The thesis of the present paper is that the child acquires English expressions for space and time by learning how to apply these expressions to the a priori knowledge he has about space and time. This a priori knowledge is separate from language itself. . . . The knowledge, it will be argued, is simply what the child knows about space given that he lives on this planet, has a particular perceptual apparatus, and moves around in a characteristic manner. . . and in this sense it is "innate." (p. 2)
In our view, then, the child learns the meanings of words one piece or semantic feature at a time. The features which the child picks out as defining are determined largely by his perceptual and cognitive processes. Viewing the child's acquisition of knowledge of the meanings of individual words in terms of semantic features enables one both to account for data from the relevant diary and experimental studies and to see how this step in the acquisition of languaging competencies is closely tied to early perceptual cognitive abilities. The next question is how the child begins to put words together to form meaningful sentences; this leads to the following section on the acquisition of syntax.
So far we have discussed some of the child's first steps in language acquisition, including the production of the sounds of the language and the association of certain sequences of these sounds with particular semantic features. The next step for the child is to begin putting words together into ordered sequences-to produce utterances that have syntax. In order to do this, the child must, in some sense, learn the rules for ordering the signs of the language. Of course the child is not consciously aware of these rules, but neither is the adult. The rules are a way of characterizing or describing the knowledge the child must acquire in order to produce grammatical sentences. They thereby enable us to discuss this knowledge and its acquisition. The actual form of the knowledge within the child's brain is unknown, and probably unknowable.
Children's first multi-word utterances usually consist of two words. Children at this stage use the high-information words as "contentives" (nouns, verbs, and adjectives) in their speech, but not the low-information words or "functors" (such as articles, prepositions, auxiliaries). They also do not use the inflections that mark tense, number, and so on, in adult English. Brown and Fraser (1963) have proposed the term "telegraphic" to describe children's two-word speech, as well as the slightly more advanced speech. They note that the child uses the same types of words that adults retain when each word "costs"-as when sending a telegram.