Uniformity is not always possible, but I agree trhat distinctions without significant differences are highly undesirable.
As I have had occasion to say before, someone who knows a little the syntax of a language should be able to make plausible, almost invariably confirmed conjectures about the syntax of other, related constructs. As anyone who has had children knows, 'almost' is an important qualifier here. When a two-year-old uses 'goed' as the perfect of the verb 'to go', the response needs to be two-fold: 1) that the conjecture was reasonable and appropriate and 2) that this time, exceptionally, the appropriate form is not 'goed' but 'went'. Dumbing down has of course led to the recognition of too few regular constructs. Too few teachers and in consequence almost none of their students now realize that to drive, to shrive, to swive, etc., have the regular principal parts drive-drove-driven,shrive-shrove-shriven, swive-swove-swiven, etc. John Gilmore, Ashland, MA 01721 - USA
