Language may indeed influence thought

We discuss four interconnected issues that we believe have hindered investigations into how language may affect thinking. These have had a tendency to reappear in the debate concerning linguistic relativity over the past decades, despite numerous empirical findings. The first is the claim that it is impossible to disentangle language from thought, making the question concerning "influence" pointless. The second is the argument that it is impossible to disentangle language from culture in general, and from social interaction in particular, so it is impossible to attribute any differences in the thought patterns of the members of different cultures to language per se. The third issue is the objection that methodological and empirical problems defeat all but the most trivial version of the thesis of linguistic influence: that language gives new factual information. The fourth is the assumption that since language can potentially influence thought from "not at all" to "completely," the possible forms of linguistic influence can be placed on a cline, and competing theories can be seen as debating the actual position on this cline. We analyze these claims and show that the first three do not constitute in-principle objections against the validity of the project of investigating linguistic influence on thought, and that the last one is not the best way to frame the empirical challenges at hand. While we do not argue for any specific theory or mechanism for linguistic influence on thought, our discussion and the reviewed literature show that such influence is clearly possible, and hence in need of further investigations.