Jason Cenoz.

Pauses and hesitation phenomena in second language production

Pauses and their function and the relationship between pauses and language proficiency.

Introduction

Pauses can indicate that the speaker is taking time to plan the following utterance(s), however,
this is not always true.
Pauses functions:
1 Physiological function to allow the speaker to breath.
2 A cognitive function to allow the speaker to plan their speech.
3 a communicative function, to help the listener to identify demarcations in the speech stream.

Apart from the physiological and cognitive functions, pauses can also have a communicative
function because they can be communicative devices used in oral interaction to stablish speakers’
turns.
Silent and filled pauses can be distinguished (Formal Criteria), silent pauses: silent periods
between vocalizations (including breath pauses), filled pause: interruptions od speech flow by
non-lexical sounds such as ah, mm, er, erm, uh, um.

(Functional criteria) different internal processes, filled pauses: reflect affective states sucjh as
anxiety. Silent pauses correspond to the cognitive difficulty of the task involved.
Others: filled pauses: floor-holding function (the speaker has more to say), debunked becauses
pauses happen in lectures.

Pauses are divided into juncture (or fluent) pauses and non-juncture (disfluent/hesitation) pauses.

Juncture: mark the boundaries between syntactic units (phrase, clause and sentence).
Non-juncture: they occur within the syntactic units and are abnormal for the speaker that’s talking.
They are likely to happen in second language speech because speakers have to search for the next
idea as well as the following linguistic element(s) in a limited proficiency language.

Distribution of speech and silence and hesitation occur differently among languages.

Here we focus on cognitive function and analyses pauses as related to other hesitation
phenomena. Pauses can mean difficulties in speech production (lexical, phonological, syntactic,
etc), or act as communication strategies used to solve difficulties, gain time or more.

Pauses and Hesitation are of individual variations, we have High input Generators (HIG) who
interact intensively and tend to produce more corrections and repeats but little silence ,and Low
Input Generators (LIG) who are more accurate but need more silent and filled paises to plan their
speech.
Pauses reasons in the study: unable to find the right word, to retrieve a lexical item, time to
correct a grammatical mistake (inflexion s), problems to retrieve the right form of the words. To
plan what is going to be said. To change a word they had already produced
word#pause#correction.
Mos common hesitation phenomena; repetition, self-correction and reformulation.

Speakers with higher level of proficiency use more filled pases than silent pauses. Hesitation
phenomena us more common in the case of silent pauses. Low proficiency used more hesitation
phenomena than HP.

Results:
Non-juncture pauses and hesitations are very frequent in L2 oral production, implying that they
face problems when planning and executing utterances.

Pauses (silent or filled) and hesitations are of individual variation and different production styles.
Filled and silent pauses are used when learners face lexical, morphological and planning difficulties,
but filled pauses are more common in the case of planning.
Most silent pauses co-occur with other hesitation phenomena but most filled pauses do not co-
occur with self-corrections, repetitions or reformulations, they seem to be used as repair devices
by themselves. +

The number of pauses is not associated with proficiency in the second language, but it seems that
those who presented lower proficiency used more strategies in combination with pauses. High
proficiency need time to retrieve the right information while lower pr. L need to vocalize different
options.

Cameron, Working with spoken discourse

Ch. 6 PRAGMATICS

Jenny Thomas defines pragmatics as the study of ‘meaning in interaction’ the social and
collaborative aspects of meaning-making. → “making meaning is a dynamic process, involving the
negotiation of meaning between speaker and hearer, the context of utterance (physical, social and
linguistic) and the meaning potential of an utterance.”
Discourse is the site for the dynamic process of meaning-making or meaning negotiation.
Speaking as doing

‘speech act’ → when we say something we are always also doing something. Austin identified a
class of utterances which he called ‘performatives’, because they perform a particular action in
and of themselves. For example, “I apologize” to utter this is actually to made an apology.

These performative utterances have some peculiarities. For instance, they only ‘work’ in the first
person and the present tense. Also, utterances of this kind cannot be true or false. If I person
apologies by saying “I apologies” I cannot say that he isn’t doing that, but I can doubt id the stratus
of the apology (is the person truly sorry?).Austin suggested that while performatives cannot be
true or false, they can be more or less ‘happy’ (felicitous). The performance of an act will be
felicitous only where certain conditions are met.

Austin distinguished utterances which are performative from other proposition-making utterances,
called constative utterances. He went on to question that distinction, suggesting that every
utterance can be analysed as the performance of some act by the speaker. In recognition of the
fact that utterances can both make propositions and perform actions, Austin proposed a three-
part framework for classifying ‘speech acts’.

- Locution is the actual words a speaker utters

- Illocution is the ‘force’ of the utterance, what it is meant to be taken as (e.g. assertion,
request, apology, promise)
- Perlocution is the effect the utterance has on the hearer.(e.g if they hear “it’s raining” as
having the force of a request for an umbrella, they supply the speaker with an umbrella)

There’s a focus on the question of how we decide on the illocutionary force of a given utterance.
The distinguishing characteristic of the type of utterance Austin originally classified as
‘performative’ is not simply that it has illocutionary force (all utterances do), but that it makes its
force explicit, the verb names the act it is used to performed (I promise you). Most speech acts do
not have to be performed, and on many or most occasions are not performed, in this way.

One source of information is the utterance itself. When Jenny Thomas refers to the ‘meaning
potential of an utterance’ she is making the point that what people actually say does not have an
unlimited range of possible interpretation. But it does have the potential to be interpreted in more
than one way. In considering these possibilities, people will make use of contextual information.

Searle (philosophical standpoint) attempts to account for people’s ability to identify the
illocutionary force of utterances by positing shared rules for the definition and felicitous
performance of particular illocutionary acts, that is to say, some conditions (rules) need to be in
place in order to identify the illocutionary force. In this framework, there’s too much emphasis on
the speaker’s intentions and state of mind and as Thomas’ said “meaning in interaction” cannot be
equated solely with the intentions of individuals. Another issue that arises with Searle’s approach
to speech acts is variability, there are different performances in different societies furthermore,
within one society there can be quite significant variation in people’s definitions of common
speech acts.

Form and function: a problem in discourse analysis

One of the questions speech act theory addresses is how the (propositional) meaning of an
utterance relates to its force. How “’ll do it tomorrow” comes to be taken as a promise. How
linguistic form relates to communicative function. Researchers use formal criteria- they look for
regular patterns in the use of a given form, where does it tend to occur? What is it used to do?. In
other words, to find the linguistic forms in which the chosen function is realized. However, the
relationship of form to functions is not a simple matter of one-to-one correspondence but many-
to-many, a given function can be realized in more than one linguistic form. And a form can
communicate a different thing.

Tannen argues for what she calls ‘the relativity of linguistic strategies’, by which she means that
the same form may realize radically different or even opposite communicative functions (e.g.
interrupting someone can be a put-down or a sign of enthusiastic support).

Speech act theory is relevant to this discussion because questions about the function of an
utterance in discourse are quite often questions about what a speech act theorist would call its
illocutionary force. If members of a speech community share rules for defining and performing
speech acts, then those rules may help to explain how people in real situations do the necessary
mapping from form to function. We might be able to explain certain kinds of misunderstandings as
the result of people not sharing the same rules for defining and performing speech acts, and
consequently arriving at different conclusions about the relationship of form to function.

Even where there is no issue of cross-cultural difference, Tannen’s point about the ‘relativity of
linguistic strategies’ raises the issue of contextual differences, which poses a problem for rule-
based approaches in the tradition of speech act theory. Sociocultural knowledge is difficult to build
into the kinds of general rules which are formulated by speech act theorists in the tradition of
Searle.

Acting on principle

The originator of the idea of a ‘pragmatic principle’ was H.P. Grice. He proposed that when people
interact with one another, a ‘co-operative principle’ is in force.

Co-operative 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.
Grice broke the co-operative principle down into four ‘maxims’

1. Quantity: make your contribution as informative as is required.

2. Quality: do not say what you believe to be false or you lack evidence of.
3. Relation: be relevant
4. Manner: avoid obscurity of expression, ambiguity, be brief.
The co-operative principle and its maxims are not rules, they are principles which it is rational and
logical for people to observe if their goal is to communicate meaning. We assume that language is
used for a purpose(to mean something), we’ll try to interpret what is said in accordance with oyr
expectations of a rational communicator. When Grice uses the term co-operative about
conversation, he is talking only about the kind and degree of co-operation that is necessary for
people to make sense of one another’s contributions. Speakers are operating on similar
assumptions about communication.

There are occasions on which people clearly do not observe all of Grice’s four maxims. We do not
immediately conclude that they have abandoned the co-operative principle. Rather, we consider
the possibility that the speaker’s ‘deviant’ behaviour is itself intended to be meaningful: the
speaker is trying to convey something to us by obviously flouting normal expectations regarding
quantity, quality, relevance or manner –they expect us to infer. Grice calls the meaning which is
inferred from the fact that a maxim is flouted implicature. Utterances in which maxims are flouted
cue us to look for relevant inferences, but the inferences we make are dependent on the
assumptions and background knowledge we bring on the interpretive task.

The principles approach is more accommodating of variation in interpretation → the existence of

such variation is less of a problem in this approach. Gricean pragmatics treats utterances as
providing evidence on the basis of which the hearer can work it out. Since all hearers do not come
to the task of drawing inferences with identical assumptions and stocks of background knowledge,
their interpretations of an utterance may differ. Nor can it be said that any of their interpretations
is the final and definitive interpretation. Implicatures are meanings that the speaker has only
implied and not stated, so they can later deny what a hearer inferred.

Not every failure maxim necessarily gives rise to implicatures. Hearers only look for implicatures if
they are obvious. Grice distinguished ‘flouting’ maxims from ‘violating them’ – flouting is meant to
be noticed, violation is meant not to be (a person lying is violating the maxim as the hearer doesn’t
realize that the liar is breaking the Quality maxim). it is also possible for people to utter what seem
to be obscure or irrelevant remarks without intending to generate implicatures.

Speakers who flout Grice’s maxims may do so as a strategy for avoiding or reducing the risks
associated with saying certain things directly. The idea that communication involves risk, and that
communicators behave in certain ways because they wish to minimize risk, has been particularly
influential in discussions of another set of pragmatic principles, those associated with politeness.
Leech has suggested that there is a ‘politeness principle’: if speakers have to choose between
being co-operative (informative, relevant, truthful), and being polite, they will normally choose
to be polite.

Not everyone agrees with Leech that there is an independent ‘politeness principle’, but many
pragmaticists would agree that politeness affects the application of the co-operative principle and
is therefore an important topic in pragmatics.
Politeness and face, Brown and Levinson

‘face’ which comes from the expressions “save face” or to “lose face” → a kind of social standing
or esteem which every individual claims for themselves and want others to respect.

- Positive face, the wish to be liked and approved by others

- Negative face, the wish to be allowed to go about your business without others imposing
unduly upon you.

many of the things people need or want to do in the course of interacting with one another are
inherently face threatening acts’ (FTAs): they have the potential to cause damage to the positive
or negative face of the speaker, the hearer or both. Brown and Levinson theorize politeness s a
strategy for mitigating threats to face in verbal interaction.

if a speaker perceives some action as face-threatening, the initial choice they must make is
whether to perform it at all. If the speaker decides to go ahead with a FTA, they must then choose
whether to perform it ‘on-record’ or ‘off-record’ (by implicature).

If they decide to perform the FTA on record which is to say somethiong directly, the speaker’s next
choice is whether to perform it with or without mitigation. The ‘no mitigation’ strategy is called
‘bald on record’ for example, asking someone to move by saying “move”. In some circumstances,
it is rational to be bald, if the roof is about to fall you’d shout “move” instead of casting around for
some indirect formula to motogate face threat. A speaker may also have reasons for choosing to
be deliberately impolite. However, in many instances speakers will choose a ‘mitigated’ way of
performing a FTA, opting for a strategy that verbally displays to the hearer that the speaker is
aware of the threat and wishes to minimize it. It is a less efficient and straightforward way of
expressing what the speaker means than the bald imperative: in Grice’s terms, it is less in keeping
with the maxim of manner. But the clearer and briefer alternative is also highly threatening to the
addressee’s face: the desire to minimize face threat gives speakers a good reason not to choose
the most direct and straightforward way of expressing what they mean.

“Politeness” can be positibe or negative. “Positive politeness” involves using language to signal
liking and approval. Negative politeness involves using language to minimize imposition.

how to use the various strategies in real-life situations? B&L suggest three factors for
consideration:

1. How threatening the FTA they are contemplating is → Some impositions are greater than
others.
2. How much social distance there is between the parties
3. What the power relations between the parties are

The difference between distance and power is that distance is a symmetrical relationship whereas
power is an asymmetrical, hierarchical one. The speaker’s strategy will deped on the calculations
concerning social distance and power. We are inclined to perform FTAs differently with those we
are socially close to and those we are more distant from. We use less elabotare politeness
strategies or we may use positive rather than negrative politeness with family and friends.We are
also inclined to behave differently to our social equals and to people whose status is higher or
lower than our own in a given situation(the “power” variable).

A politeness problem: compliments

Complimenting might seem on the surface to be the opposite of a threat to the recipient’s face,
but in practice it may resent the recipient with a potentially face-threatening interactional
dilemma.

Are pragmatic principles universal?

Both the co-operative principle and the ones underlying the model of politeness are often
presented as applying very generally across cultures and languages. In B&L’s case, the claim that
their model has unversal application is made explicitly: they recognize, and demonstrate with
examples, that politeness itself takes varying forms in different cultures, but they suggest this can
be explained with reference to considerations their model treats as variable in any case. It has
been argued that the co-operative principle captures something about the intrinsic nature of
human communication and the reasoning faculty that underpins it. These are strong claims, and
they have not gone unchallenged.
There are some “High context” cultures, in which communication relies heavily on the ability and
willingness of hearers to retrieve relevant information from contextual knowlegde. The maxims
are productos of a “low context” culture in which making meaning clear is implicitly considered to
be the responsibility of the speaker.

The Gricean approach gives hearers an active role in the production of meaning through inference,
but the maxims themselves focus on speaker’s behaviour rather than hearers’, and suggest that in
the ‘default’ case, the speaker will leave the hearer with little interpretive work to do. Indirectness
and inexplicitness, which place a greater burden on the hearer, are often treated by pragmaticists
as deviations from the ‘normal’ case, and are sometimes described as ‘risky’ (the intented meaninf
may not be retrieve) or ‘costly’ (extra effort to process) . But research on talk in places like Gapun,
or the Malagasy Republic, or Japan, suggests that these perceptions are not shared by all
communities in all circumstances.

Ch 7. Sequence and structure: conversation analysis

CA deals with conversations, talk in professional and worplace settings, political speeches and
media genres.Many practitioners of Conversation Analyisis (CA) label their object of study not
‘conversation’ but ‘talk-in-interaction’. Although CA can be used to analyse many different kinds of
data, it was developed to analyse talk, and more specifically the kind of talk that is thoroughly
interactive (not monologic sequences). The fact that talking is prototypically a joint enterprise
involving more than one person, and that people normally take turns at talk, is central to the CA
approach, which is concerned above all to describe sequential patterns.

CA is a markedly ‘data-centred’ form of discourse analysis: in ‘pure’ versions, the analyst is not
supposed to appeal to any evidence that comes from outside the talk itself.

Some CA practitioners apply the principle that one should not look beyond the data very strictly.
For them, it is not just unnecesary, but illegitimate for an analyst to make use of information that
the participants themselves have not chosen to ‘make relevant’ in their ongoing interaction. For
them it is illegitimate for an analyst to make use of information that the participants themselves
have not chosen to ‘make relevant’ in their ongoing interaction.

‘Analysts should only have recourse to concepts such as power once they are forced to do so by
virtue of some otherwise unexplainable interactional phenomenon’. There are others, including
people who use the techniques of CA as well as people who prefer other approaches, take a
different view.

CA is a ‘microanalytic’ approach, which takes apparently mundane and unremarkable spoken

interactions and finds intricate patterming in the way they are organized.

Investigating the obvious: the organization of turn-taking

Conversation requires speakers to take turns, and this requirement is managed in a particular way.
At any given moment, the turn that is in progress will typically belong to a single speaker.
Participants in conversation will not usually all talk at once. That is not to say that simultaneous
speech and silence never occur in conversation. But they are often treated by participants as
problems which need to be ‘repaired’. In the case of simultaneous speech, what typically happens
is that one speaker wins the floor, while the other(s) fall silent. In the case of a silence that
becomes long enough to feel awkward, what usually happens is that someone breaks it and claims
the floor.

The floor is not given to any one speaker for the duration → speaker change recurs. The floor is
constantly negotiated and renegotiated as a conversation goes along. CA holds that talk is ‘locally
managed’, meaning that its patterns and structures result from what people do as they go along
rather than from their being compelled to folloe a course of action that has been deterined in
advance.

Sacks, Schegloff and Jefferson proposed a model of conversationalists’ behaviour. The model has
two main elements:

- Speakers are aware that a turn consists of one or more ‘turn constructional units’. People
who are listening to someone else’s speech can use their knowledge of the possible unit
types to project the end-point of the turn currently in progress. The end of a turn
costructional unit is potentially a ‘turn transition relevance place’ , a point at which speaker
change may occur. Projecting the end of a turn involves attending to a combination of
things (the content of what is said, the prosodic and grammatical structure, and aspects of
nonverbal behaviour such as the direction of the speaker’s gaze)
- A mechanism for allocating turns to particular participants in a conversation. When a turn
trasntion relevance place is reached, what nsues is not a random free-for-all, with everyone
present having an wqual chance of getting the floor next. There is an ordered set of rules
for the allocation of the next turn:
1. Current speaker selects next speaker
2. Next speaker self selects
3. Currect speaker may (but does not have to) continue

There are many ways for a current speaker to select a next speaker, including asking them a
questin, naming them or aligning your body or gaze so you are seen to be addressing a specific co-
conversationalist at the end of your own turn.

If the current speaker does not select a next speaker, the second option is for someone other than
the current speaker to select themself, by starting to speak. There is a scope for simultaneous
speech to occur because more than one speaker simultaneously self-selects. If no one self-selects,
it is open to the current speaker to continue. At the next turn transition relevance place, the same
options apply in the same order all over again.

Some cases of simultaneous speech are classed as ‘overlaps’: they result from the new speaker’s
failure to project the end of the last speaker’s turn with complete accurac. Overlaps of this type
are common, but they are typically short.they are not instances of rule-breaking so much as
unintended errors in the application of the mechanisms. In other cases, a new speaker may start
to speak at a point in the last speaker’s utterance that cannot possibly be a turn transition
relevance place. This is interruption, and it is a hostile act designed to deny the current speaker
their legitimate right to the floor.

For other analyst, the idea that simultaneous speech is either overlap or interruption is
oversimplifies, and does not accont for some quite common instances of simultaneous speech.

Simultaneous speech with supportive function (a speaker A interjects in order to say the same
thing to support speaker B in holding the floor rather than to take the floor away from them) is
particularly common in the talk of women friends. Supportive simultaneous speech can also occur
in institutional context.

Edelsky distinguished two types of floor, which she labeled F1 and F2

• F1 → one speaker speaks at a time

• F2 → involves more simultaneous speech

Adjacency pairs

The turn-taking system provides a basic framework for the organization of talk-in-interaction,
since it allows for the floor to be alterned systematically between speakers. It is not on its own
sufficient to guarantee that there is genuine and meaningful interaction.

Questions are not always followed immediately by answers; sometimes, they may be followed by
futher questions. If this is the case, that only defers and does not cancel the expectation that the
original question must be answered.

Spoken interaction is often structured around pairs of adjacent utterances in which the second
utterance is not just related to the first but functionally dpendent on it.

‘Adjacency pair’ → a pair of utterances in whci the second is functionally dependent on the first.

The two parts of an adjacency pair may become separated. This separation is normally brief and
the second adjacency pair which is inserted between the two parts of the first one is understood
to be necessary for the completion of the original transaction.

The fact that we percieve the non-return of greetings as rude, provides supporting evidence for
the claim that a greeting is not a free-standing turn but part of a larger unit, an adjacency pair.

Some first turns present the producer of the second turn with a choice. The speaker may be met
with either agreement or disagreement.

‘Preference system’ → preferred and dispreferred responses as seconds in an adjacency pair.

The preferred response to a proposal is acceptance, and it can be identified as the preferred
option because it is typically performed without hesitation or elaboration it is possible to accept
proposals more elaborately, but acceptance can be done without elaboration. Refusal, by contrast,
is almost never done in the same way. Refusal is the dispreferred response, and is performed with
hesitance to come to the point and then an elaboration on the refusal by giving a reason for it.

Preferred responses are prompt and short while dispreferred ones are hesitant and elaborate. The
preferred option is also the less threatening to the addressee’s face.

Beyond the ‘basics’: institutional talk

‘Institutional talk’ → business meetings, service encounters, doctor-patient consultations, talk in

courtrooms or in classrooms… its analysis tends to be organized around the question ‘what special
considerations apply that make talk in this institutional context different from ordinary talk?’

One special and particular constraint that often applies to some degree in institutional talk, thogh
it rarely applies in ordinary conversation, was a constrain on the right to ask questions. Defendants
were not permitted to ask questions, whereas they were obliged to respond to questions posed by
the representatives of the institution.

The ‘one rule for one and one for another’ principle applies rather consistently in asymmetrical
talk (i.e. talk in which the participants do not have equal power, status, responsibility or control).
Typically it is the dominant party who has the right (and/or the obligation) to ask questions, while
the subordinate party has a restricted right to ask questions. A question forms the first turn of an
adjacency pair, and it does not merely require an answer, it also constrains what will count as a
relevant or appropriate answer.

Institutional talk may be associated with particular inferential frameworks. In other words, people
engaged in talk that has a certain institutional goal may use their understanding of what the
institution wishes to achieve to make inferences about the meaning or function of an utterance
that they might not make in other contexts.

Cutting, J

A2. Speech acts

• Declarations → change the world by their very utterance

• Representatives → the words state what the speaker believes to be the case
• Commissives → the words commit the speaker to future action
• Directives → the words are aimed at making the hearer do something
• Expressives → the words state what the speaker feels

Felicity conditions

In order for speech acts to be appropriately and successfully performed, certain felicity conditions
have to be met. For Austin, these are that the context and roles of participants must be recognised
by all parties, the actions must be carried out completely, and the persons must have the right
intentions. For Searle, there is a general condition for all speech acts, that the hearer must hear
and understand the language, and that the speaker must not be pretending or play-acting.

Indirect speech acts

Much of the time, what we mean is actually not in the words themselves but in the meaning
implied.

Searle said that a speaker using a direct speech act wants to communicate the literal meaning that
the words conventionally express.

Someone using an indirect speech act wants to communicate a different meaning from the
apparent surface meaning: the form and function are not directly related. There is an underlying
pragmatic meaning, and one speech act is performed through another speech act.

The classification of utterances in categories of direct and indirect speech acts is not an easy task,
because much of what we say operates on both levels, and utterances often have more than one
of the macro-functions.

Speech acts and society

Social dimensions

Indirect speech acts in many languages and cultures constitute one of many forms of politeness. In
most Englishes, indirectness is so much associated with politeness that directives are more often
expressed as interrogatives than imperatives. This is especially the case with people with whom
one is not familiar.

Other factors that can make speakers use indirect directives are the reasonableness of the
task ,the formality of the context and social distance. Social distance can give speakers power and
authority, and it is generally those of the less dominant role, etc., who tend to use indirectness.

Cultural dimension

Speech acts, their linguistic realisation and their relationship to the social dimensions mentioned
above are very much culture bound.

Differences in speech act conventions can cause communication difficulties interculturally.

Limitations of Speech Act Theory

One utterance can fall into more than one macro-class.

Another problem is that it has no provision for the ‘messiness’ of everyday spoken language, like
with fillers that say very little; this lack of semantic content makes it difficult to put in any of the
classifications. Likewise, backchannels and feelback do not fit neatly into the speech act model
either. They too have a social function, but do not constitute a speech act. The same goes for
incomplete sentences.

Macro-functions

There are two main macrofunctions of talk. B&Y describe them as the transactional and
interactional functions.

• Transactional → the function which language serves in the expression of content and the
transmission of factual information
• Interactional → the function involved in expressing social relations and personal attitudes,
showing solidarity and maintaining social cohesion. Speakers establishing common ground,
sharing a common point of view and negotiating role-relationships are speaking with an
interactional purpose.

Most talk has a mitxture of the two functions. At the extreme end of the interactional is what is
known as phatic communion, language with no information content used purely to keep channels
of communication open.

A3

Conversation analysis

CA takes a ‘bottom-up’ approach: starting with the conversation itself, it lets the data dictate its
own structure. CA differs in its methodology from discourse analysis. Whereas discourse analysis
takes the concepts and terms of linguistics and then examines their role in real data, conversation
analysis takes real data and then examines the language and demonstrates that conversation is
systemically structured.

Conversation is discurse mutually constructed and negotiated in time between speakers; it is

usually informed and unplanned.

1. It is not primarily necessitated by a practical task → many linguists would argue this, and
say that most of what we say is outcome oriented. Even the most casual of conversations
have an interaction function.
2. Any unequal power of participants is partially suspended → linguist would contend this
property, by pointing out that, in all exchanges, there is unequal powe, in varying degrees,
and that conversation can occur whe there are significant power differentials between
participants.
3. The number of the participants is small
4. Turns are quite short
5. Talk is primarily for the participants, not for an outside audience.

Turn-taking
Cooperation in conversation is managed by all participants through turn-taking. Speakers take
turns, first one talking and then another. All cultures have their own preferences.

A point in a conversation where a change of turn is possible is called a Transition Relevance Place,
or TRP. When speakers do not want to wait until the TRP, this is called an interruption.

When hearers predict that the turn is about to be completed and they come before it is, this is an
overlap.

In any culture, if the pause is intended to carry meaning, analysts call it an attributable silence.

Adjacency pairs

Conversation contains frequently occurring patterns, in pairs of utterances known as ‘adjacency

pairs’. They say that the utterance of one speaker makes a certain response of the next speaker
very likely. The acts are ordered with a first part and a second part and categorised as question-
answer, offer-accept, blame-deny, and so on, with each first part creating an expectation of a
particular sevond part. This is known as preference structure: each first part has a preferred and a
dispreferred response. The pairs are endless.

The dispreferred responses tend to be the refusals and disagreements. These are the more
unusual responses, and they can be taken as meaningful or rude.

Sequences

Speakers are mutually constructing and negotiating their conversation in time, certain sequences,
which are stretches of utterances of turns, emerge. These can be pre-sequences, insertion
sequences and opening and closing sequences.

Pre-sequences prepare the ground for a further sequence and signal the type of utterance to
follow. There are pre-invitations, pre-requests and pre-announcements.

In the case of an insertion sequence, the pairs occur embedded within other adjacency pairs,
which act as macro-sequences.

Finally, there are conventional opening structures and closing structures. Openings tend to contain
a greeting, an enquiry after health and a past reference.

Limitations of CA

There is a lack of systematicity in the sense that there is not an exhaustive list of all adjacency pairs,
or a precise description of how adjacency pairs or TRPs might be recognised. CA sets out to be a
qualitative not a quantitative approach.

It does not take into account pragmatic or sociolinguistic aspects of interaction, the background
context of why and how people say what they say, the components of situation and the features
of the social world and social identity such as occupation and gender of participants. For CA
analysts, text is context.

Interactional sociolinguistics

It brings to the forefront the situational context and the context of shared knowledge about
speakers, their histories and their purpose in speaking. It looks at grammar, social structure and
cultural patterns.

Interactional sociolinguistics focuses on the fact that social groups have their own ways of
expressing meaning with their language. Language relates to context through “contextualisation
cues”. These are the linguistic features that indicate the aspects of the context relevant to what
the speaker means and that only take on their full meaning when the hearer is familiar with the
rest of the context.

Although the main goal of interactional sociolinguistics is not to describe the structure of discourse,
and that is the main goal of conversation analysis, the two approaches have come together now,
with analysts lokig at the relationship between grammar and social interaction, within the larger
schemes of human conduct and the organisation of social life.

A4- The cooperative principle

Verbal exchanges tend to run more smoothlu and succesfully when the participants follow certain
social conventions

Observing the maxims

The first maxim of the Cooperative Principle is the maxim of quantity, that says that speakers
should be as informative as is required.

The second maxim is that of quality, which says that speakers are expected to be sincere.

The third is the maxim of relation, which says that speakers are assumed to be saying something
that is relevant to what has been said before.

The last is the maxim of manner, which says that we should be brief and orderly, and avoid
obscurity and ambiguity.

Hearers assume that speakers observe the Cooperative Principle, and that it is the knowledge of
the four maxims that allows hearers to draw inferences about the speakers’ intentions and implied
meaning. The meaning conveyed by speakers and recovered as a result of the hearers’ inferences
is known as conversational implicature.

Flouting the maxims

When speakers appear not to follow the maxims but expect hearers to appreciate the meaning
implied, we say that they are flouting the maxims. The speaker implies a function different the
literal meaning of form, when flouting a maxim, the speaker assumes that the hearer knows that
their words should not be taken at face value an that they can infer the implicit meaning

Flouting quantity → give too little or too much information

Flouting quality

- Say something that obviously does not represent what the speaker thinks
- Exaggerating as in the hyperbole – often at the basis of humour
- Using a metaphor – conventional euphemisms can be put in this category too
- Irony and banter. In the case of irony, the speaker expresses a positive sentiment and
implies a negative one (Sarcasm is a form of irony that is not so friendly). Banter expresses
a negative sentiment and implies a positive one.

Flouting relation → the speakers expect that the hearers will be able to iamgine what the
utterance did not say and make the connection between their utterance and the preceding one(s).

Flouting manner → appearing to be obscure –are often trying to exclude a third party.

Violating the maxims

A speaker violates a maxim when they know that the hearer will nor know the truth and will only
understand the surface meaning of the words. They intentionally generate a misleading
implicature. The speaker deliberately supplies insufficient information, says something that is
insincere, irrelevant or ambiguous, and the hearer wrongly assumes that they are cooperating.

- Violation of the maxim of quantity → not enough information is given

- Violation of the maxim and quality → not being sincere. Not all violations of the maxim of
quality are blameworthy → white lies. In many cultures it is acceptable to violate the
maxim of quality if one does not know the hearer very well, and it may in fact be part of
polite behaviour to
- Violation of the maxim of relation → change the topic, don’t say anything relevant
- Violation of the maxim of manner → e.g. to say everything except what the interviewer
wants to know.

Other forms of non-observance of maxims

A speaker infringing a maxim fails to observe a maxim because of their imperfect linguistic
performance. This can happen if the speaker has an imperfect command of the language, if their
performance is impaired, if they cave a cognitive impairment, or if they are incapable of speaking
clearly.

A speaker opting out of a maxim indicates an unwillingness to cooperate, although they do not
want to appear uncooperative.

Limitations of the cooperative principle

An objection that one amy have to Grice’s model is that different cultures, countries and
communities have their own ways of observing and expressing maxims for particular situations.

Another problem is that there is often an overlap between the four maxims. It can be difficult to
say which one is operating, and it would be more precise to say that there are two or more
operating at once.

S&W say that all maxims can be reduced to the maxim of relation, since relevance is a natural
feature of all exchanges in which speakers have the aim of achieving successful communication.

Relevance theory, S&W

Conversational implicature is understood by hearers simply by selecting the relevant features of

context and recognising whatever speakers say as relevant features of context and recognising
whatever speakers say as relevant to the conversation.

The degree of relevance is foverned by contextual effects and processing effort. Contextual effects
include adding new information, strenghtening or contradicting an existing assumption, or
weakening old information. The more contextual effects,the greater the relevance of a particular
fact.

The less effort it takes to recover a fact, the greater the relevance.

To understand an utterance is to prove its relevance, and proving relevance is determined by the
accessibility of its relevance to the addressee.

The filling in the missing words, elaborating or ‘enriching the propositional form’ is what S&W call
explicature. This is a necessary stage before implicature. The explicature of an utterance consists
of the propositions that are explicitly communicated by the speaker, and that some of this has to
be inferred by relevance-driven processes.

Limitations of Relevance Theory

S&W feel that their principle accounts for all Grice’s maxims, and that it is without exception and
irrefutable means that that the notion of relevance is so encompassing that it loses its explanatory
force. Everything implies something that is not said, since every utterance depends on associations
and background knowledge.

This theory says nothing about interaction and does not include cultural or social dimensions, such
as age, gender, status and nationality.

A5 Politeness

Politeness and face

In order to enter into social relationships, we have to acknowledge and show an awareness of the
face, the public self-image, the sense of self, of the people that we address. It is a universal
characteristic across cultures that speakers should respect each other’s expectations regarding
self-image, take account of their feelings and avoid FTAs. When unavoidable, speakers can redress
the threat with negative politeness (that respects the hearer’s negative face), or with positive
politeness (which attends the positive face).

• If you want to avoid an FTA, you can avoid saying anything at all.

Or you can say something. You are then faced with a choice: to do the FTA on record or off record.

• Off record implies, for example, asking for help indirectly.

Indirectess in the form of indirect speech acts and cooperative maxim flouting allows a speaker to
make suggestions, requests, offers or invitations quite casually, without addressing them to
anyone in particular. The illocutionary force will most likely be understood by hearer, but they can
choose to ignore it.

Indirectness also enables speakers to address particular people, but be polite by giving them
potions and retreating behind the literal meaning of the words.

• On record – baldly

If a speaker makes a suggestion, request, offer or invitation in an open and direct way, we say that
they are doing an FTA bald on record. These are direct speech acts; such utterances tend to
contain the imperative with no mitigating devices. It leaves the hearers little option but to do as
they are told or be seen as uncooperative. This is the most face-threatening mode of action.

The directness also makes the hearer less reluctant to threaten the speaker’s face by impigning
through accepting. The firmer the invitation, the more polite it is. Besides, directness often
indicates a wish to be seen as socially close.

• On record –with negative politeness

Negative politeness strategies pay attention to negative face, by demonstrating the distance
between interlocutors, and avoiding intruding on each other’s territory. They are used to avoid
imposing or presuming, and to give the hearer options.

e.g. emphasising the importance of the other’s time and concerns, using apology and hesitation,
or a question giving them the opportunity to say no.

the extent of the option-giving influences the degree of politeness. Speakers can minimise the
imposition by making it seem smaller than it is, or by adding devices such as hedges hat mitigate
the imposition.

They can also emphasise the distance between interlocutors by impersonalising, stating the
imposition as a general rule or a nominalisation.
 • On record –with positive politeness

Positive politeness strategies aim to save positive face, by demonstrating closeness and solidarity,
appealing to friendship, making other people feel good and emphasising that both speakers have a
common goal.

B&L say that one of the main types of positive politeness strategy is claiming common ground.
Speakers can do this by attending to the hearer’s interests, wants and needs.

Relationship with the Cooperative Principle

Speaker can violate cooperative maxims if they want to show positive politeness. Speakers may
also choose to opt out of cooperative maxims to show negative politeness.

Politeness maxims –Leech

There is a politeness principle with conversational maxims. He lists six maxims: tact, generosity,
approbation, modesty, agreement and sympathy.

- Tact maxim: focuses on the hearer. ‘Minime cost to other’ and ‘maximise benefit to other’.
Related to the negative politeness strategy of minimising the imposition, and to the
positive politeness strategy of attending to the hearer’s interests, wants and needs.
- Generosity maxim: focuses on the speaker. ‘minimise benefit to self’ and ‘maximise cost to
self’.
- Approbation maxim: ‘minimise dispraise of other’ and ‘maximise praise of other’ –
politeness strategy of avoiding disagreement + positive politeness strategy of mking other
peple feel good by showing solidarity.
- Modesty maxim: ‘minimise praise of self’ and ‘maximise dispraise of self’. Modesty is
possibly a more complex maxim than the others, since the maxim of quality can sometimes
be violated in observing
- Agreement maxim : ‘minimise disagreement between self and other’ and ‘maximise
agreement between self and other’ – related to the positive politeness strategies of ‘seek
agreement’ and ‘avoid disagreement’.
- Sympathy maxim: ‘minimise antipathy between self and other’ and ‘maximise sympathy
between self and other’ – e.g. congratulate, commiserate and express condolences.

Overlaps and gaps

There is a considerable overlap between the categories of B&L’s model and the categories of
Leech’s model. There is also overlap within B&L’s model and Leech’s: the categories themselves
are not mutually exclusive.

One utterance can contain both positive and negative politeness. Similarly, one utterance can
obey two or more maxims.t
There may be endless gaps not covered by the maxims; no model can describe all human
intteractions

Politeness and context

- Form and function

Politeness is a pragmatic phenomenon in their function and intended social meaning.

Politeness is not the same as deference, which is a polite form expressing distance from and
respect for other people of a higher status, and does not usually include an element of choice. It is
rare to find it grammatically signalled in English.

- Situational context

It is influenced by elements of the context. There are two situational context factors thati nfluece
the way that we make a request. One is the size of imposition → ‘the greater the imposition, the
more indirect the language is’

The other factor is the formality of the context → ‘the greater the formality, the more indirect the
language is’.

- Social context

The choice of the politeness formulation depends on the social distance and the power relation
between speakers. When there is social distance, politeness is encoded, and there is more
indirectness; where there is less social distance, there is less negative politeness and indirectness.
The variables that determine social distance are degree of familiarity and differences of status,
roles, age, gender, education, class and occupation.

The degree of familiarity between speakers is one of the most obvious social variables that affect
how politeness is expressed.

Differences of status, roles, age, gender, education, class, occupation and ethnicity can give
speakers power and authority-

- Cultural context

The relationship between indirectness and social variables is not so simple: the wholes issue of
politeness and language is exceedingly culture-bound. The use of indirectness ‘can hardly be
understood without the cross-cultural perspective’.

The use of the maxims of tact and generosity varies greatly from country to country.

The use of the maxims of approbation and modesty are also deeply rooted in culture.
Politeness is related to the context, the language used, the speech acts, the structure of the
conversation and the principle of cooperation. Politeness is a basic form of cooperation and it
underlies all language in some way or another.