Case in German Nouns

Question

German has an interesting situation in its noun phrases - articles and adjectives reflect case, but the noun itself does not.

Der große Mann sieht das Haus. ("The big man sees the house," with "the big man" in the nominative).

Ich sehe den großen Mann. ("I see the big man," with "the big man" in the accusative).

Ich gebe dem großen Mann das Haus. ("I give the big man the house," with "the big man" in the dative).

Is it easier to do an analysis that claims that the nouns have case but do not show it, or to assume that case is a property that the other words in the noun phrase have, but the noun does not have at all?

Answer 1

Well, nouns in German do show case. In particular Mann shows case and declines as such:

     sing     plur
nom  Mann     Männer
acc  Mann     Männer
dat  Mann(e)  Männern
gen  Mann(e)s Männer

The dative is only optionally distinct from the nominative in singular, but mandatorily in the plural, the genitive singular is always distinct. It is true that for this specific noun the nominative and accusative are the same, but not for instance for Mensch:

     sing     plur
nom  Mensch   Menschen
acc  Menschen Menschen
dat  Menschen Menschen
gen  Menschen Menschen

The major problem is that different declension classes show different rules of syncreticism: for instance: all feminine and neuter nouns are always identical in nominative and accusative, in fact all feminine nouns are identical in all singular forms, but these rules differ for declension classes, which makes any analysis that tries to collapse two cases into one difficult; as can be seen from Mensch, some nouns are extremely syncretic and differ only in the nominative singular from all other forms, there are also some nouns that only differ in the genitive singular from all other forms such as those produced by the diminutuve suffix -chen.

There is however one hard rule about syncreticism in German cases, namely that the nominative, accusative, and genitive plural are always identical for every single noun, and that's pretty much it, there are no other universal rule — so it would be possible to create an analysis that simply calls this "the direct plural form" and models German nominal cases on one axis as having: nominative, genitive, dative, accusative, direct-plural, and dative-plural forms. I however don't think this is necessarily simpler than simply using a two-axis model of two numbers and four cases, in particular because this model must also be used for pronouns, adjectives, and determiners, so a different model for nouns would overcomplicate.

Answer 2

Well, I would say that, syntactically, the word "Mann" in your second example, for instance, has an accusative case attributed to it, but it isn't morphologically visible anymore as it did in earlier stages of the language. Modern High German lost many of its noun inflections, but many still exist, and, as Ljl pointed out in the comment section, some forms are still used to express dative and, specially, genitive. So, in a nutshell: the nouns do have cases, but they don't show them all the time in a morphological (and phonological) level.

I hope it helped to solve your question :)

Answer 3

I've tried finding an answer to your question, but ultimately it presents a false dychotomy.

I see absolutely no problem in saying that Mann is inflected for the nominative, and that dative and accusative are inflected identically. In a synchronic view the eroded inflection should be as insignificant for the analysis as it is for general English. I for one really don't know whether Tür-e goes in all cases as a remnant of the dual, or sing. Hund-e only in dative, or in how many ways Mann might be inflected in various dialects. I simply don't care because I chiefly elide those. I do know Männas as regional plural vocative.

While Mann and all other nouns except names can appear with different pronouns, they can never appear alone, except in telegram or headline style, e.g. Mann beißt Hund. This I think is a strong indicator that there's not much sense in making a distinction such as "the other words in the noun phrase have, but the noun does not [have inflection]". Because it's not unusual to describe Mann, der as the headlineform in dictionaries and similar lists.

For a silly argument one could say Mann has inflection -n in contrast to the indefinite third person pronoun man (like one cannot say that with a straight face), while the etymolog of the geminate -nn- is not well understood, but it's a purely orthographic distinction. Nevertheless, consonant stems on -n are quite usual, so the euphony is maintained.

In contrast to the general case in English, the determiners can stand alone as pronouns, like that. The only case in which the inflection is then markedly different is nom ein Mann vs pron ein-er, which depending on context can mean einer [der Männer] (genitivus ...) or simply quantifying einer, zwei (occasionally zwei-e when standing alone), but it's effectively ambiguous. That shouldn't detract from einer being a pronoun. It is yet in agreement with the usual agent suffix ein Schust-er, which goes to show that the nominative is in fact inflected, but strongly depending on the semantic category. Mann, for sake of the argument, is in it's own prototypical category.

There is no general rule for all cases, so one could say that the inflectional paradigm is irregular and strong. It has been simplified where case information is nevertheless present in the clitics. And one might argue that clitics are lexicalized to a degree whenever clues are missing (das / der Notebook / Laptop), or synthetic when clues like -er are present (der Computer) as far as gender is concerned. The case inflection follows regularly.

It is notable that articles developed from determiners. So it might be reasonable to analyses nouns as noun-adjuncts, complements, attributive nouns, whathaveyou to the head of the Phrase. And adverbs, if that's what that basically is, just don't inflect for case. Adjectives do. Whether that's "easier" as you put it, I don't think so, because the noun still commands the gender of the head, unlike adverbs. But there can be situations where half way into a noun phrase one suddenly has to select one of various suitable nouns (not different from any other word finding problems) so that the determiner suddenly rules the noun--more likely when the underspecified das was used.

Last but not least, compound nouns need to be considered. The fact that the adjective in groß- Mann is inflected makes it really easy, if not preferable to employ nominalization. Duden dedicates a few pages worth of writing to rules around capitalization of these constructs. I am not sure how indicative or rather arbitrary those rules are. I mean, German really get's a bad rap for compounding, if we could as well write der Große-Mann. Compare e.g. Großhändler (bulk-sales?), Klein Lisa and similar, which would impliy that in fact the bare stem is more likely to confer noun-hood, if that weren't purely conditioned by phonetics. The recommendations are summarized there (cf D 73, D 88 - 89).

tl;dr: instead of saying that the noun is not inflected, or hiding it's inflection, rather don't mention it at all. If you do mention it, it appears as a notable exception, which can be understood as irregular inflection. So the argument is self defeating. This, which might count as hiding or zero-morph'ing, is not needed because noun phrases never appear without inflection. Ultimately, it seems that stems might well count as prototypical nouns, at which point the notion of inflection becomes meaningless. Saying there was no inflection may be effectively equivalent, but not strictly the same. It's certainly still easier not to mention it.