User talk:Gilgamesh

Definition from Wiktionary, the free dictionary
Jump to: navigation, search


Hello, welcome to Wiktionary, and thank you for your contributions so far.

If you are unfamiliar with wiki editing, take a look at Help:How to edit a page. It is a concise list of technical guidelines to the wiki format we use here: how to, for example, make text boldfaced or create hyperlinks. Feel free to practice in the sandbox. If you would like a slower introduction we have a short tutorial.

These links may help you familiarize yourself with Wiktionary:

  • Entry layout explained (ELE) is a detailed policy documenting how Wiktionary pages should be formatted. All entries should conform to this standard. The easiest way to start off is to copy the contents of an existing page for a similar word, and then adapt it to fit the entry you are creating.
  • Our Criteria for inclusion (CFI) define exactly which words can be added to Wiktionary, though it may be a bit technical and longwinded. The most important part is that Wiktionary only accepts words that have been in somewhat widespread use over the course of at least a year, and citations that demonstrate usage can be asked for when there is doubt.
  • If you already have some experience with editing our sister project Wikipedia, then you may find our guide for Wikipedia users useful.
  • The FAQ aims to answer most of your remaining questions, and there are several help pages that you can browse for more information.
  • A glossary of our technical jargon, and some hints for dealing with the more common communication issues.
  • If you have anything to ask about or suggest, we have several discussion rooms. Feel free to ask any other editors in person if you have any problems or question, by posting a message on their talk page.

You are encouraged to add a BabelBox to your userpage. This shows which languages you know, so other editors know which languages you'll be working on, and what they can ask you for help with.

I hope you enjoy editing here and being a Wiktionarian! If you have any questions, bring them to the Wiktionary:Information desk, or ask me on my talk page. If you do so, please sign your posts with four tildes: ~~~~ which automatically produces your username and the current date and time.

Again, welcome! It's always good to see another person interested in Ancient Greek. Thus far, Medellia and I seem to be the only ones working on it (and Medellia is spread out over a number of languages). You may find Wiktionary:About Ancient Greek interesting, if you haven't already seen it. Any questions, feel free to ask. Atelaes 21:50, 8 June 2007 (UTC)


Could you give me an example of an entry which has been screwed up? I took a random sampling of the A. Greek nouns, and they still look ok to me. Atelaes 18:47, 9 June 2007 (UTC)

Random Greek stuff[edit]

I realize that moving to a new project involves picking up a lot of stuff as you go along, so I apologize if you've already figured out any of the following. First, when making an external link, you only need one set of brackets, not two as for internal links. Second, there should always be a space between the inflection line (the line which comes directly after the POS header, and having nothing to do with the inflection section, confusingly enough) and the definition. As an example, take a look at the changes I made to Κίλιξ. Also, for consistency's sake, I'd appreciate it if you followed the pronunciation guide set forth at Wiktionary:About Ancient Greek, or at the very least, please don't change existing entries against it. I realize that there are a number of different theories on Ancient Greek pronunciation, and even within a single scheme, there are various pronunciations across places and time sub-periods. However, I think it would be best to be consistent so we don't present the material as if differences existed where they don't. If you think that a different phoneme should represent some character (or digraph), please bring it up on the talk page of the aforementioned page. I will be the first to admit that I am no Greek scholar, and it's highly possible that the scheme could be improved. Finally, thank you very much for all the new entries. It's rather nice to not be shouldering the whole language by myself. I'm sure I'll come up with more stuff as time progresses, so I hope you won't consider me totalitarian or didactic. Atelaes 09:39, 10 June 2007 (UTC)

Oh, one other thing: The final parameter in all the POS templates ({{grc-noun}}, {{grc-pnoun}}, etc.) should be the entry title minus any diacritics (breathing marks, accents, etc.). This is because the Wiktionary software has a rather crazy ordering scheme with regards to Greek vowels versus vowels with accents and the like. This is why, if you look at Category:Ancient Greek nouns, Κίλιξ comes before Κελτός and Κένταυρος. Atelaes 19:16, 11 June 2007 (UTC)
Oh, thank you. ^_^ Then maybe the proper nouns should be all lowercase so they collate properly. - Gilgamesh 23:08, 11 June 2007 (UTC)
Yes, that is an excellent idea. A couple notes about pronunciation. First, you may find a few Classical pronunciations with acutes represented by the raising/lowering pitch signs, which are supposed to be reserved for circumflexes. This is a mistake of mine (which I have stopped making, but have not taken the time to go back and fix). Feel free to correct these as you find them, if you like. Second, I'd like to keep the /h/ for rough breathing marks in the Koine pronunciation. While it is admitted that a number of dialects dropped this during the Koine period, a number didn't. Importantly, biblical works obviously retained it (as evidenced by Semitic loan words). As the Bible (and related works) is arguably the most important work from this period, it seems important to retain this. Atelaes 23:52, 11 June 2007 (UTC)
Oh, alright. I have an idea. As it is clearly dying hard in that period, we should mark Koine /h/ as [(h)], you know, with parentheses, to indicate that. I'll edit accordingly. ^_^ - Gilgamesh 23:54, 11 June 2007 (UTC)
Ah, good idea. Atelaes 23:55, 11 June 2007 (UTC)

Romanization on AAG[edit]

Would you please take a look at Wiktionary talk:About Ancient Greek#New Romanization scheme and Wiktionary talk:About Ancient Greek#Verbs. I suppose these conversations could probably take place on one of our talk pages, as I can't imagine anyone else on Wiktionary will care, but I feel that policy conversations of this nature should be kept somewhere more public, for posterity if nothing else. As you seem to know a great deal more than I about AG pronunciation I intend to largely leave that matter at your discretion. Many thanks. Atelaes 00:36, 14 June 2007 (UTC)

Ummm.....did you read my response on Wiktionary talk:About Ancient Greek#New Romanization scheme? There needs to be only one romanization, not three. Atelaes 04:36, 14 June 2007 (UTC)
Sorry again, I tend to edit-frenzy. XD I discuss and then do sudden marathon edits until they're either finished or I burn out. XD - Gilgamesh 05:14, 14 June 2007 (UTC)
Yeah, I noticed.  :-) Again, don't worry about it. I'm just glad your million edits are well-informed and that you're capable of rational discussion. We have other editors who go on marathon the wrong direction; and that's just a mess. Atelaes 05:20, 14 June 2007 (UTC)
By the way, I was fully aware of the irony of the term "Byzantine Exceptions" when I added that section to the bottom of the table. ^_^ - Gilgamesh 05:47, 14 June 2007 (UTC)

May I ask what the diacritic in Δαρεῖος is? I don't have it in my IPA book. Atelaes 06:44, 14 June 2007 (UTC)

Oh, sorry. It's an inverted breve, the alternate diacritic for the Greek circumflex. Tilde is already used by IPA for nasalization. - Gilgamesh 07:08, 14 June 2007 (UTC)
Hmmm....I guess I'm a bit confused here. The character I originally put is not a tilde ẽ, however I don't know the name for it ê. However, I think the diacritic I meant to use is ě. It is listed at w:International Phonetic Alphabet#Suprasegmentals as "Rise". I'm not seeing the inverted breve anywhere...... Atelaes 07:19, 14 June 2007 (UTC)
Well, whatever symbol means [↗↘]. - Gilgamesh 10:42, 14 June 2007 (UTC)

sing/plural templates[edit]

Hi please read the comments and suggestion on my talk page to save creating sing and plural forms for every declension! Robert Ullmann 01:51, 19 June 2007 (UTC)

see my talk again :-) Robert Ullmann 02:12, 19 June 2007 (UTC)

Sorry about my recent absence[edit]

I do have some things to say about your recent work with AAG, but you'll have to be patient, as I intend to go through it all and try to come up with a unified response. As for -ώ/-οῦς, I cannot find the inflection for it. Sorry. By the way, thanks for loading up the Ancient Greek proper nouns, that section is starting to look......almost respectable (as opposed to the hodge-podge random assortment of a few names that it was before).  :) Atelaes 20:11, 9 July 2007 (UTC)

Okay, I'll wait. I hope I'm doing a good job. I'm still new to Wiktionary. Someone else explained to me the inflections for -ώ/-οῦς. - Gilgamesh 21:16, 9 July 2007 (UTC)

Heading level &al.[edit]

Just a quick heads up: I've noticed that you've been putting Inflection, Derived terms, and other such headers at L3 when they should be L4. Sorry for not yet responding to your guess at -κλῆς forms; I've not felt like practicing my vowel contracting. I will try to take a gander (and perhaps a guess as well) tonight. Thanks again for adding in so many proper nouns; you're doing an excellent job! Medellia 06:08, 13 July 2007 (UTC)

Oh, I'm sorry about the L3 stuff. :( What do we do about it? It seems like a lot of work. - Gilgamesh 06:13, 13 July 2007 (UTC)

I'm kind of in need of some repetitive corrections at the moment (good stress reliever!), so I think I can take care of them. Medellia 06:17, 13 July 2007 (UTC)

Thank you. ^_^ I'm going to make new entries (starting with Κορδύβη I just added) to use L4. - Gilgamesh 06:20, 13 July 2007 (UTC)


According to my textbooks, the following masculine nouns form the Vocative in -ᾰ:

  1. ethnic: Πέρσης, Πέρσα, Σκύθης, Σκύθα
  2. in -της: δικαστής - δικαστά, πολίτης - πολῖτα, but δεσπότης - δέσποτα
  3. in -άρχης, -μέτρης, -πώλης, -τρίβης, -ώνης.

A very powerful tool that I recommend is TLG, a CD which contains all Ancient Greek texts. I was able to locate the vocative of Αἰγινήτης in a few minutes. It's implicit that I'll be glad to help, if need be. --Flyax 11:03, 5 August 2007 (UTC)

Many of these more commonly-encountered stems can be easily handled with Category:Ancient Greek declension convenience templates. - Gilgamesh 00:53, 6 August 2007 (UTC)

Just making sure that...[edit]

you (and others working on Ancient Greek) saw the discussion at Wiktionary talk:About Greek#format of greek translations in english entries and have no objections. I'm happy to take silence as consent, as long as it really is! ArielGlenn 07:29, 14 August 2007 (UTC)

Well, there is certainly work to be done, isn't there now. :3 Not to mention that I'm doing most of it... - Gilgamesh 07:33, 14 August 2007 (UTC)
Oh, my mistake. I was never actually watching that page. I watch Wiktionary talk:About Ancient Greek. I actually had no part of the discussion you spoke of, and only learned about it now. I just add translations because I think it's a good idea. :3 - Gilgamesh 07:34, 14 August 2007 (UTC)
I had posted on Wiktionary talk:About Ancient Greek earlier but it must have fallen through the cracks. It's the last post... I hadn't heard anything from anyone on your side of the greek pond, so figured I'd better check in. ArielGlenn 09:13, 14 August 2007 (UTC)
Yeah, it fell through the cracks. I didn't read precisely because I don't work in Modern Greek. XD I know phonology, romanizations, some heart isn't in it. - Gilgamesh 09:24, 14 August 2007 (UTC)

Concordance:Fragments of Heraclitus[edit]

Greetings! Since you're on ancient Greek tear, I thought you might like to have at this concordance. Cheers! bd2412 T 03:48, 22 August 2007 (UTC)

My hands are full. I already have other sources enqueued. - Gilgamesh 03:50, 22 August 2007 (UTC)
Very well, understood. Cheers! bd2412 T 09:15, 22 August 2007 (UTC)

γ as ŋ in ἄγνωστος[edit]

I am not an expert in pronunciation, let alone in Koine or Byzantine Greek, however I don't think that γ before ν was ever pronounced as ŋ. γ as ŋ is to be found in modern Greek before κ,γ,χ (έγκυος, έγγραφο, άγχος). If you could give me some links about Byzantine pronunciation, I would be grateful. --Flyax 17:37, 26 August 2007 (UTC)

In Medieval Greek. In Byzantine sandhi, voiced fricatives before nasals also become nasals. This is partially described in the Medieval Greek article. This sandhi no longer strictly applies in modern Greek as it did during the Byzantine period, but there are remnants, such as συγνώμη for classical συγγνώμη. - Gilgamesh 01:18, 27 August 2007 (UTC)
Oh, I also wrote an expansive guide to pronunciation a while back. Wiktionary:Ancient Greek Romanization and Pronunciation. It may need some very minor revisions, but it's by and large good. - Gilgamesh 01:24, 27 August 2007 (UTC)

(i)The pronunciation of modern Greek συγνώμη is //siγnomi//. (ii) Changes in words like γίγνομαι to γίνομαι, γιγνώσκω to γινώσκω could suggest that γ before ν could have become a nasal (or was it simply lost?)- but these minute changes had occurred before the byzantine times. Moreover, how could we be sure that this was a general rule? It doesn't seem very likely to me that Byzantines pronounced /praŋma/ (πρᾶγμα) and /ŋnosis/ (γνῶσις) while the pronunciation both in Koine and modern Greek is /praγma/ and /γnosis/. But as I said I'm not an expert. --Flyax 07:57, 27 August 2007 (UTC)

Then there's a chance I gleaned the information improperly. What is certain in Byzantine Greek is that the phoneme combination βν became μν (and as such ευν etc. became εμν etc.), and βμ became (μ)μ. This is no longer strictly the case in Modern Greek, correct? From what I had gleaned, I thought it was logically a rule that involved voiced fricatives before nasals, which would have implicitly included γμ γν δμ δν ευγμ etc. as well. The [vn] to [mn] and [vm] to [(m)m] is unambiguously attested in the regular sandhi of Byzantine Greek. - Gilgamesh 08:43, 28 August 2007 (UTC)

Could you give me some examples? I cannot find any words with βν but few proper names (like a city in Israel (Λεβνα) or a Persian man). About "ευν", could you write for me the pronunciation in Koine and Byzantine Greek of the words εὔνους, εὔνοια? The modern Greek pronunciation is /'εvnus/ and /'εvnia/. --Flyax 12:43, 28 August 2007 (UTC)

εὔνους (eúnous) is [ˈɛβnu(ː)s] in Koine and [ˈemnus] in Byzantine. εὔνοια (eúnoia) is [ˈɛβny(ː)a] in Koine and [ˈemni(ʷ)a] in Byzantine ([ˈemnya] in early Byzantine, [ˈemnia] in most late Byzantine). I haven't yet encountered any linguistics sources that explain exactly why this sandhi no longer strictly applies today. I get the impression though that some phonetic trends were partially reversed under Katharevousa, such as a more careful distinction of κτ/χθ, πτ/φθ and ευς/εψ. Purely Demotic words containing χτ and φτ are still as they are today. - Gilgamesh 16:55, 28 August 2007 (UTC)
Excuse me for "eavesdropping", but do you have a reference for that information someplace (hopefully internet accessible but if not, I'll take what I can get)? I always thought (not an expert!) that pronunciation from Koine to modern was a pretty straight transition, so obviously I'm missing something. Thanks, ArielGlenn 23:46, 29 August 2007 (UTC)
Medieval Greek at Wikipedia. - Gilgamesh 06:00, 30 August 2007 (UTC)

Please, have a look here. εύνοστος, επίθ.· έμνοστος· εύμνοστος· όμνοστος; 4 parallel forms. Doesn't it mean that there were 4 parallel pronunciations? --Flyax 07:03, 31 August 2007 (UTC)

This illustrates to me just how much I didn't know about this subject. I'm really not sure how to proceed. We need to draft conservatively detailed prescribed IPA for the Byzantine entries to replace what I put in there. Also, someone needs to vastly improve and cite sources in the Wikipedia article where I got this to begin with. - Gilgamesh 08:20, 31 August 2007 (UTC)
I suppose that we all have learned a couple of things from this discussion. I consider that knowledge to be my personal profit from my contribution to this site. :)--Flyax 12:28, 31 August 2007 (UTC)
Thanks for finding this information, Flyax! It inspires me to try to find out more aout this whole area. I will also be interested in seeing how you figure out what ought to go in the entries, Gilgamesh. Good luck! ArielGlenn 18:13, 31 August 2007 (UTC)
Well, I don't want to use material that could be badly misinterpreted. My top priority right now is to revise Byzantine IPA, and to do that, I need more help. It is my routine, when adding new word entries, to add pronunciation keys, and I would rather not add any more unless we get it right. - Gilgamesh 05:24, 1 September 2007 (UTC)

Alright, I have some thoughts now. First of all, since we only have what is written and no recordings from the time to demonstrate how it was spoken, it would be helpful to both not make too many far-reaching assumptions, and also not assume that by not making many assumptions we are necessarily right. So, what do we know about Byzantine sandhi? What we can glean from speech patterns is only through misspellings of Atticist use and evolution of medieval demotic writings.

  • κτ/χθ/πτ/φθ/ευσ to χτ/χτ/φτ/φτ/εψ is attested from that time, but how universal is it? If it's only occasional even in demotic, it would be safer to indicate the more conservative pronunciations [kt/xθ/pt/fθ/efs]. But if the change seems to be the normal phonology, then we should indicate the pronunciations [xt/xt/ft/ft/eps].
  • The bigger question mark is my previous rational assumption of ευμ/ευν/βμ/βν/δμ/δν/γμ/γν becoming εμ/εμν/μμ/μν/νμ/νν/γγμ/γγν and then reversing under strong Katharevousa influence. The more I think about this, the more naïve I feel that I was. I can certainly dismantle this pseudo-sandhi, but what do we put in its place? Does [vn] stay as it is, or does it become [vmn] or something else? It's hard to completely disregard the attested ευν to ευμν/εμν as if it didn't exist. Was it only regional? Only occasional? Important questions.
  • Now as for the υ/οι/υι being [i(ʷ)], I'm still comfortable with that, if we're only indicating pronunciations for Byzantine as a whole. It's unambiguously [y] before the 10th century, but becomes mostly [i] after it, except in scattered locales, which the Wikipedia article only mentions to include Cumae (which is in Campania, Italy, and their pronunciation had remained fully [u]) and Megara and Aegina (which are on the western Saronic Gulf—today a relatively central location, but Greece Proper was suggested to have been a depopulated backwater during most of the Byzantine period, which at one time was almost wholly Slavic to the tip and had to be recolonized by Byzantine Greeks from Anatolia and Sicily). If the scattered locales died out quickly, I'd be willing to say the late Byzantine pronunciation was [i]—otherwise, [i(ʷ)].
  • Another question mark is the frequency of έV becoming ίV before the 13th century accent leap. Did this also mean αίV became ίV (such as, did Φώκαια become Φώκια before it became Foça)? Were there any following vowels (represented by V) that would cause έ not to become ί? And did this change reflect something that happened primarily from the Byzantine period or was it something that started in Koine? That is to say, did έ before a vowel merge with ή and shifted with η in the Byzantine period to become pronounced like ι?
  • The murkiest issue to me is ω becoming ου, as the Wikipedia article describes it as happening in "restricted" circumstances. What are these restrictions? Were they regular restrictions, or occasional/dialectual? And, for that matter, did ουε always become ο or could that be occasional/dialectual as well?
  • And how much did all these phonetic changes actually affect contemporary pronunciation of Atticist text? Afterall, I'm working on Classical Greek entries, not Byzantine vernacular language.

So, I wouldn't feel good dismissing all these questions out of hand, but I also want to supply a well-informed prescribed Byzantine pronunciation for Classical Greek words. - Gilgamesh 05:59, 1 September 2007 (UTC)

Some thoughts of mine that might help: In modern Greek both κτ/πτ/χθ/φθ and χτ/φτ are in use. In many words of learned origin (λέξεις λόγιας προέλευσης) κτ/πτ/χθ/φθ are the only choice: κτίτορας, πταίσμα, άχθος, as χτ/φτ/χτ is the only choise in many demotic words: χτικιό, φταίχτης. In addition to, there are lots of words with two parallel forms (χτήμα - κτήμα, φτάνω - φθάνω, φτηνός but πάμφθηνος. Some say αγανάκτησα, some say αγανάχτησα and those who say αγανάκτησα never write it as αγανάχτησα. So, when it comes to give to these words IPA symbols, it's quite easy and quite clear, isn't it? The written form corresponds to the pronounced one.
I suppose that the same thing happened in Byzantine times. It's highly unlikely that a Byzantine scholar pronounced the ancient words he encountered in his studies with χτ and φτ. In the same way, I believe that Byzantines scholars did not pronounce the Ancient Greek words like εὔνους as /έmnus/. The parallel forms of εὔνοστος-ἔμνοστος and particularly that of εὔμνοστος suggest that some (the more educated maybe?) followed a "traditional" (λόγια) pronunciation, some a vernacular one - and some couldn't decide.
As far as γν is concerned, I cannot find any indication that the Ancient Greek words were ever pronounced with ŋn. This -also- could lead to misunderstandings like confusing ἄγνωστος with ἄνοστος, ἄγνοια with ἄνοια etc. --Flyax 07:43, 1 September 2007 (UTC)

Then it seems clear what to do... To resist post-classical sandhi, and leave it in the realm of proto-demotic and demotic varieties of Greek. I think I can start to move forward, but I need to get to work and fix Byzantine classical phonology. First I will fix the description page, then I will fix article texts. As for Contemporary classical pronunciation, when exactly to I collapse unstressed ει/η/ι/οι/υ/υι as Template:IPAchr instead of [i]? And, if I recall correctly, the standard contemporary pronunciations of γλι/γνι/κλι/κνι/χλι/χνι are [ʝʎi/ʝɲi/cʎi/cɲi/çʎi/çɲi], right? I know this applies to γι/κι/χι alone (being [ʝi/ci/çi]), but not to λι/νι alone (unless they're before yet another vowel, such as with λια/νια being [ʎa/ɲa]). So, is Μαγνησία [maʝɲiˈsia] or just [maɣniˈsia]? Does the ν guard the γ from becoming palatal? - Gilgamesh 08:08, 1 September 2007 (UTC)

I've just edited the contemporary pronunciation of ἄγνοια. γ, κ, χ do not become palatal before λ, ν or any other consonant. So it's [ɣli/ɣni/kli/kni/xli/xni] always. --Flyax 09:40, 1 September 2007 (UTC)

Ahh, thank you for clarifying that. And οια isn't otherwise [ja]? How do I know when to use [j] or [i]? - Gilgamesh 09:41, 1 September 2007 (UTC)

Oh, and I'd also been informed that γου/κου/χου are pronounced [wu/kʷu/ʍu]. Is this also not true? - Gilgamesh 09:46, 1 September 2007 (UTC)

χοῦς = /xus/, κοῦρος = /'kuros/, γούνατα = /'ɣunata/. οι, ει, υι etc are /i/ (as far as classical words are concerned). But I need some time to reflect on possible exceptions. --Flyax 10:06, 1 September 2007 (UTC)

Good, good, these are good things to know... So the rule of thumb (for now) is to use [i]? I wish my sources weren't so contradictory. - Gilgamesh 10:09, 1 September 2007 (UTC)

I am curious about [wu/kʷu/ʍu]. Where did you find this information? Where can I find a description of these symbols? --Flyax 10:26, 1 September 2007 (UTC)

They are labialized velar IPA symbols (as [u/ʊ/o/ɔ/ɒ] are vowel equivilents of labiovelar consonants). See International Phonetic Alphabet at Wikipedia. I got this from modern Greek spellings I kept encountering—Γουλφ for Wolf, Γουάσινγκτον for Washington, Γουχάν for Wuhan, etc. I fear have no common sense... ;_; - Gilgamesh 10:32, 1 September 2007 (UTC)

Gilgamesh, one thing you can do to verify (or not) modern pronunciations of words currently in use is to check here: [1] (but they use a modified IPA, so be careful). So for example you can see that γουέστερν comes out to /γuéstern/. (I have never seen /wu/ and so on used for transcription of γού- etc.) ArielGlenn 17:00, 1 September 2007 (UTC)
More like no IPA. It's a pronunciation guide for medieval demotic. I get your point, but it's clear now that that won't be relevant—Flyax has pointed out that classicist usage demands a more enunciated pronunciation, so a lot of the assumptions of sandhi and collapsed vowels get thrown out more often than not. Don't worry, I'm using [ɣ]. - Gilgamesh 17:12, 1 September 2007 (UTC)
The site that ArielGlenn has suggested is the online version of Triantafyllidis Lexicon of Modern Greek. It shares the same layout with the online version of Kriaras's Medieval Greek Lexicon, for both are parts of the same project (Πύλη για την ελληνική γλώσσα). --Flyax 19:11, 1 September 2007 (UTC)
The particular changes to look out for in the IPA are that they use special symbols for most of the palatized consonants (κ, χ, γγ/γκ, ν , λ). You will see k with ~ over it for /c/, x with tilde over it for /ç/, g with a tilde over it for /ɟ/, and so on. (I can't paste them in here, it's some javascripty thing.) And they use j for /ʝ/. Other than that, it is straight up IPA. I hope that clears up any confusion. ArielGlenn 00:54, 2 September 2007 (UTC)
Ah yes... I like to study Greek, but I'm not fluent in it, and I struggle to remember simple phrases. What I do tend to retain well is naming. That's my forte. And yes...I did think some of the conventions were strange for medieval Greek. Good to know it's modern. - Gilgamesh 02:44, 2 September 2007 (UTC)


The word Αὐγοῦστος looks really strange to me. The word I know is Αὔγουστος, meaning both the month and the Emperor. I moved the entry but I couldn't fix the classical pronunciation, so I hid it. --Flyax 12:18, 3 September 2007 (UTC)

I've fixed the pronunciation. As for Αὐγοῦστος (Augoûstos), there are certain Google hits for it. - Gilgamesh 16:02, 3 September 2007 (UTC)
Well, I am afraid that there are a lot of mistakes on the Internet. As far as Ancient Greek is concerned, the absolutely reliable sources are the Liddell-Scott Lexicon and the texts themselves. Liddell-Scott has a lemma only for Αὔγουστος; I searched over the TLG (Thesaurus Linguae Graecae) CD and I was not able to find the word Αυγοῦστος. In the New Testament the word occurs once in the genitive Αὐγούστου. Yet see here The origin of the mistake is obvious. As for the Suida Lexicon, see here. See also the Kriaras' Lexicon. Of course there is always a possibility of a rare word, but unless we have a valid reference, we cannot accept everything we find on the internet. --Flyax 17:12, 3 September 2007 (UTC)
Certainly true. Would you all my knowledge of Greek came from the Internet? Perhaps it's amazing I've learned as much as I have. XD - Gilgamesh 17:19, 3 September 2007 (UTC)

IPA and all that[edit]

Thanks for joining in with Wiktionary:About Greek/Pronunciation - I have had to read as much as I could find on IPA/Greek - there isn't much. Please continue. —Saltmarsh 14:04, 13 September 2007 (UTC)

And I meant to add - I have moved your comment re [ε] etc to the talk page. —Saltmarsh 14:15, 13 September 2007 (UTC)

Your {{IPA2}} - I think that you should use {{IPAchar}} to fit in with naming schemes, we also have {{SAMPAchar}} and {{enPRchar}}. IPA2 could then be deleted —SaltmarshTalk 06:05, 15 September 2007 (UTC)

I've redirected {{IPA2}} to {{IPAchar}}, and replaced nearly all article uses of {{IPA2}} with {{IPAchar}}. I have not changed talk pages. - Gilgamesh 09:54, 15 September 2007 (UTC)

"Noun" not "Νoun"[edit]

You've entered a handful of \206\157oun headings, instead of using ASCII "N". Please note that those are not the same. --Connel MacKenzie 03:10, 17 September 2007 (UTC)

Oh my...I had no idea. I wouldn't even know which ones. o.o - Gilgamesh 15:05, 17 September 2007 (UTC)
Look at the last line in the table at the bottom of User:Robert Ullmann/L3 and you will see them. Robert Ullmann 13:32, 21 September 2007 (UTC)


What are you doing? Robert Ullmann 00:22, 21 September 2007 (UTC)

Oh, we've been discussing it at User:Gilgamesh/Greek IPA tokens and User talk:Gilgamesh/Greek IPA tokens. I wanted some feedback before I started the experiment. :3 - Gilgamesh 00:23, 21 September 2007 (UTC)
I see, but you are creating some 800 of these? There may be a much better way. I can think of a couple off-hand: simplest is to have one template for each phoneme, and pass it a parameter for the age (cla, koi, byz, mod). And that is just thinking about it for a moment or two. Robert Ullmann 01:45, 21 September 2007 (UTC)
But how does that even work? Lacking a more powerful templating scripting language, the argument (cla, koi, byz, mod) has to be made meaningful in some way. - Gilgamesh 01:48, 21 September 2007 (UTC)
so grc-ipatok-eeub would be: {{#switch:{{{1}}}|cla=ɛːwb|koi=e(ː)wb|byz=ivv|mod=i(v)v}}
but I can probably come up with something even better, but it is 5 AM here, must sleep. In the meantime, I'm not looking forward to manually deleting 800 templates when we do something simpler? eh? ;-) Robert Ullmann 01:59, 21 September 2007 (UTC)
This is my productive time. :( I am unfamiliar with this syntax... I suppose I could practice it. I honestly didn't think I was doing anything wrong—this approach seemed most responsible and logical to my knowledge and experience on templating. I'm sorry. :( - Gilgamesh 02:03, 21 September 2007 (UTC)

Much better, but I'd rather see them "collapsed", so each is a row in your table, without calling sub-templates. (And then you use it to make the table of course, so you can check everything ;-)

Have you written the template to call all these? Robert Ullmann 13:30, 21 September 2007 (UTC)

What do you mean? Could you rephrase those sentences? - Gilgamesh 15:15, 21 September 2007 (UTC)
Lots of those templates call other templates, instead of just having the form (like my example above). I think there should be one template for each row in your table, and it shouldn't refer to others, but just have the content of that row.
The other part is presumably you have written/will be writing a template for entries that uses these; have you done that? Can I see? ;-) Going to listen to jazz now; be back in ~4 hours. Robert Ullmann 15:21, 21 September 2007 (UTC)
Oh, I did that on purpose. In case we have more disputes over pronunciation, most templates have singular phoneme dependencies that are easily changed with single edits. Flyax and I in particular have been disputing exact pronunciations of Contemporary phonemes, and ArielGlenn is still trying to find academic material with exact values for us to reference as sources. As for embedding, I've been going through all the pages that link to {{grc-ipa}}, starting with the oldest entries. That's what I have been doing when I need to fix a lot of common problems at once. You know, using the "What links here" function in the toolbox. - Gilgamesh 15:26, 21 September 2007 (UTC)
Yes, but as you have found, it doesn't work. You have to simplify them as above. Otherwise you include hundreds of templates for each call (times 4 at each level, say 4 levels is 4^4, 256 ...). The parser follows every path, bottom up, not top-down. See my talk page. What template are you putting on the pages? (This is the third time I've asked ;-) Robert Ullmann 10:47, 22 September 2007 (UTC)
Okay, found it. So grc-ipa-rows calls grc-ipa-row 4 times, it calls templates 50 times, each template can be several hundred expansions. just using it once on an ordinary page can result in 3-5 thousand template calls. Even the null case:

Just added 200 template expansions to your talk page. Does it really need 50 tokens? How many words use more than, say 10-12? Robert Ullmann 11:01, 22 September 2007 (UTC)

Understand that #if: is not a procedural statement, the template language is not procedural. #if:cond|a|b does not mean "call a if cond, else call b". It means: "expand a, expand b, then decide which expansion to use based on cond". It always expands everything. The -rows template can be re-written to make it much better, and I suspect -guiderows as well. But the basic token templates must be simplified so they do not make "calls". Robert Ullmann 11:07, 22 September 2007 (UTC)

Damn... And I wanted to make them easy to maintain. I never dreamed there was such server drain. - Gilgamesh 15:15, 22 September 2007 (UTC)

Modern ζ -etc[edit]

  • Searching through TLG for words with the ζζ combination, I was able to find only three: Ουζζίττα (a town in Africa), Βυζζουχάρ (Hebrew) and Μιουζζ (something like a spell?). I didn't find a modern Greek word either. So, in my opinion, as long as you deal with Ancient Greek entries, you shouldn't bother about that. I don't know if there are such words in modern Cypriot dialect and what would be their pronunciation; I am sorry I can't help you with that.
  • Φλύακ-ες were the comic actors in ancient Greek cities in S.Italy and Sicily.
  • About Εὔα, is it possible to remove the classical pronunciation from the entry? --flyax 07:32, 23 September 2007 (UTC)
Well, the Septuagint is actually surprisingly old. It's from the 3rd century B.C., near the boundary between Classical and Koine, and I imagine Classical may have still had some waning phonological influence (it certainly did among the Latin Romans for a few centuries longer—that's why they adapted diphthongs αι and οι to native Latin diphthongs ae and oe for some time into the Koine period). The Septuagint was a Hellenistic Jewish text, predating Christ. To me it seems appropriate to keep the Classical pronunciation of Septuagint terms. - Gilgamesh 15:47, 23 September 2007 (UTC)
OK. Even if the Bible language certainly isn't classical, I guess you have a point here. However, I am still wondering if it's possible to remove the classical pronunciation from a lemma, if need be.--flyax 17:06, 23 September 2007 (UTC)
I could write another template, such as {{grc-ipa-rows-koine}} or something like that, to include only the last three rows. If and when you need it, ask me. - Gilgamesh 17:07, 23 September 2007 (UTC)
In this case, it would be simpler to write the pronunciation in the traditional way, without templates, I guess. --flyax 18:20, 23 September 2007 (UTC)
Well, I decided to write the template anyway. It was easy—it took a few seconds. {{grc-ipa-rows-koine}} is functional now and can be embedded with exactly the same argument syntax as {{grc-ipa-rows}}. Use it as is appropriate. - Gilgamesh 22:07, 23 September 2007 (UTC)

the Mackrige text[edit]

Finally I laid my hands on it and so now the relevant portion for pronunciation is in digital form (as images) but I don't know how to get it to you. Because it's images, it's several megabytes. Do you have any ideas? (I surely shouldn't post it here, as it's not intended for public distribution, given copyright issues.) ArielGlenn 02:11, 10 October 2007 (UTC)

If it's smaller than 20 megabytes, and you have a gmail account, then zip up the images, log into gmail website, and send the zip to me from there. My gmail address is diarmad atsign gmail period com. - Gilgamesh 02:13, 10 October 2007 (UTC)
Done, please let me know if it does *not* get there intact. ArielGlenn 07:36, 10 October 2007 (UTC)
Okay, I'm going to subtly revise pronunciation tokens to reflect the insights of the text you sent me. - Gilgamesh 09:05, 10 October 2007 (UTC)
Glad the text arrived ok. But I don't understand why you changed the pronunciation for the sequence νσ from ns to s (as reflected in for example Κωνσταντινούπολις, from the template {{grc-ipatok-ns}}. ArielGlenn 07:52, 11 October 2007 (UTC)
Because, even in active inflection, νσ and νζ do not survive in Greek of the Classical period. This is particularly noticeable in the dative plurals, where νσ becomes σ and the previous vowel (if not long already) is compensatorily lengthened. Besides, Κωνσταντῖνος entered the language significantly into the Koine period (a loan from Latin), when Greek's pronunciation made greater allowances to νσ clusters. Under the strictest of Classical rules, however, this is not permitted, and nasal-voiceless-fricative remains fickle and difficult for Greeks of most linguistic periods to pronounce. If you notice, precious few words of native Attic/Standard Classical Greek origin contain νσ, and I can only think of a few examples like Τίρυνς, located in Argolis (whose traditional dialect was not Attic anyway). The literary learned Atticism of the Koine period and onward seems more forgiving of phonetic combinations not traditionally allowed in Attic, and non-Attic names in some cases become more common (such as Ionic Σμύρνη instead of Attic and Aeolic Μύῤῥα, and Τυρσηνός instead of Τυῤῥηνός —Proto-Attic was stricter with clusters and simplified many, followed by a more forgiving trend later on). - Gilgamesh 08:27, 11 October 2007 (UTC)
If I am allowed to intrude in your discussion, please let me notice a few things. As Gilgamesh said, νσ was very rare in Ancient Greek, but this rule does not apply to modern Greek. If you search for the %νσ% combination in , you'll get a lot of results and I can assure you that these words are pronounced clearly with /ns/, at least by the more "careful" speakers. There is only one exception -as far I can tell - fορ the proper name Κωνσταντίνος and the derived Κωνσταντινούπολη, κωνσταντινάτο / κωσταντινάτο which are often pronounced - not by the "careful" speakers - without the first /n/. So, I think that the token should reflect that "proper" pronunciation of the words with /ns/. --flyax 12:02, 11 October 2007 (UTC)

Going to undo the accumulating indentation for this response. Well, if you did notice (if you view the token's source), I do the modern Greek νσ as [̃s] with a nasalization tilde, so that ωνσ is [õs]. The text said that, despite drilling on the subject, usually even the most carefully spoken of people will not quite pronounce the nasal as a full consonant in γχ μφ νθ νσ, and they usually end up nasalizing the preceding vowel if they can emphasize the nasal at all, hence the IPA tilde. So you're saying that, even despite this, a full nasal consonant should be indicated? - Gilgamesh 21:21, 11 October 2007 (UTC)

βτω, φλυαξ, ρεμεμβερ το σωιτκη υουρ ΙΜΕ ωηεν υου τυπε ιν Ενγλιση. - Gilgamesh 23:35, 11 October 2007 (UTC)

ΟΚ, Άι 'λλ ριμέμπερ δατ. :-) --flyax 16:29, 12 October 2007 (UTC)

And so the next round of indentation begins... If I understand what you are referring to (page 25, last paragraph?), he says "many" speakers... but he does not say "most". In Standard Modern Greek most people do pronounce the nasal consonants before fricatives, and you'll notice that Mackridge uses markup that reflects this, for example /ns/ for ρύπανση (p 30). ArielGlenn 14:47, 12 October 2007 (UTC)
Oh. Well that's obviously different then, isn't it. I'll make the changes. :3 - Gilgamesh 20:01, 12 October 2007 (UTC)
I reinstated full nasal consonants in Contemporary pronunciation. However, I kept it in Byzantine, as this was a major feature of Byzantine phonology. μβ was [mb] (β was [β] elsewhere—this is similar in habit to Castilian Spanish), and this was not ambiguous because μπ was still [mp]. μφ, on the other hand, had a weak nasal since the Byzantine period, and it was often dropped altogether in Byzantine Demotic. I keep it nasalized as an enunciation that the nasal exists, albeit a weak one. Before Katharevousa started influencing Modern Greek, this trend still held (as the text said), though now speakers in their best learned behavior say [ɱv mb ɱf]. Correct? - Gilgamesh 21:52, 12 October 2007 (UTC)
I don't have any insight on pronunication of medieval Greek, so I can't really comment on the Byzantine phonology issue. All I can say is that contemporary pronunciation is as you have it, [ɱf]. ArielGlenn 04:59, 16 October 2007 (UTC)


Could you please add an entry for this Ancient Greek word? – Krun 09:09, 12 October 2007 (UTC)

I don't really add new random words on demand. I've mostly just been adding proper nouns from Woodhouse's English-Greek Dictionary. To be honest, I'm still learning Greek myself, but I know enough grammar and usage to add proper nouns. I still can't inflect just any random adjective or conjugate any random verb though. - Gilgamesh 09:12, 12 October 2007 (UTC)
Ah, OK, I understand. A bit like me and Japanese, I suppose. You just seemed to be so very knowledgeable about Greek :). I'll just ask someone else. – Krun 09:22, 12 October 2007 (UTC)
It's an easy mistake. I'm autistic, and it's probably a savant skill. - Gilgamesh 09:28, 12 October 2007 (UTC)


Please see p. 25 (Mackridge) for the pronunciation of Σφίγξ. It's /sf'iŋks/, not /sfˈiŋs/. --flyax 19:14, 18 October 2007 (UTC)

Sorry, I had done that because the text said that speakers typically delete the plosive when pronouncing μψ γξ ντσ μπτ γκτ. But I changed it now. - Gilgamesh 21:30, 18 October 2007 (UTC)

MS apathy of well known bug[edit]

I'm curious about the bug you mentioned in this post. Can you please point me to some web site where I can read more CSS spec for that particular situation and about apathy MS has shown for not being compliant? Rod (A. Smith) 18:33, 12 November 2007 (UTC)

I will attempt to find the appropriate reference now. - Gilgamesh 18:35, 12 November 2007 (UTC)
It is here. And I quote: At a given element and for each character in that element, the UA assembles the font-properties applicable to that element. Using the complete set of properties, the UA uses the 'font-family' property to choose a tentative font family. The remaining properties are tested against the family according to the matching criteria described with each property. If there are matches for all the remaining properties, then that is the matching font face for the given element. As for references acknowledging the bug, I've found one here. Apparently, "font fallback" is when a font down a list is used when those at the top of the list are not available, and "glyph fallback" is the same principle, but applied to every individual character in an element. This apparently for IE is not simply a CSS issue, but an issue in their Trident (MSHTML) display engine in general, in which only simple font fallback is used but glyph fallback between fonts in a list is totally ignored. At w:Cascading Style Sheets at Wikipedia, it says that out of all the display engines, Trident is still by far the worst with a whole variety of CSS issues, not just this. - Gilgamesh 18:59, 12 November 2007 (UTC)
Ah. Thanks very much for that w3c link above. Older versions of IE did indeed misinterpret that part of the CSS spec. I know many people in the Free culture think Microsoft is apathetic towards such issues, but I assure you that's not the case in the Internet Explorer group. They are very serious about adhering to CSS standards. In any event, that bug was fixed, so modern versions of IE (all versions for the past 18 months or so) show the glyphs using the fall-back fonts, as expected. Cheers! Rod (A. Smith) 21:40, 12 November 2007 (UTC)

About e, o, r again[edit]

I saw your edit in the Wiktionary:IPA pronunciation key and unfortunately it creates new problems: the unsuspected visitor of the page would have the false impression that the Greek e sounds as the French é. Well, the sound of the French vowel is completely strange to the Greek native speaker. So, the right thing to do would be to create a new row with e̞ and put there the Greek word, then explain in the Wiktionary:About Greek/Pronunciation page why we prefer the e instead of e̞. The same applies to o. About r, I saw your edit and comment, but I am not ready yet to answer your question. --flyax 10:56, 14 November 2007 (UTC)

Okay, we can do that. I'll make a new row. But you should also realize that other languages also have this particular articulation of and and traditionally indicate it e and o. - Gilgamesh 11:12, 14 November 2007 (UTC)
In that case we could use in the new row something like e (e̞) and put in the row the examples from these other languages. --flyax 11:38, 14 November 2007 (UTC)
Yes, and that would cover Finnish and Turkish too. But I'm about to leave a comment on Wiktionary talk:IPA pronunciation key. - Gilgamesh 11:41, 14 November 2007 (UTC)

definite articles and proper nouns[edit]

Hello! It's been a long time since our last discussion, hasn't it? I recently noticed that you either omit the article in inflection of people names or add a note like this: The personal name does not take a definite article. I think that this is not 100% true. For example, personal names do take an article when they are the subject of a participle (ταῦτα δὲ τοῦ Σωκράτους εἰπόντος), with the conjunctions μἐν and δέ (ὁ μὲν Σωκράτης ... ὁ δὲ Κέβης ...) and in other cases (e.g. Plato, Phaedon, 96e: Ἐγὼ γοῦν͵ ἔφη ὁ Κέβης͵ ἡδέως ἂν ἀκούσαιμι ἥντινα δόξαν ἔχεις περὶ αὐτῶν. Οὔκουν γ΄ ἂν οἶμαι͵ ἦ δ΄ ὃς ὁ Σωκράτης ...). So, I think we could change your note to something like "The personal name rarely takes a definite article". What do you say? --flyax 22:04, 11 January 2008 (UTC)

Still studying myself, I think you would know better when it comes to grammar. In this area, I yield to your judgment in what you know about Classical Greek grammar. - Gilgamesh 15:40, 12 January 2008 (UTC)

Ancient Greek pronunciation[edit]

I've been doing a bit of work with A. Greek lately. While I've been attempting to follow the new pronunciation scheme, I think you're aware of some pronunciation subtleties that I am not, and so my recent creations (Special:Contributions/Atelaes) may benefit from you giving them a quick once-over, if you feel like it. If I'm making consistent mistakes, please feel free to let me know what they are. Thanks. Atelaes 00:24, 13 January 2008 (UTC)

Gladly. To help streamline this process, I created an entire token-based mechanism of pronunciation templates some time ago, at Category:Ancient Greek IPA tokens. I hope the short primer on how to use it is actually helpful. With this system, the four pronunciation periods can be represented by just one shorthand that embeds as sequential IPA templates. Most represent individual vowel/consonant phonemes, but some are compounds, taking into account geminated consonants, prenasalized consonants, devoiced diphthongs, sandhi, etc. I hope all the templates are named intuitively enough. - Gilgamesh 12:55, 13 January 2008 (UTC)
Yes, I saw it after I left this message. It's an amazing system, both in ease of use, and in ease of modification. Very impressive. The only trick for me, I think, is figuring out when vowels are long and short. Obviously if it's taking a circumflex, it's long.....but past that, I'm usually stumped. Any tricks on that one? Atelaes 21:44, 13 January 2008 (UTC)
Sorry for the delay. Yes, there are some tricks. A final unstressed vowel after a circumflexed vowel is always short, including -αι and -οι. However, if a stress penultimate long vowel is not circumflexed, the the final unstressed vowel must be long. Besides that, the best I can think of is study-study-study. It helps to have sources (like the old dictionary I found that fell into public domain) that frequently marks vowels as long or short. When I cannot figure out the length and I can't determine its length from its context, then it seems safer to just assume it is short, though it might be better to keep notes on these loose ends to fix later on. - Gilgamesh 23:21, 19 January 2008 (UTC)


A discussion is afoot at Wiktionary talk:About Ancient Greek#Mycenaean.......Greek? Redux. You have been invited because you participated in a previous discussion, I thought you might have a particular insight or interest in the discussion, or simply because I wanted to spam your page and irritate you. Check it out. Atelaes 09:03, 20 January 2008 (UTC)

Unfortunately, I know next to nothing about Mycenaean Greek. ^_^; - Gilgamesh 20:45, 20 January 2008 (UTC)
I guess I didn't really know who knew what concerning that, but as an integral part of the AGr. team, I thought you should at least be made aware of its existence. Additionally, I created a new conversation which I'm sure you'll have some thoughts on. I'm having a hard time linking to it, but it's right below the previously mentioned conversation. Thanks. Atelaes 00:42, 21 January 2008 (UTC)


Thanks for fixing my mistakes here. However, I don't think we should include the iota subscript in the romanization, and certainly not in the sort parameter. —This unsigned comment was added by Atelaes (talkcontribs) at 01:52, 3 February 2008.

That's how I'd always done it though. Are you saying that we shouldn't sort iota that way? As the for the romanization, I would have to insist, as the difference is certainly reflected in Classical pronunciation. I've seen dictionary etymology boxes romanize with spelling out the i too. - Gilgamesh 07:54, 3 February 2008 (UTC)
Ok, I can live with the i in the romanization, as our romanizations are, admittedly, biased towards Classical. But let's keep them out of the sort cat, shall we? I think that people looking for words would get confused by it. Atelaes 08:02, 3 February 2008 (UTC)
I don't know. It might still be a good idea in any event to sort for iota subscript on the first letter of a word if it's capitalized, as some people's polytonic Greek displays show the iota subscript of a capital letter not as a subscript, but as a normal lowercase iota letter directly to its right, e.g. looking like Αι. Also, I have observed that lowercase iotated letters without uppercase versions in Unicode convert to uppercase as two letters, including the extra whole uppercase iota, such as becoming ᾺΙ. Compare the German letter ß becoming SS on uppercase. Also, in Classical times before polytonic orthography was invented later on, the iotated diphthongs were written out as two whole letters anyway, and polytonic was invented and made use of the subscript during a later time after it had already become silent. Since our focus is the Classical language, shouldn't we collate the iota subscript as an iota just to be proper? Afterall, in this case, the classical spelling was ΚΩΜΩΙΔΙΑ, not ΚΩΜΩΔΙΑ. ΚΩΜΩΔΙΑ would be a Koine period vulgar spelling, and wasn't formal polytonic orthography only standardized in Byzantine times? - Gilgamesh 08:10, 3 February 2008 (UTC)
I don't think so. While I admit that the capitalization issue is a good argument (the font I use actually does this), I think that, in the future, this will be fixed. Besides that, I think that, generally, when people are looking for a word in a list, they are thinking of the iota subscript as more of a diacritic than anything. With regards to classical spelling, I see no reason why we can't create entries with the full iota and link them as alternative spellings of each other. While we are biased towards Classical when we have to favour something (such as in romanizations), in other areas we should try to give every era its due. And in later eras, the iota subscript really was little more than a diacritic, if anything at all. Atelaes 08:24, 3 February 2008 (UTC)

Category:Ancient Greek conjugation templates[edit]

Hey, I was wondering if you felt like taking a look at these new conjugation templates I'm writing. Unfortunately, I only have access to Attic inflection, but Attic is probably among the more important literary dialects and should represent Koine fairly well (although admittedly not perfectly). In any case, it's something, which is generally better than nothing. Hopefully we can get other forms in time. I'm incorporating a lot of the improvements you made in the nominal templates into them. As these are (hopefully) going to get incorporated in a lot of pages, I thought it a good idea to get as many eyes as possible on these, as problems will undoubtedly be easier to fix earlier rather than later. Also, I'm working on a new nominal template form, with the primary aim of making them a bit easier on the eyes (our current ones are a bit ugly), as well as perhaps add a little functionality. The work can be found at User:Atelaes/Sandbox. If you have alternative ideas, it would probably be best if you created a third (Conrad.Irwin has already created a second), so that we can compare different people's ideas. Thanks. Atelaes 19:50, 5 February 2008 (UTC)

Well, I am not fluent in any variety of Greek. Nouns are fun, but verbs are a weak point for me. However, if you are interested more in my more technical (formatting) eye, then I will try to keep an eye out. - Gilgamesh 19:54, 5 February 2008 (UTC)


I wanted to talk to you about this template. I guess to me it seems rather unecessary. The POS template should note the word to be indeclineable. After that, it seems to me that the inflection table isn't presenting any information, and is just wasted space. Is there a particular reason you want the entries to contain these tables? Again, I'm sorry that I seem to be dumping on your projects as of late. Nothing personal.  :\ Atelaes 05:02, 7 February 2008 (UTC)

It's not entirely useless. It shows the definite articles, which are inflected. ^_^ - Gilgamesh 18:25, 7 February 2008 (UTC)
Yes, but the definite article is already inflected at . I don't think it's necessary to reinflect it on every page. Atelaes 20:32, 10 February 2008 (UTC)

Hebrew names[edit]

Hi! I see you've added a lot of names to User:Gilgamesh/Hebrew names, and I'm guessing you're thinking of adding them as entries. That'd be great, and appreciated. here are some formatting issues that you probably know about already, but which I'm writing just in case you don't:

  1. The entries' titles are without vowels or other diacritics. Adding entries for both "full" and "defective" spellings, where both are attested, would be best.
  2. The headwords are with vowels, but without other diacritics. (I see you have stress markers on that page.) What I do is use the "defective" spelling for all headwords that have vowels; some others use the "full" spelling where that's the entry title. So take your pick, I guess.
  3. The transliteration should be something fairly accirate that people can read, like something out of the "Israeli" or "Academy" columns, not the "Tiberian" or "English" column, of your table.
  4. The "English" column can be listed as the translation (in the definition line), though, I suppose.

Have fun with it!—msh210 22:37, 14 February 2008 (UTC)

Thank you, but honestly I had no real plans. It was just something to do, and I realized I could use my user space page. Some of it is what might qualify as original research anyway, and I don't honestly know quite how to approach referencing it in a way that would satisfy everyone. As long as it's just on a private user page, I didn't think I would be forced into making major decisions about it on a timetable. Besides, the Tiberian transcription is historically extremely important, especially for biblical linguistic study—it's the phonology used by the Masoretic Text, and as such does not pretend to reflect modern-style Hebrew Academy or Israeli habits. It's rather like the difference between Classical Greek and Modern Greek. - Gilgamesh 00:22, 15 February 2008 (UTC)
Thanks for your reply. I don't see anything wrong with this original research (copying from the (English and Hebrew versions of) Bible and figuring out the transliterations): this isn't Wikipedia. Also, I'm not sure what your reference to the masoretic text means: the masoretic text includes vowelization and occasional keri ("read this word as if it said such-and-such, vowelized as follows"), but no indication of how each letter/vowel is to be pronounced; for all we know, the masoretes may have pronounced the words the same way a modern Israeli does or the same way or a shtetl-dweller of 1920 did. (Well, no, there are, I assume (not having studied this), indications against that, but still we don't know how the masoretes pronounced the words.) Once we have the vowelization, doesn't that incorporate all the masoretic info?—msh210 00:32, 15 February 2008 (UTC)
Well, for one thing, it is true that we may never know precisely what the 9th century pronunciation was in Tiberias when they drafted the Masoretes. What we do know is what they distinguished, and there are common scientific transcription systems for this that are also used for Classical Arabic (where similar questions of precise historical pronunciations apply). See Tiberian vocalization at Wikipedia. (It is also notable that Tiberian merged A-grade and O-grade qamez while Sephardic did not, but I (like many linguists) use multiple symbols to compensate for this: short o å and long/stressed ā ọ. However, unlike most linguists, I do not indicate the stress in Tiberian Hebrew scientific transcriptions, in part because it is still very inconvenient to do so in Unicode and most professional linguists (unlike amateurs such as me) are probably more likely to publish academic works on paper or in PDF whereas I have HTML and MediaWiki. ^_^ Instead, I treat the stressed vowel as long (as it always was said to be), and I treat the last long vowel as stressed. All vowels after the stressed vowels are treated as short in transcription even if some of them may actually be pronounced long in historical practice. You might think this might be complicated, but it only applies to five vowels: i e a å u, as no other unstressed final vowels are attested in Tiberian vocalization. Generally speaking, if the unstressed final vowel has a consonant after it, it is pronounced short, otherwise it can be pronounced long as long as the stressed syllable before it is clear.) (Sorry for the in-parentheses ramble. ^_^;) - Gilgamesh 00:44, 15 February 2008 (UTC)
Oh, in answer to your other question (my brain is slush today, but as an autistic savant skill for me, linguistics takes almost no effort), the vowel distinctions that the Masoretic Text makes are contextual. For the seven principal nequddoth (hireq, zere, seghol, pathah, qamez, holem, qibbuz), the rules of thumb are these:
  1. Zere and holem are always long, no exceptions: ē ō.
  2. If the other symbols are before a double consonant, a consonant cluster or in an unstressed final syllable, they are short: i e a å/o u.
  3. But if they are before a single consonant before another voiced vowel (including shewa na), or if they carry a methegh, then they are long: ī ẹ ạ ā/ọ ū.
  4. Tiberian vocalization, as I said, had merged A-grade and O-grade qamez and did not distinguish them in nequddoth. But in practice they can be told apart in context, as most long qamez are ā, most short qamez are o, all final unstressed qamez are å, and long O-grade qamez is very rare, usually found as an emphatic variation of short O-grade qamez.
  5. When a maqqeph (that dash-like symbol) connects more than one word, then the chain of maqqeph'ed words is pronounced as if one word, and the final vowels of pre-maqqeph words can be treated as if they aren't the word's final vowel but instead flow seemlessly into what comes after the maqqeph: kol- for כָּל־.
You may also notice that I do not transcribe different letters for the three hataphim, but this too is contextual: A short vowel before a consonant before another vocalized vowel necessarily must be either shewa na or a hateph, unless the vowel that comes later is itself a hateph (and the qualifications for that are the same). Therefore, transcriptions like ʼEḏōm for אֱדוֹם and ʼOholīḇāmā for אָהֳלִיבָמָה are not ambiguous. - Gilgamesh 00:57, 15 February 2008 (UTC)
Thanks for your reply, much of which went over my head.—msh210 18:00, 20 February 2008 (UTC)
I'm sorry. ;_; Like I said, I mostly study the classical language. I study dead languages (or dead forms of languages) in general. X3 And as long as the Masoretic Text marks vowels the way it does, I <3 Tiberian Hebrew. :3 Do you have any questions in particular? - Gilgamesh 21:13, 20 February 2008 (UTC)
I read your profile page. [ɑɻ jɨu ən ɑʃkənɑːzi ʤɨu]? ^_^ - Gilgamesh 21:21, 20 February 2008 (UTC)
Yes, I am.—msh210 22:53, 21 February 2008 (UTC)
Fabulous. ^_^ - Gilgamesh 23:02, 21 February 2008 (UTC)
Why is that fabulous? And why did you use IPA to ask?—msh210 19:17, 26 February 2008 (UTC)
Fabulous because I like meeting new and different Jewish people from different cultural and regional backgrounds. I've already met an American Karaite, a Lithuanian Jew from England, a guy from Ra'anana from a mixed Ashkenazi-Mizrahi family, and a gay Israeli who identifies as a Canaanite. As for the IPA, oh, I just have a sense of whimsy. :3 The stuff I study and work on—Hebrew, Greek, other languages—it's fun for me. A little whimsy makes life a little more interesting. :3 - Gilgamesh 06:12, 27 February 2008 (UTC)


Hey, so I was about to create κῆρ, when I saw this. However, the LSJ says that Ker should actually be at Κήρ, not Κῆρ, and I couldn't find the entry in Woodhouse. So, just wondering if you could double-check that when you get a free second. Thanks. -Atelaes λάλει ἐμοί 20:47, 22 February 2008 (UTC)

It's in Woodhouse in proper names under "Fury". It's linked to at the bottom of the entry page. However, if I made a mistake I made a mistake. Do what you believe is appropriate. - Gilgamesh 04:53, 23 February 2008 (UTC)


Ummm......what's going on with that crazy transliteration? I realize that you see a lot more in the Hebrew than I do, but all the extra marks on the transliteration aren't helping me, nor do I think they'll help anyone else. I that you've started a discussion on Wiktionary talk:About Hebrew on this matter, however you may want to clarify it a bit if you're proposing it as a transliteration standard. As it stands, it looks like simply your thoughts on Hebrew pronunciation, not a proposal for a new transliteration scheme. -Atelaes λάλει ἐμοί 17:45, 25 March 2008 (UTC)

Oh no, I've seen that convention lots of places when dealing with Biblical Hebrew. Look up the ISO 259 standard. Heh, I wish I'd thought of it first. X3 - Gilgamesh 18:00, 26 March 2008 (UTC)
Ah, well the ISO is not the standard on Wiktionary, Wiktionary:About Hebrew is. If you'd like to change it, then feel free to propose it on the talk page, but until it's adopted, I think it's best to follow the current conventions. -Atelaes λάλει ἐμοί 18:03, 26 March 2008 (UTC)
Tried that quite a while ago. No one bothered to respond. It seems that no one really cared to comment on it. And when I discussed Hebrew on IRC #wiktionary, I found a few people who worked with Hebrew, but few had done any serious Biblical Hebrew studies outside of the regional religious traditions they grew up with. ISO 259 and similar approaches are used for serious secular linguistic examinations of Biblical Hebrew. - Gilgamesh 21:15, 26 March 2008 (UTC)
It looks like Mushy caught wind of out conversation. :) I have also added my two cents, for whatever they're worth. -Atelaes λάλει ἐμοί 05:06, 27 March 2008 (UTC)

User:Gilgamesh/English vowels[edit]

I'm curious how you decided on the examples you've used. What dialect/accent do you speak? I ask because your examples do not match either British or North American vowel sounds. --EncycloPetey 02:11, 28 April 2008 (UTC)

The abstract/conservative ones are partially not in any extant accent, but they represent something that preserves detail between the extant accents. As for the accents themselves, I studied them on their associated Wikipedia articles. Anyway, it's just a user page project. Something someone on IRC #wiktionary told me might be a good idea to describe after I spoke about some of my theories. I have no intention of proposing it for adoption. Compare my other user page projects. - Gilgamesh 02:21, 28 April 2008 (UTC)


I hate to be a downer, but can I ask what's going on with the early and late Byzantine? The previous discussion (which admittedly did not involve that many people) ended with the decision not to have such a distinction. As the template affects basically all of the grc entries, it seems like such a change should have followed a renewed discussion on the topic. -Atelaes λάλει ἐμοί 07:34, 8 May 2008 (UTC)

Well, it was more experimental than anything. Besides, I only included things about Late Byzantine that are already known. But if it's still a problem, I'll just comment it out. (Damn, it's really pretty too.) - Gilgamesh 09:48, 8 May 2008 (UTC)
Yeah, I think it'd probably be best to return it to where it was yesterday. Also, bear in mind that experimenting with such a highly used template burns up the servers for a few hours. -Atelaes λάλει ἐμοί 17:33, 8 May 2008 (UTC)

grc nominal templates, redux[edit]

So, you may or may not have noticed, but I've given the grc noun inflection templates a makeover. I'm also trying to improve them in a few other ways. A few things I'm doing: 1. Sorting them into sub-cats, such as Category:Ancient Greek 1st declension templates, etc. I think it might be a bit easier to find them when they're broken up a bit. Secondly, hidden templates like this should have something on the bar while they're hidden. I've decided that the title should be the name of the declension (first, second, third, etc.), the nominative singular, and the genitive singular. I'm thinking about adding links to them, while I'm at it. While we don't have many inflected forms yet, I plan on writing a cite bot to make them sometime this summer. However, there are a few issues that I'm coming across. The first one is that if someone enters a multiple word genitive, like in Λίχας, they both get linked together. The solution I came up with is to only link if the template uses the default inflection, and not link if a custom parameter is entered. However, I suppose that, ideally, it should be isValidPageName. Also, I've been trying to figure out how to work with macrons and breves. If we link the terms, the macrons can't be linked, as they're not a part of the entry title. So, I suppose we'll have to follow Latin's example for that, but that means a lot of code. Anywho, I'm going to set up something on User:Atelaes/Sandbox with {{grc-test}} and {{grc-test2}}. If you care to take a look and offer any critique, advice, etc. I'd love to have it. -Atelaes λάλει ἐμοί 07:18, 22 May 2008 (UTC)

Ok, so I've been working with the templates for awhile now. And I'm really sorry to say that I don't think we'll be able to keep the macrons and breves in there. I tried everything I could think of (you can see the progress I made at the above links), but it just won't work if there are multiple variants of a single type (as there sometimes are). I'm waiting on a response from Robert Ullmann, but I already ran it by Conrad.Irwin, and he couldn't think of anything. So, I think they're going to have to go. I realize that you put a lot of work into inserting them. Sorry. -Atelaes λάλει ἐμοί 08:18, 23 May 2008 (UTC)
But they're such an important part of Ancient Greek phonology. :( - Gilgamesh 14:29, 23 May 2008 (UTC)
Ok, I believe I've gotten it so that we can have the best of all worlds. Robert Ullmann clued me in to a trick which allowed me to put more conditionals in, and I think I've got it up and running. Take a look at User:Atelaes/Sandbox, where the different options and how to input them are listed. -Atelaes λάλει ἐμοί 05:20, 31 May 2008 (UTC)
Very nice, I'm very much impressed. ^_^ BTW though, in the source material I've used (including Woodhouse), a short vowel before another vowel or before two or more consonants or a doubled consonant is almost never explicitly marked as short...but it is when it's long. Go figure. Either way, I just followed the convention. Marking short everywhere may not necessarily be inappropriate. - Gilgamesh 04:49, 1 June 2008 (UTC)
Yes, I think that, in general, we may as well mark everything. One of the beautiful things about Wiktionary (as opposed to traditional paper dictionaries), is that we get to spell everything out explicitly, without making the reader have to look up all the rules, thus we include full inflections and pronunciations, which are strictly unnecessary, as one could figure them out based on the rules. We should sit down and figure out exactly where we are going to include syllable length marks, though. I think we can both agree that they should go in the inflection tables :P, and obviously the vowel lengths should be noted in the pronunciation section. But, I don't think they should really go anywhere else. For example, I don't think we should get in the habit of displaying them in links, like Latin does, as they're not as important as they are in Latin. Also, Robert Ullmann was kind enough to generate this, which needs to be taken care of eventually (although I don't consider it an urgent thing). If you have time (and the inclination), feel free to knock some of them out. Otherwise, I'll make my way through them myself. Also, I was wondering if you'd be willing to reconsider the whole definite articles in the inflection templates thing. I sort of think that we've discussed this before, but I'm not really sure. The thing is, the definite article is not a necessary part of the word; any word can exist in its inflected form without the DA. Also, no other language includes its DA in its inflected form. It seems to me that this is unnecessary clutter. Finally, I just wanted to note that I haven't forgotten about my promise to re-examine the Late Byzantine pronunciation issue. I simply haven't gotten the time to go to the library yet, but I will eventually. -Atelaes λάλει ἐμοί 07:24, 3 June 2008 (UTC)

Template:ltr, Template:rtl[edit]

You'll have to rename these templates. Three-letter template names are reserved for ISO codes. --EncycloPetey 19:08, 15 September 2008 (UTC)

Eek! Okay, I'll think of something. - Gilgamesh 19:08, 15 September 2008 (UTC)
Okay, moved to Template:LR and Template:RL. If these are alright, you can delete the redirects. - Gilgamesh 19:12, 15 September 2008 (UTC)


Please see WT:BP#CheckUser votes., you may be able to help. --EncycloPetey 20:38, 15 September 2008 (UTC)

I don't really know the issues involved. Besides, I mostly edit Ancient Greek entries. - Gilgamesh 21:47, 15 September 2008 (UTC)


Note that this doesn't work. (And can't work.) There is no way to make something sort in between (say) "ata" and "atb" without recoding those entries as well as the inserted one.

(Something very close we use for Japanese, which is moving the "diacitic" to the end, which will sort something otherwise spelled the same in the correct order. But that is only for the first character.) Robert Ullmann 17:40, 16 September 2008 (UTC)

It's not for entry names. Just for sort-order for categories on the server side, within a particular language. If used appropriately, it is invisible to the end user. - Gilgamesh 17:42, 16 September 2008 (UTC)
The sort order does not work: Look at the cat now, note that aţă is sorting at the end of "at...", after atrocitate Robert Ullmann 17:44, 16 September 2008 (UTC)
Yes, that's correct. In Romanian, letters with diacritics sort after the letters they're modified from and before the next letter. So even atzzzzzzzzzzzzzzzzzzzzz < aţ < au. That's entirely intentional. - Gilgamesh 17:46, 16 September 2008 (UTC)
By the way, since you left comments on the template comment, I did too. Read them. Then we can move them to the template's talk page. - Gilgamesh 17:51, 16 September 2008 (UTC)

You are right, I did not understand. I thought you were trying to get the letters to sort together. It was utterly inconceivable to me that you would try to use an internal-use-only code and a template to try to wedge that code into the wikitext to do ordinary sorting of the character after the plain variant. After all, why ever would you not use the simple, obvious, and bog-standard way of doing it?

You really just want to sort ţ after t? Yowza. All this cruft?

Since it doesn't seem to be obvious to you: the key for aţă is atţaă. Done.

Please fix the entries you have done so we can kill the template. Robert Ullmann 23:32, 16 September 2008 (UTC)

All done, and I even did you the courtesy of putting {{speedy}} on the template. But in the future, please don't refer to me or my edits in reference as if I'm an imbecile. I may have no common sense, but I have lexical knowledge and wiki markup skills, and I use them here to the best of my ability. - Gilgamesh 01:10, 17 September 2008 (UTC)


[2] Really? I've never heard it pronounced like that. Could you please provide which accent pronounces twenty like that so the information could be added to twenty? --Yair rand 04:15, 4 January 2010 (UTC)

Large swaths of the United States. In fact, listen to the rhymes in the pop song Love Shack (Athens, Georgia accent). "I got me a Chrysler, it seats about twenty. So come along, and bring your jukebox money." I came from a very different part of the U.S., but I also grew up saying "twunny". - Gilgamesh 01:29, 5 January 2010 (UTC)


Hello Gilgamesh -- Re this edit and a similar one you made at want, the {{references}} template does not work the way you think. It is not used for footnotes. I replaced it in both of these entries with the <references/> tag, but didn't check to see if you'd inserted it anywhere else. -- Ghost of WikiPedant 15:34, 14 September 2010 (UTC)

Wiktionary talk:About Hebrew/archives/2008#Shva na transliteration[edit]

Dunno if this topic still interests you, three years on, but I recently read Edna Amir Coffin and Shmuel Bolozky, A Reference Grammar of Modern Hebrew,[3] page 22, which says that a sh'vá ná is pronounced "when a word begins with a consonant cluster the first member of which is ‎י׳‎, ‎ל׳‎, ‎מ׳‎, ‎נ׳‎, ‎ר׳‎, or when the second is ‎א׳‎, ‎ה׳‎, or ‎ע׳‎" or when it "immediately follows another shva". For the former, it gives the examples of l'vaním, y'ladím, n'shamá, m'sibá, z'hirút, r'shimá, sh'oním, and t'uná; for the latter, it gives nivd'ká and tisg'rú. This doesn't seem to be comprehensive — for example, it doesn't mention the various prepositions and prefixes where the sh'vá is always pronounced (b'-, t'-, etc.), and it doesn't mention cases like lamád'ta (where a traditional sh'vá nákh is pronounced as if it were a sh'vá ná) — but it's the most comprehensive I've ever seen, and it's certainly closest to taking the form of your table. —RuakhTALK 22:19, 8 August 2011 (UTC)

grc-ipa-rows later byzantine–contemporary accent[edit]

Dear Gilgamesh, I find the grc-ipa-rows template very good. However, I am concerned about the omission of a later Byzantine and practically modern Greek pronounciation field. The 10th AD Byzantine pronounciation is the same with the modern Greek one except for the digraph <οι>. In the 10th c. it was pronounced as [y] but from the 11th–12th cc. onwards it has been pronounced as [i] too. This has been the last change that led to the modern Greek phonology.

Undoubtly, the late Byzantine and contemporary pronounciation of ancient or medieval words is necessary because until the 19th c. there had been writers that used Greek forms from the most archaic ones, such as Attic Greek, to heavy katharevousa.

So, if there was no problem with <οι> we could simply write 10th AD Byzantine onwards. But now, I propose we include another field Late Byzantine onwards or we write 10th AD Byzantine onwards and add a footnote or something to indicate that after some time <οι> has been pronounced as [i] and not as [y], only in cases when the entry has οι. Dimboukas (talk) 15:16, 1 June 2012 (UTC)

Years ago, there used to be a Katharevousa row. But I've been recommended against doing Katharevousa in this context, because I was told it wasn't really the same as trying to do Classical Attic anymore — that rather, it was an attempt to partially atticize Modern Greek. To be honest, I'm not entirely sure what to think. There seem to be a lot of modern politics and social stigmas involved since the establishment of independent Greece in modern times, and it seems far less complicated to stick with historical linguistics. - Gilgamesh (talk) 22:51, 1 June 2012 (UTC)
It also occurs to me, that Byzantine Greek was the last stage of more or less unified Greek phonology in the eastern Mediterranean. Pontic and Griko had already split at this time. And even by the 19th century the pronunciation conventions were more different between Cyprus and the Aegean region. I'm reluctant to open a can of worms that goes so deeply into modern politics and attracts more allegations of original research or competing linguistic orthodoxies (the Katharevousa pronunciation reforms vs. organic pre-Katharevousa Greek phonology being the biggest issue that comes to mind). The 10th century is relatively uncontroversial, and can remain purely documentive. Modern Greek and Cypriot language politics seem so complicated. - Gilgamesh (talk) 22:57, 1 June 2012 (UTC)
If I recall correctly, since Katharevousa was as much a pronunciation reform as it was a state linguistic policy, there were some significant differences of Byzantine vs. pre-Katharevousa vs. Katharevousa. For instance:
  • Byzantine and pre-Katharevousa had the plosive/fricative sandhi, where πτ/φθ became φτ, κτ/χθ became χτ, ευσ became εψ, ευθ became ευτ, etc., and this was a highly active, predictable phonotactical feature. Katharevousa started to stigmatize this, so that πτ was reinforced as /pt/, φθ was reinforced as /fθ/, and both were separate from Demotic /ft/. I've been told that the historic sandhi is now generally stigmatized as country and backward, and that the Katharevousa-inspired pronunciation reforms are considered more educated.
  • Byzantine and pre-Katharevousa retained a plosive pronunciation of μβ νδ γγ as /mb nd ŋɡ/, and in Demotic texts they even became respelt as μπ ντ γκ because the sounds had merged in pre-Katharevousa when it wasn't as clear they weren't still /mp nt ŋk/ in Byzantine. Katharevousa seemed to largely retain the newer voiced pronunciations of μπ ντ γκ, but promoted a new pronunciation of μβ νδ γγ as /ɱv nð ŋɡ~ŋɣ/, which is considered the more educated pronunciation today.
  • Byzantine was losing the nasal components of μφ νθ νσ γχ, which is why I mark them as ambiguously vowel-nasalizing in the Byzantine row, here using /a/ as merely a placeholder vowel: /ãf ãθ ãs ãx/. Though the loss of the nasals in these sequences was underway, it's not clear how lost they were, and it was common to see misspellings of these sequences as φ θ σ χ. This seems to have been complete by pre-Katharevousa. But Katharevousa prescribed a renewed nasal consonant, and μφ νθ νσ γχ became /ɱf nθ ns ŋx/.
  • And, as you said, Byzantine /y/ for υ οι υι had soon become /i/ in most accents (though not in the "Old Athenian" accents insulated by the Arvanitic-speaking region, which had /(ʲ)u/).
There are reasons why the rows stop at 10th century Byzantine — it documents historical atticist usage in the time periods associated with Ancient Greek. The Byzantine period may have ended in 1453, but it started about 330, which was still Late Antiquity. After the fall of the Byzantine Empire, there wasn't much of any independent Greek society until the 19th century. Katharevousa seems to have represented the last significant push of atticist thinking, but wasn't so much a revival of Classical Attic as it was a language policy to reverse changes in Modern Greek. The atticism up to the Byzantine period was the use of Classical Greek as a literary language of educated elite, similar to the status of Classical Latin in Western Europe long after it was already a dead language (compared to the vernacular). There have been neoclassical atticist coinings, such as the 19th century addition of words like Μινώα (Minṓa) by a non-Greek archaeologist hearkening to the mythological king (and legitimately ancient word) Μίνως (Mínōs). But these are no longer specifically a Greek or Hellenist tradition, but a study of Classical Greek as a historical language by various people from around the world, who all have their own ways of pronouncing it. - Gilgamesh (talk) 23:47, 1 June 2012 (UTC)
Your points are very good. However you focus very much on katharevousa and I had no intention to confuse katharevousa and how it influenced Greek pronounciation. The fact is that across Greece and Cyprus in schools and universities only one accent is used. For example, see how at en.wikipedia only the grammar of Standard Modern Greek is analysed and mainly the phonology of Standard Modern Greek. I propose we add a Standard Modern Greek pronounciation row; after all it is the same language in different periods. For example, take the word πολύτροπος. This is an Homeric word and it was not used in the 4th or 10th cc. AD. Of course it may have been used only when reciting the Odyssey. However we include its pronounciation. The same word is used today in schools and universities in Greece and Cyprus using a universal pronounciation, that of Standard Modern Greek. Be aware that this Greek is indeed a koine of the Greek–speaking world. When one, for example, comes across an ancient word containing <κτ>, there is no other way than to pronounce it as /kt/. You know, I am afraid as you are about politics in wikis but I think that this is not a problem.

Moreover, there are hundreds of words that are exactly the same throughout the history of Greek. This is why I think we should make possible any comparison between the different pronounciations of Greek.Dimboukas (talk) 02:49, 2 June 2012 (UTC)

Perhaps there could be a row for Modern Greek, but somehow marked to show that it's treated by WT as different than Ancient Greek.

Ahh, perhaps I misunderstood. Well, if you'd like to chip in, the discussion is mainly at Wiktionary talk:About Ancient Greek. We could certainly use more and greater consensus, that's for certain. - Gilgamesh (talk) 06:15, 2 June 2012 (UTC)

Ok then. I am going to support the new change at Wiktionary talk:About Ancient Greek. Dimboukas (talk) 11:44, 2 June 2012 (UTC)

Σουσάννα descendants[edit]

Hi. Armenian Շուշան (Šušan) and Georgian შუშანიკ (šušanik) can't be from Ancient Greek. Greek does not have š. --Vahag (talk) 13:28, 2 June 2012 (UTC)

Are you certain that's relevant? Sounds can change between languages. To be honest, I don't know enough about Armenian and Georgian etymologies to know what sounds mutate into what. Alright, if they're not Greek loans, then from what language? - Gilgamesh (talk) 19:18, 2 June 2012 (UTC)
Greek σ always yields s in Armenian and Georgian. Armenian Շուշան (Šušan) is formed natively from շուշան (šušan, lily), a borrowing from Syriac or Middle Persian. Georgian შუშანიკ (šušanik) is from Armenian. --Vahag (talk) 20:34, 2 June 2012 (UTC)
I see...very well then. - Gilgamesh (talk) 00:50, 3 June 2012 (UTC)

grc-ipa-rows (2)[edit]

Hello. I just noticed your latest edits on that template. The creation of that template was a really great idea, but I am not sure that the latest version is a real improvement.
For example, I noticed that a specific consonant cluster (ξ /ks/) is rendered in a quite peculiar way; for example in πρᾶξις#Pronunciation one can read: /práːkʰsis/ > práxsis > /práksis/; well that intermediate /xs/ stage is not documented by any book on historical Greeek phonology I know, and the aspiration in /kʰs/ is probably redundant since it doesn't seem to be phonemic; plus, you distinguish á͜a from áː while they are just two different ways to depict the same sound (this is documented in the IPA handbook). You also give ρ /ɾ/ for Modern Greek, while most sources give ρ /r/.
A more general criticism I would like to make concerns the detailed periodization ("5th BC", "1st BC", etc.). While there does exist meaningful speculation about the specific changes that occured in each period, one has to be very careful when incorporating them to a historical phonology template. For example, on what source do you base the assumption that ευ was ɛʍ in the 1st c. BC, and eɸ in 4th c. 4th c. AD? These changes probably did occur at some point or another; however their exact dating that you attempted is virtually impossible, and I doubt that any reliable source on historical Greek phonology puts forth such bold claims.
I would suggest that the template be reviewed by all Wiktionary editors contributing to Ancient Greek entries and appropriate changes be made to it, and I would also suggest that we add sources to the template's documentation page. Your input will be wellcome. --Omnipaedista 09:15, 6 November 2012 (UTC)

We already had discussions about this, addressing various points. You can contribute more if you like. It's just...there's always going to be someone who has issue with something for issues of school of thought or notions of what is proper.
We use /kʰs/ and /pʰs/ because the letters were phonetic realizations of χσ and φσ respectively. We know how the Romans heard them and wrote them, and we have a general idea of when θ φ χ became fricatives, but there seems to be no real evidence that they became analyzed specifically as combinations of κσ and πσ until into the Byzantine era, when sandhi governing clusters of voiceless plosives and voiceless fricatives had started to permeate the language's phonology. We got period pronunciation data specifically from the Wikipedia article Koine Greek phonology, which is why the pronunciations indicate locations and centuries.
As for vowel ties for long vowels, we use them because editors have disagreed over what convention to use for them. Earlier, we had been using conventions like /ɛː/ for η, /ɛː́/ for ή and /ɛ́ː/ for ῆ, indicating the mora-timed position of the pitch accent. You see, some editors believed it inappropriate for an IPA diacritics ever to be placed directly over /ː/ instead of over the vowel. So the compromise was to instead indicate all long vowels (including monophthongs) as tied multi-symbol vowels, so that we could place the accent over a vowel symbol every time, and have consistent vowel transcription throughout. This way, we use /ɛ͜ɛ/ for η, /ɛ͜ɛ́/ for ή and /ɛ́͜ɛ/ for ῆ. It was the solution that offended the stylistic sensibilities of the least number of participating editors.
And for Modern Greek, we use /ɾ/ because that's how it's realized in all the modern dialects where consonant gemination is not phonemic, including the standard dialect of the Greek state. Modern Greek linguists tend to be insistent that, whether the symbol is /r/ or /ɾ/, it is always a flap consonant. And the standard symbol for an alveolar flap is /ɾ/. It has the fewest question marks, really.
Even on Wiktionary, there can be deeply entrenched dictionary politics, and Greek historical linguistics is especially fraught with politics. Now, rather than responding about this here, could you go to Wiktionary talk:About Ancient Greek and participate with the group? It's less complicated that way, and may invite more varied feedback. - Gilgamesh (talk) 09:52, 7 November 2012 (UTC)
I had read the discussions before I commented here and they do not adress any of the points I made. Regarding ξ and ψ: as I said above phonetic variations ([kʰs]) are irrelevant to phonemic transcriptions (/ks/). Allen's Vox Graeca does not support the phonemic analysis of ξ as χ and σ; it regards χσ a strictly phonetic phenomenon. Regarding long vowels: (aside from stylistic conventions) I insist that contrasting á͜a with áː (as is done in the entry πρᾶξις) makes no sense; the IPA handbook does not make such a distinction. Regarding Modern Greek: ρ is standardly rendered as a trill. Regarding politics: I am well aware of the political side of the matter, but I still believe that we can come up with a sensible way to include Modern Greek pronunciation in the template. Thank you vey much for the reply! I will continue the discussion in the relevant page. --Omnipaedista (talk) 08:09, 9 November 2012 (UTC)
For pʰs kʰs, we're mainly following sources like wikipedia:Ancient Greek phonology#Other consonants. It is our understanding that voice and aspiration of plosives were not phonemically distinct before s, making allophones of ps~pʰs and ks~kʰs. I suppose the notation for that era can be simplified without losing anything.
On á͜a vs. áː — it's rather moot because we don't use ː anywhere for vowels, and the other editors oppose using notation like aː́ where diacritics go over the ː. If the pitch happens on different parts of the two-mora long vowel, then when it falls on the second mora, a͜á was preferable to aː́. And then we used a͜a and á͜a for visual consistency. So would you rather have aː aː́ áː or a͜a a͜á á͜a? We need to pick one or the other, and we picked the latter because ː́ was considered improper for any IPA notation. Using double vowel letters as well as double consonant letters also makes clearer the mora-timed nature of the Attic language.
And it's been pretty emphatic that we cannot use modern pronunciation for any stage of Ancient Greek — Ancient Greek ends at 1453, and any pronunciation from after that period would be no more notable in modern Greece or modern Cyprus than it would be in modern France, modern Italy or modern Turkey; it gives modern Greece undue weight in studying ancient Greece. All modern styles of pronunciation of a pre-1453 language are learnt and artificial, anywhere in the world. We actually used to have modern Greek pronunciation listed, but it was delisted after this topic was discussed. You may raise this and any other discussion topic again if you wish, but please raise it at Wiktionary talk:About Ancient Greek, not here. - Gilgamesh (talk) 03:41, 11 November 2012 (UTC)

Ambox blue question.svg
This blocked user is asking that his or her block be reviewed:

Gilgamesh (block logactive blocksglobal blocksautoblockscontribsdeleted contribsabuse filter logcreation logchange block settingsunblock)

Request reason:

My addition to the crank entry of the slang "penis" usage is genuine. It has already long been featured in Wikisaurus:penis. It is independently attested at Urban Dictionary (definition #7). I have heard this slang term since at least the 1990s, and it even featured in the script of an episode of the TV series American Dad, episode "Roger 'n' Me": "I can't wait... to massage... your... Brazilian... crank." I do know it's slang, though have never been certain how vulgar it can be considered (outside of clearly established vulgar context, of course).

As for my changing of English IPA transcriptions from /ɹ/ to /r/, this is in line with phonemic transcriptions of the phoneme, as addressed in w:English phonology. The actual rhotic can differ, and [ɹ] might be relatively more accurate in a narrowly phonetic transcription of an accent. On the other hand, there is a phonemic split in General American among younger generations between /æŋ/ and /eɪŋ/, with most -ang words taking on the latter pronunciation, but /æŋ/ remaining in a few words like anger, anguish, etc. (but not words like angle), such that anguish and languish do not rhyme.

penis, dinky and D[edit]

I have reverted your addition of "dinky" and "D" from WS:penis, the two terms do not have attesting quotations in Wiktionary mainspace (for dinky, at dinky or Citations:dinky). If you can attest the terms as per WT:ATTEST, you can add them back. Note that Urban dictionary cannot be used for attestation, nor can any other dictionary anyway. The terms have to be attested in use. --Dan Polansky (talk) 17:17, 17 July 2013 (UTC)

I see. If I find additional sourced attesting quotations, I will supply them. Out of curiosity, is there a particular reason why Urban Dictionary and other dictionaries cannot be attested? That seems to rather drastically limit the available pool of reliable sources. - Gilgamesh (talk) 13:21, 18 July 2013 (UTC)

Regional forms of English[edit]

Hi Bilgamesh! I was wondering, do you think it would be possible to generate programmatically regional pronunciations of English? Something along the Ancient Greek template. Input would be some kind of lossless dialect-agnostic spelled pronunciation. Now that we have Scribunto extension installed, we can map correspondences such as the ones you listed here very easily in Lua (it's basically search/replace with regular expressions). No more template horror. --Ivan Štambuk (talk) 03:08, 16 December 2013 (UTC)

That would be a tempting idea — and while regional varieties of English often have regular patterns of difference, there are enough word-by-word vernacular exceptions that I don't think it would necessarily be a good idea to fully automate through templates. Also, Classical Greek was a relatively archaic standard even 2000 years ago, and it had no vernacular beyond its original time period, and its pronunciations in later time are more reflections of the educated phonotactics of later ages rather than actual vernacular pronunciations. Anyway, I'm not sure I fully understand what you're telling me — what's Scribunto? Also, I'm not as alert and attentive as I could be, as my region is having a dreadful winter smog that's making it hard for me to breathe or to think clearly. - Gilgamesh (talk) 03:47, 16 December 2013 (UTC)
Also, that link — User:Gilgamesh/English vowels — was an informal user page project, and I recall deciding later that I had made mistakes on it. But with the current weather, I'm not sure I have review it and explain why, right now. - Gilgamesh (talk) 03:49, 16 December 2013 (UTC)
Scribunto is this. It's an extension for MediaWiki software which runs this site. It enables us to write "modules" in a programming language called Lua. With it, one can make conversions and substitutions of characters in strings, such as "replace X with Y if X is not followed by a front vowel, and not at the beginning of a word". Or categorize entries on the basis of their phonological structure (which is usually done manually). It's just an idea. English does seem a hard nut to crack, because of its many variants and tricky spelling. It's a long-term project which would require many test cases and surveying many sources. Both diachronic and synchronic descriptions are equally exciting possibilities. I.e. the evolution of modern languages from their ancestors (like we have for Ancient Greek, but further, or Latin-Romance, Old - modern English, even protolanguages to modern), as well as the current situation ("what pronunciations can this spelling stand for"). However, the primary motivation for this is to eliminate the human factor where it can be eliminated - e.g. there are lots of irregularities with phonemic transcriptions transcribing subphonemic (non-distinctive) features, different editors using different symbols, and similar. --Ivan Štambuk (talk) 04:14, 16 December 2013 (UTC)


Would you like your empty subpages actually deleted? (If so, you might like to keep their former content offline somewhere first) SemperBlotto (talk) 11:39, 15 April 2014 (UTC)

I...don't know yet. I'm actually okay with having a lot of blank user pages. What I'm not so comfortable with is having abandoned and stale projects of poor research quality remaining here to represent me and to represent Wiktionary's user community. It's easier to wipe them so they don't show up in Google searches, but I don't know if I'm prepared to actually have them deleted.
I mean, you know, I have a sometimes irresistable urge to theorize, document, toy with linguistics ideas in new and interesting ways. But I have a bad habit of abandoning my projects or letting them go stale, and many of them feel more like glorified blog posts that feel uncomfortable trying to stand on their own, especially after so long and so many new things I have since learnt. I may still create user pages, but I'm now more interested in them being genuinely useful, even if they're still not intended as an accredited reference. Stroking my penchant for linguistic whimsy can only be so appropriate on a site as formal as Wiktionary. - Gilgamesh (talk) 11:49, 15 April 2014 (UTC)