The nature of slang and HDAS

[Frank Abate]

Ed Gates makes some very good points in his comments on this thread.  I
would like to dispute the following, though:

>>
Like Sidney, I don't think that slang is essentially oral. Most slang
usage has been oral, because it is a feature of informal communication.
Today more and more informal communication is in the electronic medium,
and it seems possible (though not likely) that its occurrence there might
overtake its oral use. Further, some in-groups communicate as much in
writing as in speech -- "God's truth" and "Hocus-pocus" were slang terms
used informally by descriptive linguists back around 1960.
<<

I still hold out, saying that slang is fundamentally and essentially oral --
in its **origin**.  Please reply if you can show that this is not the case.

The primacy of orality in language -- in an essential way for slang, and in
important ways for most other language sets (except technical sets) -- is
the point I am trying to emphasize.  This primacy (which I think most
linguists and language experts would not dispute, or if disputed they would
have a very hard time proving; see below) means that written citations, used
for any dictionary, slang or other, are essentially limited, as they look at
only a small corner of the language, and that corner includes a lot of
edited language.  Language is, first and foremost, speech.  Most language,
by far, occurs in speech.  The written version is a minority player.  For
slang, the written version is not slang at all; it is written-down language
that is using slang in it.  By definition (I contend), written language
cannot be slang; it can only employ it.  People don't write in slang, they
speak it.  Much writing (such as much email) is highly informal, of course,
and some creative and nonfiction writing attempts to recreate slang speech,
but in doing so it is not creating new slang, merely picking up and
repeating what has already become slang.  Otherwise the use of the slang in
writing would be meaningless to the readers.  This is why I think that, even
in graffiti, slang cannot be created as writing.  There may be examples, but
I strongly suspect they are very few.

As for the possibility that Ed G raises that the electronic medium might
overtake the oral use of language, please consider the following.  However
much email and other electronic communication there is, has been, or may be,
in terms of volume of words used, it does not come close the volume of words
used in speech.  If you throw in the entire body of modern English
literature, including everything from Shakespeare's day through to
grocery-store coupons, oral language still wins -- in terms of volume -- and
handily.

Some years ago, Wendy Morris and I did an informal (and quite conservative)
calculation of the number of words that may be used in English in a single
day.  It goes something like this:

1. average number of words spoken in one minute by one speaker speaking:
200

2. number of non-infant English speakers alive today, in US and UK only
(est.)          300 million

3. number of English speakers speaking at any given time (estimated average)
30 million

4. number of English words being spoken in one "average" minute (#1 x #3)               6
billion

5. number of minutes of speaking in a waking/working day (2 hours x 60
minutes)                120 minutes

6. number of English words being spoken in one "average" day (#4 x #5)                  720
billion


One might dispute or change the assumptions and averages, but even if you
make the calculation far more conservative, the "bottom line" number for a
given day is still VERY large.  I would say surprisingly large.  Easily in
the billions per day, with any set of assumptions and averages.  That is for
ONE day, and only US and UK.

The point is that dictionaries, which traditionally rely on written
evidence, are "missing" the majority of the language, and the older the
citations used (and hence the more historical the approach of the
dictionary), the more they are showing us what the language WAS, not what it
is.  What language was is very important to know and record.  I do not
dispute that in the least, nor do I mean to denigrate what historical
dictionaries have done, slang or other.

I am a fan of studying what the language IS, lexically and grammatically.
Having worked on the New Oxford American Dictionary, which took advantage of
the OUP/OED citation files, as well as much corpus evidence (American and
other), I am convinced that corpus evidence is a better way to look at and
learn what contemporary language is up to, compared to citations.  Corpora,
while obviously reliant on written evidence, too, can and do include raw
conversational speech that has been transcribed, and is unedited.  That gets
the lexicographer much closer to the "action".  Corpus lexicography has
proven that the primary (most used) senses of some very common and highly
frequent words is not at all what dictionaries have told us (and still tell
us) for many years.  Examples abound, but I defer here to what Patrick Hanks
has written on this subject, from his work at both Collins and Oxford.
(Patrick is far more convincing, and has examples at hand to show this
point.)

I have great respect for and reliance upon historical lexicography, whether
OED, HDAS, or other.  I have and will use these dictionaries every day.  But
I try to remind myself every day that they only reflect what language IS.
(Like writing dictionaries from Plato's cave.)

As I wrote this, English speakers used several billions of words, and I only
wrote a few hundred.

Frank Abate