Search

This is a report on an on-going research project sponsored jointly by the Speech and Hearing Clinic of the University of Oregon Medical School and Portland State University. The book contains a brief historical review of the study of speech in normal children in recent years. Then follows a structural description of the language used by each of 25 female and 25 male children all within 10 days of being exactly 18 months old at the time of study. The descriptions contain the vocabulary used by the child, the number of morpheme classes, i.e., words in this case classified in terms of the way the child used them, not in terms of adult English grammar, along with the rules by which the child combined words from each class to form sentences. A summary of the language skills of the average "normal" child in the Portland area is presented, as the sample drawn exactly mirrors the Portland population, as reported by the 1960 census (the sample was drawn in 1964) in terms of economic background, age of parents at age of birth, educational background, number of siblings, etc. The final chapter is devoted to the scientific implications of this study for linguistic theory as well as general educational theory. (Author/CLK)

A variety of types of evidence are examined to help determine the true nature of "deep structure" and what, if any, implications this has for linguistic theory as well as culture theory generally. The evidence accumulated over the past century on the nature of phonetic and phonemic systems is briefly discussed, and the following areas of analysis are considered: (1) how phonemes are differentiated from each other by distinctive features and as fuzzy sets; (2) how morphemes come in two fundamentally different types, some of which form crisp sets and some of which are quite fuzzy; (3) how morphemes are combined into words which are not necessarily organized into classes in the same way that morphemes are; and (4) the fact that on the so-called syntactic level of analysis there are a variety of different types of units. Since it is proposed that the only thing that is transmitted from the speaker to listener is the vocal symbol, (i.e., the form), it is concluded that all linguistic analysis must proceed from the forms. It is suggested that at the point that phonemes are identified, morphemes are identified in terms of the sequence of phoneme combined with the statistical probabilities of occurrence. Then morphemes are identified on the basis of the phonemes and a Markov process (statistical probabilities) until a word is identified. Then the phrases are identified on the basis of the words and a statistical process and so on up the line. (SW)