- Количество слайдов: 35
THE DE-ACCENTING OF GIVEN INFORMATION: A COGNITIVE UNIVERSAL? ALAN CRUTTENDEN
DISCOURSE TOPIC variations in pitch height may be used for discourse topic delineation (i) “non-recoverable from previous contexts, and (ii) persistent in the following context” BEGINNING OF DISCOURSE TOPIC A new discourse topic regularly produces a higher pitch on the first acccented syllable after the change of topic compared with the pitch of the last accented syllable of the previous topic (Brown, Currie & Kenworthy 1981; Sinclair & Brazil 1982; Brazil 1985, who refers to higher “key”). Additionally, the first intonational phrase of a new discourse topic can be marked by a jump to a higher mean fundamental frequency and the use of a wider pitch range
END OF DISCOURSE TOPIC An extra-low fall, often accompanied by a longer pause and lower amplitude, may mark the end of an intonational phrase which corresponds with the end of a discourse topic (see Lehiste 1979, 1982) These uses of pitch height for topic delineation are putatively universal.
DISCOURSE CONTINUITY/COHESION: RISING TONES A second way in which prosody is relevant to discourse continuity is in the use of rising tones. All cross-linguistic evidence from the world’s languages suggests that rises are used for a set of meanings which can be labelled “open” (Cruttenden 1981, 1986). This set of meanings includes polar question, non-finality, implication, doubt and request. The meaning that is particularly relevant here is non-finality. Hirst & Di Cristo (1998), in a survey of the intonation of twenty languages, state: “Like questions, unfinished utterances or continuatives are commonly pronounced with rising intonation in many languages. In fact a raised final pitch is perhaps even more common for continuatives than for questions”
topics can with reasonable certainty be said to be delineated by pitch height and within such topics rises may indicate continuity UNIVERSALS ? But beyond this we are very much into a speaker-optional area; different speakers may actually choose to use different numbers of continuity-indicating rises. Alternatively, differences in proportion of rise usage can be related to cross-linguistic or dialectal variation (see Cruttenden 1995). There is a much discussed tendency among young speakers of Australian and New Zealand English to use a very high proportion of rises (see, for example, Guy, Horvath, Vonwiller, Disley & Rogers 1986; Britain & Newman 1992). Such rises appear to be used as a “check” on speaker-attention (Guy & Vonwiller 1984).
DISCOURSE COHESION: SENTENCE A third way in which prosody contributes to discourse cohesion concerns the indication of the topic or the setting of a sentence. We are of course here specifically using the term “sentence topic” as opposed to the term “discourse topic” which was dealt with above DEACCENTING OLD OR GIVEN INFORMATION Fourthly, prosody is also related to discourse cohesion by the deaccenting of old or “given” information; a summary of some of the previous work in this area is given in section 2 and the rest of this chapter investigates this phenomenon in a number of European languages.
CROSSLINGUISTIC INVESTIGATION it must be made clear that none of the above relationships between discourse and intonation has ever been investigated crosslinguistically. There has always been an unwritten assumption that such things are in some way or at least to some degree common to all languages, that they represent some sort of cognitive universal. While this has never been stated explicitly, it has regularly been implied by the fact that writers rarely feel the need to say that of course their findings are limited to English or Dutch or whatever. It does indeed look from the limited number of languages separately studied for each of these phenomena that there is an element of universality but at the same time there are clearly variations. So the use of rising tones for non-finality looks to be a universal tendency; yet the need to mark non-finality or finality clearly varies across languages and dialects. A higher mean pitch for topic-beginnings may well be universal but the actual extra height of such extra mean pitch may vary from language to language. The delineation of a sentence topic by the use of a separate intonational phrase may occur more regularly in some languages than in others. The deaccenting of old or ‘given” information may not be as obligatory in some languages as it is in English.
2. De-accenting and given information The remainder of this article tries to establish a methodology for investigating this last phenomenon cross-linguistically and tentatively presents some first results. Halliday 1967: “given” and “new”; The “tonic” (alias “nucleus”) is the primary accent in an intonational phrase),
Because there are problems with the independent assessment of givenness (i. e. one which does not become circular by deciding on givenness by reference to de-accenting) in particular intonational phrases, it is important to use an unarguable case of givenness in any cross-linguistic investigation. Repetition would seem to be such a case (see Nooteboom & Kruyt 1987: 1521 for the unacceptability of accented repeated items in Dutch)). An archetypal repetition involves a lexical item repeated in an immediately following intonational phrase, e. g. The stadium where Manchester United play is in the west of Manchester. In this example Manchester, as a repeated and hence given item, is de-accented and the accent is thrown back onto the word west.
3. De-accenting and re-accenting De-accenting of repeated items is certainly obligatory for the type of English which Halliday was describing (roughly RP) and for most other dialects of English. And it is … easy to think of de-accenting as some sort of cognitive universal: “The principal linguistic effects of the givennew distinction in English, and perhaps all languages, reduce to the fact that given information is conveyed in a weaker and more attenuated manner than new” (Chafe 1976: 31, my italics). Yet such attenuation or de-accenting (and certainly its obligatory use) may not be as universal as first thought.
Brazilian Portuguese: Esti livro custa cinco dolares e esti aqui tres dolares. “This book costs five dollars and this other one three dollars”. Rumanian: [o sa vedem] ce aveţ i şi ce nu aveţ “. . . we see what we have and what we don’t have” Spanish (Chile): ¿Vienes/ o no vienes? “Can you come or not? ” ¿Con leche o sin leche? “With or without milk? ” Escribo canciones porque escucho cosas/ veo cosas/ huelo cosas. “I write songs because I listen to things, see things, smell things. ”
Some varieties of English have also been reported to use lexical reaccenting. Hawaian English: Forty t’ree per cent is gavment owned, and fifty seven per cent is privately owned. Indian English: A: You just weren’t listening. B: I was listening. Singapore English: I went to the shop to buy sweets but they had totally run out of sweets
Cruttenden’s survey: ten different setting-response dialogue types where the response involved a lexical item repeated from the setting. translated [with necessary changes] into 11 languges; native speakers of the languages were recorded reading the settings and responses. The location of the intonational nucleus (= last pitch accent) was marked auditorily by me and by a trained intonationist working independently; there was 97% agreement at the first analysis and the remaining 3% of sentences were resolved by discussion.
English 7 speakers French (principally Provence) 14 Italian (principally Sicily) 11 Lithuanian (Vilnius) 4 Arabic (Tunisia) 4 Albanian (Tirana) 4 Russian (Moscow 2, Dushambe 2) 4 Spanish (Chile 2, Castilian 1, Catalonian Spanish 1) 4 Greek (Athens) 4 Macedonian (Skopje) 2 Swedish (Skåne 3, Östersund 1) 4 German (Westphalia 2, Munich 1, Tübingen/Hamburg 1) 4.
4. Settings and responses 4. 1. Setting-response 1 The English version of the first example is as follows: A: If you don’t hurry up, you’ll be late. B: I don’t care if we are late. / (I don’t care if we’re late) most languages behaved similarly to English French (14): Je m’en fiche d’être en retard, 11 put the last accent on fiche and three re-accented retard. Spanish (4): No me importa si llegamos tarde three of the four speakers re-accented tarde. The overall conclusion for this setting-response pair is that it produces obligatory de-accenting in all the languages tested with the exceptions of optional re-accenting in French, Spanish and Swedish.
4. 2. Setting-response 2 A: Would you like to come to dinner tonight. I’m afraid it’s only chicken. B: I don’t like chicken. Six of the 14 French speakers re-accented café in the sentence Merci, je bois pas de café, whereas four preferred to accent pas and four bois All four speakers of Swedish re-accented. In other languages, one speaker re-accented in Arabic, German, Greek, Macedonian and Spanish.
4. 3. Setting-response 3 A: You need a pair of black shoes for the wedding. B: I’ve already got a pair of black shoes. all but one of the French speakers re-accenting paire in the sentence J’en ai déjà une paire and all four speakers of Spanish re-accenting negros in Yo ya tengo un par de zapatos negros. three out of four of the Swedish informants re-accented. In the first three setting-responses we have had no cases of reaccenting in Italian; this was unexpected in the light of the comments of Ladd (1990) and other informal personal communications. In this setting-response it may be that the use of the anticipatory le in Ce le ho già un paio di scarpe nere reinforces the de-accenting of the final phrase
(4) A: Why do you keep getting angry? B: Because John makes me angry. (5) A: He earns at least thirty thousand pounds. B: I think he earns nearly forty thousand. (Adjust amounts to local currency). (6) A: I make the answer twenty-six point one. B: Well I make it thirty-six point one. (An earlier version for French, Italian, Macedonian, Russian and Swedish used the contrast between 16 and 26 which had a lesser possibility of re-accenting) (7) A: I did all the work. B: You mean your sister did all the work.
(8) A: Do you like your new house? B: Yes, I'm happy, and my wife is very happy. (9) A: What was the score? B: Inter-Milan 1 Roma 1. (Use names of football teams on relevant country) (10)a A: I’m always more interested in home news. B: I think it’s important to keep a balance between national and international affairs. (10)b A: That reply is correct. B: You mean that reply is incorrect. (10)c A: I think the locals are very friendly. B: I think they're very unfriendly.
french (1) Je m' en fiche d’ être en retard (2) Merci, je bois pas de café Thank you, I don’t drink coffee. (3) J' en ai déjà une paire I already have a pair of them. (4) C' est toi qui me mets en colère It is you who makes me angry. (5) Moi je pense qu' il gagne au moins quatre-vingt mille francs I think he earns at least eighty thousand francs. (6) D'après moi c' est vingt six virgule un According to me, it is twenty six point one. (7) Tu veux dire que ta soeur a fait tout le travail You mean your sister did all the work.
(8) Oui, j' en suis très content et mon fils en est très content Yes, I am very happy and my son is very happy. (9) Paris St. Germain un, St. Etienne un. Paris St, Germain 1, St. Etienne 1. (10)a Il faut garder un equilibre entre les affaires nationales et les affaires internationales It is necessary to maintain a balance between national and international affairs. (10)b Non, cette réponse est incorrecte No, this reply is incorrect. (10)c Moi, je les trouve très impolis I find them very impolite.
(F) ITALIAN (1) Me ne frego del ritardo I don't care if we are late. (2) Non mi piace il pollo I don’t like chicken. (3) Ce le ho già un paio di scarpe nere I already have a pair of black shoes. (4) Perchè tu mi fai arrabbiare Because you make me angry. (5) Penso che ne guadagni sugli ottanta I think that he earns about eighty. (6) A me invece ventisei virgola uno. But I make it twenty six point one. (7) Vuoi dire che tua sorella ha fatto tutto You mean your sister has done everything. (8) Sì mi piace a mia moglie piace molto Yes, I like it and my wife likes it a lot. (9) Inter uno, Roma uno Inter 1, Roma 1
(10)a Erano bloccati per sciopero i voli nazionali e internazionali National and international flights are blocked by the strike. (10)b A me invece sembrano molto dissimili But they seem very dissimilar to me. (10)c Io trovo che sia molto antipatica I find them very unfriendly.
(L) SWEDISH (1) Jag struntar i om vi kommer för sent I don't care if we are late. (2) Jag tycker inte om kyckling I don't like chicken. (3) Jag har redan ett par svarta skor I've already got a pair of black shoes. (4) För att du gör mig arg Because you make me angry. (5) Jag tror att han tjänar nästan hundrafemtiotusen I think he earns nearly one hundred and fifty thousand. (6) Jag får det till tjugosex komma ett Well I make it twenty-six point one. (7) Du menar att din syster gjorde jobbet You mean your sister did the job
(8) Ja, jag är nöjd, och min fru är mycket nöjd Yes, I'm happy and my wife is very happy (9) (Not relevant) (10)a Det är viktigt att hålla balans mellan nationella och internationella angelägenheter It is important to keep a balance between national and international concerns. (10)b Du menar att han uppträder inkorrekt You mean that he behaves incorrectly. (10)c Jag tycker att de är mycket ovänliga I think that they are very unfriendly.