ὑπεκδύομαι -> ὑπεκδύς How?!

[daivid]

This is from Herodotos
ὡς δὲ κατὰ νώτου ἐγένετο ἰούσης τῆς γυναικός ἐς τὴν κοίτην, ὑπεκδὺς ἐχώρεε ἔξω, καὶ ἡ γυνὴ ἐπορᾷ μιν ἐξιόντα.

I know that ὑπεκδύς is the aorist participle of ὑπεκδύομαι but only because Steadman tells me so.
It doesn't look like a participle to me.
So why does the participle look so odd and are there other verbs that behave in the same way?

Or am I missing something that ought to be obvious?

[Scribo]

You're missing something sure but I'd hardly say it ought to be obvious. Yes it is a participle and it's exactly the same as some other aorist participles e.g φύω > ἔ-φυν > φύς, φῦσα, φύν. You should recognise say, erm, στάς < ἵστημι or βάς from βαίνω as common examples of a similar phenomenon.

Basically, its because its formed from a different type of aorist. I tend to refer to it as "strong" but some textbooks and grammars use the term "second". Its basically the ending smacked onto the stem. No thematic vowels or anything just right up in there.

Does that help?

[Qimmik]

Confirming Scribo's explanation, here's a link to Smyth sec. 418:

http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%3A1999.04.0007%3Asmythp%3D418

It's just like the second aorists of athematic (-μι) verbs:

στάς, στᾶσα, στάν
θείς, θεῖσα, θέν
δούς, δοῦσα, δόν

The endings of these athematic verbs are added directly to the root, which ends in a vowel. The vowel is lengthened in the masculine and feminine aorist (and present) participle due to "compensatory lengthening," i.e., the vowel is lengthened as "compensation" for the loss of ν before σ.

The root of δύω/δύομαι is of course δυ-. The second aorist of this verb and its compounds generally behaves like an athematic verb, adding the inflectional endings directly to the root--although, as with all too many Greek verbs, LSJ reports a bewildering proliferation of forms.

http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%3A1999.04.0057%3Aentry%3Ddu%2Fw2

[daivid]

You're missing something sure but I'd hardly say it ought to be obvious. Yes it is a participle and it's exactly the same as some other aorist participles e.g φύω > ἔ-φυν > φύς, φῦσα, φύν. You should recognise say, erm, στάς < ἵστημι or βάς from βαίνω as common examples of a similar phenomenon.

Basically, its because its formed from a different type of aorist. I tend to refer to it as "strong" but some textbooks and grammars use the term "second". Its basically the ending smacked onto the stem. No thematic vowels or anything just right up in there.

Does that help?

A root aorist! Yes that helps a lot. So the acc is φύντα and so on.
Hence for ὑπεκδύομαι the acc is ὑπεκδυντα?

I have met root aorists but because I only meet them now and then I sort of forget about them.
(I assume that calling them"second" aorist was a deliberate mistake to keep me on my toes because root aorists seem more like 1st aorists to me)
EDIT
Scratch that - I see that root aorists are a form of 2nd aorist.

I won't say everything is clear but at least now I know what I don't know.


Thanks very much.

[Qimmik]

Another feature of second aorist participles and infinitives (both root and thematic) is that the accent is on the ending--they're not recessive--like perfect participles and infinitives, but unlike most other verb forms.

[daivid]

Confirming Scribo's explanation, here's a link to Smyth sec. 418:

http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%3A1999.04.0007%3Asmythp%3D418

I'm not quite sure if I have got this right. Is δεικνύς a present participle? Has it basically got the same form of ὑπεκδύς which is an aorist participle.

I guess I should expect that from a second aorist but when the form of the aorist looks so distinct from normal participles it is that much more disturbing than when more common aorist participles look like present participles.

The root of δύω/δύομαι is of course δυ-. The second aorist of this verb and its compounds generally behaves like an athematic verb, adding the inflectional endings directly to the root--although, as with all too many Greek verbs, LSJ reports a bewildering proliferation of forms.

"bewildering proliferation" - you can say that again.

Thanks for pointing out that it is even more confusing than I suspected. Forewarned is forearmed so the help is appreciated.

[Qimmik]

Is δεικνύς a present participle? Has it basically got the same form of ὑπεκδύς which is an aorist participle.

Yes, the form δεικνύς is a present participle, but has the same form as δύς. Scroll down to the bottom and you'll see δύς, the aorist participle of δύω, subsumed under the same paradigm.

δύω has a thematic present stem, but an athematic second aorist stem.

δείκνυμι, in contrast, has an athematic present stem, formed by adding the suffix -νυ- -- which often takes athematic endings -- to the root δεικ-, but has a sigmatic first aorist, ἔδειξα. There are other verbs that follow this pattern: present stem in -νυμι and sigmatic aorist.

Many Greek verbs mix and match stems like this. You might look at Smyth

The athematic verbs [σ]ἵστημι, τίθημι and δίδωμι distinguish the present and aorist stems by reduplication (with the vowel -ι-, as opposed to reduplicated perfects with the vowel -ε-). Apart from the reduplication, the present and aorist participles and infinitives of these verbs are alike.

* * *

Some people feel this sort of information is useless, noting that native speakers of ancient Greek would not have been able to explain the underlying morphological patterns of these forms--they would have internalized the forms and would have been able to spit them out automatically in conversation and recognize them in speech or writing without thinking. Personally, however, I have to admit that I'm not a native speaker of ancient Greek, and I find this sort of information about the regularities underlying the irregularities very useful in reading ancient Greek texts. In addition to its intrinsic interest for me as a student of the Greek language, this information helps me recognize strange forms when I encounter them in writing, and it's comforting to know that there are indeed patterns that can explain the apparent irregularities.

The print, hard-bound version of Smyth published by the Harvard University Press (2d ed., revised by Gordon M. Messing and published in 1956; my copy is dated 1959) has a very thorough discussion of the patterns of Greek verbs that makes some sense out of what seems like random irregularities. The on-line version of Smith (as well as some of the editions offered for sale on-line) is based on an earlier edition that is less helpful on these points. The sections on the verbs (106-224) are rough going, but if you can make it through--if only by skimming--you will have a much better grasp of Greek verbal morphology, and you won't find it so perplexing.

The Herodotus passage is from the story of Gyges, isn't it? Candaules got what he deserved--he should never have done that to his wife. This is one of the stories that makes Herodotus so entertaining!

[Qimmik]

Just to amplify my previous posts (Scribo, correct me where I go astray):

Greek verbs exhibit two broad classes of present stems. Thematic present stems insert the vowels ε or ο between the stem and the personal endings. Athematic present stems (the -μι verbs) attach the personal endings directly to the stem, which ends in a vowel.

The athematic stems are mostly very old, basic words such as ἵστημι, but some verbs with suffix -νυ- take the athematic personal endings directly, for example, δείκ-νυ-μι, and this class of verbs is somewhat more productive than the other athematic types. However, other verbs with the suffix -νυ- are thematic, e.g., δύ-ω.

There are several classes of aorists. Most common, and productive, are the sigmatic "first aorists": the suffix -σ- and personal endings in -α are added to the root, for example, ἔδειξα = ἔ-δεικ-σ-α.

There are also "second aorists" that are formed by reducing the vowel in the root of the verb and adding inserting the thematic vowels ε or ο before the personal endings, for example, λείπ-ω/(ἔ)λιπ-ο-ν. The present root λειπ- contrasts with the aorist root with reduced vowel λιπ-. (The phenomenon of root vowel change is similar and in fact etymologically related to the English process at work in "irregular" or "strong" verbs that mark present, preterite and past participle by vowel mutations--sing, sang, sung.) This class of aorists is smaller and less productive than the sigmatic aorists.

Finally, there is a very limited class of athematic aorists that stick personal endings directly onto the root, which ends in a vowel. (ἔ)δυ-ν, with its participle δύ-ς,falls into this class.

It's worth noting that the class into which the present of a particular Greek verbal root falls doesn't necessarily determine which class the aorist -- or the perfect -- will fall. That's what I mean by "mix and match." And many verbs have alternative presents and aorists.

The ability to recognize the various classes of presents, aorists and perfects -- along with some knowledge of the phonological and spelling rules that come into play when Greek verb forms are cobbled together out of augment, roots, suffixes, thematic vowels, personal endings etc. -- makes it easier to identify strange verb forms when you encounter them, and allows you to see the patterns and consistencies that underlie the irregularities.

I hope this brings some clarity and doesn't make things more confusing.

[daivid]

Some people feel this sort of information is useless, noting that native speakers of ancient Greek would not have been able to explain the underlying morphological patterns of these forms--they would have internalized the forms and would have been able to spit them out automatically in conversation and recognize them in speech or writing without thinking.

I am sure that Classical Athens children did not learn to speak by having Sophicles read to them.
I do think that trying to absorb this kind of information raw is going to fail. Unless you encounter them in the form of actual language anything read in a grammar book gets quickly forgotten. The catch 22 is that though these forms are individually rare there are so many of them that you encounter one or other quite often so reading gets slowed down.

Personally, however, I have to admit that I'm not a native speaker of ancient Greek, and I find this sort of information about the regularities underlying the irregularities very useful in reading ancient Greek texts. In addition to its intrinsic interest for me as a student of the Greek language, this information helps me recognize strange forms when I encounter them in writing, and it's comforting to know that there are indeed patterns that can explain the apparent irregularities.

The print, hard-bound version of Smyth published by the Harvard University Press (2d ed., revised by Gordon M. Messing and published in 1956; my copy is dated 1959) has a very thorough discussion of the patterns of Greek verbs that makes some sense out of what seems like random irregularities. The on-line version of Smith (as well as some of the editions offered for sale on-line) is based on an earlier edition that is less helpful on these points. The sections on the verbs (106-224) are rough going, but if you can make it through--if only by skimming--you will have a much better grasp of Greek verbal morphology, and you won't find it so perplexing.

Maybe its time for me to think about getting my own copy.

I commented on your comment on Gyges here:
http://www.textkit.com/greek-latin-forum/viewtopic.php?f=36&t=60910

[Paul Derouda]

Yes, the form δεικνύς is a present participle, but has the same form as δύς.

Why is this present participle accented like this? This is unfair. So totally, utterly unfair!

[Qimmik]

From Smyth 425a:

b. Participles.—(1) Oxytone: the masculine and neuter sing. of the second aorist active, as λιπών, λιπόν; and of all participles of the third declension ending in -ς in the masculine (except the first aorist active), as λυθείς λυθέν, λελυκώς λελυκός, ἑστώς ἑστός, τιθείς τιθέν, διδούς διδόν, ἱστά_ς ἱστάν, δεικνύ_ς δεικνύν (but λύ_σα_ς, ποιήσα_ς). Also ἰών going from εἶμι.

This doesn't tell you why, but it shows that the oxytone accent on δεικνύς is not an isolated anomaly.

[Paul Derouda]

Well, I sort of knew about the perfect participles and second aorists. διδούς, τιθείς, ἱστάς I also knew about, although I had vaguely thought that the present reduplication was playing some sort of trick here. δεικνύς is weird. But then I guess it's just incidental that there are not many present third declension participles ending in -ς, just those couple of well known, common irregular -mi verbs and then this δεικνύς...

[Qimmik]

δείκνυμι is a -μι verb, too, of course. And it's not alone--there are other -νυμι verbs:

http://www.perseus.tufts.edu/hopper/text?doc=Perseus%3Atext%3A1999.04.0007%3Apart%3D2%3Achapter%3D19%3Asection%3D55%3Asubsection%3D55

[Qimmik]

I am sure that Classical Athens children did not learn to speak by having Sophocles read to them. . . . The catch 22 is that though these forms are individually rare there are so many of them that you encounter one or other quite often so reading gets slowed down.

Actually, the most common verbs tend to be the most anomalous. The vestigial -μι conjugation, for example, complete with its oxytone participles, includes ἵστημι, τίθημι and δίδωμι. And there are many compounds of these verbs, and many of the compounds are very common, too. You couldn't talk about getting dressed in the morning or undressing at night without knowing the athematic ("root") aorist forms of δύω/δύομαι. You couldn't report what someone said without knowing the forms of φημί. So kids would have been hearing these forms, and internalizing them, from earliest infancy--that's why they were so resistant to the process of analogical levelling--i.e., being assimilated to more regular conjugations. (The same is true in English: strong verbs like go/went/gone, come/came/come, run/ran/run, and irregular weak verbs like buy/bought/bought, think/thought/thought, are typically very common verbs, and they've resisted being assimilated to the regular weak verb pattern--go/goed/goed, etc.--because kids learn them in infancy.)

I do think that trying to absorb this kind of information raw is going to fail. Unless you encounter them in the form of actual language anything read in a grammar book gets quickly forgotten.

The idea is not to learn all the forms of every Greek verb--you'd go insane trying to do that--but rather to get the big picture--an overview of the various types and patterns of verb form, and the regularities lurking behind the irregularities, in more depth than the basic first-year texts provide (as well as the phonological and spelling rules that come into play when the various morphemes are linked together). This will help you sort out the weird forms you encounter in reading--you won't be stymied when you see ὑπεκδὺς. You will also be more likely to remember irregular forms once you've seen them if you understand how they originated and can place them in larger patterns of regularities.

[daivid]

I am sure that Classical Athens children did not learn to speak by having Sophocles read to them. . . . The catch 22 is that though these forms are individually rare there are so many of them that you encounter one or other quite often so reading gets slowed down.

Actually, the most common verbs tend to be the most anomalous. The vestigial -μι conjugation, for example, complete with its oxytone participles, includes ἵστημι, τίθημι and δίδωμι. And there are many compounds of these verbs, and many of the compounds are very common, too. You couldn't talk about getting dressed in the morning or undressing at night without knowing the athematic ("root") aorist forms of δύω/δύομαι. You couldn't report what someone said without knowing the forms of φημί. So kids would have been hearing these forms, and internalizing them, from earliest infancy--that's why they were so resistant to the process of analogical levelling--i.e., being assimilated to more regular conjugations. (The same is true in English: strong verbs like go/went/gone, come/came/come, run/ran/run, and irregular weak verbs like buy/bought/bought, think/thought/thought, are typically very common verbs, and they've resisted being assimilated to the regular weak verb pattern--go/goed/goed, etc.--because kids learn them in infancy.)

Sure they will have been encountering irregular verbs from a young age. But those mothers would have used those words in very simple sentences. When they wanted to tell their child to hurry up and get dressed they wouldn't have composed sophisticated Sophocles style verse to do it

And you are right the μι verbs are quite frequent. I have a list of 500 "most essential" words and of the 140 verbs 14 are μι verbs. However leaving aside the ειμι it is still not often I encounter one at the rate I currently read Heroditos.

I do think that trying to absorb this kind of information raw is going to fail. Unless you encounter them in the form of actual language anything read in a grammar book gets quickly forgotten.

The idea is not to learn all the forms of every Greek verb--you'd go insane trying to do that--but rather to get the big picture--an overview of the various types and patterns of verb form, and the regularities lurking behind the irregularities, in more depth than the basic first-year texts provide (as well as the phonological and spelling rules that come into play when the various morphemes are linked together). This will help you sort out the weird forms you encounter in reading--you won't be stymied when you see ὑπεκδὺς. You will also be more likely to remember irregular forms once you've seen them if you understand how they originated and can place them in larger patterns of regularities.

Getting the the big picture is what I find hardest and to the extent that I have it is through learning the individual forms. The "big picture" is too abstract for me to stick.