Within grammaticalization theory, progression on the function word > clitic
> affix cline is associated with a number of interdependent
morpho-syntactic, functional, and phonological processes. The latter can be
further subdivided into adaptation and erosion, of which especially erosion
has received considerable attention. Drawing data from a nineteen language
sample, this study aims at testing the predictions made in the literature
with respect to the correlation of cliticization and erosion.
In order to systematize the various prosodic and segmental clines
encountered in cliticization, this study establishes and defends a
rhythm-based typology of language which relies on ten parameters in
prosody, phonotactics and morphophonology. The rhythm based typology of
language allows for a number of significant predictions with respect to the
distribution of the various segmental effects of cliticization. In
stress-based languages, stress reduction and tone neutralization go hand in
hand with vowel reduction and deletion in unstressed syllables. Since such
languages exhibit a high degree of syllable complexity, junctural consonant
clusters are likely and we can encounter certain processes applying in this
context. Mora- and syllable-based languages, on the other hand, do not show
vowel reduction and deletion in unstressed syllables. Accordingly, the
unstressed vowels of clitics will be preserved or harmonized, but crucially
not reduced and deleted. Due to the low degrees of syllable complexity in
these languages, junctural vowel clusters and associated processes are
likely. Ultimately, this typology predicts different pathways for the
evolution of morphology in the different phonological climates. Whereas
morphologization in stress-based languages is accompanied by heavy
reduction and leads to subminimal morphological markers, morphologization
in mora- and syllable-based languages results in polysyllabic markers due
to the lack of erosion in grammaticalization.
The evidence compiled in this study calls for a serious reconsideration of
the role of phonology in grammaticalization. Since erosion is not a
universal concomitant of grammaticalization it cannot be considered a
defining property. A subtler conception of grammaticalization has to
incorporate the finding that associated sub-processes are subject to
cross-linguistic variation, in our case linguistic rhythm. Accordingly, the
findings of this study cast doubt on universal scenarios for language
change such as the one enshrined in ‘grammaticalization theory’.
|