Talk:مروارید

Z, do your sources comment on the reconstruction *margārīt for Parthian? How do they justify it? Parthian *margārīt would explain Armenian մարգարիտ (margarit) perfectly, which is attested too early to be borrowed from Greek or Latin, as is usually assumed. --Vahag (talk) 16:41, 13 July 2014 (UTC)

Yes, the word has been considered a Greek loanword in Iranian, but more recent studies suggest it's the other way around. My source was this article:

• غلی اشرف صادقی، «‫فاوا، مُروارید ـ مَروارید‬، گان ـ گاد‬»، فرهنگ‌نویسی، شمارهٔ ۷، ۱۳۹۲، صص ۱۰۲–۱۱۰

For Parthian, it is only attested in the late, Manichaean form morγārīd (transliterated as murgārīδ in older works). Based on an evidence in a New Persian manuscript by an author from Pars (Persis), the article concludes that New Persian of that region (and not just that of Transoxiana) also probably had marwārīd, beside murwārīd, and therefore the rarely attested Middle Persian form mlwʾlyt should be read as marwārīd, as opposed to mwlwʾlyt' (morwārīd). So Parthian may also have both *marγārīd and murγārīd. It also compares Middle Persian mwlw' (murw) and Parthian mrwg (murγ) with Pashto مرغه (/⁠marğa⁠/), Ossetian маргъ (marǧ), Old Armenian սիրամարգ (siramarg), լորամարգ (loramarg), and says the sound change a > u was regular in Iranian after labial consonants in initial position, and g > w change in Middle Persian is also regular near r, and d < δ < t is also a known sound change in that position in Middle Iranian, thus the early Parthian form would be *margārīt. --Z 22:21, 13 July 2014 (UTC)

Thanks for elaborating. Based on this I have added a little original research to մարգարիտ (margarit). I don't believe Ačaṙyan's proposal about Latin mediation, which he had to assume not knowing about the Iranian origin of the word. We have directly borrowed only about 9 words from Latin, which are terms reflecting Roman realia. There is nothing particularly Roman about the pearls. --Vahag (talk) 16:51, 14 July 2014 (UTC)