Reader ST writes to draw our attention to Bryan Garner's 1/2/2012 note on "regardless whether":
Language-Change Index — “regardless whether”* for “regardless of whether”: Stage 2.
*Invariably inferior forms.
The "Stage 2″ designation is a reference to Garner's five-stage "Language-Change Index":
Stage 1 (“rejected”): A new form emerges as an innovation (or a dialectal form persists) among a small minority of the language community, perhaps displacing a traditional usage (e.g.: “your” misused for “you’re”).
Stage 2 (“widely shunned”): The form spreads to a significant fraction of the language community but remains unacceptable in standard usage (e.g.: *”pour over books” for “pore over books”).
Stage 3 (“widespread but . . .”): The form becomes commonplace even among many well-educated people but is still avoided in careful usage (e.g.: “clinch” misused for “clench”).
Stage 4 (“ubiquitous but . . .”): The form becomes virtually universal but is opposed on cogent grounds by a few linguistic stalwarts (die-hard snoots) (e.g.: “often” pronounced “OF-tuhn”").
Stage 5 (“fully accepted”): The form is universally accepted (not counting pseudo-snoot eccentrics) (e.g.: “decimate” for inflicting large-scale destruction).
But attitudes towards usage do not always follow this script, as we can learn from Diana's comment on Garner's post:
Bryan, I’ve long looked to your writing guidance as The Word, so I must respond to your 01/02/12 blog entry on “regardless of whether.” Ouch, that just sounds so awful! NYU Law was pretty open-minded, but it was drilled into us that proper usage excludes any “of” in the phrases “regardless whether,” “the question is whether” or “the question whether.” In your writing seminar in the 90′s, the most valuable exercise was editing text to eliminate as many “ofs” as possible. True gold! Now you say it’s a Stage 2 mistake and advocate using that unnecessary, ugly-sounding ‘uv’ here — where it makes the speaker/writer sound illiterate and pompous?
And as ST points out, the Google Books ngram corpus seems to indicate that "regardless of whether" is the innovation:
In the 19th century, neither variant was very common, but "regardless whether" was much commoner:
ST notes that in the older pattern of usage, regardless is usually predicated of a person (“You were indifferent, and regardless whether you gained their good will or not”). In such cases, the whether clause is the complement of the adjective regardless. In the more modern pattern, the whole "regardless of whether S" structure is a sentence-level adverbial, and the whether clause is a headless relative doing duty as a noun phrase, e.g. in this quotation from David Brooks:
[C}ountries with high social trust have happier people, better health, more efficient government, more economic growth, and less fear of crime (regardless of whether actual crime rates are increasing or decreasing).
(Note that he could have written something like "…regardless of the fact that crime rates might be increasing".)
So perhaps, in the sentence-adjunct usage, "regardless whether" is indeed an innovation. This is a case where simple word-string counting doesn't give us the answer. We need a a two-by-two classification (syntactic function of the whole phrase, crossed with presence or absence of "of"), and one of the dimensions requires some syntactic analysis. Pending further investigation, I'll leave the question open.
My own intuitions, for what little it's worth, are on Bryan Garner's side of the usage decision. I prefer "regardless of whether" — and I just checked previous LL postings to verify that it's what I generally (always?) write. But I wouldn't judge the of-less version as a mistake on the level of "pour over" for "pore over", or indeed as a mistake at all. And I'm not sure just who has innovated what here.
In any case, we still need to understand Diana's comment, which illustrates the fact that prescriptive usage norms are not always –and maybe not often — a simple matter of the gradual acceptance of innovation.