Humanist Discussion Group, Vol. 17, No. 585.
Centre for Computing in the Humanities, King's College London
Submit to: firstname.lastname@example.org
Date: Fri, 30 Jan 2004 08:16:15 +0000
From: "Jim Marchand" <email@example.com>
Subject: KWIC again
The difference between a KWIC concordance and a concordance is
important. A KWIC concordance requires no thought. What it does
is to offer the keyword and an automatic context. Usually one
halves 80, inserts the keyword in the middle and then prints out
the `context'; this is usually sufficient, though the work, as
Patrick points out, is left to the user. Since Norm mentions
Chaucer and I know he uses Tatlock and Kennedy, it might be useful
to compare that with A Complete Concordance to the Works of
Geoffrey Chaucer, Akio Oizumi & Kunihiro Miki, vol. 6, A
Concordance to Boece (Hildesheim: Olms, 1991), the latter being a
KWIC concordance (random):
T & K, Bo. 3. m.9. 870-5: Thow knyttest togidere the mene soule.
O & M: III m 9 25 the watris. Thow / knyttest togidere the mene
soule of treble / kynde moevynge alle.
Note that you have to chop off _the watris_, which goes with
`ploungid in the watris'. You have to worry about the construction
ending with alle ( = + `thingis'), etc., whereas T & K give you the
construction, no more, no less. As Patrick points out, a KWIC
concordance is handy and easy to make. I use them all the time
when I am editing something, but I don't want to pass my problems
off on the reader. I don't think that undoctored KWIC concordances
ought to be published. They ARE quick, especially with our fast
machines. If I had a King James on my hard disk in proper format,
I could give you a KWIC concordance in no time; I could use it for
all kinds of things. Note, however (I apologize for inevitable
WEPT: Joh 11: 35: o him, Lord, come and see. Jesus *** wept ***.
Then said the Jews, Behold how h.
As you see, a KWIC concordance, which requires no thought, offers
a lot of noise and many sources of errors, and requires hard work
on the part of the customer.
This archive was generated by hypermail 2b30 : Fri Mar 26 2004 - 11:19:39 EST