OED 2 errors

[Dominik Wujastyk]

I don't know why, but about once every couple of months I write to a
dictionary about a word they have wrong.  This time it is the OED,
second edition, and since it has been mentioned here recently, I
thought I'd share my <tiny> observation.
 
The spine of volume V of the OED has a Sanskrit word on it: "dvandva",
the first entry in that volume.  It caught my eye, so I looked it
up.  I was surprised to see it there in the first place, because
it is *definitely* a Sanskrit word, never used in English as
an English word (unlike, say, "yoga" or "guru"). But maybe I have
missed some arcane point of lexicographical principle here.
 
There are two errors in this entry.  First, the Sanskrit word is
"dvandva" not "dvandva"
                  .
 
i.e., the "n" is a normal one, not a retroflex one: no underdot.  I
thought it might be a bit of fly dirt on my copy, but it's not.
 
Secondly,  the relationship between the member words of a dvandva
compound is certainly not the copula; it is the sense of "and".
 
The mistake probably comes from believing the pericope from Monier
Williams; the following one by someone else (I'm doing this from
memory, or rather the lack of) which mentions the example
"Prince-Regent" is better, but still misleading.  It could be read as
"The Regent who is also a Prince", whereas a true "dvandva" meaning
would be "Prince and Regent", or even "the Prince and the Regent".
 
In Sanskrit, dvandvas are often used just to string things together
"catsdogssheepdonkeysgoats", meaning just "cats and dogs and ...".
 
I have noticed other cases where the OUP dictionaries follow
Monier Williams uncritically.  Probably because they publish
his Sanskrit-English dictionary.  But he is wrong about a lot
of things, and you really need to know the language to use
his dictionary safely.
Dominik