Hello,
I would really like to be able, when a I meet a Latin text with unmarked vowels, to identify the long ones and mark them for my own reference, and exercise.
Do you know of any resource where I can find all the rules about this issue?
Thank you very much for your help,
Francesco
Some are short, some are long. There are few “rules.” You just have to learn which are which, word by word. Reading verse metrically helps greatly (you can think of it as a form of the so-called direct method); dictionaries give a certain amount of information but metrical context aids memory retention. With some vowels, the quantity is not known and never will be, but most are known and are quite easily learnt. (Not all, mind. I have gone through life imagining that the “a” of iam was long, but in fact it’s short.)
You’ve got the most important point, that all vowels are either long or short, not something inbetween. It’s a binary system. And you’ll know not to confuse vowels with syllables.
Vowels which are long by nature are usually so marked in dictionaries, and beginning textbooks usually do as well. Memorizing the long marks as part of the spelling when learning vocabulary and paradigms really helps later in your Latin life.
In addition to reading poetry and paying attention to the vowels as marked in textbooks and dictionaries, as the others have suggested, listening to Latin spoken or read aloud by experienced speakers will be very helpful: Wilfried Stroh, Terence Tunberg, Daniel Pettersson (latinitium.com) and Johan Winge are some of the Latin speakers I recommend you listen to.
While many speakers are not 100% consistent in terms of vowel length, and some do not pay much attention to it at all, listening will at the very least tell you which syllable of a given word is stressed, which is also an important part of correct pronunciation and can provide some information on vowel length: If you hear the word amavérunt, for example, you know that the e is long, because otherwise the paenultima wouldn’t be stressed. If you hear pópuli, you know the u is short, because otherwise it would be stressed. However, this only applies to the length of the paenultima.
I’m using Familia Romana (Lingua Latina books) and found the following YouTube complement helpful. However, not sure how “authentic” the pronunciation is.
https://www.youtube.com/playlist?list=PLSpl3cO92FvZvZa2t5ZGA_lkUVj_1bQE-
I’m sorry to say that Barry perpetuates an antiquated misapprehension when he speaks of vowels “long by nature.” All long vowels are long by nature, and none are long by position. That is to confuse vowels with syllables, a fundamental error.
Indeed, the syllable is the thing, and in using the traditional designation of a vowel as long by nature, I did not intend to imply that vowels rather than syllables could be lengthened by position. Glad I could afford the opportunity for you to make the distinction. I was mainly concerned to point out that dictionaries and teaching texts use macrons to mark vowels which are always long. Bennet is helpful:
QUANTITY
- A. Quantity of Vowels
A vowel is long or short according to the length of time required for its pronunciation. No absolute rule can be given for determining the quantity of Latin vowels. This knowledge must be gained, in large measure, by experience; but the following principles are of aid:—
- A vowel is long,[6]—
a) before nf or ns; as, infans, inferior, consumo, censeo, insum.
b) when the result of contraction; as, nilum for nihilum.
- A vowel is short,—
a) before nt, nd; as, amant, amandus. A few exceptions occur in compounds whose first member has a long vowel; as, nondum (non dum).
b) before another vowel, or h; as, meus, traho. Some exceptions occur, chiefly in proper names derived from the Greek; as, Aeneas.
B. Quantity of SyllablesSyllables are distinguished as long or short according to the length of time required for their pronunciation.
- A syllable is long,[7]—
a) if it contains a long vowel; as, mater, regnum, dius.
b) if it contains a diphthong; as, causae, foedus.
c) if it contains a short vowel followed by x, z, or any two consonants (except a mute with l or r); as, axis, gaza, resto.
A syllable is short, if it contains a short vowel followed by a vowel or by a single consonant; as, mea, amat.
Sometimes a syllable varies in quantity, viz. when its vowel is short and is followed by a mute with l or r, i.e. by pl, cl, tl; pr, cr, tr, etc.; as, agri, volucris.[8] Such syllables are called common. In prose they were regularly short, but in verse they might be treated as long at the option of the poet.
NOTE.—These distinctions of long and short are not arbitrary and artificial, but are purely natural. Thus, a syllable containing a short vowel followed by two consonants, as ng, is long, because such a syllable requires more time for its pronunciation; while a syllable containing a short vowel followed by one consonant is short, because it takes less time to pronounce it. In case of the common syllables, the mute and the liquid blend so easily as to produce a combination which takes no more time than a single consonant. Yet by separating the two elements (as ag-ri) the poets were able to use such syllables as long.
Bennett, C. E. (1908). New Latin grammar (pp. 21–22). Boston; New York; Chicago: Allyn and Bacon.
It’s painful to see Bennett’s stultifying 19th-century product commended here when understanding of Latin phonology has come such a long way since then.
Would you mind pointing out what Bennet got wrong, and what would you suggest instead?
Hey Michael, ditto Barry’s request. I didn’t know Bennett (whose work I absorbed years ago when I was focusing on vowel length just like the OP) was outmoded, and I’d love to learn the particulars.
Thank you very much for all your comments, really. I am not experieced enough to comment on Bennet’s work, but it seems to me a good starting point for what I am looking for, that is rules the would allow me to put macrons on vowels without having to check every word in the dictionary.
I know where to find vowel information. I have a good dictionary, and my textbook gives it in full (by the way, I am using Lingua Romana, and I am working now on Lesson XXXI). My point is, given an unmarked text, to be able to put macrons myself. As a skill. And this is the skill I would like to start to work on.
So, are Bennet’s rules the only ones availble or is there more? Now, forgive me, I am going to sound extremely naive, and I think I am at this stage of learning… How do dictionary and textbook authors and developers know where to put the macrons? The must rely on some rules, or not?
Thank you for your patience,
Francesco
Francesco, You ask how it’s known what vowels are long. It’s a good question. Two ways: (morpho)phonology and metre.
Take amo (“I love”). It’s known the o is long (1) because -o at the end of a word regularly is, or more narrowly because first-person singular present indicative -o regularly is (except in circumstances where it’s shortened or elided!); and (2) because it scans long, as in Catullus’ hexameter “odi et amo; cur id facias, fortasse requiris,” where the o of amo occupies the long element of the second foot just as the o of odi does in the first foot.
(1) empirically simplifies an issue of comparative and historical linguistics. (2) is relatively straightforward. So learning to read metrically helps a lot—as I said in my first post.
(Reading “odi et amo; cur id facias, fortasse requiris,” for instance, reveals the length of nearly every vowel in the line. It doesn’t reveal the length of the first two vowels of fortasse—who cares?—but it does reveal that fortasse’s final vowel, which L&S mark as long, is actually short.)
As a beginner you can use Bennett if you must, but you’ll quickly find that for the purpose of adding macrons he’s almost useless.
Barry & Randy, Well just to oblige. Objections would include:
treating vowel length as a purely acoustic matter, ignoring perceptual and conceptual frames of reference;
still worse, doing exactly the same with syllable length;
claiming random phonological phenomena as principles;
neglecting diachronic change, and downplaying synchronic variation;
taxonomic inadequacy; linguistic ignorance; and overall puerility.
But it may be that none of that bothers you.
So, what modern treatment would you recommend?
Thank you, Michael, that was very helpful.
Francesco,
Two things:
- For your benefit, let’s clarify what “Bennett” is all about. Charles Bennett was the author of a Latin grammar (“New Latin Grammar”), originally published in 1895 and followed by subsequent editions. It is one of at least two turn-of-the-twentieth-century grammars (the other is Allen and Greenough’s “New Latin Grammar for Schools and Colleges”) that are available on Google Books and are still used by many students today, especially autodidacts, as reference grammars. (For a reference grammar I mostly use A/G. If there is a more contemporary comprehensive Latin grammar (in English), I’d like to know about it for my own purposes.)
Bennett supplemented “New Latin Grammar” with a separate publication (also 1895) called “Appendix to Bennett’s Latin Grammar” (“for teachers and advanced students”). In 1907 he published “The Latin Language: A Historical Outline of its Sounds, Inflection, and Syntax”, which he intended to supersede the earlier “Appendix”. You can find all these on Google Books.
Now, how this relates to vowel quantity. At the end of “New Latin Grammar”, Bennett included a brief discussion of Latin prosody, including “Quantity of vowels and syllables: General Principles”. In “Appendix” and again in “The Latin Language” he supplemented this very brief presentation with a chapter (chapter 3, to be specific) on “Hidden Quantity”. A hidden quantity, in his definition, is the quantity of a vowel before two consonants (the corresponding syllable will always be long in verse, but the quantity of the vowel as pronounced by the Romans is “hidden” to us). And what are the hidden quantities specifically? Bennett outlines his sources and methods for determining hidden quantity (entertainingly characterized by mwh!) and then presents in the form of “General Principles” the fruits of his research. So his first gp is, for example, that “a vowel is always long before ns and nf, e.g., cōnsul, īnfēlīx.”
Many Latin students, again especially autodidacts, in their attempt to incorporate “correct” Latin vowel quantity into their pronunciation, have applied Bennett’s rules. When I was dusting off my Latin about ten years ago, for example, I spent some time reading Bennett and listening to Evan Millner on the same.
Now, whether or not all or any of Bennett’s rules are regarded by Latin scholars today as valid, I have no idea (strictly out of curiosity, I was hoping for some enlightenment, preferably in the form of citations, from mwh, our resident academic). In my own case, I am way past the point of caring - I am resigned to being īnfēlīx, if that is the case. My recommendation to someone in the very beginning stages of learning Latin pronunciation and mastering vowel quantity, like yourself Francesco, would be to be aware of “Bennett” (meaning his hidden quantity rules), if you’re curious, but set him aside and focus rather on the basics.
- And how, you asked at the outset, do you do that? When all is said and done, there are few rules, as mwh replied at the outset. The only way you can mark up Latin text with macrons is through some combination of (i) the hard work of memorizing vowel quantity word by word as you progress, and (ii) looking up ones you don’t know in the dictionary. To do (i), you have to utilize all the resources at your disposal, such as introductory Latin texts that print macrons, dictionaries, reading lots of verse, metrically, when you’re ready for that, etc. There is no shortcut.
If there were really any rules for this, a computer program would have been written to macron-ize unmarked Latin text. Man, could I use such a program! I have done some recording of Greek and Latin, and I always prepare my text with macrons and rehearse my pronunciation accordingly. It is tedious in the extreme.
[I have a very vague memory of someone having attempted such a program/script. Perhaps someone on Textkit knows.]
I use E. C. Woodcock A new latin Syntax. Bristol classical press 1959.
I also quite like Gildersleeve’s (and Lodge) Latin grammar which you missed out from your survey of turn of the century grammars.
Does a beginner need these? Probably not.
There are several computer applications adding macrons to Latin words:
- Collatinus
- Johan Winge’s macronizer: “accuracy on an average classical text is estimated to be about 98% to 99%”
- CLTK (developeres library): two years ago, when I tested it, it failed very often comparing to the L&S dictionary
Thanks Will for refreshing my memory and supplying these links for Francesco’s benefit. I’m not sure I trust these newfangled machines called computers, but have at it, Francesco, if you wish. Of course it requires you to supply/type in plain text the Latin you wish to be marked up - not sure if you were talking about marking up hard copy.
Randy
Francesco,
If you are using Roman authors, you can always copy and paste from http://www.thelatinlibrary.com/
The works of most authors are available there.