2.1 Introduction

The normal child begins to use meaningful utterances right from her birth, even though the discrete linguistic units may appear only later on. The same thing must be expected also of the hearing impaired child; in this case, the emergence of the discrete linguistic units may be delayed and their functions may not be clear, or, may not be that weighty as found in the normal child’s speech. In any case, before, during and after the emergence of discrete linguistic units, the normal child goes through a simultaneous development on the receptive side. While the ontogeny of speech production in the normal has received tremendous attention from the investigators, the ontogeny of reception in the normal has not received equal attention. In the case of the hearing impaired, both speech production and speech reception have received only very little attention. The antogenetic stages of speech production and speech reception are not discrete and the segmental and supra-segmental aspects of one stage may overlap with the next. Note also that each of these stages is characterized by previously unobserved linguistic units of various linguistic levels. Perhaps this is the only justification that one could find for establishing “stages”. This defining characteristic, however, is not tied down to any one particular item in the stage across children.

Investigators have identified several stages in the production of oral sounds. In recent publications, Cruttenden (1979) identified two stages: neonatal period characterized by development and distinction of cries, and segmentals and the babbling period (approximately 0 : 9 to 0 : 10). According to Stark (1979), the five early stages in sequence include reflexive crying and vegetative sounds (0 to 8th week), cooing and laughter (8 to 20th week), reduplicated babbling (25th to 50th week), non-reduplicated bubbling and expressive jurgon (9th to 18th month). Van Riper (1978) distinguished the following stages: crying v. comfort sounds (upto 2 months), private babbling (8 weeks), vocal play (5th or 6th month) and inflected vocal play (8th month). Reflexive vocalization, babbling (6th month), lalling (around 2nd to 6th month) and echolalia (9th to 10th month) precede true speech, according to Berry and Eisenson (1956). Blache (1978) recognized the following stages: reflex vocalization, babbling (between 2nd and 6th month) followed by stage 3 characterized by lalling, inflected vocal play, and echolalia (between 6th to 12th month). The child begins to use conventional speech patterns between 10 and 19 months (Van Riper, 1971; Berry and Eisenson, 1956), 12 to 16 months (Blache, 1978), and by the end of the first year (Cruttenden, 1979).

For some years, researchers have concentrated their observation upon the prelinguistic tendency to produce stably significant but idiosyncratic sounds (as an example, Dore, Franklin, Miller and Ramer, 1976). Werner and Kaplan (1952), for example, have used the term vocable as a general purpose indicator of the significant pre-linguistic vocalization, and sensory-motor morpheme, when gestural context, gestural significance, or, in general, non-semantic significance, is involved as well. Still others have concentrated their attention on nonverbal language development in the early stages. Gray (1977) has studied how a child learns to take an object from the mother; Clark (1978) has studied the transitions from action to gesture. A linguistic study by Oller, wieman, Doyle and Ross (1976) on normal children tends to support the notion that there is much similarity and progressive development between the early stages of vocalization and the later symbolic period.

While much work has been done in normal children on verbal linguistic aspects, longitudianal studies in the hearing impaired during the early linguistic periods are not found much. There is a need for the detailed analysis of segmental and supra-segmental aspects of the vocalizations in the hearing impaired. These results may be compared with those of normal children to facilitate the quantitative and qualitative deviancies in the hearing impaire’s oral production. The reason for the lack of such studies could have been due mainly to the difficulties faced as regards early identification of the hearing impaired children. The disorder of hearing impairment is invisible directly to the eyes. So, it is identified by many parents at a later age only. The hearing impaired babies typically stop to babble or tend to remain silent from around six months. This may be the first or the only symptom identified by the parents. This may be true especially in India, wherein there are hardly any programmes in the hospitals to identify the high-risk babies. An initial attempt to detect high-risk babies and develop a high-risk register for Indian population has been made by Ashok Kumar (1981). Unfortunately, this work remains unpublished.

The importance of linguistic analysis of the hearing impaired children’s oral productions will be recognized with greater strength, if their relationship to later linguistic development is identified. The debate amongst the researchers along these lines in discussed later in this chapter.

2.2. Vocalization Studies

Several questions may be raised when one studies the vocalizations of young children. The first question is what vocalization is and what items can be constructed as vocalizations. The other relevant points to be considered are the classification of vocalizations, identification of units of vocalizations in relation to adult linguistic vocalization units, comparison of the stages of the emergence of vocalization units in the hearing impaired with those of the normal children, and the specific relations between the units of vocalizations and the beginning of the language.

Cries and noises made through oral and nasal cavities are termed vocalizations. Though all the cries and noises made by the baby are all pregnant with communicative potential, one may posit two broad stages in the ontogeny of cries and noises. The first one is, indeed, very brief, during which the baby cries and makes noises as a self-expressive mode. In the later stage, the baby cries with a deliberate communicative intent, or at least the baby recognizes/seeks the presence of ‘the other’ for some purpose. The former may be compared to the egocentric speech of children. These early self-expressive vocalization have an activating function – these activate the process of communication, and slowly tunes the organs of speech communication. From the self-expressive role, the later vocalizations take the child to the communicative context wherein the child and some other persons/objects are involved. At this later stage there is a possibility of recognizing certain recurring patterns and identifying certain categories or manifestations of vocalizations. We already said that the cries and other types of noises from part of vocalizations. Under the other types of noises we include babbling and lalling also. Attempts have been made to analyze babbling into various sounds and link these sounds with the emergence of certain speech sounds. Attempts have also been made to categorize cries into various types. The problem faced by the investigators is to link the classification of cries with the classification and categories of babbling and lalling, and to identify the processes by which the units of one type (for instance, the cries) change into the units of another type, assuming that there is a progression is assumed, the investigator faces the problem of allocating functions to the concurrently occurring types and the units underlying them. For example, Menyuk (1972) classifies the cries of an infant into linguistic units – single cries are comparable to phonemes. They are also comparable to morphemes in adult language. The cry cycles are analogous to morpheme combinations, sentence-like semantic units. The cry sequences may be compared to a paragraph of speech. Menyuk’s attempt is rather ambitious; what we should seek are the recurring formal patterns, to begin with. The duration and intensity and, in general, the acoustical quality of the cries must be considered, instead of looking for units comparable to those of adult speech. Indeed, it is too early in the ontogenesis of speech to expect phoneme-like or even morpheme-like units in vocalizations. The analysis of vocalization is more insightful and faithful to real conditions, only when conducted on the model and basis of the emergence and analysis of thoughts in children. Children exhibit amorphous and syncretic thoughts. The amorphous thoughts are sculpted into individual thought units which are later combined to form sequential structures; the post-babbling speech is also amorphous in the sense that the form of this speech is quite different from that of later day child’s speech and the speech of the adults. The amorphous speech is slowly sculpted into units of phonology and other levels through a process of approximations to the adult language. We argue, then, that a study of cries should not look for phoneme-like and even morpheme-like units, but only for several amorphous units of cries based on duration, intensity and other acoustical qualities. These amorphous units should be seen as transforming into linguistically oriented amorphous units.

Options differ on the relevance of the studies of vocalizations, including the studies of cries, for the early detection of hearing impairment in babies. Some investigators report that the characteristics of cries of the hearing impaired children do not differ from those of normal children upto the age of four years (Sedlačkova, 1973). Some other investigators have suggested that both crying and laughter in deaf children are atypical in inflection, pitch range, and acoustic quality. It is claimed that the cries are flat in intonational patterns and that the intensity variations may be unusually great. There are two methodological issues involved here. Firstly, there is great difficulty in establishing through a single means that the subjects (young babies) under investigation are hearing impaired. Any decision about their hearing impairment status has to wait until the subjects could be put through audiological and other tests later on. The very young babies are difficult to be subjected to these tests. This situation is in itself not unscientific, although it has all the elements of the fallacy of circumlocution. Secondly, the two view listed above are based on their underlying assumption as to where there is any link between the vocalizations and later day speech. The first view is based on the assumption that there is no link between vocalizations and later day speech and the second view is based on the assumption that there is a link between vocalizations and later day speech. The present investigators argue that vocalizations may be seen linked not only to later day speech but also to the thought process. The subjects with profound hearing impairment are also seen to exhibit some sort of vocalizations in their efforts to communicate. Vocalization must, indeed, be seen as a launching pad for both speech and thought processes, a mode basic to both. Vocalization, in our view, then, is a point from which the unity of thought and speech begins to proceed on separate (perhaps parallel) lines (to meet again in language use). We view, then, vocalization as neutral to both speech development and hearing impairment. The effects of the hearing impairment must be sought, not in the basic vocalization, but in the post-vocalization wherein a transfer from vocalizations including babbling is being made to linguistic speech. Our conclusion regarding the neutrality of vocalizations in relation to hearing impairment is strengthened also by Lenneberg (1965) who made recordings to spontaneous vocalizations to deaf babies from the first month. He found that the voices of the deaf babies were similar to those of hearing children and that the cooing sounds, laughter and sounds of discomfort were not different from those of normal subjects.

2.1. Studies on Babbling

During babbling, control of the articulatory apparatus is achieved. This control is exhibited through the child’s ability to produce a few sounds that are linguistically relatable to later speech. These sounds are generally stops and nasals. The child also comes to realize that she can play with these sounds. The tuning of the organs for speech is the major function of the babbling stage. Is this tuning dependent on any building up of a communicative context or is it mainly a self-expressive mode? Is hearing in any way connected with the quality of babbling?

Lenneberg (1965) found that the hearing children were much more vocal in the presence of others than the deaf. While the deaf had a tendency to engage in certain types of noise more frequently, the hearing produced a wide range of speech sounds in the babbling period. Mavilya (1972) found that babbling was present in the 16 week old hearing impaired infants. Babbling was at its peak by 23 weeks, 17 weeks and 25 weeks in three subjects but an abrupt decline was noticed in 24 weeks, 19 weeks and 26 weeks respectively. With the decrease in babbling, there was an increase in silence, confirming what Eisenson (1958) concluded earlier that the speech development of the congenital deaf ends with the babbling stage. There are many studies which take the same position.

Some scholars (for example, Mowrer, 1960) have argued that the deaf infants do not babble or babble only very little, because there is no secondary reinforcement to generalize their own vocalizations. This view relates the development of babbling to the development of auditory feedback loop. Fry (1966, 1973) finds a strong link between the tactual and kinesthetic impressions and the auditory sensations that the child receives from his own utterances. The present investigators do not find it necessary to ascribe all such functions to the babbling stage. It has the function of tuning the speech organs and the resultant babbling stream is more basic. It is seen in all disordered children, and in some it may be the only form of ‘vocalization’ in later day ‘speech’. Accordingly, our argument is that the link between the tactual and kinesthetic impressions and the auditory sensations are built slowly with the explicit emergence of linguistic sounds in children and not connected with babbling. This is a very complicated process closely linked with the development of processes of speech perception.

The effect of hearing impairment is felt when speech communication is to begin at the fading of babbling. In other words, the question is about the role of babbling in the development of language acquisition. The essentiality of babbling for normal language development is to be demonstrated through its relation to the emergence of speech sounds of the language in question. So long, babbling is considered a universal feature, not related in any specific manner to the sounds of the language to which the baby is exposed. Secondly, if babbling is essential, the essentiality should be demonstrated in terms of its role in the child’s discovery of language rules. The second demand is rather very stringent and yet it is a reasonable one.

With regard to the relationship of babbling to later phonological development, there have been two sets of theories. We have also earlier referred to such a distinction. The two sets are: (i) the discontinuity theory and (ii) the continuity theory.

The former theorists hold the position that there is no relationship between the early vocalizations or babbling for later phonological development. Jakobson (1968) and Lenneberg (1967) are the major proponents of the discontinuity theory. In passing from pre-language stage to the first acquisition of words, the child nearly loses all of his ability to produce sounds (Jakobson, 1968).

Lenneberg (1965, 1967 and 1973) regarded babbling to be a period of biological maturation. Cooing and babbling stages are not a sort of practice for future verbal behaviour. An example of a 14 month tracheotomized child for a period of six months was presented. A day after the tube was removed, and the opening was closed, the child produced babbling sounds typical of that age. No practice or response with hearing his own vocalizations was required. Comparable observations were noted in other hospitalized children. Some of the other studies that conform to this view are those of Irwin (1947a,b, 1948), Irwin and chen (1946), and Carroll (1961). Note, however, that Gilbert (1982) presents a severe criticism of Lenneberg’s position. He says that it is, indeed, a myth to assume that deaf infants babble until the age of six months and then stop. He finds that the data provided by Lenneberg is inadequate. Gilbert’s data does not show any significant difference in cooing, crying, and fussing in children of deaf and hearing parents. Since there is no large scale study of babbling in deaf children has been conducted, the conclusions on the subjects must be treated as unfounded, according to Gilbert.

In contrast, the continuity theorists state that there is a continuity in the development of oral speech production from one stage to another. There fore, babbling is gradually shaped to meaningful words. Some of the studies that conform to this theory are those of Moskowitz (1970) and Stark, Rose McLagen (1975), Stark (1979), Stern, Jaffa et al. (1975), Bates (1973), Elliot (1982), Nakazima (1978), Maccoby (1966), Blount (1970), Cruttenden (1970), Oller, Wieman, Doyle, and Ross (1976), and De Villiers (1978). In addition to the above two theories, there are at least four more positions which also require our attention. These are: Babbling as a reinforcement (Mowrer, 1950; Winitz, 1969), Babbling as an imitation (Van Riper, 1971), Babbling as a neurological process (Salus and Salus, 1973) and Babbling as a pre-paratory process for later speech (Berry and Eisenson, 1956; Eisenson, Aller and Irwin, 1963; Berry, 1969; Hurlock, 1956.

Most of the studies reviewed above suggests that continuity exists from one stage to next stage in the first year of language acquisition. However, in our largely impressionistic survey of the hearing impaired children, we find that there is stoppage of linguistic behaviour after babbling. The hearing impaired children do not pass through the lalling stage (as defined in Berry, 1969). This is the stage in which hearing and sound production become associated. The hearing impaired do not also pass through the echolalic stage. This has been noticed by several investigators. The stoppage of linguistic behaviour in hearing impaired after babbling has instigated researchers to seek qualitative deviances at the babbling stage in the hearing impaired. Of the theories discussed earlier, the theories of reinforcement, and imitation give primary importance to auditory sense in babbling and this auditory sense is deficient in the hearing impaired, which hinders further speech development. Thus, one is tempted to conclude that within the framework of the above theories, the hearing impairment is an independent factor the influence of which becomes more apparent and drastic during (?)/after babbling. This has led to suggestions that, if external sounds are made audible to the child, then, the breakdown in the process of speech development may be prevented. Whetnall and Fry (1964) suggest that if suitable auditory prostheses are provided, the child will, then, pass through the various stages of normal development. If is also necessary that suitable reinforcements do not increase the vocalizations in infants (Weisberg, 1963). Social reinforcements such as a smile, a word, tackling the subject’s chin, etc., increase vocalizations. The studies suggest that the vocalization rates and the length of vocalizations can be altered with reinforcing stimulus. While some studies suggest that socialized reinforcements are important, others have reported that the auditory alone, visual alone, and tactile alone, may result in the change in the rate of vocalization in normal infants.

The prosent investigators are of the opinion that there is no guarantee that adequate reinforcements would lead to better vocalizations in the hearing impared. Secondly, there is no guarantee that better vocalizations in the hearing impaired would automatically lead to acquisition of speech. Thirdly, the result that may be obtained through reinforcement is of a limited nature. Stability and continued use of items learnt through reinforcement is possible only for a few items. The manipulation of the items learnt to meet newer situations is not generated by reinforcement. If appears to us that hearing must be considered independent of production – both act as parallel lines until soon after babbling, when linguistic sounds and linguistic structures begin to emerge; the emergence of these is, indeed, the combination of, or the beginning of the combination of speech production and hearing. Relationship between language and thought is an excellent model to explore the relationship/ontogenesis of speech and hearing.

2.2. Prelingual Deafness

In addition to attempts to relate certain characteristics and stoppage of babbling to prelingual deafness, researchers have tried to convert the prelingual deafness to the characteistics of rate of heart beat, state of alertness and orientation to sounds by new-born infants (Muir and Field, 1979; Eisenberg, 1976; Suzuki, 1978), and spectrographic analysis of the newborn infant cry (SirviÖ and Michelsson, 1976). Research findings of these attempts are not conclusive. Evidence, however, is mounting in support of ‘incipient affective system as well as a linguistic processing system’ (Eisenberg, 1976). And yet at the moment the best we could say about these attempts is that these findings are likely to be useful as a supplementary tool for evaluating hearing in infants.

Another important dimension of research as regards prelingual deafness is the role of cognition vis-a-vis prelingual deafness. We have referred to the possible debilitating influences the early deafness has for both normal behaviour and language acquisition, in the first chapter. The study of auditory deprivation from birth indicates possible permanent deficiency at a cortical level in the hearing of complex sounds, for example, speech (Kyle, 1978). We noticed in our impressionistic survey of young deaf children that in most of them there is a resistance to use language. They start crying, they use jargon, tantrums, reduplication and onomatopoeia in place of actual language sounds/syllables/utterances, etc. In some we found willingness to produce speech if only they have been exposed to language continuously and if they have, thus, internalized some language structure before actual production of speech. When cognition and language are viewed as separate and independent entities, cognition alone does not become a sufficient reason for language acquisition, since language exhibits utterances not connected with thought processes. That is, even as there is prelinguistic thought, there is also language without thought. However, in normal language acquisition and normal cognitive development, one easily attests the beneficial mutual influences of the two. This beneficial mutual influence is conspicuous by its absence in the early deaf. There are some problems of ideation as evidenced in our subject’s inability to use and comprehend simple abstract relations through linguistic terms and simple abstract items such as colour and possession. But this comes to be proved only at a later stage, say, around 3 years in our subjects. We were not able to identify the problems of ideation explicitly in ages prior to this age group, perhaps because deficiency at this stage is easily comparable between the normal hearing and hearing impaired children through language use. And yet we should identify if and when differences emerge in the cognitive development of deaf children as compared to hearing children and the nature of the differences.

There is some relationship between middle ear pathology early in a child’s life and subsequent auditory processing deficits (Zinkus and Gottbeg, 1978). Chronic otitis media may affect early language development, in particular perception difficulties are noticed with regard to sounds (Dobie and Berlin, 1979 and Menyuk, 1979). Shah, et al. (1978) find that mean age for suspicion of congenital hearing impairment is 16 months.

In the study of language acquisition by normal hearing children, there has been a lot of interest with regard to rate and quality of language acquisition in relation to mother’s speech uttered to children. While there are not many studies on the language used by mothers to speech to/interact with their deaf children, the available studies have brought out some significant findings. The mother of the hearing-impaired child is seen to spend more time monitoring the child and less time playing than the mother of the normally hearing child (Neinhuys and Tikotin, 1983). There are conflicting reports of the role of mother’s utterances in the early language development of deaf children. While Kenworthy (1986) reports that caregiver language input may be significantly affected by the presence of a hearing impairment in the child, Cheskin (1981) found that all mothers used most of their language to describe or control behaviour. she also found that the maternal utterances had five functions: (i) controlling child’s behaviour, (ii) prodding, (iii) eliciting, (iv) describing objects and events in the environment, and (v) engaging in incidental conversation. Cheskin (1981) reported that each mother spoke in short sentences that were usually grammatically complete, used a repetitious and restrictive vocabulary, and repeated her own utterances far more frequently than do mothers of hearing children. The mothers of deaf children miss opportunities for involving their children in verbal interactions. It is also reported by scholars (for example, Champie, 1981) that the households of deaf children do not engage the deaf children much in true symbolic communication process. Signing to children is found to improve early language skills in some studies. Lyon (1985) finds the features associated with increased child initiative and relatively less maternal control would help increase language development in the early deaf children. As regards specific language units, Shafer and Lynch (1981) report that the prelingual deaf children used single word utterances for an extended length of time. They also find that the prelingual deaf children use more semantic functions to the oral items they utter than the normal children, showing an inability to acquire necessary forms. Some linkage in the form of chaining of words is sought to be established by the prelingually deaf children in their language acquisition process. Mogford, Gregory and Keay (1979) found that while there were no differences between the pairs of hearing children and mothers on the one hand, and the pairs of hearing impaired children and mothers on the other around the age of 18 months as regards the time spent in joint attention to picture books, length of dialogue, turns per dialogue and proportion of dialogue initiated by mother and child, there began at the age of two, just within a gap of six months, an increase of quality in all items for the pair of hearing child and mother, whereas the deaf children did not show any significant increase in quality. That is, by the age of two years, there was qualitative divergence between the normal hearing and deaf children in matters such as attention to communication, length of dialogue, turns per dialogue, etc., the features that govern normal communication via language. The deaf child began to show the disruption of basic requirements for conversation. Lederberg (1984) reports that the hearing subjects and deaf children and considerable trouble maintaining their interactions. The hearing subjects tended to spend half as much time interacting with deaf children as with hearing children, although the hearing subjects did adapt their communication to the needs of the deaf children in several ways – use of more visual communicative devices, touches and simpler diction and syntax. The deaf children were noticed to the more passive and less actively involved in the interactions than the normally hearing children. Also it was noticed that the mothers of deaf children were always more dominant in interaction with their children than mothers of hearing children (Wedell-Monning and Lumley, 1980). There is generally an emphasis on control instruction and commands in the languages used by mothers of deaf children. These controls are due to the efforts at normal adaptation instigated by the breakdown in the processes of reciprocal communication (Brinich, 1980).

In a nutshell, whether mother’s interactions with her deaf child has any decided influence on the quality of language produced and comprehended or the reduction in the impact of deafness vis-a-vis language use, it is clear that mothers/ caregivers do attempt at accommodation and that their communicative interaction gets influenced by the deaf child’s willingness and ability to participate in such interactions. As regards the impact of mother’s utterances on the deaf child, there is not much dramatic change brought about by mother’s interactions.