Language endangerment: A scholarly and humanitarian concern

Outlined in broad theoretical terms and described in detailed case studies, the phenomenon known variously as “language shift” (Kulick 1992), “language obsolescence” (Dorian 1989), “language death” (Dressler & Wodak-Leodolter 1977, Dorian 1981, Sasse 1990, Brenzinger 1992, Broderick 1999, Crystal 2000), or “language extinction” (Nettle & Romaine 2000), is a rapidly escalating problem – one that has gained the attention and concern of organizations ranging from the Linguistic Society of America to the United Nations. Although there is debate (Hale et al. 1992; Ladefoged 1992; Dorian 1993, 2002; England 2002; Fishman 2002; Hill 2002; Hinton 2002; Silverstein 2003) about how best to conceptualize the problem (e.g., as a threat to biocultural diversity, or as a natural process) and consequently how best to respond to the phenomenon of disappearing languages (e.g., with intervention motivated by humanitarian concerns, or with non-intervention motivated by strict scientific observation), there is widespread consensus about the importance of documenting existing linguistic diversity. Furthermore, there is widespread agreement that particular attention and effort should be directed toward documenting and describing endangered languages.

For example, the Linguistic Society of America Committee on Endangered Languages has published a policy statement on “The need for the documentation of linguistic diversity” (1994), which points out that the study of universal grammar and linguistic typology is greatly enriched by the documentation of linguistic diversity, especially the documentation of languages on the geographical and linguistic “fringe” (LSA 1994). The policy statement recommends support for “the documentation and analysis of the full diversity of the languages which survive in the world today, with highest priority given to the many languages which are closest to becoming extinct, and also to the many languages which represent the greatest diversity” (e.g., language isolates and languages belonging to undocumented or underdocumented language families) (LSA 1994:181–82).

Benign neglect and endangerment of sign languages

Despite growing awareness and concern about the problem of diminishing linguistic diversity, many sign languages – and the speech communities associated with them – have been overlooked, ignored, or forgotten. At-risk sign languages are threatened for the same complex reasons (demographic, political, economic, social, educational, etc.) that many spoken languages are threatened.³

Underdocumented and underrepresented linguistic diversity

I have conceptually outlined the problem of benign neglect of sign languages in the previous section; this section examines a real-world example of the problem in Thailand – a linguistically diverse country poised to become a regional center of sign language linguistic research – where until recently little attention has been paid to codes expressed in the manual-visual channel and where sign language endangerment is a serious problem.⁷

This dearth of information about sign languages in Thailand is unfortunate for many reasons. To begin with, we do not know how many sign languages actually exist in the country. Preliminary linguistic survey work (Woodward & Nonaka 1997, Nonaka 2000, Woodward 2000, Nonaka 2001a), conducted first in collaboration with James C. Woodward⁹

Based on recent research he conducted in Southeast Asia, Woodward (1996, 1997, 2000) has proposed a new tripartite model of sign language varieties. He outlines three general types of sign languages – national, indigenous, and original – all of which can be found in Thailand.¹¹

National sign languages (e.g., Thai Sign Language)

Modern Standard Thai Sign Language (MSTSL), often referred to simply as Thai Sign Language (TSL)¹²

Comparative lexicostatistical analysis has revealed that TSL is a language distinct from but closely related to ASL, and that the two codes belong to the same sign language family (Woodward 1996:245). Upon first consideration, this degree of linguistic relatedness between the two sign languages would seem unlikely for obvious areal-geographic reasons. However, the two codes are indeed related because of language contact and creolization (Woodward 1996:245) that occurred some 50 years ago when ASL was introduced into Thailand by Thai educators at the School for Deaf Children at Dusit District, Bangkok – the forerunner of the institution now known as Sesathien School for the Deaf (Sathaporn 1987:283).

The history of formal deaf education, and the tendency toward pedagogical vacillation between oral and manual philosophies and practices, is a complicated one. Suffice it to say that when ASL was introduced into Thailand – notably, before the emergence of sign language linguistics – it was widely assumed that special education necessitated “giving” deaf people what they supposedly lacked: the ability to communicate. In other words, it was assumed that Thai Deaf people lacked a sign language of their own, or if they did have one, that it was inferior to ASL.¹⁴

ASL was not introduced into a sociolinguistic vacuum. In fact, as Woodward 1996 has demonstrated, there were at least two, and perhaps more, original sign language varieties (and probably a number of indigenous sign languages) already present in Thailand prior to the 1950s. It was these local Thai sign language varieties, discussed in more detail below, and their creolization with ASL (Woodward 1996) that gave rise to TSL.

In addition to the national sign language, Thailand is home to a number of indigenous and original sign language varieties. Unfortunately, both types of codes tend to be vulnerable to language endangerment and extinction. Exactly how many indigenous and original sign languages exist or have existed is unknown. However, research demonstrates the presence of at least six such sign language varieties.¹⁵

Indigenous sign languages (e.g., Ban Khor Sign Language)

In contrast to national sign languages, indigenous sign languages typically are restricted in use to small (often agrarian village or island fishing) communities (Kushel 1973, Frishberg 1987, Woodward 1982a, Groce 1985, Johnson 1991, Torigoe, Takei & Kimura 1995, Branson & Miller 1996). The cultural notion of a “big D” Deaf community and Deaf culture – which so frequently corresponds with national sign languages – usually is not found in speech communities associated with indigenous sign languages. In speech communities of the latter type, it is common to find neutral to positive attitudes toward deafness, deaf people, and sign language; widespread fluency in the local sign language among hearing people in the community; and a high degree of integration of deaf people into the mainstream of village society. As mentioned before, indigenous sign languages and the sociolinguistic environments that sustain them are frequently highly endangered.

Pasa kidd or pasa bai, translated ‘language (of the) mute’¹⁶

Original sign languages (e.g., Old Bangkok and Old Chiangmai varieties)

Original sign languages are the third type of sign language variety found in Thailand. Old Bangkok Sign Language and Old Chiangmai Sign Language (Woodward 1996) are two examples of original sign languages that pre-date the national sign language. Both languages are moribund, remembered (e.g., in formal linguistic elicitation sessions) by a few signers over the age of 45 but no longer utilized in daily conversation. It is difficult to establish with precision when (i.e., time/date estimates) and where (i.e., exact speech community boundaries) these languages developed. However, as to the “why” and “how” of their development, Woodward 1997 has hypothesized that in the case of Thailand, original sign language varieties developed in market towns and urban areas, places where deaf people had opportunities to meet and interact.

From a comparative-historical linguistic perspective, learning more about original sign language varieties in Thailand is an important and increasingly urgent matter. “The development and spread of MSTSL [Modern Standard Thai Sign Language], while providing a nationally unifying force for Thai deaf people, has at the same time endangered original sign languages in Thailand” (Woodward 1996:245). This endangerment is both ironic and unfortunate. It is ironic to the extent that Old Bangkok and Old Chiangmai (and perhaps other) sign languages were the very codes that creolized with ASL to produce TSL. Loss of these original codes is unfortunate for many reasons, but particularly because they are critical “link languages” (Woodward 2000:34) that hold important clues for constructing Asian sign language families (Woodward 1993, 2000).

Expanding the known phonological inventory of signed languages

Study of Ban Khor Sign Language promises to expand the known phonological inventory of signed languages. Ban Khor Sign Language includes phonological forms that are rare, if not unique – that is, previously undescribed. For example, consider the lexemes ‘name’ and ‘foreigner’ shown in Figure 1. In the remainder of this subsection, I will examine those two signs to demonstrate how our understanding of sign language phonology might be enhanced from study of indigenous codes like Ban Khor Sign Language.

“Name” and Foreigner” in Ban Khor Sign Language.

Traditionally, sign language phonology theory has described signs in terms of four parameters: handshape, location, movement, and palm orientation.²⁰

Selected hand configurations. (Figures reprinted from The Signs of Language, Klima, Bellugi & Battison et al., 1979.)

With regard to location and movement, the two lexemes are both common and uncommon. They are common to the extent that they are made on the face. Many signs in ASL (Battison 1978), as well as in other sign languages (Nonaka 1997), are made on or near the face. Some scholars (Siple 1978b, Mandel 1981) have even suggested that signs are “built to be ‘seen’” and “shaped by the requirements of the body” (Baker-Shenk & Cokely 1980:80, 82), arguing that prolific articulation of signs in the facial area is directly linked to maximization of visual acuity.

Regardless of whether or not visual acuity bio-evolutionarily determines or drives sign production in the facial area, scholars studying historical-comparative linguistic change in ASL (Battison 1974, 1978; Frishberg 1975, Klima et al. 1979; Woodward 1976, 1978; Woodward & De Santis 1977; Woodward & Erting 1975; Woodward et al. 1976) have identified certain patterns for directional change in sign languages over time. Predictable changes include the following tendencies: (i) two-handed signs become one-handed signs; (ii) compound signs become simplified; (iii) symmetry develops in two-handed signs; (iv) production of signs made below the face/head move toward the midline, from visually peripheral to central locations; and (v) signs made on or at the face to shift to its periphery because of a dispreference for articulation of signs that block the face.²³

Clearly the Ban Khor Sign Language terms ‘name’ and ‘foreigner’ do not conform to pattern (v), the dispreference for articulation of signs that block the face such that signs made on/at the face shift to its periphery. The two signs, especially the word ‘foreigner’, are made directly on the front, central region of the face. Moreover, they do not involve movement but instead remain stationary in that area. This blocking of the face makes the signs phonologically unusual.

The two lexemes, however, are phonologically quite rare, if not “unique,”²⁴

1. Palm touching or oriented toward the signer's face
2. Palm facing outward (to right or left) away from the signer's dominant²⁵

   As Valli & Lucas (1992:19) explain, “The term dominant will be used to refer to the hand preferred for most motor tasks, and nondominant will refer to the other hand…. Signers can be characterized as being either left-handed with respect to signing or right-handed with respect to signing. For most signers with right (left) hand dominance, their right (left) hand will assume the active role most of the time.”

   side
3. Palm facing outward away from the signer's face (back of the palm touching or oriented toward the signer's face)

Examples of these three palm orientations appear in Figures 3, 4, and 5 respectively.²⁶

“B” handshape palm-orientation #1. (Figures reprinted from An English Dictionary of Japanese Signs, Japanese Federation of the Deaf, 1991; A Basic Course in American Sign Language, Humphries, Padden & O'Rourke, 1985; The Thai Sign Language Dictionary, Suwanarat et al., 1990.)

“B” handshape palm-orientation #2. (Figures reprinted from An English Dictionary of Japanese Signs, Japanese Federation of the Deaf, 1991; A Basic Course in American Sign Language, Humphries, Padden & O'Rourke, 1985; The Thai Sign Language Dictionary, Suwanarat et al., 1990.)

“B” handshape palm-orientation #3. (Figures reprinted from An English Dictionary of Japanese Signs, Japanese Federation of the Deaf, 1991; A Basic Course in American Sign Language, Humphries, Padden & O'Rourke, 1985; The Thai Sign Language Dictionary, Suwanarat et al., 1990.)

Now consider the three preceding examples of known palm orientations of “b” handshape signs with the palm orientation of the Ban Khor Sign Language terms ‘name’ and ‘foreigner’. Articulation of the two Ban Khor Sign Language words involves rotation of the arm 90° from the palm orientation of the signs in Figure 3, 180° from the palm orientation of the signs in Figure 4, and 270° from the palm orientation of the signs in Figure 5. It is important to note that most of the signs in Figures 3, 4, and 5 either are made at “peripheral” areas of the face or involve movement away from or out of the center of the face. By contrast, the two Ban Khor signs remain stationary, produced directly on the front, central region of the face.

Just as clicks in Bantu languages and dental plosives followed by bilabial trills in the Chapakuran languages Wari and Oro Win (Ladefoged & Everett 1996) expanded the known phonological inventory of spoken languages, Ban Khor Sign Language includes unusual phonological forms that promise to broaden our collective knowledge of the phonological inventory of signed languages, and by extension all natural human languages. Interviews with Deaf native users of TSL reveal that the phonological form associated with the lexemes ‘name’ and ‘foreigner’ in Ban Khor Sign Language does not exist in TSL – a fact dramatically underscored by the new TSL sign for the village of Ban Khor, which combines the Ban Khor sign ‘name’ with the TSL sign ‘there/over there’. In addition, the contact point – ulnar side of the hand touching the face – associated with the signs ‘name’ and ‘foreigner’ in Ban Khor Sign Language is also an impossible contact point in ASL, and it would not be predicted to occur in other sign languages based on what is currently known about sign language phonology.²⁷

Basic color terminology in Ban Khor Sign Language

Color terminology has long been a standard component of anthropological linguistic inquiry. Traditionally such research has encompassed two broad domains of study – descriptive/linguistic and perceptual/psychological – that have been conducted both for code-specific linguistic documentation and for universal cross-linguistic comparison. In the 20th century, color became “the empirical domain par excellence in which to argue for (and against) the linguistic relativity thesis” (Berlin 2001:27).

The literature on color terminology and perception has grown exponentially since publication of Berlin & Kay's (1969) classic study.²⁸

Preliminary research into basic color terms in Ban Khor Sign Language appears both to confirm and to challenge aspects of the traditional theory of universal color terminology. Ban Khor Sign Language is a three-color-term language (Nonaka 2001b). Consistent with Berlin & Kay's (1969) original prediction, those three basic color terms, shown in Figure 6, are ‘black’, ‘white’, and ‘red’.

The three basic color terms in Ban Khor Sign Language.

Expression of other colors in Ban Khor Sign Language is achieved in one of two ways (Nonaka 2001b). First, a visual search of the immediate physical environment is undertaken. If an object of the appropriate color can be found, then the speaker points to that object to indicate the color in question. For example, in Figure 7 a native Ban Khor signer describes the color of a carrot by pointing to an orange notebook atop a nearby table: ‘A carrot is … [color search is launched] … orange like this notebook [consultant points to an orange notebook].’

Expression of non-basic color terms. Example: A successful color search of objects in the immediate environment “A carrot is … [color search launched] …” “… orange like this notebook [consultant points to orange notebook].”

Not all color searches, however, are successful. If the color in question cannot be located in the immediate environment, then nonbasic color terms are expressed using one of the three basic color terms – black, white, or red. As an example, see Figure 8. Having scanned the local environment without locating any object similar in color to the deep purple hue characteristic of an eggplant, the linguistic consultant code-switched into Thai Sign Language and said that she really wanted to describe the eggplant as ‘purple’.²⁹

Expression of non-basic color terms. Example: An unsuccessful color search of objects in the immediate environment “An eggplant is … [color search launched but when no object of the appropriate color is found] …” (a) [code switching to Thai Sign Language] … “I really want to say ‘purple’ but there's no sign for ‘purple’ (in Ban Khor Sign Language) …” (b) [code switching back to Ban Khor Sign Language] … “It's black.”

To the extent that Ban Khor Sign Language has three color terms, and those terms are ‘black’, ‘white’, and ‘red’, the indigenous sign language appears to conform to Berlin & Kay's theory. However, Ban Khor Sign Language also appears to contradict their theory in at least one important way: by challenging the operational definition of a “basic” color term. By definition, basic color terms should refer to pure color and not to an object representing the color; thus, gold, turquoise, and orange are not basic color terms. In Ban Khor Sign Language, however, all three basic color terms are signed by pointing to objects – body parts – that represent those colors: hair for ‘black’, teeth for ‘white’, and lips for ‘red’ (see Figure 6).

It is important to note here that ‘black’, ‘white’, and ‘red’ are fully lexicalized in Ban Khor Sign Language. Thus, the pointing³¹

Finally, studying differences and similarities between the color lexicons of the various sign languages in Thailand is useful for comparative and historical linguistic purposes. For example, there is evidence that, like Ban Khor Sign Language, both Old Bangkok Sign Language (Figure 9) and Old Chiangmai Sign Language (Figure 10) were three-color term languages.³³

The three basic color terms in Old Bangkok Sign Language.

The three basic color terms in Old Chiangmai Sign Language.

TSL has a markedly different color term system: a multi-term color lexicon. Interestingly, however, careful analysis of the various TSL color terms reveals two different types of color vocabulary. As in the country's other sign languages, the three most basic color terms in TSL refer to body parts bearing the appropriate color – eyebrow/‘black’, upper arm/‘white’, and lips/‘red’ (Figure 11). However, as the examples in Figure 12 show, other color terms (e.g., ‘pink’, ‘purple’, ‘yellow’) are “initialized signs,” meaning that they are made using a fingerspelled letter from the Thai manual alphabet.³⁵

The three basic color terms in Thai Sign Language.

Examples of initialized color terms in Thai Sign Language. (Figures reprinted from The Thai Sign Language Dictionary, Suwanarat et al., 1990.)

Most TSL color words are initialized signs. For example, the sign for ‘pink’ is made using the initialized sign “ch” for chomphuu in sii chomphuu, the Thai word for ‘pink’. Similarly, the TSL sign for ‘purple’ is made with the fingerspelled letter “m,” which is the first letter of muang in sii muang ‘purple’, and ‘yellow’, or sii luang, is initialized with the letter “l.” These three signs are typical of the general workings of the second type of TSL color term vocabulary – initialized signs that point to language contact and borrowing from Thai.

Clearly, initialized signs are of a different order than the signs for ‘black’, ‘white’, and ‘red’ in TSL; and the TSL multi-color lexicon is distinct from those of the other known sign languages found in Thailand. Thus, research on color terminology in sign languages in Thailand would appear to be promising for cross-linguistic comparative descriptive purposes, historical linguistic purposes, and perhaps for universal comparative purposes.

Baby talk or motherese in Ban Khor Sign Language

In most languages and cultures – although not all (Ochs & Schieffelin 1984, 1986) – adult caregivers, particularly mothers, use a special form of their native language that is specifically modified for use with infants and/or small children. This linguistic register is known as “baby talk” or “motherese,” and it comprises “any special form of a language which is regarded by a speech community as being primarily appropriate for talking to young children and which is generally regarded as not the normal adult use of language” (Ferguson 1964, reprinted in Dil 1971:114).

Where it occurs, baby talk or motherese is commonly assumed to simplify language learning and to contribute to the linguistic development of children. In languages and speech communities where baby talk/motherese is used, a typical or “canonical form develops” (Ferguson in Dil 1971:123). Linguistically, baby talk/motherese involves noticeable changes in phonology, syntax, and lexicon, and even in the paralinguistics of normal language, including the use of special lexical items, mostly monosyllabic words, reduplications, and lack of inflectional affixes (Ferguson in Dil 1971:116–24). Considerable comparative research has been conducted on baby talk/motherese in spoken languages (e.g., Ferguson 1964, 1977; Von Raffler-Engel & Lebrun 1976; Williamson 1979; Blount 1982; Zeidner 1983). Far less is known about baby talk/motherese in signed languages, especially codes other than ASL.

Preliminary research (Nonaka 2001, 2004) indicates that Ban Khor Sign Language has a form of baby talk/motherese. Adults use the register to communicate with infants and small children aged two years or younger. Characteristic features of the special linguistic register include (i) heightened affect, (ii) active physical stimulation of the child, (iii) signing more slowly than usual, (iv) signing close to the child to maximize visual attention, (v) signing on the child's body, and (vi) repetition.

All the features of baby talk/motherese in Ban Khor Sign Language described above are found in the 16-second segment of natural conversation transcribed below. Differences in modality (manual-visual sign language vs. printed text) make it impossible to fully depict the videotaped sequence of talk here. Nevertheless, where it was feasible, still frame-grabs were extracted and are presented in Figures 13, 14, and 15 to illustrate four of the features – heightened affect, active stimulation of the child, signing close to the child to maximize visual attention, and signing on the child's body – of baby talk/motherese in Ban Khor Sign Language.

As indicated in Line 1 and depicted in Figure 13, Mrs. Vien exhibits classic paralinguistic features of baby talk/motherese in Ban Khor Sign. Throughout her interaction with the child, Mrs. Vien demonstrates heightened affect: lots of smiling, highly raised eyebrows and widely opened eyes. She also incorporates lots of active physical stimulation of the child – rubbing the baby's face, stroking her nose, and playfully slapping her hands.

Characteristic features of baby talk/motherese in Ban Khor Sign Language, heightened affect and active physical stimulation of the child.

Characteristic features of baby talk/motherese in Ban Khor Sign Language, signing close to the child to maximize visual attention.

Characteristic features of baby talk/motherese in Ban Khor Sign Language, signing on the child's body.

Mr. Phaiwan's talk too is classic baby talk/motherese. His initial child-directed utterances (lines 1–2) are made while bending directly over Watermelon. The signs ‘you little-birdie’ and ‘feed you’ (Figure 14) are articulated very close to the infant's face, maximizing visual acuity. Furthermore, the entire sequence, especially ‘feed you’, is signed more slowly than in normal adult conversation.

In line 4, Mr. Phaiwan tells the baby, who was born prematurely, that as a result of being fed, ‘Your tummy will get big’, and ‘That's good!’. While signing ‘tummy get big’, (Figure 15) he actually starts the utterance ‘tummy get big’ by touching his hand to the girl's abdomen. Signing on an interlocutor's body is rarely used between adults for purposes of normal, everyday conversation, but the strategy is a classic feature of signed baby talk/motherese.³⁷

Interestingly, the characteristic features of baby talk/motherese in Ban Khor Sign Language also are found in ASL (Erting et al. 1990:105). Given the very different socioeconomic, historical, and cultural contexts in which Ban Khor Sign Language, an indigenous sign language, and ASL, a national sign language, developed, documentation and analysis of the baby talk/motherese registers in the two languages offer a rich opportunity not only for original description but also for cross-linguistic comparison. Moreover, original linguistic description and cross-linguistic comparison might well allow for broader theorizing about universal linguistic features of baby talk/motherese in signed languages and, by extension, in human language generally. As a working hypothesis: The defining features of baby talk/motherese in Ban Khor Sign Language will resemble those of all other human languages in form (phonological, syntactic, lexical, etc.) and function (language acquisition and enculturation), differing only in terms of modality (expression through the manual-visual channel).³⁸