Reading versions of Anabasis, Iliad

Greek Learning Greek
1

http://greek.io/texts/anabasis.html
http://greek.io/texts/iliad.html

I have been thinking about the problem of good electronic reading versions of Greek texts for a little while, and this is my first pass. Please post any issues or suggestions.

One thing that I’d like to do, hopefully in the next couple of weekends, is to mark all of the long vowels, digammas, and instances of correption or synizesis in the Iliad text. I can’t find any available versions with those markings, and it would make it much easier to read smoothly.

2

Looks good. I’d suggest you leave this version of the Iliad as is and if you want to add prosodical markings to do that in a separate version. Then people can take their choice, or work from one to the other, since presumably the goal is to learn to read properly without such crutches.

3

I’m pleased to hear that from you, Michael! I think that having multiple versions is a very good idea. Especially since I have also been thinking about experimenting with stress markings, and there are multiple suggestions for how systemic syllable stress might have existed in Ancient Greek (if it did at all).

All of these markings are a distraction for reading, but they can be useful for reading aloud, especially the vowel lengths of α, ι, and υ, until you learn the vocabulary.

And, of course, for those who need no crutches:

ΜΗΝΙΝΑΕΙΔΕΘΕΑΠΗΛΗΙΑΔΕΩΑΧΙΛΗΟΣ…

(That is a joke, but I there’s probably no reason not to include it.)

4

Well Joel you know I’m a papyrologist, so ΜΗΝΙΝΑΕΙΔΕΘΕΑΠΗΛΗΙΑΔΕΩΑΧΙΛΗΟΣ… is how I’m used to reading. :slight_smile: Only with most of the letters missing. :smiley:

If you think you can combine syllabic stress with metrical reading, good luck to you! If you put dynamic stress on the accented syllables you’ll impair the rhythm to the point of unrecognizability. If you’re capable of it (which I’m not, and very few English-speakers are), by all means try for unstressed pitch accents; but do make sure you have the rhythm down first, to which the accents are quite irrelevant. (Astonishing but true.)

If you’re seduced by Allen’s idiosyncratic thesis in his Accent and Rhythm, I recommend reading Martin West’s review of it. I’d give you the reference if I had it, which unfortunately I don’t.

Forgive so many ifs!

Michael

5

Remember the cow turns both ways! Papyrologists are very…idiosyncratic I must say, we had Dirk Obbink teach us and it was an experience. Saying that, much needed skill and often results in phenomenal work. Ha, I remember when we met Parsons, I think we were all like: :astonished:

The reference is Gnomon, 1976 (48) for the West review. See also Devine and Stephens - I know them from their work on Latin but not read the article yet (its in TAPA, 1985, 115).

Anyway, just to say, I agree. It’s best not to mark down the metre because then you start following a script rather than develop a dynamic understanding. The trick is to nail down the rules and repeatedly try - and fail- until you get it. Pick a book and just go at it.

With any experimentation watch out lest you sound like: http://www.rhapsodes.fll.vt.edu/iliad1.htm

I really want to record some stuff myself but I’ve neither the equipment/programme nor the time. Saying that I did a jokey “three lessons of conversational Greek” MP3 series a while ago (jokey as in full of verbs like binein and offensive compound nouns etc, phrases you’d never want to say to someone) and it sounded ok despite using Windows sound recorder and a Microsoft lifestyle mic.

6

Thanks for pointing us to that review of West’s. I confess I didn’t understand everything, partly because I haven’t read the book in question, partly because I’m a hopeless dilettante. I have read Allen’s Vox Graeca; I wonder if he has actually reconsidered some of his views for the newer editions, since many of the views West criticizes him for don’t ring any bell for me (or maybe I just don’t remember).

If Allen’s view is indeed that “heavy final syllables are stressed”, that seems very strange, as does the idea that the stress in Ἀχαιῶν χαλκοχιτώνων falls on ῶν .. κοχι .. νων. My main objection to this is that this makes the stress fall on the thesis and not the arsis, but is that West’s objection as well?

I’m not completely sure what are the implications when West writes “This is dangerously near to saying ‘the stress shifts to suit the metre’.” The way I see it, it is precisely the fact that ancient Greek had a pitch accent where stress plays little role that allows poetry, in a way, to deal quite freely with the stress, to rhythmically accentuate the words in different ways as long as the basic rules of syllable length are respected. The way I’ve come to imagine, in poetry (or at least in hexameter) the (natural) pitch accent and the stress act independently of each other, so that each syllable in arsis receives some stress, but this “stress” would not be necessarily found there in everyday speech, or it would not be important feature in it. As it is possible to have the same word accommodated in the meter in different ways and make the arsis fall on different syllables, I would say that indeed “the stress shifts to suit the metre” – but that would not be the stress of natural speech.

Do you see a problem with my reasoning? Probably part of the problem is that it depends on what exactly you mean by “stress”.

7

Here is my current reading of the Iliad, which I am trying to fix:

http://greek.io/audio/Iliad%201.33-42.mp3

I have marked vowel quantity in the text I’m reading from, but not meter. I had hoped that it would be more emergent (as with my Babrius recordings, which I do the same way: http://greek.io/).

I think that pronunciation and quantity and pitch are coming along (feel free to disagree, I have found everyone’s advice very helpful). But I’m stomping all over stress and meter. My feeling for Greek poetry is that the stress should sound something more Robinson Jeffers than Milton, not quite railroaded by the meter. But perhaps I should just mark all the ictus in the lines, and attempt to read. In fact I will try that now:

(It comes across very poor, but that is likely lack of practice): http://greek.io/audio/Iliad%201.33-42%20ictus.mp3

Regardless, thank you for the suggestion of the West review. I found it on JSTOR and able to read it last night.

The question at issue strikes me as solvable: Was there a weak stress accent in ancient Greek that poets attempted to align to their meter. In principle, this is something that could be evaluated with analysis.

I have seen several proposals for how that might work. C.W.E. Miller’s Pronunciation of Greek and Latin Prose is a very good read, though is not as linguistically informed as Allen. His three rules are found on pg. 183. There is some agreement between Miller and Allen, and some disagreement. I have Devine and Stephens on my shelf, and it has been recommended to me my multiple people, but I have not been able to bring myself to read it yet.

However’s West’s criticisms strike me as exactly where it would be possible to go wrong in this sort of analysis. Unless you carefully control for where words are likely to be placed absent the boundaries of meter, and unless you carefully control for where quantity makes it likely for them to be placed in meter, your analysis will go wrong.

I suppose that I will have to write some code eventually, to really find out who is right. But I think a better first step for me is to carefully evaluate the proposals of those who have already looked at the question and to do some readings based on their schemes.

8

Stephen Daitz’s recordings of Homer come with PDF’s of scanned images with macrons and synizesis marked. For digammas and other things that most people do not distinguish, there isn’t always even a standard notation. I’ve been reading Odyssey 6 and 9, printing out Geoffrey Steadman’s translation sheets and marking them for the main caesura, as well as any syllable that I have to puzzle out whether it’s short, long by position or long by nature. I also mark my idiosyncratic pronunciation habits, such as a three-way distinction in the pronunciation of -σσ-. The online sample of Daitz’s reading of the opening of the Iliad rubbed me the wrong way (as it has a lot of people), but I’m actually finding his Odyssey listenable.

For the Anabasis Book 1, the presence of macrons is a major reason why I’ve been making recordings of the adapted passages in White’s First Greek Book and only attempting silent reading of the original.

I’ve read both Allen’s Vox Graeca and Accent and Rhythm, and I did not find his proposed stress rule workable. There’s too much working back from the end of a word or phrase to make it plausible that people might apply such a rule when speaking. My own stress patterns when reading Greek are largely inspired by German prosody. The general principle I follow is that any individual morpheme can be stressed or unstressed, and generally should be to the extent that logic dictates, but there may be rules governing certain kinds of constructions (like the German rule that for prefix verbs, separable prefixes are stressed but inseparable prefixes are unstressed). In rhythm, I follow Daitz’s pattern: fairly strict rhythm, with no breaks in the middle of a line. A short syllable that rhetorically needs stress gives a nice syncopated effect to the pattern, and with tone provides the main source of variety in the meter.

9

Just 2 quick things before I really have to leave.

Don’t think of syllables as long by position or by nature. They’re either short or long (light or heavy, to avoid confusion with vowels), that’s all.

Linguists do work back from the end, and quite right too, though it can be disconcerting. That’s how accents work, after all.

And do read D&S. (That’s 3.)

10

Well, I need some notation for a short vowel in a closed syllable. You can object to the term “long by position”, but it wouldn’t make sense to think about it the same as a long vowel when reading aloud.

The Greek accent rule can be applied by going forward with a two-syllable look-ahead. When you apply Allen’s stress rule, you can only stress a heavy syllable or a pair of light ones if the following syllable isn’t stressed, so to figure that out you have to look recursively to the syllables following until you get to the end. It’s a look-ahead of an indefinite number of syllables, which I find awkward. Given that the rule puts stresses on many endings and relatively few roots, I don’t see what communicative value it would have. I would rather read with no stresses at all than with that rule.

I have read D&S, and it has become my primary guide in trying to realize the Greek tone accent. At any rate, for Joel and anybody else interested in reconstructing the Greek tone accent, I second your recommendation to read it. I don’t think the material in Allen’s books is a sufficient guide for putting together a satisfactory tone accent, and I think Allen would agree with that.

@jeidsath

Your pronunciation is definitely coming along. When reading prose, you put pauses between most words, which is fine, and I don’t think it’s necessary to go into greater depth into prosody theory as long as you’re happy doing it that way.

@Scribo

Making recordings really doesn’t take special equipment/software. I make mine on my iPhone with its built-in “Voice Memos” app, and I think the sound quality is acceptable. Half the time I just sit in the car with my phone and iPad or a book when I have half an hour free.

11

New recording, where I think that I’ve put some of the advice that I’ve been getting into practice. The meter is finally coming across, I think. The secret seems to have been putting the stress on the ictus.

http://greek.io/audio/Iliad%201.1-7%20momentum%2010.20.14.mp3

To make this, I had to work through the text enough times that I got every ictus by heart, so I’m not sure that a marked version would be useful for me for reading aloud.

12

hi, i agree you don’t need to mark the longs and shorts once you can do it naturally. really in each homeric line you’re not mentally scanning every single syllable, but instead you only need to mentally scan 5 syllables, no matter how many syllables in the line there are (and once you’ve scanned these syllables, the rest are determined) - just scan the 2nd syll, then (if the 2nd syll is heavy) the 4th syll or (if the 2nd syll is light) the 5th syll, and so on. once you have these the scansion of the rest of the line is worked out automatically.

nevertheless for people getting used to scansion, having a marked text works, i did this for iliad books 1 and 2 a long time ago, eg my doc here on iliad 2 http://www.freewebs.com/mhninaeide/IliadBScannedWestText2006.pdf

in terms of pronunciation, i only use pitch and not any conscious patterning for stress at all - i don’t try to “destress” syllables, but just don’t think about stress at all and only focus on pitch. i follow d&s, i summarised the main pattern in para 7 in that doc linked above. words like ictus, arsis, thesis i ignore and just follow the metrics and pitch markings. just my approach, i have nothing against the other pronunciation approaches. cheers, chad

13

For mwh:

http://greek.io/texts/iliad_uc.html

14

Nice, Joel! I would certainly consider it worthwhile to listen repeatedly to several hundred lines of Homer read like that.

I’ve finally made some recordings of unadapted Anabasis Book 1.
https://archive.org/details/anabasis_ariphron

My reading style is gradually improving, and I think it has reached the next level: faster, with attention to how the pitch peaks of different phrases relate to each other in a sentence, and with D&S-inspired shortening (“subordination”) of certain long vowels based on foot structure, so that my reading is no longer filled with so many syllables held unnaturally long. My recording is read from the Steadman PDF, which reprints the E. C. Marchant edition line-for-line. The glosses were more useful than macrons would have been. Hopefully by now I recognize most of the places in Anabasis A where long vowels appear. My recordings cover chapters 3 and 6: a mutiny which Clearchus handles with great political skill, and the trial of Orontas. Chapter 3 averages 7.53 s/line; on chapter 6 I pushed myself to go a bit faster, and averaged 6.74 s/line. For comparison, Joel’s July 10 recording of chapter 1 averages 6.51 s/line, and his September 25 one, which has more space between words, 8.34 s/line.

15

More audio, as usual “This is my best yet!” until I am completely dissatisfied with it next week.

http://greek.io/audio/Iliad%20Song.mp3

And a YouTube video on how to recite the Iliad. This video is mostly English. Please send me any corrections before I mislead too much of the internet.

https://www.youtube.com/watch?v=nA97pQh-4N0

16

I much prefer your previous version, the one posted on October 21st. Now you’re trying to do something, I’m exactly sure what it is, but it doesn’t work. Or that’s how I feel about it. :wink:

The one you posted on October 21st I like a lot. I think that’s the way you should go. Of course, it may be that what I like is based on the wrong theoretical assumptions, but well… I think that version sounded really natural.

Anyway, bearing in mind that it’s always easier to point out out weaknesses than strengths, a couple of remarks on the 21st October version (which I think is clearly better than the previous one I commented):

μῆνιν ἄειδε θεjὰ Πηληϊάδεω Ἀχιλῆος
οὐλομένην, ἣ μυρί᾽ Ἀχαιοῖς ἄλγε᾽ ἔθηκε,
πολλὰς δ᾽ ἰφθίμους ψυχὰς Ἄϊδι προΐαψεν
ἡρώjων, αὐτοὺς δὲ ἑλώρια τεῦχε (you say tyoukhe, ευ’s a difficult one!) κύνεσσιν (s should be long kynessssin!)
οἰωνοῖσί τε πᾶσι,

Now you’re pronouncing and extra /j/ sound between vowels that are in different syllables (i.e., in hiatus). It happens in θεὰ and ἡρώων, and it’s especially problematic in θεὰ, as it makes the ε long.

17

Allen’s book actually recommends adding that kind of glide: /j/, /w/, or /ɥ/, depending on the preceding vowel. HIs recommended pronunciation doubles the glide in a diphthong followed by vowel (giving θεια with /εjja/), so an induced glide in θεὰ is still distinguishable as short (/εja/). Personally I find the induced glides far preferable to putting glottal stops in the hiatus positions. My viewpoint is that the orthography does not indicate semivowels because they are rarely if ever distinctive and because doing so would break the convention for distinguishing long and short syllables. The rough breathing, of course, wasn’t indicated either, but that doesn’t mean it wasn’t pronounced. In all of my recordings, I consistently add glides in most hiatus positions.

18

I’m not sure I understand. If you mean Allen’s Vox Graece, I could find this statement on p. 96 about hiatus: “…where the first of the two vowels was of close or mid quality it was followed by a semivocalic [j], [w], or [ɥ] transitional glide (in the case of front, back, and front rounded vowels respectively).” (I’m not using the same letters for the sounds as Allen here).

To me this seems logical, and I would do it automatically myself as long as the first vowel is closed, that is with /i/, /u/ or /y/, and Allen also would have us do it with mid vowels, like ὃ(w)ἔγνω. He doesn’t give an example with an ε. Anyway, as far as I can see, Allen seems to be talking only about hiatus between words. But I supposed it also depends on how closed you pronounce your ε, ο, or ω.

But if we a transitional glide in word-internal hiatus, shouldn’t we have ἡρώ(w)ων instead of ἡρώ(j)ων? In the recording θεὰ was more like θεjjὰ to me – the transitional glide was clearly too long. For me, even θεjὰ is wrong, although I could accept ἡρώwων - but that is really a matter taste :laughing:

19

First of all, thank you Paul and ariphron!

The “y” or “j” was a temporary approach that I was trying last week as I did the recordings, and yes I was attempting to follow some of Allen’s advice on diphthong to vowel-junctures (however, expanding it to some vowel-vowel junctures). I have since stopped, retaining it now only for diphthongs.

I am still up in the air on reading methods. I think that Daitz may be wrong in his 1991 paper “On reading Homer Aloud: To Pause or Not to Pause.” See also http://bmcr.brynmawr.edu/1991/02.07.09.html.

Does it, like Shakespeare, sound better unperformed? Perhaps (although there is no way to find out from my recordings). The Iliad was popular song before it was every popular poetry. And the Greeks used to paint their statues.

The big question to me is “How do I properly account for caesura?” The singsong is my first solution to that. If I have time, I will do a recording where I don’t read the Greek words as separate at all, and only go by length. Caesura doesn’t matter, of course, and it sounds neat, but I’m not sure that its the way forward for me. Hopefully that will demonstrate to you what I’m trying to flee from.

20

Thanks, Paul, for finding the reference. I didn’t have a copy on hand, and was writing from memory. IIRC, there may also have been some discussion of the matter in Accent and Rhythm.

ἡρώwων is indeed the way I pronounce it. ώjω would be ῴω. I also like θεjὰ, because then I hear identical consonants in the related words “goddess” and “divine”.

On the matter of pausing, my feeling is that Daitz’s position is a better starting point than the alternatives. Certainly many of the places where modern editors put punctuation are bad places to pause. It’s easier to learn a passage with no pauses in a line and later add a few pauses for dramatic effect than to learn a passage with pauses and then run the phrases together on a trial basis, so the no-break rule is really the more flexible approach. What I’ve tried doing is holding the long syllable before the caesura longer than the other long syllables, and my feeling is that this kind of pause works just as well as breaks between words in 90% of cases, with less disruption of the rhythmic scheme.

In any case, I recommend moving forward. I will never sit down and listen to five Joel recordings of Iliad Lines 1-7, but I look forward to hearing your rendition of the complete Book One, even if it isn’t perfectly consistent in some small details.