Relative chronology and absolute certainty

[Sean Crist]

On Tue, 14 Sep 1999 X99Lynx at aol.com wrote:

> Now take a look at the logic here.  ki > ci is widely attested.  ci > ki is
> much rarer.  Accepting that, then the proto language MUST HAVE HAD a certain
> reconstruction.

> But we know that's not true.  The rarity of ci > ki only affects probabilty.
> It doesn't make the reconstruction absolutely certain at all.  In fact, it
> makes absolute certainty an absolute error.

> So why is such absolute terminology being used?  Putting aside that this fact
> about the world's languages' habits re 'palatalization' wasn't part of the
> original hypothesis, we can still ask a very important question.

> Just how widely attested is ci > ki and just how rare is ki > ci?  Especially
> in the context of discussions about algorithms and statistics, wouldn't it be
> appropriate to talk about what the occurences are here?  How often has ki >
> ci happened?  How often has ci > ki happened?  Is there a count of these
> events?

> If there was ever a place where statistics in linguistics would be helpful,
> this is a good one.  Because this is exactly the kind of assertion that
> really needs some back-up before it should be credibly accepted as the basis
> for accepting this reconstruction.

In principle, this would not be a bad thing, but I think it is impossible
in practice.

The problem has to do with counting the events.  Would you include a rule
in the count if it were /qi/ > /ci/ (voiceless uvular stop to voiceless
palatal stop)?  What about /qi/ > /ki/ (also a fronting induced by a front
vowel, but not one that results in a palatal)?  What about /kwi/ > /ti/
(which happened in Greek)?  What if there's a concomitant change of the
manner of articulation, e.g. stop to affricate, as happened in Old
English?

Are you going to count any old case where a front vowel has dragged a
dorsal consonant forward in any way, or are you going to count it only
when it's really /ki/ > /ci/?  Does the sound change count if it has not
yet become contrastive?  Would we count modern English, given that the /k/
of /ki/ is articulated somewhat further forward than the /k/ of /ku/?
If a change started in a more limited set of cases and then spread to
others, do we count it as one rule or two?

Also bear in mind that the documentation of the world's languages varies
wildly in its quality.  It's not uncommon to run up against a case where
the only grammar of a now-dead language was written in the 1920's by an
anthropologist with little linguistic training, leaving the modern
linguist to scratch his/her head and try to extrapolate as well as
possible.  Two linguists might well argue over exactly what the grammar is
trying to say- and in this case, might argue whether a ki > ci rule has
happened or not, thus leading to differences of opinion on your totals.

> (I won't go into how selective this external evidence is.  If we are going to
> bring the pattern of the world's languages into this, why are we excluding
> the possibility that /c/ in the hypothetical might represent loans - like
> intervocalic /s/'s in Latin - or other more intricate pathways.

Given that there is a morphologically conditioned /k/ ~ /c/ alternation in
our hypothetical, I don't see how this situation could have arisen thru
loan words.

> But we should not accept this reasoning.  Not if it is used to justify
> another absolute statement.  Because the reasoning only perhaps makes it
> likely.  It tells us nothing that justifies such certainty.

Once again, you're bringing in that word 'absolute'.  This was already
hashed over.

In science, no statement is absolute.  We have to always be prepared to
revise what we believe if the evidence demands it.  When a scientist make
an assertion, it is on this shared understanding.

However, there are cases where the evidence speaks so plainly to a
particular conclusion that it would be perverse not to treat it as true. I
think that the analysis I've given for our hypothetical language family is
such a case; every bit of evidence within the hypothetical points to it
being correct. If you gave this as a toy problem to a dozen competent
historical linguists, I'd be quite surprised if anyone differed in the
basic analysis of the facts.  (You'd get variation in how the facts are
formally modelled, but I'm sure they'd all agree on the relative
chronology, for example.)

> The new hypothetical Sean is referring to is:

> <<...   Language A
>         takutu          'I run'
>         tacid           'you run'
>         takil           'they run'>>

> It amazed me to learn that without "any evidence for Language B, we could
> still
> correctly and unambiguously reconstruct Proto-AB purely by performing
> Internal Reconstruction" on the evidence given above.  I presume this has
> something to do with some rule about the phonetic formation of second persons
> or something gleaned from the world's languages that allows us to recreate
> the parent with such absolute certainty.  This is an important rule and I
> wish I knew it.

It's noting special about the second person; nor in any direct way am I
extrapolating from the other languages of the world. It's simply an
application of first-year undergraduate phonology to a particularly clear
case.

> Sean continues:
> <<But if this still isn't enough for you, suppose that Proto-AB contrasted
> */k/ and */c/; thus, Proto-AB contrasted the series *ki *ke *ka *ko *ku
> against *ci *ce *ca *co *cu....In the case of monomorphemic words, for
> example, there'd be no way to undo the merger....>>

> Once again, there would be no need to undo a merger.   This is a two step
> process, between parent and daughter.  No steps left for unmerging.

You're just plain not understanding.  Consider the following words; let's
assume they're monomorphemic.

	Language A	Language B
	cim		cim		"more"
	cibi		kibi		"before"

Suppose there hadn't been a */c/ - */k/ contrast in Proto-AB; it was all
*/k/.  Okay, them how did Language B develop the contrast?  There's no
conditioning environment to tell you where you get /c/ and where you get
/k/ in Language B, so you're forced to assume a sporadic split.  This is
definitely the wrong way to go; the right thing to do is to assume that
the contrast existed in Proto-AB, and Language A has had a merger. I just
don't know how I can make this any clearer.

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