It’s snowing like crazy here in New Jersey right now. Rutgers canceled all classes today and morning classes tomorrow, so I’ve got a lot of unexpected time off. Seeing all this snow is reminding me, as always, of Navajo linguistics. Words for “snow” play a disproportionately important role in understanding the history and dialectology of the Navajo language.
As Edward Sapir noted in the paper on Navajo linguistic origins I discussed a little while ago, there are two basic “snow” terms in the Athapaskan languages. One refers to snow lying on the ground, and the other refers to falling snow. The terms don’t resemble each other at all, and there is no etymological relationship between them. In Navajo, the “falling snow” term is chííl (usually used in the verb form níchííl “it is snowing” or “the snowstorm has arrived”), and the “snow on ground” term is yas in the western dialect and zas in the eastern. As Sapir also notes, this is one of the very few native isoglosses differentiating the two dialects, which are completely mutually intelligible and differ mainly in that the eastern dialect has borrowed more vocabulary from Spanish.
This dialect difference is odd, and hard to explain. According to Sapir, and supported by the cognates he gives in other languages, yas is the older form, and forms with z are found in only a few Athapaskan languages. Most relevant to the Navajo case is that some of the Apache languages, specifically Jicarilla, Mescalero, and Chiricahua, also have zas. These are all Eastern Apache languages; Sapir doesn’t give a cognate from Western Apache, which is closer to Navajo (so close, in fact, that the two are largely mutually intelligible), but given the rest of the data the term, if it hasn’t been lost, would be yas. To me this suggests that zas is probably a loanword into the eastern dialect from one of the Apache languages. It seems odd that such a basic item of vocabulary would be borrowed, but since the languages are all closely related it becomes more plausible than it might be otherwise. Also, since the eastern dialect is known for its greater number of Spanish loanwords, it makes sense that if either dialect were to borrow an Apache word it would be this one.
So if the word zas was borrowed by the eastern Navajos from one of the Apache groups, when and where did this happen? The number of possibilities is actually considerably more limited than might be thought from the linguistic similarities, since the Navajos in historic times were more often at war than at peace with the various Apache groups. Relations were complicated, however, and trade and intermarriage are known to have occurred. I think the most likely context for this borrowing, however, is a very specific historical event where the Navajos and some of the Eastern Apaches were thrown together against their will.
I am referring, of course, to the US government’s ill-conceived attempt in the 1860s to confine the Navajos and the Mescaleros to a reservation in southeastern New Mexico on the Pecos River, near Fort Sumner, at a place called Bosque Redondo. The Long Walk in which the US Cavalry under Kit Carson rounded up the Navajos and marched them to this reservation is one of the key traumatic experiences suffered by the Navajo people, and it looms large in Navajo history. Perhaps even worse, however, were the atrocious conditions at the reservation itself, due largely to the government’s terrible miscalculation of the number of people it would have to hold. The authorities didn’t have good information about how many Navajos there were, and it turned out there were way too many for the allotted land to support in any kind of humane condition. The supplies the government provided were also grossly inadequate, and the suffering of the Navajos was acute. Malnutrition and disease were rampant, and many people died in the four years before the government realized the project was a disaster and ended it in 1868. By that time the Mescaleros, who were both less numerous and more familiar with the area, which was close to their traditional homeland, had just left and returned home. The Navajos were too far from their own home to just do that, but they did manage to negotiate a new treaty that gave them a new reservation in the area where they had been living before the Long Walk. They returned home and have been there ever since.
The four years at Bosque Redondo were traumatic, but they also involved contact with new people, products, and ideas for the Navajos. Much of what is now known as “traditional” Navajo culture developed during this period, when the Navajos were introduced to manufactured tools, processed foods, and other products initially provided by the government and later, after the return, supplied by traders at trading posts throughout the Navajo country. There was also ideological influence from the Apaches, and a couple of ceremonies taken from the Chiricahuas, and still named as such, have since become some of the most popular healing rituals among the Navajos, probably due in part to their being shorter and less expensive than traditional Navajo ceremonies. I suspect that the word for “snow” may have been borrowed at this time under similar circumstances from the Mescaleros and/or the closely associated Chiricahuas.
But if the Navajos borrowed the word zas from the Apaches at Bosque Redondo, why is it only present in the eastern dialect? Shouldn’t both dialects have it? Well, not necessarily. Although the number of Navajos at Bosque Redondo totally overwhelmed the governments expectations, there were actually quite a few groups who were able to hide out in the rugged canyon country of Utah and Arizona and escape capture by the cavalry. These groups were later joined by returnees, but they managed to preserve a slightly different variety of Navajo culture, less influence by the Hispanics and Apaches in New Mexico, which eventually managed to become today’s western Navajo cultural system. The differences between east and west, with the boundary roughly corresponding to the Arizona-New Mexico border and the Chuska-Lukachukai-Carrizo mountain ranges, are subtle but many, and they have had important effects on Navajo politics and culture in the subsequent Reservation period, extending to this day.
You can learn a lot from words if you look close enough. Sapir’s article used the “snow” words for a very different purpose, to show that the ancestors of the Navajos were unfamiliar with agriculture but very familiar with snow, and both the semantic shift he describes and the fact that there are separate words for snow indicating if it is falling or on the ground strongly point to an origin for the language in the north. The later dialect division I have discussed here, however, sheds some possible light on later developments after the arrival of the Athapaskans in the Southwest, whenever that took place. Every little piece of evidence helps to fill in the puzzle of the past.
Sapir, E. (1936). Internal Linguistic Evidence Suggestive of the Northern Origin of the Navaho American Anthropologist, 38 (2), 224-235 DOI: 10.1525/aa.1936.38.2.02a00040
Nice discussion. I looked up the meaning ‘snow’ in the newly available
World Loanword Database. Unfortunately, there are no Athapaskan languages represented as only 41 languages are listed (hopefully this will grow over time).
But, overall, the database supports the point that ‘snow’ is not a frequently borrowed word as its score is 0.24 = “very little evidence for borrowing.”
Thanks, and thanks for the link to that database, which looks very interesting. Sort of an updated version of the old Swadesh list concept. While it doesn’t have any Athapaskan languages listed, it does have Ket, which Edward Vajda (one of the compilers of the vocabulary list, I see) has recently argued is distantly related to Athapaskan-Eyak-Tlingit. The Ket word for “snow” on the list doesn’t look anything like any of the Athapaskan snow words, which isn’t really surprising. Vajda’s proposal is based mainly on morphology, and the relationship he posits is so distant that lots of terms have likely drifted semantically or been lost entirely on both ends. It is interesting that Ket seems to have very few loanwords, which is also a famous characteristic of Athapaskan languages in general.
Teofilo, I was just delighted to find your website and have greatly enjoyed your writings. Your efforts at being both thoughtful and insightful, as well as factbased, are truly a pleasure to read.
My question has to do with the idea of “loan” words between Apache and Navajo language speakers… it has long been my understanding that both languages have a common ancestor within the Athabaskan family and that the 2 split from each other in about the early 1700s. There has been some but not a lot of evolution within the languages after the split, hence the mutual intelligibility occasionally commented on. As I remember, the Apache version of the language split had to do with a ceremony. I would draw your attention to the similarities between the 2 in the words for various colors and the very great difference between the 2 versions of “black” which seem to me to be a conscious effort to diverge.
Digressions aside, I’m not sure that loaning is exactly the right word. Wouldn’t it be more likely that the parent of both languages had a couple words for snow, which the 2 descendant languages used? No borrowing is necessary as the vocabulary was part of an ancestral vocabulary.
Perhaps my understandings of the languages themselves or their relationships is flawed but I do so enjoy your reflections, I couldn’t help but attach one of my own.
Hi Cyndy,
Glad you like the site. It’s definitely true that Navajo and the various Apache languages diverged fairly recently, but it is still quite possible for them to have exchanged loanwords. There hasn’t been a whole lot of change in any of the languages since the split, but there has been some, and the y/z difference I discuss in the post is one possible example.
One important thing to keep in mind here is that while there is a language called “Navajo” there is not a language called “Apache.” Navajo is one of the Southern Athapaskan languages, and the other languages in the family (which is often called “Apachean”) are spoken by the various groups called by the name “Apache.” These groups are not all linguistically closer to each other than they are to Navajo, however, and Western Apache in particular is quite close to Navajo and the two are generally considered mutually intelligible. The other Apache languages (Jicarilla and Mescalero are the main ones still spoken) are more distant from both Navajo and Western Apache, and may or may not be closely related to each other. Thus, the proposed loan of the “snow” word from Jicarilla or Mescalero into Navajo would be between languages that are closely related and have a recent common ancestor but that are not necessarily mutually intelligible now (although they’re probably pretty close) and have diverged from each other and their common ancestral language in some relatively small but still significant ways.
For a somewhat comparable case, consider loans between Spanish and Portuguese, which diverged from each other probably a bit earlier than the Apachean languages but still pretty recently in the context of the overall history of Indo-European languages. I don’t know of any specific examples, but I’m sure there are some, and hopefully you can see how the borrowing would work and be reflected in the form of the loanwords.
That said, you could well be correct that the y/z alternation in the Apachean languages goes back to a situation in the protolanguage in which both terms were present, with some daughter languages taking one and some the other. (Indeed, this might better explain the way the Apachean languages seem to group differently when looked at from the perspective of different changes.) One of the hardest things about finding loanwords in closely related languages is distinguishing them from this kind of cognate relationship. So yes, I can’t really prove definitively that the “snow” word was borrowed into eastern Navajo from one of the Apache languages, but I do still maintain that that is one possibility for how the situation came to be as it is.