How Words Mean Mean
••I
How How Words Words Mean Mean -
For Edith Edith For Thank you you for for my my English Eng...
106 downloads
2573 Views
123MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
How Words Mean Mean
••I
How How Words Words Mean Mean -
For Edith Edith For Thank you you for for my my English English words. words. Thank
Lexical Lexical concepts, concepts, cognitive cognitive models, models, and meaning meaning construction construction
VYVYAN EVANS EVAN
OXFORD O XFORD UNIVERSITY L'NIVt "'\11 V PREss I'RI \,
OXFORD OXFORD
Contents Contents
UNIVERSITY NIVI:;RSITV PRESS PRE'S MU fu." 61)r GreatCbmWon tiarendon Street, Oxford 01.1 (JI'~.t !"Ilrm, Odurd
Oxford tnlYnllly University Pr Presaliis..ackpanlMnl department of of ttl< the University Oxford l'nI"",,,'ly of of Oxford. Oxford. It further. the University's objective of excellence in research, II furthtn tM llmvtrllty', ob,atl pf n...dlnkt m rtW.mh, scholarship. kh...,bnlll J'. and education by publishing worldwide in .nd «Ilk.1ll"n by ruhh~lIni W\lri.Jwuk m Oxford New Oxford I'<W York \ork Auckland UJ't CapeTown Town Oar 1)ar nes ~m Salaam tHong Alk.kbnd I nl Kong Knns Karachi K.lr~hl Kualalumpur 1 umpur 1.adnd Madrid Melbourne City "Nainki Mub 1dhtlutnC' Mexico klk.O (.Jly m..tM New{)rlh. Delhi Shanghai 'lawn loronto Mw h.tncN1 T'I",", rvn.tnlO
Acknowledgement Acknowledgements p",fac Preface Li Listt of of Figures List I List of of Tab Tables Abbreviations Abbreviations
viii IX XIV XVI XVII
With oftktt offices In in With Republic IFiance ArgentinaAu Austria BrailllChile • iech RtruhL _ ArJtntlfu Irt.1 8rl.1I1 hlk c:u...h r,lI'''..:' Greece (.r('(t..t Guatemala Itunllry Hungary Italy (.lUlnn.a1.l lI.aly Japan J'l"n Poland PoLlnd Portugal I\.nupl Singapore 'm..Pllr.:' South Korea Switterland Thailand Turkey South Korc. ~'lzn1.lnd Tlu.u.nd lurlry Ukraine l'mu,," Vietnam VIt1rwm registered trade mark of Oxford University Oxford is a Oxfllrd li a rtploltrN I~ muk pf Oxfmd L'mwnlly Press Prna in 1M the UK and in other countries m UK .nd m certain t.:n1.ln (11M lounlnn Published In in 1M the United Publi.haJ UmlN States !"lU.I by Oxford Oxford University by Umvnslty Press PrftI Inc.. Inc.. New 'f:W York ....ork
Part Introduction Part II Introduction I.i. Words Word and meaning meamng 2. Toward a new Towards .1 new account account of of word word meaning 3. 3. Cognitiv~ Cognitive linguistics 4. in LCCM LC M Theory Th ry 4. Word Word meaning in
I
3 27
47 65
Vyvyan Evans VlO9 moo ,• • Vyvy n fun The moral TM mUr. Will Will you you choo choose Sometim($Sometimes— M u~ As the winds use A cmck wall crack in the wall Or a drain, drain. Their joy or their pain To throughTo whistle through— Choose me, me. You Engli h words? word,? You English I know you: You 35 dreams, dream.., You are arc light light as 35 oak, oak. Tough as Preciou Precious as gold. gold, M com. As poppies and corn, Or an old cloak; cloak; weet aas our birds bird Sweet to the ear, ear. M burnet rose As the hurnet In the heat Midsummer: Of Midsummer: races tmnge as the races Strange Of dead and unborn: trange and sweet Strange Equally, Fqually. And familiar, familiar. lu eye. To the eye, As the dearest faces faces That know,. That a man knows, And home> are: are: And as lost homes But older far far But though older Than oldest old"'t yew. yew, M hll" are, arc. old, old. As our hills Worn new Worn Again and again: agam:
\treams Young aas our streams After rain: mm: And as a dear dear A eanh whiC. The The challenge challenge then, then. in III they can can shift contexts of accounting for meaning construction in in aa theory theoryof oflanguage languageunderstanding, understanding. "is to model the the nature that language language nature of the Iinguisti to be able to model linguistic knowledge that uver have access while being able able to to account account for for the theway way word word users mu mustt have access to. to, while ofuse. use. meanings shift hili in varying contexts of The received view in linguistics, Iingui tics. and and philosophy philosophyofoflanguage. ha attempted attempted language, has recondle this challenge by distinguishing d.~tinguishing between between two twokinds kindsof ofmeaning: meaning:aa to reconcile word. and a conconcontexl-independent. "timeless" meaning associated associated with words, context-independent, "timeless" text dependent meaning. meaning. That That is, is. words words contain tontain context-independent context'lI1dependentmeanmean text-dependent way> by by virtue virtue of ofthe the ings which whi h can he be interpreted interpreted in in context-dependent context·dependent ways ings application of of various variou principles principles of ofinterpretation, interpretation. e.g., e.g.•the theGricean Griceanmaxims. maxim. I'or ofthis th. "nen-Gricean" "neo-Gri can" perspective, per pective. see seeLevinson I.evinson For aa contemporary account of (2000). Thi (1000). per pective. which Recanati Recanati (2004) (1004)refers refers to to as literalism, literallSlI/. This general perspective, enshrines as axiomatic aa principled principled distinction di tinctionbetween betweensemantics scmanticsand andpragpragmati . The position that that II develop in this thi book, book. one one which which isi in inkeeping keeping with with matics. pages which which follow, follow. takes takesthe theview viewthat that much recent recent research research discussed discussed in the pages the distinction between pragmatics isi not not principled. principled.Rather, Rather.itit between semanti semantics and pragmatics is artificial. for the the inherent inherent variation variation in inword wordmeaning meaningisi My approach approach to accounting for sy.tem-thelinguistic linguistic principled separation separation between between the the linguistic Iingui ticsystem—the to posit aa principled the non-linguisnon Iinguis that words knowledge that words encode--and encode—and the conceptual system system—the tIC knowledge knowledge that that words word facilitate fa ilitate access ace to. This This distinction distinctionI Imodel modelininterms tern. tic the theoretical theoretical constructs con truct of of the the lexical concept concept and the the cognitive cognit.ve model. mode/. of the These two constructs construct are are central central to the the theory theory developed developed in in these thesepages. pages. These Hence. II refer refer to the the approach approach as as the the 71.eo'1 ofLexical LexicalConcepts Collcep"and allliCog Cogllit.ve Hence, Theory of ► itive Models Models (or LCCM LCeM Theory Theory for short). hort). Briefly, Brieny. aa lexical lexical concept i a bundle of of concept is knowledg~=ribcd in in detail detail in in Chapters Chapters 66 and and 7—which 7- which varying sorts of of knowledge—described in language. language. In In contrast, contrast.cognitive cognitivemodels model are specialized specialized for being encoded in ofcoherent coherentand andstructured tructured non-linguistic non-linguisticknowledge— knowledge-constitute a body body of constitute described d=nbed in detail detail ininChapters Chapter 99 and and 10. '0. Cognitive .ognitive models modeh consist consist of of "recorded" subjective states including in luding information information derived derived recorded" per«ptual perceptual and and subjective from sensory-motor sensory-motor perception, perception. proprioception, proprioception. and andintrospective introspectivestates, states. from including emotions, emotion. the the visceral visceral sense, vense. cognitive cngnitive states, tat .... and and so '>0 forth. forth. In In including perceptual and subjective .ubJectivestates, "ates. addition. to tobe beable Jbletotoproduce producerehearsals rehearsalsofofperceptual addition, form. the the perceptual perceptual symbols ymbols can can he be combined combinedproviding providing albeit in attenuated form, (onccptualiz.ltion\. The The re-enactments rt·cn3ltmcnts ofofperceptual pcn:cptuJIand andsubjective SUh;Cdivc novel conceptualizations. states Slales and and the the novel novel conceptualizations coru;cptualil.llions are are referred refcrrctl to as J~ simulations. 5;IIIlIlnt'O"5. 4
XII Xii
-------------------------
PREFACE PREFACE
lienee. cognitive cognitive models modelsprovide provideaa level level of ofnon-linguistic non · linguisticknowledge knowledgewhich whichisis Hence, specialized for for being being accessed accessed via via lexical lexical concepts. concepts. specialized LC M approach follows. Words Words encode encode aa core core content, content.the the The LCCM The approach works works as as follows. 1;lIgll/$';( lexical concept, (oncept. which which relates relatel. to tohighly highlyschematic schematICinformation: information: linguistic lexical (Olllml. This Thi represents represents the the core core information information associated associated with withaagiven givenword. word. content. ofnon-linguistic non"lingui ticcontent: content: In addition, addition. words word facilitate facilitate access access to to aa large large body body of In co/lccpruai content. cOII'errl. This Thi is i achieved achieved by by virtue virtue of ofaalexical lexkalconcept conceptfacilitating facilitating conceptual access to a body body of ofcognitive cognitive models, models. which which II refer refer to to as as aa word's word's semantic semantic access to a po,ell,wl. Not Not all all of ofthe the cognitive cognitivemodels model totowhich whicha aword wordfacilitates facilitatesaccess acces are are potential. in word word meaning meaningarises arise. actIvated in in any any given given utterance. utterance. Hence, lienee. the thevariability variability in activated from the the partial partial activation actIvation of of the the semantic semantic potential potential totowhich whichaaword wordfacilifacilifrom tates access. access. tates In presenting presenting LCCM LCCM Theory. the range range of of In Theory, II develop develop aa unified unified account account of of the phenomena presented (4) above. above. That That is, is. I Itreat treat the the phenomena presentedininexamples examples(I) (i) to to (4) phenomena above. distinct. as a being being continuous continuous and and hence hence being being phenomena above, while while distinct, explainable term of common set set of ofrepresentational representational and and compositional compositional explainable in in terms of aa common mechanisms. does not not mean, mean.however, however. that that IIprovide provideidentical identicalexplanexplanmechanisms. This does ations for each each of of the the phenomena phenomena IIaddress, address. as as we we shall hall see. sec. ations for Finally. to develop develop aa cognitive cognitivelinguistics linguistiCS Finally, LCCM LCCMTheory Theoryisis an an attempt attempt to of lexical lexical representation representation and meaning construction. construction. One Oneimpulse impulseinin account account of and meaning ofmeaning meaningconstruction construction cognitive lingui tics has been to to develop develop accounts accounts of cognitive linguistics has been which priVIlege non"lingui tic processes. processes. This i true true both both ofofConceptual Conceptual which privilege non-linguistic This is Metaphor Theory (Lakoff (Lakoff and Johnson 198o, 1980. 1999) 1999) and Conceptual Conceptual Blending Blending and Johnson for example. example. Indeed, Indeed. these these approaches approaches Theory (Fauconnier and and Turner Turner 2002). Theory (Fauconnier 2001), for in the the present present work. work. Any Any linguistically linguistically centred centred account account of of remain remain important important in language uch as as LCCM LCCM Theory, Theory. must must interface interface with with these, these.as as language understanding. understanding, such is to to integrate rntegrate and and discussed Neverthel • my my main main concern concern is discussed later later in in the the book. book. Nevertheless, many of of the the important important advances advances in in terms term of ofresearch research on on linguistic linguistic build build on on many semantics linguistics. and and to to incorporate incorporate semantics and and grammar grammar evident evident in in cognitive cognitive linguistics, these recent advances advances in in philosophy philosophy and and cognitive cognitive psychology, psychology. which which these with with recent have fresh impetus impetu for for an an "empiricist" "empiricist" approach approach to toknowledge knowledge have provided provided fresh .• Barsalou 1999. 2008; Prinz 2002; Gallagher representation also Gallagher 2002; sec sec also representation (e.g (e.g., 2008; Prinz Barsalou 1999, 2006; 2006; Johnson Johnson 2007). 1007)•
Who the book book for? for? Who iis the The with aa number number of of different ditTerent readers readers in rn mind. mrnd. This ThIs The book book has has been been written written with number of of challenges, challenges. in in terms terms of ofcoverage, coverage. inevitably inevitablybrings bringswith with itit aa number accessibility. linguist Firstly,the thebook hook attemp" attempts to cater for general linguists on. rirstly. accessibility, and and so so on. who linguistic;. II have ha"e attempted attempted to to situate situate who may may not not be be familiar familiar with with cognitive cognitive linguistics. LCCM lexical semanlics and compositionality, compositionality. in in LCCMTheory. Theory,as asan an account account of of lexical semantics and umptions and and approaches approaches of cognitive linguistics. linguistics. II have have term termsof of the the core core aassumptions of cognitive included many of of the the background ba. For For the study of language—far most people, is intuitively poopl•• meaning mtaninglS inluluvtlythe lhocentral «nlral issue i"".inin tho Iudy oflanguag f.r more mo'" important import.nl than th.n understanding und....l.ndingdetails dentence propo ition. isi truth literalism, sentence meaning. meaning, t",hnically technically known known aas a proposition, evaluabl although thi i ue isi potentially potentiallyproblematic. problematic.>2 That That is, is. aa sennevaluable—although this issue formed gral11matical word - i held held toto"carry" "carry" aa tenl a well tence—a well-formed grammatical string string of words—is ofaa meanmg which patterns patterns after after reference: reference: th onv nllonal assignment ignment of meaning which the conventional of affairs affair to the the complex complexlinguistic lingui ticexpression e pr ionresultr ull worldly entity tate of entity and state ing ilion of ofthe the individual mdlvldual elements el ment in informing formingaasentence. ~nten e.The The ing from from compo composition that is, is. the the meaning aassociated sociated with ntence constitutes onstitutes the proposition, propo it Ion. that with the sentence ~nten e meaning. the following following example example sentence: ntence: sentence meaning. Thus. Thus, in the J
A numbn' lh.! Pragmatic PnpNtk. tradition InJlltOn (e.g.. Cr .... Bach 8.kh 1997:(:orston 1'#970 ( nt(ln iuoz; JOOl;Recanati R«.ln.JU number of of khubn scholars WOrlonlln working in the
lUI"") tu\'t ot'twnni " often (,hm (or ft" usuallyi u~,.) the 1Mcase \:.utthat thaithe Itwlinguistic Imgul heform (nrmuttered uuntdby byan o1n zoo4) have observed tlul that IIit is tntnt,."Ulnr III " meaning. mc.mma.. Thai "Th.It is. I utterances unt'l'alkn are aIT often vitro not n..tpropositional. rn'P' Ih.. ruJ. interlekutor PolnJrnktnml underdetermines lhtthe sentence but ttmW'dpragmatic J'RPMIK intrusion. ",IN"",".such WI. hthat lh.,inferential mt.'ftntwpro. rn"nift but h.aw have Id to bt hewmrAttnJ completed br"'NI by whathol, has~ ten termed esses ;t~ In render rtndcr the tM utterance ullft'lIkC propositional rrop,lISJlk)f\Mand andhence btrk.ctruth trulhI.-valuable. ~nlwtW :Arston. cnl..!"include IIldud an anarray arrayofof
II
dlllerent 'iOrt. of of physical phy;i al entities ntlti and and abstract ab tract events—which event - which is is related related to to difkrent sorts ~Jrlc'. "background':' Understand 109 the and (6) (6) Searle'snotion notion of of "background".4 Understanding the exampl examples in in (s) and (lIves complex detailcd knowledge knowlcdg about about the the sorts son of ofscenarios enario that that In' involves complex and and detailed 1'/,nlnttWl1l'\.1with ""lIhrespet:t r I t to Il)"'~ ""hKh,inmI large) 11.I'Jt)part, J'lr1, dciermines tWtC'rml , ... the linguistic hnctuSJ ... meaning. nwanmJ.
1Z
I NT RODUCTION
WORDS AND MEANING
referring to as encyclopaedic in nature. Moreover, the meaning of open only ever appears in given contexts of use, even when these are the minimal contexts of use deployed by the linguist: a numbered "linguistic example" set off and embedded in the running text of technical articles published in academic journals.s In other words, word meaning emerges from a large semantic potential which is narrowed by the sentential (and extra-linguistic) context in which it is embedded. As such, word meaning appears to be guided by and a function of context: words, I suggest, do not mean independently of context. Thus, the fundamental problem with literalism is that it attempts to artificially divorce (word) meaning from (situated meaning in) context of use. More precisely, literalism lives in something of a fool's paradise. It holds that language users retain an idealized, timeless meaning for open which they neatly keep apart from the situated meanings of open which arise from its use in examples such as in (5) and (6). The mistake that literalism makes, then, is in being reductionist and simplistic about meaning.
An additional challenge: figurative language
' See similar arguments made by scholar. including t lark wit); CA ► ulson /00 ► ); brit% . IvIer anti Evans t1ou3). fauctmnici ( "or ) ; "nfokier 119871; Swectser 11t>90;
The challenge, then, that awaits an account of lexical representation and the ro le of words in meaning construction is to work out the difference, if any, be tween the role and function of literal and figurative word use in meaningc onstruction processes. To illustrate the nature of the challenge, let's consider the following example: (7)
2ootil;
John's boss is a pussycat
Presumably this utterance doesn't mean that John's boss is a pussycat, in the sense of a four-legged organism, with a tail and pointy ears that utters "miaow." Rather, the meanings associated with the phrases John's boss and pussycat have to be integrated with the predicate nominative construction, which ordinarily carries a class-inclusion meaning. 6 Informally, this construction has the following syntax: "SUBJECT is an NP," and means, again informally: "The subject is a type of the entity specified." To illustrate, consider the following: (8)
As we have just seen in our discussion of open, the protean nature of word meaning relates both to literal and figurative uses. A challenge for any theory of lexical representation—which is to say, the mental representations associated with words, consonant with the protean nature of word meaning discussed in this chapter—is to provide an account of literal and figurative language. Under literalism, these are treated as radically different sorts of language. It is often assumed, from this perspective, that figurative language involves the "defective" use of literal language, as argued, for instance, by Searle ((19791 1993). On this view, the use of figurative language arises from the context-dependent interpretation of literal language, and thus involves principles of pragmatic inference being applied once the context-independent sentence meaning has been derived. Put another way, figurative language is a function of language use, and thus falls under the purview of pragmatics, rather than semantics proper. The difficulty for what we might refer to as the literalism perspective on [figurative language, is as follows. This perspective predicts that understanding a literal sentence should be faster than understanding a figurative expression: we must first understand what the sentence means before we can interpret what the speaker intends us to infer by using the sentence in a non-literal way. However, as has been shown, based on investigations of psycholinguistic processing, language users often appear to be equally as efficient in computing the meaning of figurative language utterances as they are non-figurative ones (Gibbs 1994; Glucksberg 2001, 2003; See also Giora 1997, 2003).
13
John's boss is a pianist
The meaning that a language user would ordinarily derive, for an example such as this, would be that John's boss is included in the category of those who play the piano and thus constitutes a pianist, and that this situation persists through time. But, the same construction does not provide a class-inclusion reading for the previous example in (7). The challenge then, for our account of the variation in word meaning, is to be able to provide an explanation as to why (7) means something other than what it literally says, while (8) means what it does literally appear to say.
The nature of context I suggested above that the fundamental problem with literalism is that it attempts to artificially divorce meaning from context of use. Before proceeding with an attempt to identify the ingredients of a theory of word meaning and meaning construction, we must first get an initial sense of the different sorts of context which serve to narrow the meaning of a word. Accordingly, we will begin to see that the notion of context is a complex and multifaceted phenomenon crucial for language use and language understanding. Accordingly, the notion of context is fundamental to the development of LCCM Theory that I begin to sketch in the next chapter, and develop in detail in the " The nominative predicative construction involves the copular or "linking" verb fie which com• pine. with a nominal. e.g.. "a pianist." The nominal fu n ctions as the mentirl part of the Predicate: is a pianist." 1.ang.s. ker twou in his a tt.th.sts of the n ► nuttateve pretts‘ ate lion argues that he entoties the -cunt tnudtt ton through tune of a stable situation chma t ierucci only as a st.itive relation" (find. 65).
1-1 '4
INTRODUCTION INTRODUCTIO.
re'>I Ih btx)k. book.'7 Hence, Ilen«. the the account de ount of ofword word meaning m ningprovided pmvidw isi diametdiamel rest of of the rically opposed to that offered by literalism. rically oppoS«! 10 Ihal oU rw by lileralism. As Ihe the approa approach take iis uusage-based, the term lem. utterance, utteran e. rather ralher than Ihan A h I lake ge' ba5Cd. I uuse Ihe sentence, in discussing word meaning. This reflects my assumption t hat meaning. Thi refllxl my umption thaI itil isi nlenee. in di u ing taking aaccount language in counl of oflanguage in use u that Ihal we we can can hope ho~to 10fully fullyunderstand under land only by laking the ndlure nature of of word word meanong. meaning. It also follows follow from the Ihe position po ilionthat Ihalsentences, nlenc •asa Ihe understood theoretical Iheory.are ar artificial artlfidallheor lilalconstructs, (on Irllct abstracted • ab Irallw under lood in in linguistic lingui lictheory, from actual actual usag usage ev~ events, from 1 • which whi h isi to 10say, say. utterances. utterane . IIwill will have have more moreto10sayy about the distinction aboullhe di IInClion between belw nsentences nlen and andutterances utteranc ininChapter .hapler 44.
Utterance context context above, the open above. France and Of>t:1I As we saw with the A w w,lh Ihe examples exampl relating relallng to 10 Frallcr Ihe utterance utterJIK elements which occur in 10 a a given utterance utterance contribute, conlribule. in in part, part. to 10determining delerminlOg the meaning Ihe word. word. That ThaI is, i and and as a suggested ugs IW abve, above.the Iheutterance utteran e Ihe mcanlOg of the provides aa wnle.1 context whilh which aassists in narrowing the meaning 1\1 on narrowlOg Ihe meaning of the Ihe word word in in pmvides question. qu lion. To 10 illustrate, illu lral • consider con ider the the following foUowingexamples: exampl (9) a.a.OnOn May 1st, my (9) ~by I mygrandfather grandfalh rexpired PIrW b. On b. On May May'1stI my driving licence licence expired expired expired in The m meaning meaning of rxplml in each a 'h example eumpl is I aa function function of ofthe Iheutterance utterance inin expired relates to an which it whICh il isi embedded. embWdw. In In the Ih first fir Iexample, example. rxp"ed relal 10 an event 'enl to expiry involving ddeath, while in the involving alh. whil Ihe second, second. expired! expirrd relates relal 10 p'ry of of the Ihe term I rm for whkh which an individual's to drive on the for individual' right righllo Ih public public highway highway was wa; sanctioned '>3nclionw or "licensed." "lie nS«!." Now ow consider con ider another anolher example ample involving involvingaaverb. verb. This Thi involves involv _the Ihefollowfollow. alternation associated ing well-known w II known context-dependent wnle I-dependenlaitemalion a '>tltotowhich whichwords word,are areput, put, from even when one a um a highly granular lexicon: one that po it a lars even when one assumes a highly granular lexicon: one that posits a large number of ofdistinct di,tinct senses. n .This Thi, follows follow,., thenumber numberof ofdistinct di. tinctword wordsenses '>Cnsc> number as the r;(Xiated WIth it !>C1cCn . Ilowever, we can continue finding novel uses o( fasl, for (or which which Indeed, we \\c could could produce produce aavirtually virtually infinite infinitelisting. Ii ting. Indeed, the the same ' is that Ihal all all the Ihe examples exampl we we have have What considered, and might wish to consider, upon close analysis predicate in "''''idcred, Jnd might wish to (onsider, upon close analy i predi ate in aa slightly different different way. way.In In oth other words, each each unique 'Iighlly r words, unique instance in tance has ha a3 distinct di tincl utterance context, and is associated with a slightly different semantic ullerance (ontexl, and i as;oc:ialed wilh 3 hghtly different semanticvalue. value. Thus, we can conclude from this that, in principle, every instance of use of aa 'r hu" we can condud from Ihi that, III prin iple, every in tanee o( usc o( word such fast ha has aa different "Sense "ord uch aas fllSl differenl meaning. meaning. To To take lake aa" n Enumerative" Enumerative" approach10 to word word meaning meaning would would be he 10 to sanction approach sanction an an infinite infinile proliferation proh(eration of of word senses stored in memory by language users. Such a position is psycho"nrd \Cn lored in memory by languas users. u(h a po ition i I' yeho. logically untenable. untenable. logically
Wordsas ascontextual contextual expressions expressions Words observationwilh with whICh which Ihi thisbook lx)ok proceed" proceeds, Ihen, then, is is Ihal that word; words .re are never never '1The he "h...,rvallon meaningfulondependent independentof of the the ulterance utterance in in which which they they are areembedded, embedded, and and meaningful theenC)'dopaedi encyclopaedicknmvledg knowledgeand and exIra extra-linguistic context which which guide guide how the linguisti conlext how words embedded in an utterance should be interpreted. Indeed, evidence word, embcdd d in an ullerance ,hould be IIlterprell-d. Indeed, eYldencc fromIhe thepcr perspectives of;oc:ial socialp;y In Ihi' this ....... section
Emboclled Em bodied cognition I he thesis Ihe,i, of embodied cognitio n is I at at the the heart heart of ofmuch muchresearch reseafch within wlthm cognition The "n,lIve linguistics linguistics ((Evan l004a; Evans Evan and Green rccn 1006; John n1987, 1987. 2007; 100]; Evans 2004a; zoos; Johnson c ognitive ~:,LtJ developmenl Lakoff1987; 1987;Tyler TylerJnd andhan Evans1003), 2003),and andha hasbeen been influential influential in developments 'ognitive psychology p ychology (in (inparticular parti ularBarsalou Barsalou1999; 1999; Barsalou Barsalou 01 al. 1993; in cognitive et af. 3993; ,n(,knbcrg l ded .. (B 1997).'3 This thesis—also thesis-also referred referred to as a grounded gro un cogmllo n(Barsaarsa · cognition (;lenberg 3997). III" andfirst fir Ideveloped developedinincognitive cognitivelinguistics Iing"i"i~bybyLakoff \.akolTand andJohnson Johnson lou "1R()OY' The ",n sense of of flYIIIg to the NA tiC MOTION' IOTI() ] i,is restricted r tricted to the operation operalion by an an entity enllty which which can (an be be NAMIC construed aas an an agent, construed agent. and, and, moreover, moreover, to 10 entities entitie. that that can canundergo undergoselfselfpropelled aerodynamic aerodynamic motion. motion. Further, the entity entity must propelled Furlher, the mU t be be able able to to accomaceom means of of transport. Thisexplains This explains why the agent and thereby serve aas aa mean modate theagenland modale aeroplanes and and hot hot air air balloons aeroplanes balloon are compatible with this this sense, sense, but but entities em ill unable to accommodate an unable an agent agenl are are not. not. This Thi isi illustrated iIIu trated by by example example(5). (s)·
,II.
?Hie was was flying (5) ??He (s) nying the the sparrow parrow across acro the the English Engli h Channel hannel Nevertheless, entities which which un can be construed tOnMrued as as being being guided, gu.ded, or oratalleast least 'evertheless, entitic. susceptible to being u eptible to being trained trained by by aa volitional volitionalagent, ag nt,which whichnevertheless neverthel cannot cannot accommodate an accommodat an agent, agent, are are partially parlially sanctioned nClioned by this thi lexical lexi al concept, concept, as a the the following example followmg exampleillustrates: illustrales: (6) He lie succeeded u((ceded in m flying nying the the homing homingpigeon pigeonacross aao the theEnglish Fngli hChannel hannel In the case of of [CONTROL (CONTROL OF OF LIGHTWEIGHT L1GHTWE.GHT ENTITY ENT.TY] as evidenced eviden ed by by the Use use of to be restricted sense of of flying of flying flrillg in (3), (3), this Ihi sen flyillg appears appear 10 r tricted to to entities entiti that that are ar capable of becoming airborne capable of airoorne by turbulence, turbulen e, and can he be controlled coni rolled by by an .n agent on the the ground. ground. This Thi lexical lexicalconcept con eptappears appearstotobe bespecialized peeiatizedfor forobjects objects such uch as as kites kiles and and model/remote-controlled modeUremol ·conlrolledaeroplanes. aeroplanes. The final final sense, ",n"" glossed glo~ as a 'SUSPENSION ]su Pl '0 OF O. LIGHTWEIGHT L1(iIlTWllGIIT OBIECTi, ()B'Ecr], selects Ie t~ for entities enlities that that can can he be supported upported by byvirtue virtueofofair airturbulence, turbulen e,but butremain remain "connected to" wellasashair hair to" the th ground. ground.This Thi lexical lexicalconcept conceptapplies appli totoflags nag asa well and scarves, which can "fly" in the wind. and arv . , which c.n "ny" in the wind. The third third finding finding concerns concern the the position po ilionthat Ihatgrammatical grammaticalcategories calegorieshave haveaa semantic basis. as basi. In In particular, parlicular, II am am here h reconcerned concernedwith wilhlexical Ie ieal classes da a semantic categories. Until relatively recently, particularly in the previously ",manlic categorie>, Until relatively recenlly, particularly m Ihe preViously dominant tradition ofofformal linguistics dominanllradilion formallingui ti associated a-'SOCialedwith wilhthe thework workofofGeneraGenera tive live Grammar, Grammar. itit was wa assumed ~umcd that thai lexical lexicalclasses, da ~.e.g., c.g ..nouns, noun.verbs, verb,adjectives, adjct:tivcs. etc., etc, were purely purely grammatical gramm.ukal categories, C~ltegOri • determined dclcrmllll"ti on un the the basis ba\i\ of ofdistridistrin bution—where bution ·where in III the the sentence nlen the Iheform formappears—and appears-andmorphology—i morphology. ·in
TOWARD WORD MEANING MEANINC TOWARDS AA NEW NEW A(:COUNT ACCOUNT Of' OF WORD
33 33
""lI,UI.".lhc for instance, in .'an(C, word-ending, word·ending,associated associaled with withaa the kind kind ofinnection, of inflection, for particular, Ihlrn 1 For .der the following followmg English Engll hsentence: nten e: form. example,,con consider . For exampl I hc IxlY boy kicked kicked Ihe the Ix,y, boys (7) The (7 11 as a aa noun. noun.In Incontrast, conlra t,the the ,subject u[l,cd or orobject object po position. it counts Ie I(al form innected with with the the past past tense ten marker marker -ttl can be inflected -ed and and cannot lexical form kick can ,'rr it ion. Thus, Thu , it.t counts counl as as an an instance instan eof ofaaverb. verb. appear in ,ubjecl subject or or object object po position. c'" in 'Cnlallons Ihose that thaI relate rclale 10 ncepto the human conceprepresentations encoded by language and those LC Theoryisisthat thaIthe therepresentations represenlations lual de\'e1opcd in in LCCM tual d III~ x CAUSES CAl'~f' y to John sneezed the X Y TO oh,«1 from from One one object tlit iIttlrr ft:17)tier off the MOVE Z 10V'1 I()('OItion tablt location to another OInother Agent CAU I Y Tel TO T/I~ dtme Iltel; I genl acts old on The anaesthetist X CAUSES BE OME Z rtllllatd lilt a patient au iog the causing rengiered BECOME Z palle", undergo patient them to undergo ,WWtlj( lOUS l.llt a d\01nge change of of state ta le OHSCJOUS y 111t "".up JIm' mlo Entity [ntllY moves mov to 10 The wasp flew into x MOVES MOVE. Y X Il,t room a ntw location new location the room /01", llLled at lilt Agtnl dirat\ an .n fohn kicked at the Agent directs llIaUT x DIRECTS rul lilt ion towards tow.ud action rat ACTION anolher entity tntity another AT Y Y Rf( tlVt RECEIVE
zz
10l-,t Itttcr letter hwe
The of language The usage-based usage based nature af -
\, we saw ".w in III the theprevious previnu, chapter, chapter, Grice Grice (e.g., (e.g., 1989) '989) made made aadistinction di~tlll tlon As and what gIVen ",nten mean and what its use use means mean, in in aagiven given b.:twecn what what aa given sentencee means between context: the distinction di tin tion between between sentence sent ncr meaning meaningand andspeaker sprakermeaning." meaning." context: In lingui'tic ,thi di\tin tionbetween betweensentence sentenceand andspeaker sprakermeanmean In modern linguistics, this distinction IIlg i embodied in in the the disciplinary di iplinary distinction di tinction between betweensemantics semanticsand andpragprag· ing is ofthe thestudy tudyofofmeaning: meaning:literalism. literali m. matic , and infomlS the the received received view view of matics, and informs I'mm the per I""'tive of literalism, lit rali m, semantics—the semantic the study study of ofsentence senten emeanmean · From the perspective ng- i, normally nomlally considered con idered to be beprimary, primary. and and prior priortotopragmatics—the pragmati - the mg—is tudyof peaker meaning. The supremacy suprema y of of semantics semantics isi enshrined en hrined ininformal forma l study of speaker tradition which which isi directly directly descended descended from from the thelogic-inspired log.c inspired '>emanti semantics,,the the tradition work of the th ideal ideal language languag philosophers. philosophers. point out out that in fact fact However, in However, in seminal seminal work, work, Ilerbert HerbertClark Clark(e.g., (e.g.,1996) Igo) points thi\ perspective pr"f"",tive situates ituates thll1g\ wrong way way round. round. (lark larkargues argue does not l'f( 'Ir . qUl le Ihe me way: xterior o( th vase, va ,aas opposed opposed to to the .. same way: the the crack" crack is on on the the exterior of the i n quite '~. th exampl illustrate, illu trate, the th specific pecific spatiopatio. interior. A As these examples t he H.lumetri v olumetric interior. ~'~'l11ctri'. ure meaning i not not fixed, fixed, geometricdet~il; details.;;ociated associatedWIth withthe theendo enclosure meaningo( of 1/1 in is vc'(\ on th utterance utteranc context. context. but but I'is den derived in part part (rom from the IThe he .... hema that ari from these specific example leaves leaves aside a ide the the schema that arises from these specific examples «.ntc t ' I"'tilic detaIl,. R.ther, it It Siv ri"" to to aa highly highly abstract abstract spatial spatial relation relation details. Rather, gives rise con te xt-specific in\l.h ins endo ur : th ing aero alh context of use. use. It It is is involving enclosure: the commonalityari commonality arising across each context of thl' that .. tabli h the hema for for 111. 10reover, the hem a for for this,,>1llmonality commonality that establishes the schema in. Moreover, the schema "' ry little I· and th.t they they must mu t exist, exi,t, says \C very littleabout about the the nature nature o( of the the F and RO, RO, only only that in ',1\ have the the basic basic properties properti that that enable enabl enclosure. en do ure.Crucially, eru ially, and and ih.t that they they must must have tCognitive "gllltive Grammar umes that units which which populate populate the th Grammaraassumes that the the symbolic symbolic units ment.1 more than than schemas, hem., abstracted ab tr.cted from (rom language language mental grammar grammar are are nothing nothing more hall begin to see,,in Chapter 40 LC ' M Theory Theory enshrines en hrin 4, how use. We We .shall begin to in Chapter how LCCM ~ aas fundamental count of meaning construction. construction. Specifipecifi . language language uuse fundamental to to it its aaccount of meaning ",II), . mantic units, unit, the thelexical lexi Iconcepts, concept>, cally, II ,Irgue argue that that lingui'tically linguistically ennxk,(\ encoded semantic ir situated ituated interpretation. IIlterpretation. A A situated situated interpretation interpretation is, i ,by by under>pec ify (or underspecify for th their function of oflanguage language use. u . ddini tion, aa function definition,
U"'.
Semantic structure Semantic structure structure versus versus conceptual conceptual structure ('nsniti ve Iingui;t, (or example and [Anacker Langackcr(1987)—have (1987)-have Cognitive linguists—for example Talmy Talmy (1000) (2000) and di,tinguished between two types types of ofsemantic semantic knowledge: knowledge: that that which whi h isisrich rich distinguished between two which is i; schematic. -hematic. This This distinction di;tinction is i implicit implicit in in the the work work of of and that that which and L.mga,ker and in the the work work of of'lalmy. T.lmy, for (or instance, in,tance, distinguishes di'tingui he; Langacker and explICit explicit in Talmy. Taliny, meaning associated a iated with with what what he he refers refer to to as as the the closeddosed· between >chemati between schematic meaning das or nr grammatical grammatical subsystem, ubsy tern, and and the the rich rich meaning meaningassociated a soci.ted with withthe the class illu,trate,consider eon,iderthe thefollowing (ollowingexample: e .mple: open- lass or o rlexical lexi .1 subsystem. ubsy tem.To -1" illustrate, open-class
filii
(to) (111) A A popstar /'t'ps/,Ir kissed k. d the the Om
'I he forms fo rm in m bold: bold: a, a, -ed, ·ed,the, the,and and-s· are areassociated a soclatedwith WIththe thegrammatical grammati al The ub,y,tem. Their Their semantic ",m.ntic contribution contribution relates relates totowhether whetherthe theparticipants participant subsystem. (/'t'p,"'rlfiIllS) evoked by by(to) ( HI )can canbe beeasily ca ilyidentified Identilil'(\by bythe thehearer—the hearer- theuse u'"ofo( popstar/fiws) evoked the indefinite indefinitearticle article aaver u the the definite definite article arti Ie the—that the-thatthe theevent eventtook tookplace place the versus before now—the now- the use u of ofthe thepast-tense pa,t ten",marker marker -ed—and ·ed -andhow howmany manyparticipartid · before I'an" were were involved—the involved- thepresence pre>enleor orabsence ab",nccofofthe theplural plur.1marker marker ·s.. pants In In contrast, contra t,the theforms (orm ininitalics: italics:popstar, popS/ar,kiss, kiss, and andfan fall are are associated associated with with the the leXical subsystem. , ub y,tem. That That is, i , their their semantic ""manti contribution contributionrelates relate to tothe thenature natureof of lexical the participants partieip,,"" involved involved in in the the experiential experiential complex, com pIe , and and the therelationship rd.tio",hip the holdmgbetween between them, them,namely namelyone oneinvolving involvingaakiss. ki ,.InInother otherwords, word while , whil the the holding do\C'd dd , forms form encode cn((k1~ content ( Onlentrelating reloltingto10structural tnu.: turJIaspects d pct:hof ofwhat whatwe we closed-class -
These examples Th examples involve involve spatial . patial scenes en ofofslightly lightlydifferent d,ffi rentkinds, kmd where , where in '" reflects a spatial relationship between the figure (F) and the reference object retle"" patial relatiomhip between the figure (P) and the re~ r nce object (RO). (RO ). In In (9a) (9' )the the F,1-, the I'" kitten, kill"',is" fully (ully enclosed ndo ....'(\by bythe theRO, RO, the II"box. btl\. However, Ilowcver,in m the the other other two twoexamples, examples, in ill does doe> not not prompt prompt for for quite quile the the same ... me kind kmd of of relationship. rel.tion hIp. In In (9h) (9b)the theflower nowerisi;not notfully fullyenclosed endo ....'(\bybythe thevase, v. since , inleit ilpartly partly "
("lark uses the term "signal meaning" to refer to sentence (and vrord I meaning.
rnlght refer refer to to.1' the experiential xpcrientialcomplex omplexevoked, evokc..~,the theopen-class 0llCn -dJ ,forms form\give give might as the
42
INTRODUCTION
rise to detailed information concerning the nature of the participants, scenes involving the participants, and the states and relationships that hold. The distinction in types of semantic representation is also present in Cognitive Grammar (e.g., Langacker 107). As we saw above, Langacker argues that lexical classes such as nouns and verbs encode schematic: meaning. 15 Another distinction claimed to hold between nouns and verbs has to do with whether there is a temporal dimension encoded. Langacker maintains that verbs (but not nouns) relate to time, and encode the evolution of a particular event or state through time." A further schematic aspect of meaning has to do with whether a form encodes a schematic trajector (TR) and/or landmark ( LM ). For example, Langacker argues that relational lexical classes, such as prepositions, encode a highly schematic TR and LM as part of their semantic structure. 15 For instance, in terms of an expression such as: under the sofa, it is by virtue of under encoding a schematic LM that a noun phrase (NP) can be integrated with the preposition under, giving rise to the complex expression: under the sofa. 1 ^ In addition to schematic meanings of this sort, Langacker also assumes that words encode "rich" semantic content. As we saw earlier in the discussion of encyclopaedic semantics, this is conceived of in terms of J profile/base complex, in which a given form designates or profiles a given substructure within a base. The distinction between schematic versus rich aspects of meaning can also be seen in Goldberg's work on Construction Grammar. Recall that Goldberg argues that sentence-level constructions, such as the ditransitivc, have a highly schematic meaning associated with them, serving to encode argument roles. In contrast, individual words such as the verbs which populate these constructions, e.g., bake, are associated with rich frames and participant roles. In this book I argue that the distinction between schematic versus rich meaning identified by Ta!my, Langacker, and Goldberg actually reflects a distinction in types of meaning representation, and that the two distinct types of representation relate to distinct systems. Schematic meaning relates to representations that are specialized for being encoded by language. That is, such representations take a form that is highly schematic in nature, specialized for being encoded in the auditory (or signed) medium that is language. Representations of this sort are what I refer to as semantic structure. The " The distinction between verb% and nouns concerns the nature of what is being profiled: a region in a domain. in the caw of nouns. and the relations that hold between such regions. in the case of '• I angacker refers to this as sequential scanning lie distinguishes the way in which verbs encode time from the way time is encoded by other - relational - lexical classes, such as adiectives, adverbs, and prepositions (see Langacker 59117, 199th, 1999, :ANA; see Evans and I irecti 21,04s for a review). .r.111Inut 'I Indeed. this notion is important for constituency in t I* In I angacker's terms, the NI' elaborates the conceptually dependent preposition under. I will discuss what it means to be conceptually dependent in more detail in Part II of the honk. The issue of elaboration. in the sense of Langacker. will he discussed in more detail in Part III 44 the 1 ►4,4 ► Ic.
TOWARDS A NEW ACCOUNT OF WORD MEANING
43
theoretical construct I develop to model semantic structure is that of the lex ica l concept. In contrast, the rich representations are associated with the conc eptual system. and are not directly encoded by language—although language facilitates access to this level. Representations of this sort are what I refer to as conceptual structure. The theoretical construct 1 develop to model o mceptual structure is that of the cognitive model. Cognitive models involve a frame and simulations deriving from the frame. As briefly introduced, simulations are reactivations of sensory-motor, cognitive, and subjective states, based on, but not identical to the perceptual and subjective experiences that arc stored in the conceptual system. Part of my argument in this book is that semantic structure and conceptual structure form two distinct levels of representation, and do so because they inhere in two distinct representational systems: the linguistic system and the conceptual system. Following arguments presented by Barsalou et al. (forthcoming), 1 suggest that the linguistic system evolved, in part, by facilitating more effective control of the extant representations in the conceptual system. 17 That is, linguistic representations are specialized for providing a "scaffolding" to structure conceptual representations, thereby facilitating their use in communication. While the conceptual system evolved for action and perception, i.e., for non-linguistic purposes, the emergence of language facilitated the use of conceptual representations in linguistically mediated meaning construction, thereby providing cognitively modern humans with a significant evolutionary advantage. With the association of linguistic and conceptual representations, humans were able to engage in the advanced symbolic behaviours that led to the explosion of sophisticated ritual practice, material culture, art, and science around 5o,000 years ago during the later Stone Age, the period that archeologists refer to as the Upper Paeleolithic ( Mithen 1996; see also Renfrew 2007). In essence, the argument I shall be making during the course of the book is that semantic structure and conceptual structure involve fundamentally distinct sorts of representations. Moreover, it is this distinctiveness that facilitates meaning construction. It is by virtue of semantic structure facilitating access to conceptual structure that words appear to he protean in nature. That is, what we might informally refer to as the "meaning shifting" properties, so to speak, associated with words is a symptom of there being two distinct types of representation implicated in meaning construction. But to claim that there are two distinct representational systems involved in meaning construction is not to adopt a modular perspective (e.g., Fodor 1 983). Modularity holds that the mind consists of domain-specific encapsulated modules, which work by virtue of one module working on the output of another. As we shall see, in LLCM Theory meaning arises by virtue of a dynamic exchange taking place between the linguistic and conceptual " This is discussed in more detail in a hapter
INl"ROOl'CTION INTRODUCTION
44
\yMcm\. M aning construction (on IrU(lion involves IOvol ..., aa continual Oph. t •.catedthan than has previou previously been aassumed, has Iy been umed, (iii) (i.i) the theview v.ewthat thatlexical lelU alrepresentations representat.on relate, relate, in part, part, to to non-linguistic (iv) the m non Iingui"i knowledge knowkdge structure, t.ructur , (.v) the view v.ew. that that the the mental mental grammar consists of symbohc symbolic assemblies of form and rammar con iSIs of a scmbh of and meanings, meaning, rather rather than than words and and abstract abstract rules rules whi which operate on on words, words, and and (v) the ~,ord h operate the view v~ewthat that meaning con construction arises in in the the context context of meaning truction ari of language I~nguag use, which whICh.is to to say .say the situated situated and and interactional the interactional nature nature of linguistically hngUl ucally mediated med.ated communicaco.m111UnlCa tion between That is, is, meaning arises as as aa function tion between interlocutors. interlocutors. That meaning arises funct.on of .ofthe th: expression of of situated intentions. I also exprCS5.on ~itu3ted communicative ommunicati.. intention. al discussed, d.scussed, in m this th. of LCCM Theory, the the presump presumpthe hheart chapter, tm tion at at th rt ofl .Cr.~ Theory, chapter, the the pnncipled principled di distinction tion that consist tion that the linguistic lingui tie and conceptual con eptual systems tern con. t of distinct d. unct types types of of representation: the lexical concept and the cognitive model. As we shall see, rcprC\Cntation: the lexical con cpt and the cognitive A we. hall see, although distinct, these representational types interact and thereby distinct, these reprC\CntJt.onal types mteract and therebygive gIVerise nsetoto short, LCCM the apparently protean nature the apparently protean nature of of word word meanings. meanin~. In short, L 'M Theory Theory assumes a principled di distinction as,umes tinction between between semantic semantIC structure stru ture and and conceptual conceptual structure. trutture.
3 Cognitive linguistics \, the lex; al representation representat.on and Jnd semantic semanu composition ompo .tion As the present present treatment treatment of lexical "is grounded grou nded in the the perspective perspective known known as as cognitive cognit.velinguistics, linguistics,this thi chapter chapter hrlefly Iinguisti enterprise. enterpri . Cognitive Cognitive linguistics lingui tic isi briefly introduces introduces the cognitive linguistics drgu.,bly rapidly devel ping approach approach to the the relationship relation hip between between arguably the the most most rapidly developing 1.lOguage, sociocultural experience experience in in the the language language sciences, iences. language, mind. mind, and and human sociocultural .10gntlive cognitive and brain grnce. lencc. and phil""'l'hy. In In other olher words, "'nrds. the Ih Cognitive Cognili,. gence, cogniuve cognitive neum neuroscience, and philosophy. (Commitment nmnlllmenl asserts a rt that lhal models mooel of oflanguage language and and linguistic lingui Ii organization organizalionshould hould nil d whal known about aboul the Ihe human human mind, mind.rather ralherthan Ihanpurely purelyaesthetic aeslhetic reflect what Iis known ,hdJle, use of o[ particular particular kinds kind, of of formalisms formali,m, or or economy lxonomy of 01 dictates ,uch such aas Ihe the use n·pre" nlalion. aas in approach to 10lingu15u representation, in Ihe the (case of of formal approaches linguistics..'4 'I he CognillVe Commilmenlhas ha,aanumber numberof ofconcrete (Oncreleramifications. ramificallon ,Firstly, Fir Ily. The Cognitive Commitment hngu"lic 11 include include structures IruLlur or or processes prlllc"" that Ihal violate viol.le known known linguistic Iht.'Ories theories cann cannot Ihe human human cognitive (ognllive system. y I m. For I;or instance, in lance.ifIfsequential I, linguistic.organization organization and mncerned of the Ihe key key aspects aspects of oflinguistic lingui lic knowledge, knowledge. and and the the concerned Wllh with some some of ) m!>ohc nalure language which.. shall argue argue in in detail detail later laler ininthe Ihe symbolic nature of of language which, as II shall h"Cnses associated aialed with wllh over. lance. can be be criticized on similar can CTilicized on imilar grounds grounds to 10 models model of oflexical lexical semantics mantics that Ihal constitute Sense Enumerative Lexicons, discussed in Chapter I. Indeed, this con~lilUI Sen Enumerative Lcxkons. discussed in 'hapler I. Indeed. thi, is Ihe the tack will lake take III in Ihe the next where II briefly briefly review review Ih the model model of is la k II will nexl chapter, chapler. where of word meaning proposed by Brugman, Lakoff, and others. word meaning proposed by Brugman. lakofT. and olhers. Tyler and and I.1, in in the Tyler Ihe earlier earlier phase pha of ofresearch rcsearCrve d lennine what whal counts ounl as as aa distinct dislin I sense n (i.e., (i.e,. aa lexical con concept—a time in lelCical epl-a term lerm that Ihal was wa used for for the the first fir I lime in the the context conlext of of Principled Polysemy in Evans loo4a), and thus should distinguish bePrin ipled Poly""my in [van 2oo4a). and thu hould di lingui h be tween senses stored Iween sen lored in m semantic semanlic memory, memory. and context-dependent conlelCl-dependenl meanmeanings ing constructed constructed "online," "online," and and •• they Ihcy should hould establish lablish the Ihe prototypical prolotypical or orcentral cenlral sense sen associated a ialed with wilhaa particular network. This poinl point isi important particular semantic manli nelw rk. Thl Importanl because because cognitive cognilive semanticists manlicisl have have not nOI always always agreed agreed about aboul the Ihe central central senses senses of ofsemantic semanlic categories. '987) argued calegories. For For example, example. while whileLakoff lakoff((1987) argued that thai the thecentral centralsense sen'" for for over o.'rr is i the Ih ABOVE-ACROSS ABOVE-ACRO S meaning, meaning. Kreitzer Kreilzer (1997) (1997) has has argued argued more more recently that it is an ABOVE meaning. recenlly Ihal il i an ABOVE meaning. In In our our 2003 2003 hook book The Tilt Semantics Selllall"csofofEnglish EnglisllPrepositions, PreposillOllS, Tyler and II sought ughl to provide decision principles that could be applied to the entire class to provide deci ion principl that could applied to the entire cia of of English prepositions. Here I will briefly outline the principles we proposed in Engli h prepo itlon . Ilere I will briefly oUlline the pnnClples we proposed in that that work work for for the the first fir t of ofthese th issues: issues: how how totodetermine determine what what counts countsasa distinct distinct lexical lexi al concept concept associated a iated with withaagiven givenprepositional prrpo itionalform. form. We provided two criteria for determining whether a particular We provided two criteria for determining whether. particular sense n of ofaa preposition counts as a distinct lexical tont:L.1n: prcpo Ilion (ounb a J diMinlt lexical (om.:ept: i.I. For ror aa sense ~n~ to to count lounl as a distinct, di\tmc..t. ititmust must involve m"'(llve aa meaning meaning that that isis not not purely purely.spatial patial in in nature, nature. and/or and/or aa spatial ,p.lt ..1configuration configuration holding holdlllgbetween between
57 57
Ihe Referen e Ob,e.! distllllt from from the the the I·igure Figure(n (F) and and Reference Object (RO) (RO) that that iis distinct other n conventionally iated with prepo ition; and other senses conventionally aassociated with that preposition; "n., There There must that are instances of of the the sense sense that are context-independent: context-independent: must also also be be instance lI"tanc in which the di tmct sen ould not be inferred he inferred from from anoth anotherr instances in which the distinct sense could .. "''' and ontext in which itil occurs. oc urs. sense and the the context
,
1Cd or or composed, composed. giving givingrise risetoto way oflanguage. language. II also al considered con id r dthe th way wayininwhich whi hLCCM LCCMTheory Theory unit of larger units erveonal that the the language language user user may may potentially potentiallyseek seek interpersonal functions that an , an' " .. ,c ,or this thl. is , that that language language users users se 1,InSuaSe language to to express express and and ,ullil. fulfil. One One rcason reason for i„ u ~ ,n io physical matrix hiftin.g and 0 two two ::: sociophysical matrix 'hat.i that is continuallr continually shifting and e~olving. evolving. No live i n a "IC ,uall un'. feehngs. hIp. at POlllt in 10 time, lime.are areexactly exa t1y situations, feelings,or or relallon relationships, at any any g,ven given point "lIkC. We arc contlllually using language to expr . unique meanings. about a like . We are continually using language to express unique meanings, about a nique uniqu ways. way. While While language languag has ha nique states states of of affairs affairsand and relallonship relationships,• in unique u ~a r.uo~e made schema • oror ymboli h can an be to rangeuf Ofready· ready-made schemas, symbolicunits unitswhi which be combined combined to to refer refer to to and and eexPfC" ntative range press aa rcpr representative rangeof of th the !>Cenario scenarios we we may may wish wish to ,k",ribe. rily und rdet rmine the human experiexpen describe, these these nce necessarily underdetermine the mutability mutability of of human employed by by language language users u r can can only only cnle ence. AClordongly. Accordingly,the the symbolic symbolic unit units employed e,er y sanction nction (in Langacker' s ternlS) the situated SItuated way way in in which whi hthey they ever partiJll partially (in Langacker terms) the Jre 1996) observ • language are u,ed. used. As As Clark Clark((1996) observes, language use use involves involves solving solving aa coordincoordination problem. in which language uscrs must employ non onventional ation problem, in which language users must employ non-conventional """Jonatlon trategies and devic . That That,is,•language language users u r typically typiGlllyemploy coordination strategies and devices. employ ,he repertoire of lingui ti units, units, including in ludong patterns pattern of ofassema mthe (on\'entional conventional repertoire of linguistic hling ( u h as a word-order word order conventions, conv nti n •which whi hare arethemselves themselves bling linguistic linguistic unit units (such Iingui\li' non -conventional ways." ways."On Onthis thisview, Vil~. meaning, meaning,which whichisi~ linguistic unih), units), in in non-conventional a,,,,,ia,ed u ge event). is aa consequence consequen e of ofcombining combining associated with with th the utterance utterance (or (or usage event), is unit in in novel novel ways ways in in order order to to solve solvethe theparticular particular coordination coordination the ,ymboli the symbolic units pruhlcm Jt th reby facilitating facilitating communi ation. problem at hand. hand, thereby communication. We >aw previou section section that that one one reason reason for for the the protean protean nature nature of of We saw in in the the previous "",J meaning arises ari from the the non-linguistic non ·lingui tic semantic semantic potential potential to towhich which word meaning from lex ical con epts afford afford access, a«ess. and th range r.lIIge of of simulations imulationsthat thatcan canarise. arise.In In lexical concepts and the " Motion. aa second second reason reason arises arises as as lexical lexical concepts—the concepts--the semantic semanti pole pole of of addition, ,ymbo lic units—are units-are only only ever ever realized reali,ed as as part part of oflinguistic lingui ti utterances. utteran es.UtterUttersymbolic .• by . ituated. and and thus thu,form formpart partofofan Inact act ance. are arc nnecessarily ' rily (i.e ances (i.e., by definition) definition) situated, of communication. communi ation. But But in in being being so sorealized, realized. lexical lexi .1 concepts con epts give give rise rise to to of semantic contributions; contributions; as a we weshall hall see see in in Part Part II, II. the the linlin ontext. indu ed semantic context-induced gu "tic content «(lIItent encoded encoded by bylexical lex".ll concepts conceptsconsists con iSIS of of"bundl different "bundles""of of different guistic t}'f'C> of oflingui tic knowledge. knowledge. Accordingly, Accordingly. different different aspects a peel of ofthis thi knowledge knowledg types linguistic can become beeome active active in in different different contexts. onte.ts.The Theconsequence consequen eofofthis thisisi that thatlexical Ie ical can ill tow. 10/0. Rather, Rather. itit is is only only the the contextually contextually w nccpt are are never neveractually actuallyrealized re.li,ed in concepts relevant aspects aspect which whi h surface unace in inlanguage languageuse. u . relevant wecan canliken likenthe thedistincdi tinc. Borrowing Bo rrowing an an analogy analogy from from phonological phonological theory, theory.we loo n between between lexical lexical concepts concep" on onone onehand, hand.and andtheir theircontextualized contextualizedinstaninstan tion on the the other other as as akin akin totothe th distinction distinctionbetween betweenphonemes phonemesand and tiallon on tiations ju,t as as with withphonemes, phonemes.lexical lexicalconcepts concep" qua 'I'm mental mental representarepr nta aliophones. lust allophones. tlon, arc are never never actually actuJlly perceived. pcr<elVc'tl. Rather, Rath r. their their existence exi"ence isi,inferred inferrc'tl based basedon on tions the variability, variability. and and commonalities, commonaloti •ininword wordmeaning m aningacross aero (situated) (. ituated usage ) u>ag the event,. as a judged Judged over over many many instances in"anc of u . In In this, this. then, then. the the job jobofofthe the events, of use.
-,d:
■
WORD IN LCCM L eM THEORY TlIEORY WORD Mf.ANIN(i MEANING IN
I
11 This Th aaissue t uc is 11discussed dl wtNfurther lurthc-rinInChapter l h.irll."fILIL
"
I
72
INTRODUCTION INTRODUCTION
lexical semanticist ",mantid t isi to toemploy employthe thesituated Ituatl.,)semantic manticcontribution contributionofof gIVen lexical a agiven word. by by analogy analogy akin akin to to allophones, allophone. in in order order totoinfer onferthe theexistence exi ten eofofthe Ihe word, underlying lexical lexical concepts—stored concept $lored mental mental schemas—akin hema -akin totophonemes. phonem • underlying which partially partially sanction '>Jnction the thesemantic '>Cmanticcontributions contribution.which whichsurface. surface. In In view view which ofthis Ihisdistinction, di tinction.ininthe theremainder remaonderofofthe th book bookI Iwill will refrain refraon from from using u ingthe the of term "word "word meaning." meaning."Rather, Rather. IIwill will refer refer either eilher to tothe theconstruct con tructofofthe th lexical lexic.1 term conccpt - when IIam amreferring referring totothe theunderlying underlyingsemantic '>Cmanticstructure—or, 'tru tur r.toto concept—when the semantic ;em.ntic contribution ontributionof ofaagiven givenlinguistic lingui ticform—when form - whenI Iam amreferring referringtoto the in tance of ofaa lexical lexical concept. concept. ituated instance aa situated In view view of of the the foregoing, foregOIng. we we are are now now in III aa position po ition totoprovide providesome somebask ba Il In di tinctIon with with respect rCl>pcct to to meaning meaning and and use u that thatare arecentral c nlraltotoLCCM lCe 1 distinctions Theory. First Fir t of all. we we need need to to provide provid aa definition definitionofofan anutterance. utterance.This Thi is" Theory. of all, I straightforward traightfon>..d aa task t. k than might assume. a ume. As,I will define defoneit,it.aausage U\dge less than one one might I will event or or utterance utteranc has ha aa unit-like unit like status Matus in in that that itit represents reprC5Cnts the Iheexpression expr ionof of event coherent idea, idea. making making (at Cat least lea\t partial) partial) use u e of of the the conventions convention ofofthe the aa coherent languag informally. Ihe lingui lic behaviour behaviour in in aaparticular particular language—informally, the norm norms of of linguistic Iingui lic community, communily. but bUI see \ee Croft Crofl Cl000). olherwords, word.an anutterance utteranceisi linguistic (boo). InInother om<whal discrete di rele entity. enlily. However, However. II use 05e the Ih expressions expr ion "unit-like" "unil.like" and and aa somewhat •"somewhat ...omewhat discrete" d,\(rele" because beeau", an an utterance utterance isi not nolan anabsolutely ab",lulelydiscrete, di ·rele.nor nora a preci;ely identifiable identiliabl unit. unit. This Thi follows folloW5 as a utterances utteran e5involve involvegrammatical grammalical precisely forms uch as a word word order order and and lexical lexical items, ilem •lexical lexical concepts, concepl •speech speechsounds, sound,. forms such pattern, ,uch as a pitch pilch contours, COnIOuf\.slight 'lighl pauses, pau~ and and accelerations accelerallon patterns of of inlonalion intonation such and and so so forth. forth. While Whil these IhC5Cproperties pr perti converge convergeon ondiscretedi\(releand deceleralions. decelerations, and nand Ihey do nOI co-occur co·occurininfixed fixedpatterns, patterns.and andtherefore Ihereforedo donot nol ness and unily. unity, they do not provide ",I of ofcriteria crileria for for collectively colleclively identifying idenlifying an an utterance. utteran«. In In this thi. provide aa set rrespect, peel. utteranc r laled nOlion ofaa sentence. ntence. utterances differ differ from from Ihe the related notion of A in particular parti ular by by formal formallingui t •isi an an abstract abstract entity. entily. A <entence. sentence, as as defined defined in linguists, In It is i an an idealization id.ali1.Jlion that Ihal has hasdeterminate delermimle properties, properti .... often often In olher other word,. words, it taled in grammallC.1 structure. tructure. For For example. definition of of(an Can stated in ternl terms of of grammatical example, one one definition Engli h) sentence i t ofofIhe VI'. English) sentencemighl mightcon consist theformula: formula:S Sc>NPPVP. The of aa sentence, ",nlenee. while while based b.l\l.,) on on prototypical prolotypical patterns patterns found found in in The notion notion of utterance. an utterance. utteran e. Utterances Utteranc typically typically occur OCcur sponponutterances,i,is not not the the same same aas an laneou Iy. and conform to 10 the Ih grammaticality grammaticality requirements requiremenl of of. taneously, andoften often do do nol not conform a ~ell formed ",ntence a, undef\l(xld in formallingui\lic thl'ory. I'or example. well-formed sentence as understood in formal linguistic theory. For example, on i t of ingle word H;.~. aa phra in lerm, terms of of 'Iructure. structure, an an utterance utterance may maycon consist of aa single word C(Hi!), phrase C(No '0 Imy!). an incomplete ",ntence C Did YOII pilI l/or , .. n. or a ntenee way!), an incomplete sentence (Did you put the... ?), or a sentence thai that (onlJin "error,," of pronunciation or grammar btxau\e the peakcr 1\ tirl'ti. contains "errors" of pronunciation or grammar because the speaker is tired, di;tracted. While much much of offormal formailingui ti has has been been distracted,or oreXClled. excited,and andso so on. on. While linguistics concerned the properties properti of offangu,'ge Ihat enable enable us u to to produce proxluce concerned wilh with modelling modelling the language that grammatically formed \Cntencl .... utterance exhihll graded grad d gramgrammaticallywell well-formed sentences, utterances "ftcn often exhibit grammalicality .ngocker 1987:;ee hort. whil maticality (( seeILangacker 1987; seeal", alsohan Evansand and(,rcen Green1(06). 20 o6).InIn short, while aa <enlence can be preci>ely and narrowly defined. an ullerance cannot be. sentence can be precisely and narrowly defined, an utterance cannot be. \Vhilc 'nl Ihe \(xiJll'll With uttcr. While\Cntcnu,\ sentences repr represent the\tru(turc structureJassociated with aa prntntypkal prototypical utterJnlC. ance,UtlcranlC utterancesrcpn."\Cnt represent pt.*f,;ifil specificand anduniqu uniquein\tJI1'-l: instancesof ofIJnguage languageu~. use.On(e Once
sentence is given meaning, context, and phonetic realization, it becomes a WORD WORDMEANING 1EANINGININLCCM LeeMTHEORY TIIEORY
I
7373
a Il h..'.H:e i\ given meanmg, context, and phonetic realization. it lltxome a lexical ....ken) ,kcn )utterance. utterance.Accordingly, Accordingly.as asIIam am concerned concernedwith withan anaccount accounlofoflexica] ( •,po h i· 1'1"lrl''ot,. · construction . that sentation isIused, it• It repre ' ntJtiolland In dmeaning me.mmg (on\trw,:tlon th atreflects rc nlXl\how owlanguage .1nguage u\CtI· is::IU ultimately ltllll,ltcly the Ihe utterance, utterance. rather ~ather then Ihen the the idealized idealoled notion nOllonofofthe Ihesentence, \entence. ith which IIhith IJam amconcerned concernedin111the thepresent prC5Cntwork. work. w,,'th }wing provided ofof ananutterance, IIJ' 1118 prnvidedthis this(qualified) (qualofied)definition definilion utterance.weweare arenow nowinina a position I"",tion totodistinguish di ~ingui hmeaning meaningfrom fro~lexical lexicalrepresentation. repre ntJlion.My Myclaim d~imis "that that distinction between lexical representation and meaning is that the th,'essential n..ential dl linctlon between leXIcal repr.... ntatlon and meanons I Ihat while the utterance, IIhlk meaning meaningisI aa property property of oflhe utterance.lexical lexicalrepresentations repre..entation,consist c(1n,i,1ofof the Ih,' mental Illenlal abstractions abstra tio", which whichwe weinfer infermust mu tbe bestored ,toredasaspart partofofthe thelanguage language user's knowledge of language: symbolic units, together with the range u a\ knowledge of language: symbolic unit. logether with the range of of cognitive models, the semantic potential, to which a lexical concept affords "'~I1II1VC models. Ihe \emanlic pot ntial. to which a lexical coneept affords access. .I"e" . Hence, lIen( • lexical lexical representation represenlalion involves involVe!> structures )!ructur... of ofdistinct dislincttypes 1ype5 which system II hich inhere inhere in in two IWOdistinct di tinctrepresentational representationalsystems: YSlem:the thelinguistic lingui tic. y lem and the conceptual system. The interaction of these distinct types of struc.11ll1 Ihe coneeptual 'Y'tem. Th interaction of Ihe di\linct t)'PC' of tru tures gives rise Ill .." gIve, ri\e to to meaning mcaning associated a sociated with withan anutterance. utterance.The Themeaning 01 aningassoa~) conception. will refer to as a with an utterance I ciated lOa ted wilh an utterance I will refer to a a conception. for the An architecture architecture for the role role of of words words meaning in mea ning construction construction
The conclusions rhe (ondu ion to to emerge emerge from from the the previous previou discussion di u ion suggest ugse5t aa number number of of We require requirements for a theory of lexical and compositional semantics. requiremenl for a theory oflexi(al and compo,ilional ",manlics. We require
4. 1
both an an account account of hOlh of lexical lexical representation reprC'>Cntatlon and and aa theory theory of ofsemantic >emantic compos(ompo ition, which ltiun. which together togelher should hould contribute contribute to toa adescriptively d riplivelyadequate adequate and and require We psychologically reali realistic account of of meaning I',ychologically tic aecounl meaning construction. cun;truction. We require aa theory of of lexical thc~lry lexical representation repre>ent.ltlon which which provides providCl> aadescriptively dC!>er appear appear to possess. We al", also require requirean anaccounl account which which provide. provides Ja meam means of of understanding I'" \C". We undef\\.nding how lexical representations interface with conceptual knowledge, how le"ical rep ...... ntation inlerfalC with conceplu,li knowledge. which which isi,to10 their;emanlic semantic potenlial. potential. That That is, say, their \dy. i • we we require require aa theory theory that that shows how how howthe Ihe linguistic and conceptual systems interact in order to produce semantic linguistic and conceplual . y tem interact in order 10 proldULC \emantic representations. Weal'>. together wilh with Ihe the informational together informational characterizations characterization derived derived from from the thesemantic semanti potential available, available,combine combine inin order order 10 to provide I'otential provide situated situated meanings, meaning. that thai is, is. conceptions. Finally, as Ihe the .. semantic contributions associated "'nccplion,. J.inally. a, mantic c"nlrihuli,,", a""x:iated with with words word,are arc function of of specific and thus aa fun, and Ihu aa consequence consequence of ofdiscrete discrele usage usage events,the the d(c.:ouni account dcveloped developed of event. or lexical Ie iedl representation rcprc ntJtion and and semantic 'Cl1lJntic com(om As the two aspects of the thoroughly u\Jge·ba\Cd usage-based in in nature. position mu,t must be be Ihtlroughly p,"ition nature. A, the IWO a,pec" of Ih presentaa summary of the the architectheory II prc;enl presentare are relalively relativelycomplex. complex. II pre;ent theory ummary of archilecarc argued for in detail in II turebelow. below.All Alloflhe of thecomtru," constructs inlroduced introduced are argued for in tlet,liI in Parts lure Part II andIII IIIofofIhe theIl<xll.. book. and
_4
WORD MEANING IN LCCM THEORY
INTRODUCTION
LCCM
75
theory
LCCM theory consists of (1) an account of lexical representation (symbolic units and cognitive models) and (ii) an account of semantic composition: integration of lexical concepts in a way which activates, or, in my terms, provides an access route through the cognitive models to which a given lexical concept affords access. This can serve to highlight particular attributes— aspects of a cognitive model, such as properties—and structural invariants—relations holding between attributes—of a given cognitive model. As noted above, the fundamental assumption is that meaning—more technically a conception—is a property of an utterance—a situated instance of language use—which arises, in part, by cognitive operations which apply to the lexical representations—lexical concepts and the cognitive models to which lexical concepts provide access sites—deployed by language users. Thus, meaning arises by virtue of language users forming interpretations based on the lexical concepts employed, the way lexical concepts are combined, and the access routes through the sets of cognitive models—the cognitive model profile—accessed by a given lexical concept. Moreover, these interpretations are always guided by linguistic and extra-linguistic context."
txic he :tic:::if: at :11; t e'rgnreadtiiaotned meaning construction process takes place by of semantic composition. This process involves two component procs ( i) lexical concept selection and (ii) fusion. Lexical concept selection vc ei rs,Isue: invo lves selecting the most appropriate lexical concepts associated with each vehicle in an utterance, guided by utterance, discourse, and extra-linguistic c ontext. The appropriateness or otherwise of the selected lexical concept is a function of semanticality—the semantic acceptability of a conception. This is discussed in Chapter 13. Fusion, the second compositional process, consists of two further constituent processes which are held to occur in tandem: (i) lexical concept integration and (ii) interpretation. Integration involves the construction of larger lexical entities, driven by linguistic knowledge (lexical concepts). These larger lexical units, which 1 term lexical conceptual units, are then interpreted. That is, the larger unit receives what I earlier referred to as an informational characterization. As such, those parts of the cognitive model profiles (semantic potential) associated with each lexical concept in the larger unit are interpreted in a way that is in keeping with the larger unit. Put another way, integration provides (linguistic) instructions which serve to determine how the various lexical concepts are collectively interpreted, and
Lexical representation LCCM Theory
LCCM Theory holds that knowledge of language includes (i) symbolic units, and (ii) cognitive models. Symbolic units consist of bipolar assemblies of form, what, as noted in the previous chapter, 1 refer to as a phonological vehicle (or vehicle), and a lexical concept. Lexical concepts constitute linguis tically encoded concepts—that is, highly schematic knowledge encoded in a form that can he externalized via language. Lexical concepts are conventionally associated with vehicles of all kinds including words—the focus in this book— hound morphemes, idiomatic phrases, and grammatical constructions. Accordingly, lexical concepts, by definition, concern purely linguistic knowledge, as discussed in Chapters 6 and 7. A second important part of the lexical representation is the notion of the cognitive model, which is a large-scale coherent body of non-linguistic knowledge which lexical concepts provide access sites to. The range of cognitive models which are accessed, either directly or indirectly by a lexical concept, as noted above, I refer to as a cognitive model profile. Individual cognitive models consist of attributes and structural invariants. 15 These ideas are developed in more detail in Chapters 9 and to.
Lexical representation
symbolic units
vehicles
18
The role of context in scrturitic ‘.ortipo‘ii n t%
III of the book 85
wo∎ril In nun(' tleJil at Various remit
1
in Kul
cognitive models
lexical concepts
Semantic
composition
lexical concept selection
lexical concept integration
semantic representation
.
See Bamilou ltyyzu. winbl; Bars( mu
a al- 0993)-
l'ItiLlte 4.1. An overview 01 the architecture of LCCM Theory
tusan
interpretation
76 7t'
thus, th access ace route rout that that each ~ach individual indly,du,ll lexical lexical ((incept wnceptaffords afIordsthrough thro ugh thus, the itits cognitive cogniti\e model profile. profil . The Th result r""ult is i that that any any given given word word will will provide provide aa unique unique activation a semantic '>Cmantic potential pot ntial on every every occasion occa ion of of uCmantic potential- the cognitive cognitive model model semantic potential—the prolile-to which they afford aflord access, ac<e ,which t the the scene ene for forthe thedetailed detailed profile—to which which sets development the theoretical theoretical constructs constructs of the the lexical lexical concept concept and and the the development of of the ((1gl1ltive in the the next next part part of ofthe the book. book. cognitive model in begin, con consider To be'llin, ider the the following followingfour fourutterances utteran .",first fi"tdiscussed di ussedininChapter hapter1:J. (2) 2)
WORD MEANING IN 1.CCM 'EMORY
INTROOUCTION INTRODUCTION
a. h,m« isi aa country country of ofoutstanding oUl>tandingnatural natural beauty OC,lUty a. France b. !-rJnce nation in in the the European European Union Union France i,is one one of the leading nations c. France 'ew Zealand th 2007 2007 Rugby Rugby World World Cup Cup France beat beat New Zealand in the dd.. !-rance E «In tllution in in the the2oo5 200Sreferendum referendum France votl'" voted agaimt against the the EU constitution
In each each of th these exampl examples the semantic In <emantic contribution contribution associated a iated with withthe the form France is /-rlll'ff slightly distinct. distinct. That That is, i ,the mantic contribution (ontribution provided provided by by is slightly the semantic France va /-mllff ric> across aero, these the..,. distinct di lJJlct utterances. utterance,. The key key insight in 'ght of ofLCCM LCCM varies Theory that the the reason reason for for this thi variation variation isis due duetotodifferential differentialactivation activation Theory is that of non-linguistic non ·lingui,tic knowledge knowledge structures, ,truttur , the th cognitive cognitive model model profile, prolile, to to which whith the lexical lexical concept associated a\Sotiate'" with WIth /-rlll,er and France afford, affords a«"". access. The linguistic and non · linguistic processes proce that give give rise rise to tothis thi differential differential activation, which non-linguistic that relate. hngUl til contexts (nnlcxt In /:,cmu relate, In in part, to the diffcrcnc.: differences in the four linguistic in which which France Theory iis emocdded are highly highly complex. complex.LCCM LCe 1 '1 hl,(lry represents repr< 'n" aa programmatic embedded are attempt to to identify attempt id ntify the the sorts son of ofmechanisms methano m involved IIlvolved in inthis thisactivation activation process. concept (om" 4. onventionally In lhc~ cx.lmplc II am ulOlcrnoo with wilh the thc lexical Icxi.,;.)1 \.om.cpt cntionally these examples am concerned above, and as with the vehide As noted .bc,,·e, aassociated SOliated wllh a we we shall hall see detail vehicle Frtll'ff. France. A, see 111 in detail prcXC~\.
77
Ilater 'r ct hapters, h.lpte'" a lexical lexical concept «Incept constitutes ((lnstitute, aa relatively relatively complex comple. body body of of in 101 '11'tit knowledg foml aa representational repr~ntJtional unit. II identify identify these the linguistic knowledge whkh which forms long""-ntational . 'd' I bel . '01tational units, I XI al concept), can ept), by by providing proVI JOg aa label a in111small ma II units, (the lexical rer rest rel"'1a.1 1\ "'uare br.lket . Thu I..ical «Ineept associated J>><xiJted with with!the th ls within square brackets. Thus,, the lexical capit .al" ..... . concept in the the examples in (2) II gloss [FRANCE].. In which appears appear in exampl.", 111 glo as ~ ['RANt' /."mn' which fo rm France horm . II addition, concept affords a,I.I,IO" n, aJ key property is i that that aa lexical concept afford access ace"", to toaapotentially potentIa y e set of "",t of cognitive cognitive models: model" its Its cognitive cOllnltne model model profile. pro hIe. AArobust robust finding findll1g Ilarg .,. led " h cognitive psychology 'nml psyc~ologyon onknowledge know gerepresentation reprco;entatlonisI that t at recent work in co~niti~e from «<ent .h< representations whICh II1h re in 111 the conceptual can eptual system, . y tem, while whIle extremely extremely the representations which inhere are not an an unstructured unstructured assemblage." a '>Cmblage.'·Indeed Indet'"recent relent r.",.arch proco mplex, ,~)I 111,le'\ • arc . research .pro. vides tompelling compelling evidence evid nee that that rather rather than than knowledge knowledge being bell1g organized o.'ganlzed in 111 of attributes, Iterms "''' "off lists lim of attributes, a key key aspect peet of ofknowledge knowledgerepresentation representationinvolves II1volves th, rd.It,on> that oct ween discrete di"rete aspects .>peel of ofknowledge knowledge(e.g., (e.g.,Barsalou Bar",lou the relations that hold between 1992a). My ~umption, assumption, therefore, is IQ~ld). \ly i> that a lexical I~xical concept c.oncept provides provid.",access ~""'>to toaa sophisticated and and structured body of of non non-linguistic "'l'hi\licated truClured body .hngul tIC knowledge. knowl: th French hench rugby (2C), and and in in (zd) (2d) the fill fifteen players who who make make up up the rugby team team in in (2c), Ihat ofthe the French Fren h electorate electorate who whovoted voted"non" "non"when whenpresented, presented,inin that proportion of rete nt referendum, referendul11, WIth prul"'~ll to endorse endorse aa constitution constllution for for the the a recent with the the proposal IEuropean uro pean Union. In order order to to provide provide these thesedistinct di>!in"interpretations, II1terpretation •this thi Union. In Ie ical concept rve aas an an access a ces site it for for aa cognitive cognitive model model profile profile lexical concept must must serve tha t, al very least, lea t, includes II1dud.", the the sort von of ofinformation infornlatlonindicated IndICatedininFigure !-igure that, at the very 4.2. This 4!. I hi, figure figure represents repre nt an indicat the th sort sort ofofknowledge knowledgethat that an attempt attempt to indicate language u\Cr. hav access ac "'" to when speaking peaking and and thinking thinkingabout aboutFrance. France. users must have In ligure 4.2, Ihe lexical lexical concept (oncept!FRANCE' [I RANG !provide" actC\s to t(1 aa potentially potentially In Figure 4.z, the provides access large number number of knowledge structures. large tructur.",. As each ea h cognitive cognitiv model model consists can i t of a cllmplex and structured stru tured body of ofknowledge knowledge which which provides provides access a "" totoother other complex ,ur" knowlt"'ge, we we can can distinguish di t,ngui h between oct ween cognitive cognitivemodels model which whicharc are sorts IIf of knowledge, .I " etll y all"""'" mod I ,and Iho directly accessed via via the lexical lexical concept: concept: primary cognitive models, and those cognitive models mod I which which form form substructures ub tructure of those tho which which are arcdirectly directly aetC ,,"': secondary model. These rhl..,. secondary "",ondary cognitive mgni"\C models model accessed: secondary cognitiv cognitive models. Me are mdircltly indirectly al'~'\.Cti accessed via via the the IcxkJI lexical (onccpt." concept."
,It'"
I
• Sire ItarsAlkiu 8.u 'nu t(mina) lIN"" ) tot fUf a• review. ft't M"W
17
H , hghhn,lI!.. 11 U inIndetail Jrl.lllinInChapter C "It" I ~ Highlighting is discussed
1 ° 1I mAke the UIK case kl, kwltv thedl-.tln..hun dioinokin lietween t. ~ thehd'ftftnprimary r"nwv And 1J icciindAry cognitive nimbi., in ampler ux
INTRO lll'( 'ION INTRODUCTION
CONSTITUTIONAL CONSTITUTIONAL SYSTEM SYSTEM
NATIONAL NATIONAL SPORTS SPORTS
GEOGRAPHICAL GEOGRAPHICAL LANDMASS LANDMASS
WORD MEANING IN LCCM THEORY WORD MEAN IN(i IN l M TtIEO RY
HEADOF OF HEAD STATE STATE
ELECTORATE ELECTORATE
POUTICAL POLITICAL SYSTEM SYSTEM
CUISINE CUISINE
NATION NATION STATE STATE
HOliD AY HOLIDAY DEST INATION DESTINATION
(FRANCEI [FRANCE]
F,G .... 4." Panial gr aphicallaod m nor cal enlity , aanatio n state, tate, nor to a group of Fifteenrugby rugbypla)'e players whohappe happen hereprrs representing population of "IIeen " who n totobe enllng ththe entire ntir popu lation of Rather, relatestotothat thatportio portion of the voted IFrance. rance. Rathe r, it itrelat n of the French French electorate electorate that that voted against ratification theI·EUcon constitution held in 2oos. "!:din t ratific alion ofofthe litulio n inin aa referendum referendum held on 200s. whati isactiva activated hereiisththe ELF(.T cognitive model. ELECTORATE -\Accordingly, "",.dingly, what led here ORATE cogni ti\' mode l. example provides anelega elegant illustration of the the way in which Thhis • . lalast t exam pl provi des an nt illu tratio n of way in whi h activation of aUlgnl cognitive model serves to provi provide a situated interpretation of alllva llon of. live modd rv to de a .tuate d int rprctJ lion of a lexical concept byg.ving givingririsetoto an an access the semantic • lex"a l mnle pt by a«cs . route rout through throu gh the \eman ti potential. thiseexample, interpretation requires thatan anace access route is poten tial. InInthi ample, interp ret lion requi r that rout i established through the,ognl cognitive modell profil profile accessed via the lexical cone,tahlo hed Ihrou gh the live mode e de< 'oed VIJ th lexil JI",n ceptIf[FRANCE) wayth.J1 that1\iscon consistent withthe theIlexical concepts associatedlcpt R.ANt t)mintl aWoly I lent With xkallom,:c pt .. Mklo1 ted withthe theother otherIongui linguistic formsand andunilS unitsininthe the ullera utterance. The interp interpretation wllh tic fornl n e. The retali on
80 So
IN1RODUCT_~IO~~
______________________________________
INTRODUCTION
~Iated in this thi, example x.mpl has h. to to do dowith withthe theFrench Fren1 A A dOlenroses, ro~.Nina inasent senther hermother! mOl her! (7) (;) a. a.Ditransitive Dilransilivesymbolic ymbolicunit unil Topicaliyation symbolic h. b. Topicali1A1l1on symboli unit unol VP symbolic unit (. VP ymbolic unil c. d. d. NPP symbolic ymboli unit unil Ce.. Indefinite Indefiniledeterminer delerminersymbolic symboli unit unil f.f. Plural Plural symbolic unit unil g. g. dozen, doze". rose, ro$l', Nina, Nilla. send, $l'lId. mother lIIollorr symbolic symbolic units unil Of Of course, course, my my concern concern in in this this hook book isi with with the theway wayin onwhich whi hunits unitsofofsemantic semanli structure—lexical concepts—arecombined combined in in order order 10 to prompt 'Inlcture--Iexical concepls-are prompl for for the Iheconcon struction ,Iruellon of ofsimulations. Imulalions.Nevertheless, evenhel.,LCCM 1.«M Theory Theorytakes lakes from from constructional construcllonal approaches the perspective approaches Ihe perspeCtl·da vehicles .. hid",I Ihave ha'ediscussed di"u\!>CtIthus Ihu tar farhave haveananovert overlphonetic phonelle The realizalion. 1I0wever. ea h of Ihe exampl di -ussed alcnl a lalmY·Slyle analr i .n order 10 illu!>lrale Ihe Tables 6.1 and 6.2 present a Talmy-style analysis in order to illustrate the dislinCl.on in hemalic versu rich Onlenl. The lables are based on Ihe distinction in schematic versus rich content. The tables are based on the example in (2). example in (2). As is evidenl from a compari;on of "[abies 6.• and 6.2. Ihere i, a dcar As is evident from a comparison of Tables 6.1 and 6.2, there is a clear dislinction belw n the nalure f th conlenl associated with closed- "er;u distinction between the nature of the content associated with closed- versus open-clas \'Chicles. While Ihe number of c1osed·c1 vehi les required 10 'oke open-class vehicles. While the number of closed-class vehicles required to evoke Ihc experienlial complex d"'ignaled by (2) are more numerous, Ihey rela le 10 the experiential complex designated by (2) arc more numerous, they relate to lructural a pect of Ihe <eene. and rYe 10 relate differenl aspects of th structural aspects of the scene, and serve to relate different aspects of the COGNITIVE REPRESENTATION:
COGNITIVE REPRESENTATION:
The expenenbal complex evoked The experiential complex evoked by language by language
GRAMMATICAl. SUBSYSTEM
GRAMMATICAL SUBSYSTEM dellneales slructural propert.es delineates structural properties of !he COOllllwe represenlabon
of the cognitive representation
LEXICAl LEXICAL
SUBSYSTEM SUBSYSTEM
prOVIdes provides neh richconlenlful contentful delall detail of of lhe the COOllillVe represenlalion cognitive representation
Cl.OSED-CI.ASS ELEM NTS
CLOSED-CLASS ELEMENTS aSSOClaled Wllh SChemahc associated with schematic conlenl
content
OPEN-CI.ASS OPEN-CLASSELEMENTS ELEMENTS
assoc.aled associated Wllh withneh richconlenl content
he.l'.' 6.2. rht' hifun.•l1IHn In Iht' f"pro. Ion nf Iht' U)~nHI"(, rrprnenl.ttlOn 10 The bifurcation in the expression of the tognitive representation In language
IA, .1 'RE 6.2. I,JOguoige
STRUCTURE __ ----------____________SEMANTIC ~~~~~~~~__~.o~s los
cognitive the .,,!!nill'longui hlllg additionally addilionally Remote Remote Past Pasl from Irom Recent Recenl Past, Pa I. for for instance. in Ian e. The Th lIJllg"ag a nguage with wilh the Ihe most moslparameters param ler thus Ihu far farreported reportedisi,an anAfrican Africanlanguage: language: mileke-1)schaing lia ke () hang with wllh eleven. eI en. Crucially, ruoall. parameters f,arameter> are encoded enu>am ar as of n'Ory mOlor cortex Ihal pnx n ry. motor experienc (Pulvermuller 1999, 1999, Zoo;). 100.\). experience (PulvermUller In contrast, II .rgue In tOnlra,l. Ihal hngui lie content (Onlenl isi\ so \(, highly hlghlyhemalic nalure argue that linguistic schematic ininnature that it is n n.analogu : itII takes lak.. a format formal that Ihal isi not nOI analogous analogou to 10 the Ihe multimuhi . Ihal II I non-analogue: modal experiences Ihallll that it is aa.. schematization of Hence, m.>.this Ihi gives gi,evrise ri • to a qualilati,dy qualitatively very different Iype type of of information 10. ,cry dillerent informal Ionfrom from the Ih kind kond captured caplured by by conceptual conceplual structure. lruclure. To To illustrate, illu Irale, take lak< the Ih parameters param I r> Past Pa I and and Non-past . 'on ' pa I discussed with respect .I, .... " ....'11 with re petl to 10 example example (6) «(0 )above. above.These 'I hevcparameters parameler> are Jrc highly highly "schematic hcmallc abstractions ab;traClion drawn from front the Ih complex compl xrange rang ofoftemporal temporalrelationrelalion . that hold hclwccn between our experience h,p, Ihal our experience experience of ofpast, pa I. and our ourexperi nce of ofnow: now: our our ,ships temporal location as Temporal experi-. Icmfltlrall,,,alion a,. experiencing e perientlng centres tenlr of ofconsciousness. (on IOU nt> ,1cmporalexperi ence,• a • form fom. of ofsubjective ubleeli,c experience, experience. isi,extremely eXlremelyrich richininperceptual perceplualterms I rm encc (Evans 2oo4a). Yet the parameters and Non-past (h an 10(411). Y Ilhe paramelers Past Pa I and 'on ' pa I are are not not rich rich at al all. all, An important consequence n import,]"t lonM."tluc'."nlc of the the observation ob 'r"\'Jtlun that thJI linguistic Iingul\lk content luntent isi non-analogue non' analogue in nature nalure is i the Ihe following. following. II claim daim that Ihal linguistic lingui IIC content COnlenl does .I not give rise, directly, to simulations. By this n01 give ri • directl)'. 10 imulation , Ily.'hi II do not not mean m an that Ihal lingulingu . content cannot Iistic li t: (ontent (d"nnt contribute ulOtnhule to simulations, 1n'IUI.lhnn . for lor instance, in Idnf.:c. as J\ part pdrl of ofan ;jill utterance. "The uttcrancl:. I he meanings—conceptions—which mt.·anin~ ~unu·pllun whi{..h arise n from from utterances utt rJIu,c are ilrc
114 1 14
SEMANTIC STRUCTURE
LEXICAL REPRESENTATION
"
pcCd·classlexical lexical concepts con cpt provide providetopological topological longui,tic content and thus thu closed-class reference. In contrast, ontrast. the the open-class o~n-da lexical I xi,.1 concepts con epts facilitate (a ilitat access a ces toto rckrcnce. In conceptual llCd la lexical conc concepts lexlt.1 pt are also al\O shape-neutral ha~-neutral and .ndsubstance-neutral. ub tan e-neutral.To Toillustrate Iliu trate th examples exampl below, ller number of of different nalure. Referenlialily lakes differcnl forms, form, as a detailed dClailed below. l>elow, However, the defining feature to encode Ilowev ... r. Ihe fcalure is i, that Ihal lexical l... x".1 concepts ctaU\c Oil Seplemher3rd 3 rd 1939 1939 when when he he declared d lared war, war. he he was W the Ihelegally legallyappointed appoinledPrime PrimeMinister 1inisltrofofthe Ihe United Kingdom, and under the terms of the Royal Prerogative—powers Unlled Kingdom. and under Ihe lerm of Ih Royal Prerogativ power; invested in the monarch and deployed by the Prime Minister on behalf the invcsled in Ihe mOllarch alld deployed by Ihe Prime 1inister ollhehalfofoflhe monarch—he take the monarch-hewas W legally legallyentitled enlilledto 101m Ihecountry counlryto10war. war.
•
[MANTIC STRUCTURE
SEMANTIC STRUCTURE
llS 125
I·mally. el'l which Finally,Ihe thelexical lexicalcon concept whichsanctions sanctions Ihe the uuse of of tirdllrrd decidingin in (24 (24c)) relales lMENT Of os AT AT CUSTOMS] U TOM] Ilexical xieal relates10to Ihe the [ANNOUN (ANNOUNCEMENT OF OUTIARLF DUTIABLE G GOODS .oneepl. in terms I rm; of of illocutionary illoculionary point poinl and and context(s) cOlllexl(s) concept. This This iis d,slina distinct bolh both in "I lexical concepts. on<epts. This This lexical lexical concept con<epl isi~ ofuUM: from from Ihe the previou~ly previously melllioned mentioned lexical ,pecialized involving customs cu loms provision provision at al international inlemallonal specializedfor for uuse in in conlexl contexts involving horder ing.s. Its wilh signalling ignalling as a to 10 bordercn) crossings. Itscommunicalive communicativefunaion function has has 10 to do do with good ported by suing Ihe goods hemg being Iran'ported. transported, or or caused caused 10 to he be Iran. transported by Ih the perwn person iissuing the "declaralion" r pecllO reslriction on onthe Ihenature nalure "declaration"ininIhis thisspecific specific conlexl. context, wilh with respect to restrictions and/or good that Ihal may may be he transported Iransported into inlo the Ihe country counlry which which and/or amounl amount of of goods ,.."abli,he. provi,ion. and/or alld/or tax lax payable payable on on particular partl(ular goods. good .. establishes Ihe the CUSlom customs provision, Based u ion. II present presenl in in Table Table 6.4 6.4 aa summary summary of ofthe the Rasedon on th the foregoing foregoing di discussion, ~cy pragmalic poinl Ihal are are encooed a part parI of of the Ihe linguistic lingui Ii key(omponenl~ components of of pragmatic point that encoded as lImlenl each of Ihe three Ihrcc lexical lexi 31 concepls. Mu h of of the Ihe content conlenl associated a;sodaled content of of each of the concepts. Much wilh for drclared from the the con eplual conlenl with Ihelhree the three lexical lexical concepl concepts for declared comes comes from conceptual content 10 whi ·h they Ihey afford afford access. aetes . That Thai is, i •as 3\ lexical lexi al concepts conceplshave havebipartite bipartilestruc.Iru . to which lure. are each each associated assonly ttl< m.. ,.-nc-rl ~ 1'>r\.1 ",duJm. "hrnlolll( information which r particular lexical concept can Is deployed, the mrmm"t",nconcerning "or,,-"TU",the thetypes IYrft014:officio (If ""ntntin In ""hl... h. r--nk.w"r kt.I ... ~1 UU"-rJ't un ~ dc-pluynl. the
I
nature and the "",Iur~of (.fthe theparticipants r~n"'l""nt involved Inu.),,n which wh ... hmust mu thold. hcokl
A\ UllIO. I ASSERTION
I
I ANNO~NCEM£NTOF OFNEW NEW [ANNOUNCEMENT 1 f(~Al STATUS' TATl'S) LEGAL
Unr Iricltd Restricted R tn trd Unrestricted
"hangC'official offidalstate statC' Change
I MlNl OF 01 Dtrt m ITlAS' f I IANN(lllNC ANNOUNCEMENT
R trktrtl Rest ricted
1.Jkrofficial oflluJIstatement 1.. lement Make
1S1
GOOD AT ATCUSTOMS' C\iSTO GOODS
-= ...........:~---
Re"llnUN Restricted
126 '26
LEXI AL REPRESENTATION REPRE FNTATIOS LEXICAL
in (z4b) (24b) gives gives rise ri to to. rich informational informational characterization, characterization. much much in in the thesame me a rich way relating to to France France discussed di usse
7
Summary This chapter has been been concerned concerned with developing an account of of Thi' hapter has ofthe th~ nature naturenf with semanti tructur restricted restricted linguistic to its it use use potential potentia! that that isisspecific peeific to toaagiven given lingui tic knowledge knowledge relating relating to lexical It expresses expr ..... sets set of oftendencies: tendenci :patterns pallern ofofco-occurrence co-occurrence lexical concept. concept. It abstracted ab tracted from usage u~ge events. ,ent. Moreover, 1oroo,er, as a the th lexical lexicalprofile profileisi,apparent apparentinin language use, it language use, it provides provid aa "footprint" "footprint" that that can can serve serv ininidentifying id ntifyingthe the such, we As pcxifi lexkJI (oncept tholt \anltion!t oJ gi,'cn in tJm,:c of u . A \uch, we specific lexical concept that sanctions a given instance of use. might think think of ofthe thelexical lexicalprofile profileas a providing providingaadistinct di"onct"biometric" "blometric"identifier identifier for each lexical concept. parti ularly useful u ful in incases ca of ofpolysemy, poly my,where where foreacnled below. in Iud lechnique from psycholingui Ii Ie long ( ee al. 1997) as well as corpus-based tools and methodoloa al. e.g., Cuyckens e.g., Cuy,ken, ., t997) as w II as corpus-based 1001 and melhodologies (Gries 2006).
gie (Gries 2006). Lexical concepts for time
LexlCol concepts for lime
Before being able to apply the two selectional criteria just introduced, it is first IIefore being .ble 10 .pply Ihe IwO selecllonal crtleria ju I inlroduced, il is fir I necessary to develop a hypothesis as to the nature of the distinct lexical nec~ry 10 develop. hypolhesis. 10 Ihe nalure of Ihe d,slin I lexical concepts involved in particular utterances. That is, how many lexical concepts concepl involved in parlicular uller.nc . Thai is, how many lexi .1 oncepl' are involved across the utterances to be examined? To this end, consider the are involved aero Ih Ulleranc, 10 be examined? To Ihi end. consider the following examples which involve the form time:
following exam pi
which involve Ih form ""''''
(15) Time flies when you're having fun
(t5)
The lime for a deci ion i gelling do;cr
(17)
The old man' lime (
dealh(os fa I approaching
(t8) Time flows on (forever)
(IS)
Time flows on (forever)
These instances of the lexical form time all appear in the "subject" phrase Th""" in'l.nces of Ihe lexical form /lllIt .11 .ppear in the" ubjecl" phr. Moreover, the verb phrase which complements the subject phrase relates to a Morcov r,lh verb phrase which complemenl' Ihe ,ub,ecl phr.se rdal" 10 • motion event. Thus, motion is being ascribed to the entities that time motion event. Thu . molion is being ascribed to the cnlllies that ",,,t contributes in prompting for, in each example. In addition, the semantic (ontrihulC in prompting for. inappears c.l(h example. In addition, the semantiC to he distinct in each example. In time contribution associated with (ontribution a~iJted with appears tim£' appear, to he di\um:t in .u . h example. In to relate to an assessment of temporal the first example in 15), time
Ihe firsl example in (tS),
'm" appears 10 rel.te 10 an a
menl of lemporal
1 43
, I'tATRIXI. siATRIxj.
Indeed. Ihi preliminary anal is uggesl Ihal di linci lexical oncepl Indeed, this preliminary analysis suggests that distinct lexical concepts each of Ih examples (see Evan 2004")· In IInderpon underpin Ihe the usages usages of of time in in each of these examples (see Evans zoo4a). In ",der 10 lesl Ihi hypolh i. I apply the seleclional criteria. For a di linct order to test this hypothesis, I apply the selectional criteria. For a distinct Icm.lprofile (and hen e a di linct I xi al concepl) 10 be confirmed. allea I lexical profile (and hence a distinct lexical concept) to be confirmed, at least une of Ihese IwO crileria mUSI apply. In order 10 confirm whelher Ihe one of these two criteria must apply. In order to confirm whether the in,lances (tS) 10 (tS) inelu ive are sanctioned by di linct lexical time in instances of of '"lIt in (t5) to (i8) inclusive are sanctioned by distinct lexical "lIIccpls.1 begin by applying Ihe Fortnal Seleclional rilerion. To do Ihis.lel' concepts, I begin by applying the Formal Selectional Criterion. To do this, let's (on,ider Ihe kind of noun phr in whi h each u of /I/l/f appears. I Slarl by consider the kind of noun phrase in which each use of time appears. 1 start by n~long Ihallhe exa~ples on (IS) and (tS) appear, on Ihe face ofil.lo be similar. noting that the examples in (15) and (18) appear, on the face of it, to be similar. . ellher I~ pre- modIfied by a delermoner. Ilowev.r. furl her examples reveal Neither is pre-moddied by a determiner. However, further examples reveal thai wh?11 have hypolhesized 10 be a di lin I (OURATlON(lexi I onceplof that what I have hypothesized to be a distinct [DURATION) lexical concept of 'Illoe as as in on (15) (tS) can can be be determined delermined by bythe Ih definite definilearticle arlielewhen whenthe Iheassessment a menl time lit lemporal magnilude i peeifi ralher Ihan generic. while Ih u Ihal I of temporal magnitude is specific rather than generic, while the use that I hypolh iz, I? be san tioned by Ih (MATRIX I Ie i al concepl cannol be. To hypothesize to be sanctioned by the ImArRix1 lexical concept cannot be. To ".'c ,Ihal Ihl IS Ihe case, consider Ihe following in lances of ';/l/t, which are see that this is the case, consider the following instances of time, which are "mllar to 10those Iho;cinin(15) (IS)and and (18): (IS): similar
"'"t
(20)
(17) The old man's time 1= death1 is fast approaching
143
",.,gn,l we mlghl pmvi ionally glo the lexi al con epl which ude. Th.u,. ma gnitude. Thus, we might provisionally gloss the lexical concept which ....,anetions nlllon . Ihl, I,.n e ofof/1111 In (16) Ihe lexical oncepl thisoninstance timeasas(DIJRATION(. 1 ovakrioN1. In (16) the lexical concept ,.llIllloOlng 'mIt mlghl be glossed a ( 10MENT(. Thi follows, Ihe con epsa nctioning time might he glossed as [MOMENT]. This follows as the conceplIun e a a whol relal 10 a peeifi lemporal tion.,\OClaled associatedwith with Ihtheulleran utterance as a whole relates to a specific temporal III11menl when a partICular d i ion is 10 be laken. Thu • th conlrtbulion of moment when a particular decision is to be taken. Thus, the contribution of '11"" ininIhis example appears nOI 10 relale 10 a duralional elap • bUI rather a time this example appears not to relate to a durational elapse, but rather a d,,,-rcle nclion Ihi discreteon5lanl. instant.InIn(t7) (17)Ihe thelexical lexicalconcepl conceptwhich which sanctions this uuse of of "'" time JNlC.rs 10 relale 10 an evenl. whi h extra-lingul .. ic onleXI onform u ppears to relate to an event, which extra-linguistic context informs us Iis a dealh. Thu • Ihe lexi~1 con el'l inv Ived her mighl be glossed as (EVE T(. death. Thus, the lexical concept involved here might be glossed as (EVENT]. al concepl h sanclions 10 IIIIJlly, Finally,Inin(IS),lhe,leXI (18), the lexical conceptwhi which sanctionsIhi thisuse useof of ,,,,,tappear time appears to rdJle 10 an un ndong lemporal etap . In earlier work (Evan 2oo4 a• 2oo 4 b) relate to an unending temporal elapse. In earlier work (Evans woo, 2oo4b) IIde"-ribed Ihi as Ihe "malrix" lexical concepl a ialed wilh tIIIlt. in whieh described this as the "matrix" lexical concept associated with time, in which \\e under land lime 10 be tilt evenl wilhin which all olher even I occur. Thu • we understand time to he the event within which all other events occur. Thus, Ihe gloss w mighl apply 10 d ribe Ih lexi al con epl onvolved here i the gloss we might apply to describe the lexical concept involved here is
(19)
Time flies when you're having fun
(t6) The time for a decision is getting closer
(t6)
U: XICAl ONCEPTS LEXICAL CONCEPTS
-
LEXICAL REPRESENTATION
During Ihe donner dale, Ih lime seemed 10 fly
During the dinner date, the time seemed to fly
'Thetime limeflows flow The
on(forever) (fo of of "me have quite quile distinct distind Thus,,in in subject position, formal formal selectional lcetional tendencies. tendencies. Let's now now lum turn 10 to the semantic selectional tendencies associated l.et' seleclionallendencies a sociated with wilh these Ihl..., uuses IIIlIt. by applying applying the Ihe Semantic manti Selectional Selectional Criterion. Criterion. The The point pOlnl time. I do so by here iis that here thaI the nature nature of of the Ihe motion mOlion event even Iencoded encodedby bythe thelexical Ie icalconcept concept associated with the vehicle is distinct Ie i, distlncl for for each each of ofthe th uses u ininaa J5S()(iated with Ihe verb-phrase verb· phrase vehi Moreover, the the choice choice of mOlion·evenl motion-event type significant way. way. Moreover, typeisi compatible compaliblewith wilh concepts hypothesized to the nature of the various lexical concepl to sanction san tionthe Ihedistinct dislinct time. uses of 'mIt. U\CS For instanc instance,,Ihe the (DURATION' lexicalconcepl concept which which I suggest InnATloN] lexical ugg I underpins underpin the Ihe (IS), and the particular parti ular variant—which variant-which in inprevious previou work work1Irefer refer use of time in (15), COMPRESSION[ lexical concept, as it relates to 10 as a the Ihe ITI MPO.AI COMPR. 10 I lexical concept, as it relales 10 an an [TEMPORAL to lemporal magnitude magnilude which which proceeds proceed,more more"quickly" "quilkly"than Ihan aassessment men I of temporal usual 2oo4a)—co-occurs with lexical u ual (Evans (Evans 2oo4a)-un~ Count [go-ccntred motion, (' Ego-centred mOlion. cannot take take e.g .. Her fltr time isis e.g., artlel un articles,• but can approorhltlg •.• approaching... b. and and the the Semantic Semanllc Selectional Sck'Ctional must rely (:rilcrion. Criterion. Th hallmark of.a h of th= lexical le"i al concepts con«pl. isi that Ihal they they stipulate IIpulatI ..... nction the use me of of flying f/YlIIg in (5a). (sa). stoTtosi] lexical concept. concept, which. which, I suggest, sanctions capable of self-propelled If propell elements of. of a language user's As such, sanction instances instan es of oflangualle and are are conventionally conventionallyassociated associatedwith withaalexkal language and lexical form. Accordingly, form. Accordingly. they are vehicle-specific. vehicle·~pe Jre necessarily ncee Jrily language·'pelatiom manifestations in ,pccifi( specific contelCls contexts of li,e manli ,aa bran h of cognitiv linguistics, lingui ti with with" rkby byBrugman Brugman tive lexical lexical semantics, branch of cognitive work .nd IAlkoff (1988) r on eptuali1ed polysemy polysemy as as being being an an and Lakoff (198$) and and Lakoff Lakoff (1987). (tW), reconceptualized al word~ exhibit exhibil polysemy poly.emy as a aa consequence (onsequ Il( "underlying" Thai is, i words underlying" phenomenon. phenomenon. That nol of. ingle abstract abstract mental mental representation, representcmy. III Ih light light of ofLCCM ICCM Theory. Theory. Accordingly, Accordingly, this this chapter lh.pter can be be i.e. polysemy, in the vit'wed an application application of of some m of ofthe thetheoretical theorelicalmachinery machinery relating relating to to viewed as as an ",m.ntic de,-eloped in in the th last la Itwo twochapters. chaple.... semantic structure developed I'olyscmy, LCCM Theory, rel.les not to the Ihevariation variationevident evident in the Ihe Polysemy, in in LCCM Theory, relates not to situated semanli ,ltu.ted of aa word—which word-whi h arises arises due du to to the thecomposcompo semantic contribution of Ihe book. book. Rather, Rather, polysemy polysemy ininLCCM LC M ilional principles considered con idered later laler in the itional principles consequen« of of. mgle vehicle ..hide being beingassociated a sodatedwith withdistinct dislinlt -ITheory heory i,is •a consequence a single conccp~ which whi hare mantically related. related. Semantic mantic relatedness relaledness isi aa lexi,al concepts lexical are semantically m.tler degree and and isi determined determined by by the thebipartite bipartite structure lructur of oflexical lexical matter of of degree concepts. The way «>neep". way in in which open-class open ·dJ as a well well as a closed-class clo!>ed -cia polysemous poly_ mou ofshared hared or or overlapping overlapping linguistic lingui tic lexie.1 concept be related r laled is i by by virtue virtue of lexical concepts can he in tance in in terms terms of ofshared hared parameters. paramelers. The The second ,""ond way way concerns concern (Ontent. for instance content, for the the conceptual oflhe conNCTO l'R"lle,i al con epl which whi h sanctions nction the Ihe use u of of", in (9) (9) encodes n od aa schematic sehemali dimension dimen ionabstracted ab Ira led from from sensory-motor sensory·molorexperience experi neeinin which the relation which the Ihe Fr is is contained conlained by by the Ihe RO. RO. Notice 'oti ethat thallhe relalion encoded en odedisishighly highly schematic in nature; it says nothing about whether there is contact or hemali in nalure; il says nOlhing aboul whelher Ih re i conlaCI ornot nol between the F and RO as in ( u nor whether the F represents part of the RO bel ween Ihe I' and RO a In (II), nor whelher Ihe I represents part oflhe RO or or not nol as as in in(12): ('2): (11) (II) a.a.The Thflyflyis iininthe Ihejar jar(i.e., (i.e.,flying flyingaround) around) b. The fly is in the jar (i.e., stationary b. Th fly is in the jar (i.e., lalionaryon onone oneinterior Inleriorsurface) urface) (ti) a crack (12)There's There's a crackinInthe Ihevase v~ Indeed, the precise the F,r,RO, Indeed,lhe preci spatio-topological palio·lopologi.1nature nalurcenes onvolvong partial enclosure. In th scenes involving partial In the examples in (14), the Fr is only partially enclosed by exampl {t4).the i only partiallyendosed by the th. bounded boundedlandmark: landmark:only only the of the bulb is enclosed the base b.", of bulb" cnd",,'C.! by by the socket ">eket as ., illustrated ill"'trat I in in Figure Ilgure 8.i 8.1 above; .hove; and only the stem, tem •• nd not not the the whole flower, Jl(lWer. is enclosed endo\Cd by the vase vase (see ("'"Figure hgure8.3); 8 3)' and only the the umbrella umbrella handle handle isi enclosed en losed by bythe thehand hand(see (seeFigure rigure8.4). 8.4).Indeed. Ind~: —
- J'
162
.62
POLYSEMY POLYSEMY
REPRESENTATION I LEXICAL EXI AL REPRESENTATION
.63 163
~\ although aaccess c to its contents con~ents are. are. bs, I",ong be i ng pa~iall~ partiallyendosed enclosed by by the the capcap—although to its Ikn , in a .tualoon whe .. part.al en 10 ure apphes, but locat.on w.lh urety Hence LC ,in a situation where partial enclosure applies, but location with surety ,,(lC' th. (ENCLO uRFI lellical associated with with", [ENCLOSURE] lexical concept concept associated in cannot cannot bc be not, the doe s not, al'l'lol..J rhi reveal that in th ab n e of th Location with ·urety param d. This reveals that in the absence of the Location with Surety paramapp lie etccd aas a core (ore component (omponenl of nf rarallldcr Affecling parameter AffectingCondil.on Umditions 10 to be re-analysed an independ nl lexical lexical con cpl. Clearly 'I r1y aa storm lorm and and wind wind are are much muchless I an independent concept. prn""ypically endo ur • and more saliently 'l3lienlly provide provide prevailing prevailingconditions (ondilion prototypically enclosures, and more "hllh Ihereby con ,,'u,e an envimnmenl which which affects .ffecl us. u. As As such, uth,spatial palial which thereby constitute an environment given rise ri to 10 the Ih funcfuncene involving prololypical enclosures en 10 ur have given scenes involving more prototypical lilln.11 category cal (!lory Affecting 10 the Ihe formation formal ion ofofa a tional Affecting Condilion Conditions,• which which ha has led to Affe ling Conditions Condil.on parameter parameler in in semantic manliememory. 01 mory.The Theexistence e i lenlC .1I,lInel Affecting distinct "' a• distinct di linct [PREVAILING (PREVAII.NG CONDITIONS] co, O.T'ON ) lexical I xical concept, concepl. as as evidenced evidenced by by of e ample in (25), (l5). provides provide suggestive suggellive evidence evidence that Ihal such uch aadistinct di\linct Affecting Affecling examples (nnditum parameter parameterexists. c i\1 . Conditions I argue argue that thai the Ihe distinct di lin I"state" • lale" lexical lexi al concepts concept>associated aialed with wilh in //I evi.denccd in in (17) ('7) to 10 (20) (lO)encode encodethe Iheparameter paramelerAffecting AffeclingConditions, Condilion rather ,ralherthan Ihan denced Ind,,,,d. these Ih lexical lexicalconcepts conccpl are arcwhat whalI Ihave havereferred referredto10asa "state" "we" I nd ure. Indeed, Enclosure. luical (on cpt , . Ih states lal invoked invoked allII provide, provide, ininsome mesense, n affecting • affecling lexical concepts, as the wndilion" Moreover, 10rrovcr, all all these Ihese"state" " lale" lexical lexical concepts conC that Ihal are are non-spatial non· pali.,1inmnature, nalUre.such uch. lal . affecting as states. Ihe development d 'elopmenl of ofnew n lexical lexicalconcepts. con epl . I hi, leads 10 Phis to the The first firs! such uch" laiC" lexical lexical concept con el" relates relale; to 10the Ihephysical phy icalcondition condilionofof The "state" anan urgani,m which whICh thus Ihu provides pmvid an an affecting aff,,,,'ing condition. mndllion .Such Such physical phy ical conditions cond.ti n organism indud good/ill goodJill health, heallh, pregnancy, pregnane),. and and any any salient sallenl physical ph .caI aspect asptl(iJled with wilh ilL II glo,s gloss aas the the (SM:10-INTERPERSONA1. STATE [ Ie in. lexical concept Jassociated Finally, one'. habilual profe;sional activily provide an affecling condition hy Finally, one's habitual professional activity provides an affecting condition by virtue the physical phy ieal and and social SlXial interactions interaction that that are are attendant attendant upon upon such uch virtue of of the activitie\. giving rise ri\e to 10 aa lexical Ie. ic.1 concept con«pl .ictivities.This This provide provides an an affc'the the wall, and and the glue, whi which employs the \\aII, h emplo the wall wall as a a means mean of ofaffixing ~ffixing the the apple, apple, in In the the second. Hence, ><xond. lIence, the example in in (27) (27) applies applies when when there there isI both bothphysical phy ICal contact ontact between the F and the ItO, and when the latter has a role in supporting the former. hn alwoIY' hral the been unJulv unduly Ull'knnaJ concerned With with tM attounts (If of 'mlm linguistic representation. tlC'n...r. Hence, the) they h..-a\ have nut not always ".tI.hlt(\.lu~ (,f the tM conteptual (urkq'hl.aJsystem. tnn However. UUWC'\ r, such Mkh accounts ~1.(JUnl cannot c.tnnut achieve _hlt'W pcschologital f"" hC1I'~.l1 pLitisi rl..ua arthitetture of "lIltt ('nl h'l to 3 an amodal character. Such Such theories theories assume 3 ume that that knowledge knowledge representation representation involves involvesthe themanipulamanipulation ofabstract .b,tract symbols ymbol, which which are arc purely purely propositional pmpo itional in in nature, nature. and and hence hence tllm of are not grounded tate.'4 Recent Recent accounts ",count of ofknowledge knowledgerepresenrepresenarc not grounded in in bodily bodily states. tation tation are arc modal modal or or embodied, embodied. because because they they treat treat knowledge knowledge as as being being grounded in the perceptual experiences and mechanisms that result from the perceptual experiences and mechani ms that result from grounded having the lund kinds of of bodies we, humans, have. Hence, w • as a humans. Hence. conceptual conceptual knowledge knowledge arises from the modal systems that give rise to specific sorts of information. ari'CS from the y tem that give ri to pecific sorts of information. Barsalou (2oo8), in a recent II... alou (2008). recent review, revilw. refers refe" to this thi, perspective JlCr>pective on on knowledge knowledge representation as as grounded grounded cognition cognition (as representation (as II noted noted in in Chapter Chapter 2). 2). Work Work on on the th embodied (or (or grounded) grounded) basis ba i of of cognition cognition in in cognitive cognitivelinguistics, lingui ti •especially especially as a,,,,dated associated with the Lakoff and ., the work work of ofGeorge GeorgeLakoff l.lkoffand and Mark Mark Johnson John"'n( (l.JkofT and Johnson1980. too, 1999; 1999; lakofll987; Lakoff 1987;John Johnson 1987,2007) 2007) has has emphasized emphasized the n 1987. the role role Johnson of sensory-motor experience, and hence perception—the processing of exterof .. nsary· motor experience. and hence perception-the processing of external ,timuli stimuli via via ",n sensory (or modal) n.1i lry (or modal) systems 'y.tem (vision, (vi\ion. audition, audition. olfaction, olf.1(tion. haphap tics, and gustation )—action—which provides motor information relating II". and gu,tatlon)-action-which provides information relating to to bodily states via proprioception—information about movements involving llOdily tates via proprioception-information about movement involving joints and and mu",I< muscles—as ... -a well well as a the vestibular vesllbular system—which yst m- which provides provides inforinforjoint. mation as to position in space and motion trajectories.% m.lIion a po,ition in pace and motion trajectories.' However, other other cognitive cognitive scientists in addition, addition.subjective subjective Howcver. ientists have have argued that, that. in (or introspective) experiences are are ju" just a, as impo"ant important for for grounding grounding cognition. cognition. h" For (ur iotro,,,,,,t "e) «periencc instance, D.m.,io Damasio (1994) (1994) in his of some IINance. hi, review of some of ofhis hi, groundbreaking groundbreaking work work on on emotion has emotion ha emphasized mphasized aa number number of ofcategories categories of offeelings feel inS' that that arise ari from from internal body For in"an instance, body 'tate, states «n1IIOl1\) (emotions) th.t that wc we lahel label as as IOtemal body states. tat"'. I'or ••• body Happiness, Sadness, Anger, Fear, and Oi'S"'t. Disgust, give give rise rise to to phenomenologically Ilapl'ine,,, Sadness. Anger. I·c.r. and See lLAkOti 1141471lur bur.Ia dl.-t..taJc..l detailedHltl'llK' . whine tit of IUl.h slit hdllC'mlluJIhIIh:(nunl disembodied accounts. • \no ... ~ulrI11'lA7) review of the operation 01t~ thettn sensory See Fyans itorthtoming ii) for a niethanitnis responsible t ~ (VOlns Cfl.r1fk: •• mtnI b) I•• rol rn'k"W (.1 the (lper"hclJl (.1 .ry nw...h.ulUln feo.r'ltlJlhk for sense perception. lur n~ rc"n.C'phun
178
LEXICAL REPRESENTATION
real, in the sense of directly experienced feelings. I)amasio identities a further category of feeling, what he terms background feelings, which derive from internal body states. Background feelings arise from, among other things, interoceptive experience, which is to say the visceral sense—our felt sense of the internal organs and other internal bodily states. Other subjective experiences, which arc directly felt, include various aspects of temporal experience which arise from bodily states (circadian rhythms such as the wake-sleep cycle ), as well as perceptual processing, which is subserved by a wide range of neurologically instantiated temporal mechanisms ( see Evans zooagi, 2(x)41) and references therein), and consciousness (Chafe 1994; Grady 1997). Hence, and as we shall see later when I discuss abstract concepts, cognition is grounded in both sensory-motor experience and subjective experience: experience of internal bodily and cognitive states, including emotion, nuxid, and affect. In sum, contemporary accounts of embodied (or grounded) cognition assume that recordings of perceptual states form the basis of the representations that populate the conceptual system. Further, as the conceptual system has, on this view, evolved in order to facilitate perception, as well as situated action (including social interaction), and provides the necessary platform for higher-order cognitive operations such as categorization, inferencing, and conceptualization (Barsalou et aL forthcoming), perceptual states must be recoverable. Many recent accounts of embodied cognition postulate that recorded perceptual states are activated in service of the various functions the conceptual system supports. As already noted earlier in the book, these activations are referred to as simulations (e.g., Barsalou 1999, 2003; Gallese and Lakoff 2005; Kaschak and Glenberg 2000; Prinz 2002; Glenberg and Kaschak 2002; Zwaan 1999, 2004). Simulation represents, on this view, a generalpurpose computation performed by the brain in order to recover bodily states and to perform operations deploying such multimodal states. As we shall see below, multimodal states can be manipulated in simulations in order to provide conceptualizations that are not present in the recorded perceptual states themselves. For instance, phenomena such as analogical counterfactuals are a case in point, as when we say: in France, Bill Clinton would never have been harmed by his affair with Monica Lewinsky." 6 The account of Conceptual Blending Theory (Fauconnier 1997; Fauconnier and 'turner 2002) represents one attempt to show how simulations deploy a range of existing knowledge representations in order to produce novel scenarios, categories, and inferences.
Perceptual Symbol Systems The theory of Perceptual Symbol Systems (PSS Theory), as presented in Barsalou (1999), is a theory of grounded cognition which aims to account for how
CONCEPTUAL STRUCTURE
I
hl
perceptual and other lxxly-based states come to be captured in memory, and to he A vailable for (re)activation as simulations. It is well known from research on attention that during perceptual experience, the cognitive system can focus attention on individual components of the stimulus array. For instance, attention can selectively focus on the colour of an object, filtering out, for instance, its Jive, or texture, and even the surrounding objects (Garner 1974, 1978). The essential insight of PSS Theory is that, through selective attention, i n dividual perceptual components derived from modality-specific experience are recorded, in bottom-up fashion, in sensory-motor areas of the brain.? The c omponents are stored in schematic fashion. This means that it is not individual perceptual states that are stored, but rather commonalities arc abstracted across specific instances of perceptual states providing individual memories (e.g., individual memories for red, hot, and purr). In addition, knowledge is captured from other types of perceptual state, including proprioception (e.g., lift, run) and subjective experience (e.g., compare, similar, hungry). Accordingly, Barsalou uses the term "perceptual" more widely than has traditionally been the case.' These schematic memories Barsalou refers to as perceptual symbols. They are symbols in the sense that, later, in top-down fashion, they can he reactivated, or simulated, and can he used to support the range of symbolic behaviours that subserve a fully functional conceptual system. Perceptual symbols implement a conceptual system as follows. Barsalou argues that memories of similar and related components become organized into a system of perceptual symbols which exhibit coherence. This perceptual symbol system he refers to as a frame. A frame is an information structure consisting of large collections of perceptual symbols, encoding information which is stable over time as well as incorporating variability. Hence, a frame provides a unified, and hence coherent, representation of a particular entity. For instance, a frame involves numerous components that have a perceptual basis, that are related in various ways. In addition, the perceptual symbols that collectively comprise the frame can be combined in a range of ways, giving rise to an infinite variety of simulations. Hence, a system of perceptual symbols gives rise to both a frame: a relatively stable knowledge matrix and dynamic simulations. 9 Together, the frame and simulations arc referred to as a simulator (a term I first introduced in Chapter 4).
' There is compelling neuropsycholivical and neumimaging evidence which supports the view that human conciptual representations arc grounded in the modalities, and hence are perceptual in nature. For instance. categorical knowledge is grounded in sensory-motor regions of the brain (for reviews we Datitasi ► Joy; e,atnuni earl. 199s; Pulvermullet 1493 loo3). 1)ainagje to a particular sensory motor regh in serves to impair the processing of categories that use the region in question to perceive physical -
exemplars. 1 will hem (-forth use "perkeptual" in Ikitsaloti's more ink !mixt. sense, while 6
• Sit Evans and Green (2006: ch. 12) fora discussion of this atialogiLal . ► untertaktual tram the perspective 01 t Antteptu.ti Blending Theory.
1 79
At the outset of the 4. Ihipler 6
I kthiled examples Or frames are prOlo' 'Jed III the nett
pointing
the reada to
theprovisulnd
180
-
LEXICAL REPRESENTATION
Properties symbols Properties of perceptual symbols Perceptual ymbol are ar~ characterized charact~rized by by six six key kqproperties: pro~rti~: Perceptual symbols perceptual symbols • ~rceptual ymbol are ar neural neural representations r pr ntations in in the thebrain's brain'
Il'o('ryySeth%
motor areas areas
• • • • •
perceptual symbol symbols are ~rceptual ar schematic ·hematic perceptual symbols are dynamic. dynamic, not rigid ~r on. on.In Inturn, turn,the Ih domain domain shape, itits component component parts, Retail ihr div.ussion in h.iptrr
18 ,1
C
N EPTUALSTRUCTURE CONCEPTUAL STRUCTURE
LEXICAL REPRESENTATION LEXICAL _ _ _ _ _ _ _ _ _ _ _ _ __ __ REPRES£NTA~T.:: ' O~N
tSS
185
\ Illore (omple example of profile/ba organi/.3hOn i. iliuM rated by Ihe A more complex example of profile/base organization is illustrated by the ,d.llional Gramrelationallexical lexicalconcept conceptassociated associatedwith withthe the form form ""clf. uncle.InInCognitive Cognitive Gramnl"'. tru turt aSMXiatoo .n enlily with a mar,Ih< thevemanlic semantic structure associated with with Imclf uncleprofiJ profiles an entity with a ,,"11),1« hi includes Ihe following following abstract ab lract dodooulip iexdomain domain matrix. matrix.1This includes at at lca\1 least the
which, in turn is HAND understoodwith withrerespect the domain domain ARM. which. llANO i isunderstood peet totothe in turn i understood with the r pcct to the domain BODY. Thi domain i under tood. understowihpctedomainy.Ths underto, ultimately, with respect to the domain of SPACE. In this way, the relationship ultimately. wIth r""ped to the domain of '"AU. In thi way. the rdation hlp between domains reflectsmeronymic meronymic(part·wh (part-whole) relations,wi with one doI ) relation. th one do between domain reflect main being part of a larger more inclusive domain. The most inclusive main being part of a larger more in lu ive domain. The mo t in lu lVe domains are what Langacker refers to as basic domains. Basic domains are domain\ are what langaLker refer to a basic do main . 8a ic do n1.li", arc directly grounded embodied expcrien experience, and thus directly grounded inin embodied e. and thus have have aa pre-conceptual pre'Lonl eptual basis, as illustrated in Table 9.1. ba i • a illustrated in Table 9·1.Langacker argues that the semantic structure we l>dW saw in in Chapter AAs we Chapter 2.,2, langacker argues that the scmanti trulturr conventionally associated with a given vehicle, such as knuckle, consists of conventionally iated with a gIVen vehi Ie. uch '" kllIlCk/f. con i't of profile/base organization. The profile profile for for knuckle. knuckle, for for in instance, what is profile/base organization. The tance. what i\ desde ignated, consists of the various joints in one's fingers, thumb, or hand. ignated. on i t of the variou Joints III one\ fingers. Ihumb. (lr hand . human hand hand contains AA human contains fourteen fourteen knuckles. knu kI . The The base base constitutes con ti t utes aa larger larger structure, within the domain matrix, which is essential for understanding \lruClure. within Ih domain matrix. which i essential for u nderstanding what th the knu knuckle Put another .nother way, way. the the conventional conventional semantic ",manh. whal kle designates. dC\ignates. Put consists of a substructure knuckle representation associated with the form representation associated with the form klll/ckit consi t of a ub~truLlure (the profile) of a larger conceptual structure (the base), within a domain (the profile) of a larger conceptual trueture (the base). wilhin a domain matrix (a series of hierarchically linked domains of knowledge). langacker matrix (a series of hierarchICally linked domain of knowledge). L.1 ngalker suggests that eviden evidence that the the base to the uss ts that e that base relates relates 10 Ih hand, hand. rather rather than than some '>Ome other structure, e.g., the arm, comes from examples such as the following: other SIrU ture. e.g.• the arm. com from examples such a th following: ARM,
I11JIIl"": CifNEALOGY, PERSON, GENDER, lXl,;AL I TERCOURSE, BIRTH, LIFE mains: GENEALOGY, PERSON, GENDER, SEXUAL INTERCOURSE, BIRTH, LIFE Lc\'ycLE, I •• I'ARE. T/C:III1D RELATIONSHIP, Rll.ATIOS IIIP, SIBLING SI8L1NC, RELATIONSHIP, RIIATIOS'illlP, EGO. f(iO. The PARENT/CHILD
The
~.,,< baselor forIhe the~manlic semantic repr"",ntation representation for for Imclf uncleisisdrawn drawnfrom fromaero acrossaanumber number "Iofthe" domains to provide the conceived network of FAMILIAL RELATION these domains to provide the conceived network of FAMILIAL RELATIONS
Il'),fi
UNCLE
UNCLE
auditory canal; detects motion and balance) auditory f.:anal; dctat\ motion lind ~I.lm.:r) Visual system
Vi ual y I m Auditory system
AuditOry y t
"'
Tactile (touch) system Tactile (touth ) y.,tclll Pressure sensors in the skin, muscles and joints Pre ,ure n\.Or In the kin. I1lU I .Inti Joint Detection of tissue damage by nerves under the skin [)elation of w. ur ,J.Jlnas by n n. under the kin Olfactory (smell) system Olf.dic situations ,ituation and and generic generi situations. situation.The Thedistinction diMinctionbebetween episodic and generic situations is orthogonal to the distinction between t,,«n episodi and generi ituation i orthogonal to the distinction between he t() this s. individuals ""I" Idual and types. is approach, According \L alues are are therefore ;ubordinate concepts concepts within an attribute taxonomy: subordinate concepts, or values, which ", thm allrlbute taxonomy: ubordinate concept, or values, whichare are more specific, inherit properties from the superordinate concepts, or attrimore pecific, Inherit prope"i from the superordinate concepts, or attri butes, whllh which are hute" arc more more general. general. In addition, addition, attributes within a frame In attribute within fram can can be be associated so,discussed di 'U' dabove. above .. 1hat i arises That fram are are embedded embedded within withonlarger largerframes. fram .Take Take the theframe frame CAR, CAR, disc d ,,,,u, i!d frames above. A salient attribu\< a sodated with thl' t)'P from from the the phenomenon phenomenon III Another transcenden e. Thi natur of cognlllve models mooels ffllr transcendence. This relat relatesto to the the situated situated nature of cognitive types. Recall Recall that that cognitive cognitive models models of ofthis thi kind kindare are things: individuals individual and things: and types. "llxated" , ituation . In In other other words, words, cognitive cogmllve models model, for forthings thlllS'arc arc "located" in in situations. located in in the the world world model mooel at at the thepoints point atatwhich whichthey theyare areencountered. ncountered. located lienee, cognitive cognitive models mod I for for episodic episodicand andgeneric senerksituations iluationsinclude include.'epre Hence, repr scntalillns for ondividual and types. The greater the number of Ituallo", III serrations for individuals and types. The greater the number of situations to which individual and andtypes typesare arclinked, linked,the th greater great rtheir theirtranscendence transcendenceisi held held which individuals to be. be. Hence, Ilence, transcendence Iran 'endence is i, aa function function of ofhow howinterconnected interconnectedcognitive cogniti,e to models lor things thins' are arc with with those tho for for situations, situalion , and and hence hence the the events even" with with models for which they are are connected. connected. which they due to the the componential componentialnature na ture of 01 Another for chaining chaining arises ari Another molivallon motivation for due to the onenenc as co-occurring, the greater the strength of the association area.% J\ lO-O\:e,:urnng. the greater the \trenglh ofth .1\\OC.:ialion area.'
Inte raction between the linguistic linguistic and and conceptual conceptual systems systems Interaction between the IInow IUrn 10 a (on\iderJtion uf W.ly in whi,h Iingui\lu; and .lnd concep«m~cp which the linguistic now turn to a consideration of the way tual ystem interact. per pective of of LCCM LCCM Theory, Theory. this thi. concerns concern> tual systems interact. !rom From the perspective
205 20S
Sec ftarsalou 1 al. (forthcoming) for discussion of a rclated proposal. Sce also Horoditsks Prinz (forthcoming).
1
416
-
LEXICAL REPRESENTATION L RfPRESENTATION
IE I
I xic.I concept~. by being a~iated with ace it. fadhtate a c ,to largesmnticpo.Ide,nfthmairso eptan large semantic potential. Indeed. on of th~ main rea~n for the p.rot •• n n nature of word meanings, the starting point for the present enquiry (i nature of word meanin~. the start 109 pomt for the present enqUiry (10 Chapter1),isduolgbyfcneptualkowdg,h"tenial Chapter I). i due to the large body of con eptual knowledge. the "potential" which they facilitate access to.
which facihtate . chapter, the To they illustrate, let'sacee brieflyto. consider an example from an earlier To illustrate. let', brieny con ider an example from To an do earher chapter. the so, reconsider the lexical concept I RFD) associated with the vehicle red. lex teal concept IREO I a sa we might list Royal
O(.NITIVL 10DELS COGNITIVE MODELS
r.
Primary versus secondary cognitive models models Primary versus secondary cognitive The rang of agnitive model th semantl potential-to whi h a Ie i al The range of cognitive models—the semantic potential—to which a lexical con ept facilitat a c I refer to as it cognitive model profil~ term I fir t concept facilitates access I refer to as its cognitive model profile—a term I first introduced in hapter 4 we saw earher in thi chapter. in di u ing the introduced in Chapter 4. As we saw earlier in this chapter, in discussing the nature of haining within the con eptual y.tem. th cognitive model profile nature of chaining within the conceptual system, the cognitive model profile I not an unstr~ tur~ inventory of knowledge: conc ptual tructur i highly is not an unstructured inventory of knowledge: conceptual structure is highly ,tructured. I dlStlngu, h between two a pect of a lexical concept' cognitive structured. I distinguish between two aspects of a lexical concept's cognitive model profile: th primary cognitive model profile. and the secondary cog' model profile: the primary cognitive model profile, and the secondary cognltlve model modelprofile—terms profile-t rrn also alsointroduced introdu dininChapter Chapter 4 nitive 4. ,The primary cognitive model profile consists of all tho cognitive models The primary cognitive model profile consists of all those cognitive models with which a le,ical con ept i directly a socialed: the association areas which with which a lexical concept is directly associated: the association areas which make. up its a c~ ite. Hen .e~ the primary ognitive model profile may make up its access site. Hence, the primary cognitive model profile may con t,tute many d,screte cogOltlve model an ace ite may be made constitute many discrete cognitive models—as an access site may be made lip of many di tlllct a ',allon area ispersed aero variou region up of many distinct association areas—dispersed across various regions w,thlll the cO.n. eptual ystem! The cognitive model which mak up the within the conceptual system.' The cognitive models which make up the primary cogOltlve model profile I refer to a primary cognitive models. as primary cognitive model profile I refer to as primary cognitive models, as wealso al~sawwinIIIChapter Chapter4.4 we In contrast. th secondary cognitive model profile consi t. of all Iho In contrast, the secondary cognitive model profile consists of all those cognitive ".'odel what I refer to • ondary cogOltive models--with cognitive models—what I refer to as secondary cognitive models—with whICh a lexlC.1 cone pt i< not 3S'>(luated. Ilence. secondary (agnitive model which a lexical concept is not associated. Hence, secondary cognitive models do not ompri part of the alle ite of a cognitive mod I. Put another way. do not comprise part of the access site of a cognitive model. Put another way, • Soft ill'pclially domm ant. uch governance, whereby one participant is especially dominant, such as in a ledur c. univel"iilylecture. university
nl whllh M:rve 10
clcm o n tituen telements exhib it constituent abo exhibit a liviti also jomlactivities addil ion. joint In addition, which serve to In ing: follow Ih d inclu Th re. truclu wilh themwith structure. These include the following: prtlvide them provide
who carry
more parllClpanl IWOorormore involv two actlVllyinvolves )Oinlactivity PaNlopam"AAjoint participants who carry • • Participants: Ihe partic ipant con i t • . purch hop In. . lance m fllr ly. dcti" For instance, in a shop purchase, the oullh out the activity. participants consist lanl. i a ... ,I< e andlh cu;to merand Ihecustomer ofthe of the sales assistant.
uum (Aft('f' ( !.Irk 1W6: 'SOl . ~~~~ 4.:0m erwtu lO. n('v. 1'3rer ;)rude r.dlO r('POrl • ,lO Idcph Telephone conversations, newspaper artikles, radio reports,
umlln (After Clark 1996: 50). dl (IUne'continuum "l"h('discourse I "at fiLl. TABLE ILI. The --~~--~ .
ilngul'lI \1u~tlylinguistic Mostly •
•• •
so on.
andso on. and
f ~4.: to f.,l,u' (elO\ rwtill " tcl("V1 Ie')n rcryrIlItIllCu rol activityroles. joll1tactivity theirjoint about their ,ump lions about ledge and aassumptions ding know knowledge II1du including activity. j(11l11activity, point ininaajoint givm point any given At any a tivity: At jomt activity: laic of Current of the joint nt state •• Curre knowto knowrdat to Thi relates activity. This Ihe activity. of the tate of current state th current represmt the lpant represent participants partl al. tageititisi at. whatstage andwhat progr ingand a llvityisi progressing theactivity how the rning how concerning ledge conce that tv nl that variou events th various repr nt the Iso represent Partl ipanl also far: Participants 'ents so tar: Public events • Publi activity. joint activity. the joint realizing the in realizing pIa e in thu far taken place have thus
4
Imagll1 and imagine ev nt. and purcha event, hop purchase the shop recon ider the let' reconsider illustration. let's By way way of illustration, the Atthe boots.At ofboots. pair of buyaapair tobuy order to in order h p in hot shop enters a shoe cu 10m r enters Ihe customer that Ihal the enquire ordertotoenquire ustomer. ininorder th customer, approach the a i tant approaches _hopassistant th shop moment moment the cu lome r thecustomer both the pur hase. both h r purchase, effecting her in effecting a i tance in requir assistance sh requires whether whether she knowledge involv knowledge Thi involves knowledge.This ofknowledge. bodyof large body hold aa large a i tant hold ..1 assistant and sales about andabout footw ar.and purcha ofoffootwear, makingaapurchase inmaking involved in procedur involved Ihe procedures about aboui the a um assl tantassumes I assistant Ihesales in tance.the I-orinstance, a umpt ion.For II1duding assumptions. oth r. including each each other, the andthe footwear.and itemsofoffootwear, oritems item or an item purchase an 10purchase wishes to cu tome r wishes the customer that Ihal the a I Ian e off< rlo m~r indicating Ihem r trying cu lometrying th customer andthe bool and felchingboots feet.fetching ,~tomer' feet, thecustomer's measuring them meas u"?g the for paymentfor pri e.payment onprice, confirmingon required.confirming boot required, theboots onthe decl Ionon on, makll1gaadecision on.making the:.e _ Fa hofof tran-'JclionEach Ih transaction. do,in gthe andclosing pURha •and thepurchase, wrappingthe the these boot •wrapping !heboots, hicrar,hi .11 (on lItut a hierarchical and constitutes way,and rangeoforways, proct tdinina..range canproceed joint athon scan JO int actions groun d comm onground th common lal ofofthe curre nt state Thecurrent above. The d rihtd above. sequence, ~uen e. as described participant whileparticipants ,,:ached.while htcnreached, hasbeen \lag has whichstage a totowhich kn()wledge as II1dud knowledge includes
226
226
LEXICAL CONCEPT SELECTION
SEMANTIC COMPOSITIONALITY
•
AalYllyroles rolN: In In aa joint jOintactivity, .ctivny.each ~achparticipant partk'panttakes tak on onparticular partkul •o Activity publi roles. rol . These Th determine d~t~rmin how howeach ~ hparticipant parti ipantproceeds proc~ininservice rvt' public offacilitating facih~ahng the th~ joint jointactivity. actIVity. For I'orinstance, in tJnc~. inina acommercial commer"alevent~~ of tranSOlctlOn such uch as aa purchase purcha in 10 aJ shop, hop.one oneparticipant partiCipantassumes a um the the transaction rol~ of ofseller, lIer. the th~ other otherofofcustomer. eu tom r. role Publicgoals: go,./" These Th are r~ the the mutually mutually known known goals goal which whichresult r ultfrom fro mthe th •o Public joint activity, a tlVlty. such uth as effecting tTeetingaJpurchase. purch.;e. • joint Pri,·atrgoals: goals: In In addition, addition. participants participant in in aajoint jointactivity aClivitymay mayharbou ha rbourr •o Private prlvat goals. whl h Jre unknown to th oth r participants( ). privategols,whcunktoehrpaicns(). /01111 actions: aaiollS: 'flint /Oint activities achviti are are comprised comprised of, of.and andadvance advanc~through, through.joint JOIOt •o Joint achon . These These are a .. the Ihe.discrete di rct~ action action components component that tha tmake makeup upaajoint joint actions. 10 tance. in 10 making makIOg aa purchase purch in in aashop, hop.the thesales of .ffair and relation hip. in unique W.y>. WhIle ea h langu·1I unique states of affairs and relationships, in unique ways. While each language ha, a range of "ready-made" schem ymbohc UOl" whICh cal~ be com has a range of "ready-made" schemas--symbolic units which can he comhined to facilitate coordination in ; int actions-these necessanly underbined to facilitate coordination in joint actions—these necessarily underdctemline the mutability of human experience. As lang.eker put it. determine the mutability of human experience. As Langacker puts it, .. Linguisti convention cannot provid~ a fi"~. unitary :x~r '. ion for 'ery 8 "Linguistic convention cannot provide a fixed, unitary expression for every (onceivable situation that. peaker mIght WI h to descnbe (IbId. 1987: 17 ). conceivable situation that a speaker might wish to describe" (ibid. 1987: 278). \, ( lark argues. in order to overcome this. language use involv.es employing As Clark argues, in order to overcome this, language use involves employing the conventional repertoire of ymbolic unit in non- onventlona! way ~n the conventional repertoire of symbolic units in non-conventional ways in order to overtom coordination problem (see also Croft 1000). That I • order to overcome coordination problems (see also Croft i000). That is, words do not have table and fixed semantic representation whi h \urfa e words do not have stable and fixed semantic representations which surface each time they are used. Rather. word exhibit mantic variation. a function each time they are used. Rather, words exhibit semantic variation, a function rvi e of ofjoint jointactions. acti ns. . ,f situated ,ituated language language use u in in service I rom the perspective nf lCCM Theory. the protean nature of semanll From the perspective of LCCM Theory, the protean nature of semantic represent.ti n i • in part. a fun tion of (i) selecting the appropnate lelll al representation is, in part, a function of (i) selecting the appropriate lexical (oncept. and (ii) deriving a cont xt- pecific reading ~ue to the pro< .. of concept, and (ii) deriving a context-specific reading due to the processes of lexical concept integration and ",terpretation; a leXICal concept faClhtate lexical concept integration and interpretation; as lexical concepts facilitate access to the con eptual y tern. and hen e po a vast semanti potential. access to the conceptual system, and hence possess a vast semantic potential, thi, potential mu t be narrowed in service of the formation of a conception. this potential must be narrowed in service of the formation of a conception.
Joint actions JOintactivities actions proceed by virtue of the joint actions which make them up, as we Joint JOInt a(tiviti proceed by virtue of the Jt,int .etion which make them up. a., we have seen. The hallmark of joint actions is that they require coordination between have secn. hallmark of joint i, that they requIre (oordination bet,,< 'n the two (or The more) participants. Inaction this they are participatory: they involve two or (or more)who participants. In thisa part they.re participatory: they involv or the two more participants each perform in order to achieve the joint action. moreactions parti"p.nt who ealh perform p.rt in order tn achIeve the joint action. Joint can he contrasted with a autonomous actions which are nonjoint actions can be contrasted with autonomou action whichbyarc onn participatory, such as playing a flute solo. Joint actions progress virtue •. uch ofasactions playingbetween a flute participants solo. joint actIon by virtu ofparticipatory the coordination in orderprogr to overcome a of the coordination of actions between particIpants in order to overcome coordination problem. A coordination problem arises when two or more people a coordination problem. and/or A (oordination problem Jri hewhen two orbymore I"',,'ple have common interests goals which can only achieved virtue of
,,,0
have common interests and/or goals which can only be achieved by virtu of
See the related notions 411 the situation model (van Ihik and Kintsch KAI) and mental model • ~ IN- reLunt nuUnn, 01int~ n)....ki h.n PII" and "'1111 h l.,al)in.00 R)('nl.&J m".leI lohnson-Laird IstSti developed the111u.lhun psycholinguistit: literature, and Addressed slightly more detail in (:hapter t It"'mun IA,n)n. dndnrcd in Ih.:- I"'".hohntt\.l~h' hltnlIUfC'... nJ okWrnk'\! an wltf,hlly muK
,,*,,,
delolll in (hotrt«
t\.
Factors in in selection selection Factors In lhl tion I provide a brief overview of some of the main fa tors in In this section I provide a brief overview of some of the main factors in \election. Selection. like the other compo itional pro< in LCCM Theory. selection. Selection, like the other compositional processes in LCCM Theory, i, guided by context. In order to identify o;om of the main contextual factors is guided by context. In order to identify some of the main contextual factors di>cu ion follow: involved.I divide I dividethe thediscussion involved, as afollows: • fa tor\ .s> Original Original Vodka. Vodka. The The presumably aa pork pie goc goes nicely humour play in part, on the the selection I ·tion for for the the 'LOW (LOW TEMPERATURE' TEMPFRATl'RF) humour alo;o also plays, in part, Ie,icdl evokin g the Ihe (POSITIVE )po ITIVI EVALUATION' 'VAlUATlO") lexical lexi al concept. (On cpt. lexical conlepl, concept, while evoking I hat iis,, in far as the utterance uttcranct is amusing, this thi is i so because bee.u itIt evokes .,ok one one That in so far lexical on epl-advanlagenu to 10 the th addressee's addr • girlfriend—only girlfriend--ilnly to torevise revi itil lexical concept—advantageous WIth notadvantageous—thereby advdntageou Ih rebyrevealing rev linga a"wicked "wickedside". ~id ': with another-which another—which isi not
Linguistic context context linguIstic can Ihonk think of of linguistic as involving We can lingui Ik context conlexl. involvingthree Ihreelevels. level. The Thefirst fir Ilevel le\ ·1 isI that of the Ih utterance, utt rance. the Ih primary concern (tlncern of ofLCCM I.C( 1 Theory. Above this th,s is" the Ihe Ihal discourse level, which involves an arrangement of more than one, typically arrangement more than on • Iyplcally di our level. which onvolv many, utteranc utterances.. Above the di-,,:ouN discourse level many. bove Ihe lev I is the Ihe speech pcceleclion (If the lexical lexicalway. concept facilitated FORTHRIGHT ASSERTION] Ihe (fORTHRIGHT INFORMATIONAL ASSERTloNjle"ical can epl i fa( ihlalc-d by linguistic context. This follows as "my genius" is a property being ascribed by linguislic conI namely xl. Thi Oscar followsWilde, as "myby geniu .. i Wilde. a property hcing ascribc'(\ to an individual, Oscar Accordingly, this 10 an individual. namely OSCIr Wilde. by a ar Wilde. Accordingly. counts as an informational assertion, and therefore guides the selectionIhl of counts a an infonnalional assertion. and Iher fore guides Ihe selection the (FORTHRIGHT INFORMATIONAL ASSERTION J lexical concept associatedof Ihe (IORTIIRIGIIT INfOR tATIONAL ERTIONj lexi al concepl with declare. However, both linguisticASand extra-linguistic contextassociated serve to hngu"ti, and extra hngui. li( (onlexl \CrveAT10 wilh dedart. lIowever. bolh select for the lexical concept: (ANNOUNCEMENT OF DUTIABLE GOODS selecl for In Iheterms lexical con epl: context, (ANNOUNCl tENT Of DUTIABLE GOO'" AI CUSTOMS!. of linguistic the (ANNOUNCEMENT OF DUTIABLE CUSTOMS(. In lenll ofhnguisti ,onlexl. Ihe (ANNOUN(:£ tENT Of DUTIABLf I have GOODS AT CUSTOMS( lexical concept collocates with the expression (;()()l" AT (.U'TOM,j le"iwl concepl (ollocal~ wilh Ihe expr~ ion I nothing to, which forms part of the lexical profile associated with this lexical not/,,"g In 10,addition, which forms part of the lexical proftla US a socialed \exical concept. the extra-linguistic context, customs wilh post, Ihi, serves to e'lralingui lie conleXI. a U cu loms po. I. rv~ concepl. In add ilion, Ihe facilitate selection of this lexical concept. The humour that derives from this10 faeihlale sel«lion Ihi'onlexical conlep\. Theofhumour Ihal lexical derin>eleclion, seleclion for short. Selection involves the identification of con the lexical conceptsorassocifor with hurt.each Scll~tion involvc~ the idcntifit..Jtion of the Icxit:al COl1(('pt ated vehicle in a given utterance. Selection is thus one of J~\()ll the aled wilh each vehidecentral in a given ult ranconstruction e. Seleclioninb LCCM Ihu~ one or compositional processes to meaning Theory.Ihe u)mll() itiono1t pnx
c,c\
c('ntrat to
01
aning constru tion in LC M Th('()ry·
t
llX1CAL (:ON( 1 vT SU I ( "TION LEXICAL CONCEPT SELECTION
135 235
Indeed. Ihe oulpul of selection, whi h I 10 . While thi 1",,1 of integratl(ln count linguistic or grammatical information, this doesn't account for the rich level ofof tic or grammatical mformatlon, thi d n't account for the rich level Iingui multimodal information which I refer totoasa\conceptual content. ((lncCl'lUaJcontent. multml,ldal infon nallo n which I refer ' Recall the discussion of the distinction between nominal and relational lexical
•
k omrpts
in
hvlet
.. ,luptn notion of domains WA.. discussed in Chapten 3 and Ho. See also h..i os and • Lanitacker's I I• dUc:lbWd In n"'rlCTI J .n..t IQ U ..n ,.1 do.m.J'f • noCt ' I...anc;u. I006: 4. h. -1 tor an introductory livers sew. ,'V('ntn W Inln .... 7) 1uI.an dt the ( 1006 • Recall discussion oflokl the..ry ditransilive construction in Chapter 3.
• RtuIlItw- d","usunn tlllh.: dltr..wflVC' wndru ctJon
In
Uurtn ,.
6 mm
-
239 LEXICAL CONCEPT INTEGRATION
239
thi ,appe al coun t for dopa edi aaccount erg hahasananenencyclopaedic ,ker, (;oldb ,\AsWIth for this, appealwithwngJ Langacker, Goldberg ordin g to Gold mant i' . Ac. n of , 191\5) or" (191\1 1I1g ingtotorillm Fillmore's (1982, 1985)notio notion offram frame semantics. According to Goldtructl on I relallYlzed argum ent con rb that berg, berg,each eachvverb thatfu fuseswith with aa verb· verb-argument construction is relativized conce ptual knowledge a body fram mant ic frame—a rich semantic peet toto aa rich '\lth withrrespect body of of conceptual knowledge varlou yerb in qu lion. latro with ptual detail g totothe rdatin relating theperce perceptual detailsaassociated with th the various verbs in question. dear how thi I ,I of nt, it iti isnot ker's accou ver, a aswith IInwc However, withlanga Langacker's account, not clear how this level of level, gram matic al level, or grammatical lingui ti or WIth a ntatio n inten ledg repr knnw knowledge representation interfaces with the the linguistic the perceptually rich trultu r from by structure am 01 are and andwhat whatth the mech mechanisms are where whereby from the perceptually rich tructu r . gram mati al structures. fused grammatical with fused orpor atro with ti fram ,eman semantic frames bcxom becomes in incorporated ,oldb erg, th mode l langa ker and oped hy count devel ro Tobe hefair fairtototh theaaccounts developed by Langacker and ( ;oldbcrg, the models the d tail of mant I( emro with rily eon I ped wer a h deY theye they each developed were not not prima primarily concerned with the details of semantic isrti by all mplin g to d . lop a rily exerc w re prima they were Rather, they o ition. Rather, cllmp composition. primarily exercised by attempting to develop a (a "gram tructu re(a and structure organ izatIo n and gui tie organization nt oflin ntieally ba'ird 'Cllla semantically based accou account of linguistic "gramey, and the comb ina tituen con uch u i for nt accou could which , mJr) mar"), which could account for issues such as constituency, and the combinacan th n be Theo ry can LC M Theory guag .LCCM oflan a peet of formal aspects rli of tllrtal torial prope properties of the the formal language. then be co; pro pcctiv per ch rescar the Illing lem comp ,a ct re:.pc certam in d, in certain respects, as complementing the research perspectives "CWe viewed, proLee on. iTAtl organ al matic gram of count a onal tructi con u'h constructional accounts of grammatical organization. LCCM1 by such ,ided vided by itiv Con tructi on Gram ,ramm ar and itiye Grammar s from I heory differ Theory differs from Cogn Cognitive and Cogn Cognitive Construction Grammant i repr ntatio n tureo thena ith the pr isclyw nlern ed precisely that ititlSlo Illar with nature offsemantic representation is concerned mar in in that given it found ation al Moreover, given o ition.Moreover, mant I( comp hani of and and th the 01 mechanics of semantic composition. its foundational con titute di tru ture constitute con cptua l structure and conceptual struc ture and mant i structure that semantic ~umpllon that Jassumption disof di tinlt proces two it po I that follow it n, ntallo repr of kind tinct tinct kinds of representation, it follows that I posit two distinct processes of fu. ion of lingO! IIc to fusion relates to which relates integ ration , which conc pt integration, lexical concept comp o ition: lexical composition: of linguistic conte nt. l eptua con of ion fu rn conce h whi on, retati interp and (lInte nt, and interpretation, which concerns fusion of conceptual content. content,
Fusio n Fusion an ov rview of fu ion, and the
proyi d an overview of fusion, and the section isi totoprovide thi section in this purpo in My purpose My peeitic interp r tallonasa~specific andinterpretation integ rationand con cptintegration I xical concept oflexical rol of r peetive roles respective rema inder of the chapter, theremainder prOWt'I nt u.nlrol 11m I.u""rrc ... n I i . u>n 1t1",," t>f m1rrrrrt.lbun. d .... u'OlC'\lm lhenut ,twrtt'T of power or controltotkepillIM i1 4 olnUrqUerke 1111 "PrriAllun. m the next ch.tilic-r. II for detailed dttatlN presentation rrnmLUlon of o(arguments ulummhfor lurthe thtindependent uk.l~mlexistence "XI I~~ theditransilore dllrallNliVC-Lim"on ig For pitofthe 11995. otrlk.lln hon, see Goldberg 1ft' lioklbrrt flt9't).
(17) a. vehicle vehicle (17) a. b. b. lexical lexical concept concept
"DEli Ill, NP, beFINITE bdl ITE INDEFINITE-NP" INDlH ITE-NP" "DEFINITE-NP, [ATTRIBUTION IATTRIBUTION OF OF A AQUALITY QUALITY TO TO AN AN ENTITY] ENTITY)
In lerm of of deriving deroving a conception conceplion for for (16), (16), [FRANCE] IfRAN('EI is integrated iOlegraled with wilh the Ihe In terms lexical lexi .1 concept: concept: IUNIQUE PE IflEO THING], THINGI, which which is associated a ialed with wilh the the UNIQUE SPECIFIED vehi I provided (18a) rather ralher than than the Ihe entire enlir [ATTRIBUTION IATTRIBUTION OF OF AAQUALITY QUALITY vehicle provided in in OW TO AN ENTITY] TO I NTilYI lexicdl in (17b). (17b). lexical c"neepl concept in (18) a. vehicle vehi Ie (18) a.
b. icxildl concept uuu:ept h. lexical
"DIll ITI' P" "DEFINITE-NP"
I
[UNIQUE (l INIQl'I SPECIFIED \PFUIIl 0 1THINC] II INc.t
:.25~0:..._-';.: "::..':;.A:;.N,-,TI SEMANTIC 2510
OM POSITIO.:::N:::A:,:l;:.IT:,.Y= -_ _ _ _ _ _ _ _ _ _ _ __ COMPOSITIONALITY
LEXI AL CON EPT INTEGRATION LEXICAL CONCEPT
integration of of IFMANC,I FRANCE] wilh It virlOe of of the Ihe Inlegrallon Ihe IspEciFiE I P£(:IFI>O .,) with the ❑ '1111 is by virtue It i, rIIINGI part of Ihe "predicate "prodic.le nominalive" lexicalco COn. lexl al concepl h forms part lexical concept whi which of the nominative lexical n_ ceps, that receivesilits status Theme or Subject cepl, Ihal ]FRANCE] I.MAN FI receives latu aas Them Subjecl of ofaapredicatin prodicallngg (THING] lexical concepts that al concepl Ihal form form exprsion.Putahwy,js e expressIon. PUI anolher way, jusl as Ih ITlIlNGllexi part the larger larger "ditransitive" "ditran ilive" lexical Ie ical concept concept in in(5b) (Sb)encode encodedifferential differenlial part of of the conlenl,ensuring en uringthat Ihaleach ea h[THING] ITIIING] slot 101 iis di linct, so too, 100, by by virtue virtue lingul lie content, linguistic distinct, of II RANCEI occupying P slot 101 it il does does in in (16), (16), itilachieves a hieves aadistinct dlsllncl occupying Ihe the NP [FRANCE] vi -a-vis the Ihe semantic semanticvalue valueachieved achievodby bythe thelexical lexicalconcept concepi semanlic semantic value, vis-à-vis IGfOGRAPlIlCAL UGION] in the same same utterance. ullerance. (GEOGRAPHICAL REGION] ow let's lei' consider con ider aa slightly lighlly more more complex compl..example exampleofofexternal eXlemalle .. cal Now lexical concerns the integration inlegralion of of an an internally inlernally open open concepl inlegralion. Thi concept integration. This concerns lexi al con ept which. which, via internal inlernal lexical lexical concept con epl integration, inlegralion, has ha becom becomee lexical concept "dilran ilive" lexical lexical concept concepl interalycosd.Thxmpe inlernally closed. Th example involves Ihe the "ditransitive" given in (5b) (Sb) above. abov . To ider the Ihe following following utterance: ulleran e: To illuslrale, illustrate, con consider
distinguished. e1islingul hed. Th Thee first, fir I, internal inlernal lexical leXICal concept con epl integration, inlegrallon, applies appli to 10 Inlernally open lexical concepts. con pts, while while the the second, ond, external external lexical lexical concept con cpl internally open lexical ofan an internally inlernally closed closed lexical lexi al concept. on epl. Inlegralion, involves involves the Ihe integration inlegralion of integration, inOuenced by by the the lexical lexical profile profile of oflhe lexi alprofile profileassociated associalod Inlegralion is influenced Integration the lexical inlernally open and closed closed lexical lexical concepts. con epl .Both Bothsorts sortsofofintegration inlegralionare are with internally governed by Principles of of Lexical Lexical Concept Concepl Integration, Inlegralion, of ofwhich whi hthree Ihreewere were governed oncepl integration inlegralioninvolves involv inteinle presenlod in the Ihe chapter. hapler. Finally, Finally, as a lexical concept presented of linguistic lingui lie content, conlenl, itil isi primarily primarily concerned con ernod with wilh utterance ullerancecontext. conlext. gralion gration of
(19) John gave Marya abracelet braceleland andhehegave gaveJane Janeaanecklace. neeldace. (19) John gave Mary In xampl two IWO "ditransitive" "duran iti,e" lexical concepts concepts arc are being being coordinated coordinaledby by In Ihi this example ONJUNCTION WITH ZZ BY BY ADDITION OF OF Y] Y] lexical ept associated associatod with wilh the [[CONJUNCTION lexical con concept the vehicle atilt. and: (20) (2o)
a. vehi Ie "aud" a. vehicle "and" b. lexical Ie .. al concept concepl[CONJUNCTION [.The Theoutoul come, then, of language understanding, involves the activation of non-linguistic com .Ihen. oflanguage under landing. involv the actIValion of non -lingui lie representations, which arise due to represenlation which arise du 10 prompts prompls of ofthe Ih kind kindprovided providedbybylexical lexical concept selection and integration, as described in the previous two chapters. coneepl seleclion and inlegrall(lIl. as de",ribt.-d in Ihe previous Iwo chaplers. Before I reiterate Befor proceeding proceedingwith wilhthe Ih description d riplionofofinterpretation inlerpr lalionbelow, below. I reilerale the following. Once lexical concept integration has taken place, each Ihe following. Once le.ic.ll concepl inlegration ha laken place. eachlexical lexical concept receives a semantic value as part of an integrated lexical conceptual concept rC\:civt..~ J \emanlic v.lluc 0'1\ p.ut of .In inlcgratl-d lexilJI (om:eptu.ll unit. Ullll. Interpretation InlerpretJIIOn proceeds proceed by bysubjecting subjecllllgeach eachopen-class open -da lexical lexiealconcept conccpl
in in aagiven gi\'cnlexical Ie kJIconceptual (om:cptual unit unittotothe theoperation operationknown knownas amatching: matching: the cognitive model profiles of two (or more) open-class lexical concepts Ihe TilING WITII AAPARTICULAR PARTI( nAR ATTRIBUTE] ATTRIBI'Tf) lexical concon cept associated MODIFIER NOUN the vehicle vehide DETERMINER DFll RMI I R MOl>IIIl-R OU (i.e., (i,e., an an cept a ""iated with the NP), associated 1'), prior prior to to being being integrated integrated with the th more more complex om 1'1 lexical I ",cal concept con .ptaiated with the predicate predicale nominative nominative vehicle vehide which whi hcomprises compri the theentire entireutterance. utterance. The relevant lexical concepts for interpretation in (3) are !FRANCE], BEAUTI1rdl"ant lexical wn«l1t; for interpn:tallon 10 (3) arc (FRA '( 1). (81AU1 FUL I, and [COUNTRY ]. 2 The principle fULl. (COUNTRy).l Th principle in in (P4) (P4) ensures, ensUfC>,then, then.that thatI(BEAIlTIFI!L) BEAUTIFUL] and IcauwritY] are subject «-OI'NTRY) ub)CO forth. forth. In In contrast, contra t. the th lexical I "ical or culture, and so NTRY) facilitates, fa ilitat • at very least, lea t. access a{{ to cognitive cognitive models model onccpt (co at the very to (COUNTRY] concept conterning what what itit means mean to 10he be. A110, STATE, TA 1[, and aa relating to knowledge concerning a NATION model relating relating to to a• GEOGRAPHICAL L[()(;RAPIII(;AL AREA. AHA. cognitive model matching process proc pl()(ced as follows. follow A A search "h isi established abli hed in in the tbe The matching proceeds {(lgniti,. model profiles pmfiles accessed acee \cd by by the th lexical lexicJIconcepts concep"subject IIb)C h. ?John, the nation
b. ?John.
th~
nation
by the The utterances in (6) are not straightforwardly interpretable, signalled b Th ulleranc an (6) or not tralghtforwardlyant rpretable. ignallrd question marks, precisely because an informational characterization Iis not qu tion marks. pr iscly!>«au an informational characterizatIon possible for (LANDMASS) or(NATION when matched with the primary ry pcl Ibl for llANO IA s] or INATIONI when matchrd with th~ prima cognitive model profile associated with the lexical concept (INDIVIDUAL I cognItive model profil ass!l(iated wIth the I xical concept I,NIlIvml'A NAMED JOHN!. This is due to a clash in the primary cognitive model profiles NA 1m 101lNI. Thi i due to a cia h in the primary cognitive model profil one associated with the [INDIVIDUAL NAMED JOHN) lexical concept on the lin a ",xlatl'd WIth th I,NOIVIIll'AI AMIII IOIlN llexia l concept on the hand and (LANDMASS( and (NATION) on the other. This is not to say, of hand and ILANO"A ] and INATloNI on th other. Thi i, not to >Jr, course, that matching is impossible, thereby avoiding a clash. However, clash cia h course. that matching i impo ible. thereby avoiding a clash. Howev~r. resolution requires recourse to what in Chapter to 1 referred to as the level Of dill resolution rcquor rClOU"" to what an Chapter 10 I rcferrrd to th I level is the hallmark of figurasecondary cognitive models: activation at this ligura Sentaasummary, summary.ininFigure J-igure13.7, ').7. of 01 meaning meamngconstruction !
(BooR1 IBOOK)
INTEREST MEREST
READING READING ACTIVITY ACTMTY
Aftribules
Cognitive mod/ill
Lexical concept
.on,epts. nitive nuxlel, IU.l at 13.m. 1\.10 The Thcrelationship rel.lllOn hlrbetween bchH'C1l It" u. ..1I..:um . ept .L..:ognlll\C model and Jnd attri.Jun
FIGURE
butes
hutt"
,
-
INTfRPRETA nON INTERPRETATION
271 271
~IREADING ADING AerlVITY are related related by respectively.The Thetwo twocognilive cognitive models models are by virtue virtue ACTIVITY re:.~lVely. "f tru tural invariant-who inleracts with wilh the the physical phy icalartefact invariant—who interacts artefact READER—the structural of aa READE~-Ihe bv relation holds holds between between cognitive cognitive virtueof of reading reading Ihe the prinled printed text. text. Thi This relation by virtue n;"dels Chapter to. 10. II capture caplur the Ihe structural Slructural models and/or and/or ,"ributes, attributes, as as di«:msed discussed in in Chapter the specific pecifi relation relation IIlvariant invariantin in Figure Figure 1).10 13.10by byaa double-headed double-headed arrow, arrow, and and the IIwolved addition, cognilive cognitive models models READER.In In addilion, involved iis signalled signalled by by the the mnemoni mnemonic RPADrR. of a large, delalled, bUI lructured, body of knowledge. Figure 1).10 cun i I consist of a large, detailed, but structured, body of knowledge. Figure 13.10 provides BOOK) provides provides providestwo twoallribules attributesfor foreach eachof ofthe thecognitive cognitivemodel modelswhich whichIIsom] J(C ICAl STRuerURE Ihe physical physical relates 10 to the PHYSICAL STRUCTURE relat The cognitive cognitive model model PIIY access 10. to. The Jrtefa t, con i ling of, very least, least, knowledge knowkdge as as to 10 the Ihe physical physical structure Iru ture artefact, consisting of, al at th the very .nd n book. tailed knowledge knowledge concerning concerning and organizalion organization of of aa giv given book. This This indud includes d detailed Ihe peets of weighl, binding binding the malerial material aaspects of the the artefact, artefact, including including itits dimensions. dimensions, weight, forth. This This aspect a ped of ofour our knowledge knowledge about aboul books booksIIrefer refer (pa per or (paper or cloth), cloth), and and w so forth. 10 Ihe physical phy i aI organization organizalion and and construccon trueattribute.. In In addilion addition 10 to the TOME atlribul to aas Ihe the TOMh books consist consisl of oftext lex I which which isi interacted inleracted with wilh through through the Iheprocess proc lion tion of of aa book, hook, books of refer to Ihe TEXT attribute.. -rEx -r atlribut reading. Thi This II refer to aas the of reading. Th~ READING ACTIVITY cognitive model relales to 10 the the process process involved involved in on The READING ACTIVITY cognitive model relates onleracting nature of Ihe interaction interaction with wilh the the text lext interacting with with books, hooks, especially especially the the nature of the liseif. ofthis Ihi interaction inleraction isi that Ihat reading readingtakes lakes up up, period of of itself. One One consequence consequence of a period lime, which II refer refer to 10 as as the Ihe DURATION DURATION allribute. depending on on the the attribute. That That iis, depending time, which .mounl oftext text involved, involved, reading reading can can take take lesser I r or or greater greater amounts amount of oftime. time. amount of Another e of interaction with wilh books books isI the the level level of ofinterest interest that that aa Another consequen consequence of interaction given book holds holds for for the the reader. read r. This Thi II refer refer to to as as the the LEVEL LEVEL OF INTEREST OF INTEREST given hook .nribut . That while the the reader reader might might judge judg the the book book to tobe beinteresting, interesling, attribute. That iis,, while another might might be be judged judged to to be be boring, boring, and and so so on. on. another ow let's return to the i ue of highlighting. Ea h of ofthe the utterances ullerances in in (8) (8) Now let's return to the issue of highlighting. Each and (9) (9) involves involves aadistinct di lin tinformational informalional characterization characterization for for the the (nom] IBooKI and lex I I concept. con ept. This Thi isisachieved achievedby byvirtue virtueofof ea hinstance on tan of e of IBooKII being beong lexical each (BOOK con i tent with with the theutterance ullerancecontext contextsuch suchthat that, lightly interpreted in in aa way way consistent interpreted a slightly through the thecognitive cognitivemodel modelprofile profile ditTerent access a c route is i established established through different route ",,,-.sed via via IIDooK). accessed BOOK I. For instance, the conceptions conceptionsthat that result r uh from from (8) (8)have havetotodo dowith withprimary primary For instance, the actIVation of ofthe the PHYSICAL PIIYSICAl STRUCTURE TRuerURF cognitive cognitive model. mndel.However, lIowever,each ea h conconactivation ception involves Inmlv differential ditTerenlial activation activalionofofattributes allribultsassociated a -ialedwith withthis thi cogcog ception nitive model—the model-the process proc ofhighlighting. highlighting. While Whilethe theinformational informational nitive of IBOOKI inin(8a) (8a)involves involveshighlighting highlightingofofthe the chara terizalion aassociated iated with characterization with (Book' rOMI attribute, allribul ,Ihe informati nalcharacterization characterilalionassociated a sociatedwith wilhIBooK IBooKI the informational I inin TOME Ihe TEXT TEXT attribute. allribute. (8b) involves involv highlighting highlightingofofthe (8b) In contrast, contra t, the the conceptions conception that Ihatresult r ult from from the theutterances ullerancesinin(9) (9)have hav to10 In dowith withprimary primaryactivation a livallon of ofthe the READING R.AOIN.; EVENT fV'NT cognitive cognitivemodel modelaccessed ac cssed do Theinformational informationalcharacterization characterizalionassociated a ialedwith withI Board IBOOK)inon via (Book]. IBooK).The via (9a) results resuh from from highlighting highlighllngof ofthe the DURATION IlURATION attribute. anribule. The Theinformational informational (9a) lharallerilJlion associated ''''''''Ialedwith Wllh!BOOK] InooKI in in (9b) (9b)results r "Ihsfrom fromhighlighting highlightingthe Ihe characterization LEVEL OF OPINTEREST INTf.Mf~Tattribute. .ltlribule. LEVEL
272 2
27
INTI RPRf.TATI()S INTERPRETATION
SEMANTIC EMANT IC COMPOSITIONALITY OMPOS'~'.!.T!! ' O:!!N~A~L:!'.!.T!.Y_ _ _ _ _ __ _ _ _ _ _ _ _
of matching matching Types of Types now turn turn to to aa consideration of the of matching. II now con ideration of th process proc of match inS. Matching Matching Iis tthe central m mechanism in interpretation, (cntral hani\m in interpretation, and Jnd involves involves simultaneous \imultancou primary primary activationof of cognitive cognitive models a{tiv.tion model, accessed accessed via, via. or or arising ari ing from, from. distinct distinct lexical lexical concepts andlor and/or lexical lexical con conceptual units in in ord order concept eptual unit r to to produce produ e aa complex compl x informational characterization. informational characterintlon. Matchingtakes takestwo twodistinct distinct ~forms. This Iis aa consequence of the rms. Thi con;equen of the broad broad Matching distinction which holds between nominal versus relational lexical concepts distinction which hold between nominal ver u relational lexical concep" 6, and and hence. hence, th the sort sorts of of cognitive first discussed in (Chapter fir t di 'u;;cd in hapter 6, cognitive model model profiles prolik." that, in broad terms, these two types of lexical concepts facilitate access that. in broad t rms. th two type oflexical onceph fa ilit.te access to. to. Put Put another way. way,not not only only iis there th re aa distinction distinction in in the the nature nature of ofthe thelinguistic lingui tic another content that (To) EXcontent that nominal, nominal. e.g., e.g.• (ExpLostoNI, (>XPIO ION(. versus versu relational, relational. e.g., e.g.• (bo) IX lexical concepts encode, but aas each PLODE] lexical PLOOl) onccpt en ode. but ea h lexical lexical concept con cpt has ha aa unique unique access site, the cognitive it • the cognitive models models accessed accessed via via each each lexical lexi al concept concept will will be acc slightly distinct. With respect to matching, the claim in I.CCM Theory slightly di tin t. With respect to matching. the claim in teeM Theory isi, that the the distinctive distinctive nature of these classes of lexical concept—and da oflexical concept-and hence hencetheir their that nature ofthese associated cognitive model profiles—entails a differential contribution to the a iated cognitive model profiles-entail a differential contribution to the construction of a complex informational characterization—the interpretation con tructlon of a complex informational characterilatlon- the interpretatinn that ari arises from from integration integration of of a lexical conceptual that a lexical onceptual unit—due unit-IANTMlEi ADVANTAGE/ WEU·BEING WELL-BEING
173 273
PROVISION PROVISION Of OF PlEASURE PLEASURE
(0000) [GOOD] F,GUR. 1).11. Parli.1 profile for for (Goon) Goool cognitive modd model profile Partial cogni"" FIGURE 13.11.
proc profil accessed accessed via via [Goon) (Gooo) isi of ofaa process as as itit appli applies to to thtcognitive the cognitive model model profile distin t kind. illustrate con ider the partial cognitive model profile profile for for distinct kind. To To illustrate consider the partial cognitive model (1Goonl ;OOIl( as a, presented presentl'tl ininFigure l'igure13.11. 1).11. The complex informational characterization associated associated with with "good "good nun", man". The complex informational characterization for tance. might relate to notion such uch as as physical physi al beauty, beauty. honour, honour. being being for in instance, might relate to notions morally on. depending depending upon upon morally up'tanding. upstanding, providing providing for for one", one's famIly. family, and and entente.InInLCCM LeeM Theory, scmanti semanticality relatestto well· well-formed utterances, which which iis to Theory. ality relates formed utterances. to say "'Y those that tho,"" that give g.ve rise "'" to to conceptions. concept.on. Utterances tterant that that fail fail in in this thi regard n.-gard arc arc semantically anomalous.. The semanticallyanomalou The reason rea n for f< r failing failing to to achieve a hievcsemanticality semanticalityisi due due e.ther failur to conform to to the th Principles Prin il'l of ofLe i al Concept Concept Integration Integra tion either to a failure Lexical (principles (prindpl • pi-p3), p. 1'3). or or aa failure failure to to conform confoml to tothe thePrinciples Prinupl ofofInterpretation InterpretatIOn (principles 1'41"0). pa-plo), or (prindpl oraa failure failur to to conform conf< rm to toboth. both.In Inother otherwords, words.semantimanti · cality failure CIllity f.ilu re is is aa consequence con\Cquence of offailure failure totosuccessfully u c fullyundergo undergo fusion, fu ion. thereby resulting resulting m in aa string of \vehicles, but no conception. thereby tring of h.dev. but tonception. To con consider this phenomenon phenomenon consider 10 id r thi. omidersome someexamples examplesinvolving involvingthe th verbal verbal began: vvehicle hicl btgmr. (13) ('3)
a. He a. He began began the th book book b. 111e ?He began began th the dictionary dictionary c. 1111e ??He began began the t. th rock rock
On the the face fa e of of it, it. while while the the first fir t example example evidences eviden aa semantically semanti allywell-formed well· formed conception, the conception. the utterance utterante in in (1;h) (.)b) isi, not not well-formed. well fnrmed.This 'Ihi follows follnw as a dictiondi. The FRANCE]. Hence, a search occurs in the primary cognitive model profile for arch occur Ihe cognitive model profil for I FRAN E). The and Ordered Principles ofConceplUal of Concept sal Coherence and PrinCIplerdered.Search arch serve serve to 10 ensure ensure aa match malch FRANCE'. for (8) in the th primary primary cognitive cognitive model model profile profileof ofIIrRAN EI· In terms lerm of of primary primary activation aClivalion in in (8), the the Principle Principle of ofConceptual Conceplua)CoheroherNDM ASS ence ensures enee en ure, that Ihal the Ih GEOGRAPHICAL GWGRAPIIICAL LA LANDMA S cognitive cognilive model model for fo r primary aactivation. That iis,,lhi this cognitive [FRANCE] receives IFRANCll receives primary IIvallon. Thai cogniliv model model matches mal(he, the landscape". This Ihe informational inforn,alionalcharacterization charaCierizalionassociated associaledwith with"beautiful "beaUlifullandscape". This follows follow as "' there Ihere is i, aa clash d"h between bel ween the Ihe informational informalionalcharacterization characlerizalionand andthe Ihe FRANCO: other olh r cognitive cognillve models model in in the Iheprimary primary cognitive cognilive model modelprofile profilefor for [IrRAN' I): NATION STATE, STATE, and HOLIDAY 1I0LlOAY DESTINATION. Ol$TlNATION. Hence, Hen (', the the conception con eplion which whi h arises for .ri for (8) is i literal hleral as activation aelivalion occurs occurs solely \olely in in the Ih primary primary cognitive cognolive model mod Iprofile. profil • all the In contrast, conlra\l, in in (9) (9) there Iher is i a clash cia h between bel"cen .11 th cognitive cognilive models model in in the lhe primary promary cognitive coglllllYe model model profile profile associated a "xialed with wilh I(,FRANCE] "AN' , ) and the Ihe informainfornlJ tional lional characterization characlcril.lion associated a solrated the Ihe distinction dislin(lion between belweenliteral hleral and andfigurative figuraliveconcepconcepHaving JU justI illustrated lion, now provide provide aa sketch kel h of of the the meaning-construction meaning-con IruCiion processes proce Ihal give give tions, II now that rise melaphoric conceptions. conceplion,. IIfirst fiN of ofall all consider con ider metaphoric melaphoric conceptions con by application application of ofth PrinCIpleof ofConceptual Conceptual volves matching which isis guided by the Principle Coherence ensurr a match match isi achieved. achieved. The ThePrinciple Prin ipl ofofOrdered OrderedSearch Search Coherence to ensure en ures that made to match match in in the th primary primarycognitive cognitivemodel modd ensures that anempts attempts are are made profiles profile!>. Due to to profiles before before proceeding proceeding to to the secondary cognitive model profiles. the ed Clash Cia h Resolution, Resolutjon. given that that the the communicommuni . the Principle Principle of Context-indu Context-induced cative a ribe some quality quality to to the the student's student' grades, grades. itit isis the the cative intention intention is to ascribe lexical ept [WENT IWFNT up) a the the figurative figurative vehicle vehicleinmthe the lexical con concept uP1which which i\is designated as hen e which which undergoes und rgoes the thesearch search operopermeanmg truction process, proc • and hence meaning con construction ation ils secondary secondary cognitive cognitive model model profile. profile. A A partial partial cognitive cognitive model modd ation in its went up profile for wrrl/ lip iis provided below below in inFigure rigure14.5. 14.5. As w wesee. see,Iw ENT up) UP) affords a As WENT relating to a physical ph icaJ entity entit y access to knowledge relating motion. and and the the motion motionisisdirected directedagainst ag.rinst gravity gravity on o nthe the that capable of motion, that iis capable verti aI axi reprrsent at three of th primary primary cognitive cogniti"e models models vertical axis.. These These represent at least least three of the
BENEFITS DUE TO INCREASE
AND METONYMY METONYMY METAPHOR AND
~.::!.ETAPHOR
295 195
which IWfNT upl affords afford aaccess. cess. Therr are, additionally, additIonally. a small mall number number There are, [WENT UPI which of secondary cognitive cognitive models modelsillustrated iUustrated ininFigure Figure 14.5. '4.5. The first first relates relat to to of secondary IN REA P IN ases in in everyday everyday interaction inleraction with with our ourenvirenvir· INCREASE !N QUANTITY. QUANTITY. In many cases onthe thevertical v rtllalaxis axi correlates correlat with withan anincrease mcrease unment.being bemglocated locatedfurther furtherup upon onment, m quantity—for quantity-for instance, instaner.the thehigher higherthe thelevel I 'eIof ofwater waterin inaaglass, gla the themore morrthere th rr in i\. In In each of ofthese thesecases, cases, an anincrease increase ininheight heightcorrelates correlateswith withan anincrease increaseinin is. cognitive model modelconcerns concernsthe thebenefits benefitsthat Ihatnaturnatur· quantity. A further secondary cognitive aUy of greater quantity. pile of oforanges orange!> ally accrue accrue by virtue of quantity. For instaner. instance, a higher pile correlates WIth which correlates correlates with with more morefood foodand andthus thu greater greater with more oranges, which glassrelates relat upportunltyfor for nourishment. nouri hmen!.The Thegreater greaterthe theamount amountofofliquid opportunity liquid inIIIaaglass receive refreshment, refreshment, and and so SOforth. forth.In Inaddition, addition,there therrisi also alsoa a to greater ability to receive of IMPROVfM[NT h derives derive from fmm an increase in in \C(ondary cognitive model of IMPROVEMENT whi which secondary Improvement relates relates to to aa change change evaluated evaluated as a positive, positive.in inthis thi instance mstance quantity. Improvement incre~ in i.~., an amount at one one point point in in time timemeasured measured an increase in amount, amount, over time, i.e., against an increased in 'reased amount at at aa later later point. point. against Cia h resolution achieved by virtue th secondary cognitive model of of Clash resolution is achieved virtue of the Improv ment achieving a hieving primary primary activation. This Thi provides provides aa match matchbetween between improvement DES' and the informational informational characterization chara tcrildlionassociated associatedwith withSTUDENT'S IsTunfNT's GRA (iRAIlES) the secondary cognitive ognitlve profile pmfil to 10which which[WENT IWlNT uP] upl affords afford access. ace .This Thi the secondary example provid figuralive con eption. as it involves involv clash da h resolution in in aa provides a figurative conception, «l to literal Iiter.1 meaning meaning construction. construction.In of figurative as hetween J a metonymic di~'inctiol1 mclunymi,.; conception lOIlt.:cplion and and metaphoric mct~lphori( conceptions (om:cption\ distinction bctwl'Cn consider the example in discussed earlier, let's let' conSIder in (4h). (4b).
METAPHOR AND METONYMY
FIGURATIVE LANGUAGE AND THOUGHT
296
297
The Ham Sandwich Has Asked for the Bill
Metaphor versus metonymy
In an utterance of thi% kind the relevant elements that afford access to conceptual content are the lexical concept (HAM SANDWICH{ and the lexical concepts (ASK Foal and !ma As (ASK FOR] and (mil form a simpler lexical conceptual unit than the entire utterance, by virtue of the principles o integration, these lexical concepts undergo interpretation, giving rise to an informational characterization. The lexical concept (HAM SANDWICH ] then undergoes interpretation in conjunction with the informational characterization "asked for the bill". However, there is a clash between the informational characterization, and the primary cognitive model profile of (HAM SANDwicii I. After all, a ham sandwich is not, normally, conceived of as an animate entity that can ask for the hill. Due to the Principle of Context-induced Clash Resolution, the customer who ordered the ham sandwich is identified as the figurative target, and the ham sandwich is identified as the figurative vehicle. Accordingly, it is the cognitive model profile associated with the lexical concept (HAM SANDWICH] which becomes the site for clash resolution. Following the Principle of Ordered Search, the search region for clash resolution is expanded to take in secondary cognitive models associated with (HAM SANDWICH]. A partial cognitive model profile for (ham sandwichl is provided in Figure 14.6. In this example, clash resolution is achieved by virtue of a search occurring in the secondary cognitive model profile of (HAM SANDWICH]. The cognitive model which achieves primary activation is that Of RESTAURANT CUSTOMER.
As observed earlier, it has often been pointed out that metonymy, but not metaphor, has a referential function—one entity serves to stand for, or identify, another, as in a "ham sandwich" serving to identify the particular customer who ordered the ham sandwich. In contrast, previous scholars have variously argued that metaphor serves to frame a particular target in terms of novel categories, e.g., My job is a jail (e.g., Glucksberg and Keysar 1990; Carston 2o02), or analogy, e.g., Juliet is the sun (e.g., (,entner et Al., zool). That is, metaphor has what we might very loosely refer to as a predicative function.' From the perspective of LCCM Theory the distinction between metaphor and metonymy relates to whether the figurative target and figurative vehicle exhibit alignment, and hence whether the clash resolution site corresponds to the figurative target. To illustrate, let's reconsider the canonical metaphoric "docile boss" interpretation of My boss is a pussycat. In this example the figurative target is [BOSS) and the figurative vehicle is PUSSYCAT I. Following the Principle of Context-induced Clash Resolution, the cognitive model profile for ( PUSSYCATJ, the figurative vehicle, is the clash resolution site: primary activation of a secondary cognitive model takes place here. This situation differs with respect to metonymy. In the "ham sandwich" example. the "customer" corresix)nds to the figurative target, as determined by the Principle of Context-induced Clash Resolution, and the figurative vehicle corresponds to the "ham sandwich". However, both contextually salient elements are accessed via the cognitive model profile associated with a single lexical concept: HAM SANDWICH I. In other words, there is alignment, in a single cognitive model profile of the figurative target and vehicle. Hence, the site of clash resolution corresponds to the access mute for the figurative target: "customer". In sum, LCCM Theory reveals a divergence in metaphor and metonymy, which emerges as an outcome of the application of regular meaning construction mechanisms. Figurative conceptions which are labelled as "metonymic" arise due to the figurative vehicle facilitating direct access to the figurative target due to alignment of the figurative vehicle and target in the same lexical concept and cognitive model profile. In contrast, "metaphoric" conceptions arise due to a divergence between figurative vehicles and targets across two distinct lexical concepts. Based on this discussion, we see that the "bossy pussycat" interpretation of My boss is a pussycat discussed earlier is metaphor-like, in the sense that there is nonalignment of the figurative target and vehicle. After all, in that interpretation, the lexical concept [Rossi is the figurative vehicle and hence the site of clash resolution, while (PussvcATI is the figurative target. Yet, the "bossy pussycat" interpretation doesn't intuitively feel metaphoric. While this interpretation does
COW-AM VI
SHOP CUSTOMER
CAFE/RESTAURANT CUSTOMER
SHOP
CAFE/RESTAURANT
INGREDIENTS
VENUE
[HAM SANDWICH]
APPEARANCE/ COMPOSIT ION
inctaphors").01"thr type singled b 1-ikott ti ■ %lc that the LIV■ Of 4.TOSS-d1:10141t1 on.sppings -con. tilt I And (0471 appear to be ota different kind than thi ► ,e Anil II tlinvon, as exemplified by the ci,milvk s in ( 'winner. %Imbed by scholars such as Canton, t Am -
i4.t
I4.h
Partial cognitive model profile for
i HAM SA/440%61'14M I
29$
METAPHOR AND METONYMY
FIGURATIVE LANGUAGE AND THOUGHT
constitute a figurative conception, given the way tigurativity is operationalized LCCM Theory, involving, as it is dots, primary activation of a seconda cognitive model in One of the cognitive model profiles undergoing matchin the interpretation is somewhat atypical, from the perspective of the canonical discourse function associated with metaphor. As we have seen, metaphor normally has a predicative function: it says something "about" a Subject or Them e. Yet, in the "homy pussycat" interpretation, the predicative interpretation is at odds with the organization of the linguistic content as it emerges following integration. That is, while at the level of linguistic content lexical concept integration leads to the I Boss, lexical concept having the semantic value of Subject, interpretation leads to a conception in which the utterance serves to attribute the quality of bossiness to the "pussycat", the figurative target in this interpretation, rather than the "boss". For this reason, there is what we might think of as a mismatch between the output of lexical concept integration on one hand, and interpretation on the other. The net result is that such an interpretation is unlikely to feel metaphoric, although the utterance is figurative, in present terms. In the final analysis, metaphor and metonymy, rather than being neatly identifiable types of figurative language, are terms that have been applied by different scholars to a range of overlapping and sometimes complementary figurative language phenomena. What emerges from the LCCM account is that the intuitions that lie behind the use of these terms for data of particular kinds are a function of a small set of compositional mechanisms that are guided by various sorts of constraints (the principles identified in this and earlier chapters). Moreover, the application of these mechanisms and principles gives rise to a range of figurative conceptions which, in terms of discourse functions, are continuous in nature. That is, from the perspective of language understanding, while there are what might be thought of as symptoms of metaphor and metonymy, there is not always a neat distinction that can be made that serves to identify where metaphor ends and metonymy begins. What is not figurative language
In some accounts of figurative language phenomena,' examples such as the italicized lexical items in each of the following examples are taken to he metaphoric in nature: (12) a. That is a loud shirt b. They have a close relationship c. She is in love d. That took a long time For instance. we the metaphor identification criteria A% develop ed by the Praggleia/ tiroup (20071. See also disco...ion in papers in Itarcel ► na Imono).
299
In these examples, the use of /out/ refers to a brightly coloured shirt, close relates to emotional "closeness", in relates to an emotional state while long relates to extended duration. From the perspective of LCCM Theory, such usages relate to distinct lexical concepts rather than being due, for instance, to figurative language conceptions. For instance, long has at least two conventionally established lexical concepts associated with the vehicle long. !EXTENDED IN HORIZONTAL SPACE! and I EXTENDED DURATION]. Dining lexical concept selection the 'EXTENDED DURATION! lexical concept is selected, thereby avoiding a clash in the primary cognitive model profiles associated with (EXTENDED DURATION! and [TIME'. Evidence that long has (at least) two distinct lexical concepts conventionally associated with it comes from examples such as the following: 110 a. A long kiss h. A long hook "Long" in "long kiss" relates to extended duration, not to physical length—a kiss cannot, obviously, be extended in space. Similarly in (13b), we are not, or at least not typically, dealing with an oversize book, but rather with an extended reading time. Understanding the form long as relating to I EXTENDED DURATION) relates to the process of lexical concept selection, as discussed in Chapter iI. During language understanding we select the ! EXTENDED DURATION! !Mical concept in conjunction with the lexical concept I BooKI as facilitating provision of most coherent conception, as guided by context. Of course, we are helped by the frequency with which these two forms collocate and are associated with this very conception. Collocations of this kind which provide a pre-assembled conception I refer to as concept collocations. In the same way, long time represents a concept collocation. In view of the LCCM Theory account, concept collocations such as "long time" are not, then, appropriately thought of as involving "metaphor," in the sense that they do not result from the online process of clash resolution, as described above. This view of highly conventional "lexical metaphors" is consonant with the approach developed in the Career of Metaphor Hypothesis (Bowdle and Gentner 2005), which builds on the Structure Mapping Approach to metaphor developed by Dedre Gentner (e.g., 1988; (knitter et al. 2oo1). In that approach, highly conventionalized "metaphors" are treated as being polysemous sense units which are conventionally associated with the "base" term, here, long, and which are accessed via a "lexical look-up" process, rather than by establishing structural alignments and inference projections (mappings) between a base and target. This aspect of the LCCM perspective is also consonant with the work of Rachel Giora (2oo3). In her work, Giora demonstrates that certain examples of "figurative" meanings associated with lexical items appear to be stored in memory and can he more salient than so-called "literal" meanings.
FIGURATIVE LANGUAGE AND THOUGHT
;00
METAPHOR AND METONYMY
From the LCCM perspective, the interesting question in such cases is how an ( EXTENDED DURATION] lexical concept became conventionally associated with the form long. Recent work on semantic change pioneered by Elizabeth Closs Traugott (e.g., Traugott and Dasher 2004) has argued that situated implicatures (or invited inferences) can become "detached" from their contexts of use and reanalysed as being distinct sense-units—lexical concepts in present terms—which arc associated with a given vehicle. Intuitions, by som e scholar,texmpsarfigutvebd,1sgoniterfence of contextually irrelevant lexical concepts, in the case of our example the lexical concept (EXTENDED IN HORIZONTAL SPACE). I assume that such interference can occur when the temporal restrictions on language processing are relaxed, as is the case in the theoretical practice of language scientists who often appear to analyse such expressions without taking (any) account of their usage context(s). 9 In actual conversation, I would argue, talk of a "long time" is hardly ever felt to he figurative.") The [EXTENDED DURATION] lexical concept associated with long might be historically derived from contexts of communication in which reference to length can be understood as reference to duration without harming expression of the communicative intention, as in communication about "long journeys". Through repeated use of this form, with the inferred meaning, in such bridging contexts (Evans and Wilkins 2000), 11 it is plausible that long developed an (EXTENDED DURATION] lexical concept by virtue of decontextualization (Langacker 1987).
Summary This chapter has been concerned with an LCCM account of figurative language understanding. In particular, the chapter addressed the distinction between a literal conception—the meaning associated with a literal utterance—on one hand, and a figurative conception—the meaning associated with a figurative utterance—on the other. While a literal conception canonically results in an interpretation which activates a cognitive model, or cognitive models, within the default, which is to say primary, cognitive model profile, a figurative conception arises when cognitive models are activated in the secondary cognitive model profile. This takes place when there is a clash in one of the primary cognitive models involved in interpretation. A clash results in enlargement of the search domain, such that matching takes place in the secondary cognitive model profile associated with one of the relevant lexical concepts. In some cases, context serves to determine which • Sec Leezenhcrslumil and Stern (woo) for diskussion of the importante of context in metaphor understanding. See Bowdle and GenIner (200i) for • related perspective (41. the tirades, Saliense I ► pothesis o f ' 6
Giora. e.g.. 11
14, ► ' ).
Retail the dissussion in Chapter
S.
301
lexical concept is the site for clash resolution, captured by the Principle of Context-induced Clash Resolution, introduced in this chapter. A further distinction made was the distinction in discourse function associated with figurative conceptions referred to as metaphor and metonymy. It was argued that the distinction is due to whether there is alignment or not between what %.as referred to as figurative target and figurative vehicle. While the hallmark of metaphor is that there is divergence between the two, the symptom of metonymy is that there is alignment. In general terms, LCCM Theory predicts that the same set of compositional mechanisms arc responsible for literal and figurative language understanding. Hence, figurative language does not involve a distinct module or set of processes. Rather, it is continuous with literal language understanding.
TIl SEMANTICS OF TIME THE
15 The semantics of Time In Ihi chapter chaplCf I1 apply 11.1 Theory ingle domain: of In this apply L LCCM Theory10 to aa single domain: Ihe the domain domain of Time. oflen taken laken to 10 be be an an instance in. lanc. of ofan anabstract ab Iract domain, domai n. Time. A As Tim. Time is often structured in in terms P'" excel/filet. ~lru(lu~ed lerm~ of content conlenl from from more _more concrete concrele dodo.par excellence, mains Spa e. it It provides provide an an arena arena for for investigating onvestlgatong the the relationrelallonmains such such as Space, hip between between linguistic lingui tic representations representation (e.g.. lexi al concepts) concept) and and ship (e.g., lexical conceptual representation representations (cogmtive (cognitive model) models) from conceptual from the the perspective per~pe"'ve of ofaa ingle domain. Hence. my my strategy trategy in this this chapter chapter is i to to explore explnreaaspecific pe ilk single domain. Hence, dnmain (Time). structured siructured figuratively, figuratively. rather ralher than Ihan aa type type ofoffigurative figurallve domain (Time), metaphor). I1 do do so in phenomenon (e.g .• metaphor). in order order totoinvestigate inve"igale semantic semantic (e.g., represenlation. meaning-con tru lion proce se aas in on the the representation, rather rather than than meaning-construction processes previou hapter. previous_ chapter. A further further reason reason for for selecting ",Icetlng Time Time isI that Ihal this thischapter chapleralso al.soaddresses addr relat to entire domains: Space and Time..The difficulty here is that this leads to nllre doma in: Space and Time The difficulty here i Ihal Ih, lead to the position that Time, and indeed Space, are undifferentiated intern- to the poSition thaI Time . and indeed I' e, are undifTerentiated intern ally homogenous bodies of knowledge—a 100ft byMoore mad by .Isomade critici malso acriticism ally homogenoll bodi of knowl dg (e.g., 2oo6) in his analysis of space-to-time metaphors. (e.g .• 2006) in hi•• naly t; of 'I" e· tll-t,m e m tapho r;. i. The inherent structure issue i. Th inh rent struc ture i u There are two objections that can and have been levelled at the view viewofof Th re ar two ()b)eclion Ih.t ,an and have bttn It", lied at the inherent structure for Time, often attributed to Lakoff and Johnson. The on.The inher ent truetu re (or Time , oft n attrib uted to LakofT and lohl1> first relates to the kind of generalalcriticism presented in Murphy (1996). rilici m pr nted in Murp hy ('996 ). Ii", relal 10 Ih kind of gener Murphy argues that if abstract domains such as Time have little or no no hy .rgue s thaI if ab tract dom. in ""h" , TIIl,e h.ve hili or Murp inherent structure of their own, thereby requiring conceptual projection requiring con cptua l projection inh renl lruet ur. o( Iheir own. Iherebythen from source domains to provide structure, it is not clear what motivates mru."cJnmolin to prOVide IrU(lUrc. th n it i nol clear what l110livat from the projection in the first place. That is, it is not clear what motivates the llOn III Ihe Ii", pla,e_ Th., i • ", nol d ar wh.lI m()lIvatt'" the pro)ec Ih structuring of Time in terms ofo(domains Space rather rathe r Sp.,. n and M()lioand u. PAST, PRESENT, event is
I'AOTAACT£D
•• ••
Before Before accounting acwunting for f?r these these,issues, i ue;.I I first first consider con iderthe .he way way inIIIwhich whi hspatial patial conceptual content is recruited, via conceptual metaphor, to structure temconceptual content I recruited. via conceptual metaphor, to tru ture tern· poral poral cognitive (()g~'lIv models. m~e1. IIdo dososoby byconsidering comideringthe thecognitive cognitivemodel modelprofile profile accessed via the lexical concept (CHRIsTMAsl. aCl< '" v•• the leXICal ( ncel" (tI •• I\.MA I·
TEMPORAL
PROTRACTED DURATION DURATION
In (24), while also located In (24). wh.le Christmas 'hri tmas is is al located in in the the future, future. it it is is not not imminent. .mminent. Accordingly, there are three specific issues that need to be accounted for in in Accordingly. lherr are three peeific i ues that need to be accountedfor terms of explaining how LCCM Theory models the conception which arises temlS of explaining how LC .M Theory model the con eption which ari for for the the utterance utterance in in (23). (2J). These These are are summarized ummarized below: below:
rather than a rather than a spatial patial scene. ene. That That is, i •the lh utterance utteran, isi interpreted interpreted asa concerning concemlllg aa temporal temporal scenario scenario rather rather than than one oneinvolving involvingveridical veridical motion. mohon. Issue 2: /SSIl' z The The temporal temporal event esent of ofChristmas Chmtma isi located located ininthe thefuture futurewith with respect to our understanding of the present which is implicit, although r peel to our under tanding of the present which i implicit. although not not explicitly exphCltly mentioned, mentioned.ininthe theutterance utteranceinin(23). (2J). Issue y The future event of Christmas is interpreted /SSllf •.I: The future event of Chri tmas is interpretedasasbeing beingrelatively relatively imminent with respect to the present. .mmlllent w.th r peet to the present.
SYNCHRONOUS DURATION
OBJECT IN MOTION ALONG A PATH
Christmas approaching, but but is is still till aa long long way way off off Chri tma iis approaching.
•• Issue /e IIf I: The The utterance utteran e in in (23) (2J)'is interpreted IIIterpretedas asrelating relatingto toaatemporal temporal scene scene
J2J 323
Til SeMANTICS OF TIMl. THE SEMANTICS OF TIME
FIGURATIVELANGUAGE LANGUAGEAND ANDTltOUGUT THOUGHT FIGURATIVE
I
( A MPRE SSION
0
0
I I
0 I
(. •
PAST PAST
PRESENT PRESENT
FUTlM'IE FUTURE
DURATION
DURATION
RELIGIOUS
CULTlM'IALFESTIVAL FESTIVAL CULTURAL
TEMPORALEVENT EVENT TEMPORAL
RELIGIOUS FESTIVAL FESTIVAL
[CHRISTMAS) [CHRISTMAS] FIGUaf ISS. IS-S. MAW P.lfti,,-IIprimary pnm.Jrycognitive l.(lg"'t1\'~model modelprofile profilefor (or'CHRISTMAS' It ItRI\TMAS) FIGURE
324
FIGURATIVE LANGUAGE AND THOUGHT
attribute relates to the nature of the durational elapse associated with the event, which is to say its DURATION. This attribute has a number of values associated with it. Moving from right to left, the first is TEMPORAL COMPRESSION—the underestimation of time, which is to say, the experience that time is proceeding more "quickly" than usual. The second is SYNCHRONOUS DUE• A-rtoN—the normative estimation of time, which is to say, the experience time unfolding at its standard or equable rate. The final value is PROTRACTED DURATION. This relates to an overestimation of duration, which is to say the felt experience that time is proceeding more "slowly" than usual. The sorts of experiences that give rise to the range of different attributes and values represented in the conceptual system in the domain of Time are of myriad kinds. For instance, we experience the past in terms of the range of measures we deploy to record temporal "distance" from now, such as time lines, calendars, diaries, and so on. We also experience the past in terms of biological ageing, photographic records of past events, narrative and story which recount past happenings, as well as personal and autobiographical memory, and so on. The present is experienced by virtue of direct perceptual processing, the phenomenologically real perceptual moment briefly described above." The future is apprehended in terms of our experiences of intentionality and the realization of intentions, as well as our experience of waiting and the subsequent occurrence of events. Moreover, it is apprehended in terms of our experience and interaction with the recording mechanisms that we deploy in order to gauge the relative imminence of future events such as calendars, timetables, schedules, time plans, and time-reckoning systems and devices on a daily basis. Finally, we also have detailed knowledge of the range of phenomenologically real aspects of duration which we experience throughout our lives. The final primary cognitive model diagrammed in Figure 15.5 is that of Christmas as a RELIGIOUS FESTIVAL. This relates to knowledge concerning the nature and status of Christmas as a Christian event, and the way in which this festival is enacted and celebrated. In addition, the primary cognitive models for IciiiusmiAsi recruit structure from other cognitive models via conceptual metaphor. That is, as operationalized in LCCM Theory, a conceptual metaphor provides a stable link that allows aspects of conceptual content encoded by one cognitive model to he imported so as to form part of the permanent knowledge representation encoded by another. For instance, the primary cognitive model TEMPORAL EVENT is structured via conceptual metaphor in terms of a stable, long-term link holding between it and the cognitive model relating to an witucr IN MOTION ALONG A PATH. As such, the cognitive model, OBJECT IN MOTION ALONG A PATH, which is represented, in Figure 15.5, by virtue of a circle located on a path, with the arrow indicating direction of motion, provides I' See Evans (sonsit) for further detail•
TILE SEMANTICS OF TIME
'
325
the TEMPORAL EVENT cognitive model with inferential structure relating to our knowledge of objects undergoing motion along a path. The conceptual content recruited via conceptual metaphor is indicated by the dashed lines. Specifically, inferential structure from this cognitive model is inherited by the PAST, PRESENT, and FUTURE attributes, such that content relating to the region of the path behind the object serves to structure, in part, our experience of pastness, conceptual content relating to the object's present location serves to structure, in part, our experience of the present, and content relating to that portion of the path in front of the object serves to structure our experience of futurity. This is indicated by the dashed lines which map the relevant portions of the path of motion from the OBJECT IN MOTION ALONG A PATH cognitive model onto the relevant attributes: FUTURE, PRESENT, PAST. In addition, content relating to the nature of motion is inherited by the DURATION attribute. Again this is captured by the dashed arrow, which links the arrow—signifying motion—with the DURATION attribute. In Chapter to, I discussed chaining within the conceptual system—the phenomenon whereby links and associations are established such that a web of connections serves to relate cognitive models. Hence, cognitive models are related to one another, facilitating activation of knowledge as it is required, for instance, by linguistically mediated communication. In LCCM Theory, conceptual metaphors provide one of the ways in which cognitive models from other regions of the conceptual system can become linked with cognitive models belonging to the access site of a given lexical concept. By virtue of humans acting in the world, tight reoccurring correlations serve to establish connections between cognitive models associated with distinct domains in the conceptual systems of human infants prior to the onset of language (see Lakoff and Johnson 1999). The establishing of these links provides a powerful organizational device that facilitates the deployment and re-us• of multimodal knowledge in order to structure other (less easily apprehended) domains of experience.' 2 From the perspective of LCCM Theory then, as conceptual metaphors serve to establish stable links between specific cognitive models that may belong to the access sites of many lexical concepts—for example, Easter, Spring, the concert, his prime, and so forth—this leads to massive redundancy of spatial conceptual content subserving temporal concepts. That is, conceptual metaphors provide a fundamental structuring mechanism of the human conceptual system.
" It is important to note that sonteptual metaphor—the establishment of linked sognitive models which derive from unrelated domains of expetiente is but one way in which cognitive models 'niters' %truth's-v. others insludr the ► hen ►► menon isanssentlence. as well is ♦ alit Rees, discussed in Chapter 10. These pin models.
ilia also serve to essaslish links bvissern tisimitivr
Nf.
FIGURATIVE LANGUAGE AND THOUGHT
THE SEMANTICS OF TIME
Meaning construction in Christmas is approaching I now return to a consideration of how the various interpretations (issues 1- 3
discueabov), tdwih(23arse.
Issue I Firstly, how is it that the utterance in (23) is interpreted as relating to a temporal scenario rather than a spatial one? The answer is as follows. Th e TFoRlexicanpthsoeurancswholevafrm for interpreting the open-class lexical concepts—those associated with the vehicles Christmas and approaching allowing them to achieve an informational characterization relating to a temporal scene. That is, the linguistic content encoded by this TFoR lexical concept, as described above, ensures that the interpretations that arise for the lexical concepts paired with Christ► as and {approaching are a consequence of these lexical concepts undergoing integration in the context of schematic temporal, as opposed to spatial, content. Put another way, as the overarching internally open TFoR lexical concept relates to a temporal scene, this provides a schematic framework which constrains the process of interpretation, as it applies to the open-class lexical concepts that populate the larger TFoR lexical concept. —
i ssue 2 Secondly, how is it that the utterance is understood as relating to a temporal event which is "located" in the future? After all, as we saw above the experiencer-based TFoR lexical concept which licenses the utterance as a whole does not encode whether a given temporal event is situated in the past or future. The answer, I suggest, relates to a special kind of matching that involves the spatial content recruited via conceptual metaphor, which structures the cognitive model profile of [CHRISTMAS] and the primary cognitive model profile accessed via [APPROACHING]. This type of matching I refer to as conceptual metaphor matching, which is constrained by the Principle of Conceptual Metaphor Matching, summarized below: (p12) Principle of Conceptual Metaphor Matching During interpretation, (an) open-class lexical concept(s) structured in terms of conceptual metaphor(s) are subject to matching, whenever possible, in the primary cognitive model profile of relevant lexical concepts in the same lexical conceptual unit. Conceptual metaphor matching does not preclude regular matching. This principle does two things. Firstly, it ensures that in the case of (23) the spatial content to which lciikisTmAsl has access in its primary cognitive
327
model p rofile by virtue of recruitment via conceptual metaphor is matched with relevant cognitive model(s) in the primary cognitive model profile of I APPROACHING]. Secondly, this matching operation does not interfere with, and hence does not prevent regular matching, matching that takes place on conceptual content which is not recruited via conceptual metaphor. In terms of the utterance in (23), the spatial content to which iciiRISTMASI facilitates access has to do with inferential structure derived from the motion scenario involving an object in motion. This is matched with the kind of terminal motion accessed via (APPROACHING]. The cognitive model profile associated with [APPROACHING' involves motion towards an entity, and hence, the object in motion is in front of the entity with respect to which it is "approaching". As the FUTURE attribute of the TEMPORAL EVENT cognitive model accessed via (CHRISTMAS( is structured in terms of that part of the motion trajectory that is in front, the resulting match involves an interpretation in which the temporal event of Christmas is "located" in the future. In other words, this particular interpretation is a consequence of the special type of matching I refer to as conceptual metaphor matching.
Issue 3 The final issue relates to the interpretation that the temporal event of Christmas in (23) is interpreted as being relatively imminent. This interpretation arises, I argue, due to the regular process of matching as described in earlier chapters. Matching, as guided by the previously introduced Principles of Interpretation, attempts to build an informational characterization for (CHRISTMAS' and [APPROACHING] by first searching the primary cognitive models of both these open class lexical concepts. As Christmas is a temporal, cultural, and religious event, and hence something that cannot undergo the sort of veridical motion implicated by the primary cognitive model profile associated with APPROACHING 1, a clash arises. This necessitates clash resolution. Due to the Principle of Context-induced Clash Resolution, introduced in the previous chapter, [CHRISTMAS] is designated as the figurative target, and [APPROACHING' the figurative vehicle. This follows as the utterance is "about" Christmas, and specifically serves to "locate" Christmas "in" time. The consequence of the foregoing is that a search is established in the secondary cognitive model profile of (APPROACHING]. A very partial cognitive model for (APPROACHING' is provided in Figure 15.6. The cognitive model profile for (APPROACHING] includes primary cognitive models for a TARGET LOCATION, the DIRECTED MOTION OF AN ENTITY, and the IMMINENCE of ARRIVAL OF AN ENTITY. A consequence of the relative imminence of arrival of an entity is the IMMINENCE OF OCCURRENCE OF EVENT, which is a secondary cognitive model. As a temporal event such as Christmas can occur, but not (literally) arrive, -
;28
FIGURATIVE LANGUAGE AND T"OU:::G~·::".:.T_ _ _ _ _ _ __ _ __ FIGURATIVE LANGUAGE AND THOUGHT
IMMINENCEOF OF IMMINENCE OCCURRENCE OF OCCURRENCE OF EVENT EVENT
TARGET TARGET LOCATION LOCATION
DIRECTEO DIRECTED MOTION OFAN AN MOTION OF ENTITY ENTITY
IMMINENCE OF OF IMMINENCE ARRIVAL OF ARRIVAL OF ENTITY ENTITY
(APPROACHING! [APPROACHING]
FIG"Rf 1~6. P,nl.l ognlli, model profile for (APPROA "'~G! 1- 1GURF 156. Partial cognitive model profile for [APPROACHING] there is aa match match between between the the secondary ;econdary cognitive cognili.e model model IMMINENCE IMMINE U OF UF there is oc XRRENCE Of EVFNT and the primary primary cognitive cognitive model model profile profile of of OCCURRENCE OF EVENT and the Ie(CHRISTMAS!. IIRI\T tA\I. Ilence, mterpretation of the imminence imminence of ofthe theoccurO« lor Hence, the the interpretation of the renee Christm. iis due to aa metaphoric metaphoric conception coneeplion arising, ari ing. along along the the rence of of Christmas due to line discu sed in in more more detail detail in the previou chapter. lines discussed in the previous chapter.
The of conceptual conceptualmetaphors metaphorsininLCCM LCCM Theory Theory The status status of II condude by conSidering the statu or conceptual conteptual metaphors metaphor; in in conclude this considering the status of this chapter chapter by LCC 1 Theory. to answer, answer. four four questions: question: LCCM Theory.IIdo do so by by asking. asking, and and attemptmg attempting to What tatus of metaphors in in I.CCM tecM Theory? Theory? What is is the the .status of ",nceptual conceptual metaphors What role of of lexical lellical con ept and cognitive models? models? What is is the the distinctive distinctive role concepts and cognitive What the motivation motivation for for TFoR TI'oR lexical Ie i(al concepts concepts to to deploy deploy vehicles vehide What is is the relating patiallanguage? relating to to literal literal spatial language? •• What d this how about th utility utility of of using u ing semantic semantic structure tru ture in on What does this show about the language investigating conceptual (on eptual structure? tructure? language a, as aa len lens for for investigating •• •• ••
I.i. What tatu of t.phors? Whatiis the the status of conceptual conceptual m metaphors? (onleptual structurmg Conceptualmetaphor, metaphors,in in teeM LCCM Theory, Theory, proVIde provide aa Illean, means of of structuring "'gmti .. model, term of of conceptual conceptual content content recruited recruited from from cognitive cognitive cognitive models in in terms model SO. tual tual Metaphor Theory needJ needs to be rir Ily, Con eptual Metaphor 'l11eory has ha traditionally traditionally been been concerned con erned Firstly, Conceptual Metaphor Theory with motor experience e perience and and how how this Ihi gives givC\rise rise to 10abstract ab lract concepts. on«plS. with sensory· sensory-motor ror tan e, Lakoff Lakoff and Johnson have have emphasized empha ized that that notions notion such uch asas For in instance, and Johnson Anger, are, in in some som sense, n , subjective ubjective and and hence, h nce,inincertain certain Anger, Tim, Time, and and Quantity arc, Ii n . Much of of the Ihe impetus impetu behind behindConceptual Con eplualMetaphor Metaphor rrespects peclS ab Ira t nnotions. abstract Theory has has been been to show show how how these thesemore moreabstract abstrael notions notion are ar structured truelured in in term hav been referred referred 10 a concrete dimensions d,men.ion of ofexperience, experience,for for terms of of what have to as in terms t rm of of Motion Motion through through space, pa e. Anger Anger in in terms termsofofHeat, lleat. in tan, such uch as a accompanying ac ompanylng gestures, S tur... track of all the the utt.ranc utterances iissued and oth.r other signals, pro xly. and on. This 1111 constitutes con tltut the the textual textual representation. repr ntation. In In addition, ddllion. partiparti prosody, and M> so on. c.:ip.Jnl m.1IOI,JIO;J· ilu.1tion.J1 representation. rcpr nt.1t1on. cipants maintain a 'situational Transcrndence ncerns the the number number and and range range of oflocations 1000Jtion at al which whi h and andwhen when Transcendence Concerns "individu,JI • 'types, "typo. 'episodic 'C'pi\Odksituations, ItuJtion~.and olnd'generic "gen ricsituations _itu.Jtion are .lrerepresented repr nted 'individuals, in our our mental mental representation rcprC'Knt.Jtion of ofthe the world. world.The Thegreater grealer the Ihe number numberand olndrange rang of of in location. the Ihe more more transcendent troln endent the the 'cognitive "cogniti\:emodel modelininquestion. qu lion.Transcendence TranS(cndrm;e locations, un lead I ad to to aa cognitive l.ogniti\t model model becoming h«omlllg'functionally "fun'-'Ionallydetached. dC'I,Jl.het.l. can TyptS A kind A kind mod.1. TyptS art mtotal r.p""'nwlonbased basttrallTypes of of'cognitive *cognitive model. Types are mental representations ing.1c.:m partKuln"'individuals andividualsininorder ordertotoleave I .)\'Cpoints POint ofof imiLuily.AAtype tyreisithus thu a a ing across particular similarity. gentn( representation reprnent.Jtion based ~ on onaasettof ofrelated relatedindividuals. iooivldu.tl . generic Unpacking The Th.process, pro< central ,c.ntraltoto'lexical '1«lealconcept conceptintegration, integration,whereby wh rtbylinguistic lingul tic I`npacking (ontenl encoded enuxied by hy lexical Icxiul concepts con(ept\ in 10 an .lOutterance utter.1nt:e isI~integrated Intege-ollN in inthe theway woly content (on tralnN by by the thethree thr«principles prinlipl of ofintegration. integration. constrained tltrance AA somewhat somC"Whal discrete disc.:rtte entity tntltythat that has h.1 unit unll like like status taW ininthat thatititrepresents reprncnt Utterance theexpression upr ion of ofaasingle ioglccoherent (:oherentidea, ide.J.making makang(at (.1least I ol~tpartial) p.utl.ll)use uscofc.,fthe thenorms nonn the and conventions comenlloo ofoflinguistic lingul licbehaviour behaviourinin aa particular partkular linguistic linguislic community. c.:ommunaty.An An and ulleran(erepresents repr nl aa specific, ~ifi(. contextualited, c.:ontcxtualiltd. and olndunique uniqueinstance in tJn(eofoflanguage I. mgu.lge utterance u •performed performedby hy a,J language IJngu.Jgc user u 'r in 10 !J.Crvke uf Ignollhng,J p.1rti(:ul.Jr ulmmuni use, service of signalling a particular communicativeintention. inlention.Hence, HC'n(e.ananutterance ulteran(econstitutes (:on tllul a adiscrete dl ·reteusage u geevent. C'\'cnt. cative -
Valu See Set" unhult v.Jluesets. '1 . Value •Attribute-value Hhldc. Vehicle See • "Phonulogil..aI Phonological vehicle. Vehicle
REFERENCES
References Aitchison, Jean. (1996). Wards in the Mind. Oxford: Blackwell. Aliport, D. A. (1985). Distributed memory, modular subsystems and dysphasia. In S. K. Newman and R. Epstein (eds), Current Perspectives in Dysphasia, 207-44. Edinburgh: Churchill Livingstone. Allwood, Jens. (2oo3)• Meaning potentials and context: Some consequences for the analysis of variation in meaning. In H. Cuyckens, R. Dirven, and i. Taylor (eds), Cognitive Approaches to Lexical Semantics, 29-66. Berlin: Mouton de Gruyter. Alverson. Hoyt. (1994). Semantics and Experience: Universal Metaphors for Time in English, Mandarin, Hindi gird Sesotho. Baltimore, MA: Johns Hopkins University Press. Atkins, B. T. S. (1987 ). Semantic-ID tags: corpus evidence for dictionary senses. In The Uses of Large Text Databases: Proceedings of the Third Annual Conference of the New OED Centre, 17- 36. Canada: University of Waterloo.
Bach, Kent. (1997). The semantics-pragmatics distinction: What it is and why it matters. Linguistiche Berichte 8, Special Issue on Pragmatics, 33 -So. Baddeley, Alan D. (1966). Time estimation at reduced body temperature. American Journal of Psychology, 79 (3): 475-9. Barcelona, Antonio. ( z000). Metaphor and Metonymy at the Crossroads. Berlin: Mouton de Gruyter. Barsalou, Lawrence. (1991). Deriving categories to achieve goals. In (i. H. Bower (cd.), The Psychology of Learning and Motivation: Advances in Research and Theory,
CA: Academic Press. (1992a). Frames, concepts, and conceptual fields. In A. Lehrer and E. F. Kittay (eds), Frames. Fields. and Contrasts: New Essays in Lexical and Semantic Organization, 11-74. Hillsdale, NJ: Lawrence Erlhaum. (1992b). Cognitive Psychology: An Overview for Cognitive Scientists. Hillstble, NJ: Lawrence Erlhaum. (1999). Perceptual symbol systems. Behavioral and Brain Sciences, 22: 577-66 ► . (2003). Situated simulation in the human conceptual system. Language and voL
27, 1-64. San Diego,
Cognitive Processes, 18:
513-62.
(2oos). Continuity of the conceptual system across species. Trends in Cognitive Sciences, 9: 309-11.
(ioo8). Grounded cognition. Annual Review of Psychology, 59: 617-45. and Billman. I)orrit. (1989). Systematicity and semantic ambiguity. In D. Gorfein (ed.), Resolving Semantic Ambiguity, 146-203. New York: Springer-Verlag. Santos, Ava, Simmons, Kyle, and Wilson, C. D. (forthcoming). Language and simulation in conceptual processing. In M. De Vega, A. Glenberg. and A. Graesser (eds), Symbols, Embodiment, and Meaning. Oxford: Oxford University Press.
359
Barsalou, Lawrence. and Wiener-Hastings, Katja. ( zoos). Situating abstract concepts. In Pecher and R. Zwaan (eds), Grounding Cognition: The Role of Per and Action in Memory, languag•, and Thought, 129-63. New York: Cambridge University Press. Ych, Wenchi, Luka, Barbara, Olseth, Karen, Mix Kelly, and Wu, Ling-Ling. (1993). Concepts and meaning. In K. Beals, G. Cooke, D. Kathman, K. E. McCullough, S. Kita, and D. Testen (eds), Chicago Linguistics Society 9: Papers from the Parasessions on Conceptual Representations, vol. z 23-61. Chicago Linguistics Society. Bender, Andrea, Bennardo, Giovanni, and Beller, Sieghard. (zoos). Spatial frames of reference for temporal relations: A conceptual analysis in English, German, and Tongan. In B. G. Bara, L Barsalou, and M. Bucciarelli (eds), Proceedings of the Twenty-Seventh Annual Conference of the Cognitive Science Society, 220-5. Mahwah, NJ: Lawrence Erlhaum. Bennett, David. (1975). Spatial ism, Temporal Uses of English Prepositions. London: Longman. Bergen, Benjamin K. and :hang, Nancy. (zoos). Embodied construction grammar in simulation-based language understanding. In I.-0. Ostman and M. Fried (eds), Construction Grammars: Cognitive Grounding and Theoretic -al Extensions, 147-9o. Amsterdam: John Beniamins. Polley, Carl, and Wheeler, Kathryn. (forthcoming). language and inner space. In V. Evans and P. Chilton (eds), Language, Cognition and Space: The State of the Art and New Directions. London: Equinox Publishing. Boroditsky, Lera. (woo). Metaphoric structuring: Understanding time through spatial metaphors. Cognition, 75 (1): 1-28. (zom). Does language shape thought? Mandarin and English speakers' conceptions of time. Cognitive Psychology, 43: 1 - 22. and Prinz, Jesse. (forthcoming). What thoughts are made of. In G. Semin and E. Smith ( eds), Embodied Grounding: Social, Cognitive. Affective. and Neuroscientilic Approathes. Cambridge: Cambridge University Press. Boodle, Brian and Gentner, Dedre. (zoos). The career of metaphor. Psychological Review, 193-216. Bowerman, Melissa and Choi, Soonja. (2oo3). Space under construction: Languagespecific spatial categorization in first language acquisition. In D. Gentner and S. Goldin-Meadow (eds), Language in Mind: Advances in the Study of Language and Thought, 387-428. Cambridge, MA: MIT Press. Brenier, Jason M. and Michaelis, Laura A. (2o05). Optimization via syntactic amalgam: Syntax-prosody mismatch and copula doubling. Corpus Linguistics and Linguistic Theory, I: 45-88. Brugman, Claudia. (1988). The Story of "over": Polyserny, Semantics and the Structure of the Lexicon. New York: Garland. and Lakoff, George. (1988). Cognitive topology and lexical networks. In S. Small. G. Cottrell, and M. Tannenhaus (eds), Lexical Ambiguity Resolution, 477 -507. San Mateo, CA: Morgan Kaufman. Burling, Robbins. (2oo7). The Naked Ape: How language Evolved. Oxford: Oxford University Press.
REFERENCES
REFERENCES
360
Cann, Ronnie (1993). Formal Semantics. Cambridge: Cambridge University Press. Carston, Robyn. (2002). Thoughts and Utterances: The Pragmatics of Explicit Commu-
361
Croft, William and Cruse, D. Alan. ( 2004). Cognitive Linguistics. Cambridge: Cambridge University Press. Cruse, D. Alan. (2002). Aspects of the micro-structure of word meanings. In Y. Ravin
nication. Oxford: Blackwell.
and C. Leocock (eds), Polysemy: Theoretical and Computational Approaches,
Casasanto, Daniel. (forthcoming). Space for thinking. In V. Evans and l'. Chilton
30-51.
Oxford: Oxford University Press.
(eds). Language, Cognition and Space: The State of the Art and New Directions. London: Equinox Publishing. - and Boroditsky, Um. (2008). Time in the mind: Using space to think ab(nit
Cuyckens, Hubert, I)irven, Rene, and Taylor, John (2oo3). Cognitive Approaches to
time. Cognition, 106: 579 93.
-
-
Lexical Semantics. Berlin: Mouton de Gruytcr. Sandra, Dominick, and Rice, Sally. (1997). Thwards an empirical lexical semantics. In B. Smieja and M. Tasch (eds), Human Contact through Language and
Chafe, Wallace. (1994 ). Discourse. COnStiOUSlIeSS, and Time: The Flow and Displacement of Conscious Experience in Speaking and 4Vriting. Chicago: University of Chicago
I ► nguistics, 35 54. Frankfurt: Peter Lang. -
Dabrowska, Ewa. (2oo9). Words as constructions. In V. Evans and S. Pourcel (ads),
Press.
New Directions in Cognitive Linguistics, 201-24. Amsterdam: John Benjamins.
Choi, Soonja and Bowerman, Melissa. 0990. Learning to express motion events in English and Korean: The influence of language-specific lexicalitation patterns.
Damasio, Antonio. (1989). Time-locked multiregional retroactivation: A systemslevel proposal for the neural substrates of recall and recognition. Cognition, 33:
Cognition, 41: 83-121. Chomsky, Noam. (1957). Syntactic Structures. The Hague: Mouton. (1965). Aspects of the Theory of Syntax. Cambridge, MA: MIT Press.
25 62. (1994). Descartes' Error: Emotion. Reason arid the Human Bnain. London: Vintage.
(1981). Lectures on Government and Binding. Dordrecht: Foris. (0991). Some notes on economy of derivation and representation. In R. Freidin 417 54. Cambridge, (ed.), Principles and Parameters in Comparative Grammar,
Dancygier, Barbara and Sweetscr, Eve. (2005). Alental Spaces in Grammar: Conditional
-
MA: MIT Press. 1199s). The Minimalist Program. Cambridge, MA: MIT Press. Christiansen, Morten H. and Kirby, Simon. (2003). Language Evolution. OxfOrd: -
-
Constructions. Cambridge: Cambridge University Press. Davies, Paul. ( 2006). About Time: Einstein's Unfinished Revolution. London: Penguin. Deacon, Terrence. (1997 ). The Symbolic Species: The Co-evolution of Language and the Brain. New York, NY: W. W. Norton and Co. Deane, Paul. (2oo5). Multimodal spatial representation: On the semantic unity of over." In B. Hampe (ed.), From Perception to Meaning: Image Schemas in Cognitive
Oxford University Press. 4 gain. World .1; ..wf. 1 ler A
Clark, Andy. (1998). Being There: Putting Brain. Body and Cambridge, MA: MIT Press.
Clark, Herbert. (0973). Space, time, semantics, and the child. In T. Moore (ed.), Cognitive IleveIopment and the Acquisition of Language, 27 63. New York: Academic -
Press. -
(1983). Making sense of nonce sense. In G. Flores D'Arcais and R. I. larvella
(eds), The Process of Language I !nilerstatrifing, 297-332. Chichester: john Wiley. (1996). Using Language. Cambridge: Cambridge University Press. Corhallis, Michael. (2003). From Hand to Mouth: The Origins of Language. Princeton:
Linguistics, 235-82. Berlin: Mouton de Gruyter. Dijk, Teun van and Kintsdi, Walter. (1983). Strategies of Discourse Comprehension. New York, NY: Academic Press. Donald, Merlin. (1991). Origins of the Modern Mind: Three Stages in the Evolution of
Culture and Cognition. Harvard: Harvard University Press. Dunbar, Robin. ( 0996). Grooming. Gossip and the Evolution of Language. Faber & Faber.
Engberg- Pederson, Elisabeth. (1993). Space in hinish Sign I anguage: The Meaning and
Morphosyntactic Use of Space in a Visual Language.
sions of perception verbs in Australian languages. l angua ige, 76 (3): 546-92.
Coulson, Seana. (2000). Semantic Leaps. Cambridge: Cambridge University Press. Coventry, Kenny and Garrod, Simon. ( 2004). Saying, Seeing and Acting: The Psychological Semantics of Spatial Prepositions. Hove: Psychology Press.
Evans, Vyvyan. (20040). The Structure of Time: Language. Meaning and Cognition. Amsterdam: John Benjamins. (2oo4h). How we conceptualise time. Essays in Arts and Sciences. 33 (2): 13 44Polysemy, the lexicon and conceptual structure. iru meaning (1):"time": 33
Croft, William. (1993). The role of domains in the interpretation of metaphors and
-
metonymies. Cognitive Linguistics, 4: 335 70. -
(1998). Mental representations. Cognitive Linguistics. 9 (2): (2000). Explaining Language Change An Evolutionary Approik•.
Jour nal 2 Sofh Linguistics, Th
1 51 -74-
- 75.
(449 c01-s. (20 06). LiL;exical concepts, cognitive models and Linguistics
London: lA ► gman.
(2002). Radical Construction Grammar: Syntactic Theory in Typological Perspective. Oxford: Oxford University Press. (2007). The origins of grammar in the verbalization of experience. Cognitive
Linguistics, 18 (3): 339-82.
Hamburg: Signum-Verlag.
Evans. Nicholas and Wilkins, David. (2oo0). In the mind's car: The semantic exten-
Princeton University Press.
-
London:
-
ion. Cognitive meaning-const ruction.
(forthcoming a). From the spatial to the non-spatial: The "state" lexical concepts of in, on and in. In V. Evans and P. Chilton fetish I anguage,
REFERENCES
REFERENCES
362
Cognition and Space: The State of the Art and New Directions. London: Equinox Publishing. Evans, Vyvyan. (forthcoming b). The perceptual basis of spatial representation. In V. Evans and P. Chilton (eds), Language. Cognition gold Vail.: The State of the Art
and New Directions. London: Equinox Publishing. and Chilton, Paul (eds). (forthcoming). Language, Cognition and Space: The State of the Art and New Directions. London: Equinox Publishing. and Green, Melanie. (2oo6). Cognitive Linguistics: An Introduction. Edinburgh:
363
Fodor, Jerry A. 09831. The Modularity of Mind. (:ambridge, MA: MIT Press. Fraisse, Paul. (1963) . The Psychology of Time. New York Harper and Row. Gainotti, Guido, Silveri, Maria Caterina, Daniele, Antonio, and Giustolisi, Laura. (1995). Neuroanatomical correlates of category-specific semantic disorders: A critical survey. ,%fe ► ory, 3: 247-64. Gallagher, Shaun. (2006). How the Body Shapes the Mind. Oxford: Oxford University I'ress. Gallese, Vittorio and lakolf, George. zoos). The brain's concepts: The role of the sensory-motor system in reason and language. Cognitive Neuropsychology, 22: 45s-79.
Edinburgh University Press. and Tyler, Andrea. ( 2004). Rethinking English "prepositions of movement": The
Garner, W. R. (1978). Selective attention to attributes and to stimuli.
case of to and through. In H. Cuyckens, W. De Mulder, and T. Mortelmans (eds), Adpositions of Movement. Belgian Journal of Linguistics, t8: 245-70. Amsterdam:
Experimental Psychology: General, 107 ( 3 ): 287-308. Geeraerts, Dirk. 0994). Diachronic Prototype Semantics. Oxford: Oxford University
Journal of
Press.
John Benjamins. Fauconnier, Gilles. (1994). Mental Spaces. Cambridge: Cambridge University Press. (1997). Mappings in Thought and language. Cambridge: Cambridge University
Gentner, Dedre. (1982). Why nouns are learned before verbs: Linguistic relativity
Press. and Turner. Mark. (1998). Conceptual integration networks. Cognitive Science,
Language, Thought and Culture, 301-34. Hillsdale, NJ: lawrence Erlhaum. (1988). Metaphor as structure mapping: Tlw relational shift. Child Development,
versus natural partitioning. In S. A. Kuczaj (ed.), Language Development: Vol. 2.
59: 47- 59. and Boroditsky, Lera. (29m). Individuation, relational relativity and early word
22 (2): 33-187.
(2002). The iVily we Think Conceptual Blending and the Mind's Hidden
learning. In M. Bowerman and S. Levinson (eds), Language Acquisition and Con-
Complexities. New York: Basic Books. (loos). Rethinking metaphor. In R. W. Gibbs (ed.). The Cambridge Hand-
ceptual Development, 215-56. Cambridge: Cambridge University Press.
book of Metaphor and Thought. Cambridge: Cambridge University Press. Feist, Michele. (forthcoming). Inside in and on: Typological and psycholinguistic perspectives. In V. Evans and P. Chilton (eds), Language, Cognition and Space:
Bowdle, Brian. %Vial', Phillip, and Boronat, Consuclo. (2001). Metaphor is like analogy. In D. Gentner, K. 1. Holyoak, and B. N. Kokinov (eds), The AntilOgiCal
Mind: Perspectives from Cognitive Science, 199-253. Cambridge, MA: MIT Press. and Goldin-Meadow, Susan. (2003). Language in Mind: Advances in the Study of Language and Thought. Cambridge, MA: MIT Press.
The State of the Art and New Directions. London: Equinox Publishing. Feldman, Jerome. (2oo6). From Molecule to Metaphor: A Neural Theory of Language. Cambridge, MA: MIT Press.
-
two systems in processing space time metaphors. Language ea se! (:ognitive Processes,
and Narayanan, Srini. (29o4). Embodied meaning in a neural theory of language. Brain and Language, 89: 385-92. Fillmore, Charles. (1975). An alternative to checklist theories of meaning. Proceedings Odle First Annual Meeting of the Berkeley Linguistics Society, 123-31. -
(1982). Frame semantics. In Linguistics in the Morning Calm, 111-37. Seoul: Hanshin Publishing Company.
-
17 (5): 537-65. Gibbs, Raymond W. (1994). The Poetics of Mind. Cambridge: Cambridge University Press. -
(1997). lectures on Deixis. Stanford: CA: (*.MI Kay, Paul, and O'Connor, Mary Catherine. (1988). Regularity and idiomaticity
in grammatical constructions: The case of let alone. Language, 64: 501-38. l'inke, Ronald A. (1989). Principles of Mental Imagery. Cambridge, MA: MIT Press. Flaherty, Michael. (1999). A Watched Pot-. How We Experience Time. New York, NY: New York University Press. Fleischman, Suzanne. (1982). The past and the future: Are they coming or going? Berkeley Linguistics Society, 8: 322-34.
( woo). Embodiment and Cognitive Science. Cambridge: Cambridge University Press.
Giora, Rachel. (1997). Understanding figurative and literal language: The graded
(1985). Frames and the semantics of understanding. Quaderni di Seniantica, 6: 222-54.
[mai. Mutsumi, and Boroditsky, Lera. (2oo2). As time goes by Evidence for
salience hypothesis. Cognitive linguistics. 8 (3): 183-206. -
(2003). On our Mind: Salience, Context, and Figurative Language. New York: Oxford University Press. ;lenberg, Arthur. (1997). What memory is for. Behavioral and Brain Sciences, 20: 1-55. and Kaschak, Michael. (2002). Grounding language in action. Psychonomic
Bulletin and Review, 9: 558-85. Aucksberg. Sam. ( 2001). Understanding Figurative Language. New York: Oxford University Press. (2003). The psycholinguistics of meta pho r. Trends in Cognitive Science, 7: 92-6. and Keysar, Boa,. two). Understanding metaphorical comparisons: Beyond similarity. Psychological Review, 97: 3- 18.
REIERENCES
REFERENCES
;64
365
Grammaticalization, and edition.
( ;otinian. Erving. (19811. Forms of Talk. Philadelphia: University of Pennsylvania Press.
IIupper, Paid and Traugott, Ili/abeth (loss. (2oo3).
( ;oldberg, Adele. (1995). Constructions. A Construction Grammar Approach to Argu► e ► t Structure. Chicago: University of Chicago Press.
Hurford, James. (2107). Origins of Meaning. Oxford: Oxford University Press.
- (2006). Constructions at Work The Nature of Generalization i ► i I anguage. Oxford: Oxford University Press. Gonzalez-Marquez, Monica, Mittelherg, Irene, Coulson, Seam, and Spivey, Michael
1. (2007). Methods in Cognitive Linguistics.
Amsterdam: John Benjamin,.
Grady, Joseph E. (1997). Foundations of meaning: Primary metaphors and primary scenes. Unpublished doctoral thesis, Linguistics Dept.. LIC Berkeley. (2oos). Primary metaphors as inputs to conceptual integration.
Pragmatics.
Journal of
Cambridge: Cambridge University Press. Ivry, Richard B. and Spencer, Rebecca M. C. (2oo4). The neural representation of time. Current Opinion in Neurobiology; 14 (2): 225-32. lackendoff, Ray. (1983). Semantics and Cognition. Cambridge, MA: MIT Press. (1990). Semantic Structures. Cambridge, MA: MIT Press. (1992). Languages of the Mind: Essays on Mental Representation. MA: MIT Press. (mu).
Cambridge,
Foundations of Language: Brain, Meaning, Grammar, Evolution.
Oxford
and New York: Oxford University Press.
17-10: 1595 -614.
Converging evidence for the notions of "subscene" and "primary scene. Proceedings of the 23rd Annual Meeting of the Berkeley Linguistics Society, 123-36. Berkeley, CA: Berkeley Linguistics Society. Grier, Paul H. (1989). Studies in the Way of Words. Cambridge, MA: Harvard University
Jaeger, leri and Ohala, John. 0984). On the structure of phonetic categories. Proceed-
Press. (;vies, Stefan Th. (2006). Corpus-based methods and cognitive semantics: The many
January, David and Kako, Edward. ( 2007). Re-evaluating evidence for the linguistic
- and Johnson, Christopher.
2000).
ings of the loth Annual Meeting of the Berkeley Linguistics Society,
15-26. Berkeley,
CA: Berkeley Linguistics Society. James, William. ( 118901 195o). The Principles of Psychology. New York: Dover. relativity hypothesis: Response to Boroditsky (2o01). Cognition, io4: 417-26.
meanings of to run." In S. Th. Grits and A. Stefanowitsch (eds), Corpora in Cognitive Linguistics: Corpus-based Approaches to Syntax and Lexis, 57 -99. Berlin:
lasmin, Kyle. (2oo8). Declaring Genius at Customs.
Mouton de Gruyter. and Diviak, Dagmar. (2009). Behavioral profiles: A corpus-based approach to
lohansson, Sverker. ( 2005). Origins of Language: Constraints on Hypotheses.
cognitive semantic analysis. In V. Evans and S. Pourcel (eds), New Directions in
Johnson, Mark. (1987). The Body in the Mind The Bodily Basis of Meaning, Imagin-
Cognitive Linguistics,
57- 76. Amsterdam: John Benjamins.
( ;umperz, John. (1982). Discourse Strategies. Cambridge:
Cambridge University Press.
and Levinson, Stephen. (19961. Rethinking Linguistic Relativity.
Cambridge:
Haiman, John (1980). Dictionaries and encyclopedias. Lingua, so: 329-57. Hanks, Patrick. 0996). Contextual dependency and lexical sets. international Journal Harder, Peter. (2009). Meaning as input: The instructional perspective. In V. Evans and S. Pourcel (eds), New Directions in Cognitive Linguistics, 15-26. Amsterdam: Johns Benjamins. Harley, Trevor. (2oo8). The Psychology of Language: From Data to Theory,
3rd edition.
dam: Benjamins.
ation and Reason. Chicago: University of Chicago Press. (um). The Meaning of the Body: Aesthetics of Human Understanding.
Chicago:
University of Chicago Press. Press. )(Kg-Schmid, I LIM. ( 20X0). English Abstract Nouns as Conceptual Shells: From Corpus Kaschak, Michael and Glenberg, Arthur. (2000). Constructing meaning: The role of affordances and grammatical constructions in sentence comprehension. Journal of
Memory and Language, 43: 508-29. Katz, Jerrold J. (1972). Semantic Theory. New York, NY: Harper and Row. and Fodor, Jerry A. (1963). The structure of a semantic theory. Language,
Hove: Psychology Press. Harrington, Deborah L, Haaland, Kathleen Y., and Knight, Robert T. (1998). Cortical
18 (3):
Amster-
to Cognition. Berlin: Mouton de Gruyter.
75 -98.
networks underlying mechanisms of time perception.
University of Brighton.
Johnson-Laird, Philip N. (1983). Mental Models. Cambridge, MA: Harvard University
Cambridge University Press.
of Corpus Linguistics,' (t):
Unpublished MA term paper,
Journal of Neuroscience,
and Postal, Paul M. (19641.
At: Integrated
Theory of Linguistic Descriptions.
Cambridge, MA: MIT Press.
to85 -95.
Heine, Bernd and Kuteva, Tania. (2007). The Genesis of Grammar: A Reconstruction.
Kay, Paul and Fillmore, Charles. (1999). Grammatical constructions and linguistic generalizations: The What's X doing V construction. Language, 75:
Oxford: Oxford University Press. Herskovits, Annette. (1986). Language and Spatial Cognition. Cambridge:
39: 170-210.
Cambridge
University Press. (1988). Spatial expressions and the plasticity of meaning. In B. Rudzka-Ostyn 271 - 9$. Amsterdam: John Benjamins. (ed.), Topics in Cognitive
and McDaniel, Chad. (1978). The linguistic significance of the meanings of basic color terms. Language, 54-3: 610- 46. and Michaelis, Laura A. (forthcoming). Constructional meaning and compositionality. In C. Maietth ► rn. K. von lieusinger, and P. Portlier (eds), Semantics: An I ► ternati ► nal Handbook of Natural Language Meaning. Berlin: Mouton de Gruyter.
gir 366
Kemmerer, David and Tranel, Daniel. ( 21mo). A double disvociation between linguistic and perceptual representations of spatial relationships. Cognitive Neuropsychology, 17: 393- 414.
Kendon, Adam. (2oo4). Gesture: Visible Action as Utterance. Cambridge: Cambridge University Press. Kovecses, Zoltan and Radden, Gunter. (1998). Isietonymy: Developing a cognitive linguistic view. Cognitive Linguistics, q-1: 37-77. Kreitzer, Anatol. (1997). Multiple levels of schematization: A study in the conceptualization of space. Cognitive Linguistics, 8-4: 291-325. Lakotf, George. (1987). Women. Eire and Dangerous Things: What Categories Reveal about the Mind. Chicago: Chicago University Press. (1990). The invariance hypothesis: Is abstract reason based on image-schemas? Cognitive Linguistics, i (I): 39-74. (1993). The contemporary theory of metaphor. In A. Ortony (ed.), Metaphor and Thought, 2nd edition, zoz--st. Cambridge: Cambridge University Press. (1996). Moral Politics: How Liberals and Conservatives Think Chicago: University of Chicago Press. (2.006). Don't Think of an Elephant: Know your Values and Frame. the Debate. Vermont: Chelsea Green Publishing. and Johnson, Mark. (1980). Metaphors We Live By. Chicago: University of Chicago Press. (1999). Philosophy in the Flesh. New York: Basic Books. and Thompson, Henry. (1975). Introduction to cognitive grammar. Proceedings of the 1st Annual Alerting of the Berkeley Linguistics Society, 295-313. Berkeley, CA: Berkeley Linguistics Society. and Turner, Mark. (1989). More than Cool Reason: A Field Guide to Poetic Metaphor. Chicago: University of ( :hicago Press. Landau, Barbara, Dessalegn, Banchianilack, and Goldberg, And Micah. (forthcoming). Language and space: Momentary interactions. In V. Evans and P. Chilton (eds), Lanuage, Cognition and Space: The State of the Art and New Ihrectiems. London: Equinox Publishing. Langacker, Ronald W. (1987). Foundations of Cognitive Grammar: Volume I Theoretical Prerequisites. Stanford: Stanford University Press. (19914). Foundations of Cognitive Grammar: Volume 11 Theoretical Prerequisites. Stanford: Stanford University Press. (199110. Concept, Image, Symbol: The Cognitive Basis of Grammar. Berlin: Mouton de Gruyter. (1999). Grammar and Coriceptuakation. Berlin: Mouton de Gruyter. ( zixxi). A dynaiiiik usage-haled model. In M. Barlow and S. Kemnier (eds), Usagebased Models of Language, 1-64. Stanford, CA: (:SL1 Publications. 4 2008). Cognitive Grammar: A Basic Introduction. Oxford: Oxford University Press.
I eezenherg, Michael. ( zoos). Contexts of Metaphor. OxtOrd: Elsevier Science. Levinson, Stephen. (1983). Pragmatics. Cambridge: Cambridge University Press. -
REFERENCES
REFERENCES
(woo). Presumptive. Meanings: The Theory of Generalized Conversational Implicature. Cambridge, MA: MIT Press.
367
Levinson, Stephen. (2003), Space in Language and Cognition: Explorations in Linguistic Diversity. Cambric* Cambridge University Press. Livingstone, Margaret and Hubei, David. 0988). Segregation of form, color, move111e111 and depth: Anatomy, physiology and perception. Science, 240: 740-9. Lucy, John. (1982). Language Diversity and Thought: A Reformulation of the Linguistic Relativity Hypothesis. Cambridge: Cambridge University Press.
Martin, Alex. ( 2001). Functional neuroimaging of semantic memory. In R. Cabez_a and A. Kingstone (eds), Handbook of Functional Neuroimaging of Cognition, 153-86. Cambridge, MA: MIT Press. (2007). The representation of object concepts in the brain. Annual Review of Psychology, 58: 25-45. Mauk, Michael I). and Boonomano, Dean V. (zoos). The neural basis of temporal processing. The Annual Review of Neuroscience, 27: 307 40. McNeill, David. (1992). Hand and Mind: What Gestures Reveal about Thought. Chicago: University of Chicago Press. Michaelis, Laura. A. (mu). Word meaning, sentence meaning, and syntactic meaning. In H. Cuyckens, R. t)irven, and J. Taylor (eds), Cognitive Approaches to Lexical Semantics, 163-210. Berlin: Mouton de Gruyter. (2004). Type shifting in construction grammar: An integrated approach to aspectual coercion. Cognitive Linguistics, is: 1-67. and Lambrecht, Knud. (1996). Toward a construction-based model of language function: The case of nominal extraposition. Language, 72: 215 47. Miller, George and Johnson-Laird, Philip. (1976). Language and Perception. Harvard: Harvard University Press. Mithen, Steven. (1996). Prehistory of the Mind: A Search for the Origins of Art, Religion and Science. London: Orion Books. Moore, Kevin Ezra. (wooi. Spatial experience and temporal metaphors in Wolof: Point of view, conceptual mapping and linguistic practice. Unpublished doctoral thesis, University of California, Berkeley. (zoo6). Space-to-time mappings and temporal concepts. Cognitive Linguistics, -
-
17-2: 199-244.
Munnich, Edward, Landau, Barbara, and Dosher, Barbara Anne. (2001). Spatial language and spatial representation: A cross-linguistic comparison. Cognition. 81 (3): 171-207.
Murphy, Gregory. (1991). Meaning and concepts. In P. I. Schwanenflugel (ed.) The Psychology of Word Meanings, 11-35. Hillsdale, NJ: Lawrence Erlbaum. (1996). On metaphoric representation. Cognition. 6o, 173-204. Nerlich, Brigitte and Clarke, David 13. (5.007). Cognitive linguistics and the history of linguistics. In D. Geeraerts and H. Cuyckens (eds), Handbook of Cognitive Linguistics, 589-607. Oxford: Oxford University Press. Nlinel, Rafael, Moats, Benjamin, and Truscher, Ursina. (2oo6). Time after time: The psychological reality of the Ego- and Time-Reference-Point distinction in metaphorical construals of time. Metaphor and Symbol, 21: 133-46. and Sviretser. Eve. (2006). With the future behind them: Convergent evidence from Aymara language and gesture in the crusslinguistic comparison of spatial c4 ► nstruals Of time. C‘ignitive Science, 3o: 401-50.
REFERENCES
368
REFERENCES
O'Keefe, John. (1996). The spatial prepositions in English, vector grammar, and the cognitive map theory. In P. Bloom, M. A. Peterson, L. Nadel, and M. E Garrett (cds), La ► guaxe and Space, 277-316. Cambridge, MA: MIT Press. Ornstein, Robert. 019691/1997). On the experience of time. Boulder, CO: Westview Press. ()finny, Andrew. (1993). Metaphor gind Thought, 2nd edition. Cambridge: Cambridge University Press. Paivio, Allan. (1979). Imagery and Verbal Processes. New York: Holt, Rinehart and
Sandra, Dominick. (1998). What linguists can and can't tell you about the human mind: A reply to Croft. Cognitive Linguistics, 9 (4): 361 -478. Schank, Roger. (1975). Conceptual Information Processing. New York: Elsevier. (1982). Dynamic Memory: A Theory of Reminding and learning in Computers and People. Cambridge: Cambridge University Press. and Abelson, Robert. (1977). Scripts, Plans, Goals. and Understanding: An inquiry into Milian Knowledge Structures. Hillsdale, NJ: Lawrence Erlbauni. and Kass, Alex. (1988). Knowledge representation in people and machines. In U. Eco, M. Santanibrogio, and P. Violi (cds), Meaning and Mental Representations,
Winston. (1986). Mental Representations: A Dual Coding Approach. Oxford: Oxford Uni-
181-200. Bloomington, IN: Indiana University Press.
versity Press. Poppet, Ernst. (1994). Temporal mechanisms in perception. In 0. Sporns and
Schiffrin, Deborah. (1994). Approaches to Discourse: Language as Social Interaction.
(cds), Selectionion and the Brain: international Review of Neurobiology,
Searle, John. (1969). Speech Acts: An Essay in the Philosophy of anguag•. ( 'Ambridge: Cambridge University Press.
37: 185-201. - (1978). Time Perception. In R. Held, H. W. Leibowitz, and H.-L. Teuber (cds), I landbook of Sensory Physiology, 713-29. Heidelberg: Springer. Praggleiaz Group. (2007). MIP: A method for identifying metaphorically used words in discourse. Metaphor and Symbol, 22 (1): 1-39. Prinz, Jesse. (2002). Furnishing the Mind: Concepts and their Perceptual Basis. Cambridge, MA: MIT Press. Pulvermtiller, Friedmann. (1999). Words in the brain's language. (2003).
Oxford: Blackwell.
- (1983). intentionality. Cambridge: Cambridge University Press. (119791 1993). Metaphor. In A. Ortony (ed.), Metaphor and Thought, 2nd edition, 83-111. Cambridge: Cambridge University Press. Shepard, Roger N. and Cooper, Lynn A. (1982). Mental Images !mil their Transformations. Cambridge: Cambridge University Press. Shinohara, Kazuko. ( '999). Epistemology of Space and Time. Kwansci, Japan: Gakuin
Behavioral and
Brain Sciences, 22: 253 - 336.
The Neuroscience of Language: On Brain Circuits of Words and Serial
Radden, GUnter. (2003). The metaphor TIME AS SPACE across languages. In N. Baumgarten, C. BOttger, M. Isloti, and I. Probst (eds), Ubersetzen, Interkulturelk ►
el:wren
Sprachen. Festschrift fur Juliane House :um 6o Geburtstag. leitschrift Jive InterkulII relief: Fremdsprachenunterric -ht (online), 8 (2/3): 1 - 14. Recanati, Francois. (2004). Literal Language. Cambridge: Cambridge University Press. Renfrew, Cohn. (2007). Prehistory: The Making of the Human Mind.
When Time is not Space: The social and linguistic construction of time intervals in an Amazonian culture. Journal of Pragmatics.
Pustejovsky, lames (1995). The Generative Lexicon. Cambridge, MA: MIT Press.
kom In unskation, Spracherwerb and Sprach-vermittliing--clas Leber: mit
University Press. Silva Sinha, Vera da, Sinha, Chris, Zinken, Jorg, and Sampaio, Wany. (forthcoming).
Order. Cambridge: Cambridge University Press.
Sinha, Chris and Kuteva, Tania. (1995). Distributed spatial semantics. Nordic Journal of Linguistics, 18: 167-99. Sperber, Dan and Wilson, Deidre. (1995). Relevance: C0111 nicatiom and Cognition, 2nd edition. Oxford: Blackwell. Stalnaker, Richard. (1978). Assertion. In P. Cole (ed.), Syntax and Semantics 9: Pragmatics, 3ts-32. New York: Academic Press. Stern, Josef. (2000). Metaphor in Context. Cambridge, MA: MIT Press.
London:
Sweetser, Eve. ( '999). Compositionality and blending: Semantic composition, in a cognitively realistic framework. In T. Janssen and C. Redeker (eds), Cognitive-
Rosch, Eleanor. (1978). Principles of categorization. In B. Lloyd and F. Rosch (eds),
Linguistics: Foundations, Scope and fethodology, 129-62. Berlin: Mouton de Gruyter.
Weidentield and Nicolson. Cognition and Categorization,
27-48. Hillsdale, NJ: Erlhaum. Reprinted in E. Margolis and S. Laurence (eds), (1999), Concepts: Core Readings, 189-206.
Talmy, Leonard. (2000). Toward a Cognitive Semantics (2 volumes). (:ambridge,
Cambridge, MA: MIT Press. Ruhl, Charles. (1989). On Monosemy: A Study in Linguistic Semantics. New York,
Taylor, John. (wo2). Cognitive Grammar, Oxford: Oxford University Press. (2003). Linguistic Categorization, 3rd edition. Oxford: Oxford University Press.
NY: State University of New York Press. Sacks, Harvey, Schegloff, Emanuel, and Jefferson, Gail. (1974). A simplest systematics for the organisation of turn-taking for conversation. Language, so: 696-735. Sag. Ivan. (2007). Sign-based Construction Grammar. An InfOrmal Synopsis. Unpublished manuscript. Available online at: .
I
369
MA: MIT Press.
Taylor, lawrence I. and Zwaan, Rolf A. (2o09). Action in cognition: The case of language. Language and Cognition, a (t): 45-18. Thompson, Sandra A. (zoos). "Object complements" and conversation. Studies in Language 26 (i): I25-64. ThoMps00-Sihill, Sharon. (2003). Neuroimaging studies of se:multi.. memory: Infer-
ring - how" from "where". Neuropsychoiopii, 41: 280--92.
REFERENCES
REFERENCES
37t) l ► masello,
Michael. (1999). The Cultural Origins of Human Cognition.
Harvard:
Loki Semir
11992).
371
The ‘11.tial image in mind and brain. Scientific American, 267 (3):
68-76.
I larvard University Press. 711eary of Language Acquisition.
and Shipp, Stewart. (1988). The functional logic of cortical connections. Nature,
Trancl, Daniel and Kemmerer, David. (2004). Neuroanatomical correlates of locative
335: 311-17. Ziemke, Tom. (2oo3). What's that thing called embodiment? In Proceedings of the 25th
- (2003). Constructing a Language: A llsage-based
H arvard: Harvard University Press. prepositions. Cognitive Neuropsychology, 21: 719-49. Iratigott, Elizabeth (l ► ss. (mg). On the expression of spatio-temporal relations in language. In I. Greenberg (ed.), Universals of Human Language, 369-400. Stanford, CA: Stanford University Press. and Dasher, Richard. (2oo4). Re gidarity in Semantic Change.
Cambridge:
Cambridge University Press. Turner, Frederick and Piippel, Ernst. (.983). The neural lyre: Poetic meter, the brain Turner, Mark. (1991). Reading Minds: The Study of English in the Age of Cognitive Tyler, Andrea and Evans, Vyvyan. (zoo.). Reconsidering prepositional polvscmy
(zoo;). The Semantics of English Prepositions: Spatial Scenes, Embodied Meaning and Cognition. Cambridge: Cambridge University Press. Vandeloise, Claude. (1990). Representation, prototypes, and centrality. In S. Tsohat403-37.
(1991). Spatial Prepositions: A Case Study from French (trans. Anna R. K. Bosch). Chicago: University of Chicago Press. (1994). Methodology and analyses of the preposition in. Cognitive Linguistics,
comprehension. In B. H. Ross (ed.) The Psychology of Learning and Motivation,
35-62). New York, NY: Academic Press. and Kaschak, Michael P. (2oo8). Language in the brain, body and world. In P. Robbins and M. Aydede (eds), The Cambridge Handbook of Situated Cognition, and Madden, Carol. (2004). Updating situation models. Journal of Experimental Psychology:
Learning, Memory, and Cognition, Jo: 283-8 .
and Radvansky, Gabriel A. (1998). Situation models in language and memory.
Psychological Bulletin, 123: 162-85.
5 (2): *57-84. Varela, Francisco, Thompson, Evan, and Roach, Eleanor. (199.). The Embodied Mind:
Cognitive Science and Human Experience. Cambridge, MA: MIT Press. Vigliocco, Gabriella. Meteyard, Lotte, Andrews, Mark, and Kousta, Stavroula. (2o09). Toward a theory of semantic representation. Language and Cognition,' (z). Walsh, Vincent. (2oo3). A theory of magnitude: Common cortical metrics of time, space and quantity. TRENDS in Cognitive Science, 7 (It): 483-8. Wearden, John and Penton-Voak, Ian. (199s). Feeling the heat: Body temperature and the rate of subjective time, revisited. Quarterly Journal of Experimental Psychology, 48B: 129-41. Whitney, Paul. (1998). The Psychology of language. Roston. MA: Houghton Mifflin. Wierzbicka, Anna. (1988). The Semantics of Grammar. Amsterdam: John Benjamins. (1996). Semantics: Primes and Universals. Oxford: Oxford University Press. Wilson, Margaret. (low). Six views of embodied cognition. Psych ► no ► ic Bulletin and 9 (4): 625-36.
Yu, Ning. (1998). The Contemporary Theory of Metaphor: A Perspective from Chinese. Amsterdam: John Renjamins. Zakay, I tan and Block, Richard A. (.997). Temporal cognition. Current Directions in
Psychological Science,
models. Discourse Processes, 28: 81-8.
.08-81. Cambridge: Cambridge University Press.
London: Routledge.
Review,
(eds), Cognitive Approaches to Lexical Semantics, 447-94. Berlin: Mouton de Gruyter.
(2004). The immersed experiencer: toward an embodied theory of language
networks: The case of over. Language, 77 (4): 724-65.
-
Spatial
iwaan, Rolf A. (1999). Embodied cognition, perceptual symbols, and situation
Science. Princeton, NJ: Princeton University Press.
-
London: Equinox Publishing. Zlatev, Jordan. (1997). Situated embodiment: Studies in the Emergence of Meaning. Stockholm: Gout).
(2oo3). Polysemy or generality? Mu. In H. Cuyckens, R. Dirven, and I. Taylor
and time. /Wiry, 142 (5): 277-309.
zidis (ed.), Meanings and Prototypes: Studies in Linguistic Categorization,
Annual Meeting of the Cognitive Science Society, 1134-9. Mahwah, NJ: Lawrence Erlbaum. Zinken, Jorg. (forthcoming). Temporal frames of reference. In V. Evans and P. Chilton (eds), Language. Cognition arid Space: The State of the Art and New Directions.
6: 12-16.