Introduction: ME /a, o/ in modern RP

It is usually assumed that standardisation typically involves at least two major operations: elimination of variation, and codification (in dictionaries, grammars, orthoepic treatises, and other “authorities”) of the trimmed-down and “authorised” version. In the case of English this is all true enough, globally and within limits. But English is unusual in the amount of time it took, and the lateness of the prescriptive or codifying grammatical (as opposed to phonological) tradition. And, somewhat paradoxically, the even greater lateness and variational latitude of the actual codification of parts of the phonology, even though pronunciation was from the earliest times taken as one of the hallmarks of the standard variety.

The perception of a ‘standard’ or ‘best’ kind of English (as an ideal, if not an empirically localisable object) dates back at least to the sixteenth century. Leaving aside the now overfamiliar classic remarks of writers like Puttenham and Hart, here are two characterisations, one shortly before the period I'm concerned with here, and one from quite late. In the seventeenth century John Wallis (1653: 73) says he is describing ‘puram et genuinam pronunciationem linguae Anglicanae’; specifically not ‘singulas … variorum locorum dialectos, aut affectatas muliercularum ineptias, aliosve barbarismos’.