How do grammarians know that a vowel is long by nature when it is long by position?

Ciaciaufufu

New Member

Sorry for this noob question.
For example we have the verb nārrō, and in the dictionary both the "a" and the "o" are marked with a macron. Since "a" is before a double vowel it would be long by position even if "a" is a short vowel by nature. Therefore, even if we scan poetry and find that it is long, how is it possible that we know it definitely to be long by nature (as suggested by it being marked with a macron in the dictionary) instead of it just being the outcome of being before two consecutive consonants ? Others words like i in īnfirmus or u in fūrtim or e in lēctor also puzzles me similarly.
 

Laurentius

Civis Illustris

  • Civis Illustris

Location:
Lago Duria
They use etymology, look at cognate or even descended words and sometimes they even get some evidence from inscriptions I think. Someone here will probably be able to tell you more.
 

Quasus

Civis Illustris

  • Civis Illustris

Location:
Águas Santas

Quasus

Civis Illustris

  • Civis Illustris

Location:
Águas Santas

Quasus

Civis Illustris

  • Civis Illustris

Location:
Águas Santas
But alterations happen.

Mōns - mŏntis: vowels are regularly long before ns and regularly short before nt.
Pēs - pĕdis: an irregular word, here ē in the nominative is due to compensatory lengthening.
cantāmus: normal thematic vowel ā
cantăt: regular shortening in a final syllable before a single consonant...
cantās: ... but not before -s.
vōx, gen. vōcis but vŏcāre: who would have thought?
 

Glabrigausapes

Philistine

  • Civis Illustris

Location:
Milwaukee
For example we have the verb nārrō
Latin doesn't particularly 'like' a long vowel in a closed syllable (which isn't to say there aren't plenty!) when it has the option to have one or the other (that is, either a short vowel + geminate/long consonant or a long vowel/diphthong + short consonant). The literary standard picks one, while inscriptional evidence might attest the other, e.g.:
ppiter (insc. IOVPITER) < *dyēw- 'shine'
littera (LEITERA) < *lei-t- 'scratch; furrow'
 
 

Godmy

Sīmia Illūstris

  • Censor

Location:
Bohemia
A vowel is always long before nf, ns. This is a kind of compensatory lengthening because -n- eventually disappeared, cf. mensa - Port. mesa).
If I remember the vibe I was getting from it when reading up on it in Allen: 1) it was originally short 2) then "n" in nf/ns disappeared for good, thus the vowel got long 3) then, in the classical Latin there MIGHT have been in some high/educated register (perhaps artificially, of a sort, who knows) reintroduced the pronunciation of "n" but with the long vowel retained... 4) and thus you may have actually got something like "īnfāns" even in the pronunciation.

I don't say that's exactly what it says word by word, but this was always the story I was getting from it...


Mmmm... fortase tē autem Latīnē alloquī mē oportuit, ut nunc cōgitō ;P
 

Clemens

Aedilis

  • Aedilis

Location:
Maine, United States.
If I remember the vibe I was getting from it when reading up on it in Allen: 1) it was originally short 2) then "n" in nf/ns disappeared for good, thus the vowel got long 3) then, in the classical Latin there MIGHT have been in some high/educated register (perhaps artificially, of a sort, who knows) reintroduced the pronunciation of "n" but with the long vowel retained... 4) and thus you may have actually got something like "īnfāns" even in the pronunciation.
I thought the n in words like īnfāns and mēnsa represented a nasalized vowel rather than a nasal consonant, much like the singular accusative -m.
 

Quasus

Civis Illustris

  • Civis Illustris

Location:
Águas Santas
I'm afraid Ciaciaufufu may not understand us then. :)

Yes, I remember that complicated scenario, so I decided to call it "a kind of compensatory lengthening". :)
 

Quasus

Civis Illustris

  • Civis Illustris

Location:
Águas Santas
I thought the n in words like īnfāns and mēnsa represented a nasalized vowel rather than a nasal consonant, much like the singular accusative -m.
Who knows. AFAIK m and n are not confused in inscriptions.
 
 

Godmy

Sīmia Illūstris

  • Censor

Location:
Bohemia
I thought the n in words like īnfāns and mēnsa represented a nasalized vowel rather than a nasal consonant, much like the singular accusative -m.
Well, I describe sort of 4 steps of evolution (hypothesized). What you're describing would happen all in the 2nd second step. The original consonant, once upon a time, was a full nasal, probably. Then, via weaking and further nasalization of the preceding vowel, it got ultimately lost (for some time at least)...
 

Quasus

Civis Illustris

  • Civis Illustris

Location:
Águas Santas
I think it's all hypothetical. Look, Dutch is right in the process of losing its -n's but there are no nasalized vowels at the end of the words AFAIK.
 

Quasus

Civis Illustris

  • Civis Illustris

Location:
Águas Santas
At, nōnne tibi @Quasus (Quase) quoque est dormiendum? :- D Sī in eādem, atque ego, mundī parte versāris, hahahahae...

Nunc igitur potestāte dīvīnā ūtar et computātrum exstinguam! Valēte, bonam noctem!
Est sane, recte mones! Precor ut feliciter somnias!
 

Clemens

Aedilis

  • Aedilis

Location:
Maine, United States.
I think it's all hypothetical. Look, Dutch is right in the process of losing its -n's but there are no nasalized vowels at the end of the words AFAIK.
I don't understand what Dutch has to do with it.
 

Quasus

Civis Illustris

  • Civis Illustris

Location:
Águas Santas
I don't understand what Dutch has to do with it.
It's an example of -n being lost without any nasalization. What about Latin? We know that words were spelled with -m (though not always); there is vague contemporary evidence that its pronunciation was kind of unclear; in poetry -Vm is elided; -m is lost in Romance languages. Does this logically imply existence of nasal vowels? I think not. It's a plausible scenario, but a Dutch-like scenario is not ruled out, and there is no way to know for sure.
 

Glabrigausapes

Philistine

  • Civis Illustris

Location:
Milwaukee
I've always been convinced that the casual elision of -Vm belies a complex morphophonemic process. Much like my written 'What are you doing?' (which I never pronounce as such, but rather as wutchah doin'). Meanwhile, analogous evidence in Indo-Aryan & Balto-Slavic, both of which achieved nasalization at a very early stage (Germanic also, mind you), convince me of the weakness of word-final -m in Latin, whether at the end of an utterance or not. Seems like Latin authors would have written it as a matter of course, because it says 'this is accusative'.
Even something like Mycenean a-to-ro-qo for ἄνθρωπος makes me think that many IE nasals weren't so substantial as our English /m/ and /n/ may lead us to believe.
 

Clemens

Aedilis

  • Aedilis

Location:
Maine, United States.
It's an example of the loss of -n without any nasalization. What about Latin? We know that words were spelled with -m (though not always); there is vague contemporary evidence that its pronunciation was kind of unclear; in poetry -Vm is elided; -m is lost in Romance languages. Does this logically imply existence of nasal vowels? I think not. It's a plausible scenario, but a Dutch-like scenario is not ruled out, and there is no way to know for sure.
You could just as easily point to French, Portuguese, and (some versions of) Spanish to show that nasalization is a very common form of lenition. There is more than just vague contemporary evidence, there is the evidence of historical linguistics.
 

Quasus

Civis Illustris

  • Civis Illustris

Location:
Águas Santas
You could just as easily point to French, Portuguese, and (some versions of) Spanish to show that nasalization is a very common form of lenition.
Yes, this is why I just say that the evidence is inconclusive. That's plausible reasoning, not a rigorous proof. IIRC, there is a point of view that French nasalization may have to do with a Celtic substratum. Now, Celts were also present on the Iberian peninsula.
 

Clemens

Aedilis

  • Aedilis

Location:
Maine, United States.
Yes, this is why I just say that the evidence is inconclusive. That's plausible reasoning, not a rigorous proof. IIRC, there is a point of view that French nasalization may have to do with a Celtic substratum. Now, Celts were also present on the Iberian peninsula.
I'm not using French or Portuguese as an argument for nasal vowels in Latin, I'm simply pointing out that how nasals behave in other languages isn't really useful for reconstructing Latin, in that there are a variety of ways this can work.

Regarding the French nasals, the Celtic substratum probably had little to do with this, as the process of nasalization occurred from 900-1300 CE, centuries after Gaulish was widely spoken (if at all), and there was also a process of vowel quality change and subsequent denasalization—in (later) Old French all vowels before a nasal consonant were nasalized, which is not the case in Modern French. The langues d'oïl went through massive phonological changes from the Carolingian period through the Hundred Years' War, more than occurred before or after.

The evidence will always be inconclusive in historical linguistics. Plausible reasoning is as close as you can get. Regarding Latin pronunciation, I'm sure the theory will continue to evolve, but it's clear that the various spelling pronunciations which are a legacy of the Carolingian period are not representative of how ancient Romans actually pronounced their language.
 
Top