Mixing Metaphors

Mixing Metaphors

Mark G. Lee and John A. Barnden

School of Computer Science,

University of Birmingham

Birmingham, B15 2TT

United Kingdom

mgl@

jab@

Abstract

persuggeststhatsuch

mixingisamorecomplexphenomenonthanstraightmetaphors,the

perprovidesananalysisofboth

parallelandserialmixedmetaphorswithintheframeworkofanAIsystemwhichisalreadycapableof

reasoningaboutstraightmetaphoricalmanifestationsandarguesthattheprocessesunderlyingmixing

ore,anytheoryofmetaphorsmustbeabletoaccountfor

mixing.

Introduction

Thephenomenonofmixedmetaphorshasbeenlargelyneglectedbypreviousresearchinmetaphorunderstanding.

,mixedmetaphorsareoftenregardedasexamplesof(atworst)

pathologicallanguageuseor(atbest)ly,itisclearthattheunderstandingofamixedmetaphoris

more difficult that of a single metaphor, since a mix requires reasoning about several vehicle domains.

Inthispaper,wewishtoarguethattheformerassumptioniswrong:mixedmetaphorsarecommoninmundane

everydaydiscourseation,thesec-

ondassumptionisdetrimentaltolongtermprogresssincemixedmetaphoricalmanifestationsrelyonstraightmeta-

ecifically,thispapermakesthefollowingclaim:thereasoningprocessesanddatastructuresinvolved

inunderstandingmixedmore,cur-

rentresearchonmetaphorprocessingshouldbecapableofbeingextendedtodealwithmixedphenomenaandmixing

can provide valuable insight into the processes underlying straight metaphors.

Tothisend,thispaperdescribessomeinitialworkdonewithATT-Meta[Barnden,1997]tohandlevarioustypes

ofmixingandreprisesanearlierclaimfortheneedforwithin-vehiclereasoningandtheuseofconversionrulestofil-

ter the relevant connotations of a particular metaphor.

Mierthefollowingpatholog-

ical sentence, quoted by Fowler [Fowler, 1908]:

1.“This, as you know, was a burning question; and its unseasonable introduction threw a

chill on the spirits of all our party.”

Inexample1,thequestionismetaphorically“hot”.However,itsintroductionmakestheparty’sspirits“cold”.Despite

thiscontradiction,thesentencecanbeunderstoodtomeanthatthequestionwassomehowcontroversialanditsinap-

propriate introduction saddened the emotions of the party members. Fowler criticised such examples as poor style.

However,despitetheconflictbetween“hot”questionsand“cold”emotions,theconnotationofthesentencecanbe

“DIFFICULTQUESTIONSAREHOT

OBJECTS”and“SADEMOTIONSARECOLDOBJECTS”.Furthermore,itisunlikelythatmostnativespeakers

wouldevenconsiderthedisparityof“hot”questionscausing“cold”because,ineachcasewhatis

mapped is not an instance of temperature change, buta connotation with direct relevance to the tenor domain.

Inthispaper,wewillarguethatitisoftennecessarytodoextendedreasoningpriortomappingfromvehicleto

ore,acapacityforwithin-vehicle-reasoningisessentialandanyconversionmustalsoactasastrictfilter

to limit the range of metaphorical meaning.

Thepaperisstructuredasfollows:inSection2,wewilloutlineanddistinguishtwokeytypesofmixedmetaphor:

ion3,wewillbrieflyoutlineATT-Metaandprovideananalysisofeachtypeofmixedmet-

aphorwhichourprogramiscapableofdealingwithandtheninSection4extendthediscussiontoothertypesof

mixes and the wider issues facing mixed metaphor research.

2. Mixed metaphor distinctions

Itispossibletodistinguishtwotypesofmixedmetaphor:allelmixedmeta-

phor,thetenor(A)isseenasavehicle(B)andthenasasecondvehicle(B’).Inaserialmixedmetaphor,thetenor(A)

isseenasavehicle(B)whichisthenseenasadifferentvehicle(C).Thekeydistinctionisthatinparallelmixesthe

twometaphorsdonotmetaphoricallyinteractandinserialmixes,mple,considerthefollowingtwo

mixed metaphors:

2. “The critique shed light on the theory’s shakey foundations.”

3. “One part of John hotly contested the verdict.”

Theutterancein2manifeststwometaphors:“MENTALINTERACTIONASVISION”and“THEORIESAS

BUILDINGS”[Lakoff&Johnson,1980;Grady,1997].However,whatisnovelisbothmetaphorsarebeingapplied

in the same sentence. Following the definition given above for parallel mixes, the following domains are involved:

A: Domain of theories, ideas, arguments etc.

B: Domain of light/perception.

B’: Domain of buildings.

Ifideasarelightsourcesthenacritiquecanshedlightonatheorysothatobserverscan“see”

thisinstance,the“lightsource”unda-

tionsinabuildingsuggestthatthebuildingitselfmightcollapse,therefore,iftheoriesarebuildingsthentheirweak

foundationsmaycausetheentiretheorytocollapse,orliterally,vesentencecanbeunravelledby

treatingthedifferentmetaphoricalvehic

the critique and the theory itself.

Itisworthnotingthatthemetaphor“MENTALINTERACTIONASVISION”mightitselfbeacombinationof

twofamiliarmetaphors:“IDEASARELIGHTSOURCES”[Lakoff&Johnson,1980]and“UNDERSTANDINGAS

SEEING”. We will return to this point in Section 4.

Theutteranceinexample3alsomanifeststwofamiliarmetaphors:“MINDPARTSASPERSONS”[Barn-

den,1997](seealso,Lakoff’smetaphor“IDEASAREENTITIES”[Lakoff,1993])and“ANGERISHEAT”[Lakoff

&Turner,1980].Inthe“MINDPARTSASPERSONS”metaphor,themindiscomposedofdifferentpartswhichmay

havedifferentbeliefs,ningthat“onepart”ofJohncontestedtheverdictsuggests

thatthereexistsmer,thepartof

Johnreferredtocontestedtheverdict“hotly”.Inthe“ANGERASHEAT”metaphor,ore,

the“part”ingthedefinitionsgivenabove,thefollowing

domains are involved:

A: John’s mental/emotional states.

B: Domain of people/NL communication.

C: Domain of heat.

“ANGERASHEAT”metaphor(BasC)actsonthe“MINDPARTSAS

PERSONS”metaphor(AasB)ore,itisnotpossibletoisolatethe

animportantpoint:itcouldbearguedthatthe“ANGERASHEAT”metaphor

appliesdirectlytothechedefini-

tionsabove,r,viewingtheutteranceasaparallelmixmisses

tenceisambiguous:eitheronepartofJohniscontestingthe

verdictandonepartisnot,andthepartcontestingisdoingso“hotly”orbothpartsofJohnarecontestingtheverdict

butonlyoneisdoingso“hotly”.Ourintuitionssuggestthattheformerinterpretationisthedefaultoneandwewill

r,ourtreatmentissensitivetosuchdistinctions(asis

our computational implementation) and is capable of reasoning about such uncertainties.

e2hastwometaphorswhichactasparallelvehicleson

the(literal)r,inexample3,onemetaphor(BasC)actsdirectlyonanother(AasB).Thismet-

tsection,acomputationalaccountdealingwithbothkindsofmixedmeta-