Terra-Rywko

Strange pronunciations

You changed the pronuns of אתם to be /ʔatʰˈtʰimm/ for Biblical hebrew and /ʔatʰˈtʰɛm/|[ʔatˈtʰɛː.ɛm] for Tiberian Hebrew. These both look very strange to me. Where is the justification for the doubled /m/ in Biblical Hebrew, the vowel hiatus in Tiberian Hebrew and the aspirated t? I have never encountered any of these before. User:Fay Freak any idea what's going on here? Benwing2 (talk) 20:55, 31 May 2024 (UTC)

@Benwing2: The BCE pronunciation of Hebrew is much more conjectural than the Tiberian one a millennium later, for which vocalization marks were invented after all. Joshua Blau, who transcribes segol, a sign occurring in the Tiberian niqqud without equivalent in the Palestinian niqqud, œ, difference from ɛ being the rounding, on pp. 112–113 of his 2010 Phonology and Morphology of Biblical Hebrew describes it as a “cancellation of the opposition a : i”, so there we are again in the front open-mid of the vowel chart rather than the top of the vowel chart. Both Proto-Semitic فَعْل (faʕl) and فِعْل (fiʕl) suffered segolatization, so it seems quite unreasonable to me to believe segol a close vowel. Another unlikely step then it is to state Yemenite Hebrew having developed [æ] from [i]. Fay Freak (talk) 22:40, 31 May 2024 (UTC)

OK. Do you know of any other Hebrew editors who can chime in? This user Terra-Rywko added > 100 pronunciations; if they are all suspect we may need to revert them all, pending commentary from the user in question. Benwing2 (talk) 22:46, 31 May 2024 (UTC)

@Benwing2: Not sure about anyone having more than a hobbyist idea. Perhaps @Hermes Thrice Great by now? There are occasional editors around that probably have enough concern with fathoming more recent pronunciations by doing Hebrew between a palette of other languages in the diaspora. Looking at אֹזֶן, it seems like the present editor even projects forward Proto-West-Semitic pronunciations to Biblical Hebrew, you didn’t pick out the strangest one. Why did @פֿינצטערניש add the like thing on עֶצֶם though (in 2019)? The Septuagint has segolates as expected by the Masoretic vocalization already. It seems like a miscommunication where they fancy Hebrew somewhere long before the historical books and closer to the Amarna letters :/. Fay Freak (talk) 23:23, 31 May 2024 (UTC)

I do agree that there is a general tendency to make Biblical Hebrew out as much older than it was, I think I'll just stop adding/changing these entries since I am not as certain myself. I do personally think the evidence for glottalization/ejectives in Biblical Hebrew is lacking. A bigger problem IMO is that Biblical Hebrew is a pretty vague label covering everything from the 14th to the 1st centuries BC. Terra-Rywko (talk) 23:29, 31 May 2024 (UTC)

Oh I forgot to justify /ʔuzn/ lol, the vowel appearing in the Septuagint where probably still allophonic at the time is the theory I remember reading, but I'm have to reread Suchard 2019 to make sure I'm not misremembering. the split of short /i u/ [i~e u~o] into /i u/ and /ɛ ɔ/ is a post-Tiberian thing tho and definitely wasn't the case in Biblical Hebrew. Terra-Rywko (talk) 23:32, 31 May 2024 (UTC)

@Fay Freak so I went back to check if the history of Segholization is commented on and found that Kantor's freely available dissertation The Second Column (Secunda) of Origen's Hexapla: in Light of Greek Pronunciation (Kantor 2017) does actually comment on it in section 6.5.2.1. Segholates (*qVtl) pg. 354. It shows how different transcriptions from around that time will show the same word with or without epenthesis, and even when two transcriptions both show epenthesis the epenthetic vowel is different, Biblical /giml/ is attested as as ⟨γιμελ⟩ in Eusebius (4th c. AD), ⟨γιμηλ⟩ Epiphanius (4th c. AD), ⟨γιμλ⟩ Epiphanius (5th c. AD), ⟨γιμαλ⟩ Septuagint Codex Oriental 5000 (6th/7th c. AD). I note from elsewhere that the Septuagint also likes to do echo vowels like ⟨μολοχ⟩ for /mulk/ which later Jewish traditions do not do. This evidence reflects dialectal variation, optional final cluster breaking like many current Arabic and Aramaic varieties, or both.

There is variation on whether to notate these short [i~ɪ~e u~ʊ~o] vowels as /i u/ or /e o/ as you should be familiar with from Arabic dialects.

Regardless of all that, /ʔuzn/ does not reflect Proto-West-Semitic pronunciation because that would be /ʔuðn/ /lh. Terra-Rywko (talk) 20:51, 4 June 2024 (UTC)

Yemenite Hebrew, at least in Biblical readings, always turns Tiberian seghol (/ɛ/) into /a/ regardless if it's from historical *i or *a, pre-Tiberianization of all Jewish Hebrew traditions things were different, but the Babylonian distribution of vowels is no longer used (at least in readings of the Bible). Terra-Rywko (talk) 23:19, 31 May 2024 (UTC)

Oh hey, I don't think I'm any expert and I'm especially uncertain about the Biblical Hebrew entries added, I don't mind anyone issuing corrections but I'm basing most of it on my understanding of Tiberian (Khan 2020 and Suchard 2018) and Samaritan Hebrew (Ben-Hayyim & Tal 2000 and Florentin 2005) and how other traditional pronunciations mapped their phonemes on Tiberian, which could easily lead to occassional mistakes.

For /ʔatʰˈtʰimm/ I just based the doubled M in Samaritan /attimma/ which might have arisen independently instead. Terra-Rywko (talk) 23:14, 31 May 2024 (UTC)

OK. Please remove all the questionable Biblical Hebrew entries; don't expect others to do that for you, and don't do original research in reconstructions that can't be directly justified by an established source. Also where is the justification for [ʔatˈtʰɛː.ɛm]? That is the strangest of all. Benwing2 (talk) 23:50, 31 May 2024 (UTC)

This is the way Khan 2020 analyzes the way long vowels in closed syllables (/CVːC/ or /ˈCVC/) are longer and should actually be analyzed as [CVː.VC]. This also applies to furtative páṯaħ/pattāħ where the name Noah for example is analyzed as /noːħ/ [ˈnoː.aħ] (tho optionally these become [ˈnoː.waħ~ˈnow.waħ]). This transcription is also used in www.tiberianhebrew.com. Terra-Rywko (talk) 04:43, 3 June 2024 (UTC)

OK, but that site is not independent of Khan's book. I am skeptical that this is correct given the multitude of competing interpretations of Hebrew phonology and general phonological principles. (I am not familiar with Khan's book but I have seen plenty of garbage out there written about Hebrew phonology and other linguistic topics.) Wikipedia's entry on Hebrew phonology doesn't mention it for example, and it lends undue weight to Khan's interpretation to present it as *the* correct one. I would advise removing all existing phonetic interpretations of Tiberian Hebrew and sticking to phonemic ones, which are more agnostic vis-a-vis the competing interpretations. Benwing2 (talk) 05:18, 3 June 2024 (UTC)

Wikipedia has long struggled to be up to date on Semitic linguistics ngl. It still promotes (relatively) ancient theories like "South Semitic" and things like that. Khan's analysis of Hebrew is also used by big names in the field like Benjamin Suchard, Benjamin Kantor, and A.Z. Foreman, in fact the phonemic analysis I'm using is based on Suchard 2019 which used at the time unpublished information from Khan 2020 to derive a less underlying analysis than Khan's which is harder to interpret if you know less about Hebrew morphology, his phonetic transcription seems to ignore the [Vː.V] vowels of Khan but that could be because of incomplete information or just a less narrow transcription since [CVː.VC] and [CVːC] are not contrastive. Kantor who made the website earlier also published "The Linguistic Classification of the Reading Traditions of Biblical Hebrew" in 2023 and used Khan's vowels. Though to be fair these are linguists that frequently collaborate with each other and shouldn't be said to be indicative of the whole field, I am not aware of any recent scholarship specifically tackling the analysis of Tiberian Hebrew phonology, but a blog from Suchard did complain "Most scholars seem to think there just was no length. If the Masoretes didn’t invent a sign for it, they must have had a good reason." but it's hard to tell if he's talking about simply the historical consensus, or if it's an indication of the current one too.

As a point of comparison (from Khan 2020) one could look at how Arabic reading traditions for the Qur'ān also increase the length of a long vowel before a coda, and how both Classical Arabic and Tiberian Hebrew which both developed in the 9th-12th centuries avoiding CVVC and CVCC syllables in most cases.

I'm currently fixing up the Biblical transcriptions I made as you've asked of me, differing to your experience on Wiktionary, so tell me what you think. Terra-Rywko (talk) 06:14, 3 June 2024 (UTC)