Systematicity

[Dennis R. Preston]

Well, I don't agree that dialects (or languages) are idiolects
slopped together; there is a level of abstraction that won't tolerate
that purely empirical approach. If it was really that simple, we'd
have a quantitative matrix of some sort to tell us exactly how many
features would have to be shared before we could responsibly call a
dialect a dialect, and so on up the scale. We don't.

Your example is a very good one. You "infix" "like" in "forever."  (I
can't, by the way, but I'm a really old codger.)

It's systematic in many ways for infixing in English - I bet you
can't get "unlikehappy" and "carfuckinpet" (i.e., "carpet" with
"fuckin" infixed). These impossibilities all point to describable
(systematic) rules, some morphosemantic, some purely phonological.

What remains then is to see whether it's "systematic" in your grammar
(and the grammars of those most like you). That's the killer, of
course, Is this an item you have non-productively or do you have lots
of other "like" infixed items (which obey these same rules)? The
frequency with which you use such items would be a secondary
consideration, but your acceptance of forms innovated along these
same lines would surely be one indication of your having this item as
a systematic not idiosyncratic or even imitative part of your grammar.

In a 2002 MSU Linguistics dissertation entitled "This Girl Wants Out"
Erica Benson (now at UWisc Eau Claire) plays with the degree to which
needs/wants plus particle and prepositional phrases (Pull over; I
want around. Get out of my way; I need in the room.) are grammatical
for Northern, Midland, and Southern speakers. (These items have a
much wider distribution than similar constructions with the past
participle.) But the point is that the systematicity (as revealed by
grammaticality judgments) is very subtle, dipping down to the lexical
level (both need/want opposition as well as rankings among the
various particles/prepositions involved). The complexity of a rule's
"degree of systematicity," if I may, for even one speaker is,
therefore, very complex, denying the "settings" approach to a
grammar, although that approach is very much improved, in my opinion,
by the minimalist assertion that such settings are morpheme- rather
than construction-dependent (or "associated").

dInIs



>From:    "Dennis R. Preston" <preston at MSU.EDU>
>
>: My use of systematic in this case referred to the presence of an
>: element in a "dialect" (or variety), not its status in an individual
>: speaker...
>
>Same thing in the end, IMO--all that a variety is is the idiolects (using
>the term somewhat sloppily) of a bunch of speakers who share a particular
>(usu. regional) characteristic, all lumped together.
>
>: ..., but since you raise the issue, I'm afraid that I'm a kind
>: of psyycholingusitic rebel here. I believe in degrees of
>: grammaticality; that is, I believe that some constructions are not
>: as "deeply embedded" in competence - not part of a speaker's
>: "vernacular." On the other hand, it's not clear to me that frequency
>: would be a guide to this at all.
>
>I think that, actually, we're in agreement here. However, i don't think i've
>ever run across any way to *objectively* say whether something is systematic
>or not for any particular speaker *or* variety, other than, in the end,
>counting tokens. I'm very much unsatisfied with that.
>
>And in a perhaps-related note, i caught myself using an infixed form today
>that i hadn't run across before (but that, in various spellings, gets 13
>Google hits as one word, 4,156 as three): "I feel like our power's been out
>from the hurricanes forlikeever." (And, yes, i used it very much as one
>word.) So: Is -like- infixation systematic to my particular variety? How
>would i ever be able to say for, like, sure?
>
>Like i wrote earlier, this has been bugging me for some time now. For a
>concept that one would think is so central to what we do, i just feel like
>it hasn't been defined at all well.
>
>David Bowie                                         http://pmpkn.net/lx
>     Jeanne's Two Laws of Chocolate: If there is no chocolate in the
>     house, there is too little; some must be purchased. If there is
>     chocolate in the house, there is too much; it must be consumed.


--
Dennis R. Preston
University Distinguished Professor of Linguistics
Department of Linguistics and Germanic, Slavic, Asian, and African Languages
A-740 Wells Hall
Michigan State University
East Lansing, MI 48824
Phone: (517) 432-3099
Fax: (517) 432-2736
preston at msu.edu