Once Again -- Is Japanese an Isolate? Or Genetically Related to Something or Other? Or Not?

[Ross Bender]

Two recent articles by eminent historical linguists have surfaced on Academia.edu: John Whitman's 2011 paper "Northeast Asian linguistic ecology and the advent of rice agriculture in Korea and Japan", and J. Marshall Unger's 2014 paper "No Rush to Judgment: The Case against Japanese as an Isolate."

What I find interesting about these papers is the admissibility of historical and cultural material as evidence. In general I have found historical linguistic discussions on the possibility of Japanese as an Altaic language, its possible relationship to Korean, etc  very difficult to follow. This is due to the mathematical severity (and to me opacity) of the methodology.

However, both Whitman and Unger discuss the history of agriculture and particularly the Yayoi settlement of Japan as having something like equal weight to the strict algebraic formulae of historical linguistics per se. Unger even quotes Goethe!!

The appendices to Unger's paper include some colorful maps that help explain things to even one as mathematically challenged as myself. I would welcome other PMJS'ers comments on these papers and the age-old question of where Japanese came from.

I should say that when I was in grad school back in the Paleolithic, the most compelling hypothesis was that Japanese was a Mischsprache with a Polynesian substrate and an Altaic overlay. Then came Roy Miler's book Japanese and Other Altaic Languages (1971). I was studying Korean history with Gari Ledyard when he published his exciting "Galloping Along with the Horseriders." (1975). And the beat goes on....

Ross Bender

https://independent.academia.edu/RossBender

[Michael Pye]

Dear Colleagues,
Alexander Vovin has also just (re-) published a long review on the
subject on Academia.edu. which can be found via his name, in which he
fully explains and roundly declares how the case for Ur-Altaic cannot
be mounted. He argues that people for various reasons "want" it to
have been like that. (I must admit that I do find the idea
sentimentally a bit attractive.)
In my humble opinion, it seems to me that there may be an underlying
problem with the use of the word 'isolate', which should refer
(correct me if I am wrong) to a language for which no relatives can be
identified, but to no more than that. It does not mean that such a
language dropped from the sky or developed in a totally separate
evolutionary line among higher primates. If homo sapiens is a single
species (widely so perceived, I believe), there must have been a time
when, etc. etc...even if these times are now completely untraceable
and later linguistic relationships cannot be proven. In other words,
"isolate" does not have an infinite reference, only a finite one.
Note also that there are various ways in which what is later perceived
to be "a language" can come into being. Where did Bahasa Indonesia (as
"a language") "come" from, for example?
Anyway, Japanese is currently in a state of rapid destruction. It
(whatever it may at other times have been) is being replaced by
something else which is not derived from any one source but which will
not exactly be an isolate either. Such is the flow of things in this
transitory world...
So it seems.
best wishes,
Michael Pye


Zitat von Ross Bender <rosslyn...@gmail.com>:

> Two recent articles by eminent historical linguists have surfaced on
> Academia.edu: John Whitman's 2011 paper "Northeast Asian linguistic ecology
> and the advent of rice agriculture in Korea and Japan", and J. Marshall
> Unger's 2014 paper "No Rush to Judgment: The Case against Japanese as an
> Isolate."
>
> What I find interesting about these papers is the admissibility of
> historical and cultural material as evidence. In general I have found
> historical linguistic discussions on the possibility of Japanese as an
> Altaic language, its possible relationship to Korean, etc very difficult
> to follow. This is due to the mathematical severity (and to me opacity) of
> the methodology.
>
> However, both Whitman and Unger discuss the history of agriculture and
> particularly the Yayoi settlement of Japan as having something like equal
> weight to the strict algebraic formulae of historical linguistics per se.
> Unger even quotes Goethe!!
>
> The appendices to Unger's paper include some colorful maps that help
> explain things to even one as mathematically challenged as myself. I would
> welcome other PMJS'ers comments on these papers and the age-old question of
> where Japanese came from.
>
> I should say that when I was in grad school back in the Paleolithic, the

> most compelling hypothesis was that Japanese was a *Mischsprache *with a

> Polynesian substrate and an Altaic overlay. Then came Roy Miler's book

> *Japanese
> and Other Altaic Languages *(1971). I was studying Korean history with Gari

> Ledyard when he published his exciting "Galloping Along with the
> Horseriders." (1975). And the beat goes on....
>
> Ross Bender
> https://independent.academia.edu/RossBender
>

> --
> PMJS is a scholarly forum.
>
> You are subscribed to PMJS: Premodern Japanese Studies.
> To post to the list, send email to pm...@googlegroups.com
> To unsubscribe, send email to pmjs+uns...@googlegroups.com
> Visit the PMJS web site at www.pmjs.org
> Contact the group administrator at edi...@pmjs.org



.................................................................................................................
Professor (em.) of the Study of Religions, University of Marburg
Research Associate in Buddhist Studies, Otani University, Kyoto
.................................................................................................................
"Japanese Buddhist Pilgrimage" now in print! (hardback and paperback)
http://www.equinoxpub.com/home/japanese-buddhist-pilgrimage/
"Strategies in the Study of Religions" Vols 1-2 (2013):
http://www.degruyter.com/view/product/184080 and.../184339

[Hanna McGaughey]

Dear List,

Thank you for touching on such a fascinating topic. I am not a linguist, but language and linguistic research fascinates me. The question how language develops in comparison to genetic inheritance is an important one that should be regarded in such a conversation as this. Keep in mind that not all linguistic acquisition is inter-generational; Members of the list are proof that horizontal acquisition is possible. A recent article in Ars Technica (http://bit.ly/15GRXPn) references the PNAS publication of a study (earlier this year) that suggests, "migration within geographic regions shapes phoneme evolution, although human expansion out of Africa has not left a strong signature on phonemes" (http://bit.ly/1aZ7Vam). This, of course, is unlike our genome, which shows traces of our African heritage. The implications of this include that "although geographically isolated populations lose genetic diversity via genetic drift, phonemes are not subject to drift in the same way: within a given geographic radius, languages that are relatively isolated exhibit more variance in number of phonemes than languages with many neighbors." In other words, geographic proximity corresponds to phonemic similarity even among languages from different language families. Geographically isolated languages vary more greatly from other languages including their closest neighbors than less isolated languages. Also, while isolated genomes have less diversity, isolated languages grow in (phonemic) diversity. Because Japanese culture is not as isolated as it used to be (information technology and globalization of course works to overcome geographic isolation), its phonemes have increasingly resembled those of other languages its speakers interact with. For example, Japanese speakers today are able to pronounce /v/ unlike members of even a generation earlier. Like all individual studies, the results of this study should not be generalized indiscriminately, but it does emphasize that not all aspects of language can be explained using the family model. I think it's possible to maintain if not sentimentality then fascination concerning the uniqueness and complexity of linguistic development and variation in relation to geographic dispersal.

Best,

Hanna McGaughey

To unsubscribe, send email to pmjs+unsubscribe@googlegroups.com

Visit the PMJS web site at www.pmjs.org
Contact the group administrator at edi...@pmjs.org

.................................................................................................................
Professor (em.) of the Study of Religions, University of Marburg
Research Associate in Buddhist Studies, Otani University, Kyoto
.................................................................................................................
"Japanese Buddhist Pilgrimage" now in print! (hardback and paperback)
   http://www.equinoxpub.com/home/japanese-buddhist-pilgrimage/
"Strategies in the Study of Religions" Vols 1-2 (2013):
   http://www.degruyter.com/view/product/184080  and.../184339

--
PMJS is a scholarly forum.
You are subscribed to PMJS: Premodern Japanese Studies.
To post to the list, send email to pm...@googlegroups.com

To unsubscribe, send email to pmjs+unsubscribe@googlegroups.com

[Alexander Vovin]

Dear all,

Several notes.

1) Japanese is not an isolate, at least not in the sense like Zuni or Kusunda are. It is a member of a portmanteau familym which we call Japonic. Japonic includes two branches: Japanese (2 languages: Japanese and Hachijoo) and Ryukyuan (4 or 5 languages: Amami-North Okinawa and South-Central Okinawa (Northern Ryukyuan), Miyako and Yaeyama (Southern Ryukyuan). The status of Yonaguni remains unclear -- it might be an aberrant Yaeyama dialect, but some reflexes may point out to its status as an independent language). In addition, Japonic languages were once spoken in the South and Center of the Korean peninsular -- it is not clear how many, but probably between 2 and 4. All what remains from them are some glosses in Korean and Chinese historical sources.

2) In the mainstream historical linguistics no other evidence is admissible for the genetic relationship of languages besides linguistic evidence itself. As Lyle Campbell once put it, languages must speak for themselves. Historical or any other evidence cannot prove language relationship, and reliance on it is a great methodological mistake. In the bestpossible scenarionit can bgo along with the linguistic evidence, but historical etc. evidence can be easily turned around. Let me give the following example: it will be very easy to claim that Japonic is from South-East Asia just on the historical and cultural features:

(1) worship of the sun is SEA feature as opposed to the worship of the sky in the 'Altaic' world

(2) divine kingship is again SEA feature as opposed to the ruler as the leader of the band of warriors in the 'Altaic' world

(3) rice is the basis of economy vs. the monadism with some supplemental agriculture

(4) Japanese traditional houses are of SEA type, and not yurts

(5) war mounts are not gelded as in the 'Altaic' world

(6) Japanese bows are simplex like in SEA, and not composite like bowsin the 'Altaic' world

Etc., etc., etc. But what does it tell us about the origins of the Japonic as a language family? Precious nothing.

3) Sorry, Ross, but there is no other way to go except by algebraic formulas. And they are actually not that difficult -- read , e.g. Lyle Campbell's Introduction to the Historical Linguistics -- the methodology will be crystal clear. So, to prove that two or more languages are related, one has to demonstrate that they share either paradigmatic morphology or basic vocabulary (and ideally both) established on the basis of recurrent and productive-predictive phonetic correspondences. Just a very brief demonstration how it works.

                                Tokyo                   Shuri

'forest'                       mori                      mui

'rare'                         mare                     mari

'that'                         are                        ari

'this'                         kore                      kuri

'temporary'                kari                       kai

I need only 5 words above to demonstrate that modern Tokyo and Modern Shuri Okinawan are likely to be related. We get the following recurrent correspondences:

                                 Tokyo                   Shuri

                                  m                        m

                                  o                         u

                                  a                         a

                                  e                         i

                                  i                          i

                                 r _/e                     ri

                                 r_/i                       0i

It might seem that we have T e and i corresponding to only i in Shuri, which is true -- we have a merger in Shuri, and we can do nothing about it, but the two types of correspondence for T r are conditioned in Shuri by the following vowel: in fron of e, Shuri r stays, in front of i it becomes zero. Here is where the productive-predictive factor comes in: given the Shuri form we can predict its shape in Tokyo, and vice versa.

      Now, there is a sharp contrast between the relationship between Tokyo and Shuri on the one hand, and between Japonic and Korean, and even more so with 'Altaic' on the other. THe problem is that nobody so far managed to come up with the same kind of presentation for Japonic and Korean or Japonic and 'Altaic'. On the contrary, regular correspondences are lacking, and as a result every single etymology is saved in its own unique way. And there are other numerous problems, which are discussed in detail in my 2010 Koreo-Japonica monograph.

4) 'Altaic' itself is not a valid family (unlike universally recognized Indo-European, Urallic, Semitic, Mayan, Uto-Ztecan, etc.), but a Sprachbund: most of similarities between Turkic, Mongolic, and Tungusic are result of centuries-long contact, and others are "encounters of the third kind", as once put by Stefan Georg --  they are simply chance resemblances or invented etymologies. The debate still continues, but the propenents of areal rather than genetic relationship now hold the uper hand. Since one still can see in the non-specialist literature even 'Uralo-Ataic' which was disproven more than a century ago, we should give it another hundred years before 'Altaic' goes into oblivion.

5) There are no reliable Austronesian or Malayo-Polynesian etymologies for Japonic: all of them are "encounters of the third kind". This might seem puzzling, especially in the case of South Ryukyus, but the possible explanation is that the peopling of South Ryukyus from Northern Taiwan was discountinued at an early date, and when Ryukyuans finally reached South Ryukyus they found uninhabited islands.

All the best,

Sasha

Alexander Vovin
Directeur d'études, linguistique historique du Japon et de l'Asie du Nord-Est

ECOLE DES HAUTES ETUDES EN SCIENCES SOCIALES

CENTRE DE RECHERCHES LINGUISTIQUES SUR L'ASIE ORIENTALE

131, bd Saint-Michel, 75005 Paris

Adjunct Professor of Linguistics

Department of Linguistics

University of Hawai'i at Mānoa, Honolulu
HI 96822, USA

sasha...@gmail.com

To unsubscribe, send email to pmjs+unsubscribe@googlegroups.com

Visit the PMJS web site at www.pmjs.org
Contact the group administrator at edi...@pmjs.org

.................................................................................................................
Professor (em.) of the Study of Religions, University of Marburg
Research Associate in Buddhist Studies, Otani University, Kyoto
.................................................................................................................
"Japanese Buddhist Pilgrimage" now in print! (hardback and paperback)
   http://www.equinoxpub.com/home/japanese-buddhist-pilgrimage/
"Strategies in the Study of Religions" Vols 1-2 (2013):
   http://www.degruyter.com/view/product/184080  and.../184339

--
PMJS is a scholarly forum.
You are subscribed to PMJS: Premodern Japanese Studies.
To post to the list, send email to pm...@googlegroups.com

To unsubscribe, send email to pmjs+unsubscribe@googlegroups.com

[Ross Bender]

My apologies for not posting the links to the papers I referred to, Whitman, Unger, and also Vovin:

https://www.academia.edu/4418899/Northeast_Asian_linguistic_ecology_and_the_advent_of_rice_agriculture_in_Korea_and_Japan

https://www.academia.edu/8065288/No_Rush_to_Judgment_The_Case_against_Japanese_as_an_Isolate

https://www.academia.edu/6345901/The_end_of_the_Altaic_controversy

There are also some other new or recent papers on Japanese historical linguistics now on Academia.edu.

Ross Bender

https://independent.academia.edu/RossBender

To unsubscribe, send email to pmjs+uns...@googlegroups.com

[Howell, David]

Dear Sasha,

Thank you for this very clear explanation. If I might extend the conversation, what is the latest thinking on Ainu’s relationship to other languages?

Thanks,

David Howell

Harvard University

To unsubscribe, send email to pmjs+uns...@googlegroups.com

[Alexander Vovin]

Dear David and all,

Well, Ainu is even more mind-boggling case than Japonic. Another portmanteau family with two branches: Hokkaidoo-Kuril and Sakhalin, but in the past spread at much greater territory -- from North-Western Kyuushuu up to Southern Kamchatka, possibly also on Asian mainland, but this is much more controversial. The Ainu languages that were lucky enough to be recorded show much less internal diversity than Japonic, and complete absence of non-kana based transcriptions prior to the 18th century further impedes the recovery of the Ainu language history. No apparent external relatives, although similar to Japonic numerous hypotheses were proposed (none of them successful including my own attempt in 1993 to connect it with Austroasiatic (:-), also earlier suggested by Olaf Gjerdman). Some words or morphemes, like first person singular subject agreement marker ku- 'I' or ape ~ apoy 'fire' look like Austronesian, but there are very few of them, certainly not enough to establish any reccurrent predictive-productive correspondences, and in addition there are other problems. Thus, e.g. apuy is Malayo-Polynesian form, but proto-Austronesian was something like *Sapuy, but Ainu shows no traces of this *S-. If indeed related to Austronesian, we would expect correlation with Austronesian data, in short, Ainu cannot possibly be just related to Malayo-Polynesian.

         Among its neighbors, contacts with Japanese and Nivx (Gilyak) were most extensive, and they went both ways. Many words in Eastern Old Japanese are of the Ainu provenance, but even modern Japanese has Ainu loans, such as, e.g. iruka 'dolphin' or rakko 'sea otter'. Contacts with Tungusic also took place, but they were more limited.

         Besides being Subject-Object-Verb language like all its neighbors, structurally Ainu is radically different from all its neighbors, except in some respects it shares structural and material similarities in morphology with Nivx, but these are likely to be traces of Ainu influence, and certainly, typology tells us nothing about the genetic relationship.

All the best,

Sasha

Alexander Vovin
Directeur d'études, linguistique historique du Japon et de l'Asie du Nord-Est

ECOLE DES HAUTES ETUDES EN SCIENCES SOCIALES

CENTRE DE RECHERCHES LINGUISTIQUES SUR L'ASIE ORIENTALE

131, bd Saint-Michel, 75005 Paris

Adjunct Professor of Linguistics

Department of Linguistics

University of Hawai'i at Mānoa, Honolulu
HI 96822, USA

sasha...@gmail.com

[Daniel Botsman]

Dear Professor Vovin,

With apologies for my profound ignorance, does this mean that Korean, Manchū etc are also historically unrelated to Japanese? 

Thank you in advance for your willingness to enlighten the ignorant on these basic matters!

Daniel Botsman

Yale University

[robin d. gill]

The Unger article is long and reads beautifully on-site using the arrow up and down to move (or can be downloaded). The ten or so simple maps of the languages moving over time is after the bibliography and English and Japanese summaries follow the pictures at the very bottom. For non-linguists, the longer Japanese summation may be the most easy read.

 A few questions asked on this thread are answered. 

To my unstudied eyes, it seems that despite differences in the value of mathematical tables vs historical educated guestimation (though rach uses some of the other type of evidence), there seems to be substantial agreement in the gist of Prof Vovin and Ungers conclusions.  Do I have that right?

robin d gill

"Rise, Ye Sea Slugs!"

[Alexander Vovin]

Dear Prof. Botsman,

Yes, it does. Imho, the much simpler and more elegant solution is the loanword scenario. Korean-Tungusic "cognates" are limited mostly to Manchu and Jurchen -- these were the peoples whom first Koguryeo and then Parhae ruled for centuries. Ditto for Korean and Japonic: the 'cognates" are predominantly found in Central Japanese, that is Old and Middle Japanese -- exactly the dialect group that was in contact with Korean for centuries as well. Hope this helps,

Sasha

Alexander Vovin
Directeur d'études, linguistique historique du Japon et de l'Asie du Nord-Est

ECOLE DES HAUTES ETUDES EN SCIENCES SOCIALES

CENTRE DE RECHERCHES LINGUISTIQUES SUR L'ASIE ORIENTALE

131, bd Saint-Michel, 75005 Paris

Adjunct Professor of Linguistics

Department of Linguistics

University of Hawai'i at Mānoa, Honolulu
HI 96822, USA

sasha...@gmail.com

[Alexander Vovin]

No, Robin, I think you are seeing too much agreement where it really does not exists. Well, everybody seems to agree nowadays that Japonic arrived to Japan via Korean peninsula. But here it effectively ends: neither the most important question: whether the relationship is genetic or areal, and even the dating of Japonic's arrival to Japan are still subject of debate.

Sasha

Alexander Vovin
Directeur d'études, linguistique historique du Japon et de l'Asie du Nord-Est

ECOLE DES HAUTES ETUDES EN SCIENCES SOCIALES

CENTRE DE RECHERCHES LINGUISTIQUES SUR L'ASIE ORIENTALE

131, bd Saint-Michel, 75005 Paris

Adjunct Professor of Linguistics

Department of Linguistics

University of Hawai'i at Mānoa, Honolulu
HI 96822, USA

sasha...@gmail.com

[Ross Bender]

I have taken the liberty of transcribing summary statements from both Whitman and Unger, for those who do not have time to read the entire articles. These paragraphs do not capture the entire argument of either scholar, and I would urge all interested to look at the texts themselves. Also, both papers include excellent maps well-worth perusing.

Northeast Asian Linguistic Ecology and the Advent of Rice Agriculture in Korea and Japan

John Whitman

Rice (2011) 4:149-158

 

Conclusion (p. 157)

In this paper, I have sketched a specific historical scenario that attempts to explain the linguistic ecology of the non-Sinitic language families in Northeast Asia associated with wet rice culture, Japonic and Koreanic. This scenario is couched within the general hypothesis of a diffusion of agriculture from the area around the Shandong peninsula to the north and east. According to the scenario, Japonic arrives in the Korean peninsula around 1500 BCE and is brought to the Japanese archipelago by the Yayoi expansion around 950 BCE. On this view, the language family associated with both Mumun and Yayoi culture is Japonic, although the association of a culture in the archaeological sense with a single language family is almost certainly an oversimplification.

            Koreanic arrives in the south-central part of the Korean peninsula around 300 BCE with the advent of the Korean-style bronze dagger culture. Its speakers coexist with the descendants of Mumun cultivators, and thus with Japonic, well into the common era. Each of these demic diffusions, as well as the later dispersions of Koreanic and Japonic, result in founder effects which diminish the internal variety of the language family. Japonic and Koreanic, as well as possibly other Northeast Asian languages, share some agricultural vocabulary, but this shared vocabulary precedes rice farming.

 

No Rush to Judgment: The Case Against Japanese as an Isolate

J. Marshall Unger

国語研 プロジェクトレビッユー

 

NINJAL Project Review, Vol. 4 No.3 pp.211-230 (February 2014)

(Summary of Part A from Appendix I) pp. 212-13

            In part A of the outline in Appendix I, our present understanding of Japan during the 1^st millennium BCE is summarized under three main headings.

·         First, a major population replacement began sometime between about 950 and 350 BCE. The population of Final Jōmon period Japan began to be replaced by genetically dissimilar people around this time. Today, a few pre-Yayoi inputs to the Japanese gene pool remain.

·         Second, the cause of this population replacement was a migration from what is now southern Korea. The migrants brought with them the full complement of Korean Mumun or Megalithic culture, all the salient elements of which are reflected in the Yayoi culture of Japan. Settling first in northern Kyūshū, the migrant population grew, and after about 200 BCE, they expanded to the north and east rapidly. The distribution and dating of Yayoi settlements does not show a steady push east and north out of Kyūshū but rather leaps to sites well suited for wet-field rice. Also, the migration was not a single event.

·         Third, in contrast to the Yayoi expansion out of Kyūshū, which was rapid once it began, the transition from Yayoi to the succeeding Tumulus or Kofun culture, which started in the latter half of the 4^th century CE, was gradual. It did not involve a single or sudden disruption. Although we can clearly see peninsular inputs in Kofun culture, most archaeologists today reject the so-called horserider invasion theories of the Japanese historian Egami Namio (1967) and the more recent Korean research Hong Wontack (1994, 2006.)

Money quote from Unger Appendix I, 10 d. (under C. Tentative Conclusions):

“Any claim that Japanese or Korean is a true isolate must be affirmatively reconciled with non-linguistic data from archaeology and human genetics.”

This statement of Unger's does seem to be somewhat at variance with Vovin's insistence on linguistic evidence alone. To quote Vovin above:

" In the mainstream historical linguistics no other evidence is admissible for the genetic relationship of languages besides linguistic evidence itself. As Lyle Campbell once put it, languages must speak for themselves. Historical or any other evidence cannot prove language relationship, and reliance on it is a great methodological mistake. In the best possible scenario it can go along with the linguistic evidence, but historical etc. evidence can be easily turned around."

Finally, pace Vovin, as a historian I find Whitman's emphasis on rice culture and Unger's on archaeology and human genetics to be more congenial, or at least easier to comprehend, than the historical linguistic approach alone. At any rate the work of all three scholars helps to illuminate very important periods in Japanese pre-history, and demonstrates that conversation is ongoing and dynamic.

Ross Bender

https://independent.academia.edu/RossBender

[Michael Pye]

Dear colleagues,
I refer to the following quoted by Ross Bender:

*“Any claim that Japanese or Korean is a true isolate must be affirmatively
reconciled with non-linguistic data from archaeology and human genetics.”*
[Money quote from Unger Appendix I, 10 d. (under C. Tentative
Conclusions)]

I'm still bothered about the idea of an "isolate" (I do know the usual
meaning of its use in linguistics), especially when we come across the
phrase "a true isolate". Don't we have to understand the assertion
that a language is an isolate as an admission of unfinished (and
possibly unfinishable, not further traceable) business? Otherwise,
theory lurches into some kind of genetic-racially essentialist view.
If homo sapiens is a single species there can probably be no such
thing as "a true isolate". Admittedly humans can invent the same thing
independently of each other, but speech....? That would be a difficult
theory to maintain, would it not?
Michael Pye
PS The weather is now improving in wagakuni.

[William Wetherall]

Dear Michael and others on this thread:

I share your concerns. I came of age with the "genetic" and "areal" and
"isolate" metaphors in the late 1960s and early 1970s like a number of
other members of this list. I have questioned the utility of the
"genetic" and "isolate" metaphors since the 1990s. Replacing "genetic"
and "areal" with "internal" and "external" would allow us to account for
all manner of linguistic changes, some perceived as the effects of
transmission "within" a population of speakers (most essentially from
older native speakers to children), others seen as the result of
intercourse "between" populations of speakers of other languages (most
essentially in the process of direct contact between their speakers).

There is no need to "stigmatize" any language as an "isolate" simply
because the multiple trails of historical evidence that would reflect on
the extent and nature of its "internal" and "external" links with other
languages happens to run dry. The very fact that we can postulate the
existence of a "proto Japanic" language and a "para Japanic" variety --
preceding what a better body of evidence has led to our definition of
"old Japanese" -- requires that our "null hypothesis" be that "Japanese"
is not an "isolate" but merely another language that, throughout its
known "history" and its plausible "pre-history", has constantly been
exposed to the "internal" and "external" effects of "normal" human
intercourse, whether "within" the population of Japanese speakers, or
between Japanese speakers and speakers of other languages.

In other words, the end of a trail of evidence is not a cause for
exceptionalism. The "linguistic history" of Japan has not ended simply
because the trail appears to fade away in the sands of time. Hence this
"once again" discussion. The excitement continues.

The following paper by Roger Lass (Genetic Metaphor in Historical
Linguistics, Alteration, Volume 10, 2003) critically examines the
effects of using "genetics" as a metaphor for what I would call
"internal" change.

http://alternation.ukzn.ac.za/Files/docs/10.1/04%20Las.pdf

Bill Wetherall

[Ivan Rumánek]

Dear Sasha,

   I am enjoying this thread and thanks for your comments. A small reflexion on Ainu and its obvious similarity to Japanese as a SOV language: I have been wondering whether it might to a huge extent be a devastating influence from Japanese. Ainu shares with Japanese many striking syntactical features, word-order (and indeed particle order) being basically identical. I have been witnessing alien influences in near-extinction languages, even such that won the fight and turned viable, like Welsh (its alveolar T, D are clearly English-based, unlike the presumable Celtic dentals, like the non-palatalized Gaelic T, D being as dental as in all the non-Germanic languages of Europe). In the same manner, the modern efforts to bring Ainu back to life after a period of interruption when there were no more native Ainu speakers (people who learned it from their mothers), the revitalized Ainu has the undeniable Japanese U, as a result of its being studied and learned by native Japanese speakers. Examples like these led me to wondering whether the  syntactic word order so obviously parallel to the Japanese might also be the result of Japanese influence. After all, the Ainu were in contact with the Japanese from the 17th century at the latest, and the cut of the "Ainu traditional costumes" also shows an undeniable Japanese influence (they are virtually a Japanese kosode with Ainu patterns...), so we might presume that the period of bilingualism - the start to every language extinction - was fairly earlier that the first written records Russian and other travelleres made of the Ainu language, and that acutally the language they recorded was ALREADY very influenced by Japanese... What would you think of this?

Ivan Rumánek,

Masaryk Universtiy, Brno

[Unger, James]

(particularly for Michael and William)

Sasha can speak for himself, but in my paper, I use "isolate" as a non-prejudicial technical term and distinguish between epistemic and deontic isolates.

Every language is epistemically an isolate unless and until it is explicitly shown by the comparative method to be related to some other languge(s). To say that language X is an isolate in this sense is merely a comment on the current state of the discipline.

Someone who says that all traces of the language(s) from which X diverged historically have vanished since the time X moved into its present range is claiming that it is a deontic isolate. S/he is, in effect, predicting that the problem of the origins of language X is in principle insoluble because the odds of any useful evidence turning up at this point is virtually nil.

At the 2011 conference, at which Sasha, John Whitman, Thomas Pellard, I, and others presented relevant papers, Sasha argued that Japanese (or Japonic, if you prefer) should be thought of as an isolate in the deontic sense, unrelated to Korean, Manchu, Ainu, Austronesian or any other language or language family to which we have access. He may have changed his views in the meantime. Both he and I are skeptical of the old Altaic hypothesis; both of us have no problem with the concept of a young, shallow Japonic family. But from there, we start to disagree. I am not convinced that proto-Ryukyuan, though old, is necessarily a first-order daughter of proto-Japonic. I do think that proto-Korean-Japanese can be rigorously verified. And I would not rule out an eventual demonstration of a Macro-Tungusic, including Korean and Japanese.

In my paper, I discuss the role of non-linguistic evidence. Briefly, linguistic inferences cannot be drawn directly from archaeological, genetic, or cultural data, but it is legitimate to consider them when estimating which linguistic hypotheses are worth exploring and which are historically implausible.

Finally, as for areal versus genetic relations, the important thing is not the terminology, in my opinion, but remembering that languages that have diverged from a common source may later come into contact and influence one another. The hard part is sorting out resemblances and accounting for them with a robust linguistic theory that does not depend on unlikely non-linguistic events (e.g. a full-scale military invasion of Yamato by Paekche). On areal influences, Sasha and I also disgree. In his 2009 book, he argues that most of the etymologies that Martin and Whtiman advanced in support of the pKJ hypothesis ought to be rejected; the evidence is almost entirely to be accounted for as the result of borrowing. In my 2009 book, I argue that there are good criteria for distinguishing borrowings from common inheritances, and that borrowings from Old Korean into Old Japanese number only about three or four dozen. This is, in a nutshell, why we disagree about the pKJ hypothesis.

Jim Unger

[Alexander Vovin]

Dear all,

Let us start with a Ryukyuan as a first-order sister branch of Japanese. This state of affairs, I believe, is universally recognized (except Jim) by eventually all Western scholars and increasingly by Japanese scholars as well.And to the best of my knowledge, Jim has yet to present any concrete data in support of his hypothesis. The definition of Ryukyuan and Japanese as first=order branches of insular Japonic rests, of course, on the mutually exclusive innovations. And these are plenty, especially in morphophonology, morphology, and lexicon. But there are also mutually exclusive innovations in phonology and possibly in syntax. Thus, we do have at least two branches, not just one, and this contradicts the classical definition of an isolate. Peninsular languages constitute at least one more branch, possibly of even a higher order, and coordinate with insular Japonic. Thus, we deal here with a language family, not just with one language. Yes, I do not see that any evidence presented so far demonstrated that Japonic is demonstrably related to any other language family. But that makes Japonic no more isolate than Indo-European, which is also not demonstrably related to any other language family (and size does not matter).

         I trust that it is much more productive to discuss real data than personal opinions. Therefore, since Jim believes that Koreo-Japonic can be rigorously defended, I invite him to give a small presentation similar to the one I have done with modern Tokyo and Shuri. Please take five words from Old Japanese and Middle Korean and demonstrate the same kind of predictive-productive regular correspondences that are possible to establish between Tokyo and Shuri (needless to say, without any unaccounted segments or speculative semantics). Or, alternatively, please show the common paradigmatic morphology for Old Japanese and Middle Korean. Surely, either one or another or both can be done for any uncontroversially family of languages that have been demonstrated to be related. I do not think it will be possible either for Japanese or Korean, or for Japanese and Tungusic, or for Korean and Tungusic. As an ex-'Altaicist' myself, I have been there, have done that.

         On the other topic -- how far can we recover the linguistic history of Japanh. I trust that it is useful to differentiate betwee facts, possibility of recovering, and mere speculations. The genetic relationship (sorry, but I personally see no harm in this term) between Japanese and Ryukyuan is a fact. We also know that Japonic is a relatively recent newcomer to the archipelago. What do we know about the linguistic history of Japanese archipelago before the Japonic? Unfortunately, not much. There is, Ainu of course. And then there is Hayato, but we know only two words from the Hayato language, so it is practically impossible to pin it down. It would be a folly to imagine that only Ainu and Hayato were spoken throughout the islands during the Joomon period -- it is simply too long not to have a substantial linguistic variety. We know also of Kumaso and Emisi, but do not have a single word from their languages. It is further unclear whether Emisi are Ainu or not, and whether Kumaso have to be connected with Hayato. But there might be some limited recovery through the study of old placenames, provided that some of them would have translations like it is done on the occasion in Fudoki. But the linguistic history of Joomon, I am afraid, will be forever beyond of our grasp, although we can, of course indulge in the wildest speculations like that Martian was once spoken in Japan.

All the best,

Sasha

Alexander Vovin
Directeur d'études, linguistique historique du Japon et de l'Asie du Nord-Est

ECOLE DES HAUTES ETUDES EN SCIENCES SOCIALES

CENTRE DE RECHERCHES LINGUISTIQUES SUR L'ASIE ORIENTALE

131, bd Saint-Michel, 75005 Paris

Adjunct Professor of Linguistics

Department of Linguistics

University of Hawai'i at Mānoa, Honolulu
HI 96822, USA

sasha...@gmail.com

[Alexander Vovin]

Dear Ivan,

You are possibly right, but the situation is probably more complex. On the one hand, Ainu sits in a heavy SOV neighborhood, and has predominantly prefixal morphology, which does not agree well with SOV (although there are exceptions, like Navajo). On the other hand, after Ainu migrated to Hikkaidoo under Japanese pressure, the contacts before the seventeenth century were not intense. We also do not know how much Ainu was Japanicized before this migration (except one very short text which does not have overt objects), all we have are very few glosses prior to 17th c. Finally, Nivx, can be also responsible for the SOVization of Ainu. But in 17th-18th c. Japanese and Russian records we clearly have SOV. And finally, I think that there was a Japanese-Ainu bilingualism, at least in the Kantoo area, but these were apparently not the guys who migrated to Hokkaidoo -- they were assimilated to Japanese prior to migration.

Modern Ainu, before it went extinct several years ago, was of course heavily Japanicized: e,g. Ainu intervocalic -p- -t- -k- which as we know from the earlier records were photecally voiced or half-voiced, were pronounced recently like voiceless, etc.

All the best,

Sasha

Alexander Vovin
Directeur d'études, linguistique historique du Japon et de l'Asie du Nord-Est

ECOLE DES HAUTES ETUDES EN SCIENCES SOCIALES

CENTRE DE RECHERCHES LINGUISTIQUES SUR L'ASIE ORIENTALE

131, bd Saint-Michel, 75005 Paris

Adjunct Professor of Linguistics

Department of Linguistics

University of Hawai'i at Mānoa, Honolulu
HI 96822, USA

sasha...@gmail.com

[William Wetherall]

Thanks very much for the lucid summary, James. I had read your article,
but the way you clarified the nuances of "epistemic" and "deontic" in
your last post was very helpful.

Thanks also to Professor Vovin for pointing out, every bit as lucidly,
that Japanese fails to satisfy the "classical definition of an isolate"
-- so long as one accepts his view that Japanese and Ryukyuan are
offspring of a common parenthe calls "insular Japonic", apart from
whether this "insular Japonic" might itself stem from a "peninsular" or
"continental" predecessor. I'm not taking sides on the relationship
issue. I'll ride the bus and leave the arguments to the linguists who
control the data.

Of interest to me is how Sasha then qualifies the meaning of "isolate".
He argues in effect that all constellations of languages -- whether as
small as Japano-Ryukyuan or as large as Indo-European -- are equally
"isolates" relative to one another -- "size does not matter" he says.

But if size is irrelevant, then why regard even a single language, which
defies grouping on phonological grounds, an isolate? Is a single star
somehow more "isolated" from a nearby double-star, or from a cluster of
stars or a galaxy, than such larger groupings of stars are from each
other? A "lone star" perhaps, but not isolated.

Even if we view Japanese as an "isolate" in the classical linguistic
sense, among the other languages of Central and Northeast Asia, it has
nonetheless functioned as all languages must in the environment of other
languages, as a mediator of human civilization between the speakers of
the languages. It is made of the same stellar material as all other
languages -- sounds and streams of sounds that convey meaning. And the
content of its known working vocabulary at any confirmable or reasonably
conjecturable point in time will be a reflection of the variety of
civilization it mediated at the time.

Historical linguistics should, IMHO, be the study of language systems
over time -- how not only sounds, but meanings and the structures that
convey meanings, are conserved or change under various conditions, from
geographic isolation to intense intercourse with speakers of neighboring
or even non-regional languages. Why should the "relationship" between
two languages hinge on only their phonological relationship? Sure,
languages are essentially spoken, and hence their sounds -- and how
their sounds are transmitted from generation to generation, usually but
not necessarily from native-speaker adults to their own children, is an
important component of a language. And arguably a language's
phonological system is its most conservative, hence least mutable
component, and possibly this warrants giving phonology more weight in
determinations of relationship. But sounds alone do not a language make.
Meaning, and the ways in which meaning is structured for the purpose of
conveyance, are equally important to a language, in terms of the
identity and vitality of its speakers, individually or collectively. And
it seems to me that meaning and the structures that convey meaning are
essential to a determination of the "relationship" between language
systems, again meaning the populations -- the individuals, communities,
nations and whatever -- that speak the languages, i.e., mediate their
human conditions.

My main argument for abandoning the "genetic" metaphor is not that the
analogy to genes breaks down, but that speaking of a "genetic
relationship" encourages the racialization of language in
ethnonationalist ideology, just as speaking of Japanese as an "isolate"
feeds romantic exceptionalism. Everything argued by historical linguists
could be argued within a framework of "internal factors" that operate
within a language system, and "external factors" that operate between
language systems, in a way that encourages people to embrace a view of
Japanese as just another language. No big deal.

Obviously linguists need to insist on high standards of proof in all
arguments. It's easy, and fun, to speculate, but linguistic
"historiography" has to be based on sound evidence and reasoning.
Certain aspects of linguistic analysis lend themselves to quantitative
methods. Others are better approached qualitatively.

Local coinages are not the same as adaptations or borrowings, though
local coinages often depend on the effects of external elements.

A label on some heart medicine I decided to sample because it was on
sale reads 有機ぶどうのポリフェノールたっぷリッチ. To confirm that my
reading of the label was not a side effect of the medicine, I re-read it
the next morning, and it said the same thing. It is Japanese. Pure
Japanese. As purely Japanese as it could be. And yet it is a microcosm
of multilingual influences on Japanese from antiquity to the present.

Perhaps this is what Michael meant when he wrote in his first posting on
this thread that "Japanese is currently in a state of rapid destruction.

It (whatever it may at other times have been) is being replaced by
something else which is not derived from any one source but which will
not exactly be an isolate either."

Sasha wrote "I think that there was a Japanese-Ainu bilingualism, at

least in the Kantoo area, but these were apparently not the guys who
migrated to Hokkaidoo -- they were assimilated to Japanese prior to
migration."

I think Sasha is right. One of the descendants of a remnant Ainu guy, or
maybe a girl, was destined to be a prime minister. In 1986, Nakasone
Yasuharu, from Gunma prefecture, got into trouble for believing Umehara
Takeshi, his advisor on Yamato spiritualism and Ainu Jomonism, who
suggested that his, Nakasone's, furry brows might be a vestige of the
Ainu who had peopled the area around Gunma until their displacement by
Yamato migrants.

For what its worth.

Bill Wetherall

[Ross Bender]

Wetherall says " I'll ride the bus and leave the arguments to the linguists who control the data."

Personally I find myself too such a bus rider, but as I have argued with various historical linguists of my acquaintance, if they wish their findings to be understood outside of their narrow circle they will need to express themselves more in layperson's terms.

One reason that Roy Andrew Miller was so popular, whatever the merits of his methodology, was that he was able to communicate his findings to those of the meanest intelligence (and I refer to myself).

Let me put in another plug for Academia.edu where discussions of historical linguistics in general and Japanese in particular are being posted daily. The latest I have found is:

  https://www.academia.edu/3644600/How_to_prove_genetic_relationships_among_languages_the_cases_of_Japanese_and_Korean

by Claudia Ciancaglini. This appears in English but in an Italian journal that would be very far off the radar for most PMJS members.

Ross Bender

https://www.academia.edu/

" Now, you're either on the bus or off the bus. If you're on the bus, and you get left behind, then you'll find it again. If you're off the bus in the first place — then it won't make a damn.

• As quoted by Tom Wolfe in The Electric Kool-Aid Acid Test (1968)

[Charles De Wolf]

As a linguist who first learned Japanese by way of Korean, I have long been interested in the genetic question, though I confess that I have also paid inadequate attention to it in recent years—and to the field in general, in part out of growing skepticism about the “scientific” nature of linguistics, in part out of an old love for literature and the art (!) of translation, which allows for multiple answers to knotty questions.

 

Some years ago a famous Japanese linguist in the 国語学 tradition reacted to criticisms of his pet theory concerning the origins of the Japanese language by attacking those questioning his analysis as having swallowed whole the European notion of the comparative method. I remember reacting cynically and, indeed, angrily at the attempt to use Nihonjinronistic arguments to squelch dissent. Yet the fact remains that all methodologies have their limitations.

 

I am fond of telling my students that though languages are often compared to living organisms, the analogy ultimately breaks down, as languages lack DNA. As in The Wizard of Oz, they can wind up almost like the Tin Woodman, except that the Tin Woodman has memories, even if he lacks fingerprints.

 

The late Roy Andrew Miller, no gentle soul, he, when it came to polemics, conceded that Japanese might indeed be a Mischsprache (a rare entity in the narrow definition of that term) but that once that hypothesis is granted, there is little room for further discussion along the lines of traditional linguistics. He seemed to be implying that it is still better, even for those still sober, to look for the key under the lamppost. And one must admit that such is not an unreasonable suggestion.

 

English, for all its promiscuous borrowings from French, Latin, and Greek, is still a Germanic language in overall morphological form. A word such as maintain is conjugated as a weak verb: maintain, maintained, maintained. English speakers do not usually have any knowledge of maintinssions as the first-person-plural imperfective subjunctive of the French verb maintenir.

 

When I first learned Japanese, I could lean a bit on Korean for morphosyntactic clues. For example, denwa wo kaketa vs. denwa ga kakatte kita corresponded conveniently to ch’eonhwa rul keoryeossumnida vs. ch’eonhwa ga keollyeossumnida. But eventually I came to see that in strictly morphological terms the two languages were only deceptively similar. Korean, for all its derivational irregularities, is significantly more “agglutinative” than is Japanese.

 

The principal discussants here are vastly more knowledgeable than I am in these matters, but I still obstinately maintain, so to speak, that unless one resorts to Procrustean reconstruction, there is no way to explain the Japanese verb system except as involving a simple sort of inflection, with vowels varying in a way remarkably similar to Indo-European Ablaut, as seen, for example, in sing-sang-sung.

 

How do we know that such a morphological pattern did not exist in the Jōmon language(s) and that it was simply perpetuated in the Mischsprache that became Japanese?

 

I might add that I am also something of a one-time Austronesianist, having puttered about with the verb morphologies of Palauan, Chamorro, and Tagalog. Proto-Austronesian was a radically different sort of language from Japanese, though there may have been a few lexical borrowings. Besides, Japan had human inhabitants many millennia before the seafaring Austronesians set forth.

 

Finally, Nihonjinron is, I hope, almost as dead as the term groovy, which ought to mean that one can talk about Japanese being relatively (!) isolated without proposing obligatory sun-goddess worship or evoking that once dread word unique.

Charles De Wolf

---

From: "Unger, James" <unge...@osu.edu>
To: "pm...@googlegroups.com" <pm...@googlegroups.com>
Sent: Friday, April 17, 2015 10:37 PM
Subject: RE: [PMJS] Once Again -- Is Japanese an Isolate? Or Genetically Related to Something or Other? Or Not?

--
PMJS is a scholarly forum.

You are subscribed to PMJS: Premodern Japanese Studies.
To post to the list, send email to pm...@googlegroups.com

To unsubscribe, send email to pmjs+unsub...@googlegroups.com

[Alexander Vovin]

Dear all,

William Wetherall wrote:

>Of interest to me is how Sasha then qualifies the meaning of "isolate". He argues in effect that all constellations of languages -- whether as small as Japano-Ryukyuan or as >large as Indo-European -- are equally "isolates" relative to one another -- "size does not matter" he says.

>But if size is irrelevant, then why regard even a single language, which defies grouping on phonological grounds, an isolate? Is a single star somehow more "isolated" from a >nearby double-star, or from a cluster of stars or a galaxy, than such larger groupings of stars are from each other? A "lone star" perhaps, but not isolated.

I am sorry, but this is exactly the opposite to what I was actually saying. Neither Japonic nor Indo-European are isolates. One is a small family, another is a big family. Sometimes the term 'isolate' is applied -- misleadingly -- to small families. 'Isolate' is a single language with no demonstrable genetic relation to other language or families. But if we start calling small families 'isolates', then where is the cut-off point: 2, 3, 5, 7 or more? My example with Indo-European is an example ad absurdum, but it serves two purposes: to demonstrate that 1) both Japonic and Indo-European were compared to other language families and without success and that 2) if we blur the difference between the isolates and small families, and if we further claim that the only criterion for being classified as non-isolate is genetic relationship with other language families (as Jim suggests), then we find that Japonic and Indo-European are in the same boat -- with no demonstrable genetic relationship to anything else, and this is very different, from let's say Semitic, which was shown to be related to other Afroasiatic families.
 

A few words about mixed languages. The idea was very popular in 1980s and 1990s, but then it somewhat faded away. The reason is obvious: there are only two-clear-cut cases of mixed languages: Michif and Copper Island Aleut. Both, interestingly enough, are found in what is geographically New World, and even more interesting both are confined to the areas of European colinial expansion: French and Russian respectively. The third famous cases, that of Ma'a, has been disproven since then.

Of course, Japanese, like any human language does not exist in vacuum, so it does contain layer upon layer of foreign loans. But here are two cardinal problems: 1) while the distinction between two mixed components in bothy Michif and Copper Island Aleut is very clear, where is the boundary in Japonic? 2) And what are the elements that are mixed? If one of them is Austronesian and another Altaic, then where are the material cognates? The case of Japanese is pretty much like that of English, with layers of Celtic, Old Norse, and Romance, but the core is unmistakenly Germanic. But unlike English, we cannot pinpoint the external connection of the Japanese or Japonic core.

All the best,

Sasha

Alexander Vovin
Directeur d'études, linguistique historique du Japon et de l'Asie du Nord-Est

ECOLE DES HAUTES ETUDES EN SCIENCES SOCIALES

CENTRE DE RECHERCHES LINGUISTIQUES SUR L'ASIE ORIENTALE

131, bd Saint-Michel, 75005 Paris

Adjunct Professor of Linguistics

Department of Linguistics

University of Hawai'i at Mānoa, Honolulu
HI 96822, USA

sasha...@gmail.com

[Ross Bender]

Greetings to all from a fair and flowery Philadelphia where we are experiencing genial spring weather which may only be described as "groovy."

Claudia Ciancaglini's 2008 article  https://www.academia.edu/3644600/How_to_prove_genetic_relationships_among_languages_the_cases_of_Japanese_and_Korean provides a most lucid primer on the history of these attempts. She begins by explaining the comparative-reconstructive method that allowed the reconstruction of Proto-Indo-European. She explains the meaning of "genetic" in linguistic terms, discusses isolated languages, glottochronology, the lexicostatistic method, the concept of mixed languages (including Copper Island Aleut mentioned by Vovin above), why the older use of "substratum" and "superstratum" is out-dated, long-term comparativists, the "Sprachbund", areal families and many other basic matters in a clear and concise fashion. As far as I can see, she does not address the issue of Ryukuan.

Her list of references is particularly valuable, containing among many others R.A.Miller, Murayama, Shibatani, Starostin, Trask and Vovin. (There is one item I would love to get my hands on -- Dybko, A. and Starostin, G. S. (2008) "In Defense of the Comparative Method, or the end of the Vovin Controversy." In Aspekty komparativiski 3, Moskva, 119-258.)

Here are her conclusions (pp 317-318):

“As we have seen, the principal genetic theories advanced about Korean and Japanese, among which we especially have the theories connecting Japanese to Korean or both to the Altaic family, are to be dismissed as not scientifically proved. No scholar, not even Starostin, has been able to prove this theory in a scientific way. The conclusion that Japanese and Korean are isolated languages from the genetic point of view and that all the attempts of scholars to demonstrate a genetic link between Japanese and Korean, or between these two languages and the Altaic group, have been unsuccessful does not depend (only) on the inability of scholars to correctly apply the comparative method. In fact, the failure rather depends on many historical and structural reasons, many of which we have already mentioned above and that we may summarize as follows:

•  Korean and Japanese are attested too late and in a logographic writing system;

• The presumed date of their split from proto-Altaic is too old to allow reconstruction;

• The impossibility to prove cognacy in lists of compared lexemes casts doubt on any phonological  correspondence deduced from these cognate sets; this impossibility may be due, at least in part, to the lexical typology of both Japanese and Korean;

• The impossibility of finding paradigmatic correspondences in closed sets of grammatical morphemes casts doubt on any genetic hypothesis concerning Japanese and Korean.

“Under these conditions, the comparative-reconstructive method cannot be applied; as a consequence, from the genetic point of view, both Japanese and Korean cannot be considered anything but isolated languages.”

Ross Bender

https://independent.academia.edu/RossBender

" Now, you're either on the bus or off the bus. If you're on the bus, and you get left behind, then you'll find it again. If you're off the bus in the first place — then it won't make a damn."

• Ken Kesey, as quoted by Tom Wolfe in The Electric Kool-Aid Acid Test (1968)

[Unger, James]

Q: "Historical linguistics should, IMHO, be the study of language systems

over time -- how not only sounds, but meanings and the structures that
convey meanings, are conserved or change under various conditions, from
geographic isolation to intense intercourse with speakers of neighboring
or even non-regional languages. Why should the "relationship" between
two languages hinge on only their phonological relationship?"

A: The comparative method begins with lexical comparisons and seeks to find a body of etymologically related words in which phonemes are related by a natural, coherent set of regular phoneme correspondences because the lexicon is the part of the language in which the arbitrary relationships of form to function (meaning) are most highly concentrated. Therefore, the more etymologies that can be identified, the less likely the odds that the regular phoneme correspondences upon which they depend are due to chance or borrowing; hence, by process of elimination, one is forced to accept the hypothesis of common origin. Finding lexical matches is generally harder than finding resemblances in morphology or syntax, in which the relationships of forms to functions are generally less arbitrary. This is why unrelated languages may show gross typological similarities, and why such typology is not a good starting point for comparisons. But if it can be shown that two languages are related lexically, we expect to find some systematic correspondences in morphology and syntax, and evidence of changes in one language or the other when such correspondences are absent.

[Michael Pye]

Dear Sasha, and Fellows of pmjs,

This discussion is absolutely fascinating. I want to become an
undergraduate again and enroll in a new course. When I did "mod and
med languages" the theoretical side was left to our own devising. And
now I am so torn between the analytical aspect, and the historical
aspect brought in by Bill Weatherall and Ross Bender. Some of the
lines of thought remind me quite a lot of theoretical/methodological
debates in the comparative and historical study of religions, by the
way. But now I have three questions.

Sasha, you write:

"The case of Japanese is pretty much like that of English, with layers
of Celtic, Old Norse, and Romance, but the core is unmistakenly
Germanic. But unlike English, we cannot pinpoint the external
connection of the Japanese or Japonic core."

Quite so, if so put. But, now I'm as bothered about the use of the
word "core" as I was about "isolate" (which from the various
quotations doesn't seem to have been used quite consistently by all
specialists; but I accept your clear definitiion, for which thanks).
But "core" reminds me of the word "essence" which has been used and
misused in the study of religions. OK I know that "religions" are just
great big wishy-washy socio-cultural things compared with languages
(or at least I know that some might think that). But my question is
(remember I'm a first-term student in linguistics) how do you
delineate the "core" of a language? Just SVO versus SOV obviously
isn't enough. Is is something to do with how we count up to five? Or
perhaps up to four? (Some linguists I have come across are incredibly
keen on that). Or the number of tenses, e.g. just two? Or a weighted
poll of vocabulary excluding known loan words? So what is the "core"
of "a" language? Is there really any such thing? (I'm toying with the
idea of extracting a metaphor for talking about the identity of
specific religions, or their lack of it, but you don't need to go down
that road.)

Second, what is the advantage of the word "Japonic" rather than
"Japanese"? Are you trying to refer to an ancient, (dare I say)
essential core of early Japanese as historically known? Would there be
a terminus ad quem for that? Obviously (I understand) that any search
for ancient affinities doesn't just start with Japanese of any period,
but with some kind of archaic form of the language. But how archaic?
And how does one approach the identification of "a core"? Could it be
that identifiable "Japonic" is such a small body of known linguistic
materials that I could learn it in a couple of weeks? And why do you
call it Japonic rather than Japanic? Is it something to do with Paris?
Or is there a botanical analogy lurking behind this choice of term. (I
do quite like the word "Japonic"...not that liking it is theoretically
elevant.)

Third, what is the standing, or your view, of Jens Rickmeyer's work on
morphosyntax in Japanese? Does it have any relevance to these matters,
or is it only relevant to modern Japanese?

Sorry to ask for instruction when you are doubtless busy enough.
best wishes
Michael Pye
PS My wife and I have just been reading the instructions for a new
washing machine and are wondering what language it is written in (cf.
Bill Weatherall's medical example). Perhaps we can call it "new
Kata-Japanic" ....

Zitat von Alexander Vovin <sasha...@gmail.com>:

> Dear all,
>
> William Wetherall wrote:
>
>> Of interest to me is how Sasha then qualifies the meaning of "isolate". He
> argues in effect that all constellations of languages -- whether as small
> as Japano-Ryukyuan or as >large as Indo-European -- are equally "isolates"
> relative to one another -- "size does not matter" he says.
>
>> But if size is irrelevant, then why regard even a single language, which
> defies grouping on phonological grounds, an isolate? Is a single star
> somehow more "isolated" from a >nearby double-star, or from a cluster of
> stars or a galaxy, than such larger groupings of stars are from each other?
> A "lone star" perhaps, but not isolated.
>
> I am sorry, but this is exactly the opposite to what I was actually saying.
> Neither Japonic nor Indo-European are isolates. One is a small family,
> another is a big family. Sometimes the term 'isolate' is applied --

> misleadingly -- to small families. 'Isolate' is a *single* language with no
> *demonstrable* genetic relation to other language or families. But if we

> start calling small families 'isolates', then where is the cut-off point:

> 2, 3, 5, 7 or more? My example with Indo-European is an example *ad
> absurdum*, but it serves two purposes: to demonstrate that 1) both Japonic

> sasha...@gmail.com <alexand...@ehess.fr>

[Unger, James]

The reason for the term proto-Japonic is that those who wish to take one or more speech varieties previously called dialects (方言) of Japanese as separate languages can't recycle "Japanese," since it already is a language name (the standard translation of 日本語).

The reason that proto-Ryukyuan is probably a second- or third-order daughter of proto-Japonic rather than a first-order daughter is that the pan-Ryukyuan innovations that, for instance, Pellard (2015, available at academia.edu) cites as evidence for a historically coherent proto-Ryukyuan do not, in themselves, establish the position of proto-Ryukyuan in the Japonic tree (Sprachbaum), even though its probable date of separation may have been as early as the Yayoi-Kofun transition. (As Pellard explains, non-linguistic evidence suggests that proto-Ryukyuan, i.e. the first kind of Japonic spoken by permanent settlers in the islands, probably arose from a Kyushu dialect whose speakers migrated in the 10th c. CE.) On the other hand, the historical reconstruction of Japanese pitch accent systems in by Elisabeth de Boer (2010, Harrassowitz) convincingly shows that the pRy accent system is not a direct development from the pJ system, but involves at least one intervening stage of development (probably two, perhaps more).

The significance of the Hayato and other non-Japonic speakers mentioned in early texts is not, in my opinion, so much the one or two words we may be able to identify of their language(s) as the broad implication that they successfully resisted the spread of Japonic in Kyushu for several centuries. Japonic probably did not abruptly wash over Kyushu in a single wave spreading outward from the original Yayoi settlement area, but more likely became dominant in small peripheral pockets (perhaps geographically disjoint) at different times. If so, we would expect there to have been many opportunities for dialect differentiation in Kyushu before pre-proto-Ryukyuan speakers established proto-Ryukyuan somewhere in its present range.

As for pKJ etymologies, as I mention in my NINJAL article, a new set of etymologies and sound correspondences will be forthcoming in the next year or so.

Jim Unger

[Alexander Vovin]

Dear Michael and all,

Please forgive me for responding below by citation method this time -- I think it will be clearer this way.

On Sun, Apr 19, 2015 at 10:32 AM, Michael Pye <p...@staff.uni-marburg.de> wrote:

Dear Sasha, and Fellows of pmjs,

This discussion is absolutely fascinating. I want to become an undergraduate again and enroll in a new course. When I did "mod and med languages" the theoretical side was left to our own devising. And now I am so torn between the analytical aspect, and the historical aspect brought in by Bill Weatherall and Ross Bender. Some of the lines of thought remind me quite a lot of theoretical/methodological debates in the comparative and historical study of religions, by the way. But now I have three questions.

Sasha, you write:

"The case of Japanese is pretty much like that of English, with layers of Celtic, Old Norse, and Romance, but the core is unmistakenly Germanic. But unlike English, we cannot pinpoint the external connection of the Japanese or Japonic core."
Quite so, if so put. But, now I'm as bothered about the use of the word "core" as I was about "isolate" (which from the various quotations doesn't seem to have been used quite consistently by all specialists; but I accept your clear definitiion, for which thanks).

It is not mine. I believe it was explicitly introduced by Lyle Campbell. I just follow him here.

 

But "core" reminds me of the word "essence" which has been used and misused in the study of religions. OK I know that "religions" are just great big wishy-washy socio-cultural things compared with languages (or at least I know that some might think that). But my question is (remember I'm a first-term student in linguistics) how do you delineate the "core" of a language? Just SVO versus SOV obviously isn't enough. Is is something to do with how we count up to five? Or perhaps up to four? (Some linguists I have come across are incredibly keen on that). Or the number of tenses, e.g. just two? Or a weighted poll of vocabulary excluding known loan words? So what is the "core" of "a" language? Is there really any such thing? (I'm toying with the idea of extracting a metaphor for talking about the identity of specific religions, or their lack of it, but you don't need to go down that road.)

Well, the term is not good -- I was trying to respond to Ross's plea to speak in more familiar terms, and apparently ended up with more confusion.. What I meant here is basic paradigmatic morphology and basic vocabulary, i.e. body parts, nature phenomena, generic names of animals and plants, basic verbs and adjectives, etc. In demonstrable language families we normally have huge percentage of cognates for this kind of words (let me add that this is not, imho, the strongest argument for the genetic relationship, but for a brief dfemonstration it is quite graphic. So let us have a look at three very stable words for body parts in Old English and Church Slavonic (I avoid on purpose taking examples from Sanskrit, Greek, or Latin in order not to be accused that these are considerably older than Old Japanese and Middle Korean):

                                      OE             ChSl        Indo-European

nose                               nosu           nosŭ        *nās-

eye                                 ēage           oko         *okw-

ear                                  ēare           uxo          *aus-

Not let us look how Old Japanese and Middle Korean would fare:

                                      OJ               MK

nose                               pana            koh

eye                                 me2             nun

ear                                  mi1mi1        kuy

It is quite apparent that we don't even have to bother to try to establish regular phonetic correspondences here, because none are possible. I haste to add that not all cognates are transparent and look similar -- this would be a great mistake. Some are very well may be in hiding. I always give my students an example of German was and Russian chto -- these are cognates, no matter how dissimilar they look. If we look again at the OE and ChSl, the word for 'nose' is very transparent, but the word for 'ear' is not. But knowing the phonological histories of both Germanic and Slavic, it becomes quite apparent that both are cognates as well. Not so in the case of OJ and MK: our knowledge of their histories throws no light at the the possibility of establishing any regularity of correspondences here.

Second, what is the advantage of the word "Japonic" rather than "Japanese"? Are you trying to refer to an ancient, (dare I say) essential core of early Japanese as historically known? Would there be a terminus ad quem for that? Obviously (I understand) that any search for ancient affinities doesn't just start with Japanese of any period, but with some kind of archaic form of the language. But how archaic? And how does one approach the identification of "a core"? Could it be that identifiable "Japonic" is such a small body of known linguistic materials that I could learn it in a couple of weeks? And why do you call it Japonic rather than Japanic? Is it something to do with Paris? Or is there a botanical analogy lurking behind this choice of term. (I do quite like the word "Japonic"...not that liking it is theoretically elevant.)

The term 'Japonic was introduced by Leon Serafim in early 1990s as a 'hat' designation of both Japanese and Ryukyuan. Its obvious advantage is that it helps to avoid terminological confusion of the name for family (Japonic) and the name of one of its branches (Japanese), which will necessarily arise if we use the same term 'Japanese' for both.

Third, what is the standing, or your view, of Jens Rickmeyer's work on morphosyntax in Japanese? Does it have any relevance to these matters, or is it only relevant to modern Japanese?

I think that Jens Rickmeyer worked successfully on both Classical and modern Japanese, but to the best of my knowledge he was not involved in the problems that we discuss here. He has also prepared two brilliant younger German scholars: Sven Osterkamp and Heiko Narrog, the first of whom to the certain extent is involved in this field.

Sorry to ask for instruction when you are doubtless busy enough.
best wishes
Michael Pye
PS My wife and I have just been reading the instructions for a new washing machine and are wondering what language it is written in (cf. Bill Weatherall's medical example). Perhaps we can call it "new Kata-Japanic" ....

Yes, this is very funny (as well as Bill Weatherall's example). Let me add to the pot in conjunction with the basic vocabulary. Modern Japanese borrowed several body part terms from English: ヒップ、ウェスト、バスト and ネール. But the first three are peripheral basic vocabulary, and the fourth one has never replaced native tsume.

All the best,

Sasha

[Alexander Vovin]

Dear Jim and all,

To the best of my knowledge (please correct me if I am wrong), so far Jim did not present any linguistic evidence in favor of his point of view that Ryukyuan is not a first-order daughter of Japonic, we are just told that Pellard is mistaken. I, for one, will be immensely obliged if Jim could kindly demonstrate to the list with the discussion of specific data why he believes that Ryukyuan and Japanese mutually exclusive innovations that Pellard 2015 presents are inadmissible? Incidentally, in his 2015 article Pellard provides just a small fraction of relevant information, and he very clearly states that this is not the place to discuss it at length. Much more can be found in his 2009 PhD dissertation: Ōgami: Éléments de description d’un parler du Sud des Ryūkyū. Ph.D. thesis, Paris: École des hautes études en sciences sociales. And there is much more, but let's discuss Pellard's data first.

We have to see the new set of etymologies and sound correspondences promised by Jim before passing the judgment whether they are valid or not (:-).

On a different topic: the article by Claudia Ciancaglini that Ross Bender sent a link to is an excellent one. It demonstrates very well why a scholar thoroughly trained in Indo-European views negatively the Koreo-Japonic. I have only one essential critical comment on it: contrary to what she asserts, it is not necessary to have ancient languages to demonstrate a genetic relationship. Uralic, Austronesian, Na-Dene, Algonquian, Uto-Aztecan, Kadai, etc. were all demonstrated to the satisfaction of general historical linguists without or almost without any reliance on pre-modern languages.

All the best,

Sasha

Alexander Vovin
Directeur d'études, linguistique historique du Japon et de l'Asie du Nord-Est

ECOLE DES HAUTES ETUDES EN SCIENCES SOCIALES

CENTRE DE RECHERCHES LINGUISTIQUES SUR L'ASIE ORIENTALE

131, bd Saint-Michel, 75005 Paris

Adjunct Professor of Linguistics

Department of Linguistics

University of Hawai'i at Mānoa, Honolulu
HI 96822, USA

sasha...@gmail.com

Jim Unger

[Ross Bender]

The paper by Thomas Pellard to which Unger refers above is a chapter in the 2015 Handbook of the Ryukyuan Languages,ed. Heinrich, Miyara and Shimoji, Berlin: Mouton de Gruyter

https://www.academia.edu/11943222/The_linguistic_archeology_of_the_Ryukyu_Islands

Below is my summary:

“The history of the Ryukyuan languages is a fundamental issue that still remains by and large unsolved. The study of the prehistory of human populations is not the preserve of archaeology, but is located at the interface with other disciplines such as anthropology and linguistics…” (p.13)

“Though it is generally assumed that the Japonic family has two main branches (Katō 1977), a Japanese one and a Ryukyuan one, some scholars have challenged the very idea that the Ryukyuan languages form a valid subgroup (Unger 2009:94-106)." [The Role of Contact in the Origin of the Japanese and Korean Languages, U. Hawai’i Press] (p.15)

Pellard’s tree diagram (p.14) shows Japonic with three sub-branches: Ryukyuan, Japanese, and possibly Hachijō. Ryukyuan has Southern and Northern branches. Southern has Macro-Yaeyama and Miyako, with Yonaguni and Yaeyama as sub-branches; Northern has sub-branches of Okinawan and Amami. Thus in this scheme Ryukyuan and Japanese are sisters, daughters of Japonic.

Section 5 of the chapter is “Toward a New Synthesis.” It reviews archaeological and anthropological evidence and proposes “A Unified Synthesis.” (Section 5:3, pp. 26-28) Here Pellard reviews the “Ocean-Road,” the “Hayato,” and the “Proto-Gusuku” scenarios. The first suggests “an early settlement of the Ryukyus by Japonic agriculturalists before the Yayoi period.” The second scenario “associates the spread of Japonic in the Ryukyus with a migration before the 8^th century CE of the Hayato ‘barbarians’ of Kyushu depicted in the Old Japanese chronicles." Lastly, the “Proto-Gusuku hypothesis (Asato & Doi 1999, Takamiya 2005) convincingly argues that the only event that can meaningfully be associated with a Japonic expansion in the Ryukyus is the migration around the 10^th century that led to the formation of the Gusuku culture.”

“Refining the Proto-Gusuku hypothesis, we can say that the Ryukyuan languages form a sister branch of Japanese and that their ancestor separated from Japanese probably during the first centuries CE. This is likely to have happened at the end of the Yayoi period, just before the rise of the Kofun culture and the emergence of strong polities that led to the formation of the Yamato state in Central Japan, the homeland of Old Japanese….then, around the 10/12^th century…a group of merchants from Kyushu, accompanied by craftsmen and farmers who settled there, entered the Ryukyus…”

In the Conclusion, Pellard notes that although the main line of the proposed Proto-Gusuku hypothesis is clear, “many details remain to be settled.” “For the time being, the scenario of the expansion [of] Japonic in the Ryukyus presented here is no more than a plausible narrative, though a rather convincing one.” (p. 28)

Ross Bender

https://independent.academia.edu/RossBender

 

[Alexander Vovin]

Dear Ross and all,

Just a brief note. It is highly unlikely that Hachijoo presents a third first order branch in Insular Japonic. It is probably the only surviving descendant of Eastern Old Japanese, and more specifically as my former student John Kupchik once proposed, of a EOJ dialects of Suruga (possibly also of Kamitupusa and Simotupusa, I should add).

All the best,

Sasha

Alexander Vovin
Directeur d'études, linguistique historique du Japon et de l'Asie du Nord-Est

ECOLE DES HAUTES ETUDES EN SCIENCES SOCIALES

CENTRE DE RECHERCHES LINGUISTIQUES SUR L'ASIE ORIENTALE

131, bd Saint-Michel, 75005 Paris

Adjunct Professor of Linguistics

Department of Linguistics

University of Hawai'i at Mānoa, Honolulu
HI 96822, USA

sasha...@gmail.com

[Smits, I.B.]

Dear all,

I've been following this thread with interest, and I would like to draw your attention to a research project by our new Leiden colleague, Martine Robbeets, who has just landed a ERC grant to deal with exactly this hotly debated question.  See:

http://www.hum.leiden.edu/news-agenda/using-erc-grant-with-beans-and-millet.html

Best wishes,

       Ivo Smits

**********************************************************

Ivo Smits

Professor of Arts and Cultures of Japan

Leiden University Institute for Area Studies (LIAS)

Leiden University

P.O. Box 9515

2300 RA  Leiden

The Netherlands

Tel   +31-71-527 2545 (direct)/ 2539 (secr.)

Fax  +31-71-527 2215

E-mail:  i.b....@hum.leidenuniv.nl (日本語もどうぞ)

http://www.hum.leiden.edu/lias/organisation/japanese/smitsib.html

http://www.hum.leidenuniv.nl/japans/

**********************************************************

---

From: pm...@googlegroups.com [pm...@googlegroups.com] on behalf of Alexander Vovin [sasha...@gmail.com]
Sent: Monday, April 20, 2015 12:33 AM
To: pm...@googlegroups.com
Subject: Re: [PMJS] Once Again -- Is Japanese an Isolate? Or Genetically Related to Something or Other? Or Not?

[Michael Pye]

Dear all,
That's a very interesting project being set up in Leiden. The thing
won't lie down, will it. Indeed, the concept of triangulation of
methods, or the triangulation method, seems interesting and evidently
provides opportunity for travel into difficult landscapes.
But, what about that map showing where "Inner" Mongolia is, which
shows China plus several named countries of SE Asia, but has
completely wiped out Korea and Japan! A bit more triangulation is need
in that regard.
Michael Pye
PS Thanks Sasha, for your patient teaching. I appreciate the
accessible clarifications.

Zitat von "Smits, I.B." <I.B....@hum.leidenuniv.nl>:

> sasha...@gmail.com<mailto:alexand...@ehess.fr>

> sasha...@gmail.com<mailto:alexand...@ehess.fr>

>
> On Sun, Apr 19, 2015 at 4:12 PM, Unger, James
> <unge...@osu.edu<mailto:unge...@osu.edu>> wrote:
> The reason for the term proto-Japonic is that those who wish to take
> one or more speech varieties previously called dialects (方言) of
> Japanese as separate languages can't recycle "Japanese," since it
> already is a language name (the standard translation of 日本語).
>
> The reason that proto-Ryukyuan is probably a second- or third-order
> daughter of proto-Japonic rather than a first-order daughter is that
> the pan-Ryukyuan innovations that, for instance, Pellard (2015,

> available at academia.edu<http://academia.edu>) cites as evidence

> pm...@googlegroups.com<mailto:pm...@googlegroups.com>

> To unsubscribe, send email to

> pmjs+uns...@googlegroups.com<mailto:pmjs%2Bunsu...@googlegroups.com>
> Visit the PMJS web site at www.pmjs.org<http://www.pmjs.org>
> Contact the group administrator at edi...@pmjs.org<mailto:edi...@pmjs.org>

>
>
> --
> PMJS is a scholarly forum.
>
> You are subscribed to PMJS: Premodern Japanese Studies.
> To post to the list, send email to

> pm...@googlegroups.com<mailto:pm...@googlegroups.com>

> To unsubscribe, send email to

> pmjs+uns...@googlegroups.com<mailto:pmjs%2Bunsu...@googlegroups.com>
> Visit the PMJS web site at www.pmjs.org<http://www.pmjs.org>
> Contact the group administrator at edi...@pmjs.org<mailto:edi...@pmjs.org>

>
>
> --
> PMJS is a scholarly forum.
>
> You are subscribed to PMJS: Premodern Japanese Studies.
> To post to the list, send email to

> pm...@googlegroups.com<mailto:pm...@googlegroups.com>

> To unsubscribe, send email to

> pmjs+uns...@googlegroups.com<mailto:pmjs%2Bunsu...@googlegroups.com>
> Visit the PMJS web site at www.pmjs.org<http://www.pmjs.org>
> Contact the group administrator at edi...@pmjs.org<mailto:edi...@pmjs.org>

>
>
> --
> PMJS is a scholarly forum.
>
> You are subscribed to PMJS: Premodern Japanese Studies.
> To post to the list, send email to pm...@googlegroups.com
> To unsubscribe, send email to pmjs+uns...@googlegroups.com
> Visit the PMJS web site at www.pmjs.org
> Contact the group administrator at edi...@pmjs.org
>
> --
> PMJS is a scholarly forum.
>
> You are subscribed to PMJS: Premodern Japanese Studies.
> To post to the list, send email to pm...@googlegroups.com
> To unsubscribe, send email to pmjs+uns...@googlegroups.com
> Visit the PMJS web site at www.pmjs.org
> Contact the group administrator at edi...@pmjs.org

[Ivan Rumánek]

Dear Sasha,

   talking of Martian :-) , could you please remind me of the two Hayato words preserved to us? I am sure I must have gone through this data in your workds but somehow it has not stuck in my mind...

Nice spring days!

Ivan Rumánek

[Alexander Vovin]

Dear Ivan and all,

Two Hayato words survived in the Oosumi Fudoki in the Fudoki fragments (風土記逸聞): kusira (久西良) 'comb' and pi1si (必志)  'off-shore sand-bank'.

They occur in the following contexts:

髪梳者隼人俗語云久西良

海中洲者隼人俗語云必志

See Fudoki in NKBT edition, p. 526.

These two words were discussed at length by Murayama Shichiroo  in the volume Hayato from the Nihon kodai bunka tankyuu edited by Oobayashi Taryoo (Shakai shisoosha, pp. 251-55). Murayama compared Hayato pi1si with Malay bətiŋ 'id.', but the problem is that this word is isolated in Malay, and we have certainly no evidence for direct contacts of Malay with Kyuushuu. The parallels in both Ryukyuan and Ainu might be tempting, but drawing conclusions on the basis of one word is not methodologically sound.  Kusira 'comb' may be just Japonic kusi 'id.' + diminutive -ra, but the problem is that OJ diminutive -ra has a very ,imited distribution: it is used exclusively with terms of kinship: imo1 'younger sister', se 'elder brother', etc. Hope this helps,

All the best,

Sasha

Alexander Vovin
Directeur d'études, linguistique historique du Japon et de l'Asie du Nord-Est

ECOLE DES HAUTES ETUDES EN SCIENCES SOCIALES

CENTRE DE RECHERCHES LINGUISTIQUES SUR L'ASIE ORIENTALE

131, bd Saint-Michel, 75005 Paris

Adjunct Professor of Linguistics

Department of Linguistics

University of Hawai'i at Mānoa, Honolulu
HI 96822, USA

sasha...@gmail.com

[Ross Bender]

Martine Robbeets' article "The Historical Comparison of Japanese, Korean and the Trans-Eurasion Languages" precedes Ciancaglini's in the same volume. As of this moment she has not uploaded a copy to her Academia page.

She begins:

"The question whether Japanese, Korean and the Trans-Eurasian languages are genealogically related is among the most disputed issues of language history. Linguistic literature reflects a wide range of opinions on this matter, ranging from a negative stance to a neutral or agnostic attitude to a positive stance."

"As far as the affiliation of Japanese and Korean is concerned, the majority view today (Lewin 1976, Yi 1973, Sohn 1980, Whitman 1985, Unger 1990, and Martin 1991) holds that they are likely to be related to each other and that the historically adjacent Tungusic languages are the likeliest candidates for further relationship."

The paper is highly technical, but concludes:

"We are lead [sic] to the conclusion that it is more likely that the etymologies presented here are the result of genealogical retention than that they are induced by language contact. For all the borrowing that has undeniably taken place, the Trans-Eurasian languages are most probably also related in the genealogical sense." (p.282)

Robbeets provides a massive list of references, including all the greats. The list includes Unger's Yale dissertation, and also Vovin's 2005 article "The end of the Altaic controversy" which she critiques on p. 273.

The date of publication is 2008, so neither Unger's nor Vovin's 2009 books are included in the discussion.

Ross Bender

https://independent.academia.edu/RossBender

[Ross Bender]

For a welcome change of pace, I would highly recommend the article Wetherall cited. It is the most elegant and entertaining and possibly profound piece I've read all week, even though it shows an unholy deference to Darwin, Dawkins and Dennett.

Genetic Metaphor in Historical Linguistics

Roger Lass

Alternation 10.1 (2003) 47-62

From Section 2. Historical Linguistics as a Metaphor-hijacker

“Subject-envy is not uncharacteristic of relatively new disciplines, especially ones whose ontological substrate is ill-defined. Linguistics has tried at various times to make itself look like biology, physics, mathematics, cognitive science, or all of them. And when in the 1930s and 1940s it looked like becoming about the best thing of its kind around, less secure subjects like anthropology began to borrow some of its concepts, imagery and terminology. Linguistics is certainly the source of the fruitful notion of an ‘emic’ level of organization, or what we might call ‘structuralism’ in general. The spirit of Saussure brooded over the face of the waters for a long time, and social scientists like Lévi-Strauss and Piaget were certainly among those who felt the (cold or warm) breath of linguistics.

          “But why should we linguists keep doing it, since others are apparently so enchanted by us that they have borrowed our lexicon and imagery? The answer I suppose is that we are not (or not yet) an empirical or hard science; we seem to rest uneasily somewhere among the ‘human sciences’ or ‘Geisteswissenschaften’, if with a widespread yearning for a different status.” p.49

http://alternation.ukzn.ac.za/Files/docs/10.1/04%20Las.pdf

Also, I am currently debating whether to spring for a book with the intriguing title How to Sound Intelligent in Japanese by one Charles De Wolf. It's selling for about 14 bucks on Amazon. My quandary is deciding whether sounding intelligent in Japanese is a) a plausible hypothesis b) a lost cause c) a waste of time or d) a waste of money.

Any advice will be appreciated.

Ross Bender

https://independent.academia.edu/RossBender

[John Creamer]

Might linguists also allow readers of literature to participate in "hijacking" by interpreting or attempting to categorize "kushi" as a potential term of familial kinship and "ra" as a diminutive through the use of a "folk etymology"? Personally, I find the grammar of folktales, as suggested by Propp, for example, to be a helpful and productive way of analyzing tales like "Taketori monogatari", as shown by Okada and others. Remember the story of Susanoo killing the eight-headed snake that had been terrorizing a family and their beautiful daughter. (I may be remembering the story from Chusei Nihongi and not Nihon shoki) Susanoo turns the daughter into a comb (kushi) and sticks her into his hair. The kushi is not only a comb but signifies taking her as a wife and into his clan, which might be predicted by the word kushi's origins as being related to family. (Kushi can also mean sake and he puts out eight barrels of sake to lure the snake into his trap) This is not really an etymology as linguists define it, but it might still reveal a shared understanding of the origins of the word kushi.

Personally, I don't share the outrage about sharing metaphors. I guess I am a liberal.

[Ross Bender]

Thanks to all who have contributed to this thread and in particular to Jim Unger and Sasha Vovin for answering hard questions in simple language.

As is obvious, this discussion still has many miles to travel, and it will be interesting to follow Robbeets in particular with the massive ERC project.

In closing, I would just like to ask a brief favor of our linguists. In a new thread, I will ask the question "What is the difference between Old Japanese and Middle Japanese?" in 25 words more or less.

The link below is to a Youtube video -- Laurie Anderson's "Language is a Virus" from Home of the Brave. If you watch to the end, Ms. Anderson will contribute a few of her thoughts on Buddhism.

"Language is a virus from outer space" -- William S. Burroughs 

https://www.youtube.com/watch?v=8bhldlJ0Aus

Ross Bender

https://independent.academia.edu/RossBender

[Alexander Vovin]

Dear Michael and all,

 

Unfortunately, you are right -- this is not going to go away easily, although regarding ‘Altaic’, two strongholds of its supporters now remain: Moscow and Leiden. The rest of the world grew considerably more critical in the last thirty years or so.

 

Incorrect ideas may persist and persevere for a long time. You might remember that I mentioned earlier in this discussion that Ural-Altaic was disproven at the end of nineteenth century -- early twentieth century. Yesterday night I was reading a book on Vikings’ history and on one of the first pages I ran into the following: “the Finnish language, belonging to the Finno-Ugrian group of languages is closely associated with Estonian, spoken on the south coast of the Gulf of Finland, and is more distantly related to Hungarian and Turkish”. Everything up to Hungarian is roughly right, but then Turkish is neither Finno-Ougrian nor Uralic. It is ‘Altaic’. What a bummer ().

 

It is true that Martine Robbeets is pursuing this ‘Transeurasian’ (her own term for ‘Altaic’), but ‘Transeurasian’ is a misnomer, because very unlike Indo-European and Uralic that have branches in both Europe and Asia, all ‘Transeurasian’ language families have originated in Asia -- this goes without saying for Japonic, Koreanic, and Tungusic, but even those Turkic and especially Mongolic languages that are nowadays located in Eastern Europe resulted from very late migrations from Asia.

 

It should be also mentioned that while Martine Robbeets claims to investigate the question, she has, in all her publications since 2000 and then 2004 and 2005 and up to this date always declared the question settled in favor of ‘Altaic’, which resembles more a crusade than a purely scholarly endeavor. And she never responds to any criticism. So far all Martine Robbeets’s publications dealt exclusively with ‘Altaic’ (a.k.a. ‘Transeurasian’): she has not yet produced anything on Japanese linguistics or Korean linguistics. In order to form an opinion about her work and methods, it will be necessary, alongside her own writings, also to carefully inspect the reception these received and most of those range from extremely negative to mildly negative (the list below is far from exhaustive):

 

Reviews of Robbeets 2005 monograph

Georg, Stefan 2008. Review of Robbeets, Martine Irma (2005): Is Japanese related to Korean, Tungusic, Mongolic and Turkic? Wiesbaden: Harrassowitz. Bochumer Jahrbuch für Ostasienforschung. B. 32: 247-278.

Kara, György 2007. Review of Robbeets, Martine Irma (2005): Is Japanese related to Korean, Tungusic, Mongolic and Turkic? Anthropological Linguistics 49.1: 95-98. It is quite surprising that Martine Robbeets herself counts this as a "positive" review of her book and tells people that Kara is a "former" skeptic of the hypothesis, now converted. Read it yourself to see how wrong she is on this ().

Knüppel, Michael 2006. ‘Ein Beitrag zur Japanisch-Koreanisch-Altaischen Hypothese’. Wiener Zeitschrift für die Kunde des Morgenlandes 96: 353-64.

Miller, Roy A. -- it is in Ural-Altaische Jahrbücher (Harrassowitz) -- I do not have an exact reference, because the photocopy of this article did not survive my move from the USA to France. Roy and myself disagreed on everything in this field, but I am glad to follow the old Roman saying “about the dead, either good or nothing”, and say something good about him in this respect: it was the first time that I saw Roy turning completely against someone who was defending the ‘Altaic’ hypothesis in his review. And our reviews agreed quite independently on several issues, including the greatest of them -- plagiarism. Actually, Roy’s negative position to this is very important. A criticism from one’s own camp always counts more.

Rozycki, William -- it is in Mongolian Studies, and this is another one that did not survive my move. This is the only positive review I am aware of.

Vovin, Alexander 2009. ‘Japanese, Korean, and other ‘Non-Altaic’ Languages. Review article of Robbeets, Martine Irma (2005): Is Japanese related to Korean, Tungusic, Mongolic and Turkic? Central Asiatic Journal 53.1: 105-47 (available from my page at academia.edu: https://www.academia.edu/6081007/Review_article_Robbeets2005

 

Reviews of Martine Robbeets’s more recent work and general issues connected with it:

Georg, Stefan 2011. ‘The Poverty of Altaicism’. Presentation at the INALCO conference in Paris (available from Stefan Georg’s page at academia.edu) https://www.academia.edu/1638942/The_Poverty_of_Altaicism

Georg, Stefan 2014. Review of: Martine Robbeets/Hubert Cuyckens (eds.): Shared Grammaticalization. With special focus on the Transeurasian languages, Amsterdam 2013 (available from Stefan Georg’s page at academia.edu) https://www.academia.edu/4477900/Review_of_Martine_Robbeets_Hubert_Cuyckens_edd._Shared_Grammaticalization._With_special_focus_on_the_Transeurasian_languages_Amsterdam_2013

 

I do not mean it to be a degrading comment, but there is another thing that bothers me considerably in the work of not only Martine Robbeets, but other supporters of Japanese-Altaic as well. If they ever go beyond dictionaries, all their citations from premodern texts are drawn verbatim from the secondary literature. As far as Japanese goes, there is no evidence of an independent ability to handle any Old Japanese or Middle Japanese texts written in man’yoogana or hentaigana. And it also looks like that the actual knowledge of premodern languages is not required. Personally for me it makes no sense: I came to this field from Indo-European, and it was a common knowledge that you need to know how to read Greek, Latin, Sanskrit, Hittite, Old Persian (or Avestan), Church Slavonic, Gothic, Old Irish, and ideally also Classical Armenian and/or Tocharian before you can do anything at all. The ‘Altaic’ case is much easier: one has to master only Old Japanese, Middle Korean, Middle Mongolian, Manchu and Ewenki (for Tungusic), and Old Turkic to be on the same level as IEpeanists are. But for some reason, which completely escapes me, in the ‘Altaic’ field it is considered OK just to pull a dictionary or a grammar from the shelf (one shining exception in the Koreo-Japonic field, who is a very staunch supporter of the genetic relationship is John Whitman, who can read both OJ and MK texts).

All the best,

Sasha

Alexander Vovin
Directeur d'études, linguistique historique du Japon et de l'Asie du Nord-Est

ECOLE DES HAUTES ETUDES EN SCIENCES SOCIALES

CENTRE DE RECHERCHES LINGUISTIQUES SUR L'ASIE ORIENTALE

131, bd Saint-Michel, 75005 Paris

Adjunct Professor of Linguistics

Department of Linguistics

University of Hawai'i at Mānoa, Honolulu
HI 96822, USA

sasha...@gmail.com

[Ivan Rumánek]

Dear Sasha,

   thank you for the detailed explanation. Kushira indeed looks familiar, and it could  be inferred that it belonged to a dialect of Japanese where they just used -ra as a suffix, just as there still are remnants of such suffixes in Japanese dialects today (funakko for fune, hashikko for hashi as a cosy corner...) - in this way, they might use -ra... Who knows...

Best,

Ivan

[Alexander Vovin]

Dear Ivan,

Of course, this is possible, since Kyuushuu Old Japanese (KOJ) is known to us only from one poem (Fudoki kayoo 11 from the Hizen Fudoki, see NKBT 3: 231), but the question is how likely is it? In Eastern Old Japanese the usage of the diminitive -ro2 is not limited to family relationships or ko1 'girl', so most likely you are right -- it probably was the narrowing of -ra functionality in Western Old Japanese, But in the light of the absence of any textual attestations in KOJ we will be never able to tell for sure. Sigh...

All the best,

Sasha

Alexander Vovin
Directeur d'études, linguistique historique du Japon et de l'Asie du Nord-Est

ECOLE DES HAUTES ETUDES EN SCIENCES SOCIALES

CENTRE DE RECHERCHES LINGUISTIQUES SUR L'ASIE ORIENTALE

131, bd Saint-Michel, 75005 Paris

Adjunct Professor of Linguistics

Department of Linguistics

University of Hawai'i at Mānoa, Honolulu
HI 96822, USA

sasha...@gmail.com

[John Creamer]

So are you saying that it is more likely that there was a dialect called Hayato, based on the evidence of two words, one of which we might not be able to use as evidence? This is where the logic of linguistics eludes me. You can create rules and categories in what seems to be a scientific way, but then there are always exceptions. And for those exceptions you create new categories.

I can't say it's not fun to puzzle over where to divide morphemes, etc., but in science if your hypothesis about something fails to be supported by evidence or if you fudge data, you don't get your degree or your paper published. And in the recent case of a Harvard-trained biologist and her mouse switching, there are more serious consequences. 

[Alexander Vovin]

Dear Dr. Creamer,

I usually prefer to be non-confrontational on the e-mail lists that so many people can read (my reviews in print are quite a different matter). But the tone of your last message merits an adequate response, if not in tone but in essense. It also pains me to admonish a student who was once in one of my undergraduate classes, but pretty much like my hero Uesugi Kenshin, my patience can be stretched only to certain limits.

I am very sorry to see that your understanding of linguistics did not progress at all since you were in my Classical Japanese class (10 years ago?) and insisting on reading  texts "on the fly" with the complete disregard of grammar. May be reading an introduction to general linguistics (John Lyons) and then an introduction to historical linguistics (Lyle Canpbell) would help. Unless you learn some basics of linguistics, its logic will always elude you. Ditto for the grammatical analysis or either Old Japamese or Middle Japanese

Linguistics actually is between science and humanities. And even more so with comparative linguistics.It has its axioms, but you have to prove your theoremes using the rules, and not wishy-washy impressions.

Overall, I think that you are a bright and capable person and quite possibly a victim to this unnatural separation of linguistics and literature, which is de riguer now in the USA, but thabk Gid not everywhere in good old Europe.

And as a final advice: please read other people's postings carefully before replying to them with gibberish: neither I nor Ivan Rumasnek claimed that Hayato is a "dialect", and I specifically said that we cannot do anything on the basis of these two words.

Thank you,

Alexander Vovin
Directeur d'études, linguistique historique du Japon et de l'Asie du Nord-Est

ECOLE DES HAUTES ETUDES EN SCIENCES SOCIALES

CENTRE DE RECHERCHES LINGUISTIQUES SUR L'ASIE ORIENTALE

131, bd Saint-Michel, 75005 Paris

Adjunct Professor of Linguistics

Department of Linguistics

University of Hawai'i at Mānoa, Honolulu
HI 96822, USA

sasha...@gmail.com

[Cynthea Bogel]

All, There have been many unfortunate bloopers and exchanges on the List this past week. We are probably about to go viral.

How about a breather for a day? Private exchanges?  

I would hate to see damage done (or more damage done) to anyone among us if it can possibly be avoided.

I am not suggesting I am immune to outbursts, ill-considered responses, or bloopers, and I acknowldedge the measured replies to perhaps less-measured comments. I am simply waving a kusunoki branch from Kyushu (and it is heavy!)

Cynthea Bogel

Kyushu University