Wolfgang Schulze W.Schulze at LRZ.UNI-MUENCHEN.DE
Tue Oct 6 04:40:57 UTC 2009

Dear Enrique,
I think, Indo-European is a slightly different case. Diego talked about  
his observation that in
> Bocotá, a language of Panama, the marker -/du /has been described as a 
> marker of perfective aspect (Gunn 1975), while in Buglere, a dialect 
> of Bocotá, spoken in Southwestern Panama and Costa Rica, it 
> clearly behaves as a marker of middle voice.
Hence, we are faced with a change on the conceptual layer only, whereas 
the formal (expression) layer remains unchanged. In Indo-European, 
however, the way seems to have been the other way round: Here, one 
hypothesis is, that the formal layer showed gradual merger of two 
distinct elements (morphemes) resulting in the blending of the two 
conceptual layers. If we start from a very rudimentary paradigm (that 
neglects many specifics and refers only to the singular forms), we can 
describe the following for the Perfect/Stative:

1   *-H2e (~ *-a)
2   *-tH2e (~ *tha)
3   *-He (~ -e)

It seems reasonable to assume that these forms once represented 
Dative-like variants of the 'agentive' series

1   *-m(-i)
2   *-s(-i)
3   *-t(-i)

Hence, *-m reads 'I (the doer)', *-s (you:SG (the doer), *-t ((s)he (the 
doer)' or so, while *-H2e meant 'for me', *-tH2e 'for you:SG', and -He 
'for him/her' or so (I use 'for' as a cover expression for the Dative 
function). Now imagine we had a verb like *ghwen-m 'I hit' 
(perfective/aorist). By adding the Dative clitic (?) *-H2e, the verb 
would mean 'I hit for me' (*ghwen-m-H2e). As a result, a kind of middle 
construction emerged that included both the agent marker and the 
Dative-based subjectification marker. The two elements (historically 
clitics?) then merged to one form resulting in the corresponding blend 
of the agent/subjectification layers:

1    *-m-H2e (~ *-m-a)
2    *-s-tH2e (~ *-s-ta)
3    *-t-He (~ *-t-e)

The third person element perhaps was effected by ablaut resulting in 
*-t-o. In a final step, the subjectification/reflexive paradigm became 
harmonized based on the third person (as it is of the case with such 
paradigms), hence generalizing *-e > -*-o. The final output would have 
been the classical middle forms of the singular):

1   *-m-o (instead of *-m-a)
2   *-s-o (instead of *-s-ta)
3   *-t-o

For some more details see Wolfgang Schulze 1990. Prototypik in der 
Diachronie. Zur Frage der indogermanischen Diathesen. In: FoliaLing. 
Hist . X/1:1-19.

Best wishes,



*Prof. Dr. Wolfgang 
Schulze    *                                                               


/Primary contact: 

Institut für Allgemeine & Typologische Sprachwissenschaft     

Dept. II / F 13                  

Ludwig-Maximilians-Universität München                                 

Ludwigstraße 25                                         

Postanschrift / Postal address: Geschwister-Scholl-Platz 1

D-80539 München                                  

Tel.: 0049-(0)89-2180-2486 


Fax:  0049-(0)89-2180-16567 // 

Email: W.Schulze at 
<mailto:W.Schulze at> /// Wolfgang.Schulze at 
<mailto:Wolfgang.Schulze at>                                      




Katedra Germanistiký        

Fakulta humanitných 

Univerzita Mateja Béla / Banská 


SK-97401 Banská 


Fax: (00421)-(0)48-4465512   

Email: Schulze at 
<mailto:Schulze at>                                                                             




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <>

More information about the Lingtyp mailing list