IS SENTENCE VIABLE The 3 rd International Conference

IS SENTENCE VIABLE? The 3 rd International Conference on Cognitive Science Moscow, June 21, 2008 Andrej A. Kibrik (kibrik@comtv. ru) Vera I. Podlesskaya (podlesskaya@ocrus. ru ) 1

Does spoken language consist of sentences? § Sheer facts: § Spoken language is the primary form of language § Spoken language does not contain periods, question marks and other explicit signals of sentence boundaries § Research question: § Is sentence, as a theoretical construct, as identifiable and as basic for the primary form of language as it is (or as it is thought to be) for written language? 2

Sentence in spoken language § § Position 1: sentence is a universal and basic unit of language § Assumption typically held by not only by linguists but also by other cognitive scientists § “With no more than 50 to 100 K words humans can create and understand an infinite number of sentences” (Bernstein et al. 1994: 349 -350) § Psycholinguistics: “Sentence processing” § But sentence is very far from being obvious in spoken language Position 2: avoidance of the issue, typical of discourseoriented linguists § If so, how could sentences become so much entrenched in written language? 3

Night Dream Stories § Corpus of spoken Russian stories § Speakers: children and adolescents § Subject matter: retelling of night dreams § Discourse type: monologic narrative (personal stories) § Speech act type: declaratives 4

Two basic features of spoken discourse § Segmentation § Transitional continuity 5

Segmentation § Elementary discourse units (EDUs) § Identified on the basis of a conjunction of prosodic criteria: § § § Tempo pattern Loudness pattern Integral tonal contour Presence of an accentual center Pausing pattern § Speakers tend to organize EDUs as clausal units 6

Example of segmentation Z 54 /мы с= || ехали на автобусеw. /my s= || exali na avtobusew. We rode on bus Discourse transcription . . . (0. 6) /Я /первая села в автобус. . (0. 6) /Ja /pervaja sela v avtobus. I first got on bus. . (0. 4) А A And /тогда /togda then уже uže already д= ||. . (0. 2) d= . . (0. 1) и i and /’Аня /Anja не ne not –успела –uspela managed закрывались zakryvalis’ were. closing двери, dveri, doors сесть. sest’. get. in . . . (0. 7) Иw мм(0. 4) /когда-а. . (0. 2) ’’(0. 3). . (0. 4) {ЧМОКАНЬЕ 0. 2}. . (0. 4) когда я приехала на нашу /остановку’, . . . (0. 7) IW mm(0. 4) /kogda-a. . (0. 2) ’’(0. 3). . (0. 4) {SMACKING 0. 2}. . (0. 4) kogda ja priexala na našu /ostanovku’, And when I arrived to our station 7

Transitional continuity § Term by J. Du. Bois et al. 1992 § Alternative term by Sandro V. Kodzasov: phase § Discourse semantic category: ‘end’ vs. ‘non-end’ § § § (=expectation of a forthcoming end) Hierarchical nature of phase End of tentative sentence – falling tonal accent Non-end – rising tonal accent 8

A canonical example of the transitional continuity distinction z 57: 15 -16 § § . . (0. 4) /Мы-ы’. . (0. 4) как бы за них /взя-ались, . . (0. 4) /My-y’. . (0. 4) kak by za nix /vzja-alis’, We sort of at them got. hold ØRising (“comma”) ØNon-end . . . (0. 5) и-и ввь= ||. . (0. 2) полетели вве-ерх. . (0. 5) i-i vv’= ||. . (0. 2) poleteli vve-erx. and flew upward ØFalling (“period”) ØEnd § If things were that easy, sentence would be uncontroversial 9

The problem of two kinds of falling § The existence of non-final falling may call relevance of sentence into question § However, the distinction between two kinds of falling is very systematic § The two kinds of falling: § are prosodically distinct § have distinct discourse functions 11

Prosodic criteria of the final vs. non-final falling distinction § Primary criteria: 1. Target frequency band 2. Post-accent behavior 12

Criterion 1: Target frequency band § Final falling (“period”): targets at the bottom of the speaker’s F 0 range § Non-final falling (“faling comma”): targets at level several dozen Hz (several semitones) higher 13

F 0 graph for the “lake” example 12 10 12 8 5 ozero, malen’koe nebol’ takoe, šoe. brevno kakoe mosta. -to, 14

Non-final falling (210 Гц), final falling (170 Гц), rising, post-rising falling Z 54: 4 -5 170 Hz 210 Hz. . (0. 4) А A And /тогда /togda then уже uže already д= ||. . (0. 2) d= . . (0. 1) и i and /’Аня /Anja не ne not –успела –uspela managed закрывались zakryvalis’ were. closing двери, dveri, doors сесть. sest’. get. in . . . (0. 7) Иw мм(0. 4) /когда-а. . (0. 2) ’’(0. 3). . (0. 4) {ЧМОКАНЬЕ 0. 2}. . (0. 4) когда я приехала на нашу /остановку’, . . . (0. 7) IW mm(0. 4) /kogda-a. . (0. 2) ’’(0. 3). . (0. 4) {SMACKING 0. 2}. . (0. 4) kogda ja priexala na našu /ostanovku’, 15 And when I arrived to our station

Criterion 2: Post-accent behavior § Final falling (“period”): steady falling on the post-accent syllables § Non-final falling (“comma”): lack of falling on post-accent syllables, often rise of tone (V-curve) 16

V-curve z 26 260 Hz 240 Hz. . (5. 7) /Домик. . . (0. 6) был /около реч↑ки, . . (5. 7) /Domik. . . (0. 6) byl /okolo reč↑ki, Little. house was near creek 235 Hz . . (3. 3) /рядом были –родник-ки, . . (3. 3) /rjadom byli –rodnik-ki, nearby were springs. . (0. 4) и –ле-ес. . . (0. 4) i –le-es. and forest 17

Secondary criteria 3. 4. 5. 6. Pausing pattern Reset vs. latching Steepness of falling Interval of falling 18

The final vs. non-final falling distinction § A speaker’s prosodic pattern must be identified § On its basis the difference between final and non-final falling distinction can be identified with a high degree of robustness 19

Contexts of non-final falling § Anticipatory mirror-image adaptation § Inset § Stepwise falling 20

Anticipatory mirror-image adaptation §. . (1. 8) Когда Kogda when §. . . (0. 5) что-о čto-o that я услышала, ja uslyšala, I heard /бомба гремит, /bomba gremit, bomb growls 21

Stepwise falling § . . (1. 5) то, /Озеро. . . (0. 5) § . . (0. 3) (Или § или /озеро, § но § потому что’. . (0. 2). . . (0. 6) § какое- /речка, по-моему § озеро, как-то-оw маленькое § такое, § § § небольшое. ) § . . (1. 5) to, . . (0. 3) (Ili /Ozero. . . (0. 5) kakoe- Lake some /rečka, river 210 Hz Either ili /ozero, or lake no po-moemu ozero, but I guess lake potomu čto’. . (0. 2) kak-to-o. W because 190 Hz somehow. . . (0. 6) malen’koe takoe, 160 Hz small such nebol’šoe. ) minor 23

Representation of EDU continuity types in corpus 24

The status of sentence § In the speech of most speakers final falling is § § § clearly distinct from non-final patterns Final intonation, expressly distinct from non-final intonation (both rising and falling), makes the notion of sentence valid for spoken discourse Speakers “know” when they complete a sentence and when they do not Apparently, spoken sentences are the prototype of written sentences 25

Functions of sentences § Ease the processing by creating intermediate informational chunks § Chafe: superfoci of consciousness 26

However § § § Identification of sentences is possible only on the basis of a complex analytic procedure It is dependent on prior understanding of a speaker’s prosodic “portrait” There are prototypes of final and non-final fallings, but there are intermediate instances, therefore sentencehood may be a matter of degree A significant tune-up is necessary to apply the procedure to a different discourse type or a different language Therefore, sentence is an elusive, intermediate, nonbasic unit of language 27

EDUs vs. sentences: degree of variability EDUs: distribution in terms of number of words Sentences: distribution in terms of number of EDUs 53% – 3± 1 80% – 3± 2 28

EDUs vs. sentences: degree of variability § Unlike EDUs, sentences are highly variable § Speakers with short sentences § Speakers with long sentences equaling stories § Clause chaining 29

Conclusions § Sentence is an intermediate hierarchical grouping between a whole discourse and an EDU (roughly, clause) § Sentence is very far away from being a basic unit of spoken language 30

Acknowledgement Member of our project Nikolay Korotaev 31