Reworking on LANGUAGE AREA HYPOTHESIS:

1
Reworking on LANGUAGE AREA HYPOTHESIS
consequences
Emeneau CentenaryInternational Conference Jan
1-4, 2005 CIIL, Mysore

• Udaya narayana singh
• Ciil, mysore

2
The Concerns

• To this audience, I need not underscore the
  importance of the historical moment when Prof.
  Emeneau had made his observations on India as a
  linguistic area.
• Many scholars, particularly, Colin Masica and
  others, have expanded it further.
• Prof. Ashok Kelkar has already spoken about its
  historical context.
• And Prof K.V. Subbarao in his keynote address
  will give us specific instances of the predictive
  power of this hypothesis.
• But of late, many question marks have been raised
  about the tenability of the hypothesis.
• While that is bound to happen in an emerging area
  and in an eveolving discipline like Linguistics,
  we need to be careful about possible
  over-generalization and consequent dangers.
• Lets look at some of these concerns

3
Definational problems

• We all know that linguistic areas are infamously
  messy things, with notoriously fuzzy pictures
  (Thomason Kaufman 198895, Heine
  and Kuteva 2001396, Tosco 2000332).

• Ever since Trubetzkoy (1928) floated the idea in
  his Proposition 16  (Acts of the First
  International Congress of Linguists 17-18 
  Leiden), our understanding about linguistic areas
  is depressingly meager.
• The difficulties were brought out very clearly by
  Sarah G Thomason (2001) in Language Contact An
  Introduction.
• First, it was difficult to decide whether a
  particular region constitutes a linguistic area
  or not.
• Secondly, in spite of prolonged efforts to define
  linguistic area, there is no general agreement on
  it.
• Even for the most widely accepted linguistic
  areas, scholars do not agree wholly on which
  languages belong to the area,  what linguistic
  traits characterize the area, and what its
  precise geographical extent is, says Lyle
  Campbell in Areal Linguistics  a Closer
  Scrutiny.
• There are other problems, too.

4

• For instance, K. H. Ebert L. Neukom1 (2000) as
  well as Ebert2 (2001) claimed that if we
  seriously evaluated the features of the South
  Asian sprachbund, we would find that

• (i) Many of the alleged sprachbund features
  also found in Turkic and Mongolian languages of
  Central Asia
• (ii) They seldom spread to Munda and
  Tibeto-Burman languages of South Asia
• (iii) It is a fact that many smaller languages
  of the area which may hold important clue to the
  phenomenon still remain undocumented
• (iv) Instead of talking about a common
  linguistic area, it may seem more profitable and
  viable to look at sub-areas
• (v) There seems to be a more or less sharp
  dividing line around the 84th meridian. Languages
  to the East of this line show many Dravidian and
  Austroasiatic traits, but few of the sprachbund
  features.
• Thats why Assamese, Bengali, Nepali, Oriya
  adopted many features from neighboring Austric
  and Tibeto-Burman that must have outnumbered them
  once.

--------------------------------------------------
--------------------------------------------------
--------------------------------------------------
---------------- 1. Ebert Neukom. 2000. Towards
an areal typology of relative clauses in South
Asian languages. ASAS WP No. 9 2. K. H. Ebert
(2001) Südasien als Sprachbund. Handbücher zur
Sprach- und Kommunkationswissenschaft. Bd. 11.2
Language Typology and Language Universals.
Berlin de Gruyter. 1529-1539
5
The Convergence Project

• The project 'Convergence and Linguistic Areas' of
  UK (from Sept 2003) tries to connect to ongoing
  discussions in areal linguistics and areal
  typology by raising the following questions
• Are some language functions specially prone to
  convergence in linguistic areas?
• What is the relation between structural borrowing
  (of actual forms) and convergence (adaptation of
  whole structural, or the replication of a
  pattern, without replicating a form)?
• How do borrowing hierarchies relate to
  hierarchies of convergence?
• How can a theoretical notion of convergence
  accommodate both shared typological features
  (that are a part of language structure), and
  features that emerge because of strategies of
  bilingual speech (language function features)?

• Professor Yaron Matras Romani, Languages of the
  Middle East, German dialects
• Dr Jeanette Sakel Amazonian languages (in
  particular Mosetenan), Greenlandic
• Samantha Truman Uralic languages
• http//lings.ln.man.ac.uk/Info/staff/JS/JScon.html
  

6
FRANZ BOAS HIS THEORY

• To recollect, in many ways, Franz Boas (1917,
  1920, 1929) was the scholar who had initiated
  areal linguistics.
• Boas talked about acculturation and
  absorption when he had a difficulty in
  differentiating between what was inherited and
  what was diffused.
• This was the beginning of the areal-typological
  approach.
• Boas compared the structural traits of languages
  in a particular region with their neighbors to
  determine this possible diffusion.
• It was this line of thinking had influenced the
  Prague school and Trubetzkoy. 

7
The language union concept

• The seeds of the concept of language area (called
  language union) could be seen in Trubetzkoy
  (1923), where he had said
• It happens that several languages in a region
  defined in terms of geography and cultural
  history acquire features of a particular
  congruence, irrespective of whether this
  congruence is determined by common origin or only
  by a prolonged proximity in time and parallel
  development.  We propose the term language union
  (jazykovoj sojuz) for such groups which are not
  based on the genetic principle (Trubetzkoy
  1923116, quoted from Toman 1995204.)
• Notice that even at this stage, area or space
  does not figure anywhere.

8
Sprachgruppe Sprachbünde

• Trubetzkoy (1928) talks of Sprachgruppe which he
  called a collection of languages bound to one
  another by a number of systematic agreements.
• He further sub-divided it into two types,
  families of genetically related languages and
  Sprachbünde.
• What is Sprachbünde? It is a group of languages
  which show a high degree of similarity with
  respect to syntax, similarity in the principles
  of morphological construction, and which offer a
  large number of common culture words, sometimes
  also an outward similarity in the phonological
  inventories.
• But they possess neither systematic sound
  correspondences, nor have any correspondences in
  the phonological make up of the morphological
  units nor any common basic lexical items.

9
Arguments against Morphological Borrowing

• If the Indian linguistic area has to fit into
  this mould that Boas-Trubetzkoy-Emeneau had been
  trying to evolve, we must be able to prove,
  beyond doubt, that morphological processes or
  hitherto unavailable syntactic patterns travel
  across unrelated languages in this sub-continent.
  
• But many scholars (including Rajendra Singh) have
  put a question mark on the usual examples that
  are advanced as morphological borrowing.
• So, the point would be to discover examples that
  would satisfy such objections.
• As for syntax, it is difficult to push the point
  beyond a certain limit as earlier texts and
  inscriptions are woefully inadequate corpus about
  which anything can be said conclusively. Unless a
  construction has a high frequency rate, it is
  unlikely to pop up in a text like Caryaapada or
  Varna-ratnaakara.

10
Northwest Coast Sprachbund ?

• Building on the earlier work of Frachtenberg
  (1920) and Morris Swadesh (1953a,b), David Beck
  (U Alberta) in his essay titled Grammatical
  Convergence and the Genesis of Diversity in the
  Northwest Coast Sprachbund (Anthropological
  Linguistics 42.2 1 - 67) claims that Mosan in
  the NW coast of USA boasts one of the world's
  most extensive Sprachbunds and the many languages
  of the area have engendered proposals for a
  number of controversial genetic phyla, including
  the Mosan phylum uniting the Salish, Wakashan,
  and Chimakuan language families.

Then there are claims like the NWC Sprachbund.
Such claims are many.
How does one resolve them is the question.
11
The Best-known Case So Far

• The Balkan (Alexandru Rosetti) as a linguistic
  area with Albanian, Modern Greek, several
  Romance languages and at least three Slavic
  languages (Bulgarian, Macedonian Serbian
  Torlak) has been long known as a model of
  structural similarities due to geographic
  proximity and shared living space
  (www.e-paranoids.com).

• But many questions are being asked about such
  spaces, too.
• Partly because of the fact that languages do not
  remain constant, they keep drifting away.
• But partly also because one is discovering many
  mismatches later as we grow in our understanding
  of the space.

12
Models

• All these doubts in the mind of todays scholars
  often raise this question whether the Linguistic
  area concept is still valid today?
• And if so, how do we understand the concept and
  make use of it at a time when there seems to be
  explosion of data from many hitherto unknown
  languages from within the same space
  challenging all neat and mutually exclusive
  configura-tions and classifications.
• It is common to find this kind of grey areas
  where both traits prevail.

13
But there are possible models that are more
complicated than that
But linguists have so far been very hesitant to
admit a configuration like these, fearing the
collapse of neat classifications
14
To Conclude

• But now that we know that all notions that were
  so far taken as given and therefore
  unquestionable (beginning from Subjecthood to
  neat categories like SVO, SOV and VSO, I.,e.
  subject precedence, or even genetic classes in
  many cases), it is important for us to think
  about more complicated landscapes for human
  languages.
• It is possible that the same space has different
  kinds of configurations depending on whether we
  are looking at the linguistic area or literary
  area or semantic area.

Thank you!