27.2902, Review: Comp Ling; Historical Ling: Lewis, Pereltsvaig (2015)

[The LINGUIST List via LINGUIST]

LINGUIST List: Vol-27-2902. Fri Jul 08 2016. ISSN: 1069 - 4875.

Subject: 27.2902, Review: Comp Ling; Historical Ling: Lewis, Pereltsvaig (2015)

Moderators: linguist at linguistlist.org (Damir Cavar, Malgorzata E. Cavar)
Reviews: reviews at linguistlist.org (Anthony Aristar, Helen Aristar-Dry, Robert Coté, Sara Couture)
Homepage: http://linguistlist.org

*****************    LINGUIST List Support    *****************
                       Fund Drive 2016
                   25 years of LINGUIST List!
Please support the LL editors and operation with a donation at:
           http://funddrive.linguistlist.org/donate/

Editor for this issue: Michael Czerniakowski <mike at linguistlist.org>
================================================================


Date: Fri, 08 Jul 2016 13:49:17
From: David Stifter [david.stifter at nuim.ie]
Subject: The Indo-European Controversy

 
Discuss this message:
http://linguistlist.org/pubs/reviews/get-review.cfm?subid=36119377


Book announced at http://linguistlist.org/issues/26/26-2286.html

AUTHOR: Asya  Pereltsvaig
AUTHOR: Martin W.  Lewis
TITLE: The Indo-European Controversy
SUBTITLE: Facts and Fallacies in Historical Linguistics
PUBLISHER: Cambridge University Press
YEAR: 2015

REVIEWER: David Stifter,  

Reviews Editor: Robert A. Cote

SUMMARY

“The Indo-European Controversy” by Asya Pereltsvaig and Martin Lewis takes an
article (Bouckaert et al. 2012; hereafter referred to as Gray & Atkinson 2012)
from the journal Science as the starting point for their wide-ranging
evaluation of non-linguistic contributions to this debate. Chapter 1,
“Ideology and interpretation from the 1700s to the 1970s”, contains a concise
history of the reception of Indo-European studies since their beginning in the
late 18th century, its (mis)use by scholars from extralinguistic disciplines,
and its impact or lack thereof on political ideologies in the 19th and 20th
centuries; while in Chapter 2“Anatolia vs. the Steppes”, Colin Renfrew’s
influential contribution of 1987 is evaluated. His so-called ‘Anatolian
hypothesis’ about the spread of the language family via demic diffusion,
accompanying the spread of agriculture in the neolithic, has been fervently
rejected by basically every linguistic expert conversant with the actual facts
of Indo-European Studies, but it continues to exert its appeal on
non-specialists. The driving force behind this debate is not so much a
linguistic one than one informed by historical and archaeological ideologies,
without being rooted in the hard facts of linguistics. 

In the central section of the book, “Part II. The failings of the Bayesian
phylogenetic research program”, the authors take the methodological fallacies,
fundamental misapprehensions, and practical errors apart that riddle Gray &
Atkinson’s work. In Chapter 3, “What theory we want and what theory we get”,
the authors demonstrate that creating trees of language relationships on the
basis of lexical comparisons alone is an approach that is bound to lead to
erroneous results. Diachronic phonology offers more reliable criteria to model
language relationships. 

Chapter 4 “Linguistic fallacies of the Bayesian phylogenetic model”
meticulously lists errors and misunderstandings that beset the fundamental
assumptions of Gray & Atkinson’s work: the failure to distinguish innovations
from retentions, including the failure to understand that because of the
nature of linguistic data, especially because of universal phonological
tendencies, there is in many instances an intrinsic directionality in change.
Phonological change is not unlike more traditional physical events in that
they are predicated by the arrow of time and their results may, mutatis
mutandis, be viewed as an increase in entropy. For this reason, what is a
linguistic innovation can by itself be evident for the expert, whereas a
statistical calculation only based on frequencies will unavoidably lead to a
wrong picture. Other potential pitfalls are an over-reliance on lexical data,
which is intrinsically imprecise because its significance can be blurred by
factors such as borrowing, unidentified or only inadequately identified
borrowing, or divergent lexical usage, and not awarding sufficient importance
to phonology and grammar, which offer more reliable and consistent information
about linguistic cognacy. Another stumbling block is the unreflected use of
Swadesh lists of core vocabulary for language comparison with their very
subjective, if not random character of word selections.

Chapter 5, “Dating problems of the Bayesian phylogenetic model” addresses the
inherent difficulty of assigning clear-cut dates to linguistic developments.
While glottochronology in the traditional sense has long fallen out of favour,
the Bayesian approach advocated by Gray & Atkinson offers only little progress
because distorting factors (such as undetected borrowings) can easily skew the
statistics. If the amount of data is not large enough, small changes in the
input will have massive repercussions on the output. The authors of the
present book point out that, while Gray & Atkinson make use of ostensibly
objective calibration points, i.e. historically secured events, the
significance of such dates is misunderstood. While historical events may be
the trigger for sociolinguistic developments that will eventually result in
linguistic change, they cannot be identified with the dates of those changes
as such. The inherent fuzziness of the data leads to distortions in the
calculations that add up to a 3000-year timing difference between the
computational model and the consensus or near-consensus view with which
scholars in the field actually operate. In Chapter 6, “The
historical-geographical failure of the Bayesian phylogenetic model”,
Pereltsvaig & Lewis demonstrate how the static and dynamic maps produced by
Gray & Atkinson (printed and online) ignore simple historical facts about the
distribution of languages and their speech communities and how they end up
with historically and politically bizarre scenarios. 

Chapter 7, “Unwarranted assumptions”, questions the many fundamental concepts
that underlie Gray & Atkinson’s theories of languages spread, i.e. their ideas
about migration vs. diffusion, how language spread interacts with population
spread, topographical factors (such as sea coasts), and a simplistic view of
the contiguity of languages over specific regions. Although images of mass
population movements in the old sense of ‘Völkerwanderungen’ are surely
simplistic, historical sources abound in evidence for migrations and
transplantations of speech communities. The migration model thus receives
factual support from countless real-life examples from across the globe,
against the theory-driven diffusion model; the crucial point being that even
diffusion requires the operation of ‘agents’ to actually take place (Vogl
2012). If the agents are human beings or groups of humans, contingencies come
into play that will disrupt any neat numerical model. The final section in
this chapter is devoted to those “fallacious and unexamined assumptions” (155)
by which biological evolution tends to be used as an analogy for linguistic
evolution. Arguing that the similarities are superficial and only pertain to
the use of cladistic trees, albeit for quite different purposes in each of the
two disciplines, the authors state that the fundamental procedural principles
of linguistic and genetic differentiation could in fact not be further apart.

After all the deconstruction, the third section of the book, “Searching for
Indo-European origins” , is devoted to the assembling of facts that allow
linguists to make constructive statements about the where and the when of the
Proto-Indo-European language and its break-up. In Chapter 8, “Why linguists
don’t do dates?  Or do they?”, Pereltsvaig & Lewis give an introduction to
linguistic paleontology and then roll out the major arguments associated with
the problem of wheeled transport as an example of this approach. They arrive
at the conclusion that Proto-Indo-European could not have started to break up
before the origins of wheeled transport ca. 3500 B.C. Another equally
important dating criterion of material culture is not mentioned here even
though it pins Proto-Indo-European effectively to the same chronological
horizon: the presence of the words for ‘sheep’, *h2ou̯i-, and ‘wool’,
*h2u̯l̥h1/2no/eh2-, and in particular the fact that the latter word seems to
be derived from the former (the initial sequence h2u̯ is identical with the
consonantal skeleton of *h2ou̯i- ‘sheep’), give evidence for the acquaintance
of the Proto-Indo-European people with woolly sheep. Sheep with sufficient
wool for economic exploitation were not known before the 4th mill. b.c.
(Mallory & Adams 2006: 238).

Chapter 9, “Triangulating the Indo-European homeland” exploits linguistic
archaeology and language contact data to find clues about the localisation of
the homeland. The authors argue that, while all the evidence is not always
fully conclusive, its cumulative drift is in favour of the traditional theory
that Proto-Indo-European was situated in the Pontic Steppes in the 4th
millennium BCE In Chapter 10, “The non-mystery of Indo-European expansion” ,
the authors sketch a sociolinguistic and anthropological scenario to explain
the apparent ‘success’ of Indo-European languages. Chapter 11, “Whither
historical linguistics?”  is concerned with possible alternative methods for
establishing and computing language relatedness, e.g. the Parametric
Comparison Method (PCM), developed in the ‘Language and Gene Lineages project
headed by Giuseppe Longobardi (University of York), a method that is based on
generative syntax and that compares fundamental parameters of languages and
which holds a promise for exciting future results. In the book’s conclusion, 
“What is at stake in the Indo-European debate”, the authors find very harsh
words against Gray & Atkinson and the journal “Science” (esp. p. 230), the
central accusation being that their theory is unempirical.

EVALUATION

A  lot of professional frustration permeates the introductory pages, but it is
the sting of this frustration that motivates Pereltsvaig & Lewis to engage
critically with the methodological premises of the allegedly scientific
approach. The authors employ a very polemic, pointedly formulated style which
at times reminds the reader of lawyers pleading their cause in a criminal
court. Right from the ‘Introduction’ (1–15), Pereltsvaig & Lewis leave no
doubt about their own position and their intention to refute every bit of the
article of contention. Speaking of “incorrect and … incoherent linguistic
information” (2), they accuse Gray & Atkinson of building on “erroneous and
unexamined suppositions about language differentiation, distribution, and
expansion” (3). They view extralinguistic attempts at solving central
questions of Indo-European antiquity in a very critical light. Pereltsvaigs’s
and Lewis’s basic tenet is that the distribution of human languages is “only
vaguely analogous to organic evolution”, “has nothing in common with the
spread of viruses” (3), and should therefore be described with different
models altogether.

On a positive note, the authors draw from an admirable knowledge of sources,
not infrequently making reference to fairly obscure publications, and they
employ a broad range of facts from diverse scientific fields. They convert
their ‘ammunition’ into a forceful attack not only against Gray & Atkinson,
but against the nonchalant reception and treatment of languages, and of
historical linguistics in particular, by the public and big media. Their own
position, on the basis of which they critically assess alternative hypotheses,
reflects the broad consensus of scholars in Indo-European Studies. Therefore,
the book is unlikely to change the opinion of anyone who already works in the
field, simply because, judging by my own professional experience, most
colleagues in the field hold this view anyway. Within the field, the authors
would be preaching to the faithful. In fact, much of what is said in the book
is  commonplace in historical linguistics and in Indo-European Studies in
particular. A big achievement of Pereltsvaig & Lewis, though, is to argue the
case from so many diverse angles, and for this reason the book can be useful
even to specialists.

Bayesian methods have become a popular and powerful tool in many disciplines
where large data has to be analysed. In fact, the critique which Pereltsvaig &
Lewis direct against the ‘Bayesian phylogenetic model’ does not actually
address the mathematical principles of the Bayesian approach as such at all,
but is rather concerned with the erroneous application and the non-expert
handling of the linguistic input data by Gray & Atkinson. It would, in fact,
be interesting to see what results can be achieved if experts in historical
linguistics, who have a well-grounded understanding of the data, work together
with experts in Bayesian methods.

I want to finish on a more positive note. New facts such as advances in
scientific disciplines outside of historical linguistics (e.g. Haak 2015) lend
unexpected, but very welcome support to the basic tenets Pereltsvaig & Lewis.
The very fact that a book like this had to be written, even though all the
facts have been on the table for decades, makes a disillusioning statement
about the lack of public impact of the work of historical linguists. It is
therefore to be hoped that the book’s main audience will be scholars in
neighbouring disciplines, or even further away, who may easily be blinded and
led astray by the outwardly shining ‘scientistic’ arguments used by Gray &
Atkinson. If it achieves this objective, it will be a very important
contribution to the scholarly debate about the origins and the expansion of
the Indo-European languages, even though the style in which it is written may
at times come across too polemically.

Occasionally, minor errors catch the eye, e.g.: in the satem-group of
Indo-European, there is no uniform outcome *s of the Proto-Indo-European
palatal *k’ (p. 65), but the precise nature of the sibilant differs from
language to language, ranging from plain [s] in Iranian (providing the source
for the modern term ‘satem’), to a variety of palatal sounds that remained
separate from s in most of the other languages in the group.  The
self-designation of the Spanish language is not “espagnol” (p. 111), but
español. Also, the ruling house of the Austro-Hungarian Empire were the
Habsburgs, not the “Hapsburgs”. The (pre-)PIE word *Hok̑tō(u̯) ‘tetrad’, the
putative basis of the Proto-Kartvelian loan *otχo-‘four’, was by no means lost
in later Indo-European (p. 192), but survives across the board as the numeral
‘eight’. Finally, on p. 198, the presentation of the facts is misleading: the
reader gets the impression that Indo-European *pork̑os ‘piglet’ with a
palatalised *k̑ had been borrowed into Proto-Finno-Ugric; however, the
Proto-Finno-Ugric word is *pɔ̄rš́ɔs borrowed from already satemised
Proto-Indo-Iranian.

REFERENCES

Bouckaert, R., Lemey, Ph., Dunn, M., Greenhill S.J., Alekseyenko A.V.,
Drummond A.I., Gray, R.D., Suchard M.A and Atkinson, Q.D. 2012. ‘Mapping the
Origins and Expansion of the Indo-European Language Family’. Science 337 no.
6097 (24.8.2012), 957–960 [DOI: 10.1126/science.1219669].

Haak, W. et al. 2015. ‘Massive migration from the steppe was a source for
Indo-European languages in Europe’, Nature 522 issue 7555 (11.6.2015, publ.
online 2.3.2015), 207–211 [DOI: 10.1038/nature14317]. Precis (12.2.2015) at:
http://www.nature.com/news/european-languages-linked-to-migration-from-the-eas
t-1.16919.
Mallory, J.P. and Adams, D.Q. 2006. The Oxford Introduction to
Proto-Indo-European and the Proto-Indo-European World, Oxford – New York:
Oxford University Press.
Renfrew, C. 1987. Archaeology and Language: The Puzzle of the Indo-European
Origins. London.

Ringe, D. et al. 2002. ‘Indo-European and computational cladistics’,
Transactions of the Philological Society 100, 59–129.

Vogl, G. 2012. ‘Fundamentals of diffusion and spread in the natural sciences
and beyond’, in: Migrations. Interdisciplinary Perspectives. Eds. M. Messer,
R. Schroeder, R. Wodak, Wien: Springer 2012, 261–266.


ABOUT THE REVIEWER

David Stifter, Professor of Old Irish at the Department of Early Irish at
Maynooth University, Ireland. This review was written as part of the research
project Chronologicon Hibernicum
(https://www.maynoothuniversity.ie/chronologiconhibernicum) that has received
funding from the European Research Council (ERC) under the European Union’s
Horizon 2020 research and innovation programme (grant agreement No. 647351).
Part of the project will be to use Bayesian methods for the dating of Early
Irish language developments.





------------------------------------------------------------------------------

*****************    LINGUIST List Support    *****************
                       Fund Drive 2016
Please support the LL editors and operation with a donation at:
            http://funddrive.linguistlist.org/donate/

This year the LINGUIST List hopes to raise $79,000. This money 
will go to help keep the List running by supporting all of our 
Student Editors for the coming year.

Don't forget to check out Fund Drive 2016 site!

http://funddrive.linguistlist.org/

For all information on donating, including information on how to 
donate by check, money order, PayPal or wire transfer, please visit:
http://funddrive.linguistlist.org/donate/

The LINGUIST List is under the umbrella of Indiana University and
as such can receive donations through Indiana University Foundation. We
also collect donations via eLinguistics Foundation, a registered 501(c)
Non Profit organization with the federal tax number 45-4211155. Either
way, the donations can be offset against your federal and sometimes your
state tax return (U.S. tax payers only). For more information visit the
IRS Web-Site, or contact your financial advisor.

Many companies also offer a gift matching program, such that
they will match any gift you make to a non-profit organization.
Normally this entails your contacting your human resources department
and sending us a form that the Indiana University Foundation fills in
and returns to your employer. This is generally a simple administrative
procedure that doubles the value of your gift to LINGUIST, without
costing you an extra penny. Please take a moment to check if
your company operates such a program.


Thank you very much for your support of LINGUIST!
 


----------------------------------------------------------
LINGUIST List: Vol-27-2902	
----------------------------------------------------------
Visit LL's Multitree project for over 1000 trees dynamically generated
from scholarly hypotheses about language relationships:
          http://multitree.org/