Re: Payngo = Fist & Austronesian Numerals

On 22 Dec, 05:07, "benli...@xxxxxxxxxx" <benli...@xxxxxxxxxx> wrote:
Sorry, sent previous post before completing.

On Dec 22, 3:17 am, richard01 <richardparke...@xxxxxxxxx> wrote:

On 21 Dec, 02:00, benlizross <benli...@xxxxxxxxxx> wrote:

[snippage]





A very eminent Austronesian linguist recently sent me the draft for
his (in press) Numbers chapter from his forthcoming tome - 'The
Austronesian Languages'.

It included this:
"Seimat, alone among the languages of the Admiralty islands , also has
a true quinary system, with the structure 1, 2, 3, 4, 5, 5+1, 5+2,
5+3, 5+4, 2x5.   Although this may seem like a drastic departure from
the decimal system that these languages inherited from a remote common
ancestor, even more drastic innovations in numeral systems are found
in some AN languages of New Guinea , where they clearly reflect Papuan
contact influence."

Seimat is 100 miles across open ocean from the only 'other Admiralty
Island', Manus, where some 18 languages share a very unique number
system, and has no relation whatsoever with them, although it has been
squeezed into the same language grouping.

"where they clearly reflect Papuan contact influence" - I haven't yet
come across a single instance where a local Austronesian language has
borrowed number words directly from any still-existing neighbour.
(Except for the Ekagi words for 6-9, which I am still working on).

----------------------------------------
"..may seem like a drastic departure from the decimal system that
these languages inherited from a remote common ancestor, even more
drastic innovations in numeral systems are found in some AN languages
of New Guinea"...

How, exactly, do any very obviously less-developed number systems
become 'innovations'?

You need to rethink your assumption that composite numerals are
necessarily survivals from an earlier stage.  

'Composite numerals' definitely represent an intermediate stage in the
evolution of numeral systems. There's no way that someone's going to
re-invent or 'innovate' 'five-and-two' as a substitute for 'seven'

That's exactly the assumption I'm questioning, commonsensical though
it may sound.

What do we find in northern Vanuatu? For numerals 6-9

Language A: 6.ono, 7.pitu, 8.walu, 9.siwa
Language B: 6.lima-rave, 7.lima-rave-rua, 8.lima-rave-tolu, 9.lima-rav-
vati

(These are made up so I don't have to go check the exact forms.)
These languages are closely related and spoken next door to each
other.

There is no possibility of A having "evolved" its numerals in situ,
since they are cognate with numerals all over Austronesian.
There are no Papuan languages in the area, and no evidence of a pre-AN
population in Vanuatu which might have acted as a substrate.

In the Vanuatu languages I have data for (excluding a couple of
Polynesian outliers) just 20 have a recognisable simple PAn *pitu word
for number 7, and 98 have compound terms like 'lave-rua', 'luenemu',
etc.

So, you reconstruct, from those sets of cognate forms, a 'proto-
Vanuatu' which has a compound term for 7 (as Alex Francois has indeed
done for proto-North Vanuatu). Do you infer from that reconstruction
that most immkigrants in Vanuatu 'innovated" a new form for that word?
(And the complete set of 6-9 number words that go with it?).

Or do you infer that the earliest arrivals (the longest-surviving
existing languages) have kept their original system, and that later
arrivals or influences brought in monomorphemic, closely related
words? That's exactly what Codrington thought, over a century ago. To
save going through his whole study, here's the relevant page of the
scan: .

His is the latest comprehensive study of Vanuatu numbers as complete
systems that I've come across. Ray and Ivens may have more in books
that I can't access, although I've read most of their available
papers. Tryon, when he attempted a Vanuatu grouping, gave up on the
numbers.

In Indo-European, the problem (or the opportunity to study it) simply
doesn't arise - out of 295 languages only one group, Vedda, hunter-
gatherers in Ceylon, still has a 5+1, etc system, not a monomorphemic
decimal system. Now please don't tell me that they're innovative.

In Western Malayo-Polynesian, from which the original *PAn was
reconstructed, there is also only one group, Ilongot, also hunter-
gatherers, in Luzon, who have the 5+1 construction, so there's not
much of an incentive to look at number systems there, either.

In Formosa, 8 out of 22 languages have partial or full compound word
constructions from 6-9. Only one, Pazeh, uses a straightforward 5+1
system. The other seven are a bit more complex, preserving traces of
6- and 4-cycles, and so on. But those I have any further information
on all seem to have a decimal system from 10 onwards.

Seems to me you have to admit the possibility that people can replace
a deeimal system with something that looks "less evolved" to us.

Looking at different systems, in any languages, the numeration
evolutionary path becomes very evident, from 1-3, 1-5, 1-5-20, 1-5-10.
(Forgetting body part tallies, 4 and 6 cycle systems, etc). It also
becomes evident that different sections can get 'patched in' at
different times - ie 1-10 remains quinary, while beyond 10, or beyond
100, becomes decimal.

If you just ignore the pre-supposition that proto-language speakers
spoke all of their putative lexicon at the same time, then you can
admit the possibility that different parts of a discrete word set
developed at different times, and possibly different places, not
synchronically. Then you can search for where those new systems and
words came from.


A broader possible conclusion: Counting isn't as important as we might
tend to think.

Counting properly was absolutely vital for South Sea islanders who
wanted to out-compete each other in feast or gift-giving, or yam,
coconut, shell, or pig exchange (or even, as in a very recent paper,
how many heads a Solomon Islander collected).

Accountants aren't as important as they might think, but, by G**,
don't they have an awful effect on our lives!
(Sorry if I've offended any bean-counters).

regards

Richard

Ross Clark- Hide quoted text -

- Show quoted text -

.

Relevant Pages

  • Re: Do Eskimos count like New Guineans?
    ... but some Vanuatu languages that use *sa-puluq ... Into all languages north of Epi-Efate with just a few exceptions: ... Paamese, all Epi (exc Bieria/Vovo), Efate, Erromanga - 2-5 ... few examples) that I can find of that word order is in Tetun, ...
    (sci.lang)
  • Re: Do Eskimos count like New Guineans?
    ... but some Vanuatu languages that use *sa-puluq ... Into all languages north of Epi-Efate with just a few exceptions: ... Paamese, all Epi (exc Bieria/Vovo), Efate, Erromanga - 2-5 ... few examples) that I can find of that word order is in Tetun, ...
    (sci.lang)
  • Re: Do Eskimos count like New Guineans?
    ... but some Vanuatu languages that use *sa-puluq ... Into all languages north of Epi-Efate with just a few exceptions: ... Paamese, all Epi (exc Bieria/Vovo), Efate, Erromanga - 2-5 ... few examples) that I can find of that word order is in Tetun, ...
    (sci.lang)
  • Re: Do Eskimos count like New Guineans?
    ... but some Vanuatu languages that use *sa-puluq ... Into all languages north of Epi-Efate with just a few exceptions: ... Paamese, all Epi (exc Bieria/Vovo), Efate, Erromanga - 2-5 ... few examples) that I can find of that word order is in Tetun, ...
    (sci.lang)
  • Re: Do Eskimos count like New Guineans?
    ... but some Vanuatu languages that use *sa-puluq ... Into all languages north of Epi-Efate with just a few exceptions: ... Paamese, all Epi (exc Bieria/Vovo), Efate, Erromanga - 2-5 ... few examples) that I can find of that word order is in Tetun, ...
    (sci.lang)