Strength for Today and Bright Hope for Tomorrow

Volume 1:1 March 2001
Editor: M. S. Thirumalai, Ph.D.



1. INTRODUCTION: Motor Disability

Motor disability in cerebral palsy is a primary symptom and it affects the intelligibility of the CP child’s speech. It affects the voice production for speech, the articulation of speech sounds and their sequences, and the production of appropriate sentence melody in sentences. Thus, ultimately the motor disability in the CP child comes to hinder his oral communication.

However, we must always bear in mind that CP itself has extreme variability in its occurrence in individuals. In a manner of speaking, each cerebral palsied child is unique with his own particular abilities for speech production and comprehension. The specific sounds that are missing in a CP child’s inventory may differ from one CP child to another, as well as from one language environment to another. In other words, the overall profiles of speech disturbances in CP children as a whole may be worked out for a population, but its use in assessing the specific speech abilities of the CP individual, and training the same individual in speech production and comprehension has several limitations.


Our study of the speech of the cerebral palsied children with Kannada (a language belonging to the Dravidian family of languages, and largely spoken in the state of Karnataka in South India) as their language of environment reveals the following characteristics:

  1. Cerebral palsied children in the moderate to severely impaired category, have only a limited number of sounds when compared with normal children.
  2. Their phonological system also has only a limited number of significant sound units (phonemes) which are used to distinguish meanings. Even this limited number of sounds may be used with severe inconsistencies by these children.
  3. This inconsistency is more pronounced in the consonant sounds. CP children substitute one consonant for another, or they insert a consonant sound that is not required in normal speech.
  4. In the case of the vowels, the CP child may tend to have additional vowels, more than what the normal language has. These additional vowels may be used only marginally. These may be used also in place of or in addition to their counterparts in the phonological system of the normal language.
  5. The usual corresponding pairs of sounds we notice in the oral language of the environment need not occur in the sound system of cerebral palsied children. Consider the following pairs of sounds in Kannada: p/m, T/N (retroflex sounds), c/n (palatal sounds) and k/N (velar sounds). A Kannada speaking CP child may or may not have all the pairs of corresponding sounds.
  6. Some of the entire series of sounds in the sound system of the language of the environment may not be acquired at all by CP children. For example, all the aspirated sounds in the voiceless and voiced stop consonant series (such as ph, bh, kh, gh) in Kannada were not acquired by the children whom we observed. In addition, fricatives such as [s], [h], and [s] (the palatal sibilant) also were not acquired.
  7. Production of aspirated sounds (such as ph, bh, kh, gh in Kannada) comparatively requires a greater effort of exhalation and finer position of the parts of the speech organs. The CP child is not physiologically skilled enough for this.
  8. What is more important and most interesting to an observer is the fact that the moderate to severely impaired CP child does not produce all the sounds normally produced at a single place of articulation in the normal language. Nor does he/she produce all the sounds normally produced following a single manner of articulation.
  9. For example, Kannada has five stops: [k], [c], [T] (retroflex), [t] (dental)and [p]. A moderate to severely impaired CP child reared in the Kannada environment does not acquire all these stops.
  10. Likewise Kannada has two velar stops [k, g] as well as a velar nasal [N] in its normal use. The moderate to severely impaired CP child acquires the unaspirated voiceless and voiced velar stops [k, g] but not the velar nasal [N].
  11. We find the following places of articulation in the speech of CP children. These are presented here in descending order of frequency: (a) Bilabial sounds are more freely acquired than other sounds. These are followed in a descending order of acquisition by dental sounds [t, d], velar sounds, and semivowel sounds. (b) The following places of articulation are not used at all: Retroflex (stops, nasals, sibilants such as s (retroflex) and s (palatal), and laterals) and pharyngeal sounds.
  12. The following manners of articulation are noticed in the CP speech in the Kannada environment. These are also presented here in a descending order of frequency:
    1. Stop consonants are more freely acquired than other sounds. However, among the stop consonants, only the distinction between voice and voicelessness is acquired, not the quality of aspiration.
    2. Nasals are also found.
    3. Semivowels are acquired.
    4. Laterals are acquired to some extent.
  13. What we have suggested as the descending order of occurrence of sounds in Kannada may perhaps be valid for other major Indian languages as well. However, CP children reared in other language environments (such as those of Tibeto-Burman, and Austric languages, spoken in the north-eastern regions of India, Arunachal Pradesh, Mizoram, Manipur, Nagaland, and the Ladakh region) may or may not share this descending order of acquisition of speech sounds.


Note that the sibilants, other fricatives and trills, which require relatively finer positioning and deliberate movements of a rapid and repetitive nature, are not acquired.

In general, the CP child does not show much difficulty in the production of vowels and semivowels. In both these cases, no interruption of the breath coming up from the lungs is attempted with the manipulation of the tongue. Rather, an open mouth is the setting in which these sounds are produced. However, these sounds do require raising and lowering the tongue from its usual lying position in the mouth. In other words, the CP child does not usually have any difficulty in raising or lowering the tongue, as the availability of vowels in abundance in the CP child’s speech reveals.

On the other hand, the CP child has difficulty with manipulating his tongue for finer positions such as those we find in the production of fricatives. We notice that the CP child is able to articulate a greater number of stop consonants than other sounds. These consonants involve a single and total constriction in their production.

In comparison, the retroflex, sibilant and fricative sounds require a finer manipulation of the tongue in the sense that these require maintaining a narrow aperture, and an arduous movement (from the CP child’s point of view).

Note that the nasals and laterals that are found in the speech of the CP child are also produced with single and direct constrictions and do not involve maintaining narrow apertures in the straight path of the air stream from the lungs. In any case, all these sounds (stops, nasals and laterals) are distorted because of neuromuscular disturbance. What is most significant is that in spite of the disturbance and distortion, the stops, nasals, and laterals do occur, whereas the CP child does not produce the sibilants and other fricatives at all.

Generally speaking, the generous occurrence of the stop consonants, the negligible occurrence of non-stop consonants, the absence of sounds for the production of which one requires a finer manipulation or handling of the tongue characterize the consonant profile of a cerebral palsied child.

Comparatively speaking, the stop consonants are produced more accurately. On the other hand, even when the non-stop consonants are produced, most of them sound distorted.


The vowel profile of the CP child is characterized by an excessive discrimination of the sounds, more than what is required in the normal language of the environment. In some CP children, there may be an abnormal neutralization of long and short vowels in some positions of the word. The longer vowels in final positions are very rare, and even when present these are comparatively shorter than the long vowels in the normal speech.

Note that even in imitation tasks, CP children have difficulty producing the non-stop sounds with ease. In any case, the moderate to severely impaired CP child uses only a lesser number of sounds compared with the normal child.

Age Factor in CP Speech

It is likely that, as the CP child grows in age, there may be additions to his inventory of speech sounds. We have noticed that whereas the stop consonants, or at least most of them, are available to the young CP child around 4 or 5 years of age, the non-stop consonants which require finer positioning and turning of the tongue begin to appear later. We have noticed that CP children around 7 years of age exhibit a tendency to acquire some retroflex sounds. They also manifest the use of some fricatives including sibilants.

Generally speaking, all such late occurrences of non-stop consonants present only distorted versions of the sounds produced. Nevertheless, this possibility of emergence of several non-stop consonant sounds at a later age gives us some hope that by an assiduous and patient cultivation we may be able to help the CP child to acquire some other speech sounds as well. Thus, age is a significant factor in CP children for the acquisition of speech sounds.

In our study we have seen that younger children do not have retroflex, sibilant, and fricative sounds, whereas the older children in the age group of 7 to 10 years begin to show an acquisition of these sounds.

Moreover, as age increased, clarity also increased to some extent. There was also a change in the number and kinds of consonants produced, but the number and kinds of vowels did not change significantly. We also noticed more approximations to the normal spoken forms along with the increase in ability to utter syllables clearly, a better distribution of sounds, and a better retention of most of the shape of the forms uttered.

There was also improvement in the quality of paralinguistic features used. Quantity of speech output also increased with an increase in age. Moreover, production of a greater number of multiple word utterances was noticed, but this increase did not change the fundamental nature of CP speech as characterized by single word utterances. The increase was, however, not very impressive, as it did not drastically improve the quality of speech and language in the phonological, morphological (word), and syntactic inventories and their use. Even in terms of the length of utterances, there is no correlation found between the increase in age and the length of utterance. In almost all the cases, CP children did not go beyond the single word utterances.

Whenever a CP child gave an utterance longer than the single word, it was either an attempt during the imitation task or the multiple word utterance happened to be a rare occurrence. The self-initiated conversation may attempt at going beyond single word utterances, but this also happened very rarely. In essence, since multi-word utterances were few and generally not found in the self-initiated conversations of CP children, they were more comfortable usually with single word utterances and did not generally go beyond that.

Remember that such generalizations are not a hard and fast rule with all CP children. Remember that CP may vary extremely from individual to individual.


The neuromuscular distinction established between spastic, athetoid, and ataxia is not easy to identify or establish at the speech level. It is not surprising when we consider the fact that in a large number of CP patients, identified predominantly as spastics, athetoid features are found. Likewise, in a large number of athetoid subjects the features of spasticity are noticed.

Years ago Mecham et al (1960), and several others before them, pointed out that ‘speech characteristics are highly variable, both within the groups and within the individual CP from one time to another.’ Investigators generally agree that although we may be able to distinguish between spastic, ataxic, and athetoid speech to some extent, ‘the problems in any one type of cerebral palsy are so diverse that it is practically impossible to draw a single or composite picture of cerebral palsied children.’ While this may be largely true even today, the conclusion of Mecham et al (1960) that ‘differences in the speech of the cerebral palsied children and non-cerebral palsied children are more apparent in degree than in quality’ is not necessarily accurate.

There are qualitative differences between normal speech and the speech of the severely/profoundly affected CP child. Spastic speech is recognized by its slow rate and labored production. A spastic child faces grave articulatory problems because of the child’s inability to form fine synchronous movements of the tongue, lips, palate, and jaw. The child also exhibits a lack of vocal inflection, has a guttural or breathy voice, uncontrolled volume, and abrupt change in pitch. These characteristics are shared also by athetoid and ataxia subjects. For this reason, the finer distinctions we wish to make in the speech forms of the subtypes of cerebral palsy may not have any direct consequence for diagnosis and therapy of the neuromuscular subtypes of CP.

The parents/caregivers and the therapists must, first of all, work out a profile of speech sounds already in place in the CP child, and then find out whether any pattern in the acquisition of speech sounds can be identified.

HOME PAGE | Technical Terms in Hindi and Tamil | Information Technology and Kannada | Teaching a Literary Language | Citation Index and Its Importance to Indian Linguistics | CONTACT EDITOR

Shyamala Chengappa, Ph.D.
All India Institute of Speech and Hearing
Mysore 570006, India