<?xml version="1.0" standalone="yes"?>
<Paper uid="C96-2150">
  <Title>A - ~-fll,,e (l:~ &lt; t</Title>
  <Section position="2" start_page="0" end_page="889" type="metho">
    <SectionTitle>
3 Ilcnrictta was crossing th, c wad, when sh, e was
</SectionTitle>
    <Paragraph position="0"> hit by a bus.</Paragraph>
    <Paragraph position="1"> I have. no argument with his analysis of (1) and (3). What concerns me here is the apparent change in the contribution of the present participle marker in (2). In (2) we have a (conceptually) instantaneous (',vent, namely a hiccup. Since hiccups are generally thought of as taking no time, it; is not possible~ to be in the middle of a single hiccup and hence we are solnehow driven to conelude that Harry was in the middle of a series of hiccut)s. A similar problem arises with: 4 Allan is living in Bray.</Paragraph>
    <Paragraph position="2"> Here we have a homogeneous state where there is no result to be achieved no interesting state of affairs that arises as a consequence of reaching the end point. As such the present participle cannot be taken as an indication that the cuhnination of my living in Bray has not been reached, since there is no such cuhnination to reach. In this ease the present l)artieiple somehow transforms itself to an indicator of temporariness, so that you {:an gel; exchanges like the following: Allan's living in Bray I thought he lived in Buxton .... Yes, but he's on a visit to Ireland at the inoInent (Smith, 19911) deals with this phenomenon by appealing to a notion of &amp;quot;derived interpretations&amp;quot;, though with very little discussion of how the derivations take place. (MoSns and Steedman, 1988) deal with it by invoking a process of coercion which changes the meaning of the ast)ect markers as required by the properties of the verb to which they are attaehe(t. Much of what I want to say below follows their anMysis, with one m~&gt; jor difference.. Coercion changes tile meaning of the aspectual marker in response to the semmltic properties of the marked verb. But if items are allowed to change their meanings as a consequence of the semantic properties of other items then the principle of compositionality that &amp;quot;the mealfing of the whole is made out of simple l (:onlbinations of the meanings of the parts&amp;quot; t If we alh)w arbitrary rules of combination then we (:an include rules which make arbitrary changes  becomes rather ineffectual. We are, alter all, led to describe a word as being a homonym in exactly those cases where the meaning of what appears to be a single lexical item depends on the semantic properties of the words it is being combined with. In 5 IIe keeps his money tied up in the bank.</Paragraph>
    <Paragraph position="3"> 6 He keeps his boat tied up by the bank.</Paragraph>
    <Paragraph position="4"> the fact; that; the interpretation of bank depends on the semantic properties of money and boat is what; persuades us that the form bank is being used to realise two diffe.rent lexical items. We do not, however, want to describe the present participle marker as being ambiguous, with different interpretations which depend on the semantic context in which it occurs, unless we are absolutely forced to. The analysis in this paper attempts to show that the effects described by Mogns and Steedman (:an be achieved without any meaning-changing operations or unwanted ambiguities.</Paragraph>
    <Paragraph position="5"> The basic tool that I will use is the observation  that A U F ~ A can hold when neither A ~ A nor \]7' ~ A does, and ill particular that if P and F' are different then A U F ~ A and A U P' ~ A' can hold where A and A' are different, or even incom null patible. If we back ut) the labels representing lexical and predexical items with appropriate sets of meaning postulates then we may well find that different things can be inferred from a single itein in different semantic contexts without being forced to conclude that those items themselves mean different things. In this way the meanings of words will cooperate to convey more complex inessages than each can carry alone.</Paragraph>
    <Paragraph position="6">  This 2 a is all very well as far as it goes, but unless the consequences of saying that something is an to the meanings of the parts. If this happens then the principle has no force. &amp;quot;Simple&amp;quot; combinations are usually taken to be things like function application and set union or intersection.</Paragraph>
    <Paragraph position="7"> ~The analyses in this paper require a combination of truth functional operators and l-abstraction. I use the notation .\[x, P\] rather than txP to emphasise that I am relying (Turner, 1987)'s treatment of abstraction, where you can safely combine the two, rather than classical l-calculus where you run the risk of paradoxes if you combine them.</Paragraph>
    <Paragraph position="8"> a~X::{P}(Q) says that Q is true of the X which satisfies P. As such it performs much the same role event of type &amp;quot;hiccup&amp;quot;, or that someone is the patient of an event, and so on, are spelt, out ill detail then it is not possible to perform any non-trivial inDrences on tile basis of this interpretation (and hence not; possible to argue about whether or not it is right, so that the clainl that a fragment of natural language should be paraphrased ill a parl;icular way becomes vacuous). You might, for instance, disagree with my decision to label the sleeper as the patient of the event. Unless I spell out what this label commits me to, there is no way for me to defend it; or for you to attack it,. Simply appealing to our everyday intert)retation of tile term will not do.</Paragraph>
    <Paragraph position="9"> We therefore need to develop a collection of meanin.q postulates (MPs) to specify tile conne(&gt; dons between the terlns that will appear ill our interpre.tations. This is perhaps an obvious point, but apart from a few honourable exceptions (the attempt in (Dowty, 1988) to specify the conse~ quences of assigning all item to a thematic role is a notable case) it is too often neglected. The. central claim of the current paper is that; tile interactions between Ineaning postulates can produce sut)tle effects which you may miss if you simply label items as belonging to (:lasses or as being in relationships with one another and leave it at that if you simply say, for instance, that some event is progressive, without spelling out the MPs for progressive.</Paragraph>
  </Section>
  <Section position="3" start_page="889" end_page="893" type="metho">
    <SectionTitle>
3 Aktionsart and Aspect Revisited
</SectionTitle>
    <Paragraph position="0"> I will now look in some detail at aspect and aktionsart. For the remainder of this section I will say that; the relationship specified by an aspect marker holds between a tilne and all event type, where an event type is nothing more. t, hall all abstraction over a proposition about ewmts. What we need are tile meaning postulates that spell out the consequences of saying that a time and an event; type are in the relationship specified by some aspect marker. I make the following assumptions: * The aspect; of the core verb specifies a relationship between all instant and all event type. The details of these relationships are spelt; out via MPs.</Paragraph>
    <Paragraph position="1"> * The tense of the core verb, together with ally auxiliaries, specify a relationship between the present time now, all anaphoric reference time re\]', and the time mentioneA in this relationship. Nothing much in tile analysis below depends on the particular properties of the.</Paragraph>
    <Paragraph position="2"> time lille. The only assumption that i will make any use of is t;hat there are intervals and instants.</Paragraph>
    <Paragraph position="3"> * The MPs for the core verb specify the temporal properties of the event type. If the verb as (Barwise and Perry, 1983)'s notion of anchoring.  shares temporal 1)roI)erties with a range of other verbs, then dwse are gathered together as MPs for the (:lass as a whol(;, which is r('ferred to as an aktionsart.</Paragraph>
    <Paragraph position="4"> The first MP we, will consid('r deals with the pro- null This says that the relationshi I) prog hol(ls t)(!tween an instant /; and an event l;yi)(; P if there is a sel, }\]\] of events of t;h(; al)t)roprial;e tyl)c , at; least one of whi(:h starts })efore the insbmt and at least, one of which does not end 1)efor(~ it;.</Paragraph>
    <Paragraph position="5"> Meaning 1)osl;ulates are 1101; ileCe.ssary alia Slltli('.lent; (:on(tit;ions. They are ('.onstellations of \['acts which, serve to sl;rllCtlll'('. \[;he (;on(:et)imal st)a(:c. They do not exhaust thai; sl)a(:e , an(l they do not ne,(-essarily botl;om ()lit in sense-dal;a l)ased primitives (Carnap, \]936; Quine,, 1.960). The mosl;, and least, you ca.ii say about them is that Lhey help d(;lin(',t~te, a set of (:oncep\[;s a.n(l relatiolls \[)(&gt; t;ween con(:el)ts which can 1)e use(\] to l)oint out I;he relations that \]mid among words, and at eerta.in l)oinl;s between wor(ls an(1 exl)(',ri(m(:es. In ((buse, \] 986)'s t)hrase, l;hey e.xl)r('ss s EMANTI(; 'I'I{,A ITS state.merits about so'me of the things that tyt)Jcally follow from asserting tltat some r(;lationshi I) hohts. There is th(;reIore n() irres()lwd)le (:lash 1)etween MP prog and Asher's MPs whi(:h des(:ril)e t;he conditions under whi(;h you wouhl expect a relic event des(:ribed using the progressive Lo t)roceed to its cuhnination, and i would expect, to supplemenl; what I have to say in this 1)ap(;r wid~ his defaull; 1;real;l\[tellt of t, his ol;her issue. In t)articular, it should be. note(l l;\[lat; MI)-prog entails l;he existence of a starl; l)oint for the rei)orte(l a(;l;ioil t)llt 1lOt that of an end 1)oint.</Paragraph>
    <Paragraph position="6"> The de(:ision to talk in terms of sets of ev(;nl;s provides exl;ra tl(;xil)ilil;y, ill the same way that the de.(:ision l;o deal with NPs in terms of sets of indivi(hmls supt)orts th;xible tr('.atmenI;s of plurals and of otherwise awkward l)henoinena such as generics and bare plurals(l{amsw, 1992). We (:an always (:onstrain a set of eve.nts to t)e a singleton if we need to, so certainly nothing is, lost by talking about sets rather than individuals.</Paragraph>
    <Paragraph position="7"> SUl)l)ose w(; have the following MPs for aktionsarts and thematic roles:  fl'ht;st, arc' all straighl;forward enough. I,\]vents have sta.r\[; and end points. Telic events have results, which are charat:terised by propositions which become true at the (;ii(t point of the. ev(;nl;. A stal;e is charaeterised 1)y a prot)erty P l;hat hol(ls of the state's patient a: throughout some interwd i. A('.tions are ev(;nl;s with agents, where an agent is a being that inl;entionally causes the resull; of the eVellt \[;o \])CCOllle ti'ile,~ aild oIlly aliiHlaLe t)eiltgS ca.Ii intend to bring things about, l?atients are.just; animal, e, beings. 10,xtended events take time (th(;re ix some instant 1)etween their start an(l end points), instantaneous events do llOt (tiote that this may or not lneall that the. stm'l; anti end points of an instantaneous (;veltt are identi(:al, depending on whether we regar(l the time line as d(mse. As far as th(; (:urrent t)at)er is concerned this is a free ehoic('.). Exl;(;n(h;d leJi(: actions are. just extended evellts with results whi(:h become true at; their end t)oints and agents who inten(1 I;hose results 1;o })('~(:ore(; true.</Paragraph>
    <Paragraph position="8"> All we need to know at)out eat and hicc'u.p for the moment is that; cat denotes &amp;ii extended relic action and hiccup denotes an instantaneous evenl;:</Paragraph>
    <Paragraph position="10"> rather that we don't think about the time they take). There are many other MPs dealing with these verbs, since there is a great deal more to be said about them, but we do not need this extra detail here and hence we will omit it.</Paragraph>
    <Paragraph position="11"> Compare now the following analysis of 8 Ite is eating a peach.</Paragraph>
    <Paragraph position="13"> with the interpretation of (7) given earlier. MP prog says in each case that there must be an event whose start point is before now and an event which does not have an end point, before now. In the case of (8) this is compatible with the possibility of there being exactly one such event. Indeed, since only one peach is involved, the remainder of the MP for eat '(which would include the information that you can only eat something once) would presumably force this conclusion. It is furthermore compatible with the requirement that there should be an event whose start is before t and an event, whose, end is not before t, since eating events are extended - if they have end points then these are after their start points. In the case of (7) it is not, possible for there to be a single event, since the start and end points of a single hiccup are taken to occur with no intervening instant. We therefore find that (7) must denote a set of hiccups, simply by inspecting the MPs and without resorting to a process which turns hiccupping from an instantaneous act to a homogeneous sequence of acts. In both cases, the sentence reports a sequence of events. But in (8) there is nothing to say that this sequence has more than one member, and the fact that only on(; peach is involved suggests that it has exactly one member; whereas in (7) the temporal properties of the conceptually instantaneous act of hiccupping mean that there must be more than one such event.</Paragraph>
    <Paragraph position="14">  Why does this carry an overtone of &amp;quot;temporariness&amp;quot;? Assuming that live denotes a state, we need to look at the interactions between the MP for the progressive aspect and the MP for the aktionsart state. MP state says that the characteristic property of the state is true of its patient throughout some interval, but unlike MP relicevent it says nothing about the start and end points of that state, not even whether or not they exist. Of course in general we know that most states do have start and end points, but in many cases that is all we know about them. A speaker who is committed to the existence of a state, then, may not be concerned about the existence of the start or end point of that state they may not know when it started, they may not care whether it has ended, as far as they are concerned it may have been going on since the beginning of time and it; may continue to the emt of time. If, however, their report of tiffs state invokes the progressive aspect then they do become committed to knowing something about the start and end dates. If, for instance, we were considering Allan was living in Bray rather than Allan is living in Bray then we would assume that the speaker knew enough about the end of this state to place it before the reference point marked by the past tense of the auxiliary. Thus the use of the progressive aspect here commits the speaker to the existence of an end date for the state in a way in which commitment to the existence of tile state does not: it is this that gives (4) its feeling of being about a temporary state of atfairs 4.</Paragraph>
    <Paragraph position="15"> We now turn to the simple aspect. Consider the following pair of sentences:  9 Allan lives in Bray.</Paragraph>
    <Paragraph position="16"> 10 Mary eats a peach for her&amp;quot; lunch.</Paragraph>
    <Paragraph position="17"> (9) describes a simple homogeneous state of af- null fairs. The properties of the verb live and t, he sin&gt; ple aspect seem to collude in this case, and there is no need for anything like coercion. In (1.0), on the other hand, the.re does seem to be a problem. Eating denotes an activity with a definite final state, where what was eaten ends up inside the eater's stomach. Somehow (10) conveys the message that Mary habitually eats a peach for her lunch: note in particular that it is not the same peach or the same lunch every day! We therefore need a single MP for the simple aspect which enables us to conclude different things tbr the two cases. For (9), where the verb denotes a homogeneous state of affairs, the simple aspect supports tile conclusion that such a state of affairs does indeed hold. For (10), where the verb denotes an activity, the simple aspect supports the 4of. (Smith, 1991)'s observation that aspect provides a spotlight on some portion of the event.</Paragraph>
    <Paragraph position="18">  conclusion that such an activity hal)pens on a regular basis. The following meaning postulate says that the relationshit) simple holds between an instant t and all event type P if there is an interval I which contains t, and for any instant t' in I there is some event e of the apt)ropriate tyI)e which starts before t' and finishes after it.</Paragraph>
    <Paragraph position="20"> RemeInber that the MP for live sws nothing about the start and end points of the specilied state. Then there is nothing in MP sinlple to lead us to infe, r the existence of more than one such state, of affairs. There is also n(/dting to enable us to infer that there is no more ttmn one: I will return to this below.</Paragraph>
    <Paragraph position="21"> If, on the other hand, tile Mt ) for eat says that the start and end points of the action must be quite close together, then MP simple entails that there must be several such actions in the specified interval. Whi(:h is, after all, as much as you can infer fi'om the simple aspect itself. Note that the wide scope of the aspect oi)erator si'mple ineans that for (f0) we are conside, ring ewmt types in which there is a pea(:h, and a hmch, for every instance of the type. So unlike (8), where there was one peach and the event type we were considering dealt with eating that one peach, here there is nothing driving us to (:onclude that there is only one peach and hence that the set of events nmst be a single, ton.</Paragraph>
    <Paragraph position="22"> The combination of have and a t)ast-participle (I will call this the pmuq.~(;q'wt,; different people use different terminology \['or this) prese.nts similar problents. We can obtain the same kind of interpretation for such sent;en(:es, paraphrasing  Part of the diffcre, nce bet;ween this construction and the simple past arises from the explicit mention here of the R.l&lt;lq,;R.l:;N(llr; TIME (l{eiehenltach, 1956). Sentence, s like (1l) make reh~renee to s()nle anaphorically determined instant, and this gives them a slightly difli;rent flawmr tY=om sinqfle past sentences. But there is more to it than tha~.</Paragraph>
    <Paragraph position="23"> Consider the following examples: 12 He lived in Bray for five years.</Paragraph>
    <Paragraph position="24"> la lie has lived in Bray for five years.</Paragraph>
    <Paragraph position="25"> 14 He had lived in Bray for five years.</Paragraph>
    <Paragraph position="26"> The striking thing about these is that itt each of (13) and (:1.4) the obvious interl,retadon is tsha/; his period of living in Bray continued after the reference time, so that he probatfly live.d there for more than five years in total; whereas it is atl but impossilfle to read (112) as saying anything other than that his residence in Bray took no more or less than five years. This distinction becomes even clearer when we consider  It; seems that whereas you can have both a date and a duradon with the perfective, you Call have either but not both with the simple past. One way to account for this is to argue that the simple past deals with the end point of the event whereas the perfective deals with the end of some related interval. We have to be carefifl here. The MI' for the simph', asliect given above is designed'to be open to readings where some single past event is being report;ed and to the possibility of a &amp;quot;hat)if ual&amp;quot; reading. The perfective is also open to the same ol)tions: 18 I had read the Times for years, but had gradually come to reeognise it as a capitalist rag. We have further to acknowledge the correct intuition that for telic events the perfcctive fbcuses on the end point of the event where the sinq)le aspect; views it as a whole. My current approach takes the MP given above for simple as a basis for t)oth,  but adds an extra clause saying that for the simple case all the events in the specified set, end within  Omitting this extra clause from the MP for perf means that the set, of ewmts in question could inelude one that is not yet complete, so that 19 I have also read the Guardian for years, but I am now becoming dissatisfied with it as well.</Paragraph>
    <Paragraph position="27"> has a past habitual reading which is ()pen to contimmtion in a way that the habitual reading of the simple past cannot be. The ramifications of this require further exploration, perhaps in conjunction with a treatment of implieature, like thai; given in (Gazdar, 1979) to explain why examples like (19) generally give rise to the feeling that the event sequence in question is not yet over and done with.</Paragraph>
  </Section>
class="xml-element"></Paper>