andera 'woman' Celtic ?

[roslyn frank]

Hi Steve (and IEists)

In a message dated Tue, 30 Mar 1999  18:53:09 EST Steve Long wrote:

>In a message dated 3/26/99 04:30:24 AM, xdelamarre at siol.net wrote:

><<I confess that the celticity of _*andera_ 'woman' (found in French
>dialects) is extremely uncertain>>

{SL]

>I've been confused as to why none of the replies mention a possible Greek
>connection.  I'm may be missing something very obvious again.  Please
>forgive me ahead of time.

>The old Celtic-Greek contact point in the south of France is well
>established.  "Aner, andr-" is man (versus female, as opposed to
>"anthropos" - man versus beast.)  "Androo", to become a man or raise to be
>a man, in Classical Greek was sometimes generalized to the feminine.
>"Andris" in later Greek I believe came to be used as woman.  And in such
>terms as "anandria" (want of manhood, eunuch, unmarried woman) the term was
>extended beyond the male.  Couldn't this be the Greek word with a Romance
>feminine ending dropped on it?  What have I missed?

[RF]
Over the years the relationship between the Euskeric form <andere> which
is <andre-> in composition has been subjected to a number of
interpretations, the most recent, to my knowledge, being Vennemann's
(1998) discussion of it which appeared in his article "Andromeda and the
Apples of the Hesperides". This was published in the Proceedings of the
Ninth Annual UCLA Indo-European Conference" JIES Monograph Series No.
28. I would refer you to the section appearing on pp. 12-17.

He argues that we are dealing with a item that should be classed as
"Vasconic". According to Vennemann's scenario "Vasconic" is one of three
language families spoken in post-Ice Age Western Europe (ca. 8000 BC).
Hence, in his simulation, the item passed from the lexicon of Vasconic
to that of Euskera. It was also passed on from Vasconic to the
lexicon(s) of West Indo-European languages. Thus, we are not dealing
with a model in which an Euskeric item is posited as the source for the
apparent reflexes of it in IE languages, rather there is a higher node
that facilitates the transfer. It is from that node that the source of
the reflexes found in the lexicons of both Euskera and IE should be
sought, again according to Vennemann's model.

In reference to the section of Vennemann's article cited above (pp.
12-17), first I should state that I disagree with his interpretation of
the ending of the Aquitanian items, <anderexo> and <anderexso>
(Gorrochategi 1984: 130-132). He argues that the endings <-xo> and
<-xso> do not represent the common diminutive suffix in Euskera, namely,
<-txo>, while Larry argues that they do. I fully side with Larry: the
suffix is unremarkable from the point of view of Euskera.

Having said that, I would suggest that the overall thrust of Vennemann's
argument is worthy of consideration, namely, a model that posits a
higher node and consequently, an older source for the IE and Euskeric
items alike. In such a simulation of events, one could argue that
originally the item was used in reference to a "woman" and that over
time the term was generalized to refer to "human, person". In the final
stages, as the positions held by such persons came to be "masculinized",
i.e., as the positions of responsibility and authority (beyond the realm
of the household) became more significant and the duties associated with
them were fulfilled primarily or even exclusively by males, the
expression's referentiality would have changed, i.e., its referent would
have become a "masculinized."  Here I am speaking of positions in the
"public sphere".

This line of argument has another advantage in that <andere> does not
mean simply "woman" in Euskera, rather it is, strictly speaking, a form
of address that carries the force (more or less) of "Lady, Mistress" in
English. Its counterpart is <jaun>, as I believe Larry pointed out quite
clearly in another message. For example, when called by her name, a
schoolteacher is referred to as <so-and-so andrea> "(the) Lady/Mistress
so-and-so". I would note that as a title <andre-> is regularly used to
refer to post-pubescent females and there is some indication that it was
once reserved for women who had born children, but that distinction is
not entirely clear. Hence, if, in our model, we posit that initially the
expression functioned primarily as a form of address, the shift in
meaning would have been even simpler.

If I am not mistaken I recall reading somewhere that the translation (or
at least one of the translations of) proposed for <aner> and <andros>
(IE) was "mensch" which would imply a wider spectrum of referentiality.
Sorry I can't give a source.

In addition, there is another piece of evidence that could be brought to
bear in such a modeling of events. In Euskal Herria the importance of
the role played by the woman in the household, particularly the elder
female of the family, is quite well established. Her title was
<etxekoandre-> from <etxe-> "house", <-ko> "of". Literally translated,
it means "lady-of-(the) house" and is pronounced roughly as
<echekoandre>. Again the expression is a totally normal one in Euskera
today.

Some years back while I was doing archival work on Basque law codes, I
came across a document from Navarre written in Spanish (Navarrese?). As
I recall it was from about the 15th century. There was a section it
outlining the duties that were assigned to a group of important males,
the <chandros> of the village. Further research would be needed to
determine whether at the time the document was composed, Euskera was
still being spoken in the zone where the segment in question was
written. However, it is a region in which Euskera had been spoken in
centuries past.

These individuals called <chandros> (i.e., sing. <chandro>) are referred
to throughout the text in the masculine so there is no question about
their gender. Moreover, from the duties assigned, it is likely that the
group as a whole was composed exclusively of men. In short, in this
concrete case there is little question about the ultimate female
referentiality of the "title" of <chandro>, its derivation from
<etxekoandre>. Yet it is likely that those (monolingual)
Spanish-speakers who wrote the code in question had little or no idea
that the term's original referent was to the "lady-of-the-house."

Returning to Vennemann's thesis that <andrere>/<andre-> was passed on
both Euskera and West Indo-European languages from the earlier so-called
"Vasconic" family of languages, the simulation (if it were fully
developed, that is, obviously not as it's being presented by me here,
namely, in a totally schematic fashion) would elaborate on the evidence
available for the type of replacement outlined above. It would emphasize
the increasing importance, over long periods of time, assigned to roles
played outside the immediate household environment, e.g., outside the
socio-political structures of household, by males.

For example, for the remote period in question initially we would be
talking of small-scale structures characteristic of societal units based
on an economy of transhumanic pastoralism and primitive agriculture,
supplemented by hunting, fishing and gathering activities. In our
simulation, such a society would not have had an elaborately organized
"public sphere", i.e., one that was totally separate from the familial
one.

In the model discussed here, the anecdote cited above concerning
<chandro> serves as an example of what might have happened, naturally,
on a much wider scale, when *<andere>/*<andre-> (or a phonologically
similar prototype) as a title of respect for female elders (and mature
females) came to be used with post-pubescent males, also. And
eventually, again according to this simulation of events, the expression
ended up with an altered referent because of the societal changes taking
place, i.e., the increasing importance given to public positions that
were held primarily or perhaps exclusively by males as well as the
emergence of a more complex organizational structure for the society as
a whole.

So, Steve, in terms of what you might have missed, maybe it's the
possibility of ladies becoming gentlemen. Keep in mind the above is
merely a modeling of events in prehistory. Yours is another. It draws on
similar data but organizes it differently.

For instance, if I am reading your remarks correctly, your model asserts
that the original referent of the item was a "male human being, man" and
that this meaning was later generalized to mean "woman" given that there
appears to be evidence (cited by you) in Greek for the word's field of
referentiality to contain "woman" also. But that would be a later
development in your simulation. In the one cited above, the reverse
would be true.

A couple of final comments and queries for you (and others).

In your simulation, there is an element that is lacking, in my opinion:
a mechanism to explain why it's the secondary meaning of the Gk. word,
rather than the primary one that we find showing up in Euskera and
Celtic. And on that point I would note that in your modeling of events
you speak of a "contact point" in the south of France, which I assume
refers to the geographical location where the Gk. term passed into
Celtic (as well as being the general vicinity in which
Aquitanian/Euskera was spoken).

If I am reading your simulation correctly, that would mean that the Gk.
term's meaning --at that point in time when the linguistic contacts took
place and the expression passed into Celtic (and Euskera?)-- must have
been to the female of the human species. Right? Would it be possible for
you to provide your model with more specificity in terms of the time
frame that we might be talking about for this "contact period". Also,
what is the time-depth that should be assigned to the Gk. reflexes
according to your model?

And for others on the list, does anyone know of possible reflexes of
<chandro> in Romance, e.g., in Occitan, Aragonese or Navarrese?

Best regards,
Roz Frank

Contribution # 3.
April 2, 1999
e-mail: roz-frank at uiowa.edu
[currently on leave in Panama]