thy thigh etc.

[Larry Trask]

--On Monday, May 28, 2001 3:19 pm +0000 Douglas G Kilday
<acnasvers at hotmail.com> wrote:

[on my objection to his assertion that the lexicon of English is remaining
roughly constant in size, and my observation that desk dictionaries get
bigger with each edition]

> The tendency to equate the physical size of desk dictionaries with the
> size of the underlying lexicon (or "lexis" if you prefer) should be
> avoided. Several factors favor the continued growth of desk dictionaries
> (with the literary exception of the Newspeak dictionary in Orwell's
> "1984", which got smaller every year).

> First, and most importantly, it's much easier to attest a word than it is
> to "unattest" it. Someone once said "it's hard to write an epitaph for a
> word". I'd say "hard" is understated; it's comparable to proving the
> extinction of an obscure tropical bug. If a lexicographer asserted that a
> given word was extinct, resulting in its being pruned from the desk
> dictionary, and the word reappeared in actual use, the publisher could
> have considerable egg on his face. What sane lexicographer would
> jeopardize his job in this manner? Unless he managed to break into the
> ranks of professional Scrabble, an unemployed lexicographer (fired for
> cause) would be facing tough times indeed.

Indeed, but overly cautious.  Serious dictionaries these days are prepared
from corpora, and a word that failed to appear in the corpora for a long
time would be singled out for possible removal.  Of course, lexicographers
don't drop a word after failing to encounter it for three weeks.  Instead,
they start attaching usage labels, like 'rare', 'archaic' or 'obsolete'.
They do this for excellent reasons.  First, as you say, the word may come
back from the dead.  This seems to have happened with 'albeit', which was
virtually dead a decade or so ago (OED2 has no 20th-century attestations),
but which has suddenly reappeared with a vengeance.  Second, readers often
want to read works written fifty or a hundred years ago, and they may
encounter obsolete words which they want to look up.  A good dictionary
will try to cater to this demand as best it can, within its space
limitations.  A few dictionaries even adopt a policy of entering *all*
words found in Shakespeare, and one or two extend the same courtesy to
Milton and Spenser.

But eventually some of these words have to go.  For example, readers of
George Eliot will recall the quaint word 'pocket-pistol', denoting
something resembling a hip flask.  OED2 does not record this word later
than 1882, and not one of my desk dictionaries enters it.  It's gone from
the language, and now it's gone from the dictionaries, and readers of
19th-century fiction will have to cope as best they can, or get hold of the
OED.

> Nevertheless, words do become
> extinct (or "obsolete"). A hard-core dictionary like the OED or my
> ponderous Funk & Wagnalls Standard (1903) will give thousands of words
> marked as obsolete. It's not the job of a desk dictionary to track
> extinction.

Ah, but the OED is very far from being a desk dictionary.  Its task is to
record all the English words ever significantly attested, and nothing is
*ever* removed from later editions, apart from the odd blunder.  Nor is
(was) Funk and Wagnall a desk dictionary: it was what we Americans call an
"unabridged" dictionary: far too bulky and cumbersome to serve as a desk
dictionary.

I'm thinking of real desk dictionaries, like Collins in Britain and
Merriam-Webster in the US.  And these really do drop words after a while.
Ask a professional lexicographer.

> Second, consumers would feel short-changed if each edition were _not_
> larger than the previous one. If the new dictionary couldn't boast of
> being "new, improved, and _enlarged_", why would anyone buy it? If they
> didn't already have an older edition, they could find one at a garage
> sale. Publishers need to make money.

A fair point, but perhaps overly cynical.  Desk dictionaries really do get
bigger because there really are more and more words to be included.  I
really do not think that the total number of words pertaining to cars,
planes and spacecraft is about equal to the number of words pertaining to
oxcarts and buggies.

> Third, this enlargement can be made at marginal cost, since the core of
> the new dictionary already exists in the old one. Why bother paying
> lexicographers to tramp around the boondocks trying to verify the
> extinction of obscure words? Leave 'em in! Anyone using a _desk_
> dictionary is likely to regard it as a prescriptive authority (not as a
> bank of scholarly data, like the OED) and anything in it is _eo ipso_ a
> valid word.

Yes, but this is an attitude adopted by (many) users of desk dictionaries,
not a piece of truth.  For a linguist, a word exists if people use it,
regardless of whether it's in any dictionary at all.

I've just been marking a pile of final-year dissertations.  Some of our
students chose to write on such topics as text messaging, graffiti,
internet chat-rooms, the slang associated with drugs, and obscenities.  I
was startled to discover, in these writings, a very large number of words
which were wholly unknown to me, but which are apparently commonplace among
young people.  I haven't had the time to try looking these words up in a
couple of recent desk dictionaries, but I'll bet that quite a few of them
are not entered.

Anyway, lexicographers just *can't* "leave 'em in" all the time.  Cost, and
therefore size, is just too important, and something has to give.

Nor do lexicographers even rush to enter new words.  For example, the
language has recently sprouted a number of new formations in '-wise', such
as 'moneywise', 'clotheswise' and 'healthwise'.  These have been frequent
in speech for years, and they are not rare in journalistic writing.  But
they're still not entered in my desk dictionaries.

> It's easy enough to point to new words that don't replace anything. It's
> less easy to find extinct words that weren't replaced by anything, but
> they do (or did) exist: words pertaining to obsolete tools, equipment,
> manufacturing processes, agricultural practices, social groups, dishes,
> beverages, etc. They can be found in hard-core dictionaries which dare to
> mark words "obsolete" or give dates of final attestation. All things
> considered, net stasis of lexical size makes more sense than continuous
> expansion.

I'm afraid I can't agree.  A large and expensive dictionary can often
afford to make room for lots of obsolete words, and of course the OED must
do this by remit.  But an ordinary desk dictionary has to be cheap enough,
and hence small enough, to sell large numbers of copies -- and that means
that old words have to be pruned from time to time in order to make way for
new ones.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk

Tel: (01273)-678693 (from UK); +44-1273-678693 (from abroad)
Fax: (01273)-671320 (from UK); +44-1273-671320 (from abroad)