Before getting into the main topic, I'll just link to some observations I made in the comments to the previous post -- about a Korean royal clan claiming descent from the Xiongnu. This establishes that my parallel investigation of the Steppe, and specifically Xiongnu, origins of their Japanese neighbors is already on solid ground in Southeastern Korea.
The clan that united the kingdom of Silla, which then went on to unify all of Korea, was the Kim clan from Gyeongju. Their legend of their origins is that they descend from a Xiongnu prince, who the Chinese call Jin Midi, but who they themselves call Kim Al-chi. I immediately noticed that "alchi" is one of the variant names of the Alat tribe -- the one whose name means "piebald horse" in Turkic, and who were the ruling clan of the Xiongnu confederation.
Did the unifying clan of Korea really descend from Xiongnu rulers? Well, they made this claim themselves, it's not somebody like me 2000 years later attributing it to them. And Korean scholars note how similar the grave goods are for Silla and the eastern Xiongnu.
I think the Japanese chose to make their Xiongnu origins cryptic due to the rivalry between Wa / Yamato and Silla. Yamato was on the losing side of the Tang-Goguryeo War of the 7th C, whereas Silla was on the winning side (allying with Tang China). They wouldn't try to invade Korea for many centuries after that, and wanted to distance themselves from their geopolitical rivals. Since the Xiongnu origins of the Kim clan were well known back then, the Yamato decided not to make the same claim, lest they be seen as copying Silla, or engaging in sibling rivalry.
But all the signs are there if you look.
* * *
Which brings us to the main topic of this post -- the burial rituals of the Xiongnu and other Steppe cultures of that time, as well as the early Wa culture in Japan. (And Silla, too, of course, but I'm not focusing on that.)
To reiterate an ongoing theme, we have to try to avoid using examples of convergent evolution when linking two groups together. What causes two groups to converge on the same outcome is some kind of utilitarian, economic, materially motivated force -- like getting more calories in your diet.
Group A drinks milk, and group B drinks milk -- are they descended from a common ancestor? Maybe, but maybe not -- maybe they each independently took up pastoralism, and began consuming the dairy products of their livestock. Both of them drinking milk doesn't mean they share an ancestor, it may mean that pastoralists will end up consuming dairy products, whether they share an ancestor or not.
Likewise, quite a few pastoralist groups from the Steppe practice horse sacrifice and horse burials, including Indo-European cultures from the West and Altaic cultures from the East. But that doesn't mean they share a cultural ancestor -- it may just reflect that fact that both have adopted horses, which makes horses very important, and so, what greater sacrifice could they make than sacrificing a horse?
We want to look for examples that are not steered by a cold, clinical Darwinian, economic, utilitarian incentive -- things that are more like a shibboleth. I say "po-TAY-to", you say "po-TAH-to" -- and that proves we belong to two separate cultures, whatever else we might share. Pronouncing the word either way does not help you communicate the meaning more efficiently, it is simply a random inconsequential arbitrary coin-flip that we have constructed in order to distinguish the members of group A vs. group B.
When I get to clothing and jewelry styles, we'll really see this idea take off -- what does it matter if you close your robe left-over-right or right-over-left? The robe closes just the same. But in ancient East Asia, this seemingly meaningless distinction made all the difference between who was civilized (left-over-right) and who was barbarian (right-over-left).
And yes, people in Japan at that time, and up until the Nara period (around 700 AD), were firmly committed to wearing their robes in the Northern barbarian style, just like the Xiongnu. So were the Tocharians, an Eastern Steppe group who adopted an Indo-Euro language (the only known Eastern culture to do so), but remained true to their origins in dress.
Similarly, superstitions may have a utilitarian logic to them -- in which case, it means nothing if two groups share a superstition. I was looking up Japanese superstitions, and one of them is to not whistle indoors -- it turns out, almost every culture in the world shares this superstition. Probably because everyone perceives it as rude, as though you're trying to be a band-leader in an impromptu concert that nobody asked for. So it's frowned on all over the world, and it cannot be used to prove that Russians and Japanese descend from a common ancestral culture where this superstition was born. It was born in multiple places and at various times, independently of each other. It's convergent evolution.
So when we turn to burials -- the main material trace that is left in the archaeological record for us to study in the present about cultures from the past -- we have to look for examples that look like shibboleths, not practices that many cultures could come up with on their own.
E.g., "monumental size of the grave to reflect the elite status of those buried there" -- yeah, no shit, what else are they going to do, make tiny graves for the elite and mega-tombs for the commoners?
We also have to take into account the notion of "degrees of freedom" from statistics, or how much room for variation there is. You might think, Well, mega-size doesn't show that two cultures share an ancestor, but maybe the particular shape of their mega-tombs could play the role of a shibboleth.
Only problem is -- how many 3D shapes are there to choose from for a tomb? You've got your box-like shape, your rounded mound shape, a pyramid shape, and that's about it, for a single structure. There are far, far fewer possible shapes to build a tomb in, than there are possible sound sequences to convey the meaning of "father". If two languages share a word for "father", that's highly suggestive of shared ancestry. If they both build mounded tombs, that's only slightly suggestive.
So although mounded tombs were popular throughout the Steppe in ancient times, from the West to the East, and although the most famous tombs in all of Japanese history are indeed gigantic mounds (Kofun, giving their name to the period in which they were built, roughly 300-600 AD), that is only slightly suggestive of Japan's Steppe origins.
It certainly doesn't *contradict* the claim that Japanese culture has a Steppe component -- it's in line with the claim, but it's a weaker piece of confirming evidence than some example where there's lots more room for variation and more of a shibboleth nature to it.
* * *
Enter one of the most bizarre and distinctive burial rituals in world history -- the deliberate breaking of finely crafted, highly valuable, aesthetically adorned, built-to-last mirrors. Not just putting a crack in them with a little whack from a hammer, but breaking them into at least 4 separate fragments on average. That's no accident -- especially when mirrors in the old days were made of (polished) bronze, which is much harder to fragment than glass. And these were fairly large mirrors, around 8 inches in diameter, not a little hand mirror -- something that impressive, you'd figure they would want to preserve in order to show off as a status symbol.
This is a great example because we can rule out utilitarian, Darwinian, etc. incentives for two cultures sharing this ritual. It's breaking something useful, functional, and valuable -- it's going against the utilitarian motive. Even in the figurative sense, where the grave goods are not meant to be used by the living, but by the dead in the Otherworld, breaking the mirror deprives the deceased of its use or exchange value in the afterlife.
Imagine waking up on The Other Side, surrounded by mirror fragments -- "Gee, thanks a lot for making them worth a lot less, in case I wanted to trade them for something that's only available in the Otherworld, which you couldn't provide me with during the burial. Or in case I wanted to see what I look like dead, or if I wanted to reflect light for some reason. Whose idea was it to break them into pieces?!"
A superstition about not breaking a mirror, could arise independently through convergent evolution. Mirrors are functional, utilitarian, valuable things -- don't break them, or else bad things will follow. Breaking them on purpose is the opposite -- that must be due to some unclearly motivated shared tradition.
BTW, as for the modern American superstition about "break a mirror, and you'll get 7 years of bad luck", this is claimed with absolutely no evidence to stem from "ancient Greece and Rome" -- always a telltale sign of bullshitting. We don't come from Greco or Roman cultures, even distantly. No one can point to an author of the ancient world saying it's bad luck to break a mirror, cuz you'll get 7 years of bad luck. Or some other number of years of bad luck. Or even explicitly saying that breaking a mirror is unlucky.
So the "breaking a mirror is unlucky" superstition is likely much more modern than that, from the era when mirrors became commonplace and the targets of superstitions. At that point, multiple cultures could independently come up with a superstition against breaking mirrors, in America or wherever else.
Back to the ancient Steppe -- other cultures did in fact bury their dead with the same Chinese bronze mirrors (or imitations), but without the widespread practice of breaking them. That does *not* suggest a common origin for them -- functional, valuable, finely crafted things will be sought after as grave goods no matter who they are.
As it turns out, both the Western Steppe and Han China included bronze mirrors in their grave goods, but nobody thinks they share an ancestor. They both independently figured out that these things were valuable and impressive feats of craftsmanship, so why not include them with all the other goodies in the grave?
It was only the Eastern Steppe groups that fragmented the Chinese bronze mirrors as part of their burial rituals. See this review article of the broken bronze mirror phenomenon, which surveys the Western Steppe as well as the East.
It began with the Pazyryk culture from the Altai Mountain region during the mid-1st millennium BC. They are misleadingly called "Scythian" as though they were Indo-Euro-phone, culturally Iranian, or primarily Western Steppe, none of which is true -- they seem to be proto-Turkic if anything, the western frontier of the Eastern Steppe.
This practice continued, most famously, among the Xiongnu, around the turn of the millennium...
And as fate would have it, among the Kofun burials in Glorious Nippon, in the early 1st millennium AD (and in Silla). As with the Pazyryk and Xiongnu, not all of the Chinese bronze mirrors are broken in Kofun burials, but a large number of them are -- perhaps dozens of mirrors each broken into 3 or 4 fragments on average, within a single site. And this practice was not just one fluke site, but dozens of locations all around Japan.
Nor was it done only in peripheral or culturally deviant regions of the nascent Japanese nation -- if anything, it was done in abundance at the very heart of the soon-to-be Yamato state, around Nara. See this discovery of over 100 mirrors broken into nearly 400 fragments, from the Sakurai Chausuyama Kofun near Nara, dating from the 3rd C and belonging to a very elite individual, possibly the legendary Queen Pimiko herself.
This early date is also helpful to establish that these Steppe influences of Japanese culture did not only arrive during the course of the Kofun era -- they were there before 300 AD. And it is helpful to show that these influences are not "Korean," as though they were confined only to Korea and Japan. Rather, both of them are extensions of a broader Eastern Steppe tradition, brought by Steppe people who crossed the mountains into the Korean peninsula, the first wave going further into Japan, and a second wave remaining in Korea (the Koreanic speakers).
* * *
These bronze mirrors, and mirrors in general, are so important in Japanese culture that one of the three imperial regalia -- the special material items that legitimize each emperor, which are passed on to each new holder of that office -- is an ancient bronze mirror, the Yata no Kagami. (We'll get to the Steppe origins of another of the three, the magatama or curved jade jewel-bead, in the next post.)
Mirrors have always been important in the rituals of Shinto, as symbols of the sun. It seems like the Eastern Steppe cultures view mirrors as solar symbols -- miniature suns that you can wield with your own two hands, throwing bright beams of light wherever you please, like a demi-solar-god all by yourself. The Western Steppe cultures view them more in terms of reflecting the physical likeness of a person, not as pre-industrial spotlights.
And wouldn't you know it? While browsing the Wikipedia article on "mirrors in Shinto," I nearly fell out of my chair looking at how the mirror is displayed in a typical Shinto ritual. See the center top of this image, where the mirror lies between the 4 animal statues. The mirror itself is a disc, and it's resting on a stand that is curved very much like a crescent moon
There's that distinctly Xiongnu visual shibboleth again! A solar disc, with a crescent moon underneath, opening up to the sun, which remains in use on the national flag of Mongolia. We'll see that in Kofun-era earrings, in the post on clothing styles.
Do an image search for "Shinto mirror," and you'll see all sorts of variations on this theme, but they all involve a stand that is crescent and opening up to the solar disc. I can't determine when this combination of items began, but it has endured right up through the present.
Somewhere along the way, the Steppe fixation on the "sun and moon" duo was downplayed, and the crescent shape was carved in the shape of clouds or sea waves or some abstract thing. But who ever depicts clouds as forming a shape whose border is an upward-opening crescent? Or sea-waves depicted with that same border shape? It's obviously a crescent moon, and that border shape has been preserved -- even though its interior has been (re-)decorated to distract from its moon-focused origins.
I wonder if this Shinto mirror-and-stand configuration goes back to the Xiongnu, and the smaller golden "disc with crescent" items that are found among the Xiongnu, or the identical Kofun-era earrings, are just jewelry representations of their sun-and-moon religious rituals, which would have involved one of those large bronze mirrors as the solar disc, supported by a stand in the shape of an upward-opening crescent moon. IDK, something to think about / look into.
.jpg){kind=link}
* * *
Like many ancient facets of Japanese culture, they have been preserved or served as inspiration for even the most hi-tech and futuristic domains of contemporary Japanese culture -- like video games. In The Legend of Zelda: Twilight Princess, there's a crucial item called the mirror of twilight, and it is in the shape and color of an ancient Chinese bronze mirror -- and it also lies in 4 fragments, which must be re-assembled in order to use it as a massive light-reflector.
In an earlier game in the series, Ocarina of Time, the player uses a mirror shield to reflect beams of light in order to trigger doors opening and such. However, it doesn't resemble the ancient bronze Chinese mirrors, and does not lie in fragments. Even earlier, in A Link to the Past, the mirror shield doesn't bounce rays of light, but absorbs them. And even earlier, in The Adventure of Link, the protagonist helps a townswoman find her lost mirror and is rewarded for it -- it's not shown at all, though, let alone in an ancient Chinese form, or lying in fragments, or reflecting light rays. But these are all still part of the enduring Japanese fascination with mirrors.
* * *
Finally, what the hell *was* the reason for breaking the mirrors back then? It doesn't matter for the purposes of linking various cultures together that share the practice. But just to try to get inside their heads...
I actually came up with a similar concept as the Twilight Princess video game, before even reading about it. None of the other grave goods, whether highly valuable or not-so-valuable, are deliberately broken. They're intact, in good working order, and meant to aid the deceased in some way.
Mirrors left intact would be the same -- an aid to the deceased.
But the ones that were broken -- could have been a reflection of the corpse itself. Something that used to be a finely crafted, highly valuable, built-to-last creation -- but that now finds itself at the end of its use, decaying into pieces, losing its order and structure, never to be used again...
Until some fateful event in the future, where the dead are brought back to life, and where these mirror fragments would be supernaturally placed back together, without the awkward glue or whatever means that a person would use, but actually restored to their original state and in their original working order -- with no cracks, glue, or anything else like that to be seen.
Until that day comes, both the corpse and the mirrors will lie in their fragmented, decaying state, unable to function as they were originally created.
Maybe they didn't have the apocalyptic revival of the dead and restoration of the mirrors, as though Humpty Dumpty had been miraculously put back together again. At the very least, they could have intended the broken mirrors to stand for the broken body, broken family, broken social hierarchy, now that this elite individual has been retired from their role.
Since they viewed mirrors mainly as sun symbols, i.e. projectors of light, then a broken mirror is tantamount to a snuffed-out candle for some other culture where candles represent light projection and mini-suns.
Such an important person dying is like the sun and moon themselves going dark in the sky.
We know that ancient Altaic people used to view the sun and moon as mirrors-in-the-sky -- see the earlier post on their creation myth, which is mainly about churning the primordial sea with a divine staff in order to make land-masses out of the resulting sea-foam on the surface. One of them also mentions that during the age of creation, two mirrors were placed in the sky, which brought light to the universe -- the sun and the moon.
So, far from being a sign of disrespect toward the dead, the broken mirror was the ultimate material expression of grief from the mourners.
Wow! A Xiongnu word in Japanese, meaning "ghee" or clarified butter! In Japanese, it's "daigo", in Xiongnu it's "de-ɣɑ".
ReplyDeleteFrom Appendix 2, a list of Xiongnu words that survive in Altaic languages, by Orçun Ünal, "On *p- and Other Proto-Turkic Consonants" (2022).
http://www.sino-platonic.org/complete/spp325_proto_Turkic_consonants.pdf
Daigo was made at least through the end of the 1st millennium AD, and it was used as a metaphor in Japanese Buddhism for "creme de la creme", the perfect flavor (daigo-mi), ultimate realization, etc. Not just any ol' milk product, but clarified butter.
There's a Buddhist temple named after it, Daigo-ji, and an emperor who reigned circa 900 AD was buried there and posthumously named after it -- Emperor Daigo.
https://en.wikipedia.org/wiki/Sarpir-ma%E1%B9%87%E1%B8%8Da#In_East_Asia
Another form of dairy product from the 1st millennium AD in Japan, is So, which was supposedly imported from Baekje, the kingdom in SW Korea. But that doesn't mean it's "Korean" -- it's likely from the Xiongnu as well.
https://en.wikipedia.org/wiki/So_(dairy_product)
In the same Appendix, the Chuvash word for "butter" is "śu", which is derived ultimately from the same Xiongnu word "de-ɣɑ". Chuvash is the only surviving member of the more conservative branch of Turkic, the Oghur branch. It was close to what the Bulgar tribe spoke when they migrated west across the Steppe.
So it's possible this variant entered the Korean peninsula, then into Japan, where it was called "so" instead of "śu".
Who knew that Japan used to make cheese and butter during the 1st millennium AD?! Now we know why the Japanese have an easy time drinking milk today, unlike Chinese, Southeast Asians, and other non-Altaic cultures. They descended from the Steppe (specifically the Xiongnu), and kept producing and consuming dairy well into their recorded history in Wakoku / Yamato / Nippon.
They seem to have abandoned dairy consumption during the 2nd millennium AD, but thanks to various motives, they rediscovered it during the 20th C. Before American occupation, BTW, and under Altaic influence -- the inventor of Calpis picked up a love for fermented milk after visiting Mongolia in 1902.
So, not only did the Japanese continue dairy production long after settling into the Japanese islands, their word for clarified butter was of Xiongnu origin!
How much clearer can it be?! ^_^
To be clear, the Xiongnu confederation did not extend into the Korean peninsula, let alone the Japanese islands. So the only zone of contact between Xiongnu speakers and Japonic speakers -- is that Japonic speakers contained members who used to be former speakers of Xiongnu.
ReplyDeleteThere was no migration of Japonic speakers from the Japanese islands into the Xiongnu confederation while it was still powerful (up through circa 100 AD), and then turning around and returning back to Japan, bringing the word for clarified butter with them. That's totally ridiculous.
Rather, some members of the Xiongnu headed east, into Korea, then into Japan, bringing the word -- and the practice -- of clarified butter with them.
Glorious Nippon, grandson of the OG badass nomadic Steppe empire, the Xiongnu. ^_^
I found this article while pursuing a related line of thought, about the Scythians being from the Eastern Steppe and not Indo-Euro (at best, Indo-Euro-speaking).
ReplyDeleteAnd that led me to the Amazons -- a fairly common pattern among Eastern Steppe cultures, including Glorious Nippon of course. Not just an isolated female warrior, but a whole group of women on the battlefield.
That is simply absent from all of Indo-European history, aside from the supposedly Indo-Euro Scythians. But not rare in the Altaic cultures:
https://en.wikipedia.org/wiki/Onna-musha
Everything else about the Scythians' culture matches the Eastern Steppe, not the Western Steppe. Some of them may have had Indo-Euro DNA at 50% -- and the other 50% was Eastern Steppe.
But genetic variants do not carry cultural variants. If two genepools merge at 50-50, the culture of that synthesis group could be 100% of A and 0% of B, or 0% of A and 100% of B, or literally anything in between.
My best view as of now is that an Eastern Iranian group migrated from Bactria to the Altai Mountains, where they were culturally absorbed into a Proto-Turkic group, although genetically it was closer to 50-50. Their language is uncertain, maybe Iranian, maybe Turkic. Very little of it is actually known.
Like many future waves of Turkic cultures from the frontier between the Western and Eastern Steppe, they migrated westward across the Steppe, landing in the region between the Black and Caspian Seas. They were horse-mounted archers, drank fermented mare's milk, wore conical hats, brought the Animal Style of art with them -- although they did not break their Chinese bronze mirrors when burying their dead, unlike Eastern Steppe groups. Maybe that was one element of Indo-Euro culture that was retained in Scythians, which was otherwise heavily Eastern.
Viewing them as Indo-European, whether specifically Iranian or generically as Indo-Euro, presumes too many "once in all of history" happenings. Like Iranian speakers being near the Altai Mountains -- not even the most sprawling Iranian empire ever, the Achaemenid Empire, made it to the north and east of the Aral Sea:
https://en.wikipedia.org/wiki/File:Achaemenid_Empire_500_BCE.jpg
In fact, there are only 2 Indo-Euro languages spoken in that region in all of history -- Russian, during the Modern era, and Tocharian in pre-historic through Medieval times. But Tocharians were another example of what I'm saying the Scythians were -- an Eastern Steppe group who adopted an Indo-Euro langauge, with some other minor borrowings and/or minor gene-flow from Indo-Euro migrants. They used to be Uralic / Samoyedic speakers (Peyrot).
No Indo-European group has migrated westward across the Steppe.
Indo-Euros were more famous for using horses to pull war-chariots, not riding singly on a horse and relying on archery for warfare, which the Scythians did, much like the Turkic and Mongolic groups.
Dark Age Byzantine writers said, while observing the Turkic migration of their time, that the Turks were formerly called Scythians.
This is a whole 'nother series of posts, but just to keep adding bits to this theme going in the meantime...
Onna-musha are kakkoi, but I prefer Japanese woman pirates instead, like Hololive's goddess of sexy-ness, Marine.
ReplyDeleteSpeaking of Marine, a Japanese girl with a bubble-butt can be called a PAWG too -- Phat Ass Wajin Girl. ^_^
ReplyDeleteIt's not only a term for bubble-butt white girls! ^_^
To any Japanese speakers reading this in translation, "phat" is a positive term when referring to oshiri. Very high praise!
Maybe the "Scythians" were Uralic speakers, or former Uralic speakers who adopted an Indo-Euro language, much like the Tocharians. And like the Magyars, formally called Ungri, another Steppe group that drove westward into Eastern Europe due to horse-mounted archers.
ReplyDeleteCulturally or ethnologically, they may have been Turkic, but their language looks more Uralic. Those articles by Peyrot on the Uralic background of the Tocharians is really productive! Why not apply it to the other supposedly Indo-Euro group from that region, who culturally do not resemble the Indo-Euros at all?
There is very little known about the language(s) of the Scythians during the 1st millennium BC, during their heyday. But Herodotus recorded a few terms from some Scythian language, with a gloss in Ancient Greek.
The Scythians called the Amazons "Oiorpata," which he says segments into "oior" meaning "man" and "pata" meaning "slay" -- i.e., they were the "man-slayers".
The other is a tribal name, "Arimaspi", which he says says segments into "arima" meaning "one" and "spi" meaning "eye" -- a legendary tribe of one-eyed people in "Scythia".
One of their few widely known individuals from the 1st millennium BC, was the Queen Tomyris of the Massagetae people, who led her armies into the battle that killed Cyrus the Great, King of the Achaemenid Empire, as the Persians tried to press northward through southern Central Asia. That right there tells you the Massagetae were not Iranian -- the only cultures to block the Iranian pushes through Bactria / Sogdiana have been non-Indo-Euro, like Turkic.
https://en.wikipedia.org/wiki/Tomyris
Putting aside the intended meanings of these Scythian words, focus just on their phonology.
There are basically no voiced obstruents -- only the inter-vocalic "g" in Massagetae.
There are also no fricatives like "f" or "th" or "kh".
The syllable structure is mostly open, aside from sibilants and liquids in coda position, and a geminate sibilant in Massagetae -- suggesting a consonant distinction based on length, not voicing or aspiration. Depending on how Tomyris segments, maybe nasals in coda position as well. But generally avoiding consonant clusters with a syllable, and even at the word level.
The name of the related group, Sauromatians, was more like "Sa(u)rumata", also totally open syllable structure, lacking voiced obstruents or fricatives.
The name Scythian was originally more like "Skudha", where "dh" is the voiced interdental fricative -- the only unambiguous voiced obstruent, and a fricative as well.
The supposed consonant cluster "sk" in "Skudha" may be a Greek rendering of a word that had an initial vowel, which all of the Semitic renderings have, and so does the Urartian rendering. Perhaps the Greek rendering is closest to the original, and Semitic and Urartian are epenthesizing a vowel to fit their own phonotactics. But two unrelated language families say there's an initial vowel, and only Greek disagrees -- so I'm siding with there being an initial vowel.
It's a front vowel, although the renderings disagree about its height -- so the original should be either "Askudha" or "Iskudha" or (although the "e" is not attested in these renderings) "Eskudha".
Needless to say, none of this looks like an Indo-European language, or specifically from the Indo-Iranian branch. IE is rich in voiced obstruents, including aspirated ones, and its descendants such as Indo-Iranian are rich in fricatives. They're also comfortable with consonant-rich syllables, even if not quite at the level of Caucasus Mountain languages.
So which ancient language family best fits the picture portrayed by this handful of Scythian words? Uralic!
ReplyDeleteFirst, why not Turkic? Again, they could have been culturally Turkic, but adopted a Uralic language. I'm putting aside the whole rest of their ethnic markers, just focusing on language now.
Proto-Turkic has the voiced obstruents "b" and "d" and "g", it does not have the interdental voiced fricative (or any fricatives for that matter), it does not have a consonant length distinction, it is more comfortable with consonant clusters, including word-final clusters, and is comfortable with word-final obstruents.
Now to the real answer, Proto-Uralic:
https://en.wikipedia.org/wiki/Proto-Uralic_language#Consonants
Mostly lacking voiced obstruents and fricatives -- except for the very unique combination of those features, the voiced inter-dental fricative! (They even have a separate palatalized form of that rare consonant.)
It also has an unclear velar consonant "x", and some suggestions are that it's a "g" or a "gh" (fricative voiced velar stop, not aspirated g). So that accounts for the "g" in "Massagetae". Between vowels, a stop "g" turning into a fricative "g" is fairly common across languages (Spanish, most Turkic ones), so its precise nature doesn't matter. The point is, the mystery velar consonant from Proto-Uralic can supply the "g" in "Massagetae".
Proto-Uralic also allowed geminate consonants, as opposed to voicing or aspiration. That's in line with the double-s in "Massagetae".
The syllable structure is much more open, with only 1 consonant allowed in coda position, and no final clusters. Seemingly not initial clusters either, judging from this list of PU reconstructions:
https://en.wiktionary.org/wiki/Appendix:Proto-Uralic_Swadesh_list
This is a MUCH BETTER fit to the phonology of Scythian words recorded during the 1st millennium BC.
Scythians were Uralic speakers!
Well, sure, Uralic speakers are the only other group aside from Turkic or Mongolic to have driven westward across the Steppe, using horse-mounted archers, arriving in Eastern Europe, and even settling down there for awhile (like the Magyars).
At the very least, they were former Uralic speakers who adopted an Indo-Euro language, and their Uralic L1 acted as a filter on the Indo-Euro target L2 -- eliminating most of the voiced obstruents and fricatives and aspiration, other than the interdental voiced fricative and the mystery "x" consonant, and forcing a very open syllable structure.
That would make their scenario like the Tocharians at the least, and like the Ungri at the most. But in either case, they were not Indo-European culturally.
Uralic is also further to the west than the Altaic family, near the Ural Mountains rather than the Altai Mountains, which is in line with Scythians residing in the Western Steppe, rather than further east.
ReplyDeleteThose so-called "Scytho-Siberian" groups are not Scytho at all -- they're Turkic or other Altaic group. E.g., the Pazyryk culture.
And the zone of contact with Indo-Euro cultures matches Uralic better than Turkic or other Altaic -- the Sintashta culture, which most definitely *was* Indo-European and Proto-Indo-Iranian and genetically Western Steppe, was next to the Ural Mountains, not Altai Mountains.
So, perhaps the absorption of (Indo-)Iranians into the non-IE culture that became the Scythians, took place earlier on and near the Urals, where they got absorbed by Uralic speakers -- not as I speculated before, about an Iranian group from Bactria migrating northward and hugging the mountains until they reached the Altai region of the Steppe.
Both of those scenarios involve absorption of IE's by non-IE's, but an earlier date and closer to the Urals is a better fit for where (Indo-)Iranians could be in contact with Uralic speakers. I only said a zone of contact near the Altai under the assumption that the absorbing group was Turkophone -- but it looks more like they were Uralophone, so make it the southern Ural zone instead.
That would also explain why the Scythian burials in the Black-to-Caspian Sea region didn't break their bronze mirrors! That was a distinctly Altaic ritual, from Pazyryk to the Xiongnu to Silla to Glorious Nippon.
I neglected to mention that Chinese bronze mirrors also traveled as far as Afghanistan and Vietnam -- but these cultures didn't break them into pieces either, only Altaic speakers did.
If the Scythians were culturally more Uralic, then they would not have participated in the Altaic ritual of breaking their mirrors during burial either!
It all makes so much more sense now!
In the same way that the Xiongnu were the harbinger of the Gokturks and Mongols, the Scythians were the harbinger of the Magyars!
Finally, what about Scythian etymologies? I'll start with one example, which I started out with, the Amazons. Herodotus says their name in Scythian was "Oiorpata", segmented into "oior" (man) and "pata" (slay).
ReplyDeleteIf Scythians spoke an Iranian language, there would be a clear Iranian etymology for both of those segments of the compound. Yet there isn't. The first segment has a plausible Iranian or Proto-Indo-Euro origin, which means "man":
https://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-European/wiHr%C3%B3s
It has an initial "w" consonant, unlike "oior" -- but maybe the "oi" sound is trying to capture a bilabial approximant like "w" followed by an "i". It has a diphthong, or triphtong or several vowel-like elements at any rate. It has an "r" after that. However, after the "r" there's a vowel, from PIE to P-Indo-Iranian to P-Iranian to most of its descendants, other than the Eastern branch like Sogdian. But overall, not a bad match phonologically and semantically.
However, the "pata" segment has no Iranian or PIE derivation having to do with killing, beating, stabbing, etc. The closest that the proponents have come up with is "pati" meaning "lord" -- so the whole argument rests on the ridiculous premise that in Iranian, "lord" was linked to "slaying" rather than something about authority, social hierarchy, etc.
You don't slay your subjects -- you slay your rival elites in a power struggle, or their military underlings, not your own subjects (and generally not your rivals' subjects either -- you slay their lords, and take over their subjects afterward).
So there goes the Iranian etymology. If someone wants to offer a further back PIE etymology, I'm all ears, but none has been suggested further back than P-Indo-Iranian.
Although we've excluded Turkic on phonological grounds, it's worth showing that it too can supply the first element meaning "man", to show that I-E is not unique in that regard:
https://en.wiktionary.org/wiki/Reconstruction:Proto-Turkic/%C4%93r
However, like I-E, Turkic cannot supply the second element having to do with killing. That's what led me into that article about "initial p in Turkic" -- if it was there, it was there a long time ago, and by most of its attested history, it turned into an "h" or dropped altogether.
Even if we ignore the absence of initial "p" and look at the words related to killing, hitting, stabbing, etc., only one has a medial "t" -- "öltür", which does mean "kill" -- but it also has a liquid just before "t" and another liquid in word-final position. Not to mention the lack of an initial consonant, and its vowels both being high and rounded rather than low and unrounded. Not a good fit.
https://en.wiktionary.org/wiki/Appendix:Proto-Turkic_Swadesh_list
Enter the Uralic etymology for the Scythian name of the Amazons, "Oiorpata". This is the only family that can supply plausible candidates for both segments of the compound.
ReplyDeleteFirst, "urɜ" meaning "man, male" (the 3rd character there is a vowel, not a consonant):
https://en.wiktionary.org/wiki/Reconstruction:Proto-Uralic/ur%C9%9C
Second, "pata" having to do with the verb of cooking something in a pot (Samoyedic) or the cooking-pot itself:
https://en.wiktionary.org/wiki/Reconstruction:Proto-Uralic/pata
Amazing fit!
So this term would not have meant a generic kind of "man-slayer" but specifically "man-cooker" -- amping up the negative Other-ness by hinting at the Amazons' cannibalism, although technically you can boil someone in a pot without eating them afterwards.
Very close to the English term "man-eater" about femme fatales who are tempting aggressive baddies, but who you might regret mating with...
Also, it lessens the "yuck" factor of cannibalism, if they only cooked and ate males -- then they're like the praying mantis or other "black widow" archetype, killing and eating their mates after getting it on, rather than eating their fellow human beings in general.
So that's who the Amazons were -- the OG femme fatales, the man-cookers.
Be careful about wishing for warrior babes whose athletic bods will hump you into exhaustion -- you just might end up being their meal afterward!
That fits with the description of them only interacting with males in order to mate, and otherwise keeping company only with each other. They met the guy long enough to collect a sperm donation, then killed him afterward, possibly eating him for good measure. It's the praying mantis / black widow archetype -- not being a man-hating feminazi, going on a serial killing spree against men, etc. But being a black widow.
ReplyDeletePerhaps they also cooked the men they slew on the battlefield, not an uncommon practice for pre-historic tribes.
Or maybe they combined the two -- slew most of the men, kept enough alive to mate with, then killed those mates afterward as well.
Imagine thinking your doom had turned to good fortune -- you were in the minority of your troop to be spared by the Amazons, they give you that mischievous grin and start humping you into exhaustion, but then they give you an even more sinister grin, and it's into the cooking pot with you too!
Yeah, but at least you got to go out with a bang, as it were, unlike your comrades who just ate an arrowhead on the battlefield...
I might see what else I can come up with for Uralic etymologies of Scythian words, probably in a separate post on the topic of the non-IE nature of the Scythians. But I just had to share that one immediately.
ReplyDeletePower-humpers, but also man-cookers... lol.
If you don't want to end up in the cooking pot, stick to the pillow princesses.
ReplyDeleteSome guys like living on the wild side, though, with all the mortality risks that brings along with it...
Since the Amazons were the all-hellcat biker gang of their day, here's a little reminder that the biker baddie in Pee-Wee's Big Adventure, who says "Let me have him first" while her gangmates are plotting how to kill him, is none other than Cassandra Peterson, AKA Elvira.
ReplyDeleteStarting at 2:00
https://www.youtube.com/watch?v=gyhy7Z71qzE
Sweet Jesus, I never knew these two iconic baddies were played by the same woman -- not suprising, though. Wow!
More on the non-Indo-Euro nature of the Scythians and Sarmatians, and sticking to language, which again is not the most important aspect of ethnicity -- one of the least, given how easy it is to change languages vs. other markers of your cultural in-group.
ReplyDeleteFirst, most of the so-called Scythian studies began in the 19th C, in Eastern Europe. They're the ones responsible for conjuring up the idea of Scythians, Sarmatians, Cimmerians, etc. all representing a vast Iranian region in the Western Steppe in the 1st millennium BC.
Nobody in ancient contemporary accounts, or Medieval writers nearby like Byzantine Greeks, said that the Scythians, Sarmatians, or Cimmerians were Iranian, Persian, Persian-ish, etc. Otherwise, such observations closer to the source would be cited -- but they never are. In fact, when they do speak on the matter, Medieval Greeks said the contemporary Turks used to be called Scythians.
So, when you see some claim about the meaning of some Scythian-related term, look into it, and there's typically very shaky or no solid evidence for it. It's one guy from modern Eastern Europe grasping at straws about Iranian origins.
For example, the ethnonym Scythian, supposedly "Skuδa" in its original form. The Iranian etymology claims that this descends from Proto-Indo-European "(s)kewd" for "propel, shoot", with the intended meaning that the Scythians were named after being archers.
However, none of the descendants of that PIE term have to do with shooting, only with propelling, driving, rushing, etc.:
https://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-European/(s)kewd-
The Albanian word "hedh" means "throw" -- and the Scythians did not hurl projectiles, they shot them from bows.
The only branch that has a term meaning "shoot" is Germanic, and that connection is disputed. The Germanic term may derive from PIE "sket" for "to drive forward". In which case, there goes the connection to the "u" vowel (via the "w" in "(s)kewd"), and there goes the voicing on the final stop. "Sket" is better cuz it has an obligatory initial "s".
https://en.wiktionary.org/wiki/Reconstruction:Proto-Germanic/skeutan%C4%85
So, no, "Scythian" and "Skuδa" do not mean "shooter" AKA "archer" in PIE. It's one of the weakest claims in "Scythology," and that's saying something.
Sidebar: open-ended etymologies are always weaker claims than ones whose intended meaning is already known.
ReplyDeleteE.g., when Herodotus says the sound sequence "Oiorpata" means "man + slay" in the Scythian language, the etymologies must adhere to that. Of course, Herodotus could have gotten the meaning wrong. But he's much more reliable than someone who allows literally anything to be the meaning of "Oiorpata".
Time and again in "Scythology", these close-ended etymologies are never there. Like the ridiculous attempt to make "pati" from Iranian mean "slay", when it just means "lord". They can't wiggle their way out of these puzzles, cuz there's a clearly defined target that they must aim for. They can't shoot in any ol' direction and claim success.
Whereas my Uralic etymology for "man" and "cooking pot" is far closer to the target. Maybe I'm wrong -- but I'm way closer to the target, and any objections have to get closer to the target than I did. And obviously those won't be coming from Iranian or Indo-European, cuz they've been trying their hardest in those approaches, and have only come up with the far-wide-of-the-mark derivation involving "lord" instead of "slay".
So, unless the source tells us what the meaning of some tribe's name is, we can invent all sorts of potential etymologies. They were named after being archers -- no, they were named after being swiftly propelled i.e. by their horses -- no, they were named after a switfly flowing river that they originated from -- no, they were named after...
The best we can do for place names or personal names or tribal names, is that they must be specific -- these are determiners, modifiers, qualifiers, specifiers. They distinguish one person from another person, one place from another place, one tribe from another tribe. If they're vague, they're bad. If they're specific, they're plausible -- not guaranteed, since again we don't know the intended meaning that we're supposed to hit like a target. But at least plausible.
That brings us to the Sauromatians / Sarmatians, supposedly a group related to the Scythians. I already reviewed on non-IE their name sounds on phonological grounds. Now let's investigate the meaning side.
ReplyDeleteThe Wikipedia entry says it's cognate with "śarumant" in Sanskrit, meaning "armed with throwing darts and arrows", from "śaru" meaning any kind of missile weapon.
But then we see if this Sanskrit word is of PIE origin -- and it likely is not.
"Lubotsky claims non-Indo-European origin and connects Tocharian A śaru (“hunter”), Tocharian B śer(u)we (“hunter”)":
https://en.wiktionary.org/wiki/%E0%A4%B6%E0%A4%B0%E0%A5%8D%E0%A4%B5
Whaddaya know, the Tocharians used to be Uralic speakers before switching to an Indo-Euro language (Peyrot).
So in fact this word for "arrow" may reflect contact between an Indo-Iranian group and a Uralic group -- exactly what I'm claiming for the broad class of Scythians, Sarmatians, and others from the Western Steppe.
However, we don't know that the meaning of Sarmatian is "armed with arrows" -- this is just one plausible candidate.
I can do even better, by getting the "m" into the root, whereas the one involving "śaru" has to leave it to a compound or derivational element outside of the root.
ReplyDeleteWiki says the first reference to Sarmatians is the "Sairima" in the Avesta, which then evolved into "Sarm," meaning we don't necessarily need to worry about the "t" in Sarmatian.
My etymology is a specific term, and one that we typically find around the world -- a totem animal.
Something like "sarma", referring to various wild predatory animals from Northern Eurasia, or as a generic term for wild predatory animal. Its source is Uralic, where it's most varied, and was later borrowed into neighboring Balto-Slavic to the west, as well as neighboring Yukaghir to the east.
See "Hermes and Pan expanded" by Gheorghiu (2024) for the lists of related terms.
https://zenodo.org/records/10616826/files/Hermes%2520and%2520Pan%2520expanded.pdf
The Uralic terms, which stretch from the furthest western and eastern extremes of the family, and including the two main branches of Finno-Ugric and Samoyedic:
*śurme "wild animal" (UEW 490-91) > Saami Kld. tš́ irm; Ko. Not. tš́ ɛ̮rma, "wolf; devil" | Mari KB
sǝrmǝ, ̣U šurmaŋɣe, šurmaŋše, B šurmaŋše "lynx" | Khanty DN t́ürǝm, Ni. Śurǝm "ermine,
(snow-)weasel, marten" || Samoyedic *sarma (Janhunen 1977, 136; Helimski1997, 335) > Nenets O
sārmik "predator, wolf; animal", pīn-sārma "owl" (pi=night); Enets same "wolf"; Selkup Ta. suurem,
Ke. súuram, súurm, N húurup, Tur. sūri̮p "wild animal";Tur. tī̮mpitil sūrip "bird"; Karagassian sarma
"hazel grouse / Tetrao bonasia".
The separate Yukaghir family, occupying Northeastern Eurasia, has a word for "wolf" that is "sarimə" -- probably a borrowing from Uralic, given its sparser attestations in this family.
https://en.wiktionary.org/wiki/Appendix:Proto-Yukaghir_reconstructions
Then there's Balto-Slavic to the west, where it means "wolverine". It has been taboo-metathesized into "rosomaxa", whose Proto-Slavic root before the taboo is likely "sorma" (the "-xa" is a suffix). In a Belarussian dialect and in a Circassian Ukrainian dialect, the original order of the first 2 consonants is preserved as "saromaxa" or "soromaxa":
https://en.wiktionary.org/wiki/%D1%80%D0%BE%D1%81%D0%BE%D0%BC%D0%B0%D1%85%D0%B0
In Lithuanian, "šarmu" / "šermu" means "ermine" rather than "wolverine" -- but still a kind of weasel.
So, I propose that the "Sairima" of the Avesta were named after their clan's specific totem animal, which could have been a wolf or a wolverine -- a swift, ferocious predator treated with taboo respect by many cultures that cohabit with these species.
This totem animal's name is ultimately Uralic, and although it has been borrowed into Yukaghir and the Eastern European branches of Indo-European, it was not borrowed into Indo-Iranian. So that makes it seem like the Uralic majority of the Sarmatians brought their pre-existing totem animal with them, not that they passed it onto an Iranian group who claimed it for themselves.
ReplyDeleteIt still allows for a contact between Iranian and Uralic, as in the "śaru" = "arrow" etymology, but this one does not show further cultural flow into Indo-Iranian. Rather, the Iranian minority got absorbed without passing "sarma" = "wolf / wolverine" onto their fellow Indo-Iranians.
I don't mind the "archer" name for the tribe, but that does sound like more of an exonym than an endonym. It's what somebody who was *not* famous for archery would call the archers. An endonym is more likely to be a totem animal, and wolves and wolverines fit the bill perfectly -- especially on the Steppe, where wolf cults of all sorts have existed throughout the ages.
Note that Indo-Iranian, and PIE, *do* have their own word for wolf (and in Anatolian, its meaning was changed into another wild predator, the lion), and yet the Sarmatians were named after the Uralic version, not the Indo-Euro version, if I'm right.
But I think I *am* right -- there is very weak evidence for the cultural Indo-European-ness of any of these Scythian, Sarmatian, or Cimmerian groups. It's far more likely that they were harbingers of the Ungri / Magyars, Turks, and Mongols.
Having said that, it's possible that "sarma" = "wolf" was passed onto Indo-Iranians, in the form of the "dog of the gods" in Hinduism, Sarama:
ReplyDeletehttps://en.wikipedia.org/wiki/Sarama
She's described as the mother of all dogs, or even the mother of all wild animals -- and so, more wolf-like than a mere dog. And also as a totem animal, a mythical progenitor -- not the generic term for wolf or dog.
This name is not attested elsewhere in the Indo-Iranian branch, crucially it's not in Iranian.
Back to that Gheorghiu article, it's possible that Sarama from Hinduism is cognate with Hermes from Greek mythology, who ultimately comes from Thrace, which is on the west bank of the Black Sea, and close to the western edge of the Steppe.
Most people assume Hermes' name is not of Indo-Euro origin. Well, if it's Uralic, and is cognate with the totem animal of the Sarmatians, then there you go.
Given how thin its attestations are in Indo-Euro languages and mythologies, vs. how broadly diverse it is in the Uralic world, its origin must be Uralic, and it only entered Indo-Euro cultures through specific historically contingent contact events -- like the Thracian portal, which led into Greek, and possibly Hittite, and the Sintashta portal which led into the Rig Vedic culture (but seemingly not Iranian?).
In the IE cultures, it has a highly sacred or totemistic meaning -- whereas in Uralic, it tends to refer to mundane predators.
The direction usually goes from a mundane to a sacred referent -- elevating the wolf into a totem animal for your clan. Sacralization.
Not starting with a highly reverential attitude toward some species, and then mundane-ifying its name into the word for the ordinary species that it's drawn from.
Since the Uralic (and Yukaghir) sides show the mundane usages, that's where it originated from. Slavic shows a taboo metathesis, and it's a mythical being in Greek and Sanskrit, so it entered these cultures having already been sacralized by its source culture -- which must have been Uralic.
Last case study for today, the Parthians, who led an Iranian empire that was contemporaneous with the Romans.
ReplyDeleteRather than being Iranian or Indo-European, I think they were the harbinger of a well-established pattern -- non-Indo-Euro horse-mounted archers from the Steppe who took the southern route, into the Iranosphere, rose through the military ranks, and controlled (de facto or outright) the Iranian state / empire. And who acculturated themselves to some degree into Iranian culture.
This is different from the non-Indo-Euros who swept directly westward across the Steppe, whose only major example of acculturating themselves into the local culture is the Bulgars, who adopted a Slavic language and were the first Slavic group to adopt and evangelize for Christianity, giving up their Tengrist religion.
Typically, the westward-sweeping groups just fucked everybody up and imposed their language, a la the Magyars, or immediately returned to the east with their pillaged loot. They were more like a nomadic drive-by biker gang, whereas those who took the southern route through Bactria, into Iran, south of the Caucasus, into Anatolia, and at the height of the Ottoman Empire, even into the Balkans -- were not the biker gang type, but foreign mercenaries who rose through the ranks of their adoptive nation's institutions, taking them over, and acculturating themselves, sticking around long-term rather than turning back with their loot.
From the Dark Ages through the Early Modern period, this is referred to as Turco-Persian culture.
But perhaps this pattern extended back further, at least into the Parthian era -- with non-IE mercenaries from the Steppe (Uralic, Turkic, some combo, who knows) migrating down into southern Central Asia, where the region of Parthia is located. Parthia is not from central or western Iran, it's right on the frontier with the non-IE part of Central Asia. From the Dark Ages onward, it was mainly Turkic, sometimes Mongolic, although during the 1st millennium BC there could have been a Uralic presence as well.
The Parthians were famous for being nomadic horse-mounted archers, dressed like Steppe people, and were otherwise not very Iranian or Indo-Euro.
Ancient sources refer to them having Scythian origins, such as Justin from circa 200 AD in the Roman Empire:
"The Parthians, who are now in possession of the empire of the East, having, as it were, divided the world with the Romans, came originally from Scythian exiles. This too is evident from their name: for in the Scythian language the word Parthi signifies exiles."
"Their language is a mixture of the Median and Scythian, borrowing words from both."
Jordanes, writing in the 6th C Byzantine Empire, echoed the etymology, saying that in Scythian, "Parthi" meant "deserters" who stayed on after a battle rather than stick with their military, or alternatively, those who fled from their kinsmen. "Exiles" in either case, though.
So the Parthians were not from the region that bore their name, but from some foreign kingdom or kin group. Technically, it could have been from some other Indo-Iranian or Indo-Euro kingdom or kin group -- but given the specific mention of Scythian origins, they were likely from far outside the region, i.e. from the Steppe, and traveled some distance toward the Iranian frontier.
ReplyDeleteAnd when Justin says their language is a mixture of Median and Scythian, borrowing words from both -- that sounds more like a creole situation, not a dialect continuum as though both languages were Iranian.
Technically, Scythian could be from some other IE branch outside of Indo-Iranian, but what else is there in that part of the world? Tocharian? The Tocharians resided to the east of the Altai Mountains, probably nestled into the Tarim Basin by this point.
So that leaves the more likely explanation, that Scythian was a non-IE language that they brought from the Steppe -- meaning Turkic or Uralic.
Now, how to arrive at "Parthi" meaning "exiles"? Unlike the open-ended etymologies, this one has a clear target to aim for.
ReplyDeleteAt first glance, the origin seems to be Indo-Iranian, as in the Sanskrit word "para" which has a variety of meanings, several having to do with other-ness, foreign-ness, remoteness, alien-ness, left remaining, and stranger. All of these are very close to the target meaning of exiles, i.e. remote foreigners here in our land.
https://en.wiktionary.org/wiki/%E0%A4%AA%E0%A4%B0#Sanskrit
In the ablative case, the adjective form means "other, different", and the ablative includes a "t" after the "para" stem -- "párāt". It's not an aspirated "t" or an interdental fricative, so it's still not like the "th" in "Parthi" -- but it's still pretty close.
Well, Justin and Jordanes (who may have been copying Justin, and not independent) said that "Parthi" meant "exile" in the "Scythian" language, so doesn't the above finding mean that Scythian was Indo-Iranian? Perhaps.
But it's also possible that he was confusing Scythian for the Scythian-Median hybrid language spoken in Parthia. After all, the toponym and tribal name "Parthia" are not attested outside of that exact frontier region of eastern Iran.
So I think it's more likely that this Indo-Iranian term belonged to the hybrid language of Parthia, and came from the "Median" AKA Indo-Iranian side, rather than the Scythian side. It meant "land of the foreigners, exiles, etc." -- alluding to the presence of non-Iranians, and likely non-Indo-Euros, who had begun migrating southward from the Steppe, and taking up residence there.
This is supported by there being a variety of names for the specific tribe that led the Parthians to conquer the rest of Iran and from there the broader empire. Namely the Parni, who are also thought to have been "Scythian" i.e. from the Western Steppe somewhere who migrated southward.
https://en.wikipedia.org/wiki/Parni
Their name still has "par" in it, though not a vowel right after it, and with an "n" instead of a "t" after "par".
There are related tribes from that region of possible Scythian origin -- Sparni or Apartani and the Eparnoi or Asparioi. Again the only common element is "par", somtimes with a vowel and/or "s" before it, often with no vowel after "par", sometimes a "t" or "n" after "par".
So we don't need to rely heavily on "Parthi" and "párāt" -- as long as there's "par" at the core, it's fine. That seems to be the basic theme, with many distinct variations on it.
Of course, in "Parthi" there is no vowel after the "par" element.
ReplyDeleteSo, enter the possibility that this "par" element is Uralic -- albeit having been borrowed from Proto-Indo-Iranian at a much earlier stage, presumably from the Sintashta culture.
ReplyDeleteHolopainen (2019), "Indo-Iranian borrowings in Uralic", is exactly what it says -- reviewing the evidence for all sorts of proposed borrowings *from* Indo-Iranian *into* Uralic.
P. 190 discusses the proposal by 2 separate scholars (Katz, Redei) for II "parHa" (further, beyond, etc., as already discussed in the Sanskrit version) serving as the donor for "por" in Ob-Ugric languages, historically spoken in the Southern Urals area.
Among these speakers, "por" is the name of one of the two clans (the other being the "moś"), into which their society is split. They must marry outside of their clan. They're the lesser clan in prestige, hence the etymology suggesting they were named for being other, alien, different, strangers, etc.
Now, I'm not claiming that the "por" clan from the Ob-Ugric speakers was the source of the Parni tribe, or other Scythian tribe with "par" in their name. Maybe, maybe not.
What matters is that their example suggests that the Indo-Iranian term could have entered Uralic, and served as a tribal name within Uralic speakers.
If it happened in the case of these particular Uralic speakers, maybe it happened in some other group of Uralic speakers. Like if the Scythians were Uralophone, and absorbed some Indo-Iranians from the Sintashta culture. The Scythians could have borrowed the "parHA" term into their own Uralic language, where it was deployed to refer to tribes who were different, other, strangers, or exiles.
One of whom, in the 1st millennium BC, became the Parni tribe, and/or the relatedly named tribes referred to earlier.
And so, although ultimately of Indo-Iranian origin, the "par" element of "Parni" and "Parthi" et al may exemplify a Uralic language that had earlier been in contact with Indo-Iranian and borrowed it from them.
I'm open to being corrected, but I don't know of any major examples where this "par" element is used for tribal names -- with the meaning of "alien, exile, stranger, etc" -- that are definitively Indo-Iranian, like from present-day Iran, Afghanistan, Pakistan, and NW India.
But there *is* a major example of the "par" element being used in a tribal name, referring to alien-ness or remote / other-ness, among Uralic speakers -- it's the major division of clans going back to ancient or Medieval times, among the Ob-Ugric speakers, who are from the Southern Urals, where they could have come into contact with the Indo-Iranian Sintashta culture.
So, if we see "par" meaning "other etc" being used as a tribal name, the only precedent is from a Uralo-phone and culturally Uralic group -- the Khanty and Mansi, both of whom are split along the por vs. moś clan division.
In the absence of corresponding tribal names from the Indo-Irano-phone world, we have to presume that the Scythian tribes with "par" in their name were also Uralic rather than Indo-Iranian, although certainly they had been in contact with Indo-Iranians at some point and likely absorbed some of them into their Uralic culture.
It's funny how these two culture may have met initially near the southern Urals, then the Indo-Iranians made a chariot-steering bee-line for Iran and India, while the Uralic cultures continued hanging out as nomads on the Steppe...
ReplyDeleteOnly for some of those Uralic nomads to wander south and east to eventually reunite with the Indo-Iranians in southern Central Asia, getting the Uralic-Iranian super-group band back together for another international tour...
One that would make them not just a Steppe confederation, but a centralized empire that took over several large-scale civilizations, and rivalling the Romans...
And serving as a harbinger for the Turco-Persian phenomenon that would follow, albeit that would draw from Turkic rather than Uralic Steppe nomads. But non-Indo-Euros who migrated south to join Iran from the north and east, rise through the ranks of the military based on merit, and actually acculturate themselves into Iranian society rather than ransack the place and make off with the loot (a la the westward-sweeping Steppe people).
I'll have to see what I can come up with later for the etymology of the "Skudha" name, but so far we've seen how Uralic that whole region was at the time, so I'm sure it'll be Uralic as well.
But it'll be an open-ended one, since no one said "Skudha means such-and-such in their own language". Pretty interesting etymological trek for "Parthian" though, eh? That's why you're spending a Saturday at the Cliffs of Wisdom rather than doomscrolling AI slop. ^_^
And now time for a little leisure, diving head-first into my anime backlog -- I FINALLY found a site where I can stream or download from, and they have everything ever made, with English subtitles! All those Doraemon movies I haven't seen yet, here I come! ^_^
Well, one last mini-thought but of major importance. That is the notion of convergent evolution or adaptation leading to the same thing happening over and over again in historical dynamics.
ReplyDeleteIf we don't have an immediately clear picture of who the Scythians, Sarmatians, Parni, and Cimmerians were -- we should assume the default, based on external evidence. Sure, it could be an exception, but start with the default and see if that's good enough.
The more you have to assume an event that never happened in human history other than the event you're trying to explain, the less believable it is. The more it fits into a well established pattern, the more believable it is.
And I don't mean universal archetypes like "sedentary agrarian state vs. nomadic pastoralist confederation". I mean, specific, concrete paths like the rainfall flows in this specific direction along this specific river cuz that's how the topography of the mountains is that it's flowing down.
The Parni / Parthians took over Iran by migrating from the north, ending on the eastern side of Iran. Who else in world history ever did that? Every other example is Turkic or Mongolic -- so by default, that's who the Scythians and Parni tribe were. OK, so they were probably Uralic, not Turkic or Mongolic -- but a non-Indo-Euro nomad group from the Steppe.
No Indo-Euro polity ever took over Iran, let alone from its eastern region. The only Indo-Euro group to defeat Iran was the brief run of Alexander the Great -- whose "empire" only lasted as long as he did. But despite not being a real empire, he did conquer the fragmenting Achaemenid Empire...
But he did so from the west, and from an origin south of the Black Sea, and not traveling northward around the Black Sea (i.e. taking a Steppe route to the east, then southward into eastern Iran), but staying south of it and the Caucasus and the Caspian Sea the whole time.
Even at their imperial peak, the Russian Empire didn't conquer Iran or Afghanistan, and they were Indo-Europeans who had by the Modern era reached the Urals and further east along the Steppe.
And the Greeks also did not serve as mercenaries in the Iranian military before rising to the top and taking over the whole society, as the Parni / Parthians did. They just blazed through and fucked everybody up as fast as they could.
Only the Turkic people paralleled the Parni / Parthians in the foreign mercenary to elite military rulers pipeline. Parni / Parthians were not Turkic (maybe culturally? but not linguistically -- there's that "initial p in Turkic" problem again). They were Uralic. But still, not Indo-Euro.
And Lord knows the Tocharians never swept down into eastern Iran -- although they should have given it a shot! They were Indo-Euro-phone, but culturally Uralic or other Eastern Steppe-like. They could have fit into the established pattern. But they were snugly nestled into the Tarim Basin and only wanted contact with Iranians in the form of hiring some fiery hot Persian baddies to be their musicians and dancers. They didn't feel like migrating outside of the Tarim Basin and conquering Iran.
I should watch the Arabian Nights adaptation by Doraemon, I'm getting in the mood! I already saw the other one focused on the Tarim Basin, the Doraemon version of Journey to the West. It's the one that got me hooked in the first place!
I watched the one about The Birth of Japan last night, but will wait until I see a few of them before reporting back / reviewing them.
Iran was also conquered by the Arabians in the 7th C AD. But that was not from the north and attacking eastern Iran -- it was from the south and attacking western Iran.
ReplyDeleteOnly non-Indo-Euros from the Steppe have ever invaded and conquered Iran from its eastern region.
So the Scythians and Parni and Massagetae were not Indo-European -- they were something else. If they weren't Mongolic or Turkic, that only leaves Uralic.
Now for the Massagetae, another "Scythian"-ish group, from just north of the eastern Iranian frontier. Their Queen Tomyris defeated Cyrus the Great.
ReplyDeleteAs usual, we'll ignore the Indo-Iranian etymologies, since they're Modern wishful thinking, whereas Dark Age writers said the Massagetae were like the Huns, Turks, and Tatars -- i.e., non-Indo-Euro people of the Steppe. That was correct.
I'm not going to do a full etymology, just look at the first / main meaning-bearing element that others point to, "massa" (the "-ge" and "-ta" are explained as suffixes, in Indo-Iranian). One very clueless II etymology says it comes from "fish", i.e. that they named themselves after being fishermen -- despite being mainly nomadic pastoralists, not fishermen.
Has any other group given themselves an endonym that means "fishermen"? Seems far less common, or non-existent, compared to endonyms relating to pastoralist livestock -- cows (Bavarians, Bohemians, etc.), horses (all those piebald horse clans I've discussed), fearsome predators maybe... but not fish or fishing or fishermen, lol. Especially not in a primarily land-based region, as opposed to living on the beach and primarily fishing and sailing.
A better II etymology is that "massa" comes from Young Avestan "maṣ̌a" meaning "men", a widely popular form of endonym.
But as with II "parHa" and Uralic tribal name "por", it turns out that the other Uralic tribal name, "moś", also has a plausible II etymology, cognate with Sanskrit "manuṣ̌" or "manus" meaning "man" as well as the personal name "Manu" for the primordial man (a la Adam from Adam & Eve). Discussed in the same review of II loans into Uralic as above, under "mansi" (with accents over the n and s).
This is also the basis for the Uralic endonyms Mansi and Magyar, and though from a different potential II donor, Mari as well (also meaning "man").
So we have at least 3 ethnic-level endonyms, and one of the two clans from the Khanty and Mansi, deriving from an II term meaning "man". And yet, just like "parHa" and "por", there doesn't seem to be a widespread usage of this II term for Indo-Iranian tribes -- unambiguous II tribes, that is, from Iran through India.
And just like Uralic "por" being related to the "par" element of Parni, Parthi, et al from the region north of eastern Iran -- I link Uralic "moś" to the "massa" element of "Massagetae", meaning "man".
Although ultimately of II origin, "massa" as the main part of an ethnonym arrived near the eastern Iranian frontier from a northern source, Uralic-speaking, who had borrowed it during an earlier period of contact with II (e.g. Sintashta), where it was presumably also borrowed by the other Uralic groups who still use it where they remained, on or just north of the Steppe.
I have no idea right now what the "getae" element(s) mean, just focusing on the main one. Nobody said that "Massagetae means 'the men, the people'" -- I'm just going with it because it works phonologically, and for symmetry with Uralic "por" being linked to Parni, Parthi et al.
That was the first thing I thought when reading about the "moś" II etymology -- OK, what so-called Scythian groups have something like that inside their name? Bingo -- the Massagetae.
Aside from language, the Massagetae are clearly not Indo-Euro but one of the Eastern Steppe peoples. They worshiped the sun god only, swore by only it, sacrificed horses to it. No mention of Sky Father or Earth Mother, or even a thunder god -- not Indo-Euro.
ReplyDeleteThe sun god being alone, without also being paired with the moon, seems to be a pre-Xiongnu thing. The Deerstones culture (Eastern Steppe, needless to say, not Indo-Euro) have solar discs at the top of their monumental gravestones, but only a few have the moon -- and it's a small disc off to the side of the sun, not a prominent crescent shape opening up below the sun, a la the Xiongnu symbol.
Tengri is also mainly a sun god -- spoken of as "the sky," but in the sense of a clear sky and sunlight or daylight pouring down. Not like Zeus.
Shinto is a mainly sun-focused religion, with some focus on the moon as well. The main goddess worshipped is the sun goddess, Amaterasu.
So the Massagetae belong to the religious tradition of the Eastern Steppe, not the Indo-European tradition. Just like how the Scythians demoted Sky Father to "Papaios" meaning "Papa" or "Gramps", ranking 3rd in the hierarchy, with the top spot being claimed by a female goddess of primordial fire.
There is mention of one component tribe of the Massagetae, the Derbikes, who worshipped Earth Mother according to Strabo. And their tribal name has two voiced obstruents, including one that never appears in Scythian-ish words -- "b" -- and with "d" in initial position, which was not allowed in Proto-Uralic. They're supposed to have been from closer to the Iranian frontier, like just south and east of the Caspian Sea -- not further east and north like the main Massagetae tribes.
As with all nomadic confederations, they were likely poly-ethnic, and this one group the Derbikes seem to be Indo-Euro (whether Iranian or otherwise). But for the most part, the Massagetae were a non-Indo-Euro people originally from the Steppe, probably Uralic, though again perhaps with a pre-Turkic tribe tagging along as well. Or they were culturally Turkic but Uralo-phone, IDK. The key point being, they were not Indo-European.
If Herodotus didn't mention any Scythian group jumping over fire for good luck during the New Year holiday -- they weren't Iranian, or even Indo-European. Simple as. Especially in the good ol' days, that was highly distinctive of Indo-Euros, especially the Iranian groups, and that remains true through today.
ReplyDeleteLast thought back to the Sarmatians. In that Gheorghiu article, he tries to derive "sarma" from Indo-Euro roots, having to do with fattening.
ReplyDeleteRegardless of what the meaning is supposed to be, there's no point in taking that approach -- cuz the supposed original PIE form has left almost no descendants.
He admits the Hittite and Indo-Aryan forms are borrowed, and the Greek form is borrowed from Thracian. It's not in Iranian, or Armenian, or Albanian, or seemingly in Celtic and Italic. And not in Tocharian (lemme check on that later, though).
It's mainly in Balto-Slavic and some Germanic (the Continental ones adjacent to Balto-Slavic, not the Norse ones). And perhaps in Thracian.
All of those can be explained by borrowing from Uralic, where it is widely attested. And in Uralic, it has existed far longer, enough for subtle sound changes to have affected its form across the various members. Whereas in Indo-Euro, it's nearly restricted to "sarma" or "(h)erma", without as much variation, and therefore having had less time to accumulate random mutations.
And that explains why it's sparse or absent from most of Germanic, Celtic, and Italic -- too far away from a contact zone with Uralic. Ditto for Armenian and Iranian, although it was in contact with II much earlier on (Sintashta), whose borrowing is preserved in Sanskrit "Sarama".
For it to have been born in PIE, then borrowed early into Uralic, and then loaned back into some IE languages in contact with Uralic, that assumes that the original form was killed off within the IE descendants, almost as early as it was coined! Why bother? It's not just an extra step, it's a contradictory step.
Also, most of the IE langs that borrowed it from Uralic have a taboo or reverential treatment of it, whereas it's totally mundane in Uralic, although perhaps reverential in some Uralic langs. In fact, it must have been borrowed in sacralized tone -- otherwise we have to assume it was borrowed with a mundane tone, and multiple IE langs chose to sacralize it independently of each other.
And that, furthermore, they chose to sacralize this loan-word for wild animal or wolf, rather than their own perfectly good IE word for wild animal or wolf. Greek had already subtly taboo-deformed the word for wolf (changing the "k" into "p/f"), but Balto-Slavic did not, nor did Indo-Iranian. So Slavic and Sanskrit could have easily sacralized their mundane term for wolf, but did not. We're supposed to believe they sacralized the Uralic loanword instead, independently of each other.
No way -- they all borrowed the "sarma" term from Uralic in an already reverential or special tone, whether it was a Thracian god, a Rig Vedic god, or taboo-deformed Slavic term for wolverine.
Since the Uralic Sarmations had already chosen to name themselves after this totem animal, it was already revered in at least some Uralic groups -- and from there, it was loaned into some IE langs in contact with it. Perhaps it was already revered earlier, when the Indo-Aryans were in contact with Uralic.
Point being, the Uralic origin makes a far simpler evolution and diffusion story. If its origin is not IE, then what does it mean in Uralic? Beats me, probably something like "wily one" or "cunning one", segmented as "sar/sor" and "ma". A Uralicist can fill in that detail.
Hmmm, looks like "sarma" might be in Tocharian as well. In Tocharian B, "sārme" means "bull" -- a wild animal, though not a wolf or weasel-type animal, but then some of the Uralic words go as far as birds.
ReplyDeletehttps://cetom.univie.ac.at/?F_B_s%C4%81rme
This is unlike "sārm", which has to do with seeds and planting, since this has the extra vowel at the end -- as it should, if derived from "sarma". Vowels look like a good match, too, though I know Tocharian shook up the vowels from Indo-Euro -- and in any case, this is borrowed from Uralic, not IE.
This is excellent independent confirmation of the NON-Indo-Euro origin of this term, since according to the IE origin, it should begin with "k" (with an accent), and in Tocharian that should remain "k". Tocharian did not satemize initial velars -- neither did Anatolian. That's why Gheorghiu admits that the Hittite word must be borrowed, not an internal evolution from IE origins.
Well, same applies to Tocharian -- it's a very old off-shoot, did not satemize the velars, so if its form were an internal evolution from PIE origins, it should begin with "k" -- but it also begins with "s".
Not only was Tocharian "in contact with" Uralic -- its speakers used to *be* Uralic speakers, and retained much of their culture after switching languages.
So it's very easy to imagine this being a Uralic loan in Tocharian, more like a carry-over from their previous native language, which was Samoyedic. It matches the Samoyedic form very well, as opposed to some of the other Uralic forms, which have a "ch" or "ts" or similar change to the "s", which have more deviant vowels from "sarma" and so on.
Tocharian is the furthest-east IE language by far, originating closer to the Altai or Sayan Mountains, and that's where Samoyedic was spoken, rather than other members of Uralic.
But in any case, the point is that "Sarmatian" is not IE in origin, and neither were the people who created that ethnonym for themselves, based on their totem animal (probably wolf or wolverine).
Brief remark about the Doraemon movie on The Birth of Japan, where the Tokyo gang (and a time-displaced hunter-gatherer from prehistoric China) travel back to the days when Japan was free from people, and they have the whole landscape to themselves, to live in hunter-gatherer bliss.
ReplyDeleteNaturally, trouble finds them, and one of the henchmen of the villain is an animated Dogu statue from the Jomon era of Japan:
https://en.wikipedia.org/wiki/Dog%C5%AB
During one confrontation, the Tokyo gang breaks the statue into pieces, and walk away assuming that's the end of the henchman.
Well, after awhile the statue's broken fragments piece themselves back together again! And one of the Tokyo gang had brought a piece of the statue with them -- when they break it into tiny pieces, eventually they fit themselves back together, which clues them into the main statue having revived itself and still posing a danger.
Aside from being a cool piece of ancient Japanese art history in a mainstream movie, this makes me think that somewhere in the minds of modern culture-makers in Japan, there still resides some stories that were passed along from one generation to the next, about a valuable craftwork that is shattered and left for dead... only for it to come back to live by seamlessly fitting the pieces back together.
Putting Humpty Dumpty back together again.
It makes me suspect there was a similar idea behind the practice of breaking the bronze Chinese mirrors during ancient times. They expected them to re-assemble themselves at some point in the future, which would represent the resurrection of the dead. And until that happened, if they were still lying in fragments -- that meant that resurrection day was still in the future, and a reminder that it would come sometime.
Dogu statues are also the basis for a Pokemon, Claydol / Nendoru:
ReplyDeletehttps://bulbapedia.bulbagarden.net/wiki/Claydol_(Pok%C3%A9mon)
Glorious Nippon keeping their children familiar with their ancient history!
Eastern Steppe solar cult pattern du jour -- using the same word for "sun" and "day".
ReplyDeleteIt extends as far east as Glorious Nippon, where Proto-Japonic uses "pi" to mean both, and so does modern Japanese (also present in Queen Pimiko / Himiko's name -- "Sun Child" or "Sun Girl").
Korean is a little muddy, but still suggestive. "Hae" means "sun", "daytime", and "year", though not as a daily unit of time -- that's "nal", which however might be related to Mongolian "nar" which means "sun".
Strangely, none of the Mongolic languages take part in this trend.
Nor does Tungusic.
Good ol' Yeniseian does, though -- "xʷaj" means "sun", "day", and "maternalistic sun goddess" in the proto language. Very reminiscent of Japanese, with the addition of the sun goddess, who appears in the context of Yeniseian shamanism, like Amaterasu in Shinto. Maybe the Wa people started as Yeniseian, who were also part of the Xiongnu confederation. While in that mix of cultures, they picked up some other traits from the Altaic groups, and came out the other side as a unique mix of Yeniseian and Altaic, before absorbing the Emishi and becoming something different yet again.
In Proto-Turkic, "kün" means both. This was borrowed into Tocharian, otherwise an Indo-Euro language but culturally Eastern Steppe (former Samoyedic speakers), where "kaum" or "kom" are used for both "sun" and "day".
In Proto-Uralic, "päjwä" is used for both among the branches of Proto-Samic and Proto-Finnic. A different term "keččä" is used for both among the branches of Proto-Mari and Proto-Mordvinic. Hungarian alone uses "nap" of unknown origin, but it's still for both "sun" and "day". Seems like only Samoyedic does not participate in this trend, similar to Mongolic and Tungusic.
So, it's not a strictly Altaic thing -- in fact, 2 of the 3 main members of that club do not participate, only Turkic does. But it does stretch from Scandinavia to Glorious Nippon.
It does not cross over Beringia into the Na-Dene / Athabaskan off-shoot of Yeniseian, nor is it present in Yukaghir, nor in Chukotko-Kamchatkan, nor in Nivkh, nor in Emishi-an (Ainu). So it definitely has to do with Steppe origins, regardless of where they ended up traveling to later.
If the origins are further north toward the Arctic, they don't equate the two -- perhaps regardless of later settling into the Steppe, like if Tungusic and Mongolic speakers originally hailed further to the north than the Steppe itself. Samoyedic is pretty far north as well, although they've had a Steppe presence as well, near the Sayan Mountains. Overall, though, looks like good fit.
And it doesn't happen in Proto-Indo-European, which was born on the Steppe -- but in the West, and this is an Eastern thing, i.e. beginning around the Urals, where Uralic is from. Indo-Euros were not part of a solar cult either, whether religious / ritualistic practice or their mythology.
Only Eastern cultures would put the sun itself on their national flag, and proudly name their home "Land of the Rising Sun". ^_^
BTW, modern Japanese uses "hi" instead of "pi" (sound change over time), but it's for both "sun" and "day".
ReplyDeletePerhaps Pimiko's name points to an even deeper Yeniseian connection -- did the Japanese, or their Continental ancestors, also refer to a sun goddess with the same word as "sun" and "day"?
ReplyDeleteHer name is a bit more complicated than simply "sun daughter / child":
https://en.wikipedia.org/wiki/Himiko#Name
The Chinese scribes put the same vowels for the 1st and 2nd syllables, which seems to argue against reading them as "pime", which originally meant "noblewoman" (and derived as "sun + daughter" -- "-me" being the same suffix as in "musume" meaning "daughter", which you may know from that video game everyone is playing, Uma Musume, meaning horse daughter / girl).
Later "pime" was elevated to mean "princess", and after a p -> h sound change, became the current "hime". Then the "ko" was added to mean "child", although originally "ko" was used strictly for male children, while "me" was the suffix for daughters.
These problems can be avoided by suggesting that Pimiko segmented as "pi + miko", where "miko" is still in use as a female shaman or shrine priestess in Shinto (as in the name of Hololive's Miko). So, Pimiko would mean "sun priestess".
But, priestesses are usually referred to by the name of the deity who they serve, not a natural feature like the sun. The miko serve the sun not as a natural object, but the "sun as a goddess".
Well, that suggests that the name of the god(dess) that Pimiko was serving, was "Pi" -- i.e., the same word as "sun" and "day". So there's all three concepts in a single sound sequence -- just like in Yeniseian.
The Chinese scribes say that Pimiko used magic or sorcery to sway the population, so she was definitely some kind of shaman or priestess. Probably more of a priestess -- shamans don't rise to a high rank in a social pyramid, they're more folk-level roles. But a priest or priestess is part of a religious hierarchy, and that kind of role is more likely to catapult someone into being a political and military leader of a large unified society.
So, which god did Pimiko serve as a priestess (as a "miko")? "Pi" -- the sun-goddess, who bore the same name as the sun-as-natural-object and as the word for day.
That was in the 3rd C AD. Later, this sun goddess would acquire the name Amaterasu, among others, in the Kojiki and Nihon Shoki, written in the 8th C.
However, in the Nihon Shoki she is also referred to by the more simple and likely older term "pi no kami" -- sun god(dess). Perhaps even earlier, during the era of Pimiko and further back, when the Wa people were still trekking through Eastern Asia, this sun goddess was called even more simply "Pi", as suggested by the later name "Pimiko".
Yeniseian-Japanese connection looks even more likely! They didn't use the same sequence of sounds across both languages, of course, but both cultures used the same sounds within their own language to refer to the sun, a day, and a maternalistic sun goddess.
Kakkoi! ^_^
So I'll correct my view that Amaterasu being female reflected the Emishi-an influence on Yamato culture. I thought the Wa's sun god would have been male, a la Tengri. But he wasn't the only solar deity on the Eastern Steppe -- Yeniseian speakers focused on a sun godd-ESS, "xʷaj", who bore the same name as "sun" and "day".
ReplyDeleteSo perhaps the Wa people brought this female sun goddess, not Tengri, with them from the Steppe.
I know it's a stretch, and that Japonic languages are a synthesis between an Eastern Steppe language and an Emishi-an filter, but I'll just speculate that the two sun goddesses may share the same name after all.
Both are mono-syllabic, with an open syllable, and the "aj" from Yeniseian could have become a simple "i" vowel in Japonic -- which did not allow vowel sequences, and generally kept the 2nd vowel.
The Yeniseian consonant is a fricative, and Japonic did not allow these, so "x" would have become "k" -- which is still not "p". "K" is a velar consonant, in the back of the mouth, while "p" is in the front of the mouth.
But the Yeniseian consonant also has labialization -- which could have glommed onto the "k" in Japonic and turned it into a "p", which is a labial consonant.
Labialization could also decompose into a "kw" sequence, but this was not allowed in the early stages of Japonic -- not until some time in Middle Japanese, well after Pimiko's era.
So, what was the only phonotactically permissible way for Japonic in that era to preserve labialization? Make it a labial consonant -- either "m" or "p" was available, and since the target "x" is not tooo sonorant, they chose the stop "p" instead of the highly sonorant "m".
"P" also preserves the voicing of the Yeniseian "x", whereas "m" would alter it into a voiced consonant.
Yeah yeah yeah, it's speculative, but a lot of Japanese words have uncertain etymology. And since Yeniseian and Japanese match the cultural practice of "same sounds meaning sun, day, maternal sun goddess," then maybe those sound sequences used to be the same as well -- or descend from a common ancestor, one of which became "pi" and the other became "xʷaj".
Neat!
On a big-picture level, this is a great example of how difficult it is to look back into the history of a language that has been put through a massive filter, which the Wa people's language was when they absorbed the Emishi-an speakers.
ReplyDeleteIf there was voicing, aspiration, labialization, etc. -- gone, gone, gone, after it had to change to be acquired by Emishi-an speakers.
This is the problem of interpolation vs. extrapolation -- former is valid, latter is invalid.
Just cuz you know that a target language was put through the filter of another language, yielding a synthesis language -- you can't peer from the evidence in the synthesis and in the filter, to hazard a guess about the target. You would need further external evidence, like "What if the target were Yeniseian"?
Maybe the Wa language lacked voicing and aspiration, just like Emishi-an and Japonic. But maybe it had one or both of those features -- you can't tell, based on Japonic and Emishi-an both lacking them.
Perhaps the Wa language had voicing, and the Emishi-an filter removed it. Or perhaps Wa lacked voicing even before it was put through the Emishi-an filter. There's no way to know, it's extrapolation, and therefore invalid.
But if you have independent reasons to suspect what the target language was, then you can ask what would happen if you put it through a known filter -- would that yield the known synthesis?
Just like with Tocharian -- an Indo-Euro target put through a massive Uralic filter, to yield Tocharian.
If you only had evidence from Tocharian, could you reconstruct a picture of Indo-European? No way! The Uralic filter removed all of the voicing and aspiration of the Indo-Euro target, scrambled up the vowel system, changed the morphology to agglutinative for case "inflection", and added several wacko cases that were not in Indo-Euro.
Only if you already knew some other Indo-Euro language, you could say, "I wonder if an Indo-Euro target were passed through a Uralic filter -- would it look like Tocharian?"
But just having Tocharian and Uralic evidence, you would have very little idea what Indo-European looked like.
For Japonic, we have a variety of plausible target languages -- Turkic, Mongolic, Tungusic, Yeniseian, etc., and we know exactly what the filter was (Emishi-an, whose best guess is the reconstruction of Proto-Ainu).
So far, though, no one is pursuing the model of "Start with a Turkic / Yeniseian / Etc. target, pass it through the Proto-Ainu filter, and see if you get close to Japonic's early stage".
But that's the way to do it! The Wa people's language was not Bantu or Indo-European or Mayan or Martian -- it was from the Eastern Steppe.
And not Sino-Tibetan or Austronesian! It's insane how much brainpower has been wasted on the insane and ridiculous idea that the Japanese came from a Chinese or Southeast Asian background -- not 40,000 years ago, but just 2000 years ago. It's crazy!
Hopefully I can provide a little push in the proper direction...
I'll get back to all this nonsense about Scythians, Sarmatians, etc. later, in a separate post. There is so much craziness to wipe out there, and start almost from scratch. So I'll stop writing updates on that topic within this comment section for now.
ReplyDeleteSuffice it to say, little to no Iranian component to Scythians, Sarmatians, etc. Not even a broad vague Indo-European-ness. They're clearly from the Uralic realm or further east on the Steppe.
But so far, the main corrections to the "Iranian-Scythian" idiocy has been a Turkic view -- which is way more plausible than Indo-Euro, but for the 1st millennium BC in the Western Steppe, I'm more inclined to Uralic, and will develop that approach further in a new post.
OK, one last wrapping up of some loose ends from the "Uralic-Scythian" discussion. First, on Indo-Iranian loans into Uralic, which became Uralic ethnonyms but rarely / never Indo-Iranian ethnonyms, the same II root that gave "Mari" (meaning "mortal" in the sense of "man, who is defined by his mortality"), also can explain the "murd/t" element in "Mordvin" and "Udmurt".
ReplyDeleteThat bolsters the view that people with such elements in their ethnonym are culturally and linguistically Uralic, not Indo-European -- they merely borrowed this Indo-Iranian term awhile ago (likely when Uralic was in contact with / absorbing people from the I-I Sintashta culture). Indo-Iranians themselves don't seem to have used these elements for their own ethnonyms, whether in ancient times or the present.
I already looked at the "massa" element in "Massagetae", saying it's the "mos" clan from Uralic, still present in Mansi and Khanty today. I left the "getae" for later -- but it turns out, that's an all-purpose suffix, from the standalone term "Getae", which originally referred to various peoples from the northern Black Sea area.
Sometimes it was just Getae, sometimes with a specifier in front like "Massagetae" -- so "Massagetae" just means "the mos clan among the Getae" as opposed to some other member of the Getae.
Well, there's another one called the "Myrgetae" -- and there's that same I-I term meaning mortal that shows up in "Mari", "Mordvin", and "Udmurt". Roughly, this means "the myr clan among the Getae", with "Myr" being cognate with "Mari", all of whose ethnonyms mean "man" (and it doesn't matter that the ultimate source of the "man" term was Indo-Iranian, it was appropriated for ethnonyms by Uralic cultures, not Indo-Iranian ones).
"Getae" itself is likely not Uralic, since it has a velar consonant in initial position, and that was not allowed in Proto-Uralic. But since it became a catch-all term for Black Sea nomads, it could serve as the generic element of a compound name, whose specifier would be the real tribal ethnonym -- Massa (from mos), Myr (from mor), and so on.
All right, one final insight on the topic of Uralic, Turkic, filters, etc. I found a counterpart to the Tocharians, where a Uralic culture acquired an outsider language, put it through their own Uralic filter, and turned it into an almost unrecognizable version of the outsider language family.
ReplyDeleteThe Chuvash! They're still going 1 million strong in the Volga-Ural region of Russia.
The consonant system (no voicing, plain vs. palatalized coronal series, no fricatives), 3 bizarre cases in their grammar (instrumental, causative, and abessive), and not having roundedness vowel harmony (only front-back), clearly identify them as Uralic -- but their language is otherwise Turkic. Specifically, the rare and old Oghur branch of Turkic, whereas almost all other Turkic languages spoken for a long time have been from Common Turkic, the other major branch.
The only other viable candidate for the Oghur branch is Bulgar, which has been extinct for awhile. Yes, the language spoken by the Bulgars -- before they adopted the Slavic language of the majority they served as the military elite for, once they settled down in Thrace and became the Bulgarian Empire.
So what happened was, in the mid-1st millennium AD, the Bulgars migrated westward across the Steppe. They encountered a Uralic culture, who tried to adopt the Turkic language of the Bulgars. But they put it through their own Uralic filter, and the outcome was the Chuvash synthesis language -- no voicing, few / no fricatives, plain vs. palatalized coronal stops, only front-back vowel harmony (not roundedness as well as in Turkic), adding the 3 wacko cases from Uralic, etc.
Not that genes matter, since genetic variants do not carry cultural variants -- but the DNA profile of the Chuvash is Uralic, not very Turkic / East Asian.
The filter language was not Slavic, cuz Slavs, being Indo-Euro, LOVE voiced obstruents. The ancestors of the Chuvash were not Turkic, but they weren't Indo-Euro either -- they were Uralic. And they were culturally powerful enough to resist demographic replacement by the Turkic migrations, and to impose their own massive filter on the target Turkic language.
That strong Uralic presence throughout the Western Steppe is probably the most under-rated / overlooked factor in studying the history of Eurasia. Everyone thinks of Uralic as being Arctic forest hunters, and some of them became that -- or imposed their language on the Arctic forest hunters who they encountered who used to speak some non-Uralic Paleosiberian language. But they were originally from the Ural region of the Steppe, and they were decidedly Eastern rather than Western.
They came into contact with Proto-Indo-Iranians from the Sintashta culture circa 2000 BC, but that's it. And nobody there has ever adopted an I-I language, even if some I-I genes have made their way into the present genepool (like the Mari -- still largely Uralic, culturally, lingustically, and genetically, although with some I-I genetic inflow, but without I-I linguistic imposition or cultural replacement).
If Uralic cultures were that widespread, influential, and resistant, then they could have easily been major contributors or the leaders of the Scythian / Sarmatian world.
And to offer an example of Indo-Euros adopting Turkic in "Europe", there's the Tatars from Tatarstan. Their consonant system etc. is clearly from Slavic -- it has voiced consonants, but it has a crazy number of them, just like Russian does, not just the handful of voiced consonants that the Turkic target had.
ReplyDeleteAnd the Tatar case system does not have the 3 wacko cases that Chuvash imported from their Uralic background (instrumental, causative, abessive). Indo-Euro cases don't include those, so the former Russian speakers who ended up adopting Turkic, did not bring the 3 wacko cases with them.
If you hear present-day Tatar speech, it is heavily palatalized, just like Russian. Not just cuz "they're bilingual in Tatar and Russian" -- they used to be Russian speakers as recently as the 13th C, and they gradually adopted a Turkic target during the Golden Horde era, which they have held onto ever since -- but after putting it through a Slavic (Russian) filter.
And their Turkic target was Common Turkic, not Oghur like it was for the Chuvash.
The Chuvash were/are Uralic, and adopted Turkic in the mid-1st millennium AD. The Tatars were/are Indo-European, and adopted Turkic 1000 years later, when all Turkic languages were Common Turkic.
So don't let their "neighboring Turkic speakers" status fool you -- they are very different cultures with very different ethnogenetic histories.
OK, there's more to this Yeniseian-Japanese connection. I'm not going through systematically for now, just planting a few more little sign-posts.
ReplyDeleteI said the labialized "x" would turn into a "p", so let's try another one to see if there's a correspondence.
Proto-Yeniseian "xʷaŋ", meaning "nose".
Labialized "x" becomes "p".
Japonic only had "m" and "n" nasals, not the velar, so that changes to "n" (least movement -- only has to move to alveolar place, not further to labial).
Japonic did not allow coda consonants until Middle Japanese, so the final nasal must be followed by a vowel to make it part of an open syllable. Simplest one to insert, based on vowel harmony or reduplication, is the same vowel as on the other side -- "a".
So then it should result as "pana" -- BINGO! That's exactly the Proto-Japonic word for "nose".
I'm having trouble with the other P-Y word that begins with the labialized "x", so set that aside for now. It may work, but it would be more complicated / depends on which P-Y reconstruction to use (there are dozens of them for "xʷag" meaning "tree, wood", which is "kəi" in P-J, but there are other P-Y forms that could yield that, with more steps, though.
So onto another one -- P-Y "cam" meaning "goose", where "c" is the voiceless palatal stop.
ReplyDeleteP-J doesn't have palatal anything, so move it back to velar, while keeping it's stop manner and voicelessness -- "k".
Again, no final nasals, so insert the same vowel after it.
This yields expected "kama" -- but is actually "kamo" in Japanese, and it means "duck" rather than "goose". Pretty close, though!
And check this out, Wiktionary's etymology of the Japanese "kamo":
"Unknown. Possibly alteration of Old Japanese 浮かむ (ukamu, “to float”), modern 浮かぶ (ukabu). Alternatively, may derive from Middle Chinese 雁 (MC ngaenH, “a type of wild goose”), borrowed with the kan'on reading gan, with initial g changing to k due to the avoidance of voiced stops at the start of Japanese words in older times, and final n changing to mu due to apparent lack of final n in Old Japanese."
Maybe... but my etymology only has one little unexpected deviation -- predicted final "a" but in reality final "o". But at least that doesn't violate Arisaka's Law of vowel harmony -- "a", "u", and "o2" in the root formed a class. Maybe on further investigation, other laws would show why the prediction is actually "kamo". E.g., maybe inserted dummy vowels absorb the rounding of the preceding "m", which would change "a" into "o" or "u", and to keep the height as close as possible to the original, "o" wins over "u"). IDK.
As for the slight semantic mismatch in mine, it also exists in the one at Wiktionary, where "ukamu" only has to do with flight, not specifically birds let alone ducks. And the Chinese borrowing is also a word for goose rather than duck. So mine is no worse on this score either!
Last one for now, not perfect, but just to get a better feel for what's there.
ReplyDeleteP-Y "cajb" meaning "dog". Well, no way that can turn into "enu" meaning "dog" in P-J. But it's not as if Japonic inherited 100% of its lexicon from Yeniseian. Maybe this P-Y root showed up in a closely related context...
To aim for regular correspondences, I already said "c" should become "k".
And "aj" should become just "i" (P-J doesn't allow coda consonants, and an approximant "j" is changed into an "i" vowel instead, and vowel sequences are not preferred or allowed, so as before it keeps only the 2nd vowel).
So far, that gives an expected Japanese word beginning with "ki-" that relates to dogs. "Kitsune"! That means "fox", but again, close enough.
As usual, Wiktionary says the etymology of Jap. "kitsune" is unclear:
"The most likely [theory] is based on the root form kitsu, which may have originally been onomatopoeic for the sound of a fox's cry.[1] The final ne syllable appeared for certain by the Heian period,[2] but its meaning remains unclear."
Now, "tu" (later palatalized into "tsu") was the Old Japanese possessive particle, similar to "no". And "ne" can mean "sound, voice" or specifically "the cry of an animal", and this is a Japonic origin, not Chinese borrowing. See etymology 4:
https://en.wiktionary.org/wiki/%E9%9F%B3#Japanese
So, Old Japanese "ki + tu + ne" (later, "kitsune") would mean "the cry of a dog", an onomatpoeia, which then became the name for the animal making this sound -- but since Japonic already had a different word for "dog", they used this term for another dog-like animal, the fox.
The only problem is what happened to that final labial stop from P-Y? First, it wouldn't be "b", it would be "p" since there's no voiced obstruents in P-J.
Perhaps the morphology was "kip + tu + ne", and since P-J didn't allow coda consonants, it opted to simply delete the "p" rather than insert a vowel after it to break up the cluster. That's not what happened with word-final consonants -- but this is word-medial, so keeping it simple meant deleting an offending member of a cluster rather than inserting a vowel and making a longer word.
Then I'd have to assume that the "p" was deleted rather than the "t" -- either because the 1st one was deleted rather than the 2nd one, or consonants from particles and other bound morphemes like "tu" were not allowed to be deleted, while final ones from basic nouns were...
If you delete the "t" from "tu", it makes it very difficult to see what is meant by "kipune". Deleting the final "p" of "kip" preserves the derivational morphology, so you know how to segment it and interpret it -- "kitune" has "tu" in the middle, so it's probably that very common possessive particle, and must segment as "ki + tu + ne", meaning "the cry of an animal called 'ki'".
When you're first switching languages, the presence vs. absence of your old words is not 100% either way, they're floating around to varying degrees. And more frequently for common words like "dog". So maybe the animal beginning with "ki-" could have been filled in from familiarity -- "Oh yeah, I think they mean an animal called 'kip', i.e. a dog, not some other animal beginning with 'ki-'".
And now for a little leisure, watching another Doraemon movie. ^_^ I think I'll watch the one about The Diary on the Creation of the World...
ReplyDeleteAlso a reminder that Yeniseian languages used to be spoken in the Altai-Sayan mountain area, in northern Mongolia, whereas now they've migrated further north after being displaced by Altaic speakers, much like Samoyedic speakers have been by the same people.
Doesn't the Sayan Mountain landscape look just like the Edenic paradise in Japanese animes? It's uncanny! They know where they came from, and yearn to RETVRN...
https://en.wikipedia.org/wiki/Sayan_Mountains
I should say that the former Yeniseian speakers from the Sayan region didn't necessarily migrate northward after getting displaced by Altaic speakers. Somewhat, yes.
ReplyDeleteBut more likely, they stayed put and switched to an Altaic language as Altaic cultures became more, shall we say, influential in the Eastern Steppe.
One group of these "former Yeniseian speakers, who recently switched to an Altaic target", belonged to the Xiongnu as many of them did, and then broke off toward the east for greener pastures (over-crowded niche on the Steppe, crumbling cohesion among the late Xiongnu, something else), entered Korea, and then Japan, where they absorbed the Emishi-an speakers.
One more, looks promising but hard to do sound correspondences since there are so few consonants, and monosyllables. Just something to look into, though. Pronouns -- very unlikely to be borrowed.
ReplyDeleteP-Y "aw" for 2nd person singular pronoun
P-J "ə" ("o2") ditto
P-Y "axʷ" for 1st person singular pronoun
P-J "a" for 1st person singular pronoun
NB: P-Y has "a" as a base for both the 1st or 2nd pronoun to build on top of, and perhaps this base was inherited as the 1st person pronoun alone, in P-J, without the productive morphology to build both the 1st and 2nd person pronouns on top of it.
P-Y "wu" for 3rd person pronoun (sing or plural)
P-J has no 3rd person pronouns, it uses distal demonstratives...
However, a descendant of Eastern Old Japanese (which went extinct, as Western OJ became the standard), Hachijo, has "u-" for its distal demonstrative, and that may also be related to "ʔo", the mesial demonstrative from Proto-Ryukyuan.
Where did the "w" go in "wu"? Simple -- Japanese has never had the "wu" mora ever, and it is one of the missing entries in their syllabaries. Attempts to pronounce "wu" (e.g., from English "wood" or "would") are simply rendered as "u".
I think we're on to something here!
But now it's time to watch Doraemon for real. This stuff is just too fascinating, though! ^_^
Epiphany on Japanese language history! Speakers of Japonic used to speak a different language, something Altaic, and even before that they seem to have been Yeniseian speakers. Then they have to assimilate the Emishi-an speakers when they reach southern Korea and Japan.
ReplyDeleteThe Emishi-an filter removed voiced obstruents -- Emishi-an does not have a voiced vs. voiceless series of consonants, it's based on plain vs. palatalized (Alonso de la Fuente). So, suddenly, the lexicon of the Continental majority gets stripped of its voiced obstruents -- leading to a potential homophone problem from hell.
Their goal is to preserve the lost information in some other way, ideally with the tools they have available -- and if not, invent something new.
If the voiced obstruent was word-medial or word-final (there were word-final voiced obstruents in whatever the pre-shift language was -- Yeniseian, Turkic, Mongolic, or Tungusic), then pre-nasalize the voiceless consonant that remains after its voicing is removed.
That strategy is well known -- it must have begun in the Proto-Japonic stage, then by Old Japanese the nasal had already spread its voicing onto the following consonant, so they had voiced obstruents again, but still pre-nasalized. Then the pre-nasal dropped out, and they simply returned to having voiced obstruents word-medially (some reflecting original word-medial position, some reflecting original word-final position, with a dummy prosthetic vowel stuck on the end to satisfy the new Japonic phonotactics that don't allow for coda consonants).
That avoids the homophone problem -- there was still a distinction between the two original consonants, with original voiceless remaining the same, and original voiced now becoming voiceless but with pre-nasalization.
But that's only for word-medial or word-final position -- they could not, and did not, pursue this strategy for word-initial position. Pre-nasalization needs some airflow to get it going, otherwise it's really going to break the phonotactics that prefer a single consonant in the onset. Word-medially, that's fine. But word-initially, the plan failed.
OK, so plan A fails -- time to try something different for word-initial. Time for plan B!
No one's ever thought of this!
Word-initially, the original voiced obstruent becomes voiceless, and no pre-nasalization is possible -- so you've pretty much run out of options for re-encoding the lost information onto the consonant. Well then, re-encode it onto the following vowel!
ReplyDeleteThis is similar to the tonogenesis of the Southeast Asian languages, including Chinese but various other families as well. They lost coda consonants, and chose to re-encode that lost information onto the preceding vowel that remained, in the form of tone. Khmer did it by introducing diphthongs instead -- as I mentioned before, when they were undergoing intense ethnogenesis as they were becoming the Khmer Empire, and didn't want to sound like the sub-imperial cultures around them.
But in either case, they re-encoded the info from the lost coda consonants onto the preceding vowel.
In the Japonic case, they're losing info about initial consonants -- so they should have re-encoded it onto the following vowel.
They did not use tones, and I don't think the pitch accent is involved, cuz there are different pitch accents throughout Japan, and it doesn't lead to a homophone problem, AFAIK.
They did not alter a single vowel into a diphthong, as Khmer did. If anything, diphthongs in Proto-Japonic are close to the end of the word, not in the first syllable. And they're pretty rare, not systematic, as would be required if they were part of a plan to re-encode lost info.
So perhaps they changed the vowel altogether -- probably kept it in the same vowel harmony class, so it wouldn't sound toooo different, but just changed the vowel.
E.g., if "gako" wanted to become "kako", and "kako" is not already a word, then OK, change "gako" to "kako".
ReplyDeleteBut quite frequently, the new sequence will already be taken by a homophone.
So instead of "gako" becoming "kako", change the following vowel as well -- say, "koko". Now the problem is avoided -- "kako" and "koko" are different sounds, and preserve the distinct meanings of the originals, despite not having voiced obstruents in word-initial position. That lost info has been re-encoded onto the following vowel.
Well, what if "koko" is also already taken? Then alter the next vowel as well! From "gako", try "kako" first. If that's already taken, try "koko". If that is also taken, try "koka". Eventually, there's a free spot -- not every possible combination of syllables is already a word in the lexicon. There will be a phonotactically legal opening that has no meaning associated with it, and bingo, there's your new word -- distinct sounds for distinct meanings, despite having lost an entire class of consonants.
Now, if the language is monosyllabic, this strategy will not work well, cuz most of the other vowels that you could alter the target to, will be taken. No matter how much you move around the vowel space, it's pretty full.
That's why the largely monosyllabic languages of SEA didn't pursue this strategy, but chose an entirely new feature instead -- like tone.
In Chinese, if "ma" is already taken, and you try to change it to "mo", that is probably also taken, and so is "mi" and "mu" and "me". So, go with tone instead -- if you invent a whole new distinction, there won't be any taken spaces, cuz nothing had tone on it before the invention of tone.
But in Japonic, which is heavily polysyllabic, they're less likely to run into a bunch of already taken spaces, at the level of a word. At the level of a syllable, yes, but since their words have lots of syllables, this isn't a problem at word-level, and that's the only level that matters for the homophone problem.
As with saw with "gako" -- if it were just "ga", then just altering the following vowel won't do much. It would be like Chinese. But "gako" has 2 syllables, and that's the start for Japanese, sometimes they're 3 or 4 syllables. Each syllable *multiplies*, not adds, to the number of potential words.
So they're free to go from "gako" to "kako" to "koko" to "koka" -- sooner than later, they'll find an open spot in the lexicon.
Even more so for 3-syllable words. If the target is "gakoko", they can problem just go with the most obvious change, "kakoko". If that's taken, they alter each vowel in sequence until they find an open spot -- and they have not 1, not 2, but 3 vowels to search for an open spot.
If there are 5 vowels, a 1-syllable word only has 5 choices, for a given initial consonant. But a 2-syllable word has 25 combinations of vowels (keeping the consonants the same across these combinations), and a 3-syllable word has 125 combinations of vowels (again, without changing the consonants).
There's no way all 125 slots are already taken! Even for the 2-syllable word, no way that all 25 slots are already taken! But for a 1-syllable word, it could well be that the other 4 slots are already taken.
I have no idea how this actually played out in the history of Japanese. But it absolutely must have happened this way -- otherwise there would have been the homophone problem from hell affecting initial consonants. That's partly alleviated by the polysyllabic length of their words, but they still do have monosyllabic words.
ReplyDeleteI'll bet that's one reason why Japanese has so many unknown etymologies, and why it's hard to connect it to other languages -- the vowel after an initial consonant subject to de-voicing would have been altered in a way that is not so noticeable, as if it were subject to tonogenesis.
They were just cycling through various options for the first vowel, then the 2nd vowel if that didn't work, then the 3rd vowel if that didn't work. It really jumbles up the connection between the vowels of the pre-shift language and the post-shift language.
But if we knew what language they used to speak, we would be better able to see what the pre-shift target was, which they were trying to preserve, and see how the following vowel was altered when an initial consonant was de-voiced.
Maybe that alteration was regular, predictable, etc. -- and maybe it wasn't, subject only to the random contingent facts of which slots were already taken vs. available. That makes the vowel alteration *not* systematic, *not* a simple law like Grimm's Law or whatever, and you can only tell case-by-case how the vowels were altered between the pre-shift and the post-shift languages.
So, unlike a law that you can invert and see what the language looked like before a change, this case-by-case approach cannot be simply inverted at a language-wide level in order to see what those words all looked like before the change. The changes were all ad hoc, not systematic or rule-governed -- very loosely rule-governed, at any rate.
I only noticed this possibility when trying to go from Yeniseian targets to Japonic outcomes. What if there's an initial voiced obstruent? Well, devoice it -- OK, but that creates a homophone problem. They can't pre-nasalize it, so... maybe they altered the following vowel!
ReplyDeleteYou're not just looking segment by segment, but potentially syllable by syllable, since a whole bunch of lost information is being re-encoded onto adjacent segments. Just like tonogenesis in SEA.
There's little chance I'll figure it all out -- but at least I've discovered the right path to go down, whereas previously nobody saw this possibility, and just said, "Well, we will never know the etymology of many Japanese words".
Bullshit! You need to consider that the initial *syllable* could be the site of language-wide re-encoding.
And you need to know where to look for the lexicon they were trying to preserve -- not Sino-Tibetan, not Austronesian or anywhere in SEA. Such a ridiculous waste of time trying to portray Glorious Nippon as just another "Pacific island". It's the furthest extension of the Eastern Steppe -- so far to the east that they entered the Pacific Ocean!
Yeniseian, Altaic, etc. That's where to look.
Back to the personal pronouns in Japanese being Yeniseian in origin, are there other possible candidates for where they could be borrowed from? No, I checked.
ReplyDeleteThe core Altaic languages all have the same pronouns for 1st person, 2nd person, and the distal demonstrative for 3rd person. They also share the paradigm that the plural is based off of the singular for 1st and 2nd. That speaks to their shared geneaological origin -- no way you borrowed all those pronouns from one another. Pronouns generally are not borrowed, let alone in their entirety.
And they're all different from the Yeniseian and Japonic pronouns. Emishi-an (Ainu), and Nivkh also have their own set of pronouns, which don't overlap with anyone else's (other than Nivkh having borrowed the 2nd-person singular from Altaic).
So, there's no neighboring family that Japonic could have borrowed their pronouns from -- the only possibility is Yeniseian. I showed how strong that connection is, but left out the demonstration that there are no other contenders. It's only Yeniseian.
What that means is that at some point, speakers of Yeniseian switched to an Altaic language, but kept their pronouns. Later, they incorporated the Emishi-an speakers, and the synthesis was Japonic -- but still keeping those original Yeniseian pronouns.
Pronouns are way more sticky than other categories of words. They can survive multiple language shifts. Or lesser situations, like heavy influx of foreign vocabulary -- all that Arabic vocabulary flowing into Iranian languages after their adoption of Islam, and they still have their original Iranian pronouns, not Arabic ones.
How do we know the shift wasn't directly from Yeniseian to Japonic? From morphology and syntax. Yeniseian is very unusual for the Eastern Steppe, it looks more like a Pacific Rim language, like Emishi-an (Ainu) or Nivkh or Eskimo, etc.
ReplyDeleteYeniseian is polysynthetic, not agglutinative, and it is head-marking rather than dependent-marking. Just like Nivkh, and mostly like Ainu (Ainu being subject to Japonic influence -- it's ancestor in the 1st millennium BC was probably very head-marking).
Head-marking is a very conservative feature in languages, so if both sources of Japonic were head-marking -- Yeniseian and Emishi-an -- then its synthesis would likely be so as well. Ditto for polysynthetic vs. agglutinative.
And yet, Japonic is one of the most consistently agglutinative and dependent-marking languages on Earth.
Which other languages nearby are like that? The Altaic ones, regardless of Turkic or Mongolic or Tungusic.
So, the Continental ancestors of Japonic speakers used to be Yeniseian speakers, as judged by highly conserved lexical items like personal pronouns, but they had also clearly shifted to an agglutinative and dependent-marking language in the meantime -- an Altaic one.
Head-marking is a more unusual, marked choice around the world, so when there was a clash between head-marking Emishi-an and dependent-marking Altaic, the Altaic side won, and Japonic became dependent-marking.
Also, agglutinative is more transparent than polysynthetic when there are a bunch of L2 learners to worry about, so agglutinative won over polysynthetic, also favoring the Altaic source.
This is independently supported by the material and mythological and other ethnological evidence I've shown, that the Wa people came from the Continent as former members of the Xiongnu confederation. That was the initial burst of popularity for Altaic languages, but it also included many Yeniseian speakers as well, including among the elite clans.
Later, Yeniseian speakers would shift to Altaic languages in the Steppe region, leaving only a handful of northward-migrating Yeniseian speakers to spread their languages toward the Arctic region instead.
The Wa were one of the Yeniseian-speaking clans within the Xiongnu, indicating they were probably elite, as also indicated by one of their totem animals being a Heavenly Piebald Colt, akin the to elite Alat tribe (who, however, were Turkic-speaking by the Xiongnu era -- but perhaps they were former Yeniseian speakers who just adopted Turkic).
To clarify, Ainu and Nivkh have their own separate sets of pronouns, none of which overlap outside their own language (aside from the one borrowing from Altaic into Nivkh).
ReplyDeleteNot that Ainu and Nivkh share a set of pronouns.
Another signal of how conservative pronouns are -- they bear case markings even in languages that have otherwise lost case, like English.
ReplyDeleteI, me, my, mine
We, us, our
Etc.
In the 1995 Doraemon movie (Diary on the Creation of the World), Nobita plays god with the help of Doraemon's gadgets -- he creates a little universe in his bedroom, for the purpose of observing it for a summer science project. He and the rest of the Tokyo gang then go on adventures in this newly created parallel world, based on Earth.
ReplyDeleteAnd they include the creation myth that I wrote about earlier, whose only parallel is from Mongolia.
Doraemon gives Nobita a hi-tech staff with jewel-shaped buttons on it, just like the jeweled spear that Izanagi and Izanami wield. And Doraemon instructs Nobita to use the staff to stir up the various cosmic ingredients that have been dumped into the dark void -- just like the Japanese creator gods churning the primordial ocean, whose resulting sea-foam coagulates into landmasses on the surface of the water.
Nobita says he feels like he's stirring a stew in a pot. (Whipping eggs would be more apropos, changing the state of matter by churning it with a utensil.) Perhaps that was the initial inspiration for this very distinctive and unique creation myth -- creating the world is like cooking or baking, transforming raw ingredients into a special structured meal, through the use of tools and techniques.
And not just any ol' cooking technique -- but a typical Steppe technique, stirring various stew ingredients in a cauldron with a long utensil. The only thing missing from the stew analogy is fire heating it up. So it really is more like whipping eggs or cream, to make the matter change phase, without even heating it.
Wait -- churning milk into butter! No heat needed. It's a pastoralist metaphor after all. How fitting for the Steppe. ^_^
I didn't think of the possible cooking connection to the Mongolian and Japanese myth -- but the contemporary myth-makers in Japan, who created these anime movies, probably have a closer connection to the original material from pre-historic times, as it's been transmitted locally and face-to-face and by word-of-mouth over the generations.
Before picture books, how would a story-teller have conveyed this story? Probably they took a little break from speaking, to make physical gestures like stirring an imaginary pot with an imaginary staff, to make it come alive for the audience gathered around them.
They also include a figure in their parallel world who is based on Pimiko / Himiko, who they call Himehiko (Princess Priestess), who is an over-the-top raving shamanness, in fact they make her look like an evil crazy old witch.
But she didn't unite the entire Wa people around her by being an evil old witch. She was probably more of a charismatic priestess or shrine maiden instead.
Anyway, I was very pleased to see Glorious Nippon preserving their prehistoric culture even in a modern hi-tech medium like anime.
If Hololive ever performs a scene for a concert, or music video, or Hologra, where one of the girls is creating a new world, make sure that she uses a magical staff to stir the primordial liquid so that it changes phase from a liquid to having some solid pieces forming -- just like Izanagi and Izanami!
If Doraemon can preserve this ancient myth, so can Hololive! ^_^ Whenever the context is appropriate.
*Himemiko is the character's name.
ReplyDeleteFinns are Uralic culturally, not Uralophone Indo-Europeans. I've gotten YouTube recs from Razib about Finland being Germanic, which I haven't seen, but that's on the same theme I've been pursuing about language and culture not necessarily matching -- could be language shift.
ReplyDeleteWell, for Finland, they are definitely Uralic culturally, not Indo-Euros who adopted a Uralic language. This makes them different from the Hungarians, who *are* Indo-Euros who adopted Uralic.
First, looking at the language. Finnish does not have voiced obstruents other than some kind of "d", does not have "v", does not have fricatives other than "s" and "h", and uses gemination for their main consonant distinction. None of that is Indo-Euro, it's totally Uralic.
Hungarian does have voiced obstruents like crazy, does have various fricatives (voiced and unvoiced), does have "v", and although gemination of consonants is present, it's not the main distinction.
Also, the Hungarian word for bear is "medved" -- this carries over the Slavic practice of using a taboo-avoiding word for bear, and the exact same sound sequence for the avoiding word (meaning "honey-eater" in Slavic).
We know Hungarian was only adopted in the 2nd millennium AD, after the invasion of the Magyars, who were an elite minority, while the Slavic majority remained and adopted the Uralic language of the invaders.
Finland is not like that. It has never been documented as speaking Indo-Euro languages, let alone into historical / literate times, it's much further away from the P-I-E homeland in the Western Steppe, and it's seamlessly connected to the broader Uralic-speaking territories.
I'm sure further ethnological investigation would confirm this, but just looking at their language, it looks like a pure Uralic language, not one that was put through an Indo-Euro filter like Hungarian.
What about Estonia?
DeletePossible Chinese etymology du jour -- "tree" from Yeniseian? While hunting for Yeniseian descendants in Japanese, I thought I found a perfect example -- turns out it's a borrowing from Middle Chinese, and is not present in the Ryukyuan side of the family. Oh well.
ReplyDeleteBut then I looked at the Chinese word itself, and it has no known etymology, nor does it have any cognates throughout the rest of the Sino-Tibetan family. Hmmm...
Maybe the Chinese borrowed it from Yeniseian? That's why I thought the Japanese form -- "moku," which later changed into "boku" -- was Yeniseian. It *does* have a Yeniseian feel to it, but the route could have been Yeniseian -> Chinese -> Japanese.
There are 2 reconstructions of the Old Chinese word for "tree", "moːg" and "C.mˤok", where "C" is some unknown consonant.
https://en.wiktionary.org/wiki/%E6%9C%A8
In Proto-Yeniseian, it's "xʷag":
https://en.wiktionary.org/wiki/Reconstruction:Proto-Yeniseian/x%CA%B7ag
It's monosyllabic, ends in a velar stop, the vowel is not front or high, and the initial consonant combines a labial feature with a velar feature.
The same is true for the OC word, although the 2 reconstructions differ over whether the final stop is voiced or not, but that's not very important.
Both say there's an "m" before the vowel -- perhaps the labial feature of the P-Y word became realized as an "m" in OC. I know that OC had labialized velars, but apparently only for velar stops, not velar fricatives like "x".
"C.mˤok" has two options for where the velar feature could've gone to -- either the mystery consonant at the start, or preserved as the glottalic feature on the "m" (backness).
If the velar is C, then the original order of "velar feature + labial feature" is preserved.
If the mystery consonant is unrelated to this, and it's the glottalic feature that preserves the original velar feature, then there was a metathesis of the two elements, putting the labial feature first and velar / glottalic feature second.
Why didn't OC render "xʷ" as a "kʷ", which they did have? IDK. We only have access to the most recent common ancestor of the Yeniseian languages that there's good evidence for.
Perhaps the Yeniseian variant at the time it entered Chinese was more similar to the attested OC borrowed form -- like maybe the Pre-Proto-Yeniseian form was originally the labial feature followed by the velar / glottalic feature.
In other words, Yeniseian is the one that underwent metathesis of the two features in historical times, whereas the Old Chinese form (and its descendants) preserves the original order of the two features.
Just putting it out there... there are no better ideas, so why not look to the highly underrated Yeniseian family? It was influential enough during the Xiongnu times, so they're a good candidate for who could have fed loans into Chinese -- especially if they're not anywhere else in Sino-Tibetan. That suggests a northern donor.
I suspect it's actually the Old Chinese form that is more faithful than the modern-ish Yeniseian forms.
ReplyDeleteIf the original way back when was "xʷ", it's just too easy for the Chinese to render that as "kʷ", or if they did have labialized velars, to render it 100% accurately.
So then assume that the Yeniseian form way back then was labial feature + velar / glottalic feature, which metathesized sometime after the OC borrowing.
Given the Chinese borrowing being an "m" rather than the many other labial or labialized consonants they had, it's likely that the initial consonant was "m".
If a consonantal "m" has to become a secondary co-articulation, it's guaranteed to become a generic labializing feature like "ʷ", or perhaps nasalization onto the following vowel (but OC did not allow that).
Whereas if the labializing element were originally a co-articulation, and the borrowing language couldn't do that (which OC probably could have anyway), it could've been rendered as a consonant as "w" or "p" or "b" or an aspirated "p"... or an "m". It's possible, but not as necessary as the first scenario.
If this is true for this one word, it is likely true for a whole bunch of others -- and because Chinese borrowed them a very long time ago, they reflect a stage of Yeniseian that is otherwise invisible to its reconstructors -- more like Pre-Proto-Yeniseian.
Neat!
Therefore, maybe the Japanese "moku" is an inheritance from their former Yeniseian-speaking days after all!
ReplyDeleteThey kept the initial "m", ditched whatever the velar / glottalic co-articulation was cuz they couldn't do that in Japonic, kept the vowel, either kept the final vowel or at most devoiced it (as per yoozh in Japonic), and stuck a dummy vowel on the end to obey Japonic phonotactics.
It's in Japanese only, not the Ryukyuan side of the family -- maybe it was out-competed by the "ko" / "ki" Japonic word for "tree", since "moku" is after all a rarer and unusual form of saying "tree" in Japonic.
It's still possible that the route was Yeniseian -> Chinese -> Japanese, rather than Japanese carrying it over from their former Yeniseian-speaking days, with no Chinese mediation.
But purely from looking at the Yeniseian - Chinese connection, it suggests that maybe the very old Yeniseian form was close to "mok" already, and the Japanese could've inherited it after all.
Who knows, more systematic investigation is needed. But it just goes to show how crucial it is to understand Yeniseian -- and even it's Pre-Proto stage that cannot be reconstructed internally, but may be revealed by Old Chinese borrowings from the time of their "interactions" shall we say with the good ol' Xiongnu confederation.
In fact, the borrowings could have come from a lot earlier than the Xiongnu -- the Guifang, from the mid-to-late 2nd millennium BC (against the Shang dynasty in China):
ReplyDeletehttps://en.wikipedia.org/wiki/Guifang
They were part of the Seima-Turbino culture (northern / Steppe barbarian types), some of whom would have been Yeniseian speakers, which used to be spoken near the Altai-Sayan Mountains, before the language (although not the people) eventually became displaced by Turkic, Mongolic, Uralic, and other expansionist language families.
Jeez, imagine Yeniseian speakers in 1300 BC -- that is definitely WAAAAYYY before the reconstructed form of Proto-Yeniseian that scholars can discover based on the historically attested Yeniseian languages like Ket and so on.
Pre-pre-pre-Proto-Yeniseian... but an ancestor in the family nonetheless, and which Old Chinese or an even older ancestor of Chinese could have borrowed from, but which the other Sino-Tibetan languages would have had zero contact with, as the family is mostly Southeast Asian, and the Guifang and Yeniseian homeland is far to the northwest.
These Old Chinese borrowings from Yeniseian would also help the project to connect Yeniseian with the Na-Dene family in western North America. Their most recent common ancestor goes back to before the future speakers of Navajo even left Asia! That puts quite a bit a distance between them and their Yeniseian cousins.
ReplyDeleteBut the Old Chinese borrowings from Yeniseian could reveal a stage of Yeniseian that goes back much closer to its common ancestor with Na-Dene.
How to identify these OC borrowings from Pre-Proto-Yeniseian? Just find words that have no known etymology, no cognates elsewhere in Sino-Tibetan, and seem to be very basic common words like "tree".
Who borrows the word for "tree"? Well, maybe it wasn't "Chinese speakers" in general who came up with that word, which is absent elsewhere in Sino-Tibetan -- maybe they were former Yeniseian speakers who joined the Chinese civilization, switched to speaking Chinese, but carried over some of their old Yeniseian lexicon.
If these were not random migrants and refugess, but powerful, influential clans from a worthy fuckin' adversary, they might have punched above their weight for introducing loan words -- and other cultural loans -- into Chinese culture.
That's why Northern China is cooler than Southern China -- beef instead of poultry, wheat instead of rice, some degree of dairy vs. no dairy whatsoever, horse riding, horny Amazon women, and so on and so forth. Whereas Southern China will always just be a tropical shithole.
I really meant to talk about Tocharian pronouns heavily reflecting their Uralic -- not Indo-Euro -- background, as another case of "pronouns being resistant, and revealing older stages of cultural history, perhaps when they spoke a language from a totally different family".
ReplyDeleteBut this Yeniseian stuff is just too cool!
We'll see if I'm up for it after dinner and Doraemon... ^_^
Estonian looks 90% Uralic. Hardly any voiced obstruents, other than "v", which is strange for a non-Indo-Euro language. It also has "f" as a fricative, also unusual for Uralic. Those are the only Indo-Euro traits, and they're related ("v" being a voiced "f").
ReplyDeleteThere's a consonant distinction based on length, and plain vs. palatalized throughout the consonants in the alveolar place.
Estonian has Uralic pronouns, just like Finnish does.
And Estonian has a shit-load of cases, and ones that I-D does not. Like the comitative and abessive, which Finnish and Chuvash also have.
Estonian culture has probably seen more influence from neighboring Indo-Euros, whether Germanics, Baltics, or Slavics, compared to what Finns have experienced -- which is not much, especially before 1500 or whenever Sweden and Russia encroached on them.
But Estonian still seems fairly Uralic. It's in NE Europe, contiguous with the broader Uralic belt, paved by the Seima-Turbino culture from the Eastern Steppe:
https://en.wikipedia.org/wiki/Seima-Turbino_culture
Hungarian is the odd man out among Uralic-speaking Euro cultures. They're clearly Slavs who adopted the Uralic language of the foreign invaders of the late 1st millennium AD, the Magyars.
Found another Yeniseian word in Japanese! The number 2! It took a little while, but it's there.
ReplyDeleteIn P-Y, the number 2 is "xʷin", so we apply the same sound changes we've seen in the other cases -- change "xʷ" into a labial, keep the nasal since Japonic has "n", don't worry about the root vowel, stick on a dummy vowel at the end (and don't worry about that either).
So far, the labial has been "p" -- so we'll start there. And we can end there!
In Japanese, "pana" (these days, "hana") has to do with separating, detaching, digressing, making one thing distant or remote from another, drifting apart from one another, etc. E.g., Classical Japanese "hana + ru" (like modern "hana + reru"):
https://en.wiktionary.org/wiki/%E9%9B%A2%E3%82%8B#Etymology_2
The meaning of "apart" is one of the senses used in the Ket evidence for P-Y, so we can use it here as well.
The basic numeral 2 in Proto-Japonic is either "pu" or "puta", and although the initial labial is the expected one, it lacks a nasal later on. Maybe the nasal was lost after "pu" and this is a genuine cognate with P-Y as well, but it's probably an obscure variant on "2" that I don't know.
I only checked into "pana" after seeing the "apart" sense of the Ket evidence, and looked up that word in Japanese -- lo and behold, it has a "p" followed by an "n", with vowels after each.
There's got to be a regular correspondence going on here, this is too many to be coincidence -- and again, in such basic common words that are unlikely to be borrowed anyway.
Related to "hana(re)ru" is "hanasu", which is the active rather than mediopassive one of this verb pair. It means to separate, isolate, disconnect, keep apart.
ReplyDeletehttps://en.wiktionary.org/wiki/%E9%9B%A2%E3%81%99
To be slightly more formal, I'm proposing the Yeniseian original was of the form:
ReplyDeleteLˠ...
Where L is some labial consonant, and the secondary articulation is velarization. It doesn't matter what the secondary element is, as long as it's back -- velarization, pharyngealization, uvularization, whatever.
At the stage where Old Chinese borrowed from Yeniseian, they preserved the labial consonant, and maybe changed the fine-tune nature of the secondary feature -- maybe it was pharyngealized rather than velarized, as per the reconstruction of "tree". But still mostly in keeping with the Yeniseian target.
Japonic, when its former Yeniseian speakers were carrying over these words (through their Altaic language from between their Yeniseian and Japonic periods), kept the labial consonant as best it could, perhaps devoicing it since Japonic can't have voiced obstruents.
Unlike Old Chinese, which was rich in consonants and 2nd-ary features, Japonic was a radically simplified phonology, due to the language of the Wa people being put through an Emishi-an filter, as the two cultures were merging together into the future Yamato and Japanese culture. So Japonic just deleted the 2nd-ary feature altogether.
Yes this results in a homophone problem, as we've seen two instances of Japanese "pana" (now, "hana") stemming from similar but distinct Yeniseian roots for "nose" and "two". That's just the price they had to pay for radically simplifying things.
I thought they'd put more effort into altering the main vowel, or the final dummy vowel, in order to avoid that. But at least in this pair, they didn't, and now "nose" and the root having to do with "separation" are homophones.
Not to mention the word for "flower" -- that's "pana" (now "hana") as well! I wonder if there's a Yeniseian word that begins with "xʷ" and is followed by a nasal?
We'll see... ^_^
Then later, after Old Chinese and Japonic were done borrowing / carrying over these Yeniseian words, there was an internal change in Yeniseian, which switched the roles of primary and secondary articulation in that consonant --
ReplyDeleteVʷ...
Where V is some velar consonant, and the secondary articulation is labialization. The V was probably a fricative, to preserve the previous velarization feature, which was more continuous than stop-like. Hence all those "xʷ" examples.
Metathesis is common in the evolution of Yeniseian, both internally -- P-Y to daughter languages -- and externally -- when borrowing words.
ReplyDeleteSee "Metathesis in Yeniseian loanwords of Altaic origin" by Khabtagaeva (2015), and its review of earlier work.
In particular, Yeniseian languages prefer to put the velar before the labial, over the course of their own internal evolution.
I knew it! ^_^
So perhaps that affected not just sequences of consonants, but consonants with co-articulation as well -- switching which feature was primary vs. secondary. Labials with velarization became velars with labialization.
Old Chinese borrowed, and Japonic carried over, words from the earlier stage, whereas their reconstructions from internal Yeniseian evidence show the later stage, after metathesis of the features.
I found another Yeniseian word in Japanese! It's "ken", a uniquely Japanese unit of measurement, equal to roughly a fathom, i.e. about 6 feet. In Japanese measurements, it's equal to 6 "shaku", which is about 1 foot.
ReplyDeleteWell, maybe it's not so uniquely Japanese after all... and no, it's not from Chinese. Chinese measurements don't actually have a special term for roughly 6 feet -- they have one close to a yard or meter, i.e. 3 feet, then the next one is close to 36 feet. They skipped the fathom.
Some retarded faggot keeps deleting Wiktionary entries on Proto-Yeniseian, so pardon the awkward URL here, which goes to the pre-fag-deleted version:
https://en.wiktionary.org/w/index.php?title=Reconstruction:Proto-Yeniseian/%CF%87%C3%A4%C5%84&oldid=72517998
It's reconstructed as "χäń", from "qān" in Ket, "hen" in Kott, and "xan" in Yug.
Japonic doesn't have velar fricatives, so just harden it into a stop, while keeping place and voicing -- "k". Japonic doesn't have that strange nasal, so make it as close as possible, "n". It's word-final, so it's not a mortal sin to leave it without a dummy vowel afterward. The vowel is front, and mid-height -- and in P-J, the only front vowels are mid "e" and high "i", whereas "a" is a central (and low) vowel. So go with "e".
That gives "ken"!
Japanese uses the kanji for Chinese words whose Old Chinese forms are similar, like "kre:n", but they have an extra "r" in there, and none of the many many meanings are a unit of length measurement, let alone one equal to about 6 feet.
So, Japanese did not borrow the meaning from Chinese -- it just found a Chinese character whose sounds were close enough to "ken", which as it turns out, was carried over from their Yeniseian-speaking period!
As a reminder, the Japanese had zero contact with anyone once they landed in the Japanese islands, other than some literate borrowing from China. The only time-and-place when the Japanese were in contact with Yeniseian speakers was back when they themselves were Yeniseian speakers, in the Altai-Sayan region, starting who knows when but lasting up through the Xiongnu confederation, before they trekked over the mountains into Korea and then Glorious Nippon.
Speaking of fathoms, recall that creation myth from Mongolia that is 95% identical to the Japanese creation myth involving using a divine staff to churn the primordial ocean, in order to change its phase of matter somewhat, creating solid landmasses from liquid water.
ReplyDeleteIn the Mongolian version, the divine staff is described as being "10,000 fathoms long" -- forget the number, focus on the unit of length. It's a fathom! How many cultures even have a specially named unit of length measurement for roughly 6 feet?
Well, Japanese and Yeniseian do (and so does English). So probably those tale-tellers who Potanin collected these Siberian myths from, may have been Altaic-speaking in the 1800s, but their ancestors who created that myth were likely Yeniseian speakers.
And in fact, Yeniseian did used to be spoken in NW Mongolia and to the west of Lake Baikal, where he collected those myths.
I looked into "Yeniseian creation myths" earlier, but couldn't turn up any solid connections to Japanese mythology. Well, that's why -- a lot of Yeniseian myths are now being told and preserved by speakers of Altaic langauges, so they're categorized as Altaic rather than Yeniseian, which they were once upon a time.
And nowadays, scholars categorize "Yeniseian myths" as those belonging to people speaking these languages in the past couple centuries. Only problem is that a decent chunk of them did not used to speak Yeniseian, they spoke some language from further north and closer to the Arctic circle, but adopted Yeniseian as some of its speakers migrated northward away from their Steppe homeland.
I didn't come across the "churning the primordial ocean" myth in Turkic or Mongolic or Tungusic mythology, so in all likelihood, that came from Yeniseian cultures -- and was preserved by the Wa people even after they switched to an Altaic language, and still even after they created the Japonic language synthesis with their new Emishi-an neighbors.
The fathom is actually not so common around the world, according to the list of translations for "fathom" at Wiktionary. They could be missing some -- they have the wrong term for the Japanese one, they claim a Chinese one exists when it doesn't, so they could be overlooking some.
ReplyDeleteBut there are 4 clusters of cultures where a lot of their members have a unit of measurement that's roughly 6 feet -- 3 of whom are interacting neighbors in the central-to-western Steppe -- Indo-Europeans, Uralic, and Turkic, and as we just saw, Yeniseian.
They don't use the same term across their cultures, but internally their term is widespread. Tocharian uses a term that is closer to the Turkic or Yeniseian term, not Uralic as expected, but still Eastern rather than using the expected Indo-Euro form (cognate with English "fathom"). Well, not so expected, if you know who the Tocharians used to be before adopting an Indo-Euro language -- Easterners.
The other cluster is Austronesians, especially in the Philippines. And no, it's not even remotely similar to the Japanese term -- it's something like "dipa", not "ken". Austronesians have had literally zero influence on Japanese culture, and never will.