Wiktionary talk:About Hebrew

Archives: 2007 · 2008 2009–10 2011–16 Shortcut: WT:T:AHE

Index:Hebrew/ט

Since Conrad.Irwin (who created the language indices) has been gone for a long time, I wrote something to update the Hebrew index, while hewing to the existing design to the extent that I could reverse-engineer it. So far I've only updated Index:Hebrew/ט, to solicit some feedback before proceeding with the other 21 letters. Questions/comments/etc. welcome. —Ruakh_TALK 21:12, 10 June 2018 (UTC)

@Msh210, Wikitiki89, Enoshd — no pressure, but if you have the time and inclination to take a quick look at Index:Hebrew/ט and/or Index:Hebrew/ט?diff=49705178 and let me know if any problems jump out at you, I'd appreciate it. —Ruakh_TALK 04:55, 11 June 2018 (UTC)

I personally think that there isn't much use in keeping such an index (for any language). We have Category:Hebrew lemmas after all. And for all the redlinks, we have Wiktionary:Requested entries (Hebrew). Other than that, your update to the index looks perfectly fine. --WikiTiki89 13:38, 12 June 2018 (UTC)

I too think that indices are no longer useful for Wiktionary. Our Hebrew content needs a lot of work, and there are much better uses for your precious time here! —Μετάknowledge^{discuss/deeds} 13:54, 12 June 2018 (UTC)

Oh. Hrmph. Well, *I* found the indices useful! :-P   —Ruakh_TALK 05:17, 13 June 2018 (UTC)

@Ruakh: Just curious, what did you use them for? --WikiTiki89 19:40, 13 June 2018 (UTC)

Various things: finding translations tables that list non-lemma forms (e.g. להתקשר instead of התקשר), words that have entries but aren't listed in appropriate translations sections, etc. (So, mostly things that I find useful as a Wiktionary contributor, rather than things I'd find useful as a Wiktionary reader.) I was also thinking about adding a feature analogous to {{t+}}, where the index entries would have links to the Hebrew Wiktionary entry when it exists, which would (among other things) let me find entries that I could create relatively easily; but if no one else actually uses the indices, then I guess it's not useful to add new features to them? —Ruakh_TALK 19:57, 13 June 2018 (UTC)

@Ruakh, Wikitiki89, Metaknowledge: For the quality of translations from English into Hebrew, please check User:Matthias_Buchmeier/en-he-a (words starting with 'A', replace the final letter with any English letter). I believe it's updated regularly. --Anatoli T. ^{(обсудить}/^вклад) 00:15, 14 June 2018 (UTC)

Thanks for the link! That's really useful, though it looks like that bot isn't currently aware of the fact that (e.g.) {{t+|he|אָשֵׁר|tr=ashér}} links to [[אשר#Hebrew]] rather than [[אָשֵׁר]]. —Ruakh_TALK 00:57, 14 June 2018 (UTC)

@Ruakh: The method has a couple of other issues, yes, it doesn't seem to be aware how Module:links works. It works for Russian (knows to fix stress marks) but not for Arabic or Hebrew - diacritics are not dropped. --Anatoli T. ^{(обсудить}/^вклад) 01:06, 14 June 2018 (UTC)

None do.​—msh210℠ (talk) 20:24, 17 June 2018 (UTC)

Thanks! —Ruakh_TALK 01:34, 18 June 2018 (UTC)

Hebrew transliterations

Dear all,

Earlier this year I started a thread on Beer Parlour to call attention to the current disastrous state of Hebrew transliterations (see thread here). After that, some of us (mainly @Erutuon, Malku H₂n̥rés, Metaknowledge) have been working on a transliteration module that would allow to automatically transliterate Hebrew entries in both Biblical Hebrew (BH) and Modern Hebrew (MH), ensuring consistency throughout Wiktionary. The decision of having both BH and MH transliteration was made based on the discussion in the thread above, to satisfy the needs of users of both branches of Hebrew studies. We are still working on the final details, but it is virtually ready to be implemented. Incidentally, this module would also lay the basis for an eventual future creation of templates for the Pronunciation section that automatise the display of the different main Hebrew traditions.

You can find the module's testcases here: Hebrew transliteration module. The module uses the Sartma 1 transliteration system for BH, but it can easily be modified to represent any of the proposed systems or any eventual further changes we might want to implement.

The idea is to have the Biblical Hebrew transliteration first, followed by the Modern Hebrew Romanisation. Since the traditional Biblical Hebrew transliteration system has some ambiguity and an excess of diacritic signs (often based on etymology and not really relevant to BH per se, like the difference between long and short vowels), we worked on finding a solution that would solve those ambiguities while at the same time reducing the number of diacritics. You can find our proposals here: Hebrew transliteration. Please feel free to leave comments under the Public comments section under each proposal, or here below.

It took us quite a lot of time and effort to come up with our proposals and to build the module, so please take your time to go through the material and let us know what you think. It would be great if we could agree on a system to follow and implement it sooner than later. Sartma (talk) 10:14, 15 December 2021 (UTC)

@Ruakh, Msh210, Enoshd, Mnemosientje, Pinnerup, Yair rand, Taokailam, The cool numel, Wikitiki89, Profes.I., Lambiam and please feel free to ping any potentially interested editors whom I have inadvertently forgotten. —Μετάknowledge^{discuss/deeds} 19:13, 15 December 2021 (UTC)

In theory, I like the idea of having both biblical and modern transliterations, but I also think you'll have a helluva time getting people to agree on the system to use for biblical transliterations. I have to admit I don't like the Sartma1 system proposed. It's somewhat better than some, maybe better than most, but it still contains some of the things I hate most in Hebrew transliteration, like the underlined k. If this came to a vote, I'd likely vote against it just because of transliteration qualms — but it's worth noting that almost all of my votes are in opposition. All that said, I admire the effort and find myself hoping to someday vote in favor of some proposal with a goal like this one. (Split that infinitive like an American marriage, mmmm.) — [ זכריה קהת ] Zack. — 18:23, 15 December 2021 (UTC)

We're open to suggestions. As I said above, the module is quite flexible, we can easily modify it. I'm also not a big fan of the underlined letters, but I couldn't come up with something better. In the end, they are handy in maintaining the "visual unity" of a root despite the vocalic environment the individual letters might be in (k-t-b > kāṯaḇ, yiḵtōḇ). What would your preference be when it comes to begadkefat? Greek letters? β γ δ χ φ θ? Sartma (talk) 19:46, 15 December 2021 (UTC)

I'm not convinced that my making suggestions would ultimately be of any value. Especially if you're trying to avoid diacritics. I love most diacritics, except for the underline. — [ זכריה קהת ] Zack. — 22:12, 15 December 2021 (UTC)

Sure, no problem. Personal preference will always play a role in these things. Just to clarify, though, we don't have anything against diacritics per se. We just thought of getting rid of redundant ones (like those indicating vowel length, since the general agreement in the most recent literature on the topic is that vowel length was not phonemic) and keeping them only when deemed necessary. Sartma (talk) 22:27, 15 December 2021 (UTC)

Thanks for the ping. What an impressive piece of work you've done so far! I wholeheartedly support the effort to emend the current practice, but I do have some reservations as well:

1. Are we sure the userbase is inclined to accept that each entry will have more than one Romanization following it? I'm sure biblical scholars will love it, but to speakers/learners of Modern Hebrew (who, I assume, make up the majority of readers of these entries) it's going to be more of a hassle, and the scholarly Romanizations of BH may even seem positively outlandish to speakers of Modern Hebrew, looking nothing like their own pronunciation. I'm not entirely sure disunification (with separate entries for Biblical and Modern Hebrew) isn't preferable, analogous to the case of Icelandic and Old Norse (where the unavoidable redundancy seems to be accepted without problems).
2. I'm no great fan of the Sartma 1 system myself, I have to admit, and I'm not sure I see why Wiktionary should invent its own transliteration system instead of adopting one of the already existing ones. Personally I think I'd prefer going with the "traditional" system, that being the one used in most textbooks, articles etc. and the one that users of Wiktionary with an interest in BH would most often come by.
3. When you write that the idea is to have "the Biblical Hebrew transliteration first, followed by the Modern Hebrew Romanisation", by the latter term do you mean the system in the rightmost column on [1] (i.e. táḥaṯ for תַּ֫חַת) or do you mean the system currently used with most Hebrew entries (i.e. tákhat)? Perhaps it would be an idea to illustrate how an example entry would be formatted.

The above notwithstanding: it seems to me that you've done a great job furthering this matter, and I'll support whatever model in this regard is agreeable to the majority of users. Unlike the above contributor, I've no problems with the diacritic underlines in the בגדכפת letters. —Pinnerup (talk) 02:29, 20 December 2021 (UTC)

@Pinnerup Thanks for your comments! Here are my answers:

1. My impression is that most people who replied to the thread I quoted above were ok with the double romanization. My opinion is that as long as both romanizations are clearly marked as Biblical Hebrew and Modern Hebrew, a user would be able to ignore the one that's not pertinent to their search. Disunification was also discussed in the thread, but the majority seemed to be against it (I'm with you, though. If we had a vote, I'd vote to keep Biblical Hebrew and Modern Hebrew separated).
2. The problem with the existing BH textbook transliteration system (and its variations) is that they are imprecise/incomplete when it comes to certain details. They aim at a precise transliteration, but they don't deliver it (see Notes to our Transcription Table and Cons to the Traditional transcription). The traditional system has been criticised for quite a long time now, and there are even articles wondering why BH textbooks are sticking to it despite all its faults and its misrepresentation of BH. All recent scholar works (I'd say for at least the past 30 years, if not longer) have abandoned it and use the 7 vowel system instead. Sticking to the obsolete 5 vowel system would mean sticking to a past that has no followers in the contemporary BH academic world. So it's not entirely true that this is the system users interested in BH would most often come by (or at least, that would be restricted to beginner students). On the other hand, the Sartma 1 system is modern (inasmuch as it uses the 7 vowel system) and uses diacritics consistently (all diacritics only represents spelling, never sound. The traditional system is confusing in mixing diacritics with different functions). You can ignore vowel diacritics if you're not bothered by how a word is spelled (plene vs defective, etc), and you'd still have a correct transliteration of that word (for example Sartma 1 אֱלֹהִים (ʔɛ̆lohīm) is equal to אלהים (ʔɛlohim), אֲבִיּוֹנָה (ʔăḇiyyōnɔ̂) to אביונה (ʔaḇiyyonɔ), כּוֹל (kōl) and כֹּל (kol) to כול‚ כל (kol), etc.
3. All considerations about what transliteration system to use are about BH. MH will use the system currently in use on Wiktionary, so nothing will change there. I don't have a preference about the order, it just makes more sense to me to have a transliteration (BH) before a transcription (MH), but any order would do, really.

That said, if the vast majority of people votes for the traditional transliteration (despite its many issues and its being obsolete), that's what we'll go for. The main point of our work is to put an end to the current total lack of any consistency at all. A faulty, obsolete romanization would still be an improvement! Sartma (talk) 14:15, 20 December 2021 (UTC)

Wow! This is a great idea. Right now, I have more questions than suggestions. As a Modern Hebrew learner, I would prefer that

1. the system be simple,

2. homonyms should have the same transcription,

3. words that rhyme should have the same endings in their transcription.

But let me read and digest the stuff in the beer parlour first. This sounds exciting. Taokailam (talk) 15:46, 20 December 2021 (UTC)

2. homonyms homophones Taokailam (talk) 15:50, 20 December 2021 (UTC)

Hi! Not that many answers so far, but it’s already good to see what the Hebrew-speaking community thinks of the module. I hope there’ll be a vote soon, because people are always active during votes, and it’ll end with a direct decision.

• To @Qehath, the underline is restricted to BH. I understand you don’t like this diacritic, but what do you propound to replace it? More generally, we’ve been aware for a long time that it could be hard and long to get everyone to agree on a common romanization system, so tell us which do you like and why, what do you want to change, etc.?
• To @Pinnerup, yes the people interested in MH largely outnumber BH-interested people, that’s why I thought so far the MH romanization should go first in any case, but @Sartma seems not to agree on this. In a nutshell, every Hebrew entry is going to have an automatic MH romanization, but only some are going to have a BH romanization; thus IMO, MH romanization should be first, and the BH one be second because optional.
• To @Taokailam, assuming you’re only dealing with MH, what do you find complex or hard to read in the current MH romanization system? In my view it’s very simple, sleek and intuitive, there are no diacritics but the acute accent for stress, digraphs are used, few apostrophes are displayed for glottal stops. Technically homophones do have the same MH romanization (not in BH because spelling-based, but with MH transform they do due to numerous mergers). Can you find one or several counterexamples? And I assume 3. is just a corollary of 2. in the end of words. By the way, I don’t know which BP discussion you were referring to in the end of your message, but the August (or July I’m not sure) is surely outdated, we didn’t imagined at the moment the module as it is now. We who created the module (with Sartma, Erutuon and Metaknowledge) discussed on this big issue for a long time (weeks then months), maybe too much to sum up, so ask us your questions as soon as you have one because we’re very likely to already have a clear-cut answer to it.

Malku H₂n̥rés (talk) 21:15, 21 December 2021 (UTC)

Is the new transliteration module using the current romanization system for MH? If it is, I'm fine with that. I like it as it is. Taokailam (talk) 18:38, 23 December 2021 (UTC)

@Taokailam: yes it does, see Module:User:Sartma/he-translit/testcases/documentation then Ctrl+F "test_MH_tr" to to go the list of MH testcases. Malku H₂n̥rés (talk) 19:42, 23 December 2021 (UTC)

Thanks for the ping. What does the MH transliterator do with חָכְמָה? The reason I ask is that there are two different words spelled that way (_chochma_ "wisdom" and _chach'ma_ "she was smart"). I imagine there are other ambiguous words, too.​—msh210℠ (talk) 10:48, 29 December 2021 (UTC)

@Msh210: with this exact spelling, the MH function of the module returns "khokhmá". This is a question we expected from the Hebrew community and if I did my job well, there are no ambiguous words insofar as in each case there's a diacritic to put so as to switch from a romanization to another, making khakhemá an accessible romanization. Malku H₂n̥rés (talk) 14:17, 29 December 2021 (UTC)

@Msh210: yes, with a meteg on the khet it returns "khakhemá", which is what you wanted. Malku H₂n̥rés (talk) 15:05, 29 December 2021 (UTC)

Malku H₂n̥rés, but there is no meteg in the word. I mean, we can demand people include one so the transliterator works right, but that's worse (IMO) than not including an automatic transliteration at all in a case, like this, where the transliteration is ambiguous.​—msh210℠ (talk) 08:27, 3 January 2022 (UTC)

@Msh210: Headwords shouldn’t have this issue, since eventual ambiguities only arise in inflected forms. Sartma (talk) 13:08, 3 January 2022 (UTC)

גַּבֹהַּ is returned as gavóa by the MH function. The current tr in wiktionary is gavóah. The MH function works for נֹגַהּ returning the expected nogah. Let me know if this is the right place to post differences like these here. Thanks. Taokailam (talk) 19:25, 30 December 2021 (UTC)

@Malku H₂n̥rés Can you have a look at this when you have a minute? Also, גַּבֹהַּ is wrong, it should be גָּבֹהַּ. Also: Happy New Year! :D Sartma (talk) 16:25, 31 December 2021 (UTC)

@Malku H₂n̥rés Hi! Regarding which should be first, whether BH or MH, mine is just a preference, I'm actually fine with either. If people prefer to have MH before BH, let's go with that! Sartma (talk) 16:47, 31 December 2021 (UTC)

@Taokailam, Sartma: Happy new year! I was told that any final <-h> was deleted in MH, and so this rule is in the module. For nogah, either you saw the BH (not MH) testcase, where it's kept, or it was part of a little bug I fixed (final h was kept in a couple case with stressed vowels, anyway). So what are the rules to follow? Malku H₂n̥rés (talk) 14:04, 1 January 2022 (UTC)

@Taokailam it's fixed! The MH rule was in excess, so I simply removed it and the testcases passed. For details, the BH rules already kept final <h> provided that they had a dagesh/mapiq, which are now kept in MH romanization. Malku H₂n̥rés (talk) 20:50, 1 January 2022 (UTC)

So, this may be a stupid question, but . . . who's the intended audience of the BH/MT transliterations? Are there people who will be comfortable reading and using something like ‹ʔănɔšīm› (up to and including recognizing that the macron on the 'i' simply indicates the presence of an em kri'a yud, not actually vowel length or whatnot), but who aren't as comfortable reading and using אֲנָשִׁים? I ask because I'm wondering if the motivation for the BH/MT transliterations is really something like "It's misleading to give only MH transliterations" rather than "Some readers will find BH/MT transliterations useful". And if so, then there may be other ways to address that, e.g. by explicitly indicating that the MH transliterations are MH but not actually providing separate BH/MT transliterations. —Ruakh_TALK 08:25, 17 January 2022 (UTC)

@Ruakh: The question per se is not stupid at all. The reason is the same as why we transliterate Ancient Greek or Sanskrit here on Wiktionary, despite both languages being written with alphabets, so easily readable to whoever knows them. Plus, there's a long tradition of transliterating Biblical Hebrew, and whoever is interested in BH will definitely benefit from its transliteration, especially in the beginning of its learning process, but not only. If someone ends up opening a Hebrew page via a link in some other word's etymology, but doesn't know Biblical Hebrew, they'd still be able to get an idea of what sounds/signs the Hebrew alphabet represents.

The question of the familiarity with the transliteration system is a different topic, and whatever transliteration system we end up using, it will have to be explained on WT:AHE, like any other transliteration system of any other language here, so that's not really a problem.

I don't think it's "misleading to give only MH trasncriptions"; I don't think anybody here thinks that. But since we're keeping the two languages together, it's just fair to cover both audiences. I have no interest (at the moment, at least) in MH, so the MH transcription isn't of any help to me, for example. I'm used to find transliterations of the proposed kind in books, essays and dictionaries I use as references, and I don't see any reason why they shouldn't be given on Wiktionary too.

Anyway, from the preliminary discussion we had here, the majority of people agreed that it's a good idea to have both transcriptions, that's also why we proceeded this way. Sartma (talk) 10:09, 17 January 2022 (UTC)

Thanks for your reply!

The comparison to Ancient Greek and Sanskrit doesn't really work for me, because we only provide one transliteration for those languages, not two. Above, I wrote "reading and using אֲנָשִׁים", and you took me at my word (sorry); but really I should have written "reading and using אֲנָשִׁים (anashím)", because the question isn't "Should we have transliterations for BH/MT, or not?", it's "Should we have two transliterations for Hebrew, one that's mainly based on MT spelling+nikud and one that's mainly based on MH pronunciation?"

Re: "If someone ends up opening a Hebrew page via a link in some other word's etymology": I think that such a person will actually benefit more from transliterations like we currently have, which are optimized for readability by English speakers, than from the transliterations you're proposing, which are optimized for . . . I don't know who. (Hence my question about the intended audience.) Note that English dictionaries generally do not give super-precise transliterations of languages like Greek, but rather, they give transliterations intended for English-speakers, typically with the same level of complexity as our current Hebrew transliterations.

Re: "I have no interest (at the moment, at least) in MH, so the MH transcription isn't of any help to me, for example": Sorry, I don't get it. Can you explain a bit about why the current transliterations are of no help to you?

—Ruakh_TALK 19:41, 17 January 2022 (UTC)

The best way I could put it is that we won't be giving "two transliterations" (only the BH one is a transliteration); we will be giving the transliteration of the traditional BH word plus a transcription/romanization of the MH reading of that word. I gave you Classical Greek and Sanskrit as examples of transliterations (one-to-one character transpositions). No-one really knows how Classical Greek or Sanskrit (or Biblical Hebrew) were actually pronounced (we can only make educated guesses), so the only intellectually honest thing to do is to transliterate them, to give a sign-by-sign transposition. With modern living languages things are different, we can give a transcription/romanization, because we actually know how they are pronounced. I repeat myself, but as I said above, as long as we keep BH and MH together, we need to represent both languages.

To answer your question about why the current MH transcriptions/romanizations are of no help to me, that's because MH doesn't distinguish many BH letters or doesn't pronounce them. My need is to know whether a <t> is /t/, /ṯ/ or /ṭ/, an <'> is /ʔ/ or /ʕ/, a <kh> is /ḥ/ or /ḵ/, an <s> is a /s/ or /ś/ etc., and I don't find any of these things in the MH romanization.

I'm also not sure I understand why you think that BH transliterations are not "optimized" for English speakers, since our proposals are based on transliterations systems developed by English people and currently used in books and papers about Biblical Hebrew. In my proposal I did add a couple of tweaks to fix some issues that are present in the traditional transliteration system, but at this stage we're not really excluding any of the proposed transliteration systems, so should people prefer the traditional system (despite its faults) we can easily modify the module accordingly. We're not proposing anything "strange", "unheard of" or "revolutionary". It's all well withing the customary BH studies in English, and BH vocabularies aimed at English speakers do generally give a transliteration of the kind we propose.

Sartma (talk) 21:51, 17 January 2022 (UTC)

Re: "The best way I could put it is that we won't be giving 'two transliterations' (only the BH one is a transliteration)": Well, you'd be wrong about that — both are representations of Hebrew words in a non-Hebrew script, so are transliterations by definition — but fortunately I'm pretty sure that this also isn't the best way you could put it, so I won't take this statement as completely vitiating your argument. ;-)

Re: "as long as we keep BH and MH together, we need to represent both languages": I think this gets to the heart of the issue. Your perspective is that BH and MH are two separate languages, and that BH is getting short shrift by being bundled into a "Hebrew" section that naturally gives priority to the current living language. I do understand that perspective (though I don't really share it), but if that's the issue, then maybe a better approach would be to copy Biblical Hebrew into its own language section?

Re: "My need is to know whether a <t> is /t/, /ṯ/ or /ṭ/, an <'> is /ʔ/ or /ʕ/, a <kh> is /ḥ/ or /ḵ/, an <s> is a /s/ or /ś/ etc., and I don't find any of these things in the MH romanization": Sure, but those things aren't subtle in the Hebrew text. When you have the Hebrew text and the MH transliteration next to each other, what is the MH transliteration not providing that makes it useless to you?

Re: "I'm also not sure I understand why you think that BH transliterations are not 'optimized' for English speakers, since our proposals are based on transliterations systems developed by English people and currently used in books and papers about Biblical Hebrew": I didn't say that they're not optimized for English speakers; rather, I suggested that they're not optimized for readability. They're complicated, because they're optimized for capturing almost every last nuance of the MT spelling+nikud. And, in particular, they're not optimized for the use-case you describe above: "If someone ends up opening a Hebrew page via a link in some other word's etymology, but doesn't know Biblical Hebrew, they'd still be able to get an idea of what sounds/signs the Hebrew alphabet represents."

—Ruakh_TALK 01:22, 18 January 2022 (UTC)

@Ruakh: I'm using the term "transliteration" and "transcription" in their proper, technical sense. So no, I'm not wrong about that, I'm actually quite right about that (but I agree that it's possibly not the best way to put it, since it requires technical knowledge that can't be assumed).

It's not really "my perspective" that BH and MH are two separate languages: from a linguistic point of view they are. The rest is politics, and I'm not really interested in politics, so I'll limit myself to repeat that splitting the two languages doesn't seem to be an option for people working with Hebrew here (see thread above). If you think it's important, you can start a discussion on Beer Parlour and I can promise already you'll have my vote.

The linguistics doesn't say what you think it says. Try making a good faith effort to read the available literature on the subject, and try to understand why people are making the arguments they're making. Really get into the justification for people calling it separate or the same, and examine whether it makes sense in light of the arguments against. You'll at least end up not being so cocksure that they're separate. פֿינצטערניש (Fintsternish), she/her (talk) 15:16, 7 February 2022 (UTC)

I really don't need to do any of those things. I know enough about linguistics to make my own informed decision. But thanks for your comment. Sartma (talk) 15:23, 7 February 2022 (UTC)

Then refrain from treating falsehoods as though they are incontrovertible linguistic fact. פֿינצטערניש (Fintsternish), she/her (talk) 16:37, 7 February 2022 (UTC)

Well, if we talk about serious linguistics, I'd say they are incontrovertible linguistic facts. But hey, I am for freedom of belief, so we don't need to agree. Sartma (talk) 16:56, 7 February 2022 (UTC)

You've demonstrated yourself to be utterly incurious about serious linguistics matters, so you are not the judge of what is serious linguistics, nor of anything, for that matter. You lack the intellectual curiosity and humility required to be a scholar of anything at all, or even to have informed opinions on anything scholarly. You have a series of statements that lack any basis which you hold to be incontrovertible, but you don't know anything of the arguments that those statements are based on, nor of the arguments that contrary statements are based on. I only am talking with you about this because it's harmful to the project, which is meant to be undertaken in a rigorous, systematic way, not be a collection of things that Joe Somebody thought were true and didn't need to be justified. פֿינצטערניש (Fintsternish), she/her (talk) 18:47, 7 February 2022 (UTC)

Re: "When you have the Hebrew text and the MH transliteration next to each other, what is the MH transliteration not providing": It's not providing a transliteration of the Hebrew word. It's only providing a Modern Hebrew transcription. We have transliterations for pretty much all languages written with non-Latin letters, why should Hebrew be an exception? See for example any Thai word (just picking up a random one: กรรมาชีพ (gam-maa-chîip). Click on the link and see the transliteration next to "Orthographic", then compare it with the transcription in the "Romanization" line). The argument "the original script gives you everything already" is true for many languages, but that doesn't prevent us from giving a transliteration or romanization or both when needed. On the other hand, I could reply with your same logic and ask you: what is the BH transcription not providing? To be honest, if we had to choose only one transcription, the BH one would be best, because from that you can pretty much always derive the MH pronunciation (while the contrary is not true). If we want to have only one romanization, that should be the BH one, not the MH one.

Re: "I suggested that they're not optimized for readability.": I actually have no idea what you mean by this. The concept itself of "optimized for readability" when it comes to transliterations is quite awkward per se, but generally "readability" when it comes to human beings (and excluding font size) pretty much translates to "familiarity". As long as a human being is "used to read something" they'll read it without issues. Just think of the many crazy writing systems around the world. People read any sort of gibberish once they are used to it (English is a great example of this, too!). As I said before: there will be an explanation of the transliteration/transcription systems adopted under About Hebrew, so this isn't a real problem.

To conclude (also because I have the feeling we're going in circles now): there is nothing strange with giving a transliteration and a transcription, we do it for other languages already. Adding a transliteration doesn't take anything away from MH transcription, and I don't see why anybody would be against the inclusivity of pertinent information. Let's be inclusive and represent all sides, there's no need to discriminate.

Sartma (talk) 09:30, 18 January 2022 (UTC)

Re: "their proper, technical sense": I notice that that article's only relevant reference is one that flatly contradicts it, saying e.g. that "In Hebrew, the Jewish winter holiday is חנוכה. Its English transliteration is Hanukkah or Chanukah" [emphasis mine]. Do you have independent reason to believe that the Wikipedia article is nonetheless correct?

Re: "It's not really 'my perspective' that BH and MH are two separate languages: from a linguistic point of view they are": There's no universally-accepted "linguistic point of view" for whether two language varieties are one language or two. There are clear-cut cases, and there are more or less messy cases. Hebrew is a messy case; it's obviously changed a lot since Biblical Hebrew (especially the Hebrew of the Torah; some later books are less different from Modern Hebrew), and some linguists are so impressed by the influence of European languages that they regard Modern Hebrew as a European language (!), while others forthrightly regard it as a continuation of the same language (see e.g. Glinert's The Grammar of Modern Hebrew, page 2). (My personal sense is that many of the former camp overestimate the influence of European languages and underestimate the continuous development of Hebrew over the past several thousand years; but if you prefer to view this as just "politics", that's fine. It's clear that many of the world's languages are lumped or split for political reasons, and in the absence of any objective alternative, there's little point dismissing that.)

Re: What the MH transliteration isn't providing: OK, thanks. So, yes, I think I've correctly identified your perspective, and that seems to have made you angry, which I'm sorry about, but I'm not sure what else you want from me? You seem to want me to pretend that your perspective isn't a "perspective" but just objective fact, but if so then I refuse. I think you're a mature adult who can handle respectful disagreement, and I intend to treat you accordingly.

Re: "what is the BH transcription not providing?": (1) It's hard to read (more on this below). (2) It doesn't indicate stress placement. (3) It doesn't distinguish kamats katan from kamats gadol. (4) It represents /s/, /ts/, and /ʃ/ with the same letter, which isn't generally a problem for me as long as the Hebrew script is right there, but is a problem for casual readers trying to get a sense of the pronunciation.

Re: readability: You've interpreted me correctly — readability relates to familiarity. Our current transliteration scheme is familiar to most English-speakers. (It doesn't match actual English spelling, but it matches how foreign names are typically transliterated to English.)

Re: "Adding a transliteration doesn't take anything away from MH transcription, and I don't see why anybody would be against the inclusivity of pertinent information": This is true. The concerns that are motivating my question about the intended audience are (1) that I find the specific presentation proposal confusing, especially since a large proportion of words will only have the current-style transliteration, and (2) that I think the specific presentation proposal does take something away from the MH transliteration, by shunting it off to the side with a special label. (Unaccustomed readers will have to struggle through seeming gibberish before getting to a simpler transliteration that they can take a decent stab at.) Hence my desire to understand the goals better, so I can see what other possibilities might address them. Unfortunately, I'm starting to wonder if part of your goal here really is to demote the MH transliteration (reassurance on this front would be welcome), and if so then I suppose there's no way to accommodate that part of your goal without doing so. :-/

—Ruakh_TALK 03:38, 19 January 2022 (UTC)

Re: "Do you have independent reasons to believe that the Wikipedia article is nonetheless correct?": The Wikipedia article agrees with the canonical meaning of those terms in linguistics, so why should I believe they're not correct? I know that a lot of people use those terms wrongly (like the author of the passage you quoted from Vocabulary.com), but that doesn't change their meaning.

• Transliteration: shows sign-to-sign correspondence. Doesn't represent sounds.
• Transcription: represents sounds.
• Romanization: umbrella term, includes both of the above.

The table under Difference from transcription in the [Transliteration] article is quite clear (I kept only one Greek example and added a Hebrew one):

Word	Transliteration	Transcription
Ελληνική Δημοκρατία	Ellēnikḗ Dēmokratía	Elinikí Dhimokratía
אֲנָשִׁים	ʔănɔšīm	anashím

You can find the same definitions here:

• Romanization#Transliteration
• Transcription (linguistics)
• Crystal (2008)'s Dictionary of Linguistics and Phonetics

Re: "(1) It's hard to read (more on this below).": This is quite subjective, not really a fact. People familiar with BH are used to find things like אֲנָשִׁים (ʾănā/šî́m), so I don't think that argument stands.

Re: "(2) It doesn't indicate stress placement": That's not entirely true. No-accent is equal to final-accent. In all other cases, the accent is indicated. But if this is a problem, we can add the accent at any time. The module is flexible enough. Its sort of standard in books and papers about BH that give transliterations to only indicate the accent when non final.

Re: "3) It doesn't distinguish kamats katan from kamats gadol.": It doesn't because the script doesn't, and since it's a transliteration, it doesn't need to. Scholar works on BH use the same 7 vowel system and don't distinguish kamats katan from kamats gadol. So nothing new nor strange there either. The MH transcription will show that difference.

The reason kamats gadol and kamats katan are important is because they reflect the vowel systems of Biblical and Tiberian Hebrew, and distinguishing them is crucial to reconstructing the vowel system of Biblical Hebrew. Kamats gadol was a long A in Biblical Hebrew; kamats katan was a short O in Biblical Hebrew. Both merged in Tiberian Hebrew, Ashkenazi Hebrew, and Yemenite Hebrew, but were kept separate in Sephardi Hebrew, merging with long A and short O. Again, this would be more meaningful to you if you were the least bit curious about Hebrew. פֿינצטערניש (Fintsternish), she/her (talk) 19:46, 7 February 2022 (UTC)

Qā́meṣ gādōl and qā́meṣ qāṭān don't reflect the vowel system of Tiberian Hebrew. What you wrote is plainly wrong. You don't seem to be curious enough about Hebrew yourself, I see. As for the rest of what you wrote, I knew already, so what? I'm not sure what is it that you're trying to argue here (apart from your assumptions about the degree of my being curious about Hebrew, which for the records, are just laughable). Sartma (talk) 20:31, 7 February 2022 (UTC)

Sorry, I think you've lost track of the conversation here. You were asking why I need the MH transliteration — why the BH/MT transliteration isn't good enough for that. I gave reasons. You can't then swipe away those reasons by saying "The MH transcription will show that difference", because the premise of your question was that there wouldn't be an MH transliteration. (I'll reply to some of your other comments later — thank you — but some of them I plan to not reply to, because I think there are some threads here that have gone off the rails and, I think you'll agree, are no longer constructive.) —Ruakh_TALK 21:05, 19 January 2022 (UTC)

You're right. I got confused there. Too many topics going on in parallel.

I'm not proposing to have the BH transliteration on its own anyway, so we can drop this thread. I only wanted to point out to you that while it's possible to go from a transliteration to its transcription, the other way round is not possible. The same way you can read the Hebrew script and know when a qamats is qatan, you would also be able to do when reading the transliteration. Sartma (talk) 22:38, 19 January 2022 (UTC)

Re: "4) It represents /s/, /ts/, and /ʃ/ with the same letter, which (...) is a problem for casual readers trying to get a sense of the pronunciation.": That's not true, they are represented with different letters. S, ś, š and ṣ are different letters, and are familiar to people interested in BH. Moreover, it's not the job of a transliteration to give a pronunciation, that's the job of a transcription. A casual reader will have the MH transcription to look at, or even better the Pronunciation section, to get an idea of how the word sounds.

Re: "(1) that I find the specific presentation proposal confusing": We are here to discuss about it. We're all open to change the format, so you're more than welcome to propose something you consider less confusing.

Re: "(2) that I think the specific presentation proposal does take something away from the MH transliteration, by shunting it off to the side with a special label.": As I said below, I was actually thinking I was giving MH more visibility by using the label. If the problem is the label, we can label the BH transcription instead (if you thought that the label was shunting off the MH transcription, taking something away from MH, you should be happy to have BH in its place):

1. אלוהים / אֱלֹהִים • (elohím - Biblical: ʾĕlohīm) ...

I'm completely fine with whatever layout, as long as we have both a BH transliteration and a MH transcription. Sartma (talk) 20:26, 19 January 2022 (UTC)

Re: accent: Sorry, my bad — I didn't notice that the accent was indicated when non-final. That seems to be "cheating" given the stated goal/nature of transliteration, but I won't complain. (It's consistent with the common use of meteg for non-final accents in prayerbooks.)

Re: labeling "Biblical" instead: I do prefer to put the BH/MT transliteration second, yes, not just because I think the living language should have priority but also so that we start with the simpler/less-complicated one before we scare off the uninitiated. :-)   But I'm not sure about the specific label "Biblical"; is the idea that we'll include this transliteration only for forms that literally appear in the Masoretic Text? (So, like, if a form is first attested in the Mishna, would it only get the current-style transliteration?)

—Ruakh_TALK 05:55, 21 January 2022 (UTC)

I wouldn't oppose providing "Biblical" romanisation for Mishnaic Hebrew, but it seems somewhat unnecessary. I don't know of a Mishnaic corpus, either; one advantage of sticking to truly Biblical words is that we have a straightforward corpus and can (mostly) automate the process of adding a parameter that shows a Biblical romanisation to those lemmas which are found in the corpus. —Μετάknowledge^{discuss/deeds} 07:23, 21 January 2022 (UTC)

I was thinking of giving the transliteration for all historical terms that have a traditional Tiberian vocalisation, independently of where they come from, but I'm ok with limiting it to Biblical words. Sartma (talk) 11:22, 22 January 2022 (UTC)

Oh, and one more reason: transliterations will also help the reader since they have a much higher level of readability compared to the original Hebrew script, which depending on the font and font-size used can sometimes be not very friendly to the eye. Sartma (talk) 11:07, 17 January 2022 (UTC)

I know what you mean, but on the other hand, any 100%-precise transliteration of MT spelling+nikud will also sometimes be unfriendly to the eye. (Your current proposal, which seems very well thought out, involves two different diacritics written under letters, and five different diacritics written above letters, some of which are visually quite similar. And even with all that, it still has a few ambiguities and a few mismatches from MT spelling+nikud.) I wonder if it would be worthwhile to build an easy way for readers to get a zoomed-in/high-res view of a specific word; if we had something like that, would you still find it useful to have a separate transliteration of MT spelling+nikud? —Ruakh_TALK 19:41, 17 January 2022 (UTC)

It's unfriendly to the eye inasmuch as the reader doesn't know the transliteration rules (as any other transliteration system), that's why we generally give the transliteration rules in the "About (language)" page. The traditional transliteration system for BH already has all the diacritics you mention, so people studying BH are used to them. If anything, the main merit of my proposal is that it reduces the number of diacritics while being at the same time more precise... I'm not sure what ambiguities and mismatches you mean, unless you're talking about that couple of extremely rare cases of irregular spellings that we gave in the Hebrew transliteration page that would need to be adjusted manually and that would anyway never occur as headwords. The proposed transliteration system has no ambiguities nor mismatches in headwords.

A zoom function wouldn't hurt, I guess, but I wouldn't consider it a substitution for a good transliteration.

Sartma (talk) 22:09, 17 January 2022 (UTC)

Re: "I'm not sure what ambiguities and mismatches you mean": Exactly the ones that you gave at your proposal page. (Oh, and possibly cases like w:Qere and Ketiv, but your proposal doesn't mention those so I don't know for sure how you'd handle them.) I'm not sure why you say that יִשָּׂשכָר "would never occur as [a] headword[]". And even if that were true, I don't see how it's relevant? My point was just that, despite a great deal of effort to create what you define as a "transliteration" scheme (that is: to perfectly capture MT spelling+nikud using English letters plus various diacritics), you still ended up having to make some compromises in weird cases (cases where, I assume, the Masoretes couldn't quite match their pronunciation up with the Biblical spelling). To be clear, I don't intend this as criticism; sometimes compromises are necessary, I give you a lot of credit for calling them out explicitly, and I certainly don't recommend letting weird cases like יִשָּׂשכָר completely drive your transliteration scheme. I'm just saying that for all your s**tting on the current Hebrew transliterations, you've also had to make a few compromises, and also didn't develop a "transliteration" scheme by your own definition.

Re: "A zoom function wouldn't hurt, I guess, but I wouldn't consider it a substitution for a good transliteration": OK, fair enough.

—Ruakh_TALK 01:22, 18 January 2022 (UTC)

@Ruakh: Briefly: Qere and Ketiv will have to be adjusted manually (like all exceptions, they can't be automatised) and will possibly need a Usage note too. יִשָּׂשכָר is covered automatically by the module, so it's not an issue. Compromising is necessary when it comes to exceptions to the rules. They are exceptions, though, so it's ok to treat them as such.

I don't understand why you say that I'm "shitting on the current Hebrew transliterations". Pointing out that they are completely inconsistent and messy is just an observation, I'm not "shitting" on them. The whole effort (months of work) to create a module that can automatically generate BH transliterations and MH transcriptions goes towards fixing the problem and improving Wiktionary giving Hebrew entries the consistency they deserve. Sartma (talk) 09:44, 18 January 2022 (UTC)

Sorry, I think we're talking at cross-purposes here. You seem to be talking about the automation itself, whereas I'm talking about the proposal that the automation implements. You mention that qere and ketiv have to be adjusted manually, which is fine, but that doesn't tell me what the manually-adjusted transliteration will be. (Reading between the lines, I'm guessing that it will match the qere rather than the ketiv?) You mention that יִשָּׂשכָר is covered automatically, which is also fine, but doesn't change the fact that it's an exception to your "transliteration" scheme, which is all I was saying. (And as I've said, I'm not criticizing these inconsistencies. I regret having mentioned them; it seems to have annoyed you, which wasn't my intention, and I certainly didn't intend to spawn an entire comment-thread about them!) —Ruakh_TALK 03:38, 19 January 2022 (UTC)

I was talking about the automation as an answer to your accusation of "shitting on the current Hebrew transliterations". When I observe that they are a mess, I'm not talking about the quere/ketiv question or how other exceptions are being treated. When you say "you've also had to make a few compromises, and also didn't develop a "transliteration" scheme by your own definition." you're completely missing the point. The current state of Hebrew transliterations is a mess not because of the compromises Hebrew editors might have made, but because there are a lot of inconsistencies. I copy-paste the relevant part of my first post, so you can refresh what my complaints were about:

----

As I'm sure many of you did already, I've noticed the lack of standardisation when it comes to Hebrew transliterations. Put simply: It's a mess. All other Semitic languages seem to deal with it in a much neater way. I've generally noticed the following scenarios:

1. Simplified/Modern Hebrew transcription only: אֵל, בַּיִת (see derived terms, too), קַרְקַע, אָחוֹת, etc.
2. Accurate transliteration only:ראש (see transliteration of derived terms), *ḳarḳar-, etc.
3. Both of the above in random order: עָפָר, שיבולת, *halak-, *śamš- (note how the order in which they appear is also irregular), etc.
4. Mistakes of any sort: עֹשֶׁר (ע neither transliterated nor transcribed), *ʕaśar- (ע transliterated as /ʔ/ instead of /'/), פַּרְעוֹשׁ (missing accent), etc.

----

This has nothing to do with "compromising". It's just messy. This is the situation that our module and proposal is trying to put an end to.

And to be very clear here, it's not your mentioning the inconsistencies that annoyed me (how could it? I'm the one who acknowledged them in the first place!). It's your being offensive for no reasons that annoyed me. Telling me that I'm "shitting on the current Hebrew transliterations" when you feel I shouldn't because my transliteration system "also" needs to compromise when it comes to exceptions, as if the problem with the current state of Hebrew transliterations was just a question of compromising on exceptions...! If that was the issue, you wouldn't have heard of me at all!

Anyway, to answer your question about qere/ktiv: I'd give the qere plus the details in a Usage note. Sartma (talk) 23:42, 19 January 2022 (UTC)

@Ruakh: Your discussion with Sartma seems to have devolved before actually answering your question: why would two romanisations be a good thing? The short answer is that not everyone can read Hebrew script, and we want to be able to represent the older phonology to those who cannot. There is, logically speaking, one other way we could achieve this: every Hebrew lemma could have IPA for both (reconstructed) Biblical and Standard Israeli Hebrew. This, however, is not a solution we could extend beyond the entries themselves, whereas it would be nice to be able to eventually use this in etymology sections in other languages' entries, for example. (There's also the fact that scholars almost exclusively use romanisations similar to the one proposed here, rather than IPA.) But you might well ask, what is this older phonology and why do we want to represent it? I note that you said above that the BH romanisation "represents /s/, /ts/, and /ʃ/ with the same letter", which indicates to me that you have not considered the fact that the sounds themselves have not been constant over time! The BH romanisation in headword lines will always be accompanied by MH, so <s>, <ts>, and <sh> will still appear as you expect. But MH /s/ merges two BH phonemes, so we can also represent them separately in the BH romanisation. The distinctions lost in MH are important in the context of Hebrew as an ancient Semitic language, which is the only window through which someone like Sartma is interested in Hebrew, but also for something as simple as explaining why a spelling like Sukkoth with gemination and exists when we only give sukót as the Hebrew form. We can easily provide this without requiring that readers learn the Hebrew script. —Μετάknowledge^{discuss/deeds} 18:14, 19 January 2022 (UTC)

Thanks for your reply!

I think you misread some of the above thread, and so missed some context. (I don't blame you — it's a very long thread!) My comment about /s/, /ts/, and /ʃ/ was in response to Sartma's question about why I find the MH transliteration useful — why the BH transliteration wouldn't be sufficient — and was not intended as a criticism of the transliteration scheme as a BH/MT transliteration scheme. But the current proposal isn't that this be a BH/MT transliteration scheme; it's that this be the primary transliteration scheme, with the MH transliteration scheme being shunted off to the side and marked as specifically "Modern". (It is Modern, of course. But the implication is that the BH/MT transliteration is the main/real/basic transliteration — literally unmarked; and it seems clear from Sartma's comments above that that is indeed his/her intention.)

Your comment about Sukkoth (or more commonly Succoth) is an interesting one, and I'll have to give it some thought. In particular, it raises the question of how we'd want to handle something like Rebecca (or Ῥεβέκκα (Rhebékka)), where the English, Latin, and Ancient Greek spellings reflect a different belief about the BH pronunciation than what we see in the Masoretic Text.

—Ruakh_TALK 19:07, 19 January 2022 (UTC)

@Ruakh: You're right, the thread got to be too long and tendentious, and my eyes started to glaze over — apologies for misinterpreting you. Now you bring up the issue of "shunting", which is a stylistic, rather than substantive, concern. This is very easily solved: I can say with complete certainty that neither I nor Sartma cares which order the romanisations are in, or which is presented as primary in headwords. There is a solid argument in favour of putting Modern first: unlike Biblical, it will appear in every entry, and it is the pronunciation used for a large majority of spoken Hebrew today. But if you are on board with the substantive issues of presenting both romanisations, then everything else is simply a matter of communicating preferences. —Μετάknowledge^{discuss/deeds} 20:01, 20 January 2022 (UTC)

Re: "if you are on board with the substantive issues of presenting both romanisations": Yes, I think I am now. It took me some time to process what you were saying about Sartma's area of interest, but now that I grok it, it makes sense. I'd been picturing either people who are familiar enough with Biblical Hebrew that they wouldn't need a transliteration scheme, or people whose interest is so casual that they'd be fine with (indeed, better off with) the present one. I hadn't thought about people whose interest is really in Semitic languages in general, and how the diversity of Semitic alphabets would make transliterations useful for them. Thanks for calling that out! —Ruakh_TALK 02:23, 21 January 2022 (UTC)

New layout for Hebrew transliterations

Hi Hebrew people! Following up on the above discussion, I'd like to propose a new layout for Hebrew transliterations. Here is a mockup of what the transliteration part (first part; what follows remains unchanged) of Hebrew headword-line templates would look like:

1. אלוהים / אֱלֹהִים • (elohím - Biblical: ʾĕlohīm) ...

(updated as per @Lambiam’s and @Ruakh's comments. Sartma (talk) 19:19, 24 January 2022 (UTC)

After having considered various layouts, I came to think that this is the simplest one, and the one that makes more sense. What we've been calling BH is a transliteration in the strict sense of the word, it shows a 1-to-1 correspondence of Hebrew and Latin letters, so it makes more sense to give it straight away without any label. It's neutral, in a way, it just "spells" the dotted entry in Roman letters. After that the modern "reading" of the word. What do you think? If this looks ok, we will proceed with changing the headword-line templates accordingly. Sartma (talk) 15:32, 15 January 2022 (UTC)

I like it! Taokailam (talk) 18:54, 15 January 2022 (UTC)

Another MH test case: אָמָּנוּת returns amanút instead of omanút. Could anyone look into this? Thanks. Taokailam (talk) 19:20, 15 January 2022 (UTC)

@Ruakh, Msh210, Enoshd, Mnemosientje, Pinnerup, Yair rand, Taokailam, The cool numel, Wikitiki89, Profes.I., Lambiam, Erutuon, Metaknowledge, Malku H₂n̥rés and please feel free to ping any potentially interested editors whom I have inadvertently forgotten. —Sartma

Is there a reason for using “\” rather than “/”? The latter is common as a separator between alternatives, as seen in ו/או and in job ads, like “דרוש/ה איש/אשת מכירות ושיווק שטח”. The only use for “\” mentioned in the Hebrew Wikipedia is as a character used in the field of computing.  --Lambiam 16:24, 15 January 2022 (UTC)

@Lambiam: I’m glad you asked! I just kept it because it’s the status quō, but I really dislike it. More than happy to change it! Sartma (talk) 16:29, 15 January 2022 (UTC)

Yeah, sorry, I think that's my bad: I started using \ here many years ago, somehow having had the impression that Hebrew writing slanted the slash the opposite direction from English writing. I've been meaning to suggest that we should fix this. —Ruakh_TALK 10:42, 16 January 2022 (UTC)

If I may offer a further suggestion: a modicum of whitespace surrounding the slash, as in אלוהים / אֱלֹהִים, will improve readability, particularly for users not accustomed to reading Hebrew script.  --Lambiam 12:38, 16 January 2022 (UTC)

Definitely. We currently put spaces around the backslash, and should continue to do so after switching to the forward-slash. —Ruakh_TALK 08:25, 17 January 2022 (UTC)

Fixed all the \ in modules and templates that have "he" in the title. There might be some remaining in entries and I can't guarantee it won't cause problems if there is. — Eru·tuon 05:23, 4 September 2022 (UTC)

I'm not sure I agree that the scholarly/BH translit should have precedence merely because is a stricter and more pure translatiteration: Hebrew is primarily a living language in the form of MH, and I believe MH transliterations are simply more relevant to the user than BH. While this debatable quibble is in the back of my mind I'm not particularly bothered by the proposed order either and I like the compromise of just having both styles, and having 'modern' spelled out (maybe with all-lowercase 'modern' instead of 'Modern'?) is nice and clear. So yeah, looks good from here, no serious objections.

One thing: editors should have the freedom to include only the modern translit if desired in the proposed new headword-line template (and perhaps the word 'modern' could then be omitted in such cases). In cases such as קומוניזם (komunízm) or טלוויזיה (televízya) it would seem to be more confusing than useful to the reader to have a BH transliteration, and it would mean more work for the editor with dubious added value (especially since editors interested primarily in MH will not be acquainted with the BH translit style, as it is not relevant to them). — Mnemosientje (t · c) 13:30, 16 January 2022 (UTC)

@Mnemosientje Hi! Thanks for your comment! I can't agree with you that Hebrew is "primarily" MH, but I do understand where you're coming from and I guess that until we keep the two languages together (I would be in favour of splitting them), every one will have their own opinion on this regard. If we want to keep BH and MH together (as it seems to be the wish of the majority here), I'm afraid we'll have to compromise somewhere. I tried my best to come up with something that keeps things simple while giving all the relevant information for "fans" of both languages. The main reason why I didn't put "Biblical" in front of the BH transliteration, for example, was to try and give it less prominence in favour of the MH transcription. As for the capitalisation, this is what it would look like with lower case "m":

• אלוהים / אֱלֹהִים • (ʾĕlohīm - modern: elohím) ...

I feel that the capital letter would give more visibility to the MH pronunciation, but these things are all quite subjective. I'm fine with either, really.

Regarding the freedom of including only modern transliteration, that is indeed the plan. Words that don't exist in BH will only appear in MH with MH transcription (and in that case "Modern:" will be dropped). Sartma (talk) 14:46, 16 January 2022 (UTC)

After transcription presentations being tweaked like monthly, I don’t know what the current state is, but I warn @Starbeam2 that certainly he cannot apply “Wikipedia” standards here and that is annoying. This is not Wikipedia, this is a dictionary. Fay Freak (talk) 16:58, 25 January 2022 (UTC)

No worries i can stop using Wikipedia's method, but be aware i'm not a he but a they. Starbeam2 (talk) 17:07, 25 January 2022 (UTC)

Late to the discussion. Here are my thoughts (@Starbeam2):

• We should avoid the word "Biblical", because it is not an accurate description of what the transliteration represents. The transliteration represents a quasi-phonemic quasi-graphemic (based on the vocalization) transcription of the Hebrew term. It neither reflects Biblical pronunciation, nor is it limited in use to Biblical Hebrew. Another incorrect term for it is "Tiberian", as it does not actually reflect Tiberian pronunciation, but rather a sort of underlying representation behind the Tiberian vocalization. If we use any term to describe it, it should be something vague like "scholarly [transcription]".
• The term "Modern" or "modern" is appropriate for "elohím".
• I'm not sure which transliteration should be given priority, but I lean towards the modern one just because of the status-quo.
• If we can't avoid having a label inside the layout (as opposed to, for example, just separating the two transliterations with slashes), then perhaps the best option would be to use a small-cap abbreviation that links to a page describing the scheme? Such as this: (m: elohím / sch: ʾĕlōhīm).
• Minor note: It should be "ʾĕlōhīm" with a macron on the "o" as well.
• I do not support over-Tiberianizing the scholarly pronunciation. There is no benefit to this other than the easier automatic transliteration. It would only confuse people and make the pronunciation incompatible with non-Tiberian schemes. This means, I advocate for ā for qamatz gadol. There should be no distinction between כֹּל and כּוֹל, both should be kōl, because they represent the same pronunciation.

--WikiTiki89 20:04, 1 April 2022 (UTC)

Your small-cap example looks excellent. Fay Freak (talk) 20:19, 1 April 2022 (UTC)

@Wikitiki89: Hi! Thanks for joining the discussion!

• I agree that Biblical is not technically precise, but it's somehow handy to use as a simple counterpart to Modern. I'm happy to label it with any other thing that makes sense if people think Biblical might be confusing.
• I think most people agreed that the modern pronunciation should go first, and that the *Biblical transliteration would be given only for words actually appearing in the Bible (those with a Strong reference number, I guess)
• Re' the layout: I'm happy with whatever makes more sense to people. I like the small-cap abbreviations too (to me the layout is not a very important question, so I'm happy with whatever, as long as it's reasonable)
• About the actual transliteration, we still haven't decided which one to go for, but both me and @Erutuon are not happy with the status quo and proposed new transliteration systems. Please have a look at the beginning of this discussion for all the details (and why /ʾĕlohīm/ wouldn't necessarily need a macron on the /o/. In the traditional "scholarly" transcription, it would actually be /ʾĕlōhî́m/, with a circumflex over the /i/ and the acute accent).
• I read your concern of over-Tiberianising the "pronunciation", but what we are talking about here is a transliteration, not a pronunciation. The "scholarly pronunciation", as you call it, is not really a pronunciation either, it's just a transliteration heavily influenced by historical reconstruction (and quite dated, too). It's not meant to be a pronunciation guide anyway. Most modern scholars have abandoned that system and are already using other signs for kamatz gadol (mainly: ɔ and å). It's not really a question of automatic transliteration either (since it can handle either systems without any problem), but of giving a more honest transliteration that actually represents the script. The dotting is Tiberian, so the transliteration will reflect that. This doesn't mean that we are "Tiberianising" it. We're just giving a one-to-one correspondence, as a good, neutral, correct and honest transliteration should do.

It's very important that we are all aware that we are talking about "transliterating" here, and not "transcribing" (which would imply reproducing the sound of a word). No-one knows how Classical Hebrew sounded (and no-one ever will), and many issues are still debated/non agreed upon/just theories. That's why it wouldn't be honest to try and give a "pronunciation". We can only transliterate, and for that it's important to give the eventual difference between כֹּל (kōl) and כּוֹל (kôl). I used the traditional scholarly transliteration system there to show you that this is also what's been done up to these days. I'm not introducing anything new there. Scholars have been distinguishing those plene/defective spellings in tranliterations since forever. People interested in Biblical Hebrew are used to see that kind of difference, I can't see any logical reason to take that away. Sartma (talk) 09:23, 3 April 2022 (UTC)

@Sartma: Sorry for taking so long to get back to you.

• I agree that I should have said "scholarly transliteration" rather than "scholarly pronunciation".
• I disagree that it is dated (though some aspects of it are dated, more on that later). Rather it is abstract, and later researchers have focused more on the pronunciation of particular dialects in particular time periods, and therefore began using transliteration schemes appropriate for those dialects and time periods.
• But I believe that our transliteration should be abstract. This is because we dont't want to, as I put it before, "over-Tiberianize" our transliterations. Using ɔ or å for kamatz gadol is fine when dealing with Tiberian pronunciation, or Babylonian pronunciation, or contemporary Ashkenazi or Yemenite pronunciation. But we don't want to focus on any particular pronunciation. We want to cover Hebrew as a whole. This means using a transliteration scheme that is inclusive of the historical Palestinian vocalization as well as the centuries of Sephardic-based reading traditions.
• Keep in mind that we are transliterating, using the previous example, the word אלהים. The vowel points (אֱלֹהִים) are included as an aid, but they are not really part of the spelling of the word. And in fact, there were (and in a certain sense are today as well) alternative systems for marking the vowels in the word. A strict "transliteration" of this word would be ʾLHYM, but that is not really very helpful to our users. This is why routinely in many of the languages here on the English Wiktionary, we use transliteration schemes that give our users information that is not contained in the written word itself. For this reason, in Hebrew too, we do not simply transliterate קל as QL, but we include the vowels as well, qal or kal (depending on the scheme). But this does not mean that we need to stick to trascribing grapheme-for-grapheme one particular vocalization system.
• This is related to my parenthetical note above about the dated aspects of the "scholarly" transliteration system. There is particular reason that I specifically put a macron on both the o and the i of ʾĕlōhīm. This is because the use of macrons or circumflexes to differentiate between the presence and absence of a mater lectionis is both useless and flawed (flawed, because it doesn't actually do such a great job at making systematic differentiations), and misunderstands the whole purpose of the vocalization system. The vocalization system is an overlay on the written word that tells you how it is to be read, and thus also identifies the word from its homographs. The example I used before was כל and כול. These are in fact the same word, just written differently. If we go by the graphemic transltieration, we get KL and KWL, but once we go beyond that, we recognize that these two spellings represent the same exact thing, and we don't have to differentiate them when we write the vowels (which are in fact the same exact vowel). Therefore, kōl (or kol) are two ways of transliterating the same word, regardless of whether it is spelled כל or כול. If you care about the consonantal spelling but cannot read the alphabet, refer to KL or KWL. If you care about the vocalization, refer to kōl. If we choose not to show the consonantal transliteration KL / KWL because it's not very helpful to our users, that doesn't mean this information belongs in the vocalized transliteration. If you want grapheme-for-grapheme Hebrew-with-Tiberian-vocalization, then I'd suggest a completely different scheme, one where you write the consonantal spelling as ʾLHYM, and then annotate each consonant with some vowel symbol with a one-to-one correspondence to the Tiberian vocalization. But I don't really think that would be helpful to our users.
• Thus, I believe that if you look at our transliteration scheme as an abstraction and not a real pronunciation, and you look at the breadth of Hebrew pronunciation traditions, then I believe the scheme I advocate for, that for example distinguishes min ("from") from mīn ("species"), regardless of whether the latter is spelled מין or מן, is the scheme that makes the most sense for our "Biblical" / "scholarly" / whatever-we-want-to-call-it transliteration.

I'm willing to expand more on anything I said, or give sources or examples for any claims or hidden assumptions I made. --WikiTiki89 20:46, 19 August 2022 (UTC)

@Wikitiki89: Thanks for your reply!

My understanding (correct me if I'm wrong) is that you don't want a transliteration, but you would like a sort of ad hoc transcription. You would like a romanization system that only transliterates consonants, while for vowels it not only transliterates them but also shows their allophonic length (including the difference between qā́meṣ gāḏōl and qā́meṣ qāṭān). So for example you would keep the difference between ס (s) and ‎שׂ (ś), despite in Tiberian Hebrew they were both pronounced the same (/s/), but romanize vowels long or short depending on their position in a word (despite this being completely predictable). I've never seen this kind of romanization anywhere, it would be a completely new thing. No Biblical Hebrew textbook does that either, so it would definitely be confusing to a Biblical Hebrew student.

By the way, how would you deal with the lengthening of short vowels in tonic closed syllables? Like, how would you romanize שָׁמַר? If you romanise qā́meṣ gāḏōl as /ā/, how would you romanise a long páṯaḥ?

Don't get me wrong, I'm not against it, but I'd like to see a full detailed description of the system you're envisioning (something like our Hebrew transliteration proposals), mainly. because the devil is in the details, and your comments are not clear enough to allow a full understanding of what the system you're proposing looks like in practice. — Sartma ^{【𒁾𒁉 ● 𒊭 𒌑𒊑𒀉𒁲】} 02:27, 20 August 2022 (UTC)

Remove Template:he-given name

I have updated the links that use the template he-given name. I don't know how to remove this template or whether I have the user right to do so. Could someone remove it or point me to the relevant instructions? Thanks in advance. Taokailam (talk) 17:01, 25 May 2022 (UTC)

After reading WT:DELETE, I have added rfd to the template T:he-given name and an entry in WT:RFDO. Taokailam (talk) 00:07, 27 May 2022 (UTC)

Transliteration of ה and הּ in Hebrew roots

I was looking at this page ג־ב־ה. It contains two different roots, one triliteral proper (with pronounced final /h/ as third consonant: ג־ב־הּ) and one weak (with orthographical silent /h/: ג־ב־ה). Both their transliterations are given as "g-b-h". Shall we find a way to distinguish the two? — Sartma ^{【𒁾𒁉 ● 𒊭 𒌑𒊑𒀉𒁲】} 13:38, 23 August 2022 (UTC)

I think it is probably fine for them to remain on the same page, but pretty much all inbound links should be deeplinks in order to be clear about which root is meant.

SaryaniPaschtorr (talk) 01:33, 29 May 2023 (UTC)

Conjugation of definitions of verbs

In Hebrew dictionaries, the third-person masculine singular past tense of a verb is considered the lemma. On Wiktionary, the third-person masculine singular past tense of a verb is also the lemma. In a similar vein, as do Latin dictionaries, Wiktionary considers the first-person singular present indicative of a Latin verb to be the lemma. However, for these two languages, Wiktionary diverges in its methodology regarding the conjugation of the verbs in the definitions. Whereas Latin verbs are supplied with senses beginning with the first-person singular (e.g. "I go", "I read", "I see"), Hebrew verbs are defined in the infinitive form, just like English. My issue with the current methodology for Hebrew verbs is that it gives the impression that the third-person masculine singular past tense of the verb is the infinitive form, rather than has. Words like בָּא are defined as "to come", when it really means "he came". My suggestion is to rewrite the definitions to use the he–past tense structure and instruct editors on this policy to follow the structure, as we would the I–present indicative structure for Latin verbs. FreeMediaKid! 17:57, 27 September 2022 (UTC)

I couldn't agree more. I also find it really strange to translate the 3ms past with the infinitive. Also, not all Hebrew dictionaries do this either.

The thing is that inter-language consistency is apparently not a thing here on Wiktionary, so each language can choose what to do, independently of what other languages do. Ancient Greek, for example, uses the 1s present form but translates it with the infinitive (see for instance πονέω). Sanskrit uses the 3s present (see for instance गच्छति), and Arabic too gives the 3ms past as lemma form, just like Hebrew, but translates it with the infinitive... — Sartma ^{【𒁾𒁉 ● 𒊭 𒌑𒊑𒀉𒁲】} 20:53, 27 September 2022 (UTC)

The definitions that translate the ending (Latin verbs, Classical Nahuatl NNCs etc.) are obsolete in this respect. Citation forms are translated with citation forms. For less markedness and hence less noise, exactly so you don’t have too many impressions – the opposite practice of glossing is pointy, irrelevant for the specific purpose of a dictionary and more to type for editors. Fay Freak (talk) 21:08, 27 September 2022 (UTC)

FWIW I agree with User:Fay Freak; we should define the citation form with the English citation form (the infinitive) regardless of what particular form is used to cite a verb in a given language. Benwing2 (talk) 19:39, 2 June 2023 (UTC)

Conjugation templates

I am reviving a topic which has been discussed partially before, and only quite some time ago.

I think it would be nice for conjugation templates to give a more complete account of all parts of a verb. For instance, it would be handy to include the cohortative, the jussive, the passive participle etc., so long as those forms of a particular verb are attested in the ancient/mediaeval literature or (mostly for participles used adjectivally) current in the modern language. SaryaniPaschtorr (talk) 01:24, 29 May 2023 (UTC)

I would support that. That's what we do for Romance languages. Andrew Sheedy (talk) 14:29, 30 May 2023 (UTC)

@SaryaniPaschtorr I am in favor although I think the issue is more that you need to find someone to do the work. The difference between what you're proposing and the Romance languages is that in the latter, we just include all forms that are part of modern language usage and don't have to worry about attestation. You might look into how Old Irish verbs are handled (see beirid and dobeir for examples); for them, only the attested forms are listed because Old Irish verb morphology is so incredibly complex and irregular, and there aren't so many primary sources. Benwing2 (talk) 19:34, 2 June 2023 (UTC)

I don't think attestation will be a problem. The ancient literature using the jussive etc. frequently is fairly large and there are heaps of really good lexicons for both Biblical and Mishnaic Hebrew which give a verb's every attested part and where it turns up (plus some reconstructed forms — by the way I'm not actually in favour, but I don't think it would offend Wiktionary's principles to include any conjectured forms appearing in several important reference works).

Yes, the tricky bit is getting someone to do the work. It will have to be someone with more experience using these templates than me really.

SaryaniPaschtorr (talk) 20:44, 2 June 2023 (UTC)

@SaryaniPaschtorr If those forms are not actually productive in Modern Hebrew, I would not want to include them in the conjugation tables. I'd much more prefer to have "Neo-Hebrew" and "Classical Hebrew" (Biblical, Mishnaic, Medieval, and all the "only written" forms of pre-Neo-Hebrew) as two separate languages, and only give cohortative, jussive, passive participle, waw-consecutive, the two infinitives, etc., in Classical Hebrew entries. Keep in mind that even common forms (present, past, future in NH, imperfect, perfect, present participle in CH) actually have different meanings and grammar, depending on whether we're talking about NH or CH. It's quite absurd that we're treating the two languages as one. — Sartma ^{【𒁾𒁉 ● 𒊭 𒌑𒊑𒀉𒁲】} 14:50, 16 June 2023 (UTC)

Full Spelling vs. Defective Spelling in Translation Tables?

Hello all,

Just wondering what the policy is regarding full vs. defective spellings (i.e. כתיב מלא/כתיב חסר ניקוד/ללא ניקוד vs. כתיב מנוקד), specifically within translation tables. There is no guidance on this on this page, and I’ve noticed that usage of full vs. defective spelling varies considerably among Hebrew translations here on Wiktionary. My personal policy has been to always use the full spellings in Hebrew translations, unless Biblical Hebrew is specifically listed under the translation, since the standard orthography for Modern Hebrew uses full spellings, but I think that a standard policy should be spelled out in this document for future guidance and also to provide a reference for correcting/rewriting older translations that were written defectively/with nikudim to the full spellings, in order to provide justification in comments on edits for other editors.

Thanks!

Hermes Thrice Great (talk) 15:11, 10 July 2023 (UTC)

PS, along these same lines, in translation tables, the lemma form of verbs (i.e. 3ms past form) should always be used, correct? I see a lot of participle/present forms in the translation tables as well.

Hermes Thrice Great (talk) 12:24, 7 December 2023 (UTC)