The cognitive mechanisms of adversarial humor

34 Pages·2006·0.3 MB·English

The cognitive mechanisms
1
2 of adversarial humor
3
4
5 TONY VEALE, KURT FEYAERTS, and GEERT BROˆNE
6
7
8
9
10
11
12 Abstract
13
In this paper, we provide an in-depth cognitive analysis of a specific humor
14
strategy we coin ‘‘trumping’’, a multi-agent language game that revolves
15
around the subversion of the linguistic forms of exchange. In particular, we
16
illustrate how, in a conversational setting, agents can ‘‘reflect’’ and ‘‘dis-
17
tort’’thelinguistic-conceptualconstrualofeachothers’utterances.Because
18
this reflection or parallelism in the trumping game can be situated on
19
di¤erent levels of linguistic organization, a multi-dimensional semantic-
20
pragmatic account is proposed. Using insights from cognitive linguistics,
21
we show that adversarial agents exploit the conceptual mechanisms under-
22
lying the opponent’s utterances in order to turn the tables in the humor
23
game. In doing so, an agent can trump an adversary by demonstrating a
24
‘‘hyper-understanding’’ of the lexico-conceptual meaning of an opponent’s
25
utterance.Thissubversionofconstrualoperationslikemetaphor,metonymy
26
and salience leads to a sudden manipulation of the discourse space that has
27
been set up in the previous utterance(s) (Langacker 2001). In general, by
28
providing an analysis in terms of basic principles of semantic construal, we
29
arguethatacognitivelinguistictreatmentofhumorhasanecologicalvalid-
30
ity that is lacking inmost linguistic humor research.
31
32
Keywords: Adversarial humor; cognitive ecology; cognitive linguistics; se-
33
mantic construal; trumping.
34
35
36 1. Introduction
37
38 However one might speculate about the inner workings of humor, it re-
39 mainsaprimarilysocialphenomenon.Whileitmaybemeaningfultotalk
Humor19–3(2006),305–338 0933–1719/06/0019–0305
DOI10.1515/HUMOR.2006.aaa 6WalterdeGruyter
(AutoPDFV77/6/0608:20) WDG(148(cid:1)225mm)TimesM J-1537Humor,19:3 PMU:H(A1)24/05/2006 pp.305–338 1537_19-3_05 (p.305)
306 T. Veale etal.
of private laughter, humor is at its most potent in a multi-agent setting,
1
andatitsmostpointedwhenproducedbyoneagentattheexpenseofan-
2
other. Adversarial humor is thus an important branch of humor research
3
sinceitallowscognitiviststounderstandandmodelthesocialinteractions
4
of humor in relation to the specific goals and motivations of the agents
5
that produce it. It is the nature of these interactions, in particular how
6
they exploit and subvertlinguistic norms, thatinterests us in this paper.
7
By definition, adversarial agents have competing interests, which can
8
produce conflicting communication goals if this competition is expressed
9
verbally. Furthermore, zero-sum goals, such as an argument concerning
10
the truth or falsity of a given proposition, can be advanced by undermin-
11
ing the competing goals of other agents or by undermining those other
12
agents themselves. To therefore understand the working of adversarial
13
humor, one must understand not only the specific communication goals
14
of the agents involved, but more significantly, one must also understand
15
the personal history that leads to an agent possessing those goals. This
16
makes adversarial humor an issue of encyclopedic proportions, where
17
the boundary between speaker and utterance is blurred to the point that
18
an integrated representation is required. The framework of cognitive lin-
19
guistics, with its dynamic account of meaning in terms of comprehensive
20
conceptualization, provides a number of methodological tools to tackle
21
these complexities. In this paper, one such representation, Langacker’s
22
(2001) model of meaning construction in discourse, is used as a method-
23
ological handle onthetreatment of adversarial humor.
24
To reduce the phenomenon to a more manageable level, it is useful to
25
demarcate a sub-trope that adheres to a well-defined usage pattern while
26
nonetheless exhibiting all the creativity of the phenomenon. To this end
27
we introduce a species of adversarial humor we call trumping, which can
28
be viewed as a form of multi-agent language game that generates its hu-
29
morouse¤ectthroughsubversionofthelinguisticformsoftheexchange.1
30
Furthermore, the particular type of trumping we analyze in this paper
31
exhibits a strong lexical connection between the utterances of di¤erent
32
agents, allowing us to investigate the workings of trumping in a way that
33
sheds light onthe broader phenomenonof adversarial humor.
34
35
36 1.1. The ecology of trumping humor
37
Since humor is undoubtedly a cognitive phenomenon, and verbal humor
38
in particular a linguistic phenomenon, it is natural to conclude that a
39
(AutoPDFV77/6/0608:20) WDG(148(cid:1)225mm)TimesM J-1537Humor,19:3 PMU:H(A1)24/05/2006 pp.305–338 1537_19-3_05 (p.306)
Cognitive mechanismsof adversarialhumor 307
cognitive linguistic approach to humor research should yield the deepest
1
and most coherent insights. However, if this position is to transcend the
2
vacuityofaplatitude,orworse,anideology,itisnecessarytoshowmore
3
concrete benefits of acognitive linguistic perspective.
4
In eschewing a strict modular view of cognitive function, cognitive lin-
5
guistics encourages theories to exhibit ecological validity. Humor does
6
notoperateinasocialoracognitivevacuum,soneithercanitbestudied
7
andtheorizedinisolationfromrelatedcognitivephenomenathatmayin-
8
fluence its function or share a common representational basis. In this re-
9
spect, humor occupies one point on a continuum of cognitive behaviors
10
with which it shares a family resemblance to greater or lesser degrees.
11
For instance, the form of adversarial humor we consider in this paper
12
has strong family resemblances to social forms of language such as
13
double-grounded insults (e.g.,referring to a surgeon as a ‘‘blunt scalpel’’)
14
andargumentationbymetaphor(e.g., ‘‘you sayallmenarepigs, but you
15
certainly seem to love bacon’’), and social forms of behavior such as role
16
reversal, competitiveness and one-upmanship.
17
Inparticular, the kind ofadversarialhumorwe considerherehas asits
18
logical coretheideathatone speakermay,linguistically speaking,snatch
19
victoryfrom thejawsofdefeatbyturningthetablesonanopponent. The
20
e¤ect of this reversal is to elicit not just a sense of victory in the agent it-
21
self, but a form of admiration from any observers, while perhaps earning
22
thegrudgingrespectoftheopponent.Soourreasonsforappreciatingthis
23
kind of humor in others more quick-witted than ourselves is related to
24
ourappreciationoftheunexpectedlast-minutetriumphofthosemorehe-
25
roic than ourselves (see also Gruner 1997). In this ecological view, di¤er-
26
entphenomena(humor,competition,etc.)arerelatedbyvirtueofsharing
27
acommonmechanismorsetofmechanisms,andbyvirtueofelicitingre-
28
lated e¤ectsin the participant (cheering, jeering,laughter, etc.).
29
This notion of turning the tables on an opponent is not just a motif of
30
adversarialhumor, but isindicative of a deeper mechanism that has been
31
studied both in humor research, in the guise of figure-ground-reversal in
32
the Attardo’s General Theory of Verbal Humor (GTVH, Attardo 1994,
33
2001a),andincognitivelinguisticsasthemechanismofconceptualprofil-
34
ing(Langacker1987).IntheGTVH,figure-groundreversalisseenasone
35
ofpossiblymanydi¤erentlogicalstrategiesormechanismsforgenerating
36
ahumorous e¤ect;arepresentative exampleistheclassofjokeswherean
37
easy solution to a problem is eschewed in favor of a ridiculously hard
38
(and thus, extremely stupid) solution, such as rotating an entire room
39
(AutoPDFV77/6/0608:20) WDG(148(cid:1)225mm)TimesM J-1537Humor,19:3 PMU:H(A1)24/05/2006 pp.305–338 1537_19-3_05 (p.307)
308 T. Veale etal.
aroundasockettoscrewinalight-bulb.Cognitivelinguistics,incontrast,
1
seesthefigure-grounddistinctionasmuchmorefundamentaltothework-
2
ings of thought. Thus, in the work of Langacker and others, the profiling
3
ofaconceptualstructuretohighlightcertainelementsmorethanothersis
4
seenascentraltothemechanismthatgiveswordstheirmeanings.Forin-
5
stance, one cannot define the concept of hypotenuse without first assum-
6
ing the existence of a conceptual structure that represents right-angled
7
triangles, and without secondly assuming the ability to foreground one
8
or more of the elements in the structure (e.g., the diagonal) while back-
9
grounding others.Thefact thatprofiledaspectsof astructure will appear
10
more salient than others allows linguistically-creative people to achieve
11
humorous misdirection, causing observers to focus on the profile (or fig-
12
ure)while theessenceof thejokegoesunmarked inthebasestructure (or
13
ground).
14
One of the basic tenets of the cognitive linguistic tradition then is that
15
the figure-ground distinction is given the status of a cognitive mechanism
16
that is neither ad-hoc nor specific to humor, but which instead applies to
17
each and every lexical-conceptual phenomenon. According to this view,
18
CLdescribesmeaning asadynamicvalueresidinginthetension between
19
alinguisticunit’sprofile(figure)anditsbase(ground).Whereastheprofile
20
(also ‘‘focus of attention’’) defines the entity designated by the linguistic
21
expression, the base represents the profile’s background, in which di¤er-
22
ent kinds of conceptual structures are activated with di¤erent degrees of
23
salience. It is this ability of cognitive linguistic treatments of humor to
24
tap into a powerful cognitive substrate that transcends humor while link-
25
ing it to a spectrum of related phenomena that provide these treatments
26
with an ecological validity that is generally missing in other accounts or
27
investigative frameworks.
28
29
30
1.2. Plan of thepaper
31
32
Our goal inthispaper isto investigate the complex interplay of speakers,
33
utterances and meanings in multi-agent humor by framing a particular
34
form of adversarial language game that we call ‘‘trumping’’. This lan-
35
guage game has the advantage of being su‰ciently circumscribed to per-
36
mitafocusedanalysis,while,inthehandsofcreativespeakers,beingsuf-
37
ficiently open-ended to be non-trivial and thus scalable to other forms of
38
humorous discourse. In e¤ect, trumping represents a ‘‘horizontal’’ rather
39
(AutoPDFV77/6/0608:20) WDG(148(cid:1)225mm)TimesM J-1537Humor,19:3 PMU:H(A1)24/05/2006 pp.305–338 1537_19-3_05 (p.308)
Cognitive mechanismsof adversarialhumor 309
than a ‘‘vertical’’ slice of humorous behavior that cross-cuts through the
1
phenomenon as awhole.
2
As noted under the rubric of cognitive ecology above, trumping shares
3
many features with other cognitive processes, whether humorous or oth-
4
erwise,thoughitisbestexploredinthecontextofadversarialinteraction.
5
Inthenextsectionthen,wepresentadiscussionofthemostrelevantpast
6
research onthe topic of adversarialhumor, before we consider, ina third
7
section, the various ways in which the core trumping pattern can be
8
instantiated.Wethenorganize theseinstancestocreateatypology ofdif-
9
ferent trumping strategies. This typology will reveal the most fruitful
10
trumping strategy on which to focus our analysis e¤orts, yielding a spe-
11
cific sweet-spot in the space of adversarial humor for us to investigate in
12
this and future papers. Our analysis will draw upon some powerful con-
13
cepts from the field of cognitive linguistics, to explain how connections
14
are drawn between the utterances of di¤erent agents, and how inferences
15
canbemadebetweenwhatanagentsaysandwhatanagentintends.Ina
16
fourth section, we present an inventory of these construal mechanisms,
17
before, in section on ‘‘complex construal’’, we show how these mecha-
18
nisms interact in complex ways to produce the humorous e¤ect in di¤er-
19
ent instances of trumping. Throughout, we emphasize the fact that none
20
ofthesemechanismsarehumor-specific,butareinsteadpartofthecogni-
21
tive ecology that humor occupies as one phenomenon among many. In
22
this vein, we conclude with some remarks on the overlap of these ideas
23
with the General Theoryof Verbal Humor (GTVH).
24
25
26
2. Relatedwork on adversarialhumor
27
28
The cognitive strategy of humorous trumping, and the concept of adver-
29
sarial humor in general, is in keeping with a number of insights from
30
previous humor research. Although exploring the common ground with
31
existing research would hopelessly go beyond the scope of the present
32
paper, some key notions need to be introduced that can be used as a
33
stepping-stone.2
34
Themultitudeofhumortheoriesdevelopedinmorethantwomillennia
35
of philosophical considerations on laughter and humor can generally be
36
dividedintothreemajorfamilies:thecognitivetheoriesthatgiveacentral
37
role to incongruity and its resolution (e.g. Kant, Schopenhauer, Koes-
38
tler, Suls); social theories that highlight the importance of aggression,
39
(AutoPDFV77/6/0608:20) WDG(148(cid:1)225mm)TimesM J-1537Humor,19:3 PMU:H(A1)24/05/2006 pp.305–338 1537_19-3_05 (p.309)
310 T. Veale etal.
disparagementandtheconfirmationofsuperiorityinhumor(e.g.,Hobbes,
1
Bergson, Gruner); and the psychoanalytical tension-release models in-
2
spired by their most well-known proponent, Freud.3 Although the influ-
3
ential linguistic humor theories developed byRaskin(the Semantic Script
4
Theory of Humor, or SSTH 1985) and Attardo (the General Theory of
5
Verbal Humor, or GTVH 1994, 1997, 2001a) do not fully belong to any
6
one of the three families, they do demonstrate a very clear a‰nity with
7
the family of incongruity-resolution theories. In essence, both the SSTH
8
and the GTVH are wedded to the central notion of semantic opposition,
9
as expressed by an overlap and subsequent shift from one semanticscript
10
to another, which presupposes an incongruity from the perspective of the
11
joke recipient.
12
Onecancertainlyarguewhetherthisperceivedoppositionisacauseor
13
merely an e¤ect in generating the humorous e¤ect. In either case how-
14
ever, when it comes to adversarial humor, social factors also seem to
15
play a central motivating role in the generation of the humorous e¤ect.
16
In a multi-agent setting with opponents competing in a game of verbal
17
thrusts and parries, wittiness becomes the symbol of intellectual and so-
18
cial superiority. Within this competitive view of wit as verbal fencing or
19
jousting, Gruner (1997) develops a game-theoretic account of humor in
20
which participants can be winners (those who laugh) or losers (those
21
who are laughed at). It is argued that even wordplay and punning, often
22
treated as ‘‘non-tendentious’’ (Freud 1905) or neutral cases of humor, fit
23
naturally into the formula of ‘‘laughing is winning’’. Despite the interest-
24
ingbasichypothesis,Grunerdoesnotprovideananalyticaltool,letalone
25
alinguistic one, for afine-grained description.
26
The literature on linguistic humor o¤ers slim pickings to those seeking
27
analytical approaches to the interactional, game-like aspects of humor
28
production(acriticismleveledbyKottho¤1998andNorrick1993among
29
others), not to mention the specific case of agents in a creative verbal
30
competition.Butdespitetheseriouslackoffoundationstobuildon,there
31
are some interesting concepts that crosscut the phenomena we are focus-
32
ing on. Davies (1984: 362) in her analysis of conversational joking styles,
33
refers to the thematic principle of contradiction, which is central in cases
34
of competitive humor types (e.g., ritual insulting). In one cited example,
35
participants jointly improvise and attack each other within the same joke
36
frame. Typical of this adversarial game is the expression of contradiction
37
through ‘‘parallelism across turns’’ (1984: 362), a feature that is essential
38
to the trumping strategy that will be discussed in the present paper.
39
(AutoPDFV77/6/0608:20) WDG(148(cid:1)225mm)TimesM J-1537Humor,19:3 PMU:H(A1)24/05/2006 pp.305–338 1537_19-3_05 (p.310)
Cognitive mechanismsof adversarialhumor 311
Basically,humorousinsultsconstituteakindofteasing,which,according
1
to Drew (1987: 233) can be discriminated from other interactional strat-
2
egies by three criteria: ‘‘(i) the teases are not topic-initial utterances, (ii)
3
they are all in some way a second, or a next, or a response to a prior
4
turn, almost always the adjacent prior turn, and (iii) that prior turn is
5
spoken by the person who is subsequently teased, in multiparty as well
6
as two-party talk’’. Most important in the context of the present paper is
7
the stipulation that teases are always a response to a prior utterance in a
8
given context, which squarely places teasing in the family of adversarial
9
humor types.4 Although Drew mentions the possibility of ‘‘going along
10
with the tease’’, meaning that a second speaker, the victim of the tease,
11
can continue on a line introduced by the teaser, his data do not permit a
12
rigorous analysis of this type of reaction. Rather, in the majority of his
13
examples, the recipient of the tease responds in a serious manner. In
14
contrast, the trumping language game requires the active humorous en-
15
gagement of the recipient, remaining within the same linguistic space of
16
the instigator and assiduously avoiding what Drew terms ‘‘po-faced re-
17
ceipts’’. By elaborating upon the same discourse space, the respondent
18
not only parries but fully repudiates the original thrust, showing that
19
even if one were to fully accept the instigator’s premises, his conclusions
20
arewithout merit.
21
Venturingbeyondtherealmofteasing,Curco´ (1998)o¤ersarelevance-
22
theoretic account that extends the view of humor as a response mecha-
23
nism, proposing that all types of intentional humor be seen as indirect
24
echoes. Curco´ (1998: 305f) argues that ‘‘a great amount of intentional
25
humor, if not all, consists to a large extent in implicitly making a specific
26
type of dissociative comment about a certain aspect of the world, or an
27
attributable thought. [...] [S]peakers lead hearers to entertain mental
28
representations that are attributable to someone other than the speaker
29
at the time of the current utterance, while simultaneously expressing to-
30
wards such representations an attitude of dissociation’’. In other words,
31
if a speaker is confronted with two radically contradicting assumptions
32
at a time (incongruity), the search for relevance leads one to inferentially
33
attribute one of the assumptions to another agent, and therefore disso-
34
ciate oneself from this assumption. In e¤ect, one can say that Curco´ con-
35
siders all humor to be adversarial, since opposing assumptions are attrib-
36
utedto di¤erentagentsin away thatcausesone tobecomethe subject of
37
criticism. For Curco´, incongruity is not a defining feature of humor, but
38
simply a means of invoking the additional processing e¤ort one needs to
39
(AutoPDFV77/6/0608:20) WDG(148(cid:1)225mm)TimesM J-1537Humor,19:3 PMU:H(A1)24/05/2006 pp.305–338 1537_19-3_05 (p.311)
312 T. Veale etal.
look beyond the purely propositional content of an utterance. Presum-
1
ably,anagentthatisattunedtothehumorouspotentialofwordsandcir-
2
cumstances does not need such prompting, making such agents the most
3
skilled usersof trumping.
4
Thisnecessarilyconciseandhighlyselectivesketchofpreviousworkon
5
the linguistics of adversarial humor serves as a point of departure from
6
which to study the specifics of the trumping game. We argue, however,
7
thattheinsightsderivedfromouranalyseshaveimplicationsforthestudy
8
of other types of adversarial (and even more generally, interpersonal)
9
humor as well.
10
11
12
3. Atypology of trumpingstrategies
13
14
Themetaphorofverbalfencing,commonenoughinadversarialcontexts,
15
goes some way towards capturing the productivity of trumping as a hu-
16
morous strategy. In essence, trumping occurs when an initial utterance
17
U by an agent S (the instigating speaker) evokes a counter-utterance U0
18
from a second agent H (the responding hearer), where U0 undermines U
19
(and thus S) not by mere contradiction or non-acceptance, but by reveal-
20
ing U to be fundamentally unsuited to the communication intent of S.
21
The following isa schematicview of thestrategy:
22
23 S OpenswithanutteranceUcontaining aspecific ideaXwhere Userves
24 a communicative goal G
25 (e.g., G¼self-praise, insult,persuasion,consolation, etc.)
26
H Respondswithan utterance U’containinganideaX’thatisparallel
27
to X so that U’ serves a competing or contrary communication goal
28
sG
29
U’subverts Uand H trumps S to the extent that X’is apropos to X
30
U0 must significantly parallel the speaker’s utterance U in some key as-
31
pect,whetherphonetic,lexical,structuralorconceptual,toachievetheef-
32
fect of neutralizing U using S’s own language choices. We shall consider
33
whatitmeansforoneutterancetoparallelanotherinmorepreciseterms,
34
but for now the key point is that without a substantial parallelism be-
35
tween U0 and U, H’s response does not subvert U but is at best a mere
36
refutation of U. Trumping is a form of impromptu wit whose humor
37
arises,atleastinpart,fromourappreciationofanagent’sverbalmastery
38
insubverting the languageof an adversary.
39
(AutoPDFV77/6/0608:20) WDG(148(cid:1)225mm)TimesM J-1537Humor,19:3 PMU:H(A1)24/05/2006 pp.305–338 1537_19-3_05 (p.312)
Cognitive mechanismsof adversarialhumor 313
Parallelism is the signature character of trumping as a humor strategy.
1
Adversarial exchanges are not always humorous, nor are they always in-
2
stances of trumping, even if they are humorous or linguistically creative.
3
For instance, the following is not an instance of trumping:
4
5
(1) S: (consoling) Every cloud has asilver lining.
6
H: (angry) What do you know,you idiot!?!
7
8 H is here clearly adversarial to S, yet (1) is not an instance of trumping
9 because H merely rebu¤s S without using the form or content of U
10 againstS.Merecontradictionordisagreementdoesnotconstitutetrump-
11 ing. In contrast, parallelism can support trumping even when the initial
12 utterance U is neither adversarial nor overtly provocative. Consider the
13 following exchange, usually attributed toMr.and Mrs. Winston Church-
14 illafter an election defeat in1945:
15
(2) S: (consoling tone) Perhaps it isa blessing in disguise, dear.
16
H: (angrytone) Well, it must be a bloody good disguisethen.
17
Whereas (1) can most generously be described as petulant rebuttal, (2)
18
rises to the level of humor precisely because H manages to use S’s figure
19
of speech—a stock metaphor often used as a hollow platitude—as a
20
weapon against the well-meaning but ultimately unhelpful S. Indeed, it is
21
possible to view H’s reply as a highly compressed reduction ad absurdum
22
for S’s argument: H begins by accepting S’s proposition (signaled by
23
‘‘Well’’),whichleadshimtoexpresshiscontraryopinion(thatthereexists
24
no positive perspective on the situation) in terms of S’s metaphor (any
25
suchpositiveperspectiveissowellhiddenastobevirtuallynon-existent),
26
thereby reducing S’s utterance to absurdity. It is this ability of H to sub-
27
vert the particular metaphor employed by S, that makes H’s response so
28
witty.Thehumorof(2)arisesoutofacombinationofthisratherpointed
29
cleverness and the social dynamic of one agentdefeating another.
30
The seemingly innocuous use of the word ‘‘well’’ in (2) is actually of
31
some significance when considering the mechanics of humorous trump-
32
ing. One of the most socially cohesive aspects of trumping is that it pre-
33
supposes acceptance on the part of the hearer (H) of the premises on
34
which S bases the instigating attack. Tacit in this acceptance is the idea
35
that the hearer can defeat the speaker even using the linguistic weapons
36
(logical premises) the speaker himself has chosen. The e¤ect is to not
37
only counter the speaker’s conversational gambit, but to additionally
38
show that the speaker must surely be confused and inept, or at least
39
(AutoPDFV77/6/0608:20) WDG(148(cid:1)225mm)TimesM J-1537Humor,19:3 PMU:H(A1)24/05/2006 pp.305–338 1537_19-3_05 (p.313)
314 T. Veale etal.
under-informed, to work from such premises. Indeed, the hearer’s accep-
1
tanceneednotbesotacit,butmaybesoboldastohighlightthespeaker’s
2
own prejudices. For instance, consider (3):
3
4 (3) Sheri¤of Nottingham (S): You speak treason!
5 RobinHood (H): Fluently!
6
The riposte in (3) demonstrates not only an acceptance of the speaker’s
7
accusation,butaprouda‰rmationofit.Thee¤ectistwofold:first,pride
8
o¤ers an ‘‘o¤ensive defense’’ that blunts the speaker’s insult, since insults
9
are most often targeted at characteristics that one should find shameful;
10
second, the speaker’s implicit belief that treason is wrong (as would be
11
conveyed by a shocked tone of voice) is seriously questioned, which sug-
12
gests that the situation is more complex or subtle than the speaker can
13
comprehend. In this case, the suggestion is that when faced with corrupt
14
governance, patriotism and treason must be one and the same thing.
15
Parallelism is the key to the humorous e¤ect in these examples. Note
16
how in (2) the speaker utterance U and the hearer utterance U0 are con-
17
nected via the use of the word ‘‘disguise’’, while in (3) a connection is
18
forged via the adverbial ellipsis of ‘‘fluently’’ to ‘‘speak’’. Parallelism is
19
notasubstituteforincongruityintrumping,butrathertheframingdevice
20
through which incongruity can be focused and appreciated. In particular,
21
itisby aligningthecontentofdi¤erentutterances thattrumping achieves
22
its subversive goal, allowing the hearer to expropriate the speaker’s own
23
words and ideas and mould them to a contrary communicative goal.
24
However, parallelism is not an inherently structural operation, and is not
25
limited to the repetition of key words or the aping of syntactic form. The
26
diversity of forms that utterance parallelism can take provides the most
27
useful basis for organizing a typology of trumpingstrategies.
28
Weconsiderfirstthesimplestclassoftrumping,thosethatexploitaho-
29
mophonous parallelism between utterances. This general strategy relies on
30
the hearer being able to assign a di¤erent lexical interpretation to one or
31
more of the phonological chunks in the speaker’s initial utterance. In
32
other words, homophonous trumping relies on the speaker being able to
33
make a pun using the phonological material contributed by the speaker.
34
The strategy isillustrated by the following example:
35
(4) Chamberlain (S): I believe that Herr Hitler genuinely wants
36
peace.
37
Churchill (H): Yes, a piece of Poland, a piece of France, a
38
piece of...
39
(AutoPDFV77/6/0608:20) WDG(148(cid:1)225mm)TimesM J-1537Humor,19:3 PMU:H(A1)24/05/2006 pp.305–338 1537_19-3_05 (p.314)