Article
Reference
Intercultural pragmatics: a cognitive approach
MOESCHLER, Jacques
Abstract
The main purpose of this paper is to explore pragmatic misunderstandings caused by intercultural factors. My thesis is that misunderstandings are caused not by difficulty in drawing the intended implicature, but primarily by lack of access to the correct explicature of the utterance. The explicit vs. implicit nature of the conveyed meaning is the key to the explanation of pragmatic misunderstanding. This argument is based on some assumptions supported by Relevance Theory, namely the ostensive-inferential character of linguistic communication and the difference between explicature and implicature. Some examples of intercultural misunderstanding will be discussed, and some consequences drawn. For instance, one parameter that increases the risk of misunderstanding is the quality of the audience's linguistic knowledge. My hypothesis is that the greater the audience's mastery of the speaker's language, the greater the risk of intercultural mis- understanding. The reason is that speakers tend to attribute to non-native speakers cultural background which is in due proportion to their own mastery of language and therefore do not [...]
MOESCHLER, Jacques. Intercultural pragmatics: a cognitive approach. Intercultural Pragmatics , 2004, vol. 1, no. 1, p. 49-70
Available at:
http://archive-ouverte.unige.ch/unige:109768
Disclaimer: layout of this document may differ from the published version.
a cognitive approach*
JACQUES MOESCHLER
Abstract
The main purpose of this paper is to explore pragmatic misunderstandings caused by intercultural factors. My thesis is that misunderstandings are caused not by di‰culty in drawing the intended implicature, but primarily by lack of access to the correct explicature of the utterance. The explicit vs. implicit nature of the conveyed meaning is the key to the explanation of pragmatic misunderstanding. This argument is based on some assump- tions supported by Relevance Theory, namely the ostensive-inferential character of linguistic communication and the di¤erence between explica- ture and implicature. Some examples of intercultural misunderstanding will be discussed, and some consequences drawn. For instance, one parameter that increases the risk of misunderstanding is the quality of the audience’s linguistic knowledge. My hypothesis is that the greater the audience’s mas- tery of the speaker’s language, the greater the risk of intercultural mis- understanding. The reason is that speakers tend to attribute to non-native speakers cultural background which is in due proportion to their own mas- tery of language and therefore do not necessarily imply the right explica- ture of the utterance. This thesis will constitute the starting point for a useful contribution about what the minimal conditions for successful inter- cultural communication should be. In conclusion, a sketch of the empirical field for intercultural pragmatics will be given.
1. Introduction
In a workshop on commercial interaction, an expert on international ne- gotiation told the following story: diplomats had di‰culty in obtaining minimal positive results in negotiations with Iranian partners (this fact was independent of the political regime). The expert analysed the situa- tion in the following way: in ordinary negotiations, speakers tend to
obtain an agreement at the optimal level located between the minimal stage (floor) and the maximal stage (ceiling), with the main purpose of the negotiations being the adjustment of the negotiators’ reciprocal goals.
The shared presupposition is that an optimal negotiation process gives positive results for both negotiators. Our expert observed that in negotia- tions with Iranians this presupposition was not shared. On the contrary, ordinary Persian negotiations are driven by the necessity to triumph over the partner, which means that the foreign negotiator’s optimal intended stage cannot be reached and should be lowered.
What is the significance of this short story? The main point is that as soon as a non-Persian negotiator (in trade or politics) begins the nego- tiation process with a Persian partner, he must, in order to gain his opti- mal intended stage: (i) claim a higher optimal stage than the intended one, and (ii) manifest his disappointment to his partner at the end of the negotiation.
What kind of knowledge are we talking about here? It is certainly knowledge about language use and customs, that is, cultural knowledge.
The general setting of this kind of interaction is typical of intercultural exchanges, and thus the description and analysis of such data belongs to intercultural pragmatics.
As a first definition, I will therefore describe the domain of inter- cultural pragmatics as those facts implied by the use of language that do not require access to mutually manifest knowledge, but to specific con- textual knowledge necessary for understanding the speaker’s intention.
In other words, intercultural pragmatics aims at understanding the extent to which non-shared knowledge a¤ects and modifies the retrieval of in- tended meaning. I will give two more brief examples of such situations, the first one dealing with a non-intercultural misunderstanding, the sec- ond one with an intercultural misunderstanding.
One day, while writing this introduction at the Institute for Cognitive Studies in Lyon, France, I went out to buy lunch at a co¤ee shop near the Institute. Back at the Institute, I did not find the salmon quiche I had or- dered. I went back to the co¤ee shop, where the attendant said that she hadn’t put the quiche into the bag, but on the counter, so it would not be crushed by the beverages and sandwiches. I replied that she had not mentioned that to me, so I could not possibly know it. What was the cause of the misunderstanding? My attention was on the bag containing the beverage and the sandwiches, and not on the quiche on the counter, because no communicative act—that is ostensive act—was manifest to me. What surprised me was that this fact was not relevant to the waitress.
In other words, it was obvious to her that putting the quiche on the bar was mutually manifest. My hypothesis is that intercultural misunder-
standings are simply special cases of pragmatic misunderstandings like this one.
To what extent do intercultural misunderstandings di¤er from the situ- ation described above? My hypothesis is that the main di¤erence lies in the causes responsible for the erroneous inference. To be more precise, I will provide my final example, which demonstrates a true intercultural misunderstanding. Here is the setting: I was invited to Morocco to give a series of lectures at the University of Rabat. My plane was due to land at the Casablanca Airport, which is about a hundred kilometres from Rabat. Beforehand, I had emailed my Moroccan colleague asking:Can you tell me how to get from the airport to Rabat?Her answer was:You can take the train at the airport, with a change at Ain sbaaˆ station and you’ll arrive at the Rabat downtown station.
What was the outcome of this example? You might object that there was no misunderstanding, as I asked a question and received a very pre- cise answer. The problem, however, is that French native speakers tend to address their audience with indirect requests, either conventionally or conversationally. The more the request involves the audience, the more indirect the request will be.
The explanation is that my speech act was not merely a request for in- formation, it was a request for help:Could you please pick me up at the airport?The answer to my second request (I don’t know Morocco, I have no time to plan my trip, can you please come and pick me up at the air- port?) received a literal answer, like the first request: Someone will come and pick you up at the airport. So the answer was not delayed because no one wanted to help me, but because I did not ask for such help explicitly.
I will now give a general formulation of the causes of pragmatic mis- understanding in intercultural communication:
Intercultural misunderstanding (1)
In intercultural communication, misunderstandings are due to false inferences caused by false explicatures.
This hypothesis states exactly the contrary of what classical pragmatics would state about misunderstandings in general and intercultural ones in particular. In classical pragmatics, misunderstandings are not due to misunderstanding about what is said, but about what is implicated. Of course, in the example about Morocco, what I intended to communicate was not overt, but covert. But a problem arose when the addressee re- covered the following explicature: ‘‘Jacques Moeschler wants to know how to get from the airport to Rabat,’’ and stopped processing my words further.
This paper will argue and further support the thesis that intercultural communication is not made complex and risky because of its ostensive- inferential property (Sperber & Wilson 1995), but because speakers make false choices on the explicature/implicature status of their intended meaning. In section 2 I will address the question of why linguistic com- munication is a special case of ostensive-inferential communication. Sec- tion 3 shows how explicature/implicature distinction plays a crucial role in successful communication. In section 4, I will give some examples of pragmatic misunderstandings, and in section 5 I will return to the previ- ous examples and give a more precise analysis to them. Finally, as a conclusion (section 6), I will sketch out what might be the empirical field for intercultural pragmatics from a cognitive perspective.
2. Ostensive-inferential communication and pragmatics
Within cognitive pragmatics, as illustrated for instance by Relevance Theory (RT), one of the main concepts is ostensive-inferential commu- nication (Sperber & Wilson 1995). Linguistic communication is defined (RT) as a mixed process, implying both a coding-decoding device (the code model) and an inferential process based on old and new information (the inferential model, Wilson & Sperber 2004). In isolation, neither cor- rectly describes how linguistic communication works. Linguistic commu- nication is obviously based on a code, a specific language, defined as a set of sound-meaning pairs. However, knowledge of the code, although nec- essary for optimal linguistic communication, is not a su‰cient condition for successful communication. It is not a su‰cient condition primarily because the retrieval of the speaker’s intention implies much more than a common code: world knowledge, as well as knowledge about the situa- tion of communication, are crucial to the enrichment of linguistic mean- ing encoded within utterances.
The division of labour between these two models of communication can be explained in a functional way. From this point of view, it seems reasonable to devote the e‰cient and rapid linguistic parsing to a very specialised module (the linguistic module), without expecting a complete interpretation of the verbal stimulus. On the other hand, the capability of mixing outputs of the linguistic parsing (LOGICAL FORMS) with other sources of information (old information retrieved from long-term mem- ory, new perceptual information coming from the physical environment, mid-term memory information resulting from previous utterance parsing) can increase the e‰ciency of the cognitive system in inferring new infor-
mation, since the inferential process is not subject to the time pressure inherent to the linguistic parser.
Linguistic communication as defined previously has two aspects: it is ostensive on the one hand and inferential on the other. Ostensive com- munication is achieved by an act of ostension from the communicator, even if ostensive communication can be non-linguistic, that is, conveyed by gestures, glances, smiles, or any other stimuli made recognisable by the communicator as an intentional means to convey his intention. The use of language in utterances gives rise to ostensive communication be- cause the processing of an utterance follows the addressee’s recognition of the speaker’s intention to convey his informative intention in terms of an utterance. If language were a perfect communicative device, perhaps lin- guistic communication would be restricted to a special kind of communi- cation, that is, ostensive communication. Since speakers can convey more than what they say, as in (1), the ostensive dimension of verbal commu- nication is completed by inferential processes: the addressee must infer from what is said and from other accessible information (the context) the speaker’s intended meaning:
(1) Jacques:Axel, please go and brush your teeth!
Axel:Dad, I’m not sleepy.
In this case, the addressee’s recognition of the speaker’s communica- tive intention allows him to make inferences in order to understand the speaker’s informative intention:
‘‘Informative intention
The intention to inform an audience of something.
Communicative intention
The intention to inform an audience of one’s informative intention’’ (Wilson &
Sperber 2004: 611).
This analysis of communication as an ostensive-inferential device is an explicit version of Grice’s theory of non-natural meaning (Grice 1957), and is the core concept of any pragmatic approach to verbal communi- cation. The issue which divides pragmatic theories from each other (for instance Sperber & Wilson’s Relevance Theory from Levinson’s Theory of Generalized Conversational Implicature—TGCI—Levinson 2000) is the nature of the principles responsible for the processing of inferred meaning. For instance, RT resorts to the single Principle of Relevance, whereas TGCI triggers implicatures either from the Q-Principle (derived from the Gricean Maxims of Quantity, Grice 1975) or the I-Principle (principle of informativeness, derived from the Maxims of Quality and
the Maxim of Relation). In the course of this paper, I will restrict my discussion to ostensive-inferential communication as described in RT, and show how the Principle of Relevance plays a crucial role in the re- covery of the speaker’s informative intention.
As previously mentioned, RT claims that this criterion is the Principle of Relevance (PR). PR is a generalisation of the Gricean Maxim of Re- lation (‘‘Be relevant’’), and also includes the maxims of Quantity and Quality. Being relevant implies not only giving information about what is said in a conversation, but also giving an appropriate quantity of in- formation as well as satisfying the Gricean maxim of quality (Wilson &
Sperber 2000). For instance, giving the precise number of children in (2a) is more relevant than to give a true number, entailed by (2a) as in (2b):
(2) a. Anne has four children.
b. Anne has three children.
The same argument can be used for the maxim of Quality. For instance, asserting (3) while walking in the rain can be relevant even if the state- ment is false, if the speaker’s utterance mentions in an echoic way a pre- vious thought or utterance stated during sunny weather:
(3) What a beautiful sunny day!
In RT, relevance is a comparative concept, defined through the cognitive e¤ects produced by the utterance in a specific context, as well as through the cognitive e¤ort implied by the processing of the utterance:
Relevance
a. All things being equal, the more cognitive e¤ects an utterance produces, the more relevant it is.
b. All things being equal, the more cognitive e¤ort an utterance requires, the less relevant it is.
How can relevance and PR play the role of the expected criterion re- sponsible for the inference of the informative intention? In the first edition of Relevance (Sperber & Wilson 1986), the principle of relevance simply states that ‘‘every act of ostensive communication communicates the pre- sumption of its own optimal relevance’’ (Sperber & Wilson 1986: 158).
The recent version of RT, as stated initially in the second edition of Rel- evance (Sperber & Wilson 1995) has split PR into two principles, cogni- tive PR and communicative PR. Relevance is a concept which plays a role both in cognition and in communication, but its extension is not ex- actly the same. As far as cognition is concerned, the key concept ismaxi-
mal relevance: if cognition plays a crucial role in inference processes, then what triggers the search for the relevant information is the requirement for maximal relevance. In other words, human cognition is attracted by relevant information and is able to make a sharp distinction between rel- evant information and irrelevant information. But, the requirement for maximal relevance is balanced by what really happens in communication:
How can addressees look for maximal relevance if speakers have no means to bring forth, together with the utterance, relevant information, or if they are reluctant to do so? The answer is that the communicative PR simply states that the presumption of optimal relevance is constrained by the speaker’s abilities and preferences.
Here are the precise formulations of the cognitive and communicative PR:
‘‘Cognitive Principle of Relevance
Human cognition tends to be geared to the maximisation of relevance’’ (Wilson &
Sperber 2003: 610).
‘‘Communicative Principle of Relevance
Every ostensive stimulus conveys a presumption of its own optimal relevance’’
(Wilson & Sperber 2004: 612).
‘‘Presumption of Optimal relevance
The ostensive stimulus is optimally relevant to an audience if:
a. It is relevant enough to be worth the audience’s processing e¤ort.
b. It is the most relevant one compatible with communicator’s abilities and preferences’’ (ibid.)
This shows why the audience looks for a relevant interpretation, that is, an interpretation that resembles the speaker’s informative intention, even if his search for relevance can be complicated by the speaker’s abilities and preferences.
If we turn back to intercultural misunderstandings, we can draw a preliminary conclusion based on the general cause of ordinary misunder- standing:
Misunderstanding (general)
A misunderstanding is triggered, either intentionally or unintentionally, by the speaker’s abilities and preferences, which allow erroneous interpretation by the addressee.
Misunderstanding (intercultural)
An intercultural misunderstanding is due to an erroneous evaluation of the com- municator’s abilities and preferences by the addressee.
In order to refine this claim, I will now introduce an additional concept from RT, the explicature/implicature distinction.
3. Explicit and implicit communication
Since Grice, pragmatic studies on linguistic communication have tackled one major issue: explaining how and why speakers do not literally convey their informative intentions. Many solutions have been proposed, two of them being very popular in the literature.
The first solution (Stalnaker 1977; van der Auwera 1979) resorts to the notion of common background beliefto explain why linguistic communi- cation does not need to be literal and explicit. Information belonging to the common background is not to be reset, and thus makes communica- tion much more e‰cient.
The second solution (Atlas & Levinson 1981; Levinson 1983, 1987, 2000) relies on the notion of noncontroversial statement and common ground(maxim of relativity):
‘‘Maxims of Relativity
1. Do not say what you believe to be highly noncontroversial, that is, to be en- tailed by the presumptions of the common ground.
2. Take what you hear to be lowly noncontroversial, that is, consistent with the presumptions of the common ground.’’ (Atlas & Levinson 1981: 40)
Both theories imply a principle of economy in verbal exchanges: Speakers do not have to say what is presupposed to be true, that is, information belonging to the common ground. But neither of these maxims explains whyit is more economical and e‰cient to behave like this. Sometimes, it is not; manyunsuccessfulinstances of communications are caused by not having asserted what is presumed to belong to the common ground.
Moreover, reference to common ground implies that background in- formation is a necessary and su‰cient condition for successful com- munication. As Sperber and Wilson (1982, 1986) have shown, common ground defined as mutual knowledge can be neither a necessary condition (otherwise communication should always be successful) nor a su‰cient one (because background information can be inferred) for successful communication.
The answer to the question of non-literal communication must there- fore be looked for elsewhere. In RT, non-literal communication does not contrast with literal communication in terms of economy, but in terms of contingency. Literal communication, defined as total overlap between the set of implications drawn from the thought of the speaker and the set of implications drawn from the speaker’s utterance, is a very uncommon case. The usual situation implies a partial overlap between these two sets of implications (analytical and contextual). Hence, in verbal communica- tion, the normal state is one where the intended meaning is not literally
communicated (and therefore not fully economical), but pragmatically inferred from (and therefore contingent on) contextual information and the utterance.
The crucial point for my argument is that the intended meaning is in- ferred rather than conveyed literally. The point I would like to develop now is the nature of what is inferred. It will be shown that the nature of the inferred meaning is the key to the understanding of pragmatic misunder- standing in general, and to intercultural misunderstanding in particular.
Gricean and neo-Gricean pragmatics define inferred meaning as con- versational implicatures, either generalized or particularized. This defini- tion presupposes that what is said in the utterance is not the speaker’s meaning (what he wants to convey), but the sentence meaning (what his words linguistically mean, Searle 1979), even if the literal meaning (sen- tence meaning plus background knowledge) is a by-product of linguistic meaning and of background assumptions.
I would like to propose the following assumption: What is inferred is not restricted to implicatures, but contributes to the explicatures of the utterance (Sperber & Wilson 1986; Wilson & Sperber 2004). An explica- ture results from the enrichment of the logical form, that is, the proposi- tional form of the utterance. A propositional form is a complete proposi- tion, in which referents are attributed to referential expressions, and the sentence is disambiguated (basic explicature in Wilson & Sperber 2004).
The explicit part of the intended meaning can be completed by higher- level explicatures which specify the illocutionary force of the utterance and the propositional attitude of the utterance. In (4), (5a) is the basic explicature and (5b) and (5c) the higher-level explicatures:
(4) How should I go from Casablanca Airport to Rabat?
(5) a. Jacques is going from Casablanca Airport to Rabat at 10 p.m.
on Saturday, April 10.
b. Jacques is asking how to get from Casablanca Airport to Rabat at 10 p.m. on Saturday, April 10.
c. Jacques wants to know how to get from Casablanca airport to Rabat at 10 p.m. on Saturday, April 10.
Some aspects of the speaker’s informative intention are not overt, but covert, and need some additional contextual premises to be understood.
For instance, (6) is a typical implicated premise allowing the drawing of implicated conclusion (7):
(6) a. If Jacques is asking how to get from Casablanca Airport to Rabat, then Jacques does not know how to go to Rabat from the airport.
b. Jacques would prefer not to go to Rabat alone.
(7) Jacques is asking for someone to pick him up at the airport in order to go to Rabat.
The analysis that gives the status of the implicated conclusion in (7), that is, the implicature status, will be more thoroughly discussed in section 5. What is relevant here for my argument is that some implicatures are strongly implicated, whereas others are weakly implicated. In RT, an im- plicature is strongly implicated if ‘‘its recovery is essential in order to arrive at an interpretation that satisfies the addressee’s expectations of relevance’’ (Wilson & Sperber 2004: 620), whereas it is weakly implicated
‘‘if its recovery helps with the construction of such an interpretation, but is not itself essential because the utterance suggests a range of similar possible implicatures, any one of which would do’’ (idem).
So it seems that in order to recover the intended meaning, addressees have to pass through several stages, beginning with determining the ex- plicature, and proceeding to the implicatures (implicated premises and im- plicated conclusions). This procedure has been fully described in Wilson &
Sperber (2004: 615) as the sub-tasks in the overall comprehension process:
‘‘Sub-tasks in the overall comprehension process
a. Constructing an appropriate hypothesis about explicit content (EXPLICA- TURES) via decoding, disambiguation, reference resolution, and other prag- matic enrichment processes.
b. Constructing an appropriate hypothesis about the intended contextual as- sumption (IMPLICATED PREMISES).
c. Constructing an appropriate hypothesis about the intended contextual im- plications (IMPLICATED CONCLUSIONS).’’
The final relevance-theoretic concept that is required is a general com- prehension procedure. One crucial issue for pragmatic theory is to explain why and when addressees stop processing, that is, why they do not seek further weak implicatures, and at what point they stop processing. For instance, the question is why my addressee, when processing (4) How should I go from Casablanca Airport to Rabat? understood my informa- tive intention as restricted to (5b) Jacques is asking how to get from Ca- sablanca Airport to Rabat at 10 p.m. on Saturday, April 10and not to (7) Jacques is asking for someone to pick him up at the airport in order to go to Rabat. RT gives a comprehension procedure, the path of least e¤ort (Wilson & Sperber 2004: 613):1
‘‘Relevance-theoretic comprehension procedure
a. Follow a path of least e¤ort in computing cognitive e¤ects: Test interpretive hypotheses (disambiguation, reference resolutions, implicatures, etc.) in order of accessibility.
b. Stop when your expectations of relevance are satisfied.’’
We now possess all the necessary theoretical tools to return to our initial examples and to account for the nature of intercultural misunderstanding (section 5). However, before giving a pragmatic analysis of intercultural misunderstanding, I would like to provide a more precise description of how pragmatic misunderstandings work.
4. A relevance-theoretic analysis of misunderstanding
In this section, I propose a general model of pragmatic misunderstand- ings based on the hierarchy of levels of comprehension and on the RT- comprehension procedure. In order to arrive at a complete comprehen- sion of a speaker’s utterance, the addressee must access the following levels of comprehension:
Hierarchy of levels of comprehension Access the following layers:
a. basic explicature b. higher-level explicatures c. implicated premises
d. strongly implicated conclusion e. weakly implicated conclusion.
This hierarchy does not imply a linear order of linguistic and pragmatic processing. Its main function is to define the minimal conditions for suc- cessful communication. My hypothesis is that the determination of the explicatures (basic and higher-level) is a minimal condition for the re- covery of the speaker’s informative intention.
As a first illustration, I will give examples from ordinary communica- tion where these five levels of comprehension are not reached and whose main e¤ect is the failure of communication. I will begin with the basic explicature. Di¤erent possibilities occur: failure in decoding, failure in disambiguation, and failure in reference resolution, as shown in (8), (9) and (10)2:
(8) Anne:Peux-tu surveiller le bain?
Jacques:Le pain? Tu as fait du pain aujourd’hui?
Anne:Non, pas le pain, le bain des enfants.
(Anne: Can you keep an eye on the bath?
Jacques: The bread? Did you bake bread today?
Anne: No, not the bread, the children’s bath.)
(9) Anne: As-tu vu que le gouvernement fe´de´ral va donner 100 millions aux universite´s pour les e´quipements informatiques?
Jacques:100 millions pour CHAQUE universite´? Fantastique!
Anne:100 millions pour TOUTES les universite´s suisses.
(Anne: Have you heard that the federal government allots a hun- dred million francs to universities for their computer equipment?
Jacques: A hundred million for EACH university? Fantastic!
Anne: No, a hundred million for ALL the universities in Switzer- land.)
(10) Jacques:As-tu vu le nouvel habit de Nath? Il est e´le´gant.
Anne:Qui est e´le´gant? Nath ou l’habit ou les deux?
(Jacques: Did you see Nath’s new suit? He/It looks elegant.
Anne: Who is elegant? Nath or his suit or both?)
All these examples show that the mistakes occurring in linguistic de- coding, disambiguation, and reference resolution have immediate con- sequences on successful communication, and can therefore be quickly repaired.3(8) is a classical confusion between a voiced (bain) and an un- voiced (pain) consonant. Notice that the linguistic context (to keep an eye on) makes both interpretations plausible. In (9), the world knowl- edge should point to a collective interpretation (‘‘Swiss universities as a whole will receive a hundred million’’) and not to the distributive reading (‘‘Swiss universities will each receive a hundred million’’), though the lat- ter is semantically possible. Finally, in (10) a classical reference ambiguity (possible in French because of the unmarked opposition of human/non human in French gender morphology) cannot be resolved linguistically.
These examples all require more than simple linguistic decoding: the access to the relevant context, to some information about the world, and to the speaker’s intention favour one or the other reading. But the crucial fact is that in cases where the basic explicature is not correctly deter- mined, comprehension on the part of the addressee does not correspond to the speaker’s informative intention. It can be concluded that the de- termination of the basic explicature is a necessary condition for successful communication. The correlate of this conclusion is the following: in in- tercultural communication, the addressee must have access to the basic explicature in order for that communication to go on. However, the most important problems in intercultural communication arise not from missed basic explicatures, but from missed higher-level explicatures.
Let us give some examples of deficient higher-level explicatures. The first concerns the determination of the illocutionary force; the second concerns the attribution of propositional attitude.
(11) Father:Don’t you have any homework for Monday?
Son:No.
Father:I’m not asking that, I want you to study this weekend.
(12) Father:I want you to study more seriously during the last semester.
Son:OK, dad, I will try.
Father:I’m not asking you to try to work, but to work.
In (11), the intended illocutionary force is not correctly understood by the son. As semanticists have shown, yes/no questions are oriented. For in- stance, positive questions are negatively oriented (they presuppose the falseness of the proposition), whereas negative question are positively oriented, as in (11) (cf. Anscombre & Ducrot 1983). What happens in (11) is that, since the question is positively oriented (you have some homework for Monday), the illocutionary force conveyed is not a request for infor- mation, but a request for action (the speaker wants his audience to study on the weekend).
In (12), the son’s propositional attitude is not the expected one. When someone is asked to do something, his mental state cannot simply corre- spond to a willingness to try: he should be willing to actually do it. That is exactly what the father’s reply wants to convey.
The status of higher-level explicatures is slightly di¤erent from basic explicatures because the illocutionary force and the propositional attitude are not conventionally implied by any particular linguistic form. The il- locutionary force is generally inferred, and the risk of not understanding the precise illocutionary force of the speech acts is rather high. The prop- ositional attitude is determined by the attribution of an illocutionary force to the utterance. So background information is crucial for this level of information, even if higher-level explicatures are not implicatures in relevance-theoretic terms. In the next section I will bring further evidence that higher-level explicatures are the Gordian knots of intercultural com- munication and misunderstandings.
Before presenting a general explanation of intercultural misunder- standings, I will examine the cases of implicature from (c) to (e) in the hierarchy of levels of comprehension. In RT, an implicature can be a premise or a conclusion, and an implicated conclusion can be strong or weak. Some examples of pragmatic misunderstanding based on implica- ture failure follow:
(13) A: My son voted for the first time in his life.
B: But he is still in high school.
A: Yes, but he is 18 now.
(14) A: My son has got his baccalaure´at.
B: So he is going to university next fall.
(15) A: My son has got his baccalaure´at.
B: I guess you’re not going to buy a new car next fall.
In (13), the required premise for A’s first move is ‘‘A’s son has reached the age of 18 and can vote.’’ This premise is made explicit to B in A’s second move, which reminds B that A’s son is older than he thought.
Notice here that implicated premises do not need to be new information.
In general, implicated premises are reminders of old information, which are not accessible in the current speech situation for some reason.
In (14), there is neither entailment nor a logical link between A’s and B’s utterances. Nevertheless, what B understands is that A’s intention is not only to inform B about a new fact, but about its practical con- sequences as well. In a context where secondary school leads to univer- sity, B is a relevant move making A’s implicature explicit. If B is true, and corresponds to A’s informative intention, then B’s utterance is the ex- plicature of what A’s utterance strongly implicated.
Suppose now that in the same context (A and B are talking about their children’s’ futures), B answered (15B) instead of (14B) to A. (15B) may be true, but it is B’s responsibility to draw this conclusion. In relevance- theoretic terms, B makes explicit what A weakly implicates. If A and B are visiting the Geneva car exhibition, (15B) could be the explicature of A’s strong implicature.
The di¤erence between strong and weak implicatures is a question of context and intentions. Thus, on the one hand, A strongly implicates B if A belongs to the set of propositions the speaker wants to communicate.
On the other hand, A weakly implicates B if B does not belong to the set of the propositions the speaker wants to communicate.
It is now possible to make several predictions about the crucial layers of meaning which play a role in intercultural communication:
Prediction about intercultural communication
1. Basic explicatureis the minimal level of communication. If the development of basic explicature fails, then ordinary misunderstanding will occur.
2. Higher-level explicatureis the middle level of communication. If the develop- ment of higher-level explicature fails, then strong misunderstanding will occur.
3. Implicated premiseis the first higher-level of communication. As implicated premises require knowledge of the world, access to implicated premises can be made more di‰cult when social, cultural, and behavioural assumptions di¤er among speakers.
4. Implicated conclusionis the second higher-level of communication. As impli- cated conclusions are about speakers’ intentions and the strength of their in- tended meaning, failure in recovery of strong implicature is worse than failure in recovery of weak implicature.
In other words, I predict that failure may occur atanyof these levels, and that failure at a lower level is easier to resolve than failure at a higher-
level. A minimal requirement for successful intercultural communication is the correct identification of basic and higher-level explicatures. The main purpose of the next section is to test this prediction.
5. A relevance-theoretic analysis of intercultural misunderstanding As stated in the beginning of this paper, the first hypothesis which comes to mind is exactly the opposite: viz., that the cause of intercultural mis- understanding is a failure to recover implicatures.4I would now like to provide evidence for the main thesis of this paper and show that what speakers engaged in intercultural communication primarily need is to ac- cess both basic and higher-level explicatures. In other words,explicatures, not implicatures, are the key level for communication in general and for intercultural communication in particular.
Let us begin with another example of intercultural misunderstanding, which could have had serious consequences. Anyone who is travelling to Kenya is warned about two things: vaccinations and urban violence, es- pecially in Nairobi. As a guest at the University of Nairobi, I was warned about and became aware of these two dangers. Having a whole afternoon to spend in Nairobi, I asked my host the following question and received the following answer:
(16) A: Can I go shopping downtown?
B: No problem, downtown is dangerous at night.
Here is what happened that afternoon. I left my key at the hotel reception and walked to the exit. The receptionist ran after me asking:
(17) C: Mr. Moeschler, where are you going?
A: Downtown. My host told me it was safe.
C: Please, don’t walk alone. I can order a car for your shopping at the Hilton.
The problem resides in B’s answer in (16). Was B lying to me, or was he ashamed of answering negatively? These are not very reasonable re- sponses, because the consequences of B’s utterance were too important. I would like to propose another answer, based on the implicated premises of B’s utterance (18):
(18) a. Nairobi is not dangerous in daytime.
b. Nairobi is dangerous for white women in daytime.
C’s implicated premises in (17) run counter to (18), and more precisely (18a):
(19) Nairobi is dangerous for white men.
What is at stake in this example? It seems to me that one aspect of ‘‘what is dangerous’’ for a human being in Nairobi was not grasped by B. His conceptual hierarchy of danger was the following:
(20) night>day (white woman)
On the contrary, the receptionist’s hierarchy was di¤erent:
(21) tourists>natives
As a non-native, I was an ideal target for theft or other events. This example could confirm the classical pragmatic view assigning a crucial role to implicatures in intercultural communication, and also provide an argument against the explicature-thesis. I will depart from this, and make my intention while uttering (16A) more explicit. In fact, this utter- ance came after some discussion of the afternoon’s events, and my host was supposed to come shopping with me downtown. Thus, my ques- tion was NOT a request for information, it was a request for safety war- ranty. What happened was that this high-level explicature (to be retrieved through disambiguation) was not grasped because of the implicated premise (20). The simple conclusion that can be drawn from this example is that, although implicated premises play an extremely crucial role in the understanding of the speaker’s informative intention, the role played by explicatures may be even more crucial.
Now, I would like to turn to the airport example, by giving it a more complete description. Here is the more complete record of the exchange of emails:
(22) A: Bonjour, ma re´servation d’avion est faite. J’arrive a` Casa- blanca le 10 avril a` 20h40, et je repartirai de Casa le 14 a` 14h.
Pouvez-vous me dire comment aller de l’ae´roport a` Rabat? Je compte sur vous pour les re´servations d’hoˆtel ou de logement a`
Rabat. . . .
B: . . .Pour ce qui est du transport de l’ae´roport de Casa a` Rabat, vous pourrez prendre un train a` l’ae´roport, avec un changement a` la gare de Ain sbaaˆ et vous arriverez a` la gare de Rabat-ville a` 2 mn de l’Hoˆtel Terminus ou` une chambre vous est re´serve´e.
(A: Hello, my plane reservations have been made. I will arrive at Casablanca on April 10 at 8:40 p.m., and will leave Casa on 14 April at 2 p.m. Can you tell me how to get from Casa- blanca Airport to Rabat? I’m counting on you for the hotel reservation at Rabat.
B: . . . Concerning travelling from Casa Airport to Rabat, you can take the train at the airport, with a change at Ain sbaaˆ station and you’ll arrive at Rabat downtown station, at 2 mn from the Terminus Hotel where a room has been booked).
I would like to make two preliminary remarks. First of all, the exchange was rather explicit, and my Moroccan colleague’s mastery of French was perfect. But the outcome of this example must be recalled. As a guest in Morocco, I did not know exactly how to proceed, and I was trying to get my addressee to understand my illocutionary point (Searle 1979), that is, to understand my utterance, repeated in (23), as conveying (24):
(23) Can you tell me how to get from Casablanca Airport to Rabat?
(24) Can you pick me up at the airport to go to Rabat?
Here again, the ‘‘implicated premise’’ thesis could be mentioned as an explanation of why (23) conveys (24) through the implicated premises (25):
(25) a. Someone arriving in a foreign country needs some help.
b. To travel downtown alone from the airport at night is not a good idea.
c. To ask how to go from A to B is to ask for some help to go from A to B.
The crucial implicated premise is of course (25c), and at least in Western European culture, the role of the host is to manage and keep practical worries as minimal as possible.
The question is thus the following: Why, in spite of the high accessibil- ity of (25)5, is (24) not answered, and certainly not grasped? In other words, why is the implicated conclusion (26) not inferred?
(26) Jacques is asking for someone to pick him up at the airport to go to Rabat.
Here is the answer. In order to understand this implicated conclusion, it would have been necessary for my addressee not to stop processing after obtaining the higher-level explicature (27):
(27) Jacques is asking how to go from Casablanca Airport to Rabat.
It is important to remember the path of least e¤ort which directs prag- matic processing and its second clause: ‘‘Stop when your expectations of relevance are satisfied.’’ This leads to a preliminary answer: As soon as he grasped the higher-level explicature (27), my addressee achieved su‰cient relevance to balance his cognitive process.
This analysis explains why a literal interpretation, based on a higher- level explicature, does not yield the implicated conclusion from the higher-level explicature and the implicated premise. But it does not say anything about the reasons why the speaker does not explicitly ask for some help at the airport, if he expects this.
Two reasons can be mentioned here. First, the speaker might be reluc- tant to express his wishes explicitly. Second, he may have thought that his intention was clear enough to be understood. Although the first reason is plausible, I will concentrate on the second one, which is much more interesting. There is an obvious fact that makes intercultural communi- cation much more risky than in ordinary exchanges: speakers share a higher-level use of language without belonging to the same culture. This fact, far from harmless, can be fraught with danger. My hypothesis is therefore the following:
Intercultural misunderstanding (2)
In intercultural communication, the higher the level of mastery of the shared lan- guage, the greater the risk of attributing to the addressee the same beliefs and knowledge as one’s own.
In this respect, the first definition for intercultural misunderstanding re- peated here now receives an explanation:
Intercultural misunderstanding (1)
In intercultural communication, misunderstandings are due to false inferences caused by false explicatures.
If my addressee arrived at a relevant interpretation by inferring a higher- level explicature, his answer had as main consequence to avoid a further misunderstanding, because I realised that my addressee’s answer was not a refusal to assist, i.e., he reached the appropriate higher-level explicature and went on to draw the appropriate implicature.
I will now propose a third and final definition of intercultural mis- understanding:
Intercultural misunderstanding (3)
Intercultural misunderstandings occur when false assumptions lead to false higher-level explicatures. False inferences deriving from higher-level explicatures are caused by false attributions of shared beliefs and knowledge.
In other words, intercultural misunderstandings do not simply come up because speakers do not share common beliefs and knowledge, but be- cause they attribute beliefs and knowledge to each other which they in fact do not entertain. I want to emphasize that this situation is reinforced when speakers share a common language at a higher-level.
6. Conclusion: The empirical field of intercultural pragmatics
In conclusion I would like to make some proposals about what could be the empirical domain of intercultural pragmatics. Though I have given a general sketch of how intercultural misunderstandings could occur, in this last section, I would like to make some suggestions about how in- tercultural pragmatics could define new empirical fields for pragmatic studies.
As a guideline, I will use the ‘‘sub-tasks in the overall comprehension process’’ given in section 3, that is, the layers of EXPLICATURE, IM- PLICATED PREMISE and IMPLICATED CONCLUSION.
To me, the first layer (explicature) is the core layer for investigating in- tercultural pragmatics. Following are some arguments which show that it is a central issue.
The first domain to investigate is to what extent, in a given situa- tion, speakers of di¤erent languages and cultures will use di¤erent means to convey their intention either explicitly or implicitly. For example, French culture is based on a way of communicating with implicatures whereas American culture tends to expect speakers to express their in- tentions much more explicitly. This issue is an extremely empirical one, and should be investigated with sound comparative methods. Diplomatic negotiation, trade, academic cooperation, as well as social contacts could benefit from such research.
The second domain of investigation would be that of indirect speech acts and the conventional/conversational way of conveying illocutionary forces. Here the possible convergence and divergence between languages and cultures allows the investigation of the relationship between types of illocutionary force (for instance, why and when a request for information becomes a request for help, etc.).
The second layer isimplicated premise. In order to understand speakers’
meaning it appears worth concentrating on the types as well as on the nature of implicated premise under specific settings (family, social rela- tions, professional relations, politics, etc.). For instance, because of the ubiquitous nature of Hollywood films, a person knows how to behave if an American policeman orders him to stop his car. But who would un- derstand when a Kenyan policeman said:
(28) A page of your driving license is missing.
What is missing, in fact, is a bank note, but this cannot be understood unless the following implicated premise is accessible:
(29) A bribe is usually requested by a policeman.
The third layer isimplicated conclusion. The relevant question here is not the nature of what is communicated explicitly or implicitly, as was the case of the first example, but the nature of what is strongly or weakly implicated. Here intercultural pragmatics has the same goals as inferen- tial pragmatics, as much work has been focused on the conventional/
generalized conversational nature of implicatures. These aspects of prag- matic meaning are not simply lexical or non-lexical; they imply knowledge about the contexts in which they can be used. I would like to give a final example which illustrates one aspect of strong implicatures conveyed by the speaker and which corresponds to a bad lexical choice.
A Swiss-born and American-educated student in Geneva was sitting at a table over a cup of co¤ee with some friends. Suddenly she stood up and said before leaving:
(30) Bon, je me casse.
(Well, I’m splitting)
The problem is that the verbse casserstrongly implicates that the person who is leaving is unhappy, angry, furious, etc. Se casseris a special way of leaving social groups, especially those including friends. This implica- ture was perfectly understood by the French native speakers who were sitting at the table with her, but she clearly did not intend to leave in an angry way. What she meant to strongly implicate, through a familiar way of speaking, was her very friendly relationship with her friends. But she failed, not because she did not know the context or the meaning of the word, but because she did not know what it implicated.
*Acknowledgement
Many thanks to Maria Hadjimarkos, Gabriela Soare and Julia Coryell for their revisions, to the anonymous reviewers for their comments and stylistic revisions and to my Moroccan colleague for her very useful comments on example (4).
Notes
1. In the classical version of RT, the path of least e¤ort was the notion of an interpretation consistent with the principle of relevance: ‘‘Let us say that an interpretation isconsistent with the principle of relevanceif and only if a rational communicator might have ex- pected it to be optimally relevant to the addressee’’ (Sperber & Wilson 1986, 166).
2. For convenience reasons, I give examples in French with their translations.
3. One conventional way to uphold mistakes in decoding, disambiguation and reference resolution is to create aquiproquo(mistaken identity), which is a very classical routine in French comedy (from Molie`re to Feydeau). The resolution of thequiproquois inten- tionally postponed because of its function in the plot.
4. The implicature-thesis would imply that
a. implicated premises are necessary to draw implicated conclusions, and
b. strong and weak implicatures correspond to the speaker’s informative intention.
This thesis locates intercultural misunderstandings (IM) in the non-mutual access of implicated premises, and describe IM as caused by di¤erent sets of (cultural) back- ground knowledge. Conditions (a) and (b) make intercultural communication very improbable and di‰cult to success.
5. This is a minimal presumption due to the intentional stance (Dennett 1987). We give here our formulation taken from Reboul & Moeschler (1998: 47):
‘‘The intentional stance (strate´gie de l’interpre`te) consists of an individual predicting others’ behaviour from two simple premises:
1. Others are rational agents.
2. They have beliefs, desires and other mental states.’’
References
Anscombre, Jean-Claude and Oswald Ducrot. 1983. Interrogation et argumentation. In L’Argumentation dans la Langu. Bruxelles: Mardaga. 115–137.
Atlas, Jay D. & Stephen C. Levinson. 1981.It-clefts, informativeness, and logical form:
Radical pragmatics (revised standard version). In Peter Cole (ed.),Radical Pragmatics.
New York: Academic Press. 1–61.
Auwera, Johan van der. 1979. Pragmatic presupposition: Shared beliefs in a theory of irre- futable meaning. In Choon-Kyu Oh and David A. Dinneen (eds.),Syntax and Semantics 11: Presupposition. New York: Academic Press. 249–264.
Dennett, Daniel. 1987.The Intentional Stance. Cambridge, MA: The MIT Press.
Grice, Herbert P. 1957. Meaning.The Philosophical Review57: 377–388.
— . 1975. Logic and conversation. In Peter Cole and Jerry L. Morgan.Syntax and Semantics 3: Speech Acts. New York: Academic Press. 41–58.
Levinson, Stephen C. 1983.Pragmatics. Cambridge: Cambridge University Press.
— . 1987. Minimization and conversational inference. In Jef Verschueren and Marcella Ber- tucelli-Papi (eds.),The Pragmatic Perspective. Amsterdam: John Benjamins. 61–129.
— . 2000.Presumptive Meanings. The Theory of Generalized Conversational Implicature.
Cambridge, MA: The MIT Press.
Reboul, Anne. 2001. Aux sources du malentendu. InLe Langage, Nature, Histoire et Usage, Paris, Editions Sciences Humaines. 67–72.
Reboul, Anne and Jacques Moeschler. 1998.Pragmatique du Discours. Paris: Armand Colin.
Searle, John R. 1979.Expression and Meaning. Cambridge: Cambridge University Press.
Sperber, Dan and Deirdre Wilson. 1982. Mutual knowledge and relevance in theories of comprehension. In Neil V. Smith (ed.),Mutual Knowledge. New York: Academic Press.
61–85.
— . 1986.Relevance. Communication and Cognition. Oxford: Basil Blackwell, 1stedition.
— . 1995.Relevance. Communication and Cognition. Oxford: Basil Blackwell, 2ndedition.
Stalnaker, Robert. 1977. Pragmatic presupposition. In Andy Rogers, Bob Wall, and John P.
Murphy (eds.),Proceedings of the Texas Conference on Performatives, Presupposition and Implicatures, Arlington, Center for Applied Linguistics. 135–147.
Wilson, Deirdre and Dan Sperber. 2000. Truthfulness and Relevance,UCL Working Papers in Linguistics12: 215–254.
— . 2004. Relevance Theory. In Laurence R. Horn and Gregory Ward (eds.),Handbook of Pragmatics. Oxford: Blackwell. 607–632.
