Powell’s Map of Native American Languages.

Rebecca Onion posts at Slate about a map so gorgeous and interesting I can’t resist bringing it here:

John Wesley Powell, explorer, geologist, and scientist, produced this map while he was the head of the Bureau of American Ethnology, as part of an 1890 Annual Report. According to Powell’s description of the project, the map plotted “linguistic stocks of American Indians,” as they were situated “at the time when the tribes composing them first became known to the European.”

The map was a culmination of decades of work, Powell wrote in the section of the bureau’s 1891 annual report that described its provenance. “The writer’s interest in linguistic work and the inception of a plan for a linguistic classification of Indian languages date back about 20 years, to a time when he was engaged in explorations in the West,” Powell wrote […]

In his description of the map, Powell exuded scholarly modesty: “[The map] is to be regarded as tentative, setting forth in visible form the results of investigation up to the present time, as a guide and aid to future effort.” But the project was a big deal, writes historian Donald Worster in his biography of Powell: “The classification and map were Powell’s most important achievement as bureau director … and they set the standard for linguists well into the twentieth century.”

Take a look; I’m sure a lot of it is out of date (and I hope marie-lucie will weigh in), but it’s a feast for the eyes. Thanks, Trevor!


  1. It’s quite a famous map, but it’s good to take a look at it once again. It’s not 100% correct, but not far from 90%, either. Many families now bear different names, and there are inaccuracies of territory boundaries: Yokuts is “Mariposan” in Powell, and doesn’t stretch over the mountains; What Powell marks as Siouan in the Carolinas includes Catawba, which is a sister to Siouan, etc. But the picture is overall correct, and even most small language groups (Esselen, Columbia River Athabascan) are marked.
    It does make me sad to think of so much that has been lost after Powell’s time, though of course pretty much everyone on that map had already been impacted then, some badly and some catastrophically. Some language families on the map were already gone by Powell’s time (e.g. Timucua).

  2. I’ve seen the criticism raised (here, perhaps?) that maps like this depict a situation that never was, because the distribution in the east had already been disrupted by the time the time the distribution in the west came about. Is this criticism fair or overblown?

  3. It’s kind of a compromise. This, and similar maps, aim at an idealized view of pre-European distribution, though of course that was dynamic too. If you look at where people were, say in AD 1600, the picture would not be very different from Powell’s map. The Comanches are one group on the map who weren’t there then, and there are probably some others, though I couldn’t say which.

    The other extreme, used by the Ethnologue language maps, strictly shows the present-day distribution, that is a few small reservations in a sea of English and Spanish. That picture is not accurate either, but is an idealization in its own way.

  4. Along those lines, I’ve seen at least one professional map (by National Geographic) that showed a present-day distribution of language families in Africa, Europe and Asia – easily identifiable by the presence of IE in Siberia and Southern Africa – paired with a pre-contact distribution in the Americas and Oceania. The effect is kinda trippy.

  5. Greg Pandatshang says:

    Lazar and Y, I especially wondered about the distribution of Siouan speakers on the great plains. I had the impression that they moved into the area from the Midwest quite recently … but perhaps rather it was Sioux (i.e. Lakota/Dakota) people moving into a region that was already inhabited by speakers of other related languages? Actually, the Great Plains in general are a question mark to me. My understanding is that the economics of living on the plains changed more dramatically there than for indigenous people elsewhere in North America, due to the hunting opportunities created by the introduction of the horse. And so the demographics changed very quickly. I wouldn’t have thought there were Algonquian speakers on the Great Plains south of approx. Canada prior to the horse, either, but I’m not sure just where I got that idea.

  6. Jim (another one) says:

    “but perhaps rather it was Sioux (i.e. Lakota/Dakota) people moving into a region that was already inhabited by speakers of other related languages? ”

    That is exactly what happened. The Absaaloke were way up on the Missouri when the Lakota and Cheyenne formed their alliance to wipe them out and take the Black Hills (at least that’s the story I heard at the Little Bighorn Monument from the Absaaloke park ranger). In fact the reason so many tribal names in English are of Absaaloke origin is that they formed an alliance with the US and that was the contact Army officers had with the situation. The Pawnee did the same thing. The Mandan were another Siouan group up on the Missouri.

    Before the horse the Great Plains were almost but not quite deserted. There were Athapaskans (Apachean) all the way up into Canada because they were the ones with the cultural adaptations need to live in such a stingy environment.

    One historical warp on the map is the extent of Caddoan languages, which by Powell’s time had been nearly erased by Comanche military action. The Comanche very nearly did the same to the Spanish, and they joked they only tolerated Spanish settlements because it was nice to have a place to steal new horses.

  7. Another interesting point is the depiction of Eskimoan along the Côte-Nord of Quebec and in northern Newfoundland. Recent maps don’t show the Inuit getting that far.

  8. Greg Pandatshang says:

    Perhaps both (the Sioux enfringed on what were already Siouan lands, and the Siouans came from the Midwest in recent times). Wikipedia advises me that the Apsáalooke (=Crow people) came from the vicinity of Lake Erie, pushed west in the great chain migration. The same crowdsourced encyclopedia provides circumstantial evidence that the Mandan arrived from Wisconsin. The Great Plains populated, if sparsely, by Athabaskans makes a lot of sense: it means the Navajo and Apaches didn’t necessarily travel quite so far to end up where they are now.

  9. Are the Kiowa Apache remigrants or left-behinds?

  10. I’m not sure which maps you mean, Lazar, but there certainly are Inuit communities in Labrador and Québec, though the Labrador languages are doing worse than those of Nunavut.

  11. No, I mean the northern tip of Newfoundland (the island) and the stretch of Quebec coast to the north of the Gulf of St. Lawrence, which is called the Côte-Nord. Recent maps of North American languages, like this one, don’t show any Inuit presence in these areas.

  12. Trond Engen says:

    The Kiowa are interesting. Their language is Tanoan, but I read that their myths tell that they came from north, rather than from the pueblos. And apparently they were “culturally close” to the Apache.

  13. Oh, I see now. I think that must be indeed an error of Powell’s.

  14. Y, Lazar: there definitely existed Inuit-speaking communities where Powell’s map indicates, i.e. Southern Labrador, the Northern tip of the Island of Newfoundland and neighboring areas of Quebec’s North shore. Some fur traders in colonial times recorded various words and phrases, enough to show that the variety of Inuktitut spoken there must have been very similar to the varieties spoken further North in Labrador today.

    Jim, Greg: do either of you know whether there exists any work on the impact upon Dakota of whatever languages/dialects it replaced as it expanded? In South America the Mapuche, like the Dakota, expanded (into Argentina, in the case of the Mapuche) at the expense of other non-European ethnic groups in post-Columbian times, and there exists some very interesting work on the substrate influence found in those forms of Argentinian Mapuche that are still spoken today.

  15. Greg Pandatshang says:

    Might be harder to see the substrate if the Dakota speakers were encountering mostly their co-Siouans as they expanded. Perhaps better to look for substrate effects in Apsáalooke and Mandan? In any event, I’m afraid I certainly know nothing of such research if it has been done already. Sounds fascinating.

  16. Etienne, interesting! So there were forest Inuit?

    Re Mapuche, do you have some references? I know very little about it, but my impression was that on Chiloë, any oddness about Huilliche might be glibly assigned to Chono (which is barely documented), while in Argentina the process worked the other way, i.e. that the local languages (Gününa Yajüch and Tehuelche) were influenced by Mapudungun, so I would expect the opposite effect not to be so significant.

  17. Y: I guess you could say there were “forest Inuit”, although these groups, economically speaking, lived along the coast and were very sea-oriented.

    Complicating the issue, I dimly recall (and no, I don’t think I could find the reference) that some anthropologist(s) claimed that there had existed a non-Innu, non-Beothuk and non-Inuit ethnic group living on both shores of the Strait of Belle-Ile.

    References on substrate-flavoured Mapuche (Both in Standard Central Iberian Latin):

    Fernandez Garay, A. V. 1997. “El sustrato tehuelche de una variedad del mapuche argentino”. JORNADAS DE ANTROPOLOGIA DE LA CUENCA DEL PLATA/II JORNADAS DE ETNOLINGUISTICA. Rosario (Argentina): 199-205.

    Barros, J. Pedro Viegas. 2005. VOCES EN EL VIENTO: RAICES LINGUISTICAS DE LA PATAGONIA. Mondragon ediciones, Buenos Aires (Argentina): see pages 151 to 163 for an overview of possible substrate features in Argentinian Mapuche (which also contain references to other, more specialized studies, typically dealing with individual substrate loanwords and place-names in Argentinian Mapuche).

  18. various words and phrases, enough to show that the variety of Inuktitut spoken there must have been very similar to the varieties spoken further North in Labrador today.

    eg, komatik

  19. Jim (another one) says:

    “Jim, Greg: do either of you know whether there exists any work on the impact upon Dakota of whatever languages/dialects it replaced as it expanded? ”

    Etienne, two things to remember. one is that the Great Plains were essentially empty until the return of the horse. The second is that Lakota as a language didn’t really come into contact with other languages, for two reasons. One was that most contact was unfriendly and not all that verbal. The Lakota-Cheyenne alliance was an expansionist military machine, and the few Apacheans they encountered either died, were assimilated, or refugeed off to the Kiowa. The second if that most contact was mediated in sign language rather than either side’s spoken language.

    Greg, that’s interesting about the Apsaaloke and Mandan coming from the east. Two pieces of evidence support it. One is that the only thing that made life possible where the Mandan lived in the densities they achieved was corn, and that obviously came form the east and ultimately the south. The second is that the Siouan languages must have had some point of diffusion and that point was likely originally either the middle Mississippi or the Ohio Valley, after migration form North Carolina and perhaps Virginia. That puts them in contention with Irquoian speakers and obviously Algonkian speakers – who may have been the reason they left. We know the Osage left the Ohio Valley under Iroquois pressure and displaced the Tunica.

  20. Thanks, Etienne! I actually have V.-B.’s book, but never looked at that chapter.

    On further looking at Powell’s map, I see Inuit down the western coast of Québec all the way to James Bay, whereas now there are Inuit only as far down as the Belcher Islands. Do you know what the story is with that?

  21. Y…No idea, I’m afraid! I do know that there existed a mixed Cree-Inuit settlement on the Quebec West Coast of James Bay, and apparently there still are a handful of elderly Crees there who speak L2 Inuktitut, but otherwise there was very little Cree-Inuktitut language contact.

  22. Sorry, maybe I shouldn’t intrude here, but if you want the latest extension of John Wesley Powell’s map, go here.

  23. Don’t apologize, everyone’s welcome here — especially if they bring goodies like that!

Speak Your Mind