Quote:

 

Unfortunately, no reference is given, so it is hard to establish the
origin of the proposed etymology. Nevertheless, unless AV’s idea is that
both forms were concocted already on Korean territory from two Sino-Ko-
rean morphemes (not very likely), we would have to presume that both of
these compounds can easily be found in Middle Chinese, thus corrobora-
ting the «loanword» solution. And the first of these, 惡水 (MC *ʔâk-śwí),
really exists — but its meaning, however, is usually defined as ‛dirty
(stagnant) water’ [HDC VII: 55], with no direct connections to ‛rain’, let
alone ‛heavy rain’. The second one, to use AV’s own terminology, is a
‛ghost word’: no such compound is ever attested in Chinese texts or mo-
dern spoken language (although we do sometimes have it as a verb+object
idiom: MC ʔek śwí ‛to suppress the water(s)’, said, e. g., of Yu, subduer of
the Great Flood). Even if we suppose that such a compound may have
existed, though (but why should we?), it is hard to believe that it was used
to designate any kind of falling water, since in Chinese the word 水 shuǐ
(MC śwí), throughout all the attested epochs, has always designated
flowing rather than falling water. If the Korean words meant anything like
‛flood, inundation’, the borrowing suggestion would be more plausible.
As it is, there is nothing «transparent» about these «loans», and external
explanation is at the least equally reasonable, if not more so.

Quote:

 

PJ *kàtànà ‛sword, knife’ is derived in EDAL (p. 53) from PA *gằŕ[à]
‛sharp edge’; AV rejects the derivation in favour of an internal etymology
that analyzes the word as kata- ‛one’ + -na ‛blade’, reasonably pointing out
that the word normally refers to a single-bladed sword (as opposed to OJ
turugî ‛double-edged sword’) already in OJ. This etymology, which he calls
«quite transparent» in his review, did not, however, seem all that
transparent to him a decade earlier, when he wrote: «there is not enough
evidence for hypothetical PJ *na ‛blade’, reconstructed on the basis of -na in
katana ‛small sword’, ‛knife’ and kana ‛plane’ by Martin» [VOVIN 993a: 30].
One wonders, of course, what exactly is that extra evidence for PJ *na that
came up over the last ten years and made j~êíáå’s etymology so transpar-
ent. However, we are not really putting that against AV, since we ourselves
believe that such a word is potentially reconstructible (it is at least suggested in
several dictionaries, e. g. [JDB: 52]). Yet even taking into account that the
internal analysis goes way beyond j~êíáå (the same explanation is given in
[JDB: 93]), there is still a major problem with this analysis: in the oldest texts
(as well as later ones) the word is never transcribed with the character 片, as is
the common case with all similar compounds quoted by AV, such as 片足
kata-asi ‛one foot’, 片側 kata-gawa ‛one side’, etc. Instead, it is always written
semantographically as simply 刀 ‛sword’ or 小刀 ‛small sword’, as opposed
to tsurugi, written either as 劍 ‛(Chinese) sword’ or 大刀 ‛big sword’.
Given this consistent spelling, it can be concluded that (a) already in
the OJ period the word kàtànà was not perceived as having anything to do
with kata- ‛one’ and (b) the main difference between it and turugî was seen
as their respective size — even though sharpness of the edges might have
been a criterion as well. As for the ‛one-blade’ etymology, it may have
become popular through analogy with the well-attested compound kata-
ha ‛single blade(d sword)’, appearing in the language much later and this
time faithfully transcribed as 片刃.
To this must be added that we really know little about what kind of
weapon was designated in OJ by the term kàtànà. Its current and most well-
known incarnation — the long curved single-edged samurai sword — is actu-
ally quite late (coming into heavy use around the 5th century). Before that,
the primary fashion for single-edged swords was the tachi (太刀), but even
this is generally thought to have been carried over from China not earlier
than the 0th century (around the same time that the famous dictionary Wa-
myō ruijushō [倭名類聚抄] was compiled, where the ‘one-bladed’ definition
of kàtànà is met for the first time). And yet we already have the term kàtànà
attested throughout the Kojiki (古事記, 8th century). Whatever the ultimate
solution may be, the kàtànà of these early monuments should hardly be
associated with the iconic samurai sword of the 5th century, meaning that
AV’s confidence here is slightly anachronistic.
In the light of this, we are at peace with the idea that -nà in kàtànà is
not a suffix, but really an old word meaning ‛blade’, although it had al-
ready fallen out of independent usage long before the OJ period (and, there-
fore, could hardly have been borrowed from Korean — see below). As for
what concerns the first part of the compound, an external etymology that
links it to PK *kárh ‛sword’ (actually following MARTIN [966: 25], who re-
constructs Proto-Korean-Japanese *khal-nal ‛blade of sword’) and then fur-
ther with several semantically more distant parallels in Turkic and TM34,
seems to us a much better solution than the one advocated traditionally;
and if AV’s understanding of «respect for cultural history of languages»
(more on that in the «philology» section below) equates it with uncritical
acceptance of every instance of lexical analysis found in Japanese diction-
aries, we must respectfully disagree.