Poetic effects: a relevance theory perspective

POETIC EFFECTS

Pragmatics & Beyond New Series Editor:

Andreas H. Jucker (Jusllls Liebig Uniwrsity. Giessen)

Associate Editors:

Jacob L. Mey (Odense University)

Herman Panel (Belgian National Sdent·e Foundation, Universities of l.LJII\'Uin and Allfwerp)

Jef Ven>ehucren (Belgian National Science Fmmdation, Uniw.>rsity of Annverp)

Editorial Addre.u: Justus Liebig University Giessen, English Department

Otto-Behaghei-Strasse 10. D-35394 Giessen. Germany e-mail: [email protected] Editorial Board: Shoshana Blum-Kulka (Hebrew Uni\'ersity of Jemsalem) Chris Butler {Unin:rsit.'' Colle~:e of Ripon and York) Jean Caron tUnil·ersitf! de Poirier.,); Robyn Carston (Unil't'r.fity Colle~ is selected. but relevance theory would have a clear contribution to make here. The information used would include the most readily accessible assumptions in the context that yield contextual effects. The phenomenon of pragmatic enrichment. the narrowing of a general to a more specific meaning. has been studied and discussed in psychology. where it is more commonly referred to as "instantiation". One example discussed in the psychological literature (Halff, Ortony and Anderson 1976) is given here as (14): ( 14) A fish attacked a swimmer. The word 'shark' has been found to be a better recall cue than 'fish'. even though the latter actually occurs in the utterance and the former does not. Vicente ( 1993a) discusses this example at some lenglh. She argues that the VP provides a context which encourages an instantiation of the 'fish' category to "something like ·a shark-like creature' or ·a fish of the kind that attacks people'". The narrowed-down concept, she argues. "would be sufficient to prompt recall of a more specific term." (Vicente 1993a: 3). Here we have a

98

PoETIC Et-r'ECTS

case of a general concepl being enriched or inslantiatcd to something more specific. with some psychological evidence that thai is in fact the case. The only point of debate here is whether the instanlialed concept is inlended by the speaker and hence communicaled. This example. which depends on a panicular context. would differ from the kind of enrichment that lakes place in ( 15): (15) I don"t drink.

Here the concept DRINK is enriched lo the communica1ed concepl DRINK ALCOHOL. The enriched concept is clearly intended by the speaker and communicated lo lhe hearer. If Grice had been an advocale of the pragmalic enrichment of concepts as a contribution to 'what is said', he would have termed this 'generalised enrichment". in contrast to the 'panicularised enrichment' lhat occurs in the case of (14).

Parallel lo pragmatic enrichment are those cases where concepts are broadened or loosened. Example (16) provides an example of loosening in a context where a doctor is about to stick a needle in a patient"s arm. ( 16) This will be quite painless. While the amount of pain may be minimal. presumably it is not striclly true that there is a complele absence of pain. What is communicaled is lhat. given what the palient mighl expect or fear, lhe amount of pain will be very slight. Metaphors. as we have already seen, have been defined in lenns of such loosenings. where lhe thought communicated resembles. or shares logical and contexlual implicalions wilh, lhe proposition expressed by the utterance. According to this definition, metaphorical utterances communicate a range of implicalures. The alternative account. as already mentioned, sees metaphors as contribuling to the proposition expressed. In the case of (17) whal is C.'Ommunicated is a concept close to. but not identical with, GIANT. In ( 18) what is communicated is a concept close 10. bm not identical with. PIG. ( 17) Sam is a giant. (18) Samisapig. The concepls lhat we usc to think with. lhat are combined to form sentences in our 'language of thought". are much more flexible than natural language lexical concepls or categories. as Barsalou (1987), for example. suggests. If we wish lo communicate a thought which contains a non-lexicalised concept then we have to use a lexicalised concepl which shares some of ils propenies

METAPHOR

99

with the concept we 'have in mind' and also take advantage of other features of context to encourage the communication of the thought we intend. Example (18) may be used literaJiy, when 'Sam' is the name of a pig.lfit is clear from the context that Sam is a human being then that piece of contextual infonnation will indicate that a concept is intended that can be constructed from infonnation that the lexical (lexicalised in Barsalou's tenns) concept PIG makes available. but which is not identical to that concept. lndt.."Cd. in the case of conventional metaphors such as ( 18) it may be that one can speak of two lexical concepts. both lexicalised by 'pig'. It always remains possible, however. to exploit the connection with the original concept and use the metaphor more creatively to make salient other properties of pigs. This can be illustrated more clearly perhaps in the case of 'Richard is a gorilla·, where, as has already been suggested. aspects of Richard's appearance or the way he moves can also be made highly salient. One further argument in favour of the view that metaphors contribute concepts to the proposition expressed might be that the range of implicatures communicated is. in the case of conventional metaphors. generally quite stable across contexts. Unlike standard cases of what Grice tenns "particular· iscd conversational implicatures". the implicatures communicated by examples such as ( 18) remain much the same. In the case of more creative metaphors. as we shall see in the next section. the range of implicatures. though not standard in the same way, are very similar in meaning. It may also be the case that the intuition that metaphors are used to communicate thoughts more precisely (ao; expressed in Middleton Murry's observation quoted at the beginning of this chapter) is captured better by the view that new ad hoc concepts are fonned. It is possible to have a precise thought involving nonlcxicaliscd concepts that it is difficult to express. precisely because the concepts are non-lexicalised. The implicature account tends to encourage the view that the complex thought communicated by a metaphorical utterance is vague and indeterminate. But one would also need to consider the question of what it would mean to have a thought consisting of a set of weakly activated assumptions. The relevance theory view of metaphorical utterance interpretation. whether the implicature account or the account that metaphors contribute nonlexicalised ad /we concepts to the proposition expressed by an utterance. requires no special principles, processes or metaphor interpretation mechanism. So it is not subject to the criticisms that Gibbs and Gerrig level at the "literal

100

Pot.IIC EFFECTS

meaning hypothesis .. account of metaphor interpretation. It is more explicit about the process of interpretation and it answers Levinson's points abomlhe motivation and expressive power of metaphor: metaphors are convenient and

economical ways of communicating complex thoughts. This point will receive fuller treatment in the following section which introduces a discussion of more creative uses of metaphor.

Poetic metaphor One further advantage that relevance theory has over other pragmatic accounts of metaphor is that it is able to provide a characterisation of creative or poetic metaphor. It can also account for relative creativity and the relative success of metaphors intended to be poetic. In this sense the relevance theory account of metaphor is able to show greater sensitivity to the range of stylistic effects that metaphors can achieve, in contrast to those semantic. pragmatic and even literary theory approaches to metaphor that treat metaphors ao; though they were cut to a standard pattern and expressed a standard kind of stylistic effect. Metaphors that arc relatively conventional and metaphors that are relatively creative vary in tenns of the range and strength of the assumptions that they make more salient. According to the view that treal'i these assumptions as implicaturcs. the richer and more creative the metaphor, the wider the range of weak implicatures. By contrao;t. the narrower and stronger the range of implicatures the more conventional the metaphor. The alternative account sees these ao;sumptions as contributing to the creation of a new ad hoc concept. The new concept is more creative the less it is derived directly from any one established lexicalised concept. In (18) the new concept is derived from a subset of the properties of pigs. In (9) the new concept is derived from a subset of the properties of dumps. In the case of creative metaphors, as will be demonstrated in this section, the new concept is not derived from a subset of the properties of an existing concept. but is constructed on the basis of an interaction between assumptions derived from two or more encyclopaedic entries. In the case of creative metaphors two concepts arc brought together, the connection between which is neither well-established nor easy to achieve. The addressee (in this case, more likely. a reader) has to work harder to find assumptions that the concepts might share. A greater amount of processing

METAPHOR

101

effort is required: but the rewards in tenns of contextual effects are correspondingly higher. Easy access. as in the case of dead or conventional metaphors.lcads to relatively strong communication: a small range of assumptions are standardly made highly salient. Less easy access leads to a more diffuse range of ao;;sumptions being made weakly salient. In literary communication it may happen that the effort required is beyond the capability of the reader and he becomes confused or frustrated. The example of poetic metaphor used by Sperber and Wilson ( 1995: 237) is the following remark made by Flaubert of the poet Leconte de Lisle: ( 19) Scm encre est pale. 'His ink is pale: Here there arc no strong assumptions to the truth of which Flaubert can be said to have committed himself. The considerable processing effort involved in the search for relevant contextual assumptions is offset by the subsequent large range of implicatures weakly communicated. The contextual ao;;sumptions are obtained in the usual way through exploring the encyclopaedic entries attached to the concepts INK and PALE. This example includes an instance of metonymy. as well as metaphor. with 'ink' standing for 'writing·, in the sense of 'work' rather than 'handwriting'. The metonymy itself might be considered poetic. but if one were to make the simple substitution. for the purpose of concentrating on the metaphor. one would start with the assumption in (20): (20) Leconte de Lisle's writing is 'pale'. After searching through the encyclopaedic entry attached to the concept PALE one might derive implicatures such as: (21) Leconte de Lisle's writing is weak. By searching out and constructing contextual assumptions in this manner a wide range of implicatures can be derived. Other possible implicatures arc: (22) a. b. c. d. c.

Leconte de Leconte de Leconte de Leconte de Leconte de

Lisle's writing lacks contrast. Lisle's writing may fade. Lisle's writing is sickly. Lisle· s writing will not last. Lisle does not put his whole heart into his work.

There are an indefinite number of further implicatures one could add to (21)

102

Poi:.IIC EFFECTS

and the list in (22). all of which will be weak in the sense described, and lhere is no cut-off point lhat allows one to say thai so many implicatures are communicated and no more. It is the range and the indeterminacy of the implicatures which gives the metaphor ils poetic force. These factors explain why it is thai metaphors, especially poetic metaphors. can never be adequately translated or paraphrased. They also explain why their interpretation may differ across individuals and be subjeclto debate amongsllilerary critics.

I would now like to consider a poetic metaphor from a poem. The example I wish to discuss is the metaphor used by Seamus Heaney at the end of his poem Digging. The concluding lines of the poem are a.'i follows: The cold smell of potalo mould. the squelch and slap Of soggy peat. lhe curt cuts of an edge Through living roots awaken in my head But I've no spade to rollow men like them. Between my finger and my thumb The squat pen resls. I'll dig with it.

Ir the metaphor had been presenled on iL-. own. a-. in (23), iL'i inlended meaning would not be so clear. (23) The poet digs with his pen. The pen is a spade. Such a melaphor, as it stands, is 100 vague in its intended effects. h offers insufficient encouragement to the reader to access particular conlexlual assumptions. The reader might es1ablish links between 'pen' and 'spade' or 'writing' and 'digging'. but insorar as he is lert to do practically all the work. the melaphor is excessively vague wilh regard lo what il is supposed lo c..'Dmmunicate. The reader has no evidence that the writer had anylhing particular in mind. He mighl even consider it to be prelentious.that it prelends to lhe slatus or a creative poetic melaphor. but rails lo communicate a range or implicatures sufficient to compensate the errort the reader expends in seeking an interpretation. II might be the case here, as it no doubt orten is. that the writer herselr is vague aboul whal she wanls lo say. The working oul or implications, it should be noled, is considerably more effortrul and lime-consuming lhan is the case ror implicatures. The expressive power and aeslhetic rorce or metaphors. it will be argued la1er. depend on lhe sponlaneous. rapid way in which implicalures are processed.

METAPHOR

103

In Heaney's poem the interpretation of this central metaphor hallj, been prepared by the rest of the poem. Contextual assumptions made accessible prior to the metaphorical utterance ilself help direct the search for relevant conlextual a-.sumptions from the encyclopaedic entries of the concepls brought together in the metaphorical phra.'iC or utterance. In this way. a good poem. by activating a wide network of conlextual allj,sumptions prior to the inlerpretation of the metaphorical utterance itllj,elf. may give grealer direction to the interpretation of metaphors. enabling them to be read in a richer. more creative way than would otherwise be possible. The success of a poetic metaphor depends nol only (if at all) on ils originality, but in the creation of a contexl which encourages and guides the ex.ploralion of the encyclopaedic entries of the concepts involved. Some of the contexts set up earlier in this poem include (among much else) the informa1ion thai digging is how the poel's forefalhcrs earned lheir living, that it is an activity and occupation with a long tradition in the community, that it is hard and honest and necessary work. that it requires intense concenlration and thai it inspires awe. Many more such conlexlual assumptions mighl be accessed and the properties of 'digging' in these assumplions transferred. by inference. to the aclivity of writing poetry. A danger of slarting to make a list such as lhc one above. apart from il being incomplete. is thai it mistakenly suggests thai all these assumptions are strongly communicaled and equally strongly communicated. Many of these assumptions are weakly implicated or made marginally more manifest. Heaney writes al some length on his intentions in an essay in his book Prem:,·llpation.'l. suggesting some of lhe further contexts he had in mind (Heaney 1980: 41-43). He talks abou11he metaphor recalling the oft-quoted proverb "lhe pen is lighter lhan the spade" and suggests that "lhe poem does no more than allow 1hat bud of wisdom to exfoliale". The contrast between spade work and pen work introduces a note of guilt or unease into this metaphor as well a.-. a note of assertion and what might almost be termed defiance. He empha.-.ises the idea of digging a.-. a sexual metaphor. "an emblem of initiation.like putting your hand into the bush or robbing the nesl. one of the various analogies for uncovering and touching the hidden thing ... " It is a metaphor. he writes. thai suggests a view of poelry: "poetry as a dig. a dig for finds that end up being plant-.". It was certainly a rich melaphor as Heaney intended it. The reader is encouraged to derive a very large number of implicatures from exploring the encyclopaedic entries for the concepts DIG

104

POETIC EFFECTS

and POETRY or WRITING. If these implicatures are accessible in an on-line reading. then the metaphor is both rich and successful. Consider now an earlier metaphor in this poem. Heaney wrote in Preoccupatimu: "there are a couple of lines that have more of the theatricality of the gunslinger than the self-absorption of the digger."' He is referring here to the opening two lines of the poem: Between my finger and my thumb The squat pen rests: snug a'i a gun.

The somewhat sinister contexts prepared by this image -

'pen

a-. gun'

-

suggesting that the pen is a powerful and dangerous instrument. that the poet is a gunslinger. are not exploited in the rest of the poem: in fact they might even be said to be contradicted by the rest of the poem. '"The surprise or beauty of a successful creative metaphor", Sperber and Wilson ( 1995: 237) write. ''lies in this condensation. in the fact that a single exprcssion ... will determine a very wide range of acceptable weak implicatures". This. according to Heaney himself. is what happens for the first of the metaphors discussed here. but not what happens for the second. The idea that creativity can be described in terms of the communication of a wide array of implicaturcs suggests a theory of value. What happens is that the organisation of the language in the poem together with the metaphor itself leads to many more assumptions being simultaneously activated. and brought together at one conceptual address. It becomes a way or exploring a particular concept more extensively. A creative metaphor combines insight with depth. The reader must be able to construct the links between concepts where links are not wellestablished. If the context is too readily accessible. on the one hand. or if the search through context leads nowhere. on the other. then the metaphor fails as a creative metaphor. One thought that is bound to occur is that the greater effort demanded by poetic metaphors will surely mean that they will take longer to process. In the light of this question it is worth re-examining the psycholinguistic evidence referred to earlier that suggests that metaphorical utterances take no longer to process than literal utterances. Although Gerrig ( 1989), among others. argues that metaphorical interpretation requires no special procedures or principles to guide it. he is clearly worried by this kind or problem and raises one or two interesting points regarding creative and literary metaphon;. He points out that in the case of examples such as (5) (the 'lacy blanket' example) it is possible to

METAPHOR

105

record satisfactorily the time taken to understand the metaphorical utterances. But it is not so clear that this is possible with cases such as 'Juliet is the sun·. Like most examples of literary metaphor. if you spend more time and effort on it there is the chance of achieving greater appreciation and understanding. The point of practical criticism is to do just this. A reader cannot reasonably be said to reach a point at which he understands such a metaphor. The reader recognises that he ha... 'understood' the metaphor to a satisfactory degree (or nol) before pa... sing on. but thai il would be possible to give more attenlion toil and achieve a fuller understanding. Gerrig suggesL"i that. while there may no1 be lwo differenl processes for literal and non-literal utterance interprelation, there may be lwo different processes for convenlional or standard metaphor understanding and literary or creative metaphor underslanding: 1be flnttypc or understanding can be called time-limited L'ORlprelreluimr and is governed by the total time constr•int: the sa.ooand type can be called

lei.,llrely cm11pf'ehe11sitm and may well involve types or pr<X:essing that arc largely ·specified ror mew.pbor'. (Gcrrig 1989: 238.)

Wilhin a relevance theory framework, however. such a dislinction is not necessary. Leisurely study. analysis and re-reading - not jusr in relation lo melaphor - may help the reader to deliberately and consciously build an exlended nelwork of a.fure Princ·iple: II consists of lhrte stages: the first is when a man becomes obsessed with an emolional concept to such a degree that he is compelled to do something about it. What he docs is the second stage, namely, construct a verbal device that will reproduce his emolional concept in anyone who cares to read it. anywhere. any time. The third stage is the recurrent situation of people in diffen:nttimes and plat:cs selling off the devil.-c and ~-creating in themselves what the poet felt when he wrote it. (Larkin 1983: 80).

Given this strong connection between emo1ion and poetry in general. and. more panicularly, bel ween emotion and the use of rhetorical devices in poelry (and olher discourses such as polilical oratory). it becomes imponant to ask the queslions: 'What is an emotion?' and 'Can an accounl of emolion be introduced into a pragmalic account of poetic effects, and if so. how?' In this chapter I will consider some rccenl accounls of emotion in philosophy and cognilive science. (In doing this I will necessarily be very brief and highly selective.) Then I will return lo 1he questions raised above. In Chapler7 I will consider lhe question of how nonpropositional effecls in general mighl be included in a theory of pragmatics.

Properties or emotional states Rey {1980) characterises emotions as complex states possessing cognitive. physiological. behavioural and qualitalive propcnies. Rey mentions olher 1ypcs of propcnies 1hat may be attributed to emo1ional states. but essen1ially they reduce lo these four. The cognilive prope:nies of emolions may be characterised in terms of panicular lypcs of sets of beliefs and desires. They may be characterised. therefore. in lenns of their intenlionality: emotions are about somelhing. they refer to states of affairs and events in the world. These properties are lhe ones thai have panicularly inlercsted cognitive science and will be discussed in section 6.4.

144

POETIC EFFECTS

The physiological properties of emotions seem the most clearly theorisable. faJiing within the domain of the natural sciences (biology and chemistry). The standard view is that the evolutionary function of such physiological

effects was to prepare the body for appropriate action (for example. for fight or flight) in response to external phenomena.eventsorstatesofaffairs in the world. The behavioural properties of emotions include bodily movement and gesture. Emotional behaviour responds to the physiology of emotions but can be controlled and is subject to culturaJ variation. ll is caused by a complex interaction of physiological factors and cuhurally determined belief/desire sets. Cultural constraints may influence the degree of freedom or restraint shown in the expression of behaviour either generally or with respect to certain emotions. They may also influence the manner in which such behav· iour is expressed. Social groups recognise behavioural norms with regard to the expression of emotional behaviour. but there is room for variation and the norms are sometimes contravened. Such behaviour has to be measured in terms of broad generalisations and statistical correlations rather than in terms of causality. It is possible to develop separate and combined theoretical accounts of the physiological and cognitive properties of emotions. Rey offers the follow· ing example of how the two might interact: ... a specific cognition or constellation of cognitions, might be linked nomologically to specific qualitative and physiological states. and so forth: a gh·en emotion might be regarded a.~ some commonly occuring segment of just such a sequence. A crude but not impossible instance might be, say, depression over the collapse of one's career: this might be identified as the sequence beginning with the belief that one's career has indeed collapsed. the quite strong preference that it hadn't, a consequent depletion of norepinephrine, the effects of that depletion upon the nervous system, consequent further changes in cognition le.g .. the belief that nothing any longer is worthwhile. decreased preferences for doing anything ut all). followed by litill further depletions of norepinephrine. and further effects of this still greater depletion, various portions of this sequence being accompanied. perhaps, by that unmistakable qualitative feel... (Rcy 1980: 188.)

Qualitative properties (those unmistakable qualitative feels), however. arc more problematic because they arc subjective and elusive. Only I know what my angst and despair arc like. and even then only approximately. Although subjective and elusive. this is not to say that they must remain totally private. Literariness can be characterised in terms of the attempt to express these

EMOTION. AmnJDE AND SEtmMENTAUTY

145

relalively ineffable ··qualiullive feels". Lileralure, Calvino suggesL~ in his essay Cybemetks and Ghosts. is engaged in a conlinual anempl lo express somelhing lhal cannol properly or fully be said - or even implicaled. "The suuggle of literature i!• in fact a struggle to escape from lhe confines of language: it !io11"C"Iches out from the utmost limits of what can be said: what stirs literature is the call and attr.K.1ion of what is not in the dictionary.' (Calvino 1997:16.)

He goes on lo lalk about a "meaning thai is not patent on the linguistic plane", of a meaning that has 'slipped in from anolher level'. Despite the slruggle it is possible to provide some purcha.-.e on lhe subjective character of experience through language. These qualitative properties may relate causally to the cognitive and physiological components of emotional states, but it is difficult to see how the nalure or the causal relationships may be detennined or characterised. It may be thai 1hese properties of emotions are really properties of physiological slales. lhat when we lalk of such qualilalive properties we are really talking about conscious awareness in relation to such physiological stales. just as we talk aboul lhe sensation of an itch or a headache. At some level il may be Ihat all emotions are complex physiological states triggering different types of conscious awareness. Some physiological slates trigger conscious awareness in lhe fonn of 'lhoughts'. and some in lhe form of 'feelings'. I will relum to these issues in Chapter 7. If we assume lhat emolional stales typically involve all these properties, il is also importanl to poinl oul that il is possible to have states thai combine some but not all of lhem. As Rey puts it: a typical emotional state is not analyzablc as merely the cognitive component of the stale. since one may have the beliefs and preferences without the emotions. and the emotions sometimes without the beliefs and preferences. lRey 1980: 176).

Examples of lhe latter case (of whal mighl be lenned 'incomplele emotions'). where lhe cognitive componenl is absent. would be free-noaling anxiety or elalion. Emotional states. then. arc. typically. complex slates involving nomologicaJ intew.1ions belwcen cognitions. qualitative states. and physiological processes (should qualilative states be eliminated in favour of physiological proccsses.the interaclion would of course be only between the cognitive and the physiolog.ica1). (Rey 1980: 188).

146

POETIC EFJ-l:CTS

It is this complexity of emotional states that leads Rey to question the claim lhat machines can possess emotions. Emotions depend upon interactions between the cognitive and the physiological. {Certain cognitive states cause physiological states. which. in tum. cause cognitive states. etc.). As machines do not have human physiology these interactions clearly cannot take place_ Another consequence of not possessing human physiology is that machines cannot have the characteristic human qualitative experience of emotions.

The evolution and relevance of emotion Complex emotional states of the kind described could evolve only with a relatively advanced and flexible degree of intelligence. The circuits of the brain dealing with emotion were set up to motivate appropriate behaviour in major life-challenging circumstances. There are four systems located in the limbic system involving distinct neural pathways. parallelling four types of behaviour: (i) exploratory and appetitive; (ii) attack: (iii) flight; (iv) loss or grief. For primitive life forms specific reactions to specific stimuli were sufficient for survival. The appearance of a predator. for instance. would activate the appropriate autonomic/hormonal circuit which would send out the appropriate messages and prepare the body for flight. A simple representation of some state of affairs (maybe a visual image) would directly trigger a bodily reaction, a set of physiological responses. For more complex systems. responses need to be adapted to important non-specific life-situations where flexibility is required. The flight response is no longer triggered by a fixed number of stimuli, but results from a more complex set of circumstances. It has developed into the emotion of fear. Emotions. then. have evolved as a way of interrupting the behaviour of the organism when a goal is frustrated by a novel or unexpected situation. causing it to rethink its plans. (See Oatley (I 992) for one of the most recent accounts of this view.) As the ability to represent the external world has become more sophisticated. with the development of a rich array of concepts and a language of thought. emotions themselves have become more complex and sophisticated. Whereas the fear response was activated by a narrow range of stimuli in some animals (e.g. the shape of a predator), for humans the emotion is now triggered by a complex pattern of beliefs and desires. (I leave aside the

EMOTION. AT1111JDE AND SEtmMENTAUTY

147

queslion of whelher or lo whal exlent animals are conscious of the emotions they undergo: whether. thai is. their emotions have a qualitative component in the sense discussed. This is.'iue is still hotly debated: but sec Dawkins ( 1993) for arguments in support of animal consciousness.) The desires which are thwarted or violated become more complex as immediate desires are supplemented by time-indexed desires. (An immediate desire might be the desire to quench one's thirst: a time-indexed desire might be the desire to climb the Gran Sa'iso nexl summer.) Emotion provides a means of focussing aucntion. Certain environmental stimuli become super-relevant in lhe sense lhat survival depends on paying absolule attention lo lhem. As de Souza argues: The role of emotion is to supply the insufficiency of reason by imitilling the encapsulation of perceptual modes. For a variable but always limited time. an emotion limits the range of infonnation that the organism will take into account.thc inferences actually drawn from a potential infinity. and the set of live options among which it will choose. (de Souza 1987: 195).

This analysis of emotion is seen by deSouza (1987: 195) as "one of Nature·s ways of dealing with the frame problem ... Whal the specialised circuits of the brain are designed lo do is facilitate lhe process of fixing on one piece of infonnalion - the representation of an environmenlal stimulus or state of affairs - to 1he exclusion of all others. In relevance theory tenns. these syslems make sure thai the crucial piece of information is less cosily lo process.

Analysing emotions as belid/desire sets It is clear lhal emotions play a crucial role in motivating human behaviour and lhal they are extremely complex: that is whal makes them so interesting. One way of characlerising lhe complexily of human emotions is in tenns of what Rey (1980) tenncd its cognitive properties. For Sloman (1986: 1987) these cognilive properties can be analysed in lerms of belief/desire sets. The sophistication of our awareness of emotional slates depends on the sophistication of the ways in which the cognilive properties of emotions can be represenred. Sloman ( 1987) illustrales the view thai emotions may be partially characterised in such a way with an analysis of anger. The following is a simplified and abbrevialed version of Sloman's analysis.(For a fuller analysis of 'anger·

148

Poi:.IIC EFFECTS

see Sloman 1987: 224-226). For it to be possible lo make the claim in (5). Sloman argues. somelhing like the conditions in (6) must hold: (5) (6)

X is angry with Y

a. X believes that Y did (or failed 10 do) something. b. In doing (or failing to do) lhis something. one of x·s desires wa.. violated.

I will flesh this oul with a simple example. When Gorbachev was released from his dacha in the Crimea in August 1991, afler the abonive coup in what was lhen the Soviet Union. he was angry. One of the people he was particularly angry with was his former friend. Lukyanov. Why was Gorbachcv angry? He believed that Lukyanov had helped to organise the coup (6a) and thai Ibis violaled his desire lo go on running the country. more immediately gelling the 'All Union Treaty" signed. with lois of important etceteras (6b). But arc this belief and frustrated desire enough to make Gorbachev angry? Sloman argues that the conditions in (6) are not sufficient for anger: a further condition expressed here in (7).the creation of a new motive (a.~ Sloman calls it) or desire. is essential before one can justifiably make the claim in (5). (7}

X desires to hurt/harm Y.

For X to be angry with Y, then. a new desire must be created. Alii Sloman phrases it: "Anger involves an insislent and intense non-derivative desire to do somelhing to make Y suffer., (Sloman 86:9). This is a characleristic of all emotional stales. according to Sloman: a belief about some state of affairs or evenl in lhe world causes one desire to be frustrated, or violated. and a new desire lobe created. This new desire disposes the expericncer of the emolion to act in a c.:ertain way. In lenns of the present example, for Gorbachev lo be angry with Lukyanov. if Sloman's analysis is correct. he must also desire lo hurt or harm Lukyanov. This is nol to say that he (or any X) ha."i lo aclUally hurt or harm Lukyanov (or any Y): other desires or beliefs might interfere - fear of the consequences of such aclion. for example. The general rule. however, is thai. if one desire is thwarted by some action of Y and a new desire to hurt or hann Y is created, then X may be said lobe angry withY. Sloman's method. then. involves the conceptual analysis of key emotion terms (such as 'anger') and a fonnulation of lhe results of such analysis in term.'\ of belief/desire sets that characterise the cognilive slate of a mind

EMOTION,

A rnTIJDE AND SEmlMENTALITY

149

experiencing the emotion. The analysis of 'anger' above is very schematic. Sloman's analysis. in fact. allows for much variety and sophistication. as he argues in the following passage: Emotions like anger can vary along different quantitative and qualitative dimensions, such a~: how certain X is about what Y has done. how much X cares about it (i.e. how important and intense the violated motive is): how mm:h harm X wishes to do toY: how important this new desire is, how intense it is, how insistent it is. how long la~ting it is; how much mental disturbance is pnxluced in X: how much physiological disturbance there i~; which a.-.pects of the state X is aware of: how many secondary motives and actions are generated ... (Sloman 86:9-10).

Anger without the desire to harm Y may be interpreted a." exasperation. dismay or annoyance. Emotions may be coloured by secondary emotions. perhaps by socially sanctioned emotions. that respond to primary emotions. {I may be frightened by my anger: I may be angry at my fear.) The emotion strengthens and holds the new desire in the mind. The strength of the anger and desire to harm can vary,depending on the strength of belief in (6a) and the strength of the thwarted desire in (6b). This great range of possibilities is what can make particular emotions and feelings so individual and fine-grained. The variations in belief/desire sets are reflected in the way the emotion is experienced, in the qualitative character of the emotion. The emotion becomes much more complex in a wider context. a complexity which has a corresponding qualitative feel. This is a crucial point in the consideration of poetic effects. where pragmatic processing encourages the accessing of a very wide context. I will now suggest my own brief and sketchy analysis of 'sadness' along the same lines. In order to make the claim in (8), I would suggest that something like the conditions in (9} must hold. (8) (9)

X is sad about not Y. a. X believes the state of affairs Y to be good or valuable. b. X desires to (continue to) experience the state of affairs Y. c. The state of affairs Y is no longer accessible or available to X fort amount of time. d. x·s desire in (9b) is strengthened.

This example may be fleshed out in the following way. Gorbachev believes that being President is a good and valuable thing (for him) (9a). Gorbachcv desires to go on being President (9b). Gorbachev believes he is unlikely ever

150

POETIC EFFECTS

to be President again (9c). As he entertains these beliefs and desires, his desire to be President is strengthened (9d). And so Gorbachev is sad about not being President any more.

Whereas in Sloman's analysis of 'anger' a new desire was created. focussed on the violator of an earlier desire. here there is no new desire. but a strengthening of an earlier violated desire (9b). ll may be the ca'ic that this strengthening involves raising a hitherto unconscious desire to the level of conscious awareness. This kind of analysis allows one to make many detailed distinctions between kinds and degrees of sadness. Different intensities of sadness arc possible depending on the relative strength of the belief in (9a) and the relative strength of the desire in (9b). The state of affairs that one values and desires may be (and perhaps typically is) one in which a person. place. object or experience is considered valuable (in some sense), and in which one desires to be in contact (in some sense) with that person. place. object or experience. The extent to which one believes the person. place, object or experience to be valuable will affect the quality of the sadness. The quality of sadness will depend. as well. on whether it is a person who is gone. a place one is absent from. an object that is lost or an experience that is no longer available. Again, different intensities of sadness are possible depending on the length of time one is to be separated from the person/place/object/ experience (on how the 1 slot is completed in (9c)). If someone one loves depart.;; for a weekend. that is mildly sad: if for a summer, very sad: if forever. extremely sad. Grief is the most extreme fonn of sadness: there is absolutely no hope of being with that person again. The belief in (9a) can be about a state of affairs believed to be good or valuable by X for some other person: one can feel sad on behalf of other people. Grief can, in this sense, be a mixture of personal loss and the sense of another's Joss. There arc many other dimensions to sadness. and many ways in which sadness can mix with or give way to other emotions. It was suggested above that example ( 1), repeated here, evokes the emotion of sadness and that the verbal repetition in (I a) results in the ex pres· sion or communication of a greater degree of emotion than (I b). (I)

a. My childhood days are gone. gone. b. My childhood days are gone.

'My childhood days', like everyone else's childhood days. were good days because they were carefree and innocent. Those days have definitely gone,

EMOTION. AT1111.JDE AND SEtmMENTAUTY

151

and gone for good. except for the memories I will always treasure. I am sick of my present careworn. and far from innocent. existence. I yearn to experience life as it used to be when it was less complex. when others did lhe worrying. I don'l always feel like this. of course, but when I do remember my childhood the desire to experience those days of carefree innocence is strengthened and I feel sad. What this example does is set up the (belief/desire set) conditions for 'sadness'. Childhood is believed lobe valuable. X's desire for the experience that 'childhood' offers - innocence. freedom from responsibility. securily. etc. - can no longer be fulfilled. Concentrating on these beliefs and desires strengthens the impossible desire to experience thai kind of carefree innocence again. Compare two other examples discussed in chapler 5 and repeated here as (10) and (II}: ( 10) The pubs are closed. closed. (II) My teenage days are gone, gone. Wilh regard to ( 10). pubs arc nol quite precious enough. for mo.'\t people. for the emotion of sadness to be evoked. Also the time factor is important: the pubs will be open again tomorrow morning. Because a contextual search is initiated which has nowhere logo, the hearer has no olher recourse lhan lo distance himself from the proposilion expressed. which is understood as ridiculous and possibly as having been ironically intended. In (II) 'teenage days' do not have a similar stereotypical set of positive auitudes associated with them a.'\ 'childhood' .It is sad that they have gone, bul nola." overwhelmingly sad a.-. the fact that 'my childhood days· have gone. Examples such as (lb) vaguely evoke the emotion and feeling of sadness because the right conditions apply: the desire for what childhood represenls security. innocence.lack of worry. and other good things - is strengthened. a.-. it is recognised that lhe experience of childhood can never be auained again. The emotion and feeling are much stronger in (Ia) because the concept CHILDHOOD is much more extensively explored in the search for an interpretation consistent with the principle of relevance. Many more assumptions about childhood are activated and a greater variety of things thai are good about childhood are made more highly salient: there are simply more reasons and stronger reasons for feeling sad.

152

PoETIC EFFECTS

Emotions and attitudes Clarifying the different properties of emotional states and lhe role they play should help to avoid a certain amounl of confusion in discussing lhe role of emotion in poetic effects. An analysis of the cognitive properties of emotions can be used to show how poelic effects create a wider range of. and stronger

conditions for, evoking lhe feeling or qualirarive properties of an emotion. But there must be more to say lhan this. When poeLo,; and literary crilics claim lhat emolion is central to poetry and poetic effects. what exactly do they mean? Fruslraled Connex South East pa.o;scngcrs oflen resort to the expression of 'emorional concepts'. bul they do not necessarily achieve poetic effects when they do so. In section 6.2 I discu.o,;sed Rey's notion of emolion as a complex stale involving four distinct propenies. Ofien lhc term 'emOiion' is used loosely lo refer to just one. or several. of lhese propenies. When Larkin refers to ''emotional concepts'' ,lhen. perhaps he is using Ihe term loosely in this way. I will go on to suggesl that he is. in fact. really inleresled in lhe qualilative propenies of cmolions. Before considering lhis queslion funher. there are a number of olher frequent terminological confusions thai need to be clarifred. Sloman (1986) illuslrales the inconsistency in Ihe ordinary usage of such lerms as 'emotion·. He suggests lhat X might hold at one and the same time that: (a) (b) (c)

X loves her children deeply. Love is an emotion. X is not in an emolional state.

The apparent contradiclion here can be resolved, Sloman argues, if we draw a distinclion belween emotitm.v and allitude.f. An emotion. such as fear or anger, is a temporary slate. a response to some perceived event or stale of affairs in lhe world. This causes a physiological response and.lypically. also a behavioural and qualilative response. An altitude. such a.ril'hte 8: 103-127. 1987

194

PoETIC EFFECTS

(forthcoming) Thoughrs mrd Ulleranct's: Tht> Pragmatics of Explicit Conrmrmicatian. Oxford: Blackwell. Chomsky, Noam 1957 S)'nta,·lir Strllcllrn.f. 11lc Hague: Mouton. 1986 Knawlrdgt> of LtlngutJRe: Its Nallrre. Origin. and Use. New York: Pracgcr. 1988 Longutlgr and Problems of Knowledge: nre Managua L«tuus. Cam-

bridge. Mass.: MIT Press. 1995 UJngua~ and Thought. Wakefield. Rhode Island: Moyer Bell. Clark. Herben and Marshall. C. 1981 ''Definite reference and mutual knowledge''. In Elements of DisN111rse Understanding. A. Joshi. B. Webber & I. Sag (cds). 10-63. Cambridge: Cambridge University Pn:ss. Cohen. Ted 1978 ''Mclaphor and lhe cultivalion of intimacy''. In o, Mt'taplror. S. Sacks (cd.). 1-10. Chicago: Univer5il)' of Chicago Press. Coombes, Henry 1963 Litf.'ramre and Critirism. Hannondswonh: Penguin. Culler, Jonalhan 1975 Structuralist Poetin: Structuralism. Li11guistics and the Stud)· of Literature. London: Rootkdge and Kcgan Paul. 1983 On Orrtm.ftriU"tirm: TltMr)l and Critic-ism after Strllcturali.fm. London: Rootlcdge and Kegan Paul. CurcO. Carmen 1997 The Pragmalics of Humorous l11terpretations: A Relel'Dnce-Throretic Approach. University of London. Ph.D. thesis. Cutler. Anne 1987 '"The la.'Ok of the speaker and the ta.'ik of lhc hearer"'. Behavio11ral and Brain Sdenc-es 10: 715-716. 1992 '"AuditOI)' lexical access: Where do we stillt?"'" In l.Lxical Representatioll a11tJ Prores.f, W. Marslen-Wilson (ed.). 342-356. Cambridge. Mass.: MIT Press. 1994 '"The perception of rhythm in English'" Cogniti011. 79-81. Damasio. Antonio 1989 ··concepts in the Bruin". Mind and IAngutJge 4: 24-28. 1994 Drscartes· Error: Emotion. Reason. and tire H11man Brain. New York: Avon Boob.

Danto. Arthur 1981

The Tran.fjiguration of the Commmrplw:e. Cambridge. Mass.: Harvard University Press. ""Responses and replies··. In Damo and his Critics. M. Rollins (cd.), 193216. Oxfonl: Blackwell. Davies. Manin & Humphn:ys. Glyn W. (eds) 1993 CoiiSC"iousne.ss: P.sycholoRical and Pllilosoplriral Essays. Oxford: Blackwell.

1993

REFERENCES

195

Dawkins. Marian Stamp 1993 Tlrrough Our E,.\~S 011ly? 17lt! Swrdrfor Animal Con.n·iorLfnt!.u. Oxfonl: W.H.Frccman. Dennen. Daniel 1993 Consciousness Explai11t!d. London: Penguin. de Sousa. Ronald 1987 The Ralionali()' of Emotions. Cambridge. Ma.'l.'i.: MIT Press. Devin. Michael and Stcrelny, Kim 1987 U111guage and Rt!ali()•: An llllroduction to lire Plrilosop/1)' of Longuagt!. Oxford: Blackwell. Eaglelon. Terry 1983 Liluary· Theory: An lnlroduclion. Oxford: Blackwell. Easthope, Anthony 1985 ''Why most contemporary poetry is so bad and how post-structur.dism may be able to help". PN Rt!\'irK" 48: 3~38.

&.-o. Umberto 1979 A Thror,- of Semiotics. Bloomington: Indiana University Press. Eichenbaum. Boris 1965 "The theory of the 'Formal Method"'. In R11ssian Fannalist Crltici.fm: Fo11r Essays.L.T. Lemon and MJ. Reis (eds), 99-139. Lincoln: University of Nebraska Press. Eliot.T.S. 1957 On Poetn and Ports. London: Faber and Faber. Empson. William · 1984 St!l"ell 1',\'{WS of Anrhigui()•. London: Hogarth Press. 1985 Tire Structurr of Complt!X Words. London: Hoganb Press, Erlich. Viclor 1981 Rrmi1111 Formalism: Histor)' - Doctri11e (3rd eel.). New Haven and London: Yale Universily Press. Fabb, Nigel 1988 "Saussure and litcruy theory: From lhc perspective of linguistic."". Critical Quarterly 30: 58-72. 1995 'The density of response: A problem for liaerary criticism and cognitive sc::ienc:e", In Ungui.flic Approaches to Utemturr: papers ill literary s()•listics. J. Payne (ed.). 143-157. Birmingham: English Language Research. University of Birmingham. 1997 Ling11isticsand Uterot11re. Oxford: Blackwell. Fabb. Nigel. Auridgc, Derek. Duranl. Alan and Mac:Cabc. Colin (eds) 1987 The Li11guistics of Writi11g: Arg11me111S Betwu11 la11gUGgt! and Lileraturr. Manchester: Manchesler Universily Press. Fabb, Nigel and Durant. Alan 1987 "lntroduclion: The: linguislic:s of wriling: Relrospect and prospecl after twenty-live years". In The Unguistics of Writing: Arg111ne111s Brhl't!en UJnguagt! a11d Uteroturr. N. Fabb. D. Attridge. A. Durant and C. MacCabe (eds), 1-14. Manchcslcr: Manchester University Pres.'i.

196

PoETIC EFFECTS

Fish. Stanley 1980 1987

1991

Is th~u a tt'xt in tlris da:u? Cambridge. Ma.~~.: Harvard University Press. ''Withholding the missing portion: Power. meaning and persuasion in Freud's 'The wolf man"'. In Tlrr Linguistics of Writing: Argument.f Bt>twern UJngutJRt' and Litrraturr. N. Fabb. D. Attridge, A. Durant and C. MacCabe. Colin (eds). ISS-172. Manchester: Manchesler University Press. Doing M·hat Comes Namrall_\'. Cambridge. Mass.: Harvard University Pn:ss.

Fodor. Jerry A. 1983 Tht' Mndularil)' of Mind. Cambridge, Mass.: MIT Press. 1987 ''Modules. frames, fridgcons. sleeping dogs, and the music of lhc spheres".

1990

1992

In Modtllarity in K11ow/rdge Represe111ati011 and Natural-Lartglrage U11· dersta11ding. J. Garfield. (ed.). 26-36. Cambridge. Mass.: MIT Press. A Tlrrory· of Contt'nt muJ Othrr E:s.my,t. Cambridge. Mass.: MIT Pres.'i. ''The big idea: Can lhere be a sc:ience of mind'r Times Literar): Supplement. 3 July 1992. no.46S7: 5-7.

1993

''Dijd Vu all over again: How DaniO's aesthetics recapitulates the philosophy of mind". In Du11to and his Critics, M. Rollins (ed.). 41-54. Oxford:

Blac:kwell. Fowler. Roger 1986 Ling11istic Critidsm. Oxford: Oxford University Press. Freeman. Donald (ed.) 1981 Essa,·s in ;m·listics. London: Methuen. Fry. Roger · · I 920:1981 Visitm &: Ihsig11. Oxford: Oxford Univenity Press. Fussell. Paul 1979 Poetic Meter and Ptwtit: Form. New Yort: Random House. Gerrig. Richard 1989 "Empirical conslrilints on computational theories of metaphor: Comments on Indurkhya". Cogniti\'r Science 13:235-241. Gerrig. Richard. and Healy, Alice 1983 "Dual processes in metaphor understanding: Comprehension and appreciation". Journal of Experimenral Ps)•chology: Uarning. Mrmory· and Cogni· lion 9.4: 667-675. Gibbs. Raymond W. 1984 "Uteral meaning and psychological theory". Cogniti\·e Scie11ce 8: 275-304. 1989 "Understanding and literal meaning". Cogniti\•e Science 13: 243-251. 1994 The Poetic·s of Mind: Figurath·r Thought, La11guage a11d Understanding. Cambridge: Cambridge University Press. Gibbs, Raymond W. and Nayak. Nandini 1989 "Psycbnlinguistic studies on the syntactic behaviour of idiom.o;", Cog11iti1't' P.f)'t.'hology 21: 100-138. Goodman. Nelson 1976 Language.f of An. Indianapolis: Hackett.

REFERENCES

197

Grice, H. Paul 1957: 1989 "Meaning". In Sllldies in the Wa)· of Words. 213-223. Cambridge. Mass.: Harvard University Press. Ground. tan 1989 Art or Brmk? Bristol: Bristol Classical Press. Hawkes. Terence 1977 Structurolism arul SemiotiC's. London: Methuen. Heaney. Seamus 1980 Srlectf!d POf!ms.l..ondon: Faber&. Faber. 1980 Pnocrupations: Sel«ted prose /968-1978, London: Faber &. Faber. Hollander. John 1989 Rh)·me's RM.fOII. A g11ide toE~rglish 1-'tr.fe. New Haven and London: Yaic University Press. Humphrey. Nicholas 1992 A History of the Mind. London: Vintage. Huxley. Aldous 19S4:1971 The Doors of PerC't'ption. Harmondsworth: Penguin. lngarden. Roman 1979a The literary Work ofArt (transluted by G.G.Grabowicz). Evanston: Nonhwestcm University Press. 1979b The Cogrritimr of the literary· Wt~rk of Art (translated by R.A.Cmwlcy and K.R.Oison), Evanston: Nonhwestem University Press. Iscr. Wolfgang 1978 TlreArtofReading.l..ondon: Routledge & Kegan Paul. Jackendoff. Ray. 1987 Consciousness and the ComputatiDitQ/ Mind. Cambridgc.Ma55.: MIT Press. Jackson, Frank 1982 "Epiphenomenal qualia". Philosophiral Quarterly 32: 127-136. Jakobson. Roman 1960 "CIO!Iing statement: Linguistics and poetics", In Style irr language, T. Sebeok (cd.). 350-371. Cambridge. Ma.'is.: MIT Press. 1968 "Poetry of grammar and grammar of poetry". lirrgua 21:597-609. Jefferson. Ann and Robey. David (eds) 1982 Modem literary· Theory: A Comparati1•e /rllrotlllrtion.l..ondon: Batsford Academic. Kennodc, Frank 1989 An Appetite for Poetry·: Essays in Uterary•lnterpretatimr.l..ondon: Fonlana Press. Kiparsky. Paul 1981 ""The role of linguistics in a theory of poetry", In Essays in Sty/isrics, D.Frecmun (ed.). 9-23. London: Methuen. 1987 "On theory and interpretation". In Tlw linguistics of Writing: Ary:uments between language and literature, N. Fabb, D. Attridge, A. Durant and C. MacCabe (eds). 185-198. Manchester: Manchester University Press.

198

PoETIC EFFECI'S

l..akoff. George and Johnson. Mark 1980 Metaphor.f We Live B,·. Chicago: University of Chicago Press. Lakoff. George and Turner, Mark 1989 More than Cool Reasrm: A Field Guide to PMtic Metaphor. Chicago: Univcrsily of Chicago Press. Larkin. Philip 1983 Required Writing. London: Faber& Faber. l...awrenc:c.D.H. 1967 Selected Literary Critiri,fm. London: Heinemann. L=h.Gcoff..y 1969 A Lingui.ftic Guide to English PMtry. London: Longman. Leech. Geoffrey and Short. Mick 1981 Style in Fiction: A Linguistic llllrodtiC'tion to English Fictional Prose. London: Longman. Levinson. Stephen 1983 Pragmatics. Cambridge: Cambridge University Press. Levi-Strauss. Claude 1969 The Raw and the Cooked (translaled by J.&D.Wcightman). London: Jonathan Cape. Lodge, David 1986 Working with Strll,'tllrtrli.fm. London: Ark Paperbacks. MacMahon. Barbara 1996 ''Indirectness. rhetoric and interpretive use: Communicative strategies in Browning's My Lnst D11d1t>ss". Lnng~ragt> mrd Litt>raturt> 5.3: 209-223. Marcel, Anthony 1988 "Phenomenal experience and func:tionalism".ln Consciousness in Contemporary Sciellt.'t', A. Marcel and E. Bisiach (cds), 121-158. Oxford: Oxrord Universily Pn:ss. Marslen-Wilson. William 1987 ''Func:lional parallelism in spoken word recognilion''. Cognitio11 25: 71102. 1992 "Acces. Understanding. J. Garfield (ed.). 37-62. Cambridge. Mass.: MIT ~ss. Marslen-Wilson. William.and Zwitscrlood. Picnic 1989 "Accessing spoken words: lbe importa.nce or word onsels". Journal of Experimental Psychnlng)'.' Hrm1a11 Pt>rrtption anJ Performant't', 516-585.

REFERENCES

199

McGinn. Colin 1993 ··consciousness and cosmology: Hypcrdualism venlilalcd".ln Con.u:iousn~ss,M. Davies and G. Humphreys (cds.). 155-177. OxJord: Blackwell. Meutsch, Dielrich and Schmid!, Siegfried 1982 "On the role of convenlions in underslanding lilerary 1ex1s''. Parties 14: 551-574. Middlelnn Muny. John 1922 1'lu! Prohl~nr of Sl)·l~. London: Oxford Universily Prc5.o;. Miller. George 1979 "Images and models. similes and melaphorll" .In M~taphor and Thrn1ght, A. Onony (ed.). 202-250. Cambridge: Cambridge Universily Press. Mwray.Les 1991 Collrct~d Pornrs. Manchesler: Can:anc:l. 1992 Th~ Paperbark Tre~. Manchesler: Carc:anel. Nabokov, Vladimir 1983 Uctur~s on Lit~raturr. London: Piador. Nagel, Thomns 1974 "Whul is illike lobe a bal?'' Philosophical R~l·i~w 83: 435-450. 1986 Th~ Vi~wfrmnNowh~r~.O:dord: Oxford Universily Press. Neisser.Ulrich 1982 M~nrory• Ob.f~n·nl: R~nrmrMri11g in Natural Ctml~:rt.f. San Francisco:

Freeman. Noh. Eun-Ju 1998 Oalley. Keith 1992

The Senranlics and Pragmatics of Metarepr~s~nration ill English: A ~mnt·e-ThnJutic ApprtJGch. Univc:rsily of London PhD.Ihesis.

R~l­

B~st U.id Schenr~s: Th~ Psycholog_v of Emotions. Cambridge: Cambridge University Press. Onony. Andrew (ed.) 1979 M~taphor arid Tlrouglll. Cambridge: Cambridge University Press. O'Toole. L. Michael and Shukman, Ann 1977 "A contextual glossacy of fonnalist h:rminology''. In Rus.dan P~tics in Tronslation vol. 4: 13-48. Papafragou. Anna 1995 ''Figurative language and the semantics-pragmatics distinction". IAnguag~ and Lit~rature 5.3: 179-194. Pilkington. Adrian 1991 "Poetic Effects: A relevance theory perspective". In Lit~rory Pragmatics. R. Sell (ed.). 44-61. London: Routledge:. 1992 ''Poetic effects", Ung11a 87: 29-51. Propp. Vladimir 1968 Morplwlog,\' ofth~ Folktal~ (tr.ms. L Scotl). Austin: University of Texas. Proust, Marcel 1988 Agairl.fl Saiiii~·Belll'e a11d Oth~r Essays (trans. J.Sturroc:k). London: Penguin.

200

PoETIC EFFECTS

Pugmire, David 1998 Rt'disl'fl\'t'ring Enrotim1. Edinburgh: Edinburgh University Prcs!l. Rey.Georgcs 1980 ''Functiona1ism and lhe emotions". In Explaining Emotion.,, A. Rorty (cd.J. 163-9S. Berkeley: University ofCalirornia Pn:ss. Richards. I.A. 1924 Principle., of Utt'rory· Criticism. London: Roullcdge &. Kegan P.•ul. 1929 Prac·tirol Critici.fm. London: Roullcdge &. Kegan Paul. Riffillerre. Michael 1966 ''Describing poetic structures: Two approaches lo Baudelaire's 'Lc:s Chats'''. Yalt' Fretu·h Studir., 36-37: 200-242. Saussure. Ferdinand de 1974 Coursr in Gr11era/ Unguistics (trans.W .Ba.'lkin). London: Fontana. Schank,R. 1982 ''Reminding and memory organisalion: An inb'oduction to MOPs". In Strategies for Natural LangiUigr Processing • W. Lehnert and M. Ringle (eds.), Hillsdale: Lawrence Erlbaum. Schank. Roger and Abelson, Robert 1977 S(·ripts. Plans. Goals a1ul Under.flanding. Hillsdale: Lawrem:e Erlbaum. Schmidt. Siegfried 1982 Foundalions for thr Empiriral Study of Utrraturr: Thr Componrlll.f of a Busir Throry. Hamburg: Helmut Buske. Searle.John 1979 ''Metaphor''. In Mrtaphor and Thm1ght. A. Ortony (ed.). 92-123. Cam· bridge: Cambridge University Prc5S. Scbcok. 'I11<mw 1991 A Sign Is Jrut A Sign. Bloomington: Indiana University Press. Selden. Raman 1989 A R«Jdrr'., G11idr to Cmrtrnrporary· Utrrary Throry (2nd. ed.). Hemel Hempstead: Harvester Wheatsheaf. Sell. Roger (ed.) 1991 Utrrary Pragmatirs. London: Routledge. Shklovsky, VikiDr 1917; 1965 "Art as ICChnique".ln Ru.uia11 Forma/i.ft Critici.fm: Four Eua)'·' L.T. Lemon and MJ. Rcis (ed,.). 3-57. Lincoln: UniveBity of Nebraska Press. Shoemaker. Sydney 1991 "Qualia and consciousness". Mind, voi.C. number 4: 507-524. Sloman. Aaron 1986 "Motives. mechanisms and emotion,.... Uni,'t'rsit)• of Sussrx. Cog11iti~·r Studirs Rrsrarc-h Paprrs. 62. 1987 "Motives, mechanisms and emotions''. Cog11ititm a11d Emotion 1(3): 217233. Smith. Alexander 1835 ''The phiiO!iOphy of poetry". Blar:lcwoods (Edinburgh). 827-839.

REFERENCES

201

Sperber. Dan I 975 Ret/linking Symboli.fm. Cambridge: Cambridge University Press .. 1985 On Anthropological Knowledge. Cambridge: Cambridge University Press. 1996 Explai11i11g Culmn. Odord: Blackwell. Sperber. Dan and Wilson. Deirdre 1986a Reln'tloce: Commu~rication and Cognition. Oxford: Blackwell. 1986b "Loose lalk". PTOC'eeding., of tM Aristotelian Snciety LXXXVI: 153-171. 1987 "Precis of relevance: Communication and t:ognitionM. Belun·ioural and Brain Sciem·es 10: 687-754. 1995 Rele~'tlnce: Commu11ic'tltion and Cognition (2.... eel.). Oxford: Blackwell. Spiller. Michael 1988 Review article of "Fabb, Nigel. Attridge. Derek. Durant. Alan and MacCabc. Colin (cds) 1987 The UnRuistics ofWrilillg: Arg1m1ents Behl·een LangtfORe utrd Uteramre". Manchester: Manchester Uni"crsily Press. Parltmce 1(1): 16-19. S1rawson, Galen 1994 Mental Rea/it)'. Cambridge. Mass.: MIT Press. Swinney. Da"id I 979 "Lexical access during comprehension: (Re)consideralion of effects" .Jour· nal of Verbal Learni11g and Verbal Beha1·im.r 18.6: 645-60.

Taft,L. 1984 Prosodic Cm1Strai11ts and Lexical ParsinR Strategies. Unpublished PhD thesis. University of Massachusetts. Verdonk. Peter (cd.) 1993 Twemietlr-Century Portry: From Te.'ft to Comext. London: Rautledgc. Vicente, Begoiia 1993a "Literal and non-literal uses in speech: Some clues to the structure of concepts," (ms.). 1993b "Metaphoric combimllions and the structure of concepts". (ms.). Walcs,Katte 1992 Tire Lons11age of Jamf!.f Joyce. Basingstokc: Macmillan. Werth. Paul 1975 "Roman Jak.obson's verbal analysis of poetry", Journal of Unsuistks 12: 21-73. Widdowson. Henry 1975 Stylistic., and the TeaclrinR of literature. London: Longman. Wilson. Deirdre and Sperber, Dan 1992 "On verbal irony". Unsua 87(112): Sl--76. Wimsam. William and Beardsley, Monroe 19S4 "'The intenlional fallacy".ln Th~ Verbal lt'OII, W. K. Wimsatt (ed.), 3-18. London: Methuen. Willgcnstein. Ludwig 1958:1976 Philosophical lm·estiRations. Oxford: Blackwell. Young. Robert. (ed.) 1980 Ur~tyi11g the Te.tt: A Post·Strut·turalist Reader. London: Routledge & Kcgan Paul.

Index of Names A Abelson. Roben 113 Anderson, John 114-115

Aristotle 86 Anridge, Derek 26 Austin.J. 31

8 Barsalou,LaurenceW.97,98,107.117-

Dawkins, Marian Stamp 147, 187 Dennett, Daniel 175. I 86 Derrida, Jacques 38. 4.' de Sousa. Ronald 147

Devin. Michael43 Duehamp. Marcel4 •.H Durnnt. Alan 9-10,40.56-57

E

118 Banhcs, Roland 23, 39. 57, 5R. 68 Bartlett, Frederic 116 Baudelaire. Charles 17 Beardsley, Monroe 65 Birch. David 39, -l-1. 42. 46, 70 Black, John 114 Blakemore, Diane 125. 168

Eagleton, Terry 28, 40,42,44,45, 58. 65, 66,68 Easthope. Anthony 47 Eco, Umberto 24 Eichenbaum, Boris II Empson, William 41.-12. 46.69

Bower. Gordon 114

Fabb. Nigel 9-10, 26. 40. -II, 46. 56-57. 165. 177 Fish. Stanley 8, 32-38. 57 Aaubc:rt. Gustave 101. 106. IR2-18J Fodor. Jerry A. 3-8.21 ..15. 64. 67, 68. K2. 171-175. 186 Foucault. Michel 38 Fowler. Roger 70 Frost. Robert 109. 129-130 Fry. Roger 186 Fussell. Paul 131-133. 136. 138. 140

Budd, Malcolm 2

c Calvina, halo 145 Carston, Robyn 96 Chomsky. Noam 45. 53. 175 Clark. Herbert 61 Cohen, Ted 167 Coleridge. Samuel Taylor I Coombes. Henry 119 Cowper. William 133 Culler. Jonathan 7-8, 17. 19. 24-27. 32 Cun:O. Carmen 169 Culler. Anne 63. 132. 137

D Damasio, Antonio 153-154, 191 Danto. Arthur 3-4

G Gcrrig. Richard 89-90. 99. 104--105. 108. Ill Gibbs. Raymond W. 89-90.94.99, 108, 110-111,120.166--168 Goldsmith. Oliver 133 Goodman. Nelson 165

204

Pornc EFFECTS

Grice. H. Paul 5. 30-31. M. 71-73.8590,98. 130 Ground, lan 4

H Hagopian, John 49

Hawkes, Terence40,44,60-