Timing of Language Acquisition

The infant undergoes many physiological and cognitive changes during its development. Since the timing of these advances are variable and independent, it affects which behaviours can be observed during the period of language acquisition. During the first few months after birth, the infant is physiologically constrained with regards to the types of vocalizations it can produce. However, experiments indicate that prelinguistic infants are aware of elements contained within adult speech.

During the first five to seventeen weeks after birth, infants can discriminate phonemes (Werker and Desjardins 1995). At two months of age, the vocal production of Homo sapiens sapiens infants is constrained by both the physiological state and physiology of the vocal tract thus, they are only able to “emit reactional sounds which signal their current state of well being” (de Boysson-Bardies 1999). However, they appear to be attentive to the speech of adults as evidenced by the following lip movements, and are able to distinguish voices, with a preference for that of their mother (ibid.). Research has indicated that they are sensitive to the prosody of the language spoken in their immediate environment, and are able to distinguish foreign languages easily (Nazzi et. al. 2000; Aslin et. al. 1998).

Between two to five months of age, infants tend to vocalize only when lying on their backs (de Boysson-Bardies 1999). These vocalizations are almost uniquely made up of sounds which issue from the larynx and soft palate, and there does not appear to be defined phonation (ibid.). However, between four and five months, infants become capable of voluntarily modulating their vocalizations (ibid.). They seem to develop a series of vocal exercises, which manipulate both the prosodic cues (pitch of voice), sound level, and consonantal features (friction noises, nasal murmuring, rolled labials, uvular trills) (ibid.). On acquiring control of phonation, the infant can modulate the duration, pitch, and intensity of vocal productions, and identifiable ‘vowels’ and ‘consonants’ appear (ibid.). However, these are not characteristic of the syllables of spoken language. It has been suggested that through ‘playing’ with intonation, infants familiarize themselves with a number of routines and become capable of producing varied sound effects (ibid.). This controlled phonation allows infants to employ their vocalizations to communicate their emotions and demands (ibid.). Additionally, during this time of development, infants listen significantly longer to repetitions of their own names than of the names of other people, and appear to perceive and represent some objects and events (Tincoff and Jusczyk 1999).

Between four and seven months of age, infants begin to incorporate gestures which bring the forward part of the articulatory apparatus into play (Armstrong 1999. 1995, 1994; Tomasello and Brooks 1999). At this time, infants begin to lose their “encyclopedic hearing (de Boysson-Bardies 1999). They exhibit a representation of vowel space, which is adapted to the language of their environment, just as adults do (Blake 2000; Aslin et. al. 1998). Additionally, they begin to ignore elements which are generally absent from the phonetic structures of their linguistic environment” (ibid.). Six month olds attach the words ‘mommy’ and ‘daddy’ to their own parents, and not to women and men in general (Tincoff and Jusczyk 1999).

Toward the end of the sixth month, infants are capable of coordinating phonatory and supraglottal adjustments (Blake 2000). This voluntary interruption of vocalization is an essential element of vocal control” (de Boysson-Bardies 1999). Additionally, they can model the pitch of their vocalizations on those of the interacting adult (i.e. their voice is higher with their mother than with their father) and can imitate simple patterns of intonation on the basis of adult examples (ibid.; Tomasello and Brooks 1999).

Between six and ten months infants begin to ‘babble’. Canonical babbling, which is characterized by the production of simple syllables in a consonant-vowel sequence (i.e. [ba]) is quickly followed by rhythmic sequences formed by repeated consonant-vowel combinations (i.e. [baobab]) (Davis et. al. 2000; Locke and Pearson 1990). Consonant sounds tend to be occlusive and nasals, combined with low front and central vowels ([a], [æ]) (ibid.). However, there is great variation amongst infants due to individual, linguistic and cultural differences (de Boysson-Bardies 1999; Aslin et. al. 1998; Locke and Pearson 1990). Congenitally deaf infants vocalize like hearing infants but they do not babble (de Boysson-Bardies 1999; Petitto and Marentette 1991). By seven months, the vocalizations of deaf babies tend to diminish (ibid.). However, hearing impaired infants who grow up in an environment with a sign language, ‘babble’ manually by eight months of age (ibid.).

Around ten months of age, infants begin to lose their capacity to distinguish universal consonantal contrasts (de Boysson-Bardies 1999). This reorganization seems to be tied to the beginnings of word recognition (Tincoff and Jusczyk 1999). By this age, it is presumed that infants have selected a repertoire of consonants that reflect the statistical tendencies of their native language (Davis et. al. 2000; Aslin et. al. 1998).

By eleven months to thirteen months, infants begin to show definitive signs of word comprehension (Tincoff and Jusczyk 1999). While they have better control of their articulation, the majority of their vocal productions remain mono or disyllabic, with occlusives and nasals predominating (Davis et. al. 2000; Locke and Pearson 1990). These initial productions tend to resemble the most frequent context in which the words were modeled by their mothers (Dromi 1999:115). However, the effects of maternal input characteristics decline with time and are much less pronounced once the child has established a sizable vocabulary (ibid.).

In order to learn a word, the child must first learn four things: 1) pronunciation ; 2) the syntactic properties; 3) meaning; and 4) usage (Kuczaj 1999:136). Prior to the emergence of first words, children utilize gestures, facial expressions, intonation patterns, and nonsystematic vocalization for communicating their basic pragmatic intentions (Blake 2000; Dromi 1999). The beginning of the one word stage is marked by the emergence of systematic, repeated productions of phonetically consistent forms (Dromi 1999:99). During the one word stage, children might occasionally produce multiword combinations that do not yet reflect the productive syntactic abilities (Dromi 1999:100).

Researchers propose that during the first phase of the one word stage, functional words constitute the dominant class of words (Blake 2000; Dromi 1999; Tomasello and Brooks 1999). During this early stage, there is an indeterminate usage of the same words to refer either to an object or to an activity that is typically associated with that object, or to both (Dromi 1999:107). It is not important whether or not young children find it easier to learn nouns or verbs. Children may use a word as a noun and as a verb, so even the distinction between words that denote objects and those that denote actions or states is not always an easy on to identify (Kuczaj 1999:143). Word meaning acquisition is thus a comprehension based process. Children will not learn words that they have not heard. However, before children begin to acquire words, they have formed concepts of the world (Kuczaj 1999:153). The early words of young children appear to be based upon aspects of the world that they can directly experience, regardless of whether the words are nouns, verbs or adjectives (ibid.).

In general, by the age of two, a child uses between 50 and 600 words, and adds an average of 10 words per day to its vocabulary, “resulting in a vocabulary of approximately 14,000 words by the age of six” (Kuczaj 1999:133). However, while the child can now communicate simple concepts, ‘mastery’ of language has not yet been attained. By the age of four, language experience begins to affect phoneme perception and children use information about syntactic cues to infer the meaning of words (Kuczaj 1999:153; Werker and Desjardins 1995:78). This process has been called ‘syntactic bootstrapping’, and emphasizes the point that word meaning acquisition does not take place in a vacuum but occurs in a larger context (ibid.).

Children learn the rules governing the regular use of most bound morphemes before the age of five, although learning the exceptions may take many years (de Boysson-Bardies 1999). The acquisition of syntax is largely complete by the age of five (Blake 2000). However, aspects of sound, communicative, morphological and syntactic systems continue to develop (Kuczaj 1999:134). An example of this is the incorporation of paralanguage and propositional content to discern a speaker’s intent.

The paralanguage, or affect, of a speaker is conveyed through subtle vocal cues such as altering the speaking rate, pitch level or contour, and voice quality (Morton and Trehub 2001). Paralinguistic features may unintentionally reveal the speaker’s emotional state or, used intentionally, convey the attitude of the speaker. While paralinguistic cues toward attitude vary across cultures, those linked to basic emotions appear to be ‘universal’ (ibid.). In contrast, speakers can use propositional content to depict their feelings directly or to describe a situation that has positive or negative emotional implications.

While adults consider all available cues, but rely primarily on paralanguage when there is a conflict with propositional cues, children of all ages exhibit response latencies to utterances with conflicting cues, which indicate that they processed both sources of emotional information (Morton and Trehub 2001). When the cues conveyed by propositional content and affective paralanguage conflicted, four to ten year old children rely primarily on content and not on paralinguistic cues (ibid.). However, children were able to accurately label the affective paralanguage when the propositional cues to emotion were obscured by a foreign language, even though they failed the similar task in their native language (ibid.).

A possible explanation for this conflict may be that the communication of young children is more pragmatic than that of adults. Young children are not very successful when producing vocal deception, since adults are able to discern paralinguistic cues. It is not until the late teenage years, that humans begin to attain some semblance of affect management (Blake 2000; Loritz 1999). It is possible that children are unable, developmentally, to master this type of verbal ‘deception’, and they are unaware that adults have this ability. Alternately, it may simply be that adults, in general, do not engage in overt paralinguistic deception with children. If they do, the child may be able to discern paralinguistic cues in association with visual information.