Top Banner

of 24

On Translating - Contrastive Semantics and Pragmatics

Jun 03, 2018



Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    On Translating 'What Is Said': Tertium Comparationis

    in Contrastive Semantics and Pragmatics*

    K.M. JaszczoltUniversity of Cambridge

    Author's address:

    e!artment of "inguistics#aculty of Modern and Medieval "anguagesUniversity of Cambridge$idg%ic& AvenueCambridge C( )A


    *I would like to thank the audiences of my talks at the International Conference on Germanic and

    Baltic Language Studies, Teaching and Translation at the University of Vilnius and the Second

    International Conference in Contrastive Semantics and Pragmatics at the University of Cambridge for

    their comments on the previous drafts of this paper.


  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    1. Levels of Equivalence

    0his !a!er consists of the follo%ing !arts: it 1i2 introduces one of the main !roblems

    of theoretical contrastive linguistics3 namely the issue of the !latform of com!arison3

    1ii2 !ostulates an ade4uate unit of com!arison for contrastive semantics and

    !ragmatics3 1iii2 discusses the controversy surrounding the semantics5!ragmatics

    boundary and3 finally3 1iv2 assesses the im!act of this dis!ute on translation theory.

    #irst3 some terminological remar&s are needed. 6t has been %idely

    ac&no%ledged that contrastive analysis has to com!rise theoretical linguistic research

    on all of the levels of linguistic study3 as %ell as !sycholinguistic and sociolinguistic

    studies 1cf. $aavaara ,)789 James ,)72. 0he term 'contrastive analysis' stands for

    a!!lied contrastive studies 1henceforth: AC$2 %hich deal %ith the !ractical

    conse4uences of differences bet%een contrasted languages for teaching !ur!oses3bilingual analysis or translation. 0he theoretical linguistic com!onent of AC$ is very

    im!ortant and it gave rise to theoretical contrastive studies 1henceforth: 0C$2. 0C$

    can be !erformed on the level of !honology3 le;icon3 synta;3 semantics3 !ragmatics or

    te;t linguistics. $uch theoretical studies deal %ith a universal category and the %ays

    this universal category is realized in contrasted languages. 6n !honology3 for e;am!le3

    the functioning of !honological features is contrasted 1cf. #isia& et al. ,)nglish3 and self5denigrating in ?olish 1cf. Jaszczolt

    ,))=a3 b2. $o3 %hen %e tal& about an e4uivalence for e;am!le in translation3 it is

    necessary to distinguish syntactic3 semantic and !ragmatic e4uivalence. 6t may %ell be

    the case that there is also an overall e4uivalence %hich is a function of these 1cf.

    Kalisz ,)7,23 but in this !a!er 6 intend to concentrate on semantic and !ragmatic

    e4uivalence and on the difficulty %ith defining the unit of analysis for each level. As 6

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    shall demonstrate3 since semantics does not have a clear boundary %ith !ragmatics

    but rather they overla!3 the translator's decision as to %hether to ado!t semantic or

    !ragmatic e4uivalence %ill not al%ays be !ossible.

    6t has been ac&no%ledged at least since the ,)@s and ida 1,)@823 that

    'faithful' translation means translating the author's intentions3 assum!tions3 rather than

    structures and style 1see Bentzler ,))(3 !. =72. $o3 since ...the equivalencebet%een a

    te;t and its translation can be neither in form nor le;ical meanings3 but only in the

    experience of text receivers 1e eaugrande ,)7: +),23 %e %ant to &no% ho% this

    identity of e;!erience is to be achieved in the multi5layered !rocess of translation.

    0he semantic layer alone %on't do9 since the meaning is often culture5bound3 %e need

    the socio5cultural layer. #or instance3 a Dussian idiomatic greeting '%ith easy steam'

    can only reasonably be translated as a comment '6 ho!e you had a good bath'3 and even

    for non5idiomatic e;!ressions %e have to resolve %hat situation 1state3 event2 thesentence refers to.

    $u!erficially3 contrastive !ragmatics may seem un!roblematic: conversational

    effects should be &e!t constant and one should loo& at the contrast bet%een the %ays

    languages achieve these effects. Eere the obvious areas of study are illocutionary

    forces of e;clamations3 differences in levels and meaning of self5assertion3

    differences in terms of address3 and many other culture5bound !henomena. o%3 in

    order to contrast languages on the !ragmatic level3 one has to decide %hat the

    e4uivalence of contrasted structures on the !ragmatic level means. 0he follo%ingdefinition %as !ro!osed by Fle&sy 1,)782: !ragmatic e4uivalence holds bet%een t%o

    e;!ressions in ",and "+if they can be used to !erform the same s!eech act in these

    t%o languages. All %e have to do no% is to &ee! s!eech acts steady and loo& at the

    sets of strategies used in ",and "+to !erform these acts. 0he !roblem arises as to

    %hether to admit indirect s!eech acts. As has been fre4uently !ointed out in the

    !ragmatic literature3 the directindirect distinction for s!eech acts is untenable.

    $!erber and Gilson 1,))=: +8=2 give the follo%ing e;am!le:

    1(2 0he %eather %ill be %armer tomorro%.

    182 0he s!ea&er is !redicting that the %eather %ill be %armer tomorro%.

    $entence 1(2 can successfully function as a !rediction %ithout the s!ea&er's intending

    to communicate the information in 182. 6ndeed3 182 need not be recovered by the

    hearer at all for 1(2 to function as a !rediction. Fn the other hand3 in 1=23 the act of

    bidding has to be communicated3 either directly or by inference.

    1=2 16 bid2 t%o no trum!s.

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    $!eech acts are either institutional3 social3 li&e bidding in bridge or than&ing3 or

    !erformed %ithout being recognized as such3 e.g. %arning3 threatening3 or their

    category can be recognized and universal3 as in the case of saying3 telling and as&ing.

    Also3 as is %ell &no%n from the colla!se of the !erformative hy!othesis and the literal

    force hy!othesis 1"evinson ,)7(23 there is no reliable correlation bet%een the

    sentence ty!e and s!eech act ty!e on the one hand3 and the meaning of the

    !erformative verb and the ty!e of s!eech act on the other 1see also $!erber H Gilson

    ,))=3 !!. +8@5+8

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    inference by relevance theorists. $o3 they can be regarded as either semantic or

    !ragmatic3 unli&e !articularized conversational im!licatures %hich are conte;t5

    de!endent and certainly arrived through !ragmatic !rocesses of inference.

    earing this difficulty in mind3 6 suggest trying the follo%ing hy!otheses:

    1A,2 $emantic e4uivalence is the e4uivalence of what is said.

    1A+2 ?ragmatic e4uivalence is the e4uivalence of %hat is im!licitly communicated.

    0hese definitions are not very informative as they stand. $ince the !roblem of the

    fuzzy boundary bet%een semantics and !ragmatics is unresolved3 %e are only !ushing

    the terminological difficulty one ste! on to the e4ually !roblematic notions of '%hat is

    said' and '%hat is im!licated'. ut there is an advantage to be gained. Ghat is said and

    %hat is im!licated have been subect to e;tensive studies and heated debates in thelast decade 1Carston ,)773 ,))7a9 DIcanati ,)7)3 ,))(9 ach ,))8a3 b9 "evinson

    ,))=3 +9 Jaszczolt ,)))a3 b2. 0he starting !oint to the debate is the observation

    that Brice seriously underestimated the role of !ragmatic !rocesses in establishing the

    re!resentation of the utterance of the sentence %hich can be subect to the !rovision

    of truth conditions3 i.e. the !ro!ositional form. 6n addition to reference assignment

    and disambiguation %hich he ac&no%ledged3 there are many !rocesses of enrichment

    of the !ro!osition3 or !atching u! of the incom!lete !ro!ositional form3 %hich have to

    be !erformed in order to arrive at the relevant3 truth5evaluable re!resentation. #ore;am!le3 the sentential connective 'and' can be enriched to include the indication of

    tem!oral se4uence or causal conse4uence and this enrichment is relevant to the truth

    conditions of the !ro!osition3 %hich can be tested3 for e;am!le3 by !lacing the

    sentences in the sco!e of logical o!erators such as negation or im!lication. 6 shall

    return to the tests for contributions to truth conditions in $ection @. $uffice it to say

    that what is saidand what is implicatedseem to constitute a !romising de!arture for

    the im!rovements on the definitions of semantic and !ragmatic e4uivalence in

    translation.0o sum u!3 there have been various !ro!osals of ho% to dra% the boundary

    bet%een %hat is said and %hat is communicated. 0he main stand!oints can be

    summarized as follo%s:

    1i2 $ome sentences are semantically ambiguous. 0his traditional

    vie% %as advocated3 among others3 by Dussell.

    1ii2 0here is no semantic ambiguity. 0he differences in meaning bet%een

    the t%o 1or more2 readings can be attributed to im!licated information.

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    0his is Brice's !ostulate of unitary semantics com!lemented %ith

    conversational im!licatures.

    1iii2 0here is no semantic ambiguity and the differences in meaning belong

    to %hat is said rather than %hat is im!licated. $emantics is

    unders!ecified as to these as!ects of meaning. 0he !rocess of

    su!!lementing the semantic form %ith !ragmatic information is called

    com!letion 1saturation2 and e;!ansion 1strengthening9 cf.

    ach ,))8a3 b9 DIcanati ,)7)2.

    1iv2 0here is no semantic ambiguity3 semantics is unders!ecified3 but the

    differences in meaning belong to the middle level located bet%een

    %hat is said and %hat is im!licated. 0he main !ro!onent of this vie% is

    "evinson 1e.g. ,))=3 +2 and the idea is that there are default

    meanings arrived at through the three heuristics3 uantity36nformativeness and Manner3 corres!onding to his re5arrangement of

    Brice's ma;ims of conversation.

    6n addition3 in my other %or& 1e.g. Jaszczolt ,))

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    is im!licitly communicated. Ghat is said is a !ro!osition3 an entity %hich is either

    true or false. Eence3 %e can try the follo%ing:

    1A',2 0%o sentences are semantically e4uivalent if they corres!ond to the same


    1A'+2 0%o sentences are !ragmatically e4uivalent if they render the same set of

    im!licatures3 triggered by rules of conversational inference+.

    'Corres!onding to' the same !ro!osition can be investigated by com!aring the

    !ro!ositional form and3 usually3 the truth conditions.( A'+is obviously conte;t5

    de!endent. ut3 so is A',. #irstly3 there are intensional conte;ts such as belief conte;ts

    1'A believes that s.'2 in %hich the mode of !resentation of the referent contributes

    to the truth conditions 1cf. $chiffer ,))+3 Jaszczolt ,))7a2. $econdly3 not allsentences e;!ress full !ro!ositions3 and not all !ro!ositions e;!ressed by sentences

    are sufficiently informative to be ta&en as %hat is said %ithout enrichment3 to 4uote

    only 1@2 and 1

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    out!ut of syntactic rules %ill not suffice to dra% the boundary bet%een semantic and

    !ragmatic e4uivalence.

    $o3 all in all3 6 shall tal& about what is saidinstead and its !ragmatic

    constituents 1or: '%hat is im!licitly stated'2 com!leting and e;!anding3 res!ectively3

    the logical form of the sentence.86n order to !reserve the definition of semantic

    e4uivalence in A,3 one has to resolve these difficulties of enrichment and com!letion.

    A,is %orth !reserving because there does not seem to be a more ade4uate definition

    of sentence meaning than that !rovided by truth5conditional semantics. $o3 %hat is

    said is !artly semantic and !artly !ragmatic3 as is %hat is im!licated: im!licatures

    have their o%n logical forms and hence their o%n semantics3 but they are the !roduct

    of !ragmatic !rocesses of inference.

    3. Cultural ationalismo%3 in order to acce!t A,and A+%e need to s!ecify one further theoretical

    commitment. amely3 %e acce!t thatpropositionsare recognizable cross5culturally3

    that is3 that it is not the case that !eo!le of different cultures live in different %orlds.

    6n other %ords3 %e acce!t cultural rationalism 1and realism2 against cultural relativism

    and in this %ay give !ragmatic e4uivalence access to a universally3 cross5culturally

    valid theory of im!licature of any !ost5Bricean flavour. ut this commitment is not

    driven by the need to ease the tas& of defining !ragmatic e4uivalence3 this move

    %ould be circular. 0he ustification comes from the strength of the inde!endentarguments for realism. ?resenting them is the tas& to %hich 6 shall no% turn.

    At first glance3 it may seem that relativity is the dominant orientation in

    linguistics3 brought to the fore by cognitivists 1e.g. "a&off ,)7

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    communication normally slightly transforms the message. o%3 e;!laining cultural

    beliefs boils do%n to e;!laining the distribution of ideas in !o!ulation3 e;!laining

    ho% individual !rocesses of ado!ting an idea contribute to its dissemination. 6t is

    essentially a ar%inian a!!roach3 lin&ed to these of Cavalli5$forza and a%&ins3

    %here the ar%inian idea of selection is a!!lied to culture. $!erber calls it a

    'naturalistic a!!roach' because social issues are a!!roached through the study of

    cognition3 or through !sychological !rocesses of individuals %ho combine to form a


    6deas !ro!agate and change3 sometimes they are understood3 sometimes

    misunderstood or half5gras!ed. $ince memory and communication transform original

    ideas3 the ideas are stored as representations of the world or of the spea!er"s

    representations=. 6n other %ords3

    $ocial5cultural !henomena are 1...2 ecological !atterns of !sychological !henomena.

    $ociological facts are defined in terms of !sychological facts3 but do not reduce to

    them. $!erber 1,))@: (,2.

    0he idea is that !eo!le have re!resentations of situations in the form of memory3

    belief3 or intention. De!resentations can also be !ublic for e;am!le %ritten te;ts

    %hose function is communication are such !ublic re!resentations.

    o%3 some of these re!resentations are re!eated3 communicated more oftenthan others and become cultural re!resentations of a social grou!. 0hese are shared

    beliefs3 norms3 myths3 techni4ues3 or classifications. $uch re!resentations are

    conveyed by !roducing another re!resentation3 %hich is either a descriptionor an

    interpretation of the original re!resentation 1cf. the diagram in $!erber H Gilson

    ,)7=3 !. +(+2. 6nter!retation occurs %hen3 for e;am!le3 the receiver has to e;!and on

    the content of some tribal belief in order to ma&e it understandable.

    6n other %ords3 %e ma&e !artial and often s!eculative inter!retations in order

    to understand others3 es!ecially re!resentatives of different cultures. 0hen %e

    attribute beliefs3 desires and intentions to !eo!le or communities in a %ay %hich

    ma&es them loo& rational. And that is %hy %e need inter!retations:

    0he anthro!ologist must 1...2 go beyond mere translation: only then can she ho!e to

    understand %hat she hears3 and thus be genuinely able to translate it. $he must

    s!eculate3 synthesize3 reconce!tualize. $!erber 1,))@: ()2.

    ?utting aside the status of the modal verbs in the 4uotation3 %e can summarize

    $!erber's !ro!osal as follo%s. 0o understand cultural !henomena3 the observerhearer=So, in discussing culture we have to access cognitive psychology.

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    can either 1i2 generalize over them through ma&ing an inter!retation of one

    !henomenon %ider and %ider3 a!!licable to other !henomena9 1ii2 loo& for common

    !atterns3 themes or relations bet%een these !atterns in a structuralist manner9 1iii2

    sho% ho% a cultural !henomenon is beneficial for the grou! by %ay of a functional

    e;!lanation3 or 1iv2 ans%er the 4uestion %hy some re!resentations are more

    successful than others and become shared3 cultural re!resentations3 i.e. undergo

    e!idemiology of re!resentations 1cf. $!erber ,))@3 !. =2.

    Ghat is of interest for the search for a translation e4uivalence is the idea that

    re!resentations are not re!roductions of %hat %as said but rather construction by the

    hearer of hisher o%n thoughts %hich are normally related rather closely to the

    thoughts re!resented by %hat is said 1cf. $!erber ,))@3 !. =72. $o3 if a tribesman

    strives to understand a re!resentative of a different culture3 heshe %ill store some of

    the re!resentations in a semi5understood %ay3 ho!ing to fill them in %ith the intendedmeaning at a later stage. #or e;am!le3 if a tribesman says that there are dragons in the

    %oods3 %e ta&e this !ro!osition literally and assume that it is stored in the native's

    memory not in a fully decom!osed %ay but rather semi5!ro!ositionally3 as a semi5

    understood idea3 as a belief about some re!resentation3 rather than about a

    !erce!tually verifiable fact. Ge do that rather than um!ing to the conclusion that the

    %orlds of the t%o cultures are incom!atible. 6t is a metare!resentation3 a

    re!resentation of someone else's re!resentation. $uch rationalism allo%s for

    com!aring !ro!ositions cross5culturally.Dationalism ac4uires additional su!!ort from $!erber's account of conce!t

    formation %hich is com!atible %ith e!idemiology:

    ...%e have an innate dis!osition to develo! conce!ts according to certain schemas.

    Ge have different schemas for different domains: our conce!ts of living &inds tend to

    be ta;onomic9 our conce!ts of artefacts tend to be characterized in terms of function9

    our conce!ts of colour tend to be centred on focal hues9 and so on. Conce!ts %hich

    conform to these schemas are easily internalized and remembered. "et us call them#asic concepts. A large body of basic conce!ts is found in every language. Ff course3

    basic conce!ts differ from one language to another3 but they do not differ very much.

    0he basic conce!ts of another language tend to be com!aratively easy to gras!3 learn

    and translate. $!erber 1,))@: @)2.

    $o3 there are universal schemas for basic conce!ts3 rules for forming re!resentations

    and universal !atterns of their s!read. 0he !oint of difficulty for delimiting !ragmatic

    e4uivalence seems to be located in assessing the t$peof re!resentation formed by the


  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    o%3 %e can summarize $!erber's account by saying that humans can

    re!resent not only the sensory in!ut and e;!erience3 but also someone else's and their

    o%n re!resentations and mental states. 0his is called metarepresentingand allo%s us

    to have3 for e;am!le3 an attitude of disbelief or doubt to%ards a re!resentation.

    Metare!resenting also allo%s us to store re!resentations %hich are not fully

    understood. $ome such half5understood ideas are not ste!s to%ards full understanding

    but rather create mysteries %hich s!read in the community. De!resentations %hich fit

    best %ith other mental re!resentations3 and %hich cannotbe inter!reted3 are most

    successful. 0hey lead to %hat %e see as irrational cultural beliefs. ut3 %hen %e

    consider them to be semi5understood ideas3 then they are only a!!arently irrational

    1cf. $!erber ,))@3 !. thio!ia comes to $!erber and as&s him to &ill a dragon %hose heart is made of gold3%ho has a horn on his nec&3 is golden all over and lives some%here nearby. $!erber

    believes the tribesman to be a sensible3 res!ectable !erson3 but ho% %ould a sensible

    !erson believe that there %as a dragon living %ithin %al&ing distanceL 0his is an

    e;am!le of an a!!arently irrational belief. Many anthro!ologists %ould say that

    !eo!le of different cultures live in different %orlds3 that beliefs in dragons may a!!ear

    rational to some cultures9 others advocate symbolism. $!erber reects these vie%s.

    6nstead3 he believes in rationalism. Ee says that !eo!le can have various ty!es and

    degrees of commitment to beliefs 55 beliefs are not all held in the same %ay3 and sothe criteria of rationality can also differ. 1As if to confirm the claim by intuitions3

    $!erber's immediate res!onse %as saying a!!rehensively that he hasn't got a gunN2

    elief has normally been regarded3 at least since Dussell3 as a ty!e of

    !ro!ositional attitude3 an attitude to an entity %hich is either true or false. ut:

    ...the obects of our '!ro!ositional attitudes'3 the ideas %e hold or other%ise entertain3

    are not al%ays strictly !ro!ositional in character. Just as it %ould be mista&en to

    define 's!ea&ing' as 'utteringsentences'3 it is mista&en3 6 suggest3 to define thin&ing interms of attitudes to propositions: many of our utterances do not match sentences but

    semi5grammatical strings9 similarly3 many of our thoughts are %hat %e might call

    semi5!ro!ositional3 they a!!ro;imate but do not achieve !ro!ositionality. 1...2 6f it

    %ere true that the obects of belief necessarily %ere !ro!ositions3 then %e could only

    believe ideas %hich %e fully understand. 6 am arguing that %e can also hold as beliefs

    incom!letely understood ideas. $!erber 1,)7=: =,23

    such as the belief about a golden dragon. $o3 there are !ro!ositional re!resentations

    %hich corres!ond to fully understood ideas3 and semi5!ro!ositional re!resentations

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    %hich corres!ond to ideas %hich are not fully understood. 0hese are conce!tual

    re!resentations %hich do not identify a uni4ue !ro!osition some conce!ts are

    missing. $uch semi5!ro!ositional re!resentations are useful for storing information

    %hich is not fully understood. 0hese metare!resentations are not factual beliefs3 but

    re!resentational beliefs. Eence3 there are four !otential classes of beliefs: factual

    beliefs %ith !ro!ositional content %hich are rational3 observation5based9 factual

    beliefs %ith semi5!ro!ositional content %hich do not normally occur9

    re!resentational beliefs %ith !ro!ositional content 1metare!resentations3 including

    also scientific assum!tions not fully believed29 and finally re!resentational beliefs

    %ith semi5!ro!ositional content3 such as religious beliefs3 mysteries 1cf. $!erber

    ,)7=3 !. =72. Cultural beliefs are such re!resentational beliefs.

    6ncluding its recent develo!ments3 $!erber's e;!lanation of cross5cultural

    differences and similarities is as follo%s: there are t%o ty!es of beliefs3 intuitive1factual2 and reflective 1re!resentational2@. 6ntuitive beliefs are based on innate3

    universal mechanisms3 based on !erce!tion and inference3 and hence are similar

    across cultures. Deflective beliefs are not factual3 not basic3 they are3 so to s!ea&3 'in

    4uotes': they are believed because they are embedded in intuitive beliefs. 0hey may be

    half5understood and leading either to scientific &no%ledge or to mysteries3 often

    religious. 0hese mysterious beliefs are the ones that vary cross5culturally and may

    even a!!ear irrational from another culture's !oint of vie% 1cf. $!erber ,))@3 !!. ),5

    )+2. 6n other %ords3

    ...e;!laining cultural beliefs3 %hether intuitive or reflective3 and if reflective3 %hether

    half5understood or fully understood3 involves loo&ing at t%o things: ho% they are

    cognized by individuals and ho% they are communicated %ithin a grou!9 or to !ut it

    in the form of a slogan: Culture is the !reci!itate of cognition and communication in a

    human !o!ulation. $!erber 1,))@: )

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    Ge can suggest tentatively at this stage that %hen a belief is semi5

    !ro!ositional3 and a fortiori the sentence uttered does not have the bac&ing of a

    !ro!osition3 the translator's tas& is not to elaborate on the degree of a%areness of the

    !ro!osition9 this is the anthro!ologist's tas&. 0he translator has to substitute a

    sentence in the target language %hich is a successful re!resentative of this

    !ro!osition3 %hether it is understood fully or !artially by the s!ea&er

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    non5literally 1ach ,))8a3 b2. $imilarly3 1

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    dictated by the fact that all this information contributes to the truth conditions of the

    original !ro!osition. 0his enriched !ro!osition is called by ach impliciture:

    6m!licitures go beyond %hat is said3 but unli&e im!licatures3 %hich are additional

    !ro!ositions e;ternal to %hat is said3 im!licitures are built out of %hat is said.

    ach 1,))8a: +

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    (. &hich Equivalence)

    Coming bac& to the definition of semantic e4uivalence in translation3 %e %ant to

    ensure that what is saidis !reserved. Eence3 if it is not obvious to the addressee of the

    translation that 172 communicates an e;!anded !ro!osition3 this e;!ansion should be

    e;!licitly stated 1and this a!!lies both to s!o&en and %ritten discourse2. 0he test can

    be the obvious falsehood or the blatant truth of the literal meaning. $imilarly %ith

    !ro!ositional radicals3 if the com!letion is not obvious in the translation %hereas it is

    obvious in the original discourse3 it has to be e;!licitly stated. $emi5!ro!ositional

    beliefs have !ro!ositional form of the corres!onding sentences %hich translates

    across cultures and hence can be left unattended to: a s!ea&er %ho holds a semi5

    !ro!ositional belief communicates a !ro!osition %illingly3 &no%ing that it does not

    corres!ond to a !ro!ositional belief but rather to a belief in something being the case%here this 'something' is stored for !ossible future understanding.

    0he solution %ould be sim!le indeed if not for im!licatures. y saying 172 the

    s!ea&er communicates a range of im!licatures3 the most obvious ones being 0here is

    no need to %orry3 0he %ound is not serious3 0he %ound %ill heal 4uic&ly3 etc.

    Attaining !ragmatic e4uivalence re4uires !reserving these obvious im!licatures. And

    in order to do so3 the semantic e4uivalence may have to be com!romised: sentence 172

    may have to be re!laced %ith 1,(2 or another set !hrase dictated by !articular socio5

    cultural3 linguistic conventions.

    1,(2 6t's not a big deal3 ?eter.

    $o3 the translator has to uggle these t%o levels of e4uivalence.

    Butt 1,)),23 %ho advocates the vie% that translation is !art of communication3

    observes that translation in communication theory has normally been treated through

    the 'code model' of communication3 %here communication is a matter of encoding and

    decoding messages. ut the subse4uent suggestion of $!erber H Gilson's 1,))=2 thatinference is a more essential element of communication should3 according to Butt3

    !ercolate to translation theory 1cf. Butt ,)),3 !!. +,5++2. 0he idea is3 ho%ever3 not

    com!letely ne%. Ge can trace it bac& to ida 1,)@83 ida H 0aber ,)@)2 %ho

    develo!ed the notion of d$namic equivalence:

    ynamic e4uivalence is 1...2 to be defined in terms of the degree to %hich the

    rece!tors of the message in the rece!tor language res!ond to it in substantially the

    same manner as the rece!tors in the source language. 0his res!onse can never be

    identical3 for the cultural and historical settings are too different3 but there should be a

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    high degree of e4uivalence of res!onse3 or the translation %ill have failed to

    accom!lish its !ur!ose. ida H 0aber 1,)@): +82.

    0his %as a!!lied to ible translation but seems to be %idely a!!licable.

    Coming bac& to the stand!oints on the %hat is said%hat is communicated

    debate3 it seems that the efault $emantics 1o!tion 1v22 gains some su!!ort from the

    theory and !ractice of translation in that %hen the inferences are obvious3 they need

    not be translated. And being obvious includes being natural3 !rimary3 unmar&ed. 6t

    has to be borne in mind in the translating e;ercise is that not all inferences in

    conversation 1or %ritten te;t2 are !ragmatic. 0here are default inter!retations3 the

    level of so5called utterance5ty!e meaning3 located bet%een sentence meaning and

    utterance5to&en meaning:

    0his third layer is a level of systematic !ragmatic inference based noton direct

    com!utations about s!ea&er5intentions3 but rather ongeneral expectations a#out how

    language is normall$ used. 0hese e;!ectations give rise to !resum!tions3 default

    inferences3 about both content and force... "evinson 1,))=: )(2.

    Eence3 some e;!ansions and com!letions may !roceed through a default

    inter!retation3 guaranteed by socio5cultural and linguistic conventions and the

    !henomenon of !ragmatic constituents is not as uniform as it %ould a!!ear to be:sometimes %e need !ragmatic !rocesses3 sometimes conventions and defaults3 and

    sometimes merely semantics to arrive at %hat is said. As 6 argue else%here 1Jaszczolt


  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    All in all3 achieving com!lete inter!retive resemblance may not be attainable3

    or at least there may not be a steadfast rule as to ho% to achieve it. evices may have

    to be changed even %ithin one te;t 1cf. Butt ,)),3 !!. ,7=5,7@2. 6n general3 if

    translation aims at successful communication3 %e may not be able to distinguish

    semantic and !ragmatic e4uivalence but only communicative and cognitive. 0he

    boundary may only be constituted by the distinction bet%een these 'im!licatures' that

    can be called e;!ansions and com!letions of the original !ro!osition3 and genuine

    im!licatures %hich have their o%n logical forms. $ince im!licatures force a decision

    as to %hether to ma&e them e;!licit or ho!e they %ill be derived3 there may be a

    boundary3 if im!licatures that !ertain to e;!ansion and com!letion are treated as a

    se!arate !henomenon of im!liciture3 e;!licature3 or !ragmatic constituents of %hat is

    said3 in other %ords: as the domain of semantics. And this is the idea 6 follo%ed in

    this !a!er. 6f translation is communication3 then a theory of communication %ill dofor defining e4uivalence. 0hat is %hy %e have to distinguish e;!licit from im!licit

    communication: %hat is said and %hat is im!licated. 0his is ho% %e can assess

    motivations for choices e.g. bet%een direct and indirect translation. 'irect' means

    here literal3 !reserving not only meaning3 but also the %ay it %as e;!ressed 1cf. Butt

    ,)),3 !. ,+=29 'indirect' means free translation.

    0he idea that translation is communication and includes %hat is said and %hat

    is im!licated has very res!ectable roots in the !hiloso!hy of language3 notably in the

    %or& of Dudolf Carna!3 dating bac& to the ,)8s. Carna! analysed semantic meaningas follo%s: intralingually3 t%o sentences are e4uivalent if both are true or both are not

    true3 and the individual e;!ressions must stand for the same thing to be e4uivalent

    1Carna! ,)8+3 ,)84uivalence can be either material3 %here there is e4uivalence

    of obects and facts3 or logical3 %here the truth of the sentence can be deduced from

    the semantic rules. Ge also need the e4uivalence of intensions3 i.e. the %ay of

    thin&ing about obects and events3 not only the e4uivalence of events. #or this

    stronger e4uivalence the identity of structure is re4uired3 i.e. intensional isomor!hism.

    #or e;am!le3 'three' and 'the s4uare root of nine' are "5e4uivalent but do not have thesame intensional structure. Carna!'s a!!roach failed because it %as behaviouristic: he

    defined mental attitudes such as belief thatpas dis!ositions to assent to sentences

    intensionally isomor!hic %ithp3 %hich is obviously inade4uate. ut this is ho%

    intensions gave rise to !ossible %orlds and to contem!orary semantics: there are

    intensional conte;ts in %hich the identity of truth conditions %ill not guarantee

    semantic e4uivalence3 such as belief and other !ro!ositional attitude conte;ts. ut this

    is a to!ic for a different occasion 1see Jaszczolt ,))

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    "et us come bac& to the fuzziness of the boundary bet%een semantics and !ragmatics.

    0he category of semantic underdetermination3 sentences e;!ressing !ro!ositional

    radicals3 is some%hat eclectic. >;am!les 1,82 and 1,=2 4uoted after ach 1,))8a2 do

    not seem to be clearly underdeterminate3 as o!!osed to 1,@2 %hich is very obviously


    1,82 L$he is leaving from %hereL

    1,=2 L$he %ants a ta;i to do %hat %ithL

    1,@2 6 am too tired to do %hatL

    0he matter %ould be easier to handle if all !ro!ositional radicals had slots in their

    syntactic re!resentation to be filled3 as in the case of elli!sis in 1,

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    1+2 0hey had a baby and got married.

    1+,2 + Q + R 8

    1++2 6 li&e cheese and %ine.

    0he order of conuncts is free in 1+,2 and 1++23 but it is fi;ed in 1,)2 and 1+2 %here

    'and' reads as 'and then'. Changing the order changes the truth conditions3 as can be

    tested by 1i2 the functional inde!endence !rinci!le mentioned above or by 1ii2

    embedding the sentences in the sco!e of logical o!erators such as negation and

    im!lication 1'if...then'2:

    1+(2 6f the old &ing died of a heart attac& and a re!ublic %as declared $am %ill be

    ha!!y3 but if a re!ublic %as declared and the old &ing died of a heart attac&3

    $am %ill be unha!!y.

    1cf. Cohen ,)

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    %ithin one te;t. ut delimiting tertium comparationis3 the !latform of reference3 is a

    necessary !re5condition for ma&ing such choices.

    +. Concluding emar!s

    As is %ell &no%n3 some !hiloso!hers of language claim that obective translation is

    im!ossible. uine 1,)@2 ma&es translation de!endent on a 'translation scheme' or a

    theory,,. ut it has been %idely acce!ted that !ro!ositions3 units of meaning3 are

    language5inde!endent. $o3 they constitute3 in our terms3 tertia comparationes3 a

    universal semantic category3 realized as sentences of !articular languages. $entences

    have their o%n logical forms %hich may re4uire further develo!ment into a full

    !ro!ositional form. 0his !ro!ositional form is the unit of semantic e4uivalence in

    translation. Ghen the aim is to !reserve communicative e4uivalence3 the decisions

    may have to be made as to %hether it is best done by !reserving the semantic or!ragmatic e4uivalence3 the first relying on the syntactic e4uivalence and sameness of

    logical form.

    $emantic e4uivalence is not attainable for the com!arison of grammatical

    constructions. 6t has to be te;t5bound rather than systematic 1involving the

    com!arison of grammatical and le;ical systems3 cf. Krzeszo%s&i ,)7)3 Jaszczolt

    ,))=a3 !. +23 ust as semantics has to incor!orate the !ragmatic constituents of %hat is

    said. 6t %as also suggested that !reserving semantic and !ragmatic e4uivalence may

    not be com!atible: the decision has to be made as translation !rogresses. 0hisa!!roach is ta&en to be a!!licable both to s!o&en and %ritten discourse.


    "...manuals for translating one language into another can be set up in divergent ways, all compatiblewith the totality of speech dispositions, yet incompatible with one another." Quine (1960: 27).

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics



    ach3 K. ,))8a. '$emantic slac&: Ghat is said and more'. 6n: $.". 0sohatzidis 1ed.23

    ,oundations of Speech -ct heor$. "ondon: Doutledge. +@

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    Jaszczolt3 K.M. ,)))a. 'efault semantics3 !ragmatics3 and intentions'. 6n: K. 0urner1ed.23 he Semantics9Pragmatics 'nterface ,rom ifferent Points of:iew. F;ford: >lsevier $cience. ,))5+(+.

    Jaszczolt3 K.M. ,)))b.iscourse ;eliefs and 'ntentions Semantic efaults andPropositional -ttitude -scription. F;ford: >lsevier $cience.

    Jaszczolt3 K.M. +. '0he default5based conte;t5de!endence of belief re!orts'. 6n:K.M. Jaszczolt 1ed.23he Pragmatics of Propositional -ttitude eports.F;ford: >lsevier $cience. ,@)5,7=.

    Jaszczolt3 K.M. forthcoming. 'Against ambiguity and unders!ecification'.8ournal ofPragmatics.

    Kalisz3 D. ,)7,. he Pragmatics Semantics and S$ntax of the English Sentences with'ndicative hat Complements and Polish Sentences with ;!erimental !ragmatics and %hat is said: A res!onse

    to Bibbs and Moise'. Cognition@). ((.A. ,)@8. oward a Science of ranslating &ith Special eference toPrinciples and Procedures 'nvolved in ;i#le ranslating. "eiden: >.J. rill.

    ida3 >.A. H C.D. 0aber. ,)@). he heor$ and Practice of ranslation. "eiden: >.J.rill.

    Fle&sy3 G. ,)78. '0o%ards !ragmatic contrastive analysis'. 6n: J. #isia& 1ed.23Contrastive Linguistics Prospects and Pro#lems. erlin: Mouton. (8)5@8.

    uine3 G.T.F. ,)@. &ord and >#?ect. Cambridge3 Mass.: M60 ?ress.DIcanati3 #. ,)7). '0he !ragmatics of %hat is said'./ind 0 Language 8. +)=5(+).

    De!rinted in $. avis 1ed.23Pragmatics - eader. F;ford: FU?. )

  • 8/11/2019 On Translating - Contrastive Semantics and Pragmatics


    van eemter3 K. ,))7. 'Ambiguity and idiosyncratic inter!retation'.8ournal ofSemantics ,=. =5(@.

    Gierzbic&a3 A. ,)),. Cross5Cultural Pragmatics he Semantics of 7uman'ntraction. erlin: Mouton.