Language of the Olmecs

[ECOLING at aol.com]

Mike Gaby asked:

>Also anyone know the form of Mixe-Zoquean allegedly spoken by "Olmecs"?

This is of course a very difficult question, since the Olmecs
flourished around 1500 BC to 400 BC or so (depending on who one asks).
It may not be Mixe-Zoquean at all.

According to John Justeson and Terrence Kaufman,
it is Proto-Zoque, by the following reasoning:

a)  Loan words in Mesoamerica are from Mixe-Zoquean into other languages,
     including high-culture words
b) The Olmecs were the originators of Mesoamerican high culture.
c)  Therefore the Olmecs must have spoken a Mixe-Zoquean language

And they further claim


d)  The writing system of La Mojarra must represent the Epi-Olmec people.
     Given its date and location (Veracruz, 2nd century AD by its long
counts),
     and the lack of evidence for extensive migrations during the relevant
     time frames, this writing must be interpreted
     as the writing of the successors to the Olmecs, "Epi-Olmecs".

e)  The writing system of La Mojarra is now in large part deciphered,
     and is a Mixe-Zoquean language, specifically proto-Zoquean.

*****

This line of reasoning is plausible in its first two steps
(though we have few sufficiently knowledgeable specialists
other than Kaufman who could
attempt to argue against any of these early steps in the reasoning).
The reasoning is increasingly fragile in later steps,
for example because the Olmec might have transmitted some
high culture which originated with others, or the situation might
in other ways have been more complex.

*****

In particular, the claim in (e) is unjustified.
The history of decipherment is full of claims of decipherment
which turn out to be unsubstantiated.

A claim is not a decipherment.
In particular:

1.  Kaufman "knew" the writing of La Mojarra had to be Epi-Olmec,
a conclusion he drew from steps (a,b,c,d).  He "knew" this before
ever examining the writing system itself.

2.  Kaufman and Justeson did not actually consider alternatives,
in the sense of exploring "what if" the writing represented a language
of some family other than the Mixe-Zoquean.  So their conclusion
is really the same as the assumption they began with, or rather,
should be more accurately formulated as:

Under the assumption that the language belongs to the Mixe-Zoquean
family, it is most likely a form of Proto-Zoquean.

Once we make the assumption explicit (and Kaufman and Justeson
did not hide it, they merely did not note that it was an assumption
to which any alternatives could be conceived or should be...),
then the conclusion stands clearly as what it is.

*****

3.  There is a lack of the kind of "many things suddenly falling into place,
once the answer was discovered" which is usually the hallmark of a good
decipherment.  Rather, each bit of new text "interpreted" as proto-Zoquean
which contained new glyphs requires new subsidiary hypotheses to
support an interpretation.

4.  The "translations" provided by Justeson and Kaufman for parts of
the La Mojarra text, many still only in verbal presentations rather than in
publications, are quite different from most other texts in Mesoamerica,
in several respects (speaking of "folding cloth" and other things).
This by itself might suggest that forcing a Mixe-Zoquean interpretation
onto the text has led to contradictions or dead-ends.
(Of course we cannot know in advance what the writers of the
La Mojarra text chose to write about -- it is simply that if on our first
hypotheses at interpretation we get something rather odd, it might suggest
the hypotheses are not exactly on target.)

4.  Interpretations of the same text as some language related to the Mayan
family
work moderately well also.  This is definitely NOT to argue that
La Mojarra represented a Mayan language, simply to point out that there are
alternatives.  In at least one case, an interpretation as Mayan works better
than the interpretation Justeson and Kaufman propose as proto-Zoquean,
because there is a homonym motivation within Mayan.
Many sentence structures are the same for many languages and language
families within Meso-America, so they cannot decide.  (In fact, Kaufman
and Justeson were forced by some of their other assumptions to conclude
that the La Mojarra "proto-Zoquean" had a word order different from that
of other Zoquean languages -- possible, but not the most straightforward.)

5.  There are some in the Mesoamerican field who think
the resemblances between Olmec iconography and Mayan iconography
are such that the latter may derive in large part from the former.
That does not demonstrate a linguistic relation, but does not exclude it
either.

Conclusion:  The language is not yet known.  Two proposals have been explored,
a Mixe-Zoquean one and a Mayan one.  The La Mojarra text might represent
either of these or might represent a language from some family other than
these two.
There is no avalanch of things falling-into-place as a result of any set of
hypotheses so far presented.

Lloyd Anderson
Ecological Linguistics

*****

Four publications (others are listed in Anderson 1999):

Campbell, Lyle and Terrence Kaufman. 1976.
A Linguistic Look at the Olmecs.
American Antiquity 41:80-89
     (the presentation of the claim in (a) above).

Justeson, John and Terrence Kaufman. 1993.
A Decipherment of Epi-Olmec Hieroglyphic Writing.
Science 259:1703-1711
     (presentation of the Proto-Zoquean hypothesis,
      claims of decipherment with many specifics)

Anderson, Lloyd. 1999.
The Writing System of La Mojarra, 3rd edition.
Ecological Linguistics
     (presentation of structural analyses independent of choice of language,
      and of both Mayan and Proto-Zoquean interepretations in parallel
      and in contrast.  Currently the most comprehensive work available.)
(PO Box 15156, Washington, DC 20003.
Price within the USA $40,
Price to Europe by book rate $52.)

Houston, Stephen. 1996.
[Review of 2nd Edition of The Writing System of La Mojarra]
International Journal of American Linguistics 62 no.4 pp.429-431.