misnomer 'misconception'

[Arnold M. Zwicky]

On Oct 24, 2004, at 11:53 AM, Arnold M. Zwicky wrote:

> On Oct 21, 2004, at 5:29 PM, Jonathan Lighter wrote:
>
>> Fritz, you speak the tired language of yesterday. Nowadays you'd get
>> the salad.  See, "substitute" means "replace."  "Can I substitute the
>> fries?" "Of course. Would you care for a salad?  Or baked sweet
>> potato?"
>
> i'm not doubting you, but i'd like some cites.  this is a still further
> extension of "substitute" into the realm of "replace" from the
> developments i just described.  this is the verb with a plain direct
> object.

one of the lessons here is that it's not enough to search out examples
of a word with some meaning ("substitute" 'replace'); we have to look
at the contexts in which it occurs.  i've now gone back to denison's
manuscript, and see that things are even more complex than i made out.

to recap a bit, in slightly different form.  using denison's (hopefully
transparent) labels OLD and NEW, the original verb usages were:
   (1) substitute NEW for OLD  (NEW be substituted for OLD)
   (2) replace OLD by/with NEW  (OLD be replaced by/with NEW)

"substitute" then encroaches on "replace" territory, giving the
proscribed (but very widespread and unambiguous):
   (3) substitute OLD by/with NEW  (OLD be substituted by/with NEW)

now, the oblique objects in all of these cases are omissible, with the
result that clause types (1) and (3) collapse; potential ambiguity
results:
   (1') substitute NEW  (NEW be substituted)
   (3') substitute OLD  (OLD be substituted)
this is the development jon lighter alludes to, and i asked for cites
to, above.  (3') is a kind of reversal of simple transitive
"substitute" in (1').

the construction denison is particularly interested in is (4), a blend
of (1) (original "substitute") and (3) (encroached "substitute"), and a
source of potential ambiguity with respect to (1):
   (4) substitute OLD for NEW  (OLD be substituted for NEW)
this denision calls "reversed "substitute" ".  he believes it to have
appeared recently -- his first cites are from the BNC (1985-93) -- and
in british english, and still to be almost entirely british rather than
american.  he suggests that the usage arose in soccer contexts, and
then spread (far and fast).  here's a nice cite he sent me yesterday:
-----
I meant to give you this one.  Helen Young, national weather forecast,
6.06 am, BBC Radio 4, 21 October 2004:

Well, we can substitute rain for wind today:  it's going to be a very
windy day.  Blowing a gale is no exaggeration.

[and yes, it had been very wet the day before]
-----

this picture would be significantly deranged if there are respectable
numbers of reversed "substitute" -- note, *reversed* "substitute", not
just any occurrence of "substitute" meaning 'replace'; the oblique
object with "for" is crucial -- back some years in american english.  i
have to confess that i didn't record the details of my students' use of
"substitute" to mean 'replace', so i can't swear that some (or any, or
many) of them were reversed "substitute".

can any of you provide some relevant data?

denison reports some difficulty in making his manuscript available
on-line, but says: "I'd be happy to forward it to anyone interested who
can't get it themself" (yes, "themself", a little gift to me as the
former adviser of the "themself" maven).  his website is
   http://lings.ln.man.ac.uk/info/staff/dd/
(and i'm copying this posting to him).

as someone with a lot of handedness trouble, i'm praying that i've
gotten OLD and NEW distributed correctly in the discussion abo\ve.

arnold (zwicky at csli.stanford.edu)