Scansion of ἄρρενας

Here you can discuss all things Ancient Greek. Use this board to ask questions about grammar, discuss learning strategies, get help with a difficult passage of Greek, and more.
Post Reply
Archimedes
Textkit Neophyte
Posts: 88
Joined: Fri Dec 26, 2008 5:35 am

Scansion of ἄρρενας

Post by Archimedes »

This is another line from the Greek Anthology 9.686:
βάρβαρον οὐ τρομέεις, οὐκ ἄρρενας ἀρρενοκοίτας.
Everything before ἄρρενας scans as -uu -uu --; now if ἄρρενας could be scanned -uu, then the rest of the line would also scan -uu -uu --, a neat dactyl-dactyl-spondee. However, one of the guidelines I have for determining vowel length states that the alpha in the ultima of masculine and feminine nouns is long, and since long vowels are scanned long, ἄρρενας would have to be scanned as -u-, throwing off the meter. Is there any possible resolution to this? Thanks again.

User avatar
Paul Derouda
Global Moderator
Posts: 2292
Joined: Fri Nov 05, 2010 9:39 pm

Re: Scansion of ἄρρενας

Post by Paul Derouda »

The accusative plural ending of the first declension in -ας is always short, as far as I know. Actually, if it had been long, in Attic-Ionic the α would usually have changed to η, like in the singular (not after ι ε ρ in Attic, but even there in Ionic).

Qimmik
Textkit Zealot
Posts: 2090
Joined: Mon Mar 18, 2013 10:15 pm

Re: Scansion of ἄρρενας

Post by Qimmik »

ἄρρενας is 3rd declension accusative plural. The alpha in -ας is short.

http://www.perseus.tufts.edu/hopper/tex ... Da)%2Frshn

The alpha of the first declension accusative plural is long.

http://www.perseus.tufts.edu/hopper/tex ... ction%3D13

As Smyth notes, the first declension accusative plural ending is derived from -ανς. The nu was lost and the alpha lengthened by compensation. This sound change must have occurred after the sound change of long alpha to eta in Attic-Ionic ceased to be operative, so that the long alpha derived from -ανς did not change to eta in those dialects. This is also why the masculine nominative singular aorist active participle ends in -ας with long alpha in Attic-Ionic, derived from -αντς.

User avatar
Paul Derouda
Global Moderator
Posts: 2292
Joined: Fri Nov 05, 2010 9:39 pm

Re: Scansion of ἄρρενας

Post by Paul Derouda »

Oops. Silly me. I meant third declension of course, but somehow mechanically said first declension. I'm sorry for messing things up even more.

Btw, the word looks Attic since it's ἄρρενας and not ἄρσενας (nom. sing. ἄρσην). The origin of α must be different than with the first declension; I suppose it used to be a vocalic n or m. (/arsenms/ or the like with the m pronounced like a vowel. I didn't check this anywhere so I might be wrong.)

Qimmik
Textkit Zealot
Posts: 2090
Joined: Mon Mar 18, 2013 10:15 pm

Re: Scansion of ἄρρενας

Post by Qimmik »

The alpha of the third declension accusative ending -ας can be traced to a vocalic (syllabic) n, which typically becomes alpha in Greek. The alpha privative, for example, can be traced to the PIE negative particle -n-, which shows up in Latin as in- and in Germanic languages as un-. So the accusative plural endings of the three declensions are parallel:

-α-νς > -ας (long alpha)
-ο-νς > -ους
C-Nς > -Cας (short alpha; C = consonant; N = syllabic n)

Yes, ἄρρενας is the Attic form.

Archimedes
Textkit Neophyte
Posts: 88
Joined: Fri Dec 26, 2008 5:35 am

Re: Scansion of ἄρρενας

Post by Archimedes »

Thanks for the input. Because alpha, iota, and upsilon can be either long or short, I was casting about for some concise guidelines on determining their length. I came across this list (fourth post down in the thread), which I believe I have also seen cited on Textkit:

http://www.ellopos.net/elpenor/koinonia ... PIC_ID=542

So much for that.

User avatar
Paul Derouda
Global Moderator
Posts: 2292
Joined: Fri Nov 05, 2010 9:39 pm

Re: Scansion of ἄρρενας

Post by Paul Derouda »

There are indeed some problems with that list, at least with "α in the ultimate of male and female nouns and adjectives is long"... -α in γλῶττα and some other first declension words is short. Masculine singular accusatives in -α also end in a short syllable, like ἄρσενα. If there are exceptions, they don't come to my mind now.

Archimedes
Textkit Neophyte
Posts: 88
Joined: Fri Dec 26, 2008 5:35 am

Re: Scansion of ἄρρενας

Post by Archimedes »

Paul Derouda wrote:ἄρσενας (nom. sing. ἄρσην)
I'm really confused now. I went back to my old grammar to review the rules for accentuation, and according to a statement on the place of accent for nouns, "When the nominative singular has the accent on the penult or the ultima, it stays there." Not so in the quote above.

Does there exist such a thing as a concise, reliable guide to accentuation (and also to the determination of the vowel length of alpha, iota, and upsilon)? Or does one have to pick through Smyth each time to be sure?

User avatar
Paul Derouda
Global Moderator
Posts: 2292
Joined: Fri Nov 05, 2010 9:39 pm

Re: Scansion of ἄρρενας

Post by Paul Derouda »

Accents are difficult. I think most grammar books have some sort of introduction to the subject, but they are usually rather simplified. I don't know about Smyth, but probably even he doesn't cover every case.

For my own "I'm gonna learn my accents one day" project I have acquired a rather recent book by Philomen Probert, "A New Short Guide to the Accentuation of Ancient Greek", which many people have recommanded. I haven't read it yet though. There's also a little French book on the subject By Michel Lejeune.

I think the point in the case you mention is that the accent stays on the same syllable as much as possible counting from the beginning of the word. I don't want to go to far into specifics because probably I'll just say something that's incorrect.

Qimmik
Textkit Zealot
Posts: 2090
Joined: Mon Mar 18, 2013 10:15 pm

Re: Scansion of ἄρρενας

Post by Qimmik »

ἄρσενας (nom. sing. ἄρσην)
"When the nominative singular has the accent on the penult or the ultima, it stays there." Not so in the quote above.
The accent stays on the syllable accented in the nominative singular, i.e., ἄρ-, unless an inflexional ending forces it forward.

Post Reply