Understanding Discourse – Grice and Implicatures Part 2

In my last post, I looked at Grice’s Co-operative Principle (CP) and how the observance or flouting of the maxims creates meaning and implicatures in interaction.

In this post, I will look at the criticisms of Grice’s theories and the alternatives to CP that were proposed. I will also examine the universality of CP and how they might or might not be applicable across cultures, with a special section dedicated to the use of irony and British humour.



3.      Criticisms of Grice

Critics say that Grice’s maxims are not only inapplicable ubiquitously, but that he had not used the terms in 2.2.4 consistently. CP has also been criticised on a number of counts: lack of distinctions between the maxims and between types of non-observance when calculating implicature, and the inability to withstand evidence of real language use (Thomas 1995, Gazdar 1979, Leech 1983). However, the most substantial criticisms are as follows.


3.1     Relevance Theory


Sperber and Wilson (2004) questioned the need for CP’s maxims, saying that expectations of relevance alone are enough to guide the hearer towards speaker meaning. They suggest that the search for relevance is basic to human cognitive systems (Cognitive Principle), and when utterances are made, interlocutors combine the input with available background information, while using the least processing effort required, to derive meaning. Consider:


(xiv)     I’ll google it.


Using the input plus the background knowledge (which can also be described as one’s schema of the internet) that ‘Google’ is an online search engine, one understands that the speaker intends to search online for the information. Grice was criticised for failing to address such loose uses of language, and for treating metaphor, hyperbole and irony equally as flouting of the Quality maxim (ibid). Relevance theorists claim that besides irony, all the above loose uses of language are used to convey optimal relevance more economically (ibid), while irony involves ‘an expression of tacitly dissociative attitudes’ (ibid:272) and requires a higher order of meta-representational ability.

Not believing in the co-operative nature of humankind, Sperber and Wilson (1986) suggests a Communicative Principle where the audience would only pay attention if the stimulus is worth processing and relevant enough, and that the speaker, might have to be capable and willing to draw the hearer’s attention to his/her intentions, away from competing stimuli. Grice was criticised for seeing speaker unwillingness as a violation and thus not conveying implicatures (ibid 2004), although I worry about such a misrepresentation, considering that a violation simply comes from a speaker not wanting to convey the implicature, and does not mean the hearer cannot infer from the unwillingness. Arguably, Grice tended to focus on implicature generation and less on hearer interpretation strategies.


3.2     Politeness Principle (PP)


(xv)      I don’t suppose you’ve time to spare?


Based on claims that sentences do not always have information-bearing functions, PP is proposed as a necessary complement to CP (Leech 1983:80), to explain indirect sentences like (xv) which seem to violate the quantity maxim.  While CP enables communication through the assumption of cooperativeness, PP allows for such co-operation by regulating social equilibrium and relations (ibid:82). Thus, PP is seen to override CP at times.

The Ironic Principle (IP), however, allows for PP to be exploited in order to uphold CP (ibid:82-83). In (x), Sue is afraid of causing offence to Brian and makes light of Brian’s flaws, allowing him to infer her real meaning through indirectness, by way of irony, ‘an “honest” form of apparent deception, at the expense of politeness’(ibid:83).

While CP has been criticised repeatedly for not being universal, PP allows for the study of how such principles are variable on different dimensions and are exploited differently in different societies (ibid:84). Nevertheless, how much of CP is non-universal?


3.3     The universality of CP


I have found disagreement over whether Grice believed in the universal application of CP. Spencer-Oatey and Jiang (2003) claimed that according to Grice, they were universal principles of language use. Leech (1983:80) believed that no claim has been made that CP applied in the same way everywhere. Sperber & Wilson (1987) and Davis (1998), however, both think of implicatures as social conventions, and therefore, interlinguistic, and lists quantity implicatures and irony as common to many languages (ibid:186). Gazdar (1979:54), on the other hand, takes Grice’s claim that CP is something ‘reasonable for us to follow’ to mean that the nature of the maxims are universal, and argues that Grice’s maxims ‘cannot be defended as universal principles of conversation’ (ibid:55). Clyne (1994:12) believes the maxims are anglo-centric, of limited relevance, and need reformulation to take non-English cultures into consideration.

Examples in 2.2.1 showed how some adherence to the maxims is already instinctive in my learners, but consider (iii) again. In parts of Northern Greece, ‘children’ conveys the GCI of ‘sons’. When interlocutors do not share the same schema of the word ‘children’, misunderstandings could arise. Considering the four factors Grice (1975) purported interlocutors use to communicate through implicature: the literal meaning of the utterance, the roles and expectations of participants, the situation, context and nature of the conversation, and the world around the participants, we can deduce that people from different cultural backgrounds would have different expectations, roles and world views, and see the contexts and nature of conversations differently (Bouton, 1999).  Van Dijik’s (1977 in Brown and Yule 1983) Assumed Normality of the World suggests that our reactions to particular communicative situations are learnt through our experience of interpreting them in prior similar contexts. If the respondent’s sociocultural (Sociocultural abilities refer to the respondent’s skills at selecting speech act strategies appropriate to the culture involved, the age and sex, the social class and occupations and roles and status of the participants in interaction) and sociolinguistic abilities (Sociolinguistic abilities refer to the respondents’ skill at selecting linguistic forms to express a particular speech act strategy and their control over utterance’s register of formality) must be considered for the success of speech acts (Cohen, 1996), so it must be when interpreting implicatures.


3.3.1    Bouton’s cross-cultural study

Bouton (1988) found a significant difference between the way NS and non-native speakers (NNS) interpreted the implicatures presented within contextualised dialogues (the dialogues were presented on paper and lacked the paralinguistic and non-linguistic features that would normally be present when interpreting implicatures. This may have had some effect on his findings.), with the Germans/Spanish/Portuguese having more similar scores to NS than the Chinese/Japanese. We could perhaps conclude that certain cultures have more similar expectations and world views to the target culture than others.

Subsequently, Bouton (1994) noted that without explicit instruction, there was increased mastery of implicature types as time passes (particularly, those based on flouting the relevance maxim), although progress was slow after 17 months, and irony remained a problem even for those immersed for 54 months. The types of implicatures that remained difficult for those immersed for more than 17 months, however, were the ones that improved when a separate group was given explicit classroom instruction. Such instruction, conversely, did not help them improve on relevance-based implicatures.

Bouton (1999) believed implicatures differed in their opaqueness to NNS. Relevance-based implicatures required a lot of background information to interpret (idiosyncratic implicatures) and seemed impervious to teaching efforts, but depended on the building up of native-culture schemata over time. Formulaic implicatures, such as PopeQ (Bouton (1988) uses the term PopeQ to refer to typical ironic questions used in answer to another question. Other examples include, ‘Can ducks swim?’ and ‘Do bears sh*t in the woods?’) [see (xvi)], indirect criticism [see (viii)] and irony, however, have pattern-based structural or semantic clues, and these patterns can be taught, recognised, and used.



Deb:    Do you like ice-cream?

Derren: Is the pope Catholic?


3.3.2    Irony and English humour


Much cross-cultural misinterpretation of implicatures concerns sense of humour. Consider:



(Pointing to a bottle of Coke on the table)

Nao:    Is this coca-cola?

Me:     It’s a pizza.


The Japanese Nao, who thought I had meant to treat her like a fool, was puzzled. Some months later, Nao said, ‘I understand! British humour is about bullying others!’ Both the Japanese and Chinese have one expression to mean irony and sarcasm, and being ironic carries a negative connotation of insult. Although Nao had identified the semantic pattern in irony, her sense that the English found humour in bullying others was misplaced.

Anthropologist Fox (2004:65-66) notes that the English treat irony as ‘a constant, a normal element of ordinary, everyday conversation’ and the ‘dominant ingredient in English humour’. What makes irony even more difficult for foreigners is that a deadpan face is the expected norm. Fox sympathises with foreigners, admitting that self-parodying is part of the English psyche deeply schematised in the culture.
If spoken irony is difficult for NNS, written irony poses a bigger problem. Hatim (1997) attributes irony to the English preference for understatements and the cryptic, enabling one  to express an attitude without saying very much. Irony is hard to preserve when translating written texts. Hatim (ibid:196) suggests that Arabic is intolerant to such opaqueness, with translators of irony needing to flout the quantity maxim by being over-informative.


3.3.3    Making CP universal


Keenan’s (1976) investigation in Malagasy culture has spurred criticisms that Grice’s principles were monocentric and full of assumptions based on Anglo-Saxon norms and culture (Wierzbicka 1985, Gazdar 1979). Keenan (1976) admits the Malagasy community are not uninformative, but explains that Grice’s maxims do not hold in some societies. However, perhaps it is the critics themselves who have been monocentric and have interpreted Grice’s words from an English perspective. Arguably, being ‘as informative as required’ would stand with the Malagasy speakers if the ‘information’ expected and the schema of what ‘information’ entails differs in their culture. I believe CP could be realised in different cultures in different ways. Consider.


(xviii)               Returning an unidentified missed call,

Singaporean: Hello. Who called?


Clyne (1994:192) states that in both European and East/South-east Asian culture, the more information/knowledge provided, the better. Singaporeans, however, sometimes seem comfortable with the bare information minus the niceties, so we could argue that what is ‘required’ depends on cultural variation and situational expectations. Clyne (ibid:194) proposes a set of revised ‘maxims’ (Appendix) to make CP more universal by considering different cultural norms and expectations when applying Grice’s principles. How can teachers then apply this knowledge to help students adapt to a foreign environment?



In the last part of this series, I will be looking at the implications that Grice’s theories have on teaching and asking for ideas as to how this could be relevant to what we practitioners do in the language classroom.




Clyne (1994)’s revised maxims

Quantity:     A single maxim – ‘Make your contribution as informative as is required for

the purpose of the discourse, within the bounds of the discourse parameters of the given culture.’


Quality:        Supermaxim – ‘Try to make your contribution one for which you can take

responsibility within your own cultural norms.’

Maxims (1) ‘Do not say what you believe to be in opposition to your cultural

norms of truth, harmony, charity, and/or respect.’

(2) Do not say that for which you lack adequate evidence.’[1]


Manner:       The supermaxim can be retained in its original form – ‘Be perspicacious.’[2]

Maxims (1) ‘Do not make it any more difficult to understand than may be

dictated by questions of face and authority.’

(2) Avoid ambiguity unless it is in the interests of politeness or of

maintaining a dignity-driven cultural core value, such as harmony,

charity or respect.’

(3) ‘Make your contribution the appropriate length required by the

nature and purpose of the exchange and the discourse parameters

of your culture.’

(4) ‘Structure your discourse according to the requirements of your



Clyne, M.G. (1994) Inter-cultural communication at work: cultural values in discourse. Cambridge: Cambridge University Press.



[1] Maxim (2) of Quality and Maxim (2) of Manner have disclaimers put onto them.

For more detail, see Clyne (1994:194).

[2] I believe there is a typing error here. Grice’s original maxim of Manner was to be ‘perspicuous’, rather than ‘perspicacious’. This slightly changes the meaning of the maxim, which I assume was not the intention.


Author: chiasuanchong

I am a freelance communications trainer and a teacher trainer based in York, UK. With 13 years of experience training students from all over the world to communicate better in English (and in particular, Business English), I am also a professional blogger, materials writer and intercultural trainer.

2 thoughts on “Understanding Discourse – Grice and Implicatures Part 2”

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s