LINGUIST List 7.1170

Sun Aug 18 1996

Disc: Grammaticalization

Editor for this issue: Anthony Rodrigues Aristar <aristartam2000.tamu.edu>

Directory

1. Martin Haspelmath, Disc: Grammaticalization
2. Oesten Dahl, Re: 7.1158, Sum: Grammaticalization

Frederick Newmeyer very boldly challenges a number of widespread 
assumptions concerning grammaticalization, which I hope will spark the 
debate that these issues deserve. 

Newmeyer is obviously right that reconstructed changes cannot be taken as 
evidence for a theory of language change, so we must be very careful. 
However, the directionality of change is very often evident even if the 
change is reconstructed; and, most importantly, certain changes recur 
again and again in languages, so the sheer mass of cases counts in favor 
of certain reconstructions. If a different direction of change is much 
less plausible in a large number of cases, you begin to feel very 
confident. However, the importance of the work on African languages comes 
mainly from the fact that it shows that the kinds of changes attested in 
European languages are not genetically or areally restrcited.

But Newmeyer is wrong in denying unidirectionality. His examples of 
P-to-V changes mostly show conversion (or zero-derivation), i.e. a 
word-formation process, which is different from ordinary change. I.e. by 
coining a verb 'to down' from the adverb 'down', we add a new lexeme to 
the language. It would be quite wrong to say that the adverb 'down' turns 
into a verb 'down'. Unidirectionality thus applies only to ordinary 
changes, not to extensions of the lexicon by means of productive 
word-formation. (Though of course even here true examples of P-derived 
verbs are rare, cf. '*to for', '*to of', '*to with'; 'down', 'up', 
'behind' etc. are primarily adverbs.)

In fact, unidirectionality is one of the main reasons why 
grammaticalization is so important (I don't care if you call it 
an epiphenomenon--which it probably is, like all complex phenomena). 
By denying the unity of grammaticalization changes, one would not be able 
to account for the fact that semantic and formal changes generally go 
hand in hand, i.e. units become desemanticized and phonologically eroded 
at the same time. 
I agree with Newmeyer that grammaticalization needs to be explained with 
reference to factors such as speech production and perception (i.e. 
'performance'), but somebody has to do it (just like some physicist has 
to come up with a theory for explaining the mixing behavior of ice 
cream) -- that is, we need a theory of grammaticalization. If linguists look 
only at competence, they cannot understand the vast majority of 
language changes, and a large portion of synchronic language structure. 

To my mind, the existing functionalist literature on grammaticalization 
has not been very good at explaining the unidirectionality of most of 
language change. But the Chomskyan literature has not even recognized the 
unidirectionality of most of language change. In their recent book 
"Clause structure and language change", Battye and Roberts say that 
"language change is essentially a random walk through the space of 
possible parameter settings". If this were the case, why is 
grammaticalization so pervasive?

Martin Haspelmath (Free University of Berlin)

Message 2: Re: 7.1158, Sum: Grammaticalization

Fritz Newmeyer, like various people before him, question the
unidirectionality of grammaticalization.
I think grammaticalization is unidirectional in about the same sense as
biological processes such as growth, maturation, and ageing are. As we grow
up, we become taller; in old age, we may shrink a little. However, we would
not expect a child to start becoming shorter and shorter and finally return
to its mother's womb. Similarly, eyesight generally deteriorates with age,
but myopic persons may actually become less so due to their eye lenses
getting more rigid and compensating the myopia. In other words, the
biological processes that take place during our lives sometimes give rise to
contradictory results but there can be no doubt that they are basically
irreversible. In the same way, we would not expect, say, that the French
future tense endings would start separating from the verb, become
auxiliaries and then end up as full verbs meaning "to keep" (the presumed
etymology of Latin habere, the source of the French future.)
Fritz also says:
>Second, I am skeptical that even exists a phenomenon called
>'grammaticalization'. There are semantic changes, some of which are
>more natural than others. And there are phonological/phonetic
>changes, some of which are more natural than others. I see little
>value in dignifying the intersection of one subset of the former and
>one subset of the latter with a label, and calling it a 'process' that
>needs a 'theory' to explain it. Indeed, the fact that the semantic
>processes (metaphor, metonymy), the morphosyntactic processes
>(reanalysis), and the phonetic reductions can occur independently of
>each other -- and often do -- is a powerful argument for taking an
>epiphenomenonal approach to grammaticalization -- i.e. not regarding
>it as a distinct process.
This to me seems like saying that since love and sex can occur without each
other, they are totally different phenomena. For Fritz' argument to go
through, he would have to show not only that the processes can occur
independently but also that they are unrelated even in the well-documented
cases when they show up together. What some of us have claimed is that the
things that happen in grammaticalization do so in an orderly fashion which
not only predicts what changes can occur but also puts constraints on what
synchronic grammatical systems are found. The fascination of
grammaticalization studies is precisely that it opens up a way of explaining
grammatical phenomena that has largely been neglected in post-Saussurean
linguistics. The difficulty in fitting this way of thinking into the
Chomskyan paradigm may explain why some people react negatively to it.

- oesten dahl