Menu
Home
Log in / Register
 
Home arrow Language & Literature arrow Word and paradigm morphology
Source

Paradigm uniformity and cohesion

In the Russian noun paradigms in Table 4.1, the class of a noun and its full 12-cell paradigm can be deduced from a small number of forms, and often from the nominative singular alone. The economy of this declensional system rests in large part on its uniformity. One can, for example, speak of ‘the paradigm’ of a regular Russian noun, rather than of ‘the singular (sub)paradigm’, ‘the plural (sub)paradigm’, etc. Uniformity in this sense is independent of complexity, as one can see by contrasting Russian and German declensions. Noun paradigms

Table 4.9 Singular declensional patterns in German (cf. Duden 2005:197)

SI

Masc

S2

Neut

SI

Fem

S3

Nom

Pegel

Prinz

Segel

Regel

Acc

Pegel

Prinzen

Segel

Regel

Dat

Pegel

Prinzen

Segel

Regel

Gen

Pegels

Prinzen

Segels

Regel

‘level’

‘prince’

‘sail’

‘rule’

in German are inflectionally simpler than in Russian, with fewer cases and less form variation in general. However, they are also less uniform, due to a partial dissociation between singular and plural patterns. This dissociation leads to a comparatively large number of ‘word types’ which exhibit very weak implicational relations between singular and plural forms.

The three regular singular patterns are illustrated in Table 4.9. The diagnostic genitive forms are set in bold.[1] Pattern S3 characterizes feminine nouns, which are invariant in the singular. Neuters likewise followjust the ‘strong’ pattern Si, in which the genitive singular is marked by -(e)s. Masculine nouns exhibit the only variation, as they may follow either the strong pattern Si, marked by -(e)s, or the ‘weak’ pattern S2, marked by-(e)n.

The corresponding plural formations are illustrated in Table 4.10.[2] Each of the patterns P1-P5 is classified by the shape of its ending and whether it is, or can be, based on an umlauted variant of the singular stem. A comparison of the forms in the first two blocks of rows shows that masculine and neuter nouns both follow all five patterns, although comparatively fewer neuter nouns follow the weak pattern P2 and only a very few exhibit umlaut in P3. The bottom block of rows shows that feminine nouns do not form plurals in -er and that they tend to require umlauted stems in patterns P3 and P5.

There are various correlations between the gender and stem form of a noun and the plural pattern that it follows. These subgeneralizations are discussed in detail in reference and pedagogical grammars such as Duden (2005), as well as in theoretical studies such as Kopke (1988) and Neef (1998). Of the patterns in Table 4.10, only plurals in -s are productive across all genders.[3] Unlike English, in which the default

Table 4.10 Plural patterns in German (cf. Duden 2005:226)

Ending

Stem

P1

  • -s
  • -uml

P2

  • -(e)n
  • -uml
  • -e
  • -uml

P3

-e

+uml

P4

-er

+uml

  • 0
  • -uml

P5

0

+uml

N/A/G

Uhus

Prinzen

Hunde

Bunde

Munder

Balken

Garten

Dat

Uhus

Prinzen

Hunden

Bunden

Mundern

Balken

Garten

(Masc)

‘owls’

‘princes’

‘dogs’

‘waistbands’

‘mouths’

‘beams’

‘gardens’

N/A/G

Autos

Ohren

Jahre

Flofie

Lander

Muster

Kloster

Dat

Autos

Ohren

Jahren

Flofien

Landern

Mustern

Klostern

(Neut)

‘cars’

‘ears’

‘years’

‘rafts’

‘countries’

‘patterns’

‘cloisters’

N/A/G

Bars

Regeln

Hande

Tochter

Dat

Bars

Regeln

Handen

Tochtern

(Fem)

‘bars’

‘rules’

‘hands’

‘daughters’

the number of applicable plural strategies. If only one strategy can apply, then there is no uncertainty and knowing the gender/stem combination identifies the basic plural form. If two equiprobable strategies are applicable, the gender/stem combination reduces the uncertainty to two alternatives; if three strategies are applicable, the uncertainty is reduced to three alternatives. However, the number of alternative strategies only provides a meaningful measure of relative uncertainty or productivity if each of the alternatives is equally likely. A cell with three alternative realizations but a strong bias in favour of one has considerably less uncertainty than a cell with two equiprobable realizations. The same concerns arise for any productivity scale that correlates the ‘strength’ or ‘degree’ of productivity with the number of alternatives without considering their distributional properties.

To extend the classification proposed in Laaha et al. (2006) to a probabilistic model of correlations between gender/stem patterns and plural strategies, it would be useful to adopt an information-theoretic perspective that provides a frequency- weighted measure of uncertainty.17 Yet what would still be absent from this picture would be a correlation between the factors that condition inflectional patterns in the singular and plural. The declensional system contains exactly one case of this kind, which is cited by Laaha et al. (2006):

In reality, the choice of plural formation depends largely on gender and/or inflection class as manifested also in the expression of the four German cases in the singular. Thus if a masculine has the suffix -en in the Gen.Sg., it must also have it for the plural, e.g. der Furst ‘prince, sovereign’, Gen.Sg. des Furst-en implies the plural Furst-en. (Laaha et al. 2006:279)

The broader claim that “the choice of plural formation depends largely on gender and/or inflection class as manifested ...in the singular” is unfounded, as Table 4.11 shows. Matching the singular patterns in Table 4.9 with the plural patterns in Table 4.10 defines 19 combinations, which highlight the almost complete dissociation of singular and plural sub-paradigms in German. A particularly clear

Table 4.11 Combinations of singular and plural patterns in German

uml

Masc

SI

Neut

S2

Masc

S3

Fem

PI

-

UHU

AUTO

kamera

P2

-

STAAT

oHR

PRINZ

regel

P3

-

HUND

JAHR

P3

+

BUND

(floss)

HAND

P4

+

MUND

land

P5

-

balken

MUSTER

P5

+

GARTEN

(kloster)

TOCHTER

17 Though these statistical generalizations do not not always suffice to predict the plural reliably from gender/stem patterns, as illustrated by the contrast between the masculine pairs Hund^Hunde, Bund^BUnde and Mund^MUnder in Table 4.10.

reflex of this dissociation is the existence of nouns that are traditionally described as belonging to ‘mixed declensions’. For example, the masculine noun staat ‘state’ follows the ‘strong’ pattern Si in the singular and the ‘weak’ pattern P2 in the plural. As Laaha et al. (2006) note, the converse mixture does not occur, as masculine nouns that follow the weak pattern S1 in the singular always follow the corresponding weak pattern P2 in the plural. However, this is the only inflectional restriction on masculine or neuter nouns, since a strong masculine or neuter can in principle follow any plural pattern. Feminine nouns are all invariant in the singular and thus do not exhibit any inflectional variation that can correlate with the choice of plural patterns.

In sum, there is a weaker integration of singulars and plurals into cohesive form sets in German than in Russian. The genitive singular of a German noun identifies the remaining singular forms. The nominative, accusative and genitive forms are all equally diagnostic within the plural sub-paradigm.[4] Except in the case of the weak masculine pattern S2, singulars are not reliable predictors of plural patterns. In contrast, the singular class is predictable from the plural and stem form of a noun, apart from mixed declensions that pattern like staat. The full range of patterns is predictable from a plural and genitive singular form. Hence, the best account of this type of fractured system is one that recognizes separate singular and plural principal parts for each noun.

  • [1] This table omits the proper name class listed in Duden (2005: 297), which patterns with Sr nouns.It also excludes the pattern exhibited by name ‘name’ and a handful of other nouns, which haveaccusatives and datives in -en and genitives in -ens.
  • [2] The nominative, accusative and genitive forms are always identical, so these forms are listed justonce, in the rows labelled ‘N/A/G’.
  • [3] The ending -s is often regarded as the default plural marker in German because it is associatedwith a large and diverse class of elements (Clahsen et al. 2992; Marcus et al. 2995). This class includesloans, such as Btiros ‘offices’ or Tabus ‘taboos’, abbreviations, such as LKWs (Lastkraftwagens) ‘trucks’,proper names such as die Mtillers and hypochoristics such as Omas ‘grandmothers. An alternative isthat -s marks non-native nouns that do not conform to usual nominal phonotactics or are otherwiseanomalous in the context of the German noun system (Rettig 2972; Wunderlich 2999).
  • [4] Even the dative plural is highly informative, since it unambiguously identifies the basic plural formexcept when it ends in an -n following a schwa or liquid.
 
Source
Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
 
Subjects
Accounting
Business & Finance
Communication
Computer Science
Economics
Education
Engineering
Environment
Geography
Health
History
Language & Literature
Law
Management
Marketing
Mathematics
Political science
Philosophy
Psychology
Religion
Sociology
Travel