IE and Uralic pronouns

[ECOLING at aol.com]

I have been present when solid competent linguists admitted that
the field is "not ready" to deal with the pronoun patterns
which are the subject of recent discussion.

(For a completely traditional approach to some pronoun matches,
please see near the end of this message.)

In other words, it is likely that something genetic is going on in
these pronouns,
our sound-symbolism approach is not adequate to the data
distributions, and our tools for historical reconstruction can't
handle it either.  So what do we do?  Hide our heads!

Shame.

Rather appear to be knowing the answers,
and hide the questions we don't know the answers to?
Or declare them off-limits?
In a business firm, that would be a recipe
for near-immediate bankruptcy,
failing to act positively and strongly in moving towards the future.

Better to ADMIT to students that we need more powerful
tools for penetrating the noise of centuries and even millennia,
to set more and more students to this task, to studying how deeply,
and against which kinds of noise, each tool we have can
penetrate (estimates in all cases are fine), and to estimate
in as many situations as possible how much residue should
still be detectable despite the noise of historical change,
WITHOUT the assumptions that the sample used as controls
ARE IN FACT UNRELATED.  That is the fallacy in all such
approaches I have read.  Because if the supposed controls
ARE related, even very distantly, it may distort our estimates
of what background noise is.

Rather we need GRAPHS of the degree of noise present,
and our ability to penetrate it,
across the full range of cases we can estimate time depths for,
then extrapolate beyond that to those cases like Afro-Asiatic
where the pronoun system, for example, preserves
"distinctive oppositions in distantly related languages"
(the title of a Gene Schramm paper, if I remember).
In that case, there is a known relationship, though
not well known, it is a distant one, the pronoun sound
correspondences do not follow the normal sound laws
for these families, yet they ARE related.  Because there
is a pattern with its own laws.  The palatal / labial
opposition which is in the consonants in one family
is in the vowels in the other.  I don't remember details
of the data, but approximately this:

hi vs. hu  in one family (NW Semitic?)
sa vs. fa   in another family (Egyptian?)

Now if our standard tools for historical reconstruction
cannot deal with that, then we need to extend those tools
slightly, a little bit at a time, go back and test the change
of tools against the various things we know the answers
to, and see whether the new tools manage to extract a
slightly cleaner set of data that makes a known relationship
clearer.  If it does, then we may perhaps apply that newly
refined tool to cases where we do NOT know the answer
or only suspect it, and see what happens.

Another example of this which came to my attention in the
past year or two is this, which I will pose as a question.
Why is the second comparison (b) better than the first (a)
if we are looking for potential cognates?

(a)

sepo  in one language
teka  in the second language

(b)

sepa   in one language
teko   in the second language

I don't think most historical linguists have an IMMEDIATE
recognition that these are two wildly different proposals for
cognates.  That means we are not making the best possible
use of our tools.  (The conclusion is not a certain one,
just like any other historical inference, but it is a reasonable
one that (b) is a better comparison thatn (a).)

We are never dealing with "clean" or "dirty" data.
We are always dealing with "slightly more clean"
or "slightly more dirty" data.  If slightly cleaner
data makes a language relationship look more solid,
that is perhaps an indication (not a proof) that we may
be on the track of a real relationship.

*****

If you want to try to use traditional tools on Eurasian pronouns,
here is an example from my paper
Grammatical-Meaning Universals and Proto-Language
Reconstruction, or: "Proto-World Now", in Papers of the
Chicago Linguistics Society, 1975.
p.15 of the paper, or Fig.(26).
I am sure that paper contains a lot of wrong things and some
right things.  I only expected to find truth statistically,
not certainly.  A second example follows below.

Here is what I was able at that time to find for
a standard approach to reconstructing IE-Uralic pronoun systems.

Of course borrowing or contact may have created a false
temptation here, but such an escape from the obvious interpretation
of the data would itself require considerable work to prove:

I was discussing a relic IE alternation /m/ in plural vs. /w/ in dual.
It is matched in Uralic, and an -n- "with,and" is present in the plural,
which I hypothesized caused the assimilation.

-va-s   dual   1st person  Sanskrit
-ma-s   plural

-tha-s   dual   2nd person  Sanskrit
-tha(-na)  plural

Now compare Ziryene (Uralic),
EARLY plural forms:

-m-ny-m  1st person

-d-ny-d   2nd person.

Selkup  had these forms of the first person:

-mi-y    dual
-my-n   plural

So my conclusions:

Selkup leveled the person-marker, while
IE leveled the markers of dual/plural, and kept the originally
     allophonic variants of the person.
Uralic shows in EARLY Ziryene the structure of the paradigm
     from which this could have arisen.

I'm not sure one could get more traditional in approach.

*****

Here is the second item in that 1975 paper which really calls
out for work.  The rotation of the vowel space makes these more
than mere identities, and suggests hypotheses for further work.
(ng   for the velar nasal phoneme)

ple:-nus 'full' Latin, IE
*pling     'full', Tibeto-Burman

pla:-nus  'flat'  Latin, IE
*pleng   'straight' Tibeto-Burman

Lloyd Anderson