I want to compare future perfect active indicative and perfect active conjunctive. They look identical, apart from first person singular (cogitaverocogitaverim). But is there a difference in the length of the last -i-? It has to be short in third person singular and plural due to its position, but the remaining three persons (cogitaveris, cogitaverimus, cogitaveritis) are less clear to me.

What is the length of the -i- in these forms for future perfect and perfect conjunctive? Are the lengths different? I was taught that they are all short, but I have seen some sources mark the conjunctive long and the future perfect short, and I think I have also seen both marked long.

References to reputable grammars are welcome, but arguments about how we know the lengths would be most useful. My experience is that grammars disagree — although some may be in the minority — and it is not easy to tell what to trust.

  • 2
    Apples an oranges. Ørberg, Henle etc. are secondary, pedagogical grammars, good for studying Latin but completely useless for doing research, whereas Wallace and Weiss are serious researchers generating new knowledge.
    – Alex B.
    May 5, 2018 at 16:49
  • 2
    I will say it for the last time, Ørber, Henle, TY Latin and Wheelock are pedagogical grammars but they are of no research value. What they say is of little interest to me.
    – Alex B.
    May 5, 2018 at 17:16
  • 1
    I learned my Latin from pedagogical material, not scientific material (as almost everyone, I suppose). Pedagogical material will always cut some corners short and they will also contain occasional errors. Therefore it is of great value to understand why we believe the vowels are as long as they are, and the strongest answers to that rely on research. Alex B.'s answer indicates that both lengths are possible in both forms, and it therefore makes sense for a pedagogical author to pick the length that makes the most sense to them. Consulting several grammars is useful but gives no proof.
    – Joonas Ilmavirta
    May 5, 2018 at 17:48
  • 2
    @JasperMay Asking what happened in later or earlier Latin would make a good follow-up question. When it comes to post-classical Latin, I find it unlikely that a quantity distinction would be reintroduced, but this is only my intuition speaking. It would also make sense to ask about the origins of length choices of modern grammarians. All the details on this matter will certainly not fit in this single question, and I'm always happy to see people pick up a question or answer dig deeper in a follow-up question.
    – Joonas Ilmavirta
    May 5, 2018 at 17:59
  • 1
    @JasperMay I hope in my comment I didn't come off unnecessarily dismissive or rude. I have to admit, the word "pagan" did rub me the wrong way; it would be nice to stick to the accepted, standard terminology, which is "classical." The question you ask "why and when did such respected Latin grammars begin to teach that they should be distinguished?" is interesting and I encourage you to do research on that and report your findings here.
    – Alex B.
    May 5, 2018 at 20:02

2 Answers 2


short notes/quotes from professional, serious, research-based sources (I thought it was clear to everyone what kind of sources I use; will make small changes later):

The paradigm from Tronskii 1960:

scan from Tronskii

Weiss 2009/2011: "Although etymologically the stem vowel of the perfect subjunctive should be ī and the stem vowel of the future perfect should be ĭ, Classical Latin poets use both long and short -i- indiscriminately in the 2nd sg., 1st pl., and 2nd pl. of both paradigms" (p. 420; emphasis mine - Alex B.).

Wallace 1988/9: "In the literary dialect of Latin - the dialect which forms the base of our teaching grammars - the PS and the FP are distinct only in the 1SG" (p. 164; emphasis mine - Alex B.).

Pinkster 2015: "From roughly Cicero's time onwards, the future perfect indicative forms and the perfect subjunctive forms were no longer morphologically distinct, except in the passive" (p. 462; emphasis mine - Alex B.).

See Neue and Wagener for further details (in German).

  • 2
    This is very useful. Apparently most pedagogical authors have chosen one quantity for simplicity, out of personal preference, or due to misconceptions corrected by later research. Knowing that both are acceptable is useful for metric writing where morphological flexibility helps a lot, and helps understand discrepancies between grammars.
    – Joonas Ilmavirta
    May 5, 2018 at 17:53
  • @Draconis Thanks! The image is from Tronksii 1960, so it should be moved up - if you could help me, that'd be great! (I uploaded it from the iPhone).
    – Alex B.
    May 5, 2018 at 20:04
  • 2
    Of course in egerint the quantity of /i/ is "hidden". We do not know if it is long or short.
    – fdb
    May 5, 2018 at 22:51

[I've voted for Alex. B.'s answer because he lists the research on classical Latin. My answer discusses several respected pedagogical grammars, but I don't know why and when these started to prescribe the distinction.]

The future perfect ends in:

  • -erō
  • -eris
  • -erit
  • -erimus
  • -eritis
  • -erint

Note the similarity to the future of 'esse': erō, eris, erit, erimus, eritis, erunt.

The perfect subjunctive ends in:

  • -erim
  • -erīs
  • -erit
  • -erīmus
  • -erītis
  • -erint

Note the similarity to the present subjunctive of 'esse': sim, sīs, sit, sīmus, sītis, sint. The s here turned into r between vowels, like in floris < *flosis.

Of grammars that use macrons, Ørberg doesn't mark a difference between the future perfect and the perfect subjunctive, but e.g. Henle, TY Latin and Wheelock do. This Utah State University teacher comments on Wheelock's chapter 30 (http://www.usu.edu/markdamen/Latin1000/Presentation/transcriptions/30T.pdf, page 2):

the perfect subjunctive features long marks in its second-person forms and first-person plural (-erīs, -erītis, and -erīmus), whereas all those ī’s are short in the future perfect indicative. In other words, the Romans could hear a difference between “you have loved” (amaverīs) and “you will have loved” (amaverĭs) and so the long ī’s in the subjunctive should be mandatory long marks, technically.

He then recommends students not to worry about them in the beginning, but I assume your goal is to progress beyond the beginning.

  • As I wrote, I was taught that they're all short, so different grammars disagree. Perhaps the ones I used are wrong, but it's hard to tell what's right and wrong. Do you know how we know the vowel quantities?
    – Joonas Ilmavirta
    May 5, 2018 at 14:00
  • You can predict the vowel quantity from the forms of 'esse'. I think we know simply from tradition, but you may be able to reconstruct it from scansion.
    – Jasper May
    May 5, 2018 at 14:05
  • What is your source then, and do you trust it more than Henle and TY Latin? Any grammar with macrons will show the right vowel lengths of 'esse', but not all will distinguish the future perfect and the perfect subjunctive, e.g. at least Ørberg doesn't, I don't know of others. I think this is a slight mistake or oversight by Ørberg. I've added a quote from a teacher.
    – Jasper May
    May 5, 2018 at 15:06
  • I have used grammars and textbooks written in Finnish. I don't remember what the older grammars say (I don't have my library with me here in Cambridge), but I do recall my textbook and grammar giving a short -i-. It might well be a mistake on their part. I am willing to unlearn if I have learned something wrong, but there is a bit of resistance. // Sorry, my previous comment was wrong (and deleted); it was indeed sis with long and eris with short -i-, I was just temporarily confused by the sound changes in the perfect. So the discrepancy is only in the perfect stem forms, not esse.
    – Joonas Ilmavirta
    May 5, 2018 at 15:13
  • 1
    All right, then if you correctly recall your Finnish book's lack of these 3 macrons, it agrees with Ørberg but disagrees with Henle, TY Latin and Wheelock.
    – Jasper May
    May 5, 2018 at 15:45

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.