When a Parent Becomes a Daughter

Sean Crist kurisuto at unagi.cis.upenn.edu
Tue Dec 21 16:27:53 UTC 1999

On Mon, 1 Nov 1999 X99Lynx at aol.com wrote:

> You use the term "daughter" above.  Whatever vague boundaries you used to
> decide when a language BECOMES a "daughter" - think of why the other
> remaining part of the parent also must at the same time become "a daughter?"
> Whatever caused you to call the two languages "daughters", consider what
> makes them both BECOME "daughters" and makes the parent disappear?  So you
> can say that a parent doesn't coexist with its daughter?

This is a bit like asking what happened to the child who you used to be.
Did that child "disappear" as you gradually became an adult?  Maybe
there's a sense in which that child disappeared, but I'd rather say that
that child gradually changed into something other than what it was before.
This isn't a perfect analogy since individual humans don't fork into
multiple individuals of the same age, but it illustrates the point that
thinking of the parent language "disappearing" isn't really the right way
to think of the matter.

> Take the language that you describe an "ancestor" by whatever definition you
> use above.  Why are you assuming it also has to change status and become one
> of the "daughters" just when another daughter emerges?  And at the very same
> time (so that you can say there's no period of co-existence between parent
> and daughter?)

> "An ancestral LANGUAGE cannot co-exist with" a daughter.

> Isn't this a reification?   The only reason a parent can't co-exist with a
> daughter is because you automatically change it into a daughter when there's
> another daughter branching off.  Aren't you creating another "daughter"
> unnecessarily?

> If you would have been satisfied with a single "ancestor" (by whatever your
> definition is above) as a single language - if the branch off had not
> happened - why are you turning that "ancestor" into a new language just
> because a daughter branches off?

> Isn't that reification?  Aren't you unnecessarily creating a new "daughter
> language" (however you mean it above) when nothing more than a part of it has
> broken off?

I think it's important to remember that while it's often useful to treat a
language as a singular entity, what actually exists are multiple copies of
more or less the same information in the minds of a community of humans.
The copies are never perfectly identical, however: even in a small,
relatively homogenous community of speakers, there are always variations
from individual to individual.  Further, the linguistic norm within any
community always changes over time.

A forking in a language follows from a forking in the community of
speakers, typically because of migration leading to geographical
separation.  When the original single group splits into two groups who
wander off from each other, the copies of the language information in
the minds of the members of both groups start out basically identical.
Over time, however, the norms in the two communities drift in separate
directions, since the two groups are no longer in contact.

In this post and in others, you've conceived of language branching as a
main trunk which continues, and a daughter which separates off of this
trunk and goes its own way (e.g., when you use the phrase "...a daughter
branches off"). I'd argue that this is not looking at things the right
way.  There are only forkings.  It's not possible for just one daughter to
branch off; if there's been a forking, then you have two daughters.

Perhaps some of the conceptual difficulty here comes from taking the
biological metaphor of "parent" and "daughter" too literally.  In biology,
we are actually talking about separate individuals.  I did not gradually
develop from my mother or father by starting out as an identical copy of
one of those individuals and gradually changing into someone else. In
language, however, what we're talking about are copies of information
which gradually change separately and become progressively more different
from each other.

  \/ __ __    _\_     --Sean Crist  (kurisuto at unagi.cis.upenn.edu)
 ---  |  |    \ /     http://www.ling.upenn.edu/~kurisuto/
  _| ,| ,|   -----
  _| ,| ,|    [_]
   |  |  |    [_]

More information about the Indo-european mailing list