On 21/05/2019 21:59, Jim Thain wrote:
> The only part of your answer I have real problems with is: "most
> linguists are aware that only writing tells us anything about
> language."
> In which case the comparative method is bunk?

To which oOn 22/05/2019 10:32, BPJ replied: "I meant 'in an
archaeological context', and should have said so explicitly obviously."
Thus confirming what I suspected, BPJ was _not_ debunking the
comparative method.

But, of course, the comparative method for reconstructing PIE and other
similar proto-languages relies upon written evidence.  Therfore, the
reconstruction is only as reliable, at best, as the written evidence
allows.  As new discoveries come along, the reconstruction gets modified.

When I was a youngster the reconstructed PIE generally given at the time
was somewhat different from current model(s).  With more evidence from
the Anatolian group, the laryngeal theory/theories developed and our
understanding changed and deepened.

But it has been said on this list in the past that PIE,
Proto-Hamito-Simitic and so forth are _conlangs_ - at best they give a
general idea of what the language may have been like.  The only hard
evidence are the written records.

In the case of modern spoken languages we now have the additional
evidence of recoded speech; but, in the cases where a language has a
written record that gives us some idea of very recent changes.

The comparative method has limitation.  If e.g. all written Classical
Latin texts had been lost, there is no way the language could have been
recoonstructed.  The Romance languages allow us to posit a
reconstruction of Vulgar Latin, but that language differed in important
respects from the Classical language; our knowledge of the latter
depends solely on written records.

Consider the difficulties in (re)construction Cornish for the Cornish
revival.  It led to four or five rival Cornishes, and only now is a
common form emerging.

In short, the comparative method has limitations and depends upon
written records of descendant languages.  Nor, alas, do written records
give as much information as we would like.  If the writing is more or
less phonemic as, e.g. ancient Greek, we can get only a general idea of
its phonemics - there are still uncertainties, e.g. how word pitch
accent worked and, more than once in the past, there has been
disagreement over the Classical (5th/4th cent. BC) pronunciation of ΞΆ

But in the case of ancient Egyptian which did have two written forms,
hieroglyphics and hieratic, extending over several centuries we have
less information.  We can get a more less general idea of the
_consonants_ of the language, but we learn nothing of its vowels. We
have use comparative methods to supply this information as far as we
can.   When it comes to largely logographic scripts the effort to
recover the phonology of the language becomes much greater.

> I recently watched a program (Nova: The first Horse Warriors), in
> which they linked PIE to a culture they call the Yaminaya. I would
> call that linking tentative, we probably don't know enough about
> them to be sure if they are the tribe, or collection of tribes, that
> eventually overran most of Europe. However these people did come
> from the area that 'most' linguists agree was the IE 'homeland'. We
> can't know for sure wether that is absolutely true, but it is a
> plausible theory.

What is called circumstantial evidence; and in the past when such
circumstantial evidence has been taken as hard evidence much mischief
has resulted.

> The trouble comes with linking the right people to the right
> language.

It is. Also the notion that one language (or rather group of dialects,
for there was no standardized PIE or any other proto=language)
necessarily equals one people is a dangerous concept as the theories of
the Nazis showed only too well.  It is clear from the modern world that
languages spread to differing peoples not only through conquest but
through trade and diffusion.

> It is, in the absence of writing very difficult, but writing came
> about fairly recently. So we break things up into History, what
> happened after writing was invented, and Pre-History, which is much
> more ephemeral, even with hard archaeological evidence.
> Probably one of those things we will never be entirely sure of short
> of time travel unfortunately.

We can never be certain of pre-history; not even all written records
have been fully deciphered.  Some, like Etruscan, are imperfectly known,
and others like the Indus Valley Script still cannot be read.