Íslenskt mál og almenn málfræði - 01.01.1986, Side 120

118 Kjartan Ottósson
Kennaraháskóla íslands og Iðunnar IV, bls. 115-152. Reykjavík. [Þýðing Jóns
Gunnarssonar á „Cognitive Prerequisites for the Development of Grammar“ í
C. A. Ferguson & Slobin (ritstj.): Studies of Child Language Development,
New York 1973].
Stefán Einarsson. 1949. Icelandic. Grammar, Texts, Glossary. 2. útg. Baltimore.
Torp, Alf. 1974/1909. Gamalnorsk ordavleiding. Nyutgáva . . . av Gösta Holnr.
Scripta minora regiae societatis humaniorum litterarum Lundensis 1973-
1974:2. Lund. [Frumútgáfa í Gamalnorsk ordbok eftir Marius Hægstad og Alf
Torp 1909].
Vendler, Zeno. 1967. Linguistics in Philosophy. Ithaca.
Wurzel, Wolfgang U. 1984. Flexionsmorphologie und Naturlichkeit. Studia gram-
matica 21. Berlin.
Zaenen, Annie & Joan Maling. 1984. Unaccusatives, Passive and Quirky Case.
Westcoat o. fl. (ritstj.): Proceedings ofthe 3rd West Coast Conference on Formal
Linguistics, Stanford University, bls. 317-329.
SUMMARY
This paper deals with the so-called middle voice in Modern Icelandic and its place
in the inflection vs. word formation (i. e. derivation) dichotomy. It is pointed out at
the outset that we mustn’t a priori equate formal nriddle voice or jí-form with what
might be called semantic middle voice, i. e. we mustn’t presuppose that all j/-forms
have a common semantic denominator.
Chapter 1 starts with a short general methodological discussion (1.1), emphasizing
the need for attention to psychological reality and explaining prototypical catego-
ries. It goes on to discuss the psychological basis of the inflection — word formation
dichotomy (1.2). This dichotomy is then put into linguistic context within a Word
and Paradigm-approach by defining inflection as paradigmatic (essentially „auto-
matic“) and word formation as lexical (i. e. involving storage as lexemes). The con-
sequences of this in terms of defining properties of paradigms are elaborated and the
heterogeneous content of inflectional categories is discussed shortly.
Chapter 2 contains an overview over the discussion of the meaning of Icelandic
middle voice, taken as í/-forms. It is pointed out, that through the mediation of Jak-
ob Jóh. Smári (1920), Nygaard’s (1906) historically oriented treatment of the mean-
ing of middle voice in Old Icelandic has been the most influential factor. Other more
synchronically oriented treatments are notably that of Sigríður Valfells (1970) based
on Fillmore’s Case Grammar, and those of Kress, esp. Kress (1975), who contrasts
the „pole reduction" of passive with the „pole elimination” (i. e. elimination of the
agent argument) in the middle voice. Reflexive and reciprocal meaning of formal
middle voice, the classes postulated by all earlier treatments, are shown to be lexical.
The same goes for other active j/-forms, i. a. „deponents”, such as „characterising
verbs“. „Middle voice proper", i.e. Kress’ pole eliminating class, and j/-forms in
passive meaning, are set aside for discussion in ch. 3 and 4 as candidates for inflec-
tion.