The thinking-for-speaking hypothesis, formulated by Dan Slobin, has been considered a weak version of or an alternative to linguistic relativity since the 1990 s. Both state the influence of language on thought. However, claims regarding thinking-for-speaking are based on verbal evidence, whereas linguistic relativity claims are based on nonverbal evidence. There seems to be no interface between the two. The author argues that thinking for speaking is not an alternative but complementary to linguistic relativity by opening a window into the mind through which the influence of language on cognition is seen with observable behavior, and that the new paradigm of thinking for speaking and gesturing enables a combination of verbal and co-verbal evidence for the effect of language on cognition, as illustrated here by crosslinguistic comparisons of motion event narratives based on Talmy’s linguistic typological framework.