In this paper, we cannot go into detail with tests that partition the meaning of a sentence into pre­suppositions, assertions proper and implicatures the recipient is allowed to draw from the sentence (cf. (Lcv83) for an overview of the notions used, the tests associated with them and the problems connected to them). We directly come up with the DRSs that, to our opinion, represent the im­pact of the different readings. We begin with the FS-reading of (1):

~pëTcFnow c f TP(t)

Here, TP stands for the actual (past) temporal perspective that holds for the given utter­ance/text situation. With respect to the represen­tation and resolution of presuppositions in DRT, we relate to (vdS92). Wo slightly deviate from the approach suggested there, however, in that we use a notation for the presupposition triggers that is akin to the suggestion of (BES+94): we use so-called a-conditions which describe the presup­posed objects and their characterization. In our framework «-conditions subsume différent types of projection problems. The type is characterized by the «-index (prcsp stands for 'classical' presup­positions, dc.f for definite descriptions, ri for ref­erence time, re for reference event etc., compare (I0be95) for an overview). This index triggers the projection routine that is specific to the respec­tive resolution problem. The first argument of the condition schema highlights the distinguished discourse referent (DRF) of the structure to be projected. The representation, thus, requires a definite description style resolution for the x that is characterized as a fourth, lucky number (with respect to the given setting, x together with its description will be accomodated at the main level DRS), and if presupposes a reference event c' such that the sentence (went c. is the first event of a possible elaboration sequence of c' (cf. (Ebe92) for the computation and storing of discourse rela-

' As regards definite descriptions, the distinguished 1)R.F is the DRF of the head noun; in any case it is the DRF the «-information is centered around.

Modeltheoretically the relation between the presup-positioual part and the assertional part of a DRS can be seen as a function from information states into in­formation states, see (Kam95).

tions like elaboration).

In the EPA-reading, erst is used as a focus adverb, i.e. it structures its argument into focus and back­ground. Tn the following representation of (1), we consider the case where the numeral is focused only, not the MP containing the numeral or the entire event description in the scope of erst

Hesides the assertion that Peter points to the fourth lucky number at the temporal perspective t, the representation presupposes a sum 10 con­sisting of a sequence of events ei - ej, that are related via a non further specified relation R to predicates Pi - Pfc. These P,- are required to be alternatives of the event description in the scope of erst (which is called A e. DRSC). In this paper, we cannot go into detail with the computation of the alternatives of event descriptions. l\ G AL'Id us clwhere ALTy;,/f.s6 = {Ae.DRSe[,?0(;cl(dhs,,)\q] \ Q G ALT(FOCc,(DR,Se))}, sketches the claim that the P j are event descriptions that develop from Ae.DRSe by exchanging the focused element by an alternative (Compare (Roo85), (Roo92) for the underlying focus theory and (RE95) for an algorithm that calculates the concrete alterna­tives). As mentioned, the focused element, which is marked by the underline, is assumed to be the numeral adjective. This choice trig­gers the structuring of the Ps into the back­ground event type Xc.HAC(DRSe) (which is A c. rLiKo„I^,TruM,~x7D and the focused se­mantic constituent Xx.FO('(DRSe)[4\„] (which is peter now e t el c.'l c'A vA TP(t)

b = r,'=k e,.

e;' C e.

As for the EPA-reading, we consider the case where the numeral is focussed only:

Jî-esp (s

MODE=PLAN X=petcrU, i

it.

e:'/eigen_aui(peter,x) 4te_glückszahl(x)

s meets e'i ------optional or probably obligatory

In contrast to the EPA-reading, we assume that, in the R-reading, the predicates P, that we ob­tain from the information structure of the erst­argument are not related to a sequence of opportu­nities for doing something, but describe events e, of an expectation about the ongoing of the world ë or a plan ë. The context (1) doesn't provide fur­ther 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 in­formation about the nature of the attitude MOD. The conditions that are introduced below the dot­ted line exemplify possible resolutions. According to this, Peter and the speaker/writer (i for the distinguished DRF for the self) share the attitude of having a plan for realizing r. Such resolutions may be available by an inference component that operates over richer contextual knowledge. Again, the order of the P, and the corresponding e; is in­ferred from the implicit order of the alternatives of the focused element. From this setting and the assertion of an occurrence of Peter pointing to the fourth lucky number at the temporal per­spective t, the representation entails realizations of those events of the presupposition line that precede the counterpart of e in the presupposed sequence. As a further (possibly optional) con­straint, the R-reading introduces the implicaturc that a non further specified person or group £ expected for the perspective time t that the planned or expected sequence of events should be realized to a greater degree. Without further information about the identity of it is difficult to say some­thing more precise about the temporal location of the ^-expectation than that an instance s' of the corresponding atfitudinal state holds at some time before the actual now.

In Gorman focus adverbs cannot be topicalized as such, i.e. they cannot occur in the vorfeld po­sition without an accompanying constituent (cf. (Kön9l)). If the sentence shows such a topicaliza-tion of erst - which is marked by the inversion of the basic Subj-Vf in-order - erst can oidy be used as a time adverb, i.e. its meaning can only be the FS-reading, as exemplified by (2): (2) Erst gab Peter Marin den Uriel'.

(First, Peter handed the letter to Maria.)

Thus, this type of fopicalization disambiguates be­tween the FS-reading on the one hand and the EPA- and R-reading on the other. Whether there; are other syntactic criteria that further disam­biguate between the three readings also depends on the structural description assigned to the focus particle use. There are suggestions of analysing focus adverbs as syntactic co-constituents of their foci, and there are suggestions for analysing them only as co-constituents of the verb or its pro­jections (see (Hay85), (Hay88), (,lac89), (Jac.84), (KÖH91) among others for a discussion). We have nothing .spécifie, to say about this here. We just note that, under the common assumption that the vorfeld in Gorman introduces at most one con­stituent and under the ensuing assumption that focus adverbs modify their foci, in sentences like (3), erst must be interpreted as focus adverb. (,'i) Nrst eleu Uriel' gab Peter Maria.

(Peter only handed the letter to Maria.)

As concerns the FS-reading, we add that, provided it is syntactically licensed, this reading is strongly supported if the sentence S2 following the cre­sentence SI contains an adverb of temporal succe-siou (like dann/then, danach/'after this etc.) that modifies a verbal projection of similar tense and information structure as the ens/-argument. This latter heuristics is expensive however, in that it checks extra-sentential information. The follow­ing criteria avoid this.

If the assumption of section 2.2 is true that in the EPA-scenario the background event type is tested for specific realizations, it is natural to think of this scenario to be reasonably conceptualized ordy if the background event type merits testing. This is not the case if there is no background event type at all (i.e. \e.BAC(])ItSe) is the most general event predicate). There is no background event typo if the entire argument of erst - the verb or a verbal projection - is in focus, or, though not in focus, the verb does not carry enough substance in order to provide an (went type: this is the case; for the copula without the predicative complement. Instead of technically working out the criterion, we content ourselves with some motivating exam­ples which are critical in this respect. In order to avoid interfering effects from the syntactic struc­ture that might complicate matters with regard to determine the scope of erst, we only list examples with verb final position. In (4), the parentheses mark the argument of erst, the brackets annotated by F the focus element from which the semantic focus constituent is developed. As an example, (4.a) and (4.b) present their resulting structured event types. We omit this rather canonical struc­turing for the other examples. (4) Petra war überrascht, weil

(Petra was surprised because) a) Peter erst (in [Stuttgart]war)

a Stuttgart i :[ i n ( p c t e r, s t u 11 g a i^"]

b) Peter erst (in [Stuttgart]],- anwesend war)

s:f Aliwcsendfpetcr) ~| in(s,Ätuttgai t )

(being in Stuttgart/ being present in Stuttgart)

The indications (EPA) and *(EPA) mean that one can conceive contexts that, allow for EPA or that one can not. Without further comment, we think that the criterion is confirmed by the data.

I'ocus-Background-criterion: The ÈPA-reading is acceptable only if the scope of erst is structured into focus and background in such a way that the background is a specific event type.

The R-reading presupposes a sequence of events (conceptualized as a plan or an expectation about the ongoing of the world) and it assumes that, from the perspective of the contextual perspective time, a part of the sequence is realized, according to the ordering of the plan or expectation. The reported event refers to the event of the presuppo­sition line that marks the boundary between the instantiated and the non-instantiated event con cepts, and it does this in right the same way as definite descriptions do with respect to their an ­tecedents. Now, if this is true and if the event description contains a temporal location m the focus, this information cannot be used attribu­tively, because if contributes to the antecedent description and to the distinction of this antecedent from 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 in­formation is not new. We add that nothing of the er.s/-argument is new information against this background. New is the information about the progress of the instantiation of the presupposed event concepts. Rut 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 ex­pected (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 sup­ported by implicature) results in a contradiction. This, to our opinion, seems to be the explanation of why the R-reading is not possible in case the description in the scope of erst comes with a tem­poral location in the focus. Compare the following examples to this end.

(5) Petra war überrascht, weil (5) confirms what we have said so far. Note, by the way, that the features of the Focus-Background-criterion arc not characteristic of the R-reading.

What about temporal locations in the back­ground part? At first glance, it seems that what we have said above applies to this case also. How­ever, there are some (relatively marginal) cases that possibly contradict to this assumption. The following example (6) is an instance of this:

(because only in Munich, Peter drank a beer at three (so far))

In a context that continues this information, for instance, by und noch nicht in Köln (am nächsten lag) um 3 / and not yet in Cologne (the next day) at 3 the R-reading seems acceptable. Because of the granularity of the presuppositional event se­quence that develops from the presupposition con­struction 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 referen­tial constraint, and because of what we have said above about novelty, it is not the best attributive information also. This may explain why the ex­ample is felt to be a bit odd. The decisive feature, however, why the above argumentation for the in­compatibility 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 fix the occurrence time of the event with regard to the time frame of the presup­posed 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 the 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.b), where clearly, the adjunct serves to localize the temporal perspective.

(7) a) Erst in München war er gestern.

b) weil er gestern erst in München war. (yesterday being in Munich)

We have considered only temporal adjuncts so far. The alternative temporal localization that oc­curs in the scope of erst is the construction 'cop­ula + predicative temporal expression', which ac­cepts the R-reading. Examples are (4.c) and (4.d). We skip the complete explanatory argument here and just say that (grosso modo) the function of the copula construction is to synchronize calendar knowledge (also information about different calen­dars: R-reading of (4.d)) with the actual avail­able perspective times, whereas the function of the temporal adjunct is to relate the described event to some predefined time. We take it for granted, that this difference is the reason why the decisive conflict that we mentioned further above only arises if the temporal 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 con­stituent of the scope of erst does not contribute a temporal localization (by modification of a basic event type). In addition, the scope must not con­tain an adjunct - focused or not - that is a uniquely referring temporal location (like yesterday).

Compare the following examples:

a) Tomba sich erst (an (1er [Streitalmhütte ]p disqualifizierte).

b) erst ( [drei ]f Unterschriften genügten). (three signatures were sufficient.) *(R),(EPA).

c) Peter erst ( [vier ]p Angestellte kannte).

(8.a) cannot have the R-reading. Why? The in­telligent construction of the presuppositional se­quence of events for the R-reading outputs a num­ber of disqualification events that are located at particular places of the Ilahnenkamm downhill race in Kitzbiihel. What is specific with this se­quence is the fact that the postconditions df any of these events are such that the preconditions of the successive events never can hold. Therefore such a sequence can never be a reasonable plan or a reasonable expectation about a downhill race. I.e. a constitutive element of the It-reading cannot, be constructed in this case. In (8.b) the descrip­tions of subsequent events (states in this case) of the presuppositional line are more general predi­cates than the description of the predecessors, i.e. each such sequence collapses in its first element in essence. This cannot tridy be called a sequence. With (8.c), we encounter, so to speak, the sym­metric picture with regard to the EPA-reading: Knowing n employees entails the previously tested knowing n-1 employees. The expectation of some proposition p to be true in a specific situation sncannot be falsified, in case the validity of a partic­ular proposition q in the subsequent test situation sn+| confirms the validity of p.

We retain the following criterion: Kntailment-criterion:

Kor the It-reading to be acceptable, first, the post­conditions of each event of the presuppositional line must be compatible with the preconditions of the successor and second, (at least for homo­geneous descriptions) the description of an event must not subsume the description of the following events.

For the EPA-reading to be acceptable, (at least for homogeneous descriptions) the; event descrip­tion tested at a situation must not subsume the previously tested event description, 'there are refinements of this criterion that we must omit here.