Los Acuerdos y Convenios Colectivos Sectoriales Estatales

Most coherence-based accounts are characterised by an externalised view of language (cf. Chomsky, 1986) as they describe discourse as communicative behaviour, independent of and external to the human mind. This view is radically different from that of the relevance theorists, who are not concerned with text or discourse per se, “but rather discourse understanding, or more particularly, the mental representations and computations underlying utterance understanding” (Blakemore, 2001: 100-101). In other words, the object of study within a relevance-theoretic perspective is internal to the human mind.

Building on the work of Paul Grice (1961, 1989),53 Sperber & Wilson (1986/1995, 1987) have proposed a relevance-theoretic model of human communication, which stands opposed to the classical code model whereby information is encoded into a message, transmitted and decoded by another party, with another copy of the code. They argue that utterance Interpretation is not achieved by identifying the semantically encoded meanings of sentences, but involves inferential computations performed over conceptual representations or

53 _{Grice’s theory of meaning and communication follows the tradition of ordinary language philosophy (cf.}

4.1.1) and has become a “landmark (…) on the path towards the development of a systematic, philosophically inspired pragmatic theory of language use” (Huang, 2007: 3).

propositions - that is, the propositional content of the utterance Interpreted taken together with contextual assumptions (Sperber & Wilson, 1986/1995).

The fundamental tenets of RT are contained in a definition of relevance and two principles, one about cognition and the other about communication, to which I will now turn.

4.2.1 How relevance guides inferential comprehension

The relevance-theoretic inferential account of communication is based on a central assumption about cognitive processes: human cognition is relevance oriented (Sperber & Wilson, 1986/1995, 1987). This assumption is to be found in what Sperber & Wilson call the Cognitive Principle of Relevance (Sperber & Wilson, 1986/1995: 260):

Cognitive Principle of Relevance: Human cognition tends to be geared to the maximisation of relevance.

What, then, is relevance? Sperber & Wilson define relevance as “a property of inputs to cognitive processes and analysed in terms of the notions of cognitive effect and processing effort” (Wilson, 2000: 423). Relevance is thus an improvement of one’s overall representation of the world and is seen as a matter of degree, i.e. the degree of relevance of an input to an individual is a trade-off or balance between cognitive effects (reward) and processing effort (cost). This is made clear by Sperber & Wilson (1986/1995: 252) in the following passage:

Relevance of an input to an individual

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.

Thus, the more cognitive effects the hearer is able to derive, the more relevant the information. In other words information is relevant for the hearer to the extent that it yields

cognitive effects at low processing effort by interacting with and modifying her existing assumptions about the world.

Sperber & Wilson (1986/1995) identified three types of cognitive effect: (a) generating a conclusion drawn from old or new information together, but not from such information taken separately, which is known as contextual implication; (b) strengthening an existing assumption; and (c) contradicting or eliminating an existing assumption. Consider the following example taken from Blakemore (2002: 60-61). A bus driver is about to leave from the bus terminal, when he suddenly sees in the rear window a woman waving a bus pass in her hand, trying to cross the road to get to the terminal. In such context, the bus driver’s overall representation of the world can be improved in three ways, corresponding to the three cognitive effects discussed above. Firstly, given the assumption that a person waving a bus pass intends to catch a bus, the bus driver will derive the new assumption or contextual implication that the woman intends to travel on his bus. Secondly, the bus driver’s ‘old’ (or existing) assumption that the woman is trying to catch his bus by crossing the road may be strengthened by the assumption that she is waving her pass. Thirdly, the bus driver’s existing assumption that the woman’s intention is to catch his bus is contradicted and eliminated when he sees that the woman walks in the opposite direction, after giving her pass to a friend on the street.

Thus, as the cognitive principle of relevance suggests, in processing information people try to maximise cognitive effects; in other words, human attention and processing resources are designed to look for as many cognitive effects as possible for as little effort as possible.54 This, in turn, has an immediate consequence for the theory of communication. For Sperber & Wilson (1986/1995: 158ff.), communication means ostensive communication, where this is defined as involving both first-order informative intentions and higher-order communicative intentions; the attribution of the latter is yielded by ‘ostensive behaviour’ or ostention. Ostensive-inferential communication is often ‘triggered’ by an ostensive stimulus, which is used to give rise to the expectation of optimal relevance. In other words, such stimulus is “the most relevant one the communicator could have used to communicate” (Sperber & Wilson, 1986/1995: 158).

54 _{The comparative (rather than quantitative) nature of the trade-off of effect and effort has led some critics to}

state that RT does not provide a satisfactory account of how to measure this ‘balance’ and how to measure cognitive effect and processing effort against each other (e.g. Bach, 1999; see also Wilson and Sperber’s (2004) counterarguments).

For instance, suppose that the police officer asks an interviewee if he has run to run away from the crime scene, to which request he frowns and indicates the cast on her right leg. This is an example of ostensive behaviour and the officer recognises it as such. In particular, he infers (non-demonstratively) that the interviewee is injured and wishes to rest his case, and finally that he could not have been the one who was fleeing. The suspect has communicated a negative answer and his reasons by giving the officer some evidence of his thoughts (cf. act of indicating the cast). The ostensive nature of such behaviour could be expected to suggest to the officer that he intended to trigger this idea in his mind. The suspect thought that the idea activated, and the manifestly intentional nature of its activation, would be the starting point for an inferential process leading to the discovery of the suspect’s full meaning.55

According to RT, the very act of requesting the hearer’s attention encourages her to believe that the information given will be relevant enough to be worth processing. Hence, every act of communication creates an expectation that a hearer is entitled to have – namely, that the utterance is the most relevant one within the parameters of the speaker’s abilities and preferences. This generalisation about human communicative behaviour is expressed in the second principle of relevance (Sperber & Wilson, 1986/1995: 260):

Communicative Principle of Relevance: Every act of ostensive communication communicates a presumption of its own optimal relevance.

In other words, the speaker communicates that his utterance is the most relevant one compatible with his abilities and preferences and is at least relevant enough to be worth the hearer’s processing effort (Sperber & Wilson, 1986/1995: 270 ff.). Again it is in his interest to do so, as the less processing effort and the greater the effect, the more relevant the utterance and the more likely it is that the addressee will understand it successfully (Wilson & Sperber, 2000, 2004).

The Communicative Principle of Relevance motivates the following comprehension heuristic which, according to RT, hearers spontaneously follow in utterance Interpretation (Wilson, 2000: 423):

55 _{The suspect could have achieved a similar effect by uttering the following words: “I hurt my right leg”. This}

would have activated the idea of his being injured (this time by linguistic decoding); further, it would have done so in a manifestly intentional way.

Relevance-theoretic comprehension procedure:

Follow a path of least effort in computing cognitive effects. (a) Consider interpretations in order of accessibility. (b) Stop when your expectation of relevance is satisfied.

The speaker aiming at optimal relevance will try to formulate his utterance in such a way as to minimise processing effort for the hearer, so that the first acceptable Interpretation derived by the hearer is the one she intended to convey.56

In this section, I have analysed the two fundamental principles within RT, i.e. the cognitive and communicative principles. In particular, the latter applies at the level of both explicit and implicit communication; thus, I turn to the definition of these notions in the following section.

4.2.2 The distinction between explicit and implicit content

According to Grice’s (1957, 1969, 1989) theory of communication, a distinction is made between natural meaning (in the world) and non-natural (or linguistic) overall meaning of an utterance. The meaning n[on]n[atural] or speaker-meaning is claimed to be a matter of expressing and recognising intention (similar to RT; cf. Levinson, 2000) and is divided into what is said and what is implicated, as represented in the following diagram:

56 _{This does not imply that every act of overt communication is in fact optimally relevant (Sperber, 1994;}

Wilson, 2000). For instance, speakers can be mistaken about the relevance of the information they communicate or about the hearer’s contextual or processing resources.

Figure 1. Gricean theory of communication (from Levinson 2000: 13).

According to Grice (1989: 25; see also Carston, 2002: 114ff.), what is said is argued to be the conventional meaning of the utterance (excluding conventional implicatures)57 and the truth- conditional content of the utterance. On the contrary, what is (conversationally) implicated is defined in contrast to and derived following the input provided by what is said. However, human beings must ‘unpack’ phenomena such as ellipsis, deixis, and lexical ambiguity before they can understand what is said. In relevance-theoretic accounts of human communication, it is claimed that Grice did not fully recognise the contribution provided by pragmatics to the truth-conditional propositional content of the sentence uttered. Both Grice and Sperber & Wilson (1986/1995) agree that we should distinguish between meaning which is explicitly communicated (what is said) and meaning which is part of the implicit content of the utterance (what is implicated). However, for Grice the role of the maxims and pragmatic inference is restricted to what is implicated and plays no role in the recovery of what is said. In contrast, Sperber & Wilson (1986/1995) and Carston (2002, 2004) argue that the explicit side of communication is far more inferential than Grice envisaged: according to RT, both the explicit side of communication and the implicit side involve making inferences from

57 _{The notion of conventional implicature (the second type of implicatures within Grice’s framework) will be}

explored within the broader discussion on DMs in 4.3.1.

MEANING   What  is   implicated   conventionally   conversationally   generalised   particularised   What  is   said  

contextual assumptions on the basis of general pragmatic principles. Thus, Sperber & Wilson develop the notion of explicature which is defined in terms of an inferential development of linguistically given incomplete logical forms into propositional forms. In other words, explicatures serve to ‘flesh out’ the incomplete conceptual representations encoded by the utterances and thus yield fully propositional content. Consider, for instance, the following (straightforward) case of disambiguation:58

(13) He has a mole on his face.

The hearer of utterance (13) will be required to disambiguate or select one meaning out of two (or more) potential meanings provided by the English language system,59 _{i.e. mole as (a) a}

small cylindrical mammals adapted to a subterranean lifestyle, or mole as (b) a benign tumour on human skin, usually with darker pigment. After selecting Interpretation (b) according to the context given, the explicature he derives to complete and enrich the (incomplete) logical form is as follows:

(14) He has a benign tumour with darker pigment on his face.

It is worth highlighting that relevance theorists have defined two subtypes of explicatures. The first subtype is called higher-level (or higher-order) explicature (Wilson & Sperber, 2004) and serves to embed the truth-conditional content of the sentence uttered under higher- level descriptions, including speech act descriptions and evidentials (which are perceived as comments on the embedded propositions; see Ifantidou-Trouki, 2001). The second subtype is called basic explicature and is defined as non-higher-level explicature. Consider, for instance, the following utterance:

(15) Frankly, I’ve had enough of you.

58 _{See also Asher & Lascarides’ (1995) analysis of coherence-based heuristics for disambiguation in discourse,}

where the same criterion of consistency with the principle of relevance is shown to explain disambiguation in both isolated utterances and extended texts.

59 _{It must be stressed that in RT (contra Grice and Bach) the recovery of propositional content is not just a matter}

of disambiguation or reference or deictic resolution. See, for instance, fragmentary utterances (e.g. Nice dress!) and cases where disambiguation and reference assignment do not deliver optimally relevant propositions (e.g.

The park is some distance from my home, where assigning a reference to park and home does not lead to a proposition intended).

The use of the sentence adverbial frankly here makes clear the reliability of the evidence on which an utterance is based. The higher-level explicature that may be derived is as follows:

(16) The speaker is telling the hearer frankly that he doesn’t stand her anymore.

Unlike explicit content, the implicit content or implicature within RT is seen as an assumption which can only be derived pragmatically, i.e. via pragmatic inferences (Sperber & Wilson 1986/1995). Thus, the difference between explicatures and implicatures lies in that the recovery of the former involves both decoding and inference, whereas the latter involves only inference. Furthermore, there are two subtypes of implicatures: implicated premises and implicated conclusions. Implicated premises are contextual assumptions intended by the speaker and provided by the hearer, whereas implicated conclusions are contextual implications communicated by the speaker. Consider, for example, the following:

(17) Mark: How about going out for a walk? John: It’s raining.

In this example, the hearer is required to pragmatically enrich ‘It’s raining’ in order to ensure that she has the proposition that it is raining where both Mark and John are intending to walk (explicature). The implicated premise, instead, is that if it is raining heavily outside, it is not possible to have a walk and the implicated conclusion is that Mark and John cannot have a walk outside. In this way, the implicated conclusion follows deductively from the implicated premise combined with utterances in (17).

It should be emphasised that both explicit and implicit content are a matter of degree. Explicitness depends on a balance of labour between linguistic encoding and pragmatic inference. Since explicit content is contextually determined and involves pragmatic inference, we may also say that it can vary in strength depending on the responsibility that the hearer is given for its recovery (Carston, 2002). At the same time, the strength of an implicature may vary along a continuum ranging from weak to strong implicatures (Sperber & Wilson, 1986/1995). In particular, weak implicatures are analysed as those implicatures whose recovery is not essential to understand the speaker’s intended meaning, whereas strong

implicatures are those who are essential to utterance comprehension. For instance, the implicatures recovered from John’s utterance in (17) might include:

(18) a. John does not want to walk in the rain. b. John wants to stay home.

c. John is encouraging Mark to admit that he is also unwilling to go out.

d. John would like them to stay home as their favourite soap opera is on television.

In order to satisfy the expectation of relevance raised by an utterance, the audience must, on the one hand, develop its encoded linguistic meaning into an appropriately explicit propositional content (explicature) and on the other, use contextual assumptions made accessible by the conceptual content of this explicature in the derivation of cognitive effects. These two operations do not take place serially, but are, as Carston (2002) puts it, mutual adjustment processes, with hypotheses about context, explicit content and cognitive effects being made, adjusted, and confirmed in parallel, on-line.

To sum up, Grice’s and relevance-theoretic accounts agree that there is a level of semantic representation – or the linguistic meaning of an utterance – which belongs to the domain of semantics. However, relevance theorists posit that (a substantial) part of the notion of what is said involves a greater pragmatic intrusion than originally foreseen by Grice. In particular, they put forward the notion of explicature, parallel to the Gricean notion of implicature, which serves to ‘flesh out’ the linguistically given incomplete logical form of an utterance. After constructing an appropriate hypothesis about the explicit content of the utterance (via decoding, disambiguation, and other pragmatic enrichment processes), the hearer is expected to combine the derived explicature with contextual premises for the derivation of the implicated conclusions. Lastly, such parallel, on-line recovery of both explicatures and implicatures engendered by an utterance is guided by the communicative principle of relevance and the first satisfactory Interpretation recovered by the use of such comprehension heuristic is seen as the only satisfactory one.

In contrast with Grice’s maxims, the communicative principle of relevance is thus not a rule which can be followed or deliberately flouted; rather, it is an empirical generalisation about the way we communicate, which is in turn based on the way we process information (cf. Cognitive Principle of Relevance in 4.2.1). In the next section, we aim to further explore this

inferential theory of human communication; in particular, we shall discuss the relationship between thoughts and utterances and the notions of metarepresentation and interpretative resemblance, both of which are highly relevant to the notions of the police interpreter’s role and neutrality.

4.2.3 The relationship between thoughts and utterances

The discussion so far raises the following question: is our utterance comprehension strategy to be seen as a specialised cognitive domain with its own (innately specific) principles? In other words, do we possess a ‘pragmatic’ module in our minds?

One of the most influential theories of mind is the one formulated by Fodor (1983), influenced by Chomsky’s (1986) view of language. The fundamental thesis put forward by Fodor is that the human mind is divided into an undifferentiated central system and a series of specialised cognitive systems or modules, which mainly serve to provide input to the central system. In this context, language is seen as a dedicated module (or a cluster of modules) which feed into the central ‘processor’, the latter being responsible for general mental capacities such as problem solving and rational thought formation.60

It has been argued that Fodor’s and Chomsky’s modular view of the mind underlies the distinction between grammatically specified meaning and pragmatic meaning (cf. Blakemore, 2002: 154ff.). According to RT, grammar has a role to play in communicative events, however this role is to deliver “semantic representations which fall short of the complete interpretation intended” (Blakemore, 2001: 101), rather than representations of thoughts