Language as a Clue to Prehistory

[lpetrich]

Shevoroshkin & Sidwell (eds.) - Languages and Their Speakers in Ancient Eurasia (2002)_text.pdf

In honor of Aharon Dolgopolsky on his 70th birthday. He was born in the Soviet Union, as Aron Borisovich D, but he emigrated to Israel in 1976, and used the Hebrew version of his first name. The English version of it is Aaron.

He worked on Nostratics, along with Vladimir Illich-Svitych, and he helped Vladimir Dybo publish VIS's work after VIS was killed in a car accident in 1966.

Paul Sidwell: "I myself have more recently taken a more cautious attitude towards the long-range endeavor in linguistics, but at the same time I am more convinced than ever that we are all the more richer for its existence, and I welcome and encourage interest in the field."

I agree. That's what I like about Martine Robbeets's work on "Transeurasian" (Broad Altaic). That work ought to be extended to other macro-linguistic hypotheses.

PS: "I am confident that Nostratic studies will be with us, pursued by a small international fraternity, for many decades after fads such as Transformational/Generative Grammar (and its variants and successors) have become no more than unpleasant memories pricked by the selfserving historiographies that seem to fill many library shelves."

Is transformational/generative grammar really that horrible? I agree that the problem of parsing natural language is a very difficult one -- beyond the basics of some language, it get very difficult very quickly.

Consider this rule that English speakers take for granted but seldom explicitly state: Order force: the old grammar rule we all obey without realising | Tim Dowling | The Guardian - "The rule is that multiple adjectives are always ranked accordingly: opinion, size, age, shape, colour, origin, material, purpose." That ordering is also used by speakers of other languages with adjectives before nouns, while for adjectives after nouns, the order is in reverse. Adjective Ordering Across Languages | Annual Review of Linguistics

PS: "Blanket condemnations that paleolinguistic reconstruction is marred by false cognates, undetected borrowings and faulty citations, are unfair—these are normal problems for all comparative linguistics."

In fairness to such critics, if the signal of shared ancestry is weak, then it can easily be overwhelmed by these sources of noise.

 

[lpetrich]

Shevoroshkin & Sidwell (eds.) - Languages and Their Speakers in Ancient Eurasia (2002)_text.pdf

Nostratic Stops Revisited, by Sergei Starostin

He left out Afroasiatic, because (1) he didn't think that there is a good enough reconstruction of it and (2) even if there was, then it would not be a subgroup of Nostratic but a separate macrofamily.

He focused on initial stops, because they are relatively simple to compare. He only discussed dental (t) and velar (k) stops, though he noted that labial (p) and non-initial stops have similar patterns.

Proto-Uralic and Proto-Dravidian have only one voicing of initial stops, so they aren't very much help. So he looked at PIE, Proto-Kartvelian, and Proto-Altaic.

• IE (classic): T, D, Dh
• Kartvelian: T', T, D
• Altaic: Th, T, D
• Nostratic (classic): T', T, D

T' is glottalized or ejective, with a short pause before the consonant and the vowel.

The glottalic theory of PIE stops, by Gamkrelidze, Ivanov, and Hopper, revises the stops into T(h), T', D(h). Allan Bomhard proposed that the first two voicings of Kartvelian stops are flipped: T, T', D. But Vitaly Shevoroshkin in "Typology, Relationship, and Time" rebutted that proposal as "unsupported by evidence and based solely on abstract phonological speculations." He proposed a new one, also mentioned in TRT, and in another work:

Griffen, Toby D., 1988: "Germano-European: Breaking the Sound Law" S. Illinois University Press, Carbondale & Edwardsville

• IE (classic): T, D, Dh
• IE (glottalic): T(h), T', D(h)
• IE (new): Th, T, D(h)
• Kartvelian: T', T, D
• Altaic: Th, T, D
• Nostratic (classic): T', T, D
• Nostratic (new): Th, T, D

Where he proposes Th -> T' for Kartvelian.

This makes PIE stops much like those in the Thai language, with Th, T, and D, and also those in Proto-Altaic and Proto-Nostratic.

Both the glottalic theory and the Shevoroshkin-Griffen-Starostin theory also explain why initial b is so rare in PIE reconstructions. If it was instead a p or p', that would be much more understandable, since those sounds tend to drop out.

SS notes an additional set of correspondences: IE D, Kartvelian D, Altaic Th, with -w/o/u- often present. He proposed Nostratic Dw. Thus,

• IE: Th, T, D(h), T(w)
• Kartvelian: T', T, D, D(w)
• Altaic: Th, T, D, Th(w)
• Nostratic: Th, T, D, Dw

 

[lpetrich]

Shevoroshkin & Sidwell (eds.) - Languages and Their Speakers in Ancient Eurasia (2002)_text.pdf

Then a paper on whether the IE suffix -ter comes from Uralic. It shows up in *ph2ter- "father", *meh2ter- "mother", *bhreh2ter- "brother", and *dhugh3ter- "daughter". Though *swesor- "sister" half-fits, and *suHnus "son" doesn't.

He proposes something like atta + -r where *atta is an alternate word for father in PIE, the likely origin of Proto-Slavic *otitsi, and others. This word may be cognate with Uralic *atta, Turkic *ata, Eskimo *ataata, at least if it isn't some baby-talk sort of word.

Words like (m)ama, (n)ana, (y)aya, (p)apa, (t)ata, (k)aka.

The book then has papers on Pictish, Rhaetian, and Tartessian. There is also a longish paper comparing Hurro-Urartian to North Caucasian and Indo-European. Unfortunately, it does not have a clearly-stated conclusion. Then a paper on Sumerian words for sheep, goats, and antelopes. Here also, no conclusion.

 

[Swammerdami]

... Consider this rule that English speakers take for granted but seldom explicitly state: Order force: the old grammar rule we all obey without realising | Tim Dowling | The Guardian - "The rule is that multiple adjectives are always ranked accordingly: opinion, size, age, shape, colour, origin, material, purpose." That ordering is also used by speakers of other languages with adjectives before nouns, while for adjectives after nouns, the order is in reverse. Adjective Ordering Across Languages | Annual Review of Linguistics

This adjective-ordering is interesting. This is the first time I heard that not only do other languages have a standard order, but it's the same as English's order. Or, perhaps even more interesting, the exact(?) opposite order in adjective-after languages.

You show eight adjective classes, but I've seen up to eleven listed! The order can be memorized by memorizing a 12-word phrase. Shall we stage a contest to see who can come up with the best such phrase? Here are two such, using "A/An" for the quantity adjective.

An exquisite well-made miniature antique triangular ebony-colored French ceramic flat heating iron.

An impressive tall beautiful buxom young blond Swedish silky-skinned on-call working girl.

 

[lpetrich]

False cognates, undetected borrowings and faulty citations? I have a bit more to say.

False cognates? Here are some from Indo-European: English "to have" and Latin habére (same meaning). Latin deus and Greek theos "god".

English "to have" < Proto-Germanic *habjanan < PIE *keh2p- (*kap-) > Latin capere
Latin habêre < PIE *gheh1bh- (*ghabh-) "to grab, take". A similar root is *ghebh- "to give, take" > PGmc *gebanan > English 'to give"

Latin deus < PIE *deywós < PIE *dyew- "to be bright, to shine, sky, heaven"
Greek theós < PIE *dheh1s < PIE *dheh1- "to do" (make offerings?)

Borrowings? As to "cat", there are lots of similar-looking forms in many European languages: Dutch, Danish kat, German Katze, Norwegian, Swedish katt, Icelandic köttur, French chat /Sa/, Catalan gat, Spanish, Portuguese gato, Italian gatto, Late Latin cattus, Basque katu, Welsh cath /kaT/, Irish cat, Lithuanian kate, Latvian kakis, Bulgarian kotka, Czech kočka /kotSka/, Polish, Russian, Belarusian kot, Ukrainian kit, Finnish kissa, Estonian kass, Greek gata, Turkish kedi, ...

So one concludes that this word is a Wanderwort, a wander word, one that spread with what it names. From the sound shifts, it likely spread in late antiquity or the early Middle Ages, before many of these languages were written down. The Slavic ones are from Proto-Slavic *kotu, with a > o, consistent with Slavic speakers dispersing in the early Middle Ages. This time estimate also explains /a/ > /ae/ in the English form, /k/ > /tS/ > /S/ in the French form, /t/ > /ts/ in the German form, and various /k/ > /g/. It is also after PIE *k > Germanic h.

Looking to antiquity, the usual Latin word for cat was feles / felis, and the ancient Greek word was ailouros.

Also notable is how the Basque, Finnish, Estonian, and Turkish forms look much like the others. None of these languages are Indo-European, and that adds to the conclusion that the word "cat" is a wander word.


But one arrives at these conclusions with a lot of historical-linguistics investigation, and before one does so, one has a risk of a lot of confusion.

 

[Swammerdami]

On the topic of false cognates . . .

In his earlyish work, the late Merritt Ruhlen claimed *tik/*dek ('finger') to be one of the most ancient of words. Because of the way people count with their fingers, the word for finger ('digit') becomes 'one', 'five' and/or 'ten', as well as 'hand'. It also leads to 'indicate' (to point out with a finger). He found cognates of this word in a number of the macrofamilies; it was one of the two or three best attested of very ancient words.^*

But "standard" etymologies separate PIE *deik ('to show') and PIE *dekm ('ten').

Might they BOTH be right, with *deik and *dekm evolving from an earlier *tik ('finger')?

* - I looked for my copy of Ruhlen to display some of the alleged cognates of *tik around the world that he shows, but conclude that my book is out on long-term loan! But this page shows many such alleged cognates, especially if *tak ('foot' or 'leg') is lumped in.

 

[lpetrich]

Here's where to look. John D Bengtson is Global Etymologies by John D. Bengtson and Merritt Ruhlen. It has "tik" in it.

Tower of Babel: Databases - does Nostratic, Dene-Caucasian, Borean

Wiktionary, the free dictionary - good on the better-studied families, like Indo-European

From Wiktionary, PIE *deyk'- "to point out, to show, to speak solemnly" is securely reconstructed:

• Greek deiknunai "to show, point out", dikê "custom, law", deigma "specimen, pattern", ...
• Latin dîcere "to say", index "pointer, indicator", jûdex "judge", vindex "protector, defender", digitus "finger, toe" (from *deyg'-) ...
• Germanic *taihwôn "toe", *taiknan "token", *takijanan "to show" > E "to teach"
• Indo-Iranian

Wiktionary says that *dekm "ten" has uncertain origin.

The ToB has *deik'e-, *deig'e-, *dok- (found with meaning "show")

I checked under its Nostratic etymology and I found

• Altaic tûjku "to make a sign"
  • Turkic *Tûkrag (?) "royal sign manual"
  • Mongolic *doki- "to make a sign"
  • Tungusic *duKû- "to write"
  • Korean *tjek- "to note down, to write"
  • Japonic *tunka- "to let know, inform"
• Uralic *täkkV "to look, observe"
• Kartvelian *t'qv- "to recognize, notice"
• Eskimo-Aleut *takuv- "to check out, to visit, to see"

then

• Eurasiatic *tVjk'V "to show, point at"
• Sino-Caucasian: Sino-Tibetan *tuak/d "to consider, inspect"
• Afroasiatic *dag- (Semitic, Egyptian, Cushitic) "to look, stare")

then

• Borean *TVKV "to look, show"

That database doesn't have "ten", however.

 

[lpetrich]

Long Ranger Archive – Mother Tongue
Mother Tongue Archive – Mother Tongue

I've had to collect a list of tables of contents for the Long Ranger archive as well as the Mother Tongue one. BTW, Long Ranger was known as Mother Tongue for a long time, but its name was changed to that to avoid confusion when the Mother Tongue journal was founded.

The first one has someone addressing Aron/Aharon Dolgopolsky as Aaron. Then some long-range etymologies, like *sVn for "nose". Looking in Wiktionary, I find that Indo-European has *neh2s- (*nâs-), Uralic *nena, Turkic *burun, Mongolic *kamar, Tungusic *xonge (x = kh fricative), Korean ko, Japonic *pana, Eskimo *qanga-, Ainu etu, Georgian-Zan *c1xwir-, Dravidian *mûkku-, NE Caucasian *mara (?), NW Caucasian *pwa, Basque sudur, Sino-Tibetan *sna, Athabaskan *-nachixy, Nahuatl yakatl, Quechua sinqa, Austronesian *ujung, Proto-Thai *dang, Mon-Khmer *muus, Santali mu (Austroasiatic *muus), Hmong-Mien *mbruiH, Semitic */anp-, Egyptian fenedj, Cushitic */isngw-, Wolof bakkan, Mandinka nungo, Igboid *emi, Yoruboid *ingmu, Bantu *mphuna (?) -- I lost patience, though I'm half-thinking of writing a program that gives the family membership for each language named in a translation page.

Nice discovery in Wiktionary: these Northeast Caucasian langs: Hunzib (Tsezic) maru, Chechen (Nakh) mara, Ingush (Nakh) merazh, Andi (Avar-Andic) mahar, Avar (Avar-Andic) me'er, Budukh (Lezgic) me'el, Lak mai

Hard to find much of a pattern.

 

[lpetrich]

I used nose/translations - Wiktionary, the free dictionary Fortunately, "nose" is also in Appendix:Swadesh lists - Wiktionary, the free dictionary

Appendix: Paleosiberian Swadesh lists - Wiktionary, the free dictionary - Ainu etu, Nivkh mix, vix, ux, Yukaghir yoghul, Chukchi-Kamchatkan: Koryak /inget/em, Itelmen qeqeng

Elamite shim-, Sumerian kiri, Burushaski mosh, Siouan *pas- (?), Hokan *xu (?), Uto-Aztecan *yaka (?), Oto-Manguean *xin- (?), Tupi tî, Pama-Nyungan *mulya (?), Bantu *ijulu

I'd have to do this with words for "to smell (something)" to get a better picture. When I tried, I often ended up with "to be smelly" and "to smell like".

I also found the oddity that the Hungarian word for this body part, orr, has cognates in other Uralic languages meaning "mountain", like Finnish vuori, pointing to "mountain" > "mountain peak" > "nose".

 

[lpetrich]

In the fourth issue, back in 1987, is a quiz asking how computer-savvy its subscribers are. In it is:

I own a modem, I don't own one, I have access to a modem, what is a modem?, I know almost nothing about computers.

Reminds me of the 1990's, when I went online with a modem from my home, connecting to an analog-phone landline. Those days are long gone.

Long Ranger (Mother Tongue) issue 10 has a review by Allan Bomhard of "Typology, Relationship, and Time".

He's rather annoyed by Vitaly Shevoroshkin's attacks on his proposed Nostratic stop-consonant correspondences, and I don't want to get into PIE laryngeals - that seems like an issue that may require convincing outside correspondences to resolve.

He then discusses V. V. Ivanov: "Proto-Languages as Objects of Scientific Description." (1980).

It is quite clear from the thrust of his argumentation that Ivanov belongs to the school of Linguistics that views reconstructed languages as real languages that existed at a particular point in time and not as a mere set of correspondences. This is a position that I would wholeheartedly endorse

Some linguists have gone so far as to compose text in reconstructed protolanguages, like August Schleicher and Vladislav Illich-Svitych.

VVI had a section on "The Distinction between Proto-Languages and Intermediate States of Dialectal Evolution: The Problem of Minimizing the Number of Proto-Languages"

For N langs, a binary tree would have (N-1) nodes, while a lawn would have only 1 node. VVI wants to avoid positing too much branching, though positing too little branching is also a problem.

 

[lpetrich]

Issue 11 has a review of Colin Renfew's "Archaeology and Language" by Roger Wescott.

Colin Renfrew proposes that Indo-European was spread by the Middle Eastern Neolithic farmers into Europe and South Asia, instead of by later horse-riding animal herders from the western part of the Eurasian steppes.

RW: "So I find it disappointing to have to report that Renfrew's linguistics is, at best, simplistic and, at worst, ill informed."

Renfrew describes his approach to cultural and linguistic prehistory as "processual" rather than "migrationist." In his first four chapters, he derides the notion of an Urheimat, or primal homeland, of the Indo-Europeans. Yet, in his seventh chapter, he reverts to the concept (hedging by putting the word 'homeland' in half-quotes) and locates it in eastern Anatolia about 7000 B.C.

JP Mallory states that in "In Search of the Indo-Europeans" that a little over a century ago, prehistorians would indiscriminately propose migrations, while a little over half a century ago, opinion shifted to proposing that people stayed in place, learning and acquiring things from their neighbors.

Toward the end of his first chapter, the author expresses his regret that Indo-Europeanists and other linguistic paleontologists "make little use of verbs and adjectives."

Then claiming that some languages have no adjectives. Referring to langs with verb-like adjectives like Japanese. In PIE, however, adjectives were noun-like.

RW: "Perhaps the oddest aspect of this opus is its neglect of linguistic reconstruction." Like not having any reconstructed PIE forms anywhere in it. Not only of basic vocabulary, but also of cultural items, like the name of a god that they worshipped: *dyeus pHter, "Father Sky".

Even the title of this book is misleading. It is not a general treatment of the relation of archaeology to language or even a general treatment of archaeology alone. A fairer title would have been One Archaeological View o f the Puzzle o f Indo-European Origins. Throughout the book, Renfrew represents his view as being in opposition to "traditional" assumptions about Indo-European origins and as constituting a radical rethinking of the subject. For the most part, however, his thinking seems to me to be cautious and pedestrian to a fault. I can only hope that his next book will be as intellectually adventurous as he had apparently intended this one to be but that, if he does stray from his customary archaeological turf, he will familiarize himself more thoroughly with the non-archaeological subject matter involved than he has in Archaeology and Language.

 

[Politesse]

RW: "Perhaps the oddest aspect of this opus is its neglect of linguistic reconstruction." Like not having any reconstructed PIE forms anywhere in it. Not only of basic vocabulary, but also of cultural items, like the name of a god that they worshipped: *dyeus pHter, "Father Sky".

It's almost as though Westcott has no idea what "processual" means....

 

[lpetrich]

RW: "Perhaps the oddest aspect of this opus is its neglect of linguistic reconstruction." Like not having any reconstructed PIE forms anywhere in it. Not only of basic vocabulary, but also of cultural items, like the name of a god that they worshipped: *dyeus pHter, "Father Sky".

It's almost as though Westcott has no idea what "processual" means....

What does it mean?

 

[Politesse]

RW: "Perhaps the oddest aspect of this opus is its neglect of linguistic reconstruction." Like not having any reconstructed PIE forms anywhere in it. Not only of basic vocabulary, but also of cultural items, like the name of a god that they worshipped: *dyeus pHter, "Father Sky".

It's almost as though Westcott has no idea what "processual" means....

What does it mean?

One of the major methodological schools within archeology (sic), of which Lord Renfrew was one of the first and (mostly because of his considerable influence as an undergraduate textbook author) notorious advocates. I'm not saying they are necessarily right about things, but you would not expect someone to include a speculative language reconstruction as evidence after declaring something a processual study. Everything has to tie back to a concrete piece of empirical evidence one way or another, in this approach, and Lew Binford (who popularized the method) was especially averse to anything with that certain whiff of European idealism. Light in the East, the Three Ages of Man, etc, part of the whole purpose of the school was to eat away at the European cultural bias that had, in their view, corroded the accuracy of archeological field methods.

This was all some thirty odd years ago, I don't think there are a lot of doctrinaire New Archaeologists in the field at present.

 

[lpetrich]

Long Ranger / Mother Tongue #13:

John Bengtson:

So while there are undoubtedly lexical parallels between Basque and Afroasiatic, few of them involve the most basic, non-cultural vocabulary, and these can be explained as residue of a long-range relationship between Dene-Caucasian and Afroasiatic. When Basque, Caucasian, and Burushaski are compared, basic isoglosses become much more numerous. Furthermore, the Basque noun case endings and fossilized class prefixes have counterparts only in Caucasian and Burushaski, and the phonological systems of these three language families are interrelated to a degree that clearly distinguishes them from Afroasiatic.

...
Apart from the archaic residue (from the common ancestor of Dene-Caucasian and Afroasiatic), mentioned above, many of the parallels between Basque and Afroasiatic may be ascribed to contact of early Basque with a known ·or unknown Berber dialect, possibly in southern Iberia. This is indicated by the fact that some of the most exact parallels are "too similar" and therefore probably cultural loans, ...

Looks like one needs to do a test with some highly conserved words, like the Swadesh List or similar lists.

 

[lpetrich]

LR/MT #14:

Eric de Gralier

There is no need here for detailed comments on the graphs. One very clear conclusion emerges from comparing the two I.-S. graphs with the Bomhard one: Bomhard's "Nostratic" is not the same entity as Illich-Svitych's "Nostratic". Bomhard's statistics is heavily "loaded" by the origin of the author's endeavour: a comparison between Indo-European and Semitic. By contrast, I.-S. appears more "balanced" - but it leaves the sceptically oriented "external" (i.e. not "committed" to one "school" or the other) observer with some question-marks. It is, for instance, rather surprising that the number of "cognates" between Indo-European and Afrasian is superior to that between I.-E. and Uralian.

Without agreeing to all the objections presented by Murtonen to I.-S. "Afrasian" material (which, in fact, are more directly pertinent concerning his Semitic data). one must admit that a good number of I.-S. Afrasian "cognates" with the three families pertaining to Greenberg's Eurasiatic (I.-E., Uralian and Altaic) are at least debatable.

Trying to compare Indo-European and Semitic has been done for a long time, and not a lot has come out of it, like "6", "7", "bull" and "horn". At least for Indo-European and Uralic one has 1SG, 2SG, "water", "name", and "to hear", which are less likely to be wander words.

Here are two reconstructions of Afrasian:

• Hamito-Semitic Etymological Dictionary – Materials for a Reconstruction | Brill by Vladimir E. Orel and Olga V. Stolbova
• Reconstructing Proto-Afroasiatic (Proto-Afrasian) by Christopher Ehret - Paperback - University of California Press

Here are three of Nostratic/Eurasiatic:

• Nostratic dictionary - Google Books by Aharon Dolgopolsky
• The Nostratic Macrofamily - Google Books by Allan R. Bomhard, John C. Kerns
• Indo-European and Its Closest Relatives: The Eurasiatic Language Family, Volume 1, Grammar and Indo-European and Its Closest Relatives: The Eurasiatic Language Family, Volume 2, Lexicon by Joseph Greenberg

Issue 14 later mentioned "Borean" Nostratic/Eurasiatic, Afrasian, Dene-Caucasian, Amerind, Austric, complete with some hypotheses on their branching.

 

[lpetrich]

Issue 17 in August 1992 starts with an article about Mark Kaiser's experiences in post-Soviet Russia.

Over the course of one month I was involved in one pick-pocketing, one extortion, and the disappearance and rifling of one suitcase. Foreigners are frequently the targets of crime: the contents of one suitcase could easily bring the equivalent of six months salary.

Even more discouraging is the impression of general deterioration: new buildings that appear 50 years old, public telephones forever out of order, airline flights cancelled due to a shortage of fuel. The public sector is crumbling, and a private sector has not yet been created to take up the slack. The good news is that there are more consumer items, including food, on the shelves than one year ago. But, ...

Since price controls were removed in January, Russia has experienced hyperinflation. Prices for most goods and services have increased approximately 1000%. Some examples: a plane ticket from Siberia to Moscow was b4 roubles, now is 1000 r.; a pack of Marlboros was 20 r., now is 175 r; a loaf of bread was .40 r., now is 5 r.; a kilo of tomatos was 15 r., now is 130; etc. At the same time salaries have risen only 400% (average monthly salary is approximately 2,500 r.). Those professions that rely on direct government funding, i.e., doctors, teachers, scholars, have seen their salaries increase at a much slower rate than the national average.

To understand the burden placed on Russian families, imagine two pounds of tomatos costing S104, two pounds of sausage for S140, a transatlantic plane ticket for S72,000., or a pair of shoes for S2,400. In terms of percentage of monthly salary, that is what Russians are paying.

Then soliciting donations for their Russian colleagues. Also noting Yeltsin's Team Seems in Retreat As Its Economic Reform Falters - The New York Times - By Serge Schmemann - Aug. 2, 1992

 

[lpetrich]

Why there was no Long Ranger issue 18: From 19, "EDITOR'S NOTE: MT-19 is now too big, especially since we have crammed just about all the old MT-18 material into it."

In that issue and a previous one, it was mentioned that linguist Alexandra Aikhenvald proposed the name "Noscau" for Borean: Nostratic-Sino-Caucasian-Austric.

AA's last name here is a Roman-alphabet transcription of a Cyrillic-alphabet transcription of the German name Eichenwald "oak forest".

German Eiche - cognate with English "oak", both from Proto-West-Germanic *aik, from Proto-Germanic *aiks, with origin obscure. Latin aesculus "k. of oak" and Lithuanian ą́žuolas "oak" may be related. But all the Germanic-language words for this kind of tree are cognates.

The Latin word is quercus, and for kinds of oaks, rôbur, aesculus. A weird thing about that word is that what happened in Latin's descendants, something that I read about in JP Mallory's book "In Search of the Indo-Europeans". Italian quercia < Latin quercea "made of oak (f.)". Spanish roble, Catalan roure < rôbur, Portuguese carvalho < ?, French chêne <? Gaulish cassanos, Romanian stejar < ?.

BTW, the oak's genus name, Quercus, was bestowed by Carolus Linnaeus himself.

That word's English cognate is "fir", a kind of pinelike tree. It's from PGmc *furhô, meaning a pinelike tree. But such trees are shaped very differently.

Both Latin and Germanic are from *perkw- "oak" or "to strike" -- that root appears in the names of Baltic god Perkūnas and Slavic god Perun, both associated with the sky, storms, lightning, thunder, rain, and oak trees. Greek Zeus and Roman Jupiter were also associated with these, including oak trees. The PIE speakers worshipped a god like that - Perkwunos - alongside Dyeus Pater (simplified spelling; "Father Sky").

Oak trees sacred to a god of lightning and thunder? Because these trees get struck by lightning if they grow big enough. That may explain the connection with pinelike trees. If they grow big enough. I confess that I'm totally sunk in a mechanistic view of the Universe, and my first thought is: what physical feature of oak trees would make them struck by lightning? I quickly found out: their size. If they grow big enough, they attract lightning, and the same is true of *any* other tree, and also of artificial structures.

That aside, Irish dair and Welsh dâr "oak" are from Proto-Celtic *daru < PIE *doru "tree". Greek drus "tree, oak" also < *doru. Also, Russian dub, Polish dab, Czech dub are from Proto-Slavic *dobu "oak. Wiktionary mentions as a source PIE *dheubh- "deep", but that seems semantically implausible.

It reminds me of  Johannes Goropius Becanus and his attempts to derive other languages from his native one, the Brabantic dialect of Dutch. Like Latin quercus "oak" from werd-cou "protect form cold" (English cognate: ward-cold).

 

[lpetrich]

Issue 19 ends with an editorial that is mostly grumbling about how closed-minded many other historical linguists are. Like their claiming that beyond the time depth of Proto-Indo-European, historical linguistics cannot recover relationships. In fairness to them, it becomes more and more difficult with greater and greater time depth, but it is not impossible.

In Issue 20, Allan Bomhard states

One last point needs to be made: Reconstructed languages should be thought of as real languages in every sense of the term. This means that we should be very careful not to reconstruct anything that is not characteristic of language in general: our goal should be to strive for reality in our reconstructions, and we should not hesitate to use every means at our disposal to help us arrive at realistic reconstructions.

The ultimate test is one's ability to compose text in a reconstruction, as August Schleicher had done: "The Sheep and the Horses".

Then Harold Fleming reviews Ruhlen - A Guide to the World's Languages, Vol. 1 - Classification (1991) : Allan R. Bomhard : Free Download, Borrow, and Streaming : Internet Archive - an assessment of MR's classifications in gory detail. "Rarely do I disagree flatly with Ruhlen about major matters, but there are some differences of opinion." and "Much time is spent fighting methodological battles in Ruhlen's book. It is a necessary chore for anyone in the social sciences and especially in historical linguistics."

Also has Stefan Liedtke's review of Joseph Greenberg's "Language in the Americas". SL finds JG's book to be full of errors.

Greenberg's "Amerind Etymological Dictionary" (181-270), indicated on the inside front flap to be, along with the grammatical section, the basic part of the book, unfortunately builds, as have all its predecessors, to a great extent an unmotivated hodge-podge of combinations which are phonologically and semantically too loose, and which cannot be gone into here in detail. Just one example: comparisons like those listed under 'bite' (192-193) (Guamaca kaka 'tooth,' Kiowa k'o 'knife,' Yanoama koa 'drink,' Ticuna chi 'sting') are numerous and obviously lead nowhere.

That's Goropius territory.

The situation is entirely different in Greenberg's "Grammatical Evidence for Amerind" (271-320). This part of the work is brilliant and of incomparable value for his argument.

 

[lpetrich]

Long Ranger 21 has an article on how some linguists modified the 100-word Swadesh list for some Australian languages. In one of them, "seed" and "eye" are the same word (being small and round?), and "mountain" and "stone" also the same word (a mountain is a big rock). So they deleted some and retained some similar ones.

D: all, full, R: many. D: bark (of tree), R: skin (bark is the skin of a tree), D: cloud, rain, R: water (rain is water from clouds), D: drink, R: eat (drinking is eating liquids), D: feather, R: hair (of head) (feathers are bird "hair"), D: man, R: person (aboriginal), D: mountain, R: stone (a mountain is a big rock), D: night, R: black (everything looks black at night), D: round, seed, R: eye (small and round), D: sleep, R: lie (recline) (when one is sleeping, one is lying down).

Some of the words were derived from other words.

D: bird (often "winged one", "feathered one"), D: to come, to fly, to swim, R: to go/walk (ways of moving), D: to die, R: to fall, D: green (often "leaf-leaf"), D: red (often "blood-blood"), D: yellow (often "yellow-ocher-yellow-ocher"), D: white (sometimes "ash-ash"), D: kill, R: hit (with hand), D: know, R: ear, hear, D: new, R: now, today

Then they added a lot of meanings, like "to climb", "rotten (of meat)", "sky", "throat".