When I am in doubt about a point of style or grammar, first I always see what Pam Peters has to say. Her Cambridge Guide to English Usage miraculously foresees almost every question I want answered, and offers sensible suggestions based on Australian usage.
But I am also a big fan of Bryan Garner’s Modern American Usage, the third edition of which I received last week.
The most useful addition made in Modern American Usage‘s 3rd edition is its rating of evolving words and usage not as correct/incorrect but on a scale from new usage to universal acceptance:
Stage 1: A new form emerges as an innovation among a small minority of the language community, perhaps displacing a traditional usage.
Stage 2: The form spreads to a significant fraction of the language community but remains unacceptable in standard usage.
Stage 3: The form becomes commonplace even among many well-educated people but is still avoided in careful usage.
Stage 4: The form becomes virtually universal but is opposed on cogent grounds by a few linguistic stalwarts.
Stage 5: The form is universally accepted.
For example, Garner rates using ‘reticent’ as a synonym for ‘reluctant’ at stage 4. When authors mean ‘reluctant’ I will keep removing ‘reticent’ for as long as I am an editor, but I accept that this distinction is almost lost. ‘Alumni’ as singular he puts at stage 2 (an error that occasionally appears in U of M material). ‘Hopefully’ meaning ‘I hope’ Garner puts at stage 4; I use it this way in speech and casual writing but avoid it elsewhere. ‘Populous’ (adjective) for ‘populace’ (noun), which I’ve seen twice this month, Garner puts at stage 1.
I would put incorrect apostrophes at stage 2. It is too common a mistake by Gen Y writers to be a stage one variation. Putting footnote numbers inside punctuation is at least stage 3 – while all professional writers and editors still put the numbers outside the full stop, self-publishers like interest groups, consultants, and government departments frequently put them inside.
Of course there are legitimate debates about how much de facto arbiters of usage such as editors should resist language change. But I think the five stages approach is a good one, as it acknowledges that language conservatism (and most editors are language conservatives, because rules make their jobs easier) is more appropriate for some audiences than others.