HISPhonCog 2019: Hanyang International Symposium on Phonetics & Cognitive Sciences of Language 2019 May 24-25, 2019, Hanyang University, Seoul, Korea Connecting prosody to duality of patterning in two modalities Mary E. Beckman Ohio State University (USA) [email protected] This talk sketches an idea about what kind of models of spoken language prosody will be most useful in making cross-language comparisons, to be able to identify non-trivial universals and understand how speech might differ from the vocal communication systems of our closest non- human relatives. It is an idea that I have been trying to articulate (with varying degrees of failure) for more than three decades, starting with the book that grew out of my doctoral dissertation [1]. The idea has been difficult to articulate in large part because the alphabetic model that was foundational in the development of modern linguistics is so entrenched in our thinking about speech prosody. In laying out the idea, therefore, it is useful to look to a different modality, where the compositional structures that govern the alignment relationships among “gestures” (the basic units in the articulation of an utterance) are more transparently related to the alignment relationships among the objects of perception that are innate to the medium. That is, it is useful to incorporate observations and insights that have emerged in the last half century of research on the prosody of signed languages, such as American Sign Language [2, 3, 4] (ASL), Israeli Sign Language [5, 6] (ISL), and the newly emerging Al-Sayyid Bedouin Sign Language [7] (ABSL). I will review some of this work to make the first three of the following sequence of arguments: (1) Every human language, signed as well as spoken, has a richly compositional syntactic system, with morphemic elements that can be combined and arranged to form potentially very complex and novel multi-layered sentences. (2) Every human language also has a richly compositional prosodic system, with phonological elements that can be combined and arranged to form potentially very complex multi- layered utterances. (3) Prosody is the “phonological patterning” counterpart to syntax in Hockett’s specification of “duality of patterning” [8,9] (DoP) as the universal design feature that may differentiate human languages from the communication systems of other species. However, the universality of DoP is obscured if phoneme-sized segments are axiomatically assumed and described as the “building blocks” of morphemes, as in Martinet’s “double articulation” [10], a theory of design features that is often (incorrectly) equated with DoP. (4) The incorrect equation of Martinet’s theory of “double articulation” with Hockett’s theory of “duality of patterning” also obscures evidence of how DoP develops in ontogeny and of how DoP might have emerged in the phylogeny of our species. The evidence that (1) is true of signed as well as spoken languages comes from studies of emerging languages such as ABSL and Nicaraguan Sign Language, where syntactically complex structures such as conditionals and also conventionalized agreement morphology have emerged within a single generation of early signers [11]. The evidence that (2) is true of signed as well as spoken languages also comes from emerging languages, and there is evidence of productive complexity for at least three levels of structure. First, manual gestures can be decomposed into specifications for the properties listed in (5), and a unit analogous to the syllable in spoken languages can be defined by coordinated changes in (at least one of) location, finger position, and orientation. Second, in all sign languages studied to date, content words are typically monosyllabic, but longer forms (e.g., disyllabic compound words) can be marked as prosodic words by phonological processes of reduction and/or spreading. Third, even the youngest sign languages have rich “visual intonation” systems by which non- manual gestures are aligned to the sequence of manual gestures so as to demarcate prosodic units comparable to the intonational phrases of spoken languages, as illustrated in Figs. 1 and 2. 2022[(Fig._4)TD$IG] W. Sandler et al. / Lingua 121 (2011) 2014–2033 (5) features in ASL example of minimal pair or of sign with changing specification handshape [of dominant hand (H1) or of both hands (H1&H2) in some signs] selected fingers SCHOOL (all 5 fingers) vs. IM,POSSIBLE (thumb & pinky) finger position WHITE (all 5 fingers begin spread and then close to touch) orientation STAY (both palms up) vs NOW (both palms down) location ONION (at side of right eye) vs. APPLE (at right of chin) endpoint SCHOOL (H1 moves down from neutral space to touch H2) movement type NUDE (1 large movement) vs. AVAILABLE (2 short movements) Fig. 4. The juncture of two Intonational Phrases in an ABSL conditional sentence meaning, ‘If he says no, then nothing can be done.’ Pictured are NO] and [(Fig._5)TD$IG][NOTHING-CAN-BE-DONE. The evidence for (3) stems from a close reading of relevant parts of [8,9], and the realization that the argument against DoP in ABSL [7] is actually an argument instead against Martinet’s theory. (There will be no time to review the evidence for (4), which is part of the larger idea that has been developed in more detail elsewhere [12,13].) 4000 2000 0 da me to it ta ra ʃi ka taga nai 300 HL HL %L L% 200 %L H- 0 0.5 1 1.5 time (seconds) Fig. 5. Sample coding in stretch of signing of a younger second generation signer. Fig.1 Display of the time-aligned tags for gestures of “visual Fig.2 Display of a commonly used acoustic intonation” of an ABSL utterance meaning ‘If he says no, then measure of “intonation” in an utterance of a The number of prosodic cues and their alignment in this utterance are similarly indicative of the signing of the younger pair, and contrast with the utterancesnothing can of the be older done. signers.’ (extract The finalfrom boundary Fig. 5 in of [ the7]). first Intonational PhraseJapanese is associated translation of utterance in Fig. 1. with seven prosodic cues – raised brows, widely open eyes, large sign with hold, head down, head and torso forward, prolonged gaze at the addressee,Reference ands eyeblink.6 The final boundary of the second phrase is marked by four prosodic signals – hold and large size on the last sign, head and torso tilted back, and blink, yielding a mean number of 6 cues per Intonational Phrase in this utterance. The facial articulations that co-occur with the first constituent are clearly linguistic, determined by comparing associated meanings[1] Beckman (in this, case,M. E. a conditional(1986). Stress relationship and non between-stress the accent two constituents). Amsterdam with : Foris the particular Publications. Action Units articulated[2] and Liddel their temporal, S. K., scope& Johnson, in relation R. to theE. (2013). text. American Sign Language: The phonological base. Sign Language The two clauses are connectedStudies by, dependency64, 195-277 marking,. which consists of raised brows across the first constituent and prolonged gaze at the[3] addressee Brentari, with D forward. (1998 head/torso). A prosodic at the model end, followedof sign language by head/torso phonology retraction. Cambridge, on the second MA: MIT Press. constituent, as seen in Fig. 5. Brow raise here, as in Israeli Sign Language, signals that the information of the constituent so marked is to be completed[4] byWilbur the unfolding, R. discourse.B. (200 The9).prolonged Effects gazeof atv thearying addressee rate at theof endsigning of the first on intonational ASL manual signs and nonmanual unit captures the addressee’sm attentionarkers. Language and directs and it to theSpeech coming, 52 IP.(2 These/3), 245 signals,-285. together with the forward/backward head movements, create[5] a Meir strong, I overt. (1998 cue). toSyn thetactic contingency-semantic implicature interaction between in Israeli the clauses, Sign and,Language together verbs: with theThe case of backwards verbs. Sign meaning of the signs in eachLanguage clause, yield and a conditional Linguistics interpretation, 1(1), 3-37 (Dachkovsky,. 2008; Dachkovsky and Sandler, 2009). [6] Dachkovsky, S., & Sandler, W. (2009). Visual intonation in the prosody of a sign language. Language and 6 Gaze is not interpreted as a prosodic cue per se in other studies (Sandler, in press), and it is not marked in Fig. 5. As it is a reliable indicator of dependency in the ABSL data, we included it in theSpeech analysis,, 52 leaving(2/3 the), establishment287-314. of its membership in any particular grammatical component to future research. [7] Sandler, W., Meir, I., Dachkovsky, S., Padden, C., & Aronoff, M. (2011). The emergence of complexity in prosody and syntax. Lingua, 121(13), 2014-2033. [8] Hockett, C. F. (1955). Manual of phonology. Baltimore: Waverly Press, Inc. [9] Hockett, C. F. (1961). Linguistic elements and their relations. Language, 37(1), 29-53. [10] Martinet, A. (1957). Arbitraire linguistique et double articulation. Cahiers Ferdinand de Saussure, 15, 105-116. [11] Senghas, A., Kita, S., & Özyuörek, A. (2004). Children creating core properties of language: Evidence from an emerging sign language in Nicaragua. Science, 305(5691), 1779-1782. [12] Edwards, J., Beckman, M. E., & Munson, B. (2015). Cross-language differences in acquisition. In M. A. Redford (ed.), Handbook of speech production, pp. 530-554. Chichester, UK: Wiley-Blackwell. [13] Plummer, A. R., & Beckman, M. E. (2015). Framing a socio-indexical basis for the emergence and cultural transmission of phonological systems. Journal of Phonetics, 53, 66-78. .