Starting-point for the two pilot studies reported here was the question whether systematicity in interlanguage is the same for different types of L2 behavior. We hypothesized that:
1. If L2 learners know the correct L2 rule they make more errors (deviations from that particular L2 rule) in an oral production task in which they are pressed for time than in a task where they can speak as slowly as they want, provided that they have not yet reached the level of complete automatic mastery of the rule. In other words, under these conditions L2 learners show variability.
2. If L2 learners do not know the correct L2 rule, they make the same number of errors in the + time limit as in the - time limit task. In this case their interlanguage is invariable.
These hypotheses are consistent with both skill theory (Levelt 1976) and with an early version of the Monitor Theory, which predicts that, with increasing processing time, L2 output will improve due to the monitor. In both studies we analysed word order errors in Dutch (L2) made by adult speakers of English (LI) who live in Holland and are learning Dutch. The first pilot study consisted of two longitudinal case studies (fig. I). In the second pilot study 18 Ss where scored on 3 tests:
1. A knowledge test for the L2 word order rules under investigation. Ss had to correct sentences containing word order errors. This test yielded a criterion for classifying Ss as having + or - knowledge.
2. An oral production test (retelling in L2 of passages of LI speech) with restricted time. Response length was the same as the length of the stimulus passage, i.e. about 24 seconds.
3. The same test as under 2, the difference being that Ss had ample time for their responses.
Each group (+ and - knowledge according to test 1) made as many errors in test 2 as in test 3 (+ and - time limit): table 3. Thus no evidence was found for hypothesis I , which, however , does not mean that it has been falsified. Two possible explanations are offered, both in accordance with a later version of the Monitor Theory (Krashen 1979). First, Ss may have concentrated on communication rather than on grammatical correctness, in both oral tests. Second, word order rules may be so hard for most learners that their performance does not improve even when they have plenty of time for planning their utterances.
A Time-Limit Running Wild? Article 39(2) CISG and Domestic Limitation Periods