Although I've established that the Japonic language family belongs to the broader Dene-Yeniseian family, which presently spans Siberia to the American Southwest, I had no idea what the time-frame for this relationship was, when first exploring the matter.
But upon further investigation of linguistic, mythological, and ritual relationships, I've not only uncovered further shared shibboleths, but determined that their common-ness lasted up through fairly recent times. Their common ancestor was not from 10, 20, or 100-thousand years ago. Not every element of their cultures dates back to the exact same time-period. Still, the earliest shared cultural ancestor they have goes back no earlier than 2000 BC. And in some cases the shared ancestor only goes back to between 500 BC and 300 or so AD. They could have been in lingering contact with each other through the 1st millennium AD.
This totally blows up the notion that the culture of the New World was either brought with long-distance migration from the Old World on the order of 10s of thousands of years ago, or that it evolved only in the New World after that initial wave of migration. For some New World cultures, that's true. But for others, who speak Na-Dene languages, their bodies and haplogroups may have come over 10,000 years ago, but their present-day culture did not, nor did it evolve solely in situ after settling into the Americas. There was a transmission of Siberian culture from the Old World starting as early as 2000 BC and perhaps lasting through 500 AD.
This is a great illustration of the non-correlation between genes and culture, or between migration of bodies and migration of culture. Cultural forms can spread by contagion, from one adjacent group to the next, and so on in a chain. Group A transmits it to group B, B transmits it to C, and then C to D. The fact that it is present in A, B, C, and D does NOT show that it began with A and then A migrated through all the intervening regions, leaving it behind them at each stop. In fact, it doesn't imply that A has migrated at all -- maybe they had contact with B (without, however, migrating to replace B, just interacting with neighbors), but not C, D, or any further link in the chain. And contagion is a far faster transmission process than migration of one spreader group to all regions affected -- as the phenomenon of epidemic diseases shows.
That's not to say that cultural forms spread just like epidemic diseases -- the key difference is how "susceptible" one group is to the cultural influence of another. Pathogens don't care about cultural groups or which other groups they choose as role models vs. groups they shun / avoid. Cultural transmission is not necessarily transitive either, unlike diseases -- in some cases, A influences B, and B influences C, but C rejects the influence of A (directly). In such a case, B acts as a gatekeeper between A and C, a role that does not show up in epidemic disease transmission. This is where all of the real art-and-science of cultural transmission happens -- just saying it can spread from one adjacent group to the next is not very surprising, it's what relationships must be in place for the transmission to take place, and when it will stop.
With that big-picture in mind, let's now look at some shared cultural shibboleths in Northwest America and cultures of the Eastern Steppe, including the off-shoot that wound up in Glorious Nippon. Some of these examples are from the comments in the previous post, some are ones I discovered just in the past few days.
I'm going to try breaking these up into smaller, more digestible separate posts, instead of having a long string of 250 comments on a single post.
* * *
First, we'll start with ritual. An earlier post demonstrated that ancient Japan (starting sometime before 300 AD) shared a cultural shibboleth with the Xiongnu and the Pazyryk cultures of the Eastern Steppe (and Silla, in Korea) -- breaking precious bronze mirrors as a funeral rite, and burying them as grave goods. See that post for all the details, which begin in the 3rd section.
For the time-frame, this ritual began in the mid-1st millennium BC (Pazyryk) and lasted through the mid-1st millennium AD (Kofun-era Japan). After then, the practice seems to have faded away...
But not in Northwest America, where a strikingly similar ritual lasted up through the conquest of the frontier by the Americans. Perhaps in the Old World, gradual sedentarization and civilizing influences from China led to the abandonment of this Eastern Steppe barbarian ritual, while in Northwest America there never was a sedentary mega-state civilizing influence until the white man showed up. They certainly were not in the cultural orbit of the Aztec, Maya, or Inca empires. So perhaps the Mongolians and Japanese would still be practicing this funeral rite, if there had been no China for them to interact with...
Metallurgy was never widespread in North America before European colonization. But there was a limited amount of iron and copper production in the Northwest, where the raw material was sourced from Alaska. We'll get to the mythical or not-so-mythical legend behind that, when we look at myths!
But after they began working with copper, various groups in Northwest America, not only those speaking Na-Dene languages, began to practice a ritual of breaking precious copper ceremonial shields, sometimes as part of a funeral, although sometimes as part of a potlatch. In both occasions, though, the broken fragments were buried somehow -- either under the ground or tossed into the sea, where they sank to a watery grave.
In both occasions, the "coppers" or copper shields were not very utilitarian, they were symbols of status and wealth and perhaps a connection to the supernatural and to ancestral lineages. They were decorated as art-works. They were owned and traded as decorative status symbols, with some degree of otherworldly power -- just like those bronze mirrors in the Eastern Steppe.
The main difference in the New World ritual is the shape -- they are beaten smooth and are between half a foot to several feet in length, like the Old World round mirrors, but they take the form of a T-shaped shield (again, not used as an actual shield in battle). And the adornments are also local patterns. So there was some degree of syncreticization between the Old World ritual and existing Northwest American arts-and-crafts traditions.
They don't seem to be treated as magical for their ability to collect and reflect light, as though they were mini-suns, like they were in the Old World, where they played into the solar cult of the Eastern Steppe.
We can't tell if their use at potlatches is different from their Old World context, which was funerals -- perhaps the only difference is when potlatches were held in the Old World, like mainly during a funeral, whereas in Northwest America, potlatches were held outside of funeral contexts as well. But in both cases, their breaking and burial can be viewed as a form of conspicuous wealth destruction, as an honest signal of the large amount of wealth held by the leader. In this way, it's no different from human and animal sacrifice, on the same occasions (potlatch and/or funeral) -- it's the sacrifice of precious objects, which only a wealthy household owns.
However, it's not just a vague conceptual similarity like "sacrificing precious objects" -- they are made of copper, beaten smooth into a sheet, adorned as decorative objects, used in a ceremonial context rather than a valuable object that sees real-world use, and their method of sacrifice is the same -- breaking into fragments.
Most inclusion of precious objects in a funeral or other ceremony does NOT involve breaking them or rendering them worthless as utilitarian objects, or even as decorative objects. Indeed, if they were just decorative objects made from copper, they would not seem so similar to the Eastern Steppe versions -- it's the fact that they're ceremonially sacrificed by fragmentation that jumped out at me when reading about them.
All these unnecessary / arbitrary points of similarity show that it is a shared shibboleth, not just independent variations on a universal theme.
This New World ritual goes back no further than the mid-1st millennium BC, when it is observed among the Pazyryk culture in the Eastern Steppe. It was still practiced in Japan during the early Medieval era. When was it transmitted into the Pacific Northwest? Sometime after the prerequisite adoption of copper metallurgy, which was more limited regionally and occurred later in the New World.
I'm guessing sometime at the twilight of its use in Japan, or just afterwards -- when the Siberian transmitters would have still been familiar with the ritual, but since it was no longer de rigueur in Siberia, it was not emphasized as strictly to the New World adopters. It was likely transmitted by the same Siberians who taught the New Worlders about copper in general -- "Y'know, this isn't just utilitarian stuff. Where we come from, we beat it into sheets, decorate it, and break it into fragments during major ceremonies like a funeral. You guys should do that, too!"
Maybe they specifically said to make it into a round mirror, and the New Worlders said that wasn't as relevant to their local culture, and made them into shield-like shapes instead. Or maybe since the ritual was fading away in Siberia, the Siberian transmitters didn't insist on every element being preserved -- what the hell, the ritual was dying out anyway, just give them the basic understanding, which is that copper is valuable, it can be beaten into a smooth canvas for artwork, and this valuable object can be broken into fragments for a major ceremony like a funeral.
I doubt it was too much later after it died in the Old World, since hardly anyone would still remember it in order to pass it on. Probably not 1000 AD. But also not too early, since it only began in 500 BC, and it required use of copper or bronze. So perhaps more like 300 to 700 AD, though after being transmitted, lasting right up through the "present" (closing of the American frontier circa 1900). Pacific Northwest natives never had to worry what a civilization like China might think about them sacrificing people, animals, and precious mirrors, well after the Dark Ages...
Marine gave a great karaoke performance, with classics as well as recent hits. She kept saying, "Let's sing oldies, to make teenagers cry". ^_^
ReplyDeleteAs always with Marine, I learned a new classic Japanese song -- "Nagori Yuki" by Iruka, from the good ol' Sad Seventies. It has a very Carpenters-like fusion of folk music, a jazz-y / funky bass, and a plaintive female vocal singing about love.
So far, the only other Hololive girl to sing it is Noel, several years ago. It would really suit Irys, who loves the Carpenters. ^_^
And Marine has sung it many times -- it's clearly a favorite of hers, and I can see why. Her rich, honey-dripping voice really adds something to the original. It's not her usual seductive, sexy voice... it sounds more like a nuturing lullaby from a pretty babysitter. The high level of resonance is not for va-va-voom seduction, but like a cat purring -- to calm and soothe and reassure.
I really love when Marine lets this side of her personality show, the role of "cat purring on your lap, in order to soothe and comfort you".
And as someone who is easily possessed by the spirit of the music, she really gets into the performance. She's doesn't get stage-fright. So, even if that is not her usual personality, it's a secret part of her, and she can easily turn it up to the max by being possessed by a song that has that personality in it.
Of course, when the song is seductive, then Marine can also become possessed by *that* spirit as well, and her resonant voice does sound like a seductive cat in heat, hehe. But she's not faking it just for clicks -- it's a real side of her personality, and she can't help but become possessed by the spirit of the music, no matter what it is.
Speaking of "Ihojin" again, Marine performed that in a duet with Azki during her concert that was made in the "Showa TV concert shot on video" style. I'm so glad that someone who is as popular as Marine is using that role to preserve the classics of Japanese culture, and make them cool to newer younger audiences, as well as drawing in older audiences.
Every girl in Hololive who preserves the classics, whether it's songs or movies or TV series or video games, is part of the team effort to keep Japanese culture alive. Culture doesn't keep itself alive, and is not transmitted passively by genes -- it needs to be actively transmitted, every single generation after the other, or it dies.
Living in the collapsing American Empire that is demolishing its own culture, it's a relief to see that *some* wonderful cultures are continuing to preserve themselves, and not with a groan as though it's a chore that must be done, but with an enthusiastic smile cuz it's fun and exciting and awesome.
It's no surprise that Americans, from all walks of life, are jealous of Glorious Nippon... ^_^
The Ainu / Emishi belong to the Uralo-Siberian sphere, for all sorts of cultural shibboleths, including perhaps their language. This will require a separate post with details, but just as a preview of the main topic, here's the proposed mega-family of languages that includes Uralic, Yukaghir, Eskaleut, and maybe / maybe-not Chukotko-Kamchatkan:
ReplyDeletehttps://en.wikipedia.org/wiki/Michael_Fortescue#Uralo-Siberian_languages
I've found enough evidence to put the Ainu language family into this mega-family -- whether or not they're closely related by linguistic ancestry, and/or part of a heavy region of contact, and/or switching / conversion from one family to another while bringing over linguistic baggage from their previous language.
But as I said about Japonic and Yeniseian and Na-Dene, and Indo-European languages vs. Indo-Euro cultural shibboleths, language is not the most diagnostic domain of culture in order to see who shares a cultural ancestor. People change their languages more than they change other parts of their cultural shibboleths. Still, it's worth adding to the entire picture by looking at language.
The other domains of culture where the Ainu / Emishi resemble and must share a cultural ancestor with the other groups in the Uralo-Siberian sphere, are myth / religion, the bear cult rituals, clothing, and geometric patterns / designs for adornment.
Phenotypically, they have rounder / spherical skulls, narrower eyes, less prominent nasal bridge / wide nostrils, although unlike SE Asians who share those traits, the Uralo-Siberians are hairier. They're also tall.
So in the Japanese islands, there was a collision of two expanding cultures:
First, the Wa, representing the Eastern Steppe sphere (along with Na-Dene and Yeniseian), who entered Japan from Korea / the west, associated with the Yayoi and Kofun periods in Japan.
Second, the Emishi (later evolving into the Ainu), representing the Arctic Siberian sphere (along with other Uralo-Siberian cultures), who entered Japan from the north, associated with the Jomon period in Japan -- especially the Late / Final Jomon periods, in the 1st millennium BC. The earlier Jomon periods going back to 3, 4, 5, 10,000 BC, may not belong to the Uralo-Siberian mega-culture. But when the distinctive dogu style of statues show up in Japan, that's for sure Uralo-Siberian.
Therefore, the Uralo-Siberian culture may have arrived somewhat earlier than the Wa-Dene-Yeniseian culture, but not by very much. The U-S culture landed in the north and migrated south and west, while the W-D-Y culture landed in the west and migrated east and north. In the middle of Honshu, they collided with each other, and fought over who would rule Japan. The Wa won over the Emishi, and assimilated the losing side into a new fusion culture -- the Yamato.
ReplyDeleteAs I suggested earlier, I think "Yamato" has an Ainu etymology, meaning "chestnut father", as in the lord or provider or patron over the chestnut groves, which are crucial to the ecology of Central Japan where the Yamato court was born, and whose fruits -- Japanese chestnuts -- remain an iconic food item of Glorious Nippon still to this day.
The Emishi who did not want to assimilate, doubled-back to the north and held out in Tohoku for awhile, before migrating further north into Hokkaido, Sakhalin, the Kurils, etc., joining a Medieval Okhotsk culture, and becoming a somewhat new culture, the Ainu.
And although the Yamato society took part in a cultural and genetic fusion of the previously separate populations, there is still a more Wa flavor to the Western half of Japan, and a more Emishi flavor to the Eastern half.
There are far more Emishi / Ainu place-names and loanwords in Eastern Japan, the location of the mythical Sanzu no Kawa is supposed to be in Tohoku, and Eastern Japanese are taller, with rounder faces, narrower eyes, and less prominent noses. Probably less lactose-tolerant as well.
In Western Japan, there are fewer Ainu words, mythical locations linked to Wa-Dene-Yeniseian myths (like where Yamato Takeru was guided by a white wolf guardian spirit), and the Western Japanese are shorter, with taller faces, larger eyes, with more prominent / angular noses. More lactose-tolerant.
This is just an overview, but I have to write some of it down right away before I forget it! This will take me away from my usual focus on Wa, Na-Dene, and Yeniseian, but will circle back to my focus on Uralic! Good thing I spent so much time studying Uralic while discovering the Uralic origins of the Scythians and neighboring groups of Herodotus' time! ^_^
To clarify the significance, it's already known that there was a culturally and genetically different population in Japan before the Yayoi period, and that the Wa came from the Asian mainland.
ReplyDeleteWhat I've done is to show which precise cultures in the broader region sent an off-shoot into Glorious Nippon. The Wa were not just any ol' group of mainland Asians, they were from the Eastern Steppe, and spoke a language related to Yeniseian and Na-Dene.
What I'll show coming up, is that the Emishi / Ainu were not a cultural isolate, whose culture had evolved in Japan since forever -- but that they, too, were a cultural off-shoot of mainland Asia. And definitely *not* Southeast Asian, lol. Unlike the Wa, their roots came from further to the north, part of the Arctic culture rather than the Steppe culture. And if their language is related to any others, it's Uralic, Yukaghir, Chukotko-Kamchatkan, or Eskaleut.
The big picture is that there have been two mega-cultures in NE Asia for a very long time -- the Arctic one and the Steppe one. They have collided many times in various places, sometimes one replacing the other, sometimes leading to a fusion.
Both sent off-shoots into the New World -- the Steppe sent Na-Dene, whose speakers generally remain sub-Arctic in residence, and the Arctic sent Eskaleut, whose speakers remain Arctic in residence. These two sides have been enemies forever in the New World.
And both sent off-shoots into Japan -- the Steppe sent the Wa, who settled mostly into Western Japan, and the Arctic sent the Emishi / Ainu, who settled mostly into Eastern Japan. They used to be enemies, but they formed a fusion led by the victorious Wa, with the unassimilated Emishi retreating northward to remain culturally Arctic rather than get Steppe-ified.
And on the mainland of Asia, they clashed along their border that lies between the Steppe and the Arctic, usually forming a fusion -- like Turkic and Mongolic seemingly having both a Uralic and a Yeniseian parent. Or former Yeniseian speakers converting to Uralic language (Samoyedic).
Linguistically, on the mainland, the Arctic was victorious. In Japan, the Steppe was victorious. In the New World, they have not resolved their hostilities and have carved out separate regions of major linguistic expansion -- the Arctic vs. the sub-Arctic.
A nice display of both phenotypes of Japanese, in the singing duo Aming. Takako Okamura on the left, representing the Wa from the Steppe, and Haruko Kato on the right, representing the Emishi from the Arctic:
ReplyDeletehttps://www.gettyimages.com/detail/news-photo/takako-okamura-and-haruko-kato-of-singer-duo-amin-during-news-photo/1077481278
Kato has some Wa traits as well, but shows more of the Emishi roots, whereas Okamura looks like she just dismounted from her horse after riding in from Mongolia or Kazakhstan. Taller skull, larger eyes, more angular and prominent nose, bunny-rabbit overbite, furry eyebrows (almost a unibrow), and much shorter than her partner (hard to see in this pic where they're sitting down).
At least Okamura is from Aichi, which is right on the border between Western and Eastern Japan. IDK where Kato is from, but possibly Aichi as well.
The stereotype in Japan is that girls from Kansai are the hottest -- maybe due to them having more of a Steppe look than an Arctic look. Those striking high-relief features. The Steppe people did migrate throughout Japan and Hokkaido, but they are more concentrated in the Western half.
My grandmother was from Southern Hokkaido, but she was a Steppe descendant -- short stature, tall head, large eyes, angular nose, milk-drinker. By now, Steppe people have settled Hokkaido as well, they're just not as frequent there compared to Kansai or Kyushu.
Anyway, the only Hololive girls to sing a song by Aming during karaoke are the turbo-weebs Fuwamoco, who sang the group's #1 hit "Matsu wa" from the early '80s. At least, they're the only ones with a clip of it on YouTube.
Okamura had a long solo career after Aming broke up in the mid-'80s, but so far no girl in Hololive has sung one of her solo songs. They're pretty good, too, in the easy listening / adult contemporary / new age / world music / singer-songrwiter genres. Hard to sum her up with one label -- but who doesn't love an enigmatic woman? Especially one with such striking features? She matured into a real babe, with the sharp features of a baddie but the tender voice of a romantic. Image search her name to see for yourself... ^_^
Maybe the Koronator could cover one of her songs -- like "Yume wo Akiramenaide", which means "don't give up your dreams". She could subtly alter the first word to match her brand -- "Yubi wo Akiramenaide", or "don't give up your finger", since she's always joking about stealing a finger away from her fans. She likes mixing meme-y ironic humor with sincere heartfelt songs like that, so it would be right up her alley! ^_^
Celebrating this sudden BLAST of chilly air by opening all the windows, and BLASTING some sublime tunes while Arctic gusts rush all around me! I even sang along to a few of them, loud enough for the neighbors to hear! (1, 3, and 7) Noise complaints? Not if the music is awesome, iconic, and the singer is good! ^_^
ReplyDeleteSteal his playlist --
1. "Return to Innocence" by Engima
2. "Sadeness" by Engima
3. "Come Undone" by Duran Duran
4. "Pictures of You" by the Cure
5. "Secret" by OMD
6. "Maid of Orleans" by OMD
7. "Avalon" by Roxy Music
Thank God for this relief from 80-degree weather. My tiger-bear and I are going to get SO cuddly tonight! We already had a nice cuddly nap together around 5, but it's definitely going to be a wool blanket + body heat sleep-time tonight!
He's having a blast himself, sitting at the open windows and getting to see and hear and sniff all the wondrous gifts that the wind is bringing him. ^_^
Hmmm, curioser and curioser... perhaps Ainu, although Uralic, has a Wa-Dene-Yeniseian kind of layer underneath it. But not Japonic -- maybe another region where Dene-Yeniseian-like languages used to be spoken, way way way back when...
ReplyDeleteFirst, Ainu is confirmed Uralic -- I have pages of notes, after researching over the weekend.
But one key area where I couldn't find any Uralic cognates was personal pronouns. These are never borrowed, and are some of the most stubborn words to carry over, even after switching / converting languages. OK, so maybe they look like someone else's pronouns nearby -- but they're not like Uralic, Yukaghir, Eskaleut, Chukotko-Kamchatkan, or Nivkh.
However, they look somewhat like the Proto-Japonic pronouns -- except they look even more like Na-Dene and Yeniseian, which Wa / Japonic is sister to. So they're not a result of "recent" contact with Japonic speakers during the Yayoi period and afterward.
So I don't think there was a Wa / Japonic language spoken before the Uralic language that became Emishi and later Ainu. But it did belong to the same family as Wa / Japonic, i.e. Dene-Yeniseian. And yet it's not a clear copy-paste of Yeniseian or Na-Dene.
Perhaps it was Na-Dene or Yeniseian, but the intervening centuries and several language conversions / heavy contact with Japanese, have obscured what used to be clearly Dene or Yeniseian.
Or perhaps it was a distinct 4th branch of the family -- Yeniseian in the West, Na-Dene in the Northeast, Wa / Japonic in the Southeast, and... Jomon or whatever, due east of the Steppe. Note that the northernmost tip of Sakhalin island is due east of the passage through the Altai-Sayan Mountains, lying south of Lake Baikal, and at the same latitude as most of the Amur River region.
That would have been well before 500 BC -- maybe 1500 BC, in Sakhalin, Hokkaido, Eastern Honshu.
Then sometime in the 1st millennium BC, there's an abrupt change in culture and language -- the spread of the Arctic mega-culture, and its language family, Uralic. Call this language Emishi -- mostly Uralic, but its speakers formerly spoke a Dene-Yeniseian-type language (what I'm calling Jomon, or maybe Early Jomon). Because personal pronouns are stubborn to change, Emishi was lexically Uralic except for retaining Dene-Yeniseian-type personal pronouns.
Later, toward the end of the 1st millennium BC, the Wa speakers show up and inaugurate the Yayoi period in Japan. They bring the Emishi speakers *back into* contact with Wa-Dene-Yeniseian, and reinforce their Dene-Yeniseian pronouns, as well as converting large numbers of Emishi speakers to Japonic.
Some of these Emishi do not want to convert, so they flee to the north and settle into Tohoku, Hokkaido, Sakhalin, the Kuril islands, and become part of the Okhotsk culture. They give up whatever degree of bilingualism they had with Japonic, and start becoming monolingual in Emishi all over again, only now it's evolved to the state where it's Proto-Ainu.
And yet, all that time, those Dene-Yeniseian pronouns have remained, cuz those kinds of words are so hard to get rid of and swap in an entirely new set.
I'm not as certain of this part of the story -- it's a lot farther back, with far less evidence to base it on. And yet, how else do Ainu personal pronouns resemble nobody else's except for Wa-Dene-Yeniseian? If Ainu or Emishi was a Wa-Dene-Yeniseian language, then it's no mystery -- they share personal pronouns and much of the rest of their language with that family.
ReplyDeleteBut the rest of their language -- morphology, phonology, syntax, and most of the lexicon -- does *not* look Wa-Dene-Yeniseian. It looks Uralic. That stubborn core of the language, personal pronouns, likely reveals the pre-Uralic history of those islands. There may be very few other traces of that language in the Ainu lexicon, but pronouns are very stubborn, and they're a core part of the lexicon.
How ironic that, after switching from Wa-Dene-Yeniseian to Uralic, they should come under such heavy Wa influence after all! Most of them switched back to Wa-Dene-Yeniseian (Japanese), except for the hold-outs who fled north toward Hokkaido. But those Wa-Dene-Yeniseian pronouns could not be so easily scrubbed out from their linguistic history, and provide a faint yet keen glimmer into their far-distant cultural origins!
Another cultural shibboleth shared by Japanese and Yeniseians -- both identify as "people of the daylight", as part of their solar cult. All sorts of cultures worship the sun -- what's so unique? Well, not everybody who worships the sun refers to themselves as "people" or "children" of the "daylight", as opposed to "the sun".
ReplyDeleteFor example, Persians -- or maybe only the diasporoids -- refer to themselves as "children of the sun", harking back to their fire-worshiping Zoroastrian past. But they say "sun", not "daylight".
But several Yeniseian-speaking groups, like the Ket and their neighbors the Yug, refer to themselves as "people of the daylight".
https://en.wiktionary.org/wiki/%D0%9E%D1%81%D1%82%D1%8B%D0%BA
For example, the Ket endonym was "Kə́nasked", where the final element "ked" means "people". The first element, "Kə́n" means "light, bright". The entire name does not mean that their physical bodies are bright, glittering, etc. -- it means they are the people of light-ness, brightness, glittering-ness -- i.e., the daylight.
https://en.wiktionary.org/wiki/%D0%BA%D1%8A%CA%BC%D0%BD
The word for "light, bright" does not contain a morpheme based on their word for "sun", which is "ii", so this name is even further removed from names based on "sun".
https://en.wiktionary.org/wiki/Reconstruction:Proto-Yeniseian/x%CA%B7aj
In Japan, they do identify with a phrase containing the word for "sun", like "asahi", where "asa" means "morning" and "hi" means "sun" (from P-J "pi"). And we say that as well -- Land of the Rising Sun.
ReplyDeleteAnd yet, their main goddess in Shinto, back to the earliest Japanese texts, is named "Amaterasu", where "ama" refers to the sky or heavens, not the sun, and "terasu" from the verb for "to shine" -- not based on "pi / hi" = "sun". It means "Shining-in-the-heavens", without referring to the sun per se.
In fact, she was also known by a simpler and earlier name, "Hirume", and she was linguistically paired with a boy named "Hiruko", the first-born but malformed child of the creator gods Izanagi and Izanami, who cast out that child since it lacked bones and couldn't stand on its own.
Notice that both names share the phrase "hiru", and that the suffix "me" is a generic word for a female, and "ko" is a generic word for a son or boy. But what does "hiru" mean?
Due to some bizarre historical taboo, nobody has plainly connected the dots here. But you don't have to be a cliff-dwelling sage to figure this one out -- I'm just the first one to plainly speak this truth, while others refuse to, for some reason.
In the sources cited by Wiktionary for "hiru" in the names "Hiruko" and "Hirume", the assumption is that "hi" means "sun" and "ru" is an otherwise rare / unattested variant of the also rare genitive particle "ro". So, the names mean "girl / daughter of the sun" and "boy / son of the sun".
https://en.wiktionary.org/wiki/%E8%9B%AD%E5%AD%90
But why assume a rare variant of a rare word, for "ru"? There is a far simpler explanation -- that "hiru", or "piru" in the OJ form in which it was originally written, is that exact same word from P-J -- "piru", which means "daylight" or "daytime", and is contrasted with "yoru" = "night-time".
This is morphologically simpler, with "hi" and "ru" attaching to form a single word "hiru", and either "me" or "ko" attaching to that pair. "hi + ru + ko" has 3 elements with no grouping / nesting. And it's a more likely choice of words, not involving a rare variant of a rare word. And the interpretation is scarcely different -- they mean "daylight daughter" and "daylight son".
So now the connection to the Yeniseians is complete, referring to themselves not as simply people or children of "the sun" -- but of "the daylight". Unlike Yeniseian, Japanese does include the morpheme for "sun" inside the word for "daylight / daytime". But because it's hiding inside of a larger / nested phrase, "hiru", the connotation of "sun" is still pretty weak -- the main meaning is "daylight / daytime", and you'd have to think a bit, consciously, to think of the "sun".
This must be the origin of later focus in Tengrism, also from the Eastern Steppe, on the eternal blue sky. It's not really about a sky god, or the sky personified as a god, as in the case of Proto-Indo-European religion / mythology. It's about the sun shining clearly through on a cloudless day -- when the skies are blue, and it's daytime rather than nighttime. But although it's part of a solar cult, the reference to the sun per se is indirect and hidden. It's primarily about the bright, shining daylight.
Yeniseians, Huns, Turks, and Glorious Nippon -- all children of the daylight in a great big happy Eastern Steppe family. ^_^
Technically, "piru / hiru" refers to an interval of time during the 24-hour cycle -- noon, midday, etc., as opposed to "yoru" which is the evening, night-time, etc. It doesn't have an overt morpheme relating to "shine", like the Yeniseian phrase does, or like the Japanese name "Amaterasu" does.
ReplyDeleteBut what else are they possibly talking about, by contrasting "hiru" with other times of the day like "yoru"? The most salient thing that changes in the environment during the 24-hour cycle is the amount of light shining down from the sky. And "hiru" means "midday", when the most light is shining down.
The connection to "the sun" is a bit further removed -- you can clearly chart the course of the sun over the course of the light part of the day. And during midday, it's at its peak in the sky.
But that's not the main sensory impression we have of the 24-hour cycle -- it's the amount of light shining down, wherever it may be coming from. We are far less attuned to the position of the sun in the sky per se. You don't want to chart it too accurately anyway -- you'll hurt your eyes!
So the time-word "midday" is semantically interchangeable with "daylight" in this context, and does not involve the sun per se, only the amount of light shining down during noontime.
Also, think about who their implied opposition is -- is it "children of the moon", or "children of the night"? And even for the latter, is "night" so oppositional due to a different celestial body being in the sky, or cuz it's really dark? It's about the brightness level falling, not the changing of the guard in celestial luminary bodies.
ReplyDeleteIf they base their identity on the sun per se, then maybe their enemies are those who base it on the moon. But if it's about brightness shining down from the heavens, then they may very well worship the moon as well, just to a lesser extent since it's not quite so bright nor as reliable. And indeed they do worship the moon to a lesser extent, not treating it like some rival to their precious sun.
Morphologically, Yeniseian, Na-Dene, and Wa / Japonic treated the "moon" as the "night-sun", since they were grateful for this "source of celestial shining brilliance during the otherwise dark part of the 24-hour cycle".
It's straightforward for Dene-Yeniseian, and earlier I showed how "tukui" is a fossilized form of "night-sun" in the ancient history of the Wa language, not from the later Proto-Japonic stage. You can tell since the strategy to resolve the difficult labialized velar consonant was not to turn it into a labial stop, as in the "p" in P-J "pi", but to split its features into 2 phonemes, a velar stop and a rounded vowel, namely the "ku" in "tukui".
Ainu has the Indo-Euro word for "wolf"! I just have to share this one treasure now, before offering up the entire trove to the gods in the near future. And it's not just a "wow, that's neat" factoid, it actually resolves an open question in the reconstruction of Proto-Ainu!
ReplyDeleteWell, first I figured out that Ainu belongs to the Uralic family -- either as a yet-unnoticed branch stemming from P-U and being sister to P-Samoyedic, P-Ugric, etc., or as a far-eastern sister to P-U itself. I went on the hunt for cognates between Ainu and P-U or as far back in Uralic as I could find.
The first major attempt to reconstruct Proto-Ainu was Vovin 1993, and in it there are 2 unusual phonemes -- what he writes as "g" (voiced velar FRICATIVE), and "hd" (some weird back consonant). In recent work on P-A, Alonso de la Fuente has outlined 2 scenarios that these consonants could be. See his chapter in Bugaeva's Handbook of the Ainu Language (2022).
In scenario 1, "hd" was actually "kʷ" and "g" was "w".
In scenario 2, "hd" was "w" and "g" was "s".
Well, I found a lot more Uralic cognates using scenario 1, for both phonemes. The "hd" words turn up cognates if it's "kʷ", and the "g" words turn up cognates if it's "w". Hardly any cognates turn up if these sounds are "w" and "s", as in 2.
One of those "g" words is (simplifying the transcription), "gorkeu" = "wolf" in P-A. Most modern Ainu dialects pronounce it "horkew".
Since "g" is actually "w", this makes it "workeu". Aside from the final "u" that I don't understand, this immediately and obviously jumps out as borrowed from the Proto-Indo-Iranian word for "wolf", which is "wrkas", and "wrkah" by the time it's P-Iranian. The telltale sign is the "r", which used to be "l" in Proto-Indo-European and all other daughter's version of this word, like English "wolf". But the entire Indo-Iranian branch lost "l" and changed it to "r" way back in the Proto-II stage.
Several Uralic languages have borrowed this P-II word, which as I showed last year is behind the Slavic word for "wolfman" -- "vurkolak". That's cuz Proto-Slavic speakers used to be Uralic speakers, namely the Scythians and other neighboring groups from NE Europe and the Pontic-Caspian Steppe during Herodotus' time. It at least made it into the Mordvinic branch of Uralic, and that's the branch (especially Erzya) that most closely matches Scythian.
And who can forget my discovery that "vampire" is Uralic, meaning "soul-biter / soul-feeder"? As a lifelong fan of the Gothic, I'm still proud of myself for that one. ^_^
In the eastern branch of Uralic, Proto-Samoyedic borrowed it as "wərkə", but changed the meaning to another taboo / totem wild beast, "bear":
ReplyDeletehttps://en.wiktionary.org/wiki/Reconstruction:Proto-Samoyedic/w%C9%99rk%C9%99
Since the P-II donor form has no vowel between the "w" and "r", the inserted vowel in the borrowing languages remains underspecified / generic. In all but one of the Samoyedic descendants, the vowel is back / round, mostly "o" though one case of "u", surely due to the rounding influence of the preceding "w".
So, it's no surprise at all to see it being borrowed as "workeu" in Proto-Ainu, or more likely during its much earlier stage, which I call Emishi, when they would have been on the Asian mainland and in contact with either P-II itself, or with Samoyedic speakers who they could have borrowed it from second-hand.
However, given that it means "wolf" instead of "bear" in Ainu, they likely got it directly from P-II. How else would they know to change it from "bear" to "wolf", if they got it from Samoyedic? Unlikely.
Possibly they also got it second-hand from a further-west Uralic group, like whoever borrowed the word into "vurkolak" (e.g. Mordvinic). But that strains credulity as well -- Emishi / Ainu is as far east as you can get in the Old World, so it's more likely they got it from an eastern source. If not Samoyedic, due to the semantic double-change required, then from the Indo-Iranians themselves, right as they smacked into the Altai-Sayan region and its horse-riding nomads, and took a sharp turn due south, never to return to the chaotic Steppe ever again.
This would have been at a later date, since the P-II form has a final "s", while P-Iranian has final "h", and "h" is easier to lose or turn into a vowel / glide place-holder, as in "workeu".
Is it possible that they borrowed it 2nd-hand from Samoyedic, but before the meaning was altered to "bear"? Well, it means "bear" way back in Proto-Samoyedic, when it was borrowed, so that possibility seems to be ruled out.
That means that the homeland of Proto-Emishi was not in any Pacific island, but on the Asian mainland, and close enough to the Altai-Sayan region that they could pick up this word from the Indo-Iranians before they made a bee-line away from Mongolia. As far as we know, Indo-Iranian loans never made it east of that mountain pass -- only if the word was borrowed near there, and then spread by the borrowers further to the east.
That's what must have happened here. The Emishi speakers used to live near the southern part of Lake Baikal, they picked up this loan from Indo-Iranians on their final leg of their eastward trek across the Steppe, and later the Emishi migrated or spread their language due east, ultimately being spoken in Sakhalin, Hokkaido, and Eastern Honshu, while vanishing from the mainland.
How else did they get this Indo-Iranian word for "wolf"?!
This I-I word was not borrowed into Na-Dene, Yeniseian, or Wa / Japonic, BTW, so it reflects the Uralic-speaking stage of the Emishi, not their distant Wa-Dene-Yeniseian prehistory (judging from their personal pronouns), nor their later heavily Wa-influenced stage due to Japanese.
Fascinating stuff, isn't it?! ^_^ That's why you gladly and regularly take a break from your soul-crushing doom-scrolling and politicized bullshit to visit the Cliffs of Wisdom in the ruins of the blogosphere...
Glad I can keep you from killing yourself out of boredom for another week, hehe.
Ainu uses the Uralic number system, where 8 and 9 are derived from 2 and 1, respectively. It actually goes further and derives 6 and 7 from 4 and 3, as well.
ReplyDeleteAll the Finno-Ugric branches do this for 8 and 9, although not Samoyedic, which has its own system. In Finno-Ugric, 8 and 9 are only derived from 2 and 1 -- they don't include a morpheme for 10, as though to say "2 to 10" or "1 to 10", in the way we say it's "quarter to 2" for 1:45. They do have an extra morpheme or two as a suffix, meaning something along the lines of "remaining" until you're finished counting on your two hands.
This seems to be true for Ainu also, where 8 and 9 don't have the word for 10 in them, seemingly anyway. Just a suffix or two.
But the Ainu words for 6 and 7 actually *do* have the word for 10, with no connecting morphemes between them. So, 6 is just "4 10", and 7 is "3 10".
AFAICT, no other language families count this way for numbers under 10. Latin did at some point for 18 and 19 (2 from 20 and 1 from 20), but they had a word for 20 in there. The Finno-Ugric system doesn't include the end-point number, it just says "2 remaining / left / toward", and so does Ainu.
I spent all day tracking these things down, including looking at Indo-European. It's a pretty tricky to keep track of who borrowed what from whom -- and was it the abstract template or system, or was it particular words, or both?
Lots more to say in a bit. Just want to cement the linguistic and cultural link between Uralic and Ainu in yet another domain, and that they don't just share these systems, they are the only ones who do. They share a cultural and linguistic ancestor.
As for particular words, rather than the system of counting, Ainu has some Uralic cognates, although it seems like they're only the milestone ones, like 10 and 20. They don't share 1, 2, 3, or 4 with Uralic.
ReplyDeleteWell, even in Uralic, Samoyedic has almost completely different words from Finno-Ugric, and only shares a milestone word -- 5 in F-U is 10 in S (from P-U "witte" or "wixte"), and I think 20 in Ainu ("wot"). In Ainu, 10 is "kʷan" ("hdan" in Vovin), perhaps related to "kümmen(-ne)" = 10, from P-Finnic and P-Mordvinic, where the rounding on "kʷ" corresponds to the rounded vowel in "kü".
If anything, I think Ainu borrowed 1, 2, and 3 from.... Proto-Indo-European. Not Indo-Iranian, not Tocharian, some real early-ass stage of Indo-European.
In Ainu, 1 is "sine", but "s" before "i" alternates with "h", and in general "h" could have been "w" earlier on, due to Ainu's "w" -> "h" sound change (Alonso de la Fuente). So it's possible that "sine" used to be "wine" -- like P-IE "h₁óynos", where the "y" turned into "i", and the sequence "oi" was rendered as "wi" (typical), and the confusing laryngeal ignored by Ainu, as it is in Uralic.
In Ainu, 2 is "tu" -- just like "dwóh₁" from P-IE, except there's no voicing in Ainu, just like in Uralic. So "d" -> "t", and either the "w" or "o" or both become "u".
In Ainu, 3 is "re" -- just like "tréyes" / "tri" from P-IE, except initial clusters are banned in Ainu, just like in Uralic. So keep the consonant closest to the main vowel and junk the further-away one.
The only gap is with 4 -- but I've uncovered widespread tetraphobia, or fear / taboo about the number 4, far outside of the standard zone centered on China. Well, Ainu would be included in the East Asian zone of tetraphobia. But all sorts of Indo-Euro languages taboo-altered the first consonant
ReplyDeleteAnd even the P-IE word itself is suspicious -- "kʷetwóres", which has too many syllables to be a basic word. P-IE wants 1-syllable words, beginning and ending with a consonant, ideally. So "kʷet" could work as a base, but not the whole word. I doubt that either, though, since this complex word is probably a circumlocution to avoid the no-no number -- it doesn't have a secret word for 4 in it, it's based on some euphemism like "lucky victory" or something, IDK exactly.
Ainu has three forms of 4, two related and one distinct. The standalone number is "ine", while the combining form inside the word for 6 is just "i". There's a phrase for quarters, as in fractions, "pon emko", where "emko" means "piece, fragment, part", so "pon" must mean 4.
I'm guessing this "pon" is closer to the original Ainu / Emishi word, since it's inside a complex phrase, and as we know, taboo / banned words are easier to hide inside complex phrases, rather than exposed in the open as standalone words.
Where might "pon" come from? I'm guessing again it's borrowed from I-E, just like 1, 2, and 3.
Many daughter langauges' alteration of P-IE "kʷetwóres" change the "kʷ" to "p". Now, I'm assuming that way back when, Ainu / Emishi had "kʷ", so why didn't they borrow it as is? Maybe they didn't have this sound at the time of borrowing -- too late or too early, IDK enough about Ainu linguistic history right now. Or if they had, it would've sounded too similar to the P-A word for "to be bad", "kʷen" ("hden" in Vovin).
Or maybe Ainu also taboo-deformed that sound, just like the bona fide I-E languages did, due to tetraphobia. "OK, we'll borrow this word from you... but it's inherently evil, so just let us break it a little first, all right, there we go, it can't hurt us now..."
The "o" comes from either the rounding of the 1st consonant, or the medial "w" and/or "o".
I can't explain the final "n", other than an irregular change of the "r" for some reason.
The other alternative is that Ainu borrowed the P-IE word for 5 instead, which is "pénkʷe", and since that's a milestone word that they already had semantically in Ainu ("aski", based on "aske" = "hand", semantically similar to the etymology of the P-IE word). So they changed its meaning to fill the vacuum left by the taboo surrounding 4.
That accounts for the "p" and "n", while the "e" vowel changed to a rounded one at the same height, "o", under the rounding influence of nearby "kʷ".
That's a much better phonetic derivation, and the semantic shift isn't too hard to understand either. So I'll go with that for now.
Not only does this further establish Ainu's Uralic ancestry, it reveals some very early contact with a very early form of Indo-European -- pointing yet again to Ainu / Emishi origins being closer to the Altai-Sayan region and the southern shores of Lake Baikal, not entirely indigenous to the Japanese islands.
ReplyDeleteThanks to Peyrot, we know that the Tocharians were a fusion from a mostly Samoyedic local group converting to I-E based on visitors from the West, who were P-IE speakers, not Indo-Iranians or a late group like that. It was in the Altai-Sayan region, and then the Tocharians headed further east, unlike true Indo-Euros who never made it through that mountain pass, and wound their way down into the Tarim Basin.
Well, Ainu / Emishi contact with Proto-Indo-Euro speakers would have happened in the same place around the same time. It's just another example of an Eastern Uralic group meeting an early eastward off-shoot of the Yamnaya-type people.
Since the Ainu word for "wolf" has "r" instead of "l", we know they hung around long enough to meet Indo-Iranians. But these Indo-Euro numbers in Ainu look even older than Indo-Iranian, since the II word for 1 lost the nasal. So the Emishi / Ainu were hanging around there for quite some time, and headed eastward toward the Amur River and ultimately the Northern Pacific islands after picking up the II word for "wolf".
That wasn't in the 1st millennium AD or anything -- but it was more recent than everyone thinks, which is that the Emishi / Ainu / Etc. are an entirely indigenous, native cultural development. They're not -- somebody could have been there before the Uralic culture, but they either converted to the Uralic culture, and/or got demographically replaced by them, if the aborigines were hunter-gatherers, and the Uralic speakers had a more advanced subsistence mode.
By the time they're battling the Yamato in the mid-1st millennium AD, the Emishi are horse-riding archers. They weren't exclusively sea-faring hunter-gatherers, like the picture of the Ainu these days. They had a badass Steppe element in the good ol' days -- but after they were conquered by the Yamato, they joined them, since they were also horse-riding archers from the Steppe.
It seems like only the hunter-gatherer types among the Emishi decided to remain culturally apart from the Wa / Yamato / Nippon, even fleeing further north, where they ran into more small-scale hunter-gatherers around the Sea of Okhotsk, not another group of horse-riding agro-pastoralists.
And like I said before, perhaps even the pre-Uralic group in the Japanese islands spoke a Wa-Dene-Yeniseian language, strange as that sounds.
But it would be no different from the history of the Hungarians -- they began as Uralics, like all Northeast Europeans during Herodotus' time. Then around 500 AD they switched to Indo-Euro speaking, and Slavic was born. But then less than 500 years later, they get invaded by Uralic-speaking nomads, and they switch languages again -- but now returning BACK to Uralic all over again! Their Slavic phase lasted less than half a millennium, hehe.
ReplyDeleteBut the branches of Uralic were different -- the Southwestern branch of Uralic is Mordvinic, and that's what the Scythians spoke. Probably that's what the ancient ancestors of the Hungarians spoke. When the Magyars invaded them later on, they converted to the Ugric branch of Uralic, a whole different branch, albeit in the same family.
That could have happened in Japan. The natives started out perhaps as their own 4th branch of Wa-Dene-Yeniseian(-Early Jomon). Then they convert to Uralic when the Emishi show up, and Ainu remains Uralic. But most of them convert to Japanese after invasion by the Wa, so they wind up speaking Wa-Dene-Yeniseian all over again -- but in a different branch, Wa / Japonic, not whatever their earlier distinct branch was.
Last crucial detail: those suffixes that derive 8 and 9 from 2 and 1, in Finno-Ugric, are either "-ksa" or "ksan".
ReplyDeletehttps://en.wiktionary.org/wiki/Reconstruction:Proto-Uralic/kakteksa
If it's "san", that's also the same suffix as in Ainu for 8 and 9 -- but not 6 and 7, which have an overt word for 10, not just suffixes.
Ainu 9 is "sinepesan", and 8 is "tupesan", from "sine" and "tu". The "-p" or "-e" suffix could correspond semantically to the "-k-" suffix in Finno-Ugric, although obviously they're phonetically different (unless both descend from "kʷ"...)
But after that suffix, it's "san" in both cases -- phonetically identical to their counterpart in Finno-Ugric. They're cognates!
Vovin tries to stick the word for 10 in there instead ("hdan" in his writing, "kʷan" in my choice among Alonso de la Fuente's 2 scenarios, and "wan" in the modern Ainu dialects). That is on analogy with the words for 6 and 7, which *do* have "wan" instead of "san", and do not have the "-pe" suffix, only "iwan" and "arawan".
So perhaps the Ainu words for 8 and 9, with final suffix "-san", are fossilized forms from the very old suffix that they share with Finno-Ugric. This explains why it's only 8 and 9 -- those are the only numbers that Finno-Ugric derives in this "X remaining" way.
When Ainu took this logic further, to derive 6 and 7 from 4 and 3, it had no inherited forms from elsewhere in the Uralic family, unlike for 8 and 9's suffix. So they made it explicit this time -- 6 is "4 10" and 7 is "3 10", with no connecting affixes whatsoever, a morphological simplification over the old template that they share with Finno-Ugric.
That's why Ainu has 2 different pairs of subtractive number words -- 8 and 9 reflect the Proto-Uralic or very old Uralic form, with one or more suffixes and no overt word for 10, while 6 and 7 reflect the later Ainu-specific branch's attempt to extend the subtractive method to the other numbers between the milestones of 5 and 10. But since they were doing this on-the-fly, they didn't have all those funky, fossilized suffixes to inherit, and they just stuck two numbers together to derive a third number, more like a brute-force compound than a derivation using suffixes.
But that doesn't deny the cognate status of those suffixes for 8 and 9, between Ainu and Finno-Ugric! Too many shibboleth-y similarities for them to be coined independently -- and remember, Ainu speakers and Finnish or Erzya speakers have never been in contact with each other. Ergo, they share a common ancestor! ^_^
Some final details on Ainu & Uralic numerals. Now that I've seen subtractive number systems everywhere, I actually think the Ainu word for 4 is based on the word for 1 -- semantically, "1 left / remaining / toward the milestone".
ReplyDeleteSince 1 is "wine", 4 = "ine" can be derived from it by alteration, namely deleting the first consonant, which is already a weak consonant, as a glide. Or if 1 had already shifted to "hine", same thing -- initial "h" is pretty weak, easy to delete.
As we'll see later, I don't think it's likely to have a pair of derived numerals on the same side of the "one hand" milestone divide. 1 and 4 appear too close together in sequence, that it'd be confusing to derive 4 from 1. But if it's on the other hand, on the other side of the 5 milestone, then deriving 9 from 1 isn't so confusing.
But in this case, there's been a major alteration to 1, by removing its initial consonant -- now it's not quite so similar-sounding, and less likely to confuse you while counting on a single hand, to express 4.
Tetraphobia shows up even more when 4 is used to express 6, as "4 10", where "ine" gets whittled down to just "i" (in "iwan"). Ainu was clearly not afraid of deleting segments or entire syllables to generate variant forms of numerals.
This use of "wine" to generate "ine" (getting 4 from 1), happened after the original Ainu / Emishi word for 4, "pon", was erased due to tetraphobia. A sign that this was a late event is that it doesn't have the suffixes like "-pe" or some other endpoint like "san". This is an analytic rather than synthetic strategy, just like the word for 6 and 7 having no affixes, just a root-root compound "4 10" and "3 10". Analytic stages of language tend to happen later, and multiple layers of affixation is a sign of earlier history.
Finally, modern Ainu 20 = "hot" came from P-A "wot", which I said was cognate with P-U "witte / wixte". I favor the P-U form being "wixte", since that introduces some rounding near the 1st vowel, especially if "x" is actually labialized as "xʷ". Here's a recent attempt to simplify derivations from P-U into the daughter branches, by proposing "xʷ" as a phoneme in P-U, as the realization of the mystery back consonant "x".
ReplyDeletehttps://www.academia.edu/164962051/The_case_for_a_labialized_velar_fricative_xw_in_Proto_Uralic_draft_version_1
The meaning of the P-U word seems to be "closing the hand into a fist", since it sounds similar to verbs related to grasping, holding, carrying, etc., which involve curling your fingers and thumb around whatever you're seizing, carrying, etc.
The main suspects here both have "x" -- "wixe" = "to bring, to take somewhere, to drag, to pull, etc." and "wexe" = "to take, to grasp". IDK what the suffix "-t-" means to go from those verbs to the numeral noun meaning "closed fist".
https://en.wiktionary.org/wiki/Reconstruction:Proto-Uralic/wixe-
In the Finno-Ugric branches, "wixte" means 5, but in P-Samoyedic it was shifted to mean 10 -- perhaps thinking of "seizing with 2 hands" instead of just 1. Or based on counting to 10 on just one hand, by counting all 5 fingers in one order, then reversing that order to go from 6 to 10, which is a common way to count in East Asia. So the "fist"-based number was actually 10, not 5.
In Ainu, this was shifted further to mean 20, a new milestone, but a natural end-point as well. Maybe counting on both hands, each time going forwards and backwards a la East Asia, so that both hands being closed in order to seize, carry, etc., happened at 20.
That opened up room for the 5 milestone in Ainu to be based on the simpler noun "hand", not necessarily a closed fist, a carrying type of hand. That's where 5 = "aski" comes from, "aske" = "hand" (not necessarily closed or curling the fingers or thumb).
If you start counting with both hands closed, then at 5, one of your hands is open -- so you definitely don't want to say "fist", but the generic word for "hand", which we assume is with the fingers and thumb open. Then you reverse the order to get a closed fist on 10, then repeat this process on the other hand, so that 15 has the other hand open, and reversing the order yet again on that hand, at 20 both hands are closed into a fist again -- which is where the "seize / grasp / fist" word comes in.
Major mystery solved for numerals in Indo-Euro, Uralic, Semitic, and Kartvelian! I wasn't even planning on going outside of Ainu, but once I saw it had adopted P-IE numerals, I figured I might as well snoop around there as well. And holy moly, did I stumble down a secret passageway! Good thing I brought my torch... ^_^
ReplyDeleteThis is all about the P-IE word for 4, which as I said is way too many syllables to be a basic original word. Probably a circumlocution to avoid the old word, which was replaced due to tetraphobia.
First, what others before have intuited, discovered, and argued, and then my solving of the puzzle.
The P-IE word for 8 is "oḱtṓw", and its long "o" vowel and "w" hint that it has a dual suffix, meaning "two of something". What is the singular form? From P-IE morphophonology, it must be "oḱto", and semantically it must refer to "four" somehow.
https://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-European/o%E1%B8%B1t%E1%B9%93w
But where does this word relating to 4 come from? It has been systematically erased in all daughter languages, and taboo-replaced by the circumlocution "kʷetwóres".
https://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-European/k%CA%B7etw%C3%B3res
So does it mean anything in native P-IE roots? Or was it a borrowing? And if so, from where? And where did *that* language get it from -- did they borrow it? (if so, from whom?), or did they coin it themselves (if so, what does it mean in their native roots?).
You see, the main problem is that 4 is not a primitive numeral in any language -- they have 1, 2, and "many". Maybe logarithmic milestones to mark orders of magnitude, like 5, 10, 20, 100, 10000, etc. But no language has words for fine-scale, precise integers between 1 and 10. Even though 1 and 2 are primitive, and 5 and 10 may be milestones, that still leaves 3, 4, 6, 7, 8, and 9.
Somebody has to coin new words for these non-primitive numbers, and they must have mneumonic value -- so they are easily understood and catch on in that language. From there, maybe they get loaned into other languages, whether the borrowers understand the original etymologies or not.
So in looking at who borrowed what from whom, the buck has to stop somewhere with an original coinage of a non-primitive number, with a clear etymology in the language that coined it. Or it traces back to a primitive number, but used in a new way -- like expressing non-primitive 4 by means of primitive 1 (with the mneumonic being "1 to the milestone").
Nobody has put these pieces together yet, as they span so many language families and involve trying to reduce everything to primitive numbers, which is not intuitive to us since we take all these integer words for granted. But believe me, it wouldn't been TOTALLY intuitive to people in 2000 BC!
So let's dust off this magic carpet and explore this whole new world of numeral words! ^_^
My key original insight is that ancient languages expressed non-primitive 4 by means of primitive 1, meaning "1 to the milestone". No one has figured that out -- but in fairness to them, sometimes the word for 1 is borrowed from another language. That way, it doesn't create confusion *within* a single language -- you don't express 4 by means of your own primitive word for 1, but by a borrowed word for 1. Or, as I showed with Ainu, by notably altering your own word for 1.
ReplyDeleteBy borrowing someone else's word for 1, you don't have to use a complex mneumonic like "this is the finger that sticks out the most" (which is how 3 was coined for P-IE, referring to the middle finger, which is 3rd in sequence if you start with the pinky finger).
And who exactly is borrowing all these number words anyway? And what for? It's not hunter-gatherer societies, nor small-scale chiefdoms -- it's sedentary civilizations, with centers for commerce, taxation, currency, scales that measure things being bought and sold, how many slaves you want, court astronomers counting the number of days in the week or month, and so on. That is where the need for fine-scale integers between 1 and 10 come from.
And in such environments, there's likely to be a polyglot group of people, who buy, sell, and trade across national and linguistic borders. The everyday masses of these language families may rarely encounter people who speak foreign languages. But in the places where precise measurements are needed, it's very likely to attract a multi-ethnic group of buyers, sellers, patrons, specialists, and so on.
In such environments, they can share their primitive number words -- "In my language, 1 and 2 are bla and bloo." "Ah, interesting, in my language, 1 and 2 are da and doo." Well now, each language has already covered the primitive numbers of 1 and 2, and they each have their own milestone word for 5, which is probably related to hand, fist, grasp, seize, etc. But now they can express the non-primitive numbers 3 and 4 by borrowing each other's words for the primitive 1 and 2. Namely, use the loans for 1 and 2 to express 4 and 3 in your native language.
So the first guy's language would express 1 through 5 as:
bla, bloo, doo, da, hand-word
And the second guy's language would express 1 through 5 as:
da, doo, bloo, bla, hand-word
Then if they need to extend this logic to cover non-primitive 6 through 9, they probably need to wait awhile for 1 through 5 to regularize and get familiar. After that, they can simply express 6 through 9 as 4 through 1, perhaps with an affix to mean "until / to / toward / remaining / left", like Ainu and Uralic 9 being 1 with affixes, or perhaps including an overt word for 10, like Ainu 6 being the compound "4 10".
Hypothetically, this generates words for all non-primitive numbers, without the need to coin anything by referring to some trait it has, or how it's used in daily life, etc. The only source of new-ness is borrowing from another language -- and even then, from their *primitive* words for just 1 and 2.
The only coined words are the milestones 5 and 10 (and maybe 20), which are very intuitive -- hand, fist, end, etc. That's the easy part, coining a word for 4 or 7 or 9 is the hard part. So why coin it? Commercial types are not exactly poets known for their insightful figures of speech -- just buying and selling and haggling.
And how fitting that these primitive number words are traded between the parties, just like bartering! ^_^ Both come out of the deal with the highly valuable non-primitive number words, and all without having to wrinkle their money-grubbing brain to come up with a figure of speech.
That's not to say that non-primitive numerals cannot be loaned -- only that, it is not necessary, and we should try to look at simpler strategies before considering complex ones. Unless the answer is obvious -- like how P-IE borrowed their word for 7 from one of the languages spoken in the Cradle of Civilization, such as Semitic or Hurrian.
ReplyDeletehttps://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-European/sept%E1%B8%BF%CC%A5
That's a useful heuristic -- civilizations and networks of civilizations probably coined non-primitive numerals using figures of speech, whereas small-scale and semi-nomadic societies likely borrowed them from civilization rather than coining them on their own.
OK, so where does the original Indo-Euro word for 4 come from? Well well well, how perfectly everything falls into place when you have the right perspective! That's why I dwell up here in the sublime cliffside, after all -- there's no view like the one up here! ^_^
ReplyDeleteWith the insight that 4 can be expressed as 1, we look for a word for 1 in Indo-European... and find nothing. There are two words, but neither are remotely like "oḱto", the supposed pre-replacement word for 4 -- they're "h₁óynos" and "sḗm".
But in the international marketplace of word-bartering, Proto-Uralic has just what you're looking for, stranger! Step right on up, lookie here and marvel at our own primitive word for 1, which won't confuse your Indo-European ears when used to express 4!
It's "ükte"!
https://en.wiktionary.org/wiki/Reconstruction:Proto-Uralic/%C3%BCkte
Now is that a perfect match for "oḱto" or what?! Round vowel, not low, only backed since P-IE doesn't have front-round vowels. In the Uralic descendants, there's variation between high and low vowels, so "o" is not surprising. Also, P-IE wants only "e" and "o" as vowels, whereas high vowels "i" and "u" are more likely allophonic variants of the glides "y" and "w". And there's no glide in the P-U word, so don't use a high vowel when borrowing -- make it "o".
Perhaps the original "k" was shifted a bit forward to palatovelar "ḱ" in order to preserve the front-ness of the preceding vowel, so that info didn't get totally lost.
The "t" is fine.
And the final vowel could have been preserved as "e" as well -- we don't have access to it, only the dual-ized form for 8, and the dual uses a suffix, which could have obscured the final "e" of the singular form. Or it was originally borrowed with final "e" altered to "o" since final "o" sounds more typical of P-IE nouns vs. final "e".
Eureka! ^_^
Now, the connection to Proto-Kartvelian (Georgian's family). They have a similar word, and it's still used as 4, as in the supposed old P-IE word. It shows the opposite order for the 2 consonants, so whoever borrowed it used metathesis. It's "otxo":
ReplyDeletehttps://en.wiktionary.org/wiki/Reconstruction:Proto-Kartvelian/otxo-
If you knew nothing about subtractive counting to derive non-primitive numbers from primitive numbers, and knew nothing about Proto-Uralic, you would just notice the similarity between the Indo-Euro and Kartvelian words, and couldn't tell who borrowed it from whom.
But again, the buck has to stop somewhere with these number words -- either a primitive number is arrived at, or a coinage using native roots.
If we assume I-E borrowed it from Kartvelian, the buck doesn't stop there. In P-K, this word means 4, and that's not primitive. And as far as anyone can tell -- and they've looked at this similarity for well over 100 years -- there is no etymology using native Kartvelian roots that can explain why their word for 4 has this phonetic form.
Before I solved this mystery, though, the other side could say the same thing. If Kartvelian borrowed it from I-E, it's still a non-primitive number in I-E, and there is no convincing etymology for it using native I-E roots (it's a horrible I-E native word, beginning with a vowel).
If there are no native etymologies on either side, it means neither one of them coined it. It must be borrowed, or involve borrowing -- both from the same donor, or one borrowed it from the donor and the other borrowed it second-hand.
And that's where Uralic comes along and saves the day, once again for ancient linguistics! You really can't understand jackshit about the ancient world, yes even in NE Europe like with the Scythians, without having a decent understanding of the Uralosphere. Even today they stretch all over the place -- including Ainu! But it was far more extensive southward in the BC millennia.
It makes perfect sense to say I-E borrowed the Uralic word for 1, used to express 4 in I-E with no confusion from their own native I-E word for 1. And the phonetic match is perfect -- no metathesis involved. The only metathesis involved is between I-E and Kartvelian, which means Kartvelian borrowed the I-E word, since there's no plausible source of Kartvelian borrowing it from outside of I-E.
If we assumed that I-E borrowed it from Kartvelian, the metathesis would make sense -- but then it would be a totally unexplained coincidence that this I-E word, metathesized from Kartvelian, just so happens to match the Uralic word for 1 -- after applying metathesis -- and that 1 and 4 are related to each other in subtractive counting.
To eliminate the coincidence, it must be that I-E borrowed 1 from Uralic, with no major alterations, and then Kartvelian borrowed it from I-E with a notable metathesis.
Also, geographically I-E is in between Uralic and Kartvelian, which is a Southern Caucasus family.
Finally, the Semitic connection, and did the erased I-E word leave any traces in I-E somehwere?
ReplyDeleteThere is a claim owing to Henning 1948, who studied measurement words in the ancient Iranian language Avestan, that there's a descendant of P-IE "oḱto" in Avestan, namely "ašti". This is a unit of length, it doesn't mean "4" by itself. It refers to the breadth of the hand across the palm -- so for all we know, it doesn't involve a numeral at all, just a body part word like "palm" or something. But he interprets it to mean "the span across 4 fingers", in order to make the connection between that word and P-IE "oḱtṓw" = 8.
Nothing in my solving of this mystery hinges on the outcome of this matter, about Avestan "ašti". That word only relates to whether or not the mass-erasure of P-IE "oḱto" was 100% effective, or left a faint survivor somewhere. It doesn't relate to the source of the P-IE word for 4 -- was it borrowed or coined, if borrowed from whom, if coined with what etymology, etc.
But since I wandered down one secret passageway, I figured I might as well wander down another and unearth that one's treasure too -- to share with the world, of course, not horde it all to myself. ^_^
I actually think this is from a different source, a loan from Semitic. In Proto-Semitic, "1" is "ʕašt-" -- that first consonant is the voiced pharyngeal fricative, which begins the word for "Arabic".
https://en.wiktionary.org/wiki/Reconstruction:Proto-Semitic/%CA%95a%C5%A1t-
The phonetic alteration from this word to the Avestan one is trivial -- the uber-Saharo-Arabian sound "ʕ" is simply deleted by Indo-Euro speakers, as it is when we say "Arabic". Every other sound is retained exactly, and the final "i" in Avestan could have come from the original final vowel, which varies across the potential donors, but includes a high-front vowel in Akkadian, which is likely where they got it from.
Akkadian is East Semitic, closest to ancient Iran, it was used for civilizations and empires (Babylonia, Assyrian Empire), and would have been in contact with Avestan at least in international commercial or political / diplomatic contexts.
Although the phonetic derivation is 100% for the Avestan word being a loan from Semitic, it does carry a tiny semantic wrinkle -- namely, borrowing 1 in order to express 4 natively. But as shown way back in P-IE (and later, in Ainu), this is a natural strategy, including for speakers of I-E languages. Avestan simply repeated what their P-IE ancestors did, only borrowing 1 from Semitic rather than Uralic.
Neither theory assumes that "ašti" meant the cardinal number 4 in Avestan -- it's only attested as a unit of measurement, which may relate to the width of 4 fingers. Given the connection to measurement, I'm inclined to think it sprung up in an international mercantile context, not inheriting an otherwise wiped-out word from P-IE.
Realistically, though, the Semitic word has a more natural semantic value for Avestan "ašti" -- it did not refer to 4 in Avestan, but to 1, just like in the donor language, and was simply used to mean "1 palm-width" or "1 hand wide". It's a unit, not a grouping-of-4.
So I reject that the Avestan word has anything to do with 4 anyway, it's just 1 all the way down, not "1 used to express 4".
But just in the remote case that it does relate to 4, it's a borrowing from Semitic, not an inheritance from P-IE.
Semitic to the rescue, yet again in the ancient Cradle of Civilization! And that concludes the magic carpet ride for tonight... ^_^
More on Uralic subtractive counting and loans from Indo-Euro, and vice versa. First, when P-IE borrowed the Uralic word for 1, in order to express 4, it's possible that they were also told about its role in the Uralic word for 9, derived from 1.
ReplyDeleteMaybe the Uralics said that the word could be borrowed with the meaning of "1 until", just like their existing word for 9, so the Indo-Euros didn't have to re-invent this wheel. They simply borrowed the Uralic word with the meaning of "1 until" rather than "1 per se".
And since any suffixes expressing the "until, toward, remaining, etc." were in a foreign family, they didn't borrow importing those as well. If they intended to borrow the Uralic word for 9, then they would probably have borrowed the whole thing, with suffixes and all, since they're just interested in "9 per se". But they weren't doing that, they just heard about how Uralics use 1 to express 9, so the Indo-Euros figured, why don't we do that to express 4? Mind if we borrow your word for 1, meaning "1 until"? Thanks!
Second, it's possible that Finno-Ugric (not including Samoyedic) later borrowed the P-IE replacement word for 4, "kʷetwóres". This would have had to wait until P-IE replaced their Uralic-inspired word for 4. But since Samoyedic isn't part of this trend, it does show that it's a later development in Uralic.
In F-U, 6 is "kutte" -- suspiciously like "kʷetwóres", with Uralified phonotactics. There's no "kʷ", so preserve the rounding by rounding the following vowel, either "o" or "u" -- since there's a nearby glide "w", which exerts more of a "u" than an "o" influence, go with "u". There's the "ku-".
The "t" is fine.
The "w" after the "t" is not allowed in Uralic. Medial "w" either has to be intervocalic, or the 1st of a sequence, with only 3 exceptions -- and to match the traits of "w", none of these follows a stop or a voiceless consonant ("kodwa", "käďwä", "tälwä" -- "d" is fricative and voiced, "l" is continuant and voiced).
So just geminate the 1st member of the sequence, for "-tt-" instead of awkward "-tw-".
P-U words want to be 2 syllables with no consonants word-final, so pick a final vowel -- looks like the "e" from the final syllable of "kʷetwóres". At least in this case, the strategy was keep the first and last syllables' vowels, to capture its overall shape from one end to the other, and delete the internal vowels.
Semantically, Uralic borrowed some other family's word for 4, with the meaning of "4 until" in their own -- i.e., to express 6.
Interestingly, Samoyedic likely borrowed the I-E word for 4 on their own, but to mean 4 rather than 6.
https://en.wiktionary.org/wiki/Reconstruction:Proto-Samoyedic/t%C3%A4tt%C9%99
It was a later borrowing, not from P-IE but possibly Tokharian. That makes sense -- by the time P-IE replaced its Uralic-inspired word for 4, with a native circumlocution, Samoyedic and Finno-Ugric had begun to diverge from each other. As it turns out, each side borrowed an Indo-Euro word for 4 -- but not at exactly the same time, from the same proto vs. branch language, and not with the same meaning (4 per se vs. 4 until).
I told you this stuff gets tricky to track down!
Sempiternal Semitic Springtime! Now for subtractive number words in Semitic, Egyptian, and other Saharo-Arabian branches.
ReplyDeleteHaving learned about 8 in P-IE having a dual suffix, and therefore being based on "a pair of 4's", I turned my attention to Semitic -- and BINGO, 8 also has a dual suffix! It's "ṯamāniy", where that long "a" with or without the following "n" (as in Arabic) looks like a dual suffix. The final "-iy" is just a generic suffix like an adjective, so don't worry about that too much.
I thought, "Eureka! That must mean that Semitic used to have a word for 4 that was 'ṯam', and then they stuck the dual suffix on it, to generate a new word for 8, just like P-IE!"
And in Egyptian, there's a similar-looking word for 8, "ḫa-ma-an", with another long "a" vowel followed by "n" after the 1st element "Cam".
https://en.wiktionary.org/wiki/Reconstruction:Proto-Semitic/%E1%B9%AFam%C4%81niy-
https://en.wiktionary.org/wiki/%E1%B8%ABmnw#Egyptian
The 1st element now begins with "ḫ", but maybe that corresponds to "ṯ" in Semitic. The Semitic sound is like English "th" in "thin", and the Egyptian sound is either velar or uvular fricative, voiceless. Those are at polar opposite ends of the mouth, which made me think of the realignment of the sibilants in Spanish -- there were too many sibilants and affricates all bunched together in the center of the mouth, so the strategy was to send some of them far forward in the mouth, like "th", and send others way back into the throat, like "x" or really uvular "X" in Iberian Spanish.
So maybe these both reflect a voiceless fricative that was in the center of the mouth, and due to too many sibilants and affricates -- and emphatic consonants -- in the center, Egyptian sent it toward one extreme (back), while Semitic sent it toward the other extreme (front).
I thought it would be like "sam" or "sham". And whaddaya know...
In Cushitic (the branch that includes Somali), which did not pursue the polarizing strategy, this word retained its consonant being in the center. In Proto-Cushitic, there's a number word transcribed as "lama" but phonetically more like "ɬama", with a voiceless lateral fricative. That's just what I was thinking of!
https://en.wiktionary.org/wiki/Appendix:Proto-Cushitic_reconstructions
But it turns out this does not refer to 4, as I was expecting. In fact it means 2 -- and it is the word bearing the dual suffix, that final "a". What is this word the dual of? Probably "finger", which is "lom" and may be analyzed as "law-m". That's the only way to make these words in Cushitic connect.
I can't tell if Egyptian and Semitic borrowed this Cushitic number word, and then their native sound changes altered it into beginning with "ḫ" for Egyptian and "ṯ" in Semitic -- or if they had their own cognate of this word, which has since vanished or remains obscure at any rate, and applied their sound changes to their own cognate of the Cushitic word.
It seems like they borrowed it from Cushitic, since its dual suffix is present in all 3 words, whereas Egyptian's dual suffix is "y" (more like an "i" vowel, not "a").
After borrowing the Cushitic word for 2 (with the Cushitic dual suffix intact), both Egyptian and Cushitic employed this word for 2 to express 8. That means Egyptian and Semitic are using subtractive counting -- expressing 8 via 2, perhaps with some suffix like the generic adjective suffix in Semitic, as though to say it's "2-ish", that is not quite 2, not "2 per se".
Well, it can't mean a pair of something -- that would trigger the dual suffix, not the cardinal number relating to 2. So it must mean "2 until / toward / remaining", i.e., 3 or 8. In context, when Semitic and Egyptian already have words for 3, it must be 8.
*both Egyptian & Semitic employed...
ReplyDeleteCushitic is a wonderful help here, cuz it's proto-language only has numerals from 1 to 5, not 6 to 9 (it does have milestone 10). So we can rule out this word meaning "2 until", i.e. 8, in P-C. There are no words at all for 8 in P-C. Egyptian and Semitic borrowed it, and applied their own reasoning to make it "2 until", since it only meant "2 per se" in Cushitic.
ReplyDeleteI said the Egyptian dual not being "a" but "y" means Egyptian likely borrowed it from Cushitic... but perhaps way way way back when this word existed in Proto-Saharo-Arabian, it came with a dual suffix based on "a", which was retained in Cushitic and Semitic, while Egyptian innovated a new vowel.
But it's really not that important how exactly it got into Egyptian and Semitic -- perhaps from a native root in the parent proto-language, P-SA, or perhaps borrowed from a sister branch, Cushitic. That's not really across language families, like with Uralic and Indo-European, so I don't think it bears on the matter of borrowing numbers that sound totally strange, unintelligible, and meaningless, across language families.
Another Saharo-Arabian numeral mystery solved, thanks to Cushitic. It's 4 to express 6, or "4 until" -- that can never mean "4 until 5," since what kind of deranged freak would say "4 until" to mean 1?! That's the most fundamental number word in any language! No, "4 until" can only mean 6.
ReplyDeleteAnyway, P-C has a word for 4, "salc", which cannot be primitive, since nobody's word for 4 is. But I'm not interested right now in where this came from. I want to see what it looks like in the other Saharo-Arabian branches.
Well, in Egyptian, Semitic, Berber, and Chadic, its cognates mean 6, i.e. "4 until".
https://en.wiktionary.org/wiki/sjsw#Egyptian
We can't assume that just cuz there are 4 branches on one side, and 1 branch on the other, that the majority reflects to old ancestral state. That is impossible in this case -- that would imply using a word for 6 to express 4. But remember, P-C doesn't have *any* words for 6-9, whether native or borrowed, whether using this mneumonic or that one!
And even in languages that do have 6-9, don't you think they already have a word for 4? Unless you wanted to replace 4 due to tetraphobia, and then use 6 to replace it. But that involves too many Uno reverse cards about "until until until..." There is no way in the world anyone would express 4 by means of 6.
So what's more likely is that in the 4 branches that do have numerals between 5 and 10, they employed an old word from 1-4 to express a number between 5 and 10. But having two doublets among numbers 1 to 9, can sound a little funny, and confusing if there are no affixes to clarify "no, not the number per se, but that number UNTIL, or that number-ISH".
So rather than keep the original numeral from 1-4, they replace it, and keep the member of the doublet that has been generated in the 6-9 range. Still, it seems like 4 is most likely to get replaced, so I think tetraphobia is at work here.
Unlike the case with 2 to express 8, the various words for 6 in Egyptian, Semitic, Berber, and Chadic are not close copies of the P-C cognate. True, they could have borrowed a single donor, and applied their own distinct sound changes on it. But they're pretty wild for that to be true. More likely, each language had a cognate of this word for 4, inherited from P-SA, and that's what they applied their sound changes to.
Formally, when using X to express Y subtractively, X must be less than the next milestone after Y.
ReplyDeleteThe reason is, you don't want to burden your brain too much with calculations, while it's learning these number words. The further you have to count backwards, the more it hurts the brain. Pick the milestone that's closest to the number you want to express, and count backwards from there.
Imagine having 1, 5, 10, and 20 as milestones, but expressing 3 as 17 or 17-ish or bizarro 17!
Naturally, I'm going to claim that 7 is derived from 3! Here are the cognates for 7 in Semitic, Egyptian, and Berber:
ReplyDeletehttps://en.wiktionary.org/wiki/Reconstruction:Proto-Semitic/%C5%A1ab%CA%95-
Turning to good ol' Proto-Cushitic, there's a word for 3 that's variously reconstructed as "sazħ", "saK", or "sedeh".
This isn't quite as straightforward phonetically, although the semantics are perfect (adapting "3 per se" to "3 until", i.e. 7).
The "sazħ" form has the proper 1st and 2nd sounds (with the known sound change of Semitic making "s" into "š").
The final one is nearly the same as in Semitic, but with Semitic voicing it -- perhaps on analogy with Semitic's replacement word for 4, as well as 9, also ending in the voiced pharyngeal fricative ("ʔarbaʕ" and "tišʕ"). Or it was already voiced in Semitic's reflex of the P-SA ancestor.
The only real problem is Cushitic "z" corresponding to "b" in Semitic (and "f" in Egyptian and "b" in Berber). The others have a labial-involved consonant, which is a stop or fricative, but Cushitic has an alveolar sibilant. The voicing matches Semitic and Berber, but not Egyptian.
IDK why it's "z", but we'll just chalk that up to the difficulties of reconstructing numerals for the proto-stages of any branch in Saharo-Arabian, including the Proto of the whole family. Given the examples of 4 to express 6 and 2 to express 8, it's pretty safe to assume this is another example, where 3 is expressing 7, and only Cushitic retains a reflex of the original 3 word (or an easily identifiable reflex, at any rate).
Finally, 1 used to express 9. First, a wrinkle: although there are several similar words for 9, in a Chadic language it actually means 7.
ReplyDeletehttps://en.wiktionary.org/wiki/ps%E1%B8%8Fw#Egyptian
Whether additive or subtractive, 7 and 9 can't rely on the same number. 7 is 2 away from 5 and 3 away from 10, so it would adapt 2 or 3. 9 is 4 away from 5 and 1 away from 10, so it would adapt 5 or 1. No common numbers between them, so this can't be using the "X to express Y" method, whether additive or subtractive.
How *can* 7 and 9 coincide on the same word? If the method of counting is on the hands, and you fill up one hand before moving to the next (rather than 4 fingers, 4 fingers, 1 thumb, 1 thumb). 7 and 9 both lie on the 2nd hand.
The natural place to start counting is either end of the hand -- thumb or pinky finger. If you start with the pinky, then 9 lands on the pointer finger. If you start at the other end, the thumb, then 7 lands on the pointer finger.
So probably these cognates relate to the pointer finger, whatever the allusion is (pointing, signing, shaking, wagging, stabbing, poking, etc.).
Is this a reference to the number 1? By assumption here, it's not the finger you start counting in sequence on, but the thumb or pinky.
But perhaps it's a much earlier reference to holding up just 1 finger in the air to mean 1. Which finger would you extend? Probably the pointer finger. So maybe there is a circuitous route to this being an example of using 1 to express 9 after all (for the languages where this means 9, not 7).
Then in comes Cushitic to the rescue -- there's a word for 1 that's "laħ / liħ". Not about the pointer finger necessarily (or maybe it is, deriving from "lom / law-m"), but a cardinal number.
It has the same final "ħ" as we saw with 3 / 7, and again the Semitic word has it voiced, for the same reason.
The "l" is an alveolar and not a stop, and if it were a lateral fricative, even closer to the "š" in Semitic (which in P-SA would've been alveolar "s").
The "i" vowel is shared.
The only problem is what's the initial "t" doing in Semitic? Maybe it's attaching to its cognate of Cushitic "laħ / liħ", but interpreted more as a verb than a noun / adjective. And the "t" prefix for verbs plays a medio-passive role -- reflexive, reciprocal, that kind of thing. Some kind of mirroring involved -- perhaps to suggest that this 1 is really a mirror-image 1, where there's only 1 left, or where the pointer finger is the only finger left on the two hands. That is, using "mirror-reflected 1" to express 9.
That's my best guess, but I think this example counts as subtractive counting words as well. And only thanks to Cushitic can we make sense of all this craziness! ^_^
*9 would adapt 4 or 1.
ReplyDeleteWhat was the psychology behind subtractive counting? I think it makes better sense to call it "complimentary" counting or "reflected" counting, and that it was concrete rather than abstract. Prehistoric people were not math nerds, they were as physical and concrete as possible.
ReplyDeleteSo, it's not like they were looking toward the next milestone and counting their way backward from there to get to the number in question. For 7, they weren't looking to 10 and counting backwards 3 steps. Counting backward hurts your brain.
And in a major hint, these systems DO NOT extend this logic to the 6-9 words above 10, 20, 30, etc. They don't express 28 as "2 toward 30". Perhaps some of the 11-19 words, like Latin 18 and 19, but that's rare, and non-existent above 20. For these higher numbers, they use an additive system -- 28 is "20 and 8", where 8 may be hiding an expression for "2 toward 10", since that's below the 10 threshold.
So it only applies to the numbers you can count on your one or two own hands -- or in rare cases, if you re-use both hands in a forward-then-reverse sequence, which could cover cases like Latin 18 and 19. But nobody re-uses their hands a 3rd time.
And it begins with the words for 3 and 4, based on just 1-hand counting. When the counting goes above the 5 threshold, the same logic is applied to the 6-9 numbers.
Ready for it? When you're counting your fingers, you "use" each finger one at a time. It doesn't matter what the convention is -- if you start with them extended, then they are "used" by curling them down. If you start with a closed fist, they are "used" by extending them. It also doesn't matter which finger is chosen for the first in sequence, that's an arbitrary convention.
Well then, after a finger is "used", some are still left "free", assuming you're below the next milestone (5 / hand-word). The abstract way to represent their relationship is that the milestone equals the number of used plus the number of free, in other words "free" = "milestone" - "used".
But prehistoric people were concrete, not abstract. They were looking down at their hand as it changed shape while counting. They were struggling to come up with a word for 4, and then someone noticed that when 4 fingers had changed their shape, that left 1 finger unchanged. So why not refer to the state of the hand after counting 4 steps, by referring to the 1 finger left in its original position? Eureka!
Likewise, the state of the hand after counting 3, has 2 fingers left in original position. That takes care of the non-primitive numbers 3 and 4.
For 6-9, the same logic applies. The state of the hands after counting 6 has 4 fingers unchanged, 7 has 3 unchanged, 8 has 2 unchanged, and 9 has 1 unchanged. Assuming you already have words for 1-4 -- which you ought to by the time you're coming up with words for 6-9 -- then you're gravy, baby!
No abstract variables, no equations, no subtraction operations required! Oh, thank God -- I thought we were going to have to do MATH, and solve EQUATIONS... instead, we get to keep everything nice and physical and intuitive, like our brains prefer it! ^_^
On the linguistic side, that means a morpheme about "remaining" does not mean anticipating the next milestone on the number-line, and counting backward from there. It means number of fingers on the hand "remaining" in their original position.
ReplyDeleteThis view of it being complimentary, reflexive, reciprocal -- two groups interacting with each other, the used and the free -- also fits the initial "t" in the Semitic word for 9, based on 1. It's a medio-passive, reflexive, reciprocal prefix, which means there are two parties involved, and somehow they're doing something to each other, or one party is doing something to itself, in an inter-related and mutual way.
That doesn't mean that this will become a regular, productive, across-the-board process. None of the single-digit non-primitive numbers shows a perfectly regular morphological process, in any language. But using a medio-passive / reciprocal affix is a natural solution, so it may be one of those varied strategies.
Final thought on whether Cushitic loaned some of its 1-4 words into the other branches, vs. them being cognates inherited from P-SA. It sounds crazy to propose that Cushitic, of all branches, was the source of these loans -- they're certainly not the most culturally influential among themselves, Semitic, and Egyptian. And they're about on par with Berber. They're only more influential than Omotic.
ReplyDeleteBut recall the context where these precise integer words arise -- international trade, commerce, taxation, tribute by / from one society to another, and so on.
There were extensive trade routes around the Red Sea during the 3rd millennium BC and later, involving various regions south of Egypt in Africa, down to the Horn of Africa. Nubia, Kush, the C-Group Culture, the Land of Punt -- some or all of these could have spoken a Cushitic language as their primary language, or as one of several in a polygot commercial setting.
And they were prestigious societies, with lots of trading, kingdoms not just tribal chiefdoms, and in the case of Punt, mythologically (or truthfully?) treated as the "Land of God" that their own Egyptian royalty originally came from.
So when Egypt is trading with Punt, they need integer words. Arabian traders and merchants on the eastern side of the shared Red Sea, also need integer words to trade with Cushitic speakers in Punt.
It looks less like traders flowed in the other direction, from Punt or Horn of Africa way up into Egypt or deep into Arabia. More so, outsiders traveled to Punt and the Horn. And when in Rome, do as Romans -- and speak as Romans. The people of Punt were the gatekeepers over their highly valued resources, so if you wanted to get your royal Egyptian hands on their myrrh, ebony, and baboons (yes, an exotic prestige pet, who were even mummified by the Egyptians, that's how highly they were valued), then you had to deal with Punt people in their own language.
And since Proto-Cushitic doesn't have numerals from 6-9, they had only 1-4 to utilize anyway. It was the perfect situation to force other branches of Saharo-Arabian to adopt Cushitic words for 1-4, in order to derive their own words for 6-9.
Those spices and resins from the Horn of Africa were quite powerful motivators, not just cuz they produced a pleasant heady effect, but they were used in sacred temple rituals, much like incense in Christian churches. And all you need to do to communicate with the gatekeepers of those resources, is adopt and adapt their words for 1-4? Sold! ^_^
Looking closer into Cushitic words for 3, in order to see if we can't get a better fit to Semitic, Egyptian, and Berber words for 7. The problem is that 2nd consonant being "z" in the supposed reconstruction of P-C, vs. "b" in Semitic and Berber, and "f" in Egyptian. Looks like it should be "b", then Egyptian devoiced it and fricativized it. But it should labial, at any rate.
ReplyDeleteThere's no solid reconstruction of P-C, so let's look at each language's numerals in the branch:
https://en.wikipedia.org/wiki/Cushitic_languages#Numerals
The Central members DO have a labial consonant, a labialized uvular or velar, some stops and some fricatives, and in one case, a uvular preceded by a round vowel, which could have absorbed an earlier labialization on the consonant.
If this reflects a back fricative with labialization, that makes an easier link to the problematic members with the medial sibilant, which are in the Eastern region. The original sound lost its labialization, and fronted to alveolar position, while maintaining its fricative manner, which is not such a radical change.
There are other problematic members with medial "d", but that could be a hardening of earlier "z", which reduces them to the case of medial "z".
And in fact, one of these Eastern languages DOES have a labial, not a "d" or "z" -- El Molo has "p", and the same initial consonant and 2 vowels that its neighbors have. In this case, it resolved the labialized velar by turning it into a primary labial consonant, and a stop at that. Perhaps this same strategy was employed by Berber, Semitic, and Egyptian, from an earlier labialized velar.
In the Southern region, all of them have a labial. It's mostly "tam", where the initial "t" may be a hardening of initial "s" that the others have. And "m" retains the labialization of the labialized velar, although making it nasal is a bit funny, but it still works. Dahalo has "kʼaba", with medial "b" as hoped for, and a different hardening of initial "s" into "k'", unusual.
I don't have all the precise correspondences and sound changes worked out, but this seems to be more than enough evidence to conclude that Proto-Cushitic 3 has a medial labial and initial voiceless sibilant. Probably a final voiceless pharyngeal fricative, retained in Eastern members but lost elsewhere.
That's more than enough to suppose that this P-C word for 3 is cognate with, or perhaps was loaned into, the Egyptian, Semitic, and Berber branches to derive their words for 7.
I risk losing the lovely lady section of my audience if math takes over my subject matter, so that's why it's important to make it also about psychology, and keeping it simple and concrete. I should have worked in a reference to moms teaching babies their words for numbers by counting on their cute stubby little handsies, as part of cultural transmission. ^_^
ReplyDeleteI really do value the girls who lurk here, it's rare for there to be mixed-sex spaces anywhere these days, including online. I don't really do it on purpose, as though to lure them in to a place they'd otherwise be given the ick by. They just find something intriguing about me, and online it's not even due to having a hot tutor like in IRL.
That's not to brag, just stating the facts. Girls find something about my quirky personality and childlike wonder-stricken mind, that they enjoy being in my company, even if it's just lurking. If I had to guess, I think they're drawn to a man who hasn't had the vigor crushed out of him. Manly vigor, childlike wonder, IDK exactly what mix of traits it is -- but something inspiring and energizing, not someone who's going to be a downer or a drain on them.
You have to find something you're motivated, inspired, and energized by, and devote yourself to it -- and the girly onlookers will notice your passion, and be intrigued by it. Girls don't like being tricked or baited into paying attention to a guy, they want it to happen naturally, and -- at least in their minds -- without the guy noticing them back at first.
Or perhaps ever, if they just want to lurk and be a Platonic kind of secret admirer... hehe. I don't mind if you just want to watch, I'll pretend I don't notice you back, if that would make you feel uncovered and frighten you away. ^_^
I know, I shouldn't say this out loud, to preserve the immersion. But every once in awhile, it's worth exploring the quirks of human psychology like this, especially when all the online dIsCouRsE about the sexes is just man-hating and girl-hating slopaganda.
In our dystopian hellscape, only in the abandoned ruins of the blogosphere can a safe space be preserved for the people-likers. ^_^
Also, I dedicate this little piece of scholarship to Aicha, my foxy babe study-buddy from discrete math in college, who was from Morocco and spoke Arabic (well, she spoke Moroccan, if you know what I mean), but said her ancestry was part Berber. And she was a real quant, a computer science major (I was linguistics).
ReplyDeleteI dedicated my post on the imperial ethnogenesis of Moroccan culture to her, among a few other Moroccan babes I've known IRL or online over the years.
She left quite the impression, very silent and mysterious but quick-witted and always thinking or feeling something... just not letting it bubble over into her outward behavior. I enjoy the bubbly extraverted type of girl, but it's intriguing to find the rare species of them whose natural personality is dark, quiet, and mysterious. Not emo / angsty / stay-away-from-me, just... tantalizingly mysterious. A riddle begging to be solved -- as an explorer and detective, I just can't resist that type of girl! ^_^
I really miss our homework sessions... someone who felt out-of-my-league yet down-to-earth and who was inviting me close... I don't think guys and girls have relationships like that anymore, not for awhile. This was y2k, after social antagonism had already begun, but before it had become so widespread and intense. There were still pockets of the '80s John Hughes harmony of the sexes to be found in y2k, all the more poignantly felt and appreciated at the time, since you could already feel the social harmony slipping away.
More on Proto-Cushitic and Saharo-Arabian, on the basis of numerals. First, I'm convinced that the 1-4 loans into the other branches came from Proto-Cushitic. If they were inherited from P-SA, then each branch's own words for 1-4 would look like their own words for 6-9. But they don't -- their 6-9 words look more like P-C words for 1-4.
ReplyDeleteE.g., 7 in Berber, Egyptian, and Semitic begins with "s". So does 3 in Cushitic. But 3 in those other branches begins with "k", "X", and "th", respectively. The 2nd consonant in the other branches for 7 is labial, as is the 2nd consonant in 3 for Cushitic. But the 2nd consonant in the Berber and Semitic words for 3 is liquid ("r" and "l"), although Egyptian 3 does have a labial as the 2nd. And the 3rd consonant in 7 for Egyptian and Semitic is pharyngeal (Berber deleted the 3rd consonant altogether), just like the 3rd consonant in 3 in Cushitic. But the Egyptian and Semitic words for 3 have a final consonant that is "t" and "th", respectively.
So, the word for 7 in Egyptian, Semitic, and Berber relied on Cushitic 3 -- not their own 3 -- as the basis for deriving a "complimentary 3" word, i.e. 7.
Otherwise, you'd have to assume that the P-SA reflexes in each branch, which resemble the P-C reflex, were used to derive 6-9 words in each branch, without any borrowing across branches. And then later some branch-specific sound changes affected only the 1-4 words in each branch, leaving the 6-9 words relatively untouched, and possibly some of the 1-4 words were replaced (like 4 in Semitic). That's how the the 6-9 words look similar across branches, but not so much the 1-4 words, and why within a branch, 1-4 doesn't look so similar to 6-9.
But sound changes affect all words, not just "numerals from 1-4, and excepting numerals from 6-9". It's not a morpheme or word-level change, in which case the derived 6-9 words could be hiding out inside a complex form, if there were affixes or something. However, sound changes like "change the lateral fricative to s" is not a morpheme or word-level change, so this doesn't apply. A word-level change like "replace this word with that word" would work, but not sub-morpheme segmental sound changes. And the 6-9 words don't all have affixes or other morphological complexity to disguise the old 1-4 words hiding out inside them. So that doesn't apply.
Why they all borrowed from Cushitic instead of using their own / borrowing from some other branch, is up for discussion. But that they did, is not. It seems like the most desired exotic goods were under control of Cushitic speakers, not Egyptian, Semitic, or Berber speakers. Spices, resins, incense, tropical woods, pet baboons -- quite the incentive to use the numerals of the gatekeepers!
And they borrowed the Cushitic words *after* their own branches had already begun altering the inherited P-SA words for 1-4, which is why within a branch it's harder to see the connection between 1-4 and 6-9. But the recently borrowed loans from Cushitic had not had nearly as much time to diverge, hence they appear very similar across branches. The 6-9 numerals were "recently" derived, vs. the origin and speciation of the entire Saharo-Arabian family, where the 1-4 words came from, and that's why 1-4 look less similar across branches than 6-9.
Also, I've figured out a better way to derive 9 from 1, although the semantic logic is the same. The phonology is a bit different than I assumed at first glance. The other branches' still derived their words for 9 from the P-Cushitic word for 1, but it doesn't involve that medio-passive prefix in Semitic like I thought.
ReplyDeleteFirst, Egyptian 9 is "psḏw", which has an extra 1st consonant, making it harder to see the connection with 9 in Semitic ("tišʕ") and Berber ("tẓa"). What is this extra "p" at the beginning in Egyptian? It's an anaphoric demonstrative determiner, a way of referring back to something already mentioned, but not like a pronoun, more like "the aforementioned..."
https://en.wiktionary.org/wiki/p%EA%9C%A3
Specifically, it's the singular masculine form (like Egyptian numerals in general), and "proximal to the spoken of". Well, sure, when you're counting on your hands, and you get to 9, with 1 finger left in the original position, you express 9 by referring to 1 -- which has already been mentioned (the first word you spoke while counting), and the 9 fingers used are proximal to the 1 finger left in original position. It's not about distance to the speaker, but to what "the aforementioned..." thing is. In this case, fingers on the hand.
That confirms that the word for 9 was referring back to another number-word, and therefore using complimentary counting for 6-9, rather than any other way of coining number words.
Semitic 9 has a small wrinkle, too, which is that it has 3 consonants instead of 2, but by comparing with Egyptian and Berber, it's clear that the sequence of 2 consonants, "šʕ", corresponds to a single emphatic consonant. It was probably voiced, like Egyptian and Berber, and when splitting into 2 separate consonants, the voicing went along with the "ʕ" only, since voicelessness is the unmarked state for langauges in general, and especially for Saharo-Arabian, where the sibilants of P-SA want to be voiceless. So, the "š" in "tišʕ" just represents "dummy sibilant", after the pharyngealization had been separated into a new 3rd consonant of the word.
BTW, Egyptian "ḏ" is an affricate, not a stop, so this consonant of the word for 9 has a sibilant feature just like in Berber and Semitic.
As for the 1st consonant of 9 in these languages, it varies between "s" in Egyptian and "t" in Semitic and Berber. That suggests it's resolving an affricate that no longer existed in these three languages, but did in the language they're all borrowing from.
This leads to a much better reconstruction of 1 in P-Cushitic, as well as explain the words for 9 in Berber, Egyptian, and Semitic. Previously, there were several reconstructions for 1 -- and it's not surprising for there to be two words for 1, like based on "alone" and "whole, together, unified". But more than that, means you might be missing a common form shared by some of them.
ReplyDeleteSo I'm ignoring the previous reconstructions, and going back to the source languages themselves. Reminder:
https://en.wikipedia.org/wiki/Cushitic_languages#Numerals
Previously, there were 2 reconstructions, based on the "l" words and the "t" words, and none on the "m" words -- but they're clearly there, and require one of their own. I think the "m" words, something like "mato", are a different sense for expressing 1. IDK what it is, but it's not related to the "t" and "l" words, and not related to the words for 9 in the other branches of Saharo-Arabian, so I ignore them here. I'm ignoring the pair of examples like "inik", which don't seem related either.
We're looking for a P-C word for 1 with aspects of "t" and "s", an affricate. And the Cushitic words have "t" and "l" -- but the attested "l" is often a descendant of "ɬ", the voiceless lateral fricative, which has had its manner altered to approximant to give "l". So, the first consonant of P-C 1 must have been "tɬ".
When hearing this affricate that may have existed in P-SA but was no longer present in Berber, Egyptian, or Semitic, Berber and Semitic retained the "t", while Egyptian retained the "ɬ", though centralizing its airflow into "s" (and preserving the fricative manner and voicelessness).
This also explains the unity of the words within Cushitic -- the "t" words preserve the "t", and the "l" words preserve the "ɬ" (while changing manner to approximant -- not the same change to it as in Egyptian, but why should it be, this is Cushitic).
As for the 2nd consonant, we expect something pharyngeal or emphatic, voiced, and fricative or affricate more than a stop.
ReplyDeleteNote that in the Northern branch of Cushitic, the 1-3 words show metathesis of the 2-consonantal root, for some reason. So its word for 1, "ɡaːl", reflects "l" 1st and "g" 2nd in the P-C root.
The "t" words have a 2nd consonant that is "ʔ" (glottal stop), "k", "kk", either "g" or "q", and "kʷ" (if "vattukʷe" has a prefix "va" and the "t" word). Most vowels after the 2nd consonant are round, including the contracted forms that remove the 1st syllable, like "kow". This suggests a labialized velar or uvular stop.
The "l" words have a 2nd consonant that is "g", "xʷ", "w", "ɢ" (uvular), "ɣ", and "ŋ". This also suggests a labialized velar or uvular consonant. Softening stops to fricatives is more likely than the reverse. And in the descendants, there's a match between the manner of the 1st and 2nd consonants, although that doesn't help much since the P-C 1st consonant is an affricate, with both stop and fricative features. And since pharyngealization may be in the mix, where the voiced pharyngeal is a fricative, that could be where the frication is coming from, not the primary consonant.
The presence of uvulars suggests it was that, and sometimes fronted to velar, since backing from velar to uvular is less likely. But maybe uvular is just reflecting a mix of velar consonant + pharyngealization (further back).
Voicing... given the possibility of an emphatic, the voiced pharyngeal fricative could be in the mix, as a 2nd-ary feature, and that's where the voicing is coming from.
There seems to be voicing harmony in the two classes, where the "t" words prefer a voiceless 2nd C, while the "l" words prefer a voiced 2nd C. If this voice-matching existed in the P-C ancestor, then since "tɬ" is voiceless, so should the 2nd C be. Hard to know if that did exist in the ancestor, though.
Maybe further internal investigation would clarify these problems, but since I've made the external connection to the other branches of S-A borrowing P-C words for 1-4, in their own words for 6-9, I'll rely on that. Remember, these borrowings are fairly "recent" in the family, and that's why the borrowings look so similar to each other.
For example, there's no general agreement about the correspondences across the branches of S-A regarding the emphatic consonants, which are at play here.
Based on all 3 borrowings being voiced, I'll assume the P-C donor was voiced.
Based on 2 borrowings being fricatives and 1 being an affricate, but none being a stop, I'll assume the P-C donor was a fricative.
So that makes it "ɣʷ", but with pharyngealization -- either as a 2nd-ary feature, or as a 3rd consonant with no vowel between it and preceding "ɣʷ". That would make the P-C word for 1 tri-consonantal after all.
The borrowers de-labialized it, since they have labialized fricatives, perhaps only labialized stops. They combined the velar fricative + pharyngeal into an emphatic consonant, but moved it forward where emphatics existed in their branches, like the coronal region. Then Semitic split it into the 2 separate components.
Do you have any thoughts as to whether that Cushitic m-t root for "1" has any link with either the Egyptian root for "10" or the Semitic root for "100"?
ReplyDelete*borrowers did NOT have labialized fricatives.
ReplyDeleteReconstructing 2 in P-C is a lot easier. It's "ɬaŋʷa".
ReplyDeleteThe 2nd C is reflected in "ŋ", "m", "k" / "kk" (hardening of "ŋ"), and "b", all involving nasal / labial / velar.
The 1st C is almost always "l", and "n" in 2 cases, perhaps a nasal-matching quirk of them, since their 2nd C is "m".
The Southern languages show hardening in the coronal region, meaning they reflect earlier "ʃ" from "ɬ". They de-labialized the 2nd C, fronted "ŋ" to coronal position, and hardened it into either "d" or "r" -- perhaps reflecting earlier retroflex "d", a hardening of retroflex "n", from P-C "ŋ".
When the 3 other branches borrowed this P-C word for 2 to express their own word for 8, they centralized the airflow of the 1st C, giving "s". None had labialized nasals, so for the 2nd C, they made labialization into a primary place feature, with Egyptian maintaining nasal manner (giving "m"), and Berber and Semitic hardening it into a stop (giving "b").
As for the 3rd consonant in Egyptian, this looks like the suffix "-nw", which is a numeral suffix used to derive the ordinal from the cardinal numbers, for 2-9.
https://en.wiktionary.org/wiki/-nw
IDK what the exact sense is for applying this to the Cushitic loan, but it's clearly modifying it so it's not just plain ol' 2, and the suffix is unique to numerals, not any ol' word class. It gives it a "secondary sense" meaning, like ordinals being secondary to cardinals. It says "this word is a derivative of another number word", showing again that this was complimentary counting, not coining entirely new words for 6-9.
I assume something cognate or borrowed from Egyptian, in Semitic, which also has the "n" and a long vowel after the Cushitic loan of 2. As usual, Berber deleted this 3rd consonant, if they used it to begin with.
P-C 3 takes more work, but its consonants are "s-W-ħ", where "W" is some back labialized consonant that we'll figure out.
ReplyDeleteThe 1st C is "s" or "ʃ" (or in the Northern branch, with metathesis of root consonants, it's "h", a debuccalization of "s"). Since the Southern languages harden coronals, and they have turned it into "t" rather than "tʃ", it must be "s".
The 3rd C is either deleted, "ħ", "h", "x", or "ʔ". More likely to start very back and move forward, and to lose rather than gain pharyngealization. So it's "ħ". Voicelessness does not reflect final-devoicing, since some have a vowel after the 3rd C.
The 2nd C is reflected in "m", "p", "b", "xʷ", "qʷ", "ɢ", "ɣʷ", "s" / "ss", "z" / "zz", "d" / "dd". Clearly a labialized velar (the uvulars, as in Cushitic 1, probably reflect the 3rd C's pharyngeal feature glomming onto the 2nd C, since the uvular forms don't have a 3rd C).
As for voicing, the only one with "p" seems to have voiceless obstruents as a rule, compared to other languages in its Eastern region. So that could reduce to "b". There are both voiced and voiceless sibilants, but only voiced coronal stops, which are a hardening of a sibilant (itself a fronting of the velar). So that sibilant was "z", and "s" is a de-voicing of it, while "d" is a hardening of it. The "s" is restricted regionally to the East Highlands, while the "z" is found in multiple regions (East Dullay, East Konsoid, East Arboroid). So the East Highlands dialect innovated. The "d" is widespread as well. The velar / uvular ones come in voiced and voiceless, so no help there. Overall, looks like it was voiced.
As for manner, there's already a fricative but not a stop, among velar voiced labialized. And since the fronted version is a sibilant, more likely that it came from a fricative to begin with. Softening a velar stop when moving it forward is possible, but usually takes an affricate manner, not directly to sibilant. And that's usually motivated by a following front vowel. Here, the velar is labialized, meaning when it loses labialization, it'll be followed by a back vowel if anything. Less likely to palatalize. So, probably fricative to begin with.
That makes the 3-consonantal root of the P-C word for 3, "s-ɣʷ-ħ". The 1st vowel is front and mainly "a", sometimes "e", and rarely "i". Something low-to-mid. In the Central languages, where labialization still exists, the 2nd V is not rounded and is always "a". It's "a" or "e" in the other regions, and only "o" or "u" due to absorbing the rounding of the lost labialized consonant. So, take both to be "a".
That yields P-C "saɣʷaħ" for 3.
When borrowing this to express 7, Berber and Egyptian kept the "s", while Semitic showed its across-the-board shift of "s" to "š". The P-C donor has mismatched voicing for the 2nd and 3rd C, but Egyptian and Semitic show voice-matching in opposite directions -- both voiceless in Egyptian, both voiced in Semitic. Berber has no 3rd C, just the 2nd, but it's voiced as in the donor.
Berber and Semitic turned the 2nd C into a labial stop by making labialization a primary place feature, preserving voicing as "b". Egyptian kept the fricative manner of the donor and made labialization into semi-labial place feature, i.e. labio-dental. Since it had no "v" to preserve voicing for that choice, it defaulted to "f".
Berber deleted the 3rd C, Egyptian kept it as is, and Semitic voiced it to "ʕ", with the voicing of the 2nd C spreading to the 3rd, since P-Semitic has no vowel intervening between them. They all kept the 1st vowel as "a", and Egyptian kept the 2nd as that, too.
Finally, P-C 4, which has a bizarre form and will be a two-parter. Cushitic languages show 2 fundamentally different forms, but the ones like "afur" are regionally restricted to the Eastern region, and even that region shows some of the other form. So we'll ignore the "afur" one as a later regional coinage / borrowing.
ReplyDeleteThe 1st C is either "s" or "ʃ", and since the coronal-hardening languages of the South have "tʃ" instead of "t", this should be "ʃ" -- but these hardening languages also have "i" as the following vowel, so the vowel could have palatalized underlying "s". As we'll see in external comparisons, this is what happened, and the P-C 1st C is "s".
The 2nd C is reflected in "ɖ" (retroflex), "dʒ", "z", "dz", "l", "r", "ʕ", "g", and "y". Since these are almost entirely in the coronal region, yet there's "ʕ", that suggests this is an emphatic coronal. The "r" could be a relic of retroflex feature, and the "l" could be derived from that "r". Retroflex and pharyngealization are easily mistaken in the emphatic consonants. They're all voiced. Probably "ḍ", emphatic "d", with the affricates resolving this as an unusual "d" in their language. The one example with an unusual "d" is retroflex "d", not a retroflex affricate or sibilant. Perhaps the "g" is shifting "d" backward as an attempt to capture the pharyngeal location of the empathic feature.
The 3rd C, where it has been retained, is almost always "ħ" (and not due to final de-voicing, since it can be followed by a final vowel). In the exceptional Northern region, it's "g". In one Southern language, it's "l", which could have derived from "ɬ". This is incredibly difficult to figure out internally.
So that's where we bring in the comparisons in the other branches' words for 6. Those that have a 3rd consonant, and here Berber is atypical in having one, have "s" (Berber and Egyptian) or "th" (Semitic). The "th" in Semitic is likely a fronting kind of place assimilation to the previous "d", a de-emphatic from the donor's "ḍ".
So the P-C word's 3rd C must have been closer to "ɬ", but with a pharyngeal or other far-back feature as well, such as glottalization. Maybe it was ejective, "ɬ'". This awkward consonant was mostly junked within Cushitic, and not equated with "ɬ", hence not rendered as "l". It kept the far-back feature to distinguish it from "ɬ", and since it must also try to be voiceless and fricative, that leaves only "ħ" or "h". Perhaps "ħ" was chosen over "h" on analogy with the 3rd consonant in the P-C numeral just before 4, namely 3, which already ended in "ħ".
For the 1st vowel, the phonemically conservative Central region has "e", "i", or "ə". The South generally has "i", one exception of "a". The "a" is mostly Eastern as well, though it's in the North. And "oo" is confined to a sub-region within the East. So either "a" or "i".
ReplyDeleteThe 2nd vowel is "a".
That yields "saḍaɬ'" / "siḍaɬ'" for P-C 4.
The problematic nature of emphatics across branches, and the highly unusual ejective lateral fricative, make for greater irregularity in the borrowed forms, compared to the other more obvious borrowings.
The 1st C was preserved in Egyptian, Berber, and Chadic, and the standard Semitic shift to "š". The 2nd C was preserved in Berber, de-emphaticized to "d" in Semitic and Chadic, and rhotacized to the tap "ɾ" in Egyptian (similar to the Cushitic descendants with "r"), which later altered to "y". The exotic 3rd C had its ejective feature removed since none of the borrowers use that feature, centralized the airflow to "s" in Berber and Egyptian, and fronted a bit further to "th" in Semitic, for place assimilation with de-emphaticized "d" (which moved it forward from retroflex / pharyngealized "ḍ").
Every borrower treated the donor as having a high-front 1st V, except Egyptian which treated it as "a", but still within the range of the P-C possibilities. Most likely the norm in P-C was "i", confirmed by the palatalization of the 1st C in those Southern Cushitic hardening languages.
And that wraps it up. Well, not exactly the kind of soothing lazy river ride you expect for a Sunday, but I can't help when I get possessed! ^_^
Re: Cushitic 1, Egyptian 10, Semitic 100 sounding like mVtV -- I didn't think that far ahead. Good catch! Sounds like milestone markings to me, not fine-scale integers. Certainly for 10 and 100, but perhaps this was applied in Cushitic after the original word for 1 became an integer.
ReplyDeleteThe "mato" word would've meant "mini-milestone" or something, analogized downwards from 10 and 100 that they encountered when trading with Egyptian and Semitic speakers.
If it's based on powers of 10, you can see why it doesn't show up as 1 for awhile -- that's like saying "10 to the 0 power". What the hell is a 0 power? Not intuitive, and just defined to be equal to 1 in the axioms.
But if you were intent on expressing 1 as an order-of-magnitude milestone, rather than a fine-scale integer, then you could adapt someone else's words for 10, 100, etc.
Neat!
Japanese has complimentary counting for number words! Well, one of them, anyway. I'll explain the entire system of Proto-Japonic numerals tomorrow, it's pretty strange. I've discovered quite a bit that isn't already known...
ReplyDeleteBut for now, to connect this to the topic of complimentary counting, I discovered that the P-J word for 4 is based on a word for 1. Not the standard cardinal number-word for 1, which is "pito" -- that word is the same as "person", i.e. contrasting with "group, crowd" or in other words, "the numbers 2 and above". "Pito" is based on "individual" for its meaning of 1.
However, when re-examining the numerals, which I've already studied a zillion times to look for Dene-Yeniseian connections, I had a "Eureka!" moment, now that I'm thinking of complimentary counting. Check out the P-J word for 4, which is "yə" (AKA "yo2"):
https://en.wiktionary.org/wiki/Reconstruction:Proto-Japonic/y%C9%99
Look at the 3rd meaning, which is "same". What does "same" contrast with? "Different", or even better, "varied", "multiple", "manifold", etc. "Same" or "identical" relates to 1 -- being of 1 type, not 2 or 3 or etc. separate types.
The relationship is the same, so to speak, in Proto-Indo-European, where the English word for "same" goes back to P-IE "somHós", which is based on "sem" = "together" and also used as a numeral for 1.
So at some point in the ancient past, there was a P-J number-word "yə", which meant 1. It was a less frequent word, after the frequent word "pito" (meaning "individual"). This "yə" numeral was derived from the existing P-J word for "same, identical".
But then later, when trying to derive words for the non-primitive numbers, "yə" = "1" was used to express 4, with the same semantics as we've seen before. When using the fingers to count to 4, there's 1 finger in its original position.
It used the less-frequent word for 1, so that there aren't too many near-homophones for numbers on the same hand.
Also, we'll see tomorrow that the P-J word for 2 is based on 1 = "pito" (in a doubling pattern). So if 4 were also based on 1 = "pito", then just on the first hand, there would be 3 numbers all sounding similar -- 1, 2, and 4. They avoided this as part of their doubling pattern as well -- the word for 4 was not based on 2 (doubled). They only wanted pairs of similar-sounding numerals, not triplets, quadruplets, etc.
Otherwise, 1, 2, 4, and 8 would all sound the same except for vowels. Too confusing!
Final remark: the OJ and MJ word for "same" has irregularly dropped the initial "y" -- it's "onaji" in MJ, from "onazi" in OJ, which ultimately came from "yənati" in P-J. Nobody knows that yet -- but it's true.
In P-Ryukyuan, "same" is "yono", where "yo" comes from P-J "yə" and "no" comes from P-J "nə", the attributive particle. So why wouldn't the Japanese branch have a similar construction? It did, but it used the alternate form of the attributive particle, "na", and it mysteriously dropped the initial "y".
Then "ti" is the P-J word for "essence", which became voiced to "d" in OJ, then palatalized into "z", and hardened further into "j", a typical sequence of changes.
So, MJ "onaji" ultimately comes from P-J "yə + na + ti" = "of the same essence".
I had to add this remark since there's no etymology for "onaji" noticing its connection to Ryukyuan words for "same", but with Japanese dropping the initial "y" for no apparent reason.
Cushitic word for "head" led to numerous languages' word for "large number", based on counting herds of livestock.
ReplyDeleteThis follows up on the very helpful comment left by "fan", who notes that the regional word for 1 in Highland East Cushitic, which is not like the standard Cushitic word for 1, looks like the Egyptian and Semitic words for "large number" or "order of magnitude milestone".
This actually extends FAR outside of the Saharo-Arabian family this time! And no one has ever seen the faintest glimmer of these connections! But they're right there! In fairness to others, *I* didn't notice them until this morning either, but I don't earn a living doing this stuff, and the others do!
OK, back to the topic. The regional word for 1 in Highland East Cushitic is "mato" (final vowel uncertain). Although perhaps nobody knows this, this word comes from the Proto-Cushitic word for "head", which is "matħ", but where that final consonant is a generic body-part ending. So the root for just "head" is "mat".
https://en.wiktionary.org/wiki/Reconstruction:Proto-Cushitic/mat%C4%A7-
In Highland East Cushitic, this native root, without the generic ending "ħ", was used to derive a word for 1 -- "mato", where the final vowel could be an unspecified dummy vowel V. What is the logic here? Well, the people who speak these languages are pastoralists, so they own livestock. And not just one, but many of them -- a herd.
It is common to refer to living things that are part of a large group, by referring to a salient body part (metonymy). Like in English, "hand" for a sailor (among the entire crew, which is like a herd). Or also in English, "head" for counting livestock (among the entire herd). Or "check out all the skirts", for the vast sea of female bodies making up the secretary pool in the good ol' days.
This accounts for the regional use of "head" for "1" in Cushitic.
Now is where the fun starts. Multiple languages outside the Cushitic branch borrowed the Cushitic word for "head" -- *with* the final "ħ", cuz they don't speak Cushitic, and can't distinguish the basic root "mat" from the generic body-part ending "ħ". So to them, the Cushitic counting-word relating to "head" is "matħ".
Semantically, the borrowers interpreted it in the context of "large number, as big as a herd of livestock". The specific large number varied by the borrower, but they're all powers of 10, or generic "very large number".
Egyptian borrowed it to mean 10, pronouncing it "muːcʼaw", whose 2nd C is an ejective or emphatic voiceless coronal consonant, and spelled "mḏw". No change to Cushitic "m", and the Cushitic cluster "tħ" was rendered as emphatic "ḏ", since "ħ" is pharyngeal, and this feature attached to the previous coronal stop to make it emphatic, rather than preserve a cluster of 2 consonants.
https://en.wiktionary.org/wiki/m%E1%B8%8Fw
Proto-Berber also borrowed this word with the meaning of "10", "măraw", although rendering the Cushitic cluster somewhat differently from Egyptian. Berber also heard an emphatic coronal, but it sounded retroflex, and this retroflex feature was made into a primary "r" consonant.
Proto-Semitic borrowed the Cushitic word with the meaning of "100", "miʔat", whose final "t" is part of the feminine ending and not relevant.
https://en.wiktionary.org/wiki/Reconstruction:Proto-Semitic/mi%CA%94at-
Semitic also fused the two Cushitic consonants in the cluster, into a single consonant, but differently from Berber or Egyptian. Both consonants in the Cushitic cluster are voiceless, and so is "ʔ" (glottal stop). Semitic cared more about the place of articulation, opting for a far-back consonant, and preserving the stop manner of "t". There is no pharyngeal stop, so they opted for glottal stop as the next-best far-back location. Or they first slightly shifted pharyngeal "ħ" to glottal "h". Either way, resolving that Cushitic cluster.
But wait, that's not all! Latin and Greek borrowed this Cushitic "large number" word! Almost certainly not directly, but indirectly, likely through Egyptian and/or Semitic rather than Berber, but you never know. There were so many languages circulating among international traders and merchants around the Mediterranean coast in the old days.
ReplyDeleteIn Latin, they borrowed some Saharo-Arabian "large number" word as "mille", meaning "1000". The closest phonetic match is Egytpian, whose later development of this word is "mɛd͡ʒuː", with a high-front 1st V, and whose 2nd C is not a stop but has a somewhat continuant nature as well. It has 2 sub-segments in it, "d" and "ʒ", and Latin resolved those as a geminate "ll". Perhaps influenced by Berber "r" in that position of the word, a liquid.
The etymology of the Latin word for 1000 is currently "unknown", although I've just solved it. The only Indo-Euro derivation is totally ridiculous:
https://en.wiktionary.org/wiki/mille#Latin
https://en.wiktionary.org/wiki/Reconstruction:Proto-Italic/sm%C4%AB%C9%A3esl%C4%AB
The main root there is P-IE "ǵʰéslom", which means "heap", and that could indeed be used for "large number". That seems to be the root behind the Greek and Indo-Iranian words for 1000. But the Greek and Indo-Iranian words maintain a reflex of P-IE initial "ǵʰ" ("kh" and "h", respectively). Latin / Italic does not, there's nothing remotely like that in "mille".
Also, the tortured Latin etymology assumes an initial morpheme "sm̥-", derived from the P-IE word for "1" that we already saw, to mean "one heap". Something like this appears to be in the Indo-Iranian word, but it actually preserves the "s" and turned the syllabic nasal into a vowel, "sa". Latin "mille" doesn't begin with "s"! This assumes the "s" was deleted while the "m" was preserved. Sorry, if you're preserving only 1 consonant of the word for "1", it's "s", not "m".
And on top of it all, to get the vowels right, the tortured etymology assumes two separate instances of the feminine suffix, "ih₂" -- after each of the morphemes for "one" and "heap".
This is an insane level of over-fitting the data, or torturing it until it confesses. Too many morphemes for such a simple word -- "large number". The Greek and Indo-Iranian words don't have gender suffixes at all, let alone 2 of them. Only 2 morphemes -- one substantial word for "hand" and a derivational affix meaning "full", i.e. "a hand-ful" = "heap".
So I utterly reject this tortured Indo-Euro derivation of Latin "mille". It's an obvious borrowing from Egyptian, perhaps influenced by the cognate in Berber. That's why it begins with "m" and not "s", and why it doesn't have 17 consonants in the middle, just a single geminate (if that makes sense).
Next, Greek "myriad" reflects this Cushitic loan as well. Initially "myrios" meant "countless large number", then refined to mean specifically "10,000", as "myrias".
ReplyDeletehttps://en.wiktionary.org/wiki/%CE%BC%CF%85%CF%81%CE%AF%CE%BF%CF%82#Ancient_Greek
Note that "y" is a high-front vowel, just like the Egyptian 1st V, and just like the 1st V in Latin "mille". Like Latin, there's a liquid as the 2nd C, the same as in Berber, but not dissimilar from the medial consonant in Egyptian. It was even exotic-sounding to the Egyptians, Berbers, and Semitics, who rendered a Cushitic cluster in various ways.
The "s" ending consonant in Greek is a typical noun ending, not relevant.
Semantically, both "mille" and "myrios" mean "large power-of-10 number". They didn't need to borrow words for 10 or 100, since P-IE already had words for those, and they were inherited into both the Italic and Hellenic branches.
However, P-IE had no word for 1000, 10,000, and so on. So the daughter branches needed to come up with their own. Greek derived its word for 1000 based on native I-E roots for "handful" or "heap", just like Indo-Iranian did. But Italic did not -- they went straight to borrowing the Saharo-Arabian word for "larger power-of-10 number".
But Greek still needed a word for 10,000, or vaguely "power-of-10 beyond 1000". And by that point, they stopped bothering with native I-E roots, and went for borrowing the Saharo-Arabian "power-of-10" word as well.
This also accounts for why, at least AFAICT, only Italic and Hellenic have similar words for a "large power-of-10" number, and not the other branches of Indo-European -- these two branches are close to the Saharo-Arabian mercantile heartland of the Eastern Mediterranean, where Egyptian, Berber, and Semitic were all spoken among international traders throughout international markets.
I think the Italic and Hellenic economic connection to Cushitic speakers was probably indirect, via Egyptians, Berbers from the Sahara, or Semitic speakers of the Levant and the Hejaz / Yemen. So Italic and Hellenic didn't borrow directly from Cushitic to get their "power-of-10" number-word, but it ultimately traces back to the Proto-Cushitic word for "head", used as a counting term among livestock herders.
Another "herd" of mysteries properly corraled into the SOLVED area. ^_^
BTW, the Hellenists are honest that "myrios" has no known etymology (until my solution today), and don't bother with tortured 17-morpheme derivations from native I-E roots. The only "take a stab in the dark" guesses are based on I-E roots relating to "damp" -> "water" -> "waves" -> "countless". Yeah sure, each semantic link in the chain is totally secure! Hehe, they're just throwing something at the wall and seeing if it sticks, though, they aren't really committing to it.
ReplyDeleteThe other is based on the I-E word for "ant", as in "a countlessly large swarm of bugs". Close, but not close enough -- it is an animal-related term, but it's a metonymy for counting individual livestock within a herd, and based on "head". But in Cushitic and Saharo-Arabian borrowers of Cushitic -- not native I-E roots.
Samoyedic borrowed 5 from Indo-European phrase meaning "half-full" or "halfway to completion". Back to Uralic for a minute, but it involves Indo-Euro and their word for both "one" and "half" or "partial / fraction". Number words really get around!
ReplyDeleteAs mentioned before, the Samoyedic branch of Uralic has a quite different set of numerals compared to the other branches (Finno-Ugric). They share "wixte", based on the verbs for grasping or carrying, implying making a fist with the fingers and thumb curled around the thing you're grasping or carrying. But in Finno-Ugric, it means 5, while in Samoyedic it means 10.
The reason is probably the Samoyedic speakers, being more East Asian, had a different method of counting on the hands. You start with a closed fist, then extend the fingers one at a time, and when all 5 are extended, you reverse the order in which you extended them, curling them back. In this method, you only make a "grasping hand", i.e. a fist, at 0 -- which is not a numeral -- and 10. In Finno-Ugric, they must have begun with the fingers extended, and then curled them one at a time, so that they made a "grasping hand" at 5. Similar to the American counting game "put a finger down", where they all begin extended.
Since Samoyedic shifted "wixte" to mean 10 instead of 5 -- and the Uralic branch called Emishi / Ainu shifted it further to mean 20 (P-Ainu "wot", modern "hot") -- Samoyedic needed a replacement word for 5. There are two related forms in this branch, the basic word represented in the Southern region by "səmpulä", and the more complex phrase "səmpəläŋkə" that dominates in the North.
https://en.wiktionary.org/wiki/Reconstruction:Proto-Samoyedic/s%C9%99mp%C9%99l%C3%A4%C5%8Bk%C9%99
This is obviously a commplex phrase -- way too many vowels and syllables, compared to the Uralic preference for 2 syllables in a word. You can segment either word any way you want, and not find native Samoyedic or Uralic morphemes that they correspond to. So it must be a loan -- not surprising, given that it's a numeral, a word class that is unique in its ability to borrow across entirely different language families. That's cuz numbers and counting and arithmetic are a kind of technology, and tech can easily be loaned and borrowed around all sorts of cultures, for utilitarian reasons, not ethnic or shibboleth-y reasons.
The first element comes from P-IE "sēmi" = "half", itself from P-IE "sem" meaning 1, and based on "whole, unified". The connection there is that a fraction like "half" is "one piece of a whole". Even in modern English borrowing of this word via Latin, we use the prefix "semi-" to mean "partially", not necessarily where the denominator is 2 and numerator is 1. Just, "fraction, partial, fragment of a whole -- not entirely".
The second element comes from either of two related, probably cognate, P-IE words. First, "pleh₁", also reconstructed as "pelh₁", a root realting to "fill". Second, "pelh₂", relating to "approach" after being driven, pushed, or set into motion. These are semantically very close, since when you're filling something, you've set a sequence in motion by pouring the liquid, and it approaches completion part-way through the sequence.
https://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-European/pleh%E2%82%81-
https://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-European/pelh%E2%82%82-
In either case, the phrase rendered by Southern Samoyedic "səmpulä" means "half-full" or "half-way approaching completion". The phonetics are trivial, where the "i" of "sēmi" is deleted in order to bring the "m" and "p" together according to Uralic phonotactics, which prefer a medial sequence CVCCV over CVCVCV. And the P-IE laryngeal is rendered as an indistinct dummy vowel in Samoyedic, as "ə". Southern Samoyedic gave this vowel further rounding in the 2nd and 3rd vowels, probably influenced by the "p". But not in P-Samoyedic, where they're "ə".
The semantics are also clear -- 5 is half-way toward 10, and if 10 is full since that's when you stop counting, then 5 is half-full.
Now, what's going on with that extra morpheme in "səmpəläŋkə"? I have to conclude this is contamination from the P-IE word for 5, which is "pénkʷe". The Northern Samoyedic languages heard both the P-IE standalone word for 5, "pénkʷe", as well as the Samoyedic borrowed phrase "səmpələ" meaning "half-full". There's a sequence shared between them of "p", V, C, V. So that similarity led to carrying over the "énkʷe" in "pénkʷe", tacked on after the "l" in "pel / pəl", whose initial "p" triggered the association with "pénkʷe". And the alveolar nasal "n" of the P-IE word assimilated for place with following velar "kʷ / k", yielding "ŋ" in the Samoyedic borrowing.
ReplyDeleteSolved!
Note that P-Samoyedic 5 borrowed from P-IE, not a daughter branch. There are no "l"s in Indo-Iranian, so it wasn't them. And there are no other branches with the fraction-word "sēmi" near the Altai-Sayan region -- only Germanic, Hellenic, and Italic. Of those, only Germanic and Italic preserve the original, whereas Hellenic changed the initial to "h", and Indo-Iranian changed the 1st V to "a". Italic can be eliminated due to its word for 5 altering the P-IE initial to "kʷ" instead of "p", and Germanic altered it to "f".
ReplyDeleteSo, it was borrowed a VERY long time ago, from P-IE itself.
And since Samoyedic borrowed this phrase to replace the earlier Uralic word for 5, which it displaced to mean 10, that means that Samoyedic displacement took place during the P-IE stage. And so the P-Uralic numeral "wixte" meaning "grasping hand, fist" must go back to that stage or earlier as well. Pretty damn old.
But then, these milestones like 5 and 10 are the earliest to develop, and fine-scale integers are much later.
What was the semantic source of contamination in the longer form of P-S 5? I only mentioned the phonetic source of contamination. Semantically, 5 could be "half-full", but it could also be "one 5 out of two 5's, which together form a whole, i.e. 10 when the counting is complete". This uses the "semi, partial, fractional, half" word, as well as the numeral for "5", from P-IE.
ReplyDeleteSo the phonetic contamination was also bolstered by semantic contamination or double-entendre.
Final mini-wrinkle on the Mediterranean power-of-10 words. Both the Latin and Greek words end in a high-ish front vowel, "e" in Latin and "i" in Greek (the "-os" and "-as" are generic noun endings, not part of the root). This is the only notable difference with the Saharo-Arabian words, which have either "a" or a back vowel as the 2nd vowel, or a diphthong with both, like "aw".
ReplyDeleteI doubt Latin or Greek borrowed from each other -- Greek "r" doesn't get borrowed as geminate "ll" in Latin or vice versa. In fact, Latin did borrow the Greek, but kept it identical phonetically and semantically -- "myrias", meaning a myriad. That's not how Latin got "mille" = "1000".
They must have borrowed from the same or similar sources, and outside the I-E family, in the Med region. That is, Saharo-Arabian. But none of the standard words in S-A end in a high-ish front vowel.
I don't think it was dialectal variation among the sources, since all branches and sub-branches of Semitic have "a" or "o / u" as the 2nd V.
So it must have been an Indo-Euro alteration of the S-A vowel, perhaps to make it sound more like the vowel before a typical noun ending in Indo-Euro, like "e" or "i" before "-os / -us".
Final final mini-wrinkle -- it's possible that in P-Semitic "miʔat" originally used that final "t" as a meaningful segment, not just part of a generic feminine ending. Two possibilities:
1. They originally mis-heard the P-Cushitic source as having a final single consonant, an emphatic coronal. Then they split this single emphatic coronal into 2 separate consonants, a coronal stop and a far-back glottal stop, to suggest the pharyngeal location (only "far-back", not pharyngeal per se, but hinting at it).
This is similar to what they did with their word for 9, "tišʕ", whose final 2-consonant cluster corresponds to a single emphatic coronal in Berber and Egyptian, and all of whom ultimately trace back to an ejective (emphatic?) coronal in the P-Cushitic word for "1". Semitic chose to divide the features of a single emphatic into 2 separate C's, a plain coronal and a pharyngeal.
2. They correctly heard the P-Cushitic source as having a final sequence of "t" and "ħ". They preserved the "t" and altered the "ħ" to "ʔ" as an assimilation of voicing and manner of adjacent "t" (voiceless, stop). IDK if I buy that -- the other S-A branches mis-heard the P-C donor as a single emphatic, so Semitic probably did too, and applied their splitting strategy as in "tišʕ".
In either case, why did the far-back consonant come first, and the "t" last? Perhaps under the influence of the feminine suffix having "t" -- so make it sound more like a numeral with a feminine ending, and put the "t" last rather than middle, leaving the glottal stop to go in 2nd place.
Meant to say, "ħ" could've been altered to "ʔ" to assimilate for manner with adjacent "t". That is, making it a stop. And since there are no pharyngeal stops, the next-closest location with a far-back stop is glottal, so they chose glottal stop.
ReplyDeleteNote that in "tišʕ", both the 2nd and 3rd C's are fricatives, and they are split from a single consonant in the source. So perhaps splitting required a match for manner of articulation. Further investigation needed.
Now onto solving the mysteries of the Japonic numeral system -- the native system, not the Chinese words they borrowed during the Dark Ages. The Japonic system is very strange, and has largely been abandoned by the Japanese, in favor of Chinese words. But it's revealing about how numeral systems are constructed, how number-words are coined, and so on -- so it's worth exploring.
ReplyDeleteJudging from Wiktionary, Wikipedia, etc., almost nothing is known about Japanese numerals, but it's all very easy to figure out -- you don't have to be a cliff-dwelling sage. But if nobody else is going to solve these mysteries, I will. These are all from Proto-Japonic.
Let's start the way they did -- with the milestones 1, 5, and 10. This is how numeral systems begin, by marking orders of magnitude, and then they fill in the fine-scale integers afterward.
The only non-trivial etymology that anyone has found so far, is that 1 = "pito", which is the same as "person". This is a new strategy, not based on "alone" or "same, undifferentiated", as in other families. It's based on contrasting a single individual vs. a group of people that's 2, 3, 4, 5, etc. people in size.
There's a related ordinal word for "first", which is "patu", altering the vowels of "pito" but keeping the consonants the same.
The word for 5 is "etu / itu", which is the same word as the relative and interrogatory pronoun for "when". How has no one made the connection before? When you're counting, and you reach 5, it's time to start doing something different -- in the case of East Asia, it means it's time to reverse the state of the fingers that has happened during 1-5. You start with the fist closed, then extend them one by one -- after 5, it's time to reverse those actions, and curl them back into a fist one by one.
Using the word for "when" to announce when some action needs to be taken, is exactly the same as in the English phrase, "Say 'when'..." Like when you're pouring a drink, and you want the other person to say "when" there is enough liquid to their liking. Like, "that's it, that's enough" -- it's like the notion of filling up or completing something, that we saw from the Samoyedic numerals. Calling out the relative or interrogatory time-word -- "when" -- once a filling process has filled up as much as it can, and it's time to stop that process.
The word for 10 is "təwə", and its meaning is not as obvious as a single word appropriated from elsewhere. But it's a simple compound. The "wə" means "tail, foot of a mountain, end of something". (Ignore the false claim that it's from P-J "bo" -- P-J does not have voiced obstruents, and even when OJ got them, it was not in initial position. This reflects P-J "wo" itself.)
https://en.wiktionary.org/wiki/%E5%B0%BE#Japanese
Well, 10 is the end of what? That's what the modifier "tə" answers. This root relates to sequences and time. By itself, it is the conjunctive particle -- "and" -- that strings together words in a list. It is the first element of the word for "time" = "təki", and the word for "year" = "təsi". It marks the passage of a sequence, or counting items in a sequence. So, "təwə" means "sequence end" or "counting end". And that's just when the process of counting stops, at 10.
The only systematic thing that others have noticed about Japanese numerals is the doubling pattern, where the words sound similar for X and 2X. So, 1 sounds like 2 ("pito" and "puta"), 3 sounds like 6 ("mi" and "mu", same vowel relation as in 1 & 2), and 4 sounds like 8 ("yo" and "ya", a different vowel relation from the other 2 pairs). The notable exception is that 5 and 10 don't sound alike, cuz they're milestones and have their own mneumonics, they don't need to fit into a systematic schema.
ReplyDeleteSo why don't all words that are twice some other number, fall into this pattern? Why does 4 not sound like 2? Likewise, why does 8 not sound like 2?
Well, if they did, they would also sound like 1 -- 2 to the 0 power. Recall that 1 and 2 sound alike. So if that logic were extended beyond a pair of words, to all doubled words with a common first member, it would mean 1, 2, 4, and 8 would sound alike, and so would the ordinal word for "first". By keeping the consonants the same, and only altering the vowels, you're pretty much out of vowels if you did this!
And talk about confusing -- why would you want 4 out of 10 number-words to sound similar? You want to space out similar-sounding words if possible, to avoid confusion. And to prevent the basic act of counting from becoming a tongue-twister! Just imagine if the 1-10 words contained pito, puta, peto, and pota... ay ay ay.
So, only 2 numbers are grouped into similar-sounding pairs, not 3 or more words to a group.
When I said "non-trivial" etymologies for Japonic numerals, this is what I mean by "trivial". It's clear that 2 derives from 1, 6 from 3, and 8 from 4, by vowel alternation.
Well, as neat as this sounds, you might notice how bizarre it is. I guarantee you've never seen another number-word system where two numbers are pronounced similarly if one is a multiple of the other. That's cuz counting is additive -- you add 1 at each step in the sequence. You don't multiply.
So how can a number sound like a *multiple* of another number? How can you phonetically link numbers that are multiples of each other, using your hands and fingers to count, which is normally additive?
Behold! By using one hand to represent the smaller number of the pair, then copying the state of this first hand, onto the second hand. Surely, they used the right hand for the smaller basic number, and copied it onto the left hand. So, if only the thumb is activated on the right hand, then activate the thumb on the left hand. If 3 fingers are activated on the right, then activate the same 3 on the left. If 4 are activated on the right, then activate the same 4 fingers on the left.
ReplyDeleteBut that only gets you so far -- if you activate 3 fingers on the right hand, and copy that onto the left hand, you just have 3 fingers activated on each hand. How can you tell what number this represents, from 1 to 10? Remember, primitive people can't do arithmetic -- and neither can many civilized people, including math majors. They can't do it *in their head*, I should say. That's what these hand-counting "technologies" are for.
Now they move to a second process, which is like using tally-marks. You have to re-arrange all the elements into groups of 5, and any left-overs must be less than 5. So what's the action performed with the hands? You de-activate a finger on the left hand, and activate an unactivated finger on the right hand, one by one until either the right hand fills up to 5, or there are no more fingers left to de-activate on the left.
To be explicit, for 1 & 2. Activate 1 finger on the right, copy that same finger activation on the left. Then de-activate that 1 left finger, and activate another on the right. Now there are 2 fingers activated on the right, 0 on the left, and you can see with your own eyes what number this represents -- 2, a one-hand-only number.
For 3 and 6, activate 3 on the right, copy those same 3 onto the left. De-activate 1 on the left, activate 1 on the right. De-activate another 1 on the left, activate 1 on the right. Now all 5 on the right are activated, and this ends.
If you use two hands to count, you can tell what number this is -- a full hand and 1 finger on the other, is 6. If you use one hand, forward and then inverted, then it's a bit more work. For each activated finger remaining on the left hand, de-activate it, but now de-activate also on the right hand -- inverting the sequence for filling up 1-5. After 1 step, now there are 0 fingers activated on the left, and 1 step into the inverted sequence on the right -- that's 6.
For 4 and 8, activate 4 on the right, then copy those same 4 activated on the left. De-activate 1 on the left, activate 1 on the right. Now the right is full. If you use two hands to count, one full hand and 3 fingers activated on the other, is 8. If you use one hand, forward and inverted, de-activate each of the 3 fingers on the left, and de-activate 3 on the right, in the inverted sequence. After 3 steps into the inverted sequence, the hand is in the state of 8.
Now you see why this numeral system was abandoned by the Japanese! Sure, you can just learn the words without understanding the logic of the system that constructed it. But it just doesn't sound very intuitive when there are similar sounding multiples, but only pairs not larger groups -- why doesn't 4 and 8 sound like 1 and 2?
Just get rid of that system, and that's what they did.
That explains the etymology of the larger of the doubled pair -- 2, 6, and 8. But what about the smaller pair? Where did *those* sounds come from? Well, 1 = "pito" has already been covered. It means "individual" vs. a group of 2+. And yesterday I solved the very elusive mystery of where 4 = "yə" comes from -- it's a complimentary counting term, derived from a no-longer-used P-J numeral for 1, also called "yə", which is the same as the word for "same". It expresses 4 by way of 1 -- a less frequent, secondary kind of 1, not the main standard 1.
ReplyDeleteHow about 3 = "mi"? Now we have to see which finger is used for 3 in Japonic hand-counting, so we can see what features it has, that may be alluded to by "mi".
P-J hand-counting was of the East Asian type, where only 1 hand is used for 1-10, it begins closed, and the fingers are extended through 5, then curled back through 10. It started with 1 = thumb, then moved adjacently to the pointer = 2, middle = 3, ring = 4, and pinky = 5. Then the inversion / curling began with the pinky = 6, and moved adjacently to the ring = 7, middle = 8, pointer = 9, and finally the thumb = 10.
I didn't time-travel back to 500 BC to observe this sequence -- I figured it out cuz this is the only sequence that fits the pattern of P-J numerals, as we'll see.
So 3 arrived with the middle finger (first stage). I have to conclude that "mi" refers to the fact that the middle finger is the longest, and most visually prominent among the fingers, the most eye-catching. This means that "mi" here is cognate with the P-J word for "eye" = underlying "mai", realized as "me" in standalone form. Also, the root of the verb "to see" = "miru", where the "-ru" is the default verb suffix, and "mi" is the root. It's also the noun-like form of the verb, like "seeing" in English.
In MJ, the middle finger is called by the same term as in English, "nakayubi", where "naka" means "middle, inner". But once upon a time, it didn't have this name, and was referred to based on it being the longest and therefore most visually prominent of the fingers.
This is not so different from the etymology of the Proto-Indo-European word for 3, which is based on "tip" and "protruding", alluding to the middle finger, which sticks out further than the others. And so, it must have been 3rd in sequence -- whether they started with the pinky or thumb.
https://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-European/tr%C3%A9yes
That only leaves the etymology of 7 and 9, which are not part of a doubled pair, and not milestones. Both of them also refer to the finger involved. Both are during the second, inverting stage, during which 7 arrives on the ring finger, and 9 on the pointer finger. They're both very simple, too!
ReplyDeleteThe P-J word for 7 = "nana", and it means "no name":
https://en.wiktionary.org/wiki/Reconstruction:Proto-Japonic/na
All you vtuber fans out there remember the meaning of Nanashi Mumei -- "nameless" in Japonic roots, and in Chinese roots. The "-shi" (from older "-si") is just a generic adjective suffix, where the roots meaning "no name" are "na + na".
In other words, the ring finger is the "no name" finger -- cuz there's nothing special or salient about it. Really, only once the convention of wearing a wedding ring on that finger, did English have any helpful way to describe it. It's not the longest one -- nor is it the littlest one. It's not on either end. It isn't used to do some verb, like grasping, pointing, etc. With no distinctive traits, it's just the "no name" finger.
As it happens, both Chinese and Japanese use this "no name" sense for ring finger, although only Japanese uses the sounds "nana", whereas Chinese sounds more like "ma meŋ" (in Old Chinese).
https://en.wiktionary.org/wiki/無名指
So it's possible that the Wa / Japanese learned the method of hand-counting from the Chinese, or indirectly from the Chinese through someone else in the Xiongnu confederation, which the Wa belonged to. They may have also learned the semantics of referring to the ring finger as the one with no name -- or maybe they figured that out on their own. I figured it out on my own last night. How else do you describe it? Might as well call it "the miscellaneous finger".
However, the Wa did not pick up the Chinese *sounds* to match that meaning -- they used their own native Japonic roots, to give "nana".
Finally, the P-J word for 9 = "kəkənə". This is also the name of the finger involved, the pointer finger. As with "nana", the overt word for "finger" = "oyopi" (MJ "yubi") is not included, since it's taken for granted that you're referring to fingers during hand-counting. The number-word is just the modifier, for which finger it is.
The final "nə" here is the attributive particle, which is a standard way to connect a modifier and a noun, even if the 2nd noun isn't overtly stated.
The main root is "kəkə", which is the standard demonstrative place-word -- "here". It consists of one "kə", the proximal demonstrative -- "this" as opposed to "that". And the second "kə" is a place-related suffix, also part of the interrogative place-word "doko" = "where?"
So when referring to the pointer finger, they called it the "here finger", i.e. it's the finger you use to make the pointing gesture to indicate "here".
Later, this word was shortened to just "kono", but that too means "this" in an attributive sense, like when followed by a noun -- it means "this [noun]" as opposed to some other member from its class. And that gesture also involves using the pointer finger to, well, point.
Doesn't using mesial or distal demonstratives also involve pointing, like "there" or "over yonder"? No, it does not. First, pointing away or toward something at middle or far distance can be socially rude and taboo. But also, you don't know precisely where to point when it's far away. Only when it's close, do you know precisely where the location is, so you can point at it. "There" or "over there" or "way over there" involve vague, broad, sweeping, waving motions with the whole arm, since you're expressing a wide range, since you don't know precisely where it is. Using the finger to point? That has to be only "this" or "here".
And that solves the mysteries of the Japonic numeral system. I told you it was strange! But also worth exploring, to see how cultures use language and counting together. It's not an easy merger -- that's why most people either like math or language, not both!
But I can do both, so just leave it to me... ^_^
Last few mysteries solved with those Saharo-Arabian numerals. First, I found the source that Latin and Greek borrowed from for "mille" and "myri(-os)" -- it's Berber! There's a widespread Berber word for 100, probably going back to Proto-Berber. Here's the Central Atlas Tamazight version, "timiḍi", though others are cognate, like "t-è-med̩e" in Tuareg.
ReplyDeletehttps://en.wiktionary.org/wiki/%E2%B5%9C%E2%B5%89%E2%B5%8E%E2%B5%89%E2%B4%B9%E2%B5%89#Central_Atlas_Tamazight
It has a feminine plural prefix, "ti-" or "te-", and S-A numerals do tend to be feminine in gender, and this is 100, so it's plural in the sense of "plural power of 10", i.e. 10^2.
The root, though, is "miḍi" -- exactly what I was looking for! It's "m", high-front vowel, voiced emphatic coronal, and high-front vowel.
AFAICT, no other branch of S-A has a form like this. This root, without the "ti-" prefix of course, is what Latin and Greek borrowed from, ultimately, perhaps through an intermediary. But it seems to be a Berber word meaning "large power-of-10 number", "the size of a herd," etc. And the prefixed version "timiḍi" gives it a specific value -- 100, as opposed to other power-of-10.
Note that the root for 100 is not the exact same as the root for 10, which is "maraw" -- ultimately from the same P-Cushitic source, but the Berber roots for 10 and 100 are doublets of that loan from Cushitic. In the 10 root, the 2nd C is a bit hazier than the Cushitic source, which had a pharyngeal feature (coronal + pharyngeal sequence). The "r" in "maraw" is treating it more like a retroflex feature, but still "emphatic", in a different series than the plain consonants.
But in the 100 root, it's much closer to the original, with an emphatic. The 1st V is slightly different from the source, being "e,i" instead of "a", but there's vowel harmony in Berber that changes "a" to "i" if there are other "i"s in the root, as in this case. Namely, the 2nd V of the root has been changed to "i" from the Cushitic source, more like "a(w)" or "o,u". This vowel alternated form also helps to refine / distinguish its meaning, instead of the vague "herd-size" source.
This vowel harmony, with both V's being high-front and no "a", is also present in the Latin and Greek words, so they MUST have borrowed it from a Berber source.
Berbers were mostly associated with the trans-Saharan trade routes, but they interacted with Mediterranean coastal people as well, not to mention Egyptian and Cushitic speakers far to the south of the Med, in the Horn of Africa. Romans traded across the Sahara, and evidently the Greeks traded for goods that came from the Horn of Africa, or perhaps across the Sahara as well (gold, ivory, exotic animals).
So the next time you use the word "milestone", which has the Latin word for 1000 inside it (from "the distance covered by 1000 paces"), remember to thank the Berbers! ^_^
To further support the use of "head of livestock" words to denote power-of-10 number-words in Saharo-Arabian, here are two more cases, both distinct from the Cushitic one.
ReplyDeleteProto-Semitic 1000 is "ʔalp", which originally meant "livestock animal", likely cattle. It shares a root with the verb "ʔallip" = "to domesticate". At first this was used to mean "one", just like in Cushitic -- hence the name of the FIRST letter in the Phoenician alphabet, "aleph", used to start the sequence of letters.
https://en.wiktionary.org/wiki/Reconstruction:Proto-Semitic/%CA%94alp-
https://en.wiktionary.org/wiki/%F0%90%A4%80#Phoenician
Later, the root was used to mean "the size of a herd, measured in heads of cattle", specifically 1,000.
So there's a semantic connection between "one" and "large power-of-10", by means of a word used to count the size of a herd of livestock, which itself was a form of metonymy ("head" for the entire animal). Just like in Cushitic.
And in Egyptian, there's yet another root relating to the size of a herd and a large power-of-10 number, here 1,000 = "ḫꜣ". This is part of the phrase for "shepherd" = "ḥꜣt ḫꜣ". So originally it meant "livestock animal", then it was extended to mean "size of a herd", and finally it was specified to be 1,000. So there's another connection between "one", a single animal, and then the vaguely-defined "size of a herd", and then a specific large power-of-10 number. Just like in Cushitic.
https://en.wiktionary.org/wiki/%E1%B8%AB%EA%9C%A3#Egyptian
The claim in the entry is backwards -- it claims it meant 1000 first, and then adapted to mean size of a herd. Numbers are derived from real-world things, not the other way around.
These examples are further proof that the Cushitic word for "head" was used to refer to a "head of livestock animal", then the number "one", and then borrowed by other S-A branches to mean "size of a herd", with the different branches specifying different powers-of-10 with this borrowed "herd-size" word.
Notice that these words all relate to pastoralism, not crop cultivation. You can thank pastoralists for counting systems! The produce of crop cultivation could technically be measured in single digits, then larger powers of 10, but in reality it's not. The individual seeds planted, or individual grains harvested, are WAAAAYYY too numerous to count with integers, or even 10s / dozens.
ReplyDeleteIn practice, you treat seeds and grains as a mass noun, not a count noun -- and it's based on weight or volume, like a bushel, or a bag or a pound, or enough to fill a silo of a certain volume of space. They're tiny super-numerous particles that make up a great big undifferentiated blob of a mass noun. No counting involved -- only measuring, which is different.
In contrast, livestock animals are big enough to the naked eye, and not super-duper-numerous that they can't all be counted. They do not get defined by the total weight of the herd, or the total amount of space the herd takes up. Unlike grains, they aren't loaded onto a scale like a continuously measurable mass noun. They are counted discretely, perhaps using an exponential term after they get that big in size.
And that's where these power-of-10 words come from -- trying to express the size of a herd of livestock, which could number in the 10s, 100s, or even 1000s. Small-time herders may only need 10s, but once international trade gets involved, big-time livestock owners may in fact own 100s or 1000s of animals. Or at least, be responsible for corraling that many animals into a caravan in order to take to a big international market, or wealthy court, who will want a contract specifying exactly how many animals are being exchanged.
Seeds and grains provide no motivation to count them by powers-of-10, with discrete integers at the fine level -- rather, scalar weight / volume measurement, treated as a continuous mass noun, not count noun.
The standard story is that mathematics is a product of sedentary cultivated agriculture, who have a surplus that can support specialists like mathematicians, astronomers, and so on. Well, for providing institutional support and patronage, sure. But the mathematics itself do not come from farmers and cultivators -- it's only a figure of speech, derogatory at that, when we call accountants "bean-counters". You'd never count beans one by one, or even by powers-of-10. You'd only weigh them on a scale like any ol' lump of matter.
Rather, counting comes from the closest thing to noble savages in the post-hunter-gatherer world -- pastoralists. They don't treat their wealth like a great big undifferentiated blob of inanimate stuff. Their herd is made up of individual living breathing creatures, each one with a spirit and personality and often a name, who deserve having their individuality preserved, even as they're corraled into a herd and driven around as a herd and traded as a herd. The herd is not a soulless heap -- it's one great big exciting family made up of interconnected individual members.
Hyper-agrarian societies like China have rarely made progress in mathematics, while even fairly poor pastoralist societies have excelled at math -- especially when there are agrarians around to support them with part of their surplus. But the mathematicians themselves come from a herding background, not the matter-weighers of the agrarian economy.
Without herd-driving nomads, we would never have the rungs of the exponential counting-ladder to climb our way off of the mundane ground and up into the outer heavens...
That's why the seemingly improbable source of Cushitic provided one of the most successfully shared power-of-10 words throughout the Old World. Talk about nomadic pastoralism -- that's just about the only way of life in the Cushitic urheimat. Not much crop cultivation (until recently with coffee). Not much hunting and gathering. And it's not tropical, so there's no tropical gardening / horticulture.
ReplyDeleteIn Somalia, the camel population is so large, it's about 50% of the human population! About 8 million camels, and 20 million people. Imagine if there were 150 million camels in America... that's Somalia's per capita rate. Nobody loves animals more than they do.
Who else could be the earliest coiners of power-of-10 words based on counting the size of their vast herds of livestock? ^_^
To get another dig in at China, and praise Glorious Nippon, I think that's a major reason why Japan is so much better than China -- the Wa people came from a nomadic, livestock-herding, horse-riding way of life on the Eastern Steppe. That gives them an edge when it comes to doing mathematics (not just supporting mathematicians).
ReplyDeleteFor that matter, I'll bet the Northern Barbarians who assimilated into Chinese society, are over-represented among the paltry amount of mathematicians that China has ever produced. The hyper-agrarian elites may have patronized mathematicians, but tally marks -- and brand markings -- come from pastoralist people, to count the size of their herds.
When a mega-state needs to conduct a census for the purpose of taxation, cultivators could not provide the math necessary to do so. Counting individuals within a very large group -- why, that's like counting a herd of livestock! And governments *do* think of their subjects as being cattle...
No, they needed to recruit herders to provide the crucial mathematical insights for counting very large numbers, without resorting to scalar weight measurement or filling up space. That's when you know when governments have *really* started to degrade their subjects -- measuring them as pounds of raw material on a society-sized scale...
If the vast majority of Chinese are bad at math, then where did the American myth that Chinese Americans are good at math come from?
DeleteSpain also doesn't seem to have produced a lot of mathematicians, despite being fairly pastoralist. I remember you wrote a post about how this was the result of not having the right mix of personalities. I was also wondering if maybe some of this could be attributed to the black hole of asabiya after the collapse of the Spanish Empire?
ReplyDeleteMongolian mathematician discovered the Catalan numbers in the 18th C. Now how's THAT for a pastoralist counting synchronicity?
ReplyDeletehttps://en.wikipedia.org/wiki/Catalan_number
https://en.wikipedia.org/wiki/Minggatu
Catalan lived and worked in France, much as Minggatu worked at the Chinese Qing court -- didn't I tell you?! When you have the right insight, you can never go wrong. I'd never heard of him before, but that just confirms it.
If your ancestors invented a form a tally-marks for counting the size of your herd, you're guaranteed to discover some important series of numbers! ^_^
Also, I said Chinese can't invent or create in math, not that they aren't "good at math". You can understand things or test well, but not be able to do anything with it. Especially if it's just a bunch of memorization for tests -- you don't really understand what's going on.
Quick note on Japanese counting, where the Chinese words have replaced the Japonic ones in most cases -- except for 4 and 7, which still use Japonic words. Based on the sequence they use to count on the fingers, both of these numbers use the ring finger (4 during the first stage, 7 during the inverse stage).
ReplyDeleteIs the ring finger special to the Japanese? And if so, is it good-special or bad-special? Or is it just a coincidence? IDK, but worth noting.
Ainu myth is Uralic. I'm preparing a new series of posts on the Uralic cultural origins of the Ainu, including shibboleth-y parts like creation myths and beliefs about the afterlife. These are not subject to utilitarian calcuations, hence not arrived at by convergent evolution, pressure from a materially superior outside culture, and so on. Even when a culture adopts a new international religion, they keep these aspects of their old religion, and syncretize it into the new one.
ReplyDeleteHere are the two most important ones -- regarding the creation of man, and the nature of the afterlife. The Ainu examples are from Batchelor, "The Ainu and Their Folk-lore" (1901). The Uralic examples are from the Finno-Ugric and Siberian volume of "The Mythology of All Races" by Holmberg (1927).
I stumped Google's AI search by asking about these without mentioning a specific culture's name, just "which cultures...?" or "cultures whose myth about creation of man involves..." I prompted it to specifically mention "willow", and it still missed these examples. It just goes to show how dumb AI still is -- these books are both digitized and searchable and available for free browsing and download on archive.org.
Ignore the biases and hallucinations of AI. But never bet against a cliff-dwelling sage rummaging through the ruins of collapsed civilizations...
* * *
Creation of man involves forming the skeleton from willow sticks.
Ainu version:
'When God made man in the beginning, He formed his body of earth, his hair of chickweed, and his spine of a stick of willow. When, therefore, a person grows old, his back bends in the middle.'
Uralic version (Voguls are the Ugric-speaking Mansi):
'The Altaic peoples believe that bones were created from reeds and the rest of the body from clay. The North-West Siberian peoples, like the Voguls, relate how God "took willow-twigs, bound them into skeletons, covered them with a layer of clay, set them before him and blew into them."'
Both are alone in the world in saying that the first man's skeleton was made of willow sticks. Willow may have sacred importance in other cultures, but it's especially sacred to Uralic cultures. And "importance" is vague -- willow being the skeleton of the first man is very specific, and shibboleth-y. Therefore, the Ainu and the Mansi share a cultural ancestor.
The afterlife is lived in an upside-down world, underneath our living world, where the feet of the living and the dead touch each other's, and where the nature of things is generally upside-down or inverted in the underworld. The metaphor seems to stem from the underworld being underwater, creating a mirror-like reflection like the surface of a river or lake.
ReplyDeleteAinu version:
'[Our living world] is [called] the upper world, because there is another world under foot. That world is very damp and wet, and when wicked people die they go there and are punished. But by the side of this place there is another locality, which is called Kamui moshiri, i.e. "the country of the gods" or "heaven." It is to this place that the good people go at death. They live there with the deities and walk about upside down, after the manner of flies, so that their feet meet ours.
'When it is day upon this earth it is night in heaven, and when it is daylight there, it is dark here. Now, when it is dark in this world, men should neither do any work, nor trim one another's hair, nor cut the beard, for at that time the deities and ghosts of men are busy in their own spheres. If, therefore, the inhabitants of this world work during the hours of darkness, they will be punished with sickness and meet with an early death.'
Uralic version:
'A general belief is that the life beyond is lived under the earth. The passage occurs in a Vogul song: "The dead people go to the land below"; also, in Ostiak folk-poetry we read: "We arrive at the sea belonging to the man living in the underworld." In its nature this underworld resembles the world we live in in everything, with the exception that, seen with our eyes, everything there would appear inside out or upside down. The Lapps believe that the dead walk there with the soles of their feet against ours. According to the Samoyeds the same rivers and streams exist there, but flow in opposite directions. The tops of trees there grow downward; the sun rises in the west and sets in the east. The life of those over there runs also contrary to ours; they become younger and grow smaller with the years, until they disappear and become nothing or are born into the family again as children. In this way, the "shade" [i.e. soul of a dead person] lives as long in the underworld as its predecessor on the earth. The Ostiaks say that the dead dwindle in the end to a little beetle.'
No other cultures share this view of the underworld, and it is too specific -- about standing upside-down and having the feet of the living and dead touching -- rather than a vague notion of "inversion". None of it necessary to send off the dead in general, or to adopt a material technology or whatever else. It's a shibboleth. Therefore, the Ainu and most / all Uralic cultures share a cultural ancestor.
* * *
The Ainu or their predecessors (Emishi, Jomon) also seem to be the source of the Sanzu no Kawa myth in Japanese culture, i.e. the 3 paths across the river separating the living from the underworld. It's not like the River Styx or other Indo-Euro versions, since in the I-E versions you merely pay a toll to the guide over the barrier, whereas in the Uralic version, your conduct in life determines which path you must take to cross over into the afterlife -- no toll, only karma.
I'll go into more detail on that later, but I've quoted the Uralic versions in earlier comment threads linking Japanese myth to other Northern Eurasian cultures.
However, given that it's only Uralic cultures who share this myth outside of Japan, that means the link must be via the Ainu / Emishi / Jomon, who are also culturally Uralic. It doesn't seem to belong to Yeniseian or Na-Dene myth, which Wa / Japonic is more similar to.
And one of the real-world locations for the Sanzu no Kawa is in Northern Japan, which is where the Jomon / Emishi / Ainu were more influential.
Also, making major progress on the linguistic side, showing Ainu is Uralic. Not surprisingly, they're similar to the Eastern branches -- Ugric and Samoyedic. After discovering that the Mansi and the Ainu share a creation myth, I figured their shared ancestor must have been from thereabouts, so looked into what makes Ugric and Samoyedic distinct within the Uralic family.
ReplyDeleteMainly, the hardening of P-U "s" into "t", and filling that gap by shifting palatal "ś" forward to "s". Unlike Ugric and Samoyedic, though, Ainu shifted "š" to "s", whereas it was part of the hardening-to-"t" change in Ugric and Samoyedic. At some point, then, Ainu likely merged the weird sibilants from P-U, "ś" and "š", into a single non-standard sibilant, and shifted that forward to "s", after P-U "s" had hardened into "t" a la Ugric and Samoyedic.
In over 5 pages of notes, I found almost no examples of P-U "s" initially -- well, that's cuz it hardened into "t"! And I just didn't know that at first, not knowing which branch it would resemble, if any. It could've been its own branch with its own quirks. It probably is, but it also shares telltale Eastern changes, like this one.
There are tons of Proto-Ainu words beginning with "t", and not so much with "s", especially when not followed by "i". In P-A, "si" can be an allophonic variant of earlier "hi", and even that could have come from "wi". Point being, words beginning with "si" aren't necessarily "s"-initial words going further back. That leaves even fewer P-A words beginning with "s" ancestrally. Likely due to them hardening into "t", and swelling the ranks of "t"-initial words in P-A.
Forgot to emphasize that the creation myths also share the feature where the meat-y part of the body is made from clay, earth, mud, etc. That's not so distinctive on its own, whereas the willow-skeleton is. But it's yet another point of same-ness between Uralic and Ainu creation myths.
ReplyDeleteAlso, if you think an Ugric language from east of the Urals couldn't travel to Sakhalin and Hokkaido, remember that Hungarian is the only other member of this branch, and it traveled way the hell over into Hungary.
ReplyDeleteI'm not saying Ainu belongs to the Ugric branch -- it doesn't seem to have taken part in the softening of P-U "k" to "x" or "h" before back vowels, which is distinctive of Ugric. But the notion that it couldn't have traveled far away, is bogus, based on it traveling from east of the Urals into Hungary with the original Magyar raiders.
The Uralic family got around Eurasia, as much as Indo-European did, and as much as Saharo-Arabian did, in the Old World.
From Japan to Russia, a Uralic myth of earthquakes caused by giant fish moving its body in the watery underworld.
ReplyDeleteI'm not the first to draw connections between some of these shared myths, but nobody before -- that I can tell -- has connected them all, let alone identify the origin as Uralic.
We'll start with Japan, where earthquakes are said to be caused by Namazu (lit. "catfish"), a giant catfish that thrashes its body about somewhere in the watery depths, which cause earthquakes above in our normal world. It can be restrained, and that is the thunder god's job (Takemikazuchi), who holds it down with a real-life stone -- the Kaname-ishi, which lies in Ibariki Prefecture, on the Pacific coast in Eastern Japan. Whenever the restrainer faulters in his task, the catfish thrashes about and causes an earthquake.
https://en.wikipedia.org/wiki/Namazu
You've encountered this myth if you've played Legend of Zelda: A Link to the Past, where you get the quake medallion, which causes earthquakes, from a giant catfish. When you summon him from the watery depths, he causes an earthquake first to announce his presence, before giving you the medallion to get you to leave.
https://www.youtube.com/watch?v=5PPUz5ygxgE
There's also a Pokemon catfish character whose specialty is earthquakes (Namazun in Nihongo, Whiscash in English). And to this day, a kawaii cartoon catfish remains a national symbol for earthquakes.
However, as common as this myth is in modern Japan, it does not go back to the ancient Wa people. It does not appear in the two earliest works of mythology and history, the Nihon Shoki and Kojiki, from the 8th C. Nor from the various regional Fudoki, which were collections of local tales, customs, and so on, from around the same time. Only during the Early Modern era (15th / 16th C.), do definitive links between catfish and earthquakes appear in Japanese culture.
So where did they get this myth from? Why, from the Ainu! Or rather, from the Emishi, some of whom later fled the inchoate Yamato state and became the Ainu, and some of whom stayed put and assimilated into the Yamato.
I can't tell whether the Early Modern Japanese references to earthquakes and catfish came from the Ainu proper, or were relics from the Jomon / Emishi of Honshu that remained in circulation among the commoners, even after they stopped speaking Emishi / Ainu and assimilated into the new fusion culture of the Yamato. But it very clearly came from a Jomon / Emishi / Ainu source.
Note that Namazu is not portrayed as a cosmic fish upon whose body the entire earthly world rests. He's just some mythological fish who causes earthquakes, but he's not in the larger Japanese cosmography, and plays no role in the creation myth from the Nihon Shoki and Kojiki, which reflect the Yeniseian-like cultural origins of the Wa (since the sole other example of "creator-god stirring the oily primordial ocean with a staff, to cause the seafoam to harden into landmasses", is from Northwest Mongolia, a former Yeniseian-speaking region).
But in the Jomon / Emishi / Ainu version, it's not just any ol' mythological big fish -- it's THE big fish, the cosmic fish which our ordinary solid world rests on. To condense the somewhat lengthy account in Batchelor's "Ainu and Their Folk-lore", this cosmic fish is a trout, and is called "moshiri ikkewe chep, i.e. 'the backbone fish of the world'". When it moves, it causes earthquakes. It also causes the tides by taking in vs. expelling water from its mouth. Batchelor insightfully notes that it's not a turtle or serpent, as in other myths from India, the Middle East, etc.
ReplyDeleteTo prevent earthquakes, the creator-god sent 2 deities to restrain the fish, who stand on either side of him, and must always keep one hand on him in order to hold his body steady. If for any reason they faulter and remove their hand, its body moves and causes an earthquake.
This is clearly the source of the Namazu myth, although it has been de-sacralized -- Namazu is not the fish whose backbone supports the entire world, since that goes against the Yamato creation myth. Now, he's just a monstruously large semi-legendary trouble-making beast, with no positive qualities like supporting the world. And he's an ugly-looking fish -- catfish, not a trout -- befitting his demotion from cosmic world-supporter to mere earthquake-causer.
But the shared points are impossible to ignore, and are shibboleth-y unnecessary points, which prove a common origin. It's a fish, not a serpent or turtle or other animal. It's a freshwater fish, not a saltwater fish. Its bodily movements cause earthquakes. The main god sent lesser gods to physically restrain it, so his human creations wouldn't suffer from constant earthquakes. However, they're lesser gods, so they occasionally faulter, resulting in earthquakes.
The Ainu point about the fish causing the tides is probably a local innovation from the broader myth, since they're an oceanic culture, and the origin is Uralic, whose homeland is inland and far from beaches where the tides are salient. The Japanese removed this point about the tides, perhaps cuz that's a neutral quality, and they only wanted a negative quality for the beast that causes earthquakes, since their Yamato cosmography does not allow the causer to be the world-supporting fish.
So where else does this myth show up? We consult the good ol' Finno-Ugric & Siberian volume of "The Mythology of All Races", and discover that it's widespread in the Uralosphere, although Holmberg says they got it from somewhere else, despite the obvious differences. He rarely thought that an original Uralic mythology existed, that it was bits and pieces borrowed from various neighboring cultures, and from international religions like Buddhism, Islam, Christianity, etc. He was wrong about that, including in this case.
ReplyDeleteFirst, it appears in the Buryats, who are nowadays a Mongolic-speaking culture, but they live to the west and north of Lake Baikal, and historically included Samoyedic speakers before converting their language to Mongolic during the 2nd millennium AD.
'In certain Buriat districts, one large fish only [as opposed to 3 fish] is mentioned as the supporter of the earth. When for any reason it changes its position, earthquakes occur.'
Next he cites the Russian folklorist Afanasyev, known as the Russian counterpart of the Brothers Grimm, who collected an insane amount of fairytales and legends from East Slavic sources in the 19th C.
'The idea of one or more giant fish as supporters of the earth is general also in East European legends...'
He then misleadingly cites Biblical Hebrew lore about leviathan -- which is a serpent, from a saltwater environment, and whose body encircles the world menacingly, rather than a freshwater fish that helpfully supports the world from below. Likewise the stack of animals in Indian myth, which ultimately rest on a turtle's shell -- not a fish.
He cites other examples of a cosmic bull, which in turn rests on a fish, and where the bull's thrashing motions cause earthquakes, which gets us to the "Middle Eastern" monster called Bahamut:
https://en.wikipedia.org/wiki/Bahamut
Given the timing and location of these myths involving a bull, it seems like they come from a Turkic source during the High or Late Dark Ages, and were spread by the Turkic expansion. They're found in Iran, which was controlled by Seljuks and other Turkic tribes at the time, who were creating a Turco-Persian fusion culture. They're not part of Indo-Aryan myth, so it's doubtful they're originally Iranian either (not in Zoroastrianism). They don't have a Saharo-Arabian counterpart -- with a bull -- so it's not from that neighboring region either.
All the many cultures that Holmberg cites as having this bull version lie on, or adjacent to, the Steppe, and were conquered by all sorts of Turkic tribes during the Dark Ages. Namely, the Kyrgyz, West Siberian Tatars, Volga Tatars, Caucasian Tatars, Crimean Tatars, Volga Finns (probably meaning the Mordvins), Udmurts, and Mari. Some of them converted languages to Turkic and converted their religion to Islam, during the Turkic expansion -- but others did not, and remain Uralic-speaking and practicing Uralic paganism (at the time).
In Herodotus' time, that entire region spoke Uralic -- from Lake Baikal to NE Europe, and from the northern shores of the Caspian and Black Seas up to the Arctic Ocean, with only Yeniseian breaking things up near Lake Baikal.
The Tatars of the Volga, Northern Caucasus, and Crimea, share a deeper link with East Slavs, all of whom used to be Uralic. That's why this myth shows up in all of them, even after they converted languages and religions. It's common folk-lore, and resistant to change.
Some of these groups insisted that the bull wasn't alone, but rested on a cosmic fish -- syncretizing their old culture with the newly arrived Turkic culture. The Crimean Tatars said there was a fish under the bull, and the Mari said there was a crab under the bull. They are both ancestrally Uralic (and the Mari, currently so).
ReplyDeleteThe Kyrgyz, however, said the bull rested on a stone arising out of fog on the subterranean ocean -- not another animal, let alone a water-based one. They must not have had much of a Uralic background before they were Turkic.
Holmberg also notes the different animals involved in Northeast Siberia, where it's a mammoth, or in Kamchatka, where it's a giant dog that shakes snow off its fur and causes earthquakes. Not a water-animal, definitely not a fish. He notes that a bull is closer to a mammoth, so it's a closer match. But then, the Turkic urheimat is in the east, near the Altai-Sayan region and Mongolia.
The Uralic cultures, though, all thought the world rested on a giant fish, and its motions caused earthquakes.
That includes East Slavs, who were Uralic speakers in Herodotus' time and likely through about 500 AD, judging from toponyms north of the Black Sea cited in Jordanes. In their folklore, the world rests on a giant pike -- a freshwater fish -- whose movements can cause earthquakes. In other variants, there are 3 fish supporting the world, and like legs of a table, if one or more leave, the world as we know it will cease to exist. In "Crime and Punishment", Dostoevsky refers to the 3 fish that are the foundation of the world.
Why a freshwater fish, not a saltwater serpent? Cuz the Uralic urheimat is inland, but surrounded by rivers and lakes. The Saharo-Arabian urheimat, and that of its Semitic branch, are close to the open salty ocean, near the Horn of Africa (Red Sea, Arabian Sea, Indian Ocean). So it the Sino-Tibetan urheimat, somewhere in Greater Southeast Asia, near the Pacific Ocean. Not to mention Oceanic cultures, of course.
The fact that the Ainu, and the Emishi and Jomon before them, thought their entire world rested on a freshwater fish -- even though they literally live surrounded by the Pacific Ocean -- proves that their cultural origin is further inland, far from the shores of any sea or ocean. By comparing the shibboleth-y points of their "origin of earthquakes" myth, we discover that they're Uralic, confirming an inland origin.
Their only addition to the inland / freshwater narrative, to adapt it to their newfound saltwater environment, is giving the fish a second role as well, as causing the tides by sucking in and spitting out a giant-sized amount of water.
PS, the fact that this myth does not extend further to the east of Lake Baikal, except among the Ainu and Japan, also shows that it is specifically Uralic, not just "Eastern" or "Siberian". It's absent from Northeast Siberia, where Uralic is not spoken. It's absent from China and Indochina and Oceania, where Uralic is not spoken (and where if anything, it's a serpent or dragon, not a fish, that causes earthquakes).
ReplyDeleteIt's also absent from all of the New World, where Uralic is not spoken. Some of them, along the Pacific Ring of Fire, from the Athabaskans to the Haida down to the Andeans, do have myths about a world-supporter being the cause of earthquakes, although it's never a fish. And it isn't always due to the motion of its body, but sometimes getting distracted by some other being, slacking off on the job, before righting itself. It's not an angry thrashing motion.
We can eliminate the Yamato as the origin, since it only goes back to the Early Modern era there. That only leaves the Ainu as the old-time exception to the spatial distribution of this myth. That means they must be an island of Uralic cultural legacy, stranded far from their urheimat -- much like the Tokharians being a far-flung island to the distant east of the Indo-Euro urheimat.
But also like the Tokharians, the Jomon / Emishi / Ainu were likely not long-distance migrants from the Uralosphere, just like the Tokharians were not long-distant migrants from the Western Steppe.
Rather, a small core of influential migrants brought the culture with them, and converted the locals. It's pretty well established that Tokharians used to speak Uralic, specifically Samoyedic, before converting to Indo-Euro, and bringing a bunch of Uralic baggage with them. The same likely applies to the Jomon / Emishi / Ainu, who brought some pre-Uralic local baggage with them after converting to Uralic.
Still, I find it hard to believe that Uralic-speaking migrants got all the way over into Sakhalin island and only there did the conversion of locals take place. I'm guessing the locals they converted were still from the Eastern Steppe or Amur region or perhaps a bit further north, and then after converting to Uralic, they migrated a short distance into Sakhalin, Hokkaido, and then Honshu, where their telltale dogu statues were made during the later stages of the Jomon era.
Ainu borrowed Proto-Indo-European words for "fly" and maybe "cloud". I'll post about mythology in a moment, but I came across another instance of Ainu borrowing from P-IE. In Proto-Ainu, "fly" (insect) is "mos" or "mus". Vovin says "mOs", which in Alonso de la Fuente's refinement means it's "mos". It's a dead ringer for the P-IE word "mus", with the same meaning.
ReplyDeletehttps://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-European/mus-
Coincidence? I don't think so -- it's not typical Ainu phonotactics. It's only 1 syllable, and words of shape [nasal] [vowel] [sibilant] are uncommon in P-A. The nearest match for "mos" in P-A is "mosma" = "other than, besides, anything else", which has nothing to do with "fly", and actually has a proper Ainu (and Uralic) word-shape. Even for "mus", the only other match is "emus" = "sword", which also has nothing to do with "fly", and also has 2 syllables.
So it's reasonable to suspect it being a loanword -- and given the striking resemblance, it must be from P-IE. Not from the daughter branches, all of which add some IE suffix. Seemingly from P-IE itself, borrowing the root "mus".
I speculate the original "u" was slightly lowered to "o" in Ainu due to some phonotactic dislike for "mu" vs. "mo", which has twice as many words in initial position. Maybe to dissimilate roundness, where "u" is more canonically round and "w"-like than "o" (even though "o" is also round). Or Vovin's reconstruction of that vowel for this word is slightly off, and it should be "mus". IDK which.
The P-IE word has not been loaned outside of the I-E family, except for Armenian into Georgian, not a surprise given how close they are. And Pontic Greek "loaned" it into Turkish -- i.e., after Byzantine Greek-speakers converted to speaking Turkic, they kept this old word of theirs.
It did not get loaned into any other Uralic language, so Ainu would be the only example. Still, P-U does not have a word for "fly" or "mosquito" -- nor does Finno-Ugric, only the narrower grouping of Finno-Volgaic but excluding Permian. Since Ainu is from Eastern Uralic, it doesn't use that word, but could have borrowed one instead.
It could not have been mediated by Yeniseian or Turkic, since those families ban initial nasals. Mongolic and Tungusic languages didn't borrowe it either. The only family in close contact with P-IE was P-Uralic, so the branch that Ainu belongs to must have picked it up early on, and it's the only branch in which it survives.
Very few other families have words for "fly" beginning with "m" and having a sibilant as the next consonant -- Northeast Caucasian is another, but that's adjacent to P-IE. So it's not the kind of sound sequence that any ol' language will come up with by imitating the sound of flies buzzing.
The linguistic ancestors of Ainu were in contact with P-IE and borrowed the word for "fly", then brought it all the way over into Ainu-land.
Briefly, "sky, cloud" may also be a loan from P-IE to P-Ainu, where it's "nis" -- another unusual word-shape, like "mos". One syllable, [nasal] [vowel] [sibilant]. It has a few phonetic matches in P-A, but the connection to "sky, cloud" is not so clear. "Nis" also means "to be strong, to be hard", no hit there. It also means "to draw water, to ladle, to dip" -- rain-clouds hold liquid, but clouds don't have anything to do with drawing water or using a ladle to dip into a source. "Nisap" means "shin", no hit there, nor with "nispa" = "rich man".
ReplyDelete"Nisat" means "dawn", which is a bit closer, though dawn is more about the sun rising than the clouds parting.
The P-IE word is "n̥bʰrós" or "n̥bʰrís" = "rain-cloud", the 2nd being the potential donor here due to the "i" vowel instead of "o". Ainu, like Uralic in general, bans initial clusters, so the 3-consonant cluster will have to be dealt with, and deleting offending consonants is more common than inserting zillions of vowels, across languages. There's nothing like "bʰ" in Ainu, so that can't be it.
Why "nis" instead of "ris"? Phonotactics again -- although the word-shape is not typical of Ainu in either case, "nis" at least has 2 homophones, whereas "ris" only has perhaps 1 ("to pick, nip, pluck") or perhaps 0 (since Vovin also says that could be "dis", or "lis" per Alonso de la Fuente). So it sounded better to Ainu ears.
https://en.wiktionary.org/wiki/Reconstruction:Proto-Indo-European/n%CC%A5b%CA%B0r%C3%B3s
This one isn't as clearly a loan as "mos" for "fly", but it came up when I looked for other words of [nasal] [vowel] [sibilant] shape, so worth noting.
OK, back to Ainu mythology being Uralic. Well, in this case, it's more like Eastern Siberian, but including Eastern Uralic cultures like the Samoyeds. We're trying to look for the intersection or center-of-gravity for all the different cultures it resembles, across a variety of examples.
ReplyDeleteEarlier, I discussed the common myth about the origin of mosquitos, where they are somehow the remnants or re-birth of a cannibalistic monster that was killed. The logic being, mosquitos bite you and feed on your blood, so their ancestor must have been a man-eater in some way as well. But after being slain, the ogre was reduced to small pieces -- usually involving a fire that burns its body to ashes, and these are scattered into the wind, where they turn into gnats, flies, mosquitoes, etc.
Japan has a weak version of this, where Buddhist monks believed that mosquitoes were the re-incarnation of people who had behaved wickedly in a previous life. Not an ogre or monster, not a cannibal, no slaying, no burning or reducing to small pieces... it's pretty weak, but it is a weak version of it.
However, the Ainu have a full-fledged version of this myth, whose closest match is the Samoyedic version. I don't have time to format long quotes of Batchelor and Holmberg yet again, but suffice it to say the Ainu version involves a man-eating cyclops, a hunter who without knowing it, stumbles upon the monster, summons the courage to aim his bow at it, strikes it right in its eye, the only vulnerable part of its body, and for good measure, burns it under a big bonfire, and for even better measure, throws its ashes into the winds, where they become gnats, flies, and mosquitoes.
The Samoyedic version involves a hero who sets off deliberately on a quest to slay the man-eating ogre, though it's not described as a cyclops. He kills it, burns it, and the wind scatters its ashes, which turn into mosquitoes.
There are versions of the myth in the New World, which crossed over with the Na-Dene, seemingly, and then loaned into various other cultures as well. But they don't involve a cyclops, and it's more of a disembodied head that must be slain.
Since the closest match is Samoyedic, this example shows yet again the distinctly Uralic nature of Ainu myth.
To return finally to the Indo-Euro word for "fly" being loaned outside the family, Holmberg mentions one version of the mosquito myth among the "Altai Tatars", which isn't as close to the Samoyedic and Ainu versions, but does involve a hero slaying a monster associated with mosquitoes.
ReplyDeleteAlthough found among Tatars, this myth cannot be Turkic, since all the characters involved have names beginning with "m", or compounds using a word beginning with "m", which is banned in Turkic.
The 3 heroes are named Mandyshire, Tyurun-Muzykay, and Maidere, while the ogre is named Andalma-Muus -- where the 2nd word there must derive from P-IE "mus" = "fly", and "alma" is Steppe-region word for ogre, monster, evil spirit (IDK what the "and-" means, though).
I came up with Uralic etymologies for all of their names -- the ogre just above. Mandy means "missing the mark" ("mentä") and "shire" means "old" ("serä"), presumably the eldest hero who missed when he shot. Maidere derives from "majδ̕a", relating to forest hunters. "Turun" is Turkic for "grandchild; niece or nephew" (from "young of an animal"), including in Turkic languages spoken by former Uralics, like Crimean Tatars. And Muzykay from "musɜ" = "to perform magic, tell fortunes, sacrifice".
Maybe there are better etymologies, but aside for "turun", these words cannot be Turkic. So even though the myth is described as belonging to "Altai Tatars", it was not spread by Turkic speakers, and is a relic of their Uralic-speaking past. And that Uralic heritage is shared with the Ainu, as seen in this myth.
But more importantly, this Tatar myth looks like another example of the P-IE word for "fly, mosquito" being loaned outside of the family, in the name of the mosquito-linked monster Andalma-Muus. Since Turkic could not have borrowed that word without altering the initial consonant, someone else did -- and in the Tatar-speaking world, they could only have been Uralic speakers, as hinted at by the names of the other characters.
So it's less unbelievable that it got loaned into the Ainu branch of Uralic.
Ainu source of Japanese "3-path river in underworld" myth CONFIRMED. I finally tracked it down! In fairness to myself, Batchelor stuck this way in the end of his 600-page book, away from the very early discussion about the Ainu underworld being a watery mirror-realm, where people walk upside-down with their soles touching ours. He should've had a single underworld chapter!
ReplyDeleteAnyway...
I've been talking about the "Three Path River" (Sanzu no Kawa in Nihongo) for years now, it's one of the most striking parts of all of Japanese mythology, and it ties into their funerary ritual of including coins with the deceased, another very distinctive part of Japanese culture.
In short, when the dead leave this world behind, they must cross over a river of the underworld. There are 3 paths to cross it -- a safe sturdy bridge, a shallow part that can be waded through, and deep monster-infested waters.
The logic of this myth is that your actions when you were alive *determine* which of those 3 paths you'll take when you're dead -- if you want to take the safe path, behave nobly while you're alive, and don't act wicked, unless you want to be forced to take the dangerous path when you're dead.
However, Japan has a separate ritual, where six coins are placed with the deceased during burial, so that they can pay the fare to the ferryman who escorts the dead across the river of the underworld.
This is a clear parallel to the ancient Indo-European (Greek) ritual of putting coins with the dead so they can pay the ferryman Charon to guide them across the River Styx.
There's another close parallel in Indo-Aryan religion, where a cow (sacred animal) is worshiped and donated well before the dead needs to cross the Vaitarani River in the underworld. This is another form of paying a toll to cross. In this myth, the only influence of your actions while living, is how you experience the river -- positively if you behaved well, being terrified of it if you were wicked. But as far as crossing it, there are no separate paths, and your living conduct does not determine which path you take. As long as you pay the cow-toll, you're good to cross.
Do you see how these two parts of Japanese culture are contradictory? If you need to pay a toll, then your living conduct is irrelevant. And if your living conduct determines which path you take, there's no point in paying a toll to a ferryman. One is superfluous to the other.
This suggests that Japanese myths about the underworld come from two separate sources, and have been folded into a fusion culture. The "pay the toll to the ferryman" myth clearly comes from an Indo-European source, and given that it appears in India, the birthplace of Buddhism, this myth certainly arrived in Japan when they converted to Buddhism during the High and Late Dark Ages. It's not in the earliest Japanese texts on history, culture, myth, and religion -- the Nihon Shoki and the Kojiki.
However, where did the other myth come from -- about the 3-path river that cannot be crossed by paying a toll, but where your entire behavioral track-record from your life determines which path you must take? That is not in the Nihon Shoki or Kojiki either.
ReplyDeleteI figured that out last year -- it's Uralic, as extensively detailed in chapter 5 of the Finno-Ugric section of "The Mythology of All Races". In some form or another, it's present among speakers of the Sami, Finnic, Mordvinic, Mari, Ugric, and although he doesn't specifically mention them, probably the Permic and Samoyedic branches of Uralic.
The land beyond being an underworld rather than up in the skies, it being watery, in some remote northern sublime / wasteland, with a great river to cross, three paths of varying danger, which path you take is determined by your living conduct rather than a toll, guard dogs to deter wicked people from taking the wrong path, a gate around the entire place that must be opened, and the wicked being punished by either a freezing watery drowning death or a burning underwater death (a fiery maelstrom, a cauldron of burning resin, etc., even in the watery underworld).
The paradise section of the afterlife is full of luxuriant plant-life, bountiful food, and villages centered around families, just as in the living world.
All of these shibboleth-y details of the crossing from the land of the living and into the land of the afterlife are present in the Ainu culture. The Ainu do not have a ferryman who collects a toll, there is no particular sacrifice or toll to pay while alive, and no coins or other toll included with the dead during a funeral, to make sure they can cross. The Ainu only have the "three-path river, whose crossing is determined by your living conduct" myth.
The afterlife is indeed an underworld, not a sky-world. Around the whole place is a great iron metal gate which must be opened. There are 3 roads leading into the center of the underworld -- 1 leading from the living-world, and 2 forking toward the punishment section and toward the paradise section of the afterlife. There are guard dogs along the path to make it dangerous for the wicked to go down the wrong path.
Which path the dead takes is based on the entire history of their living conduct -- and in an Ainu innovation, the dead can object if they are condemned to the punishment section. But then, the sun goddess herself appears and displays pictures of the person's entire life, since she has a picture-perfect memory of everyone's behavior, and this evidence leaves no room for objection -- on to the punishment section you go!
As with the other Uralic cultures, the Ainu view the punishment section as watery, like the deep monster-infested water path of the Japanese Sanzu no Kawa. And like the other Uralics, they allow for the punishments to be both wet and freezing, as well as fiery and burning, even while underwater.
Ainu families are reunited in family-centered villages in the afterlife, just as in the other Uralic versions. The landscape is beautiful, food is bountiful, nobody wants for anything. But most importantly, they have re-created their clan-based residential pattern in the afterlife, as it was in the living world.
The Ainu afterlife is 100% Uralic in origin, with only a few local additions or flourishes. And it does not have any of the Indo-European-derived "pay the ferryman a toll" elements that Japan received along with Buddhism from India.
ReplyDeleteTherefore, the Sanzu no Kawa from Japanese culture came from a non-Wa source -- it was from the Ainu, the Emishi, the Jomon, or whatever we call the pre-Wa society of the Japanese islands. And the Jomon in turn got it from a Uralic source, along with their language. Nobody else shares these shibboleths about the nature of the afterlife.
The focus on rivers in the afterlife only began to appear in Japanese religion around 1000 AD, and the specific view of 3 paths of varying levels of danger, was only codefied by around 1300 AD.
https://grokipedia.com/page/Sanzu_River
Where did this highly specifically Uralic view of the afterlife come from, between the arrival of the Wa, and the earliest writings of the 8th C, and the sudden fascination with the 3-path river during the early 2nd millennium AD?
It can only have come from the Jomon / Emishi / Ainu, who still held to the Uralic-only view when Batchelor visited them in the 19th C. That means the emerging Yamato state and culture incorporated an earlier Uralic myth among the islands' native population, as they succeeded in subduing the Emishi and began to assimilate them, politically and culturally, into a new fusion culture led by the Wa-descended Yamato.
But, just cuz the Wa were the leaders, didn't mean the Jomon could be left by the wayside -- to assimilate them, some elements of their indigenous culture needed to be incorporated into the new fusion culture, so that they had representation, and an easy transition from their old culture into the new one. The new fusion was not totally alien to them -- it had crucial elements of their old culture. And it was a sign of good faith by the victors --
"We aren't going to erase your culture and treat you like worthless slaves, we can dignify your culture as well, but we'll all be fusing into a single new culture, not preserving 100% of either of our earlier cultures."
I think the fact that this Uralic afterlife myth entered Japanese culture through the Buddhist, rather than Shinto, sector of society, also reflected the tension between Wa cultural domininance -- reflected more in Shinto, which was brought over with the Wa -- and Wa-Emishi co-evolution, which was reflected more in Buddhism, which did not originate with either the Wa or the Emishi. Rather, it was a new universal religion, converting both Wa and Emishi away from their former religions, into a single religion.
ReplyDelete*That* was the way for Emishi culture to enter the fusion culture -- if they tried to incorporate it into Shinto, it would have seemed "off" to the traditional Shinto practitioners. But Buddhism was not traditional or familiar to either side -- it was recently imported from India, via China. To syncretize it with local cultures, both the Wa and the Emishi could include elements of their earlier cultures into the Japan-specific form of Buddhism.
In fact, the 2nd-most common school of Buddhism in Japan, Nichiren, was born in Eastern Japan rather than in the imperial core of Kansai and the West. And it was founded by a self-described outcast -- perhaps he was from a Jomon / Emishi ethnic background, or an unfortunate cast-off from the Wa. Either way, it appealed to leveling influences, and was challenging the Buddhist establishment from Kansai.
I have to believe this was a kind of struggle between the more Wa-dominated West vs. the more Emishi-dominated East. Even as the Emishi assimilated into the inchoate Yamato state and culture, they fought to preserve some of their old ways, and to ensure they held an influential role in shaping the fusion culture, not just abandon their old culture and go along with whatever the victorious Wa told them to do.
And that is also why the agreed-upon location of the Sanzu no Kawa is near Mount Osore, way up in the North, in Aomori, right across from Hokkaido. It reflects the Jomon / Emishi / Ainu source of the Japanese fusion culture, and must be located near their homeland, as a reminder, and as a way to dignify their culture with important cultural landmarks.
By now, Japanese people don't think of themselves as Wa vs. Emishi -- their fusion culture has been solidified for the better part of a millennium. But in figuring out where their origins lie in the distant past, we have to untangle the tight knot that have tied, and follow the threads wherever they lead.
As it turns out, there is a crucial Uralic element in Japanese culture, it comes from the Ainu / Emishi / Jomon, who must have made a serious culture-wide conversion to Uralic sometime before the Wa showed up during the Yayoi era. No later than 500 BC.
A far-eastern island of Uralic culture in the middle of the Pacific Ocean -- amazing! ^_^