Alexander Kim sent me a link to “Massive migration from the steppe is a source for Indo-European languages in Europe” by Wolfgang Haak, Iosif Lazaridis, Nick Patterson, et many al.; here’s the abstract:
We generated genome-wide data from 69 Europeans who lived between 8,000-3,000 years ago by enriching ancient DNA libraries for a target set of almost four hundred thousand polymorphisms. Enrichment of these positions decreases the sequencing required for genome-wide ancient DNA analysis by a median of around 250-fold, allowing us to study an order of magnitude more individuals than previous studies and to obtain new insights about the past. We show that the populations of western and far eastern Europe followed opposite trajectories between 8,000-5,000 years ago. At the beginning of the Neolithic period in Europe, ~8,000-7,000 years ago, closely related groups of early farmers appeared in Germany, Hungary, and Spain, different from indigenous hunter-gatherers, whereas Russia was inhabited by a distinctive population of hunter-gatherers with high affinity to a ~24,000 year old Siberian6. By ~6,000-5,000 years ago, a resurgence of hunter-gatherer ancestry had occurred throughout much of Europe, but in Russia, the Yamnaya steppe herders of this time were descended not only from the preceding eastern European hunter-gatherers, but from a population of Near Eastern ancestry. Western and Eastern Europe came into contact ~4,500 years ago, as the Late Neolithic Corded Ware people from Germany traced ~3/4 of their ancestry to the Yamnaya, documenting a massive migration into the heartland of Europe from its eastern periphery. This steppe ancestry persisted in all sampled central Europeans until at least ~3,000 years ago, and is ubiquitous in present-day Europeans. These results provide support for the theory of a steppe origin of at least some of the Indo-European languages of Europe.
Longtime readers will know that I am deeply skeptical of attempts to mix linguistics with genetics, perhaps more than I should be. I don’t have the brains to actually read the paper right now, so I’ll just sit back and see what my readers have to say.
Addendum. Brice Russ, Director of Communications for the Linguistic Society of America, wrote to let me know that “the LSA very recently released a paper from our upcoming issue of Language, where four UC-Berkeley linguists use some new methodologies to examine the same issue.” He sent me this link:
“Ancestry-constrained phylogenetic analysis supports the Indo-European steppe hypothesis”, by Will Chang, Chundra Cathcart, David Hall and Andrew Garrett, will appear in the March issue of the academic journal Language. A pre-print version of the article is available on the LSA website.
I can’t wait till I’m well enough to look at it!
Alexander Kim’s own post on this http://sarkoboros.com/2015/02/10/today-on-biorxiv has if you scroll down some multi-colored charts showing different modern European populations and the proportions of 3 different hypothesized origin populations in their genes. So you can see that even if IE comes from Yamnaya, the currently non-IE-speaking populations on the list (Estonian, Hungarian, Basque) occur at different points along the continuum and are not clustered tightly together, much less clustered together toward the least-Yamnaya-percentage end of the continuum. The champion there are the Sardinians, who are plausibly largely descended from whoever was living in Europe before IE speakers arrived but nonetheless speak an IE language and have done so since, well, I don’t even know if there’s a theory about what was spoken on Sardinia before some version of Vulgar Latin was. But those “counterexamples” don’t necessarily undermine the point – language is not transmitted genetically as such, but in order to figure out why particular languages ended up being spoken where they did you have to know how different population groups moved around in time and space, because the current L1 speakers of IE languages who do not descend genetically in large part from PIE speakers do descend from ancestors who came into some relevant sort of contact with IE speakers (maybe one generation ago, maybe centuries or millenia ago) that resulted in language shift.
A detailed exposé and many comments on Dienekes’ Anthro blog. To me, it’s like an anthropologist listening to linguists talking.
I don’t even know if there’s a theory about what was spoken on Sardinia before some version of Vulgar Latin was.
Price’s Encyclopedia of the Languages of Europe says “it seems likely that we are dealing with more than one language, one of which had links with Iberian while another was in some way connected with North African languages”; you can get more at Wikipedia. Thanks for your very helpful comment!
And of course thanks to Y for the link to Dienekes’ Anthro blog; I don’t have the capacity at the moment to do more than gaze at the pretty colors on the maps, but I’m glad to know it’s there.
A new paper by Chang et al., Ancestry-constrained phylogenetic analysis supports the Indo-European steppe hypothesis has just been posted. It has a new, improved phylogenetic analysis of the IE languages, whic concludes that PIE is not as old as Gray and Atkinson had infamously concluded.
What this reader says is: that prose is impenetrable.
Sardinian split from the rest of VL about the year -100.
Totally off topic, but I figured a lot of the readers here would enjoy the twitterstorm following the announcement of the motto for the Russian Defense Forces – «Никто лучше нас».
http://tjournal.ru/paper/nobody-is-perfect
There are certain advantages in paleogenetic studies which emerged just recently, so perhaps you can slowly move to give more consideration to it, LH. Ancient DNA is increasingly in play, so we start seeing snapshots of the migrations of peoples of the past millennia, rooted deeper in time than recorded languages. Inheritance of all of the ancestors is evaluated, and timing of admixture events is good. Genes migrate with people and so do languages and cultures, and of course their paths are never identical. But there ought to be a good deal of overlap between the genetic and the cultural migrations, so the more data is out there, the more it comes together to help us understand the past?
The paper by Chang, etc is the one referred to, that will be published in the journal Language.
Shelley: that prose is impenetrable : The article is written for a specialist audience and as such it is very technical, in both linguistic and genetic terms. I skipped most of the parts referring to DNA and such, but even in the linguistic portions I encountered unfamiliar terms (which were not exactly defined but made intelligible through examples). If you can get past those things, the authors seem to take pains to be fair and to give neutral explanations of the two major theories of PIE origins (Steppe and Anatolian) and of the pros and cons of their methods and results, pointing out that they do not have to be complete opposites and each side has some points, although the Steppe is definitely the more likely winner. Not really bedside reading (perhaps as a sleeping aid for some), but worth tackling if you are into that kind of thing (I am, apart from the genetic details).
The U.S. Army’s 2nd Infantry Division uses the punning slogan “Second to None”. Note that neither slogan claims absolute superiority, only equality with the best.
“Genes migrate with people and so do languages and cultures, and of course their paths are never identical. But there ought to be a good deal of overlap between the genetic and the cultural migrations, so the more data is out there, the more it comes together to help us understand the past?”
Makes sense to me.
Sorry I haven’t done this earlier, but I hope your recovery proceeds smoothly and rapidly.
Thanks, Bill!
Ooh, I’ll need to read these papers!
There are a few words that Sardinian has in common with Basque. And then there are features like the prefix ti- on Latin-derived animal names that come out of nowhere as far as anyone can tell.
This book supposedly finds numerous proto-Basque parallels in Sardinian toponims, left over from before Latin.
“Totally off topic, but I figured a lot of the readers here would enjoy the twitterstorm following the announcement of the motto for the Russian Defense Forces – «Никто лучше нас».”
What a great motto for an army. Of course it’s laughable at the moment, but “fake it till you can make it” has worked for a lot of people. I can’t follow the twitter storm but anyway, what could anyone have against this motto.
“The champion there are the Sardinians, who are plausibly largely descended from whoever was living in Europe before IE speakers arrived but nonetheless speak an IE language and have done so since, ”
The other champions who are undeniably pre-Indo-European are the other people with Y-DNA haplotype I (all men of course but it reflects the ancestry of the population as a whole.) – Swedes and Balkan peoples.
http://en.wikipedia.org/wiki/Genetic_history_of_Italy#mediaviewer/File:Haplogroups_europe.png
what could anyone have against this motto.
Like its English equivalent, it can be interpreted as meaning “(even) nobody is better than us”; there have been many amusing variants.
I’ve read the article twice now (barring the appendixes) and I think I follow most of it now. My comments:
1) The “steppe” and “Anatolian” theories are pretty much reduced to “late breakup of IE” and “early breakup of PIE” respectively. There’s no archaeological evidence discussed here, and the only geographic point is that if PIE broke up in Anatolia, it’s weird that the Anatolian branch is the most out of touch from the others: if the non-Anatolian languages dispersed to the four winds, how did they stay in touch and share sound-changes that didn’t affect Anatolia?
2) The input to the phylogenetic software is something unusual, but in this case justified. The goal here is to find out how long ago PIE broke up, not to classify it. Indeed, a small amount of pre-existing classification is presumed, of the form Latin > French, Latin > Italian > Spanish, Old Irish > Irish, Old Irish > Scottish Gaelic, etc., and this is used to constrain solutions. Elapsed time is estimated using lexical replacement (see below), and it does not matter whether the replacement is a new word for a meaning, or a new meaning for a word. Consequently, the unit of interest is the pair (root, meaning), where “root” is a PIE root. For classification you want to eliminate loanwords, as they are noise, but here, loanwords are just as interesting as internal semantic change.
3) The authors face head-on the well-established fact that lexical replacement does not happen at a constant rate. Instead, they assume it is log-normal, and allow the values of the parameters to arise out of the data rather than a priori. The choice of a log-normal distribution may be incorrect, but it’s not obviously wrong nor perverse.
Thanks very much — that’s extraordinarily helpful.
The lognormal seems a pretty obvious choice to me. The replacement rates for different types of lexical features are very likely products of many factors that are also random variables. If there really is enough data to calibrate the parameters with any certainty is a different issue.
Asya Pereltsvaig also reviewed these papers recently, and she raises a few interesting points.
One is the reasoning that if so significant masses of migrating people moved to a new territory that they reshaped the genetic makeup and the culture there, then we must assume that the language of the migrants also took over. (Although subjugated / culturally disadvantaged majorities occasionally lose their original language & switch to the language of the minority rulers, the migrants which possess superior military technologies and transform their new lands both culturally and genetically must be assumed to be the rulers, immune to such a loss).
(Needless to say, it’s harder to prove that the early IE migrating people were *from* Yamnaya as opposed to their being “a close genetic kin of Yamnaya from somewhere else”. It just takes considerably more data to resolve such questions – it’s easy to tease apart very dissimilar ancestors, but harder to figure out which of the more closely related groups ended up where.)
Another is the whole Afanasievo / Tokharian argument rehashed (which remains without any genetic support AFAIK)
JC: thank you for your contribution to understanding the article. I forgot to mention the math, which was beyond me as well as most of the genetics.
Dmitry: if so significant masses of migrating people moved to a new territory that they reshaped the genetic makeup and the culture there, then we must assume that the language of the migrants also took over
The new territory may be inhabited by a linguistically fragmented population (even if largely uniform in genetics and culture), for whom the language of a large incoming migrant group becomes at first a lingua franca before being adopted as the single language of the territory.
I still haven’t read the paper, but it looks to me from your explanation like they’re simply doing molecular dating – they just understand better what they’re doing than Gray & Atkinson did.
From the genome paper: Geographically, these came from Germany (n=41), Spain (n=10), Russia (n=14), Sweden (n=12), Hungary (n=15), Italy (n=1) and Luxembourg (n=1) (Extended Data Table 2).
Even if this is an impressive achievement, both geographically and temporally these are just a few scattered glimpses into ancient European genetics, and the resolution will probably be vastly improved in the next few years. It strikes me that the apparent initial dominance of “Yamnaya” genes before WHG genes bounced back in later genomes could be due to sampling bias if the incoming population, because of e.g. burial practices or preferences in living environment, was better preserved, until cultural and genetic exchange let WHG genes appear in the samples with a frequency comparable to “Yamnaya”.
Caveat: I have just started reading the genome paper in full preview.
Here’s another thought: While I’ve seen strong evidence for the steppe origin, even Anthony couldn’t completely convince me that PIE itself wasn’t actually spoken in one of the neighbouring more culturally advanced regions, Tripolye in the west or Maykop in the south. The apparently mixed “Eastern Hunter-Gatherer” / “Non-Old Europe Middle East” origin of “Yamnaya” should move the interaction with Maykop and other Caucasus cultures to prominence in the genesis of PIE society while Tripolye and Old Europe should be toned down.
“One is the reasoning that if so significant masses of migrating people moved to a new territory that they reshaped the genetic makeup and the culture there, then we must assume that the language of the migrants also took over.”
Turkish? English in America and Australia? Spanish in South America? Arabic in North Africa?
if so significant masses of migrating people moved to a new territory that they reshaped the genetic makeup and the culture there, then we must assume that the language of the migrants also took over.
This seems trivially true. If the aborigines have been replaced genetically, they are unavoidably going to be replaced linguistically and culturally too. However, m-l and George W. seem to be talking about linguistic/cultural replacement without genetic replacement, which is definitely possible. Another example of that is Bengali, where almost all the speakers have fairly recent ancestors who spoke other languages.
Put another way, I think the claim would be that when the incoming group is a demographic minority but politically dominant, both scenarios in which the conquered adopt the conqueror’s language and scenarios where it’s the other way around (Manchus in China, Lombards in Italy etc) are possible, but when the incoming group is both a demographic majority and politically dominant, language shift is only going to happen in one direction.
I think that the Manchus in China, like the Franks in Romanized Gaul, were politically dominant but not culturally dominant, so that they were the ones who adopted the local language.
Assimilation of Manchus is very complex.
First of all, Manchus were not only an ethnic group, but also a caste of military servicemen, like, say, Cossacks in Russia.
This meant that absolute majority of Manchu adult men served in military garrisons stationed in China, Mongolia, Xinjiang, Tibet and Manchuria proper. Obviously, in many of these places, especially in Chinese cities, these banner troops eventually intermarried with natives and lost their native speech and adopted Chinese language.
The birthplace of Manchu nation, the Manchuria region was closed for Chinese agricultural colonization until 19th century, but once these laws were relaxed, waves after waves of Chinese peasants flooded Manchuria, until Manchus became a tiny minority in their own land. Since universal military service for Manchu men meant that they all already spoke Chinese, linguistic assimilation was quite rapid and by the end of 19th century was essentially complete.
To use analogy with Franks, there was complete assimilation in conquered region (France/China), but unlike with Frankish example, there was also a colonization of the original homeland by the conquered peoples.
The original homeland of Franks as we know did not experience anything like that and remained Germanic-speaking region until present.
The language of the Franks is still spoken there. It’s called Dutch now.
“However, m-l and George W. seem to be talking about linguistic/cultural replacement without genetic replacement, which is definitely possible.”
Not me. I was referring to both genetic replacement (or major infusion) and linguistic/cultural replacement. However, I am not suggesting that both are absolutely predictable, just a common phenomenon.
George W: Your examples seem to sort out like this:
Turkish: Language replacement with genetic mixture.
English in America and Australia: Both language replacement and genetic replacement.
Spanish in South America: Language replacement without genetic replacement.
Arabic in North Africa: Language replacement with genetic mixture.
None of those are examples of the type that supposedly can’t happen (and I agree that it can’t): genetic replacement without language replacement.
Spanish in South America: Language replacement without genetic replacement.
A large generalization. There was major genetic replacement in the Southern Cone (old hunter-gatherer zone).
And language replacement too.
“Genetic replacement without language replacement”—why not? If the displacers are speaking the same language as the displaced, say in modern-day gentrification.
Y: Sidney Morgenbesser, Sidney Morgenbesser.
Y: “Genetic replacement without language replacement”—why not? If the displacers are speaking the same language as the displaced, say in modern-day gentrification.
I’d be very surprised if the incoming gentry took on the -lect of the pre-gentrified neighbourhood.
JC—?
Trond, what I meant is, you have situations where the displacers and the displaced speak the very same lect.
Y: Sorry, I forgot to link it.
Y: Well, probably, but that’s the limit state where the set of linguistic features of the incoming population is approaching identity to the one held by the original inhabitants and there’s replacement by identical set.
-Genetic replacement without language replacement
Typical example would be Argentina.
A country with population of primarily Spanish speaking mestizos became a country of white immigrants from Europe (most of them non-Spanish speaking) within a few decades in late 19th-early 20th centuries.
Perfect example of genetic replacement without language replacement (immigrants dropped their native Italian, Portuguese, German and Welsh in favour of Spanish)
Perfect example of genetic replacement without language replacement (immigrants dropped their native Italian, Portuguese, German and Welsh in favour of Spanish)
I don’t find it perfect – it didn’t include social domination of the new arrivals, nor cultural shifts to the newcomers’ culture. The ‘tanos were largely an urban downtrodden mass even if they contribute greatly to the genetic mix
And in Paraguay, very incomplete language replacement with great genetic replacement.
The Argentinean case is the same as the North American 19C case: no one immigrant group dominates, so no chance that any of their languages can replace English. Immigrants aren’t invaders. The Paraguayan case is more interesting; I wish I had access to that paper to see how the “historical events” mentioned in the abstract are connected up.
Language shift is not the same phenomenon when it takes place in a preliterate culture, or even in one which has a literate class but is largely oral at the popular level.
In those circumstances, minorities of immigrant conquerors are almost always assimilated by their more numerous subjects, not vice versa.
Eg., England does not speak Norman French. In fact, despite the total replacement of the English elite by French-speakers after 1066, within about a century French was a second, learned language among their grandchildren and great-grandchildren, maintained at court only by continuous close contact with the European mainland.
The crucial factor is what language -children- learn, and absent formal mass education, that tends to be set by a combination of their parents (particularly their mothers, in most cultures), their caregivers (servants, nannies) and their playmates.
In other words, to preserve a language it must be a majority tongue, at least at the level of the (extended) household and immediate local community.
Add in that in every single case of language shift during the medieval period, the records attest some migration of ordinary peasant-artisan households settling in groups.
Proto-Indo-European has a relatively narrow temporal “window”; given the vocabulary items which can be securely dated to the late Neolithic, the period of PIE linguistic unity must have been between 4000 and 3000 BCE.
The degree of linguistic divergence between the earliest recorded IE languages indicate between one and two millennia of separation — a bit more for the Anatolian group. Eg., Homeric Greek and Vedic Sanskrit, which are so close that they actually share poetic tropes.
The “Anatolian Hypothesis” never did make any sense from a linguistic p.o.v., but the ancient DNA (and isotope analysis) is increasingly driving nails into its coffin-lid. The DNA evidence here, which basically proves that the whole Corded Ware horizon is of Yamnaya origin, fits the time-window precisely. Further resistance is pretty much a matter of identity-protective cognition.
In essence, Gimbutas (and for that matter Coon) were broadly right.
This at least saves us from what I once thought was the inevitable humorless graduate student OD’ing on Renfrew “proving” that the first humans in Britain were not something so uncouth as immigrants, but were instead purely indigenous reindeer, symbolically transformed by the “reception of a cult package”.
Incidentally, there was substantial genetic replacement in Latin America as well as language shift. For example, Mexicans are roughly 65% European (almost all Iberian) genetically; more so, if you leave out communities which still speak indigenous languages. This is broadly typical of Spanish-speaking communities in most of mainland Latin America, outside the ‘southern Cone’. The usual pattern is for the male lineages to be much more European than the female; a case of differential reproductive success.
@s. M. Stirling: That’s interesting. Coon, working long before DNA was even a word, thought Mexicans were only about 15% European.
Welcome back to Language Hat, Mr. Stirling; you haven’t posted here since 2003. Allow me to say how much I admire your work, especially the Nantucket trilogy (which has excellent linguistic aspects!)
Mr Stirling:
While I quite agree with your broader point, I must quibble with one detail: the shared poetic tropes between Homeric Greek and Vedic Sanskrit may or may not be a shared Indo-European inheritance: I brought up the issue here at Casa Hat a number of years ago:
http://languagehat.com/who-were-the-indo-europeans/
I’ve been reading some interesting papers by Margalit Finkelberg. I have no idea what her credibility is, but she has several papers presenting different aspects of a single thesis, namely that the predecessor culture to the “Indo-European” invasion (the “Old European” of Gimbutas) was itself Indo-European-speaking, specifically Anatolian-speaking, representing an early radiation of IE from Asia Minor into Europe. This has three aspects:
1) The “pre-Greek” stratum of names in Greece is in fact Anatolian: she suggests in particular that Parnassos < Luwian parna or some cognate + a possessive adjectival suffix.
2) Linear A can be seen as an unknown but definitely Anatolian language. It had four vowels (the -o glyphs of Linear B have no counterpart in A), and therefore cannot be non-Anatolian IE (not enough vowels), Luwian (too many vowels) or Semitic (again, too many vowels). She looks at a dozen morphological comparanda, and identifies Lycian as the closest Anatolian language to Linear A, though not exactly the same.
3) With two radiations into Europe, Renfrew’s archaeological dates represent the first, the linguistic evidence the second, and so they can be reconciled.
Point (1) has been around for decades, including the Parnassos – parna idea. Finkelberg is a reputable scholar; I’ve only read some of her Greek dialectology work, and remember thinking it had some serious problems, but it’s been a few years and I can no longer reconstruct my objections.
I should have mentioned that these papers are from the 1995-2000 timeframe. What’s new at academia.edu isn’t well-correlated with what’s new in actual academia.
@SFReader:
Aggrieved Celtic pride prompts me to say that my grandfather, who was born in Argentina, was an L1 Welsh speaker. I’ve never been quite clear how much Spanish the family actually spoke; they certainly used Welsh as their everyday language. This was in the first couple of decades of the twentieth century.
And then there is that other guy (on academia.edu as well, but I’m not looking up the reference without coffee) who thinks there is an IE-related substratum of culture words in Sumerian. Would/could that be the same ‘Old Europeans’?
(EDIT: The paper was mentioned here, within the last year or so, but searching is hard).
According to L.A.Gindin, analysis of toponyms in Greece shows four layers – non-Indo-European (Aegean), Anatolian Indo-European, Pelasgic Indo-European (which he identifies with form of Thracian language) and Greek.
Tentative dates of arrival in Greece:
6500-6000 BC – Aegean languages
3000 BC – Anatolian languages
2000 BC – Pelasgic language
1600 BC – Greek
Schrijver or Kroonen or both think “Old European” might be Hattic, or in any case a non-IE language of Anatolia.
I approve of the Hattic theory.
I thought you might.
You would.
It seems to be well corroborated that the Hittites found the Hattians in Northern Anatolia when they invaded from the south around 3800 BP, and that Hattic is very far from being an Indoeuropean language.
Let us assume that the Hittites left the homeland of the earliest PIE speakers a bit before that, like 5000 BP, and that that was in the steppes.
But, for the sake of wild speculation, is there any reason why an ancestor of PIE could not have been present in Mesopotamia and Anatolia even earlier than that, with sister languages to PIE evolving in place or taking part in the Out-of-Anatolia movements after 8500 BP? Providing an adstrate to Sumerian, pre-Anatolian place names for Greece and Old European toponymy for the rest of Europe. Spawning off Tyrrhenian too, and Basque if you like your speculation really wild…
I suspect the answer is that there is absolutely no way to get evidence either way.
Dear Lars,
You are acting crazy lately. Of course all human languages likely merge at some point in pre-history. But even that can not be proven.
It would be interesting if Basque is very distantly related to another language spoken by Neolithic farmers. Maybe Afro-Asiatic? But probably not.
The most recent paper shows that population turnover was very common in Europe. Iberia has always been a refuge for survivors from the previous times.
that other guy […] who thinks there is an IE-related substratum of culture words in Sumerian
Gordon Whittaker, “The Case for Euphratic”. We discussed it two years ago.
that other guy […] who thinks…
I rarely meet that guy…
Dear Rick,
I have no horse in this race, and I am not arguing for such a theory — and as far as I can recall this is my first comment about who might have spoken what and where in that time frame, in this or any other thread — but I’ve seen enough academics who are not clearly lunatic fringe arguing for different potential X-stratal traces of an ancestor of PIE in various languages from the Pyrenees to the Caucasus to make me curious.
Do you have any arguments except those from contempt that such could not be the case? I may have missed something obvious. (I know it can’t be proved, but it might be possible to disprove).
(I think I even read a paper by Carrasquer Vidal once outlining how Basque consonants could have evolved from a system like the PIE three way contrast. But maybe it was so long ago that the PIE system he compared it to was a strain of glottalic — and I don’t remember if he actually got any potential cognates from the exercise).
P.S. Thanks John, I know we discussed it, but I was sure it was more recently than that so I didn’t look at old enough Google results. (And I don’t think the site search here looks at comments).
@Lars
“I have no horse in this race,”
I imagine you have a personal stake in the matter on account of being Etruscan, though.
And I don’t think the site search here looks at comments
No, it doesn’t, which is why I use Google: site:languagehat.com followed by the search term.
a personal stake: Give me a credible PIE etymology for the Etruscan name Lars, and I’ll owe you. I’m stuck with a guy from a village with some bay trees.
(My impression is that Tyrrhenian is another of those cases where most people with knowledge of both that and IE are unwilling to bet against a deep connection. David, is it in your version of Nostratic?)
@Hat, I did just that. But I think I was misled because the thread in question actually started out being about Sumerian, and I thought I remembered it was a case of topic drift… or maybe I just assumed it since so many threads here drift almost instantly.
(The alternative is that someone takes a good grip and hits the topic out of the grounds for a home run. Baseball season started yet?)
I suspect that either the derivation of Laurentum from laurus is folk etymology, or that laurus itself is Tyrrhenian; the connection with Greek daphne looks shaky to me, even apart from dubious l/d alternation.
Dubious… or lubious??
@Lars
If you mean me (which you may well not, as I have zero expertise in this matter) my personal Nostratic is Indoeuropean plus zilch. Minimalist Nostratic. But then I’m a dyed-in-the-wool splitter anyhow. Not convinced by Niger-Congo, even. What do I know?
If somebody ever comes up with a robust way of really, truly demonstrating links inaccessible to the comparative method, I wouldn’t be at all surprised if Tyrrhenian turned out to be a cousin of IE, just as I wouldn’t be at all surprised if Japanese turned out to be related to Turkish. But I’d be more interested in the wonderful new method of securely demonstrating affinity than the actual links it demonstrated.
“dubious l/d alternation”
Neighbouring Latin does it (lacrima, dakru; lingua, tongue); and Etruscan didn’t have contrastive voiced stops, so representing foreign d+ by l+ mightn’t be such a stretch.
Latin “littera” from “diphthera” came via Etruscan if Giuliano Bonfante’s book is right.
I think the Fulfulde word for “book”, “deftere”, is also ultimately from “diphthera.” Not that that has anything to do with it …
That’s cool! How did it get down there?
@JC:
On actually reading your comment with due care and attention, I see that I am in fact agreeing with you …
Dingua > lingua is probably influenced by lingere ‘lick’. And Etruscan does distort the words it borrows from Greek and passes to Latin in weird ways: Ganymede > Etruscan > catamitus, or prosopon > Etruscan > persona. But /pʰn/ > /r/? What kind of sound-change is that? Diphthera > littera is entirely plausible by comparison.
In Lubious Battle would have been a great title. But dubiosus < duo ultimately, and only crazy Armenians and Germans would change that.
@Lazar:
I would think via Arabic. Islam has been a great vector of culture in the Savanna zone of West Africa for a long time, including to groups that aren’t themselves Muslim. Not too surprising when you consider that in that part of the world before the European invasions literacy was basically literacy in Arabic for much the most part. (Not many Kusaasi are Muslim and very few indeed know any Arabic, but one of the folktales in a collection I’ve got is recognisably the same as the Pardoner’s Tale, and presumably got there via the Arabic version of Kalila and Dimnag.)
Dingua > lingua is probably influenced by lingere ‘lick’
Something similar happened in Lithuanian (liežuvis) and Armenian (lezu). The IE ‘tongue’ word undergoes weird irregular changes in several languages. The Oscan cognate of lingua is fangva; needless to say, there is no such correspondence as Lat. l- : Osc. f-. What seems to have happened is aspiration metathesis: *dn̥ǵʰwā > *dʰn̥ǵwā, with a subsequent regular change of *dʰ > f.
In Tocharian, the opposite may have happened, namely metathesis of place features leaving aspiration in place: *dn̥ǵʰwā > *ǵn̥dʰwā > TochB kantwo. (At least, this is one possible explanation; simple metathesis to *ǵʰn̥dwā runs into the problem that PIE *d was lost before consonants in Tocharian, so should not have survived, but you can get around this by having the metathesis postdate d-loss).
Sure; “I” and “me” are mi and mini, which is really close to perfect. What little else is understood about the grammar is at least not crazy in Nostratic comparison. Now if only someone would find the dictionary & grammar by Emperor Claudius…
The relations of Basque are less restricted to guesswork. Basque even shares non-IE, non-AA agricultural vocabulary with Caucasian and Burushaski. Not unlikely, the Basque language arrived with the Neolithic farmers, so that pre-IE Europe may have spoken lots of languages between Basque and… Hattic. Hattic, too, appears to be related to Basque, Caucasian and Burushaski, though attempts to interpret it as specifically West Caucasian appear to have failed. Of course, the trouble here is that Hattic is… apparently, what we have is better understood than Etruscan, but the known vocabulary is smaller, the grammar remains unclear in large parts, and we don’t even know how inadequate the cuneiform writing system is – the only hint is a unique sign combined from wa and a, apparently /va/.
The claim is that they’re not inaccessible to the comparative method, just that most people haven’t tried hard enough. 🙂 And what exactly do you mean by “really, truly demonstrating”?
@David, no it was that David.
@that David, yes, it’s a minor tragedy that that grammar is lost. But maybe it never existed in more than one copy, would have been a miracle for it to survive.
@John, I think the suggestion is that Greek and Etruscan got the bay word from the same non-IE source, not that it came to Etruscan from Greek. I could see something like [daw̥l̥V] go to either [dapʰn-] or [lawr-] when borrowed.
The IE ‘tongue’ word turns out to be even messier than I thought. It’s really only Latin and Germanic that agree at all; besides the metatheses and reformations mentioned above, Celtic has an unexplained initial t-, Balto-Slavic has an unexplained initial zero, and in Sanskrit jihvá only the -hv can be related to *dn̥ǵʰweh₂. Greek γλῶσσα seems to be doing its own thing, but maybe not: apparently there is a German dialectal form Zungen-zolch which, if related, could point to something like *dl(ō)ǵʰ-, which doesn’t look that different from *dn̥ǵʰ-. (Hittite has, lālas, obviously onomatopoeic.) De Vaan handwaves it all away as “taboo deformation”, but why should a word for tongue be taboo?
Of course.
I have always thought it odd (and sad) that Claudius’s works do not survive–and I have a hard time believing that the reason was that very few copies were made. Obviously, it’s possible; if they really weren’t any good that there would have been no demand for his writings. But you would think there would be some interest in books written by the emperor himself.
The absence of Claudius’s writings in the historical record was what prompted Robert Graves to write I, Clavdivs. Graves figured that Claudius, who certainly seems to have been the most intellectual of the early emperors, would have had some real insights about the early Roman empire. Moreover, it’s actually a recurring theme in Graves’s account that Claudius is always trying to get people interested in his books, and hardly anyone cares. (There’s a wonderful scene in the television version where Claudius gets quite indignant that the copies he has ordered of his history of the Second Punic War are being illustrated with silly pictures of elephants.) Graves ended the first volume of the story with Claudius’s ascension to the throne (perhaps he stopped there because he couldn’t figure out what to do with Claudius’s reign, since almost all we know about it comes from Suetonius’s gossip-mongering; eventually he just ran with Suetonius’s account, making the problems with Messalina the central story of Claudius the God), and Claudius’s first thought after being pulled from behind the wall hanging is that he can finally get people to read his books.
Personally, Claudius’s book on Etruscan is the lost book I would second most like to see recovered. The first is Q, the inferred original book of quotations from Yeshua.
What we have is mostly scandalous histories, comedies and sacred writings, and then Cicero and Caesar because their works were held as the epitome of style — the sort of stuff you’d find in the possession of any up and coming merchant.
Compare Dan Brown’s sales figures with Don Ringe’s and you’ll see what I mean. If Caligula ‘happened’ to burn down the TIB CL CAES AUG GERM memorial library one night because his feet were cold or something, most copies could have been lost there and then. Libraries burn too well. If they aren’t flooded.
Well, we do have quite a lot of straight history (e.g. Livy, Tacitus) and some grammatical works (e.g. Quintilian). You’d think being authored by an emperor would aid a book’s chances of survival — cf. Marcus Aurelius — but maybe not when that emperor was Claudius. Do we know how late his works survived? Claudius was, to say the least, not popular during the reign of Nero, so I wonder if they were partly or wholly suppressed as early as that.
I wish we had more ancient Dan Browns, actually.
Tacitus’ works only survived by a whisker.
http://www.historyofinformation.com/expanded.php?id=4199
Livy is a bit different. He was Uplifting …
Several people in that field have stopped inferring it. They now think that Luke is an attempt to improve on Matthew the same way that Matthew is an attempt to improve on Mark.
http://www.markgoodacre.org/Q/farrer.htm
is a (the?) famous essay on this very subject.
It’s a pleasure to read, whatever views you actually hold on the subject
Regarding Austin Farrer’s essay, “On Dispensing with Q,” I would like to point that the editor of its book of collected essays, the Rev. Prof. Dennis Nineham, recently passed away. An obituary is here.
Well, that’s pretty convincing, although it omits any arguments against the converse hypothesis, that Matthew had Mark and Luke to hand, and kept the bits from Luke that he wanted, while using Mark plus his own ideas for structuring. This appears in Farrar’s article only in the parenthesized words “(or vice versa)”, and in the rest of Goodacre’s website not at all.
Although Farrer was known for having, as his intellectual biographer once put it, “contemned the footnote,” the converse hypothesis (Mark – Luke – Matthew) wasn’t really a live option then, while the Mark – Matthew – Luke order which Farrer favored had already been mooted by a couple of Americans. Goodacre does not directly address the converse Matthean posteriority hypothesis, but the logic of his editorial fatigue article should militate against it.
(Disclaimer: Goodacre’s my Doktorvater.)
wasn’t really a live option then
Is it now?
It’s probably been two decades since I delved deeply into these matters, so I have doubtless forgotten a lot, and I have not kept current with new theories. However, I did remember quite clearly that Farrer’s essay was brimming with strawman arguments. I did not feel like reading the whole thing again, but I skimmed enough of it to find a glaring one: Farrer apparently believed that if Q existed, it must have contained everything present in Luke that was not In Mark. He makes particular mention of the story of John the Baptist. Maybe in the 1950s, some people took seriously the notion that the story of John the Baptist was from Q, but I think that was thrown out pretty quickly after people who weren’t Christian theologians began studying the gospels quasi-scientifically. As far as I know, it is generally assumed that John’s story must have taken from other contemporary writings about John (who many believe was a significantly better known figure during their lifetimes than Yeshua; indeed, it has been remarked many times that it’s curious that John ended up as a saint in Yeshua’s religion, rather than vice versa).
Undercover continuation of the cult of John under a new name during Herodian persecution?
@John Cowan is it now?
There’s actually a recent article by Alan Garrow arguing for it this year. But there maybe have been 2 or 3 proponents of it since Farrer wrote, one without any argumentation for it. It’s one of those ideas that seem fine in the abstract but bog down in the details.
@Brett
There have been some changes over the past two decades. The biggest one is that the main competitor to the Two-Source hypothesis used to be the Griesbach hypothesis (Matthew – Luke – Mark) and now it is the Farrer hypothesis.
The John the Baptist material was and is still considered to be part of Q. It is in the _Critical Edition of Q_ (James M. Robinson et al. eds., 2000). And it pretty much has to be part of Q, as usually defined, on philological (rather than theological) grounds, because the verbatim agreements between Matthew and Luke in that material are so strong that there has be some non-Markan literary connection between them. If Luke got that material from Matthew (or vice versa), then much of the rationale for Q disappears. If Luke got that material from another independent source, then we’re not looking at a Q document per se but rather an unknown number of multiple documents — and this runs against the fact that much of the Q-material in Matthew and Luke follows the same order. Before Streeter (1924), quite a few scholars did put the John the Baptist material, not in Q, but in a hypothetical Ur-Markus, supposed to be the source for all three canonical synoptics, of which Mark is an abridgment. But now the synoptic source critics that hold to Q think it is simpler to assign that material to Q and just have one hypothetical source.
Certainly William would have us believe that Luke got his non-Mark material from Matthew, if the converse is excluded.
William?
Of Ockham.
Gotcha.
Well, we do have quite a lot of straight history (e.g. Livy, Tacitus) and some grammatical works (e.g. Quintilian). You’d think being authored by an emperor would aid a book’s chances of survival — cf. Marcus Aurelius — but maybe not when that emperor was Claudius.
On the whole, what texts survive from antiquity reflects the interests of the medieval copyists – authors that were part of the canon that marked an educated Roman, grammars that teach how to write good Latin, and philosophers that could be used to support the Christian faith. A grammar of a language that was probably dead or dying when Claudius wrote it and was mostly associated with pagan rituals and divination simply didn’t tick any of the right boxes.
Good catch!
Come to think of it, Claudius wrote another lost book about the history of Roman religion. There’s another funny scene in I, Clavdivs where it gets mentioned. Claudius is receiving some medication from a Greek physician, and he asks what prayers he’s supposed to say when he takes it. The (apparently atheist) doctor points out that Claudius is both the Pontifex Maximus and the author of a treatise on religion, so he probably can do a better job figuring that out himself.
You’d think being authored by an emperor would aid a book’s chances of survival — cf. Marcus Aurelius — but maybe not when that emperor was Claudius.
One more thought on this – if you read Suetonius’s Life of the Twelve Caesars, you’ll find that he lists works composed by several of them near the end of their biographies (e.g. here are the works of Augustus). AFAIK, very few of these works have survived. Most of them probably were the equivalent in literary value to the ghostwritten “masterpieces” of our contemporary politicians, but it would still be nice to have them…
The (apparently atheist) doctor
That’s been a trope for a long time: Thomas Browne’s Religio Medici was, I believe, a deliberately paradoxical title.
Autotheist doctors are a trope, but atheist ones?
From The Medical Fortnightly 5:6-7 (1894), p. 170: “How is it that you doctors are all atheists?”
Interesting. I can’t get that page to show with any amount of TLD trickery.
How about this?
Try putting [doctors are all atheists] in the Google Books search. But take my word for it, atheist doctors have been a meme for a long, long time.
atheist doctors have been a meme for a long, long time.
Which is slightly odd, given that doctors (at least in the US) are nearly as religious as the general population (in contrast to scientists).
The US is pretty irrelevant here; the meme was established before the country existed. Doctors cut into bodies and didn’t see any souls, and their profession encouraged them to deal with the facts on hand rather than traditions and superstitions (though of course many of them were subject to those latter; it was a tendency, not a universal truth).
We’re monkeys wrapped in cotton, our mind and soul is the brain, an organ subject to failure, were it not for the skull as vulnerable to insult and catastrophic destruction as any other. We die and the only real posterity we leave is our descendants. (With a medium-term view, the single exception is the author of Gilgamesh; with a long-term view, no exceptions).
From which I take; have some kids. You might even get grandchildren, all the enjoyment, practically none of the hassle.
Nope. In that edition I’m not allowed to see anything. The other four let me see a few lines on p. 144, one of which (two in the one placed at the right) contains the word “doctors”; that’s it.
I’m taking your word, though…
P. N. Kardulias, “The Ecology of Bronze Age Flaked Stone Tool Production in Southern Greece: Evidence from Agios Stephanos and the Southern Argolid,”