linguistlist.org>
Editor for this issue: Martin Jacobsen <martylinguistlist.org>
fs.tufs.ac.jp>If it is in fact the case that comparative/historical linguistics has come to be unappreciated or misunderstood, this may be because historical linguists have not made adequately clear the nature of the discipline and its limitations. Neo-grammarian rhetoric tends to obscure the fact that language classification and reconstruction are always, necessarily, neither more nor less than the drawing of INFERENCES about the unattested past on the basis of attested langauge data. Hist. Linguistics is based on the assumption that particular patterns in language data (a set of sound correspondences between languages, a pattern of morphomenic alternation within a language, etc.) necessarily, or plausibly, or possibly imply a certain historical process or a certain pre-attestation state of affairs. What basis, indeed what right, do we have to draw such inferences? The best answer, the correct answer, is or should be that we base our inferences on a body of case studies of actually observed changes, or on more general principles of change derived from this body of evidence. That is, given a particular pattern in the data (a sound correspondence, for example), we conclude that it must have come about as a result of one particular process, or could only have come about as a result of one of two or three possible processes, because in all cases in which this pattern is found in languages whose history is known, this is the process, or these are the processes by which it has developed. The classic fallacy in historical linguistics is the assumption that a particular pattern in the data (VSO word order in Celtic and Berber, for example) necessarily or plausibly implies something (that the languages are related, for example) which it perhaps only possibly implies or doesn't imply at all. Unfortunately historical linguistics generally fail to follow this procedure, and too often draw inferences on the basis of criteria which are not explicit. For these reasons, I have to say that I disagree most strongly with Professor Teeter's comments: > In fact, it seems to me that the linguist seeking to study > genetic relationship does very much what the linguist approaching a > new language does. One gathers data and writes a grammar to account > for it. In descriptive studies this results in a grammar of the > language, in comparative studies a grammar of a > protolanguage. Historical/comparative linguistics is the > construciton of grammars for protolanguages. It is I think a grave methodological error to treat historical linguistics as an extension of descriptive linguistics. There is a big difference between describing a body of data and drawing inferences on the basis of the probabilistic implications of that data. In practical terms, the procedure outlined by Professor Teeter would lead us to treat any features shared by the majority of a set of related languages as retentions from the proto-language-- whereas actual language history shows that there are two other posssible explanations for such features-- namely drift and areal diffusion. Was it Meillet who said that Proto-IE could not have been reconstructed on the basis of modern French and Russian and perhaps the relationship between the languages could not even have been recognized? I don't know if this is true. But essentially this is the kind of problem we deal with in Afroasiatic (and other big language families with shallow attestation). Hausa, Tamazight, Oromo, and Egyptian Arabic are about as different from each other (impressionistically) as French, Russian, Albanian, and Bengali, but in the former case we don't have historical documentation except for Arabic. I suspect that there are a number of widely attested features in AfAs langauges which do not go back to the proto-language, and some weakly attested features which probably do. Finding explicit criteria for making a judgement is the challenge. I would be very interested to see some Indo-Europeanist make the effort to try to reconstruct IE on the basis of only the modern languages-- just to see what couldn't be recovered and where one would go astray. It would I think be very useful to have a clearer sense of what the limitations of the method are. As far as classification is concerned, there is a simple systematic procedure for demonstrating a genetic relationship among languages that will be universally accepted. 1) Given that there are some points of similarity between two (or more) languages, it has to be shown that these are not due to chance. Proving that something is not due to chance is a mathematical problem and it has to be formulated in explicitly mathematical terms. 2) If a similarity is not due to chance, it is either due to historical circumstance, or to universal properties of language systems. In order to make a decision at this point we need to know more about language universals. I believe that one reason Altaic became a question again is because of the work by Greenberg and others on implicational universals. So what were thought of as, say, five separate word-order features shared by these languages:-- SOV, AdjN, Postpositions, RelN, GN--are now reduced to a single shared feature. 3) If the similarites are not due to chance or to universals, then they must be due to historical circumstance, but here too there are various types of historical relationships that might obtain between or among speech communities, so the would-be geneticist has to show that the similarties are not due to contact of some kind. / \ chance non-chance / \ universal historical / \ contact genetic Anyone who follows this procedure should be able to establish genetic relationships which are uncontroversial. But of course it is impossible to follow this procedure entirely, because we don't know enough about language universals, we don't know enough about contact, and we haven't advanced far enough in establishing the mathematical foundations of historical linguistics. And that finally may be another reason why classification has perhaps diminished as a research field. For some linguists, at least, these three areas of research-- universals, contact phenomena, and the mathematicization of probability claims-- are all more interesting than the ostensible 'goal' of classification. +++++++++++++++++++++++++++++++++++++++++++++++++ Robert R. Ratcliffe Senior Lecturer, Arabic and Linguistics, Dept. of Linguistics and Information Science Tokyo University of Foreign Studies Nishigahara 4-51-21, Kita-ku Tokyo 114 JapanMail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue
umich.edu>I cannot agree with Benji Wald's suggestion (if I am reading him correctly) that classification of languages is in some kind of methodological crisis. To be sure, there are occasional discussions of methodological issues, but by and large, these are tangential to the real work (much as in theoretical linguistics!). And I submit that the record is one of slowly but steady progress, of which it is a characteristic feature that early classificatory proposals are poorly supported and yet presented with excessive faith, that they elicit essentially valid but often hugely overblown and overly pessemistic criticism, and that after anywhere from a few years to a few decades or even a century, the dust settles and the correct classifications are acepte (these being the minority of those ever proposed) and the others definitely cast out. The recognition that Analolian lgs like Hittie and Lycian are IE took a quarter of a century or more. The acceptance of the Algic family took about the same or more. The acceptance of the connetion of Tlingit to Athabaskan took a half-century. And so on. There is of course methodological progress, but only as part and parcel of the real work (again just as in theoretical ling!!). It took a LONG time for people to see how tonal and nontonal languages could be related and until then the classifiction of teh languages of East and SE Asia was a mess. It took even longer to separate "race" and typology from comparative linguistics, and only when Greenberg did this could African lgs be classified correctly. And so on. The real "problem" is not that comparative linguists do not "agree" on how to classify languages, but that the real story of this (to my mind) not wholly uninteresting part of ling is not taught. I bet no student of lx reading these lines will tell us that they have in fact been taught how the classifications I mentioned above were arrived at and how they came at lenght to be accepted. Moreover, whereas once upon a time the whole subject was more or less ignored or touched on only briefly in courses which focused on other aspects of historical ling, nowadays we e are deluged with disinformation. AND--although I emphatically do NOT mean this to apply to Benji, much of the recent talk about methodology in this area has been nothing but a smokescreen hiding either an effort to discredit work in this area or (when we are lucky) inability or unwillingness to address the substantive issues. This is striking in the public discussions of Nostratic, Amerind, and (among non-Altaicists) of Altaic, but it extends further. Some time ago I had occasion to debate on another list a number of Semitic scholars who could not bring themselves to accept the validity of such concepts as Proto-Semitic and found the same combination of long on methodology but short on linguistics. Of course, other areas of linguistics have experienced similar problems. Much of the resistance, now finally essentially broken, to the laryngeal theory of IE phonology was couched in methodological terms. Much of the resistance to generative grammar in general and to such particular points as rule ordering in phonology or transformations in syntax in particular was for decades couched in methodological terms. In both cases the methodological issues were spurious and utlimately did not matter a whit (and I say this even though I continue to oppose much of what generative grammar has wrought--for SUBSTANTIVE reasons!). Of course, methodology is all-important but it has usually been the case in all sciences I think that methodological progress goes hand in hand with substantive progress. This has clearly been the case with language classification. For example, the correct classification of the Eskimoan languages was achieved by the same person who invented the method whereby this was done (Swadesh), and there any other number of such examples. In any case, the "methodological" attacks on for example Sapir, Greenberg, Illich-Svitych, and so on, have produced nothing of lasting value and indeed time and again have been simply laughable (I wonder how many linguists know that there is an electronic forum of probabilists where they post and sometimes discuss outrageous abuses of probability theory in various sciences, and that recently a widely-hailed work of a criic of Nostratic has earned the honor of being included in this roll of dishonor). Alexis MRMail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue