194,203,226,235,
241,248
J
ll//lli/l/lll//ll//l/l///ll///l//l/

- bald on record 158, 171 Judgment (Appraisal Theory) 231, 232,

- negative 158, 171 233, 234,236,237,252
-positive 158, 171
Indexicals 80-90
L
- and context 80-82 /II//I/I///I//I I/I//III/////I//II//

Indirectness 18-21, 23, 29, 50, 55-58, 61, Lexicon 221-226, 238,251,252
62,63,64,65, 77,115,117,167,188,234, Lingua franca 24
236
Linguistics
lnfelicity 42-43, 52 - computational 13, 62, 65, 245-247, 251,
- abuse 43, 52, 64 253
- misexecution 42--43, 64 - generative 15, 16, 59, 60, 65, 216, 221
- misinvocation 42, 52, 64 - historical 242-245
Inference 19, 23, 29, 91-114, 115, 116, Locution 45-46, 51, 64
304 Index /ll//l/l/lllll////ll/ll/////ll!II//III/I///////I//I//I///I//I

M - literal 3, 16, 50-51, 56-57, 99, 101, 106,

/I///II/IIIII/II/I/I/II/II/IIIII//I 165
Macropragmatics VIII, IX, 14-15, 28, - -nn 91, 99-100, 104, 116
216,243,252 - procedural 195-198, 200-202, 208,229,
Management,rapport132, 134,143,170 241, 244-245
Marker -propositional 5, 198-199
- discourse 21, 22, 34, 44, 83-84, 88, 196, -timeless 2, 27, 36, 46
201,208,229,233, 243-245,253 - truth-conditional 198, 245
-pragmatic 21-23, 50, 198,201,218 - utterance-token 111
Maxim 100-113, 137-141, 143 - utterance-type 111
-Agreement 138, 141 Metaphor 115, 242, 248, 253
- Approbation 138, 140 Metonymy 75, 224, 242
- Generosity 138-140, 143
Micropragmatics VIII, IX, 14-15, 28,216,
-Manner 25, 100-102, 106, 110, 112, 117, 252
131,182
M-implicature 113-114
-Modesty 138, 140
Model person 161
- of the Cooperative Principle 25, 101,
Mood 44, 47-49, 60, 64, 138, 140-141,
110,194
150,155,196, 199-200,202,208,229
-the Cooperative Principle, non-
M-Principle 113, 117
observance of 102-104, 110, 117, 136
Multimodality 204, 249
- Quality 21, 25, 100-101, 103, 117, 131,
182
- Quantity 25, 100-101, 103-104, 110-112,
117,131,156,182
N
/I//II///IIII/I/II//I/I//II//III/II
- Relation 25, 100, 102, 110, 117, 131,
Narrowing 224
182,228
- Q- 224-225, 252
- Self-Denigration 143
-R-225, 252
- Sympathy 138, 141
Nondetachability (of conversational
-Tact 137-138, 140, 143
implicatures) 104
Meaning
Nonpoliteness 157
- conceptual 195-198, 201-202, 208, 229,
244-245
- conventionalized indirect 19-21 0
- grammaticalized indirect 21-23 /1//I//IIIIIIII/I/IIIIII/II///I//I/
- Grice's classification of 107-109 Omission of politeness, involuntary 164
- implicit 43, 91 Optimal Relevance, presumption of 184
- indirect 12, 18-23, 29, 56, 62, 117 Ordinary language philosophy 73
l/!//III/IIIII//IIII//II/IIII//////II/I//II/I/IIIII//II/III/I Index 305

p Power (P) 151-153, 155

III//II//I/IIIIIIIIIII/II/I//III/// Pragmaphilology 243, 253
Payoff 4, 154, 259 Pragmatics
Performative -Anglo-American School of 13-14, 28,
- collaborative 41 216
- explicit 43-45, 64 - and cognition 15-18
-group 39, 41-42, 64 - and culture 23-27
- implicit 43-45, 63 - central topics of 12-15
- metalinguistic 39, 41 - cognitive 16
- ritual 40-42 - component view of 13
- types of 39-42 - computational 245-24 7
Perlocution 45-4 7 - definition and scope 1-29
Perspective on politeness - diachronic 243, 253
- emotive 132, 170 - European Continental School of 13,
28,216
- relational 153
- experimental 24 7-249
Politeness 129-172
- historical 242-245
- absolute 138, 141, 170
- main schools of 12-15
- bald on record 149, 154, 155, 171
- perspective view of 14
- first-order 132
Premises, implicated 190, 195, 208
-formal 136
Presumption
- informal 136-137
- of optimal relevance 184
- intimate 136
- of relevance 183, 190, 194
- mock 159, 171
Presupposition 13, 14, 91, 92-99, 100,
-neg-139
105, 108, 116, 117
- negative 137, 139, 148, 150, 154, 161
- and grammar 96-99
- non-verbal 169
- connotative 93-96, 116
-off record 148-150, 154,161,171
- existential 93, 96, 116
-pos-139
-factive 93-94, 98, 116
-positive 137, 139, 148-149, 154, 155,
- types of 93-96
171
Principle
- relative 138, 141
-Balance 143, 180
- scale, pragmalinguistic 141-142, 170
- conversational 110
- scale, sociopragmatic 141-142, 170
- Cooperative (CP) 25, 100-102, 106,
- second-order 132
110, 117, 131-133, 136, 137, 170, 182,
- unmarked 157 183, 194
- withhold 159, 166, 171 - Politeness or Principle of Politeness
Positivism, logical 37, 63 (PP) 132, 137-143, 170
306 Index /l!l///////l/ll/ll/l/l///l///l/l//ll!/l////!l/l!//ll////!II!!

- of Relevance 182, 183-185, 189, 191, - indefinite 76, 77, 115, 218
194 - specific 77, 115
- of Relevance, Cognitive 184 - types of 76-80
- of Relevance, Communicative 184 Relevance Theory (RT) or Theory of
Pronouns Relevance 181-208
- deictic use of 80-90 Reinforceability ( of conversational
- non-deictic use of 86, 90, 115 implicatures) 106, 108, 117
Propositional form 190, 194, 195 Rights
Prosody 197, 226-230, 235, 252-253 - association 135
- eq_,uity 135
R inference 111
Q Routines, conversational 26, 134
//II/I///I///I///I///II/I/IIII///II
R Principle 110-111, 114, 117, 224-225
Q-Principle 112
Rudeness 20, 24-25, 164, 168, 172
Ql-implicature 110
Rules
Q2-implicature 110
- of speech acts, constitutive 52, 137,
Q-inference 111
193
QI-maxim 110
- Politeness 24, 131, 146, 170
Q2-maxim 110, 112
- Pragmatic Competence 136, 170

R s
/II///I/I/II/I/I/III/III/II/II/I/II //I/II/II/I////I////I///II/II///III

Ranking (R) 152-153 Salience 202-204

Rationality 100-101, 116, 130-131, 143-
145, 147,153,161,169,171
Reference 74-80
- anaphoric 77, 79, 83, 86, 115
- assignment 192
- bridging 77, 79, 115
- cataphoric 77, 79, 83, 115
- deferred 74, 75
- definite 76-77, 78, 93, 96, 115, 116, 218
- esphoric 78, 79, 115
- exophoric 78, 79, 85, 86, 115
- generic 77, 85, 115
- homophoric 78, 79, 115
- ideophoric 78, 79, 115
Sarcasm 16, 62, 142, 149, 159, 161, 171
Semantics, truth-conditional 2, 37-38
Sociality rights, management of 134-
135
Speech act 35-64
- commissive 54, 138, 140
- communicated 193
- constative 38, 43-46, 48-51, 55, 64
- conventional indirect 61-62
- declaration 38, 54-55
-direct 60
- directive 53, 54, 140
- expressive 11, 54, 138
- indirect 18, 55-58, 61, 63, 65, 188
lll//lllll/l/l//ll/l/lllll/l!/IIII/II/I//II/II//III/IIII/II/I
- infelicitous 42, 52-53
- non-communicated 193
- performative 37, 38-46, 48-49, 51, 55,
59, 61, 63, 64-65
-primary 56-57
- representative 53-55
- secondary 56
- types of 45-4 7
Speech act theory (Austin) 36-51
Speech act theory (Searle) 51-58
Speech act theory , post-Searle studies/
work 58-63
Stance 230, 232-233, 235, 237, 240, 251 ,
252
Stint on politeness
- involuntary 164
-voluntary 166
Strategy 21, 139, 170-171
-politeness 144, 148-159
Stress, nuclear 228, 235
Supportiveness, interpersonal 133
Syllogism 91, 105
Syntax IX, 2, 4, 12, 13, 16, 28, 36, 43,
59-60, 144, 192, 195, 206, 216-221 , 223,
226,227, 235, 238, 240, 252
T ///II//I////I/JJJJ/IJ/////I/I//J/JI
I//II////II////I///I//II//////I//II
Talk
- Emotion 237
- Emotional 237
Theory , practice 133
Theory of preferred interpretation 111,
117
Topic 228, 229, 242
Index 307
u
//J/I//J/////I///I//I////I///II////
Underdeterminacy 28, 99, 182, 189, 199-
200, 207, 208, 222-223, 252
- and context 9-12
Universality 25-26, 29, 57, 60, 106-107,
108,117, 131, 142, 143, 154-155, 167,171,
187
Utterances, grammatical form and
function of 47-51
V
II//JJ//IJ/II/II///II/I/I//////III/
Verb,modal109, 144,153, 189, 208, 235
View of politeness
- conversational-contract 132-133, 170
- conversational-maxim 132-133, 135-
143, 170
-face-saving 132-133, 143-154, 171
- rapport management 132, 134
- social-norm 132, 133
View of pragmatics
-Anglo-American 13-14, 28, 216
-European Continental 13, 28, 216
w
Wants, face 144-145, 152, 158
Wastebasket, pragmatics as 217
Weightiness (W) 152
Word order 48, 218, 220-221, 253
Work, relational 133, 168-169, 172, 233
 Mackenzie

cxberpragmatics

ISBN : 978-84-486-0760-9 Editorial

I
111111111111111 II III Ill
9 788448 607609 www.mheducation.es