Corresponding to Chapter 1 (Pragmatics: 1

Definition and Scope) of base book:

Alba-Juez, Laura & Mackenzie, J. Lachlan (2016).

Pragmatics: Cognition, Context and Culture.

Madrid: McGraw Hill

 Pragmatics is a complex, dynamic discipline whose boundaries are far from
clear, and which is therefore not easy to define in static, absolute terms.
 It is a sub-discipline 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 acquired until the first half of the 20th
century, when Morris (1938: 21) wrote about the “pragmatic dimension of
semiosis” as the “relation of signs to interpreters”.
 Pragmatics deals with meaning, but meaning in pragmatics goes much
further than in the field of traditional, truth-conditional semantics. It is to be
regarded as a crucial aspect of the whole social, cultural, and cognitive
contexts.

2
 We should therefore 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 that expression in a given situation

It is important to bear in mind, then, that pragmatics is defined with respect to a

speaker or user of the language.

 In spite of these differences between semantics and pragmatics, the boundaries

between them are not clear-cut. They share a great deal of common ground, and far
from being opposed, they complement each other and even overlap in some respects.

3
 Therefore, when doing Pragmatics we are taking a theoretical and
methodological approach to language that focuses more on context and
actual linguistic behaviour 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.

4
 Mackenzie (2015) 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 intended meanings of utterances and/or whole
texts.
 Pragmatic knowledge allows us, among other things, to interpret the same
utterance in completely opposite ways if said in different situations. Consider
a) and b):
a)
Susan: Ann has finally worked out the solution to that difficult math problem.
Peter: How clever of her!
b) (Situation: Peter does not like John)

Susan: Ann has finally decided to marry John.

Peter: How clever of her!

5
 Pragmatic knowledge is not only necessary for understanding complex and
intricate ironic meanings, but also for simple cognitive/linguistic processes
such as disambiguating the real person or thing behind a referring term or a
deictic pronoun.
 For instance, knowing the syntactic properties of pronouns in English is not
enough to interpret you or him in utterances like the following:
a) You and you, but not him, will come to London with me.
b)

 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: a skillful use of these resources will most
surely bring about payoffs of various kinds.

6
 In conclusion:

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.

7
 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
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 interaction.

8
 But what do we exactly mean by context, 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
Alba-Juez & Mackenzie (2016: 8):

9
 The linguistic context 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.
 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 question.
 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.
 So  utterances are both constrained by contexts and themselves
build new contexts.

10
 The social context is broader, including such questions 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 ),
which for instance determine who leads a conversation, who asks the
questions, 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).
 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 manoeuvering towards the
achievement of their personal goals.

11
 The cultural context reflects how the variables of the social context are
interpreted from the perspective of a particular culture.
 A culture may align 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 different groups
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, ‘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 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.
 But cultures are very far from being static or monolithic. 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.

12
 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.
 For instance, 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 subsequently 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 equipment 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.

13
 Alba-Juez & Mackenzie (2016) add a fifth type of context 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 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).

14
 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.
 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
at also shape our cognitive system and our emotions in some way or another.
15
 Hence, it is apparent that utterances cannot be treated
in isolation, but as a function of their possible
contexts.
 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.).
 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.

16
 In conclusion, the concept of context is a multifarious and
dynamic one, which changes continually and always
influences the linguistic product of a discourse interaction.
Consequently, 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.

17
The underdeterminacy of language and its relation to context

 Any given utterance could have more than one possible meaning,
depending on the conditions surrounding the speech event.
 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.

18
Example: I always remember how much at a loss I felt the first time I
had to pay for my groceries at an American supermarket, when the
cashier asked the underdetermined question “Paper or plastic?” :

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-l-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”?
19
 In the previous example the meaning encoded in the linguistic
expressions used underdetermines the proposition expressed (i.e.
what is said). Some authors (e.g. Carston 2002) argue that this is the
only type of underdeterminacy that should be covered in Pragmatics.
However, other authors (e.g. Grundy 2008 or Huang 2015) include
other types within its scope: they also include instances in which
a) the linguistic meaning underdetermines what is meant, or
b) what is said underdetermines what is meant.

An example of a) would be the sign “Child psychiatrist” (Grundy 2008:

9) on the door of an office in a clinic, where the most retrievable and
appropriate interpretation would be that it refers to a psychiatrist who
treats children and not to a psychiatrist who happens to be a child, but
if we only considered the possibilities given by the logic and grammar of
the English language, both meanings are equally possible.

20
 2) below is an example of b) above, i.e. what is said underdetermines what is
meant. This type of underdeterminacy includes instances of irony,
understatement and the like. Compare 1) to 2):
1)
A: Welcome to fabulous Las Vegas (the famous sign at the entrance of the city
of Las Vegas, Nevada, USA)

2)
A: I didn’t know there was so much crime and vice in this city.
B: Welcome to fabulous Las Vegas.

21
 Underdeterminacy is related to ambiguity. Both are properties of everyday
language that are important and dealt with in Pragmatics. Huang (2015)
provides interesting examples of lexical and syntactic ambiguity which also
present underdeterminacy if taken out of their contexts, such as a) and b):

a) John is looking for his glasses

b) They are cooking apples

There are at least two possible interpretations for a): 1) John is looking for his
spectacles , or 2) John is looking for his drinking vessels .

As for b), it could be the answer to at least two different questions: 1) What are
they doing in the kitchen? Or, 2) What kind of apples are those?

22
 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 by having access to the context(s) in which the expression is

used can we choose one of these four (or any possible other)
alternatives.
23
 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.
a) 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). The central topics of inquiry include implicature,
presupposition, speech acts, deixis and reference”. 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.

24
 The European Continental conception of pragmatics 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 consequently advocates a broader view, presenting a functional
perspective on all levels of linguistic description and beyond.
 As Bublitz and Norrick (2011: 4) put it, within this broader point of view,
Pragmatics is seen as “the scientific study of all aspects of linguistic
behaviour”, and as an area of research that is “fundamentally concerned with
communicative action in any kind of context”.
 Consequently, according to the Continental view, Pragmatics is not confined
within the boundaries of just one linguistic component, but is ‘omnipresent’
in all components and every aspect of linguistic behaviour. This broader view
has been more inclined than the component view to consider pragmatic
phenomena as linguistic data that necessarily have to be studied in relation
to other domains, giving rise to new sub-areas of research such as
computational, historical, anthropological or experimental Pragmatics.

25
 But the above dichotomy of views does not in fact present the
real current picture: as Huang (2015) points out, there has been
some convergence between the Anglo-American and the
Continental traditions, for there has been important work done
on topics such as implicature or presupposition from a
Continental perspective, and at the same time the Anglo-
American school has delved into further research on the relation
of Pragmatics to areas such as cognitive psycholinguistics,
computational linguistics or neuro-linguistics, just to name a
few.
 This convergence was a necessary and fruitful one, for it joins
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.

26
 The scope, interests and methodology of the Anglo-American school could be
identified with what has been called micro-pragmatics, whereas those of the
Continental European School, with the so-called macro-pragmatics.

 So, when doing micro-pragmatics we would be concerned with specific

theoretical discussions about topics such as implicature, reference or deixis,
while when doing macro-pragmatics 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.

27
 But, again, this dichotomization of the field of
pragmatics presents its problems and raises some
important questions, such as:

a) Can a researcher do micro-pragmatics and relate her

findings on the topics of implicature, presupposition,
etc. to other non-linguistic social, cultural or clinical
phenomena?
b) If macro-pragmatics 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?

28
 To the first question 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 topics and their contexts
of whatever type. 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 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 convergence of both approaches
seems desirable, the all-embracing character of macro-
pragmatics seems too broad to make Pragmatics a necessary
sub-field of Linguistics, distinct from any other.
29
 Along this line of thought, then, we feel more prone to think that
both the findings and objectives of micro and macro-pragmatics
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.
30
 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, questions
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
Analysis in which the relation of utterances to their
social, cultural and cognitive settings is of central
importance.
31
 An important contributor to the study of the cognitive context of
communication is Bara (2010). 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, for example those that concern the nature of
the ongoing communication: a teacher and her pupil, 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 pupil during the interaction without communicating that assessment;
and the pupil may be secretly daydreaming about his favorite football team.

32
 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
consequence 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.

33
 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.
 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”, “quite 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 años (‘I have n years’); other languages have yet other constructions. The
question 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 quite conventionalized – in English,
after all, we could say “I have n years”, but we happen not to.

34
 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.
 Bosque (2010) discusses how the Spanish lexicon does not make the
distinction between shame and embarrassment, using only the word
vergüenza for both, and how (in our translation) “the best dictionaries
tell us that Spanish cariño is love, affection, fondness and tenderness,
equivalents 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 cariño and the verbs amar and querer in
Spanish; yet no Spanish-speaker would fail to make a distinction
between them.
 Bosque 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 because each culture’s emotional
life has been shaped by the social history of the speech community.

35
 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.
 For instance, when I say the following to my husband, I normally mean
something more than what I say. My statement works in fact as a request:

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?

36
 Indirectness can also be unintentional (e.g. 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).

37
 In conclusion, a great deal of the meanings we want to
communicate are not “said” literally, and therefore have to
be reached by the hearer(s) through some kind of cognitive
process, namely an inference of some sort. And precisely,
what makes it possible for the hearer to make these
inferences – apart from his cognitive capacity to process
information – is his pragmatic knowledge of the whole
discursive situation, i.e., his interlocutor’s identities, 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.

38
Conventionalization: Some constructions which were originally
indirect –and therefore in the past required some processing
cognitive effort on the part of the hearer to be understood– do not
require such effort any longer because the indirect meaning
intended has been conventionalized, i.e., it has become an
accepted and conventional way of expressing the indirect
meaning in question.

E.g.: Could you open the door, please?

Here the speaker is not asking about the ability of the hearer to
open the door, but just requesting him to open the door as a favor
to the speaker.

39
More examples  short-circuited implicatures in some ironic/sarcastic
utterances:
1)
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)

2)
If he is Mr. Universe, I’m the King of France. (= I don’t agree with you or
with the proposition expressed by your utterance = NO)

3) A: Do you think our government is corrupt?

B: Is the Pope Catholic? (= The answer to your question is obvious: YES)

40
Grammaticalization:
 The process of conventionalization can even go further, to the point that the original
indirect meaning becomes deeply ingrained in the grammar of the language.
 Invariability and incompleteness (or 'reduction') are typical signs of what is called
grammaticalization, 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.
 For instance, in English, please is used as a basic pragmatic marker to preface or
follow a polite request (a & b), but can also occur inside a clause, even a
subordinate clause (c & d):

a) Please, tidy up this mess.

b) Tidy up this mess, please.
c) Will you please tidy up this mess?
d) He asked me to please tidy up the mess.

Here 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 Shakespeare's
Two Gentlemen of Verona, for example) has been worn down to the single word
please.
41
 Grammaticalization affects many different areas of the language, not just
pragmatic markers.
 Examples of other grammaticalized units are How about ...? for suggestions,
Let's ... for proposals or You('d) better for recommendations.
 In all cases, grammaticalization involves the going beyond conventionalization
in the sense that inference is no longer involved: what was historically an
indirect meaning has become the new direct meaning.

42
 Pragmatic studies cannot be independent from cultural studies, given
the fact that pragmatic meanings are always 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
tends to be less ground for misunderstanding or pragmatic misfires,
considering that all participants generally share the same language
and the same modes of interaction. 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
questions about how 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 and b) Intercultural

43
Communication involving
interactants who share a common
culture.

A typical situation is when

compatriots communicate in their
own language, making use of each
other’s knowledge of their shared
culture as a contextual resource.

E.g.: A session in the British

Parliament.
44
Communication where a non-
native member interacts with a
native member of a particular
culture.

A typical situation is when

someone travels to another
country and communicates with
native members of the local
culture in their language.

45
Communication when the interactants
do not share a common culture.

A typical situation is when someone

from one culture meets someone from
another culture on neutral ground and
communicates in a lingua franca.

46
Any communication that isn’t intracultural.
This is an umbrella term invented by
Grundy for those situations in which we
don’t need to discriminate between the
issues that arise in cross-cultural and
intercultural situations. This term is also
useful because the distinction between
cross-cultural and intercultural
communication isn’t always easy to draw.
For instance, in a conference in the
United States where there are Brazilian
and Chinese people, describing the
ensuing three-way communication as
trans-cultural saves us attempting neat
distinctions that it’s hard to be certain
about.

47
http://www.youtube.com/watch?v=ryqNQnzAmBs (Intercultural Team
Experiencing a Culture Clash in the Workplace)

http://www.youtube.com/watch?v=haohj1sVnyk&feature=related (Cultural
Gaffes Beyond Your Borders)

http://www.youtube.com/watch?v=WcEfzHB08QE&feature=related
(Sociology. Cultural differences)

And on the humorous side…

http://www.youtube.com/watch?v=f5FFZ5KQiYg&feature=related (Just for
laughs – Cultural shock)

48
Comparative pragmatic studies:

 Comparing pragmatic systems is no easy task. 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. (We’ll see this in
detail in Unit 4).
 Grice advocates for the existence of four main maxims (Quality,
Quantity, Manner and Relevance) of conversation that are
universally followed by any speaker of any language, and Brown
& Levinson argue in favor of the universality of their model of
politeness, which is based on the abstract notion of face.
49
 But contrary to Grice or Brown & Levinson, there are other 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 consequently
cannot be used as the basis for any kind of comparison.

50
 Coulmas (1981) suggests that most conversational language is predictable, in
the sense that our ability to determine how and what to say 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’.

51
 Scollon, Wong Scollon & Jones (2012) point to the fact that what people refer
to as ‘cultures’ (be it the Spanish, the Dutch, the English or any other
culture) may differ significantly from each other, 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).

52
 Discourse systems are thus complex systems which interact with one
another and in which each speaker participates along 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 question, 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, in favour of interdiscourse system communication (2012:
278).

53
 The purpose of this unit has been to offer a first impression of Pragmatics.
We have seen how this branch of linguistics can be approached both from 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 say that these meanings are underdetermined.
 We have also seen the fundamental role of context, and it is here that the
breadth of Pragmatics shows itself: so many different levels and types of
context conspire in complex ways to invest our words with a richer meaning
than they themselves can carry. This has led to the realization that meanings
are interwoven with the nature of human cognition and with the social
groups and/or cultures in which we live and have our being. Pragmatics thus
cannot but overlap with the cognitive and social sciences.

54
Corresponding to Chapter 2 (Pragmatics: 1
Definition and Scope) of base book:

Alba-Juez, Laura & Mackenzie, J. Lachlan (2016).

Pragmatics: Cognition, Context and Culture.

Madrid: McGraw Hill

 Words are not just words: As speakers of a human language, we all know
that our words can praise, hurt, convince or disappoint, etc.
 E.g.: 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.
 Consider this other example. Depending on (here mainly) the emotional
context, A may be doing different things with the same utterance (You’re
different), irrespective of the dictionary meaning of the word different:

A: You’re different.
B: Is that a compliment?

2
 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. Put simply, his main claim was that we do not just say things, but we
DO things with words.

3
 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
stream 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.
 Logical positivists maintained that the only meaningful statements are those
that can be tested empirically, and they were consequently and mainly
interested in the properties of sentences which could be evaluated in terms of
truth or falsity.

4
 Within this frame of mind, the only meaningful sentences would be those that
can be assessed as true or false, such as (a), and other sentences such as John’s
utterance in (b) would be meaningless because, strictly and logically speaking, it
is not true that the speaker is “all ears”.

(a) The Nile River is in Africa.

(b) Peter: I want to tell you something.
John: I’m all ears.

 In opposition to this view, Austin claimed that a sentence like John’s in (b) IS
meaningful, and that people are normally able to communicate without major
problems by using language just as it is, thereby calling the attention of scholars
to the idea that language is not imperfect or defective, but on the contrary, is a
magnificent tool whose functioning should be studied in order to understand how
it is that humans manage to communicate so well by using it. Indeed,
Austin’s(1962) first observation was that most utterances have no truth
conditions, an observation that emerged from his analysis of the behavior of what
he termed performative acts.

5
 Austin pointed to the fact that certain types of utterances, such as (c) and (d),
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.
(c) I pronounce you man and wife
(uttered by a priest at a wedding ceremony).

(d) We declare war upon the Empire of Japan

(uttered by U.S. President Roosevelt, on December 8, 1941).

6
 In order to identify a performative act, it is useful to test whether the adverb
hereby can be inserted between the subject and the verb. If it can, then it is
most likely that the utterance in question is an example of a performative act.
Compare (f) and (g), 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 sneeze is not, considering that inserting hereby before this verb would
be ungrammatical and inappropriate:
(f) I hereby certify that this document is an authentic copy of the original
diploma.

(g) *I hereby sneeze.

7
 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 (d), (e) and (f)
constitute a sub-type (declarations, according to Searle’s 1975 later
classification) of what he called illocutionary acts.

TYPES OF PERFORMATIVES
 All performatives have common characteristics. They are self-verifying
(containing their own truth conditions), self-referential (the verb refers to the
action the speaker is doing) and non-falsifiable (they can never be untrue).
 It is worth clarifying here that what can be self-verified is that the action
named by the performative is the action being uttered, e.g., in the utterance I
say that Tom is innocent, what is self-verifying is that the speaker says it,
NOT that Tom is innocent. This last statement (“Tom is innocent”), however,
is subject to truth conditions and thus not self-verifying.

8
 Thomas (1995) observes that not all
performatives are the same in nature, and she
classifies them into four types:

1) Metalinguistic
2) Ritual
3) Collaborative
4) Group

9
 Metalinguistic performatives are the most direct and clear kind of
performatives, because the verb in itself is very explicit as to what type of
action the speaker is performing by means of her utterance. The following
examples illustrate this type:

I apologize for having been disrespectful to you.

I object to his contract being renewed.

 Compare them now to their non-performative counterparts (which do not

contain the corresponding performative verb, and are therefore not self-
verifying, but subject to truth conditions):

I’m sorry for having been disrespectful to you.

I do not want his contract to be renewed.

10
 The difference between the pairs of utterances above, however, 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:

Jane (to her boyfriend): I’d like to go to the movies tomorrow.

Leonard: O.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):

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!

(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 or try to get out of delicate
discourse situations ‘unscathed’).

11
 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. The following are
prototypical examples of ritual performatives:

I name this ship The Conqueror.

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

 Ritual performatives are more dependent on felicity conditions than any other
type of performative.

12
 Collaborative performatives are the kind of performatives that require the
collaboration of another person for its success to be guaranteed. So in the
following examples both the bet and the challenge will only be successful if
the other person accepts the bet and the challenge, respectively.

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

13
 Group performatives are the type of performative speech acts that are normally
uttered by more than one person, and they only succeed if they are performed on
behalf of all the people involved. Group performatives may also belong to any of
the three other categories (metalinguistic, ritual or collaborative), so the following
combinations are possible:
 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 Tranquility, 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.
 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.
It can even be argued that a given performative belongs to three or even all four of
the categories, such as the act of bequeathing, which is collaborative because in
order for it to be successful in legal terms, the children will have to accept the
goods that are bequeathed to them by their 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
bequeathing implies some rituals without which it could not be ‘happily’
performed.
14
 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, and therefore the act would not be accomplished.

15
 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 consequential
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 subsequently.

16
 When one or more of these conditions are not met, 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 therefore the purported act is disallowed (e.g.: an unauthorized
person saying the words of a marriage ceremony).
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
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 question. Prototypical examples of
abuses would be insincere promises, fake apologies or congratulations,
unfulfilled invitations, etc.

17
 But of course there are some issues in pragmatics
which are hard to determine (how do I know if someone
is really sorry??). This fact has not discouraged
pragmaticians in the least. On the contrary, it has
generated further research and work into realms which
had previously been thought of as unconquerable. One
of these realms, in which pragmaticians specialize, is
the realm of implicit meanings.

18
 In his lectures, Austin warns his audience that there is an internal
evolution to his argument, so that what is proposed at the
beginning will be rejected by the end: On the one hand, what is
viewed at the beginning as a special class of sentences with unique
syntactic and pragmatic properties (performatives) is presented at
the end as a general class of act that includes both explicit and
implicit performatives (practically any other kind of utterance).

 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.

19
 Compare the following pairs of examples, where (a) is the explicit performative
and (b) is a possible implicit counterpart:

(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)

(a) I order you to leave this room immediately.

(b) Leave this room immediately. (Device used to indicate it is an order:
Imperative mood)

20
 It should be noted, however, that 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 following
utterance could be a threat, a promise or a complaint (among other things):

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.
 As already mentioned, the development of Austin’s argument led him to the
conclusion that there is no substantial distinction between constatives and
performatives of any kind, be they explicit or implicit. He therefore concluded that
they should all be included within the broader category of illocutionary acts.

21
 A relevant observation within Speech Act Theory is that utterances not only
have sense and reference, but also force. For instance, in (a) Lucy is not only
referring to her dislikes but also rejecting Peter’s invitation to get on one. This
rejection constitutes the force of the utterance and is what Austin called
illocutionary force. Likewise, Peter’s words are not limited to their sense or to
what they refer to; they also have the illocutionary force of an invitation.
(a) (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. But
this is not the only kind of act that is performed when language is put to use.

22
 Austin noticed that there are three main kinds of acts that are
characteristic of all utterances (which include both his original
performatives and constatives):

1) Locutionary acts  the acts involved in the construction of the

utterance, such as choosing certain words and pronouncing
them, as well as arranging these words according to the
grammatical rules of, and with the sense and reference
determined by, a particular language.
2) Illocutionary acts  the acts done or performed in speaking.
They convey the force or intention behind the words, and for
that reason they are the central type of acts in Austin’s theory.
3) Perlocutionary acts  the by-products of illocutionary acts,
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.

23
 Example:
(John and Susan have just finished dinner)

John: Would you help me with the dishes?

Susan: Sure! No problem.

Locution  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.

Illocution  Request

Perlocution  To convince or persuade Susan to help him

(Other prototypical examples of perlocutionary acts could be: comforting,

inspiring, inciting or humiliating.)

24
When John Malcovich (God) in the following
commercial
(http://www.youtube.com/watch?v=EzgGeZS2
BrE&feature=related ) says to George Clooney:

“Heaven can wait, George, but not for its

capsules”

God’s words may look as an assertion, but if

you read between the lines in order to
understand the humor of the commercial, you
realize that in fact, what God is doing with his
words is to threaten George.

So in this example:
25
1) The locutionary act is the act of
uttering the words “Heaven can
wait, George, but…”, which have a
given sense and reference.

2) The illocutionary act is a threat

(i.e. “If you don’t give me the
capsules, heaven will not wait for
you to visit us up there” (and
consequently we’ll let you die)

1) The perlocutionary act performed

by God is obliging George to give
him the Nespresso capsules (in order
to ‘postpone his visit to heaven’).

26
 A way of testing whether an act is illocutionary or perlocutionary is to use the
following frame:

Speaker PERLOC Hearer by ILLOC.

 Applying this to the above example and its possible alternative, we may get
John convinced/angered Susan by requesting that she help him with the
dishes; which confirms that convincing and angering are perlocutions while
requesting is an illocution.
 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 formula, while perlocutions are not;
i.e. in our example John could make his illocution explicit by saying I hereby
request you to help me with the dishes, but he could not do so by saying I
hereby convince you by asking you to help me.

27
 English grammar distinguishes a number of moods of the verb, each of which
roughly aligns with a different broad group of illocutions. The moods of
English are three in number:

1) The imperative mood is associated with a range of illocutions that involve

getting the addressee to do something (or stop doing something). The
illocutions may range from a recommendation (as in a) through a request
(as in b) to an order (as in c):

a. Change your password regularly.

b. Pass me the salad.
c. Get out of this room.

28
 The subjunctive mood is not applied very much anymore but when it is
used the illocutionary force involves the expression of a wish, a hope or a
prayer, as in d and e:

d. God save the Queen.

e. Thy will be done.

 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.

29
 MOOD CLAUSE TYPE ASSOCIATED EXAMPLE
ILLOCUTION
INDICATIVE Declarative Asserting, stating,… John has written a
fine novel.
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, Please come home
requesting… soon.

30
 Form does not always map so neatly onto function. Many of the moods and
clause types discussed above can at times be used to convey an illocution
that is basically associated with another mood or clause type. The indicative
mood, for example, can be used to express the illocutions normally associated
with the imperative mood, as in the declarative (a) or the interrogative (b),
which are more polite ways of saying ‘Tie your shoelaces’ and ‘Pass me the
salad’, in the imperative mood:
(a) Your shoe laces need tying.
(b) Can you pass me the salad?

 With a rising intonation, a declarative clause type can be interpreted as an

interrogative, albeit one in which the speaker not only asks a question but
also indicates some skepticism, as in (c):

(c) John has written a novel?

31
 Especially in conversation, the relation between the literal meaning of an
utterance and its illocutionary force is often quite 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.

E.g.: 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 question at all but
reproving Charlene for employing an inexperienced person.

32
 The conclusion here 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.

33
 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 as p, and the latter (i.e. those
features that determine the literal illocutionary force, which he called
Illocutionary Force Indicating Devices – IFIDs) as F. This he represented with
the formula F(p), and it is prototypically illustrated by the following example:

I promise I will come F(p)

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

34
 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, these are two of the felicity conditions
that the act of promising must meet:

1) The promise (Pr) is to be uttered only if S intends to do A.

2) The utterance of Pr counts as the undertaking of an obligation
to do A.

35
 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.
 According to Searle, there are five main kinds of speech acts:

1) Representatives
2) Directives
3) Commissives
4) Expressives
5) Declarations

36
1) Representatives:
Acts which commit
the speaker to the
truth of the expressed
proposition

E.g.: Assert, conclude,

boast, complain, deduce,
recite…
37
2) Directives: Acts which
reflect the attempt by the
speaker to get the hearer
to do something (in
varying degrees).

E.g. Ask, order, command,

request, beg, forbid, instruct,
urge, warn, plead, pray,
entreat, invite, permit, advise,
dare, defy, challenge… 38
3) Commissives: Acts whose
point is to commit the
speaker (in varying degrees)
to some future course of
action.

E.g.: Promise, threaten, offer,

guarantee, pledge, vow,
undertake, warrant, swear,
volunteer…

39
4) Expressives: Acts which
express a psychological state
specified in the sincerity
condition about a state of
affairs specified in the
propositional content.

E.g. Thank, congratulate,

deplore, apologize, condole,
detest, welcome, appreciate,
regret...
40
5) Declarations: Acts which bring
about immediate changes in the
institutional state of affairs and thus
tend to rely on extralinguistic
institutions.

E.g. Appoint, nominate, christen,

declare (e.g. war), excommunicate,
sentence (e.g. someone to death),
pronounce (e.g. a couple husband and
wife), resign…

41
http://www.rtve.es/alacarta/videos/especiales-
informativos/juramento-del-rey-felipe-vi-ante-cortes-
generales/2622781/ (Juramento del Rey Felipe VI ante las cortes
generales)

https://www.youtube.com/watch?v=yFOoNpVZhHg (Boat
christening)

http://www.youtube.com/watch?feature=endscreen&v=gTtqXtZdvjM
&NR=1 (Matt Cullinan and Jessica Weakley - I now pronounce you
husband and wife)
42
 But 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 commissives and
directives: the act of inviting in the following example is a
directive in the sense that it reflects Mary’s attempt to get Tom
to come to her house for dinner, but it can also be seen as a
commissive in the sense that Mary commits herself to preparing
some kind of dinner for Tom and receiving him in her house on
the agreed date and at the agreed time.

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.

43
 Examine the following question made by Leo:

Leo: Henry, could you open the door? (in a situation where Leo has his arms
loaded with boxes and Henry is near the door)

Imagine Henry’s answer to the question were:

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?

44
 It is obvious that Leo’s utterance is an indirect request, and not a question about
Henry’s ability to open the door.
 In his analysis of indirect speech acts, Searle introduced the notions of primary
and secondary speech (illocutionary) acts. For instance, in:

Lily: We should pay the gardener today.

William: We have no cash in the house.

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.

45
 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.

 Can you think of more examples of indirect speech acts?

Discuss them with your classmates in the forum for Unit 2.

 The topic of indirectness is key to understanding many

pragmatic phenomena such as implicatures, relevance or
politeness, all of which we will study in this course.

46
 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.

 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.

47
 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:

Sincerely, this is not a trick.

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.

48
 We saw above 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 question 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, quite
understandably, to be very widespread distinctions and may even be
universal.
 The remarks in the previous paragraph concern grammatical distinctions,
but there are also massive cultural distinctions in the use that is made of
speech acts:

49
 In Muslim cultures, although there are differences here between Shia and
Sunni Islam, for a man to announce “I divorce thee” three times (Hinchcliffe
2000) suffices to bring about the end of his marriage; clearly this has no
equivalent in other cultures in which divorce is not a performative verb.

50
 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 situation may trigger different speech acts in different cultures: in
English speaking cultures, one will apologize after stepping on someone’s
toes; in the West-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, he 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 were 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 requests. Argentinian Spanish speakers were found to be the most
direct, and Australian English speakers the most indirect!

51
 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
helped to inspire the communicative or functional-notional approach to language
teaching that arose in the 1970s. This approach stimulates the use of authentic
language in identifiable contexts of communication with particular functions in
human interaction and with specific notions in play.
 In psycholinguistics, efforts were 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 (like 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:
I’m trying to watch this program on TV.

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

52
 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:
Do you know when the next train to Canterbury leaves?

she is not inquiring 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’.

53
 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.

54
 Some of the limitations of classical speech act theory have to do
with the following facts:

It concentrates too much on IFIDs rather than on the speech

act in its entirety;
It relies too strongly on native-speaker intuitions; and it has
no developed theory of context.

 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.

55
 Thanks to Searle’s systematization of speech act theory, that
concept was to have a profound impact on how linguists now
understand language in context, with expressions such as
‘illocutionary 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.
56
Corresponding to Chapter 3 (Pragmatics: 1
Definition and Scope) of base book:

Alba-Juez, Laura & Mackenzie, J. Lachlan (2016).

Pragmatics: Cognition, Context and Culture.

Madrid: McGraw Hill

 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 mechanism makes crucial use of the speaker’s and the hearer’s shared
knowledge. Besides, reference is often achieved through the use of proper
names, which normally have a pragmatically determined conventional
association with particular people, places and objects known to a speech
community. All of this makes reference an essential pragmatic
phenomenon.

2
 Reference is not the same thing as denotation. It is an action
carried out by speakers and therefore alludes to the relationship
created by a speaker between words and specific entities.
 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”. Consider
utterances (1) and (2), both said by a hotel receptionist to a
workmate:

(1) Room 25 has to be cleaned by 12:00 p.m.

(2) Room 25 has complained about the noise at night.

3
 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 (nº 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. This is an instance of the so
called deferred (or metonymic) reference.

4
 Referential expressions are normally noun phrases, but verbs,
whether nominalized or not, can be also used as referents of
particular actions or activities, such as driving and to watch TV in
the following examples:

Driving is not my favourite thing.

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 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.
5
 In sum, the main point so far has been to note
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.

6
 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
dog/(s)he/it).
 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.

7
 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.
 When we refer to a particular individual, the referring expression constitutes
an instance of specific reference. When we refer to an object or person as
pertaining to a group or class, as shown by the italicized noun phrases below,
we are dealing with what we call generic reference:

A washing machine is a very useful appliance.

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

8
 Martin & Rose (2003: 169) explain that “whenever the identity of
a participant is presumed, that identity has to be recovered”,
and this recovery can be achieved in various ways, depending on
where the relevant information is. Sometimes the information is
found by looking backward in the text or discourse, as in the
following example. This is what we call anaphoric reference:

That’s my friend Suzanne. She is an actress.

9
 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 the
example below, where the food refers indirectly
backwards to a new Italian restaurant:

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

10
 Some other times we may have to look forward in the text in
order to recover the participant or thing being referred to in
which case we have an instance of cataphoric reference:

E.g.:

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

11
 In some cases, we may find the referent within the
same noun phrase. In the example below, the speaker
makes clear that the house she is referring to is the
one that is “in the woods”, thereby preventing the
hearer from having to track the referent somewhere
else in the text or context:

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.
12
 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.:
E.g.:
Michelangelo was considered the greatest artist of his time.

13
 In some cases, the reference looks out of the text but not to
famous people or obvious, shared cultural knowledge, as in the
example below, where the red motorbike looks outside the text,
to someone or something that can be perceived with the senses.
This is a case of so-called exophoric reference:

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. Here
the red motorbike is introduced for the first time, and therefore it
is not mentioned anywhere else in the text, but the interlocutors
can see the referent in their physical environment.

14
 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 the
examples below, where Mrs. Dalloway and Dick Gibson are presented by the
respective authors in the opening line of their respective novels. We have called
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.
Mrs. Dalloway said she would buy the flowers herself.
(Virginia Woolf, Mrs. Dalloway,1925)
When Dick Gibson was a little boy he was not Dick Gibson.
(Stanley Elkin, The Dick Gibson Show, 1971)
 In these cases, once the characters in question have been introduced, the
successive reference made to them becomes homophoric, for they are now part
of the authors’ and readers’ shared knowledge. Thus, the phenomenon of
reference is not a static one for the way of referring to the same participant or
entity may change according to the conditions and circumstances of the
ongoing discourse.
15
Reference as a discourse process:

 When reference is viewed as a discourse process, referring

sequences (not just isolated referring terms or expressions) may
become crucial for the analysis.

 Thus, when analyzing a text such as a conversation, a letter, an

e-mail message or a recipe, it is important to look at 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.

16
 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) require 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.

17
 The use of indexicals in discourse is one of the processes that
clearly shows the relationship between language and context,
because utterances like the following 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.

I’ll phone you next week.

You and you, come here now!

 In effect, the referents of deictic expressions shift depending on the

contextual conditions; therefore, the you of the above examples is not going
to be the same person in all its possible interpretations. It will, of course,
depend on who the speaker and addressee are in each one of its possible
contexts.

18
 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.

19
 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 2006):

4) Discourse and
5) Social deixis.

20
 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.

 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.

21
 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 proximal (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 the medial place deixis, encoded in the demonstrative ese,
the proximal and distal ones being este and aquel, respectively.
 In both English and Spanish, verbs such as come and go (Sp. venir
and ir) or bring and take (Sp. traer and llevar) encode spatial motion to
or from the deictic center, and thus are prototypical examples of
place-indexical expressions.

22
 Discourse deixis is observed when certain expressions (e.g. this (situation), as
pointed out before, in the previous chapter, in the next section, etc.) are used to refer
to some portion of the preceding or forthcoming 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, 2006: 119).
Thus, the demonstrative That in (a) is discourse deictic (because it refers back to
the whole of David’s utterance), while the one in (b) is just an example of
anaphora, which is co-referential with the house and the one:

(a) David: Get ready, Mary. We’re off to the Maldives!

Mary: That’s what I’ve been waiting to hear for ages!!

(b) I like the house and that is the one I want to buy.

23
 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. tu/vous; tú/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.

24
 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);
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.

25
 Deictics are semantically deficient, and therefore many times, with all five kinds
of deixis, gestures are necessary to help us identify the referent. This is what we
call gestural deixis:

Teacher (pointing to Daniel and Karla)

You and you will come with me to the Headmaster’s Office!!

 But sometimes no gesture is required because the knowledge of the basic

parameters of the speech event are sufficient for the hearer to identify the
referent. In these cases we speak of a symbolic usage of deixis,
(Martha enters the dining room to find her husband standing on the table)
Martha: What are you doing???

26
 It is interesting to note that some deictic expressions can have non-deictic
usages. In some cases the expression is used in a generic, non-specific way,
as in:
With people like these, you never know how to react.

 Some uses of the demonstratives which are recognitional or empathetic

(Levinson, 2006: 108), are non-deictic too:

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 question)

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)

27
 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 aquel in Spanish)
has to do with immediate perceptions of the speaker’s
environment, specifically concerning the question 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.)

28
 Person deixis in many languages, especially in the third person, is historically
derived from spatial deixis: Spanish él and ella, for example, derive historically
from Latin illum and illam, meaning ‘that one (masc.)’ and ‘that one (fem.)’
respectively; English he can ultimately be traced back to the Proto-Indo-European
(PIE) *ki- (‘here’) and she to *so- (‘this, that’). As for the origins of the first person
pronoun I, these are lost in the mists of time, but notice how the deictic here is
still associated with the speaker:
Give it here! (= Give it to me).
Same here! (= So am I)
 Time deixis is also derived from spatial deixis. The present time is correlated with
the proximal deictic this and the past time with the distal deictic that; this is
clearly because the present time coincides with the deictic center and thus is
near, while the past is remote:
These days far too many young people are out of work.
In those times, the power of religion was much stronger in society.

29
 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 deictic this and the distal
deictic that take on discourse roles. The proximals are associated with
agreement, and the distals with disagreement. These are also examples of what
were originally spatial deictics being used in discourse deixis:

A: The company needs to avoid lay-offs.

B: Here I agree with you completely.

A: We need a new government immediately!

B: That’s typical of you – always blaming the government.

30
 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 Excelência” (‘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 obsequious usage, as in:

Did Sir enjoy his lunch?

where Sir is historically ‘lord’ (cf. Spanish Señor) and the third person is used (cf.
his lunch) to avoid the possible unwanted intimacy of your.

31
 The past and present tenses as well as the markers of future time are 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*, which is non-deictic since it
does not involve the deictic center but simply relates events to each other.
Consider the following examples:

(a) Martha has now finished her new novel.

(b) Martha had finished her new novel before she started on a new one.
(c) Martha will have finished her novel by the end of the year.

 In (a), the absolute tense is the present tense (as is shown by the form has and
the adverb now) but the relative perfect tense indicates that the moment of
finishing preceded the time identified as ‘now’.
 In (b), 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 (c), the same applies, but in the future.

*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
– analysed as ‘aspect’ rather than ‘tense’. This debate has no bearing on the matter discussed here.

32
 In sum:
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.

33
b) Song: England Skies (Shake Shake Go)

Listen to and read the lyrics of this song and try to spot
and classify the referents (deictic or non deictic) used. Then
discuss them in the virtual class with your classmates:

https://www.youtube.com/watch?v=JLyDCtoqra8

Enjoy!!
In both speaking and writing, our discourse contains many
implicit meanings that our interlocutors apprehend through the
mechanism of inference. In fact, linguistic communication could
not exist without the parallel cognitive process of making
inferences, which may be based on different types of meaning.

Types of inference:

Some inferences derive chiefly from contextual assumptions,

while others are mainly built into the structure of the linguistic
units that trigger them. The former are said to be pragmatic in
nature, while the latter are considered to be primarily semantic.

36
Examples:
Pragmatic inference  conversational implicature (Grice’s meaning ‘nn’)
To a friend, after the speaker discovers that his friend has been disloyal:
You are a fine friend indeed!
Conversational implicature: You are not a fine friend.

Logical, semantic inference  syllogism (Grices’s ‘natural’ meaning)

All men are mortal.
Socrates is a man.
Therefore, Socrates is mortal.

Both semantic and pragmatic inference  presupposition

I regret not having gone to visit my aunt before she died
Presupposition: I did not visit my aunt before she died

37
 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:

E.g.:
I regret not having accepted the invitation
I do not regret not having accepted the invitation
Presupposition  I did not accept the invitation

38
 However, this definition presents some problems, because it fails to
describe the presuppositions of some sentences such as the following,
whose apparent negation seems to presuppose just the opposite:

Why did you buy that car? (Presupposition  You bought that car)
Why didn’t you buy that car? (Presupposition  You did not buy
that car)

 This and other issues have led scholars to reformulate the definition
in pragmatic terms, 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.

39
There are three main kinds of presuppositions
that are considered to be representative of the
phenomenon:

1) Existential

2) Factive

3) Connotative
40
Existential presuppositions are inferences made in
relation to the existence of people or things normally
described in definite terms. Thus the utterance below
presupposes that Peter exists, and that he has a brand
new Mercedes Benz car:

I went for a ride in Peter’s brand new Mercedes.

41
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/amazed that, etc., or the subject
complements of mean, prove, be obvious/ fortunate, etc., as shown in
the utterances below, 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:

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 obvious.
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.

42
Inchoative* verbs as in (a), cleft constructions as in (b), wh- questions
as in (c), adverbial and relative clauses introduced by the corresponding
subordinators as in (d) are also associated with factive presuppositions
(FP):

a. Fred discovered that his friend had lied to him. (FP: Fred’s friend
lied to him)
b. It was in London that we went to see Tom Jones (FP: we went to see
Tom Jones)
c. Why did you call so late at night? (FP: You called late at night)
d. When he recited the poem, everyone started to cry. (FP: He recited a
poem)

(*Inchoative or ‘inceptive’ verbs are those verbs which show a process of beginning or
becoming.)

43
Presuppositions can also be associated with counterfactive verbs
such as pretend or wish in (a) below or counterfactual
conditionals, as shown in (b), in which case the complement of
the verb and the proposition expressed by the if-clause
(respectively) are presupposed NOT to be true:

(a) 1. He pretended that he was German. (FP: He is NOT German)

2. I wish I had a house in Miami. (FP: I do NOT have a house
in Miami)

(b) If I had obeyed my mother, I would be happier now. (FP: I did

NOT obey my mother)

44
 Connotative presuppositions (CP 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 ‘B is bad’, both of which may or may
not be presupposed, depending on the verb:

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)

45
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 below, are also the triggers of certain
connotative presuppositions (CP):
a) My neighbour is playing his bagpipes again. (CP: My neighbour 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).
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).
f) The regime in that country has to come to an end. (CP: The leader of that
country is a dictator)

46
Notice the following:

• In (a) above we again see that connotative presupposition is variable:

after all, to a lover of bagpipe music, the presupposition could be quite
different.
• In (c) 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!
• In some cases the same expression can be the trigger for different
kinds of presupposition at the same time. For instance, in (f), 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.
• All is not used when the set has only two members. It would be
ungrammatical to say something like “All my ears are big”.

47
Presuppositions are sparked off by the use of various grammatical
constructions. The specific linguistic items that have this effect are called
presupposition triggers.
E.g.: Every time you use a definite noun phrase 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.
One frequently mentioned kind of grammatical construction serving as a
presupposition trigger is the factive predicate, 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 (a), 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.

48
Other kinds of presupposition trigger are:
• The ‘implicative predicates’ manage and fail, which presuppose effort on
the part of the subject, differing only in whether success was achieved or not.
In the following examples, the presupposed content is that Henry tried to get
the car to start:
Henry managed to get the car to start.
Henry failed to get the car to start.

 Cleft sentences. These have the form It + be + focus + presupposition, as in:

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’ (Cecilia, in this case).

49
Notice the following:
 Certain emotion predicates that are followed by a complement clause do not
involve a presupposition:
Elizabeth feared that Fritz was too clever for her.
The speaker here does not presuppose that Fritz was too clever for Elizabeth;
in fact, she takes no position on the question. We therefore say that fear is a
nonfactive predicate.
Not only emotion predicates can be nonfactive in this sense. Verbs of cognition
and communication like think, believe, say, state, claim and many more share
this property. In the following utterance, for example, the speaker reports
Guy’s belief without presupposing that it is true:
Guy believed that Hyacinth was a vampire.

 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 below, where the speaker presupposes that the little girl’s
bedroom was in fact not a castle:
The little girl imagined that her bedroom was a castle.

50
 One pragmatic use that is made of presuppositions is to make a request more
polite, by presupposing that it has already been complied with. Thank you for is
a factive presupposition trigger, as in:
Thank you for believing in me.
where the speaker presupposes that the addressee believed in her.
 Another pragmatic use is found in academic writing. Frequently 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:
a) Jones (2008: 123) observes/shows/proves that Binding Theory has a fatal
flaw.
b) Jones (2008: 123) claims/states/argues/assumes that Binding Theory has a
fatal flaw.

51
 In any of the variants of (a), the speaker presupposes that Jones is
correct; in any of the variants of (b), she makes no presupposition
about the correctness of Jones’s position.
 Barristers tend to be skilled at exploiting presuppositions, as in the
famous ‘loaded question’ example:
So, tell the court, when did you stop beating your wife?

 In sum:
We have seen here 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.
52
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.
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 Implicature, 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.

53
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: 45-46):

The cooperative principle:

Make your contribution such as is required, at the stage at which it

occurs, by the accepted purpose or direction of the talk exchange in
which you are engaged.

54
Maxims of the Cooperative Principle:

The maxim of Quantity

1. Make your contribution as informative as is required (for the current purposes
of the exchange).
2. Do not make your contribution more informative than is required.

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 adequate 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.
55
Speakers do not always follow the Gricean maxims to the letter. Very
frequently 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 conversational implicature, which is not
coded in the language and has its basis in the rational shared assumptions of
the interlocutors.

56
MAXIM OF QUALITY:
Being ironic is one of the
prototypical strategies which flout
the Maxim of Quality. Consider
this utterance, said by a woman
to her friend after the friend said
something stupid:

What a clever idea!

by means of which the woman

implies that the idea was just the
opposite, i.e., rather stupid or
NOT clever at all.
Example:

A: I think Sam and Sarah

are not being honest with me,
and I believe dishonest people
are losers.
B: Well, I think Sarah is
honest.

By not including Sam in her

utterance, B may be
conversationally implicating
that Sam is dishonest, and that
consequently, Sam is a loser.
Example:

A: Have you ever met

Richard?

B: No, what’s he like?

A: Well, he’s not what one

would call ‘handsome’.

In her last utterance, A may be

implicating, by being a bit
obscure, that Richard is ugly or
not very good-looking.
Example:

A: Would you like to come to London

with me on Tuesday?

B: I have a medical appointment on

Tuesday.

B’s utterance does not exactly answer

A’s question by saying “yes” or “no”,
and it might seem irrelevant at first
sight, but any competent speaker of
English can work out the implicature
that B’s answer is negative, because
one would expect B not to miss his
medical appointment on Tuesday.
Flouting a maxim is not the only way of non-observance of the maxims of the
Cooperative Principle. Grice also pointed out that there are many occasions on
which speakers fail to observe the maxims even though they have no intention
of generating an implicature.
Grice writes about three other such types of non-observance:
1) Violating a maxim,
2) Opting out of observing a maxim, and
3) Infringing a maxim.
Thomas (1995: 72) points out that several writers since Grice have argued the
need for another category, namely

4) Suspending a maxim.

61
1) Violating a maxim:
Grice clearly defines ‘violation’ as the unostentatious 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 no 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 from flouting this maxim and being ironic, 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’. For instance, when in (a) B is not telling the complete truth:
(B bought a new motorbike with his parents’ money without telling them)
B’s father: Have you bought a bicycle without telling us?
B: No, I haven’t bought a bicycle, Dad.

62
2) 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 requires, but does
not wish to generate a false implicature or appear uncooperative.
Examples of this type of non-observance are frequently 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.

63
3) 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 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.

64
4) Suspending a maxim:

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 quantity of words does
not generate any conversational implicatures.

65
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 (a) and (b):
(a) He is poor but honest.
(b) Even teachers can afford one of these laptops.

By using the connector but, the speaker in (a) 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 (b) the meaning conventionally implicated by even tells us
that the speaker commits herself to the idea that teachers cannot afford
to buy expensive computers.

66
 Conventional implicatures, being more attached to the linguistic form
of the utterance than conversational implicatures, deal with non-
cancellable 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.

 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.

67
While conventional implicatures are mainly derived from the conventional
meanings 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):

 Cancellability/defeasibility: This means that the inference derived from the

flouting of the maxims can be cancelled or defeated by adding some premises.

 Nondetachability: By saying that conversational implicatures are

nondetachable, 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. In contrast to conversational implicatures, conventional implicatures
are always detachable from the content of the utterance.

68
 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 question. 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.

 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.

69
Levinson (2000: 15) adds the following features to Grice’s list, also taking into
consideration the works of Sadock (1978) and Horn (1991):
 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 (a) below, 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.
A: I think Robert and Eunice are very honest people.
B: Well, I think Robert is honest, yes.

 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 equivalent form should
carry the same standard implicatures. So, for instance, the utterance “I have
three nieces” carries the universal generalized implicature that ‘I have no more
than three nieces’ in any language into which this utterance can be directly
translated.
70
 Generalized and particularized conversational implicature

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 utters (a), 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 to
the speaker. Similarly, in any possible context, an utterance like (b)
will always tend to generate the implicature that the speaker cannot
run the mile in less than 4 minutes.

(a) I got into a car.

(b) I can run the mile in 4 minutes.

71
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 the
example below, where B’s flouting of the Relevance 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:

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)

72
The example below illustrates and tries to clarify 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 neighbours already have one (PCI), but on the
other, perhaps they don’t need one, for not all their neighbours have
installed one (GCI):

A: Do we really need to install a water regulator?

B: Some of our neighbours already have one.

GCI: Not all of our neighbours have installed a water regulator.

PCI: Perhaps we also need to install one, yes.
73
As to the differences between GCIs and conventional
implicatures (something you 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 that while GCIs are cancellable, nondetachable,
reinforceable and universal, conventional implicatures have none
of these properties.

74
75
But this model does not show that there are other kinds of
inference, such as those based on interactional politeness or
conversational organization that were later to be treated by other
authors (e.g. Leech 1983, Brown & Levinson 1987), nor does it
include the neo-Gricean interpretation of the theory of
implicature, which –among other things– points to the difficulties
found in making the distinction between what is said and what is
implicated, and within which not all of the Gricean maxims
receive equal attention, as we shall now explain

76
 Horn (1972, 1984, 2006) and Levinson (1995, 2000) have devoted much of their
work to a refinement of Grice’s theory of implicature, above all by exploring the
concepts of scalar implicature and generalized conversational implicature.
 Scalar (or Quantity) implicatures derive from utterances in which a scalar
term is used to suggest that the speaker had a reason for not using a more
informative or stronger term on the same scale. Many utterances display scalar
properties:
a. Some students do not like Professor James.
b. Sam should be coming back from Paris now.
In these three utterances, some and should are members of different scales. The
scale to which some belongs includes <all, many, some, few>, and using 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 (but not to its
right) being excluded from the meaning of (b) by scalar implicature.
77
 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 Q1 implicature, its meaning does not need to be lexicalized.
 The important contribution of Horn’s work in this respect was to show that
the Q1 Maxim (“Make your contribution as informative as is required”) 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 favours the idea that other phenomena might also be explained by
examining such pragmatic conversational principles.
 A Q1 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
required”. 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 required”.

78
 In later works, Horn (1984, 2006) 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 Q1 maxim, and the R
Principle is a reformulation of the Q2 maxim and the Relation and Manner
maxims, as reproduced below:
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. (Horn, 1984: 13).
 Example (a) shows an utterance from which the addressee will most probably
recover the Q inference that the woman in question was a stranger, or at least
not a person close to the speaker, since he didn’t say ‘my wife’, ‘my mother’, etc.
because it is assumed that he has said ‘as much as he can’:
(a) I met a woman at the supermarket yesterday.
 In (b) 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 R
Principle of ‘not saying more than you must’.
(b) There will be seven of us for lunch today.

79
 Another neo-Gricean author is Stephen Levinson who, within
his theory of preferred interpretation (1995, 2000), defends
the notion of generalized conversational implicature as an
essential explanatory notion for a great variety of linguistic facts,
such as those exhibiting the lexicalization constraints studied by
Horn and exemplified above in this section.
 Levinson 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 sentence-meaning nor to that
of speaker-meaning, for they are midway phenomena,
influencing both grammar and speaker-meaning.

80
Levinson identifies three types of meaning:

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

2) Utterance-type meaning (which is context-free and

inferential), and

3) Utterance-token meaning (which is context-sensitive and

inferential).

81
 The Theory of Preferred Interpretation concentrates 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.

 In contrast, an utterance token is an utterance that has a single inferred

interpretation which depends upon its context. Thus, particularized
conversational implicatures 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:

82
Q-principle (Q1)
Speaker’s maxim: Do not provide a statement that is informationally 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 clausal* Q-implicatures, as illustrated
in (a) and (b):
(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)

*Clausal implicatures are those derived “from contrasts between one expression that entails its embedded
sentence(s) and another that does not”. (Levinson, 2000: 76-77)

83
I- Principle (or Principle of Informativeness, based on Grice’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).
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 m-intended* point, unless the speaker has broken the
maxim of Minimization by using a marked or prolix expression. (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-implicatures tend: “I-implicatures
are inferences from the lack of further specification to the lack of need for it,
whereas Q-implicatures are inferences from the lack of informational richness to the
speaker’s inability to provide it” (2000: 116).

*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.

84
 Some examples exploiting the I-principle are found below, which show that I-
implicatures are based on principles of economy:

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)

85
 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 I-implicatures. Thus compare the M-implicatures below with their
corresponding I-Implicatures above:
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.)

86
 As you 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 equivalent’
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 to be predictive in the
sense that it can make predictions about presumptive meanings
at different levels of generalization (morphological, lexical, etc.).

87
George Carlin (1937-2008) was a comedian who liked to reflect
upon language use. One of his favorite topics was the use of
euphemisms. Listen to this part of an interview in which he
mentions the use of the euphemism “the golden years”:

http://www.youtube.com/watch?v=KITIt2O3Z8Q

There are many more video clips which you can watch where he
speaks about euphemisms in real life, which I invite you to
watch in order to reflect upon the workings of I-inferences, as
well as to have a lot of fun.
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.
89
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 kilometres 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 centre 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’. This will be the topic of our next learning unit.

90
END OF UNIT 3

91
Corresponding to Chapter 4 of base book: 1
Alba-Juez, Laura & Mackenzie, J. Lachlan (2016).
Pragmatics: Cognition, Context and Culture.
Madrid: McGraw Hill
 Speakers normally expect interlocutors not to oblige them to do
anything or to insult them  expectations of rationality and
tactfulness  Rules and strategies of social and personal interaction
that linguists have studied under the name of Politeness.
 But these expectations are sometimes not fulfilled: people may opt to
be impolite, simply because they feel unbalanced, annoyed or
attacked, or because the particular discourse situation requires the
speakers to disregard their interlocutors’ wishes and purposefully take
them out of their ‘comfort zone’.
 Example  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.

2
 Politeness is about putting ourselves into our interlocutor’s shoes, i.e.,
about 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 .
 Thephenomenon of (im)politeness is very much related to the
Cooperative Principle and its maxims , for it has to do with what we
human beings expect and consider to be rational or ‘civilized’ behaviour.
 Although (im)politeness as a general phenomenon has been claimed to
be universal by many linguists, these expectations will vary according to
the different cultures, communities, social groups or discourse systems
involved in the interaction: even within the same culture or language,
different groups or even different families may have different rules of
politeness to function within their inner circles.

3
 From all the above considerations it is easily deduced
that the phenomenon of (im)politeness is essentially a
pragmatic phenomenon, for it would be impossible to
even conceive of it without taking into account the
people and all the variables of the context around the
language used.

 CLARIFICATION:
When we speak of politeness in Linguistics we are not
speaking of just social ‘good manners’: Linguistic
politeness involves the management and handling of
many and various discourse-pragmatic variables, which
may make the difference between what is considered
appropriate and coherent language and what is not.

4
 Green’s (1989: 141) definition

“Strategies for maintaining or changing interpersonal relations”.

Grundy’s (2008: 187) definition:

“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’s (1983: 104) Principle of Politeness

“An important missing link between the Cooperative Principle and the problem of
how to relate sense to force”

5
 Watts (1989, 2003), distinguishes between:

• First-order politeness  the everyday sense of the term and the ways in
which it is evaluated and commented on by lay members of society
• Second-order politeness  the abstract characterization or scientific concept
which transcends the everyday notion of (im)polite behavior.

This distinction was later relabeled by Eelen (1999, 2001) as politeness1

and politeness2 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 politeness1).

6
 The main four approaches to (im)politeness identified at
the end of the 20th century are outlined by Fraser (1990)
as follows:

1) the social-norm view,

2) the conversational-contract view,
3) the conversational-maxim view,
4) the face-saving view.

7
 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, 2006a), Locher and 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).

8
1) Social-norm view  associated with the Prof. Bruce Fraser (BU)
understanding of politeness as “good manners”
(Jespersen 1965, Quirk et al 1985). 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.
2) Conversational-contract view  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 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, the
contract thus having a dynamic quality, which has
to be adjusted in accordance with the unfolding
conditions of the ongoing conversation.

9
3) Conversational-maxim view  based on
Grice’s Cooperative Principle and its
maxims, (Robin Lakoff 1973, Geoffrey
Leech 1983, 2014, Yueguo Gu, 1990, 1997.

4) Face-saving view  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.

10
5) Arndt & Janney’s (1979, 1983, 1985) perspective  a general model of
emotive interactional behavior: 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.
6) Relational perspective  a postmodernist, discoursal approach to
(im)politeness. Its forefather 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”. 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”.

11
7) Frame approach (Terkourafi 1999: 107)  defines politeness in terms of
“culture-specific ready-made patterns”. It looks into 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, requests 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.
8) Rapport management view (Spencer-Oatey, 2000)  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 of face and management of sociality rights.
Spencer-Oatey (2008) characterizes face 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.

12
 The conversational-maxim view of politeness has been one of
the two most influential perspectives on politeness. The authors
supporting this view take Grice’s Cooperative Principle (CP) and
its maxims 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 Unit 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.

13
 Robin Lakoff
Two Rules of Pragmatic Competence (Lakoff, 1973: 296):
1) Be clear
2) Be polite
These rules may reinforce each other, but at times may
come into conflict with each other: 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 A feel Good – Be friendly

Lakoff explains that each of these rules requires a different

type of politeness. Thus Formal Politeness is required if we
want to follow Rule 1, Informal Politeness is required for
Rule 2, and Intimate Politeness for Rule 3.
14
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) and therefore proposed a Tact
Maxim (1977, 1983) and a Politeness Principle
(PP) as complementary to the CP.
 The PP presents an idea of politeness in which,
other things being equal, people prefer to be polite
rather than impolite. In the same fashion as the
CP, and unlike grammar rules, the PP can be
observed, breached, suspended or flouted, and
can also be classified into maxims (Tact,
Generosity, Approbation, Modesty, Agreement and
Sympathy maxims) (Leech, 1983).

15
 One aspect that has been controversial in Leech’s (1983) model is the
fact that it distinguishes between
a) Relative Politeness  the politeness which varies according to
specific situations and,
b) Absolute Politeness  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’).

16
 Revised version of Leech’s model of politeness (2014)  responds to all the
criticisms directed at his model of 1983, and presents a more detailed and
refined account of his PP, which he now reformulates as the General Strategy
of Politeness (GSP), in response to the criticism (by B&L (1987) and others)
about the “expansionist” approach of his PP, which in their opinion, introduced
too many maxims.
 The GSP is a “single superconstraint” (2014: 90) which includes all the
maxims previously formulated in the 1983 model, which Leech defines in the
following way (where O refers to the 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 favorable value with what pertains to O or associates
an unfavorable value with what pertains to S (S= self, speaker).” (2014: 90)

17
Maxims (expressed in an Related pair of maxims Label for this maxim Typical speech –event type(s) EXAMPLE
imperative mood)

(M1) give a high value to O’s Generosity Commissives A woman to her friend: Let me pay for this meal. It’s
wants Generosity, Tact my treat!
(M2) give a low value to S’s Tact Directives A woman to her friend: If it would not bother you
wants too much, could I borrow your dress for the party?

(M3) give a high value to O’s Approbation Compliments Son to his mother: Your tiramisu is the best I have
qualities Approbation, Modesty tried in my whole life, Mom.

(M4) give a low value to S’s Modesty Self-devaluation A student to her teacher: I’m so stupid I don’t even
qualities understand how to calculate square roots!

(M5) give a high value to S’s Obligation (of S to O) Apologizing, thanking An employee to his boss: I’m terribly sorry, but I
obligation to O will have to ask you to leave early today because my
Obligation wife is sick and I have to take her to the doctor’s.

(M6) give a low value to O’s Obligation (of O to S) Responses to thanks and A: I’m very sorry to have bothered you.
obligation to S apologies B: No problem at all. It’s all right.

(M7) give a high value to O’s Agreement Agreeing, disagreeing A: Jane is a wonderful teacher, don’t you think?
opinions Opinion B: Yeah, she’s excellent!
(M8) give a low value to S’s Opinion Giving opinions A woman to her friend: My humble opinion is that
opinions reticence you should end that relationship, but of course I
may be completely wrong.

(M9) give a high value to O’s Sympathy Congratulating, commiserating A woman to her friend, who has won a scholarship
feelings Feeling to go abroad: Congratulations!!! I’m sooo happy for
you!!!!
(M10) give a low value to S’s Feeling reticence Suppressing feelings A to his girlfriend, after she bumped into a rock
feelings and fell down: Are you O.K.?
B: Yes, my ankle hurts a little bit but it’s nothing;
don’t worry.

18
 Two important aspects of the restatement of Leech’s theory in 2014:
a) 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.
Pragmalinguistic scale  refers to the fact that utterances can be ordered on a scale of
politeness while keeping 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 equal, (1) is
less polite than (2):
(1) Give me $10.
(2) Would you be so kind as to give me $10?
Sociopragmatic scale  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. E.g.: the otherwise
polite form Would you please…, as used in (3), will most surely be interpreted as
sarcastic in the context of a family conversation:
(3) Would you please let me speak for a while?
19
b) Leech’s revision presents a 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 B&L’s and his own model (e.g. by
Wierzbicka (2003)), by giving authentic examples from these
and other foreign languages. As to the related question of
whether his model is universal or not, his position is not very
different from B&L’s: although he does not press for “universal
principles, he still argues 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

20
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,
something that Leech did when he restated his model in 2014.
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 (in 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 Western civilization
and the collectivist ethos of the Eastern world.

21
 Whereas Lakoff and Leech see politeness as implying rules, principles
or maxims that are additional to Grice's Conversational 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.
 B&L 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.

22
 In B&L’s writings, face is used as a technical term and derives
from the work of Goffman (1967). It has to do with the public
image that all of us have as members of society
 In our social life we invest a lot of emotional effort in protecting
face and trying to improve it; if we lose face, we may feel
embarrassed or even humiliated and will suffer as a
consequence.
 In general, it is in everyone’s interests to respect the face of
other people. B&L (1987: 70) refer to this as ‘anointing’ the face
of the other. There are, however, cultural or group variations
here, some of them investing more effort in ‘facework’ than
others, but rationally speaking no society could survive without
its members attending to one another’s faces.

23
 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 B&L 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)

24
 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:

a) criticisms, ridicule and insult E.g. You’ve burned the toast again, you
blockhead!
b) challenge and disagreement E.g. You’re totally wrong about that.
c) violent displays of emotion E.g. I’m f*****g FURIOUS with you.
d) mention of sensitive subjects E.g. Did you ever consider gastric band
surgery?
e) bad news E.g. You’ve failed the exam.
f) boasting E.g. I’ve always succeeded in everything I
do.

25
 Some possible threats to the hearer’s negative face –given the appropriate
circumstances– are exemplified in the following set of examples:

a) orders, requests E.g. Give up your seat on the bus to elderly

passengers.
b) suggestions, advice E.g. You should not wear that color of lipstick.
c) reminders E.g. I told you the boss needs the report by 4
o’clock, remember?
d) threats, warnings, dares E.g. I dare you to go into that field with the bull
in it.
e) offers E.g. I’d like to make you a loan of $5000.
f) promises E.g. We promise to visit you next time we’re in
town.

26
 Notice that speech acts can also be threatening to the speaker’s
own negative or positive face. Some examples of the former are:

a) thanks E.g. I am very grateful for the box of

socks you sent me.
b) excuses E.g. I’m really sorry I parked in your
spot; I didn’t realize.
c) acceptance of offers E.g. I’m very glad to be getting this ride
home.
d) responses to a faux pas E.g. Don’t worry about it, you
weren’t to know I had cancer.

27
 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.

a) apologies E.g. I’m so sorry I spilled wine

over you new fitted carpet.
b) acceptance of a compliment E.g. Really, I just threw that cake
together, nothing to it.
c) loss of physical control E.g. I’d like to introduce you to
my [HICCUP] parents.
d) self-humiliation, acting stupid E.g. My talk will focus on the
human orgasm …err…
organism.

28
29
If S uses an on record strategy, there
is only one interpretation of his/her
intention and there is no room for
ambiguity. Bald on record strategies
are considered to be in conformity
with Grice’s Maxims and they are
used when maximum efficiency is
required. Consider these examples:
• Help!! (Urgent, desperate situation.
Compare to the non-urgent and non-
desperate Could you help me with the
washing-up, please? )
• Don’t move!! ( If S sees a Boa
Constrictor approaching H)
• Hands up!! (When the police find a
criminal)

30
If the situation is not desperate,
does not call for urgency or does
not require maximum efficiency,
S can still go on record but
with either positive or negative
politeness. Here are some
examples:

• What a beautiful sweater

you’re wearing! (On record with
positive politeness)
• Can you pass the salt? (On
record with negative politeness –
conventionally indirect request).

31
If S wants to do an FTA, but for some reason
wants to avoid the responsibility for doing it,
s/he will most probably go off record and
leave the interpretation to the addressee. By
going off record S will always be flouting one
or more of the Gricean Maxims.
E.g.:

• What a beautiful dress! = Please, buy me

that dress. (Off record strategy: Give hints,
which flouts the Relevance Maxim).

32
• John is a bit silly
(= John is very silly)

(Off record strategy:

Understate, which
flouts the Quantity
Maxim).

33
•John is a real genius (=
John is stupid)

(Off record strategy: Be

ironic, which flouts the
Quality Maxim).

34
• I’m going you-know-
where
(= I’m going to the toilet)

(Off record strategy:

Be vague, use
euphemisms, which
flouts the Manner
Maxim).

35
 B&L (1987: 60) explain that the more an act threatens S’s or H’s
face, the more S will want to choose a higher-numbered strategy,
because in their view, 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 scholars 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.

36
The sociological variables D, P & R
 B&L recognize three sociological) variables that impinge upon communication in
ways that are relevant for how speakers and hearers deal with threats to their face.
1) DISTANCE (D)  the social (relative) Distance between S and H (or between H
and S, which is the same thing). 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.
2) POWER (P)  ‘the relative Power of S and H’, which, in contrast with D (which is
symmetrical) represents an asymmetrical type of relationship. S or H can exert P
as a result of material or metaphysical control. E.g.:Doctors have 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.
3) RANKING OF IMPOSITION OF THE PARTICULAR CULTURE (R)  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. E.g.: 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.

37
 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 equation 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 S and H, the P of H over S and
the cultural/situational R of x. The equation is merely
indicative, since none of the variables can be given
numerical values.
38
 Some possible formulations for different settings of the
weightiness of an FTA:
a) D low, P low, R moderate (e.g. to S’s brother): Gimme
your cell for a couple of minutes!
b) 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?
c) 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 quick call
home?

39
 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
to discuss 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.

 For a detailed account of all the criticisms made by

different authors, go to pages 154-57 of the base book.

40
Watch the following videoclips of the British comedy series
“Yes, Minister”, and try to find examples of the different
politeness strategies (following B&L’s taxonomy). Pay
attention to the flouting of Grice’s maxims, and discuss it all
with your classmates in the virtual class:

http://www.youtube.com/watch?v=DGscoaUWW2M&feature
=fvst (Who reads the papers?)
http://www.youtube.com/watch?v=7xYI3M2098w&feature=r
elmfu (Complete confidence)

41
 In spite of the undeniable impact and influence of B&L’s Theory of Politeness,
very 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,
unconsciously, 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. This wider scope
allowed researchers to avoid dichotomizing the phenomenon in terms of two
poles (politeness vs. impoliteness) in favor of the consideration of a
continuum along which numerous types of behavior can be found, for which
different terms have been used (e.g. Watts’s (1989) politic behavior,
Terkourafi’s (2001b) unmarked politeness, or Kerbrat-Orecchioni’s (2011)
non-politeness).

42
 Jonathan Culpeper (e.g. 1996, 2005, 2008, 2010, 2011, 2012, 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 Soldier 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.

43
 CULPEPER’S APPROACH TO
IMPOLITENESS
Culpeper (1996) points out that some
areas of politeness are not well
represented in B&L’s politeness model,
and he therefore proposes an
impoliteness framework where each of
B&L’s politeness superstrategies has
its opposite impoliteness superstrategy
which attacks face instead of
enhancing or supporting it.

44
 Culpeper’s taxonomy includes the following strategies:

1) Bald on record impoliteness  The use of strategies with a

clear intention of attacking face in a direct way.
E.g. (The emotional context here is that of a heated argument,
and where B is very angry at her interlocutor)

A: I don’t understand why you’re so stubborn and don’t

wanna listen to me!
B: Shut up, you idiot!

45
2) Positive impoliteness  The use of strategies designed to damage the
addressee’s positive face wants.

E.g.: 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.

E.g.: To ask someone about his private life with the clear intention of
invading his private space and making him feel ashamed, or to scorn or
ridicule an interlocutor.

46
4) Sarcasm or mock politeness  the use of politeness strategies
that are obviously insincere . These strategies refer to the use of
mock politeness with the aim of provoking social disharmony.
E.g.: Sarcastic use of honorifics:
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!

47
5) Withhold politeness  the absence of politeness
work where it would be expected.
Prototypical example: when someone does not thank
somebody for a present and deliberately remains silent
in order to hurt and offend him.

48
Non verbal politeness: Culpeper also
points out that B&L’s politeness model does
not take into account some aspects such as
paralinguistic or non-verbal politeness.
E.g.: The bow that Japanese people make
when saluting others, shouting and
avoiding eye contact, 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.

49
 In a later publication, Culpeper provides the
following definition:
“Impoliteness comes about when: (1) the
speaker communicates face-attack
intentionally, or (2) the hearer perceives
and/or constructs behaviour 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 21st century, the
study of (im)politeness has mainly been
approached from the relational perspective,
taking into account Halliday’s (1978)
interpersonal metafunction.

50
 KAUL DE MARLANGEON’S APPROACH TO
IMPOLITENESS

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 way 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’s
taxonomy of politeness strategies may be the
base for a taxonomy of impoliteness
strategies, both taxonomies complementing
each other.
51
 According to Kaul de Marlangeon, impoliteness occurs in any of the
following eleven instances (2008a: 258-59 [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)

52
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) When H:
1) interprets S’s behaviour as an intentional face attack that
induces him/her to accept the attack or reject it through defence
or counter-attack. (10)
2) remains silent intentionally, in order to indicate
disagreement/ discontentment with S’s utterance. (11)
53
 Based on this “endecatomic” definition, she proposes a scale of impoliteness
acts, according to the growing intensity of the impoliteness force transmitted by
the acts in question. For the intended acts, the higher the number, the stronger
the impoliteness force, i.e., the higher the degree of impoliteness.

54
1) 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.

E.g.: the famous utterance ¡¡Viva México, cabrones!! (“Long live Mexico, you
bastards!!”), which is pragmatically used by Mexicans to 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.

55
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: E.g. 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:

Biden: 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=C2mzbuRgnI4 September 9, 2008. (in
Kaul de Marlangeon & Alba-Juez 2012: 79-80)

56
b. S’s involuntary stint on the politeness expected by H:
E.g.: Cases in which the speaker unintendedly falls short of politeness, as
in the following 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:

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)

57
c. Involuntary omission of politeness:

E.g.: when children or teenagers forget to thank adults, which

brings about the inevitable subsequent remark (in retrospective
awareness) by their mothers/fathers/caregivers: (Johnny) what
do you say (to Mrs X for having invited you to her home)?
(Everyday experience tells us that it is possible that the adult in
question 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).

58
3) Self-impoliteness acts  Acts by means of which people use impolite or rude
language toward themselves. Self-impoliteness may be a) authentic or b)
feigned.
E.g.: While in (a) the speaker is sincere about his feelings, in (b) he strategically
manipulates his message with the aim of eliciting a Face-Flattering Act (Kerbrat-
Orecchioni, 2004) from the hearer.

(a) Charles realizes he has made a big mistake in his Math exam, and sincerely
expresses:
Charles: Damn! What an idiot I am!!

(b) 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 certainly a smart student!

59
4) Formally polite acts with an impolite purpose  Here politeness forms are
paradoxically used as a means to achieve impoliteness.
E.g.: All ironic uses of polite forms of address, as well as some manifestations of
cynicism, as in Norton’s last utterance below:

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 questions?
Convict: When do we eat?
(Cued by Norton’s glance, Hadley steps up to the con and screams right in his face):
Hadley: 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):
Norton: Any more questions? (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 movie The Shawshank Redemption, in KdL & AJ, 2012: 82)
60
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.
E.g.: (Humphrey’s impoliteness is brought about by his reluctance to tell Bernard
his secret)

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)

61
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.
E.g.:

A: I think Mr. White is a wonderful teacher.

B: (Overwhelming silence).

The silence is here interpreted as a form of disagreement, equivalent

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.
62
7) Fustigation impoliteness acts  verbal
aggression in a confronting or challenging
situation, where the main and 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 (a) and (b) respectively:
(a)
Hadley: What’s your malfunction you fat fuckin’
barrel of monkey-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 you a lullaby!
(From The Shawshank Redemption, in KdL & AJ, 2012:
86)

63
(b)
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 you said she was thin.
(From Bridget Jones’s Diary, in KdL & AJ, 2012: 87)

64
 Most of the authors (e.g. Leech, Culpeper, Kaul de Marlangeon) who began
writing about (im)politeness at the end of the 20th century have continued to
do so in the 21st 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) 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.

65
 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, something in-between the two marked poles,
which, like polite behavior, is also appropriate.

66
 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: the unmarked relational 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.
 To conclude: 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 and 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.

67
END OF UNIT 4

68
Corresponding to Chapter 5 of base book: 1
Alba-Juez, Laura & Mackenzie, J. Lachlan (2016).
Pragmatics: Cognition, Context and Culture.
Madrid: McGraw Hill
 This unit focuses on the argument set out by Dan
Sperber and Deirdre Wilson in their book
Relevance: Communication and Cognition (1986)
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:

1) The claim that an essential feature of most

human communication is the expression and
recognition of intentions.
2) The claim that utterances automatically create
expectations which guide the hearer toward the
speaker’s meaning. As we saw in Units 3 and 4,
Grice described these expectations in terms of
the Cooperative Principle and its Maxims
(Quality, Quantity, Relation and Manner).

2
 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 Gricean
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.

3
 E.g.: 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
question 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 mom for that matter.
If, however, the same question were asked his mother by a complete stranger, she
would most probably lack the contextual information to interpret the question with
accuracy, and therefore the question would turn irrelevant – or at least, much less
relevant than in (1) –.

4
 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.
 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.

5
 More than one 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).

6
 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.

7
 There are two main factors that govern the degree of relevance of
an ostensive act: contextual effects and processing effort.

 S&W (1983: 118) explain that the notion of a contextual effect is

essential to a description of the 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 consequences of adding this assumption to a
set of assumptions that have themselves already been
processed” (1983: 118).

8
 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) cancelling 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.

9
 The less effort it takes to recover a fact, the
greater its relevance.
E.g.:
An officer on a train: Tickets, please!

Most relevant interpretation (= the one involving the

least processing effort given our background
knowledge and experience of what normally
happens on trains)  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)  The
officer wants to buy a ticket. (This would have no
connection with anything already known, and
therefore would not be worth processing).

10
 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 (e.g.: the
confirmation of a contextual implication as a true conclusion constitutes a
positive cognitive effect. False conclusions are also considered to be cognitive
effects, but not positive ones).
 Most important type of cognitive effect  contextual implication (“a
conclusion deducible from input and context together, but from neither input nor
context alone” (2007: 608). E.g.: Ann’s final utterance, 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):
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.m.
Ann: It’s 7:00 p.m. and she hasn’t called. I don’t think we should wait any longer,
then.

11
 But there are other types of cognitive effects, different from contextual
implications, such as the strengthening, revision or abandonment of available
assumptions (E.g.: 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.

 W&S (2007: 609) formulate the relevance of an input to an individual in the

following manner:
a. Other things being equal, 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 equal, the greater the processing effort expended, the lower
the relevance of the input to the individual at that time.

12
E.g.: Consider the following three 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):

(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) × 0 is not 0

 Most relevant response  (a), because it is the one with the greatest positive
cognitive effects and requires the least processing effort,
 Less relevant (b), because Myriam’s friend will have to work out an implicature in
order to conclude that Myriam is free on Friday,
 Least relevant  (c), 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 equation, and therefore it would require the greatest processing effort. The
three utterances will derive the same cognitive effects, but these are easier to derive
from (a) than from (b), and more difficult to derive from (c) than from (b).

13
So…
 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
Relevance Theory 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 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.

14
 Wilson (1994) distinguishes three types of communication:

1) Accidental (E.g.: 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),
2) Covert (the kind of communication where intended intentions
are backgrounded), and
3) Overt (that in which the intended intentions are
foregrounded).

15
 The act of simply showing something to someone is a case of
ostension, and this is true, too, of intentional communication.
 Ostensive communication  communicative behaviour 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.

16
 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.

 So 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).

17
 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.

 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)

18
EXPLICATURE AND IMPLICATURE
Explicature  an enrichment of the
information that is explicitly encoded in an
utterance to a fully elaborated propositional
form.
E.g.:
Inspector: Tickets, please!  Could you show
me your tickets, please?

Children on Halloween day: Trick or treat? 

Do you want a trick (on our part) or do you
prefer to give us a treat instead?

19
 Cruse (2004) illustrates four aspects of explicature in which the Principle of
Relevance guides the recourse to inference:

1) Disambiguation: We normally do not notice ambiguities in language

because the context and the Relevance Principle guide the process of
disambiguation. E.g.:

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?

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.

20
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 the following example, 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:
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:

So, how was it??!! Was it too difficult?

21
3) Enrichment: The fleshing out of 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. E.g.:
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?]

22
b) the resolution of semantic incompleteness.
E.g.:

(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.

23
4) Higher-order explicatures: this kind of explicature has to do with the
speech act of the utterance in question. In Relevance Theory 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 (Cruse, 2004: 387). 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. E.g.:
(A teacher to her student, when she sees that the student (John) is upset
about something his classmate Tony has said to him, and looks like he wants
to hit him):
Think before you act, John!
Higher order explicature (made by the student)  The teacher is reprimanding
me/ or giving me advice (depending on the context).

24
Implicature 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 the example below,
where Jeremy wants to tease his girlfriend Paula, and therefore replies
to her question in an ironical way:

Paula: Do you love me?

Jeremy (smiling and hugging her): No, I don’t!

Explicature  No, I do not love you (reconstructed by means of

enrichment),
Implicature  Yes, of course I love you. How could you even ask such
a thing?

25
 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).

 From the relevance-theoretic viewpoint, implicatures may be of

two kinds:
1) implicated premises
2) implicated conclusions

26
Example:
Jenny: Have you read Fahrenheit 451?
Robert: I don’t like science fiction books.

Implicated premise  Fahrenheit 451 is a science

fiction book.
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.

27
 The distinction between conceptual and procedural meaning is one of the
most influential notions to have emerged from Relevance Theory.
 Blakemore (1987, 2002) 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 affect the conditions under which an utterance is true)
and others encode procedures (and they do not affect the truth conditions of
utterances).
 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.

28
 But procedural meaning is not only found in connectives.
Further research on the topic 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 and 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 utterance like I like you).

29
 E.g.:
Thus, on the one hand, the pronoun in (a) 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 (b)) based on contextual
information and expectations of relevance (procedural meaning):

a. He is my father’s best friend.

b. Peter Brown is my father’s best friend.

30
 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
acquired.
 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 (1983), procedural
meanings are more difficult to translate than conceptual
meanings.

31
 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 the utterance below,
Wow Is a pragmatic marker which shows the surprise and admiration
of the speaker, but does not contribute to the propositional meaning
or to the truth conditions of the proposition expressed:
Wow, Andrew’s new car is a Ferrari Maranello!

32
 Relevance Theory 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.

E.g.: Modal verbs are typically analyzed as being polysemous. Consider

the three uses of the verb can below:
a. My three-year-old can already write her own name  dynamic
meaning, indicating ‘ability’;
b. If you have a ticket, you can now enter the auditorium  deontic
meaning, indicating ‘permission’;
c. It surely can’t rain on my birthday! epistemic meaning, indicating
‘possibility’.

33
But… how does H know which of the meanings is intended by the
speaker?
According to RT, the interpretation of can arises from H’s
understanding of the propositional meaning (in which can has an
underdetermined meaning) against the background of relevant
contextual assumptions.
In (a), 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’.
(See more examples in base book).

34
 We always make inferences of some kind to interpret what other speakers
say, and 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.
 So when we consider more than one possible explicature or implicature it
is the principle of salience that helps us choose the most relevant one.
E.g.:
(a) I need to buy some olive oil at the supermarket.
(b) I need to buy some baby oil at the supermarket.
In (a) 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 oil for olives
(because, among other things, our knowledge of the world tells us that olives
cannot eat or consume oil). However, in (b), the most salient explicature is that
what the speaker needs to buy is oil for babies, NOT an oil made from babies.

35
 But even though the notion of salience usually allows us to
interpret people’s utterances, human interaction is very
complex, and it is not always simple (as in the above example) 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 (E.g., when
someone interprets a comment as a criticism,
mistaking/misunderstanding the higher-order implicature
intended by the speaker).

36
 It frequently happens that we communicate by means of images, both
moving pictures like movies and videos and still pictures like
photographs. Now, Is RT is equally applicable to pictorial
communication?
 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 the one below 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:

37
 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. Consider only the pictorial
part of the above sign:

38
 In ordinary verbal communication, we know who the communicators are.
In the case of a sign like this, 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
question 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  the sign is not addressed only to people whose
heads float free from their shoulders and is not addressed to males only,
despite the fact that the human figure is 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.
 Higher-order explicature  “Use the exit near this picture” (Directive)

39
 So… Can pictures can trigger explicatures?
Forceville and Clark (2014) 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 “exit” one,
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, its
most relevant explicature 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.

40
 RT has been applied prolifically as a way of understanding the many
forms of 'overt and covert communication' that surround us in our daily
lives.
 E.g.: Jokes 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.
41
 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 Relevance Theory: for
example, the confident claim that it is always the most relevant
interpretation that hearers reach first is ripe for testing.

42
1) Choose any of the jokes told in any of the two following
YouTube videos and analyze it from the RT perspective.
Explain which interpretation would be the most relevant
(given its context), as well as the processes by means of
which the audience will arrive at such an interpretation
(explicature, implicatures, procedural/conceptual meaning,
etc.)

https://www.youtube.com/watch?v=bN_uU3wceNs

https://www.youtube.com/watch?v=TYiXyQxEuzo

43
2) Explain the relevance of this ad by analyzing its explicature.

44
END OF UNIT 5

45