Íslenskt mál og almenn málfræði - 01.01.1979, Page 245

JANEZ ORESNIK
On the pronunciation of modern Icelandic
rövl(a) and slafneskur
1. Kiparsky 1973 formulated the phonological universal (la) and an
alternative stronger statement (lb):1
(1) (a) Neutralization processes apply only to derived forms.
(b) Non-automatic neutralization processes apply only to derived
forms.
Since nothing in my presentation hinges on the notion NON-AUTO-
MATIC added in (lb), I shall ignore this alternative in what follows,
although it would become critically involved if further pertinent ex-
amples should be discussed. Kiparsky’s universal has been revised by
Ringen 1977, in a way that is not relevant to the subject-matter of the
present paper; this revision will therefore be ignored in what follows.
The criticism of the universal by Kenstowicz & Kisseberth 1977 is in-
conclusive, for the reasons stated in Clayton 1979.
To illustrate the universal, I shall briefly repeat one of Kiparsky’s
own examples, namely the Finnish rule of t -» s / —i. This rule is
neutralising in the following sense: it creates an output, namely s before
i, which is also present in the input to the rule, i.e. it exists at the stage
of derivation immediately preceding the application of the t-io-s rule.
(Exampie of underlying s before i: silta ‘bridge’.) The rule applies (a)
across a morpheme boundary (e.g. /halut + i/~^halusi ‘wants, impf.’),
(b) morpheme-internally, with derived i (e.g. /vete/-»/veti/-» vesi
‘water’). The rule does NOT apply (c) morpheme-internally, with
underlying i (e.g. /koti/~* koti ‘home’).
To account for this state of affairs, Kiparsky introduced the notion
DERIVED FORM or DERIVED INPUT, which he defined as follows:
(2) I will refer to an input which is created either by combining mor-
phemes through derivation or inflection, [. ..] or by applying a
phonological rule, [. . .] as a DERIVED input.
1 My thanks are due to Miss Margaret G. Davis, who has corrected my English.
L