I'm not sure how important "langauge" is for cognition. I think it is clear that the cognitive facilities that, in humans, we attribute to language, are a signi

I'm not sure how important "langauge" is for cognition. I think it is clear that the cognitive facilities that, in humans, we attribute to language, are a significant part of human cognition. However, when we test for language in other animals, we do not test for these underlying cognitive abilities. Rather, rather, we test for the animals ability to express these facilities through some communication system with other animals (either in the same species, or with humans). It is conceivable to have an animal that has all of the cognitive abilities we associate with language except the ability to serialize/deserialize [0] it. While this ability is significant at the species level, it does not seem like it would be particularly significant at an individual level. Further if you take the view that language was evolved primarily as a tool for thinking (rather than a tool for communicating), this possibilities seems even more likely. Under this view, we would first develop the cognitive aspects, then develop the communicative aspects on top of them (in the same way that we evolved our "speech organs" on our digestive (tongue) and respiratory (lungs/airway) ones. It also seems plausible that it was this adaptation that allowed humans to pass the singularity point necessary to reach modern society (which grew up very quickly on an evolutionary timeline). If this were the case, then I would expect to see that the rest of the cognitive abilities associated with speech are actually present and similar in related species. [0] Our current model of human language is that of a tree. We do not see processes that act on a linear structure until we get into phonology.