German determiner inflection

  Morphologically speaking, there are three major and two minor classes of determiner (used here to include the articles and numerals) in German. The first major class is both the most basic and the easiest to describe since it does not involve any inflection at all. All we need to say is that members of the class are words whose syntactic category is determiner. We can describe it thus:

Det_0:
    <> == Word
    <syn cat> == det.

An example instance of Det_0 is provided by zwei (`two'):

Zwei/det:
    <> == Det_0
    <phn syl1 onset> == t s v
    <phn syl1 peak> == a i.

Nothing has been said in either of these node definitions to give a value (or values) to <mor suffix ..> paths. Since nothing is said, all these paths default to the null sequence (the default ``zero formation'' in Zwicky's (1985) parlance). And that, in turn, means that all twenty four morphosyntactic specifications map to a bare monosyllabic root:

Zwei/det:
    <mor word nom sing masc> = t s v a i
    ..
    <mor word dat plur neut> = t s v a i.

The remaining four classes are morphologically more interesting. The first three have inflectional suffixation patterns that are wholly determined by one or other of the three strong declensions. As these declension patterns have already been defined, the definitions for these determiners are extremely straightforward:

Det_1A:
    <> == Det_0
    <mor suffix> == "Strong_1:<phn root form>".
Det_2:
    <> == Det_0
    <mor suffix> == "Strong_2:<phn root form>".

Det_3:
    <> == Det_0
    <mor suffix> == "Strong_3:<phn root form>".

The first (minor) class Det_1A displays a pattern of inflection that is identical to that shown in the strong forms of adjectives. Hammer (1977) cites sämtlicher (`each and every') as an example of this class:

Saemtlicher/det:
    <> == Det_1A
    <phn> == Disyllable
    <phn syl2 onset> == z
    <phn syl2 peak> == E
    <phn syl2 coda> == m t
    <phn syl1 onset> == l
    <phn syl1 peak> == I
    <phn syl1 coda> == x.

The forms induced for Saemtlicher/det are:

Saemtlicher/det:
    <mor word nom sing masc> = z E m t l I x @ r
    <mor word nom sing femn> = z E m t l I x @
    <mor word nom sing neut> = z E m t l I x @ s
    <mor word acc sing masc> = z E m t l I x @ n
    <mor word acc sing femn> = z E m t l I x @
    <mor word acc sing neut> = z E m t l I x @ s
    <mor word gen sing masc> = z E m t l I x @ n
    <mor word gen sing femn> = z E m t l I x @ r
    <mor word gen sing neut> = z E m t l I x @ n
    <mor word dat sing masc> = z E m t l I x @ m
    <mor word dat sing femn> = z E m t l I x @ r
    <mor word dat sing neut> = z E m t l I x @ m
    <mor word nom plur masc> = z E m t l I x @
    <mor word nom plur femn> = z E m t l I x @
    <mor word nom plur neut> = z E m t l I x @
    <mor word acc plur masc> = z E m t l I x @
    <mor word acc plur femn> = z E m t l I x @
    <mor word acc plur neut> = z E m t l I x @
    <mor word gen plur masc> = z E m t l I x @ r
    <mor word gen plur femn> = z E m t l I x @ r
    <mor word gen plur neut> = z E m t l I x @ r
    <mor word dat plur masc> = z E m t l I x @ n
    <mor word dat plur femn> = z E m t l I x @ n
    <mor word dat plur neut> = z E m t l I x @ n.

Many other members of the minor Det_1A class belong also to the Det_2 class and this leads to an alternation between -en (Det_1A) and -es (Det_2) for genitive singular masculine and neuter forms. The choice is apparently not determined by morphosyntactic features, declensional class or indeed any linguistic factors, but is, according to Hammer (1977,48), ``largely determined by ear''.

There is a further minor class of determiners, Det_1B, containing just the lexemes viel and wenig (`much/many' and `little/few') whose inflection is similar to that of the Det_1A class except that they are uninflected for all their nominative and accusative singular forms. Viel and wenig also occur as members of syntactic categories other than determiner but they do not inflect at all in these occurrences. We are only concerned with their determiner instances here. Zwicky's determiner classes as well as sharing inflection patterns also share syntactic information about the inflectional patterns of the adjectives with which they combine. We do not include this information in our fragment, restricting ourselves to inflectional information, but our classes map onto Zwicky's in the following way: Det_0 $\rightarrow$ Det_I, Det_1A $\rightarrow$Det_II, Det_2 $\rightarrow$ Det_II, Det_3 $\rightarrow$Det_III. Our Det_1B class cannot be categorised in this way since the words in this class only appear in inflected forms when no adjective is present. We can encode the Det_1B class as follows:

Det_1B:
    <> == Det_1A
    <mor suffix nom sing> == Det_0
    <mor suffix acc sing> == Det_0.

And here, by way of example, is the lexical entry needed for viel:

Viel/det:
    <> == Det_1B
    <phn syl1 onset> == f
    <phn syl1 peak> == i:
    <phn syl1 coda> == l.

This leads to the following forms:

Viel/det:
    <mor word nom sing masc> = f i: l
    <mor word nom sing femn> = f i: l
    <mor word nom sing neut> = f i: l
    <mor word acc sing masc> = f i: l
    <mor word acc sing femn> = f i: l
    <mor word acc sing neut> = f i: l
    <mor word gen sing masc> = f i: l @ n
    <mor word gen sing femn> = f i: l @ r
    <mor word gen sing neut> = f i: l @ n
    <mor word dat sing masc> = f i: l @ m
    <mor word dat sing femn> = f i: l @ r
    <mor word dat sing neut> = f i: l @ m
    <mor word nom plur masc> = f i: l @
    <mor word nom plur femn> = f i: l @
    <mor word nom plur neut> = f i: l @
    <mor word acc plur masc> = f i: l @
    <mor word acc plur femn> = f i: l @
    <mor word acc plur neut> = f i: l @
    <mor word gen plur masc> = f i: l @ r
    <mor word gen plur femn> = f i: l @ r
    <mor word gen plur neut> = f i: l @ r
    <mor word dat plur masc> = f i: l @ n
    <mor word dat plur femn> = f i: l @ n
    <mor word dat plur neut> = f i: l @ n.

Returning to the two remaining classes of determiner, Det_2 and Det_3, we provide some instances (Jeder (`each/every'); solcher (`such'); jener (`that'); dieser (`this'); welcher (`which'); ihr (`your/her'); euer (`your'); unser (`our')) with explicit definitions given only for the phonological realisation of the onset, peak and coda. The node names here are the citation forms, not the roots, which are mostly monosyllabic. Because many of these determiners share phonology with corresponding pronouns (for which, see the section on pronouns), we have mostly chosen to encapsulate the common phonology in separate nodes:

Jeder/det:                       Solcher/det:
    <> == Det_2                      <> == Det_2
    <phn syl1 onset> == j            <phn syl1 onset> == z
    <phn syl1 peak> == e:            <phn syl1 peak> == O
    <phn syl1 coda> == d.            <phn syl1 coda> == l x.

Jener:                           Jener/det:
    <> == Syllable                   <> == Det_2
    <phn syl1 onset> == j            <phn> == Jener.
    <phn syl1 peak> == e:
    <phn syl1 coda> == n.
Dieser:                          Dieser/det:
    <phn> == Syllable                <> == Det_2
    <phn syl1 onset> == d            <phn> == Dieser.
    <phn syl1 peak> == i:
    <phn syl1 coda> == z.
Welcher:                         Welcher/det:
    <> == Syllable                   <> == Det_2
    <phn syl1 onset> == v            <phn> == Welcher.
    <phn syl1 peak> == E
    <phn syl1 coda> == l x.
Ihr:                             Ihr/det:
    <> == Syllable                   <> == Det_3
    <phn syl1 peak> == i:            <phn> == Ihr.
    <phn syl1 coda> == r.
Euer:                            Euer/det:
    <> == Disyllable                 <> == Det_3
    <phn syl2 peak> == O y           <phn> == Euer.
    <phn syl1 peak> == @
    <phn syl1 coda> == r.
Unser:                           Unser/det:
    <> == Disyllable                 <> == Det_3
    <phn syl2 peak> == U             <phn> == Unser.
    <phn syl2 coda> == n z
    <phn syl1 peak> == @
    <phn syl1 coda> == r.

The determiner lexeme definitions yield paradigms such as those given in full below for dieser (an instance of Det_2) and ihr (an instance of Det_3):

Dieser/det:
    <mor word nom sing masc> = d i: z @ r
    <mor word nom sing femn> = d i: z @
    <mor word nom sing neut> = d i: z @ s
    <mor word acc sing masc> = d i: z @ n
    <mor word acc sing femn> = d i: z @
    <mor word acc sing neut> = d i: z @ s
    <mor word gen sing masc> = d i: z @ s
    <mor word gen sing femn> = d i: z @ r
    <mor word gen sing neut> = d i: z @ s
    <mor word dat sing masc> = d i: z @ m
    <mor word dat sing femn> = d i: z @ r
    <mor word dat sing neut> = d i: z @ m
    <mor word nom plur masc> = d i: z @
    <mor word nom plur femn> = d i: z @
    <mor word nom plur neut> = d i: z @
    <mor word acc plur masc> = d i: z @
    <mor word acc plur femn> = d i: z @
    <mor word acc plur neut> = d i: z @
    <mor word gen plur masc> = d i: z @ r
    <mor word gen plur femn> = d i: z @ r
    <mor word gen plur neut> = d i: z @ r
    <mor word dat plur masc> = d i: z @ n
    <mor word dat plur femn> = d i: z @ n
    <mor word dat plur neut> = d i: z @ n.

Ihr/det:
    <mor word nom sing masc> = i: r
    <mor word nom sing femn> = i: r @
    <mor word nom sing neut> = i: r
    <mor word acc sing masc> = i: r @ n
    <mor word acc sing femn> = i: r @
    <mor word acc sing neut> = i: r
    <mor word gen sing masc> = i: r @ s
    <mor word gen sing femn> = i: r @ r
    <mor word gen sing neut> = i: r @ s
    <mor word dat sing masc> = i: r @ m
    <mor word dat sing femn> = i: r @ r
    <mor word dat sing neut> = i: r @ m
    <mor word nom plur masc> = i: r @
    <mor word nom plur femn> = i: r @
    <mor word nom plur neut> = i: r @
    <mor word acc plur masc> = i: r @
    <mor word acc plur femn> = i: r @
    <mor word acc plur neut> = i: r @
    <mor word gen plur masc> = i: r @ r
    <mor word gen plur femn> = i: r @ r
    <mor word gen plur neut> = i: r @ r
    <mor word dat plur masc> = i: r @ n
    <mor word dat plur femn> = i: r @ n
    <mor word dat plur neut> = i: r @ n.

The overall structure of our theory of German determiners, as presented up to this point, can be seen below:

  Thus far, we have not said anything explicit about the indefinite article ein. Inflectionally, it is just a member of the Det_3 class and has exactly the suffixes one would expect of a member of that class. But it also belongs to a family of rhyming determiners and pronouns for which it makes sense to use the phonology node (Ein) specifically to capture the generalization that can be made about the peak and coda common to members of this family ( Dein (`your'); kein (`no'); mein (`my'); sein (`his/her/its')). The definitions for this family are shown below. The observant reader will spot that our morphology provides plural forms for ein. We make no apology for this. Although it is straightforward to augment the definitions so as to eliminate the plural forms, that would be a bad analysis. Ein is not morphologically exceptional in only having singular forms - it is entirely regular morphologically. However, it is inherently singular as a matter of syntax (and semantics) and thus has no use for its (regular) plural forms.

Ein:                             Ein/det:
    <> == Syllable                   <> == Det_3
    <phn syl1 peak> == a i           <phn> == Ein
    <phn syl1 coda> == n.            <syn number> == sing.

Dein:                            Dein/det:
    <phn> == Ein                     <> == Det_3
    <phn syl1 onset> == d.           <phn> == Dein.
Kein:                            Kein/det:
    <phn> == Ein                     <> == Det_3
    <phn syl1 onset> == k.           <phn> == Kein.
Mein:                            Mein/det:
    <phn> == Ein                     <> == Det_3
    <phn syl1 onset> == m.           <phn> == Mein.
Sein:                            Sein/det:
    <phn> == Ein                     <> == Det_3
    <phn syl1 onset> == z.           <phn> == Sein.

Ein also has an adjectival usage, but in this it is entirely unremarkable, inflecting as a regular adjective. Indeed, in this usage we also find a plural form in the perhaps idiomatic die einen ... die anderen, `some ... others'. Adding a definition of the adjectival form of ein is simple:

Ein/adj:
    <> == Adj_1
    <phn> == Ein.

There are a significant number of compound determiners in German, such as irgendwelcher (`some or other'). We construe such determiners as lexemes that, in effect, suffix the inflectable determiner that forms the second element of the compound to the phonological root of the first element of the compound. All other properties of the compound (such as syntactic category) are determined by the determiner component. We can formalize this analysis of such compounds in DATR node definitions that have the following form (this discussion is simply intended to show how compound determiners might be handled in the context of our theory of German inflection - it is not intended to constitute a general theory of compounding in German):

First_Second:
    <> == Second
    <mor word> == "First:<phn root form>" "Second:<mor word>".

In our example, the first element is irgend (which occurs in many such compounds, as well as in its own right as an adverb) and the second is welcher.

Irgend_Welcher:
    <> == Welcher/det
    <mor word> == "Irgend:<phn root form>" "Welcher/det:<mor word>".

This object will now decline in exactly the same way as any other member of the Det_2 determiner class:

Irgend_Welcher:
    <mor word nom sing masc> = i: r g @ n t v E l x @ r
    ...
    <mor word gen sing masc> = i: r g @ n t v E l x @ s
    ...
    <mor word dat plur neut> = i: r g @ n t v E l x @ n.

A rather different kind of behaviour is exhibited by the pronoun derjenige which inflects both on its first and second part. This lexeme behaves exactly as if it were the definite article (der) combined with an adjective (jenig). The latter component takes the weak endings, as one would expect on such an analysis. However, jenig does not occur as a free-standing adjective in its own right in modern German (Jenig is listed in Grimm's Deutsche Wörterbuch, Vol 4.2, 1877, and described as being derived from jener in the same way as einig is derived from ein). We therefore define jenig so as to inherit the morphological properties of a standard inflecting adjective:

Jenig:
    <> == Disyllable
    <mor> == Adj_1
    <phn syl2 onset> == j
    <phn syl2 peak> == e:
    <phn syl1 onset> == n
    <phn syl1 peak> == I
    <phn syl1 coda> == g.

We can now define derjenige as follows:

Der_Jenige:
    <> == Pronoun
    <mor word> == "Der/det:<mor word>" "Jenig:<mor word weak>".

In the absence of a syntactic treatment of determiner-adjective congruence in German, we have simply stipulated the weak inflection on Jenig here. However, the existence of such a lexeme points to interesting questions about the intralexical realization of phrasal syntax.

From the Der_Jenige node, all the relevant forms now follow

Der_Jenige:
    <mor word nom sing masc> = d e r j e: n I g @
    <mor word nom sing femn> = d i:  j e: n I g @
    <mor word nom sing neut> = d a s j e: n I g @
    <mor word acc sing masc> = d e n j e: n I g @ n
    <mor word acc sing femn> = d i:  j e: n I g @
    <mor word acc sing neut> = d a s j e: n I g @
    <mor word gen sing masc> = d E s j e: n I g @ n
    <mor word gen sing femn> = d e r j e: n I g @ n
    <mor word gen sing neut> = d E s j e: n I g @ n
    <mor word dat sing masc> = d e m j e: n I g @ n
    <mor word dat sing femn> = d e r j e: n I g @ n
    <mor word dat sing neut> = d e m j e: n I g @ n
    <mor word nom plur masc> = d i:  j e: n I g @ n
    ...
    <mor word gen plur masc> = d e r j e: n I g @ n
    ...
    <mor word dat plur masc> = d e n j e: n I g @ n
    ...