LINGUIST List 5.1128

Sun 16 Oct 1994

Disc: Comparative method

Editor for this issue: <>

Directory

1. Jacques Guy, Re: 5.1116 Comparative Method
2. "Dr R.M. Blench", Afroasiatic
3. , loss of vocabulary

Message 1: Re: 5.1116 Comparative Method

 Alexis Manaster:
>
> (a) It is not the same thing to say (as Jacques does) that there
> may be some in principle limit on the CM although we have no idea
> what it is, and saying (as Nichols, Bender, Kaufman, etc. have done)
> that there is a very shallow time limit of somewhere between 7000
> and 10000 years.

I took my calculator and asked myself: if any two languages have
been as conservative since the year dot as Bergsland and Vogt
have observed, say, Armenian, to have been for the past 1500
years, or Icelandic for the past 1000, what is the time limit?
About 40,000 years. They will still have 20% vocabulary in
common, which ought to be enough for reliable reconstruction,
unless, of course, phonological evolution has obscured everything.
But what if they have evolved at the rate of Eastern Greenlandic?
4,000 years at the most, after which they would have only 11%.
And the rate of Muyuw? (30% replacement in one generation)
About 100 years (yes: ONE HUNDRED YEARS). Need I say more?

Can I say that I have been kindly forwarded some of the recent debate on
the time-depths, especially of Afroasiatic, and it has spurred me to join
the list to add my pennyworth.

I have been working intensively on Afroasiatic during the last 2 years and
I am absolutely mystified by all these bold statements about its time
depth. I have been in touch with most of the scholars who are proposing
either hypotheses about its internal structure or reconstructions and I
have no idea at all where these dates come from. Who says Afroasiatic is
7,8 or 10,000 years old?

Debates on Afroasiatic are crystallised into those who subscribe to what
may be called Semitocentric theories which have it that AA somehow
originated in the Near East, and the (probably more numerous) school who
consider an African homeland, presumably Ethiopia, more likely.
Dates of 7-10K might come from the Semitocentrics, as this adds something
to the known dates of written texts.
The other point of view essentially comes from looking at Omotic which is
deeply internally divided and which takes a great deal of hard thinking
to establish how it fits together. Luckily, new data coming from
Ethiopia, especially, from the SIL surveys and the like is beginning to
give us the tools we need.
However you take it, such a deep, complex branch as Omotic must be
substantially older than Semitic, which is relatively tightly
inter-related and looks rather 'young' as it were. But of course how much
older Omotic is is very hard to guess at present and I don't know of any
Omoticists who have laid themselves on the line with this one.

Just a plea for slightly fewer wild unsupported estimates of time-depth.
Of course we also cannot know whether the CM will work with Omotic, because
the data available are still too scanty.

Roger Blench
RMB5cam.ac.uk

 Something that came up in the recent discussion of how far back in
 time historical reconstruction can be used, is the notion of the
 gradual loss of vocabulary items. (The idea that this loss happened
 at a constant rate formed the basis of glottochronology, but the rate
 of loss is an issue that I don't want to get into!)

 I have no doubt that languages lose lexical items over time. Reading
 Shakespeare or Chaucer will convince one of that, even if you factor
 out the forms that have suffered enough changes so as to be difficult
 to recognize. What I would like to know is, why should a language
 lose words?

 The only answer I can think of is that a language gains words from
 contact with some other language, words which have more or less the
 same meaning as native words, and that for some reason the loan words
 replace the native ones. (Rather like the introduction of placental
 mammals to Australia resulted in the replacement of marsupial mammals,
 although there is presumably no evidence in the linguistic world of
 better fitness on the part of one word or the other!) Apart from such
 replacement by loan words, a loss of lexical items would result in a
 gradual reduction in the size of the vocabulary, something which
 doesn't appear to happen (or is that the reason for the drop in the
 last 30 years in SAT scores? :-)). (I am assuming that native words
 don't often get replaced by newly coined words.)

 If vocabulary loss results from contact, would that imply that
 relatively isolated language groups (if there be such over a long
 period of time) tend to retain original vocabulary more than do groups
 with a great deal of contact? Immigrant groups would presumably be at
 the forefront of vocabulary loss. (I'm talking here of groups where
 there are still native speakers, not those where a later generation
 has adopted the local language and doesn't become fluent in their
 parents' language.)

 Or is there some other explanation for vocabulary loss?