Language as a Clue to Prehistory

[lpetrich]

The authors then grumble about the dismissive approach many Western linguists have toward macro-linguistics, like point out mistakes, then wave hands about areal effects.

Then some estimated dates of origin from glottochronology. kya = thousand years ago

• 
  [*[Sino-Caucasian: 10 kya, North Caucasian, Sino-Tibetan 6 kya, Yeniseian 2 kya, Basque, Burushaski modern.
• Nostratic/Eurasiatic: 12 kya, Altaic 8 kya, Indo-European 7 kya, Uralic 6 kya, Dravidian 5 kya, Kartvelian 4-3 kya, Eskimo 2 kya
• Afroasiatic: 12 kya, Cushitic 9 kya, Omotic, Chadic, Semitic 7 kya, Berber 3 kya
• Austric: 10 kya, Austroasiatic: 7 kya, Austronesian, Tai-Kadai 6 kya, Miao-Yao 4 kya

These dates are rough estimates, and for Indo-Euorpean, the date is likely an overestimate. But all these macrofamilies date back to the beginning of the Holocene, nearly 12 kya.

Borean, their possible ancestor, goes back even further, to 15 - 17 kya. That's close to the Last Glacial Maximum, about 26 - 20 kya.

Looking elsewhere in the world, sub-Saharan African langs divide up into Niger-Congo, East Sudanic, Central Sudanic, Kordofanian, Khoisan, and smaller ones like Songhay and Atlantic.

In the Americas, at least three:

• Almosan (Algic, Salishan, Wakashan, ...) likely also Chukchi-Kamchatkan (Chukotko-Kamchatkan), Nivkh of NE Asia -- Beringian superfamily? Resemblances with Borean?
• Some others (Penutian, Hokan, Mayan, Mixe-Zoque, Maipuran, Pano-Tukanoan, ...) also with resemblances to Borean.
• Some others without detectable external relations (Siouan, Gulf, Otomanguean, ...)

Not much research into long-distance relationships in Australia and New Guinea.

 

[lpetrich]

Back to the EHL 2014 document

The 50 meanings that they use: ashes, bird, black, blood, bone, claw/nail/, die, dog, drink, dry, ear, eat, egg, eye, fire, foot, hair, hand, head, hear, heart, horn, I, kill, leaf, louse, meat, moon, mouth, name, new, night, nose, not, one, rain, smoke, star, stone, sun, tail, thou, tongue, tooth, tree, two, water, we, what, who

They then propose Eurasiatic > Nostratic instead of the more common Nostratic > Eurasiatic. (> superset, < subset).

Eurasiatic: Indo-European, Uralic, Altaic: age 10 - 12 kya
Nostratic: Eurasiatic, Kartvelian, Dravidian: age 12 - 14 kya

Eskimo-Aleut: Eurasiatic
Chukchi-Kamchatkan: Eurasiatic + Beringian?
Yukaghir: Eurasiatic (Uralic) + Beringian?
Beringian: Nivkh, Algic, ...

Dene-Sino-Caucasian is much easier:
( ( (Basque, North Caucasian), (Burushaski, Yeniseian) ), (Sino-Tibetan, Na-Dene) ) age 10 - 12 kya

Afro-Asiatic: (Omotic, (Cushitic, (Semitic, Berber, Egyptian, Chadic, age 8 - 10 kya) ) ) age 10 - 12 kya

Austric: (Austronesian, Tai-Kadai), (Austroasiatic, Miao-Yao), both age 9 - 10 kya. Austric itself may be > 14 kya

One finds (Austric, (Dene-Sino-Caucasian, (Afroasiatic, Nostratic) ) ) -- though Amerindian langs likely also fit in this tree.[/url]

 

[lpetrich]

Khoisan: Juu-Taa and Khoe-Kwadi-Sandawe, both around 10 - 8 kya -- very little evidence of a relationship, and a relationship with Hadza

Nilo-Saharan: East Sudanic, Central Sudanic: each one 8 - 6 kya and together 12 kya. Several difficult-to-classify members like Saharan, Fur, Songhay, ...

Niger-Congo: 12 - 10 kya

Then to the Americas. First, Almosan / Beringian, then a grouping that may be called West Amerind: Mayan, Mixe-Zoque, Uto-Aztecan, Hokan, ... with some S American families like Quechuan possibly included. 12 - 10 kya.

Looking at First-person n and second-person m in Native America: a fresh look by Raoul Zamponi, and combining the lists, I check on which ones have n-m personal pronouns:

• Penutian: ((Maritime)) Tsimshian (?), Chinook (n-m), (Coast Oregon: (n?-.)) Alsea (n?-.), Siuslaw (n-.), Coos (n?-.), ((Inland: n-m)) (Yok-Utian: n-m) Utian (n?-m), Yokuts, (n-m) (Great Basin, Oregon: n-m) Maidu (n-m), (Plateau: n-m) Sahaptian (n-m), Molala (n-m), Klamath (n-m)
• Hokan: Chimariko (n-m), Yana/Yahi (.-.), Karuk (Karok: n-m), Shasta–Palaihnihan (.-m), Pomoan (.-m), Washo (.-m), Esselen (n.-), Salinan (.-m), Yuman (n-m), Seri (.-m), Coahuilteco (n-m), Comecrudan (n?-.), Tequistlatecan (.-m), Jicaquean (Tolan: n-.)
• Penutian (n-m), Hokan (n-m), Mixe-Zoque (n-m), Maipuran (Arawakan: n,t-p), Pano-Tucanoan (.-m), Uto-Aztecan (n-m), ...

So this group has n-m. Could we call it Nimian? In analogy with Nostratic/Eurasiatic being called Mitian.

Beringian: Algic (n-k), Kutenai (u-i), (Mosan) Wakashan (n-s), Salishan (n-'an), Chimakuan (.-.), Chukotko-Kamchatkan (m-t), Nivkh/Gilyak (n-tS)

CK has Eurasiatic m-t pronouns instead of Beringian n-?

The remaining one:

• Gulf: Muskogean (.-.), Natchez (.-.), Tunica (n-.), Atakapa (.-.), Chitimacha (.-.)
• Siouan (.-.), Gulf (.-.), Otomanguean (n?-.)

So n-m is absent from these ones.

 

[lpetrich]

Returning to EHL 2014 it next gets into New Guinea and Australia.

New Guinea has possible macrofamilies Trans New Guinean and West Papuan.

Australia has Pama-Nyungan (most of the continent) and Gunywinygan (western north coast): age 7.5 kya.

Australia - New Guinea: age from comparison: 20 kya -- with almost no resemblance to West Papuan.

The EHL people disclaimed any attempt to find humanity's original language. They want to see how far they can go while being as bold as is reasonably feasible.

Evolution of Human Languages - News -- this progress report is the only item listed there.

 

[lpetrich]

Evolution of Human Languages - Main Research Focus Areas

With macrolinguistic research for all over the world, even if without a lot of recent results.

 Trans–New Guinea languages and  Proto-Trans–New Guinea language -- that family is very recognizable? Yet another mid-Holocene dispersal?

 Pama–Nyungan languages and  Proto-Pama–Nyungan language -- a mid-Holocene dispersal.

I had to do an Internet search, but I found the canonical spelling:  Macro-Gunwinyguan languages in the western north-end projection -- in the  Macro-Pama–Nyungan languages alongside the Tangkic and Garawan families, of north Australia between the two projections.

 

[lpetrich]

John D Bengtson -- mostly on Dene-Sino-Caucasian, especially Basque and North Caucasian, but with some Proto-World: The Proto-Sapiens Prohibitive/Negative Particle *Ma (2020) and Global Etymologies

The second one has the potential problem of counting only the hits and ignoring the misses. One ought to look at all the negation markers to see if this isn't some case of randomly making some m's amidst a lot of other sounds. That can happen through Jespersen's cycle, which I posted about earlier: negation markers can become weakened from using them a lot, then someone has the idea to strengthen them, like no + thing -> nothing. That can be reduced again, sometimes leaving the "thing" part instead of the "no" part.

The first one? Fortunately it gives the semantics of every word and protoform cited.

*tik 1: one, five, ten, only, finger, toe, hand, foot, to point, to show

*pal 2: two, pair, twin, half, one of a pair, other, part, portion, side

Rather broad semantics, even if plausibly related semantics.

 

[lpetrich]

Mother Tongue Journal splash page

Pierre J. Bancel & Alain Matthey de l'Etang -- those two have been working on global etymologies for works for relatives, especially *kaka "older male relative". That one is significant because /k/ is not a typical baby-talk sound. They find more such words for older relatives:

Male: *papa, *tata, *kaka
Female: *mama, *nana, *yaya

with the initial consonants sometimes lost. Consider Proto-Semitic *?abw- "father" (Arabic ?ab, Hebrew ab), *?imm- "mother" (Arabic ?umm, Hebrew em).

Sometimes a word gets applied to the opposite sex:

For "father", Georgian mama < Proto-Kartvelian *mam- Also, Proto-Austronesian *amax (*amah)

For "mother", Georgian deda < Proto-Kartvelian *ded- Though Proto-Austronesian has *ina

If one looks at older relatives, one finds older brothers and sisters, uncles and aunts, grandparents, ... Yes, some languages have separate words for older and younger siblings. So one has a big semantic field yet again.

 

[lpetrich]

Entry *kuan seems to be semantically well-controlled, with most of the entries meaning "dog" or "wolf" and sometimes other canids. There are some different ones, like "lynx" (an Amerind one) and "hyena" (Khoisan). Most of the entries are Nostratic and Afroasiatic and Dene-Sino-Caucasian and Amerind, meaning that this word form may date back to the domestication of dogs.

So I went to Starling macrofamily databases and looked for "dog" and "wolf" and "fox"

Indo-European *k'won- "dog" and *wlkw- *wlp-, *lup- "wolf, fox" I've seen the theory that this variation is taboo deformation.

Eurasiatic *K'uyna "wolf, dog"
from IE,
Altaic *káŋV "dog" from Turkic *KAŋ-čɨk "female dog, female", Tungusic *kači-kān "puppy", Korean *kàŋ- "dog, puppy"
Eskimo-Aleut *qǝnʁa- (~ *qiHǝnʁa-?)

Afroasiatic *kwVHen- from Berber *kun-, W Chadic *kwin-H-, E Chadic *kany-, Omotic *keHen- 'dog'

Sino-Caucasian *χHwĕ́je from N Caucasian *χ_Hwĕje, Sino-Tibetan *qhʷīj ( / *qhʷīn), Yeniseian *ʔɨʔɨn (~x-,-G-,-χ-) "puppy", Burushaski *huk "dog", Basque *hor

Amerind *(a)kuan "dog

Austric PAN *u(ŋ)kuq 'puppy'? (POc. *nkaun 'dog'?) -- Austronesian, Oceanic

Borean *KVNV "wolf, dog"

-

Turning to IE *wlp- "wolf", I find a lot more variation.
Eurasiatic *wVlpV from IE and:
Altaic *ū́ĺpe is from Turkic *ǖĺ "lynx", Mongolian *olbo / olbi "flying squirrel", Tungusic *ulgu-kī "chipmunk", Japanese *bǝsǝ ( ~ -ua) "otter"
Dravidian pùl- ? "tiger"
They are four-legged and furry, but that's about it -- it lumps together large predators and small rodents.

 

[lpetrich]

I've found various numbers for when dogs were domesticated, ranging between 40,000 and 15,000 years ago, and likely around 25,000 years ago, around the Last Glacial Maximum, when the Ice Age glaciers were at their farthest extent.  Domestication of the dog

The word for dog need not be inherited in every case. If domestic dogs spread over Eurasia not long after their domestication, a word for these animals may have spread with them, thus being a Wanderwort, a wander word.

dog/translations - Wiktionary, the free dictionary

English "dog" goes back to Old English, but can't be traced back any further. Something like present-day English informal "pooch". The original word became "hound", a hunting dog. That word is from Old English hund, Proto-Germanic *hundaz, and PIE *kwon-

This word form has plenty of descendants in Indo-Europeandom, though it has often been replaced over the last millennium, as far as I can tell. Spanish has inherited can, from Latin canis, but the usual word is now perro, with informal chucho, both with obscure origin. Slavic has *pisi, with obscure origin, with descendants like Czech pes, Polish pies, Serbo-Croatian pas. Russian sobaka is borrowed from medieval Iranian, in turn a descendant of the PIE form. Hindi has kutta, also with obscure origin. But descendants of the PIE form are well-represented in Germanic, Celtic, Romance, Armenian, and Persian.

Elsewhere, Proto-Uralic has *penä, but being replaced in Finnish, koira, and in Hungarian, kutya.

In the rest of Nostratic, Proto-Turkic has *köpek, Proto-Mongolic *nokai, Korean gae, Proto-Japonic *enu, Inuktitut qimmiq, Proto-Kartvelian *zh'aghl-, Proto-Dravidian *naH-

Afro-Asiatic: Proto-Semitic *kalb-, Coptic ouhol < Egyptian whr, Proto-Berber *aydiʔ, Somali ey

Dene-Sino-Caucasian: Basque txakur < Proto-Basque *zakuR, Proto-Sino-Tibetan *d-kwai-n, Na-Dene: Haida xa, Tlingit keitl, Navajo łééchąąʼí

Austric: Proto-Tai *hma:, Proto-Mon-Khmer *co?, Proto-Hmong-Mien(Miao-Yao) *qluw, Proto-Polynesian *kuli, Proto-Malayo-Polynesian *asu, *anzing, the former from Proto-Austronesian *(w)asu

Amerind: Proto-Algonquian *athemwa, Proto-Siouan *washuke, Proto-Iroquoian *kîr, Chickasaw ofi, Hopi pòoko, Nahuatl chichi, itzcuintli, Mapudungun xewa, Quechua alkho, allqu, allgo, Aymara anu, ...

Niger-Congo: Proto-Bantu *mbóà, Proto-Yoruboid *á-byá, several others < Proto-Atlantic-Congo *-bwa, *-boa

Quite a lot of variation.

 

[lpetrich]

I'll look at some chronology.

The previous interglacial was the  Eemian one, 130 to 115 kybp (thousand years before present), and it was a little warmer than today, though with less CO2. We are living in the  Holocene interglacial.

The  Last Glacial Period is from 115 to 11.7 kybp. Like other glacial periods, it had several  Dansgaard–Oeschger event - warming to almost interglacial temperatures over some 30 - 40 years, followed by cooling back down again over a few centuries.

Evidence of  Behavioral modernity appeared after anatomical modernity, first in Africa roughly 100 kybp, then in northern and western Eurasia around 40 kybp. This would involve full-scale language, since that is a human universal. As I've posted elsewhere, the emergence of our species would likely have been accompanied with speaking only one language.

Some time around 70 - 50 kybp was the departure from Africa of the ancestors of all of the rest of humanity.  Recent African origin of modern humans There is some evidence of a genetic bottleneck - The great human expansion | PNAS - around then, meaning that it was a relatively small population, a population that likely spoke only one language.

The  Last Glacial Maximum, 26 - 20 kybp was a cold and dry period, with expansion of deserts and semideserts. Canada, Northern Europe, the Alps, and the Tibetan Plateau were iced over, and there was much less forest, whether boreal or temperate or tropical. Also Human population dynamics in Europe over the Last Glacial Maximum | PNAS - humanity moved southward for the LGM, then moved northward as the Earth thawed.

This thawing had some reversals on the way:

•  Last Glacial Maximum -- 26 - 20 kybp -- cold
•  Oldest Dryas -- 18.5 - 17 kybp to 15 - 14 kybp -- cold
•  Bølling oscillation -- 14.7 to 14.1 kybp -- warm
•  Older Dryas -- about 200 - 400 years around 14 kybp -- cold -- dividing the  Bølling–Allerød warming
•  Allerød oscillation -- 13.9 to 12.9 kybp -- warm
•  Younger Dryas -- 12.9 to 11.7 kybp -- cold
•  Holocene -- 11.7 kybp to present -- at present temperature by 10 kybp

The beginning of the B-A warming was almost as warm as the beginning of the Holocene and the middle of the YD was almost as cold as the LGM.

 

[bilby]

Some time around 70 - 50 kybp was the departure from Africa of the ancestors of all of the rest of humanity.

Well, that depends what you mean by "departure".

Sure, there must have been a first person who stepped (or sailed) over whatever you select as "the border" from Africa into another continent. But the average rate of migration for humans as they populated the world was around a kilometre per year, or three metres per day.

Land migration was almost certainly not a journey, as we would understand it, with tearful parents waving goodbye to their intrepid children as they marched off to a new continent. It was just a case of each generation roaming a few hundred metres further than their predecessors.

There probably were some moderately long single journeys, such as crossing of straits between landmasses; But in the most part, this wasn't so much a migration as a diffusion; Nobody would ever have died very far away from where he was born, and likely nobody (with the possible exception of those strait crossings) ever felt that they had "departed" from their homelands.

 

[lpetrich]

How far can a person walk in a day? I am interested in knowing both the distance that a fit person could walk, and also an average person could walk. - Quora
For walking a whole day, it is something like 20 miles / 30 kilometers.

If one alternates between a day of foraging - hunting, gathering - and a day of traveling, then one's effective speed is 15 km/day. Much the same is true if one does agriculture. One has to let one's crop plants grow and to let one's animals feed themselves.

If it's 2/3 foraging or farming or herding and 1/3 traveling, then it's 10 km/day, and 5/6 and 1/6, then it's 5 km/day.

Looking at early Indo-European migrations, from the IE homeland of roughly Ukraine (Yamnaya), the first one was to central Anatolia (2,000 km) around 4,000 BCE, and the next two were east to just north of the west end of Mongolia (Afanasievo) (2,500 km), and north then west to Amsterdam (Corded Ware) (3,000 km), about 3,000 BCE.

Yamnaya - Afanasievo I measured with Google Maps, while for the Anatolian one I did Zaporizhia - Istanbul - Ankara, and for the Corded Ware one I did Zaporizhia - Minsk - Amsterdam.

So one could make all these journeys in two years.

 

[lpetrich]

For Bantu speakers, I'll use Douala, Cameroon - Nairobi, Kenya - East London, South Africa -- over 9,000 kilometers. At 5 km/day, that is 5 years of traveling.

Austronesian speakers did some impressively long sea journeys using Neolithic technology. From mainland China to Taiwan, their homeland, it is 180 km. The next step, from Taiwan to Luzon, the northernmost Philippine island, it is 400 km. One can then island hop from there to the rest of Philippines, Indonesia, New Guinea, and the Solomon Islands, going some 100 - 400 km with each hop. Going further into the Pacific Ocean requires some 1,000 - 2,000 km or more from island to island. From the Marquesas Islands to Hawaii is 3,600 km and to Rapa Nui 3,700 km, and from the Cook Islands to the North Island of New Zealand 2,800 km.

Turning to the colonization of Madagascar, the colonists departed from the Barito River in south Borneo, and likely went a long way, looking for a place to settle, until they discovered Madagascar. A direct trip would take over 7,000 km, while going along the coast of the Indian Ocean would be much easier -- one could take on supplies and repair one's boats as one went. But longer: 16,000 km.

How long to travel? A typical sailboat speed is about 5 knots or 9 km/h or 200 km/day. Rowing is slower, at about 2 knots or 4 km/h or at 8 hours of rowing per day, 30 km/day.

So if one can get a good wind, one only needs a day or two to go island hopping out to the Solomon Islands, but a week to a month between islands in Polynesia.

 

[lpetrich]

Turning to the  Southern Dispersal the first step was crossing the [wikil]Bab-el-Mandeb[/wiki] strait between East Africa and the Arabian Peninsula. At present, the distance between Africa and Perim / Mayyun Island is 21 km, with 4 km the rest of the way. Its maximum depth is 186 meters at present, and with an Ice-Age drop in sea level of around 100 m, that means 86 m.

I also found OpenSeaMap - The free nautical chart in the tradition of OpenStreetMap and OpenRailwayMap, but it isn't very detailed. Turning to the I-boating charts - I estimate a width of about 10 km, with Perim connected to the mainland.

I must note that 5 km/day is about 1,800 kilometers/year.

Going some 1,400 kilometers northward, the next step is to cross the  Strait of Hormuz, about 39 km at present with a maximum depth of about 100 m in its shallower parts. During the last Ice Age, it wasn't an extension of the Indian Ocean but instead of the Tigris and Euphrates Rivers. So it was easy to cross it by boat. Measurement of tidal and residual currents in the Strait of Hormuz - ScienceDirect and Shoreline reconstructions for the Persian Gulf since the last glacial maximum - ScienceDirect

From the Strait of Hormuz to Bali, Indonesia along the south coast is a distance of around 12,000 - 13,000 kilometers, and would take some 7 years to walk.

 

[lpetrich]

Back to paleolinguistics, it's somewhat tempting to associate the early part of the Bølling-Allerød warming with the dispersal of some language macrofamily, from it being warm enough for its speakers to more easily disperse.

But that was about 14.7 kybp and the beginning of the Holocene, at a similar temperature, was at about 11.7 kybp. But the error bars I've found for lexicostatistics calculations usually have larger relative sizes.

Rapid radiation of the inner Indo-European languages: an advanced approach to Indo-European lexicostatistics is typical -- some 1,000 to 1,500 out of 5,000 to 6,000 years.

Going back farther gives larger error bars:
Triangulation supports agricultural spread of the Transeurasian languages | Nature

Our results indicate a time-depth of 9181 bp (5595–12793 95% highest probability density (95% HPD)) for the Proto-Transeurasian root of the family; 6811 bp (4404–10166 95% HPD) for Proto-Altaic, the unity of Turkic, Mongolic and Tungusic languages; 4491 bp (2599–6373 95% HPD) for Mongolo-Tungusic; and 5458 bp (3335–8024 95% HPD) for Japano-Koreanic (Fig. 1b). These dates estimate the time-depth of the initial break-up of a given language family into more than one foundational subgroup.

 

[bilby]

If one alternates between a day of foraging - hunting, gathering - and a day of traveling, then one's effective speed is 15 km/day. Much the same is true if one does agriculture. One has to let one's crop plants grow and to let one's animals feed themselves.

My point is that these speeds, which represent the maximum plausible rate of travel, massively exceed the actual speed at which "migration" actually occurred, to the point where it's unlikely that any of the people involved actually felt like they were travelling at all.

A population that is spreading at one or two km per annum looks, to its inhabitants, like it's not moving at all. Yet it can populate the entire world in less time than archaeological evidence suggests was actually taken to do so.

Humanity came out of Africa and took over the world; But no individual human did anything of the sort - they could have all died within a few hours walk of their birthplaces, and still have achieved enough to fill the world with people.

 

[lpetrich]

Fair enough.

About the  Southern Dispersal it would have stopped at  Sundaland for a while, the western Indonesian islands with exposed continental shelf as a further extension of Southeast Asia. But once one learns how to travel a few hundred kilometers by boat, one can travel to the eastern Indonesian islands --  Wallacea -- and to  Sahul -- Australia and New Guinea as connected land.


I've found Journal of Language Relationship - Вопросы Языкового Родства - Voprosy YAzykovogo Rodstva - Questions of Linguistic Affinity

Lots of articles, mostly in English, but some in Russian, and a few in French and German.

Also Journal of Language Evolution | Oxford Academic

 

[lpetrich]

 Dolgopolsky list - noting Gipoteza drevnejšego rodstva jazykovych semej Severnoj Evrazii s verojatnostej točky zrenija - The hypothesis of the ancient origin of the language families of Northern Eurasia with the probable point of maturity

English translation: Shevoroshkin & Markey (eds.) - Typology, Relationship, and Time (1986) : Allan R. Bomhard : Free Download, Borrow, and Streaming : Internet Archive

In the middle is this paragraph with his results.

В результате составлен список значений, расположенных по рангу сохраняемости морфем. В начале списка располагаются следующие значения:
1) 0 замен: «пять», «три», «четыре», «шесть», 1-е лицо ед. числа;
2) 1—1,5 замены: «два», «семь», «восемь»;
3) 2—2,5 замены: 2-е лицо ед. числа;
4) 3—3,5 замены: «кто»;
5) 4—4,5 замены: «десять», 1-е липо мн. числа, «один», «язык» (в смысле нем. Zunge), 2-е лицо мн. числа, «девять»;
6) 5—5,5 замен: «имя»;
7) 6—6,5 замен: «сто», «что»;
8) 7—7,5 замен: «глаз», «двадцать», «сердце»;
9) 8—8,5 замен: «зуб», «не» (запретительное), «не» (отрицательное при глаголе), «гнида»;
10) 9—9,5 замен: «ноготь», «вошь», «месяц», «слеза»;
11) 10—10,5 замен: «вода», «мертвый», «рука» (в смысле нем. Hand);
12) 11—11,5 замен: «ночь», «кровь»;
13) 12—12,5 замен: «рог», «полный», «солнце», «ухо», «соль».

Google:
As a result, a list of values was compiled, arranged according to the rank of morpheme persistence. At the beginning of the list are the following values:

Bing, Babylon, FreeTranslations:
As a result, a list of values arranged by the rank of persistence of morphemes is compiled. At the top of the list are the following values:

DeepL:
The result is a list of values arranged according to the rank of morpheme persistence. The following values are located at the beginning of the list:

Reverso:
As a result, a list of values is made, located according to the ranking of morph persistence. At the top of the list are the following values:

PROMT.One:
As a result, a list of values located according to the morpheme storage rank was compiled. At the beginning of the list are the following values:

Elan, Translate.com:
As a result, a list of values arranged by the rank of persistence of morphemes is compiled. At the top of the list are the following values:

TRT:
As a result, a list of semantic values was compiled in which all values were ranked according to their degrees of morphemic stability. The list is headed by the following semantic values:

"Morpheme" is a word or word part that cannot be analyzed further.

I'd earlier posted the translated version in TRT, and for the most part it agrees with the autotranslators. I also checked the translations with this online dictionary: Russian Dictionary

0: 5, 3, 4, 6, 1SG
1: 2, 7, 8
2: 2SG
3: who?
4: 10, 1PL, 1, tongue (body part), 2PL, 9
5: name
6: 100, what?
7: eye, 20, heart (body part)
8: tooth, prohibitive NEG, verbal NEG, nit (louse egg)
9: finger/toenail, louse, Moon, tear (from eye)
10: water, dead, hand (body part)
11: night, blood
12: horn, full, Sun, ear (body part), salt

For the pronouns, the autotranslators all agreed on abbreviating the ordinals as 1st and 2nd, but the rest they stumbled over. Also, for what I've listed as "Moon" the original was "month" and TRT has "new moon, crescent of the moon".

 

[lpetrich]

(PDF) Borrowability and the Notion of Basic Vocabulary
(journal paper) Borrowability and the notion of basic vocabulary:

This paper reports on a collaborative quantitative study of loanwords in 41 languages, aimed at identifying meanings and groups of meanings that are borrowing-resistant. We find that nouns are more borrowable than adjectives or verbs, that content words are more borrowable than function words, and that different semantic fields also show different proportions of loanwords.

In the authors' sample:

Word type	Loanword fraction
Nouns	31.2%
Verbs	14.0%
Adjectives and adverbs	15.2%
Content words	25.2%
Function words	12.1%
All words	24.2%

For some curious reason, adjectives and adverbs are borrowed at nearly the same rate as verbs are, and function words a little bit less.

Borrow rate by semantic field:

Religion and belief 26 41.2%, Clothing and grooming 59 38.6%, The house 47 37.2%, Law 26 34.3%, Social and political relations 36 31.0%, Agriculture and vegetation 75 30.0%, Food and drink 81 29.3%, Warfare and hunting 40 27.9%, Possession 46 27.1%, Animals 116 25.5%, Cognition 51 24.2%, Basic actions and technology 78 23.8%, Time 57 23.2%, Speech and language 41 22.3%, Quantity 39 20.5%, Emotions and values 48 19.9%, The physical world 75 19.8%, Motion 82 17.3%, Kinship 85 15.0%, The body 159 14.2%, Spatial relations 75 14.0%, Sense perception 49 11.0%, All words 1391 24.2%

One reviewer noted that some fields have a larger fraction of verbs than others, and that that may make a difference -- sense perception is high in verbs.

The authors then create their  Leipzig-Jakarta List -- it has a lot of overlap with Swadesh's and Dolgopolsky's lists but not complete overlap. It must be noted that stability against borrowing is not complete stability, because there may be some words that are seldom borrowed but instead often internally replaced.

 

[Swammerdami]

If one alternates between a day of foraging - hunting, gathering - and a day of traveling, then one's effective speed is 15 km/day. Much the same is true if one does agriculture. One has to let one's crop plants grow and to let one's animals feed themselves.

My point is that these speeds, which represent the maximum plausible rate of travel, massively exceed the actual speed at which "migration" actually occurred, to the point where it's unlikely that any of the people involved actually felt like they were travelling at all.

A population that is spreading at one or two km per annum looks, to its inhabitants, like it's not moving at all. Yet it can populate the entire world in less time than archaeological evidence suggests was actually taken to do so.

Humanity came out of Africa and took over the world; But no individual human did anything of the sort - they could have all died within a few hours walk of their birthplaces, and still have achieved enough to fill the world with people.

I think the net speeds were even faster than your estimate. I've attached a wiki image that shows 2000 years net travel time from Canada to the tip of South America: that's over 5 km/year. That's NET speed; some migrations would have been much faster. And many experts believe humans were in Monte Verde, Chile by 14,800 BP, even earlier than in the Wiki graphic.

(I seem to get an error when I enclose the following URL in Image tags.)

https://upload.wikimedia.org/wikipedia/commons/8/8b/Early_migrations_mercator.svg

Y-DNA haplogroups shed some light on migration speeds, but the most interesting examples might involve travel by ship or wagon.