Folie 1

1
Typology and the history of languages
Bernd Heine Lyon, 12 May 2008
2
Typology and the history of African languages

• A gloomy past
• The history of African linguistics, at least up
  until the 1960s, suggests that linguistic
  typology has played a role in reconstructing
  language history that was not really beneficial.
  Consider a few selected examples
• (1) Morphological typology was used to justify
  not only genetic classifications but also
  language evolution.

3
Typology and the history of languages

• Carl Meinhof imposed the 19th century typology on
  African languages
• Isolating languages Sudansprachen
• Agglutinating languages Bantusprachen
• Inflectional languages Hamitensprachen
• Meinhof, Carl 1910. Die moderne
  Sprachforschung in Afrika. Berlin.

4

• Westermanns (1935) notion Sudansprachen
• A Labial-velars (kp, gb)
• B Lexical tone
• C Structure CV in word stems
• D Lack of morphology, reliance on syntax
• E Preposed objects (SOV), preposed genitive,
  postpositions
• F Dative expressed by give
• G Widespread S V O word order
• H Nominal plural formed by third person plural
  pronoun, also by reduplication
• I Lack of grammatical gender, animate vs.
  inanimate instead
• J Lack of passives
• K Use of plural-forming pronouns i and a
• L Use of a- to form de-verbal nouns.
• Westermann, Diedrich 1935. Charakter und
  Einteilung der Sudansprachen. Africa 8, 2
  129-48.

5

• Meinhofs hypothesis was adopted by subsequent
  generations of Africanists, but he himself went
  even one step further He proposed an
  evolutionary scenario according to which Africas
  isolating languages constitute the first stage of
  linguistic evolution, followed by the
  agglutinating stage, and the peak of linguistic
  evolution was reached when people were blessed
  enough to invent inflection in Africa and
  elsewhere.
• Meinhof, Carl 1910. Die moderne
  Sprachforschung in Afrika. Berlin.
• ----- 1912. Die Sprachen der Hamiten.
  Hamburg.

6

• (b) A. N. Tuckers classification of the Kuliak
  languages (Ik) as being related to Ancient
  Egyptian rested on the basis of typological
  similarities in the paradigm of personal
  pronouns.
• Tucker, A. N. 1967. Erythraic elements and
  patternings Some East African findings.
  African Language Review 6 17-25.

7

• (2) Presence vs. absence of a sex-based gender
  system has been taken to suggest questionable
  genetic classifications, such as following
• Carl Meinhofs (1912) classification of Hamitic
  languages, where the presence of a gender system
  was taken as criterial for membership in this
  family.
• Meinhof, Carl 1912. Die Sprachen der Hamiten.
  Hamburg.

8

• (b) Bernhard Struck and Diedrich Westermanns
  distinction between Niloto-Sudanic and
  Hamito-Sudanic, and subsequently the split by
  Tucker and Bryan (1956 1966) between Nilotic
  ( Western Nilotic) and Paranilotic ( Eastern
  and Southern Nilotic).
• Westermann, Diedrich 1940. Die Sprachen
  Afrikas. In H. Baumann, R. Thurnwald D.
  Westermann 1940, Völkerkunde Afrikas. Essen. Pp.
  375-453.
• Tucker, A. N., Margaret Bryan W. Leslau 1966.
  Linguistic analyses. Handbook of African
  Languages Oxford University Press.

9

• (c) The genetic division between Chado-Hamitic
  and Chadic languages by Johannes Lukas was
  motivated to quite some extent by the fact that
  the former have, while the latter lack a gender
  system.
• Lukas, Johannes 1939. Die Verbreitung der
  Hamiten in Afrika. Scientia 33 108-18.

10
A brighter future?

• What then is the present role of language
  typology in reconstructing Africas linguistic
  history?
• One major innovation We are now able to
  distinguish between two historical factors
  genetic and areal relationship.

11
We have a fairly good overview of the genetic
relationship patterns among the languages of
Africa. But what about areal relationship?
Greenberg, J. H. 1963. The languages of Africa.
The Hague Mouton.
12

• Previous work (Greenberg 1959 1983 Larochette
  1959 Welmers 1974 Gregersen 1977 Meeussen
  1975) was characterized by a search for what
  following Meeussen (1975) we call Africanisms
  (Heine Leyew 2008). With this term we are
  referring to properties that satisfy the
  following set of criteria
• African areal properties according
• to Greenberg (1983 3) are "those
• which are either exclusive to Africa,
• though not found everywhere within it,
• or those which are especially common
• in Africa although not confined to that
• continent. Joseph Greenberg

Greenberg, Joseph H. 1959. Africa as a linguistic
area. In Bascom and Herskovits (eds.) 1959. Pp.
15-27.
13

• The questions one may ask relate on the one hand
  to a comparison of different kinds of linguistic
  relationship
• Typology as a diagnostic
• Does typology reflect genetic relationship?
• Sometimes yes, sometimes no.
• - Does typology reflect areal relationship?
  Sometimes yes, sometimes no.

14

• Typology as a methodological tool
• - Can typology be of help in discovering genetic
  relationship? The answer is flatly no typology
  is more of an obstacle than a useful tool for
  genetic reconstruction.
• Can typology be of help in discovering areal
  relationship? Yes, it is the most important tool
  for the reconstruction of areal relationship.
• BUT There is no one-to-one relationship between
  typological and areal structuring.

15

• Typology and areal classification

16

• Characterization of a linguistic area
  (sprachbund)
• a There are a number of languages spoken in one
  and the same general area.
• b The languages share a set of linguistic
  features whose presence can be explained with
  reference to neither genetic relationship, drift,
  universal constraints on language structure or
  language development, nor to chance.
• c This set of features is not found in languages
  outside the area.
• d On account of (b), the presence of these
  features must be the result of language
  contact.
• Heine, Bernd and Derek Nurse (eds.)
  2008. A linguistic geography of Africa.
  Cambridge Cambidge University Press.

17

• Widely recognized sprachbund-like units
• - the Balkans (for convenient summaries, see
  e.g. Joseph 1992 Feuillet 2001),
• - Meso-America (Campbell et al. 1986),
• - Ethiopia (Ferguson 1976 for a critical
  review, see Tosco 2000),
• - South Asia (Masica 1976 Emeneau 1980 Ebert
  2001),
• - the East Arnhem Land (Heath 1978 1981),
• - the Amerindian Pacific Northwest Coast (Jacobs
  1954 Sherzer 1976 Beck 2000),
• - the Vaupés basin of northwest Amazonia
  (Aikhenvald 1996 2002 2003),
• - Standard Average European (Haspelmath 1998
  2001),
• - and the Daly River area of Australia (Dixon
  2002 674-9).

18

• Linguistic areas in Africa?

19
Phonological zones in Africa (Clements Rialland
2008)
20
The Macro-Sudan Belt labial-velar consonants
(Güldemann 2008)
21

• The marked-nominative languages of Africa
  (König 2008)

22
The Tanzanian Rift Valley area (Kießling, Mous
and Nurse 2008)
23

• Problems notoriously associated with sprachbunds
• The isoglosses that make up sprachbunds do not
  all coincide. Which isoglosses should be included
  or excluded is a matter of much dispute.
• Sprachbunds are hypothesized to be the product of
  events in the past. Unlike genetic groupings,
  however, their potential for reconstructing
  history is limited hardly any one of them has
  contributed significantly to a better
  understanding of the history of the area
  concerned.
• Their contribution to synchronic linguistics has
  also been moderate.

24

• MetatypyAn ideal linguistic area would be one
  having in particular the following
  characteristics
• (a) The languages making up the area share the
  same organization of semantic structure.
• (b) They also share the same patterns in which
  morphemes are concatenated to form sentences,
  phrases, and words.
• (c) The grammatical constructions are equivalent
  across the languages of the area.

25

• (d) What distinguishes the languages is that each
  uses different forms but, on account of (a) and
  (b), each form has an exact structural and
  semantic equivalent in the other languages.
• (e) Accordingly, the languages are
  intertranslatable to the extent that the task of
  the translator or language learner is simply
  confined to inserting the appropriate lexical and
  grammatical forms to move from one language of
  the area to another.
• (f) The linguistic area is the result of language
  contact, that is, of a clearly definable
  historical process of linguistic assimilation.

26

• Linguistic communities that are said to have
  undergone metatypy (Ross 1997 146)
• The Indian village Kupwar (Gumperz Wilson 1971)
• Northwestern New Britain, New Guinea (Thurston
  1987 1982)
• The Gangou dialect of Chinese and the Mongolic
  language Minhe Monguor (Yongzhong, Chuluu,
  Slater, Stuart 1997)
• Arvanítika, the Albanian dialects spoken in
  central Greece, and Greek (Sasse 1985)
• The Oceanic language Takia and the Papuan
  language Waskia of Papua New Guinea (Ross 1996
  2001).

27

• Metatypy situations compared to sprachbunds
• (1) Metatypic situations are characterized by a
  much more pervasive degree of structural
  isomorphism.
• (2) They usually consist of a severely limited
  number of languages, sometimes no more than two
  (even though there is no theoretical limit as to
  how many languages are required).
• (3) There usually is fairly detailed information
  on the sociolinguistic and historical factors
  that contributed to the rise of metatypy.
• (4) While directionality in linguistic diffusion
  is frequently a controversial issue in the case
  of sprachbunds, in most cases of metatypy there
  is solid information on the patterns and
  directionality of linguistic transfer across
  languages.

28

• Resumé
• Many situations in Africa are suggestive of
  metatypy but, to my knowledge, no clear case has
  been documented so far.

29
Any questions?