The enclitic -que in the words neque and atque can be shortened to produce nec and ac. Are there other instances where -que can turn into -c? Can this be productive, or can it only happen in very special cases? (Bonus question: Are there examples where scansion suggests that -que must be read as -c?)

It seems that ac is only used before consonants in classical Latin, whereas atque, nec and neque could be used anywhere, but this is irrelevant here.

  • 3
    The question title is pretty different from the body: I think it's a pretty obvious no that -c cannot replace -que "in general." I would be surprised if there were even a couple more examples of this, much less a codified rule.
    – brianpck
    Commented Dec 10, 2016 at 22:42
  • @brianpck Good point. I edited the title. I suspect that it cannot happen in general. My question is whether there is any amount of productivity or whether the phenomenon is found at all outside ac and nec.
    – Joonas Ilmavirta
    Commented Dec 11, 2016 at 12:57
  • 2
    I've seen a lot of these questions here: When the question is "are there more examples of this", and yet there are no more examples of this, what's a good answer? A simple "No, I haven't found any and no one mentions it"?
    – cmw
    Commented Dec 11, 2016 at 17:28
  • 1
    @C.M.Weimer Questions of that kind are indeed tricky. I think "No, I looked up this and this source and no one that I remember mentions it" is a good answer. A very definitive negative answer is hard to give, but I am willing to accept a "no" and change my mind if someone comes with positive evidence later on. While "no" can be a good answer (and probably the correct one here), it does call for some justification.
    – Joonas Ilmavirta
    Commented Dec 11, 2016 at 21:06
  • @brianpck You commented first, you want the first go for the (tentative) negative answer?
    – cmw
    Commented Dec 11, 2016 at 21:33

1 Answer 1


I wrote a longish post attempting a negative answer, and as a last precaution consulted a list of all Latin words ending in -c. One word stuck out to me like a sore thumb, and further research indicates that it may, in fact, be a contraction of -que to -c:

dōnĕc: conj. [shortened from ante- and post-class. form dōnĭcum , from old dative doni (dioni; for root, etc., v. dies) and conj. cum.... —In the Inscr. of Orell. 4370 DONIQVIES is i. q. DONIQUE IS, and donique = donicum

It appears, thus, that at least some authors saw donec as a contraction of donique. This helps with the parsing of Lucretius, De Rerum Natura, 2: 1116

...ignem ignes procudunt aetheraque aether,
donique ad extremum crescendi perfica finem
omnia perduxit rerum natura creatrix;

Leonard's translation:

...and fires, as on a forge,
Beat out new fire; and ether forges ether;
Till nature, author and ender of the world,
Hath led all things to extreme bound of growth

I will include my other answer, since I believe it otherwise helps shed some light on this:

It appears that this kind of substitution of -c for -que is only present in these two conjunctions:

  • atque: ac
  • neque: nec

It is difficult to prove a negative, and I am willing to be corrected, but one fairly convincing inductive line of reasoning would be to choose similar words (i.e. conjunctions with "que") and check if a similar substitution is attested. Here's a list of such words from a brief search:

  1. cumque
  2. itaque (not really a conjunction, but I'll include it)
  3. namque
  4. sicutque (one attestation in Siculus, Eclogae 6.46)
  5. utque

None of the corresponding pairs (cunc, itac, nanc, sicuc, uc) occurs in a classical corpus, even though some (especially at/ut) seem pretty exact equivalents.

In short: this substitution does not appear to occur in any other words except ac and nec (EDIT) and donec.

  • 1
    I am unclear on whether donec < donique is a false etymology or not. L&S's "donique = donicum" seems pretty cryptic...
    – brianpck
    Commented Dec 12, 2016 at 18:42
  • I looked up donec in de Vaan but he doesn't list it at all, unless I somehow missed it.
    – TKR
    Commented Dec 12, 2016 at 19:00
  • 5
    Oh, it's under a different heading (p. 161). He derives dōnec from a form * dō-ne-kwe, and dōnicum from * dō-ne-kwom, but then says in the next sentence that "Lat. dōni/eque is probably a recent remake on the model of nec/neque". Not sure why the latter statement, but it does look like the -c of dōnec is from -que.
    – TKR
    Commented Dec 12, 2016 at 19:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.