Prosody consists of intricate patterns of rhythmic and intonational structure. Rhythmic grouping, stress (within and above the word), and intonation all interact with each other in this system, and they interact with phonology, morphology, syntax, semantics, and pragmatics as well. Some important functions of prosody are:
a. to delimit syntactic and semantic units within sentences and discourse
b. to indicate what information is focussed in a sentence
c. to convey pragmatic notions like illocutionary force (question vs declarative)
d. to convey nuances of meaning.
Since these functions are all obviously important, we may conclude that prosody is an essential part of human communication. A story about an English professor and his students illustrates this point well. The professor is said to have written the following string of words on the board, and asked his students to punctuate it.
WOMAN WITHOUT HER MAN IS NOTHING
According to the story, the male students provided the following punctuation:
The female students saw things differently, and supplied the following punctuation:
Phrasing similarly distinguishes restrictive and nonrestictive relative clauses. Intonation (melodies comprised of rises and falls in the pitch of the voice) can also make important contributions to meaning, in some languages (e.g., Hebrew) minimally distinguishing declarative sentences from yes/no questions. See Ladd (1996) for a recent detailed study of phonological intonation and for references.
Since it is well established that sign languages have phonology and syntax (see Sandler and Lillo-Martin, in press, for a recent overview), it is reasonable to expect that they may have some system that is comparable to prosody as well.
My recent work on Israeli Sign Language (ISL), some of it with Marina Nespor, has shown that that language has a prosodic system that can profitably be compared to the prosody of spoken languages. Following Selkirk (1984) and Nespor and Vogel (1986), we assume that there is a hierarchy of prosodic constituents,
From the level of the prosodic word through the level of the intonational phrase, each of these constituents coincides to some extent with a morphological or syntactic constituent, such as the word, the phrase, or the clause. That this correspondence is not absolute is one of the observations that justifies an independent level of structure for prosody. Nevertheless, it is through this correspondence that prosody performs its role of interpreting the structure of our utterances.
The work of Nespor and Sandler (in press) shows that there are specific prosodic cues that mark the levels of this hierarchy that we investigated: the prosodic constituents Phonological Phrase and Intonational Phrase exist in ISL. We constructed 30 sentences for translation to ISL, that were designed to elicit certain syntactic structures in the language. These were signed by three different signers and videotaped. We then delineated the phonological phrases based on syntactic criteria, and coded the sentences to learn whether or not there were prosodic clues to this structure. We were interested in learning about any cues to rhythmic structure and/or intonation-like elements.
In that study (Nespor and Sandler, in press), we found the following phonetic correlates to the phonological phrase: reduplicated final signs, holds at the end of the final sign, or pause after the last sign of the phrase. We also discovered a rule of external sandhi that appears to have the phonological phrase as its domain. This rule involves the spread of the nondominant hand from a two-handed sign trigger to the end of the phrase.
Intonational phrases were normally marked by a change in head position and changes in all elements that comprise facial expression. We hypothesize that facial expressions are the equivalent of intonation in spoken language (cf. Wilbur, in press, for ASL). These may be bound to the intonational phrase or the phonological phrase, as in spoken language, but unlike spoken language, the facial articulations of sign language can be simultaneously layered on each other, and they characterize entire prosodic constituents rather than occurring only at the edges and on focused words as they do in spoken language. Thus, we find both similarities and differences between languages in the two modalities.
At a lower level of the prosodic hierarchy, I found in a different study that ISL cliticizes pronouns and deictics (indexing signs) in certain positions within phonological phrases to form single prosodic words (Sandler 1999a). Phrase initially, the handshape of pronouns may assimilate that of the following host word, while phrase-finally, two-handed (double dez) host words and the following pronoun or deictic may coalesce to form a single syllable. Both of these cliticization processes invoke constraints on prosodic words that exist in the lexicon. However, I argue that at the postlexical level where cliticization takes place, these same constraints have different rankings with respect to each other and to other constraints (see McCarthy and Prince, 1993, for the theoretical foundation of this approach). In both kinds of cliticization, signers tend to mouth the host word only and not the pronoun, and, importantly, the mouthing of the host word extends over the signing of both the host and the prononimal clitic.
Details of these studies can be found in Nespor and Sandler (in press), Sandler (in press), Sandler (1999a), and Sandler (1999b). In order to arrive at the analyses presented in those studies, we developed a coding system with the categories shown in the chart below. The coding was done by native signers together with a research assistant conversant in ISL, by viewing videotapes of the elicited sentences repeatedly in slow motion.
(1) Coding Categories
GLOSS
brows
eyes
cheeks
mouth
mouthing (of spoken language word)
tongue
head
torso
hold
speed
size
reduplication
handshape assimilation
nondominant hand spread
Using this system, we drew lines in different colors across the page under the glosses, showing the extent of the manual or nonmanual marking. In cases of particular facial configurations, for example of the brows or mouth, we adopted a system of little icons to represent them, which we placed next to the lines showing their extent. With this coding system, we could see where prosodic cues occurred and how long they remanied in the signal. The following example from Nespor and Sandler (in press) illustrates the coding. `P´ stands for Phonological Prase, and `I´ stands for Intonational Phrase.
(2) Example of coded sentence
| [[book-there ] P | [he write ] P ] I | [[interesting] P ] I | |
| brows | up----------------------- | ------------ | down-------- |
| eyes | squint--------- | droop---------- | |
| cheeks | |||
| mouth | `O´----------- | down ---------- | |
| tongue | |||
| head | tilt----------------------- | -------------------- | |
| mouthing | `book´--------- | `interesting´ | |
| torso | lean--------------------- | -------------------- | |
| hold |
|
||
| reduplication | -1 | x 3 | x 4 |
| pause | |||
| speed | slow | ||
| size | big | big |
(This example does not include some categories shown in (1) that were added at later stages of the research.)
This work offers support for the approaches of other researchers on other sign languages which encode details of such categories as facial expressions, mouthing, body postures, reduplication, size, and speed, by demonstrating that these phonetic properties can play a linguistic role in sign languages. What this work also suggests is this: It is important to distinguish between citation forms and signs in context for the purpose of compiling data bases. That is, if the input data are signs that are not in citation form but in utterances, it is important to distinguish what may be attributed to the sign itself and what may be attributed to prosody that comes from higher levels of structure. For example, a sign may appear with three reduplications, or raised brows, but these characteristics may not be part of the lexical item. Rather, they may be related to position or function within some prosodic constitutent larger than the word, or to some sentence- or discourse-level meaning or intent. The work also provides analyses of prosody in one sign language that may serve as a basis for comparison and further elaboration in other sign languages.
*This work is supported by grant no. 820/95 from the Israel Science Foundation.
References
Ladd, D.R. (1996) Intonational Phonology. Cambridge: Cambridge University Press
McCarthy, J. and A. Prince (1993). Prosodic Morphology I: Constraint Interaction and Satisfaction. Ms. University of Massachusetts, Amherst, and Rutgers Unviersity.
Nespor, M. and I. Vogel (1986) Prosodic Phonology. Dordrecht. Foris.
Nespor, M. and W. Sandler (in press). Prosodic Phonology in Israeli Sign Language. Language and Speech.
Sandler, W. (1999a) Cliticization and prosodic words in a sign language. in T. A. Hall & U. Kleinhenz (eds.). Studies on the Phonological Word. Amsterdam: Benjamins. (Current Studies in Linguistic Theory).
Sandler, W. (1999b). The Medium and the Message: Prosodic Interpretation of Linguistic Content in Israeli Sign Language. paper presented at the Theoretical Issues in Sign Language Research conference. Gallaudet University, Washington, D.C.
Sandler, W. (in press). Prosody in two natural language modalities. Language and Speech
Sandler, W. and D. Lillo-Martin (in press). Natural sign languages. In M. Aronoff and J. Rees-Miller, (eds.), The Blackwell Handbook of Linguistics. Oxford: Blackwell
Selkirk, E. O. (1984) Phonology and Syntax. Cambridge, Mass. MIT Press.
Wilbur, R.B. (to appear) Prosodic structure of American sign language. ms. M. Lynch, ed., The Cognitive Science of Prosody. Amsterdam: Elsevier
Posted: 18.01.2000