From gonzalor at jhu.edu Thu Oct 1 13:22:30 1998 From: gonzalor at jhu.edu (Gonzalo Rubio) Date: Thu, 1 Oct 1998 09:22:30 EDT Subject: Q: oblique cognates In-Reply-To: Message-ID: ----------------------------Original message---------------------------- In his four vol. work _Comparative Bantu_ (Farnborough: Gregg International Publ., 1967-1971), Malcom Guthrie talked about SKEWED correspondences or reflexes (of "proto-forms"). These skewed forms are very probably connected because of their formal and semantic similarities, but they do not exhibit the regular correspondences one would expect. These skewed correspondences cannot be explained by any specific sound law or known internal development: Meinhof's rule (deletion of voiced stops in nasal contexts: C --> 0 /N___VN[C]); Kwanyama law (like Meihof's rule, but affecting only /mb/ and /nd/); Dahl's law (dissimilation by voicing of /k/ --more or less, the Bantu version of Grassmann's law), etc. Sometimes, it may be the case that we are missing information about the internal history of a concrete Bantu language, but this unknown rule/event would affect just one or two words... unlikely if it's not by analogy. Moroever (and more likely) a skewed form might be a loanword from another Bantu language, instead of a reflex from a "common ancestor" (call it "Ursprache" if you wish), which would explain the irregularity. Nevertheless, many skewed forms seem to exhibit different patterns, suffixes or prefixes, etc. In sum, for some reason we are missing, these skewed forms seem to belong to a coherent set of cognates but exhibit some irregularities. As far as I know, the only use of SKEWING as a concept in historical linguistics besides Guthrie's (who coined it, I think) can be found in a recent book by Patrick R. Bennett, _Comparative Semitic linguistics: A manual_ (Winona Lake, Eisenbrauns: 1998), pp. 30-31 (see p. 31 for examples of skewing in Semitic). In many examples from Bennett's, the skewing depends on the use of different nominal patterns (similar to the caput/cabeza problem), different vocalizations, different genders, truncation, expansion, etc. Other examples from Bennett's present just the usual "irregularities" we all know well (metathesis, prosthesis, etc.). BTW, Bennett's book offers a very useful repertory of materials to teach Semitic linguistics (lists of cognates, paradigms in many languages, maps with lexical isoglosses, etc.). The fist part of the book is a sort of general introduction to historical linguistics, but using Semitic examples. And the rest is just a very practical collection of lists, maps, and so on. I am teaching Comparative Semitic Linguistics this semester, and I recommended the book to my students since all of them know some Semitic languages (Akkadian, Hebrew, Aramaic, etc.) but few are familiar enough with historical linguistics. I hope I didn't skew with my reply, and this is, more or less, what Larry Trask was thinking of. ____________________________ Gonzalo Rubio Near Eastern Studies The Johns Hopkins University Baltimore MD-21218 gonzalor at jhu.edu ____________________________ From dyen at hawaii.edu Thu Oct 1 22:41:57 1998 From: dyen at hawaii.edu (Isidore Dyen) Date: Thu, 1 Oct 1998 18:41:57 EDT Subject: Q: oblique cognates In-Reply-To: Message-ID: ----------------------------Original message---------------------------- It appears that you have run up against the distinction between lexical cognation, the cognation between words implying the existence of a word or d lexeme in their last common protolanguage on the one hand and root cognation or perhaps better put, morpheme cognation in their last common protolanguage. What is causing some confusion is that morpheme cognation is likely to have had a longer history or past life than a lexical cognation. To turn to your 'tooth' example, there appears little reason to doubt that the forms cited continue in some way what was in the protolanguage a single lexeme whose base varied in different inflectional combinations in terms of which later different stages (either dialects or daughter languages) reached their own reorganizations into inflections through analogical changes. The results of such changes can be said to be obliquely cognate or cognate in any other way that one likes, but it is best to keep in mind that they are each aproduct of different uninterrupted contiuities. On Wed, 30 Sep 1998, Larry Trask wrote: > ----------------------------Original message---------------------------- > I am looking for a term for a certain non-canonical type of > cognation. > > One non-canonical variety is as follows. Latin `tooth' > requires a PIE *. English `tooth' and Greek require > a PIE *. Gothic requires a PIE *. The several > forms are therefore not strictly descended from a single ancestral > form, but rather from variant forms of a single root. Such forms as > the Latin, English and Gothic ones have been called `oblique > cognates' in the literature. Fine. > > But there's another case. English `head' is directly cognate with > Latin `head'. However, Spanish does not descend > directly from , but rather from a suffixed derivative of this. > Therefore the English and Spanish words are not directly cognate, > even though they are indirectly cognate in an important way. Is > there a label for this kind of cognation? What would you prefer to > call the relationship between the English and Spanish words? > > Larry Trask > COGS > University of Sussex > Brighton BN1 9QH > England > > larryt at cogs.susx.ac.uk > From Roger.Wright at liverpool.ac.uk Fri Oct 2 11:57:06 1998 From: Roger.Wright at liverpool.ac.uk (Roger Wright) Date: Fri, 2 Oct 1998 07:57:06 EDT Subject: Q: oblique cognates In-Reply-To: Message-ID: ----------------------------Original message---------------------------- In cases like Spanish *cabeza* ("head") - Could we just say that the root morpheme is cognate with the others? That is, Latin *caput* and English *head*; French *chef*, Italian *capo* and Spanish *cabo* ("end") are all cognate; and so is the root morpheme *cab-* of Spanish *cabeza*. RW (Spanish *cabeza* was invented, because of the semantic change that had overcome *cabo* < CAPUT, as a word specifically meaning "head"; the *-eza* was available in the suffix inventory, but seems to have meant nothing at all, and just to have been used as a formal device to create a form different from *cabo*.) On Wed, 30 Sep 1998, Larry Trask wrote: >But there's another case. English `head' is directly cognate with >Latin `head'. However, Spanish does not descend >directly from , but rather from a suffixed derivative of this. >Therefore the English and Spanish words are not directly cognate, >even though they are indirectly cognate in an important way. Is >there a label for this kind of cognation? What would you prefer to >call the relationship between the English and Spanish words? From bwald at HUMnet.UCLA.EDU Tue Oct 6 02:38:42 1998 From: bwald at HUMnet.UCLA.EDU (bwald) Date: Mon, 5 Oct 1998 22:38:42 EDT Subject: Cladistic language concepts Message-ID: ----------------------------Original message---------------------------- Sorry I did not have time to reply to Isidore Dyen's message sooner. He writes: >one of the main factors in linguistic change and >perhaps the main factor is the drive for efficiency in communication which >is dialectically resisted by the need for clarity so that efficiency does >not actually increase and the actual level of clarity does not change. I think there is good reason to keep these two traditional factors in mind in any attempt to understand constraints on linguistic change *overall*. There are, however, other factors involved in promoting change beside efficiency of communication, if that is intended to refer specifically to communication of *referential* information. There is also the factor of maintaining some kind of (local) social identity, viewable as a shorthand for shared knowledge of the local society, and therefore contributing to communication (esp what can be understood within the social group *without having to be said*). An important issue is the extent to which the continual nature of the (socio)linguistic reformation of social identity, as society continually changes, promotes linguistic change, even overriding wider concerns with communication of referential information. This may not be a constant at all times. Maybe sometimes it is more active than at other. It is not clear that there is a constant balance between social factors promoting linguistic change and the counter-pull of intelligibility (to whom? ALL speakers of the "language"? Doubtful.) One possibility is that there are constantly at least TWO regsiters (or styles or whatever) for all mature speakers, one for wider communication, e.g., standards and other lingua francas, and one more susceptible to linguistic change due to the local social identity effect (usually called the "dialect" or "vernacular"). The more lingua-franca like register is also susceptible to change, not to mention interpenetration of what might be viewed as (at least) two registers/styles. However, it remains to be seen how change in one register affects change in another. In any case, it affects the relation between the two more traditional factors maintaining "balance" between "efficiency" and "intelligibility" remains an issue in need of more consideration in understanding the shaping of linguistic change. ID continues: >I believe your characterization of mutual intelligibility as being an >arbitrary criterion is a misconception. After all it concerns >intercommunication, the primary function of language. The difficulty with >mutual intelligibility lies rather in applying it and improvements in that >area could be achieved if the importance of distinguishing languages from >each other could reach the level of attracting financial support. I wouldn't dispute that. It's worth noting, however, that intercommunication as the primary function of language should lead to more concern to promote multilingualism than it does in societies such as the US, among others. This tells us something more generalisable about concerns with intercommunication and its mitigation by the social identity factor. Even so, multilingualism attracts more financial support than mutual intelligibility within what is considered one language. The Ebonics controversy was an interesting issue in trying to move a variety (of language) from one category (dialect variation) to another (multilingualism). ID concludes: >As for entropy it could not be expected to be found in the structure of a >language since the energy input to maintain clarity prevents observable >change in the direction of disorganization. However within a language >regarded as a closed system, there is (for all practical purposes) >observable changes in the direction of disorganization in >dialectalization as diversification tending toward the shattering of a >language. The opposing force is the rate of interlocution; as that rate >is high it militates against diversification and if it is high >enough, promotes homogeneity and when it is low or decreases is >accompanied by increased dialectalization and if it reaches zero, may >be followed by language fission. I found the entropy analogy attractive when I read it in an earlier message. The rate of interlocution is a more complex notion. It may not only be frequency, but also, in some currently hard to specify sense, diversity and/or quality of communication among different groups of interlocutors. Note that *inter*locution already dismisses passive absorption of linguistic norms from the media, whether TV, books or whatever, in favor of mutual communication, even though passive absorption of media norms may be more frequent than active communication for many speakers. One must seek to understand why this is so, to the (large) extent that it is. I still maintain that measures of mutual intelligibility are difficult to interpret in the real world because linguistic distance is not the only factor involved in intelligibility in the real world. Familiarity and motivation to understand (just as in the case of unquestionable bilingualism, and either one, arguably, even without motivation to imitate (accomodate?)) are additional factors. From larryt at cogs.susx.ac.uk Wed Oct 7 15:30:04 1998 From: larryt at cogs.susx.ac.uk (Larry Trask) Date: Wed, 7 Oct 1998 11:30:04 EDT Subject: Sum: `oblique cognates' Message-ID: ----------------------------Original message---------------------------- A few days ago I posted a query about terminology for labeling certain kinds of indirect cognates. I received about a dozen responses from eight different people. The terms put forward are summarized below. Most of them have apparently appeared in print, though not all. (1) `direct cognates': forms which are fully cognate in every detail, and are independently descended from a single reconstructible ancestral form. (2) `lexical cognates': direct cognates (in the above sense) which are word-forms. (3) `paradigmatic cognates': forms which are directly descended, without additional material, from paradigmatic alternants of a single item: what I called `oblique cognates', as in the IE words for `tooth'. (4) `doublets': etymologically related forms descended from variants of a single proto-morpheme. (3) and (4) are perhaps identical. (5) `derivational cognates': forms which are derived by morphological processes from directly cognate stems but which may contain additional non-cognate morphological material of a derivational nature. (6) `partial cognates': forms which contain cognate material but at least some of which contain further material which is not cognate. (7) `morpheme cognates': forms which share at least one cognate morpheme but which also contain additional material which is not cognate (at least in some forms). (8) `root cognates': forms whose roots are cognate but which contain additional material which is not cognate (at least in some forms). (9) `root etymology' (German `Wurzeletymologie'): same as the preceding, but often used in the past as a dismissive term indicating skepticism. (6) to (9), and perhaps also (5). are very similar, and in some cases identical. (10) `deformed cognates': forms which are clearly of a common origin but which require irregularly different etyma, such as the IE forms variously requiring PIE * or * `wolf, fox', particularly when it can be shown that one of the reconstructed forms is original and the other has arisen from an irregular development. These might also be termed `oblique cognates' if the original form cannot be identified. (11) `skewed cognates' or `skewed correspondences': forms which are clearly cognate but which exhibit irregular developments or correspondences. (12) `word family': a set of items which clearly contain cognate material (especially cognate roots) but whose precise relation is obscure. (13) `allofams': the members of such a word family. I'll try to organize these terms as best I can in my dictionary. Further comments welcome. My thanks to Richard Coates, Isidore Dyen, Ralf-Stefan Georg, Martin Huld, Carol Justus, Harold Koch, Gonzalo Rubio, and Roger Wright. Larry Trask COGS University of Sussex Brighton BN1 9QH UK larryt at cogs.susx.ac.uk From dyen at hawaii.edu Fri Oct 9 19:34:47 1998 From: dyen at hawaii.edu (Isidore Dyen) Date: Fri, 9 Oct 1998 15:34:47 EDT Subject: Cladistic language concepts In-Reply-To: Message-ID: ----------------------------Original message---------------------------- I will take up only one point, the one regarding social resistance to change based on the desire to maintain identity. The facts seem to indicate that such resistance is usually coupled with some other basis of resistance such as a political, religious, or economic motivation. The omplication is that such resistance tends to disappear with the disappearance of the motivation. I am not sure that a case can be cited of such resistance apart from such extraneous considerations, but perhaps you have examples. On Mon, 5 Oct 1998, bwald wrote: > Sorry I did not have time to reply to Isidore Dyen's message sooner. He writes: > > >one of the main factors in linguistic change and > >perhaps the main factor is the drive for efficiency in communication which > >is dialectically resisted by the need for clarity so that efficiency does > >not actually increase and the actual level of clarity does not change. > > I think there is good reason to keep these two traditional factors in mind > in any attempt to understand constraints on linguistic change *overall*. > There are, however, other factors involved in promoting change beside > efficiency of communication, if that is intended to refer specifically to > communication of *referential* information. There is also the factor of > maintaining some kind of (local) social identity, viewable as a shorthand > for shared knowledge of the local society, and therefore contributing to > communication (esp what can be understood within the social group *without > having to be said*). An important issue is the extent to which the > continual nature of the (socio)linguistic reformation of social identity, > as society continually changes, promotes linguistic change, even overriding > wider concerns with communication of referential information. This may not > be a constant at all times. Maybe sometimes it is more active than at > other. It is not clear that there is a constant balance between social > factors promoting linguistic change and the counter-pull of intelligibility > (to whom? ALL speakers of the "language"? Doubtful.) > > One possibility is that there are constantly at least TWO regsiters (or > styles or whatever) for all mature speakers, one for wider communication, > e.g., standards and other lingua francas, and one more susceptible to > linguistic change due to the local social identity effect (usually called > the "dialect" or "vernacular"). The more lingua-franca like register is > also susceptible to change, not to mention interpenetration of what might > be viewed as (at least) two registers/styles. However, it remains to be > seen how change in one register affects change in another. In any case, it > affects the relation between the two more traditional factors maintaining > "balance" between "efficiency" and "intelligibility" remains an issue in > need of more consideration in understanding the shaping of linguistic > change. > > ID continues: > > >I believe your characterization of mutual intelligibility as being an > >arbitrary criterion is a misconception. After all it concerns > >intercommunication, the primary function of language. The difficulty with > >mutual intelligibility lies rather in applying it and improvements in that > >area could be achieved if the importance of distinguishing languages from > >each other could reach the level of attracting financial support. > > I wouldn't dispute that. It's worth noting, however, that > intercommunication as the primary function of language should lead to more > concern to promote multilingualism than it does in societies such as the > US, among others. This tells us something more generalisable about > concerns with intercommunication and its mitigation by the social identity > factor. Even so, multilingualism attracts more financial support than > mutual intelligibility within what is considered one language. The Ebonics > controversy was an interesting issue in trying to move a variety (of > language) from one category (dialect variation) to another > (multilingualism). > > ID concludes: > > >As for entropy it could not be expected to be found in the structure of a > >language since the energy input to maintain clarity prevents observable > >change in the direction of disorganization. However within a language > >regarded as a closed system, there is (for all practical purposes) > >observable changes in the direction of disorganization in > >dialectalization as diversification tending toward the shattering of a > >language. The opposing force is the rate of interlocution; as that rate > >is high it militates against diversification and if it is high > >enough, promotes homogeneity and when it is low or decreases is > >accompanied by increased dialectalization and if it reaches zero, may > >be followed by language fission. > > I found the entropy analogy attractive when I read it in an earlier > message. The rate of interlocution is a more complex notion. It may not > only be frequency, but also, in some currently hard to specify sense, > diversity and/or quality of communication among different groups of > interlocutors. Note that *inter*locution already dismisses passive > absorption of linguistic norms from the media, whether TV, books or > whatever, in favor of mutual communication, even though passive absorption > of media norms may be more frequent than active communication for many > speakers. One must seek to understand why this is so, to the (large) > extent that it is. > > I still maintain that measures of mutual intelligibility are difficult to > interpret in the real world because linguistic distance is not the only > factor involved in intelligibility in the real world. Familiarity and > motivation to understand (just as in the case of unquestionable > bilingualism, and either one, arguably, even without motivation to imitate > (accomodate?)) are additional factors. > > > From hubeyh at montclair.edu Sat Oct 10 17:16:21 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Sat, 10 Oct 1998 13:16:21 EDT Subject: Cladistic language concepts Message-ID: ----------------------------Original message---------------------------- > On Mon, 5 Oct 1998, bwald wrote: > > > Sorry I did not have time to reply to Isidore Dyen's message sooner. He > writes: > > > > >one of the main factors in linguistic change and > > >perhaps the main factor is the drive for efficiency in communication which > > >is dialectically resisted by the need for clarity so that efficiency does > > >not actually increase and the actual level of clarity does not change. > > > > I think there is good reason to keep these two traditional factors in mind > > in any attempt to understand constraints on linguistic change *overall*. > > There are, however, other factors involved in promoting change beside > > efficiency of communication, if that is intended to refer specifically to > > communication of *referential* information. There is also the factor of > > maintaining some kind of (local) social identity, viewable as a shorthand > > for shared knowledge of the local society, and therefore contributing to > > communication (esp what can be understood within the social group *without > > having to be said*). An important issue is the extent to which the If the factors all do not operate on the same time scales, the effects will not necessarily cancel out or mutually re-enforce each other. A good example of a similar problem is in the Brownian motion problem solve in the early part of this century. Atoms of a liquid are in continual motion but we cannot see them. The time scale in which they act/react/move is also very short compared to motion at slightly larger scales. The motion we can observe (with a microscope) is the motion of larger objects (like dust particles). The motion of these particles is due to correlated motion of atoms. The basic idea is that like throwing up 1 million pennies. Almost all the time, about half will be heads, and half tails. Similarly of all the atoms banging into these particles, about half will be in one direction and half in the other so that their combined effects will cancel out and there will be no observable motion of the particles. However, just as there will be cases in which about 900,000 coins can be Heads or Tails, there will instants in time in which most of the atoms will be moving in one direction (which is the "correlated motion" of the huge ensemble of atoms) and that effect will be seen in the motion of the particles suspended in the liquid. That motion is Brownian motion. IT is still random just like the underlying random motion of the atoms. But the fluctuations of the atoms cancel out almost all the time at their own time scales. The forces which have a propensity to create linguistic change also occur at various temporal and spatial scales. The rapid and local fluctuations in speech do not normally have permanent global effects. Every once in a while, there will a larger scale changes correlated in space and time, and it is those changes that we track in historical linguistics. IT would probably be a good idea to categorize the changes mentioned into different classes based on time and space scales. -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From bwald at HUMnet.UCLA.EDU Sat Oct 10 17:16:49 1998 From: bwald at HUMnet.UCLA.EDU (bwald) Date: Sat, 10 Oct 1998 13:16:49 EDT Subject: Cladistic language concepts Message-ID: ----------------------------Original message---------------------------- In response to my last message, Isidore Dyen wrote: >I will take up only one point, the one regarding social resistance to change based on the desire to maintain identity. The facts seem to indicate that such resistance is usually coupled with some other basis of resistance such as a political, religious, or economic motivation. The omplication is that such resistance tends to disappear with the disappearance of the motivation. I am not sure that a case can be cited of such resistance apart from such extraneous considerations, but >perhaps you have examples. I did not mean to focus on social resistance to change (which I take to mean change emanating from outside the local community), though that also happens to the extent that vernaculars/dialects endure. I was focussing on the opposite, where local identity (or, more accurately, *local interests*) promotes change (encourages change originating within the community). The general idea is that groups or networks within a local community are continually affected by changes in the social composition of that community, often immigration of new populations into the community. As a matter of course, the newer populations most often assimilate to the linguistic norms of the local community. However, older segments of the community already have established networks, etc., and they are often relatively closed to members of the newer groups. Linguistically, then, the older networks are propelled to change some features of their local language in order to preserve their more exclusive networks. All this happens below the level of consciousness (the conscious analog would be how "slang" continually changes in an "in-group" as its older norms spread to outgroups who adopt it for some reason or other). I am necessarily oversimplifying the nature of social change and how it effects (and affects) linguistic change, but there is a large (socio)linguistic literature illustrating the point I am making, starting with (actually predating, with less rigorous methods of demonstration) Labov's study of linguistic (phonetic) changes among groups on Martha's Vineyard (ca. 1963). The focus, then, is that resistance to changes in the (previous) social structure of a LOCAL community actually promotes (rather than retards) linguistic change. ID is quite right that in the long run a specific social motivation and alignment of interest groups will change, but in the meantime it has its effect on the changes to which the local language is subjected, these are often irreversible, they do not consider mutual intelligibility with external groups (presumably their effect is contrary to such a concern -- ultimately because it weakens what can be understood locally without having to be said, the ultimate in communicative efficiency), and finally, for all intents and purposes, there are almost ALWAYS, i.e., at ANY time, some social changes within any community promoting linguistic changes to differentiate various local interest groups, even though those interest groups continually change and realign in the long run. The last point -- that changes in the social structure of a local community are virtually always going on -- vitiates the notion that social factors are "extraneous", if this is intended to imply that in the long run they can be discounted or are minor in considering whether the two more traditionally acknowledged factors of the balance between "(linguistic) effort" and "intelligibility" are sufficient to account for the direction of (cumulative) linguistic change within a "language". On the contrary, up to the present, history indicates the continual long-term fragmentation of languages and destruction of mutual intelligibility. The vanescence and short-sightedness of LOCAL SOCIAL factors promoting any particular linguistic change (operating below the level of consciousness) no doubt accounts for most of the fragmentation of languages and loss of mutual intelligibility (among the fragments) in the long run. In view of the above, it seems to me that any "balancing" pairs or sets of changes to preserve mutual intelligibility are independent factors which apply primarily WITHIN any particular LOCAL variety of a language, with *questionable* inevitability, and have little if any effect on the eventual fragmentation of a language into mutually unintelligible newer "languages", with (virtually) *unquestionable* inevitability. We are certainly interested in these balancing or "equilibrium-maintaining" mechanisms, but they seem to be quite separate from possibly short-term constantly changing social factors as the MAJOR factor in linguistic fragmentation and loss of mutual intelligibility across local communities. From Roger.Wright at liverpool.ac.uk Mon Oct 12 13:08:12 1998 From: Roger.Wright at liverpool.ac.uk (Roger Wright) Date: Mon, 12 Oct 1998 09:08:12 EDT Subject: Cladistic language concepts In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Benji Wald says: >On the contrary, up >to the present, history indicates the continual long-term fragmentation of >languages and destruction of mutual intelligibility. Yes; nicely put. "Up to the present" -- There's a good case for saying this may never happen again. RW From Roger.Wright at liverpool.ac.uk Mon Oct 12 13:07:58 1998 From: Roger.Wright at liverpool.ac.uk (Roger Wright) Date: Mon, 12 Oct 1998 09:07:58 EDT Subject: Cladistic language concepts In-Reply-To: Message-ID: ----------------------------Original message---------------------------- On Fri, 9 Oct 1998, Isidore Dyen wrote: >----------------------------Original message---------------------------- >I will take up only one point, the one regarding social resistance to >change based on the desire to maintain identity. The facts seem to >indicate that such resistance is usually coupled with some other basis of >resistance such as a political, religious, or economic motivation. The >complication is that such resistance tends to disappear with the >disappearance of the motivation. Sometimes it seems, to us looking back, like unnecessary resistance, at other times it seems like unnecessary innovation. But at the period concerned, it usually takes the form of a motivated choice between existing variants. Malkiel, for example, pointed to several cases in which sixteenth-century Portuguese had two variants available (in morphology or phonetics, but it also applies to vocabulary), both indigenous, and they - perhaps consciously - chose the one that was least like Spanish, asserting their identity that way. Well, sometimes they chose the least evolved variant (thereby appearing to be resistant to change) and sometimes thereby they chose the most evolved (thereby appearing to be changing more than expected) but in neither case was the choice of variant based on relative archaism. But it doesn't look as if they chose forms, archaic or innovative, that didn't actually exist at all at the time; it was a choice between already existing variants. This train of thought and motivated choice of variants is still happening within the non-Castilian regions of Spain, particularly in vocabulary; where Catalan, or Galician, has two words that are for practical purposes synonymous, one of which is like the Castilian word for the same meaning and the other of which is not, the dictionaries and the standardizers of Catalan and of Galician have tended to prefer the one that isn't like Castilian. (In order to annoy the Castilians, but of course in practice it just annoys many Catalans and Galicians, since the Castilians couldn't care less). This is known as "diferencialismo". RW From dyen at hawaii.edu Tue Oct 13 12:40:28 1998 From: dyen at hawaii.edu (Isidore Dyen) Date: Tue, 13 Oct 1998 08:40:28 EDT Subject: Cladistic language concepts In-Reply-To: <361E906D.4C61ACB8@montclair.edu> Message-ID: ----------------------------Original message---------------------------- I like yor contribution to the discussion. When I spoke of a main factor in linguistic change, I was trying to get at the notion that there will be linguistic change regardless of extraneous factor s such as those that bwald is insisting on. The point is, as I see it, that linguistic change is built into the way the community interacts with its language, whereas some aspects of linguistic change are conditioned by the social changes that are going on in the community. The latter type of change, since it is local and temporary I thought could be excluded from being regarded as a 'main factor', but I suppose it gets to be a matter of defintion. The advantage of looking at the matter the way I have been doing it is that the effects of linguistic change in the fragmentation of a linguistic community can be related to the rate of intercommunication among the speakers. Of course this rate is itself dependent on social factors, but can be regarded apart from those factors as a matter of objective observation, even though I would not seriously recommend that anyone undertake to do it, even Labov. In the absence of other factors, we would expect a community to become linguistically disparate and finally mutually unintelligible if it were to separate into two sets that did not intercommunicate with each other for some long period, say a thousand years. Social factors other than the separation could be disregarded even if they could not be actually excluded, human beings being what they are. On Sat, 10 Oct 1998, H.M.Hubey wrote: > ----------------------------Original message---------------------------- > > On Mon, 5 Oct 1998, bwald wrote: > > > > > Sorry I did not have time to reply to Isidore Dyen's message sooner. He > > writes: > > > > > > >one of the main factors in linguistic change and > > > >perhaps the main factor is the drive for efficiency in communication which > > > >is dialectically resisted by the need for clarity so that efficiency does > > > >not actually increase and the actual level of clarity does not change. > > > > > > I think there is good reason to keep these two traditional factors in mind > > > in any attempt to understand constraints on linguistic change *overall*. > > > There are, however, other factors involved in promoting change beside > > > efficiency of communication, if that is intended to refer specifically to > > > communication of *referential* information. There is also the factor of > > > maintaining some kind of (local) social identity, viewable as a shorthand > > > for shared knowledge of the local society, and therefore contributing to > > > communication (esp what can be understood within the social group *without > > > having to be said*). An important issue is the extent to which the > > If the factors all do not operate on the same time scales, the effects > will > not necessarily cancel out or mutually re-enforce each other. > > A good example of a similar problem is in the Brownian motion problem > solve in the early part of this century. Atoms of a liquid are in > continual motion but we cannot see them. The time scale in which they > act/react/move is also very short compared to motion at slightly larger > scales. The motion we can observe (with a microscope) is the motion of > larger objects (like dust particles). The motion of these particles is > due to correlated motion of atoms. The basic idea is that like throwing > up 1 million pennies. Almost all the time, about half will be heads, and > half tails. Similarly of all the atoms banging into these particles, > about half will be in one direction and half in the other so that their > combined effects will cancel out and there will be no observable motion > of the particles. However, just as there will be cases in which about > 900,000 coins can be Heads or Tails, there will instants in time in > which most of the atoms will be moving in one direction (which is the > "correlated motion" of the huge ensemble of atoms) and that effect will > be seen in the motion of the particles suspended in the liquid. That > motion is Brownian motion. IT is still random just like the underlying > random motion of the atoms. But the fluctuations of the atoms cancel out > almost all the time at their own time scales. > > The forces which have a propensity to create linguistic change also > occur at various temporal and spatial scales. The rapid and local > fluctuations in speech do not normally have permanent global effects. > Every once in a while, there will a larger scale changes correlated in > space and time, and it is those changes that we track in historical > linguistics. IT would probably be a good idea to categorize the changes > mentioned into different classes based on time and space scales. > > -- > Best Regards, > Mark > -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= > hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey > =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= > The information transmitted is intended only for the person or entity > to which it is addressed and may contain confidential and/or privileged > material. Any review, retransmission, dissemination or other use of, > or taking of any action in reliance upon, this information by persons > or entities other than the intended recipient is prohibited. If you > received this in error, please contact the sender and delete the > material from any computer. > =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= > From eirikur at rhi.hi.is Tue Oct 13 12:41:38 1998 From: eirikur at rhi.hi.is (Eirikur Rognvaldsson) Date: Tue, 13 Oct 1998 08:41:38 EDT Subject: New publication Message-ID: ----------------------------Original message---------------------------- New publication: Linguistic Studies Historical and Comparative by Hreinn Benediktsson On the occasion of Professor Hreinn Benediktsson's 70th birthday, October 10th, 1998, the Institute of Linguistics, University of Iceland will publish a collection of his papers (two volumes, a total of approx. 700 pages). Most of the papers have appeared elsewhere, but the papers that were not published originally in English have been translated into English for this publication and all the papers have been reedited and retypeset. In addition, the collection contains an Introduction, a general bibliography and an index for all the papers. This edition is being offered at a special pre-publication price to those subscribing before October 25th, 1998. More detailed information, together with the table of contents and a subscription form, can be found at the following website: http://www.hi.is/~nordconf/lingstud.htm --------------------------------------------------------------------- Eirikur Rognvaldsson voice: +354-525-4403 Department of Icelandic fax: +354-525-4242 University of Iceland e-mail: eirikur at rhi.hi.is IS-101 Reykjavik, Iceland URL: http://www.rhi.hi.is/~eirikur From dyen at hawaii.edu Tue Oct 13 21:40:51 1998 From: dyen at hawaii.edu (Isidore Dyen) Date: Tue, 13 Oct 1998 17:40:51 EDT Subject: Cladistic language concepts In-Reply-To: <9807129029.AA902948340@casmail.calacademy.org> Message-ID: ----------------------------Original message---------------------------- Please forgive me for breaking into this discussion of cladistic concepts and linguistics. There is a very sharp difference between the characters of cladistics and the cognates of genetic linguistics. Homologous characters, if I understand cladistic talk, are features that are candidates for having the same origin in some common ancestor. There three analogous structures that I am aware of and that deal with procedures for inferring genetic evolution. One of these is the language family-tree, and the other two are respectively the cladistics of biology and the use of cultural traits in anthropology. 'Traitisitics' has a poor reputation and deservedly so. What cladistics and the linguistic family-tree have in common is the assumption that their basic units do not mix. Once a species or a language becomes distinct, that distinction is indissoluble. Where traits are concerned, such an assumption would lead nowhere because there are too many instances in which cultures have amalgamated in a way that precludes attributing the continuity to just one of the contributing cultures. In other words, cultures do not behave that way. As for cladistics the difficulty surrounds th issue of definable characters. Although broadly the difference between characters is usually easy to see, the discrimination between differences and identities in extreme cases seem to present difficulties. Now that there is competition from the study of DNA, it seems likely that cladistics wil serve to be ancillary support to inferences from the more intrinsic DNA agreements and differences. As for linguistics, it has the advantage of the comparative method which is based on the inferences that can be drawn from the systematic correspondences between phonemes that characterize related languages and raise the likelihood of inferences based on those correspondences, thus giving the inferences scientific standing. On Wed, 12 Aug 1998, Ghiselin, Michael wrote: > ----------------------------Original message---------------------------- > Dear Dr. DeLancey: > Thank you very much for casting your vote. It would be > nice if more linguists would do the same because the sample > as it exists is small and perhaps not representative. > In spite of that the preliminary results are very > interesting. Not only has a cladistic language concept been > generally presupposed but, as you say, linguists do not even > consider the topic particularly interesting. Why should > this be? One possibility is that where linguists have a > written record it lacks the fragmentary nature of the fossil > record that results from accidents of preservation and the > like. Another is that what the linguists do perceive as > important is trying to find older and older common > ancestries and the genealogical relationships are all that > they need. Linguists do not have the elaborate system of > categories, such as phylum, class, order etc., that we > zoologists do. And unless I am mistaken (please correct me > if I am) they do not believe that there are important > differences that need to be expressed by giving a taxon a > higher rank, as when our own species has been put in a > separate order or even kingdom. > Linguists must have methodological problems with > respect to paraphyly, parallelism and convergence. But so > far as I can tell, they treat these as problems to be > overcome in reaching a strictly genealogical arrangement. > MG > From dyen at hawaii.edu Tue Oct 13 21:41:21 1998 From: dyen at hawaii.edu (Isidore Dyen) Date: Tue, 13 Oct 1998 17:41:21 EDT Subject: Cladistic language concepts In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Your point is interesting and relevant, but the question would remain whether Malkiel or anyone else for all of their approach to omniscient are able to control all the facts. What is at issue for theories concerning linguistic change is what is intrinsic to natural language as an instrument and what is extrinsic. This question concerns, I suppose, what we would expect to happen in a language in the absence of abnormal social events like conquests and enslavements, restrictive laws including persecutions, that sort of thing. Under the least interference I would expect linguistic change to go on, unevenly in proportion to the extent of the community with homogeneity depending on the rate of interlocution ( or intercommunication). Such a theory which distinguishes between linguistic change that is inherent and that which reflects social stress may suffer because the distinction between the two effects cannot always be made and perhaps can never be made, but would have the value of recognizing one of the components or factors in linguistic change. Resistance to maintain identity comes into play when the threat to the identity involves some loss and in this sense the insistence on the identity is a pursuit of advantage. Although this pursuit is omnipresent, questions of identity do not usually arise without association with loss or gain. Perhaps what I am trying to say is that in a community in which threats or advantages related to identity did not arise, lingusitic change would still go on and in this sense such threats and advantages are extraneous social factors, no matter how close to universal. On Mon, 12 Oct 1998, Roger Wright wrote: > ----------------------------Original message---------------------------- > > > On Fri, 9 Oct 1998, Isidore Dyen wrote: > > >----------------------------Original message---------------------------- > >I will take up only one point, the one regarding social resistance to > >change based on the desire to maintain identity. The facts seem to > >indicate that such resistance is usually coupled with some other basis of > >resistance such as a political, religious, or economic motivation. The > >complication is that such resistance tends to disappear with the > >disappearance of the motivation. > > > Sometimes it seems, to us looking back, like unnecessary resistance, at > other times it seems like unnecessary innovation. But at the period > concerned, it usually takes the form of a motivated choice between > existing variants. Malkiel, for example, pointed to several cases in > which sixteenth-century Portuguese had two variants available (in > morphology or phonetics, but it also applies to vocabulary), both > indigenous, and they - perhaps consciously - chose the one that was > least like Spanish, asserting their identity that way. Well, sometimes > they chose the least evolved variant (thereby appearing to be resistant > to change) and sometimes thereby they chose the most evolved (thereby > appearing to be changing more than expected) but in neither case was the > choice of variant based on relative archaism. But it doesn't look as > if they chose forms, archaic or innovative, that didn't actually > exist at all at the time; it was a choice between already existing > variants. This train of thought and motivated choice of variants is > still happening within the non-Castilian regions of Spain, particularly > in vocabulary; where Catalan, or Galician, has two words that are for > practical purposes synonymous, one of which is like the Castilian word > for the same meaning and the other of which is not, the dictionaries and > the standardizers of Catalan and of Galician have tended to prefer the > one that isn't like Castilian. (In order to annoy the Castilians, but of > course in practice it just annoys many Catalans and Galicians, since the > Castilians couldn't care less). This is known as "diferencialismo". > RW > From maxw at cogs.susx.ac.uk Wed Oct 14 15:20:09 1998 From: maxw at cogs.susx.ac.uk (Max W Wheeler) Date: Wed, 14 Oct 1998 11:20:09 EDT Subject: Cladistic language concepts In-Reply-To: Message-ID: ----------------------------Original message---------------------------- On Mon, 12 Oct 1998, Roger Wright wrote: > > choice of variant based on relative archaism. But it doesn't look as > if they chose forms, archaic or innovative, that didn't actually > exist at all at the time; it was a choice between already existing > variants. This train of thought and motivated choice of variants is > still happening within the non-Castilian regions of Spain, particularly > in vocabulary; where Catalan, or Galician, has two words that are for > practical purposes synonymous, one of which is like the Castilian word > for the same meaning and the other of which is not, the dictionaries and > the standardizers of Catalan and of Galician have tended to prefer the > one that isn't like Castilian. (In order to annoy the Castilians, but of > course in practice it just annoys many Catalans and Galicians, since the > Castilians couldn't care less). This is known as "diferencialismo". > RW Can you actually substantiate this claim, with regard to Catalan? It is true that such an effect may appear to someone who doesn't know about the history of the languages (obviously not you); in several cases standardizers prefer a form with a longer established literary tradition. Note that people from French Catalonia have accused the standardizers of preferring those forms which were MORE similar to Spanish. Max Wheeler ___________________________________________________________________________ Max W. Wheeler School of Cognitive and Computing Sciences University of Sussex, Falmer, Brighton BN1 9QH, UK Tel: +44 (0)1273 678975; fax: +44 (0)1273 671320 ___________________________________________________________________________ From bwald at HUMnet.UCLA.EDU Thu Oct 15 15:43:41 1998 From: bwald at HUMnet.UCLA.EDU (bwald) Date: Thu, 15 Oct 1998 11:43:41 EDT Subject: Cladistic language concepts Message-ID: ----------------------------Original message---------------------------- Roger Wright quotes from my last message. >Benji Wald says: > >>On the contrary, up >>to the present, history indicates the continual long-term fragmentation of >>languages and destruction of mutual intelligibility. >Yes; nicely put. "Up to the present" -- >There's a good case for saying this may never happen again. I'm glad that Roger appreciated the qualification I put on what I was saying. I deleted a further paragraph on how I invite such views as he is proposing about a sharp discontinuity between the past and the future, since I see little reason to suppose that such a discontinuity has come about in the 20th c. or will in the foreseeable future, despite impressive advances in communication technology (at least impressive to us current beings) and increasing sharing of various kinds of literacies. Meanwhile, the same old problems of miscommunication that have always existed (and have occasionally been reported in the past) persist. (Would it be surprising if "human nature" exists not only in our linguistic devices but in how we use them?) I think the main thing that would interest me among Roger's proposals is what changes in the nature of human society (as a whole or in its various various parts) he envisages or suggests to have relatively recently arrived which will override the steady and unrelenting effects of social localism that in the past (presumably) have been the major cause of linguistic fragmentation and loss of mutual intelligibility. I am skeptical, and suspect he is underrating the long-term cumulative effects of localism, and overrating the stability of centralised power, but I am open to hearing interesting proposals about the "changing" relation of social change to linguistic change. With regard to Isidore's latest message, I share his appreciation for Hubey's comments, but paused at the following passage: >The point is, as I see it, that linguistic change is built into the way the community interacts with its language, whereas some aspects of linguistic change are conditioned by the social changes that are going on in the community. The latter type of change, since it is local and temporary I thought could be excluded from being regarded as a 'main factor', but I suppose it gets to be a matter of defintion. I'm not sure I understand the intent of "local and temporary". The "temporary" part seems to suggest that the local changes are eventually undone, as if afterwards they seem to have never occurred (i.e., no *lasting* harm done to mutual intelligibility). If that is not what is meant, then they have had their effect in changing the local language AWAY from other local varieties. This seems more than a matter of definition (of "main factor"?) to me, but of the cumulative consequences of local and temporary (temporally bounded?) changes. Of course, most changes do spread beyond the temporary local interest group that intiates them, so their effects don't go away but continue, and establish the typical mosaic patterns we commonly find in dialect geography. I cannot address all fronts at once, so when I emphasised fragmentation and loss of mutual intelligibility as the traditional and still usual focus of historical linguistics (at least on the elementary level), I did not complicate that traditional picture with diffusion and convergence which takes place across languages as well as in them, due to bilingualism, bidialectalism, register/stylistic complexity, or whatever level of analysis is appropriate at the time. I am continually struck by cases in which it is not clear which family or (even more commonly *in the literature*) branch of a family some language (group) or other belongs to because of convergence and sharing of features. Such things most strikingly occur at the margin of isogloss bundles. Isidore (and no doubt Roger) is certainly right that channels and interests promoting communication ("mutual" intelligibility) *across* local groups is also a factor in change. The issue remains: will languages continue to fragment and produce mutual unintelligbility, or will the whole world eventually abandon its local variety in favor of some "homogeneous commercial or totalitarian English" or whatever (no doubt incomprehensible to us current beings). I think not. From mcv at wxs.nl Thu Oct 15 15:38:30 1998 From: mcv at wxs.nl (Miguel Carrasquer Vidal) Date: Thu, 15 Oct 1998 11:38:30 EDT Subject: Cladistic language concepts In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Max W Wheeler wrote: >On Mon, 12 Oct 1998, Roger Wright wrote: > >> where Catalan, or Galician, has two words that are for >> practical purposes synonymous, one of which is like the Castilian word >> for the same meaning and the other of which is not, the dictionaries and >> the standardizers of Catalan and of Galician have tended to prefer the >> one that isn't like Castilian. (In order to annoy the Castilians, but of >> course in practice it just annoys many Catalans and Galicians, since the >> Castilians couldn't care less). This is known as "diferencialismo". >> RW > >Can you actually substantiate this claim, with regard to Catalan? It is >true that such an effect may appear to someone who doesn't know about >the history of the languages (obviously not you); in several cases >standardizers prefer a form with a longer established literary >tradition. Note that people from French Catalonia have accused the >standardizers of preferring those forms which were MORE similar to >Spanish. For Catalan, most of the standardizing work was done in the first half of this century, primarily by the orthographical and grammatical standards of the Institut d'Estudis Catalans and the dictionary of Pompeu Fabra. Fabra talks about these issues in a 1924 article ("L'obra de depuracis del catal`") and admits that "hi hagui una hpoca en quh, de dues formes, de dos mots, de dos girs en concurrhncia, es creia que sistem`ticament calia donar la preferhncia al mis allunyat de l'espanyol. En tota coincidhncia entre l'espanyol i el catal`, es veia un castellanisme, i bastava que un mot s'assemblis massa a l'espanyol corresponent perquh se li cerquis mis o menys arbitr`riament un substitut" (There was a time when it was thought that if there were two competing forms, words or expressions, one had to prefer the one furthest from Spanish. In every agreement between Spanish and Catalan a castilianism was seen, and a word only had to look too similar to the corresponding Spanish one in order for more or less arbitrary substituions for it to be sought.) [The above passage contains three such cases in relation to my own dialect: I would say and quite possibly write for "words", for "would search", although I would not write (but possibly say) for "less"]. While trying to steer clear of such exaggerations as described above by Fabra, standard Catalan has drastically reduced the number of castilianisms, often by harking back to the Medieval literature. Many castilianisms of course still remain in the spoken language, and a considerable number have also been maintained in the standard, as it would be silly (and impossible) to deny that contact with Castilian ever existed [after all, there are also catalanisms in Castilian]. In many of these cases where standard Catalan has perforce a castilianism (more or less catalanized), a gallicism is likely to be used in Rossells. I suppose that's what the complaints may have been about. ======================= Miguel Carrasquer Vidal mcv at wxs.nl Amsterdam From martha_ratliff at wayne.edu Fri Oct 16 15:15:08 1998 From: martha_ratliff at wayne.edu (Martha Ratliff) Date: Fri, 16 Oct 1998 11:15:08 EDT Subject: 7th annual workshop on comparative linguistics Message-ID: ----------------------------Original message---------------------------- 7TH ANNUAL WORKSHOP ON COMPARATIVE LINGUISTICS November 21-22, 1998 at The University of Illinois at Urbana-Champaign Room 210, Illini Union THE RECONSTRUCTION OF CULTURE Speakers and topics include: Hans Henrich Hock: "Use and misuse of linguistic prehistory: India and beyond" Frederick Schwink: "The reconstruction of IE gender as a reflection of culture" Mary Niepokuj: "IE poetics" Rex Wallace: "Sabellian" Martha Ratliff: "Vocabulary of environment and subsistence in Proto-Hmong-Mien" William Baxter: "Evidence for early Austronesian-Chinese contact in China" Colleen Reilly: "Gender and sexuality in Anglo-Saxon" Graham Thurgood "What proto-Chamic reconstructions tells us about early Chamic culture" Craig Hilts "Vocabulary for flora and fauna of Mixe-Zoquean" Co-sponsored by: English, Program in South Asian and Middle Eastern Studies, Germanic Languages and Literatures, Linguistics, Classics, Program in Comparative Literature. For information, contact Frederick Schwink Dept. of Germanic Languages University of Illinois at Urbana-Champaign 3072 FLB 707 South Mathews Ave. Urbana, IL 61801 From sw271 at cus.cam.ac.uk Fri Oct 23 11:25:42 1998 From: sw271 at cus.cam.ac.uk (Sheila Watts) Date: Fri, 23 Oct 1998 07:25:42 EDT Subject: r and s Message-ID: ----------------------------Original message---------------------------- Hullo everyone, Could anyone help me with the answer to a small question which has been interesting me for some time? There are alternations between r and s in different parts of the paradigm of some Latin words (only verbs?), and these are apparent in borrowings into English such as adhere vs. adhesive, adhesion. Now these alternations are similar to those found as a result of the operation of Verner's Law in words like English was - were, German war - gewesen, kiesen - kor and so on. Is there a link, or is this phenomenon typologically common? Thank you, Sheila Watts _______________________________________________________ Dr Sheila Watts Newnham College Cambridge CB3 9DF United Kingdom phone +44 1223 335816 From sw271 at cus.cam.ac.uk Fri Oct 23 15:53:53 1998 From: sw271 at cus.cam.ac.uk (Sheila Watts) Date: Fri, 23 Oct 1998 11:53:53 EDT Subject: r and s Message-ID: ----------------------------Original message---------------------------- Thank you for helpful postings to date. Apologies for suggesting that the Latin development was peculiar to verbs - I had lost sight of the wood for the trees. If I have it right, English adhere comes from a Latin verb of which the infinitive is hærere and the fourth principal part hæsum, so Latin rhotacism must be more complex than just 'inlautend zwischen Vokalen', which was the description I had found. What had struck me was the similar effect (to Germanic) of having a change in consonant across the principal parts of a verb. Sheila Watts _______________________________________________________ Dr Sheila Watts Newnham College Cambridge CB3 9DF United Kingdom phone +44 1223 335816 From ph1u+ at andrew.cmu.edu Fri Oct 23 15:53:31 1998 From: ph1u+ at andrew.cmu.edu (Paul J Hopper) Date: Fri, 23 Oct 1998 11:53:31 EDT Subject: r and s In-Reply-To: Message-ID: ----------------------------Original message---------------------------- On Sheila Watts' query about s/r alternations in Germanic and Latin (so-called rhotacism): I can't think of other examples of it outside these two groups, but the two processes were the results of independent changes s > z, followed by z > r and a lot of analogical restoration or generalization of the original s. The environments for s > z are a bit different, since in Latin it occurred between vowels regardless of accent. The oldest Germanic documents still have the original sibilant. z > r doesn't seem to be a commonly attested change, and it is an interesting coincidence that it should occur independently in two families often seen as quite close within Indo-European. I don't have my copy of C D Buck's Comparative Grammar of Latin and Greek with me here, but I'm sure there's a good clear presentation of the Latin situation there, and E Prokosch, A Comparative Germanic Grammar would be a good place to go for the Germanic details. Paul Hopper Thomas S. Baker Professor of English & Linguistics Carnegie Mellon University Pittsburgh PA 15213, USA Excerpts from mail: 23-Oct-98 r and s by Sheila Watts at cus.cam.ac. > ---------------------- > Hullo everyone, > Could anyone help me with the answer to a small question which has been > interesting me for some time? > There are alternations between r and s in different parts of the paradigm > of some Latin words (only verbs?), and these are apparent in borrowings > into English such as adhere vs. adhesive, adhesion. Now these alternations > are similar to those found as a result of the operation of Verner's Law in > words like English was - were, German war - gewesen, kiesen - kor and so > on. > Is there a link, or is this phenomenon typologically common? > Thank you, From mcv at wxs.nl Sat Oct 24 15:17:35 1998 From: mcv at wxs.nl (Miguel Carrasquer Vidal) Date: Sat, 24 Oct 1998 11:17:35 EDT Subject: r and s In-Reply-To: <4qA86Xe00WBM02EGE0@andrew.cmu.edu> Message-ID: ----------------------------Original message---------------------------- Paul J Hopper wrote: >On Sheila Watts' query about s/r alternations in Germanic and Latin >(so-called rhotacism): I can't think of other examples of it outside >these two groups There are the Sanskrit and Nuorese (Sardinian) sandhis, where -s > -r before a voiced consonant at the start of the following word. ======================= Miguel Carrasquer Vidal mcv at wxs.nl Amsterdam From Georg at home.ivm.de Sat Oct 24 21:17:22 1998 From: Georg at home.ivm.de (Ralf-Stefan Georg) Date: Sat, 24 Oct 1998 17:17:22 EDT Subject: r and s In-Reply-To: <363101ee.35251404@mail.wxs.nl> Message-ID: ----------------------------Original message---------------------------- At 11:17 Uhr -0400 24.10.1998, Miguel Carrasquer Vidal wrote: >----------------------------Original message---------------------------- >Paul J Hopper wrote: > >>On Sheila Watts' query about s/r alternations in Germanic and Latin >>(so-called rhotacism): I can't think of other examples of it outside >>these two groups > >There are the Sanskrit and Nuorese (Sardinian) sandhis, where -s > -r >before a voiced consonant at the start of the following word. And there is the Bolghar-Turkic Rhotacism, which changes Common Turkic *-z-, -z to -r-, -r in the languages of the Bolghar branch, viz. Volga Bolghar and Modern Chuvash. St.G. Stefan Georg Heerstrasse 7 D-53111 Bonn FRG +49-228-69-13-32 From mcv at wxs.nl Sun Oct 25 17:51:33 1998 From: mcv at wxs.nl (Miguel Carrasquer Vidal) Date: Sun, 25 Oct 1998 12:51:33 EST Subject: r and s In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Ralf-Stefan Georg wrote: >And there is the Bolghar-Turkic Rhotacism, which changes Common Turkic >*-z-, -z to -r-, -r in the languages of the Bolghar branch, viz. Volga >Bolghar and Modern Chuvash. Of course! But can we be sure it went from z to r, and not the other way around? ======================= Miguel Carrasquer Vidal mcv at wxs.nl Amsterdam From hubeyh at Montclair.edu Sun Oct 25 17:54:01 1998 From: hubeyh at Montclair.edu (H. M. Hubey) Date: Sun, 25 Oct 1998 12:54:01 EST Subject: r and s Message-ID: ----------------------------Original message---------------------------- Ralf-Stefan Georg wrote: > > > And there is the Bolghar-Turkic Rhotacism, which changes Common Turkic > *-z-, -z to -r-, -r in the languages of the Bolghar branch, viz. Volga > Bolghar and Modern Chuvash. > St.G. > Unless I am wrong the word for 5 in Chuvash is pilek (which to me sounds a lot like 'bilek' (wrist, hand) and which sounds to me as if it is older than Common Turkic 5 (besh). Secondly the weakness of the liquids in Turkic and Altaic (and Dravidian) is well known. It makes more common sense to believe that l~r Turkic (Chuvash) is more archaic and these sounds changed to sh~z because the eastern languages did not have the liquids, than to assume that it is a rhotacism. > > Stefan Georg > Heerstrasse 7 > D-53111 Bonn > FRG > +49-228-69-13-32 -- M. Hubey Email: hubeyh at Montclair.edu Backup:hubeyh at alpha.montclair.edu WWW Page: http://www.csam.montclair.edu/Faculty/Hubey.html From s_nickn at cassius.its.unimelb.EDU.AU Sun Oct 25 17:54:40 1998 From: s_nickn at cassius.its.unimelb.EDU.AU (Nick Nicholas) Date: Sun, 25 Oct 1998 12:54:40 EST Subject: TOC: History of Language 4.2 Message-ID: ----------------------------Original message---------------------------- The latest issue of _History of Language_ (formerly _Dhumbadji! Journal for the History of Language_) is now available, featuring: H.M. Hubey: Quantitative Approaches to Historical Linguistics with Example Application to *PIE/IE E.F.K. Koerner: On the Historiography of the Polish Contribution to the Understanding of Language Change See http://www.lexicon.net/opoudjis/Work/ahl.html for more information. -- ____.____1____.____2____.____3____.____4____.____5____.____6____.____7____. When in doubt, Nick Nicholas, Linguistics & Applied Linguistics, --- DOUBT! N.Nicholas at linguistics.unimelb.edu.au University of (- me) http://www.lexicon.net/opoudjis Melbourne, Australia From Georg at home.ivm.de Sun Oct 25 22:24:44 1998 From: Georg at home.ivm.de (Ralf-Stefan Georg) Date: Sun, 25 Oct 1998 17:24:44 EST Subject: r and s In-Reply-To: <363888cd.21120723@mail.wxs.nl> Message-ID: ----------------------------Original message---------------------------- >>And there is the Bolghar-Turkic Rhotacism, which changes Common Turkic >>*-z-, -z to -r-, -r in the languages of the Bolghar branch, viz. Volga >>Bolghar and Modern Chuvash. > >Of course! But can we be sure it went from z to r, and not the other >way around? I expected some sort of objections to this, and my answer has to be that we can*not* be sure. In fact, within Turkology, two positions exist, one taking the sibilant as original (the "Rhotacists") and the other - more at home with the proponents of the Altaic theory, but also supported by its chief critic, Gerhard Doerfer, that the Rhotic is (the "Zetacists"). While consensus is still to be waited for, it might be a safe approach to watch out for loan-words which entered either branch of Turkic *before* the particular sound change (-z-/-z > -r-/-r or the other way round) occurred. For the zetacistic position, this seems to be lacking (i.e. a clear example of an Iranian/Chinese/Tokharian, you name it) word with -r-/-r ending up zetacised in Common (= non-Chuvash) Turkic. For the rhotacistic position, there seem to be a few examples. Among those which I regard as most convincing is Chuvash /pir/ "linen" or the like, going back to Arabic baZZ and ultimately to Greek byssos (a Mediterranian-Oriental Wanderwort which really went places). I don't want to make this mootest of Turkological problems seem more easy than it is (it isn't), the bottomline should, however, be that both positions are defendable at the moment and that any account which states that the problem is "solved" once and for all, is therefore in error (hereby I explicitly complain about some pro-Altaistic works of the 90's which simply forget to tell their readers that the other position exists at all, or, in more belligerent passages, that it is a malevolent idee fixe of rabid anti-Altaicists; that this is not the case should be clear from Doerfer's position). And *even* within the framework of "zetacistic pro-Altaistics", it is of course commonplace that the input for Commot Tk. /z/ and Bolgh. /r/ cannot be simply /r/ (since Common Tk. *does* have /r/ in native words and morphemes), but something usually labeled as /r2/, /r'/ or the like, sometimes with the implication that it might have been something like Czech /R/ in DvoRak. Maybe. But isn't that multiplicare entia praeter necessitatem ? Rhotacism is able to make do with *one* proto-phoneme, viz. /z/, zetacism needs two /r, r'/. The whole /z-r/-mess is paralleled by a similar Lautgesetz, by which Common Tk. /sh/ and Bolgh.-Chuv. /l/ are connected. Again, the direction of the change is the object of often fierce discussions (between lambdacists and sigmatists, you guessed it), though good (and old) loanwords which might help to decide seem to be lacking even more (A. Rona-Tas presented some candidates in the corpus of words which he views as Tokharian loan-words in Proto-Turkic, but, although I personally happen to like some of them, I don't expect them to stand a real chance of ever getting generally accepted; it might surprise few at this stage that they seem to support lambdacism). And: the rhotacistic process has been called rather exceptional, not very wide-spread in this thread. I'm not so sure whether we've already found all prominent examples, I think there must be more. On the other hand, what is the evidence for r > z being more wide-spread" or, well, "natural" ? What parallels can be found ? For the general subject of this thread the conclusion is: *part* of the Turkological community views the Chuvash-Common-Turkic /r,z/ relation as just another example of Rhotacism. St.G. says: right they are, though the opposite opinion is never far away. That's life in Altaic studies. St.G. Stefan Georg Heerstrasse 7 D-53111 Bonn FRG +49-228-69-13-32 From mcv at wxs.nl Sun Oct 25 22:25:03 1998 From: mcv at wxs.nl (Miguel Carrasquer Vidal) Date: Sun, 25 Oct 1998 17:25:03 EST Subject: r and s In-Reply-To: <36326DD3.7F6D84F@Montclair.edu> Message-ID: ----------------------------Original message---------------------------- "H. M. Hubey" wrote: >Unless I am wrong the word for 5 in Chuvash is pilek (which to me sounds >a lot like 'bilek' (wrist, hand) and which sounds to me as if it is >older than Common Turkic 5 (besh). Secondly the weakness of the liquids >in Turkic and Altaic (and Dravidian) is well known. It makes more common >sense to believe that l~r Turkic (Chuvash) is more archaic and these >sounds changed to sh~z because the eastern languages did not have the >liquids, than to assume that it is a rhotacism. But the "eastern" languages do have liquids. The question is whether Proto-Turkic (cq. Proto-Altaic) had two of each (*r > r/r; *r1 > r/z; *l > l/l; *l1 > s^/l) or only one. ======================= Miguel Carrasquer Vidal mcv at wxs.nl Amsterdam From mcv at wxs.nl Mon Oct 26 10:53:46 1998 From: mcv at wxs.nl (Miguel Carrasquer Vidal) Date: Mon, 26 Oct 1998 05:53:46 EST Subject: r and s In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Ralf-Stefan Georg wrote: >And: the rhotacistic process has been called rather exceptional, not very >wide-spread in this thread. I'm not so sure whether we've already found all >prominent examples, I think there must be more. There sure must be, although for some reason I can only come up with syllable/word final examples. Armenian: *-is and *-us > -r in certain cases (erkir < *dwis, erir < *tris, u-stem nouns and adjectives like barjr, asr, t`anjr etc.) Basque: sl > rl, e.g. irla "island" < Spa. isla; erle "bee" < *ezle, cf. ezti "honey", ezko "wax". [NB. Basque is (laminal) /s/, not /z/!; -le is an agentive suffix] Hausa: the ind. obj. pronoun "to him" has a variant ma/\r (< *mas) [the /, \ and /\ are high, low and falling (high->low) tones]. ======================= Miguel Carrasquer Vidal mcv at wxs.nl Amsterdam From mcv at wxs.nl Mon Oct 26 10:54:10 1998 From: mcv at wxs.nl (Miguel Carrasquer Vidal) Date: Mon, 26 Oct 1998 05:54:10 EST Subject: r and s In-Reply-To: <3633B3EA.B6D8AC09@Montclair.edu> Message-ID: ----------------------------Original message---------------------------- "H. M. Hubey" wrote: >Neither Turkic nor Dravidian have words beginning with the liquids. I don't think initial position was mentioned previously (and I believe the Turkic z ~ r alternations occur everywhere but initially). As to r-, it is also practically absent from, say, IE and Basque in the West. You may have a case for l- (not extremely common in IE or Basque, but hardly absent), but note the "compensatory" paucity of initial n- in the West (both in IE and Basque [also Kartvelian?]). ======================= Miguel Carrasquer Vidal mcv at wxs.nl Amsterdam From hubeyh at Montclair.edu Mon Oct 26 10:54:39 1998 From: hubeyh at Montclair.edu (H. M. Hubey) Date: Mon, 26 Oct 1998 05:54:39 EST Subject: r and s Message-ID: ----------------------------Original message---------------------------- Miguel Carrasquer Vidal wrote: > ----------------------------Original message---------------------------- > "H. M. Hubey" wrote: > > >Unless I am wrong the word for 5 in Chuvash is pilek (which to me sounds > >a lot like 'bilek' (wrist, hand) and which sounds to me as if it is > >older than Common Turkic 5 (besh). Secondly the weakness of the liquids > >in Turkic and Altaic (and Dravidian) is well known. It makes more common > >sense to believe that l~r Turkic (Chuvash) is more archaic and these > >sounds changed to sh~z because the eastern languages did not have the > >liquids, than to assume that it is a rhotacism. > > But the "eastern" languages do have liquids. The question is whether > Proto-Turkic (cq. Proto-Altaic) had two of each (*r > r/r; *r1 > r/z; > *l > l/l; *l1 > s^/l) or only one. > Neither Turkic nor Dravidian have words beginning with the liquids. KOrean (or Japanese) has only a single liquid and [l], and [r] merely allophonic realizations. As can be seen the liquids become less and less used from west to east. The gradient runs east to west. So if anything the easterners substituted other sounds for the liquids from the west and managed to pick up a single liquid along the way (in the far west). The state of the language in the ME before the AA and IE spread is similar; lots of confusion of l and r and especially in the beginning of words (see von Soden). > > ======================= > Miguel Carrasquer Vidal > mcv at wxs.nl > Amsterdam -- M. Hubey Email: hubeyh at Montclair.edu Backup:hubeyh at alpha.montclair.edu WWW Page: http://www.csam.montclair.edu/Faculty/Hubey.html From hubeyh at montclair.edu Mon Oct 26 10:58:15 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Mon, 26 Oct 1998 05:58:15 EST Subject: r and s Message-ID: ----------------------------Original message---------------------------- Ralf-Stefan Georg wrote: > > ----------------------------Original message---------------------------- > > >>And there is the Bolghar-Turkic Rhotacism, which changes Common Turkic > >>*-z-, -z to -r-, -r in the languages of the Bolghar branch, viz. Volga > >>Bolghar and Modern Chuvash. > > > >Of course! But can we be sure it went from z to r, and not the other > >way around? > > I expected some sort of objections to this, and my answer has to be that we > can*not* be sure. In fact, within Turkology, two positions exist, one > taking the sibilant as original (the "Rhotacists") and the other - more at > home with the proponents of the Altaic theory, but also supported by its > chief critic, Gerhard Doerfer, that the Rhotic is (the "Zetacists"). We are never sure of anything, even the sun rising tomorrow. That is the famous induction problem of Hume. But by this time we should all be reasonably used to working under uncertainty. There are reasons why it is not necessary to posit rhotacization or lambdacization, or zetacization etc etc. > While consensus is still to be waited for, it might be a safe approach to > watch out for loan-words which entered either branch of Turkic *before* the > particular sound change (-z-/-z > -r-/-r or the other way round) occurred. > For the zetacistic position, this seems to be lacking (i.e. a clear example > of an Iranian/Chinese/Tokharian, you name it) word with -r-/-r ending up > zetacised in Common (= non-Chuvash) Turkic. For the rhotacistic position, > there seem to be a few examples. Among those which I regard as most > convincing is Chuvash /pir/ "linen" or the like, going back to Arabic baZZ > and ultimately to Greek byssos (a Mediterranian-Oriental Wanderwort which > really went places). I don't want to make this mootest of Turkological > problems seem more easy than it is (it isn't), the bottomline should, > however, be that both positions are defendable at the moment and that any > account which states that the problem is "solved" once and for all, is > therefore in error (hereby I explicitly complain about some pro-Altaistic > works of the 90's which simply forget to tell their readers that the other > position exists at all, or, in more belligerent passages, that it is a > malevolent idee fixe of rabid anti-Altaicists; that this is not the case > should be clear from Doerfer's position). Since the present de facto standard (which should be called the zeroth order approximation) does not take into consideration anything other than a model of linguistic descent which is biologically similar to asexual descent which is similar to tracing the mtDNA, many things cannot be explained without going into a more realistic position. I do not believe in this model of constant divergence. So it is just as easy to believe that some people whose language was rich in the liquids moved west to east and the peoples of the east got an incomplete dosage of the liquids. It seems that most of the Altaic arguments are about whether there really was a state of some language which via constant divergence gave rise to Japanese, Korean, Tungus, Turkic, etc. It is not necessary for such things to occur in order for these languages to form a family. Semitic and IE were formed from a mixture of some ancient languages and it is patently clear in the case of Semitic that it got "frozen" into some semi-regular or quasi-regular state which can be seen in the binyanim. The verb gradations of Hittite are exactly the same kind of evidence. They belong to the same type of phenomena as that of the irregular verbs in English. It is easy for some people to see layer upon layer in Japanese, such as proto-Korean, or proto-Altaic over Austronasian etc but I can see the layering in English (i.e. Latin layer over a Germanic layer over a non-IE layer, mixed in with some Celtic) but the IEnists cannot see these layers. > And *even* within the framework of "zetacistic pro-Altaistics", it is of > course commonplace that the input for Commot Tk. /z/ and Bolgh. /r/ cannot > be simply /r/ (since Common Tk. *does* have /r/ in native words and > morphemes), but something usually labeled as /r2/, /r'/ or the like, > sometimes with the implication that it might have been something like Czech > /R/ in DvoRak. Maybe. But isn't that multiplicare entia praeter > necessitatem ? Rhotacism is able to make do with *one* proto-phoneme, viz. > /z/, zetacism needs two /r, r'/. > The whole /z-r/-mess is paralleled by a similar Lautgesetz, by which > Common Tk. /sh/ and Bolgh.-Chuv. /l/ are connected. Again, the direction of > the change is the object of often fierce discussions (between lambdacists > and sigmatists, you guessed it), though good (and old) loanwords which > might help to decide seem to be lacking even more (A. Rona-Tas presented > some candidates in the corpus of words which he views as Tokharian > loan-words in Proto-Turkic, but, although I personally happen to like some > of them, I don't expect them to stand a real chance of ever getting > generally accepted; it might surprise few at this stage that they seem to > support lambdacism). Goethe said a long time ago that "When an idea is lacking a word can always be found to take its place." Here is something that could have happened. A superstratum P dominates over a substratum B. At first only the words of P gets into writing. Since they don't have TV and forced education for 12 years the B people can't learn this language too well, but over a period of time start to manage. The offspring of P mix with B over time. Because of the large numbers of B, eventually the offspring of P also start to speak their ancestors' language like B people. By this time the B also make it to the top (mixing) and the language (written of course) suddenly starts to display some strange characteristics. These characteristics, of course, are the imcomplete learning of P by B people. Almost as if they did it on purpose things like rhotacization, lambdacization, zetacization, voicing etc appear out of nowhere in the language (as evidenced by writing) centuries and millennia later when our intrepid linguists start to decipher these codes. Voila! Changes appear out of nowhere and some nice linguist decides to become famous by making up a name for the process. That is how these things are born. That is much more believable than turning a complex reality into a comical textbook version only to feed the incomplete theories so that they can linger on and on and on....instead of letting them go the way of the dinosaurs. Making up a name is nice but is it necessary? What does it explain? So then what is its value other than holding up a toy model of language change? > And: the rhotacistic process has been called rather exceptional, not very > wide-spread in this thread. I'm not so sure whether we've already found all > prominent examples, I think there must be more. On the other hand, what is > the evidence for r > z being more wide-spread" or, well, "natural" ? What > parallels can be found ? If there are apparently such changes it probably means that it arose out of some kind of an interaction of peoples speaking different languages. What all this means is that instead of using principle A (which says always attribute everything to "natural" change -- what on earth does natural mean?) create a new principle; one that says (no effects without causes). If there is some very good reason to suspect that for some reason people find it easier to create /r/ than /z/ then we should read about it. Ditto for lambdacization. > For the general subject of this thread the conclusion is: *part* of the > Turkological community views the Chuvash-Common-Turkic /r,z/ relation as > just another example of Rhotacism. St.G. says: right they are, though the > opposite opinion is never far away. That's life in Altaic studies. If this rhotacization seems to happen a lot in the West, maybe there is a good reason for it. Ooops, there it goes again; another name invented already called 'areal' phenomena. Could this not simply be due to the fact that the indigeneous inhabitants of some region who spoke related languages got invaded from all over the place by people speaking similar languages and reacted to these new languages similarly simply because their phoneme/phone repertoire was only capable of producing some sounds and not others? What other reason could there be? Certainly nothing the air, the water or the mountains! > > St.G. > > Stefan Georg > Heerstrasse 7 > D-53111 Bonn > FRG > +49-228-69-13-32 -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From l.campbell at ling.canterbury.ac.nz Mon Oct 26 10:59:58 1998 From: l.campbell at ling.canterbury.ac.nz (Lyle Campbell) Date: Mon, 26 Oct 1998 05:59:58 EST Subject: r and s in Turkic Message-ID: ----------------------------Original message---------------------------- In view of the several messages on the problem of Turkic sound correspondences involving r (and liquids) and s (and sibilants), I'll throw into the mix a small bit about it taken from a recent paper of mine evaluating Dolgopolsky's Nostratic. Since this bit starts with one of Dolgopolsky's specific proposed Nostratic forms, the content relevant to the discussion of the Turkic problem becomes clear only towards the end of it all. (I claim no first-hand knowledge of the topic, and attribute most of the significant content to Juha Janhunen, except for any mistakes I may have made -- sorry for any diacritics which do not come across). Lyle Campbell [48] *p'oK'ü 'wild cattle, pack' (Indo-European *pek'u / *pek'we- 'cattle'; Altaic *p'ok'ür'- 'bovine animal, bull'). This set clearly involves borrowing. The Altaic *p'ok'ür'-, represented only by Turkic, is a clear example of a documented loan, involving one of the strong points among the arguments of those who oppose the Altaic hypothesis. Proto-Turkic *s split into s and z, and *S ("sh") into S and Z ("zh") in specific environments (involving roots of two syllables and with long vowels), and then in the highly influential Bulgaric (Chuvash) branch of the family z > r, and Z > l. As a result, words in Mongolian (and Tungusic) which have an r or l corresponding to s, z, S, or Z in other Turkic languages can only be borrowings from this branch of Turkic, not true cognates to other Altaic languages (or they are accidental similarities). There is a sizeable number of these in the Dolgopolsky's putative Altaic lexical comparisons. In this case, in set [48], the word involved is Proto-Turkic *pöküs 'bovine', borrowed from Bulgaric into Mongolian and from there on into Tungusic (Janhunen 1996a:240-1, 255). This set would be questionable in any case, given the important role of cattle in the prehistoric cultures from the area of the Proto-Indo-European homeland and in the territory of the various so-called Altaic languages. I should add here that Starostin and Dolgopolsky (in discussion in the symposium) disputed this interpretation of the Turkic facts, preferring reconstructions of Proto-Turkic which reflect the liquids rather than the sibilants and in this way they deny that borrowing is a problem for these "Altaic" forms. This interpretation would require assuming that the liquids (l/r) were original and changed to sibilants in certain of the Turkic languages, a kind of sound change seldom seen in the world's languages, though changes in the other direction are common (as in rhotacism). There is considerable literature precisely on this topic. Among Turkologists, those who believe in the Altaic hypothesis (as well as Doerfer, who opposes Altaic, though he holds Mongolian forms in these comparisons to be Turkic loans) postulate original liquids (which then would make the sibilants of other Turkic languages the results of later sound changes); those who oppose the Altaic hypothesis (with the exception of Doerfer) hold the sibilants to be original (which makes the liquids the results of later sound changes). Put differently, those with a grasp of phonological systems and phonetic plausibility all postulate a change of *s > z > r and *S > Z > l, where the steps in the change are seen as incremental, intimately interrelated, and natural. No one with a sense of phonology postulates the reverse, the unnatural and implausible changes of *r > z/s and *l > Z/S, which are almost unknown in languages elsewhere (Shherbak 1986b, Janhunen personal communication). [[From: Campbell, Lyle. In Press. Nostratic and Linguistic Palaeontology in Methodological Perspective. In: Nostratic: Evaluating a Linguistic Macrofamily, ed. by Colin Renfrew and Daniel Nettle. Cambridge: The McDonald Institute for Archaeological Research.]] From hubeyh at montclair.edu Mon Oct 26 11:00:58 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Mon, 26 Oct 1998 06:00:58 EST Subject: r and s Message-ID: ----------------------------Original message---------------------------- Miguel Carrasquer Vidal wrote: > > "H. M. Hubey" wrote: > > >Neither Turkic nor Dravidian have words beginning with the liquids. > > I don't think initial position was mentioned previously (and I > believe the Turkic z ~ r alternations occur everywhere but > initially). The initial r and l are treated almost as if they were consonant clusters, except in Turkish into which they have settled. > As to r-, it is also practically absent from, say, IE and Basque in > the West. You may have a case for l- (not extremely common in IE or > Basque, but hardly absent), but note the "compensatory" paucity of > initial n- in the West (both in IE and Basque [also Kartvelian?]). Lahovary has a huge list of words re: Basque, Etruscan, Sumerian, Caucasian languages, Nilo-Saharan and AA and IE. He is making a good case for a related set of languages being spread around the Mediterranean thru India (Dravidian). I had noticed some of these strange things but Lahovary has done a much better job. It is now easier to find links amongst Dravidian, Turkic and Sumerian (and probably Basque) because of that. Chuvash seems to figure more prominently than the other Turkic languages. At least that is my present pre-impression. > ======================= > Miguel Carrasquer Vidal > mcv at wxs.nl > Amsterdam -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From Georg at home.ivm.de Mon Oct 26 11:46:36 1998 From: Georg at home.ivm.de (Ralf-Stefan Georg) Date: Mon, 26 Oct 1998 06:46:36 EST Subject: r and s In-Reply-To: <363101ee.35251404@mail.wxs.nl> Message-ID: ----------------------------Original message---------------------------- Regarding the question how common rhotacistic phenomena are, here are some further examples from Greek dialects: - in intervocalic position -s- gets rhotacized to -r- in the Eretria and Oropos local varieties of Ionic (pairin = paisin) - only in Auslaut position rhotacism is found in later Laconian, which is (partly) retained in modern Tsakonian - and in later Elian (Dior = Dios, tir = tis etc.) Othe examples or near-examples incude: In the Mongolian language Dagur several stop consonants (b, G, g, d) + s are changed > r in syllable final position. In Tundra Yukaghir, under certain assimilatory conditions including secondary intervocalic position in compounds, initial s- is rhotacized > r-. >From East Caucasian languages: the Akush and Urakh dialects of Dargwa seem to respond with -r'- (r + glotal stop) and -r to a proto-Dargwa input -z-/-z. In Tibetan dialects, in some positions (always involving an initial sC- cluster) s- may yield to r-: Written Tib. sna "nose" > Panakha (Banag, a NE dialect) rna, WT sku "body" > Panakha rku, WT stag "tiger" > Golok rtag. But this process depends strongly on the following C, instances of rC- > sC- are also found. So far, Latin rhotacism has been mentioned, but not the similar phenomenon in Umbrian (but not Oscan), which, however, might be seen in connection with the Latin process (though the precise nature of this connection, i.e. the chronology, is unclear to me). St.G. Stefan Georg Heerstrasse 7 D-53111 Bonn FRG +49-228-69-13-32 From hubeyh at montclair.edu Mon Oct 26 17:01:46 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Mon, 26 Oct 1998 12:01:46 EST Subject: r and s in Turkic Message-ID: ----------------------------Original message---------------------------- Lyle Campbell wrote: > > > results of later sound changes). Put differently, those with a grasp of > phonological systems and phonetic plausibility all postulate a change of *s > > z > r and *S > Z > l, where the steps in the change are seen as > incremental, intimately interrelated, and natural. No one with a sense of > phonology postulates the reverse, the unnatural and implausible changes of > *r > z/s and *l > Z/S, which are almost unknown in languages elsewhere > (Shherbak 1986b, Janhunen personal communication). To put it mildly differently, as can be seen everywhere in phonology and phonetics the sounds /ptksn/ are "rarely absent" from the world's languages. If the "natural" change to /r/ and/or /l/ was so natural we would have had no /s/ left anywhere. Perhaps those who are pushing rhotacization on us have latched on to some local phenomena (local in both time and space) and think it is universally applicable or rhotacization does not exist as it is said to exist but is a substratum/superstratum phenomena and is once again a local phenomena and not a global one. OTOH, if you are looking at a mixing of a language A with strong consonant-clusters and rich in fricatives and sibilants with another language B which is not clustered and is weak in fricatives perhaps for a period of time, (local in space and time) you could obtain the (in)famous rhotacization, lambdacization etc. PS. REad Lindblom's works on phonetics and the physics and/or optimization of sound patterns. > > [[From: Campbell, Lyle. In Press. Nostratic and Linguistic Palaeontology > in Methodological Perspective. In: Nostratic: Evaluating a Linguistic > Macrofamily, ed. by Colin Renfrew and Daniel Nettle. Cambridge: The > McDonald Institute for Archaeological Research.]] -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From tuitekj at ERE.UMontreal.CA Mon Oct 26 16:59:34 1998 From: tuitekj at ERE.UMontreal.CA (Kevin Tuite) Date: Mon, 26 Oct 1998 11:59:34 EST Subject: r and s yet again In-Reply-To: Message-ID: ----------------------------Original message---------------------------- In a recent posting, Ralf-Stefan Georg asked "what is the evidence for r > z being more wide-spread" or, well, "natural" ? What parallels can be found ?" Phonetic changes similar in some respects to the shift r > z postulated by some Turkologists have been attested. One of them is an interesting chapter in the sociolinguistic history 13th-16th century French, of a sound change that subsequently was beaten back. One of the few traces remaining in the standard language of this event is the word "chaise", which I discuss in a forthcoming article (see excerpt below). Another potentially useful case is the women's pronunciation of Chukchee, as described by W. Bogoras in the Handbook of American Indian Languages (sic): "Women generally substitute /sh/ for /ch/ and /r/". It is, I believe, important to note that both these sound shifts (1) uvulaire (not alveolar) /R/ becomes a fricative; (2) are sociolinguistically sensitive. Any ideas why? Kevin Tuite ******** (from "Au-delà du Stammbaum : Théories modernes du changement linguistique"): Le mot 'chaise' remonte, comme l'indiquent les dictionnaires, au latin 'cathedra', lui-même emprunté au grec. 'Cathedra' a subi une succession complexe de lénitions successives pour aboutir, plusieurs siècles plus tard, à 'chaire'. La transformation de la forme de ce lexème a été l'effet de lois phonétiques régulières (cf. le sort du lat. 'fraxinum' > 'frêne', ou de 'pectinum' > 'peigne'). C'est la toute dernière étape, [Se:R] > [Se:z], qui nous pose problème. Si la transformation de la dernière consonne de 'chaise' représentait le résultat d'une loi phonétique, comment doit-on expliquer la prononciation actuelle de 'père', 'mère', 'arrière' et de celle de dizaines d'autres mots présentant le même contexte phonologique (y compris les homophones 'chair' et 'chaire')? Heureusement, l'histoire nous a conservé quelques témoignages pertinents. Déjà au 13ème siècle, des documents provenant des régions centrale et sud-est de la zone du français proprement dit (la "langue d'oïl") indiquaient "une tendance à l'assibilation de r intervocalique"; deux siècles plus tard, la même tendance a fait son apparition "dans le parler vulgaire de la capitale", où elle a même attiré l'attention d'Érasme en 1518 "Idem faciunt hodie mulierculae parisinae pro Maria sonantes Masia, pro ma mere, ma mese" (Fouché 1961: 603). Paradoxalement, un deuxième observateur à Paris, quinze ans après Érasme, a pris note de la prononciation de courin pour 'cousin' - cette fois, c'est le [z] qui a cédé sa place à l'[R]. Encore plus paradoxalement, toutes les deux substitutions, l'une exactement à l'inverse de l'autre, se notent dans une expression relevée en 1529 à Bourges: Jerus Masia! (Meyer-Lübke 1974: 408). Un siècle plus tard, un écrivain parlait au passé de cette vague de substitutions de phonèmes qui avait frappé la langue de Villon: "Nos Parisiens mettoient autrefois (mais cela ne se fait plus ou c'est fort rarement et seulement parmi le menu peuple) une s au lieu d'une r et une r au lieu d'une s" (loc. cit.) Que s'est-il passé? Le post mortem offert par W. von Wartburg (1988: 156) est instructif; à son avis, il s'agissait "d'un mouvement avorté, parce que rejeté par les classes supérieures". Rapidement après que cette "contagion" linguistique eut gagné la capitale, "les classes supérieures résistèrent, elles n'acceptèrent pas ce changement venu d'en bas. Cette réaction eut même pour conséquence que les gens du peuple ne savaient plus quand il fallait prononcer r. Par peur de s'exposer à des critiques on se mit à remplacer par r même les z qui étaient justifiés étymologiquement. On trouve donc des formes comme rairon et courin. Mais en général le rétablissement se fit correctement. On sait qu'un doublet est resté: chaire - chaise". En lisant cette histoire trois décennies après la publication des premiers ouvrages de W. Labov, on a le sentiment de déjà vu. Les phénomènes qui, selon la description de von Wartburg, accompagnait cette guerre entre "l'inconscient" et "la raison" portent aujourd'hui des noms qui font désormais partie du vocabulaire technique de chaque linguiste (et dont l'un, au moins, aurait sonné familier au romaniste allemand): le "changement par le bas" et l'"hypercorrection" ************************************************************** Kevin Tuite 514-343-6514 (bureau) Département d'anthropologie 514-343-2494 (télécopieur) Université de Montréal C.P. 6128, succursale centre-ville Montréal, Québec H3C 3J7 tuitekj at mistral.ere.umontreal.ca ************************************************************** From mcv at wxs.nl Mon Oct 26 23:17:42 1998 From: mcv at wxs.nl (Miguel Carrasquer Vidal) Date: Mon, 26 Oct 1998 18:17:42 EST Subject: r and s in Turkic In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Lyle Campbell wrote: >No one with a sense of >phonology postulates the reverse, the unnatural and implausible changes of >*r > z/s and *l > Z/S, which are almost unknown in languages elsewhere Whatever happened in Turkic, and I have no opinion, what keeps the debate going (apart from the lack of hard evidence either way, despite Stefan's pir < baZZ) would seem to be that both changes are equally likely. The Altaicist (and Nostraticist) argument is of course not based on plain *r and *l, but on palatalized/fricative *r^(> z) and *l^ (> s^), cf. Polish and Hebrew, respectively. ======================= Miguel Carrasquer Vidal mcv at wxs.nl Amsterdam From DISTERH at UNIVSCVM.SC.EDU Tue Oct 27 20:00:55 1998 From: DISTERH at UNIVSCVM.SC.EDU (Dorothy Disterheft) Date: Tue, 27 Oct 1998 15:00:55 EST Subject: rhotacism from Ray Hickey Message-ID: On the reverse of rhotacism and other *unlikely* changes ======================================================== The present discussion of rhotacism and its possible reverse has been centered on Altaic and the Germanic/Latin data so that perhaps the data has prevented us from making some general points. Taking Lyle's suggestion that the reversal of rhotacism is very unlikely as a starting point: the issue to home in on in such changes is the SECONDARY ARTICULATION of the segments involved. A fricative /r/ (apical trill or continuant) can easily become a full sibilant, i.e. re-align itself phonologically as /z/, as has happened in Polish with the Slavic PRE-prefix and as seen in Czech in the trilled /r/ (evident e.g. in _Dvorak_, the composer's name and indicated by a superscript hacek). It also happens sub-phonemically in Stockholm Swedish and forms of Western Irish when /r/ is phonetically palatalised and where the raising of the apex to the palate leads to assibiliation. I might push this point and suggest that secondary articulation is the natural bridge leading across a phonological divide to a phoneme on the other bank, so to speak. The classic case is velarised /l/ [l-] to /u/ which is hardly worth commenting on, it is so common: Polish, colloquial forms of southern British English, Brazilian Portuguese; historically: southern English in general, French, etc. The secondary articulation becomes primary, the original primary articulation is dropped and bob's your uncle. Ray Hickey English Linguistics Essen University Germany r.hickey at uni-essen.de From jhewson at morgan.ucs.mun.ca Tue Oct 27 18:02:47 1998 From: jhewson at morgan.ucs.mun.ca (John Hewson) Date: Tue, 27 Oct 1998 13:02:47 EST Subject: r and s In-Reply-To: Message-ID: ----------------------------Original message---------------------------- In the flurry of comments on rhotacism, Sheila Watts original question seems to have been lost: is the s/r alternation in Latin attributable to something like Verner's Law? Paul Hopper did comment that the change in Latin affected all instances of intervocalic /s/, and that leaves us with Sheila's question as to why haereo/haesi and English adhere/adhesive? The answer is that intervocalic /s/ in Classical Latin comes from original clusters, mostly /ss/, /ts/, /st/, and so forth. Latin haesi is obviously a sigmatic perfect (Latin merged the perfect and the sigmatic aorist, leaving remnants of both morphologies), similar to scribo/scripsi, nubo/nupsi, rideo/risi Message-ID: ----------------------------Original message---------------------------- >Whatever happened in Turkic, and I have no opinion, what keeps the >debate going (apart from the lack of hard evidence either way, >despite Stefan's pir < baZZ) first I'd like to make a minor correction of "my" (of course, neither the word, nor the etymology belong to me, though I'd not be against it ;-) baZZ, which should have been given as /bazz/. And there are some more quite likely LWs, which bespeak rhotacism and might amount to something in the way of "hard evidence": Chuv. /hIr/ 'pinus abies', most probably from some Volgaic or Permic source close to Mordva /kuz/, Mari /koZ/, Komi /koz/, or Udmurt /köz/, which in turn are deeply rooted in FU (fi. kuusi, chanty kool, man'si xavt etc.) and possibly U (nenc. xaadI). Chuv. tAvAr 'narrow' goes together with CTk. *tIqIz 'id.', *but*: the non-change of /t/ > /ch/ in front of /I/ is understandable (only, I'd say), if the Chuv. word is a Kipchak loan (i.e. it entered Bolghar after t>ch/_I and before z>r. The other changes are the expected ones. Chuv. tir- 'to arrange in rows', Ctk. tiz- 'id.', same reason for likelihood of borrowing from another Tk. language. There is further the possibility that some Chuvash personal names contain Persian /niya:z/ 'request, prayer othl.' in the form /-never/, but this is admittedly not too strong. These are not hundreds of examples, but I think rhotacism is alive and kicking, and I do think that it will be possible to seperate Chuv. /pir/ from Ar. bazz and Gk. byssos (the b>p, a>i changes are all regular and expected, the meanings are identical, and byssos is one of the most successful Wanderwörter of Mediterranian origin in the whole of Asia ( matched only by diphthera and nomos, possibly); it would be bewildering *not* to find it in Chuvash somehow (and hardly imaginable that it would end up as anything but /pir/ there) ! On the putative palatalized l2 and r2 of proto-Turkic (and 'Altaic'), we should not forget that these values are posited *in order to* allow for a host of words being Tu.-Mo. etc. cognates rather than being borrowings, in which case proto-Turkic simply would not need those phonemes. Id multiplicatio entiorum praeter ullam necessitatem mihi esse videtur ! St.G. Stefan Georg Heerstrasse 7 D-53111 Bonn FRG +49-228-69-13-32 From vovin at hawaii.edu Tue Oct 27 14:17:36 1998 From: vovin at hawaii.edu (Alexander Vovin) Date: Tue, 27 Oct 1998 09:17:36 EST Subject: r and s in Turkic In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Dear Lyle: Let me throw my two cents into the discussion, as it seems to me you might oversimplify certain issues. (1) For the Altaic theory, it really does not matter whether one reconstructs *z and *sh or *r2 or *l2 (note that *none* of the Altaicists reconstructs just plain *r and *l here, as it may seem from you posting: Starostin, for example, following Ramstedt, reconstructs palatalized *r' and *l'). I, for one, am not convinced that the protolanguage had liquids, and I though I usually write *r2 and *l2, it is rather due to tradition, than to an assumption that these are actual liquids. You are certainly right in pointing out that r > z is less typical than z > r. But I fail to see how recognition of proto-Turkic *z rather than *r2 is going to "disprove" Altaic. It still can correspond to -r- in other languages, right, and our cognates do not have to be complete look-alikes, right? (2) I don't think any of turkologists except Shcherbak nowadays support his idea that *z < *s. There are too many exceptions to the "rules" of *s > *z that you list, thus, e.g. *ka:z (Chuvash xur) 'goose' does have a long vowel, but *k"uz (Chuvash ker) does not. Etc., etc. (3) The theory of wholesale borrowing from Bulgaric to proto-Mongolic faces many obstacles (some of them unsurmountable in my opinion). First, it is cyclic by its logic: Turkic "loans" in Mongolic have /r/, therefore they must be Bulgar loans. But there should be, a second independent evidence for the fact that these words are indeed Bulgaric loans, apart from the the /r/ itself. As a matter of fact, this evidence does not exist. Let me demonstrate it by using the same word for "ox". (a) First, while there is an initial h- in Middle Mongolian form, which allows us to reconstruct pre-Mongolic *p-, there is no evidence for reconstruction of proto-Turkic *p"ok"uz with initial *p-. The only Turkic language that has initial h- that can be claimed to represent an earlier *p- is Khaladj, but as far as I know the word is not attested in Khaladj (or not recorded) (But not eveyone accepts even Khaladj h- < *p-). There is an initial h- in Uzbek in this word, but it is prothetic, as it doesc not prtesent a regular reflex of *p- (contrary to Doerfer's claim), e.g. does not show up in such word as ar "man", where we should expect it if Uzbek h- were regular. Thus, we end up with Turkic *"ok"uz. (b) The only actual Bulgar form that we know is Chuvash vAxAr 'ox' (If I rememnber correctly, the word is not attested in Bulgar inscriptions), and it *does* represent a series of problems. While initial prothetic v- is in all probability a late development (it looks like there is no v- in Bulgar inscriptions, or at least it is difficult to tell), there is no way to claim that Chuvash vocalism (very different from Common Turkic, and even today not satisfactorily explained) is a late development, too. Thus, what basis do we have to say that MM h"ok"ur is a Bulgar loan? It appears that none, except the final -r-. (c) summing up, the hypothesis that Mongolic word is a loan from Bulgar s based on three (sic!) unproven hypotheses: that Bulgaric had *h- (no traces of it in any of Bulgaric languages) (1), that Chuvash vocalism is secondary as compared to Common Turkic (2), and that it represents a late development (3). Thus, the loanword explanation, a hypothesis itself, is relying on three other hypotheses to be true. It is much more complex solution than a single hypothesis that words in question represent cognates. Finally, the wholesale loanword scenario is based on one more hypothesis, again not proven. It presumes that Mongols were in contact of all Turks with Bulgars. But Bulgars are... the westernmost branch of Turkic, and there is a zero independent evidence that they ever were in touch with Mongols (until 13th c.), who came to the territory of Mongolia from the territory to the east. The only evidence presented for this brave claim is again Bulgar "loanwords" in Mongolic with -r. Circularity again, isn't it? Sasha ======================================= Alexander Vovin Associate Professor of Japanese Department of East Asian Languages and Literatures 382 Moore Hall 1890 East-West Road University of Hawaii at Manoa Honolulu, HI 96822 vovin at hawaii.edu fax (808)956-9515 (o.) t.(808)956-6881 (o.) On Mon, 26 Oct 1998, Lyle Campbell wrote: > ----------------------------Original message---------------------------- > In view of the several messages on the problem of Turkic sound > correspondences involving r (and liquids) and s (and sibilants), I'll throw > into the mix a small bit about it taken from a recent paper of mine > evaluating Dolgopolsky's Nostratic. Since this bit starts with one of > Dolgopolsky's specific proposed Nostratic forms, the content relevant to > the discussion of the Turkic problem becomes clear only towards the end of > it all. (I claim no first-hand knowledge of the topic, and attribute most > of the significant content to Juha Janhunen, except for any mistakes I may > have made -- sorry for any diacritics which do not come across). > Lyle Campbell > > [48] *p'oK'� 'wild cattle, pack' (Indo-European *pek'u / *pek'we- > 'cattle'; Altaic *p'ok'�r'- 'bovine animal, bull'). This set clearly > involves borrowing. The Altaic *p'ok'�r'-, represented only by Turkic, is > a clear example of a documented loan, involving one of the strong points > among the arguments of those who oppose the Altaic hypothesis. > Proto-Turkic *s split into s and z, and *S ("sh") into S and Z ("zh") in > specific environments (involving roots of two syllables and with long > vowels), and then in the highly influential Bulgaric (Chuvash) branch of > the family z > r, and Z > l. As a result, words in Mongolian (and > Tungusic) which have an r or l corresponding to s, z, S, or Z in other > Turkic languages can only be borrowings from this branch of Turkic, not > true cognates to other Altaic languages (or they are accidental > similarities). There is a sizeable number of these in the Dolgopolsky's > putative Altaic lexical comparisons. In this case, in set [48], the word > involved is Proto-Turkic *p�k�s 'bovine', borrowed from Bulgaric into > Mongolian and from there on into Tungusic (Janhunen 1996a:240-1, 255). > This set would be questionable in any case, given the important role of > cattle in the prehistoric cultures from the area of the Proto-Indo-European > homeland and in the territory of the various so-called Altaic languages. > I should add here that Starostin and Dolgopolsky (in discussion in > the symposium) disputed this interpretation of the Turkic facts, preferring > reconstructions of Proto-Turkic which reflect the liquids rather than the > sibilants and in this way they deny that borrowing is a problem for these > "Altaic" forms. This interpretation would require assuming that the > liquids (l/r) were original and changed to sibilants in certain of the > Turkic languages, a kind of sound change seldom seen in the world's > languages, though changes in the other direction are common (as in > rhotacism). There is considerable literature precisely on this topic. > Among Turkologists, those who believe in the Altaic hypothesis (as well as > Doerfer, who opposes Altaic, though he holds Mongolian forms in these > comparisons to be Turkic loans) postulate original liquids (which then > would make the sibilants of other Turkic languages the results of later > sound changes); those who oppose the Altaic hypothesis (with the exception > of Doerfer) hold the sibilants to be original (which makes the liquids the > results of later sound changes). Put differently, those with a grasp of > phonological systems and phonetic plausibility all postulate a change of *s > > z > r and *S > Z > l, where the steps in the change are seen as > incremental, intimately interrelated, and natural. No one with a sense of > phonology postulates the reverse, the unnatural and implausible changes of > *r > z/s and *l > Z/S, which are almost unknown in languages elsewhere > (Shherbak 1986b, Janhunen personal communication). > > [[From: Campbell, Lyle. In Press. Nostratic and Linguistic Palaeontology > in Methodological Perspective. In: Nostratic: Evaluating a Linguistic > Macrofamily, ed. by Colin Renfrew and Daniel Nettle. Cambridge: The > McDonald Institute for Archaeological Research.]] > From manaster at umich.edu Tue Oct 27 14:16:25 1998 From: manaster at umich.edu (manaster at umich.edu) Date: Tue, 27 Oct 1998 09:16:25 EST Subject: r and s yet again In-Reply-To: Message-ID: ----------------------------Original message---------------------------- If I may, I'd like to say hi to Kevin, and say that as always great minds do think alike. I cite this very French sound change (I dont mean it is very French; I mean this very one) in a recent paper in JIES (along with an exmaple from Polish) to show that such a direction of change is indeed possible. The context may be of more general interest. A leading Russian linguist, Serebrennikov or Shcherbak I forget which, had attacked Nostratic among other things on the grounds that some of the sound changes from Nostratic to daughter languages posited by Illich-Svitych were impossible. This was one of them. Alexis MR On Mon, 26 Oct 1998, Kevin Tuite wrote: > ----------------------------Original message---------------------------- > > > In a recent posting, Ralf-Stefan Georg asked "what is the evidence for r > > z being more wide-spread" or, well, "natural" ? What parallels can be found > ?" Phonetic changes similar in some respects to the shift r > z postulated > by some Turkologists have been attested. One of them is an interesting > chapter in the sociolinguistic history 13th-16th century French, of a sound > change that subsequently was beaten back. [snip] From hubeyh at montclair.edu Tue Oct 27 14:15:46 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Tue, 27 Oct 1998 09:15:46 EST Subject: r and s yet again Message-ID: ----------------------------Original message---------------------------- Kevin Tuite wrote: > > > generally substitute /sh/ for /ch/ and /r/". It is, I believe, important to > note that both these sound shifts (1) uvulaire (not alveolar) /R/ becomes a > fricative; (2) are sociolinguistically sensitive. Any ideas why? Yes, it has to do with a lot of things. Basically, assuming that it is definitely not a strata problem, then it is the interaction of the phonotactics of the language with physical constraints. There is an easy way to visualize what these physical constraints are. Conceptually it is about inertia & acceleration effects mixed with a predetermined goal of creating a given set of sounds. Words like assimilation, metathesis, etc are used to describe these complex of maneuvers. If you want to see both an acoustic, physical and a perceptual view of these processes in a single 3D space (a higher dimensional space is really needed), you can do so for the cost of $8.95 at 1stBooks.com. Just enter my name 'Hubey' at the prompt and you can download and electronic copy of my book which hopefully will be published in paper in the west. For those who do not want to spend the money, a copy of a paper which discusses this can be found on my website. (It's something like "phonological spaces"). > Kevin Tuite > > -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From hubeyh at montclair.edu Wed Oct 28 12:44:18 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Wed, 28 Oct 1998 07:44:18 EST Subject: r and s in Turkic Message-ID: ----------------------------Original message---------------------------- Ralf-Stefan Georg wrote: > > from Ar. bazz and Gk. byssos (the b>p, a>i changes are all regular and > expected, the meanings are identical, and byssos is one of the most > successful Wanderwvrter of Mediterranian origin in the whole of Asia ( > matched only by diphthera and nomos, possibly); it would be bewildering > *not* to find it in Chuvash somehow (and hardly imaginable that it would > end up as anything but /pir/ there) ! Speaking about Wanderworter, I have been wondering about one myself. The word 'ak' in Turkic (these days meaning 'white' and 'to flow') seems to be one of these also. For example, aqua, eku (Hittite, to drink), ich (Turk. to drink), akar (that which flows) > watar?, aryk (water canal), and even shows up as lots of words with 'ar' root in them in Dravidian (Lahovary) and I think 'Mediterratean' as 'ar', and 'sar', etc. Is that possible? -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From hubeyh at montclair.edu Wed Oct 28 12:45:05 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Wed, 28 Oct 1998 07:45:05 EST Subject: uc and ucak Message-ID: ----------------------------Original message---------------------------- I am going to ask something that sounds very strange. I noticed accidentally that a word like 'uchakka' (that is reasonably close) meant 'to fly' in Sanskrit. (I hope that was Sanskrit, and not Indic or Hindi). In any case, it is very strange for me to see this since 'uch' means 'to fly' in Turkic. The choices are 1. Accident 2. There is something we are missing 2.a) Uch was borrowed into Turkic 2.b) "uch" is protoworld 2.c) The root comes from the Ancient ME (ANE) There might be more but these are good enough for a start. 1) with odds of 1 to 100 or 1 to 1000, it behooves not to believe this at first. 2a) There are too many pieces of evidence against it. But this could be one sided because I do not know Sanskrit. This is a common phenomena. If I am asked "Is a qumquat more like a quince or an apple?" I could not answer this correctly unless I knew all three fruits. So the Sanskrit part has to be provided by IE scholars, however I will do my little part from the Turkic angle, and even more. I will also take this to the ANE :-) There are too many Turkic words that have something close to this and they fall into a pattern. First there is 'uc' meaning 'edge'. But there are similar words (/r/ words) like /Or/ (where O=high rounded o, i.e. o-umlaut) (having to do with height, /Orle/ (to climb), /Orge/ (upwards), /Oreley/ (standing up), etc. And these can be found in East Turkistan as well as in the North Caucasus. Secondly the /z/ version is there; i.e. Uzre, Uzeri , or Ust, UstUnde, UsUnden, UstUN, etc having to do with 'top, above' etc. Then there is 'oz' (to pass, to overtake), and /Os/ (to grow (high)). 2b) 'uch' or something related (see below in 2c) is protoworld, but it goes thru lots of changes. 2c) The word for /bird' is /kush/. Something like k>x>h>0 would produce /ush/ > /uch/. And the k>x can be seen between Turkic languages right now. Further, /mushen/ is /bird/ in Sumerian and m>k is Sumer > Turkic (as can be seen in Tuna's book). Furthermore in the OI books on Hittites, the word /ar/ shows up as 'height', or 'high'. I think something like /arma/ was 'moon'. This is said to be an Asianic word, not IE. We note that a > O, gives /Or/. Furthermore r > y gives /ay/ which is the word for moon in Turkic. Furthermore r > y can be found to be a change observed in Dravidian > Turkic (I think, I have to check Lahovary, again). Where does that leave Protoworld? -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From manaster at umich.edu Wed Oct 28 12:45:40 1998 From: manaster at umich.edu (manaster at umich.edu) Date: Wed, 28 Oct 1998 07:45:40 EST Subject: r and s in Turkic and the Altaic Theory In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Sasha Vovin is entirely right that there is no connection whatever between the different views on the phonetic values of the two Proto-Turkic phonemes at issues AND the different views on the Altaic theory. Leading opponents (e.g., THE leading anti-Altaicist, Doerfer) as well as proponents of Altaic (like Sasha or in my own small way, me) agree on this point, though there are some people who have not seen the light. From afaefk at upe.ac.za Wed Oct 28 12:46:39 1998 From: afaefk at upe.ac.za (Prof E F Kotze') Date: Wed, 28 Oct 1998 07:46:39 EST Subject: rhotacism from Ray Hickey Message-ID: Dear Collegues, I perhaps should have made it clearer when I posted Ray Hickey's remarks yesterday that the text was entirely Ray's. Dorothy Disterheft ----------------------------Original message---------------------------- Dear Dorothy An interesting synchronic phenomenon in Afrikaans, which seems to be spreading rapidly, is the changing of syllable-final /s/ from apical/alveolar to retroflex when preceded by /r/, to such an extent that /r/ and /s/ coalesce into something which is phonologically similar to the Czech r hacek (but voiceless). Examples in Afrikaans are kinders ('children') (with accent on the first syllable), verseker ('ensure') (accent on the second) and nors ('grumpy'). Your comments below (or are they Ray's?) led me to draw this comparison: >A fricative /r/ (apical trill or continuant) can easily become > a full sibilant, i.e. re-align itself phonologically as /z/, as has > happened in Polish with the Slavic PRE-prefix and as seen in Czech in the > trilled /r/ (evident e.g. in _Dvorak_, the composer's name and indicated > by a superscript hacek). It also happens sub-phonemically in Stockholm > Swedish and forms of Western Irish when /r/ is phonetically palatalised and > where the raising of the apex to the palate leads to assibiliation. Another case of the promotion of secondary articulation? Seemingly. Ernst Kotze' ================================== Prof. Ernst F. Kotze Dept. Afrikaans & Nederlands, UPE Posbus 1600 ZA-6000 Port Elizabeth Afrique du Sud Tel. +27 41 5042226 (W), 533230 (H), 5042574 (F) From tuitekj at ERE.UMontreal.CA Wed Oct 28 13:05:53 1998 From: tuitekj at ERE.UMontreal.CA (Kevin Tuite) Date: Wed, 28 Oct 1998 08:05:53 EST Subject: Jerus Masia! Message-ID: ----------------------------Original message---------------------------- My assertion that the Chukchee /r/ has uvular articulation is based on the Bogoras grammar in HAIL, which describes its pronunciation "as in French (hard trill, 'roue')", contrasted to a "dental r with weak trill" (p 645). (On the other hand, in a table on the preceding page, both /r/'s are classed as "alveolar", god knows why). I had assumed that the Chukchi "French r" would sound roughly the same as the uvular R of Eskimo. Ralf-Stefan Georg kindly brought to my attention two recent grammars of Chukchi which describe the sound in question as being, indeed, alveolar or perhaps alveo-palatal ("perednejazychnyj, kakuminal'nyj"). I remain intrigued by the sociolinguistic aspects of the Chukchi and French /r/ > sibilant shifts, and in particular the association of both changes with women's speech (Recall Erasmus' observation that the French shift was led by the "mulierculae parisinae "). Could the sibilant articulation of an earlier rhotic have "feminine" associations of the sort that Matthew Gordon and Jeffrey Heath propose for vowel shifts (see the latest issue of Current Anthropology, vol 39 #4)? Kevin Tuite PS Greetings to Alexis. ************************************************************** Kevin Tuite 514-343-6514 (bureau) Département d'anthropologie 514-343-2494 (télécopieur) Université de Montréal C.P. 6128, succursale centre-ville Montréal, Québec H3C 3J7 tuitekj at mistral.ere.umontreal.ca ************************************************************** From Georg at home.ivm.de Wed Oct 28 17:15:38 1998 From: Georg at home.ivm.de (Ralf-Stefan Georg) Date: Wed, 28 Oct 1998 12:15:38 EST Subject: r and s in Turkic (long and focussing on Altaic) In-Reply-To: Message-ID: ----------------------------Original message---------------------------- No, I don't mean to convert Histling into AltaiNet - and I did notice that this thread has left its original focus long ago, but I think some points maybe of general interest, and the Altaic details can, if need be, discussed by us further in a different forum. Sasha, old friend, good to hear from you again. While it will come as no surprise to you (nor to me) that a discussion of these matters will see us two, as so often, on the opposite sides of the table, I might be allowed to comment a bit on your assertions. First, let me agree with you (and Alexis, of course) that the whole rhotacism/zetacism has no immediate bearing on the Altaic question. The languages can wlel be related (or not), regardless what the original consonant qualities were. This is, e.g., the position prominent in most writings of Gerhard Doerfer, who is, as we know, a critic of Altaic and a zetacist (by which I mean that he views /z/ as the end of the story). However, it is quite surprising that pro-Altaicists usually defend zetacism like the crown-jewels, each time a rhotacist appears lumping him/her with the anti-Altaicists. In view of what you said yourself, this is an oversimplification, or, better, plainly wrong. I don't follow Lyle Campbell where he formulates that " No one with a sense of phonology postulates the reverse, the unnatural and implausible changes of *r > z/s and *l > Z/S, which are almost unknown in languages elsewhere" etc., since the opposite sound-change *can* happen (few things really *don't ever* happen), as we see in Kevin Tuite's French example (already quoted in Shcherbak somewhere and in writings by Rona-Tas); I remember you yourself bringing an example from Vietnamese - if I'm not mistaken - to my attention. Yet I subscribe to the view that rhotacism is far more common (see previous examples from a wide variety of languages). Yet, phonetics apart, and it is not basically a phonetic problem I'm dealing with, there *are* difficulties with either scenario (for *both*) positions which have at least potential bearing on Altaic studies (not on the "Altaic question", no question like that can be decided on by relying on a single point like that; so, in a word, there can be no talk of "disproving" "Altaic" by dealing with the liquids alone, I think we all can agree on that). >(3) The theory of wholesale borrowing from Bulgaric to proto-Mongolic >faces many obstacles (some of them unsurmountable in my opinion). First, >it is cyclic by its logic: Turkic "loans" in Mongolic have /r/, therefore >they must be Bulgar loans. But there should be, a second independent >evidence for the fact that these words are indeed Bulgaric loans, apart >from the the /r/ itself. As a matter of fact, this evidence does not >exist. Let me demonstrate it by using the same word for "ox". With this I have to take issue, since, I'm afraid, it contains a piece of oversimplification on your side. -r- instead of CT -z- is *by no means* the only criterion for the "Bolgharicity" of some words. There are other criteria, involving other pieces of consonantism and, especially, vocalism, which is all too often overlooked. A word can have (more than one) typical Bulghar trait, even if it doesn't contain anything remotely resembling -r- (< -z- or -r2). I have recently (in press) argued *against Doerfer* (who took the words as chance similarities) for a loan-scenario involving the Turkic and Mongolian words for "heart" (although there *is* an -r- in this word, it's of the /r1/-type) on the evidence of non-rhotacistic criteria alone. I'll not try to get on everyone's nerves here by giving the whole story of this etymology, but I'm ready to share it with everyone, who wants to hear about it. On rhotacism: zetacism does carry along one difficulty for Altaic studies: whily in a rhotacistic scenario this process happened only once (in Bolghar, carried over by LWs to Mong. and Tung.), the zetacist scenario would hold /r/ as the proto-Altaic sound, maintained almost everywhere, but shifted to /z/ in each and every non-Bulghar Turkic language. While this is OK as far as it goes, it brings about the need to view zetacistic Tk. languages as *one and only* primary branch of Turkic (languages from Turkish to Yakut via Tuvan, that is). This may well be the case, but on other occasions we two had a hot debate on the number of *primary* branches of Turkic, with me voting against and you voting for such a branch as "Sayanic" and others, being as primary as all the others (Kipchak, Oghuz). Now the least thing you'll have to do is to admit a *primary* branch "Common Turkic" with the shift r2 > z as the one common innovation. Maybe you are prepared to do so, but don't talk to me about "primary Sayanic" again ;-) On the "bovine"-etymology you do a variety of things. You reconstruct proto-Mong. *with* *p-, using the Mmo. evidence, which is OK with me. Then you deny *p- (> h-) in Turkic, reconstructing *"ok"uz only (I take it that this means CT, or you accidentally forgot to write -r2 ;-). This should now make borrowing less likely then cognacy. I admit that I don't follow (in fact, the Starostinian sound-laws expect proto-Altaic p- - evidenced by Mong. - to be reflected by Turkic h- as well; so, taking Starostin's sound-laws for granted, you are actually advocating for the Turko-Mong. etymology to be given up ! Or else, where's the h- ? Calm down, it is there, see below.). In fact, by removing the h- from the Turkic form you strengthen the case for an old (and long ago given up) borrowing scenario which wanted to derive Turkic *"ok"uz from sthl. Tokharian okso (this etymology had been given up, largely because h- got in the way, which was not well known in Turkic at that time, as well as the Mmo. data). Now I wouldn't call this borrowing scenario nonsense, but I don't subscribe to it, since I do think that the h- is there after all. The question of Turkic h- is a difficult one, and it can't be put aside by a laconic "but it is prosthetic" and a mere "contrary to Doerfer's claim". While I'm not going to present the fifty-odd papers D. devoted to the problem here, readers interested in the question might wish to consult G. Doerfer: "Materialien zu t"urk. h-" I, UAJb N.F. 1/198193-141, II: 2/1982, 138-168 (oops, it's eighty-odd pages). There a (imho strong) point is made for the attestation of original h-s in some Turkic lgs. and prosthetic ones also, together with some criteria to distinguish between them. I hope interested parties will read this to form an opinion of their own. Now, for the "bovine"-word, h- is attested from more than one source. While it is true that Khaladzh is silent on this word, we find h- forms in Turkmen dialects (R"as"anen), Khorasanli Turkic, Modern Uighur, Uzbek (which you mentioned) and the older literary language Chaghatay. So, h- is there, the Tokharian loan scenario is again to be forgotten, and the Turkic and Mong. words belong together. (Please note that, while I'm far from presenting everything Doerfer says as the plain gospel, I'm unwilling to further discuss the matter of Tk. h- on the base of anything but this paper. It is a most thorough investigation of this difficult problem and every discussion which does not depart from it (maybe by disagreeing, but then the whole of the data and the methods employed should be discussed in detail, which we certainly cannot do here) is simply uninformed. >does not show up in such word as ar "man", where we should expect it if >Uzbek h- were regular. Thus, we end up with Turkic *"ok"uz. And: the word for "man" *is* attested in Uzbek with h- (albeit not in the literary standard, but rather in Kipchak-Uzb. dialects; and it is so attested in Old Turkic in Brahmi-script, TT VIII). >(c) summing up, the hypothesis that Mongolic word is a loan from >Bulgar s based on three (sic!) unproven hypotheses: that Bulgaric had *h- >(no traces of it in any of Bulgaric languages) (1), that Chuvash vocalism >is secondary as compared to Common Turkic (2), and that it represents a >late development (3). Thus, the loanword explanation, a hypothesis itself, >is relying on three other hypotheses to be true. It is much more complex >solution than a single hypothesis that words in question represent >cognates. (2) and (3) are only one hypothesis. It is true that Bulghar does not show any trace of h-. But this is common practice: Germanic loans in Balto-Finnic sometimes show traces of Proto-Germanic, which are nowhere attested in Germanic, but only recoverable through comparative reconstruction; Finnish words often show that this reconstruction was right. There is no reason I can see why an older form of language A should not be observable in A loans in B. > Finally, the wholesale loanword scenario is based on one more >hypothesis, again not proven. It presumes that Mongols were in contact of >all Turks with Bulgars. But Bulgars are... the westernmost branch of >Turkic, and there is a zero independent evidence that they ever were in >touch with Mongols (until 13th c.), who came to the territory of Mongolia >from the >territory to the east. The only evidence presented for this brave claim is >again Bulgar "loanwords" in Mongolic with -r. Circularity again, isn't it? But there is another argument here, which is simply dangerous: that of geography. Well, Bulghar is the Westernmost branch of Turkic, so what ? Baluchi is one of the easternmost Iranian languages, yet it is West Iranian, Ossetic is one of the westernmost languages of this family, yet it is East Iranian. Iranian loanwords are present in a variety of Finno-Ugric languages, Baltic loanwords are to be found as far east as Mordva, it is linguistics which can *claim* that for these borrowings previous contact *has* to be assumed (and it is, in the case of Baltic in the East, seconded by toponymy). You don't assume that Bulghar Turkic speaking people have always been living on the Volga, do you ? To put it more precisely, changing seats for argument's sake: given the Altaic theory were correct, and an original proto-Altaic speech community split up somewhere in Asia: we don't have to assume that the Japanese end up on the Japanese islands immediately after that, as well as the Bulghar Turks on the Volga, do we ??? The disintegration of an original speech community is a complicated process, certainly allowing for a great deal of inter-branch borrowing going on for some time, at least as long as geographical contiguity is maintained or contacts are not blocked by social and other reasons. *Of course*, the Bulghar loan hypothesis *does* make a strong claim for prehistoric contacts between Bulghar Turkic and Mongolian, that's its strength, that's where the linguist's work becomes interesting for historians, that's part of the success story of I.E. linguistics in Western Universities, I'd say. The only thing we can say that Bulghar *writing* started in the West. There is a host of early Inner Asian nomad confederacies which are linguistically "unlumpable" into one of the known language families. There can be no reason to exclude that one or some of them contained a "Bulgharoid" element (though I don't dare to forward a specific hypothesis on which one). And: given that most pro-Altaicists locate the Altaic "homeland" pretty far in the East (you say yourself that the Mongols probably entered the steppe from the east), the Bulghars had to find *some* way to their present habitats through all that landmass, hadn't they ? OK, my apologies for getting verbose (as always when things Altaic are at stake), I'll leave this thread to itself now and see what happens (I'm ready to continue with you, Sasha, or anyone else interested on the other fora, we are subscribed to). Stefan Stefan Georg Heerstrasse 7 D-53111 Bonn FRG +49-228-69-13-32 From mcv at wxs.nl Wed Oct 28 17:13:48 1998 From: mcv at wxs.nl (Miguel Carrasquer Vidal) Date: Wed, 28 Oct 1998 12:13:48 EST Subject: Jerus Masia! In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Kevin Tuite wrote: >I remain >intrigued by the sociolinguistic aspects of the Chukchi and French /r/ > >sibilant shifts, and in particular the association of both changes with >women's speech (Recall Erasmus' observation that the French shift was led >by the "mulierculae parisinae "). Could the sibilant articulation of an >earlier rhotic have "feminine" associations of the sort that Matthew Gordon >and Jeffrey Heath propose for vowel shifts (see the latest issue of Current >Anthropology, vol 39 #4)? While Emesal, the Sumerian "women's speech" [a literary dialect used primarily in direct speech of female characters/goddesses] is not distinguished by assibilation of /r/, there is a "shift" /n/ > /s^/, which might be interpreted in a similar way. However, not all /n/'s are affected: in most Emesal words /n/ = /n/ in the standard dialect. That should caution against making too much of this. The issue whether Emesal actually *is* a women's language (and not a local or late variety) is itself unresolved (in the Sanskrit drama, Prakrits are generally only spoken by female characters, yet the Prakrits are not "women's speech"). ======================= Miguel Carrasquer Vidal mcv at wxs.nl Amsterdam From larryt at cogs.susx.ac.uk Wed Oct 28 17:09:55 1998 From: larryt at cogs.susx.ac.uk (Larry Trask) Date: Wed, 28 Oct 1998 12:09:55 EST Subject: uc and ucak In-Reply-To: <363694DD.2C8B85C4@montclair.edu> Message-ID: ----------------------------Original message---------------------------- On Wed, 28 Oct 1998, H.M.Hubey wrote: > I am going to ask something that sounds very strange. > I noticed accidentally that a word like 'uchakka' (that is > reasonably close) meant 'to fly' in Sanskrit. (I hope that was > Sanskrit, and not Indic or Hindi). > In any case, it is very strange for me to see this since > 'uch' means 'to fly' in Turkic. The choices are > 1. Accident > 2. There is something we are missing > 2.a) Uch was borrowed into Turkic > 2.b) "uch" is protoworld > 2.c) The root comes from the Ancient ME (ANE) > There might be more but these are good enough for a start. > 1) with odds of 1 to 100 or 1 to 1000, it behooves not to > believe this at first. No, not so, I'm afraid. If you ask the question "what is the probability that a form will mean `fly' in both Turkish and Sanskrit?", then the *a priori* odds against are simply enormous. But that's the wrong question. The right question is this: what is the probability that *some* short form will turn up in *some* two languages with similar meanings in both? And this time the answer is "as close to 100% as you care to get". With 6000+ languages available, all of them with thousands of words, and with only a small number of distinguishable consonants and vowels available to construct those words, it is statistically inconceivable that we should fail to find many, many chance resemblances or identities like this one. Hence stumbling across one is evidence for nothing at all, except that the laws of probability are not taking the day off. Failure to appreciate this constitutes what somebody has dubbed "Koestler's fallacy", since the writer Arthur Koestler was notoriously prone to it. Koestler was constantly impressed by the observation that *some particular* coincidence had occurred, reasoning that the *a priori* odds against that particular coincidence were astronomical, and hence concluding that Something Deeply Significant was going on. What he always failed to realize is that the probability that *some coincidence or other* would occur purely by chance was effectively 100%. Larry Trask COGS University of Sussex Brighton BN1 9QH UK larryt at cogs.susx.ac.uk From jrader at m-w.com Wed Oct 28 16:56:12 1998 From: jrader at m-w.com (Jim Rader) Date: Wed, 28 Oct 1998 11:56:12 EST Subject: rhotacism from Ray Hickey Message-ID: ----------------------------Original message---------------------------- Another unexotic example of assibilation of a rhotic sound can be found in several varieties of New World Spanish, where orthographic , the lengthened trill, becomes a voiced palato-alveolar fricative in certain positions. I recall being struck by this in the speech of an instructor from Co/rdoba, in the interior of Argentina, at a time when I knew nothing about the phonetics of Spanish. In the dialects of Irish I have heard, palatalized /r/ retains a rhotic quality, though there is a tendency to devoice and assibilate either "broad" (i.e., non-palatalized) or "slender" (palatalized) /r/ before a voiceless stop (in words like 'strength,' 'hen'). In Scottish Gaelic dialects, however, slender /r/ has some very diverse realizations; at least some speakers from Lewis in the Outer Hebrides have a voiced interdental fricative (at least that's what it sounds like to me). Jim Rader > > The present discussion of rhotacism and its possible reverse has been > centered on Altaic and the Germanic/Latin data so that perhaps the data has > prevented us from making some general points. Taking Lyle's suggestion that > the reversal of rhotacism is very unlikely as a starting point: the issue > to home in on in such changes is the SECONDARY ARTICULATION of the segments > involved. A fricative /r/ (apical trill or continuant) can easily become > a full sibilant, i.e. re-align itself phonologically as /z/, as has > happened in Polish with the Slavic PRE-prefix and as seen in Czech in the > trilled /r/ (evident e.g. in _Dvorak_, the composer's name and indicated > by a superscript hacek). It also happens sub-phonemically in Stockholm > Swedish and forms of Western Irish when /r/ is phonetically palatalised and > where the raising of the apex to the palate leads to assibiliation. > >.................... > Ray Hickey > English Linguistics > Essen University > Germany > > r.hickey at uni-essen.de > From mghiselin at casmail.calacademy.org Thu Oct 29 13:31:14 1998 From: mghiselin at casmail.calacademy.org (Michael Ghiselin) Date: Thu, 29 Oct 1998 08:31:14 EST Subject: Cladistic language concepts Message-ID: ----------------------------Original message---------------------------- Dear Dr. Dyen, I must apologize to list members if something happened when I tried to forward your message. Also please forgive my delay in answering your commentary. Cladistics does run into problems where there is as you say some kind of amalgamation of lineages. Sometimes the amalgamation is not so complete that lineages cannot be detected. An interesting example has been the origin of mitochondria and chloroplasts from symbiotic bacteria. As to DNA, it turns out that the techniques that we use with it are pretty much the same sort of thing that we have used all along in comparative anatomy. I spent a lot of time working with ribosomal RNA sequences and they worked pretty well. But they did not show the clearness of resolution that we hoped for. Some parts of the genome change very slowly, others more rapidly. What we try to do for the very old relationships is find something that evolves very slowly. Linguists try to find that sort of feature too. The frustrating thing for us is that branching can occur repeatedly within a very short period of time, and the branches may not be evident in the slowly-evolving molecules. We get a sort of bush instead of a tree. I suspect that this has happened in linguistic evolution too. What the DNA does for us is give a lot of additional evidence and it often helps. We will just have to keep plugging away. Sincerely, Michael Ghiselin From hubeyh at montclair.edu Thu Oct 29 13:31:48 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Thu, 29 Oct 1998 08:31:48 EST Subject: r and s in Turkic (long and focussing on Altaic) Message-ID: ----------------------------Original message---------------------------- Ralf-Stefan Georg wrote: > > But there is another argument here, which is simply dangerous: that of > geography. Well, Bulghar is the Westernmost branch of Turkic, so what ? Is it not a fundamental principle in genetics as well as historical linguistics, that the region in which one finds the greatest variety must be the ancestral homeland? Is this not this principle which people like Cavalli-Sforza (and other geneticists as well as historical linguists) use to posit that Africa is the homeland of hominids and that the region in which the largest number of languages of a family cluster in the Urheimat? If Bulgharic is in the west, Khalaj is in the west, Oguz is in the west, Kipchak is in the west, where does that point? Is there Khalaj or anything in that family (assuming it can be considered a family) in the east? Is any Bulgharic ever found in the east? > Iranian loanwords are present in a variety of Finno-Ugric languages, Baltic > loanwords are to be found as far east as Mordva, it is linguistics which > can *claim* that for these borrowings previous contact *has* to be assumed > (and it is, in the case of Baltic in the East, seconded by toponymy). You > don't assume that Bulghar Turkic speaking people have always been living on > the Volga, do you ? Maybe they lived there long enough to have other people name the river after them :-) > To put it more precisely, changing seats for argument's sake: given the > Altaic theory were correct, and an original proto-Altaic speech community > split up somewhere in Asia: we don't have to assume that the Japanese end > up on the Japanese islands immediately after that, as well as the Bulghar > Turks on the Volga, do we ??? It does not follow. Assuming that Altaic is a family and that it did split up, we don't know where the split occurred and where the Turkic branch developed or when. Even for IE for which much more data is in existence, the Urheimat still ranges from the Balkans to Central Asia or Anatolia. But nobody wants to name someplace where there are no IE speakers the outlying regions where there were only one or two languages. That cuts out Indo-Iranian regions, the British Isles, or Germany, France, etc. Even if Altaic homeland is given as Mongolia, it does not follow that Turkic homeland is the same place. The disintegration of an original speech > community is a complicated process, certainly allowing for a great deal of > inter-branch borrowing going on for some time, at least as long as > geographical contiguity is maintained or contacts are not blocked by social > and other reasons. *Of course*, the Bulghar loan hypothesis *does* make a > strong claim for prehistoric contacts between Bulghar Turkic and Mongolian, > that's its strength, that's where the linguist's work becomes interesting > for historians, that's part of the success story of I.E. linguistics in > Western Universities, I'd say. The only thing we can say that Bulghar > *writing* started in the West. There is a host of early Inner Asian nomad > confederacies which are linguistically "unlumpable" into one of the known > language families. Bulghar-Mongolian contacts (even if it happened) imply that Bulghar Turkic originated in the East or any other Turkic originated in the east. There are different intensities of language mixing. If we normalize this intensity as a number between 0 and 1 (i.e the interval [0,1]) then there are two favorite types of mixtures. One is at one extreme, say 0, in which the superstratum language disappears almost totally (like Mongols in Iran, or Russia or the Turkic Moguls in India). At the other extreme, say 1, is where the superstratum's language eventually becomes dominant (the changes all allegedly having nothing to do with the substratum). This is the linguists' equivalent of the 'ideal gases' of physics. What about a mixture of type 0.5? What kind would that produce? That is rather easy to guess for me. It would produce a language which is full of quasi-regularity or partial-regularity. For example Semitic is exactly this kind of a language. It has 15-20 "regular" thingamajics called "binyanim". It is the English regular-irregular verb formation to the nth degree. The language "froze" at a point in which it could not go to one or the other extreme but got stuck in between. The Hittite "grades" of verbs is the same phenomena. So that adds more evidence to a rather thorogh mixing process happening over a period of severall millenia in that region. That came out of a mixture of the previous non-IE and non-AA languages with some other intruders. The Ural-Altaic-Dravidian languages share traits which are left over from many millenia ago and also borrowed from each other. Of course, nothing like this can be "proven" because nothing in linguistics can be proven. You can only convince people. Similar things (modelable by similar differential equations) happen to things like creation of steel alloys. What determines the properties of the resulting steel alloy is not only what was added to iron but also the rate of cooling. Slow cooling produces soft metals, because it has more time for the diffusion processes to settle and the stresses to relieve themselves. Rapid cooling by quenching in water produces hard steels. A thorough mixture of 2 or more languages intensely over a period of time in which many people become bilingual or even trilingual is probably a rare occurrence but it is the type of mixture which can cause great changes in all aspects of langauge including its phonology, syntax, and typology. The present conventional wisdom that relationships can be classified as genetic, typological and areal is doing an injustice to languages and is an affront to science. Alas, things persist. There is the long story about why the British railroads are so narrow. It apparently goes back to the standard for Roman roads. That is how slowly some things change. There can be no reason to exclude that one or some of > them contained a "Bulgharoid" element (though I don't dare to forward a > specific hypothesis on which one). And: given that most pro-Altaicists > locate the Altaic "homeland" pretty far in the East (you say yourself that > the Mongols probably entered the steppe from the east), the Bulghars had to > find *some* way to their present habitats through all that landmass, hadn't > they ? Without doing a thorough statistical analysis of what kinds of sounds changes have taken place over centuries and accross the world, it is difficult to make strong statements, however, there must certainly be something to why m>k occurs in Sumerian >Turkic, and Dravidian > Turkic, or why n >y occurs in Sumerian > Turkic and Mongolian > Turkic. I think there is also l>y in Dravidian>Turkic etc. All of this has to be also put in the context of phonological systems of the world's languages circa 12,000-8,000 years ago because very important things were happening around that time. > OK, my apologies for getting verbose (as always when things Altaic are at > stake), I'll leave this thread to itself now and see what happens (I'm > ready to continue with you, Sasha, or anyone else interested on the other > fora, we are subscribed to). One good turn deserves another. Since this can't be discussed on Altainet, it is probably better to discuss it in the context of sound changes that occur in the main theater of history; Eurasia and North Africa and in terms of IE, AA and Altaic, Uralic, and the isolates like Basque. -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From vovin at hawaii.edu Thu Oct 29 13:32:48 1998 From: vovin at hawaii.edu (Alexander Vovin) Date: Thu, 29 Oct 1998 08:32:48 EST Subject: r and s in Turkic (long and focussing on Altaic) In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Dear Stefan, Good to hear from you, too. Let me just jot several points below, because the specifics should be discussed on AltaiNet. (1) Although my expertize within the eastern part of the family, Chuvash is the Turkic language I know really well (used to speak it fluently 10 years ago, unfortunately this is no longer the case). Nevertheless, I fail to see any Chuvashoid features in alleged "Bulgaric loans" in Mongolic, including vocalismthat you mention: Chuv. vAkAr, MM h"ok"ur, OT "ok"uz: obviously Chuvash has a different vocalism. May be you can come up with some hard-core evidence on AltaiNet. I will, for one, will be interested to hear tour etymology of Turco-Mongolian heart, but I honestly can't see how vocalism in Chuvash cEre 'heart" is going to prove that it is loan in Mongolic. (2) As I said in my reply to Lyle, I agree with him that z >r is more natural change than r > z. Yes, I usually cite the example of Hanoi Vietnamese where orthographical r- is pronounced nowadays as /z-/. The fact that it was once /r/ is easily deduced from the historical spelling (based on 17th c. Portuguese spelling) + on modern Saigon dialect where most speakers still keep have /r-/ for /r/. (3) I do reconstruct *"ok"uz for PT (or *"ok"ur2, if you like it better). We can even do *[h]"ok"ur2/z, if you like. But, imho, h- ain't there. I am familiar with Doerfer's article you cite, but it does not persuade me: all we have there is a collection of random h- cases in various Turkic languages. There is no a single word that would have h- in all languages Doerfer cites. Thus word X will have h- in A, D, E, word Y in B and C, word Z in E only, etc. etc. I do not buy it, and I must say that I am really surprised that you do. Taking Qypchak "dialect" of Uzbek hardly improves the picture: that vividly reminds me of one "oficial" PRC linguists, who had to prove relationship of Chinese and Tai against all odds for apparent political reasons, so he manipulated between languages at his free whim. Fortunately, such manooevres are observed mostly in Turcology (:-). (4) Bulgars Of course *could* live near Mongols in spite the fact that they are nowadays westernmost Turks, but this is no more than a speculation that is given without any solid proof. There is no evidence apart from the alleged loanwords that it was ever so, and in this case it is your word against mine. You do believe that there did live there, I don't. OK, can you come up with any kind of evidence, apart from "loans" to show that this is the case? Any place names, archeology, burial practices, anything? The problem of anti-Altaicoists is that they often take their hypotheses to be axiomatic truths. One of Shcherbak's works starts with line: "When Chuvash ancestors lived in Siberia..." O, ya, that is very entertaining, but who has demonstrated that they ever did? (May be thei lived in Huanghe valley: after all Chuvash word for person c,yn looks like a Chinese loan, does not it?) (5) Starostins 1991 book is an important contribution to Altaic studies. But it is not the Bible. The fact that he does not have PT *h- or pre-Turkic *p- is not going to wreck either this etymology or Altaic studies. Cheers, Sasha ======================================= Alexander Vovin Associate Professor of Japanese Department of East Asian Languages and Literatures 382 Moore Hall 1890 East-West Road University of Hawaii at Manoa Honolulu, HI 96822 vovin at hawaii.edu fax (808)956-9515 (o.) t.(808)956-6881 (o.) From dyen at hawaii.edu Thu Oct 29 21:46:50 1998 From: dyen at hawaii.edu (Isidore Dyen) Date: Thu, 29 Oct 1998 16:46:50 EST Subject: Cladistic language concepts In-Reply-To: <9810289096.AA909625474@casmail.calacademy.org> Message-ID: ----------------------------Original message---------------------------- Can you segregate the slowly developing parts of the genome from the oth ers, or more broadly, can you classify sections by their rates of branching? On Wed, 28 Oct 1998, Michael Ghiselin wrote: > > Dear Dr. Dyen, > I must apologize to list members if something happened > when I tried to forward your message. > Also please forgive my delay in answering your > commentary. > Cladistics does run into problems where there is as you > say some kind of amalgamation of lineages. Sometimes the > amalgamation is not so complete that lineages cannot be > detected. An interesting example has been the origin of > mitochondria and chloroplasts from symbiotic bacteria. > As to DNA, it turns out that the techniques that we use > with it are pretty much the same sort of thing that we have > used all along in comparative anatomy. I spent a lot of > time working with ribosomal RNA sequences and they worked > pretty well. But they did not show the clearness of > resolution that we hoped for. Some parts of the genome > change very slowly, others more rapidly. What we try to do > for the very old relationships is find something that > evolves very slowly. Linguists try to find that sort of > feature too. The frustrating thing for us is that branching > can occur repeatedly within a very short period of time, and > the branches may not be evident in the slowly-evolving > molecules. We get a sort of bush instead of a tree. I > suspect that this has happened in linguistic evolution too. > What the DNA does for us is give a lot of additional > evidence and it often helps. We will just have to keep > plugging away. > Sincerely, > Michael Ghiselin > > > From dyen at hawaii.edu Thu Oct 29 21:45:46 1998 From: dyen at hawaii.edu (Isidore Dyen) Date: Thu, 29 Oct 1998 16:45:46 EST Subject: uc and ucak In-Reply-To: Message-ID: ----------------------------Original message---------------------------- The point that you make is quite right, but I believe that what you are dealing with is likelihood instead of mathematical probability. What lies at the bottom of the problem is that the lay word probability most commonly has the meaning 'likelihood' that is the relation between the respective probabilities associated with each outcome. ID. On Wed, 28 Oct 1998, Larry Trask wrote: > ----------------------------Original message---------------------------- > On Wed, 28 Oct 1998, H.M.Hubey wrote: > > > I am going to ask something that sounds very strange. > > > I noticed accidentally that a word like 'uchakka' (that is > > reasonably close) meant 'to fly' in Sanskrit. (I hope that was > > Sanskrit, and not Indic or Hindi). > > > In any case, it is very strange for me to see this since > > 'uch' means 'to fly' in Turkic. The choices are > > > 1. Accident > > 2. There is something we are missing > > 2.a) Uch was borrowed into Turkic > > 2.b) "uch" is protoworld > > 2.c) The root comes from the Ancient ME (ANE) > > > There might be more but these are good enough for a start. > > > 1) with odds of 1 to 100 or 1 to 1000, it behooves not to > > believe this at first. > > No, not so, I'm afraid. > > If you ask the question "what is the probability that a form will > mean `fly' in both Turkish and Sanskrit?", then the *a priori* odds > against are simply enormous. But that's the wrong question. > > The right question is this: what is the probability that *some* short > form will turn up in *some* two languages with similar meanings in both? > And this time the answer is "as close to 100% as you care to get". > > With 6000+ languages available, all of them with thousands of words, and > with only a small number of distinguishable consonants and vowels > available to construct those words, it is statistically inconceivable > that we should fail to find many, many chance resemblances or identities > like this one. Hence stumbling across one is evidence for nothing at > all, except that the laws of probability are not taking the day off. > > Failure to appreciate this constitutes what somebody has dubbed > "Koestler's fallacy", since the writer Arthur Koestler was notoriously > prone to it. Koestler was constantly impressed by the observation that > *some particular* coincidence had occurred, reasoning that the *a > priori* odds against that particular coincidence were astronomical, and > hence concluding that Something Deeply Significant was going on. What > he always failed to realize is that the probability that *some > coincidence or other* would occur purely by chance was effectively 100%. > > > Larry Trask > COGS > University of Sussex > Brighton BN1 9QH > UK > > larryt at cogs.susx.ac.uk > From anna-karin.strobel at swipnet.se Thu Oct 29 18:29:21 1998 From: anna-karin.strobel at swipnet.se (anna-karin.strobel) Date: Thu, 29 Oct 1998 13:29:21 EST Subject: The languages of Gibraltar Message-ID: ----------------------------Original message---------------------------- Hej, I read somwere that in Gibraltar there is a small community where the inhabitants speak genovese-ligurian - is this correct and does anyone has any information or reference abour this? The common language is english but the most of the people also speak a dialect of castilian - which. And has this dialect influenced the local english language or does they just speak standardenglish. Thanks in advance for any information, Steve Lando Halland, Halmstad, Sweden, Europe, Tellus. From Roger.Wright at liverpool.ac.uk Thu Oct 29 16:23:29 1998 From: Roger.Wright at liverpool.ac.uk (Roger Wright) Date: Thu, 29 Oct 1998 11:23:29 EST Subject: South American /r/ & [h] In-Reply-To: <14492328618037@m-w.com> Message-ID: ----------------------------Original message---------------------------- And in Brazilian Portuguese, the phoneme /r/ (or, at least, the one written with the letter "r") is usually [h] (or [x]). A secondary articulation, I suppose; Lisbon has an aspirated sort of [rh] or [hr]; further East, nearer the Spanish border, it doesn't. RW On Wed, 28 Oct 1998, Jim Rader wrote: >----------------------------Original message---------------------------- >Another unexotic example of assibilation of a rhotic sound can be >found in several varieties of New World Spanish, where orthographic >, the lengthened trill, becomes a voiced palato-alveolar >fricative in certain positions. I recall being struck by this in the >speech of an instructor from Co/rdoba, in the interior of Argentina, >at a time when I knew nothing about the phonetics of Spanish. From tonybreed at juno.com Fri Oct 30 11:50:39 1998 From: tonybreed at juno.com (D. Anthony Tschetter-Breed) Date: Fri, 30 Oct 1998 06:50:39 EST Subject: s > r Message-ID: ----------------------------Original message---------------------------- In Rebecca Posner's "The Romance Languages" (Cambridge 1996, p.230), she cites an s>r transformation in Sard, with the plural definite article "sas": "Even in Sard we find assimilation of -s before voiced consonants: sar dentes 'the teeth', sa mmanos 'the hands', for sas dentes/manos." I don't know if this transformation exists throughout Sard or in limited areas. In that section, Posner talks about the general weakness of /s/, transforming, variously, to /j/, /h/, /S/, and /x/ via /S/, as well as dropping completely in certain cases (in French), in addition to the Sard example. -Tony Breed ___________________________________________________________________ You don't need to buy Internet access to use free Internet e-mail. Get completely free e-mail from Juno at http://www.juno.com/getjuno.html or call Juno at (800) 654-JUNO [654-5866] From sgbrady at ucdavis.edu Fri Oct 30 11:51:18 1998 From: sgbrady at ucdavis.edu (Sean Brady) Date: Fri, 30 Oct 1998 06:51:18 EST Subject: Cladistic language concepts In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Just a note on DNA sequence data. The rate of change in a particular sequence (e.g. a gene) must be conceptually kept separate from the rate of branching. This latter rate involves speciation, while the former rate probably has nothing to do with speciation. People have certainly tried to tie these two together, but such a link, if it exists, has not been demonstrated yet. In terms of using DNA sequence as comparative evidence for the history of speciation (i.e. building a tree), the main concern is finding a sequence of DNA that changes slow enough to preserve comparative differences, but without changing so fast that these differences 'pile up' on each other and obscure their history. Unfortnately, if the true history is more like a bush than a tree, as Ghiselin discusses, any gene we may pick will probably contain too few changes (i.e. be too slow) to reconstruct this history. A crucial question in both biological and linguistic reconstruction seems to be this: Does our analysis yield a bush-like structure 1) because there really is a bush-like structure; or 2) because our data and/or methods of analysis cannot resolve the (true) tree-like structure. This is a huge, and unsolved, problem in biological systematics. We simply cannot distinquish between these two scenarios yet, although with the rapid influx of DNA data, some people are beginning to tackle this problem. I wonder what the state-of-the-art is on this issue regarding linguistic reconstruction. \>----------------------------Original message---------------------------- >Can you segregate the slowly developing parts of the genome from the oth >ers, or more broadly, can you classify sections by their rates of >branching? > ************************************************************ Sean Brady sgbrady at ucdavis.edu Population Biology Graduate Group tel. (530) 752-9977 Department of Entomology fax. (530) 752-1537 University of California One Shields Avenue Davis, CA 95616-8584 ************************************************************ From hubeyh at montclair.edu Fri Oct 30 11:51:44 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Fri, 30 Oct 1998 06:51:44 EST Subject: uc and ucak Message-ID: ----------------------------Original message---------------------------- Isidore Dyen wrote: > > ----------------------------Original message---------------------------- > The point that you make is quite right, but I believe that what you are > dealing with is likelihood instead of mathematical probability. What lies > at the bottom of the problem is that the lay word probability most > commonly has the meaning 'likelihood' that is the relation between the > respective probabilities associated with each outcome. ID. Actually the biggest problem with probability calculations involving linguistics (especially historical linguistics) is not about technical matters but with setting up the problems correctly. 1. First, the simplest thing to do is the use uniform or equal probabilities when other info is lacking. 2. Second, the simplest thing to do is to assume things like independence, uncorralated sequences, etc. 3. Third, some people, who are allegedly experts in this field, trip over simple things. Notice, for example, the Ringe and Manaster-Ramer exchanges. However, the main problem, as in all probability problems is to be able to clearly know/define the sample space. The sample space is not only those languages that are now in existence, nor only those languages that were in existence but all languages that could have possibly existed but did not. It is this last part that is the biggest problem. Also as a part of this problem is the tendency to extrapolate from simple results linearly without any justification for it. For example, someone might run a simple simulation problem with 2 languages, each having 5 phonemes, and 100 words and getting 15 matches by accident and then extrapolating that extending this to 1,000 words with 50 phonemes etc will probably produce 500 matches. Or starting with calculations using the binomial distribution with p=0.001 (which might be appropriate for 1,000 words) and then using a semantic shift of 25 (because you can find 25 words having to do with "eating" in English) while ignoring that these 25 words came from the full English language with 100,000 to 400,000 words. The other big problem is to assume that word distributions are all totally uncorrelated and independent when we know that they are best modeled as correlated and as Markov processes. Even when simple models are used we should not use independence. Despite all this, there are even more elementary problems. For example, if you toss a penny 5 times and get 4 heads (H) it does not mean that since getting 5 heads is so rare, the penny sort of owes that it should pop up as tails. It does not. It has no memory. The prob is 1/2 at each throw. As another example, think about bus schedules. They arrive in dense intervals during morning and evening. Suppose the avg time between bus arrivals is M. If a passenger pops up at a bus stop at random. What is is expected wait? Surely, we know it is longer than the average time between the bus arrivals. As another example, just because someone is guaranteed to win a lottery of 1 million numbers, does not mean that the guy who won is not lucky. His chances of winning was still 0.0000001. Related to this is the fact that just because you are bound to find gold someplace in the ground because there is so much of it, it does not mean that just because you found a gold nugget when digging some place that there is no more there because it was just an accident. No. There is probably more there, because this is not a purely random event. Where there is some, there is probably more. The same holds for words. Every word you find, has to be thought of as being improbable because you see, the biggest part of the problem is missing. I don't know any Sanskrit. I just read maybe 20-40 words in some book by accident and one of them hit! If I watch a movie with some Eskimos and all of a sudden, 3 words hit me suspiciously just like some other words, anyone who tries to convince me that it happens all the time better spend some time learning to use probability theory to solve some real problems in the real world. Next time you watch the natives of X-land on TV, listen to their words carefully and see if you can spot any English! If you do, chances are great that they will be speaking English. That is a Markov process. PS. The total sample space is not 6,000 languages. IT is maybe 1,000,000 languages that might have existed but did not (or disappeared). PPS. Just because 80% of the world's population is white does not mean that the fact that they resemble each other is not genetic. Of course it is. Physical characteristics are genetic. The most difficult part of using any kind of math is in knowing when to use which formula. Those who botch things in the 8th grade or earlier say "I hate those word problems". Some people are like Duracells :-) They don't hate word problems, not in the 8th grade, not 12th, not in the MS level and not in the PhD level, and not even 20 years after the PhD level :-) > On Wed, 28 Oct 1998, Larry Trask wrote: > > > ----------------------------Original message---------------------------- > > On Wed, 28 Oct 1998, H.M.Hubey wrote: > > > > > I am going to ask something that sounds very strange. > > > > > I noticed accidentally that a word like 'uchakka' (that is > > > reasonably close) meant 'to fly' in Sanskrit. (I hope that was > > > Sanskrit, and not Indic or Hindi). > > > > > In any case, it is very strange for me to see this since > > > 'uch' means 'to fly' in Turkic. The choices are > > > > > 1. Accident > > > 2. There is something we are missing > > > 2.a) Uch was borrowed into Turkic > > > 2.b) "uch" is protoworld > > > 2.c) The root comes from the Ancient ME (ANE) > > > > > There might be more but these are good enough for a start. > > > > > 1) with odds of 1 to 100 or 1 to 1000, it behooves not to > > > believe this at first. > > > > No, not so, I'm afraid. > > > > If you ask the question "what is the probability that a form will > > mean `fly' in both Turkish and Sanskrit?", then the *a priori* odds > > against are simply enormous. But that's the wrong question. > > > > The right question is this: what is the probability that *some* short > > form will turn up in *some* two languages with similar meanings in both? > > And this time the answer is "as close to 100% as you care to get". > > > > With 6000+ languages available, all of them with thousands of words, and > > with only a small number of distinguishable consonants and vowels > > available to construct those words, it is statistically inconceivable > > that we should fail to find many, many chance resemblances or identities > > like this one. Hence stumbling across one is evidence for nothing at > > all, except that the laws of probability are not taking the day off. > > > > Failure to appreciate this constitutes what somebody has dubbed > > "Koestler's fallacy", since the writer Arthur Koestler was notoriously > > prone to it. Koestler was constantly impressed by the observation that > > *some particular* coincidence had occurred, reasoning that the *a > > priori* odds against that particular coincidence were astronomical, and > > hence concluding that Something Deeply Significant was going on. What > > he always failed to realize is that the probability that *some > > coincidence or other* would occur purely by chance was effectively 100%. > > > > > > Larry Trask > > COGS > > University of Sussex > > Brighton BN1 9QH > > UK > > > > larryt at cogs.susx.ac.uk > > -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From manaster at umich.edu Fri Oct 30 11:52:11 1998 From: manaster at umich.edu (manaster at umich.edu) Date: Fri, 30 Oct 1998 06:52:11 EST Subject: h- in Turkic In-Reply-To: Message-ID: ----------------------------Original message---------------------------- I must say that I am somewhere between Sasha Vovin and Stefan Georg on the question of Turkic h-. I agree with Sasha (and like him am surprised anyone would not agree) that the examples cited by Doerfer from languages other than Khalaj have all the earmarks of sporadic secondary developments of no great antiquity. By the way, does anybody know why so many languages have these weird "inorganic" initial h's--and has there been any work on this? (What I mean is that some languages, e.g., Tubar in the Uto-Aztecan family, Polish or at least some dialects of it, and any number of others) prefix h- to some but not all words that etymologically begin with a vowel, without any regard to the doctrine of regularity of sound change. Why? Anyway, there is one Turkic language where I am not sure that I would share Sasha's view entirely, namely, Khalaj. It is obvious if one looks closely that even here the initial h- cannot be simply the reflex of Altaic (for anti-Altaicists, pre-proto-Turkic) *p- that Doerfer (the anti-Altaicist) as well as Poppe and Dolgopol'skij (pro-Altaicists) took it to be. There are altogether too many cases where Khalaj has h- but where the non-Turkic Altaic languages do not have (reflexes of) *p-. I have long suspected that the other h-'s in Khalaj are reflexes of an otherwise unreconstructed Altaic *w-, but have not done the detailed work required to see how well this idea works. One problem with this idea is that the only good test of it I know of at the moment is to compare Khalaj to Nostratic, which of course will not please people who are closed-minded on the subject of Nostratic. But of course no one on this list could be closed-minded, could we, so perhaps I can get someone to look at this problem... Alexis From sw271 at cus.cam.ac.uk Fri Oct 30 11:52:30 1998 From: sw271 at cus.cam.ac.uk (Sheila Watts) Date: Fri, 30 Oct 1998 06:52:30 EST Subject: r and s again Message-ID: ----------------------------Original message---------------------------- Since this is turning into a whole collection of r and s/z stories, here's another one. English names with a first syllable in CVr- are frequently turned into an affectionate form with an -l replacing the -r: Mary to Molly, Harry to Hal, Dorothy to Dolly and, for more modern examples, Derek to Del and Terry to Tel. However, there are also names in which the r tends to become z - Sharon to Shaz(za) and Barry to Baz(za), for instance (alongside names where the z is simply voicing, as in Gazza from Gascoigne). Is anyone aware of any work on this - what triggers the choice between l and z, for instance?There is clearly a dialect factor of some kind: if I judge it correctly, these are London forms, certainly not current in my native Ireland. Sheila Watts _______________________________________________________ Dr Sheila Watts Newnham College Cambridge CB3 9DF United Kingdom phone +44 1223 335816 From vovin at hawaii.edu Sat Oct 31 16:33:27 1998 From: vovin at hawaii.edu (Alexander Vovin) Date: Sat, 31 Oct 1998 11:33:27 EST Subject: h- in Turkic In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Alexis, I think your message regarding initial h- raises one important methodological issue. Namely, can we allow a reconstruction of a segment for a proto-language that is preserved in a single language of otherwise big language family, and for which there is no second independent evidence? It seems that you would answer that question in the affirmative in this particular case, although I remember that once you yourself were bashing (quite justifiable, in my opinion, a person X from Moscow Nostratic school for search of IE and Nostratic accent distinctions uniquely preserved in Bengali). I would hate to disagree with you on Khaladj h-, but I think I have to. I would answer in the negative to the question I posed above, although I think that some exceptions could be allowed when a language that unikely preserves segment X, is on the top of the branching. In all other cases it is much safer to reconstruct something, especially something radical, like PT *h- on the basis of two independent pieces of evidence. Khaladj is probably *close* to the root of Turkic tree, but it does not represent primary branching, I think. It will be dangerous enough to reconstruct PT *h- on its sole evidence (although I think that this might eventually turn out to be true -- let us see), but looking for the traces of something Nostratic in khaladj *only*, does not seem to be very realistic. Cheers, Sasha ======================================= Alexander Vovin Associate Professor of Japanese Department of East Asian Languages and Literatures 382 Moore Hall 1890 East-West Road University of Hawaii at Manoa Honolulu, HI 96822 vovin at hawaii.edu fax (808)956-9515 (o.) t.(808)956-6881 (o.) From hubeyh at montclair.edu Sat Oct 31 16:33:53 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Sat, 31 Oct 1998 11:33:53 EST Subject: Cladistic language concepts Message-ID: ----------------------------Original message---------------------------- Sean Brady wrote: > > In terms of using DNA sequence as comparative evidence for the > history of speciation (i.e. building a tree), the main concern is finding a > sequence of DNA that changes slow enough to preserve comparative > differences, but without changing so fast that these differences 'pile up' > on each other and obscure their history. Unfortnately, if the true history > is more like a bush than a tree, as Ghiselin discusses, any gene we may > pick will probably contain too few changes (i.e. be too slow) to > reconstruct this history. There was recently a science article in which the short-term mtDNA mutation rate was found to be much higher than the long-term mutation rate was thought/accepted to be. In the article it said/hinted that it could be because there was a limited number proteins that could occur and that some mutations went back to some older ones. What is the latest on this? I can see how something similar can account for the great disagreement in linguistics. For example, there is nothing to prevent a change r > z to change later to z >r and then look as if nothing happened. When we look at the results of the diffusion equation (the one that creates the Gaussian i.e. normal density) we see that if the changes in linguistics also obey this equation at least approximately (and they probably do because almost all math models implicitly use this Gaussian distribution) then if we examine intensities of change, then the largest number is always no change. IOW, if we could measure change and could assign numbers like level 1 change, level 2 change etc, then level 0 change (i.e. no change) is always greater than level 1 change, and it is also greater than level 2 change etc. So this means that it is a mistake to assume that if we see two words that seem exactly alike after 5,000 years that it is due to chance because after all, we expect more of no-change than one-sound-change, and more no-change than 2-sound-changes etc. This is a common belief that is exacerbated because many linguists also seem to assume that all changes in all languages should be occurring at the same rate as in the IE languages whereas other languages might be many times more stable. > A crucial question in both biological and linguistic reconstruction > seems to be this: Does our analysis yield a bush-like structure 1) because > there really is a bush-like structure; or 2) because our data and/or > methods of analysis cannot resolve the (true) tree-like structure. This is > a huge, and unsolved, problem in biological systematics. We simply cannot > distinquish between these two scenarios yet, although with the rapid influx > of DNA data, some people are beginning to tackle this problem. This is also a huge unsolved problem in linguistics :-) > Sean Brady sgbrady at ucdavis.edu > > Population Biology Graduate Group tel. (530) 752-9977 -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From gonzalor at jhu.edu Thu Oct 1 13:22:30 1998 From: gonzalor at jhu.edu (Gonzalo Rubio) Date: Thu, 1 Oct 1998 09:22:30 EDT Subject: Q: oblique cognates In-Reply-To: Message-ID: ----------------------------Original message---------------------------- In his four vol. work _Comparative Bantu_ (Farnborough: Gregg International Publ., 1967-1971), Malcom Guthrie talked about SKEWED correspondences or reflexes (of "proto-forms"). These skewed forms are very probably connected because of their formal and semantic similarities, but they do not exhibit the regular correspondences one would expect. These skewed correspondences cannot be explained by any specific sound law or known internal development: Meinhof's rule (deletion of voiced stops in nasal contexts: C --> 0 /N___VN[C]); Kwanyama law (like Meihof's rule, but affecting only /mb/ and /nd/); Dahl's law (dissimilation by voicing of /k/ --more or less, the Bantu version of Grassmann's law), etc. Sometimes, it may be the case that we are missing information about the internal history of a concrete Bantu language, but this unknown rule/event would affect just one or two words... unlikely if it's not by analogy. Moroever (and more likely) a skewed form might be a loanword from another Bantu language, instead of a reflex from a "common ancestor" (call it "Ursprache" if you wish), which would explain the irregularity. Nevertheless, many skewed forms seem to exhibit different patterns, suffixes or prefixes, etc. In sum, for some reason we are missing, these skewed forms seem to belong to a coherent set of cognates but exhibit some irregularities. As far as I know, the only use of SKEWING as a concept in historical linguistics besides Guthrie's (who coined it, I think) can be found in a recent book by Patrick R. Bennett, _Comparative Semitic linguistics: A manual_ (Winona Lake, Eisenbrauns: 1998), pp. 30-31 (see p. 31 for examples of skewing in Semitic). In many examples from Bennett's, the skewing depends on the use of different nominal patterns (similar to the caput/cabeza problem), different vocalizations, different genders, truncation, expansion, etc. Other examples from Bennett's present just the usual "irregularities" we all know well (metathesis, prosthesis, etc.). BTW, Bennett's book offers a very useful repertory of materials to teach Semitic linguistics (lists of cognates, paradigms in many languages, maps with lexical isoglosses, etc.). The fist part of the book is a sort of general introduction to historical linguistics, but using Semitic examples. And the rest is just a very practical collection of lists, maps, and so on. I am teaching Comparative Semitic Linguistics this semester, and I recommended the book to my students since all of them know some Semitic languages (Akkadian, Hebrew, Aramaic, etc.) but few are familiar enough with historical linguistics. I hope I didn't skew with my reply, and this is, more or less, what Larry Trask was thinking of. ____________________________ Gonzalo Rubio Near Eastern Studies The Johns Hopkins University Baltimore MD-21218 gonzalor at jhu.edu ____________________________ From dyen at hawaii.edu Thu Oct 1 22:41:57 1998 From: dyen at hawaii.edu (Isidore Dyen) Date: Thu, 1 Oct 1998 18:41:57 EDT Subject: Q: oblique cognates In-Reply-To: Message-ID: ----------------------------Original message---------------------------- It appears that you have run up against the distinction between lexical cognation, the cognation between words implying the existence of a word or d lexeme in their last common protolanguage on the one hand and root cognation or perhaps better put, morpheme cognation in their last common protolanguage. What is causing some confusion is that morpheme cognation is likely to have had a longer history or past life than a lexical cognation. To turn to your 'tooth' example, there appears little reason to doubt that the forms cited continue in some way what was in the protolanguage a single lexeme whose base varied in different inflectional combinations in terms of which later different stages (either dialects or daughter languages) reached their own reorganizations into inflections through analogical changes. The results of such changes can be said to be obliquely cognate or cognate in any other way that one likes, but it is best to keep in mind that they are each aproduct of different uninterrupted contiuities. On Wed, 30 Sep 1998, Larry Trask wrote: > ----------------------------Original message---------------------------- > I am looking for a term for a certain non-canonical type of > cognation. > > One non-canonical variety is as follows. Latin `tooth' > requires a PIE *. English `tooth' and Greek require > a PIE *. Gothic requires a PIE *. The several > forms are therefore not strictly descended from a single ancestral > form, but rather from variant forms of a single root. Such forms as > the Latin, English and Gothic ones have been called `oblique > cognates' in the literature. Fine. > > But there's another case. English `head' is directly cognate with > Latin `head'. However, Spanish does not descend > directly from , but rather from a suffixed derivative of this. > Therefore the English and Spanish words are not directly cognate, > even though they are indirectly cognate in an important way. Is > there a label for this kind of cognation? What would you prefer to > call the relationship between the English and Spanish words? > > Larry Trask > COGS > University of Sussex > Brighton BN1 9QH > England > > larryt at cogs.susx.ac.uk > From Roger.Wright at liverpool.ac.uk Fri Oct 2 11:57:06 1998 From: Roger.Wright at liverpool.ac.uk (Roger Wright) Date: Fri, 2 Oct 1998 07:57:06 EDT Subject: Q: oblique cognates In-Reply-To: Message-ID: ----------------------------Original message---------------------------- In cases like Spanish *cabeza* ("head") - Could we just say that the root morpheme is cognate with the others? That is, Latin *caput* and English *head*; French *chef*, Italian *capo* and Spanish *cabo* ("end") are all cognate; and so is the root morpheme *cab-* of Spanish *cabeza*. RW (Spanish *cabeza* was invented, because of the semantic change that had overcome *cabo* < CAPUT, as a word specifically meaning "head"; the *-eza* was available in the suffix inventory, but seems to have meant nothing at all, and just to have been used as a formal device to create a form different from *cabo*.) On Wed, 30 Sep 1998, Larry Trask wrote: >But there's another case. English `head' is directly cognate with >Latin `head'. However, Spanish does not descend >directly from , but rather from a suffixed derivative of this. >Therefore the English and Spanish words are not directly cognate, >even though they are indirectly cognate in an important way. Is >there a label for this kind of cognation? What would you prefer to >call the relationship between the English and Spanish words? From bwald at HUMnet.UCLA.EDU Tue Oct 6 02:38:42 1998 From: bwald at HUMnet.UCLA.EDU (bwald) Date: Mon, 5 Oct 1998 22:38:42 EDT Subject: Cladistic language concepts Message-ID: ----------------------------Original message---------------------------- Sorry I did not have time to reply to Isidore Dyen's message sooner. He writes: >one of the main factors in linguistic change and >perhaps the main factor is the drive for efficiency in communication which >is dialectically resisted by the need for clarity so that efficiency does >not actually increase and the actual level of clarity does not change. I think there is good reason to keep these two traditional factors in mind in any attempt to understand constraints on linguistic change *overall*. There are, however, other factors involved in promoting change beside efficiency of communication, if that is intended to refer specifically to communication of *referential* information. There is also the factor of maintaining some kind of (local) social identity, viewable as a shorthand for shared knowledge of the local society, and therefore contributing to communication (esp what can be understood within the social group *without having to be said*). An important issue is the extent to which the continual nature of the (socio)linguistic reformation of social identity, as society continually changes, promotes linguistic change, even overriding wider concerns with communication of referential information. This may not be a constant at all times. Maybe sometimes it is more active than at other. It is not clear that there is a constant balance between social factors promoting linguistic change and the counter-pull of intelligibility (to whom? ALL speakers of the "language"? Doubtful.) One possibility is that there are constantly at least TWO regsiters (or styles or whatever) for all mature speakers, one for wider communication, e.g., standards and other lingua francas, and one more susceptible to linguistic change due to the local social identity effect (usually called the "dialect" or "vernacular"). The more lingua-franca like register is also susceptible to change, not to mention interpenetration of what might be viewed as (at least) two registers/styles. However, it remains to be seen how change in one register affects change in another. In any case, it affects the relation between the two more traditional factors maintaining "balance" between "efficiency" and "intelligibility" remains an issue in need of more consideration in understanding the shaping of linguistic change. ID continues: >I believe your characterization of mutual intelligibility as being an >arbitrary criterion is a misconception. After all it concerns >intercommunication, the primary function of language. The difficulty with >mutual intelligibility lies rather in applying it and improvements in that >area could be achieved if the importance of distinguishing languages from >each other could reach the level of attracting financial support. I wouldn't dispute that. It's worth noting, however, that intercommunication as the primary function of language should lead to more concern to promote multilingualism than it does in societies such as the US, among others. This tells us something more generalisable about concerns with intercommunication and its mitigation by the social identity factor. Even so, multilingualism attracts more financial support than mutual intelligibility within what is considered one language. The Ebonics controversy was an interesting issue in trying to move a variety (of language) from one category (dialect variation) to another (multilingualism). ID concludes: >As for entropy it could not be expected to be found in the structure of a >language since the energy input to maintain clarity prevents observable >change in the direction of disorganization. However within a language >regarded as a closed system, there is (for all practical purposes) >observable changes in the direction of disorganization in >dialectalization as diversification tending toward the shattering of a >language. The opposing force is the rate of interlocution; as that rate >is high it militates against diversification and if it is high >enough, promotes homogeneity and when it is low or decreases is >accompanied by increased dialectalization and if it reaches zero, may >be followed by language fission. I found the entropy analogy attractive when I read it in an earlier message. The rate of interlocution is a more complex notion. It may not only be frequency, but also, in some currently hard to specify sense, diversity and/or quality of communication among different groups of interlocutors. Note that *inter*locution already dismisses passive absorption of linguistic norms from the media, whether TV, books or whatever, in favor of mutual communication, even though passive absorption of media norms may be more frequent than active communication for many speakers. One must seek to understand why this is so, to the (large) extent that it is. I still maintain that measures of mutual intelligibility are difficult to interpret in the real world because linguistic distance is not the only factor involved in intelligibility in the real world. Familiarity and motivation to understand (just as in the case of unquestionable bilingualism, and either one, arguably, even without motivation to imitate (accomodate?)) are additional factors. From larryt at cogs.susx.ac.uk Wed Oct 7 15:30:04 1998 From: larryt at cogs.susx.ac.uk (Larry Trask) Date: Wed, 7 Oct 1998 11:30:04 EDT Subject: Sum: `oblique cognates' Message-ID: ----------------------------Original message---------------------------- A few days ago I posted a query about terminology for labeling certain kinds of indirect cognates. I received about a dozen responses from eight different people. The terms put forward are summarized below. Most of them have apparently appeared in print, though not all. (1) `direct cognates': forms which are fully cognate in every detail, and are independently descended from a single reconstructible ancestral form. (2) `lexical cognates': direct cognates (in the above sense) which are word-forms. (3) `paradigmatic cognates': forms which are directly descended, without additional material, from paradigmatic alternants of a single item: what I called `oblique cognates', as in the IE words for `tooth'. (4) `doublets': etymologically related forms descended from variants of a single proto-morpheme. (3) and (4) are perhaps identical. (5) `derivational cognates': forms which are derived by morphological processes from directly cognate stems but which may contain additional non-cognate morphological material of a derivational nature. (6) `partial cognates': forms which contain cognate material but at least some of which contain further material which is not cognate. (7) `morpheme cognates': forms which share at least one cognate morpheme but which also contain additional material which is not cognate (at least in some forms). (8) `root cognates': forms whose roots are cognate but which contain additional material which is not cognate (at least in some forms). (9) `root etymology' (German `Wurzeletymologie'): same as the preceding, but often used in the past as a dismissive term indicating skepticism. (6) to (9), and perhaps also (5). are very similar, and in some cases identical. (10) `deformed cognates': forms which are clearly of a common origin but which require irregularly different etyma, such as the IE forms variously requiring PIE * or * `wolf, fox', particularly when it can be shown that one of the reconstructed forms is original and the other has arisen from an irregular development. These might also be termed `oblique cognates' if the original form cannot be identified. (11) `skewed cognates' or `skewed correspondences': forms which are clearly cognate but which exhibit irregular developments or correspondences. (12) `word family': a set of items which clearly contain cognate material (especially cognate roots) but whose precise relation is obscure. (13) `allofams': the members of such a word family. I'll try to organize these terms as best I can in my dictionary. Further comments welcome. My thanks to Richard Coates, Isidore Dyen, Ralf-Stefan Georg, Martin Huld, Carol Justus, Harold Koch, Gonzalo Rubio, and Roger Wright. Larry Trask COGS University of Sussex Brighton BN1 9QH UK larryt at cogs.susx.ac.uk From dyen at hawaii.edu Fri Oct 9 19:34:47 1998 From: dyen at hawaii.edu (Isidore Dyen) Date: Fri, 9 Oct 1998 15:34:47 EDT Subject: Cladistic language concepts In-Reply-To: Message-ID: ----------------------------Original message---------------------------- I will take up only one point, the one regarding social resistance to change based on the desire to maintain identity. The facts seem to indicate that such resistance is usually coupled with some other basis of resistance such as a political, religious, or economic motivation. The omplication is that such resistance tends to disappear with the disappearance of the motivation. I am not sure that a case can be cited of such resistance apart from such extraneous considerations, but perhaps you have examples. On Mon, 5 Oct 1998, bwald wrote: > Sorry I did not have time to reply to Isidore Dyen's message sooner. He writes: > > >one of the main factors in linguistic change and > >perhaps the main factor is the drive for efficiency in communication which > >is dialectically resisted by the need for clarity so that efficiency does > >not actually increase and the actual level of clarity does not change. > > I think there is good reason to keep these two traditional factors in mind > in any attempt to understand constraints on linguistic change *overall*. > There are, however, other factors involved in promoting change beside > efficiency of communication, if that is intended to refer specifically to > communication of *referential* information. There is also the factor of > maintaining some kind of (local) social identity, viewable as a shorthand > for shared knowledge of the local society, and therefore contributing to > communication (esp what can be understood within the social group *without > having to be said*). An important issue is the extent to which the > continual nature of the (socio)linguistic reformation of social identity, > as society continually changes, promotes linguistic change, even overriding > wider concerns with communication of referential information. This may not > be a constant at all times. Maybe sometimes it is more active than at > other. It is not clear that there is a constant balance between social > factors promoting linguistic change and the counter-pull of intelligibility > (to whom? ALL speakers of the "language"? Doubtful.) > > One possibility is that there are constantly at least TWO regsiters (or > styles or whatever) for all mature speakers, one for wider communication, > e.g., standards and other lingua francas, and one more susceptible to > linguistic change due to the local social identity effect (usually called > the "dialect" or "vernacular"). The more lingua-franca like register is > also susceptible to change, not to mention interpenetration of what might > be viewed as (at least) two registers/styles. However, it remains to be > seen how change in one register affects change in another. In any case, it > affects the relation between the two more traditional factors maintaining > "balance" between "efficiency" and "intelligibility" remains an issue in > need of more consideration in understanding the shaping of linguistic > change. > > ID continues: > > >I believe your characterization of mutual intelligibility as being an > >arbitrary criterion is a misconception. After all it concerns > >intercommunication, the primary function of language. The difficulty with > >mutual intelligibility lies rather in applying it and improvements in that > >area could be achieved if the importance of distinguishing languages from > >each other could reach the level of attracting financial support. > > I wouldn't dispute that. It's worth noting, however, that > intercommunication as the primary function of language should lead to more > concern to promote multilingualism than it does in societies such as the > US, among others. This tells us something more generalisable about > concerns with intercommunication and its mitigation by the social identity > factor. Even so, multilingualism attracts more financial support than > mutual intelligibility within what is considered one language. The Ebonics > controversy was an interesting issue in trying to move a variety (of > language) from one category (dialect variation) to another > (multilingualism). > > ID concludes: > > >As for entropy it could not be expected to be found in the structure of a > >language since the energy input to maintain clarity prevents observable > >change in the direction of disorganization. However within a language > >regarded as a closed system, there is (for all practical purposes) > >observable changes in the direction of disorganization in > >dialectalization as diversification tending toward the shattering of a > >language. The opposing force is the rate of interlocution; as that rate > >is high it militates against diversification and if it is high > >enough, promotes homogeneity and when it is low or decreases is > >accompanied by increased dialectalization and if it reaches zero, may > >be followed by language fission. > > I found the entropy analogy attractive when I read it in an earlier > message. The rate of interlocution is a more complex notion. It may not > only be frequency, but also, in some currently hard to specify sense, > diversity and/or quality of communication among different groups of > interlocutors. Note that *inter*locution already dismisses passive > absorption of linguistic norms from the media, whether TV, books or > whatever, in favor of mutual communication, even though passive absorption > of media norms may be more frequent than active communication for many > speakers. One must seek to understand why this is so, to the (large) > extent that it is. > > I still maintain that measures of mutual intelligibility are difficult to > interpret in the real world because linguistic distance is not the only > factor involved in intelligibility in the real world. Familiarity and > motivation to understand (just as in the case of unquestionable > bilingualism, and either one, arguably, even without motivation to imitate > (accomodate?)) are additional factors. > > > From hubeyh at montclair.edu Sat Oct 10 17:16:21 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Sat, 10 Oct 1998 13:16:21 EDT Subject: Cladistic language concepts Message-ID: ----------------------------Original message---------------------------- > On Mon, 5 Oct 1998, bwald wrote: > > > Sorry I did not have time to reply to Isidore Dyen's message sooner. He > writes: > > > > >one of the main factors in linguistic change and > > >perhaps the main factor is the drive for efficiency in communication which > > >is dialectically resisted by the need for clarity so that efficiency does > > >not actually increase and the actual level of clarity does not change. > > > > I think there is good reason to keep these two traditional factors in mind > > in any attempt to understand constraints on linguistic change *overall*. > > There are, however, other factors involved in promoting change beside > > efficiency of communication, if that is intended to refer specifically to > > communication of *referential* information. There is also the factor of > > maintaining some kind of (local) social identity, viewable as a shorthand > > for shared knowledge of the local society, and therefore contributing to > > communication (esp what can be understood within the social group *without > > having to be said*). An important issue is the extent to which the If the factors all do not operate on the same time scales, the effects will not necessarily cancel out or mutually re-enforce each other. A good example of a similar problem is in the Brownian motion problem solve in the early part of this century. Atoms of a liquid are in continual motion but we cannot see them. The time scale in which they act/react/move is also very short compared to motion at slightly larger scales. The motion we can observe (with a microscope) is the motion of larger objects (like dust particles). The motion of these particles is due to correlated motion of atoms. The basic idea is that like throwing up 1 million pennies. Almost all the time, about half will be heads, and half tails. Similarly of all the atoms banging into these particles, about half will be in one direction and half in the other so that their combined effects will cancel out and there will be no observable motion of the particles. However, just as there will be cases in which about 900,000 coins can be Heads or Tails, there will instants in time in which most of the atoms will be moving in one direction (which is the "correlated motion" of the huge ensemble of atoms) and that effect will be seen in the motion of the particles suspended in the liquid. That motion is Brownian motion. IT is still random just like the underlying random motion of the atoms. But the fluctuations of the atoms cancel out almost all the time at their own time scales. The forces which have a propensity to create linguistic change also occur at various temporal and spatial scales. The rapid and local fluctuations in speech do not normally have permanent global effects. Every once in a while, there will a larger scale changes correlated in space and time, and it is those changes that we track in historical linguistics. IT would probably be a good idea to categorize the changes mentioned into different classes based on time and space scales. -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From bwald at HUMnet.UCLA.EDU Sat Oct 10 17:16:49 1998 From: bwald at HUMnet.UCLA.EDU (bwald) Date: Sat, 10 Oct 1998 13:16:49 EDT Subject: Cladistic language concepts Message-ID: ----------------------------Original message---------------------------- In response to my last message, Isidore Dyen wrote: >I will take up only one point, the one regarding social resistance to change based on the desire to maintain identity. The facts seem to indicate that such resistance is usually coupled with some other basis of resistance such as a political, religious, or economic motivation. The omplication is that such resistance tends to disappear with the disappearance of the motivation. I am not sure that a case can be cited of such resistance apart from such extraneous considerations, but >perhaps you have examples. I did not mean to focus on social resistance to change (which I take to mean change emanating from outside the local community), though that also happens to the extent that vernaculars/dialects endure. I was focussing on the opposite, where local identity (or, more accurately, *local interests*) promotes change (encourages change originating within the community). The general idea is that groups or networks within a local community are continually affected by changes in the social composition of that community, often immigration of new populations into the community. As a matter of course, the newer populations most often assimilate to the linguistic norms of the local community. However, older segments of the community already have established networks, etc., and they are often relatively closed to members of the newer groups. Linguistically, then, the older networks are propelled to change some features of their local language in order to preserve their more exclusive networks. All this happens below the level of consciousness (the conscious analog would be how "slang" continually changes in an "in-group" as its older norms spread to outgroups who adopt it for some reason or other). I am necessarily oversimplifying the nature of social change and how it effects (and affects) linguistic change, but there is a large (socio)linguistic literature illustrating the point I am making, starting with (actually predating, with less rigorous methods of demonstration) Labov's study of linguistic (phonetic) changes among groups on Martha's Vineyard (ca. 1963). The focus, then, is that resistance to changes in the (previous) social structure of a LOCAL community actually promotes (rather than retards) linguistic change. ID is quite right that in the long run a specific social motivation and alignment of interest groups will change, but in the meantime it has its effect on the changes to which the local language is subjected, these are often irreversible, they do not consider mutual intelligibility with external groups (presumably their effect is contrary to such a concern -- ultimately because it weakens what can be understood locally without having to be said, the ultimate in communicative efficiency), and finally, for all intents and purposes, there are almost ALWAYS, i.e., at ANY time, some social changes within any community promoting linguistic changes to differentiate various local interest groups, even though those interest groups continually change and realign in the long run. The last point -- that changes in the social structure of a local community are virtually always going on -- vitiates the notion that social factors are "extraneous", if this is intended to imply that in the long run they can be discounted or are minor in considering whether the two more traditionally acknowledged factors of the balance between "(linguistic) effort" and "intelligibility" are sufficient to account for the direction of (cumulative) linguistic change within a "language". On the contrary, up to the present, history indicates the continual long-term fragmentation of languages and destruction of mutual intelligibility. The vanescence and short-sightedness of LOCAL SOCIAL factors promoting any particular linguistic change (operating below the level of consciousness) no doubt accounts for most of the fragmentation of languages and loss of mutual intelligibility (among the fragments) in the long run. In view of the above, it seems to me that any "balancing" pairs or sets of changes to preserve mutual intelligibility are independent factors which apply primarily WITHIN any particular LOCAL variety of a language, with *questionable* inevitability, and have little if any effect on the eventual fragmentation of a language into mutually unintelligible newer "languages", with (virtually) *unquestionable* inevitability. We are certainly interested in these balancing or "equilibrium-maintaining" mechanisms, but they seem to be quite separate from possibly short-term constantly changing social factors as the MAJOR factor in linguistic fragmentation and loss of mutual intelligibility across local communities. From Roger.Wright at liverpool.ac.uk Mon Oct 12 13:08:12 1998 From: Roger.Wright at liverpool.ac.uk (Roger Wright) Date: Mon, 12 Oct 1998 09:08:12 EDT Subject: Cladistic language concepts In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Benji Wald says: >On the contrary, up >to the present, history indicates the continual long-term fragmentation of >languages and destruction of mutual intelligibility. Yes; nicely put. "Up to the present" -- There's a good case for saying this may never happen again. RW From Roger.Wright at liverpool.ac.uk Mon Oct 12 13:07:58 1998 From: Roger.Wright at liverpool.ac.uk (Roger Wright) Date: Mon, 12 Oct 1998 09:07:58 EDT Subject: Cladistic language concepts In-Reply-To: Message-ID: ----------------------------Original message---------------------------- On Fri, 9 Oct 1998, Isidore Dyen wrote: >----------------------------Original message---------------------------- >I will take up only one point, the one regarding social resistance to >change based on the desire to maintain identity. The facts seem to >indicate that such resistance is usually coupled with some other basis of >resistance such as a political, religious, or economic motivation. The >complication is that such resistance tends to disappear with the >disappearance of the motivation. Sometimes it seems, to us looking back, like unnecessary resistance, at other times it seems like unnecessary innovation. But at the period concerned, it usually takes the form of a motivated choice between existing variants. Malkiel, for example, pointed to several cases in which sixteenth-century Portuguese had two variants available (in morphology or phonetics, but it also applies to vocabulary), both indigenous, and they - perhaps consciously - chose the one that was least like Spanish, asserting their identity that way. Well, sometimes they chose the least evolved variant (thereby appearing to be resistant to change) and sometimes thereby they chose the most evolved (thereby appearing to be changing more than expected) but in neither case was the choice of variant based on relative archaism. But it doesn't look as if they chose forms, archaic or innovative, that didn't actually exist at all at the time; it was a choice between already existing variants. This train of thought and motivated choice of variants is still happening within the non-Castilian regions of Spain, particularly in vocabulary; where Catalan, or Galician, has two words that are for practical purposes synonymous, one of which is like the Castilian word for the same meaning and the other of which is not, the dictionaries and the standardizers of Catalan and of Galician have tended to prefer the one that isn't like Castilian. (In order to annoy the Castilians, but of course in practice it just annoys many Catalans and Galicians, since the Castilians couldn't care less). This is known as "diferencialismo". RW From dyen at hawaii.edu Tue Oct 13 12:40:28 1998 From: dyen at hawaii.edu (Isidore Dyen) Date: Tue, 13 Oct 1998 08:40:28 EDT Subject: Cladistic language concepts In-Reply-To: <361E906D.4C61ACB8@montclair.edu> Message-ID: ----------------------------Original message---------------------------- I like yor contribution to the discussion. When I spoke of a main factor in linguistic change, I was trying to get at the notion that there will be linguistic change regardless of extraneous factor s such as those that bwald is insisting on. The point is, as I see it, that linguistic change is built into the way the community interacts with its language, whereas some aspects of linguistic change are conditioned by the social changes that are going on in the community. The latter type of change, since it is local and temporary I thought could be excluded from being regarded as a 'main factor', but I suppose it gets to be a matter of defintion. The advantage of looking at the matter the way I have been doing it is that the effects of linguistic change in the fragmentation of a linguistic community can be related to the rate of intercommunication among the speakers. Of course this rate is itself dependent on social factors, but can be regarded apart from those factors as a matter of objective observation, even though I would not seriously recommend that anyone undertake to do it, even Labov. In the absence of other factors, we would expect a community to become linguistically disparate and finally mutually unintelligible if it were to separate into two sets that did not intercommunicate with each other for some long period, say a thousand years. Social factors other than the separation could be disregarded even if they could not be actually excluded, human beings being what they are. On Sat, 10 Oct 1998, H.M.Hubey wrote: > ----------------------------Original message---------------------------- > > On Mon, 5 Oct 1998, bwald wrote: > > > > > Sorry I did not have time to reply to Isidore Dyen's message sooner. He > > writes: > > > > > > >one of the main factors in linguistic change and > > > >perhaps the main factor is the drive for efficiency in communication which > > > >is dialectically resisted by the need for clarity so that efficiency does > > > >not actually increase and the actual level of clarity does not change. > > > > > > I think there is good reason to keep these two traditional factors in mind > > > in any attempt to understand constraints on linguistic change *overall*. > > > There are, however, other factors involved in promoting change beside > > > efficiency of communication, if that is intended to refer specifically to > > > communication of *referential* information. There is also the factor of > > > maintaining some kind of (local) social identity, viewable as a shorthand > > > for shared knowledge of the local society, and therefore contributing to > > > communication (esp what can be understood within the social group *without > > > having to be said*). An important issue is the extent to which the > > If the factors all do not operate on the same time scales, the effects > will > not necessarily cancel out or mutually re-enforce each other. > > A good example of a similar problem is in the Brownian motion problem > solve in the early part of this century. Atoms of a liquid are in > continual motion but we cannot see them. The time scale in which they > act/react/move is also very short compared to motion at slightly larger > scales. The motion we can observe (with a microscope) is the motion of > larger objects (like dust particles). The motion of these particles is > due to correlated motion of atoms. The basic idea is that like throwing > up 1 million pennies. Almost all the time, about half will be heads, and > half tails. Similarly of all the atoms banging into these particles, > about half will be in one direction and half in the other so that their > combined effects will cancel out and there will be no observable motion > of the particles. However, just as there will be cases in which about > 900,000 coins can be Heads or Tails, there will instants in time in > which most of the atoms will be moving in one direction (which is the > "correlated motion" of the huge ensemble of atoms) and that effect will > be seen in the motion of the particles suspended in the liquid. That > motion is Brownian motion. IT is still random just like the underlying > random motion of the atoms. But the fluctuations of the atoms cancel out > almost all the time at their own time scales. > > The forces which have a propensity to create linguistic change also > occur at various temporal and spatial scales. The rapid and local > fluctuations in speech do not normally have permanent global effects. > Every once in a while, there will a larger scale changes correlated in > space and time, and it is those changes that we track in historical > linguistics. IT would probably be a good idea to categorize the changes > mentioned into different classes based on time and space scales. > > -- > Best Regards, > Mark > -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= > hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey > =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= > The information transmitted is intended only for the person or entity > to which it is addressed and may contain confidential and/or privileged > material. Any review, retransmission, dissemination or other use of, > or taking of any action in reliance upon, this information by persons > or entities other than the intended recipient is prohibited. If you > received this in error, please contact the sender and delete the > material from any computer. > =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= > From eirikur at rhi.hi.is Tue Oct 13 12:41:38 1998 From: eirikur at rhi.hi.is (Eirikur Rognvaldsson) Date: Tue, 13 Oct 1998 08:41:38 EDT Subject: New publication Message-ID: ----------------------------Original message---------------------------- New publication: Linguistic Studies Historical and Comparative by Hreinn Benediktsson On the occasion of Professor Hreinn Benediktsson's 70th birthday, October 10th, 1998, the Institute of Linguistics, University of Iceland will publish a collection of his papers (two volumes, a total of approx. 700 pages). Most of the papers have appeared elsewhere, but the papers that were not published originally in English have been translated into English for this publication and all the papers have been reedited and retypeset. In addition, the collection contains an Introduction, a general bibliography and an index for all the papers. This edition is being offered at a special pre-publication price to those subscribing before October 25th, 1998. More detailed information, together with the table of contents and a subscription form, can be found at the following website: http://www.hi.is/~nordconf/lingstud.htm --------------------------------------------------------------------- Eirikur Rognvaldsson voice: +354-525-4403 Department of Icelandic fax: +354-525-4242 University of Iceland e-mail: eirikur at rhi.hi.is IS-101 Reykjavik, Iceland URL: http://www.rhi.hi.is/~eirikur From dyen at hawaii.edu Tue Oct 13 21:40:51 1998 From: dyen at hawaii.edu (Isidore Dyen) Date: Tue, 13 Oct 1998 17:40:51 EDT Subject: Cladistic language concepts In-Reply-To: <9807129029.AA902948340@casmail.calacademy.org> Message-ID: ----------------------------Original message---------------------------- Please forgive me for breaking into this discussion of cladistic concepts and linguistics. There is a very sharp difference between the characters of cladistics and the cognates of genetic linguistics. Homologous characters, if I understand cladistic talk, are features that are candidates for having the same origin in some common ancestor. There three analogous structures that I am aware of and that deal with procedures for inferring genetic evolution. One of these is the language family-tree, and the other two are respectively the cladistics of biology and the use of cultural traits in anthropology. 'Traitisitics' has a poor reputation and deservedly so. What cladistics and the linguistic family-tree have in common is the assumption that their basic units do not mix. Once a species or a language becomes distinct, that distinction is indissoluble. Where traits are concerned, such an assumption would lead nowhere because there are too many instances in which cultures have amalgamated in a way that precludes attributing the continuity to just one of the contributing cultures. In other words, cultures do not behave that way. As for cladistics the difficulty surrounds th issue of definable characters. Although broadly the difference between characters is usually easy to see, the discrimination between differences and identities in extreme cases seem to present difficulties. Now that there is competition from the study of DNA, it seems likely that cladistics wil serve to be ancillary support to inferences from the more intrinsic DNA agreements and differences. As for linguistics, it has the advantage of the comparative method which is based on the inferences that can be drawn from the systematic correspondences between phonemes that characterize related languages and raise the likelihood of inferences based on those correspondences, thus giving the inferences scientific standing. On Wed, 12 Aug 1998, Ghiselin, Michael wrote: > ----------------------------Original message---------------------------- > Dear Dr. DeLancey: > Thank you very much for casting your vote. It would be > nice if more linguists would do the same because the sample > as it exists is small and perhaps not representative. > In spite of that the preliminary results are very > interesting. Not only has a cladistic language concept been > generally presupposed but, as you say, linguists do not even > consider the topic particularly interesting. Why should > this be? One possibility is that where linguists have a > written record it lacks the fragmentary nature of the fossil > record that results from accidents of preservation and the > like. Another is that what the linguists do perceive as > important is trying to find older and older common > ancestries and the genealogical relationships are all that > they need. Linguists do not have the elaborate system of > categories, such as phylum, class, order etc., that we > zoologists do. And unless I am mistaken (please correct me > if I am) they do not believe that there are important > differences that need to be expressed by giving a taxon a > higher rank, as when our own species has been put in a > separate order or even kingdom. > Linguists must have methodological problems with > respect to paraphyly, parallelism and convergence. But so > far as I can tell, they treat these as problems to be > overcome in reaching a strictly genealogical arrangement. > MG > From dyen at hawaii.edu Tue Oct 13 21:41:21 1998 From: dyen at hawaii.edu (Isidore Dyen) Date: Tue, 13 Oct 1998 17:41:21 EDT Subject: Cladistic language concepts In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Your point is interesting and relevant, but the question would remain whether Malkiel or anyone else for all of their approach to omniscient are able to control all the facts. What is at issue for theories concerning linguistic change is what is intrinsic to natural language as an instrument and what is extrinsic. This question concerns, I suppose, what we would expect to happen in a language in the absence of abnormal social events like conquests and enslavements, restrictive laws including persecutions, that sort of thing. Under the least interference I would expect linguistic change to go on, unevenly in proportion to the extent of the community with homogeneity depending on the rate of interlocution ( or intercommunication). Such a theory which distinguishes between linguistic change that is inherent and that which reflects social stress may suffer because the distinction between the two effects cannot always be made and perhaps can never be made, but would have the value of recognizing one of the components or factors in linguistic change. Resistance to maintain identity comes into play when the threat to the identity involves some loss and in this sense the insistence on the identity is a pursuit of advantage. Although this pursuit is omnipresent, questions of identity do not usually arise without association with loss or gain. Perhaps what I am trying to say is that in a community in which threats or advantages related to identity did not arise, lingusitic change would still go on and in this sense such threats and advantages are extraneous social factors, no matter how close to universal. On Mon, 12 Oct 1998, Roger Wright wrote: > ----------------------------Original message---------------------------- > > > On Fri, 9 Oct 1998, Isidore Dyen wrote: > > >----------------------------Original message---------------------------- > >I will take up only one point, the one regarding social resistance to > >change based on the desire to maintain identity. The facts seem to > >indicate that such resistance is usually coupled with some other basis of > >resistance such as a political, religious, or economic motivation. The > >complication is that such resistance tends to disappear with the > >disappearance of the motivation. > > > Sometimes it seems, to us looking back, like unnecessary resistance, at > other times it seems like unnecessary innovation. But at the period > concerned, it usually takes the form of a motivated choice between > existing variants. Malkiel, for example, pointed to several cases in > which sixteenth-century Portuguese had two variants available (in > morphology or phonetics, but it also applies to vocabulary), both > indigenous, and they - perhaps consciously - chose the one that was > least like Spanish, asserting their identity that way. Well, sometimes > they chose the least evolved variant (thereby appearing to be resistant > to change) and sometimes thereby they chose the most evolved (thereby > appearing to be changing more than expected) but in neither case was the > choice of variant based on relative archaism. But it doesn't look as > if they chose forms, archaic or innovative, that didn't actually > exist at all at the time; it was a choice between already existing > variants. This train of thought and motivated choice of variants is > still happening within the non-Castilian regions of Spain, particularly > in vocabulary; where Catalan, or Galician, has two words that are for > practical purposes synonymous, one of which is like the Castilian word > for the same meaning and the other of which is not, the dictionaries and > the standardizers of Catalan and of Galician have tended to prefer the > one that isn't like Castilian. (In order to annoy the Castilians, but of > course in practice it just annoys many Catalans and Galicians, since the > Castilians couldn't care less). This is known as "diferencialismo". > RW > From maxw at cogs.susx.ac.uk Wed Oct 14 15:20:09 1998 From: maxw at cogs.susx.ac.uk (Max W Wheeler) Date: Wed, 14 Oct 1998 11:20:09 EDT Subject: Cladistic language concepts In-Reply-To: Message-ID: ----------------------------Original message---------------------------- On Mon, 12 Oct 1998, Roger Wright wrote: > > choice of variant based on relative archaism. But it doesn't look as > if they chose forms, archaic or innovative, that didn't actually > exist at all at the time; it was a choice between already existing > variants. This train of thought and motivated choice of variants is > still happening within the non-Castilian regions of Spain, particularly > in vocabulary; where Catalan, or Galician, has two words that are for > practical purposes synonymous, one of which is like the Castilian word > for the same meaning and the other of which is not, the dictionaries and > the standardizers of Catalan and of Galician have tended to prefer the > one that isn't like Castilian. (In order to annoy the Castilians, but of > course in practice it just annoys many Catalans and Galicians, since the > Castilians couldn't care less). This is known as "diferencialismo". > RW Can you actually substantiate this claim, with regard to Catalan? It is true that such an effect may appear to someone who doesn't know about the history of the languages (obviously not you); in several cases standardizers prefer a form with a longer established literary tradition. Note that people from French Catalonia have accused the standardizers of preferring those forms which were MORE similar to Spanish. Max Wheeler ___________________________________________________________________________ Max W. Wheeler School of Cognitive and Computing Sciences University of Sussex, Falmer, Brighton BN1 9QH, UK Tel: +44 (0)1273 678975; fax: +44 (0)1273 671320 ___________________________________________________________________________ From bwald at HUMnet.UCLA.EDU Thu Oct 15 15:43:41 1998 From: bwald at HUMnet.UCLA.EDU (bwald) Date: Thu, 15 Oct 1998 11:43:41 EDT Subject: Cladistic language concepts Message-ID: ----------------------------Original message---------------------------- Roger Wright quotes from my last message. >Benji Wald says: > >>On the contrary, up >>to the present, history indicates the continual long-term fragmentation of >>languages and destruction of mutual intelligibility. >Yes; nicely put. "Up to the present" -- >There's a good case for saying this may never happen again. I'm glad that Roger appreciated the qualification I put on what I was saying. I deleted a further paragraph on how I invite such views as he is proposing about a sharp discontinuity between the past and the future, since I see little reason to suppose that such a discontinuity has come about in the 20th c. or will in the foreseeable future, despite impressive advances in communication technology (at least impressive to us current beings) and increasing sharing of various kinds of literacies. Meanwhile, the same old problems of miscommunication that have always existed (and have occasionally been reported in the past) persist. (Would it be surprising if "human nature" exists not only in our linguistic devices but in how we use them?) I think the main thing that would interest me among Roger's proposals is what changes in the nature of human society (as a whole or in its various various parts) he envisages or suggests to have relatively recently arrived which will override the steady and unrelenting effects of social localism that in the past (presumably) have been the major cause of linguistic fragmentation and loss of mutual intelligibility. I am skeptical, and suspect he is underrating the long-term cumulative effects of localism, and overrating the stability of centralised power, but I am open to hearing interesting proposals about the "changing" relation of social change to linguistic change. With regard to Isidore's latest message, I share his appreciation for Hubey's comments, but paused at the following passage: >The point is, as I see it, that linguistic change is built into the way the community interacts with its language, whereas some aspects of linguistic change are conditioned by the social changes that are going on in the community. The latter type of change, since it is local and temporary I thought could be excluded from being regarded as a 'main factor', but I suppose it gets to be a matter of defintion. I'm not sure I understand the intent of "local and temporary". The "temporary" part seems to suggest that the local changes are eventually undone, as if afterwards they seem to have never occurred (i.e., no *lasting* harm done to mutual intelligibility). If that is not what is meant, then they have had their effect in changing the local language AWAY from other local varieties. This seems more than a matter of definition (of "main factor"?) to me, but of the cumulative consequences of local and temporary (temporally bounded?) changes. Of course, most changes do spread beyond the temporary local interest group that intiates them, so their effects don't go away but continue, and establish the typical mosaic patterns we commonly find in dialect geography. I cannot address all fronts at once, so when I emphasised fragmentation and loss of mutual intelligibility as the traditional and still usual focus of historical linguistics (at least on the elementary level), I did not complicate that traditional picture with diffusion and convergence which takes place across languages as well as in them, due to bilingualism, bidialectalism, register/stylistic complexity, or whatever level of analysis is appropriate at the time. I am continually struck by cases in which it is not clear which family or (even more commonly *in the literature*) branch of a family some language (group) or other belongs to because of convergence and sharing of features. Such things most strikingly occur at the margin of isogloss bundles. Isidore (and no doubt Roger) is certainly right that channels and interests promoting communication ("mutual" intelligibility) *across* local groups is also a factor in change. The issue remains: will languages continue to fragment and produce mutual unintelligbility, or will the whole world eventually abandon its local variety in favor of some "homogeneous commercial or totalitarian English" or whatever (no doubt incomprehensible to us current beings). I think not. From mcv at wxs.nl Thu Oct 15 15:38:30 1998 From: mcv at wxs.nl (Miguel Carrasquer Vidal) Date: Thu, 15 Oct 1998 11:38:30 EDT Subject: Cladistic language concepts In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Max W Wheeler wrote: >On Mon, 12 Oct 1998, Roger Wright wrote: > >> where Catalan, or Galician, has two words that are for >> practical purposes synonymous, one of which is like the Castilian word >> for the same meaning and the other of which is not, the dictionaries and >> the standardizers of Catalan and of Galician have tended to prefer the >> one that isn't like Castilian. (In order to annoy the Castilians, but of >> course in practice it just annoys many Catalans and Galicians, since the >> Castilians couldn't care less). This is known as "diferencialismo". >> RW > >Can you actually substantiate this claim, with regard to Catalan? It is >true that such an effect may appear to someone who doesn't know about >the history of the languages (obviously not you); in several cases >standardizers prefer a form with a longer established literary >tradition. Note that people from French Catalonia have accused the >standardizers of preferring those forms which were MORE similar to >Spanish. For Catalan, most of the standardizing work was done in the first half of this century, primarily by the orthographical and grammatical standards of the Institut d'Estudis Catalans and the dictionary of Pompeu Fabra. Fabra talks about these issues in a 1924 article ("L'obra de depuracis del catal`") and admits that "hi hagui una hpoca en quh, de dues formes, de dos mots, de dos girs en concurrhncia, es creia que sistem`ticament calia donar la preferhncia al mis allunyat de l'espanyol. En tota coincidhncia entre l'espanyol i el catal`, es veia un castellanisme, i bastava que un mot s'assemblis massa a l'espanyol corresponent perquh se li cerquis mis o menys arbitr`riament un substitut" (There was a time when it was thought that if there were two competing forms, words or expressions, one had to prefer the one furthest from Spanish. In every agreement between Spanish and Catalan a castilianism was seen, and a word only had to look too similar to the corresponding Spanish one in order for more or less arbitrary substituions for it to be sought.) [The above passage contains three such cases in relation to my own dialect: I would say and quite possibly write for "words", for "would search", although I would not write (but possibly say) for "less"]. While trying to steer clear of such exaggerations as described above by Fabra, standard Catalan has drastically reduced the number of castilianisms, often by harking back to the Medieval literature. Many castilianisms of course still remain in the spoken language, and a considerable number have also been maintained in the standard, as it would be silly (and impossible) to deny that contact with Castilian ever existed [after all, there are also catalanisms in Castilian]. In many of these cases where standard Catalan has perforce a castilianism (more or less catalanized), a gallicism is likely to be used in Rossells. I suppose that's what the complaints may have been about. ======================= Miguel Carrasquer Vidal mcv at wxs.nl Amsterdam From martha_ratliff at wayne.edu Fri Oct 16 15:15:08 1998 From: martha_ratliff at wayne.edu (Martha Ratliff) Date: Fri, 16 Oct 1998 11:15:08 EDT Subject: 7th annual workshop on comparative linguistics Message-ID: ----------------------------Original message---------------------------- 7TH ANNUAL WORKSHOP ON COMPARATIVE LINGUISTICS November 21-22, 1998 at The University of Illinois at Urbana-Champaign Room 210, Illini Union THE RECONSTRUCTION OF CULTURE Speakers and topics include: Hans Henrich Hock: "Use and misuse of linguistic prehistory: India and beyond" Frederick Schwink: "The reconstruction of IE gender as a reflection of culture" Mary Niepokuj: "IE poetics" Rex Wallace: "Sabellian" Martha Ratliff: "Vocabulary of environment and subsistence in Proto-Hmong-Mien" William Baxter: "Evidence for early Austronesian-Chinese contact in China" Colleen Reilly: "Gender and sexuality in Anglo-Saxon" Graham Thurgood "What proto-Chamic reconstructions tells us about early Chamic culture" Craig Hilts "Vocabulary for flora and fauna of Mixe-Zoquean" Co-sponsored by: English, Program in South Asian and Middle Eastern Studies, Germanic Languages and Literatures, Linguistics, Classics, Program in Comparative Literature. For information, contact Frederick Schwink Dept. of Germanic Languages University of Illinois at Urbana-Champaign 3072 FLB 707 South Mathews Ave. Urbana, IL 61801 From sw271 at cus.cam.ac.uk Fri Oct 23 11:25:42 1998 From: sw271 at cus.cam.ac.uk (Sheila Watts) Date: Fri, 23 Oct 1998 07:25:42 EDT Subject: r and s Message-ID: ----------------------------Original message---------------------------- Hullo everyone, Could anyone help me with the answer to a small question which has been interesting me for some time? There are alternations between r and s in different parts of the paradigm of some Latin words (only verbs?), and these are apparent in borrowings into English such as adhere vs. adhesive, adhesion. Now these alternations are similar to those found as a result of the operation of Verner's Law in words like English was - were, German war - gewesen, kiesen - kor and so on. Is there a link, or is this phenomenon typologically common? Thank you, Sheila Watts _______________________________________________________ Dr Sheila Watts Newnham College Cambridge CB3 9DF United Kingdom phone +44 1223 335816 From sw271 at cus.cam.ac.uk Fri Oct 23 15:53:53 1998 From: sw271 at cus.cam.ac.uk (Sheila Watts) Date: Fri, 23 Oct 1998 11:53:53 EDT Subject: r and s Message-ID: ----------------------------Original message---------------------------- Thank you for helpful postings to date. Apologies for suggesting that the Latin development was peculiar to verbs - I had lost sight of the wood for the trees. If I have it right, English adhere comes from a Latin verb of which the infinitive is h?rere and the fourth principal part h?sum, so Latin rhotacism must be more complex than just 'inlautend zwischen Vokalen', which was the description I had found. What had struck me was the similar effect (to Germanic) of having a change in consonant across the principal parts of a verb. Sheila Watts _______________________________________________________ Dr Sheila Watts Newnham College Cambridge CB3 9DF United Kingdom phone +44 1223 335816 From ph1u+ at andrew.cmu.edu Fri Oct 23 15:53:31 1998 From: ph1u+ at andrew.cmu.edu (Paul J Hopper) Date: Fri, 23 Oct 1998 11:53:31 EDT Subject: r and s In-Reply-To: Message-ID: ----------------------------Original message---------------------------- On Sheila Watts' query about s/r alternations in Germanic and Latin (so-called rhotacism): I can't think of other examples of it outside these two groups, but the two processes were the results of independent changes s > z, followed by z > r and a lot of analogical restoration or generalization of the original s. The environments for s > z are a bit different, since in Latin it occurred between vowels regardless of accent. The oldest Germanic documents still have the original sibilant. z > r doesn't seem to be a commonly attested change, and it is an interesting coincidence that it should occur independently in two families often seen as quite close within Indo-European. I don't have my copy of C D Buck's Comparative Grammar of Latin and Greek with me here, but I'm sure there's a good clear presentation of the Latin situation there, and E Prokosch, A Comparative Germanic Grammar would be a good place to go for the Germanic details. Paul Hopper Thomas S. Baker Professor of English & Linguistics Carnegie Mellon University Pittsburgh PA 15213, USA Excerpts from mail: 23-Oct-98 r and s by Sheila Watts at cus.cam.ac. > ---------------------- > Hullo everyone, > Could anyone help me with the answer to a small question which has been > interesting me for some time? > There are alternations between r and s in different parts of the paradigm > of some Latin words (only verbs?), and these are apparent in borrowings > into English such as adhere vs. adhesive, adhesion. Now these alternations > are similar to those found as a result of the operation of Verner's Law in > words like English was - were, German war - gewesen, kiesen - kor and so > on. > Is there a link, or is this phenomenon typologically common? > Thank you, From mcv at wxs.nl Sat Oct 24 15:17:35 1998 From: mcv at wxs.nl (Miguel Carrasquer Vidal) Date: Sat, 24 Oct 1998 11:17:35 EDT Subject: r and s In-Reply-To: <4qA86Xe00WBM02EGE0@andrew.cmu.edu> Message-ID: ----------------------------Original message---------------------------- Paul J Hopper wrote: >On Sheila Watts' query about s/r alternations in Germanic and Latin >(so-called rhotacism): I can't think of other examples of it outside >these two groups There are the Sanskrit and Nuorese (Sardinian) sandhis, where -s > -r before a voiced consonant at the start of the following word. ======================= Miguel Carrasquer Vidal mcv at wxs.nl Amsterdam From Georg at home.ivm.de Sat Oct 24 21:17:22 1998 From: Georg at home.ivm.de (Ralf-Stefan Georg) Date: Sat, 24 Oct 1998 17:17:22 EDT Subject: r and s In-Reply-To: <363101ee.35251404@mail.wxs.nl> Message-ID: ----------------------------Original message---------------------------- At 11:17 Uhr -0400 24.10.1998, Miguel Carrasquer Vidal wrote: >----------------------------Original message---------------------------- >Paul J Hopper wrote: > >>On Sheila Watts' query about s/r alternations in Germanic and Latin >>(so-called rhotacism): I can't think of other examples of it outside >>these two groups > >There are the Sanskrit and Nuorese (Sardinian) sandhis, where -s > -r >before a voiced consonant at the start of the following word. And there is the Bolghar-Turkic Rhotacism, which changes Common Turkic *-z-, -z to -r-, -r in the languages of the Bolghar branch, viz. Volga Bolghar and Modern Chuvash. St.G. Stefan Georg Heerstrasse 7 D-53111 Bonn FRG +49-228-69-13-32 From mcv at wxs.nl Sun Oct 25 17:51:33 1998 From: mcv at wxs.nl (Miguel Carrasquer Vidal) Date: Sun, 25 Oct 1998 12:51:33 EST Subject: r and s In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Ralf-Stefan Georg wrote: >And there is the Bolghar-Turkic Rhotacism, which changes Common Turkic >*-z-, -z to -r-, -r in the languages of the Bolghar branch, viz. Volga >Bolghar and Modern Chuvash. Of course! But can we be sure it went from z to r, and not the other way around? ======================= Miguel Carrasquer Vidal mcv at wxs.nl Amsterdam From hubeyh at Montclair.edu Sun Oct 25 17:54:01 1998 From: hubeyh at Montclair.edu (H. M. Hubey) Date: Sun, 25 Oct 1998 12:54:01 EST Subject: r and s Message-ID: ----------------------------Original message---------------------------- Ralf-Stefan Georg wrote: > > > And there is the Bolghar-Turkic Rhotacism, which changes Common Turkic > *-z-, -z to -r-, -r in the languages of the Bolghar branch, viz. Volga > Bolghar and Modern Chuvash. > St.G. > Unless I am wrong the word for 5 in Chuvash is pilek (which to me sounds a lot like 'bilek' (wrist, hand) and which sounds to me as if it is older than Common Turkic 5 (besh). Secondly the weakness of the liquids in Turkic and Altaic (and Dravidian) is well known. It makes more common sense to believe that l~r Turkic (Chuvash) is more archaic and these sounds changed to sh~z because the eastern languages did not have the liquids, than to assume that it is a rhotacism. > > Stefan Georg > Heerstrasse 7 > D-53111 Bonn > FRG > +49-228-69-13-32 -- M. Hubey Email: hubeyh at Montclair.edu Backup:hubeyh at alpha.montclair.edu WWW Page: http://www.csam.montclair.edu/Faculty/Hubey.html From s_nickn at cassius.its.unimelb.EDU.AU Sun Oct 25 17:54:40 1998 From: s_nickn at cassius.its.unimelb.EDU.AU (Nick Nicholas) Date: Sun, 25 Oct 1998 12:54:40 EST Subject: TOC: History of Language 4.2 Message-ID: ----------------------------Original message---------------------------- The latest issue of _History of Language_ (formerly _Dhumbadji! Journal for the History of Language_) is now available, featuring: H.M. Hubey: Quantitative Approaches to Historical Linguistics with Example Application to *PIE/IE E.F.K. Koerner: On the Historiography of the Polish Contribution to the Understanding of Language Change See http://www.lexicon.net/opoudjis/Work/ahl.html for more information. -- ____.____1____.____2____.____3____.____4____.____5____.____6____.____7____. When in doubt, Nick Nicholas, Linguistics & Applied Linguistics, --- DOUBT! N.Nicholas at linguistics.unimelb.edu.au University of (- me) http://www.lexicon.net/opoudjis Melbourne, Australia From Georg at home.ivm.de Sun Oct 25 22:24:44 1998 From: Georg at home.ivm.de (Ralf-Stefan Georg) Date: Sun, 25 Oct 1998 17:24:44 EST Subject: r and s In-Reply-To: <363888cd.21120723@mail.wxs.nl> Message-ID: ----------------------------Original message---------------------------- >>And there is the Bolghar-Turkic Rhotacism, which changes Common Turkic >>*-z-, -z to -r-, -r in the languages of the Bolghar branch, viz. Volga >>Bolghar and Modern Chuvash. > >Of course! But can we be sure it went from z to r, and not the other >way around? I expected some sort of objections to this, and my answer has to be that we can*not* be sure. In fact, within Turkology, two positions exist, one taking the sibilant as original (the "Rhotacists") and the other - more at home with the proponents of the Altaic theory, but also supported by its chief critic, Gerhard Doerfer, that the Rhotic is (the "Zetacists"). While consensus is still to be waited for, it might be a safe approach to watch out for loan-words which entered either branch of Turkic *before* the particular sound change (-z-/-z > -r-/-r or the other way round) occurred. For the zetacistic position, this seems to be lacking (i.e. a clear example of an Iranian/Chinese/Tokharian, you name it) word with -r-/-r ending up zetacised in Common (= non-Chuvash) Turkic. For the rhotacistic position, there seem to be a few examples. Among those which I regard as most convincing is Chuvash /pir/ "linen" or the like, going back to Arabic baZZ and ultimately to Greek byssos (a Mediterranian-Oriental Wanderwort which really went places). I don't want to make this mootest of Turkological problems seem more easy than it is (it isn't), the bottomline should, however, be that both positions are defendable at the moment and that any account which states that the problem is "solved" once and for all, is therefore in error (hereby I explicitly complain about some pro-Altaistic works of the 90's which simply forget to tell their readers that the other position exists at all, or, in more belligerent passages, that it is a malevolent idee fixe of rabid anti-Altaicists; that this is not the case should be clear from Doerfer's position). And *even* within the framework of "zetacistic pro-Altaistics", it is of course commonplace that the input for Commot Tk. /z/ and Bolgh. /r/ cannot be simply /r/ (since Common Tk. *does* have /r/ in native words and morphemes), but something usually labeled as /r2/, /r'/ or the like, sometimes with the implication that it might have been something like Czech /R/ in DvoRak. Maybe. But isn't that multiplicare entia praeter necessitatem ? Rhotacism is able to make do with *one* proto-phoneme, viz. /z/, zetacism needs two /r, r'/. The whole /z-r/-mess is paralleled by a similar Lautgesetz, by which Common Tk. /sh/ and Bolgh.-Chuv. /l/ are connected. Again, the direction of the change is the object of often fierce discussions (between lambdacists and sigmatists, you guessed it), though good (and old) loanwords which might help to decide seem to be lacking even more (A. Rona-Tas presented some candidates in the corpus of words which he views as Tokharian loan-words in Proto-Turkic, but, although I personally happen to like some of them, I don't expect them to stand a real chance of ever getting generally accepted; it might surprise few at this stage that they seem to support lambdacism). And: the rhotacistic process has been called rather exceptional, not very wide-spread in this thread. I'm not so sure whether we've already found all prominent examples, I think there must be more. On the other hand, what is the evidence for r > z being more wide-spread" or, well, "natural" ? What parallels can be found ? For the general subject of this thread the conclusion is: *part* of the Turkological community views the Chuvash-Common-Turkic /r,z/ relation as just another example of Rhotacism. St.G. says: right they are, though the opposite opinion is never far away. That's life in Altaic studies. St.G. Stefan Georg Heerstrasse 7 D-53111 Bonn FRG +49-228-69-13-32 From mcv at wxs.nl Sun Oct 25 22:25:03 1998 From: mcv at wxs.nl (Miguel Carrasquer Vidal) Date: Sun, 25 Oct 1998 17:25:03 EST Subject: r and s In-Reply-To: <36326DD3.7F6D84F@Montclair.edu> Message-ID: ----------------------------Original message---------------------------- "H. M. Hubey" wrote: >Unless I am wrong the word for 5 in Chuvash is pilek (which to me sounds >a lot like 'bilek' (wrist, hand) and which sounds to me as if it is >older than Common Turkic 5 (besh). Secondly the weakness of the liquids >in Turkic and Altaic (and Dravidian) is well known. It makes more common >sense to believe that l~r Turkic (Chuvash) is more archaic and these >sounds changed to sh~z because the eastern languages did not have the >liquids, than to assume that it is a rhotacism. But the "eastern" languages do have liquids. The question is whether Proto-Turkic (cq. Proto-Altaic) had two of each (*r > r/r; *r1 > r/z; *l > l/l; *l1 > s^/l) or only one. ======================= Miguel Carrasquer Vidal mcv at wxs.nl Amsterdam From mcv at wxs.nl Mon Oct 26 10:53:46 1998 From: mcv at wxs.nl (Miguel Carrasquer Vidal) Date: Mon, 26 Oct 1998 05:53:46 EST Subject: r and s In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Ralf-Stefan Georg wrote: >And: the rhotacistic process has been called rather exceptional, not very >wide-spread in this thread. I'm not so sure whether we've already found all >prominent examples, I think there must be more. There sure must be, although for some reason I can only come up with syllable/word final examples. Armenian: *-is and *-us > -r in certain cases (erkir < *dwis, erir < *tris, u-stem nouns and adjectives like barjr, asr, t`anjr etc.) Basque: sl > rl, e.g. irla "island" < Spa. isla; erle "bee" < *ezle, cf. ezti "honey", ezko "wax". [NB. Basque is (laminal) /s/, not /z/!; -le is an agentive suffix] Hausa: the ind. obj. pronoun "to him" has a variant ma/\r (< *mas) [the /, \ and /\ are high, low and falling (high->low) tones]. ======================= Miguel Carrasquer Vidal mcv at wxs.nl Amsterdam From mcv at wxs.nl Mon Oct 26 10:54:10 1998 From: mcv at wxs.nl (Miguel Carrasquer Vidal) Date: Mon, 26 Oct 1998 05:54:10 EST Subject: r and s In-Reply-To: <3633B3EA.B6D8AC09@Montclair.edu> Message-ID: ----------------------------Original message---------------------------- "H. M. Hubey" wrote: >Neither Turkic nor Dravidian have words beginning with the liquids. I don't think initial position was mentioned previously (and I believe the Turkic z ~ r alternations occur everywhere but initially). As to r-, it is also practically absent from, say, IE and Basque in the West. You may have a case for l- (not extremely common in IE or Basque, but hardly absent), but note the "compensatory" paucity of initial n- in the West (both in IE and Basque [also Kartvelian?]). ======================= Miguel Carrasquer Vidal mcv at wxs.nl Amsterdam From hubeyh at Montclair.edu Mon Oct 26 10:54:39 1998 From: hubeyh at Montclair.edu (H. M. Hubey) Date: Mon, 26 Oct 1998 05:54:39 EST Subject: r and s Message-ID: ----------------------------Original message---------------------------- Miguel Carrasquer Vidal wrote: > ----------------------------Original message---------------------------- > "H. M. Hubey" wrote: > > >Unless I am wrong the word for 5 in Chuvash is pilek (which to me sounds > >a lot like 'bilek' (wrist, hand) and which sounds to me as if it is > >older than Common Turkic 5 (besh). Secondly the weakness of the liquids > >in Turkic and Altaic (and Dravidian) is well known. It makes more common > >sense to believe that l~r Turkic (Chuvash) is more archaic and these > >sounds changed to sh~z because the eastern languages did not have the > >liquids, than to assume that it is a rhotacism. > > But the "eastern" languages do have liquids. The question is whether > Proto-Turkic (cq. Proto-Altaic) had two of each (*r > r/r; *r1 > r/z; > *l > l/l; *l1 > s^/l) or only one. > Neither Turkic nor Dravidian have words beginning with the liquids. KOrean (or Japanese) has only a single liquid and [l], and [r] merely allophonic realizations. As can be seen the liquids become less and less used from west to east. The gradient runs east to west. So if anything the easterners substituted other sounds for the liquids from the west and managed to pick up a single liquid along the way (in the far west). The state of the language in the ME before the AA and IE spread is similar; lots of confusion of l and r and especially in the beginning of words (see von Soden). > > ======================= > Miguel Carrasquer Vidal > mcv at wxs.nl > Amsterdam -- M. Hubey Email: hubeyh at Montclair.edu Backup:hubeyh at alpha.montclair.edu WWW Page: http://www.csam.montclair.edu/Faculty/Hubey.html From hubeyh at montclair.edu Mon Oct 26 10:58:15 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Mon, 26 Oct 1998 05:58:15 EST Subject: r and s Message-ID: ----------------------------Original message---------------------------- Ralf-Stefan Georg wrote: > > ----------------------------Original message---------------------------- > > >>And there is the Bolghar-Turkic Rhotacism, which changes Common Turkic > >>*-z-, -z to -r-, -r in the languages of the Bolghar branch, viz. Volga > >>Bolghar and Modern Chuvash. > > > >Of course! But can we be sure it went from z to r, and not the other > >way around? > > I expected some sort of objections to this, and my answer has to be that we > can*not* be sure. In fact, within Turkology, two positions exist, one > taking the sibilant as original (the "Rhotacists") and the other - more at > home with the proponents of the Altaic theory, but also supported by its > chief critic, Gerhard Doerfer, that the Rhotic is (the "Zetacists"). We are never sure of anything, even the sun rising tomorrow. That is the famous induction problem of Hume. But by this time we should all be reasonably used to working under uncertainty. There are reasons why it is not necessary to posit rhotacization or lambdacization, or zetacization etc etc. > While consensus is still to be waited for, it might be a safe approach to > watch out for loan-words which entered either branch of Turkic *before* the > particular sound change (-z-/-z > -r-/-r or the other way round) occurred. > For the zetacistic position, this seems to be lacking (i.e. a clear example > of an Iranian/Chinese/Tokharian, you name it) word with -r-/-r ending up > zetacised in Common (= non-Chuvash) Turkic. For the rhotacistic position, > there seem to be a few examples. Among those which I regard as most > convincing is Chuvash /pir/ "linen" or the like, going back to Arabic baZZ > and ultimately to Greek byssos (a Mediterranian-Oriental Wanderwort which > really went places). I don't want to make this mootest of Turkological > problems seem more easy than it is (it isn't), the bottomline should, > however, be that both positions are defendable at the moment and that any > account which states that the problem is "solved" once and for all, is > therefore in error (hereby I explicitly complain about some pro-Altaistic > works of the 90's which simply forget to tell their readers that the other > position exists at all, or, in more belligerent passages, that it is a > malevolent idee fixe of rabid anti-Altaicists; that this is not the case > should be clear from Doerfer's position). Since the present de facto standard (which should be called the zeroth order approximation) does not take into consideration anything other than a model of linguistic descent which is biologically similar to asexual descent which is similar to tracing the mtDNA, many things cannot be explained without going into a more realistic position. I do not believe in this model of constant divergence. So it is just as easy to believe that some people whose language was rich in the liquids moved west to east and the peoples of the east got an incomplete dosage of the liquids. It seems that most of the Altaic arguments are about whether there really was a state of some language which via constant divergence gave rise to Japanese, Korean, Tungus, Turkic, etc. It is not necessary for such things to occur in order for these languages to form a family. Semitic and IE were formed from a mixture of some ancient languages and it is patently clear in the case of Semitic that it got "frozen" into some semi-regular or quasi-regular state which can be seen in the binyanim. The verb gradations of Hittite are exactly the same kind of evidence. They belong to the same type of phenomena as that of the irregular verbs in English. It is easy for some people to see layer upon layer in Japanese, such as proto-Korean, or proto-Altaic over Austronasian etc but I can see the layering in English (i.e. Latin layer over a Germanic layer over a non-IE layer, mixed in with some Celtic) but the IEnists cannot see these layers. > And *even* within the framework of "zetacistic pro-Altaistics", it is of > course commonplace that the input for Commot Tk. /z/ and Bolgh. /r/ cannot > be simply /r/ (since Common Tk. *does* have /r/ in native words and > morphemes), but something usually labeled as /r2/, /r'/ or the like, > sometimes with the implication that it might have been something like Czech > /R/ in DvoRak. Maybe. But isn't that multiplicare entia praeter > necessitatem ? Rhotacism is able to make do with *one* proto-phoneme, viz. > /z/, zetacism needs two /r, r'/. > The whole /z-r/-mess is paralleled by a similar Lautgesetz, by which > Common Tk. /sh/ and Bolgh.-Chuv. /l/ are connected. Again, the direction of > the change is the object of often fierce discussions (between lambdacists > and sigmatists, you guessed it), though good (and old) loanwords which > might help to decide seem to be lacking even more (A. Rona-Tas presented > some candidates in the corpus of words which he views as Tokharian > loan-words in Proto-Turkic, but, although I personally happen to like some > of them, I don't expect them to stand a real chance of ever getting > generally accepted; it might surprise few at this stage that they seem to > support lambdacism). Goethe said a long time ago that "When an idea is lacking a word can always be found to take its place." Here is something that could have happened. A superstratum P dominates over a substratum B. At first only the words of P gets into writing. Since they don't have TV and forced education for 12 years the B people can't learn this language too well, but over a period of time start to manage. The offspring of P mix with B over time. Because of the large numbers of B, eventually the offspring of P also start to speak their ancestors' language like B people. By this time the B also make it to the top (mixing) and the language (written of course) suddenly starts to display some strange characteristics. These characteristics, of course, are the imcomplete learning of P by B people. Almost as if they did it on purpose things like rhotacization, lambdacization, zetacization, voicing etc appear out of nowhere in the language (as evidenced by writing) centuries and millennia later when our intrepid linguists start to decipher these codes. Voila! Changes appear out of nowhere and some nice linguist decides to become famous by making up a name for the process. That is how these things are born. That is much more believable than turning a complex reality into a comical textbook version only to feed the incomplete theories so that they can linger on and on and on....instead of letting them go the way of the dinosaurs. Making up a name is nice but is it necessary? What does it explain? So then what is its value other than holding up a toy model of language change? > And: the rhotacistic process has been called rather exceptional, not very > wide-spread in this thread. I'm not so sure whether we've already found all > prominent examples, I think there must be more. On the other hand, what is > the evidence for r > z being more wide-spread" or, well, "natural" ? What > parallels can be found ? If there are apparently such changes it probably means that it arose out of some kind of an interaction of peoples speaking different languages. What all this means is that instead of using principle A (which says always attribute everything to "natural" change -- what on earth does natural mean?) create a new principle; one that says (no effects without causes). If there is some very good reason to suspect that for some reason people find it easier to create /r/ than /z/ then we should read about it. Ditto for lambdacization. > For the general subject of this thread the conclusion is: *part* of the > Turkological community views the Chuvash-Common-Turkic /r,z/ relation as > just another example of Rhotacism. St.G. says: right they are, though the > opposite opinion is never far away. That's life in Altaic studies. If this rhotacization seems to happen a lot in the West, maybe there is a good reason for it. Ooops, there it goes again; another name invented already called 'areal' phenomena. Could this not simply be due to the fact that the indigeneous inhabitants of some region who spoke related languages got invaded from all over the place by people speaking similar languages and reacted to these new languages similarly simply because their phoneme/phone repertoire was only capable of producing some sounds and not others? What other reason could there be? Certainly nothing the air, the water or the mountains! > > St.G. > > Stefan Georg > Heerstrasse 7 > D-53111 Bonn > FRG > +49-228-69-13-32 -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From l.campbell at ling.canterbury.ac.nz Mon Oct 26 10:59:58 1998 From: l.campbell at ling.canterbury.ac.nz (Lyle Campbell) Date: Mon, 26 Oct 1998 05:59:58 EST Subject: r and s in Turkic Message-ID: ----------------------------Original message---------------------------- In view of the several messages on the problem of Turkic sound correspondences involving r (and liquids) and s (and sibilants), I'll throw into the mix a small bit about it taken from a recent paper of mine evaluating Dolgopolsky's Nostratic. Since this bit starts with one of Dolgopolsky's specific proposed Nostratic forms, the content relevant to the discussion of the Turkic problem becomes clear only towards the end of it all. (I claim no first-hand knowledge of the topic, and attribute most of the significant content to Juha Janhunen, except for any mistakes I may have made -- sorry for any diacritics which do not come across). Lyle Campbell [48] *p'oK'? 'wild cattle, pack' (Indo-European *pek'u / *pek'we- 'cattle'; Altaic *p'ok'?r'- 'bovine animal, bull'). This set clearly involves borrowing. The Altaic *p'ok'?r'-, represented only by Turkic, is a clear example of a documented loan, involving one of the strong points among the arguments of those who oppose the Altaic hypothesis. Proto-Turkic *s split into s and z, and *S ("sh") into S and Z ("zh") in specific environments (involving roots of two syllables and with long vowels), and then in the highly influential Bulgaric (Chuvash) branch of the family z > r, and Z > l. As a result, words in Mongolian (and Tungusic) which have an r or l corresponding to s, z, S, or Z in other Turkic languages can only be borrowings from this branch of Turkic, not true cognates to other Altaic languages (or they are accidental similarities). There is a sizeable number of these in the Dolgopolsky's putative Altaic lexical comparisons. In this case, in set [48], the word involved is Proto-Turkic *p?k?s 'bovine', borrowed from Bulgaric into Mongolian and from there on into Tungusic (Janhunen 1996a:240-1, 255). This set would be questionable in any case, given the important role of cattle in the prehistoric cultures from the area of the Proto-Indo-European homeland and in the territory of the various so-called Altaic languages. I should add here that Starostin and Dolgopolsky (in discussion in the symposium) disputed this interpretation of the Turkic facts, preferring reconstructions of Proto-Turkic which reflect the liquids rather than the sibilants and in this way they deny that borrowing is a problem for these "Altaic" forms. This interpretation would require assuming that the liquids (l/r) were original and changed to sibilants in certain of the Turkic languages, a kind of sound change seldom seen in the world's languages, though changes in the other direction are common (as in rhotacism). There is considerable literature precisely on this topic. Among Turkologists, those who believe in the Altaic hypothesis (as well as Doerfer, who opposes Altaic, though he holds Mongolian forms in these comparisons to be Turkic loans) postulate original liquids (which then would make the sibilants of other Turkic languages the results of later sound changes); those who oppose the Altaic hypothesis (with the exception of Doerfer) hold the sibilants to be original (which makes the liquids the results of later sound changes). Put differently, those with a grasp of phonological systems and phonetic plausibility all postulate a change of *s > z > r and *S > Z > l, where the steps in the change are seen as incremental, intimately interrelated, and natural. No one with a sense of phonology postulates the reverse, the unnatural and implausible changes of *r > z/s and *l > Z/S, which are almost unknown in languages elsewhere (Shherbak 1986b, Janhunen personal communication). [[From: Campbell, Lyle. In Press. Nostratic and Linguistic Palaeontology in Methodological Perspective. In: Nostratic: Evaluating a Linguistic Macrofamily, ed. by Colin Renfrew and Daniel Nettle. Cambridge: The McDonald Institute for Archaeological Research.]] From hubeyh at montclair.edu Mon Oct 26 11:00:58 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Mon, 26 Oct 1998 06:00:58 EST Subject: r and s Message-ID: ----------------------------Original message---------------------------- Miguel Carrasquer Vidal wrote: > > "H. M. Hubey" wrote: > > >Neither Turkic nor Dravidian have words beginning with the liquids. > > I don't think initial position was mentioned previously (and I > believe the Turkic z ~ r alternations occur everywhere but > initially). The initial r and l are treated almost as if they were consonant clusters, except in Turkish into which they have settled. > As to r-, it is also practically absent from, say, IE and Basque in > the West. You may have a case for l- (not extremely common in IE or > Basque, but hardly absent), but note the "compensatory" paucity of > initial n- in the West (both in IE and Basque [also Kartvelian?]). Lahovary has a huge list of words re: Basque, Etruscan, Sumerian, Caucasian languages, Nilo-Saharan and AA and IE. He is making a good case for a related set of languages being spread around the Mediterranean thru India (Dravidian). I had noticed some of these strange things but Lahovary has done a much better job. It is now easier to find links amongst Dravidian, Turkic and Sumerian (and probably Basque) because of that. Chuvash seems to figure more prominently than the other Turkic languages. At least that is my present pre-impression. > ======================= > Miguel Carrasquer Vidal > mcv at wxs.nl > Amsterdam -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From Georg at home.ivm.de Mon Oct 26 11:46:36 1998 From: Georg at home.ivm.de (Ralf-Stefan Georg) Date: Mon, 26 Oct 1998 06:46:36 EST Subject: r and s In-Reply-To: <363101ee.35251404@mail.wxs.nl> Message-ID: ----------------------------Original message---------------------------- Regarding the question how common rhotacistic phenomena are, here are some further examples from Greek dialects: - in intervocalic position -s- gets rhotacized to -r- in the Eretria and Oropos local varieties of Ionic (pairin = paisin) - only in Auslaut position rhotacism is found in later Laconian, which is (partly) retained in modern Tsakonian - and in later Elian (Dior = Dios, tir = tis etc.) Othe examples or near-examples incude: In the Mongolian language Dagur several stop consonants (b, G, g, d) + s are changed > r in syllable final position. In Tundra Yukaghir, under certain assimilatory conditions including secondary intervocalic position in compounds, initial s- is rhotacized > r-. >From East Caucasian languages: the Akush and Urakh dialects of Dargwa seem to respond with -r'- (r + glotal stop) and -r to a proto-Dargwa input -z-/-z. In Tibetan dialects, in some positions (always involving an initial sC- cluster) s- may yield to r-: Written Tib. sna "nose" > Panakha (Banag, a NE dialect) rna, WT sku "body" > Panakha rku, WT stag "tiger" > Golok rtag. But this process depends strongly on the following C, instances of rC- > sC- are also found. So far, Latin rhotacism has been mentioned, but not the similar phenomenon in Umbrian (but not Oscan), which, however, might be seen in connection with the Latin process (though the precise nature of this connection, i.e. the chronology, is unclear to me). St.G. Stefan Georg Heerstrasse 7 D-53111 Bonn FRG +49-228-69-13-32 From hubeyh at montclair.edu Mon Oct 26 17:01:46 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Mon, 26 Oct 1998 12:01:46 EST Subject: r and s in Turkic Message-ID: ----------------------------Original message---------------------------- Lyle Campbell wrote: > > > results of later sound changes). Put differently, those with a grasp of > phonological systems and phonetic plausibility all postulate a change of *s > > z > r and *S > Z > l, where the steps in the change are seen as > incremental, intimately interrelated, and natural. No one with a sense of > phonology postulates the reverse, the unnatural and implausible changes of > *r > z/s and *l > Z/S, which are almost unknown in languages elsewhere > (Shherbak 1986b, Janhunen personal communication). To put it mildly differently, as can be seen everywhere in phonology and phonetics the sounds /ptksn/ are "rarely absent" from the world's languages. If the "natural" change to /r/ and/or /l/ was so natural we would have had no /s/ left anywhere. Perhaps those who are pushing rhotacization on us have latched on to some local phenomena (local in both time and space) and think it is universally applicable or rhotacization does not exist as it is said to exist but is a substratum/superstratum phenomena and is once again a local phenomena and not a global one. OTOH, if you are looking at a mixing of a language A with strong consonant-clusters and rich in fricatives and sibilants with another language B which is not clustered and is weak in fricatives perhaps for a period of time, (local in space and time) you could obtain the (in)famous rhotacization, lambdacization etc. PS. REad Lindblom's works on phonetics and the physics and/or optimization of sound patterns. > > [[From: Campbell, Lyle. In Press. Nostratic and Linguistic Palaeontology > in Methodological Perspective. In: Nostratic: Evaluating a Linguistic > Macrofamily, ed. by Colin Renfrew and Daniel Nettle. Cambridge: The > McDonald Institute for Archaeological Research.]] -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From tuitekj at ERE.UMontreal.CA Mon Oct 26 16:59:34 1998 From: tuitekj at ERE.UMontreal.CA (Kevin Tuite) Date: Mon, 26 Oct 1998 11:59:34 EST Subject: r and s yet again In-Reply-To: Message-ID: ----------------------------Original message---------------------------- In a recent posting, Ralf-Stefan Georg asked "what is the evidence for r > z being more wide-spread" or, well, "natural" ? What parallels can be found ?" Phonetic changes similar in some respects to the shift r > z postulated by some Turkologists have been attested. One of them is an interesting chapter in the sociolinguistic history 13th-16th century French, of a sound change that subsequently was beaten back. One of the few traces remaining in the standard language of this event is the word "chaise", which I discuss in a forthcoming article (see excerpt below). Another potentially useful case is the women's pronunciation of Chukchee, as described by W. Bogoras in the Handbook of American Indian Languages (sic): "Women generally substitute /sh/ for /ch/ and /r/". It is, I believe, important to note that both these sound shifts (1) uvulaire (not alveolar) /R/ becomes a fricative; (2) are sociolinguistically sensitive. Any ideas why? Kevin Tuite ******** (from "Au-del? du Stammbaum : Th?ories modernes du changement linguistique"): Le mot 'chaise' remonte, comme l'indiquent les dictionnaires, au latin 'cathedra', lui-m?me emprunt? au grec. 'Cathedra' a subi une succession complexe de l?nitions successives pour aboutir, plusieurs si?cles plus tard, ? 'chaire'. La transformation de la forme de ce lex?me a ?t? l'effet de lois phon?tiques r?guli?res (cf. le sort du lat. 'fraxinum' > 'fr?ne', ou de 'pectinum' > 'peigne'). C'est la toute derni?re ?tape, [Se:R] > [Se:z], qui nous pose probl?me. Si la transformation de la derni?re consonne de 'chaise' repr?sentait le r?sultat d'une loi phon?tique, comment doit-on expliquer la prononciation actuelle de 'p?re', 'm?re', 'arri?re' et de celle de dizaines d'autres mots pr?sentant le m?me contexte phonologique (y compris les homophones 'chair' et 'chaire')? Heureusement, l'histoire nous a conserv? quelques t?moignages pertinents. D?j? au 13?me si?cle, des documents provenant des r?gions centrale et sud-est de la zone du fran?ais proprement dit (la "langue d'o?l") indiquaient "une tendance ? l'assibilation de r intervocalique"; deux si?cles plus tard, la m?me tendance a fait son apparition "dans le parler vulgaire de la capitale", o? elle a m?me attir? l'attention d'?rasme en 1518 "Idem faciunt hodie mulierculae parisinae pro Maria sonantes Masia, pro ma mere, ma mese" (Fouch? 1961: 603). Paradoxalement, un deuxi?me observateur ? Paris, quinze ans apr?s ?rasme, a pris note de la prononciation de courin pour 'cousin' - cette fois, c'est le [z] qui a c?d? sa place ? l'[R]. Encore plus paradoxalement, toutes les deux substitutions, l'une exactement ? l'inverse de l'autre, se notent dans une expression relev?e en 1529 ? Bourges: Jerus Masia! (Meyer-L?bke 1974: 408). Un si?cle plus tard, un ?crivain parlait au pass? de cette vague de substitutions de phon?mes qui avait frapp? la langue de Villon: "Nos Parisiens mettoient autrefois (mais cela ne se fait plus ou c'est fort rarement et seulement parmi le menu peuple) une s au lieu d'une r et une r au lieu d'une s" (loc. cit.) Que s'est-il pass?? Le post mortem offert par W. von Wartburg (1988: 156) est instructif; ? son avis, il s'agissait "d'un mouvement avort?, parce que rejet? par les classes sup?rieures". Rapidement apr?s que cette "contagion" linguistique eut gagn? la capitale, "les classes sup?rieures r?sist?rent, elles n'accept?rent pas ce changement venu d'en bas. Cette r?action eut m?me pour cons?quence que les gens du peuple ne savaient plus quand il fallait prononcer r. Par peur de s'exposer ? des critiques on se mit ? remplacer par r m?me les z qui ?taient justifi?s ?tymologiquement. On trouve donc des formes comme rairon et courin. Mais en g?n?ral le r?tablissement se fit correctement. On sait qu'un doublet est rest?: chaire - chaise". En lisant cette histoire trois d?cennies apr?s la publication des premiers ouvrages de W. Labov, on a le sentiment de d?j? vu. Les ph?nom?nes qui, selon la description de von Wartburg, accompagnait cette guerre entre "l'inconscient" et "la raison" portent aujourd'hui des noms qui font d?sormais partie du vocabulaire technique de chaque linguiste (et dont l'un, au moins, aurait sonn? familier au romaniste allemand): le "changement par le bas" et l'"hypercorrection" ************************************************************** Kevin Tuite 514-343-6514 (bureau) D?partement d'anthropologie 514-343-2494 (t?l?copieur) Universit? de Montr?al C.P. 6128, succursale centre-ville Montr?al, Qu?bec H3C 3J7 tuitekj at mistral.ere.umontreal.ca ************************************************************** From mcv at wxs.nl Mon Oct 26 23:17:42 1998 From: mcv at wxs.nl (Miguel Carrasquer Vidal) Date: Mon, 26 Oct 1998 18:17:42 EST Subject: r and s in Turkic In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Lyle Campbell wrote: >No one with a sense of >phonology postulates the reverse, the unnatural and implausible changes of >*r > z/s and *l > Z/S, which are almost unknown in languages elsewhere Whatever happened in Turkic, and I have no opinion, what keeps the debate going (apart from the lack of hard evidence either way, despite Stefan's pir < baZZ) would seem to be that both changes are equally likely. The Altaicist (and Nostraticist) argument is of course not based on plain *r and *l, but on palatalized/fricative *r^(> z) and *l^ (> s^), cf. Polish and Hebrew, respectively. ======================= Miguel Carrasquer Vidal mcv at wxs.nl Amsterdam From DISTERH at UNIVSCVM.SC.EDU Tue Oct 27 20:00:55 1998 From: DISTERH at UNIVSCVM.SC.EDU (Dorothy Disterheft) Date: Tue, 27 Oct 1998 15:00:55 EST Subject: rhotacism from Ray Hickey Message-ID: On the reverse of rhotacism and other *unlikely* changes ======================================================== The present discussion of rhotacism and its possible reverse has been centered on Altaic and the Germanic/Latin data so that perhaps the data has prevented us from making some general points. Taking Lyle's suggestion that the reversal of rhotacism is very unlikely as a starting point: the issue to home in on in such changes is the SECONDARY ARTICULATION of the segments involved. A fricative /r/ (apical trill or continuant) can easily become a full sibilant, i.e. re-align itself phonologically as /z/, as has happened in Polish with the Slavic PRE-prefix and as seen in Czech in the trilled /r/ (evident e.g. in _Dvorak_, the composer's name and indicated by a superscript hacek). It also happens sub-phonemically in Stockholm Swedish and forms of Western Irish when /r/ is phonetically palatalised and where the raising of the apex to the palate leads to assibiliation. I might push this point and suggest that secondary articulation is the natural bridge leading across a phonological divide to a phoneme on the other bank, so to speak. The classic case is velarised /l/ [l-] to /u/ which is hardly worth commenting on, it is so common: Polish, colloquial forms of southern British English, Brazilian Portuguese; historically: southern English in general, French, etc. The secondary articulation becomes primary, the original primary articulation is dropped and bob's your uncle. Ray Hickey English Linguistics Essen University Germany r.hickey at uni-essen.de From jhewson at morgan.ucs.mun.ca Tue Oct 27 18:02:47 1998 From: jhewson at morgan.ucs.mun.ca (John Hewson) Date: Tue, 27 Oct 1998 13:02:47 EST Subject: r and s In-Reply-To: Message-ID: ----------------------------Original message---------------------------- In the flurry of comments on rhotacism, Sheila Watts original question seems to have been lost: is the s/r alternation in Latin attributable to something like Verner's Law? Paul Hopper did comment that the change in Latin affected all instances of intervocalic /s/, and that leaves us with Sheila's question as to why haereo/haesi and English adhere/adhesive? The answer is that intervocalic /s/ in Classical Latin comes from original clusters, mostly /ss/, /ts/, /st/, and so forth. Latin haesi is obviously a sigmatic perfect (Latin merged the perfect and the sigmatic aorist, leaving remnants of both morphologies), similar to scribo/scripsi, nubo/nupsi, rideo/risi Message-ID: ----------------------------Original message---------------------------- >Whatever happened in Turkic, and I have no opinion, what keeps the >debate going (apart from the lack of hard evidence either way, >despite Stefan's pir < baZZ) first I'd like to make a minor correction of "my" (of course, neither the word, nor the etymology belong to me, though I'd not be against it ;-) baZZ, which should have been given as /bazz/. And there are some more quite likely LWs, which bespeak rhotacism and might amount to something in the way of "hard evidence": Chuv. /hIr/ 'pinus abies', most probably from some Volgaic or Permic source close to Mordva /kuz/, Mari /koZ/, Komi /koz/, or Udmurt /k?z/, which in turn are deeply rooted in FU (fi. kuusi, chanty kool, man'si xavt etc.) and possibly U (nenc. xaadI). Chuv. tAvAr 'narrow' goes together with CTk. *tIqIz 'id.', *but*: the non-change of /t/ > /ch/ in front of /I/ is understandable (only, I'd say), if the Chuv. word is a Kipchak loan (i.e. it entered Bolghar after t>ch/_I and before z>r. The other changes are the expected ones. Chuv. tir- 'to arrange in rows', Ctk. tiz- 'id.', same reason for likelihood of borrowing from another Tk. language. There is further the possibility that some Chuvash personal names contain Persian /niya:z/ 'request, prayer othl.' in the form /-never/, but this is admittedly not too strong. These are not hundreds of examples, but I think rhotacism is alive and kicking, and I do think that it will be possible to seperate Chuv. /pir/ from Ar. bazz and Gk. byssos (the b>p, a>i changes are all regular and expected, the meanings are identical, and byssos is one of the most successful Wanderw?rter of Mediterranian origin in the whole of Asia ( matched only by diphthera and nomos, possibly); it would be bewildering *not* to find it in Chuvash somehow (and hardly imaginable that it would end up as anything but /pir/ there) ! On the putative palatalized l2 and r2 of proto-Turkic (and 'Altaic'), we should not forget that these values are posited *in order to* allow for a host of words being Tu.-Mo. etc. cognates rather than being borrowings, in which case proto-Turkic simply would not need those phonemes. Id multiplicatio entiorum praeter ullam necessitatem mihi esse videtur ! St.G. Stefan Georg Heerstrasse 7 D-53111 Bonn FRG +49-228-69-13-32 From vovin at hawaii.edu Tue Oct 27 14:17:36 1998 From: vovin at hawaii.edu (Alexander Vovin) Date: Tue, 27 Oct 1998 09:17:36 EST Subject: r and s in Turkic In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Dear Lyle: Let me throw my two cents into the discussion, as it seems to me you might oversimplify certain issues. (1) For the Altaic theory, it really does not matter whether one reconstructs *z and *sh or *r2 or *l2 (note that *none* of the Altaicists reconstructs just plain *r and *l here, as it may seem from you posting: Starostin, for example, following Ramstedt, reconstructs palatalized *r' and *l'). I, for one, am not convinced that the protolanguage had liquids, and I though I usually write *r2 and *l2, it is rather due to tradition, than to an assumption that these are actual liquids. You are certainly right in pointing out that r > z is less typical than z > r. But I fail to see how recognition of proto-Turkic *z rather than *r2 is going to "disprove" Altaic. It still can correspond to -r- in other languages, right, and our cognates do not have to be complete look-alikes, right? (2) I don't think any of turkologists except Shcherbak nowadays support his idea that *z < *s. There are too many exceptions to the "rules" of *s > *z that you list, thus, e.g. *ka:z (Chuvash xur) 'goose' does have a long vowel, but *k"uz (Chuvash ker) does not. Etc., etc. (3) The theory of wholesale borrowing from Bulgaric to proto-Mongolic faces many obstacles (some of them unsurmountable in my opinion). First, it is cyclic by its logic: Turkic "loans" in Mongolic have /r/, therefore they must be Bulgar loans. But there should be, a second independent evidence for the fact that these words are indeed Bulgaric loans, apart from the the /r/ itself. As a matter of fact, this evidence does not exist. Let me demonstrate it by using the same word for "ox". (a) First, while there is an initial h- in Middle Mongolian form, which allows us to reconstruct pre-Mongolic *p-, there is no evidence for reconstruction of proto-Turkic *p"ok"uz with initial *p-. The only Turkic language that has initial h- that can be claimed to represent an earlier *p- is Khaladj, but as far as I know the word is not attested in Khaladj (or not recorded) (But not eveyone accepts even Khaladj h- < *p-). There is an initial h- in Uzbek in this word, but it is prothetic, as it doesc not prtesent a regular reflex of *p- (contrary to Doerfer's claim), e.g. does not show up in such word as ar "man", where we should expect it if Uzbek h- were regular. Thus, we end up with Turkic *"ok"uz. (b) The only actual Bulgar form that we know is Chuvash vAxAr 'ox' (If I rememnber correctly, the word is not attested in Bulgar inscriptions), and it *does* represent a series of problems. While initial prothetic v- is in all probability a late development (it looks like there is no v- in Bulgar inscriptions, or at least it is difficult to tell), there is no way to claim that Chuvash vocalism (very different from Common Turkic, and even today not satisfactorily explained) is a late development, too. Thus, what basis do we have to say that MM h"ok"ur is a Bulgar loan? It appears that none, except the final -r-. (c) summing up, the hypothesis that Mongolic word is a loan from Bulgar s based on three (sic!) unproven hypotheses: that Bulgaric had *h- (no traces of it in any of Bulgaric languages) (1), that Chuvash vocalism is secondary as compared to Common Turkic (2), and that it represents a late development (3). Thus, the loanword explanation, a hypothesis itself, is relying on three other hypotheses to be true. It is much more complex solution than a single hypothesis that words in question represent cognates. Finally, the wholesale loanword scenario is based on one more hypothesis, again not proven. It presumes that Mongols were in contact of all Turks with Bulgars. But Bulgars are... the westernmost branch of Turkic, and there is a zero independent evidence that they ever were in touch with Mongols (until 13th c.), who came to the territory of Mongolia from the territory to the east. The only evidence presented for this brave claim is again Bulgar "loanwords" in Mongolic with -r. Circularity again, isn't it? Sasha ======================================= Alexander Vovin Associate Professor of Japanese Department of East Asian Languages and Literatures 382 Moore Hall 1890 East-West Road University of Hawaii at Manoa Honolulu, HI 96822 vovin at hawaii.edu fax (808)956-9515 (o.) t.(808)956-6881 (o.) On Mon, 26 Oct 1998, Lyle Campbell wrote: > ----------------------------Original message---------------------------- > In view of the several messages on the problem of Turkic sound > correspondences involving r (and liquids) and s (and sibilants), I'll throw > into the mix a small bit about it taken from a recent paper of mine > evaluating Dolgopolsky's Nostratic. Since this bit starts with one of > Dolgopolsky's specific proposed Nostratic forms, the content relevant to > the discussion of the Turkic problem becomes clear only towards the end of > it all. (I claim no first-hand knowledge of the topic, and attribute most > of the significant content to Juha Janhunen, except for any mistakes I may > have made -- sorry for any diacritics which do not come across). > Lyle Campbell > > [48] *p'oK'? 'wild cattle, pack' (Indo-European *pek'u / *pek'we- > 'cattle'; Altaic *p'ok'?r'- 'bovine animal, bull'). This set clearly > involves borrowing. The Altaic *p'ok'?r'-, represented only by Turkic, is > a clear example of a documented loan, involving one of the strong points > among the arguments of those who oppose the Altaic hypothesis. > Proto-Turkic *s split into s and z, and *S ("sh") into S and Z ("zh") in > specific environments (involving roots of two syllables and with long > vowels), and then in the highly influential Bulgaric (Chuvash) branch of > the family z > r, and Z > l. As a result, words in Mongolian (and > Tungusic) which have an r or l corresponding to s, z, S, or Z in other > Turkic languages can only be borrowings from this branch of Turkic, not > true cognates to other Altaic languages (or they are accidental > similarities). There is a sizeable number of these in the Dolgopolsky's > putative Altaic lexical comparisons. In this case, in set [48], the word > involved is Proto-Turkic *p?k?s 'bovine', borrowed from Bulgaric into > Mongolian and from there on into Tungusic (Janhunen 1996a:240-1, 255). > This set would be questionable in any case, given the important role of > cattle in the prehistoric cultures from the area of the Proto-Indo-European > homeland and in the territory of the various so-called Altaic languages. > I should add here that Starostin and Dolgopolsky (in discussion in > the symposium) disputed this interpretation of the Turkic facts, preferring > reconstructions of Proto-Turkic which reflect the liquids rather than the > sibilants and in this way they deny that borrowing is a problem for these > "Altaic" forms. This interpretation would require assuming that the > liquids (l/r) were original and changed to sibilants in certain of the > Turkic languages, a kind of sound change seldom seen in the world's > languages, though changes in the other direction are common (as in > rhotacism). There is considerable literature precisely on this topic. > Among Turkologists, those who believe in the Altaic hypothesis (as well as > Doerfer, who opposes Altaic, though he holds Mongolian forms in these > comparisons to be Turkic loans) postulate original liquids (which then > would make the sibilants of other Turkic languages the results of later > sound changes); those who oppose the Altaic hypothesis (with the exception > of Doerfer) hold the sibilants to be original (which makes the liquids the > results of later sound changes). Put differently, those with a grasp of > phonological systems and phonetic plausibility all postulate a change of *s > > z > r and *S > Z > l, where the steps in the change are seen as > incremental, intimately interrelated, and natural. No one with a sense of > phonology postulates the reverse, the unnatural and implausible changes of > *r > z/s and *l > Z/S, which are almost unknown in languages elsewhere > (Shherbak 1986b, Janhunen personal communication). > > [[From: Campbell, Lyle. In Press. Nostratic and Linguistic Palaeontology > in Methodological Perspective. In: Nostratic: Evaluating a Linguistic > Macrofamily, ed. by Colin Renfrew and Daniel Nettle. Cambridge: The > McDonald Institute for Archaeological Research.]] > From manaster at umich.edu Tue Oct 27 14:16:25 1998 From: manaster at umich.edu (manaster at umich.edu) Date: Tue, 27 Oct 1998 09:16:25 EST Subject: r and s yet again In-Reply-To: Message-ID: ----------------------------Original message---------------------------- If I may, I'd like to say hi to Kevin, and say that as always great minds do think alike. I cite this very French sound change (I dont mean it is very French; I mean this very one) in a recent paper in JIES (along with an exmaple from Polish) to show that such a direction of change is indeed possible. The context may be of more general interest. A leading Russian linguist, Serebrennikov or Shcherbak I forget which, had attacked Nostratic among other things on the grounds that some of the sound changes from Nostratic to daughter languages posited by Illich-Svitych were impossible. This was one of them. Alexis MR On Mon, 26 Oct 1998, Kevin Tuite wrote: > ----------------------------Original message---------------------------- > > > In a recent posting, Ralf-Stefan Georg asked "what is the evidence for r > > z being more wide-spread" or, well, "natural" ? What parallels can be found > ?" Phonetic changes similar in some respects to the shift r > z postulated > by some Turkologists have been attested. One of them is an interesting > chapter in the sociolinguistic history 13th-16th century French, of a sound > change that subsequently was beaten back. [snip] From hubeyh at montclair.edu Tue Oct 27 14:15:46 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Tue, 27 Oct 1998 09:15:46 EST Subject: r and s yet again Message-ID: ----------------------------Original message---------------------------- Kevin Tuite wrote: > > > generally substitute /sh/ for /ch/ and /r/". It is, I believe, important to > note that both these sound shifts (1) uvulaire (not alveolar) /R/ becomes a > fricative; (2) are sociolinguistically sensitive. Any ideas why? Yes, it has to do with a lot of things. Basically, assuming that it is definitely not a strata problem, then it is the interaction of the phonotactics of the language with physical constraints. There is an easy way to visualize what these physical constraints are. Conceptually it is about inertia & acceleration effects mixed with a predetermined goal of creating a given set of sounds. Words like assimilation, metathesis, etc are used to describe these complex of maneuvers. If you want to see both an acoustic, physical and a perceptual view of these processes in a single 3D space (a higher dimensional space is really needed), you can do so for the cost of $8.95 at 1stBooks.com. Just enter my name 'Hubey' at the prompt and you can download and electronic copy of my book which hopefully will be published in paper in the west. For those who do not want to spend the money, a copy of a paper which discusses this can be found on my website. (It's something like "phonological spaces"). > Kevin Tuite > > -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From hubeyh at montclair.edu Wed Oct 28 12:44:18 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Wed, 28 Oct 1998 07:44:18 EST Subject: r and s in Turkic Message-ID: ----------------------------Original message---------------------------- Ralf-Stefan Georg wrote: > > from Ar. bazz and Gk. byssos (the b>p, a>i changes are all regular and > expected, the meanings are identical, and byssos is one of the most > successful Wanderwvrter of Mediterranian origin in the whole of Asia ( > matched only by diphthera and nomos, possibly); it would be bewildering > *not* to find it in Chuvash somehow (and hardly imaginable that it would > end up as anything but /pir/ there) ! Speaking about Wanderworter, I have been wondering about one myself. The word 'ak' in Turkic (these days meaning 'white' and 'to flow') seems to be one of these also. For example, aqua, eku (Hittite, to drink), ich (Turk. to drink), akar (that which flows) > watar?, aryk (water canal), and even shows up as lots of words with 'ar' root in them in Dravidian (Lahovary) and I think 'Mediterratean' as 'ar', and 'sar', etc. Is that possible? -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From hubeyh at montclair.edu Wed Oct 28 12:45:05 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Wed, 28 Oct 1998 07:45:05 EST Subject: uc and ucak Message-ID: ----------------------------Original message---------------------------- I am going to ask something that sounds very strange. I noticed accidentally that a word like 'uchakka' (that is reasonably close) meant 'to fly' in Sanskrit. (I hope that was Sanskrit, and not Indic or Hindi). In any case, it is very strange for me to see this since 'uch' means 'to fly' in Turkic. The choices are 1. Accident 2. There is something we are missing 2.a) Uch was borrowed into Turkic 2.b) "uch" is protoworld 2.c) The root comes from the Ancient ME (ANE) There might be more but these are good enough for a start. 1) with odds of 1 to 100 or 1 to 1000, it behooves not to believe this at first. 2a) There are too many pieces of evidence against it. But this could be one sided because I do not know Sanskrit. This is a common phenomena. If I am asked "Is a qumquat more like a quince or an apple?" I could not answer this correctly unless I knew all three fruits. So the Sanskrit part has to be provided by IE scholars, however I will do my little part from the Turkic angle, and even more. I will also take this to the ANE :-) There are too many Turkic words that have something close to this and they fall into a pattern. First there is 'uc' meaning 'edge'. But there are similar words (/r/ words) like /Or/ (where O=high rounded o, i.e. o-umlaut) (having to do with height, /Orle/ (to climb), /Orge/ (upwards), /Oreley/ (standing up), etc. And these can be found in East Turkistan as well as in the North Caucasus. Secondly the /z/ version is there; i.e. Uzre, Uzeri , or Ust, UstUnde, UsUnden, UstUN, etc having to do with 'top, above' etc. Then there is 'oz' (to pass, to overtake), and /Os/ (to grow (high)). 2b) 'uch' or something related (see below in 2c) is protoworld, but it goes thru lots of changes. 2c) The word for /bird' is /kush/. Something like k>x>h>0 would produce /ush/ > /uch/. And the k>x can be seen between Turkic languages right now. Further, /mushen/ is /bird/ in Sumerian and m>k is Sumer > Turkic (as can be seen in Tuna's book). Furthermore in the OI books on Hittites, the word /ar/ shows up as 'height', or 'high'. I think something like /arma/ was 'moon'. This is said to be an Asianic word, not IE. We note that a > O, gives /Or/. Furthermore r > y gives /ay/ which is the word for moon in Turkic. Furthermore r > y can be found to be a change observed in Dravidian > Turkic (I think, I have to check Lahovary, again). Where does that leave Protoworld? -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From manaster at umich.edu Wed Oct 28 12:45:40 1998 From: manaster at umich.edu (manaster at umich.edu) Date: Wed, 28 Oct 1998 07:45:40 EST Subject: r and s in Turkic and the Altaic Theory In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Sasha Vovin is entirely right that there is no connection whatever between the different views on the phonetic values of the two Proto-Turkic phonemes at issues AND the different views on the Altaic theory. Leading opponents (e.g., THE leading anti-Altaicist, Doerfer) as well as proponents of Altaic (like Sasha or in my own small way, me) agree on this point, though there are some people who have not seen the light. From afaefk at upe.ac.za Wed Oct 28 12:46:39 1998 From: afaefk at upe.ac.za (Prof E F Kotze') Date: Wed, 28 Oct 1998 07:46:39 EST Subject: rhotacism from Ray Hickey Message-ID: Dear Collegues, I perhaps should have made it clearer when I posted Ray Hickey's remarks yesterday that the text was entirely Ray's. Dorothy Disterheft ----------------------------Original message---------------------------- Dear Dorothy An interesting synchronic phenomenon in Afrikaans, which seems to be spreading rapidly, is the changing of syllable-final /s/ from apical/alveolar to retroflex when preceded by /r/, to such an extent that /r/ and /s/ coalesce into something which is phonologically similar to the Czech r hacek (but voiceless). Examples in Afrikaans are kinders ('children') (with accent on the first syllable), verseker ('ensure') (accent on the second) and nors ('grumpy'). Your comments below (or are they Ray's?) led me to draw this comparison: >A fricative /r/ (apical trill or continuant) can easily become > a full sibilant, i.e. re-align itself phonologically as /z/, as has > happened in Polish with the Slavic PRE-prefix and as seen in Czech in the > trilled /r/ (evident e.g. in _Dvorak_, the composer's name and indicated > by a superscript hacek). It also happens sub-phonemically in Stockholm > Swedish and forms of Western Irish when /r/ is phonetically palatalised and > where the raising of the apex to the palate leads to assibiliation. Another case of the promotion of secondary articulation? Seemingly. Ernst Kotze' ================================== Prof. Ernst F. Kotze Dept. Afrikaans & Nederlands, UPE Posbus 1600 ZA-6000 Port Elizabeth Afrique du Sud Tel. +27 41 5042226 (W), 533230 (H), 5042574 (F) From tuitekj at ERE.UMontreal.CA Wed Oct 28 13:05:53 1998 From: tuitekj at ERE.UMontreal.CA (Kevin Tuite) Date: Wed, 28 Oct 1998 08:05:53 EST Subject: Jerus Masia! Message-ID: ----------------------------Original message---------------------------- My assertion that the Chukchee /r/ has uvular articulation is based on the Bogoras grammar in HAIL, which describes its pronunciation "as in French (hard trill, 'roue')", contrasted to a "dental r with weak trill" (p 645). (On the other hand, in a table on the preceding page, both /r/'s are classed as "alveolar", god knows why). I had assumed that the Chukchi "French r" would sound roughly the same as the uvular R of Eskimo. Ralf-Stefan Georg kindly brought to my attention two recent grammars of Chukchi which describe the sound in question as being, indeed, alveolar or perhaps alveo-palatal ("perednejazychnyj, kakuminal'nyj"). I remain intrigued by the sociolinguistic aspects of the Chukchi and French /r/ > sibilant shifts, and in particular the association of both changes with women's speech (Recall Erasmus' observation that the French shift was led by the "mulierculae parisinae "). Could the sibilant articulation of an earlier rhotic have "feminine" associations of the sort that Matthew Gordon and Jeffrey Heath propose for vowel shifts (see the latest issue of Current Anthropology, vol 39 #4)? Kevin Tuite PS Greetings to Alexis. ************************************************************** Kevin Tuite 514-343-6514 (bureau) D?partement d'anthropologie 514-343-2494 (t?l?copieur) Universit? de Montr?al C.P. 6128, succursale centre-ville Montr?al, Qu?bec H3C 3J7 tuitekj at mistral.ere.umontreal.ca ************************************************************** From Georg at home.ivm.de Wed Oct 28 17:15:38 1998 From: Georg at home.ivm.de (Ralf-Stefan Georg) Date: Wed, 28 Oct 1998 12:15:38 EST Subject: r and s in Turkic (long and focussing on Altaic) In-Reply-To: Message-ID: ----------------------------Original message---------------------------- No, I don't mean to convert Histling into AltaiNet - and I did notice that this thread has left its original focus long ago, but I think some points maybe of general interest, and the Altaic details can, if need be, discussed by us further in a different forum. Sasha, old friend, good to hear from you again. While it will come as no surprise to you (nor to me) that a discussion of these matters will see us two, as so often, on the opposite sides of the table, I might be allowed to comment a bit on your assertions. First, let me agree with you (and Alexis, of course) that the whole rhotacism/zetacism has no immediate bearing on the Altaic question. The languages can wlel be related (or not), regardless what the original consonant qualities were. This is, e.g., the position prominent in most writings of Gerhard Doerfer, who is, as we know, a critic of Altaic and a zetacist (by which I mean that he views /z/ as the end of the story). However, it is quite surprising that pro-Altaicists usually defend zetacism like the crown-jewels, each time a rhotacist appears lumping him/her with the anti-Altaicists. In view of what you said yourself, this is an oversimplification, or, better, plainly wrong. I don't follow Lyle Campbell where he formulates that " No one with a sense of phonology postulates the reverse, the unnatural and implausible changes of *r > z/s and *l > Z/S, which are almost unknown in languages elsewhere" etc., since the opposite sound-change *can* happen (few things really *don't ever* happen), as we see in Kevin Tuite's French example (already quoted in Shcherbak somewhere and in writings by Rona-Tas); I remember you yourself bringing an example from Vietnamese - if I'm not mistaken - to my attention. Yet I subscribe to the view that rhotacism is far more common (see previous examples from a wide variety of languages). Yet, phonetics apart, and it is not basically a phonetic problem I'm dealing with, there *are* difficulties with either scenario (for *both*) positions which have at least potential bearing on Altaic studies (not on the "Altaic question", no question like that can be decided on by relying on a single point like that; so, in a word, there can be no talk of "disproving" "Altaic" by dealing with the liquids alone, I think we all can agree on that). >(3) The theory of wholesale borrowing from Bulgaric to proto-Mongolic >faces many obstacles (some of them unsurmountable in my opinion). First, >it is cyclic by its logic: Turkic "loans" in Mongolic have /r/, therefore >they must be Bulgar loans. But there should be, a second independent >evidence for the fact that these words are indeed Bulgaric loans, apart >from the the /r/ itself. As a matter of fact, this evidence does not >exist. Let me demonstrate it by using the same word for "ox". With this I have to take issue, since, I'm afraid, it contains a piece of oversimplification on your side. -r- instead of CT -z- is *by no means* the only criterion for the "Bolgharicity" of some words. There are other criteria, involving other pieces of consonantism and, especially, vocalism, which is all too often overlooked. A word can have (more than one) typical Bulghar trait, even if it doesn't contain anything remotely resembling -r- (< -z- or -r2). I have recently (in press) argued *against Doerfer* (who took the words as chance similarities) for a loan-scenario involving the Turkic and Mongolian words for "heart" (although there *is* an -r- in this word, it's of the /r1/-type) on the evidence of non-rhotacistic criteria alone. I'll not try to get on everyone's nerves here by giving the whole story of this etymology, but I'm ready to share it with everyone, who wants to hear about it. On rhotacism: zetacism does carry along one difficulty for Altaic studies: whily in a rhotacistic scenario this process happened only once (in Bolghar, carried over by LWs to Mong. and Tung.), the zetacist scenario would hold /r/ as the proto-Altaic sound, maintained almost everywhere, but shifted to /z/ in each and every non-Bulghar Turkic language. While this is OK as far as it goes, it brings about the need to view zetacistic Tk. languages as *one and only* primary branch of Turkic (languages from Turkish to Yakut via Tuvan, that is). This may well be the case, but on other occasions we two had a hot debate on the number of *primary* branches of Turkic, with me voting against and you voting for such a branch as "Sayanic" and others, being as primary as all the others (Kipchak, Oghuz). Now the least thing you'll have to do is to admit a *primary* branch "Common Turkic" with the shift r2 > z as the one common innovation. Maybe you are prepared to do so, but don't talk to me about "primary Sayanic" again ;-) On the "bovine"-etymology you do a variety of things. You reconstruct proto-Mong. *with* *p-, using the Mmo. evidence, which is OK with me. Then you deny *p- (> h-) in Turkic, reconstructing *"ok"uz only (I take it that this means CT, or you accidentally forgot to write -r2 ;-). This should now make borrowing less likely then cognacy. I admit that I don't follow (in fact, the Starostinian sound-laws expect proto-Altaic p- - evidenced by Mong. - to be reflected by Turkic h- as well; so, taking Starostin's sound-laws for granted, you are actually advocating for the Turko-Mong. etymology to be given up ! Or else, where's the h- ? Calm down, it is there, see below.). In fact, by removing the h- from the Turkic form you strengthen the case for an old (and long ago given up) borrowing scenario which wanted to derive Turkic *"ok"uz from sthl. Tokharian okso (this etymology had been given up, largely because h- got in the way, which was not well known in Turkic at that time, as well as the Mmo. data). Now I wouldn't call this borrowing scenario nonsense, but I don't subscribe to it, since I do think that the h- is there after all. The question of Turkic h- is a difficult one, and it can't be put aside by a laconic "but it is prosthetic" and a mere "contrary to Doerfer's claim". While I'm not going to present the fifty-odd papers D. devoted to the problem here, readers interested in the question might wish to consult G. Doerfer: "Materialien zu t"urk. h-" I, UAJb N.F. 1/198193-141, II: 2/1982, 138-168 (oops, it's eighty-odd pages). There a (imho strong) point is made for the attestation of original h-s in some Turkic lgs. and prosthetic ones also, together with some criteria to distinguish between them. I hope interested parties will read this to form an opinion of their own. Now, for the "bovine"-word, h- is attested from more than one source. While it is true that Khaladzh is silent on this word, we find h- forms in Turkmen dialects (R"as"anen), Khorasanli Turkic, Modern Uighur, Uzbek (which you mentioned) and the older literary language Chaghatay. So, h- is there, the Tokharian loan scenario is again to be forgotten, and the Turkic and Mong. words belong together. (Please note that, while I'm far from presenting everything Doerfer says as the plain gospel, I'm unwilling to further discuss the matter of Tk. h- on the base of anything but this paper. It is a most thorough investigation of this difficult problem and every discussion which does not depart from it (maybe by disagreeing, but then the whole of the data and the methods employed should be discussed in detail, which we certainly cannot do here) is simply uninformed. >does not show up in such word as ar "man", where we should expect it if >Uzbek h- were regular. Thus, we end up with Turkic *"ok"uz. And: the word for "man" *is* attested in Uzbek with h- (albeit not in the literary standard, but rather in Kipchak-Uzb. dialects; and it is so attested in Old Turkic in Brahmi-script, TT VIII). >(c) summing up, the hypothesis that Mongolic word is a loan from >Bulgar s based on three (sic!) unproven hypotheses: that Bulgaric had *h- >(no traces of it in any of Bulgaric languages) (1), that Chuvash vocalism >is secondary as compared to Common Turkic (2), and that it represents a >late development (3). Thus, the loanword explanation, a hypothesis itself, >is relying on three other hypotheses to be true. It is much more complex >solution than a single hypothesis that words in question represent >cognates. (2) and (3) are only one hypothesis. It is true that Bulghar does not show any trace of h-. But this is common practice: Germanic loans in Balto-Finnic sometimes show traces of Proto-Germanic, which are nowhere attested in Germanic, but only recoverable through comparative reconstruction; Finnish words often show that this reconstruction was right. There is no reason I can see why an older form of language A should not be observable in A loans in B. > Finally, the wholesale loanword scenario is based on one more >hypothesis, again not proven. It presumes that Mongols were in contact of >all Turks with Bulgars. But Bulgars are... the westernmost branch of >Turkic, and there is a zero independent evidence that they ever were in >touch with Mongols (until 13th c.), who came to the territory of Mongolia >from the >territory to the east. The only evidence presented for this brave claim is >again Bulgar "loanwords" in Mongolic with -r. Circularity again, isn't it? But there is another argument here, which is simply dangerous: that of geography. Well, Bulghar is the Westernmost branch of Turkic, so what ? Baluchi is one of the easternmost Iranian languages, yet it is West Iranian, Ossetic is one of the westernmost languages of this family, yet it is East Iranian. Iranian loanwords are present in a variety of Finno-Ugric languages, Baltic loanwords are to be found as far east as Mordva, it is linguistics which can *claim* that for these borrowings previous contact *has* to be assumed (and it is, in the case of Baltic in the East, seconded by toponymy). You don't assume that Bulghar Turkic speaking people have always been living on the Volga, do you ? To put it more precisely, changing seats for argument's sake: given the Altaic theory were correct, and an original proto-Altaic speech community split up somewhere in Asia: we don't have to assume that the Japanese end up on the Japanese islands immediately after that, as well as the Bulghar Turks on the Volga, do we ??? The disintegration of an original speech community is a complicated process, certainly allowing for a great deal of inter-branch borrowing going on for some time, at least as long as geographical contiguity is maintained or contacts are not blocked by social and other reasons. *Of course*, the Bulghar loan hypothesis *does* make a strong claim for prehistoric contacts between Bulghar Turkic and Mongolian, that's its strength, that's where the linguist's work becomes interesting for historians, that's part of the success story of I.E. linguistics in Western Universities, I'd say. The only thing we can say that Bulghar *writing* started in the West. There is a host of early Inner Asian nomad confederacies which are linguistically "unlumpable" into one of the known language families. There can be no reason to exclude that one or some of them contained a "Bulgharoid" element (though I don't dare to forward a specific hypothesis on which one). And: given that most pro-Altaicists locate the Altaic "homeland" pretty far in the East (you say yourself that the Mongols probably entered the steppe from the east), the Bulghars had to find *some* way to their present habitats through all that landmass, hadn't they ? OK, my apologies for getting verbose (as always when things Altaic are at stake), I'll leave this thread to itself now and see what happens (I'm ready to continue with you, Sasha, or anyone else interested on the other fora, we are subscribed to). Stefan Stefan Georg Heerstrasse 7 D-53111 Bonn FRG +49-228-69-13-32 From mcv at wxs.nl Wed Oct 28 17:13:48 1998 From: mcv at wxs.nl (Miguel Carrasquer Vidal) Date: Wed, 28 Oct 1998 12:13:48 EST Subject: Jerus Masia! In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Kevin Tuite wrote: >I remain >intrigued by the sociolinguistic aspects of the Chukchi and French /r/ > >sibilant shifts, and in particular the association of both changes with >women's speech (Recall Erasmus' observation that the French shift was led >by the "mulierculae parisinae "). Could the sibilant articulation of an >earlier rhotic have "feminine" associations of the sort that Matthew Gordon >and Jeffrey Heath propose for vowel shifts (see the latest issue of Current >Anthropology, vol 39 #4)? While Emesal, the Sumerian "women's speech" [a literary dialect used primarily in direct speech of female characters/goddesses] is not distinguished by assibilation of /r/, there is a "shift" /n/ > /s^/, which might be interpreted in a similar way. However, not all /n/'s are affected: in most Emesal words /n/ = /n/ in the standard dialect. That should caution against making too much of this. The issue whether Emesal actually *is* a women's language (and not a local or late variety) is itself unresolved (in the Sanskrit drama, Prakrits are generally only spoken by female characters, yet the Prakrits are not "women's speech"). ======================= Miguel Carrasquer Vidal mcv at wxs.nl Amsterdam From larryt at cogs.susx.ac.uk Wed Oct 28 17:09:55 1998 From: larryt at cogs.susx.ac.uk (Larry Trask) Date: Wed, 28 Oct 1998 12:09:55 EST Subject: uc and ucak In-Reply-To: <363694DD.2C8B85C4@montclair.edu> Message-ID: ----------------------------Original message---------------------------- On Wed, 28 Oct 1998, H.M.Hubey wrote: > I am going to ask something that sounds very strange. > I noticed accidentally that a word like 'uchakka' (that is > reasonably close) meant 'to fly' in Sanskrit. (I hope that was > Sanskrit, and not Indic or Hindi). > In any case, it is very strange for me to see this since > 'uch' means 'to fly' in Turkic. The choices are > 1. Accident > 2. There is something we are missing > 2.a) Uch was borrowed into Turkic > 2.b) "uch" is protoworld > 2.c) The root comes from the Ancient ME (ANE) > There might be more but these are good enough for a start. > 1) with odds of 1 to 100 or 1 to 1000, it behooves not to > believe this at first. No, not so, I'm afraid. If you ask the question "what is the probability that a form will mean `fly' in both Turkish and Sanskrit?", then the *a priori* odds against are simply enormous. But that's the wrong question. The right question is this: what is the probability that *some* short form will turn up in *some* two languages with similar meanings in both? And this time the answer is "as close to 100% as you care to get". With 6000+ languages available, all of them with thousands of words, and with only a small number of distinguishable consonants and vowels available to construct those words, it is statistically inconceivable that we should fail to find many, many chance resemblances or identities like this one. Hence stumbling across one is evidence for nothing at all, except that the laws of probability are not taking the day off. Failure to appreciate this constitutes what somebody has dubbed "Koestler's fallacy", since the writer Arthur Koestler was notoriously prone to it. Koestler was constantly impressed by the observation that *some particular* coincidence had occurred, reasoning that the *a priori* odds against that particular coincidence were astronomical, and hence concluding that Something Deeply Significant was going on. What he always failed to realize is that the probability that *some coincidence or other* would occur purely by chance was effectively 100%. Larry Trask COGS University of Sussex Brighton BN1 9QH UK larryt at cogs.susx.ac.uk From jrader at m-w.com Wed Oct 28 16:56:12 1998 From: jrader at m-w.com (Jim Rader) Date: Wed, 28 Oct 1998 11:56:12 EST Subject: rhotacism from Ray Hickey Message-ID: ----------------------------Original message---------------------------- Another unexotic example of assibilation of a rhotic sound can be found in several varieties of New World Spanish, where orthographic , the lengthened trill, becomes a voiced palato-alveolar fricative in certain positions. I recall being struck by this in the speech of an instructor from Co/rdoba, in the interior of Argentina, at a time when I knew nothing about the phonetics of Spanish. In the dialects of Irish I have heard, palatalized /r/ retains a rhotic quality, though there is a tendency to devoice and assibilate either "broad" (i.e., non-palatalized) or "slender" (palatalized) /r/ before a voiceless stop (in words like 'strength,' 'hen'). In Scottish Gaelic dialects, however, slender /r/ has some very diverse realizations; at least some speakers from Lewis in the Outer Hebrides have a voiced interdental fricative (at least that's what it sounds like to me). Jim Rader > > The present discussion of rhotacism and its possible reverse has been > centered on Altaic and the Germanic/Latin data so that perhaps the data has > prevented us from making some general points. Taking Lyle's suggestion that > the reversal of rhotacism is very unlikely as a starting point: the issue > to home in on in such changes is the SECONDARY ARTICULATION of the segments > involved. A fricative /r/ (apical trill or continuant) can easily become > a full sibilant, i.e. re-align itself phonologically as /z/, as has > happened in Polish with the Slavic PRE-prefix and as seen in Czech in the > trilled /r/ (evident e.g. in _Dvorak_, the composer's name and indicated > by a superscript hacek). It also happens sub-phonemically in Stockholm > Swedish and forms of Western Irish when /r/ is phonetically palatalised and > where the raising of the apex to the palate leads to assibiliation. > >.................... > Ray Hickey > English Linguistics > Essen University > Germany > > r.hickey at uni-essen.de > From mghiselin at casmail.calacademy.org Thu Oct 29 13:31:14 1998 From: mghiselin at casmail.calacademy.org (Michael Ghiselin) Date: Thu, 29 Oct 1998 08:31:14 EST Subject: Cladistic language concepts Message-ID: ----------------------------Original message---------------------------- Dear Dr. Dyen, I must apologize to list members if something happened when I tried to forward your message. Also please forgive my delay in answering your commentary. Cladistics does run into problems where there is as you say some kind of amalgamation of lineages. Sometimes the amalgamation is not so complete that lineages cannot be detected. An interesting example has been the origin of mitochondria and chloroplasts from symbiotic bacteria. As to DNA, it turns out that the techniques that we use with it are pretty much the same sort of thing that we have used all along in comparative anatomy. I spent a lot of time working with ribosomal RNA sequences and they worked pretty well. But they did not show the clearness of resolution that we hoped for. Some parts of the genome change very slowly, others more rapidly. What we try to do for the very old relationships is find something that evolves very slowly. Linguists try to find that sort of feature too. The frustrating thing for us is that branching can occur repeatedly within a very short period of time, and the branches may not be evident in the slowly-evolving molecules. We get a sort of bush instead of a tree. I suspect that this has happened in linguistic evolution too. What the DNA does for us is give a lot of additional evidence and it often helps. We will just have to keep plugging away. Sincerely, Michael Ghiselin From hubeyh at montclair.edu Thu Oct 29 13:31:48 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Thu, 29 Oct 1998 08:31:48 EST Subject: r and s in Turkic (long and focussing on Altaic) Message-ID: ----------------------------Original message---------------------------- Ralf-Stefan Georg wrote: > > But there is another argument here, which is simply dangerous: that of > geography. Well, Bulghar is the Westernmost branch of Turkic, so what ? Is it not a fundamental principle in genetics as well as historical linguistics, that the region in which one finds the greatest variety must be the ancestral homeland? Is this not this principle which people like Cavalli-Sforza (and other geneticists as well as historical linguists) use to posit that Africa is the homeland of hominids and that the region in which the largest number of languages of a family cluster in the Urheimat? If Bulgharic is in the west, Khalaj is in the west, Oguz is in the west, Kipchak is in the west, where does that point? Is there Khalaj or anything in that family (assuming it can be considered a family) in the east? Is any Bulgharic ever found in the east? > Iranian loanwords are present in a variety of Finno-Ugric languages, Baltic > loanwords are to be found as far east as Mordva, it is linguistics which > can *claim* that for these borrowings previous contact *has* to be assumed > (and it is, in the case of Baltic in the East, seconded by toponymy). You > don't assume that Bulghar Turkic speaking people have always been living on > the Volga, do you ? Maybe they lived there long enough to have other people name the river after them :-) > To put it more precisely, changing seats for argument's sake: given the > Altaic theory were correct, and an original proto-Altaic speech community > split up somewhere in Asia: we don't have to assume that the Japanese end > up on the Japanese islands immediately after that, as well as the Bulghar > Turks on the Volga, do we ??? It does not follow. Assuming that Altaic is a family and that it did split up, we don't know where the split occurred and where the Turkic branch developed or when. Even for IE for which much more data is in existence, the Urheimat still ranges from the Balkans to Central Asia or Anatolia. But nobody wants to name someplace where there are no IE speakers the outlying regions where there were only one or two languages. That cuts out Indo-Iranian regions, the British Isles, or Germany, France, etc. Even if Altaic homeland is given as Mongolia, it does not follow that Turkic homeland is the same place. The disintegration of an original speech > community is a complicated process, certainly allowing for a great deal of > inter-branch borrowing going on for some time, at least as long as > geographical contiguity is maintained or contacts are not blocked by social > and other reasons. *Of course*, the Bulghar loan hypothesis *does* make a > strong claim for prehistoric contacts between Bulghar Turkic and Mongolian, > that's its strength, that's where the linguist's work becomes interesting > for historians, that's part of the success story of I.E. linguistics in > Western Universities, I'd say. The only thing we can say that Bulghar > *writing* started in the West. There is a host of early Inner Asian nomad > confederacies which are linguistically "unlumpable" into one of the known > language families. Bulghar-Mongolian contacts (even if it happened) imply that Bulghar Turkic originated in the East or any other Turkic originated in the east. There are different intensities of language mixing. If we normalize this intensity as a number between 0 and 1 (i.e the interval [0,1]) then there are two favorite types of mixtures. One is at one extreme, say 0, in which the superstratum language disappears almost totally (like Mongols in Iran, or Russia or the Turkic Moguls in India). At the other extreme, say 1, is where the superstratum's language eventually becomes dominant (the changes all allegedly having nothing to do with the substratum). This is the linguists' equivalent of the 'ideal gases' of physics. What about a mixture of type 0.5? What kind would that produce? That is rather easy to guess for me. It would produce a language which is full of quasi-regularity or partial-regularity. For example Semitic is exactly this kind of a language. It has 15-20 "regular" thingamajics called "binyanim". It is the English regular-irregular verb formation to the nth degree. The language "froze" at a point in which it could not go to one or the other extreme but got stuck in between. The Hittite "grades" of verbs is the same phenomena. So that adds more evidence to a rather thorogh mixing process happening over a period of severall millenia in that region. That came out of a mixture of the previous non-IE and non-AA languages with some other intruders. The Ural-Altaic-Dravidian languages share traits which are left over from many millenia ago and also borrowed from each other. Of course, nothing like this can be "proven" because nothing in linguistics can be proven. You can only convince people. Similar things (modelable by similar differential equations) happen to things like creation of steel alloys. What determines the properties of the resulting steel alloy is not only what was added to iron but also the rate of cooling. Slow cooling produces soft metals, because it has more time for the diffusion processes to settle and the stresses to relieve themselves. Rapid cooling by quenching in water produces hard steels. A thorough mixture of 2 or more languages intensely over a period of time in which many people become bilingual or even trilingual is probably a rare occurrence but it is the type of mixture which can cause great changes in all aspects of langauge including its phonology, syntax, and typology. The present conventional wisdom that relationships can be classified as genetic, typological and areal is doing an injustice to languages and is an affront to science. Alas, things persist. There is the long story about why the British railroads are so narrow. It apparently goes back to the standard for Roman roads. That is how slowly some things change. There can be no reason to exclude that one or some of > them contained a "Bulgharoid" element (though I don't dare to forward a > specific hypothesis on which one). And: given that most pro-Altaicists > locate the Altaic "homeland" pretty far in the East (you say yourself that > the Mongols probably entered the steppe from the east), the Bulghars had to > find *some* way to their present habitats through all that landmass, hadn't > they ? Without doing a thorough statistical analysis of what kinds of sounds changes have taken place over centuries and accross the world, it is difficult to make strong statements, however, there must certainly be something to why m>k occurs in Sumerian >Turkic, and Dravidian > Turkic, or why n >y occurs in Sumerian > Turkic and Mongolian > Turkic. I think there is also l>y in Dravidian>Turkic etc. All of this has to be also put in the context of phonological systems of the world's languages circa 12,000-8,000 years ago because very important things were happening around that time. > OK, my apologies for getting verbose (as always when things Altaic are at > stake), I'll leave this thread to itself now and see what happens (I'm > ready to continue with you, Sasha, or anyone else interested on the other > fora, we are subscribed to). One good turn deserves another. Since this can't be discussed on Altainet, it is probably better to discuss it in the context of sound changes that occur in the main theater of history; Eurasia and North Africa and in terms of IE, AA and Altaic, Uralic, and the isolates like Basque. -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From vovin at hawaii.edu Thu Oct 29 13:32:48 1998 From: vovin at hawaii.edu (Alexander Vovin) Date: Thu, 29 Oct 1998 08:32:48 EST Subject: r and s in Turkic (long and focussing on Altaic) In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Dear Stefan, Good to hear from you, too. Let me just jot several points below, because the specifics should be discussed on AltaiNet. (1) Although my expertize within the eastern part of the family, Chuvash is the Turkic language I know really well (used to speak it fluently 10 years ago, unfortunately this is no longer the case). Nevertheless, I fail to see any Chuvashoid features in alleged "Bulgaric loans" in Mongolic, including vocalismthat you mention: Chuv. vAkAr, MM h"ok"ur, OT "ok"uz: obviously Chuvash has a different vocalism. May be you can come up with some hard-core evidence on AltaiNet. I will, for one, will be interested to hear tour etymology of Turco-Mongolian heart, but I honestly can't see how vocalism in Chuvash cEre 'heart" is going to prove that it is loan in Mongolic. (2) As I said in my reply to Lyle, I agree with him that z >r is more natural change than r > z. Yes, I usually cite the example of Hanoi Vietnamese where orthographical r- is pronounced nowadays as /z-/. The fact that it was once /r/ is easily deduced from the historical spelling (based on 17th c. Portuguese spelling) + on modern Saigon dialect where most speakers still keep have /r-/ for /r/. (3) I do reconstruct *"ok"uz for PT (or *"ok"ur2, if you like it better). We can even do *[h]"ok"ur2/z, if you like. But, imho, h- ain't there. I am familiar with Doerfer's article you cite, but it does not persuade me: all we have there is a collection of random h- cases in various Turkic languages. There is no a single word that would have h- in all languages Doerfer cites. Thus word X will have h- in A, D, E, word Y in B and C, word Z in E only, etc. etc. I do not buy it, and I must say that I am really surprised that you do. Taking Qypchak "dialect" of Uzbek hardly improves the picture: that vividly reminds me of one "oficial" PRC linguists, who had to prove relationship of Chinese and Tai against all odds for apparent political reasons, so he manipulated between languages at his free whim. Fortunately, such manooevres are observed mostly in Turcology (:-). (4) Bulgars Of course *could* live near Mongols in spite the fact that they are nowadays westernmost Turks, but this is no more than a speculation that is given without any solid proof. There is no evidence apart from the alleged loanwords that it was ever so, and in this case it is your word against mine. You do believe that there did live there, I don't. OK, can you come up with any kind of evidence, apart from "loans" to show that this is the case? Any place names, archeology, burial practices, anything? The problem of anti-Altaicoists is that they often take their hypotheses to be axiomatic truths. One of Shcherbak's works starts with line: "When Chuvash ancestors lived in Siberia..." O, ya, that is very entertaining, but who has demonstrated that they ever did? (May be thei lived in Huanghe valley: after all Chuvash word for person c,yn looks like a Chinese loan, does not it?) (5) Starostins 1991 book is an important contribution to Altaic studies. But it is not the Bible. The fact that he does not have PT *h- or pre-Turkic *p- is not going to wreck either this etymology or Altaic studies. Cheers, Sasha ======================================= Alexander Vovin Associate Professor of Japanese Department of East Asian Languages and Literatures 382 Moore Hall 1890 East-West Road University of Hawaii at Manoa Honolulu, HI 96822 vovin at hawaii.edu fax (808)956-9515 (o.) t.(808)956-6881 (o.) From dyen at hawaii.edu Thu Oct 29 21:46:50 1998 From: dyen at hawaii.edu (Isidore Dyen) Date: Thu, 29 Oct 1998 16:46:50 EST Subject: Cladistic language concepts In-Reply-To: <9810289096.AA909625474@casmail.calacademy.org> Message-ID: ----------------------------Original message---------------------------- Can you segregate the slowly developing parts of the genome from the oth ers, or more broadly, can you classify sections by their rates of branching? On Wed, 28 Oct 1998, Michael Ghiselin wrote: > > Dear Dr. Dyen, > I must apologize to list members if something happened > when I tried to forward your message. > Also please forgive my delay in answering your > commentary. > Cladistics does run into problems where there is as you > say some kind of amalgamation of lineages. Sometimes the > amalgamation is not so complete that lineages cannot be > detected. An interesting example has been the origin of > mitochondria and chloroplasts from symbiotic bacteria. > As to DNA, it turns out that the techniques that we use > with it are pretty much the same sort of thing that we have > used all along in comparative anatomy. I spent a lot of > time working with ribosomal RNA sequences and they worked > pretty well. But they did not show the clearness of > resolution that we hoped for. Some parts of the genome > change very slowly, others more rapidly. What we try to do > for the very old relationships is find something that > evolves very slowly. Linguists try to find that sort of > feature too. The frustrating thing for us is that branching > can occur repeatedly within a very short period of time, and > the branches may not be evident in the slowly-evolving > molecules. We get a sort of bush instead of a tree. I > suspect that this has happened in linguistic evolution too. > What the DNA does for us is give a lot of additional > evidence and it often helps. We will just have to keep > plugging away. > Sincerely, > Michael Ghiselin > > > From dyen at hawaii.edu Thu Oct 29 21:45:46 1998 From: dyen at hawaii.edu (Isidore Dyen) Date: Thu, 29 Oct 1998 16:45:46 EST Subject: uc and ucak In-Reply-To: Message-ID: ----------------------------Original message---------------------------- The point that you make is quite right, but I believe that what you are dealing with is likelihood instead of mathematical probability. What lies at the bottom of the problem is that the lay word probability most commonly has the meaning 'likelihood' that is the relation between the respective probabilities associated with each outcome. ID. On Wed, 28 Oct 1998, Larry Trask wrote: > ----------------------------Original message---------------------------- > On Wed, 28 Oct 1998, H.M.Hubey wrote: > > > I am going to ask something that sounds very strange. > > > I noticed accidentally that a word like 'uchakka' (that is > > reasonably close) meant 'to fly' in Sanskrit. (I hope that was > > Sanskrit, and not Indic or Hindi). > > > In any case, it is very strange for me to see this since > > 'uch' means 'to fly' in Turkic. The choices are > > > 1. Accident > > 2. There is something we are missing > > 2.a) Uch was borrowed into Turkic > > 2.b) "uch" is protoworld > > 2.c) The root comes from the Ancient ME (ANE) > > > There might be more but these are good enough for a start. > > > 1) with odds of 1 to 100 or 1 to 1000, it behooves not to > > believe this at first. > > No, not so, I'm afraid. > > If you ask the question "what is the probability that a form will > mean `fly' in both Turkish and Sanskrit?", then the *a priori* odds > against are simply enormous. But that's the wrong question. > > The right question is this: what is the probability that *some* short > form will turn up in *some* two languages with similar meanings in both? > And this time the answer is "as close to 100% as you care to get". > > With 6000+ languages available, all of them with thousands of words, and > with only a small number of distinguishable consonants and vowels > available to construct those words, it is statistically inconceivable > that we should fail to find many, many chance resemblances or identities > like this one. Hence stumbling across one is evidence for nothing at > all, except that the laws of probability are not taking the day off. > > Failure to appreciate this constitutes what somebody has dubbed > "Koestler's fallacy", since the writer Arthur Koestler was notoriously > prone to it. Koestler was constantly impressed by the observation that > *some particular* coincidence had occurred, reasoning that the *a > priori* odds against that particular coincidence were astronomical, and > hence concluding that Something Deeply Significant was going on. What > he always failed to realize is that the probability that *some > coincidence or other* would occur purely by chance was effectively 100%. > > > Larry Trask > COGS > University of Sussex > Brighton BN1 9QH > UK > > larryt at cogs.susx.ac.uk > From anna-karin.strobel at swipnet.se Thu Oct 29 18:29:21 1998 From: anna-karin.strobel at swipnet.se (anna-karin.strobel) Date: Thu, 29 Oct 1998 13:29:21 EST Subject: The languages of Gibraltar Message-ID: ----------------------------Original message---------------------------- Hej, I read somwere that in Gibraltar there is a small community where the inhabitants speak genovese-ligurian - is this correct and does anyone has any information or reference abour this? The common language is english but the most of the people also speak a dialect of castilian - which. And has this dialect influenced the local english language or does they just speak standardenglish. Thanks in advance for any information, Steve Lando Halland, Halmstad, Sweden, Europe, Tellus. From Roger.Wright at liverpool.ac.uk Thu Oct 29 16:23:29 1998 From: Roger.Wright at liverpool.ac.uk (Roger Wright) Date: Thu, 29 Oct 1998 11:23:29 EST Subject: South American /r/ & [h] In-Reply-To: <14492328618037@m-w.com> Message-ID: ----------------------------Original message---------------------------- And in Brazilian Portuguese, the phoneme /r/ (or, at least, the one written with the letter "r") is usually [h] (or [x]). A secondary articulation, I suppose; Lisbon has an aspirated sort of [rh] or [hr]; further East, nearer the Spanish border, it doesn't. RW On Wed, 28 Oct 1998, Jim Rader wrote: >----------------------------Original message---------------------------- >Another unexotic example of assibilation of a rhotic sound can be >found in several varieties of New World Spanish, where orthographic >, the lengthened trill, becomes a voiced palato-alveolar >fricative in certain positions. I recall being struck by this in the >speech of an instructor from Co/rdoba, in the interior of Argentina, >at a time when I knew nothing about the phonetics of Spanish. From tonybreed at juno.com Fri Oct 30 11:50:39 1998 From: tonybreed at juno.com (D. Anthony Tschetter-Breed) Date: Fri, 30 Oct 1998 06:50:39 EST Subject: s > r Message-ID: ----------------------------Original message---------------------------- In Rebecca Posner's "The Romance Languages" (Cambridge 1996, p.230), she cites an s>r transformation in Sard, with the plural definite article "sas": "Even in Sard we find assimilation of -s before voiced consonants: sar dentes 'the teeth', sa mmanos 'the hands', for sas dentes/manos." I don't know if this transformation exists throughout Sard or in limited areas. In that section, Posner talks about the general weakness of /s/, transforming, variously, to /j/, /h/, /S/, and /x/ via /S/, as well as dropping completely in certain cases (in French), in addition to the Sard example. -Tony Breed ___________________________________________________________________ You don't need to buy Internet access to use free Internet e-mail. Get completely free e-mail from Juno at http://www.juno.com/getjuno.html or call Juno at (800) 654-JUNO [654-5866] From sgbrady at ucdavis.edu Fri Oct 30 11:51:18 1998 From: sgbrady at ucdavis.edu (Sean Brady) Date: Fri, 30 Oct 1998 06:51:18 EST Subject: Cladistic language concepts In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Just a note on DNA sequence data. The rate of change in a particular sequence (e.g. a gene) must be conceptually kept separate from the rate of branching. This latter rate involves speciation, while the former rate probably has nothing to do with speciation. People have certainly tried to tie these two together, but such a link, if it exists, has not been demonstrated yet. In terms of using DNA sequence as comparative evidence for the history of speciation (i.e. building a tree), the main concern is finding a sequence of DNA that changes slow enough to preserve comparative differences, but without changing so fast that these differences 'pile up' on each other and obscure their history. Unfortnately, if the true history is more like a bush than a tree, as Ghiselin discusses, any gene we may pick will probably contain too few changes (i.e. be too slow) to reconstruct this history. A crucial question in both biological and linguistic reconstruction seems to be this: Does our analysis yield a bush-like structure 1) because there really is a bush-like structure; or 2) because our data and/or methods of analysis cannot resolve the (true) tree-like structure. This is a huge, and unsolved, problem in biological systematics. We simply cannot distinquish between these two scenarios yet, although with the rapid influx of DNA data, some people are beginning to tackle this problem. I wonder what the state-of-the-art is on this issue regarding linguistic reconstruction. \>----------------------------Original message---------------------------- >Can you segregate the slowly developing parts of the genome from the oth >ers, or more broadly, can you classify sections by their rates of >branching? > ************************************************************ Sean Brady sgbrady at ucdavis.edu Population Biology Graduate Group tel. (530) 752-9977 Department of Entomology fax. (530) 752-1537 University of California One Shields Avenue Davis, CA 95616-8584 ************************************************************ From hubeyh at montclair.edu Fri Oct 30 11:51:44 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Fri, 30 Oct 1998 06:51:44 EST Subject: uc and ucak Message-ID: ----------------------------Original message---------------------------- Isidore Dyen wrote: > > ----------------------------Original message---------------------------- > The point that you make is quite right, but I believe that what you are > dealing with is likelihood instead of mathematical probability. What lies > at the bottom of the problem is that the lay word probability most > commonly has the meaning 'likelihood' that is the relation between the > respective probabilities associated with each outcome. ID. Actually the biggest problem with probability calculations involving linguistics (especially historical linguistics) is not about technical matters but with setting up the problems correctly. 1. First, the simplest thing to do is the use uniform or equal probabilities when other info is lacking. 2. Second, the simplest thing to do is to assume things like independence, uncorralated sequences, etc. 3. Third, some people, who are allegedly experts in this field, trip over simple things. Notice, for example, the Ringe and Manaster-Ramer exchanges. However, the main problem, as in all probability problems is to be able to clearly know/define the sample space. The sample space is not only those languages that are now in existence, nor only those languages that were in existence but all languages that could have possibly existed but did not. It is this last part that is the biggest problem. Also as a part of this problem is the tendency to extrapolate from simple results linearly without any justification for it. For example, someone might run a simple simulation problem with 2 languages, each having 5 phonemes, and 100 words and getting 15 matches by accident and then extrapolating that extending this to 1,000 words with 50 phonemes etc will probably produce 500 matches. Or starting with calculations using the binomial distribution with p=0.001 (which might be appropriate for 1,000 words) and then using a semantic shift of 25 (because you can find 25 words having to do with "eating" in English) while ignoring that these 25 words came from the full English language with 100,000 to 400,000 words. The other big problem is to assume that word distributions are all totally uncorrelated and independent when we know that they are best modeled as correlated and as Markov processes. Even when simple models are used we should not use independence. Despite all this, there are even more elementary problems. For example, if you toss a penny 5 times and get 4 heads (H) it does not mean that since getting 5 heads is so rare, the penny sort of owes that it should pop up as tails. It does not. It has no memory. The prob is 1/2 at each throw. As another example, think about bus schedules. They arrive in dense intervals during morning and evening. Suppose the avg time between bus arrivals is M. If a passenger pops up at a bus stop at random. What is is expected wait? Surely, we know it is longer than the average time between the bus arrivals. As another example, just because someone is guaranteed to win a lottery of 1 million numbers, does not mean that the guy who won is not lucky. His chances of winning was still 0.0000001. Related to this is the fact that just because you are bound to find gold someplace in the ground because there is so much of it, it does not mean that just because you found a gold nugget when digging some place that there is no more there because it was just an accident. No. There is probably more there, because this is not a purely random event. Where there is some, there is probably more. The same holds for words. Every word you find, has to be thought of as being improbable because you see, the biggest part of the problem is missing. I don't know any Sanskrit. I just read maybe 20-40 words in some book by accident and one of them hit! If I watch a movie with some Eskimos and all of a sudden, 3 words hit me suspiciously just like some other words, anyone who tries to convince me that it happens all the time better spend some time learning to use probability theory to solve some real problems in the real world. Next time you watch the natives of X-land on TV, listen to their words carefully and see if you can spot any English! If you do, chances are great that they will be speaking English. That is a Markov process. PS. The total sample space is not 6,000 languages. IT is maybe 1,000,000 languages that might have existed but did not (or disappeared). PPS. Just because 80% of the world's population is white does not mean that the fact that they resemble each other is not genetic. Of course it is. Physical characteristics are genetic. The most difficult part of using any kind of math is in knowing when to use which formula. Those who botch things in the 8th grade or earlier say "I hate those word problems". Some people are like Duracells :-) They don't hate word problems, not in the 8th grade, not 12th, not in the MS level and not in the PhD level, and not even 20 years after the PhD level :-) > On Wed, 28 Oct 1998, Larry Trask wrote: > > > ----------------------------Original message---------------------------- > > On Wed, 28 Oct 1998, H.M.Hubey wrote: > > > > > I am going to ask something that sounds very strange. > > > > > I noticed accidentally that a word like 'uchakka' (that is > > > reasonably close) meant 'to fly' in Sanskrit. (I hope that was > > > Sanskrit, and not Indic or Hindi). > > > > > In any case, it is very strange for me to see this since > > > 'uch' means 'to fly' in Turkic. The choices are > > > > > 1. Accident > > > 2. There is something we are missing > > > 2.a) Uch was borrowed into Turkic > > > 2.b) "uch" is protoworld > > > 2.c) The root comes from the Ancient ME (ANE) > > > > > There might be more but these are good enough for a start. > > > > > 1) with odds of 1 to 100 or 1 to 1000, it behooves not to > > > believe this at first. > > > > No, not so, I'm afraid. > > > > If you ask the question "what is the probability that a form will > > mean `fly' in both Turkish and Sanskrit?", then the *a priori* odds > > against are simply enormous. But that's the wrong question. > > > > The right question is this: what is the probability that *some* short > > form will turn up in *some* two languages with similar meanings in both? > > And this time the answer is "as close to 100% as you care to get". > > > > With 6000+ languages available, all of them with thousands of words, and > > with only a small number of distinguishable consonants and vowels > > available to construct those words, it is statistically inconceivable > > that we should fail to find many, many chance resemblances or identities > > like this one. Hence stumbling across one is evidence for nothing at > > all, except that the laws of probability are not taking the day off. > > > > Failure to appreciate this constitutes what somebody has dubbed > > "Koestler's fallacy", since the writer Arthur Koestler was notoriously > > prone to it. Koestler was constantly impressed by the observation that > > *some particular* coincidence had occurred, reasoning that the *a > > priori* odds against that particular coincidence were astronomical, and > > hence concluding that Something Deeply Significant was going on. What > > he always failed to realize is that the probability that *some > > coincidence or other* would occur purely by chance was effectively 100%. > > > > > > Larry Trask > > COGS > > University of Sussex > > Brighton BN1 9QH > > UK > > > > larryt at cogs.susx.ac.uk > > -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From manaster at umich.edu Fri Oct 30 11:52:11 1998 From: manaster at umich.edu (manaster at umich.edu) Date: Fri, 30 Oct 1998 06:52:11 EST Subject: h- in Turkic In-Reply-To: Message-ID: ----------------------------Original message---------------------------- I must say that I am somewhere between Sasha Vovin and Stefan Georg on the question of Turkic h-. I agree with Sasha (and like him am surprised anyone would not agree) that the examples cited by Doerfer from languages other than Khalaj have all the earmarks of sporadic secondary developments of no great antiquity. By the way, does anybody know why so many languages have these weird "inorganic" initial h's--and has there been any work on this? (What I mean is that some languages, e.g., Tubar in the Uto-Aztecan family, Polish or at least some dialects of it, and any number of others) prefix h- to some but not all words that etymologically begin with a vowel, without any regard to the doctrine of regularity of sound change. Why? Anyway, there is one Turkic language where I am not sure that I would share Sasha's view entirely, namely, Khalaj. It is obvious if one looks closely that even here the initial h- cannot be simply the reflex of Altaic (for anti-Altaicists, pre-proto-Turkic) *p- that Doerfer (the anti-Altaicist) as well as Poppe and Dolgopol'skij (pro-Altaicists) took it to be. There are altogether too many cases where Khalaj has h- but where the non-Turkic Altaic languages do not have (reflexes of) *p-. I have long suspected that the other h-'s in Khalaj are reflexes of an otherwise unreconstructed Altaic *w-, but have not done the detailed work required to see how well this idea works. One problem with this idea is that the only good test of it I know of at the moment is to compare Khalaj to Nostratic, which of course will not please people who are closed-minded on the subject of Nostratic. But of course no one on this list could be closed-minded, could we, so perhaps I can get someone to look at this problem... Alexis From sw271 at cus.cam.ac.uk Fri Oct 30 11:52:30 1998 From: sw271 at cus.cam.ac.uk (Sheila Watts) Date: Fri, 30 Oct 1998 06:52:30 EST Subject: r and s again Message-ID: ----------------------------Original message---------------------------- Since this is turning into a whole collection of r and s/z stories, here's another one. English names with a first syllable in CVr- are frequently turned into an affectionate form with an -l replacing the -r: Mary to Molly, Harry to Hal, Dorothy to Dolly and, for more modern examples, Derek to Del and Terry to Tel. However, there are also names in which the r tends to become z - Sharon to Shaz(za) and Barry to Baz(za), for instance (alongside names where the z is simply voicing, as in Gazza from Gascoigne). Is anyone aware of any work on this - what triggers the choice between l and z, for instance?There is clearly a dialect factor of some kind: if I judge it correctly, these are London forms, certainly not current in my native Ireland. Sheila Watts _______________________________________________________ Dr Sheila Watts Newnham College Cambridge CB3 9DF United Kingdom phone +44 1223 335816 From vovin at hawaii.edu Sat Oct 31 16:33:27 1998 From: vovin at hawaii.edu (Alexander Vovin) Date: Sat, 31 Oct 1998 11:33:27 EST Subject: h- in Turkic In-Reply-To: Message-ID: ----------------------------Original message---------------------------- Alexis, I think your message regarding initial h- raises one important methodological issue. Namely, can we allow a reconstruction of a segment for a proto-language that is preserved in a single language of otherwise big language family, and for which there is no second independent evidence? It seems that you would answer that question in the affirmative in this particular case, although I remember that once you yourself were bashing (quite justifiable, in my opinion, a person X from Moscow Nostratic school for search of IE and Nostratic accent distinctions uniquely preserved in Bengali). I would hate to disagree with you on Khaladj h-, but I think I have to. I would answer in the negative to the question I posed above, although I think that some exceptions could be allowed when a language that unikely preserves segment X, is on the top of the branching. In all other cases it is much safer to reconstruct something, especially something radical, like PT *h- on the basis of two independent pieces of evidence. Khaladj is probably *close* to the root of Turkic tree, but it does not represent primary branching, I think. It will be dangerous enough to reconstruct PT *h- on its sole evidence (although I think that this might eventually turn out to be true -- let us see), but looking for the traces of something Nostratic in khaladj *only*, does not seem to be very realistic. Cheers, Sasha ======================================= Alexander Vovin Associate Professor of Japanese Department of East Asian Languages and Literatures 382 Moore Hall 1890 East-West Road University of Hawaii at Manoa Honolulu, HI 96822 vovin at hawaii.edu fax (808)956-9515 (o.) t.(808)956-6881 (o.) From hubeyh at montclair.edu Sat Oct 31 16:33:53 1998 From: hubeyh at montclair.edu (H.M.Hubey) Date: Sat, 31 Oct 1998 11:33:53 EST Subject: Cladistic language concepts Message-ID: ----------------------------Original message---------------------------- Sean Brady wrote: > > In terms of using DNA sequence as comparative evidence for the > history of speciation (i.e. building a tree), the main concern is finding a > sequence of DNA that changes slow enough to preserve comparative > differences, but without changing so fast that these differences 'pile up' > on each other and obscure their history. Unfortnately, if the true history > is more like a bush than a tree, as Ghiselin discusses, any gene we may > pick will probably contain too few changes (i.e. be too slow) to > reconstruct this history. There was recently a science article in which the short-term mtDNA mutation rate was found to be much higher than the long-term mutation rate was thought/accepted to be. In the article it said/hinted that it could be because there was a limited number proteins that could occur and that some mutations went back to some older ones. What is the latest on this? I can see how something similar can account for the great disagreement in linguistics. For example, there is nothing to prevent a change r > z to change later to z >r and then look as if nothing happened. When we look at the results of the diffusion equation (the one that creates the Gaussian i.e. normal density) we see that if the changes in linguistics also obey this equation at least approximately (and they probably do because almost all math models implicitly use this Gaussian distribution) then if we examine intensities of change, then the largest number is always no change. IOW, if we could measure change and could assign numbers like level 1 change, level 2 change etc, then level 0 change (i.e. no change) is always greater than level 1 change, and it is also greater than level 2 change etc. So this means that it is a mistake to assume that if we see two words that seem exactly alike after 5,000 years that it is due to chance because after all, we expect more of no-change than one-sound-change, and more no-change than 2-sound-changes etc. This is a common belief that is exacerbated because many linguists also seem to assume that all changes in all languages should be occurring at the same rate as in the IE languages whereas other languages might be many times more stable. > A crucial question in both biological and linguistic reconstruction > seems to be this: Does our analysis yield a bush-like structure 1) because > there really is a bush-like structure; or 2) because our data and/or > methods of analysis cannot resolve the (true) tree-like structure. This is > a huge, and unsolved, problem in biological systematics. We simply cannot > distinquish between these two scenarios yet, although with the rapid influx > of DNA data, some people are beginning to tackle this problem. This is also a huge unsolved problem in linguistics :-) > Sean Brady sgbrady at ucdavis.edu > > Population Biology Graduate Group tel. (530) 752-9977 -- Best Regards, Mark -==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= hubeyh at montclair.edu =-=-=-= http://www.csam.montclair.edu/~hubey =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer. =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=