THE AUTHOR FEELS THAT TO APPROACH CHILD LANGUAGE TRANSFORMATIONALLY IS TO USE A TECHNIQUE SUITED TO PROVIDING ADDITIONAL INSIGHT INTO A WELL-KNOWN LANGUAGE FOR TREATING AN UNKNOWN, OR AT BEST LITTLE KNOWN LANGUAGE. SHE MAKES THE FOLLOWING CRITICISMS OF TRANSFORMATIONAL ANALYSIS OF CHILD LANGUAGE--(1) NOTHING CAN BE DIRECTLY INFERRED WITH REGARD TO THE PROCESS OF LANGUAGE ACQUISITION FROM TENTATIVE INSIGHTS INTO THE MECHANISM OF NATURAL LANGUAGE, (2) THE HUMAN BRAIN DOES NOT WORK LIKE A COMPUTER, AND (3) CONCENTRATING ON SYNTAX (THE CORNERSTONE OF TRANSFORMATIONAL RESEARCH) IN CHILD LANGUAGE ANALYSIS IS "TOTALLY MISLEADING." FURTHERMORE, THE CHILD'S FIRST VERBAL EXPRESSIONS ARE DESIDERATIVES, WITH INTERROGATIVES APPEARING VERY EARLY. ACCORDING TO GENERATIVE THEORY, THE PRODUCTION OF THESE FORMS REQUIRES MORE THAN ONE TRANSFORMATION. THE AUTHOR ALSO POINTS OUT THE CONSTRUCTIVE ROLE OF THE GENERATIVE APPROACH--(1) IT HAS CONTRIBUTED TO A MORE PRECISE FORMULATION OF THE MANIFOLD PROBLEMS OF FIRST LANGUAGE ACQUISITION, (2) BY BRINGING TO LIGHT THE FRUITLESSNESS OF USING MODELS BEFORE STRUCTURING DATA, IT HAS LEFT LINGUISTS FREE TO TRY OTHER ROUTES, AND (3) IT HAS GIVEN IMPETUS TO NEW RESEARCH ON CHILD LANGUAGE. THIS PAPER WAS PRESENTED AT THE UNIVERSITY OF KENTUCKY FOREIGN LANGUAGE CONFERENCE, LEXINGTON, IN APRIL 1968. (DO)