your mail

[Isidore Dyen]

----------------------------Original message----------------------------
 
It seems to me that the variety of new formations thatare involved in the
discussion are all of recent origin. It would be interesting to examine
whether the stump types or whatever can be exemplified before the present
highly literate period and then the frequency of the contaminative
types (i.e. analogy-like new formations where the models involve a nonce
analysis) are common at earlier times anywhere. If not, there is
circumstantial evidence that literacy is playing a role along with
sportswriters, Madison Avenue, and other word-spinners.
The fact that change is inevitable in the continuation of a language is
not a peculiarity of language, since it is a pretty widely distributed
characteristic. In fact it is pretty difficult these days to cite anything
that is inlabile. The only reason for mentioning that lability is a
language universal is that languages, though dialectally variable, give
the impression of being stable.
 On Sat, 1 Aug 1998, bwald wrote:
 
> ----------------------------Original message----------------------------
> Isidore Dyen writes with respect to such constructions as "workaholic":
>
> >I think that it might be important to add to the discussion that these
> >forms are artificialloy constructed and in that respect fall in with words
> >like AIDS or is it AIDs and CIA and G-man and the gamut that have
> >sprung up in at least a partial connection with writing and thus
> >differ from the types of analogical phenomena that appear in comparative
> >studies.
>
> I think ID raises a very interesting issue, but one which is more
> problematic than he suggests.  To begin with, he no less than any other
> reader recognises that the examples he has given above are different from
> any of the examples we had been previously discussing, in that his examples
> make reference in one way or another to spelling.  That is not true of the
> examples, call them "condensed compounds" for the moment, that we have been
> discussing.  Nevertheless, an interesting *burden of proof* issue arises
> between those like ID who suggest that the types of condensations that give
> rise to "workaholic" and other examples -- indeed what ARE other examples
> and what are NOT? -- can only arise from some kind of literate perception
> of the condensed words (this is certainly NOT obvious), and those like me,
> who might say: PROVE IT! or at least, WHAT MAKES YOU THINK SO?
>
> For example, take "workaholic".  ID's idea seems to be that the -aholic is
> abstracted by some literate means.  I can't really follow this, and even
> the variation in spelling of the unstressed vowel *a*holic vs.
> (alc)*o*holic seems to argue against this.  But I don't insist this is a
> strong argument against his assumption.  A stronger argument against, I
> think, has to do with such diminutive derivatives as "alkie" for
> "alcoholic" (= a person who drinks too much), which, without any hint of
> literacy, abstracts, "back-clips" (thanks, Jim Rader and Hans Marchand) or
> "stumps" the first "syllable", leaving -oholic free by implication.  OK.
> Why do I say -oholic is free by implication?  Well, I haven't totally
> thought this through, but I think the stumping technique and its residual
> can be compared to the Romance (and more generally learned) stratum in
> English where speakers, whether literate or not, can generally perceive
> multi-morpheme combinations that recombine, but do not have the slightest
> idea what the constituent morphemes "mean" -- and indeed that poses a
> problem for us linguists to decide how to deal with them, cf. defect,
> perfect (say the verb), effect, detain, pertain, retain, etc etc.  Again, I
> think people are quite clear about the constituent parts but quite unclear
> (and unconcerned) about their meanings in a great many Romance formations
> in English.  (malaprops also show perception of constituent parts of
> Romance formations, e.g., defuse for diffuse, prevert for pervert, or
> whatever ones actually occur).  Thus, such Romance forms provide a
> precedent (if that's historically relevant) to splitting things like
> "alcoholic" into constituents morphemes "alc+oholic".  In fact, the latter
> are much more motivated than the Romance forms, since "alkie" gives the
> "alcohol content" of the alc-, and leaves -oholic for "addict", etc.  I
> agree with ID that many forms like "workaholic" might have been coined by
> literate users for literate purposes (cf. the journalese proliferation of
> "-gate" forms for political scandals), but not that the technique
> presupposes literacy and could not happen/start without literacy.
>
> "cheeseburger" might be quite different from "workaholic", in that "burger"
> alone "fore-clips" "hamburger".  No literacy seems necessary, certainly no
> more than in "gator" for "alligator" (cf. "gatorade", wherever that came
> from -- Florida?  And NB "telephone" > "phone" does not presuppose the
> morphological independence of "tele", why not also for "microphone",
> "megaphone" and what-not, whoops "phone" also means "allophone" in one
> obscure jargon.)  Thus, "cheeseburger" seems to be simply a convenience for
> "cheese hamburger", and not as complex as "workaholic" in its origin.  All
> of this, of course, depends somewhat on whether "burger" as an independent
> word came into existence BEFORE OR AFTER "hamburger" (something we may
> never know, though it might seem almost within our reach to know, ESP, and
> NOTE THIS, if the origins of the two were ORAL, and their order of written
> appearance is relatively close -- as is probable -- and arbitrary. In fact,
> except for "dialect representation", we might expect "burger", like
> "gator", to be suppressed in written language until it had spread quite
> widely, leaving doubt about its true chronological relationship to the
> emergence of "cheeseburger" and the rest of the (-)burger family. EG one
> abridged 1994 dictionary gives "cheeseburger" from 1938 but does not
> acknowledge "burger" EXCEPT as a combined = dependent form, ignoring that
> "burger" has been short for "hamburger" to my ears for
> I-don't-know-how-long, many decades at least.  Surprisingly, this
> dictionary, despite its intentional incompleteness, does list "gator" from
> 1844, no initial apostrophe or hyphen.)
>
> A final comment on this matter, which may be relevant to the "literacy"
> issue, but I'm not sure how, is that "workaholic" and many such structures
> have a certain humor or cleverness about them that suggests conscious
> manipulation.  That in itself does not suggest literacy to me.  But it
> contrasts the technique with the stump compound, which seems to be most
> often just a convenience which reduces or *abbreviates* -- something also
> done by such uncontroversially literate techniques as those which produce
> AIDS (acronyms) or CIA, LA etc.(crude initial abbreviations), without any
> humor and, in fact, often with a *bureaucratic tediousness* about them (but
> there is something humourous to me, at least, in referring to people by the
> initials of their names, e.g., BW and ID etc etc, so immediate context is
> important too -- similarly, humour or cleverness was a feature of many
> acronyms in 1960s-80s political discourse, cf. NOW!, and a crude kind of
> rhyming humour in things like "high fi" and "sci fi", the long vowel of
> "fi" in both cases also depending on spelling).  But, in general, stump
> compounds and the spelling-reference types are relatively boring,
> convenient for users and opaque to non-users.  Not so for either the
> "workaholic" or "cheeseburger" type (with regard to opacity), or for the
> "monokini" and "glitterati" types (with respect to specialised humour)
> about which more could be said.    It may be that ID is confusing ALL (or
> TOO MANY) condensation techniques involving multi-word expressions with
> those which have a transparent literate base, though I find it hard to
> believe that he might have no more basis for his claim.  As I said above,
> if he has more of a basis, WHAT IS IT?
>
> In a separate matter, ID replied to my comments on GG (or UG) and
> historical ling with:
>
> >How about contemplating whether language change is inevitable. If it is
> >not, theren should be some stable languages somewhere, If it is, then it
> >must be inherent in all languages and thus a universal.
>
> To which Jan Terje Faarlund replied:
>
> >..Change in itself cannot be part of
> >the system. The only interesting connection between universals and change
> >is the fact that no change can lead to a result which violates UG.
>
> I had been tempted to respond in a similar way, but I thought that ID was
> writing this tongue-in-cheek, and also that he might escape by insisting
> that "universal" does not necessarily mean "linguistic universal", since he
> might deny, like many, that there are specific linguistic universals
> unrelated to some more general cognitive or whatever universals, and that
> "change" is one of them ("linguistic change" being part of "change", as
> Faarlund observed with regard to social and biological change, i.e., life
> and death, in his longer reply).  In view of this, I would agree with
> Faarlund's use of "interesting" in the above passage as keeping comments
> about LINGUISTIC change *on track*, but experience suggests to me that ID
> would object to the adjective and say that "interesting" is not an argument
> but an expression of taste.  I'll leave it at that.
>