Some new approaches to the assessment and treatment of language handicaps suggest the need to examine relationships between verbal creativity and language performance. Data were collected from 40 normally developing white fourth, fifth, and sixth grade children, drawn from both urban and rural communities and from a middle socioeconomic level. Subjects were given the Torrance Tests of Creative Thinking--Verbal and six commonly used measures of language performance including the Peabody Picture Vocabulary Test--Revised, and an expressive language subtest of the Clinical Evaluation of Language Function. An oral language sample was judged for mean length of utterance, semantic complexity, and verbal fluency. Statistical analyses yielded only one moderate and positive correlation between verbal creativity (the Verbal Flexibility subtest of Torrance) and one language performance measure (the Peabody). Results of a stepwise regression indicated that verbal creativity measures--the Verbal Fluency, Verbal Originality, and Verbal Flexibility Torrance subtests--could not be used to predict selected language performance measures. Results suggest a distinction between the skills needed for encoding and decoding of language and the creative or divergent use of language. Programs for linguistically handicapped persons need to emphasize language use in novel, varied, and more generalized situations. (DB)