Folie 1

1
Using findings on language contact for
linguistic reconstruction Bernd
Heine Lyon, 13 May 2008
2

• The grammaticalization of reflexives
• in Africa grammaticalization areas

3

• Mauritian (Morisien, French-based creole Carden,
  1993 107)
• a Zorz fin pan?di
  li. (him)
• b Zorz fin pan?di limem.
  (himself)
• c Zorz fin pan?di so
  lekor. (his body)
• George COMPL hang himself
• 'George hanged himself.

4
Main strategies to develop reflexive markers
Label Strategy
P Pronoun strategy uR unmarked reflexive Use personal pronouns
I Intensifier strategy Add an intensifier to P
N Noun strategy Use a body-noun
X Non-transparent (opaque) reflexives
5

• Mauritian (Morisien, French-based creole Carden,
  1993 107)
• a Zorz fin pan?di li.
  Pronoun s.
• b Zorz fin pan?di limem.
  Intensifier s.
• c Zorz fin pan?di so lekor.
  Noun s.
• George COMPL hang himself
• 'George hanged himself.

6

• Intensifiers (variously referred to as an
  emphatic reflexive, intensive reflexive,
  adverbial reflexive, appositive pronoun, emphatic
  pronoun, emphasizing clitic, intensive reflexive,
  or identifier), serve mainly to evoke
  alternatives to the focus they refer to, in
  particular
• To present an unexpected participant
  (paraphrasable by even),
• To present a participant in addition to the
  expected one (also),
• To exclude other possible participants (only),
• To express a contrast vis-a-vis other possible
  participants ('X, rather than Y'),
• To express emphasis or discourse prominence.
• König, Ekkehard and Peter Siemund 2000.
  Intensifiers and reflexives A typological
  perspective. In Frajzyngier, Zygmunt and Traci
  S. Curl (eds.) 2000. Reflexives Forms and
  functions. Amsterdam, Philadelphia Benjamins.
  Pp. 41-74.

7
Africa The noun strategy (N)

• Efik (Benue-Congo, Niger-Congo Essien 1974 11,
  14)
• a Árìt éye? ídém.
• Arit has body
• 'Arit has a beautiful body.'
• b Árìt óyòm n?díwòt ídém ésie?.
• Arit want kill body her
• 'Arit wants to kill herself.
• Yoruba (Kwa, Niger-Congo Awolaye 1986 4)
• Nwosu rí ara r?.
• Nwosu saw body his
• Nwosu saw himself.

8
Nominal sources of reflexive markers in African
languages (Sample 46 African languages, 49
forms).
Nominal source Frequency Percentage
body 25 61.0
head 6 14.6
soul/life 4 9.8
Other body parts (e.g. skin, heart) 6 14.6
Total 41 100
9
Nominal sources of reflexive markers across the
world (Sample 89 languages Schladt 2000 112).
Nominal source Frequency Percentage
body 71 79.8
head 13 14.6
Other body parts 5 5.6
Total 89 100
10
Nominal sources for reflexive markers in African
languages
11
The Western Sahel head-reflexives

• Fulani or Peul (Atlantic, Niger-Congo
  Klingenheben 1963 141)
• ?o mbari hoore maako.
• he killed head his
• He killed himself.
• Koyra Chiini (Songhai, Nilo-Saharan Heath 1999
  329-31)
• yer ta bine nda yer tun nda
• 1.PL TOP TOP if 1.PL.S arise with
• yer bomo kul, .
• 1.PL head all
• As for us, if we get up by ourselves, .

12

• Hausa (Chadic Afro-Asiatic Kraft Kirk-Greene
  1973231)
• Sun kashè ka?n- sù.
• they kill head- their
• They have killed themselves.
• (Lit. they have killed their head)

13

• Possible evidence for a diffusion hypothesis is
  provided by the following observation Whereas
  Fulani (Peul) has only one reflexive marker,
  derived from the noun hoore head, many of the
  languages spoken in the Western Sahel belt have
  two markers, where one is a head-reflexive and
  the other a body-reflexive (Lele, Margi, Mina,
  Pero) or an unmarked reflexive (Koyraboro, Koyra
  Chiini).

14
The languages of the Western Sahel

• These languages are spoken in one specific area,
  the sub-Saharan belt of West Africa, roughly
  between Senegal and Cameroon, and they belong to
  three different language phyla Niger-Congo,
  Afroasiatic, and Nilo-Saharan. Niger-Congo
  languages are Fulani and Diola (West Atlantic),
  while Afroasiatic is represented with two of its
  branches, namely Chadic (Hausa, Margi, Mina,
  Pero, Kwami, and Lele) and Berber (Tamazight
  Schladt 2000). The only Nilo-Saharan languages to
  be found in West Africa are those of the Songhai
  group and both, Koyraboro and Koyra Chiini, have
  a head-reflexive, both being spoken in the
  Western Sahel (Heath 1999a 1999b).

15
Possible motivating forces

• Two main historical forces appear to have
  contributed to this areal clustering. At least
  eight hundred years ago, Fulani people from the
  Senegal region of the extreme west end of Africa
  began to migrate eastward across the sahel belt
  in search for new pastures for their cattle,
  founding empires and extending their sphere of
  influence up to the Lake Chad region. The Western
  Sahel belt, where there is a clustering of
  languages having a head-reflexive, roughly
  coincides with that of the Fulani expansion. A
  second factor was presumably the trans-Saharan
  trade where nations in the Western Sahel played
  an important role. Finally, the spread of Islam
  in western Africa may also have contributed to
  the process.

16
Nominal sources for reflexive markers in African
languages
17
The Eastern Sahel soul-reflexives

• Languages using a noun whose meanings include
  soul and/or life as a source cluster
  significantly in north-central and north-eastern
  Africa, roughly between Lake Chad and River Nile.
  
• How come?

18

• a Arabic is widely spoken as a first or second
  language in the region.
• b Arabic varieties spoken in this general region
  make use of noun stems for soul or life (nefs
  or ru?u) to express reflexivity.
• c A language spoken in the Eastern Sahel belt
  between Lad Chad and River Nile is likely to have
  a life/soul- reflexive, in addition to some
  other reflexive marker.

19

• Examples
• In Maba there is ndu skin next to néfès soul,
  person (Trenga 1947 64-5).
• In the Saharan language Teda-Daza, the noun ?ro
  life plus possessive adjective serves as a
  reflexive marker, in addition to the body-nouns
  kasar body and daho head (Le Coeur 1956 94).
• In Nile Nubian there is áy heart next to
  newerti life, soul as reflexive markers (Werner
  1987 128).

20

• Maba (Nilo-Saharan Trenga 1947 64)
• ti néfès ténèn tuia?.
• he soul his killed
• He has killed himself.
• Trenga, Georges 1947. Le bura-mabang du
  Ouadai notes pour servir à létude de la
  langue maba. (Travaux et Mémoires de
  lInstitut dEthnologie, 39.) Paris  Institut
  dEthnologie.

21

• Diffusion in this region does not appear to be
  restricted to grammatical replication ( transfer
  of meaning or structure) but also involved
  borrowing ( transfer of form-meaning units).

22

• Chimwini (Bantu, Niger-Congo Abasheikh 1976
  12)
• Musa xad?a?- ile ruhu- ye.
• Musa cheat- PAST soul- his
• Musa cheated himself.

23
On motivation for diffusion

• The Arabic expansion In a similar fashion does
  the distribution of languages having
  grammaticalized a soul/life-reflexive coincide
  geographically with the Arabic expansion in
  northeastern Africa. Arabs reached the Lake Chad
  basin over a thousand years ago and from then on,
  the region between Lake Chad and the Red Sea
  became a sphere of Arabic influence.

24
Nominal sources for reflexive markers in African
languages
25
The Ethiopian area

• The use of a noun for head to express
  reflexivity is very widespread in the Ethiopian
  are, most of all for intensifiers, but also for
  reflexive markers. The roots involved are ras and
  ?rs head, which are widely distributed over the
  Ethio-Semitic area (Böhm 1984 97 Goldenberg
  1991). But diffusion also appears to also have
  affected other languages of the area. The Central
  Cushitic language Kemantney has developed its
  noun -a?wäy head into an intensive reflexive
  (Zelealem 2003 181), and much the same is
  reported for the Nilo-Saharan language Kunama,
  where the noun ana head is said to have given
  rise to a reflexive marker (Böhm 1984 97).

26
Grammaticalization areas

• By grammaticalization area we understand a
  group of geographically contiguous languages that
  have undergone the same grammaticalization
  process as a result of language contact. In order
  to identify a grammaticalization area it is
  therefore important to rule out factors other
  than language contact, such as genetic
  relationship, drift, and chance.
• Heine, Bernd and Tania Kuteva 2005. Language
  contact and grammatical change. Cambridge
  Cambridge University Press.

27

• Advantages of grammaticalization areas over
  sprachbunds
• No mismatch between isoglosses is possible
  because there is only one isogloss.
• Unlike a sprachbund, a grammaticalization area
  can be assumed to be motivated by some specific
  historical event. Thus, there likely is a
  one-to-one match between linguistic and
  extra-linguistic processes.
• Accordingly, unlike sprachbunds,
  grammaticalization areas offer a tool for
  historical reconstruction.

28
Conclusions

• What the observations made in this talk suggest
  is that the development of grammatical forms is
  not independent of the socio-cultural environment
  in which it takes place. Given any unknown
  African language one will expect that if a new
  reflexive marker evolves, most likely it will be
  based on the grammaticalization of a noun for
  body, and this is also worldwide the most
  probable option. But in addition to this
  conceptual option there is at least one more
  factor that determines the choice of reflexive
  markers, namely areal influence.

29
In concluding, mention should be made
of the following observation that surfaces from
the above discussion On the one hand, the cases
examined are all hypothesized to be due to
language contact on the other hand, they can be
described as well as involving language-internal
developments. That external and internal
linguistic changes are by no means mutually
exclusive is not really new (see Heine 2005
2006), but what exactly this means with reference
to our understanding of linguistic change and of
language structure is an issue that would seem to
require much more attention in future
research. Heine, Bernd Tania Kuteva 2005.
Language contact and grammatical change.
Cambridge Cambridge University Press. ----
2006. The changing languages of Europe. Oxford
Oxford University Press.

30
Any questions?