Disambiguation by Information Structure in DRT 
Kurt Eberle * 
Institut fiir Maschinelle Sprachverarbeitung 
Universitgt Stuttgart 
Azenbergstr. 12 
D-70174 Stuttgart 
Abstract 
~i%xt understanding and high quality 
machine translation often necessitate 
the disambiguation of ambigous struc- 
tures or lexical elements. Drawing in- 
ferences from the context can be a 
means for resolving semantic ambigu- 
ities. However, often, this is an ex-. 
pensive strategy that, in addition, not 
always comes up with a clear prefer- 
ence for one of the alternatives. In 
this paper, we argue that in a num- 
ber of cases deep semantic analyses can 
be avoided by taking into account the 
constraints that the alternative readings 
impose onto the information structure. 
To this end, we present a study of the 
arnbigous German adverb erst and point 
out the particular circumstances under 
which the given information structure 
disambiguates the adverb without fur- 
ther semantic analysis. 
1 Introduction 
German erst is ambigous. Consider the following 
examples: 
(1) Peter zeigte erst auf die vierte Gliickszahl. 
a) Und dann auf die zweite. 
(Peter first pointed to the fourth lucky number. 
And then to the second.) 
b) Nicht zuvor auf die erste, zweite oder dlqtte. 
(Peter only pointed to the fourth lucky number. 
Not to the first, second or third.) 
c) Noeh nieht aufdie Nnfte. 
(Peter only pointed to the fourth lucky number 
so far. Not yet to the fifth.) 
The alternative contexts a) - e) determine the 
meaning of the first sentence of (1) according to 
*This paper describes research done within the 
Sonderforschunsbereich 3~0 at IMS. I would like to 
thank Anette Frank, Hans Kamp, Michael Schiehlen 
and the other members of the IMS semantics group 
for helpfull discussion. 
the disambiguating translations presented. 'rhe 
example testifies the following three uses of crsl: 
• In the context (1.a), the recipient under- 
stands the introduced event as the first of 
a sequence of events that he expects to be 
completed by the following text. We call this 
reading: 
the First of a Sequence-reading (FS). 
• In the context (1.b), the recipient under- 
stands erst as a signal of the speaker/writer 
that the occurrence of the reported event is 
not preceded by the occurrence of similar (al- 
ternative) events. We call this reading: 
the Exclusion of Preceding Alternatives- 
reading (EPA). 
• In the context (1.c), the recipient under- 
stands the event as element of a sequence of 
events, and the realization of the sequence, 
in particular the reported realization of the 
event at the textual perspective time, seems 
to be in retardation, with regard to some 
(previous) expectation about the realization 
dates of the sequence. We call this reading: 
the Retardation-reading (R). 
As can be seen from the example, the contextual 
disambiguation not only is needed for understand- 
ing the text, but is a prerequisite for high quality 
translation. 
In the literature, different lbrmalizations have 
been discussed ((KSn79), (LSb89), (Kgn91), 
(HR81) (the latter one for the similar noch and 
schon) and others). With respect to the focus ad- 
verb use (the cases (1.b) and (1.c); (1.a) being 
an example of the temporal adverb use), rnod- 
ellings are prevailing that associate ers* with dif- 
ferent scales (cf. (I~5b89)). However, a precise 
evaluation of the context that can decide about 
the relevant reading (for instance, what infor- 
mation defines which scale) is still missing. We 
tackle this problem in the framework of Discourse 
Representation Theory (DRT) (Kam81), assum- 
ing that discourse representations (DRSs) may be 
augmented by information structure. 
334 
2 The cr.d.-readings, their 
entailments, presuppositions and 
implicatures 
2.1 The l,'irst of a ,%qucuce intert)re, tation 
Ill this paper, we cmmu)t go into detail with tests 
that partition tile meaning of a sentence into pre- 
sui)positions , assertions proper and inlplicatures 
the recipient is allowed to drmw from the sentence (of. (Lev8a) 
for ~m overview of the notions used, 
the tests associated with them and the problems 
connected to them). We directly come up with 
tile I)RSs that, to our opinion, represc.nt the im- 
pact of the different remdings. We begin with the 
FSeeading of (1): 
(1 .a) 
~7t77~ 
~\['P(t) t -4 now 
e: zeig~~) 
e starts e 
llere, TP stands for the actual (past) temporal 
perspective that holds for the given utter- 
anee/te×t sil;uation. With respect to the represen- 
tation and resolution of presuppositions in DR/F, 
we relate to (vdS92). We slightly deviate from 
the approach suggested there, however, in that 
we use a notation for the llresui)position triggers 
that is akin to the suggestion of(BgS + 94): we use 
so-calle<\[ ~-condilions which describe the presup- 
posed objects and their characterization. In our 
framework (,-conditions sllhsHlrte (tilDrent types 
of i)roje.ction problems. The type is characterized 
by the r.>in(tex (prcsp strands for 'classical' presup- 
positions, dcf for detlnite descriptions, rt for ref- 
erettce tilne: ~"e \['or re\['ereucx~ event et(:.~ cOm|lare 
(1%e95) for an overview). This index triggers the 
projectiou routine that is sl)ecific to the respec- 
tive resolution l>robhm,. The first argument of 
the condition schema highlights the distinguished 
discourse referent (I)I{F) of the structure to be 
projected, t The representation, thus, requires a 
definite description style resolution for the x that 
ix characterized as a fourth lucky number (with 
respect to the given setting, x together with its 
description will he accomodated at the main level 
I)R.S), and it pre.supposes a reference event c' such 
that the senteuce (;vent c is the tirst event of m 
possible elabormtiou sequence of c' (cf. (Ebe92) 
\['or the computation mad storing of discourse relm- 
I As regards delhtite descriptions, the distinguished 
t)I{.F is the l)l{.\]" of the heard \[(oull; in a.tty case it is 
the I)RF the r~-information is (:entered a.round. 
Modeltheoretic~d\[y the rel~ttion between the presup- 
positional p~rt ~md the asse.rtional part of a DRS can 
be seen as a function from information states into in- 
formation st~ttes, see (Kam95). 
tions like elaboration). 
2.2 The Exclusion of Preceding Alternatives 
interpretation 
In tile EPA-reading, crsl is used ms a focus mdverb, 
i.e. it structures its argument into focus and back- 
ground. In the following representation of (l), wc 
consider the case where the numeral ix focused 
only, not the NP containing the numeral or the 
entire event description in the scope of crM: 
(t % 
peter now ~.~.,Te4 
TP(t) t -4 now 
P e: zeigen_auf(peter,x) 
e(at 
1~ = ":--t ek 
A i=k--I f=l (e i -4 ei+t) 
A/=k--1 Pi ~O Pi+I i=l 
e C e4 
\]4 .... ..g . ( , ( , ! 
imp~t.',l *: AT'r({,<I,~XI',| now -4 ~-l>) 
Besides the assertion that Peter points to the 
fourth lucky number at the temporal perspective 
t, the representation presupposes a sum l!', con-. 
sisting of a sequence of events el - ek that are 
related via a non fln'ther specified relation 1{. to 
l)redicates I'~ - l)~. These It'; are required to be 
alternatives of the event descriptiou in the scope of 
crst (whic.h is called/~ e. I)RS~). In this paper, we 
cannot go into detail with the computation of the 
alternatiw:s of event descriptions. Pi 6 ALTD,t&, 
where AlfF,)ltS ={Ac. Dte,c;~\[~,~o(:o,(~t~s,OtO\] \] Q 
AlfI'(l"O(7~,t(l)R.,~e))}, sketches the claim that 
the Pi are (;vent descriptions that develop fl:om 
Ae.I)RS~ by exchanging the focused element by 
an ,,lte,',,ative (Conq, are (,~.oo85), (l~.oo.(~2)ror 
the underlying focus theory and (BE95) for an 
algorithm that calculates the concrete alterna- 
tives). As mentioned, the focused elemeut, 
which is marked by the underline, is assumed 
to be the numeral adjective. This choice trig- 
gers the structuring of the Ps into the brock- 
ground event type $c.13AO(I)\[L%.) (which is 
k e. ~2.uf(petetlx ) \]) and the fo(:used se- 
mantic constituent lx.l;~O(/(DR,S'e)\[4l n\] (which is 
335 
I I" \] AX. O~def(X, nte-~liicksze~hl(x) \]) ' where n is a number from the set of alternatives of 4). The relation R 
has to be understood as characterizing the e i as 
opportunities for Peter to point to (specific) num- 
bers. In (1), these opportunities may be situations 
that Call be described by: The firsle~ / second~ / 
lhird~ number is presented to Peter. Tile order of 
the oppertunities is inherited from the order of the 
Ps, -<o, which conforms to the intrinsic order of 
the set of Mternatives of the focused element (i.e. 
O=order(ALT(FOC~.I(DI~S~)))). In (1), this is 
the canonical order of the numbers. 
The negation test, which is comlnonly used to 
detect presuppositions, supports these strueturM 
assumptions. 
Now, we think that the EPA-reading inter- 
pretes tile asserted event, which is backed by the 
described scenario, as the lirst one that; is indeed 
realized within the range of possible instantiations 
that the sequence of opportunities provides, i.e. 
the asserted event presents the first positive out- 
come to tile test about the instantiation of the 
Ae.BAC(DRS~)-type that is connected to the ei- 
sequence, where each test situation el is charac- 
terized by its own specific additional test criterion 
kx. F O(;(1) I£Se )\[41n\] '2 
Further linguistic tests, that we must omit 
here, support the assumption that the information 
about the negative tests is an entailment. There- 
fore, in the representation, the negated DRSs for 
the pointing-opportunities el - e3 are part of the 
main DRS. 
In the literature the representation of the fo- 
('using use of ersl (and corresponding uses of noch 
and schon) often comprises the information that 
tile reported reMization of the event is earlier or 
later (depending on the reading and the adverb) 
than the speaker/writer and/or the recipient (or 
even a third person) would have expected. We 
think that such an expectation, in the case of 
the EPA-reading of erst, is only optional. Fol- 
lowing (K6n91), we think that, if present, it is not 
a presupposition but a conventional implicature. 
In (1.b), we use the specific c~i,,~,l-format and the 
representation convention of (Ram95) for attitudi- 
nal states in order to express the EPA-expectation 
of a previous test to be successful. 
2.3 The Retardation interpretation 
As for the EPA-reading, we consider the case 
where tile numeral is focussed only: 
2Depending on the tocus structure of tile phrase in 
the scope of erst in (1) and depending on the contex- 
tuM restrictions of the admissible alternatives, other 
sets of Ps might result. It is clear that depending 
on this choice, the focus conditions may characterize 
a thematic role., as in the described example, or the 
event wriable. It is ~dso clear, as we will argue in the 
next section, that not all of these sets of alternatives 
can accept the El~A-reading. 
(1.c) 
peter now e t 
TP(t) 
t -4 now x 
e: zeigen_auf(peter,x) e@t 
OZpresp(S, s:ATT(x , , <MODE, 
eE 
Ai=k --1 i=1 (el -< ei+l) 
AI:  P,(°,) 
i=1 
MODE=PLAN X=peterlA i 
realization_of(e,e4) I 1, o I'" \] 
/ 
F~ ~j3 e ~J = i=1 i 
A:-? reMiz~ti°n-°f(ei',eO 
S ineets e'l 
el' ~ e2' ~ e3' -~ e 
...... optional or probably obfigatory ..... 
Ottmpl(. ~ :A'PT(~ <lt \],:L, I~*l > 4 ) 
In contrast to the EPA-reading, we assume that, 
in the R-reading, the predicates Pi that we ob- 
tain from the information structure of the erst- 
argument are not related to a sequence of opportu- 
nities for doing something, but describe events ei 
of an expectation about the ongoing of the world 
or a plan e. The context (1) doesn't provide fin:- 
tiler information about the identity of the person 
or persons X to whom the introduced attitudinal 
state has to be ascribed, to the speaker/writer, to 
the recipient, to Peter, to someone else or to some 
group of salient people. Also it doesn't provide in- 
formation about the nature of the attitude MOI). 
The conditions that are introduced below the clot- 
ted line exemplify possible resolutions. According 
to this, Peter and the speaker/writer (i for the 
distinguished I)I{.F for the self) share the attitud(; 
of having a plan for realizing //. Such resolutions 
may be available by an inference component that 
operates over richer contextual knowledge. Ag~fin, 
tile order of the Pi and tile corresponding ei is in- 
ferred from the implicit or(ler of the alternatives 
of the focused element, l,'rom this setting and 
the assertion of an occurrence of Peter pointing 
to the fourth lucky number at the temporal per- 
spective t, the representation entails reMizations 
of those events of the presupposition line that 
precede the counterpart of e in the presupposed 
sequence. As a further (possibly optionM) con- 
strain|;, the l~-reading introduces the implicature 
that a noi~ further specified person or grout) ~ cx- 
336 
petted h)r the perst)ective time t that the plamtcd 
or expected sequence of events shouhl be realized 
to a greater degree. Without further information 
about the identity of ~, it is ditllcult to say some- 
thing more precise about the temporM location of 
the ~-expectation than that an instance s' of the 
corresponding attitudinal state holds at some time 
I)efore the actual now. 
3 Disarnbiguating Criteria 
3.1 Tim Syntax Criterion 
In (.;crman focus adverl)s cannot be topicMized as 
such, i.e. they cannot occur in the vorfeld po- 
sition without an accompanying constituent (cf. 
(KSn!)l)). If th('. sentence s|n)ws such a topicaliza- 
tion o\[' ersl-- which is marked by the inversion of 
the basic Subj-Vlin-order - ersl can only be used 
as a time adverb, i.e. its meaning can only be the 
li'S-reading, as exemplitied by (2): 
(2) Erst g'ab Peter Maria den Brief. 
(First, l)(!ter handed the letter to M~ri;t.) 
'Fhus, this tyt)e of topicMization disambiguates be- 
tween the I,'S-reading ou the one hand and the 
El'A- and R-reading on l,hc~ other. Whether there 
are other synta('tic criteria that further (lisam 
1)iguate 1)etween the thre.e readiugs Mso del)ends 
on tire structnral description assigned to th(' focus 
l)article use. There are suggestions of analysing 
focus adverbs as syntactic co-coustituents of their 
feet, and there, are suggestions for analysing them 
only as co-(:onstitu(mts o\[' the verb or its pro- 
jections (see (Bay85), (Bay88), (,lac89), (.lac84), 
(K5,,91) among others for a discussion). We have 
nothing sf)eciiic, to say about this here. We just 
note that, under the conuno,, assuml)tion that the 
vorfeld in (h'.rmau introduces at most one con- 
stituent and under the ensuing assuml)tion that 
\['ocus adverbs modify tht'ir loci, in sente.nc(;s like 
(3), crsl must 1)e interpret('d as focus adverl). 
(3) Iqrst den Ib'ief g'ab l'ctcr Maria. 
(l>eter only h~tnded the. letter to M~trim) 
As concerns the I,'S-reading, we add that, provided 
it is syntactically licensed, this reading is strongly 
supported if the sentence $2 following the (r,sl- 
sentence S1 contains an adverb of temporal succe- 
sio,t (like d,L,zn/thcn, dauach,/after thi.s etc.) that 
rood|lies a verbal projection of similar tense and 
in lbl'mation strut:tm:('~ as th(: crsl-argument. This 
latter heuristics is expensive howew~r, in that it 
checks extra-sententiM information. The follow- 
ing criteria avoid this. 
a.2 Th(,. D)cus-Backgr()und Criterion 
If the asstLlnption of s(;ction 2.2 is true thai, it\] the 
I!;I)A-sccnario the |)ackground event type is tested 
for specitic reMizations, it is natural to think of 
this scenario to be reasonably con(:eptualizt~d ouly 
if the, I)ackg;rouud eveut type merits testing. This 
is not the case it ~ there is no background event I;ype 
at all (i.e. Ae.IiA6'(I)ItS~) is the most geucral 
event predicate). There, is no backgrouud ew'nt 
type if the entire argument of erst - the verb or a 
verbM projection - is in focus, or, though not in 
focus, the verb does not curry enough substance in 
order to provide an event type: this is the case for 
the COl)ula without; the t)redicativc ('olrtl)\]elneHl.. 
Instead of technically working out l,he criterion, 
we content ourselves with sore(; motivating exam- 
pies whi('h are critical in this resl)ect. It, order to 
avoid interfering effects from the syntacti(" struc- 
ture that might eoutplicate matters with regard t.() 
deterntine the SCOl)e of crsl, we only list examples 
with verb tinM l)osition. In (4), the parentheses 
mark the argument of crst, the brackets an notated 
by F the focus elen,ent fi'om which the semantic 
focus constituent is deveh)pe(l. As an example, 
(4.a) and (4.b) present their resulting structur(:d 
event types. We t)lnit this r~ther cmlonicM struc- 
turing for the other examples. 
(4) I'etra war {ibc.rrascht, wcil 
(l'etr~t w*ts surpris(M Imca, use) 
a) (,,'st (i,, wa,') 
b) l',,tcr cr, t (i,, \[St,,ttg'art\]r a,,w('.s~,,,,l wa, 9 
t (being in fituttg~u:t/ I)eing pl'eSeltt itl ,qtHttga.rt) 
,:) os erst  a,9 *O':Pa) 
d) cs erst (it, \[,~tuttgart\],,, 12.00 wa,9 (El'A) 
(being 12 Cclock / bt',ing \[2 o'do(:k iu Stuttgart) 
c) l'etcr c,'st (\[arbeitctc\],., ) *(l';l'a) 
t) Peter crst (i,, \[St,,ttgart\]r arbeitct@ (lqPa) 
(working in Stuttgart) 
The indications (El'A)and *(El'A)mean that o,e 
cmJ conceive contexts that allow tbr I')I)A or that 
(me can not. Without fllrther comment, we thiuk 
that the criterion is cent|trot'x\[ by the data. 
l"ocns- 13ackgr(mnd-cr\] terion : 
The F,l~A-reading is ac('(;l~tM)h~ only if the SCOl)C 
of c, rst is structured into focus and bac.kgrouml in 
such a way that the background is a specitic event 
type. 
3.3 The T(:nqmral Location Crlte.ri(m 
'l'he R-reading i)resul)poses a sequence of events 
(concel)tuMized as a plan or an expectation about 
the ongoing of the world) and it assumes that, 
fl'om the t)erspective of the contextual l)ersl)eCl, ive 
time, a part of the sequence is reMized, at.cording 
to the ordering of the plan or expectation. The 
rel)orted event refers to the event of the presul)l)O 
sition line that marks the boundary l)(%wcen the 
instantiated nnd tire nol>instantiated event cou- 
cepts, and it; does this in right the satlle way as 
delinite descriptions do with resl)ect to their ml- 
tecedents. Now, if this is true and it' the ev(ml; 
descrit)tion contains a teml,orM location hi t.hc 
focus, this information cannot be used att,'ibu 
tively, because it contributes to the antecedenL de 
337 
scription and to the distinction of this antecedent 
fi'om its alternatives. Because of this setting, it 
acts as a restriction on the referring expression 
that helps to pick up the right antecedent from 
the presupposition line. This means that, against 
the background of the presupposition, this in- 
formation is not new. We add that nothing of 
the ersgargument is new information against this 
background. New is the information about the 
progress of the instantiation of the presupposed 
event concepts. But then, stating that an event 
of the corresponding antecedent type indeed was 
realized (the assertional impact of the R-reading), 
and stating that it occurred at a time as was ex- 
pected (consequence of the specific description of 
'antecedent' and 'anaphor'), and simultaneously 
insinuating that it could have been realized earlier 
(presuppositional structure of the R-reading sup- 
ported by intplicature) results in a contradiction. 
This, to our opinion, seems to be the explanation 
of why the I{-reading is not possible in case the 
description in the scope of erst comes with a tem- 
poral location in the foens. Compare the following 
examl>les to this end. 
(5) Petra war iiberrascht, well 
a) Peter erst On \[Stuttgart\]\]+, war) (R) 
b) Peter erst (urn \[12.00\]v in Stuttgart war)*(R) 
(5) confirms what we have said so far. Note, by the 
way, that the features of the Focus-Backgronnd- 
criterion are not characteristic of the R-reading. 
What about temporal locations in the back- 
ground part? At first glance, it seems that what 
we have said above applies to this case also. How= 
ever, there are some (relatively marginal) cases 
that possibly contradict to this assumption. The 
following example (6) is an instance of this: 
(6) ... well Peter erst (in \[Mfinchen\]v um 3 
ein Bier getrunken hat) 
(be¢:anse only in Munich, Peter drank a 
beer at tt~ree (so fa~)) 
In a context that; continues this information, for 
instance, by und noch nicht in Kb'ln (am ndchsten 
7'ag) um 3 / and not *,let in (Jolo.qne (lhe next day) 
al 3 the R-reading seems acceptable. Because of 
the granularity of the presuppositional event se- 
quence that develops from the presupposition con- 
strnction in such cases - in (6), the iteration must 
satisfy to a one-day-rhythm at least- the temporal 
adjunct cannot truly act, as a restrictive referen- 
tial constraint, and becanse of what we have said 
above about novelty, it is not the best attributive 
information also. This may explain why the ex- 
ample is felt to be a bit odd. The decisive feature, 
however, why the above argumentation for the in- 
compatibility of the R-reading and the presence 
of temporal localizations does not go through, is 
the fact that the background temporal localization 
does not uniquely \[ix the occurrence time of the 
event with regard to the time fralne of the presup- 
posed plan or expectation. This, of course, is so, 
only if the localizing predicate allows for multiple 
(periodic) instantiation. (Times of day allow for 
this, also adjuncts like after lunch etc.). 
We stress that what we have said relates to 
temporal adjuncts in tile scope of erst. The 
following (7.a) allows for the R-reading, because 
the most natural analysis gives wide scope to the 
temporal adjunct, i.e. the sentence is analysed like 
(7.h), where clearly, the adjunct serves to localize, 
the temporal perspective. 
(7) a) Erst in Miinchen war er gestern. 
b) well er g estern erst in Miin&en war. 
(yesterday being in Munich) 
We have considered only temporal adjuncts so 
far. The alternative temporal localization that oc- 
curs in the scope of erst is the construction 'COl>= 
ula + predicative temporal expression', which ac- 
cepts the R-reading. Examples are (4.c) and (4.d). 
We skip the complete explanatory argument here 
and just say that (grosso mode) the function of 
the copula construction is to synchronize calendar 
knowledge (also information about different calen- 
dars: R-reading of (4.(1)) with the actnM avail- 
able perspective times, whereas the function of 
the temporal adjunct is to relate the descibed 
event to some predefined time. We take it for 
granted, that this difference is the reason why the 
decisive conflict that we mentioned fllrther abow'~ 
only arises if the temporM location is introduced 
by modification, i.e. in case it is introduced by an 
adjunct. 
We retain the following criterion: 
Temporal-Location-criterion: 
The R-reading is acceptable only if the focus con- 
stituent of the scope of erst does not contribute a 
temporal localization (by modification of a basic 
event type). In addition, the scope must not con- 
tain an adjunct - focused or not - that is a uniquely 
referring temporal location (like yesterday). 
3.4 The Entaihnent Criterion 
Compare the following examples: 
(8) ... well 
a) Toml, a ~id~ er~t (a, aer \[St,'eif~,lml,~tte IF 
disqualigizierte). 
(T. disqualified himself at the S.) *(R),(EPA) 
I,) erst (\[drei \]F Unterschriften genfigten). 
(three signatures were sntticient.) *(R),(gI°A). 
e) Peter erst ( \[vier \]r Angestellte kannte). 
(P. knew font' employees.) (I{),*(EeA) 
(8.a) cannot have the R-reading. Why? The in- 
telligent construction of the presuppositional se- 
quence of events for the H.-reading outputs a num- 
ber of disqualification events that are located al; 
particular places of the Ilahnenkamm downhill 
race in Kitzbfihel. What is specific with this se- 
quence is the fact that the postconditions of any 
of these events are snch that the preconditions of 
the successive events never can hold. Therefore 
such a sequence can never be a reasonable plan or 
338 
a reasonable: expectation el)out a <townhil\] race. 
I.e. a constitutive element of the R-reading cannot 
h(; constructed in this ease. In (8.1)) the descrip-- 
tions of subsequent events (states in this case) of 
the presuppositional line are more gcnerM predi- 
cates than the description of the predecessors, i.e, 
each such sequence collapses in its tirst element in 
essen<;e. This cannot truly he called a sequence. 
With (8.c), we encounter, so to speak, the sym- 
metric picture with regard to the El)A-reading: 
t(no'wing n employees entails the previous|y tested 
knowin( t n-\[ cmployccs. 'l'he exl)('<'tation of some 
l)rOl)osition p t;o be true in a specific situation s,~ 
cannot t>e falsilied, in case the wdidity of a parti<:- 
ular prol)osition (1 in the subsequent test situation 
s~+l confirms the wdidity of p. 
We retain the following criterion: 
I';ntailment-criterion : 
For the I{,:reading to l)e acceptable, first, the l)ost: 
conditions of each event of the t>resupt)ositional 
line must he compatible with the preconditions 
of the successor and s<;cond, (at least for homo-. 
geneous deseril>tions) the description of an event 
must not subsmne the des<'ril)tion of the following 
(:Velits. 
\[,'or the I,;I)A-reading to l>e acceptable, (at least 
for homogeneous descriptions) the event descril>- 
tion tested at a situation must not subsume the 
previously tested evt'Att description. 
There are refinements of this criterion that we 
tnust omit here. 
4 Final Remarks 
'l'he four criteria of the last sectiou can be used 
in order to exclude readings <)f crsl. I1; is <)nly the 
last (entailment) <;ritel:ion that necessitates some 
economic semantic infcrencing, The <)t, hcrs c<)rre- 
spon<l more<>v less to +~structural lookup. Using 
I;he convincing structural interd<:t>endencies thai; 
(Liib89) shows for a subset of the German focus 
adverbs containing crsl, the generalization of tire 
approach suggested here to other ambiguous ad: 
verbs seems very promising. 
References 
J. Bayer. Adiazenz und Kettenbildmlg. Be- 
merkungen zur Syntax der (leui;s(:hen (fred= 
partikch,, 1985. (ms.)Max-lqanck-lnstitut f/Jr 
l)sycholinguistik. 
J. Bayer. Rightward movement and the syn- 
tax of <luantilicational particles in German. In 
V. I{oscn, editor, l'apcrs fl'om the 7'cnth Scan- 
dinavian Conference of Linguistics 1, Bergen, 
Norway, 1988. 
l)el;er I. Blok and Kurt Eber|e. What is the alter- 
native. In Peter Bosch and Rob wm der San(It, 
e<litors, \["ocus and Natural Language Process: 
ing, Proceedings of the Interdisciplinary Con: 
J'crence in Cclcbration of the lOth Anniversary 
of the Journal of Semantics. IBM, \[leidelberg, 
1995. 
J. Bos, E.Mastenbroek, S.M(.Gias~an, S.Millies, 
and M.Pinkal. The Verbmobi\[ semantic formal- 
ism. VM report 6, Universit'~t des Saarhmdes, 
1994. 
Kurt Eberle. On rel)resenting the temporal stru(> 
ture of a natm'a\[ language text. In Proceedings 
of Coling92, Nant<;s, 1992. 
Kurt Eberle. Zu einer Setnantik \[iir I)ialogverste- 
hen und Ubersetzung. Verbmobil-l(,el)ort, 8;I, 
U niversitaet Stuttgart, 1995. 
Jakob Hoel>ehnan and Christian tl,ohrer. I{e- 
marks on noch and cohen in (lerman. Syntaa: 
and Semantics, 14:10f{ 126, 1981. 
J. Jacohs. '\['he syntax of bound focus in German. 
(;roninqcr Arbciten zur Gcrmanistischcn Lin- 
guislik, 25:172 200, 1984. 
.1. Jacobs. l"okus und Skalcn. Zur Syntax 
und Scmanlik dcr (D'adparlikeln im Deulschcn. 
Niemeyer, Tiibingen, 1989. 
Ilans Kaml). A theory el' truth and semantic rel)- 
resentation. In J.A.G. (Iroenendijk, T.M.V. 
Janssen, and M.ILJ. Stokhof, (~(litors, l"ormal 
Methods in the Study of Languagc. Mathemat- 
ical (~entre Tract, Amsterdam, 1981. 
Hans Kemp. Some elements of a 1)\[~SP-based the 
ory of the representation of mental states and 
verl>al comnmnication. (ms.), I MS, Universit:~t 
Stuttgart, 1995. 
Ekkehard K6nig. A sen,antic analysis of (h~rnmn 
'erst'. In Raincr llihlerle, U. Egli, and Aruim 
wm Stechow, editors, Semantics firm DiJ.fcr- 
cnl Points of Vie'w, pages 1d8 160. Springer, 
Berlin,\[\[eidelberg,New York(lbkyo, 1979. 
Ekkehard K6nig. 7'hc Meaning of I"ocus l'arti- 
clcs: A Comparative Perspective. I~outle(lge, 
London, New York, 1991. 
Stephen (I. I,evinson. Pragmalics. Canfl>ri(lge 
(hfiversity Press, (;ambridge, New York, 1,()83. 
Sebastian 1,5bner. Schoil - erst-noch: An inte- 
grated analysis. In Werncr Abraham, editor, 
Discourse Parliclcs, pages 85 140. Niemeyer, 
'l'iibingen, 1989. 
Mats E. I/=oot, h. Association with t,'ocus. Phi) 
thesis, University of Massachusetts at Amherst, 
1985. 
Mats E. Rooth. A theory of focus interpreta- 
tion. Nalural Language ,qcmanlics, 1(1):75 
117, 1992. 
Roh van der Sandt. I)resupposition projection 
its anaphora resolution. Journal of Sam.antics, 
9(4), 1992. 
339 
