14.934, Review: Historical Linguistics: Carpelan, et al. (2002)

LINGUIST List linguist at linguistlist.org
Sat Mar 29 23:18:07 UTC 2003

LINGUIST List:  Vol-14-934. Sat Mar 29 2003. ISSN: 1068-4875.

Subject: 14.934, Review: Historical Linguistics: Carpelan, et al. (2002)

Moderators: Anthony Aristar, Wayne State U.<aristar at linguistlist.org>
            Helen Dry, Eastern Michigan U. <hdry at linguistlist.org>

Reviews (reviews at linguistlist.org):
	Simin Karimi, U. of Arizona
	Terence Langendoen, U. of Arizona

Home Page:  http://linguistlist.org/

The LINGUIST List is funded by Eastern Michigan University, Wayne
State University, and donations from subscribers and publishers.

Editor for this issue: Naomi Ogasawara <naomi at linguistlist.org>

To give you an incentive to donate, many of our Supporting Publishers
have generously donated some amazing linguistic prizes. As a donor you
are automatically entered into this prize draw. To find out what's on
offer and the rules etc., visit:


We still have a long way to go, however, to reach our target of
$50,000. Please make a donation at:


The LINGUIST List depends on the generous contributions from
subscribers like you; we would not be able to operate without your

The moderators, staff, and student editors at LINGUIST would like to
take this opportunity to thank you for your continuous support.

What follows is a review or discussion note contributed to our Book
Discussion Forum.  We expect discussions to be informal and
interactive; and the author of the book discussed is cordially invited
to join in.

If you are interested in leading a book discussion, look for books
announced on LINGUIST as "available for review." Then contact
Simin Karimi at simin at linguistlist.org.


Date:  Sat, 29 Mar 2003 15:00:04 +0000
From:  John Hammink <John.Hammink at F-Secure.com>
Subject:  Early Contacts between Uralic and Indo-European

-------------------------------- Message 1 -------------------------------

Date:  Sat, 29 Mar 2003 15:00:04 +0000
From:  John Hammink <John.Hammink at F-Secure.com>
Subject:  Early Contacts between Uralic and Indo-European

Carpelan, Christian, Asko Parpola and Petteri Koskikalio, ed. (2002)
Early Contacts between Uralic and Indo-European: Linguistic and
Archeological Considerations, Finno-Ugrian Society.

Announced at http://linguistlist.org/issues/13/13-829.html

John Hammink, F-Secure Corporation

Over the years, there has been much speculation about the early
identity and roots of the Uralic languages, and indeed, in recent
years there seems to be a renewed interest in the topic.  In fact,
since the 1980's, linguists have begun to make concessions to
archeologists, breaching an intellectual gulf that was considered
pretty daunting up until that point, particularly due to differences
in methods of study and dating.  In addition, a well-known culture of
academic cliquishness has always endured (as in practically every
academic field) which lends credence to some ideas over others, while
not necessarily putting either into the correct perspective.

But, particularly in Finland, such topics have seen great national
interest since the early 19th century, beginning with Castren and
continuing up to recent times with conferences and seminars on
Finno-Ugric enthnohistory.  One such meeting was the 3 day symposium
on ''Contacts between Indo-European and Uralic speakers in the
Neolithic, Eneolithic and Bronze Age in the light of linguistic and
archeological evidence'', which was held at Tvärminne Research
Station at the University of Helsinki in January, 1999.  At that
conference, the 18 papers and 4 abstracts that comprise this book were

The first paper, ''Persistent Identity and Indo-European Archaeology
in the Western Steppes'', presented by David W.  Anthony, deals with
the problem of reconciling what could be seen as societal polymorphism
with the notion that PIE language would appear to be linked to
specific tombs and settlements.  Just as ''there is no necessary
relationship between the way people speak and the way they make pots
or stone tools'' there is a reluctance among western archaeologists to
affiliate linguistic and archaeological cultures.  In the larger
scheme, this article seems to set the scene for things to come, and so
it appears in the right place at the beginning of the volume.

The author argues that there should be a middle ground in the debate,
that, in fact, stable ethno-linguistic (pre- state) frontiers were
possible, and he cites the Northern Iroquois as an example.  One
practical example also cited was the western boundary of the
Pontic-Caspian region, which, in several stages, remained a stable
frontier for several millennia.  The other Pontic-Caspian frontiers
were discussed as well.

Christian Carpelan outlines a scenario for the emergence of
Uralic/Finno-Ugric speaking groups in ''Late Paleolithic and
Mesolithic Settlement of the European North Possible Linguistic
Implications.''  To accomplish this he describes ''an archaeological
culture as a sphere of internal communication probably based on a
common identity''.  In his scenario, the European population north of
the Alps retreated to the southeast and southwest due to the Last
Glacial Maximum (LGM) which has been dated around 18000 BP (BP =
radiocarbon years before AD 1950) or approximately 22000 calBC.  This
created an Eastern and a Western Block.

The Western Block represented (among others) the Magdalenian cultures,
which repopulated the depopulated territories starting around 14500BP
/15400 calBC; and the aptly-named Hamburg culture, which appears to be
an extension of the Magdalenian cultures.  Starting from 13200 BP /
13900 calBC, the Hamburg culture spread over an area including what is
nowadays northern Germany, Poland, Southern Sweden and Britain and
part of the North Sea, which was then dry land.

The Eastern Block represented much of the widespread Gravettian techno
complex (which was displaced in Southern France by the Magdalenian
cultures).  Most of the discussion here seems to center on place
rather than specific cultures, although it is suggested later that
''people from the west infiltrated the [Eastern Block] region, which
was already inhabited by groups representing the Eastern Block.''

The author continues to trace the transition to the Mesolithic period
through the initial colonization of the Scandia Peninsula around 10300
7150 BP / 10100 6000 calBC.  When discussing the initial colonization
of the Baltic, Northern Russia and Fennoscandia, the author discredits
Bryosov's scenario that ''the northern part of Russia was initially
colonized from the East.''  The article goes on to summarize events in
Eastern Fennoscandia (8050 2400 BP / 7000 500 CalBC).  It would appear
that the Volga-Oka area eventually became a sort of cultural boiling
pot, which presumably also carried linguistic and genetic influence.
Of the 7 cultural horizons, the Lyalovo or Pitted Ware was
highlighted, as it seems to be the culture that, evolving to Combed
Ware 2, expanded to a significantly large area, encompassing Sweden,
the Gulf of Bothnia, the White Sea and extending eastward to the
Urals. Of course, there were several cultures that appear to have
intersected with it, putting their own unique stamp on things.  The
article contains short sections on Craniometry and Genetics as well as
Languages.  The latter would appear to be a more speculative take on
Uralic roots.  The author doesn't appear to propose a specific
geographical locale for the Uralic origins, but rather proposes only
that the language has started from the Eastern Block settlers. It
would appear unlikely that Ahrensbergian influence in western Russia
at the end of the Paleolithic would have sparked Uralic development,
although it may have left substrate elements.

Christian Carpelan and Asko Parpola continue the linguistic line of
reasoning with ''Emergence, Contacts and Dispersal of PIE, PU, and PA
in Archaeological Perspective.''  ''When and where were
Proto-Indo-European (PIE), Proto-Uralic (PU), and Proto-Aryan (PA)
spoken?  And when and where did they split into their main branches?''
While archeological findings can be mapped and dated, as no written
remains exist, any correlation with definite languages is problematic.
The article begins by discussing problems in distribution with
radiocarbon dating when Europe (south of 54 degrees) as compared to
the Volga-Oka confluence up to the Urals do not contain a
proportionate amount of published dates, which appears to make
checking either extremely laborious or practically impossible.  The
dispersal of late PIE is discussed.  As all Indo-European languages
possess inherited vocabulary related to wheeled transport and the PIE
daughter languages have not borrowed them from one another after the
dispersal, one can assume that PIE speakers knew and used wheeled
vehicles.  As wheeled transport was first invented sometime during the
second quarter or middle of the fourth millennium BC, and dispersed
over the next two centuries or so, then the dispersal of PIE cannot
have taken place much earlier than 3500 calBC.  The article also
discusses PIE in relation to the following branches: Tocharian,
Anatolian, and Proto- Northwest-Indo-European.  The Pit Grave culture
is considered as a central group for late PIE, while the Khvalynsk
culture is considered to possibly possess the language that was the
immediate predecessor of early PIE.

Early Indo-European Loanwords in Uralic languages are also considered,
and some existing theories are disputed.  Of notable mention here is
Juha Janhunen's reluctance to accept PIE loanwords into PU because of
his placement of the PU homeland in central Siberia.  Because this
idea has no basis in the archaeological record, the authors dispute
it.  Also considered here are the Uralic language family and its main
branches. When discussing the disintegration of PU, the authors
concede that all Finno-Ugric languages ''appear to have been
originally spoken in the forest area west of the Ural mountains.  Thus
the homeland of the Finno-Ugric protolanguage has been considered to
be in one of three more or less adjacent or overlapping regions: 1)
The area of the mid Volga; 2) The area between the Volga, Kama,
Pechora and the Urals, and 3) The entire region between the Baltic sea
and the Urals.  Curiously, the Volga-Oka interfluve is again cited
here as a region that ''continuously created both cultural and
demographic surplus'' and the pottery styles produced here eventually
appear to have made their way to Fennoscandia.  But temporally
speaking, the Lyalovo culture represented by Pitted Ware (c. 5000-3650
calBC) is seen as a better candidate for PU.  The Lyalovo culture
sprung from the upper Volga region and spread to the Onega region of
Russian Karelia.  Their Pitted Ware influence was felt all the way
down south to the forest steppe between the Dnepr and Don.  Lyalovo
culture seems to have extended barely as far the Kama basin but not
into it.  As such, there seems to be no possibility of Samoyedic
splitting off to Siberia, but there is ample evidence that this
happened later.  The article also discusses the corded ware cultures
as a basis for northwest Indo-European loanwords.  The Balanovo and
Fat'yanovo cultures expanded eastward into the Volga-Oka interfluve.
It is reasoned that the oldest Baltic- loanwords in Proto-Finnic came
with the late Neolithic Kiukainen culture (c. 2300-1600 calBC).

The interesting high-water mark of this article appears in the chapter
called ''The early Aryan Loanwords for 'Honey' and 'Bee''' In it, the
authors assert that it is ''generally accepted that Proto-Finno-Ugric
(PFU) *mete 'honey' (distributed in Finnic, Saami, Mordvin, Udmurt,
Komi, and Hungarian) is borrowed from PIE=Pre-Proto-Aryan (PA) *medhu-
(which became *madhu- in Proto-Aryan).''  It is also probable that
'''bee' as a compound meaning 'honey- collector' was borrowed into PFU
before the PA sound change *e > a took place.''  Beeswax is discussed
in a following section as ''A new Indo-Aryan Etymology for a
Volga-Permic word.''  Early Indo-European and Aryan Loanwords are also
found in Proto-Samoyedic (PS), for example, Northern Samoyedic *jäe
'meal, flour' is related PA *yeva < PIE *yewo.  Finally, the authors
present a summary of their article.

H. P. Francfort's ''The Archeology of Protohistoric Central Asia and
the Problems of Identifying Indo-European and Uralic-Speaking
populations'' addresses the archaeological identification of
linguistic groups in Central Asia in the Bronze Age.  The author
presents two cases that represent the problems of
1. tagginglinguistic groups by using available archeological
   data; and
2. the lack of any material representing non-Indo-European
   speaking populations.

The civilizations studied are the Oxus Civilization and the
Afanasevo/Okunevo sequence. The Oxus Civilization existed between ca
2500 and 1500 b.c. in the area that included present-day Afghanistan,
Turkmenistan, southern Uzbekistan, Tajikistan and Eastern Iran.  The
Indian, Iranian and steppe connections are considered, along with
other ethnolinguistic attributes.  Finally, iconography and symbolic
systems are considered as some evidence that points to
Non-Indo-European influence.

The Afans'evo-Okunevo Complex/Sequence is considered in the context of
its territory and chronological sequence.  The Afanas'evo culture,
dating from 3rd millennium, ''was widespread in western Mongolia,
northern Xingjian, southern Siberia, eastern and central Kazakhstan,
with connections or extensions in Tajikistan and the Aral area.''
They were herdsmen and hunter-gatherers, they buried their dead in
conic or rectangular enclosures, typically in a supine position, not
unlike their Yamnaya counterparts in the European steppes.  As with
the Oxus Civilization, the article considers time-space location.
Also discussed are iconography and symbolic systems, again pointing to
non- Indo-European worlds, possibly either Uralic or Altaic.  The
article contains several interesting drawings: In one: an Okunevo
masked figure is shown; others show tomb slabs; yet others depict
Okunevo/Afanas'evo masks, monsters and petroglyphs.

Kaisa Hakkinen's paper ''Prehistoric Finno-Ugric Culture in the Light
of Historical Lexicography'' examines vocabulary elements
representative of ''the oldest lexical stratum common to the Uralic
languages, whether originally indigenous or borrowed.''  Apparently
much data is derived from hunting cultures, for example terms for
hunting and fishing equipment and game animals.  Much less derived
from terms relating to agriculture, which apparently have a fairly
narrow geographical distribution.

Firstly, the article defines the age of available lexical material.
Oldest Finno-Ugric written documentation dates only from the middle
ages, so a comparative lexicological study of related languages is
necessary.  Once can assume a word to be of early origin and
indigenous if the words are restricted to Finno-Ugric languages yet
have a wide distribution.  Likewise, if a word appears in other
languages, it may be, in fact a loan.  Lastly, words restricted to
Finno-Ugric languages, but with a narrow distribution, can be assumed
to be fairly new.

Next, the article considers the oldest common lexical stratum of the
Uralic languages.  Otto Donner's 1882 study of the common cultural
stock of Finns and Mordvinians is considered the pioneering work here.
Also mentioned here is Björn Collinder's 'Fenno-Ugric Vocabulary',
from 1955, as a source of study of the oldest common Uralic lexicon.
In the following section, the lexical domains of the oldest common
vocabulary are discussed.  These include time, sensations, flora and
fauna, trade and transport, quality notions, nature, hunting and
fishing, processes and states, building and construction, nourishment,
body parts, speech and thought, family, space and time relations,
pronouns, and other miscellaneous items.  The items are quantified,
first in total; secondly in Uralic, thirdly in Finno-Ugric, and lastly
in Finnish itself.  Interestingly enough, Uralic languages dominate in
pronominals and family relationship terms; while Finno-Ugric seems to
record more words overall than Uralic.  Lastly, before an appendix,
the article evaluates distribution, and thus, certainty of etymology
of the word items.

In his article, Eugene Helimski outlines ''Early Indo-Uralic
Linguistic Relationships: Real Kinship and Imagined Contacts''.  He
begins with the following premise: ''there may be many riddles but no
wonders in linguistic prehistory.'' Or, in other words, ''all too
often, the early prehistory of languages is viewed as a terra
incognita with its own unknown rules (which are therefore invented by
some scholars freely and with vivid imagination).  While on one hand
stating that early linguistic relationships can be studied with the
same relevance and methodology that historical linguistic treats more
recent events, linguists can only be used in the same way in this
context if
1. Human language underwent no fundamental changes since
   Upper Paleolithic/Mesolithic times;
2. Language prior to these eras possessed structural types
   which lay within the same typological limits as in the
   contemporary world;
3.  Language, prior to these eras, evolved similarly as those
   attested in the contemporary world.

That said, the author begins by outlining some of the types of
language kinship.  The first discussed (and most common type) is
direct kinship determined by divergent evolution.  In this scenario,
one uniform language spoken on a small territory spreads (due to
historical circumstances) ''far beyond the original territory''.  Over
time, the distributed languages become dialectically and
sociolectically distinguished, until distinct states are formed.
Eventually each regional variant undergoes its own distinct and
independent evolution.  The most obvious example of this is the
pre-classical Latin of Latium (which evolved to French, Italian,
Romanian, Portuguese etc.).  Other language families are mentioned
here, too (Slavic => Common Slavic, Aver/Slavic, eventual rise of
Polish, Russian, Bulgarian), also Mongolic, Permic, Germanic,
Indo-Aryan, Bantu, Samoyedic, Turkic, Polynesian, etc.  In all of the
above cases, while accompanying circumstances (e.g.  substratum,
koine, secondary interaction) are dissimilar between each case, ''they
do not introduce any major changes into the picture of kinship''.

Another type of language kinship is lateral kinship, characterized by
relexification and typical of ''mixed'' languages.  This is generally
a sociolinguistic phenomenon where speakers of a source need or want
to speak a target language, but ''being adults grown up with another
language'' are unable to handle the target grammar.  So instead, they
simply repopulate their source grammatical system with the target
language's words.  Some cases are Anglo-Romany, Ma'a or Mbugu.  A more
complex form does all of the above plus adds some simpler grammatical
elements from the target as well (Copper Island Aleut).  Another form
of lateral kinship happens when the lexicon of the target is
superimposed on a synthetic, simplified version of the source language
grammar (like with most pidgins and creoles).  These processes
''produce only new languages in which grammar and vocabulary are of
different origin''.  They also produce only one ''mixed'' language (as
opposed to a new language family).  Most, importantly, these languages
typically do not evolve and are short lived.  I would also add (feel
free to prove me wrong) that these languages are typically also spoken
in situations where interaction between speakers of the target and
source languages are necessary; they are not ''native'' languages (or
mother tongues) as such.

So by applying simple inductive reasoning, it is common enough to
reach the conclusion that PIE and PU share direct kinship.  For
example, Kaisa Häkkinen's paper demonstrates ''the direct
relationship between the stability of words (stems) and their
occurrence in the common vocabulary of several Eurasian language
families.''  One group of 18 items demonstrated ''100% etymological
certainty'', while a second group of 23 items demonstrated 90%
etymological certainty,. meaning that ''their counterparts were
missing or dubious'' in only 10% of the sampled languages.

Thus, the author considers a lateral kinship between PIE and PU
unlikely.  First, there are broad and frequent occurrence of both
vocabulary cognates and grammatical structures across all the
languages.  Second, ''nothing in the linguistic structures of PIE or
PU implies a 'mixed' past.  Lastly, the author considers the
statistical rarities of a lateral kinship.  The author neither accepts
or dismisses the Indo-Uralic (Nostratic) kinship in this article.
There is a subjective need for a proof at some level, but simply a
lack of comparable data.  The rest of article is devoted to rejecting
alternative treatments of PIE/PU kinship.

In ''Indo-Uralic and Ural-Altaic: on the Diachronic Implications of
Areal Typology'', Juha Janhunen attempts to locate the prehistorical
homeland of Uralic, Indo-European, and Altaic by studying typological
relationships among regional languages.  In the first section, he
reviews the Indo-Uralic typological discrepancy.  The most obvious
difference between the two is phonology.  PIE, for example, may have
had from 1-5 vowel sounds, and close to 30 consonants, while PU seems
to have many more vowel sounds (around 8) while much fewer consonants
(probably less than 17).  Morphologically, the languages actually seem
to share a number of similarities, but differences include the fact
the PIE employed affixation and flexion (including Ablaut), while
Uralic used only a sort of mechanical suffixation (no prefixation or
infixation).  Syntactically, the data available would seem to remain
relatively inconclusive, with SOV word order prominent in both PU and
PIE, while PIE also seems to show some use of SVO.

The Ural-Altaic Typological Parallelism is also discussed.  A
reference is made to the Nostratic framework in which this work is
carried out.  The Altaic languages are divided into two groups (both
of which appear typologically similar to Uralic).  Again, phonology,
morphology, and syntax are discussed.

Uralic is then viewed in the Eurasian context.  The available data
from Indo-European, Semitic, and Caucasian parallels seems to suggest
that PIE originates from an area which includes Anatolia, Mesopotamia,
and the Caucasus.  However, archaeology puts it in a slightly
different area, the Pontic steppes, north of the Black sea.  It is
more difficult to locate the original geographic center of the
Ural-Altaic complex.  The author suggests that around 2000 BP, all
''currently known Altaic entities were located in the northern part of
the Far East'' that is, the region encompassing what is now Mongolia,
Manchuria, and Korea.  However there were already, by this time
branches including those in the Yenisei-Baikal region going all the
west to the Baltic Sea (Finnic and Saami).  Lastly, the author
proposes a framework for future study.

In his article, Petri Kallio raises the question: ''Phonetic Uralisms
in Indo-European?''  He deals with those features in IE languages
which may attest to Uralic influence.  The languages considered here
are Proto-Balto- Slavic (PBS), Proto-Germanic(PG), Proto-Indo-Iranian
(PII) and Proto-Tocharian (PT).  The author begins with the idea of
tracing phonetic Uralisms to another language, e.g. as a Finn speaks
English, and extends the analogy that indeed, ''it may be possible to
hear pronunciation errors made by ancient Uralians thousands of years

The article contains a number of tables.  The first of such shows
phoneme substitutions in late PIE loanwords in PU/PFU/PFP.  It is easy
to note from this that sound shifting between IE and Uralic was, in
itself, quite minimal, but taken as a whole, could add up to
significant differences between source and loanwords.  Also, it's
noted that ''Late PIE stops had three manners of articulations and 5
places of articulation; where as PU stops had one manner of
articulation and three places of articulation.''  This seems to
support Janhunen's assertion about number of consonants in PIE as
opposed to PU, but an exact mathematical correlation cannot be
determined from this.

The second table lists PIE and Uralic/Finno Ugric Phonemes.  One can
easily see from this that PIE simply does contain many more consonants
and fewer vowels than PU.  The third table lists Tocharian and PU
phonemes.  One interesting point mentioned here is the fact that
places of articulation of affricates are identical between PII and
PFU.  The author follows with an in-depth discussion of the four
languages mentioned earlier.

Earliest Indo-European Loanwords in Uralic/Finno-Ugric are on the
docket in Jorma Koivulehto's ''The Earliest Contacts between
Indo-European and Uralic Speakers in the Light of Lexical Loans.''  In
the first section, the author asserts that:
1. IE loanwords in Uralic are distributed across a wide area;
2. Reconstruction suggests that the words were borrowed
   into Uralic at a proto-stage for both languages;
3. ''in most cases the reflexes of the PU counterpart in
   later Uralic daughter languages do not show any internal
   irregularities which would point to borrowings transmitted
   from one secondary Uralic dialect/language to another''.

Thus, it is probable that earliest loan words into Uralic may have
been adopted when the speech area of Uralic did not exceed an area of
1,000,000 square kilometers.  The author then continues with examples
of the actual loanwords.  Interestingly, one word, which is listed, is
'vesi', or 'water' in Finnish, which may or may not have been
borrowed.  It would seem to have been asserted either way, although
the author's etymology of the word is pretty compelling.

The subsequent sections are titled: ''Indo-European Loanwords in
westerly Finno-Ugric''; ''Early Contacts with Pre-Aryan and Early
Proto Aryan''; and ''A Proto-Iranian Feature: Reflex of an Early
Depalatalization of the Common-Aryan Palatal Affricates.''  The
article concludes by making several assertions about the cultures and
chronologies where the borrowings and interactions occurred.  The
author refers to several colleagues' articles, which also appear in
the volume.

In ''The Neolithic Period of North-Western Siberia: The Question of
Southern Connections'', L. L. Kosinskaya investigates the age-old
question of southern connections to North-Western Siberia, evident,
for example, in the similarities of regional pottery styles from more
or less the same time periods.  Data is compared, mostly
archeological, and the locations of some of the early cultures in this
area are pinpointed.  Lastly, ethnolinguistic implications are

In her article, E. E. Kuz'mina elaborates on these ideas, with focus
on mythological as well as archeological data.  ''Contacts between
Finno-Ugric and Indo-Iranian speakers in the light of archeological,
linguistic, and mythological data'' begins with a discussion of
Finno-Ugric languages and known branches.  Secondly, linguistics and
archeology are discussed in the light of Indo-Iranian borrowings (not
surprisingly related to economic production, social relations and
religious beliefs).  Finally, some mythological comparisons are drawn
between Indo-Aryan/Indo- Iranian and some Finno-Ugric beliefs.  Of
notable mention are the ''horse-breeding together with mythical ideas
and rituals comparable to the Indian asvamedha'' adopted by many
Uralic peoples.

Alexander Lubotsky's article is titled ''The Indo-Iranian
Substratum,'' and in it, he discusses the study of loanwords as a
means of ''determining prehistoric cultural contacts and migrations'',
particularly as they apply to loanwords of Proto-Indo-Iranian,
''before it split into two branches.''  Apparently PII maintained
basically one dialect, up to the time the Indo-Aryans crossed the
mountains and lost contact with their roots. Mayrhofer's EWA corpus is
used as a base for the study.  In the second to last section parallels
are drawn between the loanword evidence and archeological evidence.
This places the Indo-Iranians as having moved from the Eurasian
steppes (Pit-Grave culture 3500-2500 BCE), first to the lower Volga
(Pottapovo culture 2500-1900 BCE) and finally to central Asia
(Andronovo culture, after 2200 BCE).  Lastly, one section discusses
the contact between Uralic and Indo-Iranian speakers.  An appendix at
the end of the article lists Indo-Iranian isolates.

In the most controversial and adversarial paper in the whole
collection, Janos Makkay addresses the problem of ''The Earliest
Proto-Indo-European-Proto-Uralic Contacts: An Upper Paleolithic
Model''.  He starts out by examining the problem of surplus scientific
production, and the asyncronicity between available archaeological vs.
linguistic research and methodology.  Fortunately, with the passing of
time, the problem appears to be getting better, or at least more,
treatment.  The author is able to find much contradicting and
inconsistent evidence in the other author's work in the volume. He
also historically surveys different prevailing theories.  Based on
these, he is able to provide evidence for quite a variation of early
Uralic homelands, each of which he discredits in turn.  One solution
he seems to be content with is the Paleolithic model of the late
Miklos Gabori presented at the nice congress in 1976.  In this theory,
the idea that the first Eastern Baltic population was PU/PFU speaking
runs into problems, because there were already presumably PIE speakers
residing there as early as Paleolithic times.  ''According to this
model, the separation of Indo-Iranian from the parental stock began
already during Upper Paleolithic times.''  Thus any PU/PIA contact
must have occurred after this separation, given the loanword stock.

''Uralics and Indo-Europeans: Problems of Time and Space'' is
J. P. Mallory's article.  Presumably, we are much more able to
determine Uralic homelands than Proto-Indo-European for a number of
reasons, which Mallory points out.  He outlines 5 assessment
principles for determining a solution to a linguistic homeland
problem: Temporal-spatial plausibility; Exclusion principle;
Relationship principle; Total distribution principle; and
Archaeological plausibility.

Temporal-Spatial plausibility presupposes the idea that ''Time and
place in homeland research are dependent variables, i.e., there is no
meaningful concept of one without the other.''  This sheds some light
on the sort of different ideas people have about the Uralic homeland:
either there seems to be ''Deep time depth with broad territory'' or
there is ''Shallow time depth with confined territory.''  A comparison
is then drawn between Uralic and Northern Athapaskan and Algonquian
languages, and their distributions are collectively superimposed on a
map of North America.

In subsequent sections, the author considers glottochronology (and
dismisses it); Estimation (a time estimation for earliest
differentiation among Uralic languages); and external contact dating
(as it applies to loanwords in Uralic).  The linguistic cultural date
of Uralic is considered in the context of Corded Ware (Fat'yanovo)
cultural distribution.  In this light, it is more easy to see where
Uralic contact to IE may have occurred, and also, which loanwords
could have been borrowed.  The Total Distribution Principle requires
that, as far as the Uralic homeland problem is concerned ''all the
pieces of the puzzle fit together with no exceptions. ''  The total
distribution principle is forfeit whenever one argues that one can fix
a segment of the protolanguage if one can anchor that segment to a
fixed, given area.  In what could lead to the next section, Milton
Nunez (1987) is quoted: ''It seems logical to assume that major
migrations should be reflected in the archaeological material.  But
there is no evidence for a major migration that could have brought a
Finnic language to Finland other than that connected with the
Mesolithic colonization of the country.  ''  Archaeological
plausibility is the idea that the archaeological record acts as a
proxy to the linguistic one, in the absence of the latter.

Vladimir Napol'skih's article ''Tocharisch-Uralische Berührungen:
Sprache und Archäologie'' focuses on Uralic borrowings from the
Tocharian language branch.  There is some discussion in the article
about the Uralic languages that had already diverged by the time of
borrowing: (Proto-Permic; Proto-Samoyedic; Proto-Ugric; and Finno-
Volgaic).  There is also some discussion about the cultural and
technological contexts in which the Tocharian borrowings fit
(metallurgy and horse breeding).  The Afanas'evo culture seems to
represent the target language for many of the borrowings.  In this
light, a number of etymologies (many for current words) are discussed.

Tapani Salminen's article is titled: ''The rise of the Finno-Ugric
Language Family.''  In it, he acknowledges that there are radically
different ideas about the source of Finno-Ugric languages.  As such,
there are a number of problems which need to be addressed by scholars
attempting to study and classify the languages:
1. How are Finno-Ugric Languages related to each other,
   and how are they classified?
2. What is the oldest center of expansion of the Finno-Ugric
3. When did the first contacts between Finno-Ugric and IE
   take place?
4. What are the possibilities for a distant relationship
   between Finno-Ugric and IE?

The sections in the article would seem to correspond to the various
problems and the way that Salminen addresses them.  The
''Classification'' deals with, firstly, the traditional binary
classification of the Uralic family.  Firstly, of course, came
Proto-Uralic, which would seem to have diverged into Proto-Saami,
Proto-Finnic, Proto-Mordvin, Proto-Mari, Proto-Permian,
Proto-Hungarian, Proto-Mansi, Proto-Khanty, and Proto-Samoyed, but
with several overlaps (e.g. Ugro-Samoyed, Finno-Volgaic) which may be
a real genetic units instead of actual transitional languages.  The
fulcrum here seems to be that ''whatever the value of the proposed
innovations is, the crucial thing is that they are very few; so few
that even their cumulative effect is not sufficient to make a lowest
level intermediate protolanguage (e.g. Proto-Finno-Saami(PFS))
different from a higher-level one (i.e. PU).''  There is some
subsequent discussion about which sort of model is actually best for
classifying languages taxonomically, a circle, wave or tree model. (It
would have been useful at this point to see some examples, rather than
a reference to another work).

In subsequent sections, the concept of Urheimat is discussed in the
context of shared lexicon; as Finno-Ugric languages are distributed
across chain-like across a single ecological zone, one can assume that
the proto-languages began somewhere close to the center of this zone
(where Mari, Udmurt, and Mordvin are spoken).  The section ''Indo-
European Contacts'' considers the idea that ''it does not matter much
if the primeval Finno-Ugric and Indo-European centers of expansion are
thought to have been located next to each other or not, because even
at the time of a relatively late first contact, the dialects within
the protolanguage continuums had not differentiated much.''  This
means, among other things, that some early language
(e.g. Proto-Samoyedic) had split from the proto-language at a point
before many (if not all) of the earliest contacts with PIE.  Also, the
Saami word for water has a cognate only in Khanty; this give more
weight to the idea that the so-called Uralic word for water is

In the next article, Pekka Sammalahti discusses Indo- European
Loanwords in Saami.  He examines the implications of early IE sources
and distributions.  Surprisingly, ''the concept of IE loanwords in
Saami is fairly new'' dating from the 60's or so.  The article
contains etymologies of 18 items.  Following this, the discussion
turns to those loanwords borrowed from IE idioms with PIE phonetic
traits.  There is an indexed table of PIE loanwords in Saami which
demonstrates a variety of attributes concerning the substratum,
including distribution and source.  Another table charts the
distribution of the oldest indigenous words in Saami, which is divided
by Uralic indigenous; Finno-Ugric indigenous; and Finno-Permic
indigenous.  content words in Saami.  Lastly, the relevant
implications of the distributions are discussed.

In the very last article, Peter Schrijver discusses ''Lost Languages
in Northern Europe.  By this, he's referring to ''the nature and
origin of words of non-Indo-European stock in northern Indo-European
languages (Germanic, Celtic) which have cognates in Lappish or
Finnish.''  As it happens, there has been some progress in recent
years regarding the identification of non-IE substratum languages.
The article acknowledges three such substratums: 1.) The ''Old
European Hydronomy'' language; 2.)the ''language of bird names'' as he
calls it; and 3.)The ''language of geminates'' which the author seems
the most concerned with. Apparently this substratum appears in
Germanic, Celtic and Balto-Slavic.  This language would appear to have
been localized in Northern Europe.  As I understand this article,
neither FU nor Uralic would appear to be the source of this language
of geminates.  However, there is apparently some limited evidence for
Finno-Ugric loanwords in Indo-European.

The book also contains a number of abstracts.  ''Ancient Metallurgy in
Northern Eurasia: On the problem of Contacts between the Indo-European
and Uralic-Speaking peoples'' discusses some of the archeological and
regional implications of this problem.  ''Chronology of the Volga-Oka
Valley; Neolithic and the Lyalovo Migrations'' looks at current
evidence to ascertain ''information on ethnic history and migration in
the Neolithic period.''  ''Migrations, Diffusion and Uninterrupted
Development in the Stone Age of the Forest Zone of Eastern Europe:
Some remarks'' considers these ideas in the context of post- glacial
northeast Europe.  Lastly, ''The Problem of Interaction of Cultural
Traditions in the Bronze Age in Central Russia (Volga-Oka Basin)''
examines ''an analysis of the ethno cultural situation in central
Russia in the Bronze Age.''

In my opinion the book without a doubt illustrates the range of ideas
and methodologies available to the modern scholar in this area.  While
this book could never be read as even anything resembling a popular
introduction to the subject, it sheds a clear and resonant light on
just how divided the camps are when considering such topics as the
overlap between archaeological and linguistic evidence.  While I was
able, in this review, to highlight in summary, some articles
containing points in the discussion I personally found interesting,
writing this review was extremely difficult in the sense that it
wasn't always so easy to summarize some of the articles, as some of
the main points in them were not always so clear.

Regardless of these shortcomings, I'd strongly recommend this book to
any scholar who is considering a pursuit in these areas.


John Hammink holds a B.Sc. in Linguistics from Eastern Michigan
University.  He has been working for the last six years in the areas
of data-security software validation.  His work focus has ranged from
test automation and document inspections which apply corpus-linguistic
methodologies.  Most recently, he has developed and taught seminars to
address the problem of linguistic ambiguity in software requirements
and the downstream cost effects these have.  His other interests
include linguistic evolution and the Uralic Language family.


If you buy this book please tell the publisher or author
that you saw it reviewed on the LINGUIST list.

LINGUIST List: Vol-14-934

More information about the Linguist mailing list