Oftenly

[Arnold M. Zwicky]

On Aug 26, 2007, at 2:56 AM, i wrote:

> ... characteristically, the frequency of any *particular* #2-type
> combo
> is low (even "herrible", which i've collected as a clear inadvertent
> slip -- the speaker corrected himself, in embarrassment -- is not all
> that frequent, even though it appears every so often as a quote from
> Winnie the Pooh, in "herrible heffalump" and allusions to that
> phrase, rather than as a slip of the tongue), ...

the example i intended to cite here was "horrifle".  there are only
33 google webhits (with dupes removed), and many of these are
irrelevant or are deliberate coinings.  some are probably errors,
either true blends or classical malaprops for "horrific" (and some
might be typos).  ("herrible" is also attested, and also at a low
frequency, but the influence of WtP gets is an interfering factor, so
"horrifle" is a better example.)

the hit rate for "fastly" and "oftenly" looks to be *at least* two
orders of magnitude greater than this.

> ... my argument against a #2-combo analysis in this particular case
> doesn't follow from terminological considerations, but from the
> details of this case.  i brought up the terminology (*after*
> conceptual analysis, i note) only because i think you're bewitched by
> it: you're inclined to take all sorts of combos ("blends" in a very
> broad sense) to be at root #2 combos ("blends" in a narrow sense),
> and that just muddles things up hopelessly.

when we went through this the last time, back in 2006, i tried to
make careful distinctions, only to have these rejected in a statement
from jerry cohen that clearly takes all sorts of combos to constitute
a unified category, and uses "blends" as the name of this larger
category: "Blends are no less blends if they arise
inadvertently."  (this also seems to assume that i was claiming that
#2 combos are not blends, while i was in fact treating #2 combos as
an object of study on their own, separate from other types of combos,
and was reserving the term "blends" for these and these only.)  at
this point i threw up my hands, despairing of ever being understood,
and abandoned the thread.

but i should have commented on something from this same posting of
jerry's (of 8/15/06), namely his citation of his book on syntactic
blends:

   >_Syntactic Blends In English Parole_ ("parole" as used by
Saussure, i.e., everthing not part of "langue," i.e., not part of the
standard language; "parole" includes speech errors)<

now, in fact, the examples in the book are almost all clearly speech
errors -- a fact that made the data in the book suitable as a corpus
for liz coppock's 2005 stanford qualifying paper "Alignment in
Syntactic Blending", which is *entirely* about (inadvertent)
production errors and the production mechanisms that might give rise
to them.  the paper is available at

   http://www-csli.stanford.edu/~zwicky/CoppockSyntacticBlends.pdf

the reference to Saussurean "parole" muddles things up; nothing in
the book suggests that speech errors of the interwining/combining
sort (#2 combos, which are inadvertent) and advertent but non-
standard variants together constitute any sort of coherent class of
phenomena, and i know of no evidence that indicates that they do.

arnold

------------------------------------------------------------
The American Dialect Society - http://www.americandialect.org