Investigating grammatical difficulty in second language learning: Implications for second language acquisition research and language testing

Both second language acquisition (SLA) researchers and language testers collect data in order to make statements about what learners have learned. Many researchers and testers consider the ideal data for this purpose to be naturally occurring language use. This paper examines whether data elicited by instruments designed to provide separate measures of implicit and explicit second language knowledge afford a valid basis for determining what learners have learned. It reports on a study that tested predictions derived from Pienemann's Processability Theory regarding the learning difficulty of four grammatical structures. The results showed that the predictions were borne out in the data from the tests of implicit knowledge but not in the data from the tests of explicit knowledge. The study suggests that experimentally elicited data can be used to examine interlanguage development (i.e. how learners’ implicit knowledge develops) and to make statements about learners’ grammatical proficiency. It also indicates that what constitutes learning difficulty needs to be considered separately for implicit and explicit knowledge. The implications for SLA research and language testing are considered. 1