I'm thinking about which diacritics to use in Latin to give pronunciation hints without writing the length of all the vowels (which I find very noisy). My main aim is to avoid homographs that are not homophones.

How common are minimal pairs, distinguished only by the length of the vowel in an unstressed non-last syllable? In other words, is it enough to be unambiguous in the majority of cases, to mark stressed syllables and final vowels (see this previous question of mine: Which vowel lengths and stressed syllables are marked in the Vatican Lexicon?).

  • Vowel length has an effect on stress, but I assume you want the syllable in question to be unstressed in both words, not just one.
    – Joonas Ilmavirta
    Commented Apr 24 at 13:00
  • @JoonasIlmavirta Yes you are correct, thank you for the point
    – user14825
    Commented Apr 24 at 15:37

3 Answers 3


The verb nĭteō "shine" is only used in the active, and the verb nītor "strive" only in the passive (it's deponent), so at first glance it seems like there won't be confusion between them. But both of them have participles and gerunds, leading to confusion between nītentēs and nĭtentēs, or nītendum and nĭtendum.

These sorts of minimal pairs aren't especially common, but they do exist.


Here are two more examples:


It barely matters if it's stressed. Sometimes the stressed vowel is short.

I really recommend this channel: https://www.youtube.com/@ScorpioMartianus

These are quite common homographs:

āeris: "of the air",
aeris: "of copper", "of coin"

malum: "bad thing",
mālum: "apple"

venī: "come" as imperative,
vēnī: "I came"

pedīs: "of foot",
pēdīs: "of louse"

pēditum peditum: "fart of the infantry"

lēgō3: "I delegate",
legō1: "I read"

Because you aim to give pronunciation hints:

Let's put Ecclesiastical Latin aside because the "speakers" but rather writers, use macrons inconsistently, since after around 900AD people weren't speaking the language of administration and didn't really know which were the long/short vowels. Stress became more dominant in their speech and writers of different neo-vulgar-latin regions often used diacritics where they thought it might be important.

In Classical times vowel length was (imo still is) a key feature of the language. By "key feature" I mean, not only the last vowel was important, which indicates the case of the word many times, and not the vowels of differing lengths in quasi homographs, but every vowel. It gives the rhythm of the whole language and does that of the metric poems.

So, to give pronunciation hints, you must indicate EVERY long vowel (even the stress but that could be inferred). There are a lot of complete homophones concerning vowel lengths so help the reader wherever you can.

The Classical Romans, by the way, used "accent" marks (á,é,í,ó,ú) and not macrons (ā,ē,ī,ō,ū) for long vowels and never marked the stress or the short vowels.

At last, don't find it noisy, I'm Hungarian and our language has short and long vowels. It doesn't matter if somebody misplaces the stress, but exchanging a vowel length can lead to funny situations. A decent Hungarian writes out all the length marks!:D

  • 3
    I don't think it's quite fair to compare vowel length markers in Latin and in Hungarian. For the latter, they're feature of the alphabet; for the former, they're mostly used for pedagogical purposes.
    – brianpck
    Commented Apr 26 at 2:38
  • 1
    The Romans didn't use the apex with i (í), they used the long i instead (ꟾ).
    – Cairnarvon
    Commented Apr 26 at 3:43
  • 2
    As far as I can tell, the point of this question is that examples of the requested kind are hard to find, not that others wouldn't matter.
    – Joonas Ilmavirta
    Commented Apr 26 at 6:31
  • 1
    @brianpck It does not for the orthography, but it sort of does for the perceived importance of the vowel length. As a speaker of Czech, I can certainly relate. Your stress is on different syllables? You just sound a little bit strange, perhaps from the regions bordering Poland. You use the wrong vowel length? You are using completely different or non-existent words. That is still true even when I (as I often do) do not mark the length on the internet or in informal e-mails. Commented Apr 26 at 13:16
  • 1
    @VladimirFГероямслава I agree it matters for pronunciation, but I took the final sentence ("a decent Hungarian writes out all length marks") to be suggesting that a decent Latinist should write them out as well, which just isn't true.
    – brianpck
    Commented Apr 26 at 13:23

Your Answer

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