View Our Catalog

Join Our E-Mail List

What's New

Sign Language Studies

American Annals of the Deaf

Press Home

Primary Movement in Sign Languages: A Study of Six Languages

Donna Jo Napoli, Mark
Mai, and Nicholas Gaw

Part One

The past fifty years have witnessed a flowering of research on sign languages, largely on their phonology and morphology but in more recent years increasingly on their syntax and semantics. The first decade of this century also experienced rich comparative work across sign languages. For example, the Sign Language Typology Research Group at the University of Central Lancashire in Preston, United Kingdom, often in cooperation with the Max Planck Institute for Psycholinguistics in Leipzig, Germany, has been and is presently instrumental in multiple projects. These projects range from cataloging and describing endangered and little known sign languages in a browsable corpus to studies of specific topics, such as negative and interrogative constructions, possessive and existential constructions, numeral incorporation, and agreement systems. The Sign Language Typology Research Group has also organized international workshops in which researchers of sign typology can get together and discuss their results. Ulrike Zeshan (2004a, 2004b, 2006) has been at the forefront of much of this work, particularly on interrogatives and negatives.

Additionally, there has been considerable work on word order in particular sign languages, (from the seminal work of Fischer [1975] and the classic work of Volterra et al. [1984] to many of the articles in Brennan and Turner [1994] and the considerable work since), although several factors seem to stand in the way of a word-order typology for sign. Although sign languages vary in many ways syntactically (see Perniss, Pfau, and Steinbach 2007), typically they make substantial use of classifier predicates. (We have read about only two exceptions. One is Adamorobe Sign Language, used in an Akan village in eastern Ghana, which lacks classifiers for motion and location [Nyst 2007]. The other is Indo-Pakistani Sign Language, which Zeshan [2000, 27] originally reported to have “no systematically arranged paradigm of classificatory handshapes” but Zeshan [2003] later reported to have limited use of whole entity classifiers.)

Once we enter the realm of classifier predicates, we undoubtedly find movement from one indexed position to another, with all characteristics of the signing—from handshape, to palm and fingertip orientation, to location, to movement, to nonmanuals—potentially being determined by setting up a framework in which each physical element in the signed message is analogous to some action or participant (active or passive, including locatum) in an event (see, among others, McDonnell 1996; Vermeerbergen 1996; Sutton-Spence and Woll 1999; Leeson 2001), and this includes nonpresent referents (Engberg-Pedersen 2004). Additionally, it appears that context plays an enormous role in word order in sign languages, as seen in spontaneous conversation (compared with elicited data)—a fact that makes firm statements about particular word orders difficult to maintain (among others, see discussion in Deuchar 1983; Johnston et al. 2007; Jantunen 2008). So we expect much in common syntactically across sign languages in such utterances, which we do indeed find ( Johnston 1989; Woll 2003; Vermeerbergen 2006; Napoli and Sutton-Spence n.d.), with questions of comparative word order receding in importance. Nevertheless, the field is fertile, and we look with optimism at the search for typological characteristics both at the level of more specific constructions being examined in the studies alluded to in the previous paragraph and at the overarching level of word order.

Phonetic Typology

In this book we look for overarching characteristics for typologizing sign languages by studying another component of the grammar: phonetics. It is often possible just from overhearing a snippet of spoken conversation to recognize that a language we ourselves do not speak belongs to some larger group, such as Chinese, Slavic, or Athabaskan, based solely on sound properties, whether phonetic or phonological. In fact, this common observation is not trivial; artificial intelligence has been using prosody analyzers for language recognition for years (Waibel 1988). Likewise, when we hear a nonnative speaker of English speak English, we can often guess at the larger group her or his mother tongue (L1) belongs to, just from phonetic and/or phonological properties carried over in the transfer from mother tongue to a second language (L2)—in this case English. Although influences from L1 on L2 are complex, there is general agreement that phonemic inventories, allophonic variations, phonotactic constraints, and prosody are all likely to be involved (Flege 1987; Rochet 1995; Boula de Mareüil, Marotta, and Adda-Decker 2004), sometimes to such an extent, particularly with respect to vowel quality and prosody, that intelligibility is threatened (Munro and Derwing 1995; Mayfield Tomokiyo and Waibel 2001; Burleson 2007).

With that in mind, we set out to see if we could typologize sign languages by phonetic characteristics, in particular by characteristics of the paths of primary movement. We chose to look at this particular component of the sign for several reasons.

Some scholars have argued that movement in sign is comparable to vowels in spoken language (Liddell and Johnson 1989; Perlmutter 1992). And some have argued that the distinction between full and reduced movement in sign is comparable to the distinction between strong and weak vowels in speech (Wilbur 1985). Additionally, in syllables that contain final holds, movement accounts for 55% of the duration and the final hold accounts for 45% (Wilbur and Nolan 1986), a finding that suggests movement may figure prominently in the perception of rhythm and stress (Wilbur 1990). Consistent with these findings, many have claimed that movement represents a visual analogue of sonority (Brentari 1990; Corina 1990b; Perlmutter 1992; Sandler 1993). Building on much of this work, Brentari (1998) offers the prosodic model of sign syllables, analyzing the sign as (1) two sets of features organized in a hierarchical feature geometry (where a feature geometry is independently motivated for sign languages—see Sandler 1986, 1987, 1989; Corina 1990a; Sandler and Lillo-Martin 2006); (2) inherent features (including handshape and location), which are comparable to consonants in speech; and (3) prosodic features (movement, both primary and secondary—a distinction we address in chapter 2 in the section “Primary Movement Only”), which she compares with vowels in speech. Regardless of whether one assumes the prosodic model, the recognition of movement as (somehow) vocalic and (somehow) relevant to prosody has shed light on phenomena in a variety of sign languages, including the appearance of something comparable to vowel harmony in the acquisition of BSL (Morgan 2006) and the accentual prosody (speed, intensity, and manner of movement) relevant to poetic form in LSF nursery rhymes (Blondel and Miller 2000, 2001).

Next Page