This article hypothesizes that language repetition of young children (in the sense used by Kobashigawa and Snow) does not help language acquisition. The evidence comes from the results of a prior study in which no indication was found that mothers who repeat themselves a great deal have children who acquire language more quickly. However, non-correlations may occur because there is no relation between the variables. In the present study the relationship between repetition and language growth is examined inferentially, by examining the short-term consequences of repetition on comprehension. In particular, a premise of the repetition hypothesis is examined: according to the hypothesis, the child must process repetition sequences as repetitions, rather than as sequences of unrelated utterances. On the basis of two studies, one naturalistic and the other experimentally controlled, the child gives no evidence of responding to repetitions as repetitions. It is concluded that it seems unlikely that repetition per se could have a beneficial effect on language acquisition. (NCR)