« Word final *n in indo-european n-stems | Main | Because Japanese reduplicates too! »



Glen Gordon

I can't believe nobody's commenting on this. Unfortunately I can't say that my forté involves the details of modern Semitic languages and I have no clue what a better explanation of nunation in Arabic would be in light of the spelling conventions you mention in the older language.

However, on a potentially distracting tangent, I've wondered rather generally and vaguely as to whether it's possible that Proto-Semitic had mimation in the singuilar and nunation in the plural. Has this already been theorized or is it assumed that Proto-Semitic only had mimation? When I think about it, the word *sabʕ-at-u-m "seven", which of course was attested as a loan in PIE, would seem to suggest that mimation was also used in the plural... unless, mimation here was meant to convey collectivity, as in "_a_ group of seven"? I notice that *n is used in pronominal plurals both in Semitic and Ancient Egyptian. I haven't investigated this in depth yet (so many questions, so little time) but my spidey senses are pointing me to this general hunch. Am I off track?





I sometimes suspect that using Vox was one of the worst choices I couldn't made, and therefore nobody is commenting. Maybe I should finally switch to blogspot.

As for nunation-mimation issues, I honestly have no idea, I only know quite a bit of Arabic (that is, a year's study and then some, a little bit of Hebrew and a tiny little bit of Akkadian (All that is necessary to make sense of Hittite :P). All I know of proto-semitic is common knowledge, or stuff that seems fairly obvious.

Arabic of course only displays nunation, with no clear signs of mimation. Word final *m > n. Also in the plurals we find endings like -uun(a) -iin(a). Though pronouns seem to ignore this ('antum) but maybe there was something behind these pronouns in Proto-Semitic, I'm unsure.

Then if we look at Hebrew we find no sign of nunation or mimation in the singular, but we find the plural ending -im which points to a mimated plural.

But this could be analogy.

Akkadian has the following masculint paradigm:

nom beelum
gen beelim
acc beelam
nom beeluu
gen-acc beelii

The femal has the follwoing:
nom beltum
gen beltim
acc beltam
nom beeleetum
gen-acc beeleetim

(Long vowels indicated with double vowels for convinience since it's 5AM, and don't feel like looking for the right characters).

All this evidence seems to point towards mimation. If anything, there was nunation after long vowels.

In traditional Arabic grammar paradigms aren't treated so much as having a disctinction between plural and singular. The sound masculine plural is simply considered a noun class. Same for the sound feminine plural (and the dual).

Broken plurals though, follow normal patterns. Like rajul 'man' and rijaal 'men' both follow the pattern as presented in the above post. Besides that maybe we should keep in mind the incredibly vage distinction between plurals and singulars in at least Arabic. shajar 'trees (in general)'  shajara 'tree' shajaraat 'trees'. The first two obviously act like sound singulars, and here the difference between shajar and shajaraat is still semantically there, but there's obviously nouns where this becomes a lot less meaningful.

From that point of view, I'd say it doesn't makes sense to differe between nunation and mimation as a distinction of plural and singular, as plurals only form one class of noun conjugation. Besides that, most evidence I have access to seems to point towards mimation, also in plurals.

But as I said, I have not much expertise in this field, I'm pretty much thinking out loud.

Glen Gordon

However what about the presence of nunation in the Akkadian dual alongside the aforementioned mimation in the singular and plural? Why then does Akkadian have both?


Good question. I'm not expert on Akkadian I'm afraid, but I could imagine a shift of *m > *n /V:_#

Or in human language: Word final *m that turns into *n before a long vowel at the end of a word.

But this would make the absence of any nunation or mimation in the plural difficult to explain though. It currently sort of looks like an idaafa plural. But I honestly don't know enough of Akkadian to actually say anything sensical about its case and indefiniteness markers.

Glen Gordon

I could be wrong but from what I've personally scanned through on Semitic linguistics, the question of the presence or absence of nunation in Proto-Semitic is an open one. Note also that Eblaite also employs nunation rather than mimation in the dual.


Phoenix: "I'm not expert on Akkadian I'm afraid, but I could imagine a shift of *m > *n /V:_#"

That this is a plausible sound shift in general worldwide there's no doubt. However, I'm skeptical of it in Proto-Semitic since *-m# and *-n# are otherwise phonemically distinct and they remain distinct in Akkadian even after long vowels (e.g. inān "two eyes" versus rabām "great (acc.sg.)").


Andrew Brehm

I know I am late, but I just found this now.

Could both nunation and memation be remnants of a nasal vowel? That would explain the shift between /n/ and /m/ as well as why they are not always there and why they might not be written at all times.

A nasal vowel based on /a/ could be written with a symbol for /a/, or symbols for both /a/ and /m/ or both /a/ and /n/ depending on the exact quality of the nasal vowel. And in an abjad without vowels, the nasal vowel might have been written with a symbol for /m/ or /n/.

This would also explain the construct form for regular masculine plural in Hebrew. If another word follows, writing the Mem after the Yud that is actually a nasal /i/ might not make a lot of sense.

The regular female singular construct form would have a totally different explanation. In that case the non-construct form would be the result of not proncouncing a Tav that actually belongs there.

I am thinking

chaver (non-construct) to chaver (construct, why not)

chaverat (actual felame form) to chavera (non-construct form because proncouncing a regulat /t/ is unecessary)

chaverim (non-construct) to chaveri (because the /im/ is really a nasal /i/ vowel anyway and proncouncing what is only nearly an /m/ sound is unecessary when another noun follows or in a posessive form).

chaverot (broken plural form of chaverat, only a vowel changes) to chaverot (because chavero would be too similar to chavera and the plural is the one that ended up more distinctive).

I think that would explain the Mem and Nun and the lack of it at times.

Now, the big question is, why does Kana3an have a Nun at the end? It's Hebrew or Phoenician and hence wouldn't be subject to Nunation. And the root KN3 doesn't have a Nun at the end. So where does the Nun come from?


Hey there! Thanks for the reply sorry I took so long but here's my thoughts:

I personally doubt the nasal -m/-n come from a nasal vowel. I've never heard of such a shift; And actually I have never heard of nasal vowels that didn't originally come from nasal consonants following vowels.

Since the m/n distribution is neatly divided among languages, it's probably safe to assume the original element was -m, which became -n word finally in Arabic. This is a common shift, like this we see the Acc Sg. of indo-european *-om > *-on in Greek.

Then there's a little problem in your theory about non-construct/construct stuff. Sounds don't just disappear because it's 'unnecessary'; Sounds disappear due to (usually regular) sound changes, there's often a well grounded phonetic reason, not just 'necessity'. Of course there's exceptions where it does happen, but I'd venture such things don't happen when not at least a similar regular phonetic process has taken place to which it can base the process.

With that said I'm not completely clear on what triggered all these shifts But it seems to me that word final -t was lost. I also doubt chavero would be too similar to chavera. They're very different vowels even in Hebrew, which is a language which is actually pretty good in not screwing around with its vowels for a semitic language. Unlike Arabic where especially the i/a sometimes switch around for rather mysterious reasons.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Your Information

(Name is required. Email address will not be displayed with the comment.)