More on Persian kinship terms; "daughter" and the laryngeals
« previous post | next post »
Following up on "Turandot and the deep Indo-European roots of 'daughter'" (3/16/20), John Mullan (student of Arabic, master calligrapher, and expert chorister) writes:
As someone who’s studied a bit of Persian and a few other Indo European languages, I’ve always found it odd that most all of the kinship terms in Persian—mādar, pedar, barādar, dokhtar, pesar (cf. ‘puer’ in Latin and ‘pais’ in Greek, I assume)—have easy equivalents to my ear, /except/ ‘khāhar,’ sister. Wiktionary suggests it’s still related.
One quite recent finding of mine in PIE. As you probably know, 'Baghdad' is not an Arabic name, but a Persian one. It's composed of 'Bagh,' God (not the word used today), and 'Dād,' Given/Gift. Now I'm familiar with Bagh, ultimately, from listening to way too much Russian choral music and hearing Church Slavonic 'Bozhe.' Similarly, in the deep corners of my Greek student mind I remember names like 'Mithradates'—gift of Mithra or something along those lines—popping up as rulers/governors of city states in Classical Anatolia. What I /didn't/ pick out was the exact same construct as 'Baghdad' hiding in front of my eyes all along. There are two active NBA players named 'Bogdan(ović).' It's the same name as the city, only it's popped up in Serbo-Croatian. Funny stuff.
John Colarusso weighs in on the laryngeal-2 as it pertains to "daughter":
"Daughter' is interesting because it gives strong evidence for the nature of the laryngeal-2.
Greek has an /-a-/ for *h-2, and no change on the preceding /-g-/, while Sanskrit has an /-i-/ and voiced aspiration of the preceding /-g-/.
The only sound that can split into back low as well as front high vowels is a pharyngeal.
Pharyngeals have strong low first formants (acoustic resonances) mimicking high vowels (hence pharyngeal or emphatical softening as Trubetzskoy called it; note Phoenician Ba9al (<9> for voiced pharyngeal) 'lord', vs. Hebrew Be?el).
But pharyngeal articulation involves contraction and lowering of the tongue root, hence it can give low vowels.
So, Greek went with the acoustics of PIE *h.2, while Sanskrit went with the articulation along with some friction reinterpreted as belonging to the preceding voiced sound.
The kinship suffix may have been *-h.2t-er, akin to the full grade *h.2et-el- (-el- a diminutive of endearment) as in OE athel-, Mod E Ethel, Mod G Edel 'noble'.
Some of our most intimate and familiar words hold within them the most profound linguistic secrets.
Philip Taylor said,
March 18, 2020 @ 1:51 pm
"Church Slavonic 'Bozhe'" — also Polish Bóg, Boże.
Twill said,
March 18, 2020 @ 2:40 pm
Re: "Gift of God" as a name, see here.
Chris Button said,
March 18, 2020 @ 2:43 pm
I've always liked the Starostin/Pulleyblank prosodic analysis of the Old Chinese division of syllables into two types (conventionally type A and type B) since it is typologically reasonable and has convincing associations with tense/lax alternations elsewhere in the region noted by Pulleyblank/Ferlus.
Nonetheless, I have always found Norman's pharyngealization hypothesis interesting as an alternative (albeit typologically pretty unlikely) alternative. However, I don't believe the notion of "emphatic softening" has been discussed in that regard. It could throw an interesting wrinkle on it.
Somewhat tangentially, but sticking with the guttural theme and its coloring effects, the association of uvulars with labialization is something I feel hasn't always received the scholarly attention it deserves.
Victor Mair said,
March 18, 2020 @ 2:50 pm
From Geoff Wade:
Might the surname of Cypriot tennis player Marcos Baghdatis (Μάρκος Παγδατής) perhaps also be tied to the Persian 'Bagh' God and 'Dād,' Given/Gift ?
Apparently he is of Lebanese descent and his full name is Marcos Omar Baghdatis (not confirmed) which would further suggest some Persian connections
John Swindle said,
March 18, 2020 @ 4:00 pm
In Sanskrit there are presumably-related Bhagavān and dāna. I don't know Sanskrit and don't know whether they combine.
Chris C. said,
March 18, 2020 @ 5:17 pm
CS Bozhe is vocative. Since one addresses God rather often in prayer, one hears it fairly often in Russian Orthodox choral music. The nominative is Bog'.
One might also note Sanskrit "Bhagavan", which came immediately to mind.
Michael Watts said,
March 18, 2020 @ 9:35 pm
Those are just localizations of Theodore, the Greek version of the name. But for these purposes, I wouldn't say the Hebrew version of Theodore is Theodore; I would say it's Nathaniel.
Apollodorus and Diodorus deserve mention too.
Twill said,
March 18, 2020 @ 10:53 pm
@Michael Watts Whoops, most the calques or equivalents are above the section I linked: Nathaniel, Adeodatus, Godiva, Matthew, Johnathan etc. etc.
Francesco Brighenti said,
March 19, 2020 @ 1:40 am
The exact Sanskrit cognate of Persian Baghdad and Slavic Bogdan is the personal name Bhagadatta ("he who is given by Bhaga," a god who bestows wealth).
Scott P. said,
March 19, 2020 @ 2:56 am
Interestingly, Latin gives us both Adeodatus and Deusdedit (both Papal names, too).
Andreas Johansson said,
March 19, 2020 @ 3:54 am
"Mithradates" (the form "Mithridates" is an order of magnitude more common online) is Mehrdad in Modern Persian, with the same form of the second element as in Baghdad.
Tom Dawkes said,
March 19, 2020 @ 9:08 am
Persian khāhar is actually written khwāhar, and this matches Latin sorority. Both can be derived from *swesor, with medial -s- becoming respectively Persian -h- (cf. Greek s>h) and Latin -r- (< *-z-, parallel in Old English, where freosan 'to freeze' alternates with gefroren 'frozen').
Vincent Daly said,
March 19, 2020 @ 9:13 am
@Victor Mair: Just speculation, but I would guess that Mr Baghdatis's Lebanese ancestors were called Baghdadi after some earlier Baghdad-born ancestor who settled in Lebanon. So no direct connection to Persian (if I'm right).
John Swindle said,
March 19, 2020 @ 5:43 pm
@Francesco Brighenti: Thanks!
R. Fenwick said,
March 19, 2020 @ 9:09 pm
@Chris Button:
Somewhat tangentially, but sticking with the guttural theme and its coloring effects, the association of uvulars with labialization is something I feel hasn't always received the scholarly attention it deserves.
Indeed. The use of the Semitic qoph for labialised stops in both early Greek and Etruscan (and thence Latin) comes immediately to mind, but also, while recently reading about the etymology of khan I came across Alexander Vovin's suggested etymology of the Xiongnu titles 單于 and 護于, which relies on reconstructing Han-dynasty Chinese *darɣʷa and *xʷaʔɣʷa, proposing these as attempts at phonetically rendering Yeniseian terms with uvular stops (he reconstructs Xiongnu *darqā "crown prince" and *qaʔqā "supreme ruler", and suggests a cognacy of the element *-qā with Ket qɨ̄ːj "ruler", Yugh kɨ̄j "id.", Arin kej "boss, power"). I don't know how much sense this makes from a historical Chinese phonology point of view, though. What are your thoughts on this hypothesis?
Victor Mair said,
March 20, 2020 @ 8:28 am
From Sanping Chen:
Regarding "daughter" and theophoric names, the Iranian _dukht_ is a rather unique case in Indo-Iranian or Indo-European onomastics, in the sense that "god-given/Dieudonné" is the dominating IE naming format. My favorite is Herodotus "Hera's gift." In contrast, "god's child" is very rarely seen, at least among mortals. However, "god's daughter" was (still is?) a rather popular Iranian format, typified by Adurdukht "daughter of Fire (god)." Rüdiger Schmitt, in a personal communication, has a fascinating interpretation for this exception. He suggests that it was an "emergency measure" due to the loss of the terminal vowel from Old to Middle Iranian. Hence the "daughter" was added to signify a feminine name.
I do not know whether the Iranian "god's daughter" names were the impetus to "god's child" becoming the dominating format when China imported theophoric names. Though "god's child" was frequently used in the Ancient Near East, it is hard to imagine Chinese borrowing it without an "Iranian interface." Nevertheless, "god's wife" used in the ANE yet seemingly not in ancient Iran became very popular in medieval China. Wonder if this format was adopted by other cultures.
Chris Button said,
March 20, 2020 @ 3:01 pm
@ R. Fenwick
That's a really interesting observation. I would reconstruct 于 as *ʁàɣ (giving Early Middle Chinese wuă) and 護 as *ʁáks (giving EMC ɣɔʰ).
So they both would have had the OC uvular onset *ʁ- to which I attribute the emergence of the rounding in the EMC reflexes (the EMC onsets are w- and ɣ- due to the syllable distinction of -à- versus -á-, which it doesn't seem Vovin is incorporating).
The articulation in Han times would have been somewhere between the OC and EMC forms. At least uvular ʁ- has no additional labial coarticulation and its place of articulation correlates with uvular q-, but we do still have the difficulty with the manner of articulation since q- is a voiceless stop.
Chau said,
March 20, 2020 @ 4:08 pm
@Victor Mair
I would like to say, I thoroughly enjoyed the scholarly book “Multicultural China in the Early Middle Ages 木蘭與麒麟” by the erudite historian Samping Chen. Each chapter of the book, varied in topics, is a fascinating, thoroughly researched, page-turning read. Chapter 6 on “Son of Heaven and Son of God” is particularly relevant to the discussion we are having here.
The Iranian bagapuhr /Sogdian βαγαpūr (written βγpwr) ‘son of god’ suddenly appeared in Chinese histories after the influx of the northern “Barbarians”, names such as mohefu 莫何弗 or mofu 莫弗 (e.g., Wei Shu 魏書 40.902) were the Chinese renditions of the Iranian word. Also, Chinese naming practices changed dramatically with the arrival of the Steppe peoples and the coincidental introduction of Buddhism, with theophoric personal names becoming extremely popular, spreading from the Steppe people to the Han Chinese. Names such as 神佑 ‘god protected’, 天佑 ‘heaven protected’, 天賜 ‘heaven’s gift’, 天送 ‘heaven delivered’, among others, have since been common place. For example, 詹天佑 was a pioneer railroad builder of China at the beginning of the 20th century.
The Chinese m- initial in mohefu 莫何弗 or mofu 莫弗 resulting from transcribing Iranian bagapuhr is consistent with the ancient Greek which rendered the Indo-Iranian baγa ‘god’ as Mαγα and Mεγα, as well as with other languages listed in Chen’s book, p. 128 (Footnote 60).
BTW, I think I can account for the change from Iranian bag- to Sinitic mo-: I have found a pattern of sound correspondence of foreign -ag changing to Taiwanese -an, supported by at least 60 examples. For example, Germanic dag ‘dawn, day’ > Tw tàn 旦 ‘dawn, day’. Thus, the first syllable bag- of Iranian baga- could first change to *ban-, which with *bVn- being equivalent to *mV- would turn into *ma-. And finally, the /a/ sound becomes /o/ in Sinitic today, which happens commonly to words transmitted from India and Central Asia along the Silk Road, e.g. Skt dharani > 陀羅尼 ‘voice of permanence, truth’. Thus, the derivation goes as follows: baga- >*bag- > *ban > *ma > mo.
R. Fenwick said,
March 20, 2020 @ 7:06 pm
@Chris Button:
At least uvular ʁ- has no additional labial coarticulation and its place of articulation correlates with uvular q-, but we do still have the difficulty with the manner of articulation since q- is a voiceless stop.
Vovin does also deal with that in his paper (I can send you the reference if you're interested in reading the whole thing). Heinrich Werner reconstructs Proto-Yeniseian with only an aspirated-unaspirated contrast, with voicing originally and in many cases still merely allophonic. In modern Ket, in particular, the uvular stop is conventionally rendered as /q/, but that's merely a phonological rendering and the phoneme also has the usual allophones [ɢ ~ ʁ] intervocalically and after /ŋ/. Also, glottalisation in Yeniseian and especially Ket seems to be in a weird liminal space between segmental and suprasegmental (a bit like Danish stød), so I just don't know whether a putative compound *qaʔqā would undergo that intervocalic voicing or not (I'm no Yeniseian expert, after all).
Chris Button said,
March 21, 2020 @ 7:06 am
@ R. Fenwick
Thanks–you can reach me at hotmail dot com with a dot between my first and last name (as used in my posts here)
I was wondering about intervocalic voicing, but how would an account be made at the start of the word?
Incidentally, I do have a copy of Vovin's "Did the Xiongnu Speak a Yeniseian Language? Part 2: Vocabulary" where he talks a bit about 護于.
He also discusses 孤塗 "son" and mentions, but ultimately rejects, Pulleyblank's proposed association of Ket qalek "younger son, grandson", for which Vovin seems to cite qal/qaloq "grandson". In that regard, all I can say is that 孤 would indeed have originally had a uvular *q- onset in Old Chinese as I reconstruct it and since there would unlikely be any syllables at this stage still attesting the earlier Old Chinese *-l coda, using 塗 (which presumably would still have had a lateral onset and not yet have become EMC *d-) could make sense.
Vovin sometimes comes on LLog to comment, so if we're lucky he might chime in…