Wikipedia talk:Naming conventions (Greek)

From Wikipedia, the free encyclopedia

Contents

[edit] Transcription of Classical ει

ει tends to be transcribed as any of i, ei, or e. Ideally, we need a set of rules to cover this. Either we need to look at the context (e.g. before alpha vs interconsonantal etc) or we need to consider the origin (eg ε+ε vs lengthened ε vs diphthong) --Nema Fakei 14:55, 28 February 2006 (UTC)

[edit] Modern γ

Shoud γ be written 'g' or 'gh'? 'gh' is more phonetically correct, but 'g' is less unwieldy.--Nema Fakei 18:28, 6 March 2006 (UTC)

γ is pronounced /j/ (ie a 'y' sound) before ε or ι eg γηρος is 'yeeros'. 202.82.109.151 12:27, 2 August 2006 (UTC) ςς

[edit] Modern Greek

See Transliteration of Greek to the Latin alphabet for some more transcription schemes. What I've been using for Greek geographic names is the UN/ELOT system, but there are several more in use, including ones in which γ is 'gh'. An advantage of the UN/ELOT system is that it is nearly reversible, i.e. you can (with some exceptions) get the original Greek from the transcription. A disadvantage is that the transcription doesn't tell you how to pronounce it (if you don't know modern Greek). Markussep 01:37, 12 March 2006 (UTC)

Hm, thankyou. Hopefully, we can eventually include IPA for the pronunciation.--Nema Fakei 02:25, 12 March 2006 (UTC)
See also Greek alphabet, the modern (and ancient) pronunciations are shown there. Adding IPA is a very good idea, much better than "pronounce: BUH-nee". Markussep 11:12, 12 March 2006 (UTC)

[edit] Proposal: UN/ELOT

To make this policy rather than a reproduction of the encyclopaedia article, and also to be in any way useful, we need to set a standard for WP articles so that others can refer to this page and reolve disputes: listing alternatives is no good. I'd say use the UN/ELOT system for reversibility. Pronunciation is secondary, as we can use IPA. Comments/objections? --Nema Fakei 17:06, 22 May 2006 (UTC)

I agree: use UN/ELOT, and give the IPA pronunciation as well. I think we should invite some others to give their opinion as well, we can't decide this policy just between the two of us. I'll post it at RfC and Talk:Greece. Markussep 19:08, 22 May 2006 (UTC)

[edit] Diacritics

I think we should say something about diacritics in article titles. See e.g. the discussion at Talk:Haliacmon. The official UN/ELOT transliteration keeps both the acute accent (ý) and the diaeresis (ÿ). Currently, virtually none of the Greek article titles have diacritics, maybe it's best to keep it that way. Markussep 11:49, 27 July 2006 (UTC)

[edit] IPA for Ancient Greek?

I don't see the point of adding pronunciations (IPA or otherwise) for Ancient Greek. It is especially pointless for placenames, the pronunciations of which participated in all the sound changes documented in Ancient Greek phonology. --Macrakis 22:56, 22 May 2006 (UTC)

Well, I think they should be optional. You're right that it's probably inappropriate for placenames. But some Greek names just look weird to those unused to them - there's often no intuitive pronunciation, and WP should fill that gap. --Nema Fakei 00:21, 23 May 2006 (UTC)

[edit] "Scientific Transliteration"

I would strongly suggest making "scientific transliteration" (given at Transliteration of Greek to the Latin alphabet) the preferred transliteration for Ancient Greek. It allows for nearly perfect one-to-one conversion between the original text and transliterated text (not considering polytonic orthography of course). Ardric47 04:02, 1 June 2006 (UTC)

I second that, without any expectation that it would ever be implemented. The scientific transliteration is applicable for any form of Greek and it preserves the accurate and reversible orthographical structure of the text, allowing anyone to reproduce the original spelling and to read it however they like (whether in Erasmian or modern Greek). It should be noted that there is now a tendency to substitute kh for ch, which is justified by both derivation (since χ is the aspiration of κ) and by the gross mispronunciation of Greek words rendered with ch by even the specialists who ought to know better (the beginning of Chalkedon has nothing to do with English chalk, nor does the beginning of Chephren resemble French chef).
However, there is a general dislike for diacritics by English readers, and it may be difficult to enter them in search engines and the like. This can be surmounted by transcription (as opposed to transliteration), omitting the diacritics (e.g., o for ō, ω, and e for ē, η). This system is largely that adopted by the Oxford Dictionary of Byzantium for transcribing those Greek names that do not have a common English equivalent, e.g., Andronikos, Leontios, Nikephoros, Spandounes. Something like this is gradually becoming more common in treatments of Ancient Greece as well. From among the books that I use with my own students, I will just mention the Penguin Atlas of Ancient Greece and Dillon and Garland's Ancient Greece: Social and Historical Documents from Archaic Times to the death of Socrates. However, it should be noted that these works generally preserve "household" names such as Homer, Hesiod, Thucydides, Plato, Aristotle, Ptolemy, Philip, and Alexander, while rendering others more faithfully as Peisistratos, Kleisthenes, Aristoteles (Battos I of Cyrene), Themistokles, Perikles, Alkibiades, Ptolemaios (for a non-royal Ptolemy). The basic problem with these systems is where to draw the line. One of the books mentioned above uses Socrates but the other Sokrates. Here this problem is actually minimal, because of a feature unique to Wikipedia: hyperlinked redirects. Ultimately the question is less whether an article is called Socrates or Sokrates, but whether the reader wants to see a Latinism, Anglicism, or something like the original Greek in the text.
Best, Imladjov 17:41, 4 June 2006 (UTC)
Can I point out that some Greek-derived english words are normally pronounced reasonably properly anyway - /kiˈɹopɘdist/, not /tʃiˈɹopɘdist/ (and note that in the first case, [kʰ] is a member of /k/)? I'd like diacritics to be more common, too, but when writing articles, it's just not practical to use them all the way through.--Nema Fakei 19:51, 4 June 2006 (UTC)
I think the real problem with diacritics would be article names, which must be readily searchable. Other than that they may merely be tedious to put in (it is much easier in a word processor like MS Word). I just stumbled across an illustration of the diacritic-less transliteration in use: [1]. Best, Imladjov 04:37, 5 June 2006 (UTC)
I'd also like to throw my support behind something more like the "scientific" transcription for Ancient Greek. Latinisms and Anglicisms are all very well, but in articles devoted to AncGreek culture it seems most appropriate to reproduce AncGreek words using a transcription which is as near as possible to a one-to-one transformation of Greek letter to Latin without being too great a challenge to read or type.
The "scientific" approach has the advantage of employing a single diacritic (the macron, ¯) with only two letters (e and o) to distinguish ε from η and ο from ω. The macron is probably among the least off-putting of diacritics to most English-reading readers as it does not suggest a "foreign" sound so much as specifying one "native" variety. Some readers will even properly recognize the macron as indicated a long vowel — though what long and short mean to them may not apply to AncGreek vowels at all. The purpose, however, is to allow recovery of the original spelling, not to provide any guide to pronunciation.
To make the "scientific" approach easier to type, my suggestion would be to only include the macrons in the first use of a given term in its own article, with subsequent occurrences presented without macrons. That would provide readers with the extra detail the diacritics can provide, without requiring editors to repetitively interrupt their typing to insert the macrons, and without requiring readers to repetitively process or ignore them.
I could be persuaded either way as far as article names go. My own experience with search engines is that they disregard diacritics in both queries and indices — so that ā matches a in both directions. And redirects can make up for any remaining gaps. On the other hand, I have a certain attachment to URLs without encodings.
I'm curious, though. Recent comments here suggest a degree of consensus. How do we think we might proceed? The convention can be edited, but what are the best avenues to communicate (or suggest) it to the editors of the sprawling family of articles where AncGreek words occur? I actually started looking for the convention because I'm tired of passersby "correcting" the spelling of names in Atlas (mythology), and wanted to know if there was a convention in use. It wasn't that easy to find. If some form of the "scientific" transliteration were adopted here, how would anyone else know about it? --Americist 16:25, 16 June 2006 (UTC)
The big problem I have with the scientific transliteration is that you end up with very unfamiliar names: Thoukudidēs for Thucydides. General WP naming policy is to always use the most common name, and the 'traditional' Romanisation is almost invariably the best known. Therefore, what about just using the traditional notation throughout the article and for the title, but always add the scientific name as an alternative in bold? e.g.
Thucydides, Thoukudidēs, Greek: Θουκυδίδης, (between 460 and 455 BC–circa 400 BC), was an ancient Greek historian...
To be honest, if you're looking for faithfulness to the Greek, why not just make Θουκυδίδης the article title? --Nema Fakei 18:00, 16 June 2006 (UTC)
I'd like to throw some qualified support behind the "scientific" transcription; the qualification is exactly what Imladjov mentions above, that the Romanized spelling should be retained for household names like Homer, Hesiod, Thucydides, etc. For these household names, we can follow the format Thucydides, also Thoukudidēs, (Greek: Θουκυδίδης), using "Thucydides" in the rest of the article. For less commonly known figures, we could use scientific transcription throughout.
The problem, of course, is what the line is between household figures and lesser-known; for this I have no useful input, as every person who sprang to mind (Kleon, Stesikhoros, Rhianos) appears in Wikipedia under Latin transcription (the influence of the 1911 Britannica shows here). Given WP's general policy of using the most common name, I don't think I'd be too upset if every article used the "Roman standard", as long as the scientific transcription was also given (e.g., the Achilles article, which starts "In Greek mythology, Achilles, also Akhilleus or Achilleus (Ancient Greek Ἀχιλλεύς)..." --Akhilleus (talk) 04:57, 4 July 2006 (UTC)

Could I argue for the form "Greek Σωκράτης Sōkrátēs" (as I originally put it in the Socrates article, though it has since been changed)? This only needs to appear once and the familiar form can then be used afterwards. The advantage is that those that can read Greek have the correct form, and those that cannot get the closest possible (without loss of information). Those that abhor diacritics only have to face them once. m.e. 12:38, 2 August 2006 (UTC)

I agree that there is definitely a consensus on the issue in general, but there are still details to sort out. I'm therefore making a detailed statement of what I see as the remaining issues.
First of all, I strongly support "scientific transliteration" for names used in the Archaic, Classical, Hellenistic, and Roman periods (I say "used in" to catch mythological names too).
On Latinisms: there is a good reason to retain some Anglicised forms (like Homer, Aristotle), since English is the language of the articles; but I feel there is no good reason to retain Latin forms, except for one case which I note below.
On diacritics: I oppose the use of diacritics in article titles, as (1) it's really really hard to type a URL with macrons; (2) they are not universally used by those who use "scientific transliteration". But in article contents, I would say that the use of macrons to represent eta and omega should be accepted as an optional thing. I oppose the practice of using both tonal accents and macrons in transliterated names (as m.e. suggested), because Unicode doesn't provide for a vowel to have both a macron and all possible accents unless you use combining diacritics; but combining diacritics are extremely poorly supported by existing software and fonts.
On the Byzantine period: there may be more grounds for debate over what direction the policy should take here. Aldux has pointed out to me a dispute that arose not long ago over Byzantine rulers: Wikipedia:Mediation Cabal/Cases/2006-05-23 Names of Byzantine rulers. In that case, arbitration decided on "scientific transliteration" based on Classical phonology (though it looks like some people are still not happy with the arbitration). Personally I would have thought transliteration according to modern phonology would be more appropriate, but I'm not a Byzantinist so can't really state a preference. I suggest a policy of tolerance until there are enough Byzantine enthusiasts around to swing it one way or the other.
On the letter υ outside diphthongs: I favour "y" in preference to "u", because (1) "y" is the letter that represents the sound of Classical Attic υ in many European languages that use the Roman alphabet, especially Germanic languages; and (2) English "u" already represents too many different sounds. At present, English-language academic writings vary on how they render this letter; German-language writings universally use "y". I suggest there's no reason to create a conflict with that.
On the letter χ: in academic writings that use "scientific transliteration", "ch" is more commonly used than "kh". Could people perhaps be persuaded to accept that, either for that reason, or as a concession to Latin-style transliterations? I really don't think there are going to be many people who will get confused with the English pronunciation of "ch" as /tʃ/.
As a summary of how to apply transliteration schemes, I suggest the following (though I should note that I do not feel quite as strongly about the Roman period as the others, but for complicated reasons):
  • Bronze Age: use existing conventions for Mycenaean Greek.
  • Archaic, Classical, Hellenistic, and Roman periods: "scientific transliteration", with no macrons in article titles. Exceptions:
    • Anglicisms: exceptions can be made for names where no one ever uses the Greek form in English, e.g. Homer, Helen, Priam, Penelope, Lucian, Pindar, Aristotle.
    • Latinism: the one solitary name I can think of where everyone, but everyone, always uses the Latin form, is Plato.
  • Byzantine period: not many strong feelings have been expressed in the discussion so far. For rulers, the nature of WP seems to make the arbitration process I mentioned above authoritative. For Byzantine writers, I propose that it be optional whether to use "scientific", Latin, or modern Greek transliteration, until such time as there are enough Byzantine enthusiasts around who have an overwhelming preference for one or the other.
  • 1453 and later: use existing conventions for Modern Greek.
With regard to implementation once a policy has been made, I suggest a policy of tolerance. New articles should of course follow the policy given by the page on naming conventions, whether the current one or whatever replaces it; old articles to be moved at discretion and convenience.
Nema Fakei and Akhilleus, while unfamiliarity is an issue, I suggest that that's exactly what redirect pages are for. Spellings like "Thoukydides" and "Aischylos" are already used by many academics, and the usage is increasing (except in journals with strict editorial policies); I really don't see any reason why they should not be accepted as the norm. But if there are going to be exceptions, there should probably be a centralised list -- not an exhaustive one, just an ad hoc one to prevent move-wars.
Anyway, there's a consensus, but details need to be worked out. Could I invite suggestions on the following specific issues?
  1. The chronological division I outline above, esp. w.r.t. Byzantine period -- this needs more discussion
  2. Use of diacritics -- this seems up in the air, esp. w.r.t. article titles and use of tonal accents
  3. Transliteration of υ (and perhaps χ) -- this seems up in the air
  4. Should Anglicisms be condoned? -- the answer so far seems to be yes
  5. Should any Latinisms be condoned? -- there's been a couple of yeses, but I'd suggest no
Cheers, Petrouchka 21:23, 5 October 2006 (UTC)
Edit: one clarification needed, I think, and I want to retract something I said earlier.
(1) diacritics: the problem with using macrons and tonal accents in transliterated names is the combination (vowel + macron + circumflex/perispomene). Unicode provides for combinations with grave or acute accents: 1E14-1E17 (for "e") and 1E50-1E53 (for "o").
(2) I retract what I said about being against any Latin forms of names. It's not remotely realistic, and it would inevitably upset people. Petrouchka 01:59, 6 October 2006 (UTC)
Petrouchka, this may be dense of me, but could you clarify what you mean by "Anglicism" and "Latinism"? I assume that by Latinism, such phenomena as -us for Greek -ος, "Aeneas" instead of "Aineias" is meant. "Helen" instead of "Helene" is an Anglicism. But I'm unsure whether "Achilles" is a Latinism, Anglicism, or both. I'm guessing that what I referred to before as "household names" are Anglicisms, for the most part (e.g. Thucydides, Alexander, Hesiod). "Ajax", though, is a Latinism, right?
Anyway, on your point #2, WP seems to avoid diacrictics in titles, for the most part (see Wikipedia:Naming_conventions_(use_English)#Disputed_issues), so I think titles should be diacritic-free. I can think of some problematic cases, though--what if someone decides to write an article on timē? Within the body of the article I'm inclined to let editors do as they see fit--personally, I'd mark eta as ē and omega as ō, and not give accents. The problem you note with combining circumflexes can thus be avoided.
On point 3, though I prefer "kh" for χ I can live with "ch" instead. "y" for upsilon seems ok in general, but I wouldn't want to move hubris to hybris.
On 4 and 5, as you note, straying from well-known forms will upset people; I'm not sure whether it's worth compiling a list of names/terms where we use an anglicized/latinized form instead of a transliteration, as there are many (Homer, Hesiod, hubris, Helen, etc.) --Akhilleus (talk) 03:39, 6 October 2006 (UTC)
I'll just pop up again briefly to clarify what I said above about Anglicisms vs. Latinisms: by Anglicisms I meant forms which have been changed after being borrowed from Latin. E.g. Greek Ὅμηρος => Latin Homerus => English Homer, i.e. dropping inflected endings altogether; I didn't have in mind names like Thucydides, where the English spelling follows the Latin. I suppose you could argue that a name like "Thucydides" is an Anglicism, in the sense that it's pronounced very English-ly. Perhaps, if the policy were to change, Anglicised pronunciation could be a useful criterion? Petrouchka 04:24, 7 October 2006 (UTC)
Please keep in mind that for ancient Greek the use of Latinized forms is overwhelming; changing would mean to pass to have to rename literally thousands and thousands of articles, and create an endless number of redirects. Also, the use of Latinized forms is: 1) promoted by this naming conventions guideline 2) use by most sources; for example the Oxford Classical Dictionary.--Aldux 11:54, 6 October 2006 (UTC)
Latinization is the standard English approach to Greek. The Hellenist has no particular reason to care how Θουκυδιδης is transliterated; he knows who is meant, and will choose transliterations for his purpose. The Greekless anglophone will expect Thucydides, if anything; and be confused by anything else. Septentrionalis 16:23, 20 October 2006 (UTC)
Apologies - I've only just realised about this section (if I did see it, I have quite forgotten). In addition to echoing Septentrionalis, let's see, Petrouchka...
First, can I point out that because, as you note, following a scientific translation forces us to write Mycenaean as Mycenaean (otherwise the whole thing would be a bit silly), we then have to write Orestes as Orestas and other things that no academic outside of mycenaean epigraphical studies writes.
Second, while a select few people do write all these funny transcriptions, they generally *say* Ajax, not Aias.
Third, all 'household names' come via Latin. If we're writing about half of our Greeks with Latin-based names and the other half with scientific transliterations, then articles are going to look silly. Most academics who write Teukros also write Akhilleus.
Fourth, what's a household name? Plato, sure, but Zeno? Phaedo? Crito? Meno? Yet I've never seen any of these called Menon or whatever - in English, anyway. That's just asking for debate.
I would love it if we all had better systems than the one that's most common, but this is supposed to be an encyclopaedia for everyone to read, and it's no good if we fill it with transliterations that make best sense mostly just to people who know a lot about the subject already. For starters, they're not so much the ones who'll be relying heavily on WP.--Nema Fakei 21:17, 20 October 2006 (UTC)
I just want to register support for direct transliteration, excluding Latinisms but allowing common Anglicisms. Academic and general practice is (slowly) moving this way: see for example the Cambridge Dictionary of Classical Civilization (2006). Current practice on post-640 Byzantine articles in Wikipedia (after fierce debate, and subject to review) uses a similar direct transliteration.
But I think, in your box above, Petrouchka, you have to add Strabo to Plato. Andrew Dalby 09:49, 25 October 2006 (UTC)

[edit] Italics in Greek characters

Is there any agreement on whether to italicize or not Greek characters ? If not, I propose to adopt a guideline advising against the use of italics in these cases:

  • Italics are not necessary, since the difference with "normal Latin text" is obvious.
  • Italics hinder readability, at least for those of us not used to those funny characters :-)

Of course, there would be exceptions, as for the "Bibliography" and "References" sections, where italics in these scripts do tend to make sense.

I imagine something very simple, along the lines of:

Do not use Italics for the following cases:
  • Foreign language words and texts in Greek characters, such as Ελληνικό.

I'm posting this in Wikipedia talk:Manual of Style (text formatting)#Italics in Cyrillic and Greek characters & Wikipedia talk:Naming conventions (Cyrillic)#Italics in Cyrillic characters too. - - Regards, Evv 03:16, 12 October 2006 (UTC)

I'd agree with this, with the more general consideration that italics are unnecessary if you're switching to any alphabet other than Roman. Petrouchka 07:26, 12 October 2006 (UTC)
I agree, my mistake :-) For that reason, I guess that the main discussion will be in Manual of Style (text formatting), and that here we sould probably talk about "Greek only" issues, if there are any.
At the same time, it may be a good idea to add a simple "on Greek characters only" guideline to this project page, in a last section titled "Style recommendations" or similar (providing there a link to Wikipedia:Manual of Style (text formatting).
Best regards, Evv 03:59, 13 October 2006 (UTC)

[edit] Greek characters with diacritics

The article Olga Vasdeki has been moved to Ólga Vasdéki. The Greek name given in the article is Όλγα Βασδέκη, i.e. with diacritics. However, there seems to be no precedence in Wikipedia for having diacritics in latinized Greek names, do you think the move should be reverted? Punkmorten 10:18, 2 February 2007 (UTC)

Currently there is a rule in this naming convention that for personal names the transliteration preferred by the person in question should be used. I'm not sure whether that extends to the use of diacritics. Anyhow, I think we should add a rule about diacritics to this naming convention. Currently, there are virtually no city or town articles with accents in the title, and very few articles about persons. There has been some discussion about this at Talk:Haliacmon and Talk:Greece/Archive 1#Naming convention. My personal opinion is that the accents are not necessary in titles, except maybe the diaeresis, since it modifies the sound of the vowels. If we add a rule that the name as written in Greek alphabet should always be given, people will know where the accent is anyhow. Markussep 10:50, 2 February 2007 (UTC)

A small test of how other media handle this: encyclopedias:

news sites:

This doesn't really make things clearer, I know. Markussep 16:52, 22 February 2007 (UTC)

Some greeks prefer not to use the accents in Latin transcriptions, and we should assume no accents unless we have evidence that they prefer it (e.g. a book published in their lifetime). As the talk page lists no discussion of the move whatsoever, I rate it as enforcing preferences alongside changing -ise to -ize or vice versa with no good reason. --Nema Fakei 23:54, 22 February 2007 (UTC)
My opinion is that accents are not commonly used in transliteration from Greek (and also make searches more difficult) and it's much better to omit them, certainly from titles of pages. I think I'm agreeing with Markussep and Nema Fakei. As observed above, if we give the Greek script, anyone who looks will see where the accent is. Andrew Dalby 09:56, 23 February 2007 (UTC)

Based on the above comments, I reverted all his page moves. If he wants do move these pages, he can discuss it here. Moving without discussion is henceforth considered disruption. Punkmorten 10:27, 23 February 2007 (UTC)

I'll add a comment about using accents to the (still proposed?) naming convention. Markussep 15:33, 23 February 2007 (UTC)
There are no different search results (in Google at least) for Ioánnina and Ioannina (or whatever accented word). I don't really think accents are an issue for searching and/or copy-pasting. On the other hand, the little accent shows the syllable the reader should stress in Greek, if they want to pronounce correctly. I see no reason why this should be an issue (haven't noticed any edit war over something so trivial), and I propose that it is based on the editor's judgment for the complexity of the word. For example, in English we would say "PhilhEllenism", while the Greek original is "PhilhellinismOs". I think the editor should be free to stress the ...stress if needed. NikoSilver 16:18, 28 February 2007 (UTC)
Sure, the stress (and pronunciation) should be indicated in the article text (first line). The discussion was whether the article title should carry an accent. It's not a big issue, I agree. Markussep 17:31, 28 February 2007 (UTC)
The issue is consistency. If one name is changed, all must be changed.. Punkmorten 11:28, 1 March 2007 (UTC)

[edit] Diacritic vs. monotonic Greek

When should diacritic or monotonic Greek be used in articles? – Zntrip 03:07, 28 February 2007 (UTC)

My opinion: polytonic Greek should be used when dealing with ancient and medieval (Byzantine) Greek names:
  1. ancient because the different accents and breathings mark real differences in the pronunciation at that period
  2. Byzantine because they wrote more-or-less ancient Greek using the accents, even if they didn't pronounce it the ancient way
Monotonic Greek is currently standard and makes sense in dealing with modern names. Andrew Dalby 14:14, 28 February 2007 (UTC)
Retrieved from "http://en.wikipedia.org../../../n/a/m/Wikipedia_talk%7ENaming_conventions_%28Greek%29_b147.html"