File Information

File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/intro/01/w01-1306_intro.xml

Size: 4,493 bytes

Last Modified: 2025-10-06 14:01:17

<?xml version="1.0" standalone="yes"?>
<Paper uid="W01-1306">
  <Title>Telling apart temporal locating adverbials and time-denoting expressions</Title>
  <Section position="3" start_page="0" end_page="1" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> In this paper, Portuguese and English - chosen as examples of the Romance and Germanic families - will be the object languages in an attempt to draw a dividing line between two semantically close subcategories of temporal phrases - time-denoting expressions and temporal locating adverbials (henceforth TDEs and TLAs, respectively). Nevertheless, the advocated hypotheses are expected to apply to comparable expressions in other languages as well. The formal framework for analysis will be the Discourse Representation Theory (DRT).</Paragraph>
    <Paragraph position="1"> In general informal terms, the difference between TDEs and TLAs can be put as follows: the former include (direct or indirect  ) representations of intervals, or sets of intervals, while the latter include phrases which locate entities (e.g. eventualities) on the time axis. Often, the distinction between them is unproblematic:  (1) a. Portugal became a Republic [in [1910]]. b. Several epidemics swept through  Europe [during [the 14th century]].</Paragraph>
    <Paragraph position="2"> In these examples, the expressions within the inner brackets [?] 1910, the 14th century [?] are clearly TDEs, while the prepositional phrases that contain them as complements [?] in 1910,  Indirect representation of intervals is performed, e.g., by temporal subordinate clauses or situational NPs, which primarily describe eventualities, but can act as derived TDEs when used as complements of temporal prepositions - cf. e.g. Rohrer (1977) and Hamann (1989).</Paragraph>
    <Paragraph position="3"> during the 14th century [?] are TLAs, which temporally locate the eventualities represented in the remainder of the clause.</Paragraph>
    <Paragraph position="4"> Sometimes, however, the dividing line between these two categories is unclear. This is due namely to the existence of expressions that can surface with exactly the same form in the typical contexts (to be defined below, in section 2) of both TDEs and TLAs, hence appearing to be ambivalent. This is the case of relatively simple expressions like English now, then, yesterday, and last week, for instance (or their approximate Portuguese counterparts agora, entao, ontem, and a semana passada, respectively). See the following pairs of sentences, where the expressions in italics occur in a typical context of a TDE (namely as the complement of since or until) in the a sentences, and in a typical context  of a TLA (namely in adverbial position) in b: (2) a. John has been in Paris since yesterday. b. John left the hospital yesterday.</Paragraph>
    <Paragraph position="5"> (3) a. Until then, John had been happy.</Paragraph>
    <Paragraph position="6"> b. Then, John felt very miserable.</Paragraph>
    <Paragraph position="7"> (4) a. John has been in Paris since last week.  b. John left for Paris last week.</Paragraph>
    <Paragraph position="8"> More interestingly, this is also the case of relatively more complex expressions, which include a temporal preposition (or prepositionallike particle), such as English before, after, between, when, ago or from (or their approximate Portuguese counterparts antes, depois, entre, quando, ha and de (...a) respectively):  (5) a. John has been a clerk since before 1980. b. John graduated before 1980.</Paragraph>
    <Paragraph position="9"> (6) a. John was a clerk until three years ago. b. John graduated three years ago.</Paragraph>
    <Paragraph position="10">  This distribution raises a categorisation issue that I will try to tackle in this paper, namely whether this kind of apparently ambivalent phrases should be regarded as temporal locating adverbials, as time-denoting expressions or as both. My contention here will be that they should (all) be taken as mere time-denoting expressions, i.e. expressions that denote intervals, or sets of intervals, but that do not just by themselves - locate eventualities on the time axis, therefore requiring the presence of an - explicit or implicit - temporal preposition like in, during, since, until, or the like, when used adverbially. Before proceeding to the arguments, let us briefly consider some basic semantic and syntactic properties of the categories under analysis.</Paragraph>
  </Section>
class="xml-element"></Paper>
Download Original XML