A paper prepared for Frank Barnes School,
London, 17th February 1995
For many years we have studies the development of hearing children in their acquisition of spoken language. We have proposed many different theories. In very simple terms these theories propose that
No single module fully describes what children do in achieving a fluency in their native language by he age of 5 years. It is a major task. Whenever a child does not follow the pattern exactly or where the learning proceeds more slowly, then we become concerned. Different countries have different responses, but in the UK the child is assessed and given special support. The child has special needs. The problem with this is that it suggests that the only language which has to be learned is English. It ignores the situation where a child may develop another language.
Deaf Children create a double problem - first they do not develop spoken language directly because they do not hear the language around them and so have special needs; but secondly, they seem to develop another form of communication altogether - sign language. For many years there has been confusion about this second option. Signs were thought to be concrete and limited. Deaf people were thought to be limited by this impoverished language. However, our linguistic research on deaf adults show that t the language is extensive and rich. In Bristol, we use the language to teach University students - Linguists, Psychology, Education and so on.
But we still have the problem of knowing what it is that a child does when he or she learns to sign. We need to know about this so that we can advise parents and teachers and we need to know about it to determine when a child is progressing too slowly and it will also be a contribution to our knowledge about languages in general. In Bristol we have studied this development in deaf families of deaf children. This constitutes the natural language situation comparable to that of hearing families. It is the stating point for understanding the development of sign.
The research and the findings
Working on this topic has been problematic. One of the reasons has been the difficulty of recording sign acquisition. In Britain, deaf children who will develop sign language occur in only one in 1500 births, and those born into deaf families(deaf children with deaf parents - dcdp) occur perhaps once in 40,000 births. The situation of parent passing on the language to the child is rare. Studies of individual children are still the norm and continuous studies of even five children (eg Ackerman et al, 1990) are large. Where it has been possible to record deaf children from an early age a number of striking findings can be reported. Each area below will be illustrated by some examples on video,
Joint reference: The ability of a hearing parent to discuss an object or event while a child is observing it , is an important feature of interaction in infancy and early childhood; if a deaf child engages with an object however, the communication channel(visual) is cut off. Deaf mothers overcome this problem partly by signing rather less than hearing mothers speak, but also by referring to objects prior to engaging with them. So in a task, which involves directing the child's gaze, deaf mothers refer to the object when they have eye contact with the child and then point to it. While the child is looking at the object, the mother does not communicate and only as the child's eye contact is returned, is there any further information provided by the mother. The rule is simple - deaf mothers always refer to the object before they point to it. They will ensure that the child has some information before they have to look. Hearing mothers tend to point and try to add the comments while the child is looking.
The purpose of this sort of interaction which appears slower by hearing-speaking standards, is to establish attention routines in order that much more extended interaction and turn-taking can occur in the second year. This is vital to the subsequent development of language. The fact that this interaction is often unsatisfactory in older deaf children in hearing homes is indicative of this earlier failures in dealing with divided attention. It is noticeable that hearing mothers of deaf children do not naturally develop these skills. Efforts to develop bilingual approaches or to introduce early signing to hearing families has to take into account these facts.
Motherese: Just as we detect baby talk in hearing interaction, we find similar characteristics in dcdp. Signs are lengthened in duration, extended in space and manipulated in space(perhaps to increase attention); signs are displaced from their normal location to locations appropriate to the child's attention eg the mother will sign key concepts on top of a picture book or near an object with which the child is playing.
Signs can also be made "on the child" and manipulated as in a game eg the sign DUCK (made by opening and closing the hand in the manner of a duck's beak) can be used as if the imaginary duck was pecking at the child. One further significant feature is that not only does the mother model sign for the child, she can model the signs on the child - she takes the child's hands and moves them in an appropriate way for that sign. This can be done when the mother is facing the child and also, perhaps more commonly, when the mother is sitting side-by-side with the child.
Emerging Signs: Just as with hearing children, signs begin to emerge reliably around one year. First signs tend to be object signs(seen around 11 months of age, Ackerman et al, 1990) although in the earliest stage these are difficult to distinguish from gestures (hearing children also gesture and these can be detected from 15 months - Acredolo and Goodwyn, 1988). These early action gestures are probably part of general actiona schemata.
These action schemata are the child's attempt to represent the world of objects, actions and events and they become the tools for communication. We know that only deaf children make the leap to productive use of these action schemata in the combinations of representational gestures which form the basis of the language which they will eventually master.
These early signs are poorly articulated in terms of handshape, location and movement. Parents interpret the sign from the context and frequently provide a correction. It appears that signs are pre-lexical(as with words)(see also Nelson and Lucariello, 1985) until around 16 months of age when they begin to be generalised and where over-extension can be noted. This means that these first signs tend to occur in game-playing routines(such as "what's this?") or in turn-taking (where the mother creates an expectation of a child response by presenting an object or picture). The fact that the child learns to produce the same response on different occasions does not mean that the child has internalised the true lexical and conceptual meaning. That will come a little later.
Sign Combinations: As with words in hearing children signs come to be combined, but appear to pass through an intermediate stage prior to emerging as two-sign combinations. This intermediate stage is one of bracketing. In this, dcdp combine single signs with POINTS ie a gesture signifying a location, usually made with the index finger as
POINT sign POINT or sign POINT sign
before they move to SIGN1 SIGN2. Since we are not yet absolutely clear as to when two contiguous words become a two-word utterance rather than individual words(the issue is not clear cut in sign).
The signs begin to develop into two and three sign utterances. However, it would also seem that sign utterances tend to be shorter than corresponding length of utterance in words. This may be partly due to the fact that the signs are more densely packed as units but is also likely to be a function of the relatively shorter conversation length possible for a young child when he/she has to engage in eye contact throughout.
Question Development: An interesting feature of sign competence is the use of question forms. Sign languages have wh-forms, but users are most likely to indicate questions by the use of non-manual markers(notably use of eyebrows). Initial work on the acquisition of question forms suggest that deaf children and their caregivers tend to avoid such question marking and the only distinguishing feature of a question appears to be the "terminal hold" - the holding of the last sign for an extended period while maintaining eye contact(Kyle, Woll and Ackerman, in preparation) . One possible explanation is that deaf children tend to learn the emotive connotation of the face before they learn its use for non-manual grammar(McIntire and Reilly, 1985). Early sign questions if marked by knit brows would be interpreted emotionally as admonition by the young child. As a result question marking in the early stages has to adopt a different framework.
Further Sign Development: Beyond these early stages we see the full developmental patter unfold. Deaf children interact in the way that hearing children do and we see extended conversations developing. Deaf children negotiate the meaning in the same way that hearing children do in speech. By the age of three and four years we can see the same level of sophistication in deaf children’s signing as we do in hearing children’s speech. The video examples will show this.
Sign in School: Beyond this period there has been more concern with the applied field of education as the deaf child will be admitted into educational programmes from as early as two years of age. Although we can see sophisticated use of sign by dchp(deaf children from hearing families) there can be significant problems. Pidgin or sign supported speech forms are used by the hearing adults around the child.
Hoiting and Loncke(1990) have analysed the typically atypical language learning situation of the deaf child. It incorporates language input from both sign and speech and also may be presented in forms which mix both. Nevertheless, they conclude that the cognitive-linguistic system of deaf children is more linked to sign language than to spoken language. Deaf children do separate spoken forms from sign forms in early childhood and peer interaction is most often sign dominated.
In an examination of sign competence in deaf children in school programmes where sign was used, Kyle(1990) reports a pattern which indicates learning over time in the programme, but disappointing levels of mastery of sign grammar. Features such as the use of spatial grammar(location of actions, objects and people in space) and movement parameters(where movement and direction indicate aspect and person of the verb) are incompletely understood by deaf children up to the age of 10 years (although this does not apply to dcdp). Since a great deal of early interaction will deal with objects and story sequences this could prove to be a significant obstacle to learning. In effect, in the programmes studied, the sign acquisition of the children was not adequate as a result of the nature of the input by adults - a fact reflected in Johnson et al(1989). Loncke et al(1990) also investigated the acquisition of grammar in dchp (deaf children with hearing parents) in contrast to dcdp. They found problems in the former's use of morphosyntactic operations, while the latter seem to incorporate them in their signing from the age of 6 years onwards.
Sign development can be seen to reach levels comparable to spoken language but the process whereby one reaches this point differs between deaf and hearing children. In most education systems at present, only those from deaf families are likely to achieve natural sign competence at a rate comparable to hearing children. It can be expected that components of sign morphology and syntax will be acquired rather late(at school rather than at home) by the majority of deaf children. It can be seen that there is some urgency for the preparation of accessible information to those involved with the deaf child in bilingual programmes, as well as the need to contribute to the research debate.
Ackerman J, Kyle JG, Woll B and Ezra, M (1991) Lexical acquisition in sign and speech, in Lucas C (ed) Sign Language Research: Theoretical Issues, Washington, DC: Gallaudet University Press
Ackerman J and Woll B (1990) Deaf and Hearing children learning to sign and speak, in J Kyle (ed) Deafness and Sign Language into the 1990s, Bristol: Deaf Studies Trust
Acredolo L and Goodwyn S (1988) Symbolic gesturing in normal infants, Child Development, 59, 450-66
Bonvillian JD, Orlansky MD and Novak LL (1983) Early Sign Language Acquisition and its relation to cognitive and motor development, in JG Kyle and B Woll (eds) Language in Sign, London: Croom Helm
Deuchar M (1984) British Sign Language, London: Routledge