With few and marginal exceptions, such as may occur when mastering the overtly codified prescriptions of an archaic standard, human beings acquire their native language in complete ignorance of its history. They do so by means of a capacity for language which is part of their biological endowment, whether we think of it in rationalist terms as a domain-specific mental module, or in empiricist terms as a set of general predispositions also applied to other cognitive domains. Linguists, in contrast, do not know what the biologically endowed human capacity for language consists of, except by more or less corroborated scientific conjecture; but, as if in compensation, they have access to a rich record of historical change in a variety of languages, of which English is a notable example. Linguists can thus observe what innovations occur frequently or infrequently in the history of languages, and under what circumstances, and from these observations they can infer hypotheses about the cognitive abilities of the speakers participating in linguistic change, whether these be children in the task of language acquisition or adults in the everyday business of language use. The relevance of diachronic evidence to the general enquiry into the human faculty for language is, in this sense, tolerably clear.

Far less clear is the relevance of historical evidence to the description of a synchronic state of a particular language. How can knowledge of Middle or early Modern English, for instance, inform our synchronic understanding of a variety of present-day English, if such knowledge plays no part in the acquisition of that variety by its native speakers? One could conceivably answer that, when we have well-supported cross-linguistic generalisations about the typical course of change in some part of grammar, we can advance a hypothesis about how innovation may have unfolded in a language from a given point in its history, and the end-point in this hypothetical pathway of change may satisfyingly agree with an independently formulated synchronic description of the present-day state of affairs. Something of this sort is attempted, for a number of English phonological phenomena, in Bermúdez-Otero (2011: §2–§3, §7, §9). Such arguments have a certain force, which is however tempered by the danger of a conspiracy of error: a synchronic description and a theory of change can very well support each other and yet both be false.