Latin minimal pairs distinguished only by vowel length

[Iáson]

venīre ('come') vs vēnīre ('go for sale/be sold').

venit (pres.) and vénit (perf.)

Also for veniō, the imperative venī vs. perfect vēnī.

 

[Iáson]

ēlāte voc. sg. masc. p.p.p. efferō.
elatē fem. sg. nom. 'fir tree' (loan from Greek ἐλάτη).
Hence also ēlātīs vs. elatīs.

 

[Bitmap]

That's not a minimal pair, though, is it?

 

[Iáson]

Granted they're not distinguished by the length of a single vowel, but they are distinguished only by vowel length.

 

[Serenus]

cēdo ‘I walk’
cĕdo ’give!’

Is the last vowel of the latter short or long?

Secundum Gaffiot, corrupta est. Quoniam terminatio verbi anceps est, unâ et dimidiâ longitudine differunt.

Saepe in librōs histōriam linguae tractantēs, huius verbī mentiō fit ad praebendum exemplum correptiōnis iambicae (quae nōn poētica sed historica), eaedem quae ē duenē et malē quoque contulit bene et male. Sānē vērō remānsit quod multō ūsitātum, etsī ce dare iam nōn adhibēbātur. Orīgō enim invenītur apud *ce 'hīc' (ut in ecce, hunc, hocc(e), huiusce modī) atque *dō 'dā', imperātīvus antīquissimus verbī dare quī fōrmam *deh3 aliter exspectātam indoeuropaeam ostenderet.

 

[Bitmap]

Hasn't that been mentioned above?

 

[Serenus]

Hasn't that been mentioned above?

You mentioned iambic shortening in your post, yes, but I figured I could mention more detail while specifying the diachronic iambic shortening of duenē > bene was involved here, not the poetic one of the Plautine vidĕn hánc (u u —) or volŏ scī´re (u u — u).

 

[Bitmap]

I've always considered iambic shortening a real-life phenomenon, not a poetic licence. I would have thought that at least in volo, many speakers on the street also pronounced 2 short syllables (seeing as it just occurs to often).

 

[Serenus]

I've always considered iambic shortening a real-life phenomenon, not a poetic licence. I would have thought that at least in volo, many speakers on the street also pronounced 2 short syllables (seeing as it just occurs to often).

I think that's likely true, I hadn't thought of that before. There might not be overwhelming ancient evidence for it, but reduction in unstressed words can be easily observed in today's languages. For one ancient example, many if not most Romance languages eliminated the -g- in egō after all, and then they often reduced the -ō to a semivowel, as in Portuguese eu [ew]. In some it went away entirely, as in the stressed jé and gié found in medieval Oïl alongside éo, jéo and jo (modern French je).

The claim being made here is that cedo might've had two short vowels even in careful pronunciation, just as bene did.

 

[Quasus]

Nullum lexicon indicat o productam.

 

[Bitmap]

the -g- in egō

The o in ego is short as well (so far, I haven't found a single instance in poetry where it's long, so I guess it pretty much reflects actual speech) ... that's what I mean by iambic shortening happening so often.

 

[Serenus]

The o in ego is short as well (so far, I haven't found a single instance in poetry where it's long, so I guess it pretty much reflects actual speech) ... that's what I mean by iambic shortening happening so often.

Oh! That's interesting... I looked up the entry in L&S and it does mention "ŏ always in poets of the best age, as Cat., Verg., Hor., etc.; ō ante-class. and post-Aug., as Juv. 17, 357; Aus. Epigr. 54, 6". Most Latin textbooks, at least those in English, use the -ō variant. I was able to find an instance of it in Martial IV.72, but Martial otherwise mostly uses short -ŏ:

— u u | — — | — u u | — u u | — u u | — —
‘Aes dabŏ prō nūgīs et emam tua carmina sānus?
— — | — u u | — | — u u | — u u | —
Nōn’ inquis ‘faciam tam fatuē.’ Nec egō.
(EDIT: the last line is actually "(u) | x" for egō/egŏ, as corrected by Pacifica and Bitmap below.)

And yet again here I note the short -ŏ of dăbŏ...

 

[Pacifica]

The length of the last syllable of a line never matters, so you can't tell from that example.

 

[Bitmap]

Oh! That's interesting... I looked up the entry in L&S and it does mention "ŏ always in poets of the best age, as Cat., Verg., Hor., etc.; ō ante-class. and post-Aug., as Juv. 17, 357; Aus. Epigr. 54, 6". Most Latin textbooks, at least those in English, use the -ō variant. I was able to find an instance of it in Martial IV.72, but Martial otherwise mostly uses short -ŏ:

— u u | — — | — u u | — u u | — u u | — —
‘Aes dabŏ prō nūgīs et emam tua carmina sānus?
— — | — u u | — | — u u | — u u | —
Nōn’ inquis ‘faciam tam fatuē.’ Nec egō.

And yet again here I note the short -ŏ of dăbŏ...

That's not an example of a long o in ego. The last syllable in a pentameter can be short (Ovid, for example, has lots of pentameters ending with pede).

 

[Serenus]

Hmm, yeah. After Pacifica posted that, I found the last syllable of a dactylic pentametre is usually but not always long... The more you know. Thanks!

 

[Serenus]

Nullum lexicon indicat o productam.

Lewis & Short does, and so does Gaffiot, both alongside o correptam.

 

[Pacifica]

It's the same in all kinds of classical Latin verse, I believe, not only in pentameters. The last syllable can just be anything.

 

[Bitmap]

It's the same in all kinds of classical Latin verse, I believe, not only in pentameters. The last syllable can just be anything.

That's true.
It's also true that you sometimes observe some tendency, though. Some schemes try to account for this by putting a macron below a breve to indicate that the syllable is usually long, and vice versa to show that it is usually short ... like this one:

... although you could argue that that scheme shows more of an ideal than the actually reality ... I think in order to make the claim that the last syllable in a pentameter is *usually* long, you would have to consider all closed syllables at the end of a line to be long by position.

 

[Pacifica]

I think in order to make the claim that the last syllable in a pentameter is *usually* long, you would have to consider all closed syllables at the end of a line to be long by position.

But they are.

Though I hate that term "long by position"; it's so inexact.

 

[Bitmap]

It's the same in all kinds of classical Latin verse, I believe, not only in pentameters. The last syllable can just be anything.

That's true.

I think there are a few lines in a few lyrical metres by Horace ... like the third line in the third asclaepiad, which always finishes in a long syllable (again, if you consider closed syllables at the end to be long by position) ... but in the vast majority of metres, the last syllable is anceps.

Spoiler: c. 1,14

O navis, referent in mare te novi
fluctus. O quid agis? Fortiter occupa
portum. Nonne vides ut
nudum remigio latus,

et malus celeri saucius Africo 5
antemnaque gemant ac sine funibus
vix durare carinae
possint imperiosius

aequor? Non tibi sunt integra lintea,
non di, quos iterum pressa voces malo. 10
Quamvis Pontica pinus,
silvae filia nobilis,

iactes et genus et nomen inutile:
nil pictis timidus navita puppibus
fidit. Tu, nisi ventis 15
debes ludibrium, cave.

Nuper sollicitum quae mihi taedium,
nunc desiderium curaque non levis,
interfusa nitentis
vites aequora Cycladas. 20

Though I hate that term "long by position"; it's so inexact.

I see absolutely no problem with that term.