Do es German Satisfy the Sub ject Condition

Judith Berman

University of Stuttgart

Pro ceedings of the LFG Conference

The University of Manchester

Miriam Butt and Tracy Holloway King Editors

CSLI Publications

httpwwwcslistanfordedupublications

LFG J Berman Do es German Satisfy the Sub ject Condition

Abstract

In this pap er I argue that German satises the Sub ject Condition even if it has

clauses without a cstructure sub jectconstituent In particular I argue that expletive

sub jects are provided by the verbal agreement morphology from this it follows that the

insertion of an expletive sub ject in cstructure is not necessary to satisfy the Sub ject

Condition and therfore by Economy of Expression is prohibited

Sub ject Condition

In this pap er I will address the question whether German is a language that satises the

Sub ject Condition cf Baker ie put informally whether every sentence in German

has a sub ject

A denition of the Sub ject Condition is given in Alsina

Sub ject Condition

An fstructure with prop ositional contentmust include a sub ject as one of its gram

matical functions and no fstructure may include more than one sub ject

The Sub ject Condition is a condition on fstructures but since the fstructure is constructed

by the mapping function from cstructure the rst question is how the sub ject is identied

in cstructure in German We will see that German has clauses without a phonologically

realized sub ject hence without a representation in cstructure However since the function

is not onto a second question arises Do es the Sub ject Condition hold in German when

there is no phonologically realized sub ject and if so how is it satised

It is still controversial whether the Sub ject Condition in LFG should b e a universal or not

In the following an analysis will b e presented whichshows that German is consistent with

the universal status of the Sub ject Condition and it will b e shown that this follows without

further stipulation from the analysis of agreementinLFG

Function sp ecication in German

Languages use dierent kinds of function sp ecication in particular a distinction is made

between structural and morphological function sp ecication eg Bresnan

According to Alsina the Sub ject Condition is if not an absolute universal at least the un

marked parameter setting Mohanan concludes The status of the Sub ject Condition as

an inviolable universal is thereby called into question by this construction Given that the motivation for

the Sub ject Condition is crosslinguistically robust the condition can by no means b e abandoned The only

conclusion op en to us then is that the Sub ject Condition may b e violated under certain sp ecial circum

stances BresnanZaenen fo otnote remark that this condition may need to b e parameterized

so as to hold only for some typ es of languages BresnanKanerva p oint out the generality of the

sub ject condition due to Baker is op en to question b ecause many languages have constructions

in whichthereisnoovert sub ject It remains unclear whether these cases involveanempty nonlogical

sub ject as prop osed by Baker or whether the sub ject condition itself is languagedep endent

LFG J Berman Do es German Satisfy the Sub ject Condition

Haider Following Haider I assume that in German grammatical

functions are identied morphologically and not by structural p osition It follows that

the sub ject and the ob ject do not have to occupy distinct unambigous p ositions and

there needs to be only one functional pro jection namely the CP Consequently German

has no distinguished sub ject p osition The assumed sentence structure cf Haider

FreyTapp e and its corresp onding fstructure are given in

weil der Vater den Jungen lobt

b ecause the father nom the b oy acc praises



 



subj pred der Vater







pred lob en subj obj



CP



tense present



 



 

obj den Jungen

C VP

weil

SUBJ VP

DP

der Vater

case nom OBJ VP

DP

den Jungen V

case acc lobt

Typ es of morphological function sp ecication

Based on Nichols Bresnan schematized two general typ es of morphological

function sp ecications namely dep endentmarking and headmarking

Typ es of morphological function sp ecication

a dep endent marking

CASE k GF

b headmarking

AGR AF AGR AF

In German both typ es of morphological function sp ecication can be observed with the

sub ject it is marked nominative which is an instance of dep endentmarking and it has

to agree in numb er and p erson with the nite verb which is an instance of headmarking

The sentences in illustrate the morphological function sp ecication of the sub ject ex

emplied by the sub ject of the verb lachen

LFG J Berman Do es German Satisfy the Sub ject Condition

a Der Junge lacht

The b oy nom laugh p erssgp erspl

b Dendemdes Jungens lacht

The b oy accdatgen laugh p erssgp erspl

c Der Junge lachst

The b oy nom laugh p erssg

d Der Junge lachen

The b oy nom laugh p erspl

In a the DP is marked nominative and has the same p erson and numb er features as the

nite verb b shows that the sub ject must b e marked nominative no other casemarking

is p ossible In c there is numb er agreement but no p erson agreement in d there is

p erson agreement but no numb er agreement

Hence for the sub ject in German the schema can b e sp ecied as in

CASE NOM SUBJ AGR SUBJ AGR SUBJ

Consequently the Sub ject Condition is always satised if there is a nominativemarked DP

in cstructure that agrees with the verb

There are however certain typ es of sentences in German which lack a nominativemarked

agreeing DP In the following it will b e argued that even in these cases the Sub ject Condition

is satised I claim that this pattern results from the interaction of casemarking and verbal

agreement morphology in German

Clauses without a nominative DP

Default vs Lexical casemarking

In German there is a distinction between default casemarking and lexical casemarking

As was shown by ZaenenMalingThrainsson default case marking is sensitive

to surface grammatical relations and results in nominative sub jects and accusative ob jects

while lexical casemarking is an idiosyncratic prop erty of the lexical item and is asso ciated

with a particular thematic role In German lexically casemarked arguments are obligatorily

mapp ed onto OBJ ZaenenMalingThrainsson

Consider the contrast between the verbs loben to praise and helfen to help Loben is a

verb whichhas no lexically casemarked arguments According to the mapping principles

formulated in BresnanZaenen see the agent which is intrinsically marked

o is mapp ed onto the sub ject and the theme which is intrinsically marked r is

mapp ed onto the ob ject

Grammatical functions are dened by the features r thematically restricted or not and o ob

jective or not A sub ject corresp onds to
o
r an ob ject to o
r ob ject to o r and oblique

to
o r

LFG J Berman Do es German Satisfy the Sub ject Condition

Mapping Principles BresnanZaenen

a Sub ject roles

i is mapp ed onto SUBJ otherwise

o

ii is mapp ed onto the SUBJ

r

b Other roles are mapp ed onto the lowest compatible function on the markedness

hierarchy SUBJ OBJ OBL OBJ

lob en hagent theme i

o r

subj obj

The sub ject is asso ciated by default with nominative case the ob ject is asso ciated by

default with accusative case

weil der Vater den Jungen lobt

b ecause the father nom the b oy acc praises

In the passive construction the agent is suppressed so the theme argument being intrin

sically marked r is mapp ed onto the sub ject and satises the Sub ject Condition as

shown in

lob en hagent theme i

o r

subj

Hence in the passive the theme must b e realized as a nominative DP

weil der Junge den Jungen gelobt wird

b ecause the b oy nom the b oyacc praised is

In contrast lexical case is asso ciated with a sp ecic thematic role In German casemarked

thematic roles are obligatorily mapp ed onto OBJ So in the passivevoice helfen do es not

sub categorize for a SUBJ but for an OBJ



There are no quirky sub jects in German see ZaenenMalingThrainsson for a comparison of

Icelandic nonnominative marked sub jects and German dative marked DPs which can not function as a sub ject

LFG J Berman Do es German Satisfy the Sub ject Condition

helfen h agent b eneciary i

o oDAT

OBJ

Lexical case do es not change when the clause is passivized In it is shown that the

dative is also retained in the passivevoice

a weil der Vater dem Madchen hilft

b ecause the father the girl dat helps

b weil dem Madchen geholfen wird

b ecause the girl dat help ed is

So in sentences like b there is no nominative marked DP

The same thing is seen with certain verbs in the activevoice whichhave only one lexically

casemarked argument and oblique functions

a cf Reis

weil mir an einer schnellen Losung liegt

b ecause me dat at a quick solution lies

b ecause I am interested in a quick solution

b weil mir davor graut

b ecause me dat of it dreads

b ecause I dread it

c weil mich friert

b ecause me acc coldis

b ecause I am cold

Again due to lexical casemarking these verbs do not sub categorize for a SUBJ but an

OBJ Therefore sentences with such verbs as in b and violate the Sub ject

Condition unless there is an alternativewaytoprovide a sub ject

Imp ersonal Passive

Another case where the Sub ject Condition seems to b e violated is the imp ersonal passive

of intransitiveverbs where no argument is left as in

weil gestern im Wald getanzt wurde

b ecause yesterday in the wo o ds danced was

b ecause there was dancing in the wo o ds yesterday

LFG J Berman Do es German Satisfy the Sub ject Condition

When intransitiveverbs are passivized the only argument is suppressed and the verb do es

not sub categorize for any function at all This results in sentences containing only the

predicate and optionally adjuncts

In contrast to German in many languages with imp ersonal passive an expletive sub ject

must be inserted For example in Danish Swedish and Norwegian the sub ject p osition

mustbelledby an expletive sub ject as shown in

Sten Vikner pc

a Vielleicht wird getanzt German

p erhaps is danced

b Maske bliver der danset Danish

p erhaps is there danced

c Kanskje blir det danset Norwegian

p erhaps is it danced

d Kanske dansas det Swedish

p erhaps dancedis it

In the following it will b e argued that this dierence is due to the role of verbal morphology

in the dierent languages

The role of the verbal agreement morphology

We have seen that in German the sub ject agrees with the nite verb in p erson and num

ber Following the analysis of sub jectverbagreement in KaplanBresnan and

Bresnan chap the sub jectverb agreement is enco ded on the verbal ax

The verbal ax contains on the one hand information ab out tense and mo o d and on

the other hand it sp ecies that its sub ject must have certain number and p erson values

The verb is able to sp ecify features of the sub ject b ecause it is the fstructure head of the

clause The entry of a verbal ax is given in

t V TENSE present

inf l

MOOD indicative

SUBJ

PERS

NUM sg

CASE nom

The p erson number and case features of the sub ject provided by the verbal agreement

morphology are unied with the features of the sub ject DP and guarantee sub jectverb

agreement For example the agreement features of the verbal ax t are compatible with

the agreement features of the pronoun er The lexical entry of er is given in

LFG J Berman Do es German Satisfy the Sub ject Condition

er N PRED pro

BIND

NUM sg

PERS

GEND mas

CASE nom

Hence the information of the ax t and the pronoun er can b e unied as shown in

weil er lacht

b ecause he laughs







pred pro









bind









pers





subj







gend mas





VP







num sg

 





case nom



SUBJ





pred lachen subj

DP VP

 

tense present

V

er lacht

Sub jectverbagreement is an instance of headmarking The sub ject is sp ecied on the

head of the clause the verb But since the verbal ax sp ecies only p erson number

and case features of the sub ject a corresp onding DP is usually obligatory But then the

question is What happ ens in sentences like b or when no sub ject DP is

presentandthus no unication can take place

There are languages in which arguments do not have to b e represented in cstructure For

example BresnanMchombo claim that in Chichewa the optional absence of the

sub ject NP can b e analyzed as an instance of prodrop The ob ject in Chichewa however

is sp ecied only by the verbal morphology and is an example of pronoun incorp oration

Hence in this case no ob ject NP can o ccur To see whether clauses without a cstructure

sub ject in German can b e analysed along these lines I will briey summarize the analysis

of prodrop and pronoun incorp oration given in BresnanMchombo and Bresnan

Pronoun incorp oration and prodrop

Bresnan chap suggests that pronoun incorporation or pronominal inection

can b e analyzed in this way an incorp orated pronoun or pronominal inection is a b ound

morpheme that sp ecies a complete pronominal fstructure The functional sp ecication



On pronominal incorp oration and prodrop see also Mereu and BorjarsChapman

LFG J Berman Do es German Satisfy the Sub ject Condition

of a pronoun is incorp orated with the functional sp ecications of the stem to which the

morpheme is b ound

In Chichewa the ax wa sp ecies an ob ject fstructure containing semantic and agreement

features This is illustrated by the following example

Bresnan chap b

Njuc hi zinawa luma

b ee SPSTObiteFV

The b ees bit them

The ob ject marker wa carries the following information

OM V OBJ

inf l

PRED pro

AGR

Since wa sp ecies the PREDfeature of the ob ject and since PREDfeatures are treated

as instantiated symb ols which cant b e unied even if they are identical KaplanBresnan

no ob ject DP can o ccur

The sub ject in Chichewa is also sp ecied by the verbal morphology with the dierence

that in this case the semantic feature is optional

SM V SUBJ

inf l

PRED pro

AGR

This gives us two p ossibilities If the semantic feature is present the sub ject is sp ecied

by the morphology If the semantic feature is absent it is provided by the corresp onding

DP and wehave sub jectverbagreement

In German prodrop and pronoun incorp oration are not p ossible since a thematic sub ject

can never b e omitted as shown in

weil er lacht

b ecause he laughs

This indicates that in German the verbal agreement morphology do es not provide a PRED

feature either optionally or obligatorily Hence in the sub ject DP providing the

semantic feature has to b e present and wehave sub jectverbagreement

However it was shown that there are verbs in German which do not sub categorize for a

sub ject namely certain verbs with lexicallycasemarked arguments and imp ersonal pas

sives In the following I will argue that in these cases the verbal morphology can sp ecify a

sub ject more precisely an expletive sub ject



Except in the case of topicdrop where a sub ject or an ob ject in sentenceinitial p osition maybeom itted under certain pragmatic conditions cf Huang

LFG J Berman Do es German Satisfy the Sub ject Condition

The expletive sub ject provided by the morphology

Wehave seen that in German the verbal morphology do es not provide a semantic feature

Notice however that the verbal agreement morphology nevertheless sp ecies a sub ject

containing only p erson numb er and case information as shown in







pers









subj num sg

 



t V TENSE present

inf l



case nom



SUBJ

 

tense present

PERS

NUM sg

CASE nom

Normally this information is unied with the features of the sub ject DP But what happ ens

now if no sub ject DP is present In this case to o the verbal agreement morphology

intro duces a sub ject in the fstructure namely a sub ject without a semantic feature A

sub ject which has no semantic feature but only agreement features is an expletivesubject

Ifaverb sub categorizes for a thematic sematically contentful sub ject like the verb lachen

the sub ject provided by the morphology do es not satisfy Completeness since it has no

semantic feature However if the verb do es not sub categorize for a thematic sub ject the

sp ecication of an expletive sub ject is ne and furthermore required to satisfy the Sub ject

Condition Notice that the insertion of the expletive sub ject by the verbal morphology is

just a consequence of sub jectverb agreement and do es not need to b e stipulated On the

contrary further constraints would be necessary in order to prevent the insertion of the

expletive sub ject

The mapping from c to fstructure of a sentence without a sub ject DP is illustrated in

Grauen has a lexically casemarked argument which is mapp ed onto OBJ The Sub ject

Condition is satised by the verbal agreement morphology

weil mir graut

b ecause me dat dreads



I put aside the question whether sentences mayhave the sub ject function or not since this would take me to o far aeld

LFG J Berman Do es German Satisfy the Sub ject Condition





pers









subj num sg



 





case nom







pred grauen obj





tense present CP









pred pro





C VP





pers





obj



weil



num sg



 

 

case dat

OBJ VP

DP

mir

V

grau t

On this analysis every nite clause satises the Sub ject Condition since either the sub ject

sp ecied by the agreement morphology is unied with a sub ject DP or the agreement

morphology sp ecies an expletive sub ject

Notice that sentences like

a weil lacht

b ecause laughs

b weil sieht das Haus

b ecause sees the house

do satisfy the Sub ject Condition under this analysis but violate Completeness since lachen

and sehen sub categorize for a thematic sub ject

However there is a restriction an expletive sub ject can only be intro duced by the ax

which is marked rd p erson singular All the other morphological forms lead to ungram

maticality

a weil getanzt wird p erssg

b weil getanzt wirst p erssg

c weil getanzt werde p erssg

d weil getanzt werden p erspl

e weil getanzt werdet p erspl

But this is in accordance with the observation that the expletive DP in other languages

is also rd p erson singular as for example Swedish or Norwegian det or Icelandic Ba It

seems that this is the only sp ecication that allows an interpretation without semantic

content This observation can be captured by the following wellformedness condition on fstructures

LFG J Berman Do es German Satisfy the Sub ject Condition

All fstructures must have a PREDfeature unless they are sp ecied for rd p erson

singular

f f PRED f NUM sg f PERS

Another question which arises is why the sub ject intro duced by the verbal morphology

do es not violate Coherence since the verb do es not sub categorize for a sub ject Compare

where grauen sub categorizes only an OBJ or the passive of an intransitive verb as

in where tanzen in the passivevoice do es not sub categorize for any function at all

weil getanzt wird

b ecause danced is

b ecause p eople danceare dancing





pers











subj num sg

 





case nom







pred tanzen





passive

 

tense present

Alsina claims that grammatical functions can b e licensed in three dierentways

Coherence Alsina

a

b a direct function must b e licensed either i by a mapping principle ii bya

general principle such as the Sub ject Condition or iii by a lexical stipulation

Following Alsinas claim that nonthematic functions are not represented in astructure

pp an expletive sub ject cannot b e licensed by a mapping principle but b eing a

sub ject it is licensed by the Sub ject Condition itself This is the case for expletive sub jects

intro duced by the morphology as in German as well as for the expletive sub ject DP as

for example in Norwegian

Kanskje blir det danset

p erhaps is it danced

Also in the Norwegian case the expletive sub ject DP det is not sub categorized by the verb

but it is licensed by the Sub ject Condition

LFG J Berman Do es German Satisfy the Sub ject Condition

Predictions of the analysis

The analysis presented here makes a prediction for other languages a language which

do es not have sub jectverbagreement cant intro duce an expletive sub ject via the verbal

morphology In order for such languages to satisfy the Sub ject Condition the expletive

sub ject must b e intro duced in another way

This is the case for languages like Danish Norwegian and Swedish Danish Norwegian

and Swedish do not have sub jectverb agreement Hence the verbal morphology do es not

sp ecify an expletive sub ject rather the sub ject must b e identied structurally in IPSp ec

The examples are rep eated for convenience

a Maske bliver der danset Danish

p erhaps is there danced

b Kanskje blir det danset Norwegian

p erhaps is it danced

c Kanske dansas det Swedish

p erhaps dancedis it

Yiddish and Icelandic do have sub jectverbagreement So the expletive sub ject can be

intro duced by the verbal agreement morphology The prediction of the analysis presented

here that no expletive DP o ccurs is b orne out The examples are given in

Sten Vikner pc

a Efsher vert es getantst Yiddish

p erhaps is it danced

b Kannski er Ba dansa Icelandic

p erhaps is it danced

An apparentcounterexample is English which has b oth sub jectverbagreement and a struc

tural expletive sub ject

Perhaps there dances a man



In case there is a sentenceinitial lo cative PP or a lo cativeadverb the expletive sub ject is optional cf

Vikner for Danish Falk and Kersti Borjars pc for Swedish Helge Dyvik p ointed out to

me that it also holds in Norwegian For example Sten Vikner pc

i I huset bliver der danset in house is there danced Danish

ii I huset blir det danset in house is it danced Norwegian

iii I huset dansas det in house dancedis it Swedish

However in this case the PP functions as a sub ject and is an instance of lo cative inversion whichhas a

sp ecic discourse function cf Bresnan

LFG J Berman Do es German Satisfy the Sub ject Condition

However English has only partial sub jectverbagreement mainly in the present tense

rd p erson singular so we can assume that in English the verbal morphology is to o weak

cf Rohrbacher Vikner to intro duce a sub ject by itself Weakness of the

morphology can b e expressed by constraining equations on the agreement features This is

illustrated by the English verbal ax s

s V TENSE present

inf l

SUBJ

PERS

c

NUM sg

c

CASE nom

Thus in English the agreement features of the verbal morphology must be licensed by a

co o ccurring DP Consequently in English imp ersonal passive is ungrammatical

Perhaps there was danced

German as a semiprodroplanguage

The prop osal that German has a phonologically unrealized expletive sub ject recalls the

claim made by eg Sar Cardinaletti Grewendorf Vikner

that German is a semiprodroplanguage disallowing an argumental null sub ject pro

but licensing an expletive pro The expletive pro o ccupies the IPSp ecp osition and the

Extended Pro jection Principle Chomsky is fullled

However the analysis presented here and the semiprodrop analysis are dierent in several

asp ects

First the prop onents of the semiprodrop analysis claim that the expletive is represented

in phrasestructure by the empty category pro as shown in

Vikner

weil pro getanzt worden ist

IP VP

b ecause danced been is

b ecause there has b een dancing



On the analysis of agreement in sentences like see eg Alsina Baker Bresnan



Dutch is more problematic under this analysis Dutch has sub jectverbagreement and a structural

sub ject expletive er there But in contrast to English in Dutch imp ersonal passive is grammatical

i Misschien wordt er gedanst p erhaps is there danced

Constraining equations on the verbal agreement morphology wouldnt capture the Dutchdata since the

expletive er could not license them However notice that there is a dierence b etween the weakness of

the English and the Dutch morphology While English has neither p erson nor numb er agreement in all

tenses Dutch has numb er agreement but no p ersonagreement in all tenses cf Vikner This might

suggest that the analysis of the Dutch morphology must b e captured in a dierentway from the English

morphology There are also consequences for the formulation of HoweverIleave these issues for future

research

LFG J Berman Do es German Satisfy the Sub ject Condition

But since the claim defended here is that the functions in German are identied morpho

logically an empty category lacking any morphology couldnt b e identied as the sub ject

at all Note that even if one assumed that German has structural function sp ecication in

addition to morphological function sp ecication cf Choi an empty category in the

sub ject p osition would b e ruled out by Economy of Expression see b elow So in contrast

to the analysis prop osed by Vikner and others on the present analysis the expletivesubject

is not represented in cstructure byanempty element but is intro duced in fstructure by

the verbal agreement morphology

Second the prop onents of the semiprodropanalysis assume that in the case of an ergative

or passivized verb the argument is basegenerated in the ob ject p osition and the sub ject

p osition has to b e o ccupied by the expletive pro Grewendorf prop oses struc

tures as in

a weil pro dem Hans ein Fehler unterlaufen ist INFL

S i VP i

b ecause the Hans dat a mistake nom happ ened is

b weil pro dem Hans der Arm verbunden wurde INFL

S i VP i

b ecause the Hans dat the arm nom bandaged was

However on the analysis b eing presented here a nominativemarked agreeing DP is iden

tied as the sub ject indep endent of its thematic role and its structural p osition Therefore

also in sentences like the nominative marked DP is identied as the sub ject and agrees

with the nite verb

Is there an expletive sub ject DP in German

Wehave seen that with imp ersonal passive and with a few verbs in the activevoice the Sub

ject Condition is satised bytheverbal agreement morphology Notice that the additional

insertion of the expletiveDPes with imp ersonal passive is ungrammatical

weil es getanzt wurde

b ecause it danced was

This follows from the present analysis straightforwardly

The sub ject is already sp ecied by the verbal agreement morphology Hence the o ccurrence

of an additional structural expletive sub ject would give only redundant information and

therefore violate Economy of Expression

Economy of Expression Bresnan chap

All syntactic phrase structure no des are optional and are not used unless required by

indep endent principles completeness coherence semantic expressivity

The same is true for some verbs in the activevoice

LFG J Berman Do es German Satisfy the Sub ject Condition

weil es mir an einer schnellen Losung liegt

b ecause it me dat at a quick solution lies

becauseIaminterested in a quick solution

Again a structural sub ject expletive cant o ccur since in contrast to the sub ject provided

by the morphology the expletiveDPwould require additional phrase structure no des

This means that according to the analysis presented here we exp ect that German has no

expletive sub ject DP at all since every nite verb p otentially sp ecies an expletivesubject

and hence the expletive es would always violate Economy of Expression

The prediction of the analysis that es in the middle eld cant be an expletive is in

agreement with the view taken by Cardinaletti Vikner and Fortmann

following Ho ekstra who claim that es in the middle eld is always a quasi

argument However in the following I will present some examples where the status of es as

a quasiarguments is questionable and it will b e shown how an expletive sub ject DP could

b e handled by the present analysis

Evidence for es as a quasiargument

Es can o ccur in the middle eld either obligatorily a and b or optionally c

and d

a weil es regnet

b ecause it rains

b weil es keine Honung gab

b ecause it no hop e gave

b ecause there was no hop e

c weil es mich friert

b ecause it me coldis

b ecause I am cold

d weil es mir graut

b ecause it me is dreadful

b ecause I dread it

Note that es in CPSp ec is grammatical

i Es wird getanzt there is dancing

ii Es kamen viele Linguisten there came a lot of linguists

However in this case es is required to ll CPSp ec whichmay not remain empty in a declarativesentence

Following Grewendorf Fortmann and the analysis of the traditional grammar I assume that

es in sentences like i and ii do es not have sub ject function

In traditional grammar the German sentence is divided into three elds the VorfeldtheMittelfeld and

the Nachfeld The Mittelfeld middle eld is that part of a main verbsecond clause b etween the nite verb

and any nonnite verbal forms or a separable verb prex or that part of an emb edded verbnal clause

between the complementizer and the verb complex A constituent that precedes the nite verb in a main

clause o ccupies the Vorfeld preeld those following the verbal complex o ccupytheNachfeld p osteld

LFG J Berman Do es German Satisfy the Sub ject Condition

Following Bennis Cardinaletti Vikner and Fortmann supp ort their claim that es

here is an argumentby showing that es can o ccur in control constructions

a Fortmann a

gestern hat es geblitzt ohne zu donnern

yesterday has it ashedlightning without to thunder

yesterday there was lightning without thundering

b Fortmann b

gestern hat es geregnet anstatt zu schneien

yesterday has it rained instead to snow

yesterday it rained instead of snowing

However with other verbs than weather verbs it is more doubtful whether control is

p ossible

Christine Czinglar handout GGS Stuttgart example a

Es gibt in dieser Gegend viele Tump el ohne jemals richtig zu regnen

It gives in this area a lot of p onds without ever really to rain

There are lots of p onds in this area without ever really raining

It is b eyond the scop e of this pap er to discuss the interpretation of these data in detail We

have seen that the analysis presented here supp orts the claim that there is no structural

expletive in German However if it turned out that German had an expletive sub ject DP

in cases like the expletive do es not have to b e inserted to satisfy the Sub ject Condition

b ecause this could b e done bytheverbal agreement morphology So I assume that if there

is an expletive sub ject DP the verb has to sub categorize for it and the verb has sp ecial

requirements on the realization of its sub ject namely it requires a nonthematic sub ject

and the sub ject must b e realized by es In this regard they act like an idiomatic expressions

Formally this could b e captured by a constraining equation

geb en V PRED geb en OBJ SUBJ

SUBJ FORM es

c

The sub ject in nonnite clauses

Finally nonnite verbs provide further evidence for the claim that the agreement mor

phology sp ecies an expletive sub ject Compare the following examples

a die Moglichkeit dass getanzt wird

the p ossibility that danced is

the p ossibility that there is dancing

b die Moglichkeit getanzt zu werden

the p ossibility danced to b e

LFG J Berman Do es German Satisfy the Sub ject Condition

The imp ersonal passive is only p ossible in nite clauses but not in nonnite clauses The

nonnite clause in b would b e an example of arbitrary control With arbitrary control

the sub ject is intro duced by the following rule

Rule of Functional Anaphora Bresnan

For all lexical entries L assign the optional pair of equations SUBJ PRED

pro FIN to L Bresnan

c

In German we can assume that the equation SUBJ PRED pro is optionally asso ciated

with the ax zu as shown in

zu V SUBJ PRED pro

inf l

The nonnite clause zu tanzen therefore has the fstructure given in

zu tanzen

to dance



 

subj pred pro



 

pred tanzen subj

The optional insertion of a thematic sub ject it contains a PREDfeature by the ax zu

explains why the examples in are grammatical although in cstructure no sub ject is

realized b shows further that a passive of a transitiveverb is grammatical

a die Moglichkeit zu tanzen

the p ossibility of dancing

b die Moglichkeit verstanden zu werden

the p ossibility understo o d to b e

the p ossibility of b eing understo o d

The imp ersonal passive where an expletive sub ject would have to b e inserted to satisfy the

Sub ject Condition is however ungrammatical as we have seen in b This follows

from the present analysis straightforwardly

Since nonnite verbs do not show agreement with the sub ject the verbal morphology cant

sp ecify an expletive sub ject So a nonnite verb never intro duces an expletive sub ject in

the fstructure Hence in example b the Sub ject Condition is violated

It could b e ob jected that the controlled sub ject of a nonnite clause must always b e thematic Haider

fo otnote p oints out in the framework of GB that it is only a stipulation that PROhasbe

thetamarked and there is no explanation provided In LFG the intro duction of the sub ject by the ax zu

must b e optional to account for functional control and coherent innitives with zu So the insertion of an

expletive sub ject must b e imp ossible for other reasons

LFG J Berman Do es German Satisfy the Sub ject Condition

Conclusion

It was argued that German is a language which despite having clauses without a structural

sub ject do es satisfy the Sub ject Condition and is therefore consistent with the claim that

the Sub ject Condition is universal In particular I argued that the expletive sub ject is

provided by the verbal agreement morphology which follows from the analysis of sub ject

verbagreement without further assumptions

Acknowledgements

IamgratefultoSteve Berman Kersti Borjars Joan Bresnan Miriam Butt Stefanie Dipp er

Christian Fortmann Jonas Kuhn Gereon Muller Sten Vikner and twoanonymous referees

for helpful discussions and comments Thanks to Steve also for correcting my English This

research was supp orted by the DFG Graduiertenkolleg Linguistische Grundlagen fur die

Sprachverarb eitung at the University of Stuttgart

References

Alsina A The Role of Argument StructureinGrammar CSLI Stanford

Baker M Ob jects themes and lexical rules in Italian In L Levin et al eds Papers in

LexicalFunctional Grammar

Bennis H Gaps and Dummies Foris Dordrecht

Borjars KChapman C Agreement and prodrop in some dialects of English Linguistics

Brandner E The pro jection of categories and the nature of agreement In G Fanselow

ed The parametrization of universal grammar Benjamins Amsterdam

Bresnan J Control and Complementation Linguistic Inquiry

Bresnan J Lo cativeInversion and the Architecture of Universal Grammar Language

Bresnan J LexicalFunctional Syntax In J BresnanL Sadler Model ling Dynamic

Interactions between Morphology and Syntax Reader of the th Euop ean Summer Scho ol in

Logic Language and Information Saarbruc ken

Bresnan JKanerva J Lo cativeInversion in Chichewa A case Study of Factorization in

Grammar Linguistic Inquiry

Bresnan JMchombo S Topic Pronoun and Agreement in Chichewa Language

Bresnan JZaenen A Deep UnaccusativityinLFG In K DziwirekPFarrell E Mejas

Bikandi eds Grammatical Relations A CrossTheoretical Perspective CSLI Stanford

Cardinaletti A Impersonal constructions and sentential arguments in German Unipress

Padua

Choi HW Optimizing Structure in Context Scrambling and Information Structure CSLI Stanford

LFG J Berman Do es German Satisfy the Sub ject Condition

Chomsky N Some Concepts and Consequences on the Theory of Government and Binding

MIT Press Cambridge

Falk C Sub jectless Clauses in Swedish Working Papers in Scandinavian Syntax

Fortmann Ch Wetterverben und ahnliche unpersonliche Pradikate Ms Universitat

Stuttgart

FreyWTapp e H T Zur Interpretation der Xbar Theorie und zur Syntax des Mittelfelds

Ms Universitat Stuttgart

Grewendorf G Ergativity in German Foris Dordrecht

Grewendorf G Small pro in German In G GrewendorfW Sternefeld eds Scrambling

and Barriers Benjamins Amsterdam

Haider H PROBLEME In G FanselowS Felix eds Strukturen und Merkmale syn

taktischer Kategorien Narr Tubingen

Haider H Deutsche Syntax generativ Narr Tubingen

Ho ekstra T The distribution of sentential complements In H BennisWUS van Lessen

Klo eke eds Linguistics in the Netherlands Foris Dordrecht

Huang CT J On the Distribution and References of Empty Pronouns In Linguistic

Inquiry

Kaplan RBresnan J LexicalFunctional Grammar AFormal System for Grammatical

Representation In J Bresnan ed The Mental Representation of Grammatical Relations

MIT Press Cambridge

Mereu L For a lexicalfunctional representation of agreement axes In M ButtT King

eds Proceedings of the LFG Conference University of California San Diego

Mohanan T Argument Structure in Hindi CSLI Stanford

Nichols J Headmarking and dep endentmarking grammar Language

Paardekooper P Beknopte ABNSyntaxis Zevende druk sterk uitgebreid Uitgave in

eigen b eheer Eindhoven

Reis M Zum Sub jektb egri im Deutschen In W Abraham ed Satzglieder im

Deutschen

Rohrbacher B The Germanic Languages and the Ful l Paradigm aTheory of V to I Raising

PhD University of Massachusetts Amherst

Sar K Syntactic Chains Cambridge University Press Cambridge

Vikner S Verb Movement and Expletive Subjects in the Germanic Languages Oxford

University Press Oxford

o o

Vikner S V toI Movementand Inection for Person in All Tenses In L Haegeman

ed The New Comparative Syntax Longman London

Zaenen AMaling JThrainsson H Case and Grammatical Functions The Icelandic

Passive Natural Language Linguistic Theory