Language as a Clue to Prehistory

[lpetrich]

Vyacheslav Vsevolodovich Ivanov of the IE glottalic theory, which he worked on with Tamaz Valerianis dze Gamkrelidze, has an article on protolanguages as objects of scientific description, as opposed to systems of sound correspondence. Rather arcane, I must say. Much of it was about the possibility that Proto-Nostratic distinguished between volitional and non-volitional verbs, being "active-passive", and that this distinction had traces in some of its descendants.

Then Aharon Dolgopolsky's "A Probabilistic Hypothesis Concerning the Oldest Relationships Among the Language
Families In Northern Eurasia"

That's a GREAT article.

A primary difficulty in resolving this issue resides in the lack of a mathematically rigorous procedure to substantiate linguistic relationships. When dealing with closely related languages whose genetic affinity is rather obvious, as for example, in the case of Turkic vs. Indo-European, we may dispense with such rigorous proofs. It is quite another matter, however, with distance linguistic relationships. Here, we require mathematically rigorous methodologies which will permit us to distinguish between languages that are actually genetically related and those that are merely fortuitously similar.

1. "Comparison of several languages" -- that gives better statistics.
2. "Statistical selection of semantic values represented by morphemes (word parts) which are relatively impervious to change." -- that's what's behind the Swadesh list and similar lists. AD essentially rediscovers that kind of list.

The first step is to find those stable word forms. As an example of different stability, AD compared words for "star" and "lightning" in several language families. "Star" was much more stable than "lightning" in most of them.

AD then looked in several language families and for each meaning, counted replacements. Using language histories would have been ideal, but not many langs have long written histories.

He investigated 140 Eurasian langs, and he wanted to also use langs from elsewhere, but their data was sometimes limited, no words noted for "heart", "louse", "horn". Their stability trends were similar, however.

He then consulted Carl Darling Buck's Dictionary of Indo-European Synonyms, using only those with at most 10 replacements, and Robert B. Lees's list of relatively stable meanings for langs with long written histories. These, along with some others that it might seem good to check, were collected into a list of 250 meanings.

He found some 40 meanings with highly stable word forms, with at most 12 replacements for each one. He then trimmed his list down further, using various criteria, and he decided that "who" and "what" deserve only one entry, unlike in the Swadesh lists. That left him with 15 meanings.

Not much for good statistics, it seems to me. Martine Robbeets used 254 meanings for Transeurasian (Macro-Altaic, Broad Altaic), discovered in a similar manner.

 

[Swammerdami]

Do vulgar words sometimes persist longer than expected? English S**t, F**k, etc. are relatively old yet persist. When I was young, F**k was a word that children (in our town) NEVER heard, yet is now one of the most common words in English, perhaps 2nd-place behind "the."

C*nt and (French) pute trace back to proto-IE I think, yet are now mostly obscene.

Thai pronouns are VERY complex. Read what Wiktionary has to say about the Thai pronouns that trace back to proto-Austric:

Pronoun​

กู • (guu)

1. (now considered vulgar and offensive) a first person pronoun: I.

Pronoun​

มึง • (mʉng)

1. (vulgar, derogatory, offensive) a second person pronoun: you.

These words are the ONLY residue from the ancient proto-Autric and proto-Daic pronouns. As an example of change, /khun/ -- a polite 'you' -- derives from a Sanskrit word meaning 'virtue.'

I sometimes use these vulgar pronouns (guu, meung) playfully, to feign anger. Thais understand I'm joking but always object and say that I should NEVER use these words.

These pronouns are NEVER used except in vulgarity, or among children addressing their same-age friends! I asked my informants if these vulgar words, never taught in school, were likely to disappear. They laughed at such a stupid suggestion.

 

[lpetrich]

Do vulgar words sometimes persist longer than expected? English S**t, F**k, etc. are relatively old yet persist. When I was young, F**k was a word that children (in our town) NEVER heard, yet is now one of the most common words in English, perhaps 2nd-place behind "the."

No idea. The first one was earlier a more neutral word, like present-day "dung", and it goes back to Proto-Germanic. It is likely from PIE *skeyd- "to split, divide", a root which also gives us present-day English "to shed".

The second one earlier had meanings like "to hit, strike", without anything sexual.

C*nt and (French) pute trace back to proto-IE I think, yet are now mostly obscene.

The first one can be traced back to Proto-Germanic, and the second one is possibly from a Latin word for "girl".

Words for "to fart" go back to PIE: *perd- "to fart loudly" and *pesd- "to fart softly"

Thai pronouns are VERY complex. Read what Wiktionary has to say about the Thai pronouns that trace back to proto-Austric:

กู - Wiktionary, the free dictionary - guu
มึง - Wiktionary, the free dictionary - meung

These words are the ONLY residue from the ancient proto-Autric and proto-Daic pronouns. As an example of change, /khun/ -- a polite 'you' -- derives from a Sanskrit word meaning 'virtue.'

I sometimes use these vulgar pronouns (guu, meung) playfully, to feign anger. Thais understand I'm joking but always object and say that I should NEVER use these words.

These pronouns are NEVER used except in vulgarity, or among children addressing their same-age friends! I asked my informants if these vulgar words, never taught in school, were likely to disappear. They laughed at such a stupid suggestion.

A List of All Common Thai Pronouns and How to Use Them Like a Pro - although Thai pronouns are not inflected, but they vary a lot by social context.

กู (guu):

From Proto-Tai *kuːᴬ (“singular first-person pronoun (weak)”).
Cognate with Lao ກູ (kū), Lü ᦅᦴ (kuu), Tai Dam ꪀꪴ, Zhuang gou.
Compare Proto-Hlai *ɦuː (“I”), Proto-Austronesian *aku (whence Malay aku, Tagalog ako and Javanese aku).

มึง (meung):

From Proto-Tai *mɯŋᴬ (“singular second-person pronoun (weak)”).
Cognate with Lao ມຶງ (mưng), Lü ᦙᦹᧂ (mueng), Tai Dam ꪣꪳꪉ, Zhuang mwngz.
Compare Proto-Hlai *C-mɯː (“you (singular)”).

Both of them go back to  Proto-Kra–Dai language
I (1.SG) *akuː
thou (2.SG) *isuː; amɤː

 Austronesian personal pronouns
Identical with Proto-Malayo-Polynesian ones unless indicated otherwise
1SG *i-aku
2SG *i-(ka)Su -- PMP *i-kahu
1PLinc *i-(k)ita
1PLexc *i-(k)ami
2PL *i-kamu -- PMP *i-kamu, *ihu

 Austro-Tai languages
# Gloss p-Austronesian p-Tai p-Kam-Sui p-Hlai Gelao (Anshun) p-Kra-Dai tone
49 I *aku *kuu (*ju) *hou (Baoding) kuu (Buyang) *A
50 you *kamu *mɯŋ *maa (Lakkia) *meɯ maa (Buyang) *A

Also to  Proto-Austroasiatic language and  Appendix:Proto-Austroasiatic reconstructions
1SG *ʔaɲ
2SG *miːʔ

So one may conclude Proto-Austric *aku and *mi

 

[lpetrich]

Aharon Dolgopolsky's second step was to find common sound changes, and to encode them by dividing consonants into consonant classes, by point of articulation.

His third step is to compare some language families using reconstructed protoforms as much as possible, with their sounds encoded with consonant classes.

He then calculated the probability of getting the amount of match that he found, though some more recent work uses a probability measure that better captures the statistics of sound frequency: calculated matches for scrambles of the word forms.

He concluded that Indo-European, Uralic, Altaic, Chukchi-Kamchatkan, Kartvelian, and Afroasiatic were likely related, and Sumerian less likely related to these ones.

Half a century later:

• Ultraconserved words point to deep language ancestry across Eurasia | PNAS - Mark Pagel et al. - 2013
• Support for linguistic macrofamilies from weighted sequence alignment | PNAS - Gerhard Jäger - 2015
• Bayesian phylolinguistics reveals the internal structure of the Transeurasian family | Journal of Language Evolution | Oxford Academic - Martine Robbeets et al. - 2018
• Triangulation supports agricultural spread of the Transeurasian languages | Nature - Martine Robbeets et al. - 2021

 

[lpetrich]

Then VV Ivanov's reviews of the Nostratic Etymological Dictionary of Vladislav Markovich Illich-Svitych.

"The main distinction between Illic-Svityd's work and other earlier or even contemporaneous attempts at a broader comparison of the principal language families of the Old World lies in the exceptional precision of his methodology."

TRT also includes an article from a skeptic about Nostratic, BA Serebrennikov, stating that some of VIS's comparisons are flawed, that it's hard to get Semitic C-C and C-C-C roots from Nostratic CVC, that Nostratic's consonants and vowels are reconstructed from different descendants, etc. Shevoroshkin and Markey rebutted him in their notes.

Finally, TV Gamkrelidze and VV Ivanov give us their glottalic reconstruction of IE's stop consonants' voicing.

The traditional system is T - D - Dh, inspired by the traditional big three of Indo-European studies: Latin, Greek, and Sanskrit.

But *b is very rare, and between /b/ and /p/, it is always /p/ that drops out. So the system must be T1 - T2 - D, like in Germanic and Armenian. So they propose T - T' - D where T' is glottalic or ejective (short pause between the consonant and the following vowel).

This accounts for the absence of IE roots of form T'1-T'2 (trad. D1-D2) where 1 and 2are different, a common restriction on glottalic consonants in roots. The non-glottalic ones appear together only as T1-T2 (trad. T1-T2) or D1-D2 (trad. Dh1-Dh2), though any of them can occur with a glottalic one in either order. In D1-D2 roots, one of them must be aspirated: Dh1-D2 or D1-Dh2.

S and M then give their own theories about these consonants' voicings, and how they relate to other members of Nostratic.

 

[lpetrich]

Internet Archive: Shevoroshkin - has Vitaly Shevoroshkin's collections of macro-linguistics papers.

Allan Bomhard has generously uploaded those and some more macro-linguistics books:
[https://archive.org/search?query=creator:"Allan+R.+Bomhard"]Internet Archive: Allan Bomhard's uploads[/url]

Among them is "Lamb & Mitchell (eds.) - Sprung from Some Common Source (1991)" and "Ruhlen - A Guide to the World's Languages, Vol. 1 - Classification (1991)" and "Ruhlen - The Origin of Language - Tracing The Evolution of the Mother Tongue (1994)"

"Sprung from Some Common Source" also discusses some early Indo-European culture, like how being a sexual bottom was a disgrace for a man, yet a man could redeem himself by killing some big predator, like a bear. Presumably demonstrating his virility.

Merritt Ruhlen's "Guide to the World's Languages" is the first volume of a planned series, a series which never materialized. What might MR have discussed? Phonology? Grammar? Semantics?

But that book does have some history of historical linguistics, complete with the difficulties it had in early modern times. Like trying to derive European languages from Biblical Hebrew.

His "The Origin of Language - Tracing The Evolution of the Mother Tongue" is a less-technical introduction to historical linguistics and macro-linguistics. A nice feature of it is some short vocabulary lists from various languages and language families, so one can try one's hand at language classification.

 

[lpetrich]

I have several collections of papers from the First International Interdisciplinary Symposium on Language and Prehistory (November 8-12, 1988, Ann Arbor, Michigan)

Shevoroshkin (ed.) - Reconstructing Languages and Cultures (1989)_text.pdf

VS started off with grumbling about American linguists being reluctant to accept that families like Penutian and Hokan are families by ancestry rather than proximity (areal effects, Sprachbund). Even the N-M pronoun pattern was often dismissed by them.

Later 1 understood that this negative attitude was a reaction to some unbridled comparisons and reconstructions of the 1950"s and 1960"s. These "reconstructions'" were apparently made by scholars trained as anthropologists who considered linguistics to be a step-daughter of anthropology. No method whatsoever was used in these works. It was acceptable to compare anything with anything (both semantically and phonetically).

VS then noted that many American linguists had never heard of Vladislav Illich-Svitych and his work on Nostratic.

The book has a biography of VIS himself, "Slava", noting his careful work, and noting the huge volume of notes that his colleagues worked from after his death from being hit by a car.

He worked on Afroasiatic,, Kartvelian, Indo-European, Uralic, Dravidian, and Altaic.

John Bengtson has an article on the "fallacy" of "diminishing returns" in long-range comparison. It's not an outright fallacy, and JB argued that with suitable methods, one can get better results than what one might expect. Like comparing several languages and language families and not just two.

Vaclav Blazhek has a bit of Proto-World, mostly words for body parts and "hot", "fire", "rain". But with a lot to choose from, there is still a strong possibility of coincidence.

 

[lpetrich]

Shevoroshkin (ed.) - Reconstructing Languages and Cultures (1989)_text.pdf

Vladislav Illich-Svitych discovered a remarkable Nostratic sound correspondence, between Indo-European and "East Nostratic" velars -- EN including Uralic, Altaic, Dravidian.

IE *Kw (labiovelars) -- EN *Ku/o/ü/ö (back vowels)
IE *K' (palatovelars) -- EN *Ki/e/ä (front vowels)
IE *K (plain velars) -- EN *Ka (central vowels)

The vowel became merged into the consonant, making it vary.

"West Nostratic" - IE, Kartvelian, Afroasiatic (I like "Afrasian" for that one)

A mention of Dravidian prehistory.

Dravidian speakers mostly live in southern India, but there are a few elsewhere, speaking Brahui in the northwestern Indian Subcontinent and Kurukh and Malto in the northeast. The three branches split around 3000 BCE, and the southern branches around 2000 BCE.

From a combination of reconstructed agricultural vocabulary and archeological evidence, the South Dravidian homeland is likely in western India, near present-day Ahmedabad.


Toward the end, a list of Vladislav Illich-Svitych's Nostratic reconstructions, each one sourced with reconstructions for the families that he worked from, like:

**barA 'take': IE *bher- 'take, carry'; Alt. *bary- 'take'.

 

[lpetrich]

Shevoroshkin (ed.) - Explorations in Language Macrofamilies (1989)_text.pdf

Sergei Starostin has "NOSTRATIC AND SINO-CAUCASIAN" in it, asking lots of questions, like "does there exist a distinct Altaic branch of Nostratic?" and can Nostratic be divided into Eastern and Western branches?

Altaic he concludes to be a well-defined group, but Afroasiatic is a different story, being internally divergent as the rest of Nostratic.

He concludes that VIS over-reconstructed the back-of-tongue sounds, distinguishing between k (velar; upper back) and q (uvular; lower-back). That is only from Kartvelian, and he thinks that that was areal, that distinction emerging from North Caucasian influence.

Also, that Core Nostratic (N - AA), Afroasiatic, and Sino-Caucasian are members of some larger family, sometimes called Borean, though he doesn't think that we are in good position to reconstruct it.


Then an article on Nivkh/Gilyak and Chukchi-Kamchatkan as Almosan-Keresiouan languages by Oleg Mudrak and Sergei Nikolaev. Eskimo-Aleut, however, is Nostratic, closest to Altaic.

Then a list of "Kamchukchee Roots"

The following list represents the reconstruction of Kamchukchee (- Chukchi-Kamchatkan) roots by 0. Mudrak. According to Mudrak and Nikolaev, Kamchukcheee, along with Gilyak (= Nivkh), is related to Almosan-Keresiouan, a phylum of Amerind (in Greenberg’s terminology). According to Dolgopolsky, Greenberg, et. al„ Kamchukchee is related to Nostratic.

Also has three detailed VIS reconstructions, like for *bari "to take", from Afroasiatic (Semitic, Berber, Cushitic Chadic), IE, Altaic (Turkic, Mongolian), Dravidian (?), listing every attested form that the reconstructions are based on.

 

[lpetrich]

Shevoroshkin (ed.) - Proto-Languages and Proto-Cultures (1990)_text.pdf

In his introduction, Vitaly Shevoroshkin mentions comparisons between Nostratic, Sino-Caucasian, and Amerind, and he continues:

... I compared Salishan languages with Nostratic, and I got some correspondences which, apparently, will hold. But as soon as I started to compare Salishan with Sino-Caucasian, I saw that practically all stablest roots (pronouns T, 'thou'; numerals 'two', 'three'; terms for body parts, etc.) show clear matching between Salishan and Sino-Caucasian (mostly between Salishan and North-Caucasian). This kind of relationship is much closer than that between Salishan and Nostratic.

John Bengtson: "Notes on the Sino-Caucasian Affinity of Sumerian" -- though I've seen Sumerian compared to Nostratic.

Vladimír Orel and Sergei Starostin: "Etruscan as an East Caucasian Language" -- though I've seen Nostratic also. They find comparisons for about 60 Etruscan words, about half of those with a confidently-interpreted meaning. "What we do not discuss in detail here is the problem of morphological analysis due to the poor and contradictory state of Etruscan grammatical description."

I've also come across a book, Igor Diakonoff and Sergei Starostin: "Hurro-Urartian as an Eastern Caucasian Language"

 

[lpetrich]

Shevoroshkin (ed.) - Proto-Languages and Proto-Cultures (1990)_text.pdf

Then Joseph Greenberg on Indo-European vowel alternations. He proposes that those alternations are the result of breakdown of an earlier system of vowel harmony.

An example of vowel harmony is in Turkish plurals. The plural suffix is -lar or -ler, depending on the previous vowel:

• Back: a, ï, o, u -lar
• Front: e, i, ö, ü -ler

I use ï for Turkish dotless i.

JG notes Chukchi vowel harmony, with harmony by height:

• High: i, ä, u
• Low: e, a, o

with a schwa in between.

As appears from the preceding exposition the Indo-European variants i ~ e, e ~ o, u ~ o show the same characteristic of intradialect variation, cross language variation, semantic differentiation and occasional grammaticization which characterized languages with formerly functioning vowel harmonic systems.

i ~ e and u ~ o are high-low, while e ~ o is front-back.

JG then mentions that the usual reconstructions of IE don't have /a/ as a primary vowel sound, a very atypical feature. Some Indo-Europeanists have noted that /o/ often shows up as /a/, and they propose that e ~ o ought to be e ~ a, with a > o instead of o > a.

Thus making all three alternations high-low:
i ~ e, e ~ a/o, u ~ o

much like Chukchi.

e ~ a/o was grammaticalized as ablaut. It's present in the noun and verb inflections of the older IE langs and it survives in some

I note in passing another form of vowel harmony: umlaut. It's in the Germanic langs, with the root vowel getting shifted to approximate some suffix vowels with the resulting vowel often written with two dots over it. It's in the opposite direction from Turkish vowel harmony.

 

[lpetrich]

Shevoroshkin (ed.) - Dene-Sino-Caucasian Languages (1991)_text.pdf

Has Sergei Starostin's paper proposing Sino-Caucasian: North Caucasian, Yeniseian, Sino-Tibetan

Also Sergei Nikolaev's follow-up that added Na-Dene. He also mentioned Algonquian and Salishan as SC members.

Then John Bengtson's Notes on Sino-Caucasian.

• Likely members: Basque, North Caucasian (+ Hurro-Urartian, Hattic), Sumerian, Burushaski, Yeniseian, Sino-Tibetan, Na-Dene
• Possible members: Etruscan, Nahali, Nivkh/Gilyak, Chukchi-Kamchatkan, Almosan-Keresiouan

He makes an odd claim:

Negative evidence is irrelevant. Thus, the claims that Indo-European and Uralic, or Basque and Burushaski, canaot be related because the respective pairs of languages share few or no numeral isoglosses, агеto be dismissed. Linguists who make such claims ignore the possibility of vocabulary replacement over mапу millennia, and the fact that nunerals, beyond 'two' or 'three', have evolved within relatively recent times. (Even in Indo-Suropean, there is no single common vvord for 'one'!) The comparative linguist cannot expect long-range language groups to behave in every respect like the familiar Indo-European.

All negative evidence? But I agree on some sorts of negative evidence, like words for numbers. People at low levels of technology tend to have only a few number words, like 1 and 2, sometimes 3, less likely 4. When one looks far enough to have to do macro-linguistics, one is looking back to people with low levels of technology, people unlikely to have a lot of number words.

Then mentioning possible Eurasiatic (Nostratic) - Sino-Caucasian - Amerind connections -- Borean.

He then discussed what he calls Macro-Caucasian: North Caucasian, Basque, Burushaski.

I am finding interesting things about Dene-Caucasian substratum in Europe. Besides the more obvious cases in Iberia, Balkans, Alps, and Sardinia, I think there may have been Dene-Caucasian substrata even farther north in Europe. I have been collecting Germanic words that have eluded IE explanations, but may have DC parallels, e.g.:

• Proto-Germanic *lambaz > English "lamb", etc. -- NC *tlamba-kV "sheep"
• PGmc *gatwôn > German Gasse "street" , etc. -- NC *GGwât'V "street"
• PGmc *dawwan (dawwô, dawwaz) > English "dew", etc. -- NC *dwiX/V "snow, hoarfrost, dew", Werchikwar (Bur) dau "rain"
• PGmc *hêran > English "hair", etc. -- NC *kiwrV, ST *kra

NC = North Caucasian, ST = Sino-Tibetan, Bur = Burushaski

I checked all the North Caucasian etymologies in Tower of Babel: Databases -- they are all there.

 

[lpetrich]

Shevoroshkin (ed.) - Nostratic, Dene-Caucasian, Austric, and Amerind (1992)_text.pdf

The last one of this series.

Irén Hegedüs has an article on derivational elements in Nostratic: -lV, -na, -an, -t

Then Roger Wescott on "protohistory", a sort of intermediate between prehistory and history -- rather poorly defined, it must be noted. He also considers some crackpottery, I must say -- collapsing early Mesopotamian history.

Then a lot of discussion on why American linguists have not received Nostratic with much enthusiasm. This was back in 1988, and it's been 35 years since that time.

Katherine Rowenchuk: Why aren't Americans interested in Nostratics? She claims that Americans are much less interested in historical linguistics than Russians. Also that American academics tend to have more freedom to choose what they want to work on, while Europeans tend to do more long-term projects.

Just last month, at a phonetics conference in Ann Arbor, Eric Hamp of the University of Chicago made the observation that it is often dissimilar words which are actually cognates. He gave the examples: Russian dva. Tocharian A wu and Armenian (Y)erku. Having given these examples, he cautioned Nostraticists to beware of cognates that look too similar to one another.

and

Another linguist with whom I spoke stated that he doesn't believe monogenesis at all because, he has found, specifically in Southeast Asia, that languages converge more than they diverge. Convergence, he said, should be equally powerful.

 

[lpetrich]

Shevoroshkin (ed.) - Nostratic, Dene-Caucasian, Austric, and Amerind (1992)_text.pdf

Sergei Starostin broadly agreed with KR. Responding to Eric Hamp, "It is true that dissimilar words are often cognates, but similar words are even more often cognates."

Regarding comments made by the anonymous linguist, I can only say that I have also studied languages of Southeast Asia, and while he has found that languages converge more than they diverge, I have found that they diverge more than they converge, so we should perhaps compare our findings. Here I should mention that before we use the terms divergence and convergence we should try to understand what they actually mean.

Diverging then converging can happen, like in Western Europe and in the Balkans.

Eugene Helimsky:

As far as the critical remarks made by American linguists are concerned, I can certainly agree with Professor E. P. Hamp that it is often dissimilar words which are actually cognates. But it is no less often similar words which are cognates. In connection with the example he gives, it can be added that Russian dva, Latin duo, and English two are cognates despite their similarity. It s good advice to beware of cognates that look too similar, but does this advice have any more to do with Nostratic studies than with other areas of comparative and historical linguistics?

It's not very clear how one gets from PIE *dwô to Armenian erku, so that's not a very good example of a nontrivial sound correspondence. A more typical one is for "four" - Latin quattuor, Russian chetyre, Armenian ch'ors, Irish ceathair /kyahary/, Welsh pedwar, Greek tessares, Sanskrit catur /tSatur/. It relates /p/, /f/, /t/, /tS/, /ky/, /kw/.

"Professor C. Watkins actually does not refute the Nostratic theory. He just says that it presently has very little to tell him. This is a pity."

During the so-called stagnation years in the Soviet Union, linguistics, and especially historical and comparative linguistics, provided, so to say, a good shelter for those people who were interested in the humanities, but who did not want anything to do with the all-penetrating ideology. I think that is why the search for distant kinship, being absolutely impractical, attracted more attention on the part of Soviet scholars than on the part of Western scholars who were not forced to look for that kind of scholarly niche.

 

[lpetrich]

Shevoroshkin (ed.) - Nostratic, Dene-Caucasian, Austric, and Amerind (1992)_text.pdf

Raimo Anttila: discussed US generative-grammar research and a general disdain for historical linguistics.

Coupled with it all is the American promotion treadmill which favors short irrelevant projects that produce lots of entries in bibliographies at a steady yearly rate. It is now about impossible for a scholar outside the fashionable lines (where the long term research monies go) to engage in a long-range investigation that would take any length of time. In contrast, a European linguist may remain silent for ten years or so, and no eyebrows raised.

Then on what methods the Nostraticists use. The same as Indo-Europeanists, for instance.

Merritt Ruhlen:

I would not underestimate the lack of knowledge of Russian by American academics—publishing only in Russian is certainly a good way of keeping things quiet in the U.S.A.—but it is especially effective when what you are saying is something most people would rather not hear. And the one thing that Indo-Europeanists have not wanted to hear during the twentieth century was that I-E is related to this or that family.

I agree with the first part, but the second part seems like a conspiracy theory. Being turned off by failed attempts at long-distance comparison is a more plausible hypothesis.

Rowenchuk is obviously correct in that the so-called "Transformational Revolution," being as it was—and still is—an orgy of Eurocentric ethnocentrism, could only impede long-range comparison. For Chomskyites an outgroup is another Germanic language and long-range comparison includes Slavic and Romance languages. As far as I know the MIT school has still not discovered New Guinea.

Seems like the Chomskyites started out by studying only English.

Finally I would like to add a few words on monogenesis. Here two fundamentally different questions are often confused. The first is whether all currently extant languages derive from a common source; the second is whether human language has a single or multiple origins. These questions are not the same.

There is a third issue that is often mixed up with these. Is one able to learn anything about that ancestral language? 100,000 years seems like plenty of time to garble it beyond recognition.

 

[lpetrich]

Shevoroshkin (ed.) - Nostratic, Dene-Caucasian, Austric, and Amerind (1992)_text.pdf

Then some papers on Afroasiatic comparisons and some early papers on Nostratic.

Ilya Peiros: "The Austric Macrofamily: Some Considerations"

Information on the East Asian languages is highly limited. For only a fraction of these languages have dictionaries or phonologies been made, and even those are not always available in Moscow. For mainland Asia there exist only two full-scale reconstructions--those of the Chuang-Tai family by Li Fangkuei, and those of the Viet-Muong family by N. K. Sokolovskaya.

These are subfamilies of Kra-Dai and Austroasiatic. Since then, some work has been done:
 Proto-Tai language (Chuang/Zhuang-Tai) and  Proto-Kra language and  Proto-Kam–Sui language and  Proto-Hlai language
 Kam–Tai languages includes Zhuang-Tai and Kam-Sui
 Proto-Kra–Dai language

 Vietic languages (Viet-Muong) and  Proto-Munda language and  Pakanic languages and  Katuic languages and  Bahnaric languages and  Proto-Khmeric language and  Proto-Aslian language and  Nicobarese languages and  Pearic languages
 Proto-Austroasiatic language

 Proto-Hmongic language (Proto-Miao) and  Proto-Mienic language (Proto-Yao)
 Proto-Hmong–Mien language (Proto-Miao-Yao)

 Proto-Malayo-Polynesian language (I'll skip over its subgroupings' protolanguages) and  Atayalic languages
 Proto-Austronesian language

So we have plenty to work from.

 

[lpetrich]

Shevoroshkin (ed.) - Nostratic, Dene-Caucasian, Austric, and Amerind (1992)_text.pdf

More on Austric.

The lexico-statistical counts indicate that the Miao-Yao languages and those of the Austroasiatic family are related, but not directly affiliated. They appear to form a higher-order unit (a "Miao-Austroasiatic" family) that split into those branches around the fifth millennium B.C.

Thus, Austric is (Austro-Tai, Miao-Austroasiatic).

Then a section on Proto-Hokan roots (Hokan is in California and NW Mexico), and one on comparing Proto-Ge (South America) with Nostratic and Sino-Caucasian.

Then a section on "Hypotheses", like possible Australian words in Dravidian.

Vaclav Blazhek proposes

Our analysis shows remarkable cognates between the Australian basic lexicon and Dravidian equivalents characterized by more peripheral meanings. The nature of these cognates, then, suggests a relationship that is not directly genetic; the similarities can be explained as a consequence of the relationship of Pre-Dravidian substratum of Indie Peninsula and Australian languages. This conclusion is reinforced by a presence of similar elements in Munda (Thomsen 1892), Austronesian and Indo-Pacific (von Carolsfeld 1890), perhaps also in Mon-Khmer (Rivet 1927), and, of course, also by anthropological data.

Then, is Sumerian Nostratic or Dene-Caucasian? A paper on comparing Sumerian pronouns to Nostratic ones and finding a good match.

Finally, the "Mother Tongue" - Proto-World/Human/Sapiens - the most recent ancestor of *every* documented human language. As one might expect, the proposed PW roots are mostly body parts.

That concludes this series.

 

[lpetrich]

Vitaly Shevoroshkin and Paul Sidwell have collaborated on two more recent anthologies of macro-linguistics papers:

Shevoroshkin & Sidwell (eds.) - Historical Linguistics and Lexicostatistics (1999)_text.pdf

Sergei Starostin - COMPARATIVE-HISTORICAL LINGUISTICS AND LEXICOSTATISTICS

Using Morris Swadesh's 100-word list, what cognacy rate, what fraction of cognates?

• Dialects of some language: 90% or more
• Closely-related langs (Slavic, Romance, Germanic, Turkic): 70% - 80%
• More distantly-related langs (Indo-European): 25% - 30%
• Even more distantly-related langs (Uralic, Altaic): 10% - 20%
• Very distantl-related langs (macrofamilies like Nostratic): 5% - 9%

Slavic langs. cognacy rates: >75%
IE langs from different branches: cognacy rates 20% - 35%
Austronesian, Uralic, Sino-Tibetan similar.

IE, AN, Ur, ST are some of the oldest generally-recognized language families, with only Afroasiatic significantly older.

Rates of language change can vary by sizable amounts. For instance Icelandic and Norwegian (Riksmal) are both descended from Old Norse, but Norwegian has changed more than Icelandic: replacement rates 0.2 vs. 0.04.

"It is easier to explain the accelerated development of Riksmal. This language is actually a hybrid of Norwegian and Danish, and its one hundred word list includes eleven Danish, three Swedish and two German loans."

Removing borrowings gives change rates of 0.06 for Icelandic and 0.05 for Norwegian and Faroese, and 0.04 for Swedish, Danish, and two Norwegian dialects. Less variation.

 

[lpetrich]

Shevoroshkin & Sidwell (eds.) - Historical Linguistics and Lexicostatistics (1999)_text.pdf

Harald Sverdrup and Ramon Guardans: COMPILING WORDS FROM EXTINCT NON-INDOEUROPEAN LANGUAGES IN EUROPE

Which they call Paleoeuropean for convenience. These langs vary widely in inscriptions known of them and success in interpreting them.

Northwestern Paleoeuropean: Basque (w/ Aquitanian; NE Spain, SW France), Iberian (E Spain), Pictish (Scotland), possibly also Itturian (Netherlands), Ligurian (NW Italy), Nuragic (Sardinia)

Central Paleoeuropean: Etruscan (C Italy), Rhaetian (E Switzerland, W Austria, NE Italy), Lemnian (Aegean Sea), possibly also Pelasgian (mainland Greece), Elymnian (W Sicily), N PIcene (NE Italy), Camunian (C Switzerland), Lepontic (W Switzerland)

(?) Tartessian (SW Spain, S Portugal, nearby N Africa) -- a southern Paleoeuropean branch?

"A comparison of the extinct Paleoeuropean languages show many similarities in the non-cultural lexicon. Some of the languages show similarities in numerals and pronomina." -- 1 to 10 and 100, 1SG (ni/nai, mi), 2SG (gu/ki/khu), ...

A good indicator of recognizable shared ancestry.

The authors lined up words for "house, hut", "land, world, "beautiful, lick, mouth", "city, tribe", "border, limb", "spring, drinking water", "high", "stone", "grave, hell", "hand, power", "god, sky", "head, leader", "man, chieftain", "hollow, cave", "rich, plenty", "man, male", "me, myself", "I, self", "house, at home", "son, male", "community, many", "large", "Moon, month", 'water, stream", "river, water", "hand, possession", "brother, land, soil", "4"

Only some of the meanings are cross-language high-stability ones, and the authors made some odd lumpings.

But this and Greek and Germanic substrate vocabulary are evidence that the first European farmers spoke Dene-Caucasian languages, or at least Euskaro-Caucasian ones.

 

[lpetrich]

Shevoroshkin & Sidwell (eds.) - Historical Linguistics and Lexicostatistics (1999)_text.pdf

Ilia Peiros, FAMILY EVOLUTION, LANGUAGE HISTORY AND GENETIC CLASSIFICATION

He proposes a distinction between plain language and "sociolanguage":

• One language, two or more sociolanguages: Serbian, Croatian, Bosnian
• Two or more languages, one sociolanguage: Chinese "dialects"

After a long discussion of finding language family trees which I failed to follow, he gets into core vocabulary. What meanings and word forms? Universal meanings, simple word forms (no compounds or phrases). The Swadesh list is generally good, but it has problems, like lack of word for "horn" for people who do not live among horned animals, like in Australia before the introduction of European farm animals. Also, lack of independence of meaning, like "bark (of tree)" and "skin". Both of them are surface layers.

One must use the most commonly-used and stylistically-neutral word with some meaning, like English "kill" instead of "slay" or 2PL "you" instead of "youse" or "you all". One must also use typical present-day words, like English 2SG "you" instead of "thou".