I think that an analogy with "amicable" figures in. "Amicable" and "inimicable" aren't opposites, but there's some sort of blurry relationship (somewhat an echo of Spanish "amigo" and "enemigo".)

I don't think that this is a real etymology, but a sort of partial explanation of why "inimicable" sounds right to (some) people.

Oddly, I'm more familiar with inimicable. Maybe it has to do with the books I've read.

"I'm more familiar with inimicable."

I suspect your brain is connecting it with inimiTable.

Via Google Books, once again, here's a possibly earlier citation (snippet only), from 1794: "but the people could not accuse me of being inimicable to their liberty." It's used again in the book but the snippet doesn't show the word. The book is a translation from the French memoirs of General Dumouriez (turned into Dumourier in the English translation). The GB listing, however, mentions chapters from the 1822 edition. The illustrated title page does look like it says 1794; the snippet is from page "xliv". The Yale library catalog shows this edition was printed by Samuel Harrison Smith in Philadelphia in 1794. The 1822 chapters are perhaps bound into the book scanned by Google? So the question is whether "inimicable" appears in the 1794 version or the 1822 chapters.

Elsewhere, within this discussion: " 'Amicable' is derived from 'amicabilis', a Roman legal term, according to OneLook. 'Inimicabilis', or 'inimicable' in English, would be the customary negative form of that word..."

As pointed out by Gary, it sounds likely to me that people are patterning it on "inimitable". That was certainly my first thought upon seeing the post.

Hmmm. Zwicky hardly addresses the relevant shades of meaning, except to suggest that they are there. Surveying my own inner acceptatorium, I cannot discern any stable difference in sense between inimical and inimicable. I think I would probably "correct" the latter if I found it in a text I was editing, because I would think it less likely that a client so particular as to choose such a variant knowingly would have called on me for editing in the first place!

What is the supposed difference? Is it just a matter of inimicable being an attempted direct and transparent opposite of amicable (pace John Emerson; and cf inept as an apposite opposite of apt)? That would differentiate it from inimical as a direct derivative of a Latin inimicalis (itself from inimicus, itself from in- + amicus). So inimicable would be milder: simply falling short of amicability; and inimical would be stronger: being like an enemy. Is that all there is to it? I note that OED merely reports that the two are synonyms, and finds no shade of difference. Same with the ever-subtle Webster's Dictionary of English Usage.

Intuitively, now, we might distinguish two sorts of opposites: privative and polar. Inimicable would then be a privative opposite of amicable, since it strips away amicability. But inimical would be a polar opposite of amicable (and of amical, rare in English), even though its deep root inimicus started out as a privative opposite of amicus. (Smith gives unfriendly as a first gloss on inimicus, and hostile only as a second.)

It's interesting that French has no amicable (according to Petit Robert), while English hardly has amical (OED: "now rare [...] friendly"). One of the pair seems to suffice. On balance, it seems that one of the pair inimical and inimicable will also suffice; and in such a case, with all else equal, weight of long-established usage might reasonably determine things. I would use only inimical.

Also spracht Noetica.

Another way of looking there is that we have a perfectly nice word sitting there unused, just waiting to have a specifical meaning attached to it. Who knows, even as we speak some English-language Heidegger is coining the term inimicable and giving it a specialized meaning.

"Specifical" being another potential word sitting there all alone sobbing.

Speaking of Language Log, this post is very appropriate.

While the randomness of what gets criticized is certainly the case, it is not true that these things never die out. I collect 19th century grammars and usage manuals (so as to firmly establish my geek cred...). This is the Richard Grant White and Alfred Ayres and Goold Brown crowd. I have never been able to detect any patterns in which of their complaints get carved in stone and which quietly die out. Facts and logic clearly have nothing to do with it. But we no longer read sputtering condemnations of constructions like "The house is being built." or "John was given a gift."

Sounds reasonal to me.

I want to buy some of that shit NOW.

Interesting, Richard.

But we no longer read sputtering condemnations of constructions like "The house is being built." or "John was given a gift."

Coincidentally, though, your two examples are uneuphonious, and I would probably want to alter them for that reason alone. I can only guess that the semantic redundancy in the second caused some pedantic anguish. Care to say briefly what the original objections were, in fact? Can't be mere passivity, can it? And er, know where I can get cheap Zyrtec?

"Would have been being done". Is there a longer non-compound non-complex VP?

"Would have been being done." Is there a longer non-compound non-complex VP?

"Would have been going to be about to be done". Is this non-complex, and non-compound? Explain exactly what you mean by these terms, with contrasting examples, please.

No the objection was not to passivity. That seems to have originated with Strunk & White.

The objection to "The house is being built" is to the syntactic construction of the passive progressive. Early Modern English used auxiliary verbs less than does present day English, and the various auxiliery constructions didn't come in all at once. This particular one was a latecomer, first appearing in the late 18th century. It is a little peculiar in that it strings two forms of "to be" together, but the logic of it is consistent with other constructions. The earlier construction would be "The house is building" or "The house is abuilding". The latter, of course, sounds very backwoodsy today. I could follow this with a discussion of agency and ergativity and the like, but I won't. Richard Grant White devoted an entire chapter to sputtering about this construction.

As for "John was given a gift", the objection is that it is not regularly formed from the presumed active form. Normally you can reconstruct the active form like this:

[1] The ball was hit [by someone].

derives from

[2] [Someone] hit the ball.

but

[3] John was given a gift [by someone].

does not derive from

[4] [Someone] gave a gift to John.

That is, in [1] the subject is the direct object in [2]. In [3] the subject is not the direct object in [4].

In classic prescriptivist logic, they had a model of how passive constructions worked, and if a class of passives used in the real world did not match the model, obviously the real world was wrong and must adapt accordingly.

Both of these complaints were staples of late-19th century usage manuals, but both simply died out for reasons not entirely clear.

Thanks, Richard. Most intriguing. You wrote earlier: "Facts and logic clearly have nothing to do with it." But there is logic of a kind – captious and fussy – to the objections as you elaborate them, isn't there? (Fussy logic! I like it.)

Curious, what we care about. A wrongly voiced sibilant will drive me distressed from the room ("converzation", /biznez/ for "business"); others don't even hear it. Encountering "whilst" in a text can tip me into melancholia, while others read cheerfully on.

You are right. There are logical arguments being made against these constructions. They aren't *good* arguments, being based on flaws premises, but they are there.

On the other hand, if logic really where the basis of the objections rather than merely the justification, the objections would be subject to modification as additional facts are adduced and faulty assumptions re-evaluated. This happens occasionally, but it is rare. Far more commonly, the logical argument is merely for show.