This study examined the effects of explicit instruction on the development of pragmatic competence in L2 English. The study is based on Taguchi's (2012) study conducted in an English-medium university in Japan, which revealed patterns of change in Japanese EFL students' production of requests in high- and low-imposition situations. Students showed strong development with low-imposition requests (e.g., asking a friend for a pen) over one year but almost no gain with high-imposition requests (e.g., asking a professor for an extension of an assignment). Based on these findings, in the same institution, we implemented explicit instruction on the target pragmalinguistic forms that never emerged in the students' request data (i.e., syntactic mitigation, hedging, and amplifiers). Results from a new cohort of 23 students revealed strong instructional effects. Students' production rates of the target features jumped by 40-90% from pre- to immediate post-test, and a large portion of the gain was maintained at delayed posttest given four months later.