Wiktionary talk:About Ancient Greek/Archive 1

Transliteration section needed

This page really, really needs to specify the system for transliteration of words in to Latin characters. I know that (somewhere) a page has been started to deal with Modern Greek, and it would be nice if the systems were coordinated a bit. We need some standard to rely on for creating, checking, and correcting transliterations. --EncycloPetey 16:29, 28 January 2007 (UTC)

Well, here's a list of the transliteration standards that I've been operating under so far, α=a β=b γ=g δ=d ε=e ζ=z η=ā θ=th ι=i κ=k λ=l μ=m ν=n ξ=x ο=o π=p ρ=r σς=s τ=t υ=u φ=ph χ=ch ψ=ps ω=ō. I've been ignoring accents and iota subscripts (while they were certainly pronounced in old Attic, it is my understanding that they were largely forgotten footnotes in Koine and were on the road to obscurity well before that). Certainly there's plenty of room for a non-Greek speaker to interpret these incorrectly (especially with the vowels), so I'm open to suggestions. Modern Greek would probably have a different Romanization standard (I would presume), with δ=th and most dipthongs=i, among other things. Is that a problem? I suppose it might be confusing to new users, but it might also show a lot of people the distinction between various pronunciations. Certainly we'll also have IPA's, once I figure out a system for that. Saltmarsh is working on an about Greek page, but I don't believe it has Romanization just yet. Cerealkiller13 20:27, 28 January 2007 (UTC)

I think I would have used ē for eta, though it's a matter of personal prefernce until we establish a definite system. And some people would argue for using y for upsilon. --EncycloPetey 21:19, 28 January 2007 (UTC)

I would be ok with ē for eta, but I would still prefer ā, as eta is more closely associated with alpha than epsilon and I think it will cause confusion with i for iota. I have serious reservations about y for upsilon. If a whole bunch of people are adamant about it, I'll concede, but I think u captures the basic English equivalent much better than y, although it is admitted that neither exactly match it.

I've added a table for pronunciation and Romanization standards. The Romanization is taken from ISO 843. Hopefully it will work for both modern and Ancient Greek. I took the pronuncation from a number of sources, most notably from Muke's Greek subpage, along with a few others. It's lacking a few Koine diphthongs which I couldn't find, so if anyone knows them, it would be appreciated. Also, if anyone has any problems with any of the pronunciation, feel free to discuss it. Accentuation stuff will be coming shortly. Cerealkiller13 01:26, 30 January 2007 (UTC)

For the IPA there are a few others to note, such as γγ pronounced /ŋ/. There are a couple of others that probably should be included but these are not coming to mind at the moment. As for the Koine diphthongs, have you looked in Horrocks? In addition to his description of Classical pronunciation at the outset of the book, there is a discussion of phonological shift in the Koine (with IPA) beginning on p 102. --EncycloPetey 03:41, 30 January 2007 (UTC)

The second sound of gamma is noted in the note, on the right side. Sadly I don't have a copy of Horrocks, I've never heard of him. Cerealkiller13 03:53, 30 January 2007 (UTC)

Really? You definitely should pick up a copy of his book Greek: A History of the Language and its Speakers by Goeffrey Horrocks. →ISBN Sadly, it's out of print, but you might be able to find a used or remaindered copy. He's a leading authority on the hisotry of the Greek language. --EncycloPetey 01:04, 31 January 2007 (UTC)

Thanks for the tip. That is not an easy book to find. I now have a copy (hopefully) coming my way. Cerealkiller13 04:17, 31 January 2007 (UTC)

It was a bit easier for me, since I find out about the book back when it was still in print. --EncycloPetey 04:35, 31 January 2007 (UTC)

I just looked at ISO 843 and it would be completely inappropriate for Ancient Greek β = v, for example. Also, ē for η is so common as to be a de-facto standard; Δημήτηρ is commonly called Demeter in English; transcribing it Dēmētēr would be perfectly normal; transcribing it Dāmātār would just be weird. (And anyway, while many instances of η come from older ā, not all do.) Angr 17:18, 2 February 2007 (UTC)

It admittedly is rather incorrect for Ancient β to be transcribed as a v. However, it should be noted that we're trying to come up with a Romanization system which will do both Ancient and Modern Greek together, and certain sacrifices need to be made. To be honest, I think Modern is making more concessions here than Ancient, especially with vowels. Cerealkiller13 17:27, 2 February 2007 (UTC)

I don't think we should attempt to have a single transliteration system for both Modern and Ancient Greek; they're just too different. Likewise, we shouldn't have a single transliteration system for every language written with the Cyrillic alphabet, or for every language with Devanagari (IAST works beautifully for Sanskrit but would be quite inappropriate for Hindi, Marathi, and Nepali). Ancient Greek and Modern Greek are for all intents and purposes separate languages and should be transliterated separately. Angr 20:19, 2 February 2007 (UTC)

It is quite true that they are rather different from each other. However, it should be kept in mind that, for better or for worse, Ancient Greek on Wiktionary includes Koine (for which transliterating a β as a v isn't too far off). To tell the truth, I'm rather ambivalent about the exact specifications of the Romanization system as long as we come up with one standard for transliterating Ancient Greek on Wiktionary. There does not seem to be an agreed upon "standard" system for us to use, and so I think Wiktionary can take a fair bit of liberty (within reason, of course) on this. Also, the pronunciation will be made more specific in the pronunciation section, which will have separate IPA's for both Classical and Koine. The advantage of having one system for both languages is that it keeps a sense of continuity. Whatever differences there may in fact be, Greek has changed very little considering the span of time it's been in existence. The advantage of separate systems is that each might be a bit more accurate for its respective language. Does anyone else have any strong opinions on the subject? Cerealkiller13 21:29, 2 February 2007 (UTC)

Since transcription and pronunciation are separate issues, I would prefer to see a single transcription system used. The goal of transcription is merely to make it possible for someone unfamiliar with the little squiggles to examine spellings. Having a single transcription system makes it easy for the squiggle-impaired to compare spellings between Ancient and Modern Greek. Having different transcription systems would hinder this. However, my opinion is not a strong one and a good argument for using two distinct transcription systems could sway my opinion. --EncycloPetey 23:06, 2 February 2007 (UTC)

Another reason to keep the transcription of Ancient Greek and Modern Greek separate is the fact that Ancient Greek uses diacritics that Modern Greek doesn't. For Ancient Greek we will want to show the difference between the rough breathing and the smooth breathing, we will want to show iota subscript, and we will want to show the difference between the acute, the grave, and the circumflex. For Modern Greek none of that will be necessary. Having one single romanization for both Ancient and Modern Greek is just asking for trouble; should we try to force it on the Arvanitic alphabet alphabet too? Angr 10:39, 3 February 2007 (UTC)

Rough breathing is already done with the h. I don't think that we really want to have the accentuation in the Romanization. The Romanization, by its very nature is supposed to be quick and dirty. It can do without the accents (especially tonal accents. How in the hell would you put that in a reasonable Romanization?). I'm kind of conflicted about the iota subscripts. Certainly they were relevant in older Classical pronunciation, but they were all but obsolete in Koine, and on their way out well before that. Atelaes 07:33, 6 February 2007 (UTC)

Accentuation

This section would benefit greatly from a series of examples. Also, it would help to have a table showing the various diacritical marks used and giving their English (and Greek) names. --EncycloPetey 16:32, 28 January 2007 (UTC)

I added a big, verbose chunk of prose to the accentuation section. I couldn't quite figure out how to work it into a table. But, the paragraph does look kind of long, long-winded, and uninviting. If someone can pare it down a bit, I think that would be good. Cerealkiller13 05:00, 30 January 2007 (UTC)

Can it be broken down into digestable chunks with sub-headings? —Saltmarsh 06:06, 30 January 2007 (UTC)

Headings

Suggested section headings for use in both Ancient Greek and Greek

1. are the names, order and levels correct?
2. Do we say "Inflection" or use "Declination" and "Conjugation" ?
3. Should modern Greek have "Descendants"?

==Ancient Greek==
===Etymology===
===Romanisation===
===Pronunciation===
===Noun===
====Inflection====
====Descendants====

1. Well, except for Romanisation (I prefer Romanization, but they're both correct) and Inflection, these are all standard headings in standard format, so I don't think we need to discuss them, as they're already covered in the manual of style. The Ancient Greek POS templates put the Romanization on the POS line (bet you wouldn't have guessed that!) instead of under a separate header, and I think that's for the best.
2. We've been using inflection, and I would prefer to continue in such a manner, because it allows for just one non-standard header, instead of two.
3. Modern Greek certainly can have descendants, but I imagine it will have a lot less than Ancient Greek. Greek doesn't get to take credit for all the words that have come into various languages that clearly derived from the Ancient period (often through Latin).

I already discussed most of this with Saltmarsh, but I figured it should be put here as well, so everyone can see. Cerealkiller13 08:38, 30 January 2007 (UTC)

The romanization should just be on the "POS line" as noted. There isn't any official list of L4 headers, but I'd consider Inflection standard, it is used in other places now (700+). It seems reasonable to use Conjugation/Declension/Inflection as appropriate for any given language and POS. (Not "Declination" in any case ;-) See User:Robert Ullmann/L4-5 for an un-official list of L4/5 headers in use. Robert Ullmann 11:52, 30 January 2007 (UTC)

I would include ====Related Terms==== Just before Descendants, and a not explaining that Related means "etymologically related, in the same language" (i.e. other Ancient Greek words sharing the same root), while Descendants refers to words in other languages derived from this Ancient Greek word. Many, many people have been congused about this, so I think it should be described with special care. If Alternatiev Spellings are likely to appear with regularity, then I'd add that header in as well (first one, just before Etymology). --EncycloPetey 01:09, 31 January 2007 (UTC)

I guess I was thinking that any header which has a paragraph in WT:ELE 1. Does not need to be justified or ok'd 2. is already explained there. Once we get the cool little blue policy box, we'll have a link to that, so I think we're covered. If anyone can give a good reason for having such redundancy, I'd be glad to hear it. Anway, the only two headings which have been mentioned so far which aren't there are Inflection and Romanization, which we've already covered, I think. Cerealkiller13 04:25, 31 January 2007 (UTC)

My opinion is to list the headers most likely to be used. Also, Descendants is much more important a header when it comes to Latin and Ancient Greek than it is for English; it would be good to have an example of how to set up that section. I intend to do this on the About Latin page. --EncycloPetey 04:35, 31 January 2007 (UTC)

Is there a special way to set that up? I guess I've been pretty haphazard, just listing the words (linked, of course) alphabetically with the dots (the ones produced by *).

Take a look at how I set up auscultare. --EncycloPetey 04:40, 31 January 2007 (UTC)

Oooh. I like it. I was just going to say that with Latin and Ancient Greek, it would need to be organized in some way, or else it would become one big clusterfuck. Here's a question then, if more than one word derives from a particular language, should it be:

• English: word 1, word 2

or

• English: word 1
• English: word 2

No one has officially decided that yet, so I'd say play around with possibilities and see what works. If one format looks to be significantly better, you can propose it become standard. --EncycloPetey 04:58, 31 January 2007 (UTC)

Also, would it be a good idea to use a collapsable table, like we're now using for translations? I imagine that on some of these words, like Μουσική (soon to be moved to μουσική), there will a veritable plethora of descendants. Cerealkiller13 04:49, 31 January 2007 (UTC)

Eventually, perhaps, but I wouldn't worry about that at this stage. After all, that section comes at the end, so it won't be interfereing with later entries. That was one of the reasons for initiating a special set of collapsible tables for translations. There's a move afoot to create a new set of customaizably collapsible table templates, but I can't begin to guess when sommeone will (successfully) tackle such a hairy issue. I'd just display them all for now. --EncycloPetey 04:57, 31 January 2007 (UTC)

Ok, I added a section on headers. About Ancient Greek is getting more verbose every day. I suppose what it lacks for in concision, it makes up for in preciseness. Cerealkiller13 23:40, 31 January 2007 (UTC)

I think it would look better to have example headers like above, and leave all the === out of the section headers. Maybe I'll just do that and you can see? Also the POS names shouldn't be plurals, it's confusing, since they don't appear that way. Made an edit, use it as you wish ;-) Robert Ullmann 04:01, 1 February 2007 (UTC)

And note that "Inflection" is not a non-standard L4 header: WT:ELE doesn't mention it, but it also doesn't mention Conjugation, Declension, Dictionary notes and a few others that see some use. We don't have a policy list of L4 headers. Robert Ullmann 04:07, 1 February 2007 (UTC)

Biblical Greek

I also added a section on special considerations for biblical Greek, in an attempt to nip any problems in that area in the bud. Please take a look, and see if it sounds reasonable and fair (and if there are any other things that should be included in it). Of course, any other further suggestions are most welcome. Cerealkiller13 23:40, 31 January 2007 (UTC)

I would extend the explanation to note that the Christian New Testament and the Septuagint translation of the Hebrew Bible/Christian Old Testament are in Ancient Greek. We have potential debate just as much for Jewish/Christian bias in the translation of words from the Septuagint. Otherwise this looks like a good start. I might think of some way to massage the text the weekend when I'll have less work stress. --EncycloPetey 01:54, 1 February 2007 (UTC)

Mycenaean.........Greek?

I suppose this is as good a place as any to finally nail down the question of whether Mycenaean should be called Mycenaean Greek, or simply Mycenaean (both in the header and language category). Any takers? I retain my original position that the category should be called Mycenaean Greek. For some odd reason, I sort of think that the header should remain as Mycenaean. I guess I don't really know why; it admittedly seems somewhat contradictory. But there you have it. As a further push for my POV, see [1]. Ultimately, I don't think it makes a huge deal either way, but it should be decided, once and for all. By the way, if anyone thinks this convo should take place on the BP, please feel free to move it there. Cerealkiller13 08:50, 31 January 2007 (UTC)

I think it should be Mycenaean; the category name Category:Mycenaean language and the POS cats (Category:Mycenaean nouns etc) really have to match the header.

However: If you want to use category Category:Mycenaean Greek language etc then the header should match that ... note that the gmy template should be the same, and if you want to sort these with Greek and Ancient Greek in translation tables (rather than under "M") it should be Mycenaean Greek. Otherwise any kind of automatic sort becomes overloaded with special cases, and humans looking at it will not know what to do. (Suppose a trans table has only Greek, Ancient Greek, and Mycenaean; someone wants to add, say, Japanese. How would they know that goes after Mycenaeaen? ;-) Robert Ullmann 09:07, 31 January 2007 (UTC)

Excellent points. Then I change my position to all three (header, language category, and eventually POS categories) should be Mycenaean Greek. Especially if Mycenaean is going to be considered a sub-project of Ancient Greek and alphabetized under G in translation tables along with Greek and Ancient Greek (as I think it should), it should have the Greek appended to it for clarity. Cerealkiller13 20:00, 31 January 2007 (UTC)

Section order and grouping

I have reordered and regrouped the existing sections. I think it works better to have 5 major headers rather than 10. Opinions? --EncycloPetey 01:02, 3 February 2007 (UTC)

I was hoping someone would find a more intuitive way of organizing all of that. Thank you. I think it is a distinct improvement. Cerealkiller13 01:35, 3 February 2007 (UTC)

POS headers

In the list of POS headers I see two notable omissions, but don't know enough about how Ancient Greek works to be sure they weren't left out purposefully. In particular, I don't see Interjection or Participle. Does Greek have Interjections? Do Greek Participles function like Latin ones—as a bizarre verb-form-escapee that thinks it's an adjective? If so, then it might warrant it's own POS header. I haven't decided for certain, but I'm leaning towards using that header in Latin. --EncycloPetey 04:19, 3 February 2007 (UTC)

Also, what part of speech is βρεκεκεκὲξ κοὰξ κοάξ (From Aristophanes' ΒΑΤΡΑΧΟΙ)? I'm not skilled enough to attempt to read in in the original language, but it's till one of my favorite plays. --EncycloPetey 04:19, 3 February 2007 (UTC)

This may be yet another example of my lack of mastery of the language. I guess it was my intent to treat participles simply as verb forms. Certainly they do function somewhat differently than normal verbs, but I guess I didn't feel that they were distinct enoug to merit their own header. You seem to think that Latin participles function differently enough to merit some distinction. Whether Greek participles function differently in this respect, or whether they should, in fact, get their own headings, I don't really know. I must be quite clear in that I am not a master of the Ancient Greek language. I took five semesters of Biblical Greek at a bible college. I've since studied up a bit, but nonetheless...... As for interjections, they just slipped my mind. I can only think of one off the top of my head (οὐαί), but I imagine there must be more. I'll add that category. And as for that crazy phrase by Aristophanes, I haven't the foggiest; I've never heard it. If you can point me to a copy of the original text, I'll take a look at it, see if I can figure it out. Sorry. Cerealkiller13 04:48, 3 February 2007 (UTC)

The Aristophanes phrase in question is onomatopoeic, which the second chorus (of frogs) refrains as Dionysius and his servant cross the Styx. The edition of the play I have is in the "original" Greek, titled Aristophanes: The Frogs edited by W. B. Stanford (London:Macmillan, 1958). I don't know whether a more recent edition of the Greek text is available. --EncycloPetey 04:54, 3 February 2007 (UTC)

Meow is listed as an interjection. I suppose that would be the same. If it's not, we could always add a header. Certainly it's better to add things to the AAG before it gets the blue box, but it's not as though it's set in stone once that happens. It's still a wiki. Cerealkiller13 05:08, 3 February 2007 (UTC)

To resurrect this topic: would any Ancient Greek editors find fault with adding "Particples" to the list of POS? It was brought up at the Latin page and figured it would be good to ask around here. If people do have problems with the header, any suggestions on how to approach the issue? Thanks! Medellia 17:11, 21 May 2007 (UTC)

I don't have any strong objections to it. I guess I've always considered them simply verb forms, but I have to admit that I don't really understand them too well. If we were to include "Participle" as a POS, how would we denote the participle's relation to the corresponding verb? Would it be best to simply put "Particple form of x verb", or would it get its own definition, with "Particple form of x verb" in the etymology? Atelaes 04:31, 22 May 2007 (UTC)

Here's an example of what I would ideally do under the "Participle" heading:

ἀγούσης

1. feminine genitive singular present active participle of ἄγω (ágō)

Medellia 06:02, 22 May 2007 (UTC)

That looks quite reasonable to me (although I must admit, I just can't be bothered to use the polytonic template outside of non-Greek entries). Atelaes 06:39, 22 May 2007 (UTC)

Should there be a "sub-lemma" participle page somewhere? That is, is there one participle form for each "word" that could give a definition and full table of participle inflection. If so, which participle form should it be? --EncycloPetey 15:17, 22 May 2007 (UTC)

To be honest, I would have a full inflection template at the masculine nominative singular, but nothing different in the definition line. Atelaes 16:32, 22 May 2007 (UTC)

I vote for multiple pages (or contracting templates!) because there are simply too many forms (if what we're striving for is providing full inflection, which I would assume to be true): present active; present middle/passive; future active; future middle; future passive; aorist active; aorist middle; aorist passive; perfect active; perfect middle/passive; and future perfect middle/passive. The participial system of Greek is so substantially greater than Latin that it really does need more room. Medellia 16:38, 23 May 2007 (UTC)

Linear scripts

This questions was posed on the About Ancient Greek main page, but it didn't get any feedback, and so I'm hoping by placing it here, it might generate some discussion. What should be done with words in Linear scripts? Should they be considered etymons? Alternative spellings? Cognates? How about the distinction between words in Linear B and say, the Cypriot syllabary? For the latter, it would seem that the words themselves would have to be under the Ancient Greek language header, since Arcadocypriot was simply a dialect of Ancient Greek (and largely concurrent with it), as opposed to Mycenaean Greek, which preceded other Ancient Greek dialects, such as Attic, Ionic, and Doric (and, somewhat importantly, has its own ISO code). However, the matter is further complicated by the fact that Arcadocypriot was a rather archaic dialect. And so the etymologies of many Ancient Greek words are determined by comparison with corresponding Arcadocypriot words. In my opinion, perhaps the best approach would be something along the lines of: Etymology - From *(hypothetical etymon), as evidenced by (Mycenaean word/Arcadocypriot word/etc.). Unfortunately, I can't think of any spectacular examples to illustrate my point, and so this less than helpful example will have to suffice. Any thoughts? If anyone would like to participate in this discussion, but is not up and up on Ancient Aegean scripts and whatnot, please ask and someone will fill you in. Atelaes 21:48, 14 February 2007 (UTC)

Also, it is my opinion that, when a word in a Linear script does not evidence an archaic form, but more or less simply points to the current form (such as 𐀓𐀬𐀰, which seems to simply be the same as χρυσός), it should simply be listed as an alternative spelling. I realize this will be a little weird, having an alternative spelling in a "different language", but I can't think of a better solution. Atelaes 21:51, 14 February 2007 (UTC)

We might be able to borrow a page from Dijan's book. He works often in Serbian, which is routinely written in both Roman and Cyrllic scripts. As a result, each entry form must reference the spelling in the other script. Look at час and čas, which are the same Serbian word in two writing systems. We might adopt a similar approach with Ancient Greek words that appear in different scripts. --EncycloPetey 00:57, 15 February 2007 (UTC)

Sorry to but in :) — I'm unsure of the chronology. Did the language represented by LinearB evolve into AncientGreek (like AngloSaxon became English) If so it mut be an Etymology. Or were LinearB and Grk-script used at the same time - if so I suppose it is an Alternative Spelling —Saltmarsh 09:02, 15 February 2007 (UTC)

Linear B represented an earlier form of Greek. The only problem with treating Linear B entries as straight etymologies is that sometimes it is quite obvious one word came from another, and so we'll have bouncing back and forth. For example, the etymology of Αἰγύπτιος (Egyptian) might run: From 𐁁𐀓𐀠𐀴𐀍 (ai-gu-pi-ti-jo) from Αἴγυπτος (Egypt). That doesn't make any sense because the derivation came before any script, and the Mycenaean is simply reflecting the same thing as the Ancient Greek. So, instead of saying Mycenaean is to Ancient as AngloSaxon is to English, it might be better to say Mycenaean is to Ancient what Shakespearean English is to modern English. Certainly there's been some changes, but there are more similarities. Which is what makes this so confusing, because some Mycenaean words definitely should be in the etymology, because they represent a distinctly earlier form. But a lot of Mycenaean words are just representing the same thing that the Ancinet Greek is, exactly the same thing (probably, no one really knows for absolute certain :-)). Atelaes 18:45, 15 February 2007 (UTC)

Perhaps a better way to think of it is that Mycenean is to Ancient Greek as the language of Robert Burns is to English. Yes, it is older, but it is not the immediate ancestor (or at least that's the way I understood it) – it is just the only attested dialect of an older form of Greek. With that in mind, I think the solution Atelaes propsed is the best – to say something along the lines of From Proto-Greek *xxxx ( > Mycenean yyyy). Widsith 20:26, 15 February 2007 (UTC)

ου

It has long been my understanding that the classical linguistically orthodox pronunciation of ει and ου are [eː] and [oː] respectively, and that they were considered phonetically long versions of short ε and ο (pronounced [e] and [o]), and that this is reflected in the compensatory lengthening of vowels when ν is swallowed by a following σ in Classical Attic (ᾰν+σ=ᾱσ, εν+σ=εισ, ῐν+σ=ῑσ, ον+σ=ουσ, ῠν+σ=ῡσ). It is also my understanding that ει and ου (along with οι which was either pronounced [oɪ], [oe] or [øː]) together became more closed vowels by the time of Alexander the Great ( ει=[iː], ου=[uː], οι=[yː]), and then all vowels became isometric in Koine ( ει=[i], ου=[u], οι=[y]). (Polytonic orthography became rarely used or pronounced in vulgar manuscripts by this time as well, but that's a separate issue.) For Classical οι, a conservative pronunciation like [oɪ] or [oe] should probably be used. But the written pronunciations of ει and ου should definately mirror one another—either [eː] with [oː], or [iː] with [uː], but not [eː] with [uː] nor [iː] with [oː]. [uː] for ου may have been dialectually common by the late Classical period, but we do know that while many phonemes interlapped dialectually, ου and ῡ never did, nor did they even came close (their conservative pronunciations were [oː] and [yː]). Either one of these could be realized as [uː] depending on the dialect, but no dialects merged them. Also, in the Classical period, most foreign [u] sounds were adapted with υ ( Βηρυτός=Beirut, Καπύη=Capua, Λιγυρια=Liguria, Λυσιτανία=Lusitania, Τύρος=Tyre/Sur/Sor), and ου only became commonplace for this task by the very late Classical and Koine periods as new place names came into being to be given names in Greek. Anyway, what I propose is an official change in policy to transcribe Classical pronunciation of ου as [oː], and continue its Koine pronunciation as [u]. - Gilgamesh 05:12, 11 June 2007 (UTC)

Seems reasonable, switched. Atelaes 23:06, 11 June 2007 (UTC)

Wow, slam dunk. I didn't expect it to be that easy to make my case. XD - Gilgamesh 23:10, 11 June 2007 (UTC)

New Romanization scheme

While the new pronunciation/romanization scheme is certainly more comprehensive (and I have to assume, more accurate as well), it is now too complex. Not too complex as in covering too many digraphs, or anything like that, but too complex as in it offers too many options. This page is not meant to give a summary history of the language, or all the possibile pronunciations, etc. that are feasible, but is meant to provide instructions and an SOP for doing Ancient Greek on Wiktionary. Thus, each character should have a single option for how it is to be romanized and transcribed in IPA. Now, if characters differed based on their context within a word, that is fine, we can certainly have a complex set of rules, but what we really can't have is "character x was sometimes pronounced as IPA|g, and sometimes as IPA|c. While this probably sounds worse than it is really intended, this table is not interested in how Greek was actually pronounced, but how it is pronounced on Wiktionary (although, hopefully the latter is somewhat influenced by the former :)). Where variations occur (as they certainly do), a decision needs to be made. In general, my original scheme was meant to illustrate major sound shifts which took place with the transition from Classical to Koine to modern. The same applies to romanization. Upsilon cannot be romanized as both u and y, one must be decided on (I have a very strong preference for u over y myself). Atelaes 00:18, 14 June 2007 (UTC)

Yes, I'm sorry. XD As I am high-functioning autistic, I often have a notorious inability to keep things simple. However, with your help, I am certainly willing to try to simplify it. As for rules of thumb, I can try to edit out the "sometimes" and "often" and just set forward rules. As for the romanizations, I have traditionally used a more non-latinist approach, where ξ is /ks/ (not /x/), χ is /kh/ (not /ch/), etc. This largely persists into Koine, though for Koine transcriptions I tend to drop solitary /h/ (though I can see how it can be left in), and also change Classical /u/ to /y/ and simplify the vowels to monophthongs (αι=e, αυ=av/af, etc.) However, because of certain abnormal behaviors, in Koine romanizations I keep η=ē and ω=ō, if at least for the sake of the significant Pontic dialect. For Byzantine, heavy breathing /h/ is totally gone and I tend to replace the aspirates with /f/, /þ/ and /h/ at this point. /b/ becomes /v/ completely. I retain /y/ as such because, even though by the 10th century it becomes [i] most places, it remains distinct in the Saronic Gulf region (not including Athens), where they start to spell it ιου for a time. Certain Byzantine mutation patterns are widespread and persist to the present day, but rare ones like ω=u can probably be safely ignored—to be safe, η=ē and ω=ō can be retained into Byzantine because Pontic at this time is still transitional with Constantinople. Modern Greek is outside of the scope of this project, though it should be understood that the increased fragmentation of Greek-speaking communities at this time cements the separate language status of the major dialects of Common Ottoman Greek from that of Pontic, Cappadocian, Tsakonian and Griko. Transcribing Classical Greek vowels is actually pretty easy, using α=a ε=e η=ē ι=i ο=o υ=u ω=ō αι=ai ᾳ=āi ει=ei ῃ=ēi οι=oi υι=ui ῳ=ōi αυ=au ευ=eu ηυ=ēu ου=ou ωυ=ōu. For accents, αί=aí αὶ=aì αῖ=aî ή=ḗ ὴ=ḕ ῆ=ê etc. On long diphthongs, āí=ᾴ āì=ᾲ āî=ᾷ etc. For diaeresis, experience has taught me that using Unicode umlauted vowels in transcription is not always possible with diacritics, so I simply use an apostrophe before the romanization, e.g. ϊ='i, ϋ='u, etc. In Koine, these tasks all become much simpler as all vowels either become monophthongs or +v, and there is only stress accent, and so I don't even bother with keeping transcribed vowels from touching each other without diaeresis in Koine romanization: αι=e αυ=av οι=y ου=u υ=y υι=yi ωυ=ōy. υι and ωυ have already congealed into two separate consecutive vowels (and ωυ was not native to Koine's Attic nucleus anyway), so I write them as two consecutive vowels without worries. For Byzantine, I don't change the vowel romanization system much, but I reflect the regular phonetic habits in consonants, e.g. πτ/φθ=ft, ευσ=eps etc. Atticism had a lot of clout with the literary elite in the Koine and Byzantine periods, and even influenced Katharevousa in the Modern period, but it's understood that these were archaisms that were not reflected in common speech. So maybe I can clarify this situation by prescribing a different historical transcription system each for the Classical, Koine and Byzantine columns, with occasional additional notes. It should also be understood that the main center of elite culture is different in these three periods—Athens for Classical, Alexandria for Koine, and Constantinople for Byzantine (and then back to Athens for modern standardized Katharevousa and standardized Demotic Greek), and standard pronunciations should reflect the common languages of these centers. - Gilgamesh 02:35, 14 June 2007 (UTC)

First of all, please don't feel the need to apologize. Your work has already markedly improved the Ancient Greek section (we simply have some road left to travel :)). To begin with, I think you may be muddling romanization and pronunciation. While each time period can receive its own pronunciation, each word gets only one romanization. The purpose of the romanization is a quick and dirty pronunciation, so that people not familiar with Greek characters can get a rough feel for how the word sounds when it pops up on non-Ancient Greek pages (as they often do, especially in etymologies). For this, all we want is a one-to-one correspondence of Greek characters to Latin characters. And I maintain my position that it is a bad idea to include accents within the romanization, as this helps few, and confuses many. The IPA within the entries will provide more accurate and precise pronunciation. A few starting notes on specific romanizations. I have no problem with using kh for χ, nor ks for ξ. However, I do have a problem with y for υ. The reason is that English really doesn't have a set sound for y. I think most commonly it is pronounced [ɪ], but is often [i], and is also often [ai], none of which are terribly close to [y]. However, u is either [ʊ] or [u], which is probably the closest English has to [y]. Atelaes 02:55, 14 June 2007 (UTC)

Hmm... I still think accents are a good idea, but I guess we can do without. - Gilgamesh 04:59, 14 June 2007 (UTC)

That looks fantastic. One question: Why is ᾳ transcribed as āi, shouldn't it simply be ai? Other than that, it looks excellent. Atelaes 05:15, 14 June 2007 (UTC)

Because you wouldn't transcribe ῳ as oi either. :3 ᾰ and ᾱ merge in Koine, as do ᾰυ and ᾱυ, and so neither the breve nor macron are typically written in Koine polytonic. But αι and ᾳ do not merge in Koine, and if we transcribe ῃ as ēi and ῳ as ōi, then it only makes sense to transcribe ᾳ as āi because it remains distinct from αι to modern times. - Gilgamesh 05:46, 14 June 2007 (UTC)

Verbs

By far, the major sore spot currently in Wiktionary's Ancient Greek section is verbs. If you take a look at Category:Ancient Greek verbs, you'll see that they're rather in a state of shambles. When I originally started on Wiktionary, the whole language looked like this. Medellia and I have managed to clean up the nouns, proper nouns, adjectives, and, to a limited extent, pronouns, numbers, prepositions and conjunctions. The basic verb question is one of inflection. Medellia set up {{grc-verb}} to include the six principle parts. The major problem with this is that different dialects vary considerably in their prinicple parts for a single verb, which this template can't adequately reflect. Also, Medellia wrote a rather impressive inflection template for verbs (which you can see in use at φιλέω), but its use is rather limited. I believe the inherent problem here is that the six parts inflect rather independently of each other (how the present/imperfect inflects is rather independent of how the aorist inflects). What I'm thinking might be best route is to create separate inflection templates for each principle part, and then have a super-template which merges them all together in an entry. Currently, I lack the necessary understanding to undertake this problem. While I fully intend to study up, that could take a while. Thoughts? Atelaes 00:34, 14 June 2007 (UTC)

I would say that to have separate templates for each principal part would be an excellent plan. Obviously, if the template for a standard Attic contract verb is as difficult to use as it has proved itself to be, irregular verbs would quickly become a nightmare. I can throw together a standard Attic -ω conjugation present tense template as an example. Any idea of how you'd like it to look? Medellia 01:04, 14 June 2007 (UTC)

I really like the look of φιλέω, as far as what the user sees goes. I think the coding within the entry should probably look something like {{grc-conj|pres3=φιλ|aor2=ἐφίλ|perf5=πεφίλ|perfmp2=πεφίλ|aorpass1=ἐπιλή}}. Atelaes 01:13, 14 June 2007 (UTC)

Also, what we might consider doing is to have the six principle parts be displayed on the non-expanded inflection (for example, when φιλέω has the inflection hidden, there's a little bar that says "Inflection"). That way, we could have....say....three inflection tables in a verb entry, showing the inflection of that verb within three different dialects, and each hidden inflection template would show the principle parts associated with that inflection scheme. Atelaes 01:19, 14 June 2007 (UTC)

Do you know how much difference in inflection is present in other dialects? I've mainly worked in "Epic" and Attic (and, accordingly, Ionic)... I know some tidbits about differences between Doric and Attic, but beyond that my grasp of AG dialects is tenuous at best. What with school out, I don't quite have the resources I usually enjoy... anyway, if there's difference between the Attic and Aeolic second person plural present m/p subjunctive - for instance - past differences between the principal parts, I don't know how much I'll be able to offer! Medellia 02:53, 14 June 2007 (UTC)

My knowledge of all that is scant at best. But, I do know that a number of dialects have different principle parts for a number of verbs. I simply think it wise to set up the infrastructure to accomodate such things, in the hopes that someone will, one day, come along who can make use of it. I don't have a big problem if no one currently can do all the different dialects; I just don't want to re-do all of the templates a year down the road when we have 1000 verbs, and someone comes along who knows Doric. Atelaes 02:59, 14 June 2007 (UTC)

Makes sense. I've made an example template for a first principal part with -άω contraction, if you'd like to take a look. Medellia 16:51, 14 June 2007 (UTC)

Awesomeness

Just a comment to say that I think the About Ancient Greek page is probably the best page of its kind on Wiktionary. It's fantastic. Widsith 17:01, 15 June 2007 (UTC)

I imagine any such awesomeness is likely a result of the heavier community involvement which this page has received (as compared to other pages of similar natures). There are a dozen editors in the page's (and its talk page's) history. I suppose it is fitting that the Greeks yet again set the standard for others to follow. Atelaes 18:23, 15 June 2007 (UTC)

It's my fault.  ::runs:: ^^ - Gilgamesh 21:00, 15 June 2007 (UTC)

Information overload

I think it might be time to split the pronunciation / Romanization tables off to their own page on Appendix:Ancient Greek pronunciation. Right now it's swamping out all the other information about Wiktionary format. --EncycloPetey 05:07, 19 June 2007 (UTC)

That's a good idea. I'll do that now. - Gilgamesh 05:08, 19 June 2007 (UTC)

OK, All we need here are the basics for convention, with a link to the appendix for those who need more information. --EncycloPetey 05:09, 19 June 2007 (UTC)

Breves in Templates

Maybe I'm confusing Latin policy with AG policy, but I just wanted to clarify: are we using breves? I would personally not support their inclusion as it seems self-explanatory that a vowel without a macron would be short. There are enough diacritics in Greek without the use of one that seems somewhat unnecessary. Medellia 15:50, 19 June 2007 (UTC)

No, we're not using breves in the dictionary. However, I recalled that breves and macrons are useful in inflection tables to distinguish certain otherwise minimal pairs. I plan to keep the breve-macron distinction confined to there and maybe to IPA pronunciation. - Gilgamesh 17:01, 19 June 2007 (UTC)

Hold on a second, yes we are indeed using breves in this dictionary (or at least, that's what's currently spelled out on the AAG page, and my personal preference, the two coinciding not coincidentally I suppose). Here's the reason, most of the Ancient Greek entries do not have any macrons or breves. So, a blank vowel needs to be assumed ambiguous. Thus, a blank vowel cannot be assumed to be short. Thus, breves must be allowed. The usage of macrons and breves is not as widespread in Ancient Greek references as it is in Latin ones. Even a short, beginner's Latin dictionary will have all the macrons, but a beginner's Ancient Greek one will not. So, we have to allow for editors to input entries with ambiguous vowel lengths when they don't have access to that information (as well as editors who are simply too lazy, such as myself), or for the myriad of A. Greek words which we simply don't yet know what the vowel length is. Atelaes 21:02, 19 June 2007 (UTC)

I must be thinking of Latin policy then. (Sorry for assuming that it wouldn't be in the AAG page!) Do you think you could edit the AG Edit tools to include: ᾰ ᾱ ῐ ῑ ῠ ῡ? (Do accented versions exist in unicode?) While we're on the topic of diacritics, thoughts on the diaeresis? My Homer text would mark them from time to time, most notably for "Πηληϊάδης". Medellia 23:26, 19 June 2007 (UTC)

Edit tools? Hehe, I never noticed them before. XD Most of the time when I work with polytonic text (especially before I installed the Windows Polytonic Greek IME), I used a UTF-8 text file with letters and common digraphs: ἈἀἉἁα ἌἄἍἅά ἊἂἋἃὰ ἎἆἏἇᾶ ᾸᾰᾹᾱ ΑἰαἰΑἱαἱαι ΑἴαἴΑἵαἵαί ΑἲαἲΑἳαἳαὶ ΑἶαἶΑἷαἷαῖ ᾈᾀᾉᾁᾳ ᾌᾄᾍᾅᾴ ᾊᾂᾋᾃᾲ ᾎᾆᾏᾇᾷ ΑὐαὐΑὑαὑαυ ΑὔαὔΑὕαὕαύ ΑὒαὒΑὓαὓαὺ ΑὖαὖΑὗαὗαῦ Ββ Γγ Δδ ἘἐἙἑε ἜἔἝἕέ ἚἒἛἓὲ ΕἰεἰΕἱεἱει ΕἴεἴΕἵεἵεί ΕἲεἲΕἳεἳεὶ ΕἶεἶΕἷεἷεῖ ΕὐεὐΕὑεὑευ ΕὔεὔΕὕεὕεύ ΕὒεὒΕὓεὓεὺ ΕὖεὖΕὗεὗεῦ Ϝϝ Ζζ ἨἠἩἡη ἬἤἭἥή ἪἢἫἣὴ ἮἦἯἧῆ ᾘᾐᾙᾑῃ ᾜᾔᾝᾕῄ ᾚᾒᾛᾓῂ ᾞᾖᾟᾗῇ ΗὐηὐΗὑηὑηυ ΗὔηὔΗὕηὕηύ ΗὒηὒΗὓηὓηὺ ΗὖηὖΗὗηὗηῦ Θθ ἸἰἹἱιϊ ἼἴἽἵίΐ ἺἲἻἳὶῒ ἾἶἿἷῖῗ ῘῐῙῑ Κκ Λλ Μμ Νν Ξξ ὈὀὉὁο ὌὄὍὅό ὊὂὋὃὸ ΟἰοἰΟἱοἱοι ΟἴοἴΟἵοἵοί ΟἲοἲΟἳοἳοὶ ΟἶοἶΟἷοἷοῖ ΟὐοὐΟὑοὑου ΟὔοὔΟὕοὕού ΟὒοὒΟὓοὓοὺ ΟὖοὖΟὗοὗοῦ Ππ Ῥῥρῤῥ Σσς Ττ Ὑὑυϋ Ὕὕύΰ Ὓὓὺῢ Ὗὗῦῧ ῨῠῩῡ Υἱυἱυιϋι Υἵυἵυίϋί Υἳυἳυὶϋὶ Υἷυἷυῖϋῖ Φφ Χχ Ψψ ὨὠὩὡω ὬὤὭὥώ ὪὢὫὣὼ ὮὦὯὧῶ ᾨᾠᾩᾡῳ ᾬᾤᾭᾥῴ ᾪᾢᾫᾣῲ ᾮᾦᾯᾧῷ ΩὐωὐΩὑωὑωυ ΩὔωὔΩὕωὕωύ ΩὒωὒΩὓωὓωὺ ΩὖωὖΩὗωὗωῦ. - Gilgamesh 23:34, 19 June 2007 (UTC)

Oh, alright. I wasn't using breves and macrons in entries because you guys weren't and I quietly and casually presumed it to be convention. I will start using macrons and breves where appropriate. - Gilgamesh 23:28, 19 June 2007 (UTC)

I've been thinking. Maybe the breves and macrons should remain in inflection and not as article names or anything. The problem is that Unicode doesn't yet support precombined glyphs for ᾸᾰᾹᾱῘῐῙῑῨῠῩῡ to include acute, grave, circumflex, diaeresis, breathing marks, etc. in combination. It is entirely possible that one could use the plainest of Greek letters along with a stack of combining diacritic characters, but until most widely-distributed polytonic fonts support that with elegance and auto-combine them at a font level (or unless the Unicode standard adds more precombined glyphs), it is not yet practical to include breves and macrons in any but unstressed vowels without breathing marks or diaeresis. For now, it seems to me that the most practical option is to only include breves and macrons where permitting in the inflection tables, and pick up the slack in Classical IPA vowel lengths. - Gilgamesh 06:17, 20 June 2007 (UTC)

No, they shouldn't be used in entry titles, nor when noting A. Greek words in non-A. Greek entries (such as in etymologies). They should be placed in the inflection line, and as for in the inflection tables....I am unsure. Atelaes 19:01, 20 June 2007 (UTC)

Perhaps, then, we ought to hold off on mass proliferation until accented characters are provided in Unicode, but incorporate non-accented characters in the inflection templates wherever possible? Or would the general preference be to incorporate wherever possible (entries and templates alike) and continue to add as new characters are introduced? Medellia 19:53, 20 June 2007 (UTC)

I've been doing that already. Besides, breves and macrons in inflection tables are so pretty. ^_^ I know that's not NPOV though, so...do you think it's pretty too? :3 - Gilgamesh 02:21, 21 June 2007 (UTC)

How to inflect -ω nouns

There is a class of nouns I've seen use a superficially regular inflection pattern (said to be of the third declension), but I'm only familiar with their nominative and genitive forms. I don't know how to inflect all their cases in singular, dual and plural. Here are some examples:

• Jericho: Template:grc-decl-mix-ω
• Sappho: Template:grc-decl-mix-ω
• Sardinia: Template:grc-decl-mix-ω

Does anyone know enough about this class of nouns and help me fill in the blanks? :3 Then we can design a new declension template or two to accommodate them. - Gilgamesh 12:33, 27 June 2007 (UTC)

Alright, I've found some additional singular case endings for Sappho, and have filled them in above. - Gilgamesh 12:38, 27 June 2007 (UTC)

The same seems to apply for Sardinia. Filling it in. - Gilgamesh 12:42, 27 June 2007 (UTC)

And confirmed Jericho. So that takes care of singular. What about dual and plural? - Gilgamesh 12:44, 27 June 2007 (UTC)

Unfortunately, Smyth says:

"Stems in οι, with nominative in - ώ, turn ι into unwritten ι ̯ (y) (43) before the endings beginning with a vowel. ἡ πειθώ persuasion is thus declined:

N. πειθώ. G. πειθοῦς ( πειθό-ος). D. πειθοῖ ( πειθό-ι). A. πειθώ ( πειθό-α). V. πειθοῖ. Dual and plural are wanting.

So ἠχώ echo, εὐεστώ well-being, φειδώ sparing, Σαπφώ, Λητώ, Καλυψώ. οι stems are chiefly used for women's names.

a. A stronger form of the stem is ωι, seen in the earlier form of the nominative ( Σαπφῴ, Λητῴ). The accusative has the accent of the nominative.

b. When dual and plural occur, they are of the second declension: nom. λεχοί (late) from λεχώ woman in child-bed, acc. γοργούς from γοργώ gorgon.

c. ἡ εἰκών image, ἡ ἀηδών nightingale, properly from stems in ον, have certain forms from this declension ( εἰκοῦς, εἰκώ, voc. ἀηδοῖ)."

At least it makes template building easier! Hope that helped. Medellia 15:34, 27 June 2007 (UTC)

It would seem so, but this raises a question: Are these nouns of the second declension or the third declension? Or are they a mixed declension, where the singular inflections are third and the dual and plural inflections are second? Or are they a different declension altogether (like the Attic declension)? Which is it? - Gilgamesh 18:29, 27 June 2007 (UTC)

Mixed declension: third in the singular, second in the dual and plural. Medellia 22:24, 27 June 2007 (UTC)

Dual and plural? Why would proper names of places ever appear in the dual or plural? --EncycloPetey 19:35, 27 June 2007 (UTC)

The only example off the top of my head is the Kingdom of Two Sicilies... but I could see some odd poetic something or other popping up. Medellia 22:24, 27 June 2007 (UTC)

Oh, don't forget Athens (αἱ Ἀθῆναι), Thebes (αἱ Θῆβαι) or Mycenae (αἱ Μυκῆναι). If you're forgetting about those already, you really need a review. :3 - Gilgamesh 03:53, 28 June 2007 (UTC)

Those are different, though, in that they exist mainly in the plural. While attestations of ἡ Ἀθήνη (for the city, not the goddess), ἡ Θήβη, and ἡ Μυκήνη no doubt survive, post-epic dialects preferred the plural form. As such, the words came down to Latin as Athēnae, Thēbae, and Mycēnae. Medellia 14:53, 28 June 2007 (UTC)

I've made the template. I've also fixed the examples above to make use of them. :3 - Gilgamesh 04:17, 28 June 2007 (UTC)

-κλης/-κλεους

I can't figure out how to decline -κλης (genitive -κλεους) names. Examples include Ἀμεινοκλῆς, Ἀνδροκλῆς and Ἡράκλης. Ἡράκλης in particular seems to have a lot of variant cases. After review, I determined that these nouns don't inflect like any declension pattern I've encountered so far, and I can't seem to isolate appropriate existing declension templates for them. Help? :3 I learn much more as I go along, but I appreciate that it is especially important in this environment to get it accurate and not just guess as a student would. - Gilgamesh 19:42, 1 July 2007 (UTC)

Try taking a look at Appendix:Latin_first_declension#Greek declension. Does the ending pattern resemble what you're seeing? These are declension patterns in Latin that were borrowed from Greek. --EncycloPetey 21:05, 1 July 2007 (UTC)

I see what you're saying, but no. There's already a well-documented use of -ης in Greek's first declension. But the names I'm speaking of don't reflect like those—they seem irregular, but regular among each other (probably because of the common rendering of -κλης). κλης is a contraction of κλέος, which means "glory", which is a neuter noun whose declension pattern I'm also uncertain of. Here are the few examples I've seen using the name Ἀνδροκλῆς.

Declension of Ἀνδροκλῆς, Ἀνδροκλέους

Case / #	Singular	Dual	Plural
Nominative	Ἀνδροκλῆς	?	?
Genitive	Ἀνδροκλέους	?	?
Dative	Ἀνδροκλεῖ	?	?
Accusative	Ἀνδροκλέα	?	?
Vocative	?	?	?
Dialectal forms:
Notes:	{{{notes}}}

I'm not sure what to make of the other inflections. Template:grc-decl-3rd-M-ευς seems closest to this, but it applies only to -ευς nouns and its genitive form is the slightly different -εως. - Gilgamesh 21:32, 1 July 2007 (UTC)

Somewhere in my books, I know that I've seen a discussion about the various ways in which Greek proper names inflect and how that was adopted into Latin. I'll see if I can find additional useful information. --EncycloPetey 21:35, 1 July 2007 (UTC)

It looks a lot like the Third declension Greek form (It's not in the corresponding Wiktionary appendix). I have, for example, for crater:

Declension of crater, crateros

Case / #	Singular	Dual	Plural
Nominative	crater	?	?
Genitive	crateros	?	?
Dative	crateri	?	?
Accusative	cratera	?	?
Vocative	?	?	?
Dialectal forms:
Notes:	{{{notes}}}

This matches (loosely) the endings you've noted. --EncycloPetey 21:42, 1 July 2007 (UTC)

Yeah, most Greek third declension nouns have those endings. But the devil is in the details. There are subtle differences in these nouns that I would rather have nailed before I start putting the declensions in articles and presenting them as if they are scientific fact when I am not certain of it. Besides, Latin is still a different language with its own naturally-evolved grammatical differences, though similar many of them are. Latin inflection of words loaned from Greek is not identical to Greek, as the loans often calque from the Greek declensions to similarly-used Latin declensions. Indo-European languages with conservative declension-oriented grammar have frequently transplanted words from one language in one declension to the other language in a similar declension. So I would not personally use Latin inflections of Greek loans as a reflection of how they are inflected in Greek itself—I'd rather hear it from the Greek grammarian's mouth. - Gilgamesh 22:03, 1 July 2007 (UTC)

I am aware of the shortcomings of this approach, but in this case I think the approach is stronger than you realize. The third declension "Greek" forms in Latin are exclusive to words loaned from Greek and do not follow any of the Latin inflectional patterns. In the Latin third declension, the singular genitive should end in -is, and the singular accusative shouls end in -em. For words not borrowed from Greek, they do. But for words loaned from Greek, the ending patterns don't follow the norm. Hence, thorough Latin grammars list a separate declension pattern for the words of Greek origin. That's not to say I recommend trying to work back from Latin and set it into the templates here. However, it is a valuable line of evidence that might be pursued to more definitive information. --EncycloPetey 23:10, 1 July 2007 (UTC)

One quick check before I get going on this: are you sure that a stand-alone κλῆς exists? I recall κλέος having a digamma ( κλέϝος) in its earlier form. My first thought is that it could possibly resemble the α-contract nouns, like Ἑρμῆς. I don't have any books at hand, but I'll dig one up and see if I can find out anything to add to the discussion. Medellia 23:59, 4 July 2007 (UTC)

What do you mean? -κλῆς is a suffix I've encountered on various names that means κλέος in that context. I don't know if κλῆς exists by itself as a word. - Gilgamesh 00:11, 5 July 2007 (UTC)

Ah I see. I misunderstood. Thank you for the clarification on that one! Medellia 03:30, 5 July 2007 (UTC)

I took a second to read through everything and am fairly certain I'm correct. You may want to see if Smyth discusses this type of nouns, but from all the notes I have sitting around this is what I gather. Okay... first things first: -έος cannot contract to -ῆς in Attic. What we're dealing with rather is a contraction of -έης in this situation. This is a third declension noun, nothing terribly irregular about it. Due to contraction, the endings of the singular are: -ῆς, -έους ( έεος, ε+ο to ου - a spurious diphthong), -εῖ ( έει, ε+ι to genuine diphthong ει, ε+ει then to ει genuine diphthong, circumflex for the contraction), -έᾱ ( έεα, this is the most irregular thing about the paradigm; it's a contraction that occurs I think solely in the third declension. In the first it would be an η), and -εις ( εες, ε+ε to spurious ει; accentuation is thrown back into the stem). I hope that clears this up a bit. Medellia 03:30, 5 July 2007 (UTC)

One last thing before I forget: while I'm certain we could deduce the dual and plural forms, I think the chance of a name in either number is going to be even more unlikely than our place names in the dual and plural as discussed above. I would hesitate to include information without attestation. Medellia 03:36, 5 July 2007 (UTC)

And because I can't stop beating this to death: if you are interested in exploring more forms, τριήρης is a good example of the sigmatic stem, albeit without the contraction. Medellia 07:10, 5 July 2007 (UTC)

I'd like to make templates for this. What would you gather the dual and plural forms would be? Could you try filling out these inflections to the best of your ability?

Declension of Ἀνδροκλῆς,

Case / #	Singular	Dual	Plural
Nominative	Ἀνδροκλῆς
Genitive
Dative
Accusative
Vocative
Dialectal forms:
Notes:	{{{notes}}}

Declension of Ἡράκλης,

Case / #	Singular	Dual	Plural
Nominative	Ἡράκλης
Genitive
Dative
Accusative
Vocative
Dialectal forms:
Notes:	{{{notes}}}

- Gilgamesh 23:00, 9 July 2007 (UTC)

While I could guess at this, I really don't feel comfortable presenting information that was determined by anyone less than an ancient grammarian. If I can find an attestation of a name in the dual or plural, I'll throw together a complete table. Since the issue of this -εεσ stem seems to be expressed mainly in names, there shouldn't be a problem.

Declension of Ἀνδροκλῆς, Ἀνδροκλέους

Case / #	Singular
Nominative	Ἀνδροκλῆς
Genitive	Ἀνδροκλέους
Dative	Ἀνδροκλεῖ
Accusative	Ἀνδροκλέᾱ
Vocative	Ἀνδρόκλεις

I personally see nothing wrong with the template existing solely in the singular for the time being. Now Ἡρακλέης is another monster altogether. Lest I attempt and fail to explain something which I barely understand (you'd think that for someone so embedded in the Greek consciousness, they could get his name down), I will point to LSJ on that count! Medellia 02:49, 10 July 2007 (UTC)

Alright, so what about -ης/-ους? Up to now I've assumed it to be a variation of -ης/-ου. Is it also third declension? If so, how is it inflected?

Declension of Σωκράτης, Σωκράτους

Case / #	Singular	Dual	Plural
Nominative	Σωκράτης
Genitive	Σωκράτους
Dative
Accusative
Vocative
Dialectal forms:
Notes:	{{{notes}}}

- Gilgamesh 03:12, 10 July 2007 (UTC)

It is third declension; the primary textbook example seems to be ἡ τριήρης.

Declension of Σωκράτης, Σωκράτους

Case / #	Singular
Nominative	Σωκράτης
Genitive	Σωκράτους
Dative	Σωκράτει
Accusative	Σωκράτη
Vocative	Σώκρατες

Declension of τριήρης, τριήρους

Case / #	Singular	Dual	Plural
Nominative	τριήρης	τριήρει	τριήρεις
Genitive	τριήρους	τριήροιν	τριήρων
Dative	τριήρει	τριήροιν	τριήρεσι(ν)
Accusative	τριήρη	τριήρει	τριήρεις
Vocative	τριῆρες	τριήρει	τριήρεις
Dialectal forms:
Notes:	{{{notes}}}

I've also seen an accusative Σωκράτην which I'm sure is due to some sort of pattern assimilation (first declension seems a reasonable bet). Medellia 04:45, 10 July 2007 (UTC)

Okay, I've finished purging the 1st declension -ους-gen templates of their embedding articles and migrated them to Template:grc-decl-3rd-M-ης-pax. I will use that one for penultimately-stressed -ης/-ους from now on. - Gilgamesh 05:50, 10 July 2007 (UTC)

I'm going to take a wild gander. I am not changing the templates to reflect this without educated group consensus, but I wonder if this works...

Declension of Ἀνδροκλῆς, Ἀνδροκλέους

Case / #	Singular	Dual	Plural
Nominative	Ἀνδροκλῆς	Ἀνδροκλεῖ	Ἀνδροκλεῖς
Genitive	Ἀνδροκλέους	Ἀνδροκλέοιν	Ἀνδροκλέων
Dative	Ἀνδροκλεῖ	Ἀνδροκλέοιν	Ἀνδροκλεῖωῐ(ν)
Accusative	Ἀνδροκλέᾱ	Ἀνδροκλεῖ	Ἀνδροκλεῖς
Vocative	Ἀνδρόκλεις	Ἀνδροκλεῖ	Ἀνδροκλεῖς
Dialectal forms:
Notes:	{{{notes}}}

- Gilgamesh 17:37, 10 July 2007 (UTC)

Byzantine pronunciation

It occurs to me that I've been rather arbitrary with deciding which Byzantine pronunciation choices to include or not to include. I mean, from C.E. 300, there are certain features characteristic of the entire Byzantine linguistic period that set it apart from the earlier Koine period.

• η and ε/αι were pronounced [e] and [ɛ] in Koine, but had shifted to [i] and [e] by Byzantine.
• Full labiodentalization of the Koine bilabial fricatives, where β and φ had been pronounced [β] and [ɸ], and they became pronounced [v] and [f].
• μβ, νδ and γγ in Koine had remained dental-plosive, but became dental-fricative in Byzantine.
• μφ, νθ, γχ and νσ became pronounced φ, θ, χ and σ.
• The voiced fricative nasal sandhi, where a voiced fricative before a nasal consonant itself becomes a nasal, where βν=μν, γν=γγν, ευν=εμν, etc.
• The voiceless consonant sandhi, whereby if two different consecutive consonants are both plosives or both fricatives, then the first becomes a fricative and the second becomes a plosive, where πτ/φθ=φτ, κτ/χθ=χτ, ευθ=ευτ/εφτ, etc., unless the second consonant is the voiceless fricative σ, then the σ remains a fricative and the first consonant becomes a plosive, where αυσ=αψ, etc. Existing fricative-plosive, plosive-fricative or geminated consonant combinations remain stable.
• Word-final unstressed -ον and -ος after a vowel become simply -ν and -ς, and -ιος merges with -ης.
• ε/αι before a vowel (presumably not ι/ει/η) is pronounced as ι, even when stressed. (This may be because of an earlier hypothetical but unclear merging of ε/αι with η before a vowel, and where η was pronounced e in Koine, it became i in Byzantine.)
• Combinations of ουε/ουαι become pronounced ο.
• The fate of -εος/-αιος and -εον/-αιον are uncertain as they would seem to be affected by two of the rules I mentioned above. These cases are not clearly described in my sources. I've been giving their Byzantine pronunciations as [es] and [en], but I'm not certain these are correct, and they may in fact be [is] and [in]. I just don't know.

Other developments happen later in Byzantine, after the end of Late Antiquity, and it's unclear whether these were creeping changes or sudden changes. In the 10th century, υ/οι completely merged with ι/ει/η except in some dialects along the western Saronic Gulf (where they spelt it ιου). I've included this change in the pronunciation, but I'm not sure now that this was a proper thing to do. A later change in the 13th century (that I did not include because of its very late date in the Byzantine period) was when all stressed [i] sounds before another pronounced vowel lost their accent to the next vowel. Palatalization of velars before front vowels is not quite certain in Byzantine, but definately exists by the Ottoman period—it is a feature that may have developed earlier under Slavic influence, or later under Turkish influence (or, contrastly, Turkish palatalization of velars before front vowels may have been under Greek and/or Slavic influence). Either way, as I am an uncertain of Byzantine's use of palatal consonants or of the glides [j] and [w], I cautiously do not indicate them in pronunciation. Degemination is also not certain in Byzantine, as it's a process that's not even complete today—the modern Dodecanese and Cypriot dialects maintain gemination of fricatives, and distinguish geminated plosives by pronouncing them as ungeminated aspirated plosives (where phonemically ungeminated plosives are unaspirated and sometimes even voiced). It should be remembered that Byzantine Greek has significant differences from Modern Greek, as Modern Greek was heavily influenced by the Katharevousa movement and forcefully introduced certain archaisms in the phonology—in Byzantine, the scholars wrote in straight Attic and the spoken language was not regulated as spoken languages are today, so the difference between the street and the writing desk could be vast (especially under the strict archaicist Atticist principles influential since Koine), and this is why a lot of Byzantinisms have so drastically retreated today. So, my point is...should I change the Byzantine pronunciations to include [y] (prior to 10th century), or do I change it to include later iota stress loss (by 13th century), or do I leave things as they are? And should I indicate [j] and [w] glides (and [ɥ] if [y] is retained), or should I leave them [i] and [u] as they are now? - Gilgamesh 00:36, 13 July 2007 (UTC)

Unfortunately, the thing I have the least amount of expertise in — when it comes to AG — is pronunciation. Some part of me feels that, as with an etymological progression, it makes sense to provide a variety of pronunciations. My main concern would be feasibility; as such, another part of me feels that leaving the pronunciation as it stands would enable you to make progress in other areas instead of going back to change many entries (a rather defeatist attitude, I know). How much change would you feel comfortable with? Medellia 07:06, 13 July 2007 (UTC)

I really don't know. It depends on whether [y] or [i] is more characteristic of υ in Byzantine as a period. - Gilgamesh 07:08, 13 July 2007 (UTC)

Κρεῦσις

I have encountered this third declension word, Κρεῦσις, of uncertain inflection. I know it's third declension because its genitive is Κρεύσιος, but the nominative and genitive cases are all I know for this word, as I have not encountered this inflection pattern before. It appears similar to Template:grc-decl-3rd-prx-pure-υ, with ι instead of υ, except that the ι in the nominative is short, not long. I request review and comment. - Gilgamesh 18:56, 15 July 2007 (UTC)

Done and to do

I'm Gilgamesh, the workaholic linguistic hobbyist who's been editing Ancient Greek entries with obsessive frequency for some time now. And...I think I spread myself a little thin. I will try to continue without burning myself out, but I thought I should illustrate here some of the work (both finished and unfinished) I've been doing, so that others can take note if they need to reference/check it.

• I heavily revamped Category:Ancient Greek declension templates, using templates and subtemplates and subsubtemplates to streamline many common layouts and functions into less repetitive code. After adding inflections to many noun articles, I've also created a subcategory Category:Ancient Greek declension convenience templates, which reference existing templates but require less embedding work for certain commonly-encountered stems (such as -ίδ- and -ία). Versatile blank templates are also available for irregular nouns. In nearly all but the convenience templates, it's possible to indicate an irregular case inflection with the optional arguments NS= (nominative singular), GS= (genitive singular), DS= (dative singular), AS= (accusative singular), VS= (vocative singular), ND= (nominative/accusative/vocative dual), GD= (genitive/dative dual), NP= (nominative/vocative plural), GP= (genitive plural), DP= (dative plural) and AP= (accusative plural). As much as possible, I use known breves and macrons over unstressed unbreathed vowels as Unicode Greek support allows.
• I added IPA pronunciation templates, where Template:grc-ipa-ckb allows you to indicate a word's pronunciation in Classical, Koine and Byzantine, and where Template:grc-ipa-ck allows you to indicate a word's pronunciation in Classical and Koine/Byzantine (where the pronunciation is the same in both Koine and Byzantine). The format is {{ckb|ipa|ipa|ipa}} or {{ck|ipa|ipa}}, with IPA brackets and labels already provided.
• I've been browsing, entry by entry, page by page, through the proper names section of Woodhouse's English-Greek Dictionary, an old and very detailed dictionary that has fallen into the public domain. I'm currently at "Cronus" in the English-named-collated list, and have since backtracked to try to enter more generic terms associated with the name such as for "Constantinopolitan". And in doing that, I have since backtracked yet again to fix L3 headers to L4 headers, and in doing that, I started to add more derived terms and descendants, and putting Classical Greek terms in the {{polytonic}} template. Then I started to backtrack again to add romanizations of Modern Greek descendants, and...

I can see right now that I've spread myself far too thin. There is much to do, too little people to do it, and I've tried to do everything all at once. Combined with my tendency to go off on marathon edits and not keep many personal notes, it's a recipe for turning a hobby you love into a chore you fear you may come to despise. So, I wanted to lay out my notes here, now, and continue editing on fewer goals at a less inhuman pace. I think, that next time I continue, I will stick to my original unsidetracked goal and proceed with Κρόνος. I have an obsessive compulsion to make everything complete and perfect, and I am rarely satisfied that I do. But I can at least continue entering proper noun entries sequentially from Woodhouse's. I will endeavour to write articles good enough the first time. But if I don't, I might not have the time or nerves to go back and fix each and every one of them like I've been doing. And I'm not asking anyone else to do that work. As with all the Wiki projects, material and references can be added as needed by editors who recognize the need and have the time and energy to do it. Which reminds me...one of the most recent new templates I wrote was Template:grc-wh-page, which cites one or two specific page numbers of Woodhouse's English-Greek Dictionary as references, and links to the URLs at the University of Chicago website where the scanned images of the dictionary's pages are publically stored. - Gilgamesh 01:59, 21 July 2007 (UTC)

-ις/-ιος/-ιν inflections

I've made an incomplete template for Template:grc-decl-3rd-prx-pure-ι. I don't know the dative or vocative inflections, nor the dual nor plural inflections, though I can kind of guess that the nominative plural is -ιες. Anyone know the rest of the inflections? - Gilgamesh 07:04, 30 July 2007 (UTC)

format of translation entries (ancient greek, modern)

Please see the dicussion at Wiktionary talk:About Greek since this (may) affect you too. It shouldn't be a big deal, I just want to settle on one format. ArielGlenn 04:09, 3 August 2007 (UTC)

an unusual third declension plural noun

Ἴνσομβρες ("Insubres") is strange...  If that's the plural, then what's the singular?  Google yielded zero results for  Ἴνσομψ or  Ἴνσομβερ.  I'm skipping it for now unless I know how to inflect it. - Gilgamesh 05:38, 15 August 2007 (UTC)

greek declension appendices

I just noticed these Category:Greek declension appendices that have been here for awhile, probably predating the decision about Greek referring to the modern language on enwikt. I suppose at some point these ought to be moved but there are many things that link to them. Maybe (I hope) they are included only in templates which would make it easier. Your thoughts, anyone? ArielGlenn 10:18, 18 August 2007 (UTC)

Heh, the third declension appendix page looks like material I originally wrote on Wikipedia articles on the subject. - Gilgamesh 22:28, 18 August 2007 (UTC)

oops, forgot to put that under a new header :-) Anyways, it probably *is* what you wrote; the person who brought it over said they got it form the 'pedia aricle! ArielGlenn 01:53, 19 August 2007 (UTC)

You're correct; I would assume most if not all of the links are generated by Template:grc-noun. They really should be "Ancient Greek Attic declension" and the like. I'll double check the template, but all in all I think a move is completely reasonable and appropriate. Medellia 11:11, 24 August 2007 (UTC)

Okay, I've moved the templates and am looking around to make sure all the links to the old ones are done away with. I think I still need to go and fix the categories... but yeah. Should be more or less good to go. Please post if you notice any problems! Medellia 12:24, 24 August 2007 (UTC)

I left Category:Greek declension appendices for now. I know Modern Greek declensions don't work quite the same as AG, so if the category will never be used, please feel free to mark it for deletion. Medellia 12:30, 24 August 2007 (UTC)

Katharevousa

I'm realizing that, though a very modern form of Greek, Katharevousa is still an intentional archaism, and I've been thinking for a while that, at the least, it might be good to provide Katharevousa pronunciations below the Byzantine ones. Afterall, Classical Greek was effectively extinct as a street language from the Koine period, but even well into the Byzantine and Katharevousa, Atticism remained influential in dictating what was considered proper in literary Greek use. So I think it would be appropriate to add Katharevousa pronunciation. :3 - Gilgamesh 04:13, 21 August 2007 (UTC)

Okay, it's done. - Gilgamesh 12:50, 21 August 2007 (UTC)

It seems odd to me to have modern pronunciation (effectively that's what it is) under the Ancient Greek entries, but it's y'all's call. If you're going to go that route you might want to keep in sync with Wiktionary:IPA_pronunciation_key which is in turn synced with el:Βικιλεξικό:Οδηγός προφοράς. Happy trails! ArielGlenn 13:14, 21 August 2007 (UTC)

To labialize or not to labialize

In reviewing Byzantine IPA pronunciation, I'm still rather divided on the issue of how to represent both the labialization of υ οι υι until the 10th century, and the divergence that happened after the 10th century until the end of the Byzantine period. These two parts are almost equal halves of the Byzantine timespan, and I realized it would be improper not to reflect this quirk of development. So far, I've been marking the vowels as [i], but this is simplistic as it only reflects the majority Greek heartland pronunciation of the phoneme from the 10th century onwards. Before the 10th century, it was widely agreed (either in practice or in conservative grammarian thought) that the phoneme was pronounced [y]. From the 10th century, it becomes [i] almost everywhere except in Cumae (which never stopped pronouncing it [u] in the first place and they were spelling it ου at this time), and notably central local dialects of the Saronic Islands and Megara in Attica, where, for a time as a minority dialect, the vowels υ οι υι were all respelt ιου, which ambiguously could have been pronounced [y] (retension), [iu], [ju], [iw] or even [i.u]. For the Byzantine period, I thought of doing a compromisingly broad IPA transcription [i(ʷ)], which can mean a front close vowel that may or may not be rounded. Alternatively, I thought of possibly using [i̹], which means such a vowel that is partially rounded, but it seems more neutral and appropriately less specific to do [i(ʷ)]. I am troubled with the continuing use of just [i] in these circumstances, considering that this pronunciation is only the stark majority pronunciation in the latter Byzantine period, while the Byzantine language period spans about 1200 uninterrupted years in a single linguistic period, including nearly all of Late Antiquity (qualifying it as ancient), but also stretching well into the Dark Ages (which weren't so dark in Byzantium), virtually the entire Middle Ages and the very early Renaissance (which in contrast was not as felt by the Greeks as it was felt in the West) before this monolithic language period ends and transitions to the next period in Greek language history. So, what would be more appropriate? [y] that only clear represents the first half, [i] that only clearly represents a majority of the second half, or [i(ʷ)] which is ambiguous but can apply to the entire period? - Gilgamesh 13:07, 21 August 2007 (UTC)

"Ancient" dilemma

As I endeavour to balance this project with its ancient scope, I keep running into a dilemma. Where exactly do we draw the line of "ancient"? Virtually everyone will agree that the Greek spoken during the three centuries of Late Antiquity (from 300 to 600) is ancient. The dilemma is that the Byzantine Empire was one of the exceptions that defied the stark European boundary between antiquity, dark ages and renaissance, because the Byzantine cultural and linguistic period is one period, lasting from about 330 to 1453. That's a 1123-year monolith. I thought before that it would be trivial to split up the Byzantine Greek linguistic period between before 600 and after 600, into an "ancient" and "not ancient" part. The problem is that no such boundary really existed for Byzantium, and that the distinction between Late Antiquity and the Dark Ages is largely by Western European reference, something that does not apply to Byzantium's continued infrastructure and prosperity. If it is undeniable that 300 to 600 is Late Antiquity and that almost 300 years of the Byzantine period existed at this time, then what of the rest of the period? Is it not ancient by virtue of being after 600? Or does it remain ancient by virtue of it being one institution that continued to last when the other ancient cultures had long since transitioned? One would be hard-pressed to say that the year 1400 in, say, Italy, is ancient, because it's practically the onset of the Renaissance by then. But in Constantinople in 1400, it was still Rhomania (the "Roman" Empire)—on its last legs, certainly, but at the tail end of one period that began in 330. It does not seem conscionable for me to split this one ancient-originating period artificially based on the civilizational upheaval that happened in Western Europe, but if I include its history up to 1453, can we still continue to include it in our Ancient Greek studies? I can see potential arguments for not including it in ancient but putting the period in medieval inguistics studies because the majority of its timespan was located during the Western Middle Ages. Likewise, I can see potential arguments for objecting to it being pigeonholed in medieval because it includes crucial centuries at the beginning during Antiquity. My personal opinion would be to put the entire Byzantine period in an ancient category, in part because what it preserved through its period was indeed an aging ancient culture as compared to the rest of Europe. I don't see a problem in this because Byzantine linguistic history is such an unusual case of sustained contact, literacy and infrastructure that did not experience a sharp periodic division for more than a millennium. - Gilgamesh 06:22, 22 August 2007 (UTC)

Since all Greek is a continuum - maybe we should only have one Greek in Wiktionary? —Saltmarsh 11:30, 24 August 2007 (UTC)

It is difficult. Ethnologue isn't terribly helpful here. I do think that an el-grc split needs to take place here, but as to where things go... as horrible as this sounds, might an option be to label things as "Early Byzantine" or "Late Byzantine" and lump them all in grc? Or do we need grc Byzantine and el Byzantine? We may want to open this question up to Wikt as a whole, in order that more opinions are voiced. Medellia 12:03, 24 August 2007 (UTC)

This is such a mess... It's rare to hear of a language and cultural period being 1200 years long, but this is the case. It's like Byzantium didn't even have a different Late Antiquity and Middle Ages. However, I do understand that modern Greek is largely a product of Greek culture rediscovering itself on a literary level after losing its elite literary status under the Ottoman Empire. So even the late Byzantine period has more in common literarily with ancient Koine than it has with Demotic. Greek linguistic history is seeming to defy linguistic history norms left and right. It's not often a language's written standard is frozen for so incredibly long. - Gilgamesh 13:40, 24 August 2007 (UTC)

!!! That's it! Our problem is that we've been dealing with "Ancient Greek" when its distinction as an ancient language is so blurry. Maybe we should be dealing with this as "Classical Greek", whose written form (barring the evolution of glyph styles) has been frozen since Classical Antiquity. Because, on some levels, Ancient Greek is both an ancient and not an ancient language. It's ancient in that it was standardized so long ago, but it's not ancient in that the same standards were reinforced rigidly into relatively recent times. In its most historically important form, Greek is not a spoken language—it's a written one. It would certainly make sense for a history that for most of its time was without television and radio, and successive thalassocracies using writing and the delivery of letters by ship as a chief form of non-local communication. This is comparable to Classical Arabic and Standard Arabic being virtually the same, while being very different from what is actually spoken. Similarly, Classical Latin writing standard has remained frozen in Roman Catholic ecclesiastical use. The reason it's so hard to pigeonhole Classical Greek to an ancient time period is because such a limitation simply does not exist. If you think about it, Byzantine Greek needed no separate demotic standard as the literary elite used only the classical language. But the reason modern Greek had a separate Katharevousa was because Classical Greek had lost its top-level functions under Ottoman rule, and Greek writing became less formal, and Katharevousa was a sort of reinforced Byzantine Classical Atticism, itself a reinforcement of Koine Classical Atticism, itself a reinforcement of Classicist Atticism. The reason Katharevousa eventually failed in public life was because far too much had changed to allow it to succeed in being artificially reimposed on Greek elite life. And finally, as with Latin, modern practical uses of Classical Greek are largely limited to esoteric ecclesiastical use. - Gilgamesh 13:51, 24 August 2007 (UTC)

Wisely stated. While the fact that Wiktionary deals primarily with written language is a definite limitation to the project (as language is most truly and most often realized in spoken form), this does provide us with certain advantages in classification. Greek is an excellent example of this. While spoken Greek has changed quite a bit in its history, written Greek remained conveniently static for a very long period of time, making the writing of a dictionary on it somewhat simpler. Atelaes 03:30, 25 August 2007 (UTC)

What is our dividing line between ancient and modern? Will it contain any word that a modern Greek would use or write in a book? —Saltmarsh 05:27, 25 August 2007 (UTC)

I would strictly define the beginning of the Modern period at the fall of Constantinople in 1453, as Wikipedia also does. And no, it wouldn't contain any word a modern Greek would write. Since the Koine period when Atticism took hold, it was not considered acceptable for properly educated to people to write in anything but Classical Attic. There were coinages and neologisms, but they were incorporated into the language in ways consistent with the Classical and Christian canon. In modern contexts, the closest to this would be Katharevousa, which was standardized because it was still felt by influential Atticists that the common Greek language was too changed and loaned too many words from other languages, and was thus somehow "impure" (Katharevousa means "pure one"). Katharevousa includes some very modern neologisms, but hasn't had any official sanction outside of ecclesiastical use for decades now. Modern Demotic Greek, where very decidedly non-classical writing conventions are used reflecting common language, would most certainly not be considered ancient or classical. ὀκτώπους (oktṓpous) ("octopus") is Ancient and Atticist. χταπόδι is definately modern. λέων (léōn) ("lion") is Ancient and Atticist. λιοντάρι is definately modern. Atticist use tended to try to retain very old rules of writing and very conservative vocabulary, even at times (especially during the Late Byzantine period) when phonology and sandhi took great liberties on how they were actually pronounced when spoken aloud. In addition to yielding more to modern language, Modern Greek is also now very unashamedly monotonic in writing. - Gilgamesh 13:58, 25 August 2007 (UTC)

In an ideal world we would have a separate sector for the Medieval phase of Greek language containing at least the words of "δημώδης" (dimodis), which was the "ancestor" of demotic. But this doesn't seem to me realistic, for the moment. So, the wiser thing to do is to continue our work in Ancient Greek keeping there the archaistic forms of Μedieval Greek and our work in (modern) Greek including - for the moment- the words of δημώδης. Ι hope that the Lexicon of Medieval Greek by Kriaras (Λεξικό μεσαιωνικής ελληνικής δημώδους γραμματείας του Εμμανουήλ Κριαρά) will be completed sooner or later - until now 15 volumes have been published. Then we shall be able to discuss this issue in a more stable basis. --Flyax 21:22, 28 August 2007 (UTC)

Yeah, I admit that my study of Byzantine Greek could be better than it is... Will this material be available online? Academic shelfware does not suit those of us living in proverbial Antarctica. :3 - Gilgamesh 00:33, 29 August 2007 (UTC)

A part of it is available here. For an example look here. --Flyax 12:22, 29 August 2007 (UTC)

template:polytonic

Please see Wiktionary talk:About Greek#template:polytonic. Rod (A. Smith) 22:18, 12 September 2007 (UTC)

Loaned [j]

I get the impression that, in virtually all cases, ι is never pronounced [j] in modern pronunciation of Classical Greek words. But what about words loaned from Latin that converted j to ι? What about words loaned from Hebrew that converted consonantal י to ι? Does modern Greek pronunciation of the classical words pronounce these as [j]? - Gilgamesh 23:44, 8 October 2007 (UTC)

Πυθώ

I've decided to skip this word for now (at page 1023), lacking a proper understanding of how the spelling is used. Wikipedia redirected "Pytho" to "Python (mythology)", and I can't seem to reconcile the nuances of Πυθώ with Python's claimed etymology Πυθων. So, for now, I'm moving on, but I thought I should make a note of this. - Gilgamesh 10:10, 12 October 2007 (UTC)

Στάγειρος

Skipping this for now. This page says Στάγειρος, but all Wikipedia references refer only to Στάγειρα. I don't know if the dictionary misprinted or not. - Gilgamesh 04:23, 18 October 2007 (UTC)

It's not an error. See in Scholia In Thucydidem 4.88.1: Στάγειρος Ἀνδρίων ἀποικία, ἡ τοῦ φιλοσόφου Ἀριστοτέλους πατρίς. Also in Aelius Herodianus 3,1.389 Στάγειρα πόλις Μακεδονίας. λέγεται καὶ Στάγειρος. So it's an alternative form. --flyax 12:19, 18 October 2007 (UTC)

So both are correct? Both feminine? One's just first declension, the other's second declension? - Gilgamesh 21:32, 18 October 2007 (UTC)

Στάγειρα is neuter, plural, second declension. --flyax 13:12, 19 October 2007 (UTC)

Θαλῆς

This is obviously a historically important name, but I have no idea how to properly inflect it. Genitive form is Θάλεω, that's all I know. Skipping for now. - Gilgamesh 07:13, 19 October 2007 (UTC)

Genit. Θάλεω or Θάλητος or Θαλοῦ; Dat. Θαλῇ; Acc. Θαλῆν; Voc. Θαλῆ. --flyax 13:54, 19 October 2007 (UTC)

 *scared of wackily irregular nouns* ;.; hehehe, Okay, I'll get to it when I continue adding entries. :3 - Gilgamesh 21:29, 19 October 2007 (UTC)

Romanization of MGreek descendants

I just noticed that there are multiple romanizations used for Modern Greek descendants (see Γεώργιος) but we have a single romanization scheme here, as do most languages. See WT:AEL#Romanisation. Thanks, ArielGlenn 18:00, 17 November 2007 (UTC)

The problem is that, in practice, no one quite agrees on this. Both the United Nations and Board of Geographic Names are widely-used standards for Greek romanization. The United Nations (UN) standard is used by the Greek government, and the Board of Geographic Names (BGN) is used in the United States for transcribing Modern Greek and by the important tourism industry in Greece. I'm not sure United Nations vs. United States has ever been a fun debate as it invokes high passions, so I indicate both systems to satisfy both standards. I suppose though that the "Other" transcriptions can go. - Gilgamesh 11:44, 18 November 2007 (UTC)

By "here" I mean on Wiktionary (en.wiktionary, specifically, since the other projects may adopt different systems). ArielGlenn 12:44, 18 November 2007 (UTC)

Yes, I knew what you meant the first time. - Gilgamesh 14:27, 18 November 2007 (UTC)

The point is that romanizations of Modern Greek entries here should follow that one system as described in WT:AEL. That is what that section is there for. It's true that it is a draft policy. But if you have disagreements with it (for example, if you believe we should instead put two romanizations, one with the UN system and one with the BGN system), you should say so on the policy's talk page (so that all interested parties can weigh in), rather than silently ignoring it. (Feel free to just add a link back to this page.) Thanks, ArielGlenn 18:53, 18 November 2007 (UTC)

Mycenaean.......Greek? Redux

Please see the following discussions: AAG, RFDO. This is also interesting.

Yes, yes, I know we've had this discussion before, but we've never reached a conclusion. And since Ivan Štambuk has been so industrious as to start improving the.......section, I feel we owe it to them to hammer out there details so they don't have to be changed later on. Is the language Mycenaean or Mycenaean Greek? I won't repeat my arguments as they've already been presented. Suffice to say that I am for Mycenaean Greek. Atelaes 08:57, 20 January 2008 (UTC)

So as I understand it, it doesn't matter *logistically* what we change the header and the categories to. Since Mycenaean is a (very old) form of Greek and has a good ISO number [2] as such, I am happy with "Mycenaean Greek". ArielGlenn 11:05, 20 January 2008 (UTC)

Seems fine, it is the ISO name. As I noted above, the category names need to match. Robert Ullmann 12:05, 20 January 2008 (UTC)

I was taught (centuries ago) that the language of Linear-B tablets is a very ancient dialect of Greek, related to the Arcadocypriot dialect. So, it's clearly "Mycenaean Greek".--flyax 12:07, 20 January 2008 (UTC)

Mycenaean Greek, of course. Already present articles were categorized under Category: Mycenaean Greek language, but had L2 header "Mycenaean", which was inconsistency I eliminated. I also created Template:Linb (just like official ISO 15924 name) to be used with sc= parameter of {infl}, {t} etc. Mycenaean Greek is the most archaic Ancient Greek dialect, and that fact should be reflected in it's name. That naming scheme also makes is convenient to add translations with those of Greek and Ancient Greek, like this. --Ivan Štambuk 15:29, 20 January 2008 (UTC)

I don’t have a strong preference. Wikipedia calls it w:Mycenaean language, and my book on the Aegean scripts just calls it Mycenaean. Agreed, it’s an ancient form of Greek, but it seems to me to be quite different from Classical Greek. I can see the strong connection, but I can’t make out much of it just going from a knowledge of Greek. A sentence that I have in front of me says, "hijereja hekhej kwe eukhetoj kwe etōnijon hekhehen theōj, ktojnohokhoj de ktojnāhōn khekhemenāhōn onāta hekhehen 3.95 measures" (literally, "priestess has and swears and special-landgrant to-have for-the-god, landholders but of-cultivable-lands of-communal landgrants to-have 3.95 measures" = "the priestess has and swears she has a special landgrant for the god, but the landholders say that she has 3.95 measures in landgrants of communal lands"). It seems to be rather removed from Greek, but it’s probably closer to Ancient Greek than Old English is to English. So either term is okay with me. —Stephen 17:11, 20 January 2008 (UTC)

That would look like this:

Linear B	Template:Linb	Template:Linb	Template:Linb	Template:Linb	Template:Linb	Template:Linb	Template:Linb	Template:Linb
Transliteration	i-je-re-ja	e-ke-qe	e-u-ke-to-qe	e-to-ni-jo	e-ke-e	te-o	ko-to-no-o-ko-de	ko-to-na-o
Transcription	hijereja	hekhej kʷe	eukhetoj kʷe	etoːnijon	hekhehen	theoːj	ktojnohokhoj de	ktojnaːhoːn
Gloss	priestess	has	swears-and	special.land.grant	to.have	for.the.god	landholders.but	of.cultivable.lands

Linear B	Template:Linb	Template:Linb	Template:Linb	Template:Linb "3 T 9 V 3"
Transliteration	ke-ke-me-na-o	o-na-ta	e-ke-e	GRANUM 3 T 9 V 3
Transcription	khekhemenaːhoːn	onaːta	hekhehen	GRANUM 3 T 9 V 3
Gloss	of.communal	land.grants.	to.have	GRANUM 3 57/60 measures

hijereja would be (deprecated template usage) ιέρεια, theoːj would be θεός (theós) etc. Linear B is a lossy syllabary script derived from script(s) which were originally meant to write non-IE language. Just like syllabograms for e.g. Hittite - lots of important phonetic information is lost, and homophones are present all over the place. But underneath it's really archaic Greek dialect. ISO name is also 'Mycenaean Greek'. Wikipedia article name change has already been discussed on the talk page linked above, but unfortunately 'Mycenaean Greek language' is already used as a redirect there, which non-sysop cannot redirect over. --Ivan Štambuk 22:05, 20 January 2008 (UTC)

Well, anyone can drop the content of the page w:Mycenaean language into w:Mycenaean Greek and replace it with a #REDIRECT link to the latter. >:D --flyax 22:40, 20 January 2008 (UTC)

Yes, but that would screw up the edit history. Atelaes 00:26, 21 January 2008 (UTC)

Didn't appear controversial, so I put on my WP admin hat and moved it. -- Visviva 10:29, 21 January 2008 (UTC)

Modern Greek pronunciation in {{grc-ipa-rows}}

You know I'm thinking that perhaps we ought to drop modern Greek pronunciation in the pronunciation templates for Ancient entries. My reasons for this are as follows: First and most importantly, it pretty much requires that we sync this component with any pronunciation given in the modern equivalent (if it exists). We would look pretty foolish if we were to give two modern pronunciations which differed on the same page. Secondly, it seems to me that the authority to determine pronunciation on modern entries should stay well vested within the Greek community, and the Ancient Greek community should just keep their hands off of it. Otherwise, I think problems could arise. Thirdly, pronunciation of modern languages is just a different animal than that of dead languages. With living languages, recordings of actual speakers can be used and analyzed, exceptions can be noted for specific words, etc. With dead languages, the best we can really do is create a rule based system for extrapolating based on the written word. I'd like to apologize in advance to Gilgamesh, as I'm sure they put a lot of thought, time, and effort into integrating this feature into the template (which is surely appreciated), but nonetheless, I think it a bad idea. Atelaes 00:38, 21 January 2008 (UTC)

The argument about two different modern pronunciations for the same word (and particularly on the same page) being problematic for the reader is important, I think. The rest I agree with but this is the clincher for me. So I too would like the modern pronunciation to be added only to the el entries. ArielGlenn 10:41, 21 January 2008 (UTC)

Well, the template system is adaptable, and I provide different tokens (like kj nj j) for situations where Modern Greek articulation of Classical words (as opposed to Modern Greek articulation of Modern words) has nuances that are particular to modern times. Also, in multiple pronunciations of the same word, if you're talking about the doubled consonants ([s(s)] etc.), that is a very real modern distinction for both modern pronunciation of Classical Greek and pronunciation of Standard Greek (as opposed to Greek dialects). It is meant to be inclusive of the way the modern Greek world pronounces polished Standard Greek. For example, in Athens, it is common for the prescribed [nd ŋɡ] to be articulated [d ɡ] even when a speaker is trying to be clear, isn't this right? And in Cyprus and the Dodecanese (and we're not even talking about the Cypriot Greek dialect or the Dodecanese Greek dialect, which is a different animal), it is routine to pronounce polished Standard Greek doubled consonants as geminated (including, as I have been informed, intervocalic ζ as [zz]) instead of ungeminated, the latter being more common in Greece Proper. The parentheses are merely part of a broad reflection of the Standard polished phonology, and not of the various folk dialects of varying mutual intelligibility. Anyway, getting back to the token system...keeping in mind that we are talking about pronunciation of Classical words in modern Greek lands by Greek speakers who enunciate, and not Modern Standard words, can't we just address different characteristics individually? Then, even a Standard Greek speaking editor can look at the Contemporary pronunciation and edit the pronunciation for himself to appropriately reflect how it is pronounced, and using the highly flexible token system. - Gilgamesh 19:23, 21 January 2008 (UTC)

While I admit that the token system is indeed highly flexible, it's going to have a hard time dealing with the exceptions of real speech. Imagine a word which has a singular exception to the general grapheme > phoneme rules. You'd have to write a whole new rule into the system just for that word. And while you certainly could carry this on for a while, eventually, you'd run out of characters . Imagine if you tried to make a token based system for English pronunciation. You might be able to do one which covers.....ohhh.....95% of the language, but the remaining 5 would squash everything. Now, I imagine modern Greek is probably a bit better behaved than English, but it nonetheless has exceptions and variations which all living languages do. Secondly, if we're talking about modern Greek speakers and how they pronounce Ancient words, well, then we're talking about someone saying something which is outside their native tongue, in effect. I guess I'm not completely aware of how the distinction goes between ancient and modern words. Perhaps Flyax can shed some light on this. But, if my interpretation is correct, then we may as well have a section on how Americans pronounce Ancient Greek, or Japanese. We're only interested in native pronunciation. Atelaes 20:40, 21 January 2008 (UTC)

A few months ago we discussed this same issue. At this moment the label "Katharevousa" was in use, something that could cause misunderstandings about the true nature of this "language", so I had proposed another label, as a compromise. Now, with the discussion having started again, we have the opportunity to reconsider some things. Ιn my opinion, there are two main problems with the existing markup of contemporary pronunciation. The first one has to do with [ia], [ʝa], [ça] etc. For example: while the modern Greek pronunciation of the word αλήθεια is [al'iθça], it's quite uncertain if the contemporary pronunciation of the ancient word ἀλήθεια should be [al'iθιa], as I'd prefer, or [al'iθça], as a lot of people would pronounce it. Insisting οn such details would be over-pedantic, though. The second "problem", which I have discussed a lot with Gilgamesh, without reaching an agreement, is the necessity or not of indicating geminated consonants. I have thought a lot of times that if we didn't have the contemporary pronunciation at all, I would be happier and more available to contribute in other, more useful, ways. But, we'll see. --flyax 22:57, 21 January 2008 (UTC)

These are all sensible points, and I suppose I could bend on it. But I don't think it's necessary at this time to completely scrap the contemporary token system, but leave it in as dummy code. My opinion is still to keep it, and adapt case-by-case as a need arises (as I have thus far, and it has worked well). However, if it is consensus that it should be hidden for the time being, then I'll cooperate. I suppose it comes down to how influential Atticism (intentional use of the frozen Classical language instead of the vernacular) was in Byzantine times as opposed to Contemporary times (when for a time people opted for Katharevousa instead). Are we certain there are no lasting notable Contemporary Atticist texts, if even only perhaps in religious or liturgical use? - Gilgamesh 17:09, 22 January 2008 (UTC)

When words or phrases either of Ancient Greek or Koine are in somewhat broad use, then they will go into the Modern Greek section as fixed phrases or in other appropriate ways, but there is no need to keep markups for the modern pronunciation of the rest of the corpus. Students of Ancient Greek or Koine are learning them in their historical context, not to revive them as living languages today. I don't have any preference one way or the other about keeping the token system as dummy code or removing it. ArielGlenn 21:18, 22 January 2008 (UTC)

Agreed. If a word is in modern use, it will get a modern cite, with a pronunciation determined by and discussed at About Greek. And yes, there is no need to completely scrap the code. Who knows? It may come in handy for something down the road. Atelaes 21:43, 22 January 2008 (UTC)

I agree too that we should hide the contemporary pronunciation. --flyax 23:00, 22 January 2008 (UTC)

Then it's settled. I will hide the code for the time being. - Gilgamesh 15:21, 23 January 2008 (UTC)

And it is done. The Contemporary row is commented out. - Gilgamesh 15:24, 23 January 2008 (UTC)

Many thanks Gilgamesh. It is quite the ingenious template, in case someone has not yet made that comment. Atelaes 17:47, 23 January 2008 (UTC)

I know, I know... It is actually chiefly inspired by similarly-designed Hebrew token protocols I've used in my own work for years. ^'e ph . + r a.+ y i m$ (or something similar), depending on the cipher, yields אֶפְרַ֫יִם, ʼEṗrạyim, ʔɛɸɾaːjim, Efráyim or efˈʁajim. I had a lot of practice with these principles before I even began to design Greek tokens. It can be adapted for almost any written/transcribed language, including English and reflecting differences in standard dialects too. I mean, just considering English, what are all the distinct vowel phonemes you can think of that correspond to the standard and conventional dialects? General American, General Canadian, British Received, Hiberno-English, Australian, New Zealander, or even Birmingham city dialect or Northumberland, or things like that. There are various phoneme mergers, and occasional phoneme splits in some dialects too. And how would one go about representing all the corresponding phonemes in tokens? You gather the phonetic information from all the dialects, compare correspondences, then come up with a system that represents all of them. Naturally, there are some words that are pronounced more than one accepted way, but you just indicate their pronunciation more than once in the different ways, as dictionaries typically do anyway. See, when you become accustomed to planning and analyzing in such a way, then it eventually becomes easy to design (in concept) a token system for whatever you need. Actually, the more I think about this for other languages, the more fun it sounds. Maybe I'll draft new token table pages on my user account space sometime... - Gilgamesh 01:53, 24 January 2008 (UTC)

Late Byzantine?

I know that, for the time being, we've largely dummied out the Contemporary pronunciation, because Atticism (intentional use of frozen Classical Attic for contemporary new texts) has not been influential since the fall of the Byzantine Empire. However, something dawned on me. Though the Byzantine period is one linguistic period, it is long, from 330 to 1453. There were documented changes in phonology that were relevant in the latter half (second millennium) that didn't really apply in the first half. It might be interesting to add a new row for "Late Byzantine" to reflect second millennium Atticism. What we do know is Late Byzantine had lost [y], which had become [i] in virtually the entire Greek world except for the Old Athenian pockets (Athens, Megara, Aegina, Cyme and Mani) which were insulated by other local populations that spoke other languages, especially Albanian (now Arvanitic there), but also Slavic dialects, Vlach dialects, Turkish dialects, etc. ...where it either stayed [y] or became [u]. What else is known is that ungemination had taken root in the Greek mainland (but not in the Dodecanese or Cyprus, which retain it to this day). But what else is attested in Late Byzantine Greek that would have been reasonably reflected in how they pronounced Attic? I know that by the 13th century, ι before a vowel in Byzantine Demotic dialects had lost its unique syllable and even its applicable accent to the next letter. Of course it is not reasonable to believe that this would still apply to a conservative pronunciation of Atticist Greek at the time, but it makes me wonder if palatalization ([k→c]) had begun already. Palatalization is attested from at least some centuries ago by Greek names loaned into other languages, such as Russian Юрий (Yuri), which was derived from Γιούρης (Yuris), a contemporary Greek dialectual form of Γεώργιος, and we know that Kyivan Ruś converted to Christianity and started absorbing myriad Greek names during the Late Byzantine period (though it would be nicer to know the more detailed history and development of names like Yuri). And Modern Greek pretty much started under Ottoman rule, when the literary languages of choice became Arabic and Turkish, and Atticism became more valued in the West as a curiosity than it was in the Greek world, which is probably why modern pronunciation of Classical Greek is not so significant and we decided to axe it. Later classicist Atticist coinages (like 19th century Μινώα) were wholly compatible with the ancient language, but were coined by non-Greeks: scholars, archaeologists, etc. Anyway, I ramble too easily... What we also know is that Katharevousa-style phoneme splitting was not used during the Byzantine era, or indeed later until Katharevousa, hence β is Late Byzantine [v] but μβ is Late Byzantine [mb], etc., and these retained hardenings became respelled μπ etc. later as it too became pronounced [mb] and Katharevousa also reclaimed μβ as then-unnatural (but now normal) [ɱv]. But that also raises the question—do we know if modern-style plosive voicing [mp→mb] was happening during Late Byzantine? Though that would of course merge μβ with μπ in pronunciation and I'm not certain if it had happened by the 13th century or not... - Gilgamesh 03:30, 3 February 2008 (UTC)

I thought of some more things. We can look at what we know of the Greek varieties. Pontic split from Byzantine, and it has palatalization. Cypriot split from Koine even earlier, and it also has palatalization. I don't believe palatalization was necessarily universal by early Byzantine, but it seems to have certainly existed at least in some dialects. So, I have an idea for a Late Byzantine pronunciation model. It is the same as normal Byzantine, except:

• [y] is [i] everywhere. This was not the case in Old Athenian as mentioned above, but I'm speaking of polished Byzantine Atticism, and the Old Athenian enclaves had become a largely inconsequential backwater compared to the formal Byzantine world, right?
• The velars [k ɡ x ɣ] have certainly become palatals [c ɟ ç ʝ] before front vowels before the end of the Byzantine period.
• There is greater tolerance of /i/ as [ʝ] in specific circumstances—it has even become the typical pronunciation on the demotic streets for all occurances of /i/, stressed or not, before another vowel but not after /r/. Atticist pronunciations of the time would however have been unlikely use this where Katharevousa did not.
• Gemination is lost in the mainland, but retained in Cyprus and the Dodecanese where it survives to this day. As with the Contemporary pronunciation which we dummied out, parentheses can be used for geminates when pronounced in Atticist texts, e.g. [(s)s].
• Early Byzantine bilabial fricatives [ɸ β] have settled into labiodentals [f v].

What is not so clear and I wouldn't venture, includes:

• Whether μπ etc. is [mb] etc. during the second half of the Byzantine Period. We know that μβ is [mb] at this time, and remains so into pre-Katharevousa, and many old μβ spellings became μπ to compensate. We also know (if I read correctly and didn't misread) that μπ as [mb] is universal in all parts of modern Greece and Cyprus. But were μπ and μβ already merged late Byzantine? I don't know. I'd rather know for certain whether or not this was the case. If they were not merged, then it is almost certain that μπ was [mp] and μβ was [mb]. But if they were merged, then both were [mb]. (β alone was [v] in either case.) On the other hand, if Atticism of the time was to be pronounced conservatively, then it is probable that they tried to distinguish the pairs in some way, though nasal-fricative [ɱv] as Katharevousa used seems unlikely back then, and this was unstable until Katharevousa.

What do you think? - Gilgamesh 05:33, 5 February 2008 (UTC)

Sorry for the delayed response on this. Here are my thoughts: The first (and primary) problem with this is that it's beginning to sound (to my ears) like original research. Very insightful original research, but nonetheless. If you can cite a respected source which lists such a pronunciation scheme for this period, that would be different. Secondly, I think it a bit overkill. All of the sections of our pronunciation cover vast swathes of time, and are admitted to be less than perfectly precise. However, I think it best to be a little fuzzy when dealing with hypothetical phonological reconstructions, as it is just that, hypothetical. It would be foolish to think that we can make extremely precise assertions about how people sounded two thousand years ago. Besides that, the pages will simply look too cluttered if we add too much to the pronunciation section (admittedly four won't topple the whole thing, but you get my idea nonetheless). While I have to imagine that you've put a great deal of thought into this, I think it not a good idea for our project. Atelaes 07:54, 5 February 2008 (UTC)

I have to agree with Atelaes here. In the absence of a strong published authority for these historical pronunciations, this is not a good direction for Wiktionary to go. -- Visviva 12:53, 5 February 2008 (UTC)

Alright, agreed. It was worth input. - Gilgamesh 19:35, 5 February 2008 (UTC)

ὄζον

Since participles have been on and off your mind for some time, I thought I'd throw this format at you for your critique and amusement. -Atelaes λάλει ἐμοί 23:49, 28 April 2008 (UTC)

That's along the lines I've been considering for Latin, but it doesn't really make me happy yet. The problem with participles is that they're not quite adjectives (since they have tense) and not quite verbs (since they have gender). I also can't decide which participle form (if any) should be considered the "lemma". --EncycloPetey 00:42, 29 April 2008 (UTC)

I feel absolutely no guilt in using "Participle" as the POS, because....well.....that's what they are. Participles can be used in nominal, verbal, and adjectival senses (in grc at least). Using anything other than participle is, in my opinion, misleading. While some may have and undoubtedly will again argue that the average reader will be confused by "participle," that ignores the fact that the average user will misled by anything else. Perhaps once I write Appendix:Ancient Greek participles (as part of my grand (read naïvely optimistic) scheme of creating and populating Category:Ancient Greek grammar appendices) it will soften their criticisms. Also, I see no reason to deviate from the standard of masculine nominative singular as the lemma. In any case, I'm glad the format does not strike you as odious, at least. Also, if you have some time, would you be willing to give the la terms listed on Appendix:Proto-Indo-European *h₃ed- a bit of help? I'm trying to con Ivan into cleaning it up, and I think it might be more worthwhile for him to do so if there are some nice daughter words in need of a more wholesome mother. -Atelaes λάλει ἐμοί 01:11, 29 April 2008 (UTC)

Yeah, but which masculine nominative singular should be the lemma? The passive perfect, present active, or future active? Or do I make all three lemmata? Most major dictionaries seem to list the present active and passive perfect as headwords, and if we use both then what connection should be given between the two forms? It's a headache I haven't quite wrapped my brain around (eww...).

I've already been thinking the same thing about Ivan and PIE. I've started doing some of the core substantive nouns (arbor, avis, bos, equus, etc.), and started thinking about how to get more PIE into these. The catch at the moment is that I'm already stretched thin doing WOTD, categorizing, verb cleanup, etc. I may only be able to do one quality entry every 1-3 days off the PIE appendix, at least until the verb cleanup and categorizing are completed. (...and then Conrad is eager to get the noun templates redone...sigh) --EncycloPetey 01:19, 29 April 2008 (UTC)

Yes, there simply are not enough hours in the day. Although, perhaps you should be counting your lucky stars. If I ever get a bot figured out for the import of L&S, as I intend to this summer, I'll be jumping down your throat to get every template and formatting policy for all of Latin into pristine condition.  :-) As for active vs middle vs passive vs medio-passive vs.....I haven't given that too much thought, but I think it would be best to treat each one separately. Perhaps {{grc-part}} (once created) could take parameters for the other voices, so they're consistently linked to each other. So you'd get something like:

Participle

ὄζων m, ὄζουσα f, ὄζον n, (middle/passive: ὀζόμενος)

Eh? Whaddya think? -Atelaes λάλει ἐμοί 01:30, 29 April 2008 (UTC)

But that would preclude having the feminine and neuter nominative on the inflection line, wouldn't it? Anyhoo, I've done ōdī and a bit on odor. You would pick a defective irregular verb for me to start with, wouldn't you? :P --EncycloPetey 02:17, 29 April 2008 (UTC)

Ummm.....why would the feminine and neuter need to be in the inflection line? They're in the inflection, stupid (i.e. just like a standard adjective/pronoun inflection). And did you really think I'd give you an easy verb? :P -Atelaes λάλει ἐμοί 03:10, 29 April 2008 (UTC)

Yes, but the standard ajective inflection line gives the feminine and neuter. That's the problem: the inflection line is vastly different for adjectives and verbs, so a participle has to blenderize the two very different approaches. --EncycloPetey 03:11, 29 April 2008 (UTC)

Oh yeah, forgot about that. Well, would it be pushing it too far to have all four? I mean, grc verbs have six forms in their inflection line. -Atelaes λάλει ἐμοί 03:26, 29 April 2008 (UTC)

The problem is more that the various forms are not parallel. One set are the gender inflections of the current entry, the other set are the tense forms of the current entry. It's not like other inflection lines where there's variation in just one dimension. It makes the whole thing icky. --EncycloPetey 03:32, 29 April 2008 (UTC)

Hmm....well the alternative voice forms would be set in parentheses, setting them apart a bit. But yeah, I see your point. Well, I guess I'll keep this rolling in the back of my head for awhile. It's not like grc has that many participles floating around just yet (don't really know about Latin), so it's not terribly urgent. By the way, would you object to me copying this convo to Wiktionary talk:About Ancient Greek? It seems like something like this should be archived there. -Atelaes λάλει ἐμοί 03:39, 29 April 2008 (UTC)

That'd be fine. --EncycloPetey 03:40, 29 April 2008 (UTC)

Copied with permission from User talk:EncycloPetey#ὄζον. -Atelaes λάλει ἐμοί 03:49, 29 April 2008 (UTC)

Late Byzantine (revisited)

I added information for Late Byzantine (post 10th century) in Category:Ancient Greek IPA tokens. I then briefly tested it by adding it to Template:grc-ipa-rows, and then commented it out at User:Atelaes's suggestion, pending a discussion. I thought about the issue of not using original research, and so I based it on Early Byzantine (from 330) based on existing research and simple deductive reasoning. Seeing as I am autistic, I don't know how obvious such reasoning would be to other people, but it seems pretty straightforward to me.

• Attested: [y] becomes [i] everywhere except in the increasingly geographically isolated and politically insignificant Old Athenian phonologies (enclaves surrounded by Albanian-speaking regions at the time).
• Attested: Prenasalized voiced plosives [mb nd ŋɡ] do not change to prenasalized voiced fricatives [mv nð ŋɣ]. Such a development is only attested much later on Modern Greek in the 19th century under influence from Katharevousa. If the Atticist (read: not dialectual) pronunciation of prenasalized voiceless plosives remained voiceless, then [mp nt ŋk] remain and contrast clearly with [mb nd ŋɡ].
• Deductive reasoning: Palatalization [c ŋc ŋɟ ç ʝ] of velars [k ŋk ŋɡ x ɣ] before front vowels [e i]. This exists in every modern Greek phonology descended from Byzantine. It also exists in Cypriot Greek phonology, descended directly from Koine.
• Common guess: Bilabial fricatives [ɸ β] become labiodental fricatives [f v]. This probably is original research, as it also appears in less polished phonological reconstructions of early Byzantine as well. However, I can provide no justification for Late Byzantine labial fricatives to differ from Early Byzantine labial fricatives. Revert them to bilabial?
• And that's about it. Other Early Byzantinisms conservatively retained and unchanged. As we discussed before, Atticist usage would have resisted dialectual vulgarisms, except where changes occur on a fundamental phonological level. Since palatalization certainly occurs everywhere by this time in all positions where it happens today, it seems extremely unlikely that Byzantine Greeks at the time would have done anything to distinguish palatals from velars, as there simply was no need to do so.
• (Early Byzantine) IPA^(key): /vyz̚zántion/
• (Late Byzantine)
• (Early Byzantine) IPA^(key): /kõstantinúpolis/
• (Late Byzantine)
• (Early Byzantine) IPA^(key): /xalcíðon/
• (Late Byzantine)
• (Early Byzantine) IPA^(key): /áɲcyra/
• (Late Byzantine)

- Gilgamesh 00:49, 10 May 2008 (UTC)

To begin with, I just don't feel comfortable with anything that's not posted in a book. I'll come right out and say that my understanding of Greek phonological evolution is not what it should be, so I'm a long way off from seeing the obvious jump to labiodental fricatives. This is not to say that I think it implausible, simply that I lack the understanding to make an assessment one way or the other. Additionally, I feel uncomfortable relying so much on Wikipedia for our research. While I think the 'pedia an excellent resource and commonly use it myself, I think that for this we really need to hit the books.....real books....made of paper. It is fully my intention to go to the library and get a bunch of books on Greek pronunciation and try and make a synthesis from them some time after school's done (a week from today). Would it be possible to hold off on this until we can pull in some more resources for this? What I'd like to see is something along the lines of how Saltmarsh handled the transliteration of Greek. Made a table, compared five different references on the subject, and then made a decision. I'd like to do that for our pronunciation. -Atelaes λάλει ἐμοί 01:43, 10 May 2008 (UTC)

Okay, I'll wait. But you know how I despise relying so much on academic shelfware. I'm impractically far away from libraries. Information distributed on the Internet is helpful for a very good reason—it's paperless and it doesn't have to be physically stored on a shelf. Just assume I live in Antarctica for all the academic convenience I have. :3 - Gilgamesh 01:55, 10 May 2008 (UTC)

I live on a major university campus, so for all the academic convenience I have I curl up in journal papers when I go to sleep. :) When I get the time I'll put the information up here, and we can discuss it. -Atelaes λάλει ἐμοί 02:32, 10 May 2008 (UTC)

Braggart. :3 Get the info AYEC. - Gilgamesh 07:58, 10 May 2008 (UTC)

I just wanted to add my two cents: I feel very uncomfortable with the idea of adding pronunciations based on our own speculation or conjecture, however logical those conjectures might be. In my opinion we should rely on recognized academic sources for this information, representing the concensus if there is one. If there is not a general concensus as to the pronunciation in academic circles, there is probably a very good reason for that. -- ArielGlenn 06:29, 10 May 2008 (UTC)