On the Role of Mutuality in Ironic Communication and Cognition

ZENG Yan-tao,ZENG Ha-dai,LIU Yue-mei

When ironic communication occurs， there must be a basic cognitive environment that may include inexhaustible elements， among which is a suggestion of complicity between the ironist and the hearer， which makes the process simple， convenient and fast. Complicity refers to the mutuality tacitly established or implied between the participants. This stduy explores its components， necessity and possibility in processing ironic communication and cognition.

Keywords： cognitive environment， mutuality， irony understanding

When ironic communication occurs， there must be a basic cognitive environment consisting of infinite possible information that includes such elements as participants and their mutual relations， norms， preconditions， personal characteristics， social situation， cultural environment， immediate context， etc.. For an utterance to be recognized as ironical， it seems necessary for a hearer to process all the infinite information before he arrives at the correct interpretation of it. In fact， however， it is not the case that all the inexhaustible factors enter the process. What occurs seems not so complex and arduous. Instead， the interpretation process is simpler. This is due to the fact that there seems to be a suggestion of complicity between the ironist and the hearer. Complicity， as a condition for irony recognition， is an external enabling ground for irony. The speaker who makes an utterance in an ironical manner may assume that he and the hearer share some assumptions about what is likely to be ironical and the hearer can select from the cognitive environment enough relevant contextual information which provides clues to the processing of irony understanding. The hearer assumes （and also believes） the contextual clues he chooses for processing are shared and mutually known by both the speaker and his audience. Stalnaker （1991） had a statement which he presented as a truism： “Communication， whether linguistic or not， normally takes place against a background of beliefs or assumptions which are shared by the speaker and his audience， and which are recognized by them to be so shared.… The more common ground we can take for granted， the more efficient our communication will be. And unless we could reasonably treat some facts in this way， we probably could not communicate at all” （p. 315）. We take it to be uncontroversial that this statement is appicable to ironical communication and cognition and the process of understanding ironical utterances is one with mutuality as the cognitive precondition.

Mutuality refers to the characteristic feature of a cluster of concepts that include mutual knowledge， common knowledge， mutual belief， mutual manifestness， and shared understanding. These concepts play a crucial role in the analysis of such notions as trading and bargaining （Auman， 1970； Milgrom， 1981）， norm， social practice， rule， role， social group and organization （Bach & Harnish， 1979）， definite reference （Clark & Marshall， 1981）， meaning （Schiffer， 1972）， convention （Lewis， 1969） and distributed processing （Halpern & Moses， 1984）. The analysis of cognitive interaction requires one or more of these concepts. They form an essential part of cognitive theory. Our prime concern in this paper is mutuality. The question we focus on is： why should people engaged in an ironical interaction justifiably assume that they have mutuality？ When and how can they do this？

Elements Constituting Mutuality

Although it is hard to define what constitutes mutuality， some basic constituents are indispensable for successful ironical communication， among which are situational and background knowledge， including the knowledge of participants’ mutual relation， social conventions and the awareness of communicative principles. Here， we briefly discuss participants’ mutual relations and their mutual power status， the societal politeness convention， and the roles of these factors in the establishment of mutuality in ironic communication.

Participants’ relationship plays an important role in the recognition of irony. Three types of participants can be differentiated： the speaker or ironist， the victim or hearer， and the audience （Barbe， 1995； Nash， 1985）. Although these roles are not static， the distribution of roles depends on participants’ relationship. The participants in ironic communication will differ in the degree of their particular awareness of irony depending on the closeness of their mutual relationship. The closer thir relationship， the larger the amount of mutuality between them， and the easier they may understand each other’s ironies. Private ironies are not easy to recognize due to lack of mutuality unless the ironists and other participants have familiar relationships with each other. The hearer in the lack of background knowledge may not recognize irony that victimizes the ironist himself， for instance， self-satirical irony is not easy to recognize by the audience due to the lack of mutuality. Enright （1986） comments： “Every reader will have the greatest difficulties detecting irony that mocks his own beliefs or characteristics” （p. 15）.

In normal situation， the ironist assumes his intended audience （sometimes inclusive and sometimes exclusive of the victim） shares his presuppositions or contextual assumptions， which form much of our taken-for-granted beliefs about the structure and texture of the world. The establishment of groups frequently happens as a result of the solidarity that grows from the increasing number of shared presuppositions. Ironists or speakers who try to draw out solidarity and understanding from the audience may find themselves shunned if the audience considers the irony unacceptable. In this case， the ironists misjudged the audience. In the political arena， for example， ironies often misfire because of the diversity of the audience.

Another element in close relation with mutuality between the speaker and the hearer is power relation. The power relation， especially when it is not built on culturally induced factors like status， age， money， or education， is situation-bound and subject to constant variation. Power is a relational or interactive concept. It represents an asymmetry involving at least two participants—one having more control and the other less. One participant possesses a status that another lacks and perhaps the latter acknowledges the status in a specific situation. The use of irony and the establishment and maintenance of power are interrelated. Power relation is one of the preconditional elements for ironic communication and also the result of it.

Politeness convention is also one of the essential elements in the mutuality between the ironist and the hearer. This convention regulates not only the ironist’s choice of language style but also the hearer’s understanding of it. Lakoff （1973）， Leech （1983） believe that speakers and hearers in interaction are all aware of and will conscientiously heed a common social convention， i.e.， the so-called Politeness Principle. The use of irony indicates that power-relation in ironic communication is incompatible with politeness convention. On the one hand， the ironist tries to show superiority over his audience （especially his victim）， on the other， he has to take into account politeness convention and keeps his intention not too conspicuous. The ironist is making a critical statement， but he is making it not too offensive. By using irony， the ironist hints at a conflict or an incongruence between non-aggressive ways （utterances or expressions） and aggressive intention.

So far， we have discussed the potential aspects of mutuality in the understanding of ironic communication. These aspects are presumed to be shared in the knowledge of both the ironist and his audience. If the ironist had no idea of these aspects， he would have no way to express irony. If the audience had no idea of them， they would not be able to find any way to appreciate and recognize it.

Necessity of Mutuality in Ironic Communication

In ironical communication， ironists are not just ironic： They are ironic to certain hearers. Suppose it is common ground to Zhang， Lily， Wang that none of them likes watching American Hollywood film. Now suppose that Zhang and Wang have been to a theatre and have watched an American film entitled Saving Private Ryan that they agreed was unexpectedly fascinating. As they came out of the theatre， they met Lily. Zhang said either of the following：

（1） Zhang to Wang： 這部电影真没意思，是吗？1 （Zhang to Wang： It is a boring film， isn’t it？）

（2） Zhang to Lily： 这部电影真没意思，是吗？ （Zhang to Lily： It is a boring film， isn’t it？）

With （1）， Zhang was being ironic to Wang， but not to Lily. Without knowing that they enjoyed the film， Lily could not be a party to Zhang’s irony， because as far as she could tell， he was completely serious. Zhang recognized that Lily would take him seriously， basing on their former common ground that none of them liked American film. So， when he uttered the same sentence to Lily， she would assume he was telling the truth. She had no way of recognizing his ironical intention. If， however， Lily had also watched the film， she would have doubted Zhang’s ironical intention because she would find Zhang’s utterance to be one of inconsistency with the immediate context and also against her expectation that arose out of the immediate context.

As this example illustrates， the perception of irony hinges on subtle judgements of what is mutuality to whom， so a hearer or reader not supplied with the right information may not make these judgements accurately. Understanding irony crucially depends on mutuality between the ironist and the addressee.

The importance of mutuality in the process of irony recognition is widely acknowledged. No approach so far has refuted the centrality of shared world. In the relevance theory， shared background information forms the basis for Relevance Principle. According to this principle， when communication occurs， speakers will agree on one correct interpretation if they share the same background information. Wilson and Sperber （1992） argue that for an utterance to be understood， it must have one and only one interpretation consistent with the principle of relevance. That is， it must have one and only one interpretation that would have enough effects to be worth the hearer’s attention， so that the speaker would not put the hearer to gratuitous effort in obtaining the intended effects. Hence， in their account of irony as a variety of echoic interpretive use， they assert that the audience should also choose one interpretation. However， since many factors play a role in the recognition of irony and， moreover， not all factors will be handled or understood in the same way by participants， a variety of interpretations will be possible. There are many different situations that can be construed， in which an ironic utterance will be understood differently. For instance， irony interpretation may vary with participants’familiarity with each other. If the participant knows the speaker well and likes or dislikes him/her， or is a casual acquaintance， or has just met the speaker， the ironist’s utterance is certain to be interpreted in different ways. Perhaps the principle of relevance can be applied separately to each relationship holding between participants in ironic interactions. The more knowledge speaker and hearer share， the easier and better they understand each other. The relevance principle again points only to the crucial place that background， textual， circumstantial and cultural knowledge takes in interactions in general and in ironic instances in particular.

In their experiment， Jorgensen et al.（1984） meant to test the feasibility of the Mention Theory by means of short sketches， such as example （3）. However， what they actually tested was the importance of shared information. The test subjects were divided into two groups. They had the task of an “ironic evaluation.” Both groups received identical copies of the short sketches but with one difference. The statements motivating the irony （in italics and parentheses） were only made available to one group.

（3） The party was at the Clarks’， but Joe didn’t know where Mr. Clark lived. “It’s on Lee Street，” Irma told him.“（It’s the house with the big maple tree on the front lawn.） You can’t miss it.” But Joe did miss it. He never would have found it if Ken hadn’t seen him wandering down the street and led him to the Clarks’ apartment. They lived over a store， and their apartment door was right on the sidewalk. Irma was already there when they arrived. “You are late，” she called to Joe. “The Clarks have a beautiful lawn， ” he replied. （Jorgensen et al.， 1984， pp. 112-120）

Not all the test-subjects shared the same background knowledge （or were not aware of the antecedent use）. Their interpretations of the short sketches as ironical or non-ironical varied accordingly. Thus without the sentence in parentheses， test-subjects generally did not interpret situations as ironic. Jorgensen et al. （1984） concluded that the “results tend to support the claim that people do not perceive an implausible non-normative utterance as ironic unless it echoes some antecedent use， which is the outcome predicted by the mention theory of irony” （p. 112）. However， this conclusion only proves the importance of mutuality for the recognition of irony. William （1984） notes that the very utterance that the subsequent， ironic utterance supposedly echoes provides part of the necessary background information to define the target utterance as ironic.

Like Sperber and Wilson， Clark and Gerrig （1984） also argue that the participants’ shared background-knowledge conditions the recognition of irony， “A listener’s understanding of an ironic utterance depends crucially on the common ground he or she believes is shared by the ironist and the audience—their mutual beliefs， mutual knowledge and mutual suppositions” （p. 121）. Ironic pretense must imply previously shared knowledge， or the pretense would not be recognizable as such. An ironist addresses only the initiated participants. The perception of irony often hangs on subtle judgments of what is common ground to whom， so a hearer or reader not supplied with the right information may not make these judgments accurately. Such misjudgment may have happened in Jorgensen et al’s （1984） echoic version of “The Lecture”：

（4） The instructor asked the whole class to attend a special evening lecture by a visiting professor. “How tedious！”Anne complained to Harry and Tom. Harry and Tom attended together and were both impressed by the high quality of the lecture， which was both educational and amusing. As they were leaving the lecture hall， they bumped into Anne. “Tedious， wasn’t it？” Harry said. （Jorgensen et al.， 1984， p. 120）

The story fails to give one crucial piece of information： Did all three share the knowledge that Anne， too， unexpectedly enjoyed the lecture？ If they did， Harry’s utterance would have been ironic to Anne； if they didn’t， it would not have been. The test result proves just the crucial role of mutuality or shared knowledge， about half the students in Jorgensen et al’s study saw irony in Harry’s question， and half did not. This shows that it should make little difference in this instance whether or not Anne had complained earlier. What is really critical here is the pattern of shared knowledge and beliefs but not the presence of an utterance to be echoed per se.

In pretense theory， ironists do not tell their hearers they are making a pretense but let them discover it for themselves. As Grice （1978） put it， “while one wants the pretense to be recognized as such， to announce it was a pretense would spoil the effect”. But what do they need to be able to discover it？ Again the crucial notion is common ground or mutuality. Hearers must see how the speaker’s utterance is relevant to the common ground already established between the speaker and the addressees. If they could not， they might not be able to discover the pretense. Consider again Jorgensen et al.’s （1984） story “The Party”， but with two different endings， one given by Jorgensen et al， the other by Clark et al.

（5） The party was at the Clarks’. But Joe didn’t know where Mr. Clark lived. “It’s on Lee Street，” Irma told him.（“It’s the house with the big maple tree on the front lawn.） You can’t miss it.” But Joe did miss it. He never would have found it if Ken hadn’t seen him wandering down the street and led him to the Clarks’ apartment. They lived over a store， and their apartment door was right on the sidewalk. Irma was already there when they arrived. “You’re late，” she called to Joe. Joe replied：

（a） “The Clarks have a beautiful lawn.” （Jorgensen et al’s ending， 1984）

（b） “You give wonderful directions.” （Clark et al’s ending， 1984）

When the material in parentheses is absent， Joe and Irma share no knowledge against which they can make sense of （a）. But when it is present， they do share the knowledge， and Irma can therefore discover Joe’s pretense. That is just what Jorgensen et al. found. For ending （b）， however， Joe and Irma’s shared knowledge should be sufficient for her to discover the pretense with or without the material in parentheses. Ending （b） would be judged ironic even without any previous utterance to echo. If so， the reason that （a） isn’t ironic without the previous material to echo is not that there is no previous material to echo. It is because （a） cannot be related to anything in Joe and Irma’s shared world. It violates one of Grice’s most important maxims to speakers： Be relevant.

After the status of mutuality is confirmed in the recognition of irony， when can a person engaged in an ironical interaction justifiably assume that they have mutuality？ As far as this question is concerned， there have been two different hypotheses： the mutual knowledge hypothesis and the relevance-theoretical hypothesis about mutuality. On the whole， both hypotheses propose that a hearer depends on certain information in the shared world in understanding the speaker’s utterances. The distinction between them is whether this knowledge is mutually known beforehand.

According to the mutual knowledge hypothesis， making conversational inferences must involve knowledge and beliefs that hearers and speakers mutually share. There must be mutuality between participants beforehand. In contrast， the relevance-theoretical hypothesis assumes that there need be no assumption that this knowledge is mutually known beforehand. The existence or establishment of mutuality is not a necessary condition for the comprehension of utterances； rather， it is the result of the act of interpreting.

For the moment， we do not claim which hypothesis is superior. However， we may partly agree with Sperber and Wilson’s proposal that understanding is a search for relevant contextual assumptions， but argue it is highly unlikely that hearers can correctly understand a speaker’s irony without some previously-set mutuality. Imagine the situations in which Shakespeare is talked about. In the case the participants are all native speakers and they assume that all native speakers should be in the know of Shakespeare， then Mary’s utterance in （6） is most probably recognized as ironical. However， if the participants involved are all non-native speakers， B’s utterance in （7） is not necessarily understood as ironical even though the speaker intends his utterance to be so.

（6） （Peter and Mary are talking about Bill） （Sperber & Wilson， 1986） Peter： I was told Bill has great learning. He has read quite a lot of books. Mary： Well， Bill is a well-read man indeed. He has even heard of Shakespeare.

（7） A： 李四閱读广泛，学识渊博。 （A： Li Si is a well-read and learned man.） B： 他的确博学，连莎翁都知道。 （B： He is erudite indeed. He has even learned of Shakespeare.）

One argument for mutuality is that shared world knowledge is itself highly structured and clearly related to a hearer’s knowledge of language （Gibbs， 1984）. It argues that it is impossible to clearly distinguish between linguistic and factual consideration in assessing the truth of a statement， which also suggests that it is difficult to distinguish our beliefs about a statement due to its logical form and general shared knowledge. Thus， understanding Mary’s response correctly in the above situation can only be done if it is mutually assumed that Peter knows whom Shakespeare is referred to by Mary. Without some appeal to the notion of mutuality， Mary’s mention of Shakespeare would be nothing more than a guess as to whether Peter was capable of knowing what was meant. Peter may not necessarily be consciously aware of this information in advance of Mary’s statement. All that matters is Mary believes that Peter is capable of using this knowledge at the right moment because it is assumed to be mutually known.

Even if it is assumed that mutuality is a result but not a precondition of comprehension （Sperber & Wilson， 1986）， speakers and hearers still must do something with the mutuality that is established once an utterance is comprehended. This newly established mutuality is certainly utilized during comprehension. But when？ It is undoubtedly resorted to when comprehension occurs. In （8）， it is absurd to think that the interpretation of the ironical utterance made by means of metaphor does not take advantage of the previously established mutuality between the speaker and the hearer about “Lu Deng in Lu’an County”.

（8） ….女子 （回头看见钥匙在他手里）啊，在你手里？喔，记起了，还是我交给你的。（Woman： （Turn round and see the keys in his hand） Ah， the keys are in your hand？ Oh， yes， I don’t remember that I have given them to you.）

巡警： 哼，这会儿你可也有点慌了吧？（怒目向之） （Policeman： Hump， are you also stricken with some panic now？（Stare at her with rage）

女子： 是的，有一點儿。（望着他的眼更畏惧）我真没有想到你会是这样一个不肯屈服的英雄！你倒象《潞安州》里的陆登！ （袁牧之， 1933） （Woman： Yes， a bit. （Looking more frightened at his facial appearance） I have never thought that you will become such an unsubdued hero！ You are as brave as Lu Deng in Lu An county. （Yuan Muzhi， 1933）

In many other mundane utterances， it will also be very absurd to think that the interpretation of them does not make use of any mutuality between the ironist and the hearer. If there is a wrong assessment or presumption of mutuality， it is quite likely that misunderstanding occurs. For instance， in （9）， two girls are talking about one of their classmates going to study abroad. Understanding of the reference words Ta（she）， Renja （the other =she here） may appear to be a tiny problem. But it requires that A and B both mutually recognize who “Ta and Renja” are being referred to.

（9） A： 你知道吧，她说出国就出国了。

（A： Perhaps you know， she said she was going abroad， and she did what she said！ ）

B： 咱们怎能跟人家比，人家可是傍上了洋干爹。

（B： How can we compare favorably with her？ She has a foreign nominal father.）

The interpretation of utterances such as （9） shows that mutuality is necessary in order to insure that the reference words will be properly understood. Sperber and Wilson （1982） suggest that the situation like the one in which （9） occurs is unnatural and that such misunderstandings are unlikely to occur. But scenarios like this one are not uncommon， misunderstandings often occur because the speaker and the hearer fail to correctly coordinate each other’s shared world. Misunderstandings of this type are frequently found in native / nonnative speaker’s conversation where the participants roughly speak the same language， but do not have the same underlying beliefs and cultural knowledge.

As was previously discussed， Sperber and Wilson propose an echoism hypothesis for ironic communication. They regard irony as one of a variety of utterance types in which the speaker does not express his own belief， but echoes the beliefs of some one else， and perhaps， expresses his attitude to those beliefs. Yet， they do not explain that hearers must share enough information to be able to recover the source of what is being echoed， particularly in cases where the source of the echo is based on a distant event or vague social norm as proposed by them. A speaker would not state an echo unless he assumed that the hearer shared enough knowledge with him so that he could locate the source of the echo and thus see a sentence echoed as ironic. In fact， it is speaker’s knowledge of who does and who does not share certain information that makes it possible to have “victims” of irony. For example， the speaker might compliment his cousin in appearance by saying （10B）. Although the audience might take this utterance as a compliment， and acknowledge it as so， the speaker really intended his utterance as a sarcastic pun and meant to criticize the addressee indirectly. Other participants who did not share the speaker’s world of beliefs could not recognize that the speaker was being sarcastic-ironic.

（10） （The exchange took place when the elder cousin heard that the younger cousin had forced his mother to death.）

A： 表哥，有空到我家做客。（A： Elder cousin， come and pay us a visit when you are free.）

B： 表弟這么有本事，我们怎敢高攀。（B： Younger cousin， you are so capable， how dare we claim ties of kinship with you？）

As revealed by the above analyses， the interpretation of any types of irony depends on mutuality or shared world. Without mutuality， ironic communication and cognition is not possible， either “ informative intention”or “communicative intention” （Sperber & Wilson， 1986， 1995） cannot be conveyed. Hearers use what is mutually shared at the earliest stages of linguistic processing which constrains the way an utterance is interpreted.

Possibility of Establishing Mutuality in Ironic Communication

Cues Indicating Mutuality

Participants in communication normally provide each other with a variety of cues or signals that assist in the search for mutuality and the establishment and confirmation of mutuality. The existence of these signals themselves indicate a certain degree of mutuality. Hearers discover irony only when they recognize some signals available， which guarantee more or less knowledge assumed to be mutually known or tacit. In most cases， if hearers have no recourse to certain cues or signals of irony， they will be misled. Hearers recognize irony because they notice cues that appear incongruent or mismatching with the general tenor of the statement or context. Irony is made evident to the understanding either by the delivery， the character of the speaker， or the nature of the subject. If any of these three is out of keeping with the words， it at once becomes clear that the intention of the speaker is other than what he actually says （Booth， 1974）.

Certain cues are said to be present in each occurrence of irony. These cues coincide with what Gumpers（1982） called “contextualization clues”. The cues for irony include signals of intonation （stress of a syllable or word， different pitch）， a particular way of talking， words incompatible with the context， contradictory clauses， a style inconsistent with the content of an utterance， register changes， etc.. These cues can be linguistic or extra-linguistic. They can be found at different levels： linguistic， discourse （textual） and pragmatic levels. These cues can be present simultaneously. The juxtaposition of conflicting words， actions， or words and actions also yields irony.

Actions， gestures， movements indicating irony are extra-linguistic cues. There are many examples of extra-linguistic cues， in the situation below is one example cited from K. Barbe （1995）.

（11） National Public Radio’s Penny Dennis remarked on her show Marketplace about the business executive Frank Lorenzo， who eliminated unions at Continental Airlines： “Today is Lorenzo’s birthday. The former chairman of Continental and Eastern and hehehe a big friend of unions. He turned fifty-three.” （Barbe， 1995， p. 12）

Penny Dennis clearly signaled this instance of irony with laughter hehehe. She is aware of her large and varied audience and attempts to influence their interpretation of the news as far as possible. Guided by this signal， an audience may understand that Lorenzo is not a big friend of the unions， he is rather a foe.

The linguistic cues， which might hint an ironical interpretation， include a variety of linguistic devices used in ironic communication. They are rhetorical questions， innuendo or insinuation， analogy， deliberate ambiguity， the display of incompatibles， overstated or understated use of language， echoing an utterance or thought， echoic questions， deliberate quotations， negative expression of positive meaning， positive expression of negative meaning， etc.. Linguistic devices not only mirror the speaker’s estimation of the hearer’s contextual resources but also help the hearer search for mutuality and guide the interpretation process.

Take understated use of language for example. The utterance below involves the use of an understatement. Normally， understatement is described as a form of irony in which something is intentionally represented as less than in fact it is. In Gricean terms， understatement can be characterized by quality observance but quantity breach. The quantity breach gives rise in each case to an implicature that the hearer would not normally accept as a reasonable assessment of the topic. For instance， in （12）， for an addressee in the know about the actual temperature of the soup， he can easily see that the quantity maxim is infringed and realize it is a misapplied evaluation and thus recognize the speaker’s self-ironical intention.

（12） （One drinks soup which is unexpectedly too hot. When he gets scalded， he feels embarrassed in the public. But

he disguises his embarrassment by saying something understated and self-satirical.）

這汤热了一点 （The soup is a little hotter than I expected）.

For the use of discourse cues indicating irony， consider Sperber and Wilson’s example cited from Austen（1971）. On seeing Emma playing happily with her sister’s child， Knightley counsels Emma： “If you were as much guided by nature in your estimate of men and women， and as little under the power of fancy and whim in your dealings with them， as you are where these children are concerned， we might always think alike” （Austen， 1971， p. 115）. And to this she replies：

（13） “To be sure—our discordancies must always arise from my being in the wrong.” （Austen， 1971， p. 115）

In Emma’s reply， “to be sure” and “must always” are unreasonable and the strength of these assertions of epistemic modality is a feature of her discourse sharply at odds with the accompanying declaration of inferiority and fallibility： her righteous insistence on her wrong constitutes just the discoursal cues readers need that Emma intends her mismatched or non-congruent commentary to be recognized as intended mismatched commentary. These discoursal cues put the reader on alert and make them recognize that Emma’s expressed view is ill matched with what might be a reasonable commentary in the circumstances， a mismatch that is beyond the expectation of readers （and Knightly， Emma’s addressee as well）. It’s not reasonable， in the affairs between two rational and civilized adults， for one of those adults to admit and even insist that he or she is invariably the source of error and discordance.

Discoursal cues can also be found in the case of redundant use of language， misrepresentation or false statement， intended contradiction， fallacious reasoning， stylistically signaled irony， etc.， which more or less， in this way or that alert readers or hearers to an ironical interpretation of an utterance.

The existence of some particular pragmatic acts also hints at certain degree of mutuality already set between the ironist and his audience. The pragmatic acts such as praising in order to blame， blaming in order to praise， pretended agreement with the victim， pretended advice or encouragement to the victim， pretended doubt， pretended error or ignorance， pretended omission of censure， pretended attack upon the victim’s opponent， pretended defense of the victim， making the hearer be the victim ， making a third party rather than the hearer be the victim， etc.—all hint that there is already a complicity between the ironist and the audience of what is possibly ironic. Consider the following example in which the speaker points at one but abuses another.

（14） （In a quarrel， a woman jeers at another gorgeously-dressed and clandestine woman indirectly by cursing a dog

that takes food on the sly. That is， she points at one but abuses another）

怪不得你那么豐满，原来你常常偷吃。（You often take food on the sly， no wonder you are so full and round！）

To summarize， when ironic communication occurs， mutuality is possibly embedded or indicated at cues of all levels： semantic and pragmatic levels； or， word/sentence/discourse levels， intonational levels and situational levels. The very fact that these cues exist hints at certain degree of mutuality to be already set between the ironist and the audience. In other words， there’s already complicity between the ironist and the audience of what is likely to be ironical. These cues are directive information of mutuality or common ground shared by the participants. They also guide and constraint the search for and establishment of mutuality. Without mutuality， the ironist is bound not to make use of these linguistic devices， to choose the discourse style and to adopt these pragmatic acts. However， the existence of these cues does not guarantee the final confirmation of mutuality. Instead， a hearer has to make some inference or search by following some mutuality decision criterion in order to finally confirm mutuality.

Mutuality Decision Criterion in Ironic Communication

In the philosophy of language and pragmatic theories， it is deemed necessary to explain how speakers and hearers ever coordinate what they mutually believe when there is always one more belief statement to be established and how hearers determine which pieces of knowledge of the many shared with their speakers are to be used in interpreting utterances. These concerns have been widely discussed.

One standard analysis of mutuality is what Barwise （1985） calls the iterated attitude approach. By definition， two agents， A and B mutually know some proposition P， if the following conditions are followed：

A knows that P

B knows that P

A knows that B knows that P

B knows that A knows that P

A knows that B knows that A knows that P

B knows that A knows that B knows that p And so on， ad infinitum

This analysis led Clark and Marshall （1981） to identify the mutual knowledge paradox. In this scheme， two people seem to be unable to verify mutuality for ever because it is highly unlikely that hearers can compute an infinite series of these propositions in a finite period of time. Therefore， in place of this procedure， Clark and Marshall proposed a heuristic finite decision procedure for determining when knowledge is mutual. The procedure is formulated as：

A and B mutually know that P， if and only if some state of affairs G holds such that：（a） A and B have reason to believe that G holds.

（b） G indicates to each of A and B that the other has reason to believe that G holds.（c） G indicates to A and B that P.

This formulation is normally called the mutual induction scheme， according to which if A and B make certain assumptions about each other’s rationality， they can use certain states of affairs as a basis （G） for inferring the infinity of conditions all at once. There is no need to confirm each and every one of the infinity of conditions. As for the ground G used as the basis for the mutual induction scheme， however， Clark and Marshall fail to provide precise explanations， they merely suggest that people ordinarily rely on three kinds of co-presence2. However， the weak point of this scheme is just in the assumption of triple co-presence. Garnham and Perner （1990） argue that triple co-presence is neither a sufficient nor a necessary condition for assuming mutuality， for on the one hand， even when triple co-presence is available， it does not guarantee that mutuality can be established； on the other hand， the absence of triple co-presence does not prevent the establishment of mutuality， for instance， in the instance of conversation over the telephone.

Sperber and Wilson （1986） have drawn a more radical conclusion from the mutual knowledge paradox. They argue that the solution to these problems concerning mutual knowledge is to abandon the idea that mutual knowledge plays any significant role in conversational inference， that the concept of mutual knowledge has no place in a theory of communication and no place in a cognitive theory， because no two people can ever be sure they have mutual knowledge. Therefore， in place of mutual knowledge， they put forward the concept mutually manifest cognitive environment. Indeed， Sperber and Wilson have made two valid points about mutual knowledge. First， there are many circumstances in which ascription of mutual knowledge would be open to doubt. More generally， assumptions about cognitive environments， i.e. what people ought to be able to perceive or infer are safer than assumptions about other mental states and achievements， since the latter are often mistaken. Second， mutual knowledge is not a precondition for communication， for instance， speakers who make a definite reference to a church do not necessarily assume that they and their audience have mutual knowledge of the church， they merely assume that audience will be able to work out which building they are referring to. The church need not be mutually known but only need be mutually manifest.

However， the fact that mutual knowledge is not necessary condition for successful communication does not mean the concept of mutuality can simply be dismissed. Even Sperber and Wilson themselves have to distinguish between manifestness and mutual manifestness about cognitive environments. It seems that mutual manifestness itself does not avoid the infinite regress either. So it is as hard to determine mutual manifestness as to determine mutual knowledge. Now we can see that Sperber and Wilson’s arguments are invalid just due to the fact that they rest on the unwarranted assumption that mutuality must be both defined and tested for in terms of the infinite sequence of iterated attitudes.

As a matter of fact， mutuality should not be defined in terms of iterated attitudes. Nor should the infinite series of iterated attitudes be regarded as the definitive test for mutuality. According to a psychological decision criterion—the finite mutuality decision criterion proposed by Garnham and Perner （1990）， the mutuality dilemma is likely to be solved. The mutuality criterion gets rid of the infinite regression. It is stated and formulated as follows.

Any situation S involving two participants a and b， which is perceived by one participant， a， as S（a）， provides grounds G （where G is a subset of S（a）） for a to assume that the proposition p is mutually known by a and b iff participant a has reason to believe that G satisfies the following four conditions：

C1. G

C2. G—a RG & bRG

C3. G—a Rp & bRp

C4. Whether G satisfies conditions C2 and C3 is established by common sense reasoning.

（In this scheme， the symbol ‘—’stands for material implication and ‘&’ for conjunction. If q is any proposition and x any rational person， then xRq means that person x has reason to believe that q.）

Condition C1 states the trivial fact that S can provide a with grounds G for mutuality only if it gives a reason to believe that G holds. Condition C2 is the central part of the criterion. It requires that G provides grounds for both participants a and b to know that G holds. In other words， G must be self-revealing or “open”to both participants. Condition C3 states that G must also make the target proposition p known to a and b. Condition C4 requires that the judgement about whether G meets C2 and C3 must depend on application of common sense inference rules.

To apply this criterion， now let’s have a look at how a judgement of mutuality is done in the case of ironic communication， consider （15）， in which A is the mother of B. A wants to know how B thinks of his sister’s boyfriend. A’s intention is to know whether her daughter’s boyfriend has a good moral character.

（15） A： 大為，你妹妹今天来的朋友，人好不好？

（A： Dawei， how do you think of your sister’s boy friend who has come today？ Does he have a good character？）

B： …根据报纸上官方介绍，他是天底头等的大好人，……浑身上下毫无缺点，连肚脐眼都没有。

（B. …According to the official newspaper， he is one of the best and kindest men on earth… he has no flaws

from top to bottom， he even has no navel.）

With the occurrence of the conversation， the situation can be easily accessed for both B and A. They both know whom they are talking about. The reference of “Ta （He）” in B’s utterance can be easily designated. Therefore， the situation can be formulated as：

S= （A， the hearer knows and believes that B， the speaker is giving a reply to her question. B， the speaker knows that he is replying to A.

Both B and A know and believe they are talking about the same topic.

Both B and A know and believe that they are talking about “sister’s boy friend.”

Both B and A believe what they are talking about is related to them， or at least related to one of them. etc）.

This situation S provides grounds G for both A and B. In the given circumstances， S can be equated with G. C1 is satisfied since S （S=G）. Take A’s point of view， C2 is met， since A， the hearer knows what she asks about and since A has good reason to anticipate that B will provide the message that answers her question. C2 is also met for B since B can reconstruct from what A asked about that A must have meant him to give an answer relevant to her question. Then， what about C3？ According to the mutuality decision criterion， it can be assumed that any situation S involving two （or more） participants， provides grounds G for the speaker to assume that the proposition P （B’s utterance in （15）） is mutually known. The situation S is also known by A， the hearer. S provides G for A to assume that P is mutually known. Therefore， C3 is met since S （S = G） implies that both A and B know P， that is， G has an implication that both the hearer and the speaker have reason to believe that P. Also C4 is met since the judgement that C2 and C3 apply does not depend on any special valuation or expertise； in other words， common sense reasoning can establish whether G satisfies C2 and C3.

Conclusion

The above discussion reveals the importance of mutuality in the process of ironic communication and cogniton. It shows that mutuality is not only necessary for the interpretation of any types of irony but also possible in the processing since there are cues embedded at all levels including semantic， pragmatic， lexical， morphological， syntactical， intonational and situational levels.

References

Aumann， R. J. （1970）. Agreeing to disagree. Annals of Statistics， 4， 1236-1239.

Austen， J. （1991）. Pride and prejudice. Beijing： Foreign Language Teaching and Research Press.

Bach， K.， & Harnish， R. M. （1979）. Linguistic communication and speech acts. Cambridge， MA.： MIT Press.

Barbe， K. （1995）. Irony in context. Amsterdam / Philadelphia： John Benjamins Publishing Company.

Booth， W. （1974）. A rhetoric of irony. Chicago： Chicago University Press.

Brown， P.， & Levison， S. （1987）. Some universals in language usage. Cambridge： Cambridge University Press.

Clark， H. H.， & Marshall， C. R. （1981）. Definite reference and mutual knowledge. In A. K. Joshi， B. Webber， and I. Sag （Eds.）， Elements of discourse understanding. Cambridge： Cambridge University Press.

Clark， H.， & Gerrig， R. （1984）. On the pretense theory of irony. Journal of Experimental Psychology： General， 113（1）， 121-126.

Cole， P. （Ed.）. （1978）. Syntax and semantics 9： Pragmatics. New York： Academic Press.

Enright， D. J. （1986）. The alluring problem. Oxford： Oxford University Press.

Garnham， A.， & Perner， J. （1990）. Does manifestnes solve problems of mutuality？ The Behavioral and Brain Sciences， 13（1）， 178-179.

Gibbs， R， （1984）. Literal meaning and psychological theory. Cognitive Science， 9， 275-304.

Grice， H. P. （1978）. Further notes on logic and conversation. In Cole 1978： 113-28.

Gumperz， J. J. （1982）. Discourse strategies. Cambridge， Mass： Harvard University.

Halpern， J. Y.， & Moses， Y. O. （1984）. Knowledge and common knowledge in a distributed environment. Proceedings of the Third ACM Conference on Principles of Distributed Commuting， 50-61.

Jorgensen， J.， Miller， G.， & Sperber， D. （1984）， Test of the mention theory of irony. Journal of Experimental Psychology： General， 13（1）， 112-20.

Lakoff， R. （1973）. The logic of politeness； or， minding your p’s and q’s. Papers from the 9th regional meeting of the Chicago Linguistic Society， 292-305.

Leech， G. （1983）. Principles of pragmatics. London： Longman.

Lewis， D. K. （1969）， Convention ： A philosophical study. Cambridge， MA.： Harvard University Press.

Milgrom， P. （1981）， An axiomatic characterization of common knowledge. Econometrica， 49， 219-222.

Nash， W. （1985）. The language of humour： Style and technique in comic discourse. London： Longman.

Schiffer， S. （1972）. Meaning. Oxford： Clarendon Press.

Smith， N. （Ed.）. （1982）. Mutual knowledge. London： Academic Press.

Sperber， D.， & Wilson， D. （1982）. Mutual knowledge and relevance in theories of comprehension. In Smith 1982： 61-131.

Stalnaker， R. （1978）. Assertion. In Cole 1978： 315-332.

William， J. P. （1984）. Does mention （or pretense） exhaust the concept of irony？ Journal of Experimental Psychology General， 113（1）， 127-129.

Wilson， D.， & Sperber， D. （1992）， On verbal irony. In D. Wilson and N. Smith （Eds.）， Relevance theory. Amsterdam： Elsevier Science Publishers B. V. （North Holland）.

Yang， Q. Y. （2013）. On the construction of pragmatic presupposition and ironical discourse. Foreign Language Teaching， 34（6）， 25-29.

Yuan， M. （1935）. A woman and a dog. The grand collection of Chinese new literature. Shanghai： Shanghai Literature and Art Press.

Yus， F. （2000a）. On reaching the intended ironic interpretation. International Journal of Communication， 10（1-2）， 27-78.

Zeng， Y. T. （2005）. Irony research in the past 40 years. Academic Research， （10）， 139-142.

Zeng， Y. T. （2014）. Experimental research on irony understanding. Journal of Zhejiang Foreign Studies University.

Zeng， Y. T. （2016）. Irony research in the past， present and future. Pragmatics Studies Life. Higher Education Press.

Zeng， Y. T. （2019）. Visual irony： A case study based on the trade war cartoon series， 16th International Pragmatics Conference.

Zeng， Y. T.（2021）.Perception of counter-expectational mismatch as irony trigger：A pragmatic perspective. American Journal of Humanities and Social Science.

Zhao， X. F. （2019）. Interactive integration of multimodal irony and conceptual metaphor/metonymy—Take graphic irony as example. Foreign Language Teaching， 5（40）， 8-12.

1 All the examples， unless indicated of their origins， are from field-work data.

2 Triple co-presence refers to linguistic co-presence， physical co-presence and community membership.