Neogrammarian vs. other sound change

[Brian D. Joseph]

Andrew is absolutely right that there is the one remaining issue still  
to be resolved.  At the risk of sounding self-serving, let me say that  
Rich Janda and I gave an answer to this (at least implicitly) in our  
2003 paper on a "'big bang' model of sound change" (published in the  
Benjamins volume arising out of the 2001 ICHL in Melbourne, edited by  
Barry Blake and Kate Burridge).  We located regularity in the initial  
onset of a sound change, where phonetic factors are paramount.  We  
envision that as relatively brief (hence the parallel to the "Big  
Bang" in physics) and after that, extensions along a number of  
parameters are possible.  One is further phonetic generalization (thus  
giving regularity of broader scope than at the initial "big bang" (and  
generalization to a broader set of input sounds is also possible), but  
lexical, grammatical, or social lines of generalization can be  
followed.  To our way of thinking, the crucial thing about sound  
change proper is the phoneticity -- with conditions that are purely  
phonetic in nature (which we envision as the essential starting point,  
the "big bang") regularity follows automatically since that is the  
most elemental and the most broadly applicable type of conditioning.   
Once the forces of generalization set in post-big bang (assuming they  
do -- they need not), phoneticity can disappear, and there is no  
guarantee of regularity (though it could occur, as noted with regard  
to hypercorrection and other sorts of rule-governed behavior).

--Brian

Brian D. Joseph
The Ohio State University

Quoting Andrew Garrett <garrett at berkeley.edu>:

> While I am disposed by training and impulse to agree fully with Brian's
> comments, about the distinction between Neogrammarian sound change (or
> sound change proper) and other types of changes in sounds, I think the
> jury is actually still out on one crucial question. This is the
> question of precisely when regularity actually emerges in the series of
> unfolding events that we retrospectively see as Neogrammarian sound
> change. The view that was maybe implicit in some Neogrammarians'
> thinking, and that I have taught from time to time, is that the
> regularity effect arises in innovation and not in diffusion -- for
> example, that the very first person to have a changed pronunciation
> will (presumably) vary in whether s/he uses the changed pronunciation
> on any particular occasion, but that it will be just as likely to
> appear in all words that conform to the relevant phonological
> generalization. An alternative view is that the first person to have a
> changed pronunciation really does use it only in one word but not in
> other words that are phonologically equivalent, and that the regularity
> effect arises somehow in the process of sociolinguistic diffusion, i.e.
> as the phonetic effect (sound change) acquires some sociolinguistic
> traction. This view was propounded as early as 1901 by Benjamin Ide
> Wheeler (a Neogrammarian student and incidentally at the time the
> President of the University of California); the obvious disadvantage of
> this view is that it becomes unclear why we should (at least usually)
> encounter the regularity effect, but it's a view that finds a congenial
> home in much present-day sociophonetics. To my mind, the question
> remains unresolved.
>
> The relevance of the question is that if Neogrammarian sound change,
> i.e. the regularity effect, only arises in the course of
> sociolinguistic diffusion, then it is not so clear that there is
> actually a meaningful distinction between Neogrammarian sound change
> and socially driven change (assuming that we allow the latter to be
> "unconscious"). Indeed, Wheeler's own explanation of the regularity
> effect invoked precisely what Sally referred to as correspondence rules
> (borrowing routines): his idea was that (say) you produce a fronted /u/
> in just one word, for example "cool", I hear that fronted /u/ in your
> speech and I know it corresponds to a non-fronted /u/ in my speech, I
> think you're worth imitating, and I generalize a pattern whereby
> non-fronted /u/ corresponds to fronted /u/ more generally. This
> approach also has obvious analogs in more recent Labovian work. I don't
> mean to say that I agree with this analysis, but it strikes me that we
> don't yet know where the truth lies.
>
> -- Andrew
>
> Andrew Garrett
> UC Berkeley
>> ------------------------------
>>
>> Message: 3
>> Date: Sat, 26 Jan 2008 09:51:20 -0500 (EST)
>> From: Brian Joseph <bjoseph at ling.ohio-state.edu>
>> Subject: Re: [Histling-l] 'Reversed change' dialect borrowing
>> To: thomason at umich.edu (Sally Thomason)
>> Cc: histling-l at mailman.rice.edu
>>
>> Greetings -- this discussion is most interesting, and I would
>> like to add just a few observations; even if they are not directly
>> on target, I think this is an appropriate forum to mention them
>> and they are in any case provoked by the recent postings.
>>
>> First, even though no one has actually said this, so in a
>> sense I am either responding to a straw man or preaching to
>> the choir, let me note that this discussion shows why it is
>> useful to be precise about what we mean by "sound change". In the   
>> cases brought out here, the event that was reversed was
>> a Neogrammarian-style (i.e. regular and phonetically driven)
>> sound change, what I like to call "sound change proper" or
>> "sound change in the strict sense" (or even simply "Neogrammarian
>> sound change") whereas the reversing event is a socially-driven
>> change.  Terminology is always tricky to be sure, but I would love   
>> it if the term "sound change" were restricted to "sound
>> change proper" and some other term were invented for "changes in
>> sounds that are not sound change (in the strict sense)".  I often
>> tell my classes that as paradoxically as it might seem, not every
>> change in the sounds of a word is a matter of sound change (analogical
>> change, clippings, tabu deformation, and so on are all the sorts of
>> "other events" that can lead to changes in the pronunciation of a
>> word that are not Neogrammarian sound change in the strict sense).
>>
>> Second, socially-driven events of change *can* show regularity (though
>> they need not), but typically lack the phonetic basis that is a hallmark
>> of Neogrammarian sound change.  We see this sort of thing in   
>> hypercorrection all the time (Peter's.
>>
>> Finally, it reminds me of what I said at my ICHL presentation this past
>> summer, namely that when we get right down to it, social factors can
>> trump everything else in change (or everything except for the most basic
>> immutable universal foundational aspects of language, to the extent there
>> are any and if there are to the extent we can identify them).
>>
>> --Brian
>>
>> Brian D. Joseph
>> The Ohio State University
>
> _______________________________________________
> Histling-l mailing list
> Histling-l at mailman.rice.edu
> https://mailman.rice.edu/mailman/listinfo/histling-l



----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.

_______________________________________________
Histling-l mailing list
Histling-l at mailman.rice.edu
https://mailman.rice.edu/mailman/listinfo/histling-l