May 06, 2006

No Word for Thank You

A recurring theme here on Language Log is the claim that such-and-such a language has no word for this-or-that. Such claims are easy to make fun of as they're often wrong and even when true don't have the implications people think they do. There are, however, some serious points to make about these claims. Not only do they rest on false linguistic premises, but they can be quite damaging.

One false premise is the Whorfian one on which we have often commented: not having a word for something does not mean that one lacks the concept. Indeed, there is some very good evidence that people have covert categories, classes of things for which we have concepts but no terms. The evidence for this is nicely presented in Brent Berlin's wonderful book Ethnobiological Classification: Principles of Categorization of Plants and Animals in Traditional Societies. If you haven't read it, you should do so immediately. Not only is it fascinating, but it has lovely illustrations by one of the Tzeltal people whose ethnobiology Berlin has studied extensively. (And if you're the person who has my copy, please return it.) Among other examples, Berlin points out that in Tzeltal there are no words corresponding to "plant" and "animal", but there is a variety of evidence that Tzeltal speakers nonethless classify living things into categories that correspond to "plant" and "animal". One such piece of evidence is the fact that there are classificatory suffixes that reflect these categories.

Another false idea is that if people have borrowed a word that implies that they previously lacked a word for the same idea. There is some truth to the converse of this idea - if a new idea is introduced, people are likely to create a term for it, and a common way to do this is to borrow the word from the language of the people from whom the new idea comes - but people borrow words for reasons other than the absence of an equivalent term in their own language. This is easily seen in the history of languages like Japanese which have vast numbers of doublets, one native, the other borrowed, for the same meaning. In the case of Japanese most of the loans come from Chinese. There is typically both a native Japanese word for something and a loan from Chinese. Usually, though not invariably, the loan from Chinese belongs to a more formal, literary register. One of the difficulties of reading Japanese is that if both words are written in Chinese characters it may be impossible to know which word is intended. Often you can figure this out from the morphology, but sometimes you can't. For example, if you encounter 村人 "villager", a compound of 村 "village" and 人 "person" there is no way to be sure whether to read it as murabito, which is the native Japanese word, or sonjin, the loan from Chinese. You can guess, since sonjin is much more formal than murabito, but you can't be sure.

A particularly damaging example of the No word for X fallacy is one that one hears here in Northwestern Canada. Many of the Athabascan languages of Canada have a word for "thank you" that is borrowed from French merci. In Carrier it is [mʌsi]. This fact has suggested to the ignorant that these languages previously had no word for "thank you", from which they draw the further conclusion that their speakers had no concept of gratitude. Such a people, of course, must have been sub-human savages. The conclusion is that it's a good thing that white people came to rescue them from their degraded traditional way of life. This claim is so well known that it figured in an episode of the television program North of 60, which was set in a Slave village in the Northwest Territories.

The fact is that the loan was not motivated by the lack of a native way to say "thank you". In Carrier, there are actually two different verbs for expressing thanks, one for giving thanks for what someone has said, the other for giving thanks for what someone has done. Both verbs are conjugated for both the subject (the one thanked) and the object (the one giving thanks). Here are the paradigms for giving thanks for what someone has done and for giving thanks for what someone has said. (These are in the Stony Creek (saik'ʌz) dialect.)

snaʧailjaI thank you (one person)
snaʧaɬʌjaI thank you (two or more people)
nahnaʧailjaWe (two) thank you (one person)
nahnaʧaɬʌjaWe (two) thank you (two or more people)
nenaʧailjaWe (more than two) thank you (one person)
nenaʧaɬʌjaWe (more than two) thank you (two or more people)

snaʧadindlihI thank you (one person)
snaʧadahdlihI thank you (two or more people)
nahnaʧadindlihWe (two) thank you (one person)
nahnaʧadahdlihWe (two) thank you (two or more people)
nenaʧadindlihWe (more than two) thank you (one person)
nenaʧadahdlihWe (more than two) thank you (two or more people)

The reason that the subject is the one thanked is that these verbs literally mean something like "you have done me a favour".

The verb for giving thanks for what someone has said is the appropriate verb for saying "No, thank you". Since you are refusing what is being offered to you, you are not giving thanks for receiving something. Rather, you are giving thanks for the offer, which is something that someone has said.

Far from lacking a way of saying "thank you", Carrier had, and has, a more highly articulated, finer-grained way of doing so than English or French. The loan from French is used for relatively casual thanks, and increasingly by semi-speakers and non-speakers, but truly fluent speakers still use the traditional verbs when seriously expressing gratitude.

Posted by Bill Poser at May 6, 2006 01:31 PM