12cs-raw [⌂ EvenYrbk]

12cs, PDF
Nominal and pronominal arguments in German within the framework of AS

Tamás Csontos

49

Nominal and pronominal arguments in German within the framework of Alignment Syntax

0 Introduction In the present paper I would like to account for the possible position(s) and order of nominal and pronominal arguments in German within the framework of Alignment Syntax. This is a continuation of work started in my previous essay (Csontos, 2011), which concentrated on double object constructions in German. Now the focus will be on nominal and pronominal subjects. In section 1 I will analyse the position of German verbs. In section 2 I am going to deal with nominal and pronominal arguments. At the end of the paper there is a list of abbreviations.

1 Verb-second and verb-last phenomena in German In Csontos (2011) I claimed that pronouns tend to follow verbs and be as close to them as possible. I postulated two constraints: pFv ‘pronouns follow verbs’ and pAv ‘pronouns are adjacent to verbs’. I dealt only with main clauses and finite main verbs. The issue is, however, far more complex. If we take subordinate clauses and non-finite main verbs into consideration, we will see that the assumption that pronouns are ordered with respect to verbs has to be reconsidered. The second problem with these constraints is that pronominal subjects do not follow verbs but precede them (see below). I will attempt to solve this problem as well.

1.1 Main clauses First, I would like to briefly discuss the so-called verb-second and verb-last phenomena. In German the finite verb takes the second position in declarative main clauses and wh-questions. The finite verb can be preceded by the subject, the topic, or the wh-element, see the italicised constituents in (1), (2) and (3), respectively.1 (1) Ich gehe jedes Wochenende ins Kino. I go every weekend to the cinema 1

There has been much debate on what topics are. Some would claim that the subject in first position is the topic. However, I will adopt Newson’s (2010) view, viz. that subjects in first position are distinct from topics. The Even Yearbook 10 (2012), Department of English Linguistics, Eötvös Loránd University, Budapest ISSN 2061–490X, http://seas3.elte.hu/delg/publications/even, © 2012, Tamás Csontos

Nominal and pronominal arguments in German within the framework of AS

50

‘I go to the cinema every weekend.’ (2) Jedes Wochenende gehe ich ins Kino. (3) Wohin fährst du nächste Woche? where travel you next week ‘Where are you travelling next week?’ According to Newson (2010), we miss a generalisation if we say that the [finite] verb has to follow the wh-element or the topic or the subject, in that order of preference, given that this order is already stated for the fronted elements in the ranking of separate constraints (i.e. the requirement that the wh-element be first outranks the requirement that the topic be first, which in turn outranks the requirement that the subject be first). In other words, we need constraints that determine that the finite verb must be in the second position irrespective of what the first element is. Secondly, as shown in (4), (5) and (6), we have to take into consideration that non-finite main verbs take the last position in main clauses (see the underlined elements), while inflected auxiliaries (e.g. wollen, haben or werden) are the second (see the italicised elements): (4)

Ihr wollt ihn diese Woche einladen. You[pl] want him this week to invite ‘you want to invite him this week.’

(5)

Wir haben ihn letzte Woche nicht gesehen. We have him last week not seen ‘We didn’t see him last week.’

(6)

Das Haus wurde im Jahre 1980 gebaut. The house was in year 1980 built ‘The house was built in 1980.’

Therefore, we have to make a distinction between finite and non-finite verbs. The former can be identified by a functional conceptual unit which must follow the verb. I will refer to this conceptual unit as I (for inflection – tense plus agreement). Non-finite verbs, on the other hand, are not followed by I. Admittedly, we are only scratching the surface of how agreement and tense are realized in German within Alignment Syntax: this will be a task for future research. For our present purposes, however, we do not have to go into more detail. I introduce two further constraints, namely IPDep and I*PDep. The The Even Yearbook 10 (2012), Department of English Linguistics, Eötvös Loránd University, Budapest ISSN 2061–490X, http://seas3.elte.hu/delg/publications/even, © 2012, Tamás Csontos

Nominal and pronominal arguments in German within the framework of AS

51

first is a gradient constraint which says that I wants to be the first element of the extended predicate domain (Dep), which is made up of all the dependents of a verb, i.e. arguments and adverbs.2 This constraint is violated by every member of the extended predicate domain which precedes I. The second constraint is an anti-alignment constraint, which says the inflection must not be the first element of this domain. This constraint is violated if the inflection precedes every member of the extended predicate domain. As we know, if there is a main verb in a clause, it must immediately precede the inflection see examples (1), (2) and (3). But if there is an auxiliary verb in the clause as well, it is the auxiliary verb that must immediately precede the inflection while the main verb must follow the extended predicate domain, see examples (4), (5) and (6). Thus, it seems there is a requirement which says the inflection must be adjacently preceded by some verbal element (either a main verb or an auxiliary): [+V]PI > [+V]AI (Newson, personal communication). [+V]AI is violated by every member of the whole domain which appears between the verbal element and the inflection. On the other hand, there is a condition on the main verb to follow the whole domain (D): vFD (as main verbs appear at the end of clauses with an auxiliary). It is violated by every member of the domain which follows the verb. Obviously, this constraint is ranked lower than [+V]PI > [+V]AI, because verbs in main clauses do not take the last position, but precede the inflection - see examples (1), (2) and (3). Note that at this point this constraint can also account for the fact that it is auxiliaries and not main verbs that are preferred by the inflection: violation of the verbal constraint can be avoided. (However, we will see in section 1.2 that this issue is a bit more complicated.) Consider (7) and (8): (7) → Ich gehe jedes Wochenende ins Kino. Ich -e jedes Wochenende ins Kino gehen. Gehen ich -e jedes Wochenende ins Kino.

I*PDep IPDep [+V]PI [+V]AI vFD

(!)

(!)

2

P is interpreted both as “be before something” and ”be the first element of something”. In a similar vein, F is interpreted both as ”be after something” and ”be the last element of something” (see below). Why these interpretations are not incompatible, see Newson and Szécsényi’s paper (2012) in the present volume.

Nominal and pronominal arguments in German within the framework of AS

52

(8) → Ihr wollt ihn diese Woche einladen. Ihr einladet ihn diese Woche wollen. Ihr wollt einladen ihn diese Woche. Wollen ihr -t ihn diese Woche einladen. Wollen ihr einladet ihn diese Woche.

I*PDep IPDep [+V]PI [+V]AI vFD

(!)

(!)

(!)
- (!)

1.2 Subordinate clauses If we observe examples (9) and (10), we can see that the finite verb takes the last position in subordinate clauses when there is a complementizer: (9) ….dass ich jedes Wochenende ins Kino gehe. that I every weekend to the cinema go ‘that I go to the cinema every weekend’ (10) … ob ich jedes Wochenende ins Kino gehe. if I every week to the cinema go ‘if I go to the cinema every weekend’ According to Newson (personal communication), a possible answer to this phenomenon is proposing a subsequence constraint on the inflection: IFDx, where Dx is dependent on the presence of the complementizer. That is, when there is no complementizer, there is no domain and hence the constraint is vacuous. However, when there is a complementizer, the domain is defined and the inflection (with the verb) must be at the end of it. This constraint is ranked higher than the anti-alignment (and the IPD) constraint, because if the rank order were reversed (i.e. I*PD > IPD > IFDx), the verb would always take the second position irrespective of whether a complementizer is present or not. Therefore, the order of these constraints must be IFDx > I*PD > IPD. Thus, the inflection will be the last element in the domain in the presence of a complementizer.3 3

It goes without saying that the complementizer must be the first element of the whole domain; this constraint is ranked higher than IFDx. For the sake of convenience, however, I am not going to include it in the following tables. The Even Yearbook 10 (2012), Department of English Linguistics, Eötvös Loránd University, Budapest ISSN 2061–490X, http://seas3.elte.hu/delg/publications/even, © 2012, Tamás Csontos

Nominal and pronominal arguments in German within the framework of AS

53

Finally, let us look at subordinate clauses with an auxiliary. We can observe that the inflected auxiliary takes the last position in the clause and it is preceded by the main verb: (11) …dass du jetzt trinken willst. that you now drink want-I(nfl) ‘that you want to drink now.’ The constraints we have introduced so far and their ranking however do not yield the desired results, see (12): (12) Preliminary → a) dass du jetzt wollen trinkst . b) dass du jetzt trinken willst.

IFDx I*PDep IPDep [+V]PI [+V]AI vFD

(!)

The winning candidate should be (12b) and not (12a). Consequently, we have to assume that there is another factor that has a visible effect here, while its effect was invisible above. Therefore, I postulate a constraint which says that auxiliaries must be adjacent to inflections: auxAI, see (13). It must outrank the verbal constraint, vFD. (13) Revision of (12) a) dass du jetzt wollen trinkst. → b) dass du jetzt trinken willst.

IFDx I*PDep IPDep [+V]PI [+V]AI auxAI vFD

(!) *

To sum up, we have seen how specific constraints and their ranking can determine the possible positions and the order of main verbs and auxiliary verbs both in main and embedded clauses. In the next section I would like to focus on nominal and pronominal arguments.

Nominal and pronominal arguments in German within the framework of AS

54

2 Nominal and pronominal arguments 2.1 Pronominal arguments The order of pronominal arguments seems to be strict. In order to observe how these arguments behave, we have to analyse the following examples, which are borrowed from Lenerz (1993): (14a) weil er es ihm ja wahrscheinlich gestern gab because he it him [particle] probably yesterday gave ’because he probably gave it to him yesterday’ (14b) * other word orders It can be seen that basically there is one possible order: the agent (er) precedes the theme (es) and the theme precedes the goal (ihm). It can also be observed that the pronouns appear towards the left edge of the domain; no element can precede them (except for the complementizer). We can account for this phenomenon by introducing a ranking of precedence constraints: agentpPDep: violated by every member of the extended predicate domain which precedes the pronominal agent themepPDep: violated by every member of the extended predicate domain which precedes the pronominal theme goalpPDep: violated by every member of the extended predicate domain which precedes the pronominal goal

Nominal and pronominal arguments in German within the framework of AS

55

(15) 4 IFDx I*PDep IPDep vFD agentpPDep themepPDep goalpPDep weil ihm er gestern es gab weil ihm es er gestern gab weil es ihm gestern er gab weil es er ihm gestern gab weil er ihm es gestern gab weil er es gestern ihm gab →weil er es ihm gestern gab

(!)

*

(!)

* (!)

(!)

(!)

* (!)

Our analysis however needs to be extended to nominal arguments as well. This will be done in the next section. 2.2 Nominal vs. pronominal arguments In this section I would like to take a look at nominal arguments. As I have already mentioned above, Newson (2010) says that in main clauses, where the verb takes the second position, the subject, the topic or the wh-element can precede the verb.5 Subjects are usually agents or themes, see the italicised constituents in (16a) and (16b): (16a) Mein Vater baute dieses Haus. my father built this house ’My father built this house.’

4

For the sake of convenience I do not include the constraints [+V]PI and [+V]AI as the inflection is adjacently preceded by a verbal element in all of my examples. 5

I will not concentrate on wh-elements, only on topics and subjects.

Nominal and pronominal arguments in German within the framework of AS

56

(16b) Der Ball ist die Treppe heruntergerollt. The ball is the stairs down-rolled ‘The ball rolled down the stairs.’ If they are agents (as in 16a), the explanation is straightforward: nominal agents must precede the extended predicate domain. But if both nominal and pronominal agents precede the extended predicate domain, then it is unnecessary to have two separate constraints for nominal and pronominal agents. Consequently, it is not agentpPDep that is ranked higher than themepPDep, but agentPDep. On the other hand, in case of nominal themes (as in 16b) and nominal goals, the issue is different, as they do not behave the same way as their pronominal counterparts.6 Nominal themes do not necessarily have to precede nominal goals, although, if both of them are animate or inanimate, the theme does precede the goal, see example (17). (17) Er hat [das Baby]theme [seiner Frau]goal gegeben. he has the baby his wife given ‘He gave the baby to his wife.’ However, if the arguments differ in terms of animacy, the animate arguments will precede the inanimate arguments regardless of their thematic roles. In Csontos (2011) I argued that in this case the animacy constraint (aPi) is responsible for the order of (nominal) objects. In example (18) the animate object dem Mann must precede the inanimate object die Vase, even though dem Mann is the goal and die Vase is the theme: (18a) Er hat [dem Mann]goal [die Vase]theme gegeben. he has the man the vase given ‘He gave the vase to the man.’ (18b) * Er hat [die Vase]theme [dem Mann]goal gegeben. Therefore, themePDep and goalPDep must be ranked lower than the animacy constraint, as demonstrated in (19):

6

For further details, see Csontos (2011).

Nominal and pronominal arguments in German within the framework of AS

57

(19) a) Die Frau gab die Vase dem Mann. → b) Die Frau gab dem Mann die Vase.

agentPDep aPi themePDep goalPDep

(!)

aPi on the other hand does not have any effect on pronominal themes and goals. Therefore, the constraints which are responsible for their order must be ranked higher than aPi. This also means that we need separate constraints for nominal and pronominal themes and goals: (20) agentPDep > themepPDep > goalpPDep > aPi > themePDep > goalPDep The next question to consider is what happens if there is a topic in the clause. As (21) reveals, the subject follows and does not precede the verb: (21) Am Wochenende[topic] ging die Nachbarin in die Disko. at weekend went the neighbour to the disco ‘The neighbour went to the disco at the weekend.’ This means that a topicPDep has to be recognised, which is ranked higher than agentPDep: (22) →a) Am Wochenende[topic] ging die Nachbarin in die Disko. b) Am Wochenende[topic] die Nachbarin ging in die Disko. c) Die Nachbarin ging am Wochenende[topic] in die Disko.

I*PDep IPDep vFD topicPDep agentPDep

- (!)

*

(!)

Next I will account for a phenomenon that is illustrated by example (23) (Lenerz, 1993): (23) weil es ihm ja wahrscheinlich gestern ein Mann gab because it him [particle] probably yesterday a man gave ’because probably a man gave it to him yesterday’ A nominal subject ein Mann is the last element in the predicate domain – following even the adverb – apart from the verb. In Csontos (2011) I observed a similar phenomenon with nominal objects:

Nominal and pronominal arguments in German within the framework of AS

58

(24a) Sie brachte die Briefe am Mittwoch einem Sekretär[foc] she brought the letters on Wednesday a secretary ‘She brought the letters to a secretary on Wednesday.’ (24b) Sie brachte am Mittwoch die Briefe einem Sekretär[foc] (24c)* Sie brachte die Briefe einem Sekretär[foc] am Mittwoch. I claimed that the focusFDep constraint, which says that a focused element must be the rightmost element in the extended predicate domain, was responsible for this order. Recall that Jacobs (1988) says there is a focus– background structure in German. Sentences can be divided into a focused part and a background part. The focus is something new in relation to the background, something that the hearer or listener is not familiar with. He adds that the focused part follows the backgrounded part. This is in complete harmony with what we can see in (23). The agent ein Mann is a new piece of information, introduced by the indefinite article. Pronouns belong to the backgrounded part as by their nature they are never new. Therefore, the conclusion is that ein Mann[foc] is a focused element, which follows the backgrounded part.7 The question arises: how can the ranking of the focus constraint be determined? As we will see we cannot give a definite answer. It must be ranked higher than agentPDep, otherwise the agent would occupy the first position in the extended predicate domain even if it is a focus. However, it is impossible to determine which comes first: the focus or the topic constraint, because they can never conflict.

7

There is a distinction between focus and contrast. Contrastive pronouns exist in German; otherwise it would be impossible to translate sentences like It was me (and not you) that Peter invited into German. However, I am not going to deal with contrast in this paper. The Even Yearbook 10 (2012), Department of English Linguistics, Eötvös Loránd University, Budapest ISSN 2061–490X, http://seas3.elte.hu/delg/publications/even, © 2012, Tamás Csontos

Nominal and pronominal arguments in German within the framework of AS

59

(25) IFDx I*PDe IPDep vFD focusFDe agentPDe themepPDe goalpPDe p

p

→a) weil

p

*

(!)

* (!)

es ihm gestern ein Mann[foc ] gab

b)weil es ihm gestern gab ein Mann[foc

(!)

]

c)weil es ihm ein Mann[foc ] gestern gab d)weil ein Mann[foc ] es ihm gestern gab

Note that in main clauses, if there is a focused agent but no topic and whelement, the agent will not follow the verb in spite of the high-ranked focus constraint: (26) * gab es ihm ja wahrscheinlich gestern ein Mann[foc]. This would violate the higher ranked I*PDep constraint. Therefore, the agent must take the first position in order to satisfy I*PDep in spite of violating the focus constraint. The question may arise: why is it the agent that “moves” there. The answer is simple: because the agent constraint is the highest ranked constraint that can satisfy I*PDep, see (27):

Nominal and pronominal arguments in German within the framework of AS

60

(27) I*PDep IPDep vFD focusFDe agentPDe themepPDe goalpPDe p a) gab es ihm ja wahrscheinlic h gestern ein Mann[foc] → b) ein Mann[foc] gab es ihm ja wahrscheinlic h gestern

(!)

p *

* * *

*

p

As Csontos (2011) claimed the focus constraint is ranked between themepPDep/goalpPDep and the animacy constraint, which, roughly speaking, says that animate objects precede inanimate objects. (28) pFv > pAv > themepPDep > goalpPDep > focusFDep > aPi > themePDep > goalPDep However, this is not problematic for us at all. Theoretically, we can rank the focus constraint higher, because it will never conflict with constraints that involve pronouns, because pronouns cannot be focused.

3 Conclusion In this paper certain aspects of German syntax have been discussed and some important problems have been accounted for within the framework of Alignment Syntax. I think the results that we have achieved are promising but incomplete. The German verbal and inflectional system could be analysed in more detail, with special emphasis on clauses with more auxiliary verbs and on the temporal domain. In addition, phenomena with nominal agents following pronominal arguments in embedded clauses need some discussion. These issues are interesting and definitely worth investigating.

Abbreviations: a: animate A: is adjacent to aux: auxiliary D: whole domain (which does not contain the complementizer) Dx: the whole domain plus the complementizer Dep: extended predicate domain The Even Yearbook 10 (2012), Department of English Linguistics, Eötvös Loránd University, Budapest ISSN 2061–490X, http://seas3.elte.hu/delg/publications/even, © 2012, Tamás Csontos

Nominal and pronominal arguments in German within the framework of AS

61

F: follows i: inanimate I: inflection p: pronoun/pronominal P: precedes

P: anti-precedence constraint

v: main verb [+V]: main verb or auxiliary verb

References Csontos, Tamás. 2011. An analysis of the German Double Object Construction within the framework of Alignment Syntax. Budapest: ELTE PhD Research Seminar Paper (unpublished). Jacobs, Joachim. 1988. Fokus‐Hintergrund‐Gliederung und Grammatik. In: Hans Altmann (ed.). Intonationsforschungen [Linguistische Arbeiten 200]. Tübingen: Niemeyer. 89‐134. Lenerz, Jürgen. 1993. Zu Syntax und Semantik deutscher Personalpronomina. In: Marga Reis (ed.). Wortstellung und Informationsstruktur. Tübingen: Niemeyer. 117-152. Newson, Mark. 2010. Syntax First, Words After. In: László Varga (ed.). The Even Yearbook 9. Eötvös Loránd University, Budapest. (Available at: http://seas3.elte.hu/delg/publications/even/2010.html#nm.)

Tamás Csontos Department of English Linguistics, Eötvös Loránd University, Budapest

csontostamas82@freemail.hu