3

To my knowledge every vowel is either long, short or belongs to a diphthong, there are no vowels which are medium in length. If a vowel belongs to a diphthong, it seems that modern writers will mark the second vowel either long or short. Every once in a while a vowel will have both the macron dash, meaning long and the oblique c meaning short on the same vowel, which I guess means that it is not known what the vowel's quantity was or both were possible. It follows that if a vowel is not marked long or short then it must be followed by a vowel with which it forms a diphthong. I've never seen anyone spell this out as a rule, but as far as I can see this would make the most sense. However, if this is the correct rule I've noticed a lot of writers apply this rule inconsistently. Some vowels will be neither marked long nor short and they clearly do not belong to a diphthong as they are surrounded by consonants. For example, at fatus, as you can see sometimes vowels are marked long or short but quite often neither are marked. Does this mean that the vowel is short if it is not marked? It seems that the writer wanted to mark every vowel short with an oblique c but got lazy and gave up. I realize that that site is very unreliable but it's much easier to use than Collatinus. At Collatinus it seems that the authors have taken the trouble to mark every vowel long or short but there are some cases I'm not sure about. collatinus I cannot link directly to the verb conjugation but if you conjugate "edo" you'll see that the Indicatif perfectum futur antérieur 1st person is ēdĕrō̆. I'm not exactly sure what the vowel quantity of 'o' is. Also, if you conjugate 'aequore', you'll see that e is not marked long or short, I'm guessing because it is a diphthong but this supposition is not consistent with their not marking 'u'. As far as I'm aware there are 7 diphthongs in Latin, ei, eu, oe, ae, au, ou and ui. uo is not a diphthong. So why did they not mark u either long or short? If they do believe it is a diphthong then why did they not put the mark on the first vowel?

The same is also true of the Lewis dictionary 1879. If you look at the entry for 'insula' you find "insŭla". What, does he think it is obvious what the quantity of 'i' and 'a' is but for 'u' he decided to make it entirely clear? This is very annoying.

bobsmith76
  • 2,279
  • 3
  • 14
  • 2
    “What, does he think it is obvious what the quantity of 'i' and 'a' is but for 'u' he decided to make it entirely clear?” – well, it may not be “obvious,” but yes: the 'i' before 'ns' becomes nasal and is automatically long. The 'a' is the nominative ending of the first declension and is therefore short. – Sebastian Koppehel Jun 22 '21 at 19:32
  • 2
    In other words, you, the educated Latin speaker who knows all the rules but not necessarily all the words, do not need Messrs. Lewis & Short to tell you the length of the 'i' and 'a', but the 'u' could very well be long, and so they make it clear it is short. Keep in mind that Lewis & Short's dictionary used to be a very expensive tool for professional Latinists and was not intended for pupils in Latin class. – Sebastian Koppehel Jun 22 '21 at 19:39
  • 2
    @SebastianKoppehel You should just go ahead and turn this into an answer. – cmw Jun 22 '21 at 20:52

2 Answers2

4

Many sources don’t mark all vowels with a breve or macron.

Sources that intend unmarked vowels to be understood as short will generally not use breves at all. Examples are Lingua Latina per se Illustrata and Wiktionary (the latter as a matter of general policy; some entries may go against that policy as it is a user-edited dictionary).

Sources that use macrons, breves, macron-breves and unmarked vowels usually don't intend to directly indicate the length of unmarked vowels. I'm not sure if the reasons are explicitly laid out somewhere, but usually this occurs in contexts where vowel length is either predictable, uncertain, or thought to be unimportant. There are two main contexts where this is especially common:

  • Vowels in the final syllable. The length of these vowels is in most cases predictable from the way that a word inflects or from its part of speech. Sometimes, the length of a final vowel is also uncertain or unimportant. For example, several large classes of words with final have variant forms with final ; the distribution of these forms varies depending on the era you look at, but there is no difference in meaning between the long and short vowel variant. This variation with certain words ending in o is why Collatinus shows a macron and a breve on the last syllable of "ēdĕrō̆": it means that the vowel can be found as either short ŏ or long ō.

    Here are some web pages with rules about the length of vowels in final syllables:

  • Vowels in a closed syllable (i.e. before most clusters of two or more consonants). These syllables are treated as heavy in the scansion of Latin poetry regardless of the actual length of the vowel itself, so it can be difficult to know the vowel length, and it is not necessary to know it to determine the correct scansion of a line. In a few special cases (e.g. vowels before ns or nf) the actual length of the vowel before a consonant cluster is predictable, but in most cases, it is not completely predictable.

Note: the Collatinus conjugation/declension tool seems to consistently use macrons to mark syllable weight rather than vowel length, so it won't tell you that the a in actus was long while the a in factus was short.

Diphthongs are not generally marked with either macrons or breves

“Diphthong” is not a very clear term, and I think you might be conflating a few different concepts.

The most consistent use of diphthong is to refer to something like ae, au or oe pronounced in one syllable. This type of diphthong starts with a vowel and ends in a "glide" (spelled e, i or u). The other diphthongs you mention (ei, eu, ou, ui) are very rare, and often coexist with alternative forms that do not contain a diphthong.

In my experience, traditional dictionary leave both letters of this type of diphthong unmarked with no macron or breve. Before a consonant, this kind of diphthong always creates a heavy syllable, the same as a long vowel like ā or ō or a sequence of a vowel and a coda consonant like ar, al, an.

Latin syllables could start with the approximant consonant sounds /w/ or /j/ (also called "semivowels" or "glides"), as in volo or iacit. A glide followed by a vowel is considered a diphthong in some languages, such as Spanish and Italian. However, while there may be some texts that refer to these sequences in Latin as "diphthongs", they are not usually called diphthongs in modern English descriptions of Latin. When the letter U/V or I/J represents an approximant, it is never marked with a macron or a breve. The vowel letter after a glide is marked the same way a vowel letter after a consonant would be marked (e.g. vŏlo/uŏlo, jăcĭo/iăcĭo).

Syllables starting with the consonants /k/, /g/ or /s/ could also have a glide between the consonant and the vowel; alternatively, you can analyze one or more of these combinations as a unitary sound (/kʷ/, /gʷ/, /sʷ/, spelled "qu", "gu", "su" respectively). The u in these syllables is not written with either a macron or breve.

A vowel letter before the glide /w/ is almost always marked long or short according to its actual length: e.g. căvus = /ˈka.wus/ with a short vowel, suāvis = /ˈswaː.wis/ with a short vowel. (Loans taken from Greek words with αυ or ευ may constitute a very insignificant category of exceptions where "āv" or "ēv" represents a diphthong, as in Agave /aˈgau.eː/(?).)

A vowel letter before the glide /j/ is often marked with a macron in words that probably actually had a short vowel followed by long or double /j.j/: e.g. maius or majus was most likely pronounced /ˈmaj.jus/ with a short vowel and a long approximant, and is found in some refernce works as māius. Lingua Latina per Se Illustrata doesn't generally do this, but Lewis, as well as Lewis and Short, does.

Asteroides
  • 28,832
  • 1
  • 80
  • 144
  • 1
    so if an author is marking long vowels with a macron but haphazardly marks a few vowels as short here and there should I just assume that any unmarked vowel is short? – bobsmith76 Jun 23 '21 at 02:06
  • 2
    @bobsmith76: It's usually not safe to assume that. – Asteroides Jun 23 '21 at 02:34
  • 2
    Well that's extremely unfortunate. I don't understand how I'm supposed to learn this. – bobsmith76 Jun 23 '21 at 02:36
  • 1
    @bobsmith76 Every dictionary I own tells me how long vowels are marked and how to interpret short vowels in the front matter. I guess you'll just have to read the front matter if you want to learn it, and it is often different from dictionary to dictionary. – Figulus Jun 23 '21 at 02:49
  • I use online dictionaries and usually the front matter is hidden. My favorite dictionary is the Jeaunneau dictionary available at Collatinus. They have translated examples, plus it is more nicely laid out. – bobsmith76 Jun 23 '21 at 03:23
  • If I were to use a physical dictionary I would have no chance. – bobsmith76 Jun 23 '21 at 03:24
  • However, doing a google search I have not yet figured out when the Jeanneau dict was written. It says 2017 on the software but years sometimes are wrong on the internet. Plus I don't know if the Jeanneau dictionary is a revision of the Gaffiot dictionary or what. I get very upset that people put this info on the internet and they don't tell us where they get it or how reliable it is. I'll do some more research in this question soon but not now. – bobsmith76 Jun 23 '21 at 03:30
  • 2
    @bobsmith76: There are many online resources with information about Latin vowel length, but unfortunately it can be difficult to determine their accuracy or to find the accurate ones. I will also try to update with links to further resources – Asteroides Jun 23 '21 at 03:38
  • Thanks I appreciate that. – bobsmith76 Jun 23 '21 at 03:47
  • Asteroides i would appreciate that. i'll probably be opening another thread on the issue soon – bobsmith76 Jul 03 '21 at 05:09
  • 1
    Actually, let me make sure that you're not doing work that I already know about. I have got files of 7 open source dictionaries which contained vowel quantity info these are: Lewis&Short (1879), Lewis (1890), Georges (1913), Gaffiot (1934), Pocket Oxford (1991?) - then there Jeanneau and Orchard but I'm pretty sure they just copied from the others, though they do provide some new words. I have gathered stats on how often these 5 dicts agree with one another, including the other 2 copyists. I'm trying to figure out who the odd man out is. I'll assembled the stats in an easy to read format. – bobsmith76 Jul 03 '21 at 08:41
  • Ouvrard not Orchard – bobsmith76 Jul 03 '21 at 09:46
1

many times short vowels (like in your example insŭla) are marked on not so common words to make clear where exactly the accent is

  • 1
    Welcome to the site! Just to be clear, are you saying that insula is an uncommon word, and that's why it's so marked? If not, do you have a few examples of dictionaries that do this (especially if they say they do!)? – cmw Jul 01 '21 at 18:30