The autonomy of syntax is at the center of the debates between formalist and functionalist linguistics,[1][2][3] and since the 1980s research has been conducted on the syntax–semantics interface within functionalist approaches, aimed at finding instances of semantically determined syntactic structures, to disprove the formalist argument of the autonomy of syntax.
[5] The assumption of the autonomy of syntax can be traced back to the neglect of the study of semantics by American structuralists like Leonard Bloomfield and Zellig Harris in the 1940s, which was based on a neo-positivist anti-psychologist stance, according to which since it is presumably impossible to study how the brain works, linguists should ignore all cognitive and psychological aspects of language and focus on the only objective data, that is how language appears in its exterior form.
[1] The assumption of the autonomy of syntax has been a highly controversial topic in the functionalist and formalist linguistic spheres.
[1] [2][3] A common example that is used by linguistic formalists to indicate the validity of autonomy in syntax is, "Colorless green ideas sleep furiously", which demonstrates that, in order for a sentence to be syntactically correct, it does not need to be coherent or meaningful in any way.
The main grammatical model that is in support of the Autonomy of Syntax is Generative Grammar, created by Noam Chomsky.