The German definite article

  We have now dealt, effectively, with all the German determiners bar one. Der is both the most frequently occurring determiner in German and the hardest to describe. The complexity resides in the vowel. If we were to ignore the vowel, we could analyse der as being a member of the Det_2 class with d as its root. Such an analysis would get the onset and the codas exactly right for all the inflected forms. Where it goes wrong is in its prediction that every such form will have schwa as its peak. In reality, none of the inflected forms of der has schwa as its peak. The traditional response to this observation is to deny our assumption that the peaks of the relevant inflectional suffixes are best given as schwa. Instead, it is claimed, they should be given as various full vowels that are then almost always reduced to schwa by a phonological process. They are not reduced to schwa in der, according to such an account, because the forms of der are all syllables that bear stress. We appreciate the diachronic motivation for this traditional response but we do not find it plausible as a synchronic story. There is simply no need for such abstractness in a theory like ours since the vowels that occupy the peak slot in the der paradigm can be simply and correctly captured by a rule that specifies the default peak (e) together with three rules that override that default.

Der:
    <> == Strong_2
    <phn syl1 onset> == d
    <phn syl1 peak> == e
    <phn syl1 peak nom sing femn> == i:
    <phn syl1 peak nom sing neut> == a
    <phn syl1 peak gen sing masc> == E.

Der/det:
    <> == Det_0
    <phn> == Der.

The definition of the Der node (repeated from the section on syllable structure) says that Der is an instance of the Strong_2 declension class whose onset is d and whose (default) peak is e. In addition, it stipulates that the peaks for the nominative singular feminine and neuter forms are i: and a, respectively and that of the genitive singular masculine is E. The definition of the Der/det node just says that Der/det is a determiner whose phonology is to be taken from the Der node. From these node definitions, all the various forms of determiner der now follow:

Der/det:
    <mor word nom sing masc> = d e r
    <mor word nom sing femn> = d i:
    <mor word nom sing neut> = d a s
    <mor word acc sing masc> = d e n
    <mor word acc sing femn> = d i:
    <mor word acc sing neut> = d a s
    <mor word gen sing masc> = d E s
    <mor word gen sing femn> = d e r
    <mor word gen sing neut> = d E s
    <mor word dat sing masc> = d e m
    <mor word dat sing femn> = d e r
    <mor word dat sing neut> = d e m
    <mor word nom plur masc> = d i:
    <mor word nom plur femn> = d i:
    <mor word nom plur neut> = d i:
    <mor word acc plur masc> = d i:
    <mor word acc plur femn> = d i:
    <mor word acc plur neut> = d i:
    <mor word gen plur masc> = d e r
    <mor word gen plur femn> = d e r
    <mor word gen plur neut> = d e r
    <mor word dat plur masc> = d e n
    <mor word dat plur femn> = d e n
    <mor word dat plur neut> = d e n.

Note the way in which the link between nonmasculine accusative and nominative forms is still being captured by the abstract rule of referral that was stated quite generally at the Weak node at the very top of the declensional hierarchy. Notice also the way that the plural nominative and accusative forms are linked to the nominative singular feminine (i) by the rule of referral at the Strong_1 node relating the nominative plural to the nominative singular feminine, and (ii) by the abstract rule of referral relating accusative to nominative at the Weak node.

The definite article sometimes appears fused to certain prepositions. These Verschmelzungsformen (`merged forms', e.g. im = in dem, ins = in das) are described in phonological terms by Schaub (1979); in morphological terms by Hinrichs (1986); and in terms of syntactic movement and incorporation by Beermann (1990). We remain agnostic on the matter but note that Hinrich's inflectional account could be most readily implemented in the present framework by concatenation of the preposition and the coda of the relevant form of the definite article - by a kind of compounding, in effect. However, since the facts are subtle and complex, and the literature on the topic rather extensive, we do not pursue the issue further here.