Don Ringe on Indo-European

August 15, 2010 8:04 AM

The Linguistic Diversity of Aboriginal Europe (and what happened to it)

A series of guest articles by the historical linguist Don Ringe on Language Log:

The Linguistic Diversity of Aboriginal Europe (same as above)
Horse and wheel in the early history of Indo-European
More on IE wheels and horses
Inheritance versus lexical borrowing
The linguistic history of horses, gods, and wheeled vehicles
Some Wanderwörter in Indo-European languages
some other stuff ... (gender. God and thorn clusters)

posted by nangar (35 comments total)

Don Ringe is a supporter of the Kurgan hypothesis. This is of course controversial. There's quite a bit of discussion of this in the comments.
posted by nangar at 8:09 AM on August 15, 2010

Also, I should have mentioned: Previously and previously.
posted by nangar at 8:15 AM on August 15, 2010

Wow, that's really interesting. I had thought (for some reason) that Basque-related languages dominated Europe before Indo-European, but this article's hypothesis seems to be much more scientific.
posted by Electrius at 8:41 AM on August 15, 2010

Proto-Indo-European (PIE) was a single language for which a complex grammar and an extensive vocabulary can be reconstructed[...] It follows that the speakers of PIE must have occupied a comparatively small territory. [...] For instance, the fact that a word for ‘horse’ is solidly reconstructable for PIE (with reflexes in all the earliest-attested branches of the family, including Anatolian) rules out Mesopotamia, Anatolia, and any forested part of Europe as the area where PIE was spoken

I've heard this argument before, and it has never made sense to me. Why would this be the case? Vocabulary jumps into languages easily. It could also mean that PIE-speakers encountered or traded with a horse-riding group, and adopted their word for the exotic animal. The fact that the word is non-native might even explain why the word for "horse" was so stable (the way "sushi" is more stable across IE languages than "bread" is).
posted by Casuistry at 9:11 AM on August 15, 2010 [1 favorite]

OMG THIS! :

It’s the same with the European colonial expansion of the 15th and 16th centuries. Forget the supposed superiority of European ideas; Europeans had better ships and better artillery, and some of the people they encountered didn’t have immunity to Eurasian diseases. No other explanation for European success is necessary.

In the Latin American Studies community we have been saying this for decades. It saddens me that it is still viewed a foreign concept when discussing the history of European imperialism and its bastard child "American Excecptionalism".
posted by liza at 9:16 AM on August 15, 2010 [3 favorites]

The last section, "What it all means." in the first link is a good summary and framework for the previous content. It's a great post/article, but a bit hard to take on if you don't have the background. Also, it contains this gem:

"Secondly, keep constantly in mind that all human society and all human language are single phenomena with multiple instantiations that are only superficially different."

This has implications for, well, everything. It makes language reconstruction possible, as well as refutes any bogus claims about any one language being more or less complex, logical, developed, intelligently designed, etc. than any other.
posted by iamkimiam at 9:42 AM on August 15, 2010 [4 favorites]

I notice Ringe's list does not include Gaelic languages, unless those are in the Celtic sub-family. I am no expert, but I was under the impression that Gaelic has a radically different grammar (verb-subject-object) from the IE languages and that various forms of it dominated much of Europe before the advent of the Roman empire, with only the British Isles retaining pockets after the influx of Latin.
posted by Jimmy Havok at 9:43 AM on August 15, 2010

Thanks for the post; I was of course a fascinated observer of (and occasional commenter in) those LLog threads, and I highly recommend them to anyone interested in this stuff. Don Ringe, a fellow sufferer of Yale grad school in the '70s, is what I was hoping to be if I managed to get my PhD and luck into a job; whatever your opinion on the Kurgan hypothesis, Ringe knows Indo-European backwards and forwards and his ideas should be taken more seriously than most people's. Speaking of which, I should warn everyone that the immense thread at the first link was derailed right off the bat by a kook promoting the "Paleolithic Continuity Theory," and the result was so depressing I just stayed out of it and let others engage him. He popped up a couple of times in the later threads, but they're much more dominated by sensible discussion.
posted by languagehat at 9:46 AM on August 15, 2010

Jimmy Havok, yes, Gaelic languages are Celtic languages, which are Indo-European languages.
posted by Casuistry at 9:53 AM on August 15, 2010

Moreover, most of those states officially recognize only one dialect of each of their official languages—a “standard” dialect.

This seems to be rubbish. The existence of hochDeutsch, for example, has had no discernable effect on Scwabian or that charming language that they speak in Austria.

What I find fascinating is that despite the standardisation of language, national educational systems, radio and television and the movement of people within a country, that regional dialects remain so pronounced and seemingly indelible. May it always be so.
posted by three blind mice at 9:55 AM on August 15, 2010 [1 favorite]

Don Ringe is a supporter of the Kurgan hypothesis.

The hypothesis that it's better to burn out than fade away?
posted by homunculus at 9:59 AM on August 15, 2010 [1 favorite]

Forget the supposed superiority of European ideas; Europeans had better ships and better artillery...


No. You've got to give the devil his due.

The Europeans had better ships, artillery and other military resources directly because of the Renaissance. Reason is not a pure and angelic muse, it's a tool to accomplish worldly ends - and that often means war and violence. The Europeans developed a more efficient and effective way to think about things, and then went on to conquer the world with it. Pre-renaissance Europe, even with the disease vector, would not have been a serious contender as a long-lasting cultural hegemon. (See: Vinland, Crusader States).

This may go a long way to explain the emergence and spread of PIE, even more than horses and chariots - they may have offered a widely-adopted cultural advantage that allowed the language to take root right along with it.
posted by Slap*Happy at 10:35 AM on August 15, 2010 [4 favorites]

Don Ringe is a supporter of the Kurgan hypothesis.

Well, the supporter, really...

...because there can be only one.
posted by Mr. Bad Example at 10:42 AM on August 15, 2010 [1 favorite]

for me, the thing about IE and PIE is that they are part and parcel with 19th century german romantic ideology, i.e. Goethe. however, considering an Ur-culture to go with the Ur-language gets you thrown in with Jungian psychology: artifacts of a supposed ur-culture = archetypes, and isn't considered reputable these days, maybe for good reason. anyway, the idea of an ur-language is very useful for generating an analytical framework for discussing the ancient languages (and could be considered as a logical outgrowth of classical philology) but linguists are technicians and i don't think they are very comfortable about going outside of that playground, which is a shame.

the (pre)history of the culture(s) of the ancient european world is really interesting. i am rereading herodotus, and more than half of his 'Histories' is devoted to an ethnographic portrait of the known work including speculation into pre-history and linguistics. you can get a sense of the diversity of "european" culture even in the classical greek period from reading herodotus.
posted by ennui.bz at 10:57 AM on August 15, 2010

Thanks for posting this. Just a quibble: "unobservable past" is redundant, except in science fiction.
posted by MarshallPoe at 11:02 AM on August 15, 2010

> linguists are technicians and i don't think they are very comfortable about going outside of that playground, which is a shame.

What does this mean? It sounds like you're saying the equivalent of "physicists are technicians and i don't think they are very comfortable about going outside of that playground" because physicists don't go along with UFOs or perpetual-motion machines. That "playground" is called science, and linguists don't go outside it because, being scientists, they stick to what's known or verifiable. I have no idea what it has to do with "german romantic ideology" except that both developed around the same time. And Herodotus tells great stories but his "speculation into pre-history and linguistics" is worthless in terms of anything but describing his view of the world.
posted by languagehat at 11:24 AM on August 15, 2010 [7 favorites]

> "unobservable past" is redundant, except in science fiction.

Why so? Inference based on available evidence is a valid form of observation. In fact, outside of direct sensory experience in the present moment (which is a pretty darned small window onto the universe) it's the only form of observation. The observable past is the collection of past events which left traces from which we can draw conclusions, while the unobservable past would be those past events which left no traces. That we know of. So far.
posted by jfuller at 11:40 AM on August 15, 2010

That "playground" is called science, and linguists don't go outside it because, being scientists, they stick to what's known or verifiable. I have no idea what it has to do with "german romantic ideology" except that both developed around the same time. And Herodotus tells great stories but his "speculation into pre-history and linguistics" is worthless in terms of anything but describing his view of the world.

it's exactly that attitude which leads you to dismissing herodotus, which is a shame. it's interesting, because i think, as an attitude, it's history is rooted very much in old arguments about where to put classical german philology in the 19th century scientific order: just because something is systematic and logical doesn't necessarily make it scientific. i'm not all that interested in debating whether linguistics is a science, my main point is that while it is productive, in the sense of producing linguistics arguments, to separate language from history and culture it ends up impoverishing the discussion. if your linguistics arguments come down to the anthropology of technology, as these seem to re: wheel and horse you can't really afford to only bring in history and culture at the end of your argument. a sophisticated understanding of linguistics combined with a naive view of culture doesn't produce very good "science."
posted by ennui.bz at 11:56 AM on August 15, 2010

Just to add something re: the Kurgan hypothesis, Herodotus is one of the primary sources for information about the culture and history of ancient nomadic pastorialists from the Pontic-Caspian steppe aka "the Scythians" (he describes other nomadic groups there as well... the Getae who believe themselves to be immortal, the "Black Cloaks," the cannibals....)

Apparently, the Scythians made functional and decorative leather items out of the skin of their enemies and smoked a lot of pot.
posted by ennui.bz at 12:32 PM on August 15, 2010

I've heard this argument before, and it has never made sense to me. Why would this be the case? Vocabulary jumps into languages easily. It could also mean that PIE-speakers encountered or traded with a horse-riding group, and adopted their word for the exotic animal. The fact that the word is non-native might even explain why the word for "horse" was so stable (the way "sushi" is more stable across IE languages than "bread" is).

I'm not a linguist, but there's a lot wrong with this speculation. Vocabulary jumps into languages very easily today, with the world being much smaller and with constant reinforcements from advertising, television, films, etc. But your reasoning neglects that fact that earlier peoples tended to be more isolated, and even when roaming about, people and cultures did not mix as they do today. For instance, the Roma are known to have been in certain language territories for at least a few generations, yet only managed to pick up a handful of words (and sometimes, as best is known, no words at all) from those places. The usefulness of the word "horse" can be divined from the fact that it survived in every little tentacle of PIE progeny. That makes it a little difficult to imagine that the word was adopted by *all* PIE speakers and kept by *all* descendent languages by some sort of magical coincidence. Not to mention additional support from the existence of other words that would seem to show that horse-keeping was a part of PIE. Occam's razor applies as well.

Your comparison of "bread" to "sushi" is also bizarre. (Actually, the word in question is really "loaf," not "bread.") Two years ago, a Moldovan girl asked me if I'd heard about sushi. She told me, with great glee, that it was raw fish. Clearly a new word for her. I was also In Kolozsvár, Romania, two or three years ago, when the first sushi restaurant in the city opened and no one could believe it, but you could go to the mall and get free samples for the first week or so, because this was the only way to attract a population of people who'd never had sushi before.

My point is, it's a pretty new word for most of the IE world. The earliest English citation for "sushi" is just over a hundred years ago, and it's fair to say that the word did not probably enter common language for most English-speakers until just a few decades ago. It's also pretty phonetic for most IE languages and therefore, I suspect, less likely to be altered straightaway. So to declare it "stable" is ridiculous, it's simply too recent a word to have changed much. (Although interestingly, it is used differently in English than in everyday Japanese, so there's already been that change.)

"Loaf," on the other hand, has been freakishly stable, given its millennia of life. With a little knowledge of some basic sound shifts, it's recognisably the same to me from English to my native Serbo-Croation, which isn't a very common thing.
posted by Dee Xtrovert at 12:42 PM on August 15, 2010 [3 favorites]

The Europeans had better ships, artillery and other military resources directly because of the Renaissance.

Interestingly, GDP per capita and other measures economic power were pretty much equal between India, China, and Europe until about that period. It was a sudden burst of speed on the part of the Europeans that put them over the top and allowed them to control much of the world. As a result we also got to the Industrial Revolution first, and the results of that are visible to this day.
posted by atrazine at 1:20 PM on August 15, 2010

regional dialects remain so pronounced and seemingly indelible

I'm skeptical of that claim. I've met young people from all over the USA, and they have what to me is an alarming homogeneity in their accents. They all sound to like they are from mid-California, with only shadowy traces of regionalism.

Of course, my sample is for the most part college students, and mostly grad students, so it isn't precisely representative.
posted by Jimmy Havok at 2:35 PM on August 15, 2010

The assignment of Gaelic to the IE family seems problematic to me, since its most basic grammatical structure, word order, is different from that of the other IE languages.

Given the ease with which words are coined and adopted from one language to another, grouping languages by means of shared/shifted vocabulary seems like a questionable practice. I would think that grammar would be more strongly retained than vocabulary, given human ingenuity with words.

I'm quite familiar with a language, Hawaiian Creole English, that takes almost its entire vocabulary from English, but which has a distinguishably different grammar. Like all creoles, it sprang up spontaneously within a mixed-ethnicity culture. Is it Indo-European? By the vocabulary method, it would be considered so, probably as a branch of English, but we know exactly when, where and how it actually arose, and its only relationship to Indo-European is the fact that it uses English words within a spontaneously generated structure.
posted by Jimmy Havok at 2:50 PM on August 15, 2010

Jimmy Havok,

The USA barely has regional dialects for a number of reasons including the homogenising effect of the immigration of many English speakers with different dialects + non-English speakers as well as a national media that is pretty much as old as the country itself. Swabian and Austrian German are way more different from HochDeutsch than any English dialect even within England (which has greater dialect diversity than the US - in fact, the least comprehensible dialects from isolated parts of New England can often be tracked back to a single founding population who all came from the same part of England with their dialect intact.)
posted by atrazine at 2:54 PM on August 15, 2010

my sample is for the most part college students, and mostly grad students, so it isn't precisely representative

At the Northern college I attended, there was intense social pressure on students from the South to lose their Southern accent. They worked hard at it all freshman year, and by sophomore year everyone spoke with the same generic accent. That accent is a prerequisite for being accepted as part of the educated elite.
posted by fuzz at 3:31 PM on August 15, 2010

> while it is productive, in the sense of producing linguistics arguments, to separate language from history and culture it ends up impoverishing the discussion

If what you want to discuss is cool stuff like the guys growing upside down in Herodotus and you don't care about, you know, facts, then sure, dump it all in the blender and see what you get. From my point of view, as someone in the reality-based community, it just creates fun stories; the fact is that, while language is obviously related to culture, you can't tell what that relationship is unless you can observe both at once. Since you can't do that with prehistoric languages, you can only say "Wouldn't it be interesting if..." To avoid that is not (from my point of view) "impoverishing the discussion," it's keeping it grounded in reality.

> The assignment of Gaelic to the IE family seems problematic to me, since its most basic grammatical structure, word order, is different from that of the other IE languages. Given the ease with which words are coined and adopted from one language to another, grouping languages by means of shared/shifted vocabulary seems like a questionable practice. I would think that grammar would be more strongly retained than vocabulary, given human ingenuity with words.

That's because you don't know what you're talking about. I don't want to be mean, but seriously, you're like someone saying "the idea that the earth goes around the sun is problematic to me, because I can see the sun rising and setting." As always, I wish basic linguistics were required in school so everyone would absorb the concepts and discussions could be carried on at a more interesting level, but all I can do is suggest you read some basic material on linguistics so you'll have some idea how languages work and how they change. Word order is quite frequently changed; words for small numbers, basic family members, pronouns, etc. change very rarely and make excellent material for comparison. There is no doubt that Gaelic and the other Celtic languages are IE, and by "no" I mean "zero." Sorry if that upsets your worldview.
posted by languagehat at 4:49 PM on August 15, 2010 [4 favorites]

There's intense social pressure for Northern students to lose their Northern accents, too. You won't hear many Harvard grads pahk the cah at Hahvahd Yahd.

Southerners with heavy accents sound like hicks to Northerners, true, but Northerners with heavy accents sound like mobsters to Southerners, and it's all hicks and mobsters to everyone without a heavy local accent.
posted by Slap*Happy at 6:43 PM on August 15, 2010

> Herodotus is one of the primary sources for information about the culture and history of ancient nomadic pastorialists from the Pontic-Caspian steppe aka "the Scythians"

There's also a bit of archeology. (It confirms the cannabis use, I think, but not the cannibalism.)

The Scythians described by Herodotus lived in a later time period than the Pontic steppe cultures identified by proponents of the Kurgan hypothesis as possible speakers of PIE or very early Indo-European languages. According to the Kurgan hypothesis, the Greek language itself, that Herodotus wrote in, would have been a descendant of a language spoken on the steppes - with a large influx of vocabulary from non-IE (that is, non-steppe) sources.

Cannabis/hemp is an interesting Wanderwort, by the way. It was borrowed into a number of Indo-European languages from an unknown source at a fairly early date. There's a similar word in Sumerian referring to the same plant. (Hemp and cannabis are different forms of the same word, borrowed into English at different times. H in English and other Germanic languages corresponds to C, or a K sound, in other western Indo-European languages, and P corresponds to B, in older shared vocabulary. This softening of certain sounds in Germanic occurred before the first Runic inscriptions in Germanic languages. The word "cannabis" is a modern loan from ancient Greek, without the softening.)
posted by nangar at 8:06 PM on August 15, 2010

I've heard this argument before, and it has never made sense to me. Why would this be the case? Vocabulary jumps into languages easily. It could also mean that PIE-speakers encountered or traded with a horse-riding group, and adopted their word for the exotic animal. The fact that the word is non-native might even explain why the word for "horse" was so stable (the way "sushi" is more stable across IE languages than "bread" is).

This is the exact question Ringe tries to answer in his post here. In short, regular sound change allows you to distinguish an inherited word from a loanword. This is the key paragraph:

There is one other consideration which should be discussed. Nonspecialists sometimes think of languages, including reconstructed languages, as sets of words; but that’s somewhat less than half true. Yes, every language does have a distinctive lexicon, but the structure of the language is even more distinctive; you can replace a large propor­tion of the lexicon with words borrowed from other languages without any significant effect on the language’s structure. (Modern English is an obvious example.) Historical linguists recon­struct a protolanguage’s system of sounds and system of inflectional morphology as well as its lexicon. In some cases the sound system and inflectional system turn out to be complex and intricate, and PIE happens to be one of those cases. Moreover, because we reconstruct protolanguages by exploiting the regularity of sound change, competent re­con­structions are mathematically precise. Under those circumstances, when we reconstruct a word which fits perfectly into the sound system and inflectional system, with no hint that there is anything out of line, the default hypothesis has to be that it’s an inherited word, simply because the odds that a word borrowed from some other lan­guage would fit in well are significantly lower. Of course we’d still like to know whether it could conceivably be a loanword, just to have all the bases covered; that’s why Mr. Marjanović’s question is apt. But unless there’s positive evidence that it is a loanword, linguists will regard that possibility as something of a long shot.
posted by dd42 at 8:33 PM on August 15, 2010 [1 favorite]

Wow, what a fascinating discussion! I'm fascinated by pre-Indo-European Europe and the Kurgan hypothesis, even though I had an art history teacher who was the biggest Gimbutas zealot ever. (It made for a bizarre class, that was for sure. She spent an inordinate amount of time on cave paintings, and she would wax poetic at length about the matriarchal glories of Old Europe and the evils that the Indo-Europeans brought to this goddess-worshiping Garden of Eden. I'm not really exaggerating either.)

@Languagehat-- I'm especially enjoying your comments. Can you tell me in more detail why the Paleolithic Continuity Theory is crap? I always figured it was... but I'd like to hear your take on it.
posted by suburbanbeatnik at 8:46 PM on August 15, 2010

Poor Marija Gimbutas. She had a pretty impressive career as an archeologist. She wrote amazingly well. And then she wrote a kind of speculative book about prehistoric religion, and all the wrong people liked it for the wrong reasons. Now everything she ever wrote gets condemned as "feminist archeology."
posted by nangar at 4:31 AM on August 16, 2010 [2 favorites]

Nangar, didn't at least one book come out attempting to refute many of her conclusions? I'm not sure if that level of effort is condemnation or a sign that one has finally "arrived."
posted by adipocere at 10:03 AM on August 16, 2010

> my main point is that while it is productive, in the sense of producing linguistics arguments, to separate
> language from history and culture it ends up impoverishing the discussion.

Science depends on exactly this sort of "impoverishment." It's also called "abstraction" and it consists of focusing exclusively on what happened -- what you saw, your observations -- expressed symbolically with the fewest possible terms, and excluding from consideration any side tracks about why it happened, expressed in terms at different explanatory levels from the one where you collected your observations. These are usefully excluded because "why" side tracks tend, pretty much irresistably, to lead you further and further away from your data (which is, in the last analysis, all you actually know) and end in idle speculation, handwaving, and internet arguments.

Here's an example of what I mean. Let's say you are able to track the movements of a certain library's book holdings. You tag your observational subjects by title, author, and topic; you track each one as it move around; you code your tracking data symbolically and reduce your symbol set to the lowest number of terms you can. Soon enough you'll have developed a theory (properly so called) of the likely locations of different kinds of book, with "kind" defined operationally by your tags and your tagging procedures. If you're working in the US, chances are good that your theory will end up looking a lot like the Library of Congress classification system, or if you started with a smaller library maybe the Dewey Decimal system. But, crucially, it will not (because it need not) have any explanatory references to the activities of librarians, shelvers, or patrons. You also don't need to worry about particle physics, gravity, or the calculus of stresses and strains to explain why shelves hold books up. Librarians and shelves are entities at different explanatory levels from the one where you did your work, and it's entirely legitimate to treat them as black boxes. They are extraneous distractions when it comes to what you are actually doing when you analyze your data by coding symbolically and reducing terms, which is summarizing (first for yourself and then for others) what you actually studied and what you found.

The abstracting-reducing-systematizing approach lends itself very well to identifying real and definable explanatory terms that nevertheless should be excluded from a given account because they exist and operate at different explanatory levels from the one where your data lives. Noteably, the approach also has the closely related benefit of exposing fuzzy-BS explanatory "entities" that should be excluded because they aren't definable with any precision and their "explanatory power" is specious. Grasping the need for and usefulness of systematic abstraction may be the single biggest difference between the hard and the social sciences (and the origin of the resistance is pretty obvious.) But it appears to me that many linguists are are entirely clear on the matter. Though they often study the deep past, the approach of coding symbolically, reducing, and systematizing appears to me likely to get them closer to what actually happened than any other. If what happened seems impoverished, armchair "...and here's why..." guys can provide the enrichment later.
posted by jfuller at 10:38 AM on August 16, 2010 [2 favorites]

That's because you don't know what you're talking about.

Sorry to be so stupid, but I'm always ready to learn something. You didn't address the question of creoles, though, entire languages made up of loanwords, and I still have trouble with the idea of a radical shift in word-order.

Claiming that linguistics is as settled a science as astronomy makes me think you're a little too deep in the ditch to see the landscape.
posted by Jimmy Havok at 8:47 PM on August 16, 2010

Sorry to be so stupid, but I'm always ready to learn something. You didn't address the question of creoles, though, entire languages made up of loanwords, and I still have trouble with the idea of a radical shift in word-order.

Claiming that linguistics is as settled a science as astronomy makes me think you're a little too deep in the ditch to see the landscape.

Jimmy, do yourself a favour and buy a couple of books about language and linguistics, because you really don't know what you're talking about, and you'd really benefit from a deeper understanding of the hows and whys of all this stuff, since you seem to have an actual interest. The "assignment" of Gaelic to the IE language family isn't actually debatable by anyone with knowledge of basic linguistics. Sure, there are differences between Celtic languages and other branches of the IE family, but the point you bring up has been examined for eons and completely "dealt with." Shifting word-order is widely understood and acknowledged, not just in theory but through empirical historical proof. It's a point once brought up by experts, I'm sure, but now a non-issue. Ditto your concerns about Creole - a good question (re: its linguistic classification), but utterly dealt with. Part of the problem here is that you have a very simple misunderstanding of how languages are classified; it's not simply vocabulary vs grammar. Even a brief foray into some basic reading about how Modern English is descended (quite irrefutably) from Old English would provide you with much insight into how vocabulary and grammar can change dramatically within the continuum of a language.
posted by Dee Xtrovert at 3:36 AM on August 17, 2010 [1 favorite]