--- In cybalist@..., tgpedersen@... wrote:
> "The combined lexicons of the AN languages"? "Words selected ad
hoc"? In what way does this methodological criticism differ form what
might have been raised against Rask and Grimm? This is pretty lame,
Piotr. Frankly, I'm disappointed.
------------------------------------
To get Manansala's "results" it is enough to proceed as follows: take
any Indo-European root and scan Austronesian word-lists until you
find a similar-looking item with a roughly matching meaning. The game
is an easy one, given the number of candidates. Let's suppose you've
found a match in language "A". The chances are that in languages
closely related to "A" there will be cognate (and similar) words.
E.g. a match discovered in Hawaiian may secure you further matches in
Maori, Marquesan, Tahitian, etc. With a little bit of luck, there may
be one or two other serendipitous matches in different branches of
Austronesian -- there are so many of them. Who cares if the matches
you have collected are not systematic, or if the "etyma" you list are
not reconstructible to PAN (or to Proto-Malayo-Polynesian, or to
proto-anything), or even if they are loanwords in Austronesian?
To use my own example, let's imagine you want to find an AN match for
PIE *dHeg^Hm-/*dHg^Hom- ~ *g^Hom- 'earth'. A cursory glance at one of
the pages in an Austronesian comparative dictionary will give you
Murut dagana? 'earth (as opposed to sky)' (jeez, it's pure
Hittite!!), Balinese gumi 'earth, world', Pazeh daxe 'soil, clay,
earth' -- a beginning that promises a nice collection of matches if
you cast your net wide, and note that I haven't yet begun to relax
the semantics of the comparison. Of course the more numerous word
lists or the more inclusive dictionary you have, the easier the task.
You can also, if you prefer, compare Sanskrit bHu:mi- 'earth' with
Indonesian bumi, Madurese bumi, Javanese bumi, Sasak bumi, Sunda
bumi, etc. -- all meaning "earth"! Guess what's wrong here. You might
think it's a naive parody, but some of Manansala's matches are really
of this kind.
Rask and Grimm? Boy, they must be turning in their graves. They
usually concentrated on pairwise or three-way comparison in their
arguments and looked for systematic correspondences, not mere
similarities. Let me quote from Grimm:
"In the investigation of the words, likeness or resemblance of
consonants which are in general related is less important than
observation of the historical course of gradation, which does not
become disturbed or reversed. A High German word with <p>, which
shows <b> in Gothic and <f> in Latin is originally related in these
three languages ... If, however, we were to find an <f> in a High
German word, <b> in Gothic and <p> in Latin, then the relationship
would be nonsensical ..."
By modern standards, their methods were still rather imprecise in
many ways, but unlike Manansala they were on the right track.
Grassmann, Verner and other people who really believed in
methodological rigour improved those pioneering attempts and made
comparison a scientific method. Today it would be wrong methodology
to proceed like Rask and Grimm did (after all, we have learnt from
*their* errors), but it is no methodology at all to do things
Manansala's way. It has more to do with Greenberg's "mass
comparison" -- which isn't a compliment as far as I am concerned.
Sorry, Torsten, it's all ballocks and a waste of time.
Piotr