Universal Grammar (UG)
- is the theory of the genetic component of the language faculty
- is the assumption that there is a rich set of innate features and categories with common grammar, not only in phonology
- is not a set of grammatical rules that all languages follow. Instead, it is what is in our brains that causes us to acquire language
Many of the difficulties with discussion of universal grammar suffer from the fact that it is a loaded term, which has been used to refer to a wide range of different things, and it’s not always obvious in any given discussion which one is being used. Leaving aside the various pre-20th Century uses of the term, which generally aren’t used today, some of these are as follows:
- Properties that languages necessarily have in common with one another, for whatever reason. The study of this bears a lot on other definitions, but the notion doesn’t in itself make any falsifiable claims - though it is a relatively open question whether such properties exist.
- Whatever it is that gives humans the ability to learn and use language, setting them apart from e.g. other animals. It is almost definitely the case that this is something genetic, rather than e.g. social, since there are no plausible cases of animals approaching human linguistic capacity (this includes e.g. ASL-learning apes, but a full discussion of this is beyond the scope of this post). But so long as you agree that humans are better at language than other things, which almost everyone does, this notion doesn’t so much estabilish a claim as describe a field of study - since this could just be a result of general cognitive superiority.
- The notion of a specific physical sub-organ in the brain responsible for language, which is used to explain (among other things), various features of the above notions. As others have noted, advances in the study of the brain make it pretty likely that this sub-organ doesn’t exist, since processing seems to occur in a very distributed way. Few people support this claim today.
- The notion of a cognitive module that is language-specific. This may sound a lot like the above claim, but though we are beginning to know more about the brain, we still know very little about how higher-level cognition works. It is possible (perhaps even likely) that this may be modular in a certain sense, but there’s no guarantee that these modules would correspond to distinct physical regions of the brain - in the same way that the different components of a computer program don’t correspond to distinct physical regions of computer hardware. We know so little about cognition that we have no idea whether this notion is true or not, but it’s a lot more plausible than the previous one.
Arguably the most well-formed question relating to the existence of universal grammar is: is there a language-specific component of human cognition? And this has yet to be satisfactorily answered, but it can’t be trivially falsified