Alexis Manaster Ramer

Telling general linguists about Altaic

The hypothesis of an Altaic language family, comprising the Turkic, Mongolic, Tungusic, Korean and, in most recent versions, Japanese languages continues to be a viable linguistic proposal, despite various published claims that it is no longer accepted. A strong body of research continues to appear, developing and refining the hypothesis, along with publications that argue against a demonstrated relationship among these languages. This paper shows that many of the arguments against a genetic relationship fail to address the criteria demanded in modern historical linguistics, while many of the responses from proponents of the Altaic theory have failed to address the criticisms raised. We hope that arguments focusing on the real issues of phonological correspondences and morphological systems will shed greater light on the relationship among these languages.

Capacity, complexity, construction

This was, in essence, the goal of mathematics of language at its inception (even though it would be 1984 before this name would appear). It took a long time before there appeared linguists who were comfortable enough with mathematics (and some non-linguists comfortable enough with linguistics) to see that this idea of finding the natural languages a place in the Chomsky Hierarchy could be wrong. Wasow [50] was perhaps the first, observing that if, as Chomsky seemed to be claiming, there were no natural languages among the finite, regular, or context-free tongues, then the class of natural languages was literally nowhere in the Chomsky Hierarchy, it had to be incomparable to the Chomsky classes. The same point, but buttressed by additional arguments (such as the fact that some context-free patterns like xx R are absent from natural languages, whereas some non-context-free ones like xx are quite common) was made by Rounds et al. [46]. A different tack was taken by Barton et al. [2] and Ristad [45], who argue that the appropriate way to look at natural languages is in terms of time or space complexity classes, rather than classes of structural complexity such as those of the Chomsky Hierarchy. All this might suggest that mathematics of language is a waste of time, but in reality all it means is that it should simply concentrate on some problem other than finding the class of natural languages (qua sets of all and only grammatical strings) a place in the Chomsky Hierarchy. Moving right along, Langendoen and Postal [19] and Hintikka [15] argued that natural languages are not r.e. sets, and hence are outside the Chomsky Hierarchy in a more radical sense (a possibility also discussed by Chomsky [8]). Neither of these

The distribution of /s/ vs. /š/ and related issues in aztecan phonology and etymology

Abstract Extract 1. There always seems to be more to say about the phonological development of the various attested Uto-Aztecan (UA) languages from Proto-Uto-Aztecan (PUA). Since the groundbreaking work of Sapir (1913, 1915), much progress has been made in many areas of UA studies, among them the description of individual languages, the collection of sets of cognates or possible cognates and the discovery of a number of sound laws (e.g., Voegelin, Voegelin, and Hale 1963, Miller 1967, 1988), and the beginnings of a comparative morphology (e.g., Langacker 1977, Heath 1977, 1978) which Sapir had promised but never wrote. But the goal of formulating a precise phonological system for the proto-language and a rigorous account of its developments in each of the daughter languages has not yet been fully accomplished, not even for the most widely studied UA language, Nahuatl (together with the other Aztecan dialects), 2 Unless otherwise indicated, all Nahuatl forms are from Karttunen (1983), Pochutec from Boas (1...

Lárbitraire de chine

Phonologists often look to “external” data to determine which analysis of some set of data is psychologically real for the users of the language. In this vein, Chao (1931, 1934) argued that secret language data support the analysis of Beijing palatals as allophones of the velars. However, several of the secret language rules described by Chao are not allophonic rules of Beijing phonology, and some do not even belong to the phonology at all. In particular, the process which relates palatals to velars is argued not to be allophonic and to be probably quite independent of the phonology (as suggested by data from a second secret language). Depending on how we analyze the latter, the palatals turn out to be phonemically distinct from velars, or else their phonemic status remains indeterminate. Thus, “external” data cannot always be used to settle the cases of indeterminacy (or, non-uniqueness).