Two languages, one grammar?

Another tidbit from the same book I've been stealing from lately
(>Man's Many Voices<; see "stylemes" posting for ref info):

# John Gumperz has examined the colloquial dialects of Marathi and Kannada
# in a village along the Maharastra-Mysore boundary in central India where
# these two languages come into direct contact.  Marathi is an Indo-Aryan
# language, while Kannada is Dravidian.  Historically these two languages
# go back to utterly different antecedents, but the Indo-Aryan and Dravidian
# languages have been in contact in India for several thousand years and have
# long influenced one another.  Along the borders their mutual influence has
# been profound.  In the village studied by Gumperz most speakers feel
# themselves to be bilingual, but the two village dialects share such a
# large part of their grammar that one can almost doubt whether they should
# count as separate languages.  Consider, for example, the following sentence:
#
# Kannada:  hog- i      w@nd kudri turg  maR- i      aw    t@nd
# Tags:     verb suff.  adj. noun  noun  verb suff.  pron. verb
# Marathi:  ja-  un     ek   ghoRa cori  kar- un     tew   anla
# English:  go   having one  horse theft take having he    brought
# Idiomatic English: Having gone and having stolen a horse,
#	he brought it back.
#
# All of the morphemes of the Kannada sentence are different from those of
# the Marathi sentence, but they are used according to identical grammatical
# principles.  The sentences have identical constituent structures and their
# morphemes occur in the same order.  The same kind of suffixes are attached
# to the same kind of bases.  These sentences seem by no means to be atypical
# of village usage.  In fact, one can plausibly suggest that these two
# languages (if indeed they >are< two languages) have the same grammar and
# differ only in the items filling the surface forms.  One can translate from
# one language to another simply by substituting one set of lexical items for
# another in the surface structure.
#
# Both the Marathi and the Kannada used in this village differ from the more
# literary or educated styles of the same languages, but both can be shown to
# be related to the more standard forms according to the usual criteria by
# which linguists recognize genetic affiliation.  Yet the village dialects
# have undergone such profound mutual grammatical influence as to almost
# obscure the boundaries between the two languages.  Curiously, in this case,
# it is the lexicon that maintains the separation, and after considering the
# effect of Marathi and Kannada upon each other, one can hardly maintain that
# lexicon is always the easiest component of language to borrow or that the
# true genetic affiliation will necessarily be shown by the underlying grammar.

Me again.  The relationship of village-Marathi to village-Kannada is oddly
like the relation of Lojban to (other forms of) Loglan; shared grammar, utterly
divergent lexicon.  While the history is quite different, the synchronic
situation is very much the same!

-- 
John Cowan		sharing account <lojbab@access.digex.net> for now
		e'osai ko sarji la lojban.