This Content Downloaded From 46.217.116.115 On Mon, 27 Sep 2021 04:49:31 UTC

Aspect vs.
relative tense: the case reopened

Author(s): Jürgen Bohnemeyer
Source: Natural Language & Linguistic Theory , August 2014, Vol. 32, No. 3, Special Issue:
Aspect across languages: semantic primitives, morphosyntactic representation and the
limits of cross-linguistic variation (August 2014), pp. 917-954
Published by: Springer
Stable URL: https://www.jstor.org/stable/43697744
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
Springer is collaborating with JSTOR to digitize, preserve and extend access to Natural
Language & Linguistic Theory
This content downloaded from

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
All use subject to https://about.jstor.org/terms
Nat Lang Linguist Theory (2014) 32:917-954
DOI 10.1 007/s 1 1 049-0 1 3-92 1 0-z
Aspect vs. relative tense: the case reopened
Jürgen Bohnemeyer
Received: 30 October 2012 / Accepted: 4 August 2013 / Published online: 26 November 2013
© Springer Science+Business Media Dordrecht 2013
Abstract Klein (1994) points out that within the treatment of the temporal seman-
tics of English that he proposes, there is no need to maintain the traditional distinction
between perfect aspect and anterior tense. An analysis of the semantics of perfect as-
pect in terms of placing the topic time in the post-time of the event under description
can account for the anterior tense readings of the pluperfect as well. In this article,
I argue that "Klein's Conjecture" appears more problematic once extended to other
languages, drawing on evidence from Japanese, Kituba, Kalaallisut, Korean, and Yu-
catec Maya. Languages such as Japanese have expressions of anterior tense that do
not fit Klein's analysis of perfect aspect (topic time after event time), while others -
e.g., Yucatec Maya - have expressions that fit Klein's analysis, but do not have ante-
rior tense readings. The additions to Klein's theory necessary so it can accommodate
the new evidence comprise a revised viewpoint aspect component that distinguishes
not only relations between topic and event time, but also relations between topic time
and the runtimes of states preceding and following the event in a causal chain, as
well as an updated tense module that distinguishes relations between topic time and
perspective times in addition to relations between topic time and utterance time.
Keywords Temporal semantics • Relative tense • Viewpoint aspect • Perfect •

Yucatec Maya • Japanese
1 Klein's Conjecture
The question examined in this article is whether viewpoint aspects and relative tenses
are distinct semantic categories. Informally, tense and aspect operators constrain (i.e.,
restrict) the temporal reference of utterances by expressing temporal relations that
J. Bohnemeyer (EE3)
University at Buffalo - SUNY, Buffalo, USA
e-mail: jb77@buffalo.edu
£[ Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
918 J. Bohnemeyer
partially bound the tim

pects constrain event d
temporal reference tim
post-state (further opt
strain the time interva
temporal order with r
which is the function o
zations, it is not obviou
shall call "Klein's Conjec
that within the theory
a sketch), the phenome
viewpoint aspects.
"I think that relative

notion of relative tens
the Future Perfect. W
'aspect' here. But then
imperfective and perfe
to be a particularly fo
Although Klein's Conje

the question of whethe
from one another sema
pology quite independe
described as perfect or
languages of the world
tion arises whether and
question is the central
According to Klein's C
ter analyzed as perfect
tenses is not anterior t
point, but perfect aspe
which it makes an asser
of the event.1
Traditionally, the Eng
like 'Perfect-in-the-P
persen 1924; Comrie 19
with respect to the tim
(1), the adverbial denot
of the arriving event,
denotes the time at wh
throughout this article, I a

crosslinguistically identifiabl
categories by representing th
capitalized initial letters).
£) Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
Aspect vs. relative tense: the case reopened 919
semantics of the Pluperfect, attributing the differen

tic ambiguity of the time adverbial, which is unders
(1), but as an event time adverbial in (2) (cf. Klein 1
(1) Bill had arrived at six o'clock. I arrived at six

done with his meal, so he must have gotten there
(2) Bill had arrived at six o'clock and had left aga
not get there until eight. (Comrie 1976:56)
In this article, I present evidence from English, J

West Greenlandic), Korean, and Yucatec Maya tha
Klein's Conjecture.3 In these languages, the aspect
that are conflated in the English Pluperfect and Futu
ysemous analysis are expressed separately. Yucatec h
patible only with interpretations such as the one in
reference/topic time adverbial, not with interpretatio
in which at six o'clock is understood as an event time adverbial. The same holds for
varieties of Kalaallisut according to Fortescue (1984). Yucatec also has a prospective
aspect that functions like a mirror image of a perfect aspect, placing the topic time
inside the runtime of a pre-state of the event described by the verb.
In contrast, Japanese, Kituba, and Korean have anterior tenses that occur in con-
texts such as (2) but not in those of type (1). I propose that the former are pure perfect
aspects, which place the topic time (or rather what Klein calls the 'topic time projec-
tion range' - the possible time interval in which the actual topic time of the utterance
must fall) inside the run time of some post state caused by the event described by
the verb stem. In contrast, Japanese and Kituba have pure relative past (i.e., anterior)
tenses. Unlike perfect aspects, these do not shift the topic time (projection range) to
a time after the described event and thus do not permit topic time adverbiais in cases
where the event time is not included in the topic time. To obtain the post-state aspect
readings, such anterior tenses may be combined with perfect aspects. And Korean has
a Pluperfect that, like its English equivalent, has an interpretation involving a com-
bination of deictic and anaphoric anteriority, but which unlike the English Pluperfect
lacks a Perfect-in-the-Past interpretation.
2Cf. also Demirdache and Uribe-Etxebarria (2004, 2007), who propose an LF-based account in which topic
time and event time adverbiais occupy distinct functional head positions in the syntax. The model compo-
sitions I develop in Sects. 2-7 of this article presuppose instead a generic version of non-transformational
phrase structure grammar semantically interpreted through type-driven translation into lambda calculus
(along the lines of Klein and Sag 1985). Even in such a framework, however, event time adverbiais and
topic time adverbiais will be assumed to enter the semantic composition at distinct stages and thus also to
occupy distinct syntactic positions. I take up this matter briefly in Sect. 4. There are approaches to compu-
tational semantics that allow syntax to underspecify the order of semantic composition, such as Minimal
Recursion Semantics (Copestake et al. 2005). I assume that it may be possible in such theories to compute
the two readings of the first clause of (l)-(2) without treating the clause as syntactically ambiguous.
3 An anonymous reviewer wonders what role translation plays in this article. The article is not, in fact,
based on translation data at all. The Yucatec data come from my own fieldwork on the language over a
period of two decades, relying on the analysis of recorded texts and a large battery of elicitation methods.
I assume that the same holds true for Fortescue's Kalaallisut data. The Japanese, Kituba, and Korean data
is taken from scholarly work by native speakers.
& Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
920 J. Bohnemeyer
Whether the English

mous, as Klein sugge
Sect. 4, I attempt to
The result is inconclu
strongly suggests tha
guages of the world a
have to be modified s
tenses relate topic tim
(adopting a term from
1993:593-601, whose d
tive point' parallels t
more generally, of w
also be anaphorically
erence points, and pe
phenomena (similarly
and an anaphoric com
terpretation, the valu
perspective time and
be other solutions in t
date anaphoric and co
distinct from topic an
times, however, pers
ance times and anapho
tenses such as those o
ondary topic times. C
Secondly, the tense co
tion to constraints on
absolute/deictic tense
of perspective times
complex absolute-rela
relation between these
time-relational treatm
topic time (projection
logical approach that
range) and the run ti
eventuality is embedd
(imperfective) or insi
may include the even
ary (egressive). This i
to be preceded by such
and other lexical even
in question is in fact
therein; Bohnemeyer
Several reviewers rai
the semantics of perf
gories of English as it
£ļ Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
First, this article addresses a theory of the semantics

inally proposed exclusively for English (Klein 1992,
number of semantic tense-aspect categories without a
versality of their expression across languages. Subse
what extent these semantic categories are expressed
guages, including Russian (Klein 1995), Mandarin (K
kaleut, Canada; Swift 2004), Yucatec (Maya, Mexic
2002, 2009), Badiaranke (Atlantic, Guinea, Guinea
Tonhauser 2014), Paraguayan Guaraní (Tupi-Guar
Cover and Tonhauser 2014), and the languages exam
such application of the theory to a new language can
of sorts: it tests the hypothesis that a set of assumpt
descriptions of temporality in other languages will d
language as well. This experiment usually results in a
of the theory proving in need of revision.
Such an experiment is precisely what the present
jecture claims that there is no need for a theory of
between perfect aspect and anterior tense. An analy
aspect in terms of placing the topic time in the post-
tion can account for the anterior tense readings of
article makes the case that Klein's Conjecture may b
appears to be more problematic once extended to ot
e.g., Japanese - have expressions of anterior tense t
perfect aspect (topic time after event time), while
pressions that fit Klein's analysis, but do not have
neither Klein nor I actually claim that there is any su
pect or a universal anterior tense expressed in the gr
Conjecture does not claim the universality of the se
because Klein proposed it solely for English. And th
claims, either, as the only claim it defends regarding
one: I argue that Klein's Conjecture does not hold un
guages that have grammaticalized perfect aspects al
tenses, nor do all languages that have grammaticaliz
express perfect aspect.
The paper is organized as follows. Section 2 summ
Klein's theory and introduces a standard formalizat
cusses a more explicit formulation of Klein's Conjec
monosemous account of the English Pluperfect and F
polysemous analysis, providing formal representatio
tense readings in the process and evaluating the - in
semy vs. monosemy. The following sections present
(and prospective) aspects (Sect. 5) and pure relativ
cusses Korean -essess as a deictic-anaphoric hybrid t
'upgrades' to Klein's theory that would allow it to a
sented in Sects. 5-7. Section 9 concludes.
& Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
922 J. Bohnemeyer
Kleinian variable Situation time TSit Topic time TT Utterance time TU
Labels used in this article Event time x(e) Topic time ttopc. Utterance time tuc
Corresponding
Reichenbachian variable Event point E Reference point R Speech point S
Related by (in Klein's theory) Viewpoint aspect Tense
Fig. 1 Building blocks of K
2 Klein's theory
Klein (1992, 1994) breaks with almost the entire tradition by proposing that (deic-
tic or 'absolute') tenses relate utterance times, not to event times, but to what Klein
calls topic times . These are in turn related to event times via an independent second
notional functional category, which Klein identifies as aspect. In the terminology of
Smith (1991), Klein's 'aspect* is viewpoint aspect , rather than situation aspect - a
classification of the temporal properties of situation/eventuality type descriptors -
which Klein calls 'aktionsarť. Other authors use 'grammatical aspect' for what Klein
means by 'aspect' and 'lexical aspect' for Klein's 'aktionsarť. Tense and aspect cat-
egories are semantic categories that may - but need not - be expressed by inflections
and/or function words by themselves or in language-specific packages with other
tense/aspect categories, but that are always involved in the temporal interpretation of
natural language utterances even if these utterances contain no expression of them.4
There is a structural similarity between Klein's proposal and neo-Reichenbachian
approaches such as Comrie (1981), Declerck (1991), Hornstein (1990), and Ogihara
(1996). Like the neo-Reichenbachians, Klein's theory breaks Reichenbach's ternary
arrays down into pairs of dyadic relations, as illustrated in Fig. 1. One of these is
a deictic relation involving Reichenbach's 'speech point' S (the utterance time), the
other an anaphoric relation involving Reichenbach's 'reference point' R. In Ogihara's
theory, both S and R are directly related to the 'event point' E - the runtime of the
described eventuality - whereas in the other theories, both E and S are related to R,
which serves as a link between them.
In Klein's theory, the ordering relations expressed by tense and aspect operators
constrain the range of possible values of the topic time variable. The actual topic
time of any given utterance must fall into an interval so constrained by the tense
and aspect operators that have scope over the asserted/questioned (etc.) proposition
if the discourse of which the utterance is a part is to be coherent. The topic time is
constrained vis-à-vis the 'situation time', which corresponds to E, by an aspectual
relation, and vis-à-vis the utterance time, corresponding to S, via a tense relation.
4 See Bittner (2005, 2008), Bohnemeyer (1998, 2009), Bohnemeyer and Swift (2004), and Smith et al.
(2007) for some proposals of how temporal interpretation in the absence of tense/aspect marking works.
& Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
So relative tense plays a similar role in the neo-Rei

played by (viewpoint) aspect in Klein's theory, partly
that relative tense and aspect are in fact the same thi
theory of temporal semantics can account for both in t
dering relations between topic time and 'situation time
eventuality.
In (neo-)Reichenbachian theories (which have an ancient history greatly predat-
ing Reichenbach 1947) and in the Discourse Representation Theory (DRT) tradition
adopting the reference time concept, this notion is understood as a time variable that
receives its value anaphorically from the discourse context. Klein's notion of 'topic
time' can be understood as a reinterpretation of this concept, preserving the context
dependence (through a mechanism Klein does not discuss; see Bohnemeyer 2009 for
a proposal based on Gricean pragmatics), but at the same time giving the variable
an entirely new role as the evaluation time for speech acts. For assertions, the topic
time is the time for which the speaker claims the asserted proposition to be true. For
polar questions, it is a time of which the speaker asks whether the proposition is true
during it. For commands, it is the time such that the speaker requests/instructs/orders
the addressee to make the proposition true during it. And so on. Kratzer (201 1) links
Klein's notion of 'topic time' to Austin's (1950) proposal that natural language utter-
ances are about particular situations. These 'topic situations' are not to be confused
with an eventuality of which the utterance may contain a lexical descriptor, such as
a verb. In the examples in (3), the described event is always one of Sally writing a
letter to Floyd, but the topic situation contains this event in (3a), is itself contained in
it in (3b), and precedes it (according to an intention attributed to Sally) in (3c):
(3) a. Sally wrote a letter to Floyd.

b. Sally was writing a letter to Floyd.
c. Sally was going to write a letter to Floyd.
The different aspect forms in (3) motivate Klein's observation that viewpoint as-
pect constrains the temporal relation between topic time and event time (or between
topic situation and situation under description). Table 1 gives a more comprehensive
overview of Klein's analysis of the semantics of the English tense-aspect categories.
Klein's approach is a member of a family of theories that treat viewpoint as-
pect in terms of the selection of a particular part of the event under description
(or of a larger causal chain containing this event) such that the utterance concerns
in some sense or other specifically this part (in the case of Klein's theory in the
sense that it is only this part whose realization is entailed by the truth conditions
of the utterance; cf. Bohnemeyer and Swift 2004). Other members of this family
include Bartsch (1986, 1995), Breu (1985, 1994), Chung and Timberlake (1985),
Dowty (1979), and Smith (1991). This 'time-relational' (Klein 1995) or 'frame-
selection' (Chung and Timberlake 1985) type of theory contrasts with a tradition that
treats viewpoint aspect as effecting mappings between lexical-aspectual (or 'situa-
tion aspect', in the terminology of Smith 1991) classes. In the simplest case, perfec-
tives are treated as eventive and non-perfective aspects as stative (e.g., Kamp 1979;
Kamp and Rohrer 1983; Kamp and Reyle 1993; Kamp et al. 201 1). Many authors dis-
tinguish classes of activities or processes in addition (e.g., Bach 1981; Hinrichs 1986;
& Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
924 J. Bohnemeyer
Table 1 Klein's (1994) analys

range); x(e) - situation time (t
Aspect Relation Tense Relation

Past Present Future
U op < ht tu C ttop tu < ttop
Perfective Simple Past Present Simple Future

r(e) ç ttop I wrote I write I will write
Imperfective Past Progressive Present Progressive Future Progressive
ttop C r(e) I was writing Ī am writing I will be writing
Perfect Pluperfect Present Perfect Future Perfect
r(e) < ttop I had written I have written / will have written
Prospective Past Prospective Present Prospective Future Prospective
ttop < *(e) / was going to write I am going to write I will be going to writ
Dowty 1986; Moens 1987; ter Meulen 1995; Bittner 2008, and many others). The
lation between these two types of theories is too complex to discuss here adequat
I will confine myself to this: it is my belief that a mereologically fully articu
time-relational theory such as the one sketched below - that is, a theory that
viewpoint-aspectual operators access, not merely to the described eventuality, bu
causal chains embedding it - can capture all the phenomena a "mapping"-type a
proach can. In contrast, mapping-type theories suffer from two principal short
ings: they obscure the role of lexical and viewpoint aspect in temporal anaphor
reducing it to linguistic "metaphysics" (Bach 1981); and they blur the distincti
between lexical (or situation) and viewpoint aspect.
Where mapping theories currently hold the edge is in their "implementation"
dynamic frameworks of semantics (as is the case for most of the above-mentio
studies). The only dynamic time-relational account of viewpoint aspect I am awar
is that developed in Bartsch (1995). For a sketch of elements of an implementatio
an enhanced Kleinian theory in the framework of Kamp et al. (201 1), see Bohnem
(2007). Perhaps most important for present purposes, the proponents of mapping
theories of aspectual semantics have not solved the central problem of this article
relationship between perfect (and prospective) aspect and anterior (and posteri
tense. As far as I am aware, these authors have not even addressed this problem.
Reichenbachian reference times can be understood as topic times in their capa
of being determined in context, in particular as time intervals that are anaphori
tracked across clause boundaries. None of the utterances in (3) actually involves a
erence time in the sense of an anaphorically tracked time interval. But this chang
soon as the utterances are placed into a connected discourse, as in the mini disco
in (4):
(4) a. Juan entered. Sally wrote a letter to Floyd.

b. Juan entered. Sally was writing a letter to Floyd.
c. Juan entered. Sally was going to write a letter to Floyd.
The topic situations of adjacent sentences in a coherent connected discourse will

be interpreted as related to one another. In (4a), the writing event will most likely
â Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
be interpreted as extending the situation containing

events will be understood to follow one another in s
the Reichenbachian notion of reference time in the
Partee 1984; Kamp and Reyle 1993; inter alia), this is
advancement of the reference time, with the event ti
being contained in their reference times. In contrast
ation will be understood to be the same for both cla
referential shift. The reference time of the second clau
The progressive in (4b) locates this reference time in
prospective in (4c) places it before the writing event.
At the most basic level, it could be said that Klein's
the neo-Reichbachian approaches by reinterpreting
However, on the (re-)introduction of a distinction bet
and anterior/posterior tenses proposed in this paper,
troduce the distinction between topic and a certain ki
I use the label 'perspective times' adapted from K
'temporal perspective point' via Cover and Tonhauser
primarily potential relata of topic times, either in lie
relative hybrid tenses such as those of English and K
utterance times and anaphorically determined times
those of Japanese and Kituba). Perspective times diff
latter, but not the former, are times such that propo
etc., for their truth during these time intervals. How
brid tenses such as those of English and Korean are a
spective times could really be said to be a special kind
marked for such tenses can be said to be about multi
Klein's theory also includes a treatment of eventual
with a classification of the lexical content of the eve
what Klein calls '0-state contents', '1-state contents',
contents are atemporal, corresponding to individual-
1-state contents correspond to Vendlerian 'states' an
ing a description of a 1-state content in an utterance
utterance is sensitive to a distinction between a tim
plies and times preceding and following that time. L
all state changes in non-gradient properties. Applicat
timeline into a time during which a certain stage-le
theme and one during which it does not. They diffe
they require semantically imperfective aspectual ope
an overlap relation with the first state, semantically
time into overlap with the second state, and semant
topic time into overlap with both, and thus with the t
Any other event-semantic or lexical-aspectual prope
tivity, telicity, or causal relations, remain outside the
rather minimalistic treatment of eventuality semant
ably richer ontology proposed in the following sectio
events, states, and causal relations.
Ô Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
926 J. Bohnemeyer
What are the advance

ation to attribute muc
reinterpretation of re
the semantics of many
such role and their co
textualized examples i
phenomenon exemplif
(5) a. What did you

b. There was a book
c. There was a book on the table. #It is in Russian.
Suppose in the context of witness testimony, the witness is asked (5a). Why would
(5c) be incoherent even though the book in question has clearly not in any meaningful
way stopped being in Russian? Because the question sets the topic time for the re-
sponse, and that topic time happens to be in the past of utterance time, requiring past
tense. The tense shift in (5c) induces incoherence because it can only be understood
as indicating a change in the topic situation. If utterances are about, and interpreted
with respect to, topic situations, the incoherence in (5c) becomes explainable in anal-
ogy to the effect of an abrupt change in discourse topic. Similarly, at a much larger
scale, the notion of 'topic time' makes predictions for descriptions of the temporal
semantics of utterances in tenseless languages such as Mandarin5 (Klein et al. 2000)
and Yucatec Maya (Bohnemeyer 2002, 2009) that are empirically borne out: if nat-
ural language utterances are indeed about topic situations, it follows that topic times
should play a role in the interpretation of tenseless utterances as much as they do
in that of tensed ones - and they do. In contrast, there is no obvious reason why the
semantics of tenseless utterances should involve a reference time variable.
Klein's theory has also been applied with considerable success to the semantics
of aspect systems in languages such as Mandarin and Yucatec (see the references
above) and Russian (Klein 1995) in a way Reichenbachian approaches have not. To
get a flavor for the reasons behind this success, consider the contrast between (3a)
and (3b) above. Why is it that (3a) entails completion of the letter whereas (3b) does
not? The answer is intuitively that any proposition the speaker wishes to assert or
question, etc., must concern the topic situation. And the topic situation includes the
completion of the letter in (3a), but not in (3b) (cf. also Bohnemeyer and Swift 2004;
Bohnemeyer 2012). In contrast, while it is possible to describe the contrast between
the English Simple Past and Past Progressive in terms of the relation between event
time and reference time, this in no way explains the distinct entailment patterns. Why
would inclusion of the reference time in the event time mean that the completion of
the event is not entailed? There is no obvious nexus between the reference time of an
utterance and its entailments unless the reference time is defined as a topic time.
Finally, and for similar reasons, Klein's approach has also made important contri-
butions to the theoretical understanding of finiteness and its interaction with temporal
interpretation (Klein 1998, 2006, 2009).
5Tenselessness of Mandarin is widely considered as having been established by Li and Thompson (1981),
but the tenseless analysis has recently been contested by Sybesma (2007). Cf. also the reply by Lin (2010).
Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
[Sallywas going to write a letter to Floyd]c
3e. ttopc<^ ') (sally ')

-edc [Sally BE going to write a letter to Floyd]c
XQ3e.ttopc < tw & Q(e) le. ')
[BE going]c Sally write a letter to Floyd

lP.le.tfofH < x(e) & P(e) le. write '(e)(floyd ')( letter *)( sal
Fig. 2 Simplified semantic composition for (3c) based on Klein
It could be said that Klein moves to reinterpret refe

this in turn leads him to reinterpret relative tenses
the notion of 'reference time' with that of 'topic ti
tense* with that of 'viewpoint aspect', Klein follo
the simplest theory that can account for the data he
Klein's Conjecture is an empirically testable hypothe
present article to present additional data that discou
more powerful theory, one that includes both topic t
conceptual toolkit and recognizes viewpoints aspects a
expressed by distinct functional categories in individ
Let me conclude this sketch of some of the bas
by briefly outlining a possible formalization (cf. als
Bohnemeyer 2012; Demirdache and Uribe-Etxebarria
Assume a model that includes domains De of events
uant variety), D, of individuals (of the perdurant var
with a mereological partial-order relation defined o
ders < (precedence) and C (inclusion) and the corr
c defined over D¡. The temporal trace function r . D
their run times. Natural language eventuality descript
Figure 2 diagrams the semantic composition of (3c) in
the entire lexical content of the clause as an event pr
ear order and thus circumventing the type raising o
the topic time variable tt0pC and the utterance time va
text c . All expressions whose semantics involves on
marked for this context dependence by a superscript
ator considered in Fig. 2 is represented as triggering
6There are other elements, which I ignore here since they do not d
article - above all, Klein's treatment of aktionsartãexictú aspect.
& Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
928 J. Bohnemeyer
description. The analy

the constraints on the
nent (the lower left n
with the appropriate r
3 Klein's Conjecture
Let me begin with a m

in (6) is of course min
(6) Klein's Conjectu

'viewpoint aspects'
range of possible topi
vis the runtime of t
that are neither view
exhaustively charact
times.
The motivation behind
• Traditionally, tenses
times - in the case of
relative (i.e., anaphori
• Klein's theory reinte
Reichenbach's anteced
• Replacing reference
aspect module, which
and topic times.
• Combined with the
analysis of relative te
the semantics of relat
in terms of relations
theory that is in line
the semantics of bot
times as topic times t
viewpoint aspects.
Under the above inte

Table 2.
In contrast to (6), I argue in this paper that anterior tenses and perfect aspects
express distinct semantic categories. Although these semantic categories are both ex-
pressed in English by a finite form of the have auxiliary combined with the participle
form of the main verb, which may or may not be monosemous (see below), they
are expressed separately and independently of one another in some (possibly many)
€) Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
Table 2 Relative tense categories and corresponding viewpoint

that the reference point R corresponds to topic time tt0p
Traditional relative tense category Corresponding aspectual category in

Klein (1992, 1994)
anterior (relative past) E < R Perfect z(e) < ttop

posterior (relative future) R < E Prospective ttop < r(e)
simultaneous (relative present) E and R Perfective x(e) ç ttop or
overlap Imperfective ttop C r(e)
other languages.7 The following semantic p

other and are not predicted by Klein's theor
his theory laid out in Klein (1994):
• True perfect aspects (i.e., expressions of p

stative and therefore do not combine with
• True perfect aspects describe resultant sta
by the verb. While the properties of the r
inference (cf. Nishiyama and Koenig 2010),
time is not.9
7 In a somewhat reminiscent fashion, Arche (2013) po

English complex past tenses between interpretations
owing to the synthetic aspect inflection of the auxiliar
8It was argued in Sect. 2 that one of the advantages o

point aspect was that the latter, but not the former, blur
aspect. But how can a separate treatment of viewpoint
ciled with an analysis of perfect aspects as stative? Map
in terms of mappings from events into events or states
sketched in this paper, these mappings can be explained
tuality under description or the surrounding causal chai
even a verb that is lexically dynamic in terms of the e
utterance in combination with perfective (or ingressive
include the boundaries of an eventuality of the requisite
the occurrence of change. True perfect aspects are argue
select for inclusion in topic time (and thus for assertion
description. In the same manner, prospective aspects pl
the event. Imperfective and progressive aspects are non
topic time into a phase of the eventuality during which
property of the theme of the utterance is actually com
approach affords explanations for the intuitions underl
of the fundamentally different nature of lexical and vie
9The very large body of literature on perfect aspects di

and Koenig (2010) for an overview and some discussion
experience' or 'existential interpretation', illustrated b
P. T Anderson. I assume that this is a special case of a r
to the lifetime of the subject. In contrast, I do not con
1978), as in Floyd has lived in Rochester since 2003 , a '
point aspect. The Extended-Now Perfect is a special te
viewpoint and a topic time that overlaps utterance/per
the past of utterance/perspective time. In English, it is e
& Springer

46.217.116.115 on Moff:ffff on Thu, 01 Jan 1976 12:34:56 UTC
930 J. Bohnemeyer
• True anterior tense

with expressions of d
default (in the sense
combinations, at leas
• When interpreted p
properties that disti
shift.10
To be able to account for these additional properties, the theory must be expanded
to include the following:
• Relations, not just between topic time and the event time, but also between topic
time and the runtimes of pre- and post-states surrounding the described eventuality
in a causal chain;
• Relations, not just between topic time and utterance time, but also between topic
time and some perspective time determined in context.
4 The English Perfect tenses revisited
Klein (1994) is a theory of tense and aspect in English. The main goal of the present
article is to test this theory against data from a variety of other languages and to
suggest modifications that would help improve the theory's coverage of the phenom-
ena in these languages. The goal of this section is to develop the alternative analyses
the proposed revised theory entails for the English tense-aspect forms Klein's theory
started from. Since part of the evidence in support of the revised theory is presented in
the following sections, rolling out the theory in the present section is perhaps some-
what backward. It is, however, my hope that this move will help clarify the relations
between the original theory and proposed revisions.
In Sects. 5 and 6, 1 present crosslinguistic evidence for the existence of pure per-
fect (and prospective) aspects and pure anterior (and posterior) tenses. The complex
tenses of English, however, are compatible with both aspectual and relative-tense in-
terpretations. Traditionally, this was considered a case of polysemy. Table 3 illustrates
with the aspectual vs. anaphoric (relative-tense) interpretations of the Perfect tenses
according to Comrie (1976) and Jespersen (1924). Jespersen's terms are 'Retrospec-
tive Past' vs. 'Ante-Preterit' for Comrie's 'Perfect-in-the-Pasť vs. 'Past-in-the-Pasť
and 'Retrospective Future' vs. 'Ante-Future' for Comrie's 'Perfect-in-the-Future' vs.
the Perfect Progressive of dynamic verbs. In contrast, in German, a language in which viewpoint aspect is
largely not grammaticalized, this notional category is expressed by the Present tense. In Spanish, the two
strategies apparently co-occur (María J. Arche, p. c.). And in Yucatec, which is tenseless, this notional cat-
egory cannot be expressed in a single clause at all. Single clauses represent the past beginnings of current
situations or their ongoingness, but not both.
10Some, but not all, of these properties also hold for prospective aspects and posterior tenses, respec-
tively. Matters are apparently complicated by the intensional nature of future time reference. For example,
prospective aspects, unlike perfects, are apparently compatible with event time specifications; compare
I am (now) going to meet you at five vs. I have (now) met you Cat five), and see Sect. 5. In this article,
I focus on the domain of past time reference in a bid to steer clear of these complications.
Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
Aspect vs. relative tense: the case reopened 93 1
Table 3 Traditional (Pre-Reichenbachian) analysis of the English per
Deictic component Nondeictic component

Aspect Anterior Tense
(Result focus) (Event focus)
Present Present Perfect N/A
Past Perfect-in-the-Past Past-in-the-Past
Future Perfect-in-the-Future Past-in-the-Future
'Past-in-the-Future'. The Present Perfect,

persen's terminology. Although Jespersen
under his notional 'aspect' categories, he a
core meaning is "the element of result" (1
Comrie's (1976) illustration of the contra
Past and the anaphoric Past-in-the-Past int
(1)~(2) in the introduction, repeated here f
(7) Bill had arrived at six o'clock. I arriv

done with his meal, so he must have got
(8) Bill had arrived at six o'clock and ha

not get there until eight. (Comrie 1976
Klein's Conjecture entails that these two

contrast between two distinct meanings a
a single construction, but rather are the r
ing a single general (i.e., underspecified) m
the difference in how the first clause Bill
the two examples boils down to the time
as specifying the topic time of the first s
a straightforward case of syntactic ambigu
in Sect. 2, Klein's analysis is represented in
verbial is interpreted as modifying topic t
Fig. 3, but as modifying the event time wit
difference is captured by two related fun
maps events into ordered pairs of natural n
to the minute), and clocktimej : D¡ N x
vals.11 The two adverbiais also differ in wh
the event time adverbial in Fig. 4 modifies
the topic time adverbial in Fig. 3 modifie
11 Of course, different formats are needed for diffe

of day ('morning', 'evening', etc.). And the interpre
considerable vagueness.
12 In the model compositions I present in this article
the lexical event description, and thus inside the scop
adverbiais attach outside the scope of aspect markers,
& Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
932 J. Bohnemeyer
[Bill had arrived at six o'c

^e t(op^ 'l« & clocktimeJ(t
-edc [Bill HAVE arrived at six o'clock]0

XR3e.tlopi< tu. & R(e) Xe.clocktimej(tlopJ = 6 & x(e) < t,oļH. & arrive '(e)(b
[Bill HAVE arrived]0 [at six o'clock]0

Xe. x(e) < tfnfH. & arrive' (e)(bilV) XQ.Xe.clocktimej(tUifK ) =
[HAVE ...-ed]° Bill arrive

XP.Xe.x(e) < tlofH & P(e) Xe .arrive '(e) (bill
Fig. 3 Simplified semantic
[Bill had arrived at six o'clock]0

3e.tlopc< tuc & x(e) < tU)fK & clocktime e(e) = 6 & arr
-edc [Bill HAVE arrived at six o'clock]0

XR3e.t,op< tlu & R(e) Xe.x(e) < tt0fH. & clocktime e(e) = 6 & arrive '(e)(bi
[HAVE . . .-ed]° Bill arrive at six o'clock

XQXe.x(e) < t(opc & Q(e) Xe.clocktime e(e) - 6 & arrive '(e)(bil
Bill arrive at six o'clock

Xe. ar rive '(e)(bill ') XP.Xe.clocktime e(e)
Fig. 4 Simplified semantic composition for (8) based on Klein
compositional interpretation correctly entails the arrival ev

time, specified as six o'clock, in (6)/Fig. 3, whereas it place
arrival event, said to have occurred at six o'clock, and utter
ternative is of course to left-dislocate topic time adverbiais). However, in rea

adverbiais and tense and aspect markers enter the semantic composition
from Yucatec and Kalaallisut presented in Sect. 5 suggests that pure perf
with event time specifications. This entails that viewpoint aspect marker
the event description with time adverbiais. They also affect their interpr
the event time adverbial later in / was going to call you later, but now
implication of these facts, as well as, e.g., the distribution and interpretation
clauses/projections, for the ordering of adverbiais and functional catego
in-depth study.
Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
Klein's analysis raises the question why only topic t

adverbiais, are compatible with the Present Perfect:
(9) Bill has arrived now/*at six o'clock.
Klein (1992) refers to this asymmetry as the Prese

the following explanation: by requiring topic time to
ble 1 above), the Present tense narrowly specifies top
tenses do not. Present tense marking renders topic ti
definite'. But it is pragmatically infelicitous for both
specified for p-definiteness in the same utterance:
(10) #At 7pm, Bill had arrived at six o'clock.
One might attribute the infelicitousness of (10) to

rendering the topic time specification vacuous, som
of the locative phrase in (1 1):
(11) #At the hip, Sally is 6ft tall/weighs 1201b.
Since Sally's height and weight are properties of S

sense to assert them with respect to particular parts.
the arrival event in (10) is a property of that event th
the topic time. But since the Present tense component
topic time anyway, restricting it to the deictic prese
avoided instead, since they render Present Perfect cla
There is, however, another major difference in ho
six o'clock is interpreted in (7) and (8), aside from t
(7), but not in (8), a result state of the arrival event
at the location where the speaker met him, is under
the topic time coincides with that of the second sent
link between the two (cf. Partee 1984; Bohnemeye
understood to overlap with the speaker's arrival in (
in (8). On a monosemous analysis such as Klein's, one
difference through contextual inferences and perhap
not seem entirely satisfactory, as it does not explain
had arrived at six o'clock , on the interpretation that
time, commits the speaker to the truth of the propo
at six o'clock at wherever it was he had arrived. Kle
topic time to lie in the relative future of the arrival e
having left again by that time.
In contrast, the analysis I advocate in this article d
the result state holding at topic time in (7), but not
the two meanings associated with the Pluperfect. Thi
primarily in the participle, which expresses perfect a
but a combination of anterior tense and perfect/ve a
The perfect aspect places the topic time projection ra
caused by the described eventuality. On this analysi
ated (asserted, questioned, etc.) with respect to topic
& Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC12:34:56 UTC
934 J. Bohnemeyer
[Bill had arrived at six o'

3«?.3s.f/opc< tuc & tlopc
-edc [Bill HAVE arrived at six o'clock]«

XR3e.tlopc< tuc & R(e) Xe3s.clocktimeJ(t,opo) ~ 6 & ttopiCx(s) & e» s & arr
[Bill HAVE arrived]6 [at six o'clock]«

"Ke3s.ttopcCx(s) & e*>s & arrive '(e) (bill) XQ.Xe.clocktime_t( tU)fJ =
[HAVE ...-ed]c Bill arrive

XP. Xe3s.tļofKCx(s ) & e»s & P(e) Xe. arrive '(e)(b
[Bill had arrived at six o'clock]«

3e. tn< tuc & ttopc< & x(e) C ttopc & clocktime_e(e
-edc [Bill HAVE arrived at six o'clock]c

XR3e. tuc á R(e) ^ xrc & T(e) £ hope & clocktimeje(e) = 6 & arrive '
[HAVE . . .-ed]c Bill arrive at six o'clock

XQ.Xe.ttopc< & x(e) C tlopc & Q(e) Xe.clocktimeje(e) = 6 & arrive '(e)(bilV)
Bill arrive at six o'clock

Xe.arrive '(e)(bill ') XPXe.clocktime_e(
Fig. 6 Simplified semantic composition for the Past-in-the-Past
but a (resultant) state description. Although this analysis d

the result state is the state of Bill's presence, this is the on
at least in the absence of very special contextual informati
the model theory sketched in Sect. 2 by adding a subdomai
states, and the causal relation ^ > , which defines a strict pa
In contrast, the anterior tense interpretation of the part
topic time projection range in the past of some perspective
the context, just like the auxiliary confines it to the past
the Perfect-in-the-Past interpretation in Fig. 5. A conceivab
13 The issue of the range of possible interpretations of the result state of

for decades. See Nishiyama and Koenig (2010) for a recent take.
& Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
der which both the deictic meaning component expr

anaphoric meaning component expressed by the parti
latum , one constraining it with respect to utterance ti
perspective time, conflicts with examples such as (1
(12) When you arrive at his home tomorrow, Bill

you may want to check whether the plants need
The Future Perfect clause in (12) describes an even

five days prior to perspective time, while the perspec
poral clause, is placed a day in the future of utteranc
two relations entails that the event of Bill's leaving p
days. Assuming perfective viewpoint aspect, i.e., inc
ing in the topic time, the latter should precede utter
if the finite component of the complex matrix claus
tion between utterance time and the topic time projec
ought to be in the Past tense, i.e., the complex tense
substitution of the Pluperfect for the Future Perfec
(12') #When you arrive at his home tomorrow, Bill
Assuming the finite component is always deictic, i.

relatum , the relation that correctly predicts the finite
in (12) is the relation between utterance time and per
Since this analysis entails that the finite tense c
tly different meanings under the two interpretations
topic time vis-à-vis utterance time under the Perfect
constraining perspective time vis-à-vis utterance tim
interpretation - the question arises whether the dei
the auxiliary should be considered polysemous (as fo
see below). I do not think that this is the case. A
tive times in hybrid tenses can be understood as
long as it is ensured that the auxiliary picks up the r
times utterance - the right one being the outermost o
discourse - a monosemous tense auxiliary can do the
On the anterior past/future reading, the Pluperfec
tually perfective. Independent evidence suggesting pe
that these forms support referential shift (cf. Kam
example, the events described by the non-initial Plup
ture Perfect forms in (14) are understood each as be
that follows that of the preceding verb form:
(13) Bill had arrived, unpacked, taken a shower,

(14) Bill will have arrived, unpacked, taken a show
14 Strictly speaking, the topic time could also include Bill's leaving
future. However, the Future tense of the auxiliary in (12) suggest
in the future of utterance time.
& Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
936 J. Bohnemeyer
Moreover, the fact th

adverbiais is likewise
time specifications are
fective descriptions t
inside the scope of ass
Why would the anter
At this point, I can at
answer assumes that a
tual viewpoint in orde
point is expressed or
aspect is the default i
English, owing to a s
tity. Given that the a
morphologically unma
otherwise expresses p
that this implicature -
case carries over to th
What about the Pres
cle, the reason why t
tions is that it allows
interpretation - see T
compatibilityof the P
ent incompatibility of
me address these in tur
How can it be explain
Perfect, lacks the ant
ing the present tense i
in the perspective tim
(15) a. *Bill has arrived at six o'clock.

b. 3 e.tuc C tre & hope < tre & r(e) ç tîopc & clocktime_e(e ) = 6 &
arrive 'é){bilV)
Except for the vacuous perspective time variable, this has the same truth conditions
has the Simple (perfective) Past, illustrated in (16):
(16) a. Bill arrived at six o'clock.

b. 3 e.ttopc < tuc & r(e) ç tt0pC & clocktime_e(e ) - 6 & arrive 'e){bilV)
This is how the French Passé Compose and the German Present Perfect are inter-
preted when they occur with event time adverbiais, which both categories allow. In
these languages, the domain of the present perfect extends well into the perfective
past, where it competes with past tense forms, which are in various ways more re-
stricted than the English Simple Past. Standard varieties of English in contrast block
15 As mentioned above, future time reference is different in this respect, since it allows attaching event
times to the anticipated or planned (etc.) realization of events. This explains the event time specifications
in examples such as I was finishing/ going to finish by Monday , but then my hard disk died on me.
Ê Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
this extension of the Present Perfect, mandating the us

past time reference.16
The Present Perfect is incompatible with event tim
aspect interpretation is its sole interpretation and b
is incompatible with event time specifications. The ev
comes from the pure (tenseless) perfect aspects of Y
in the next section. The explanation for this incompat
fects (i.e., perfects that actually express perfect vie
Compose and the German Present Perfect in their pe
uses do not) being semantically stative, as illustrated
On the revised Kleinian theory proposed in th
anaphoric-tense meanings of the English Pluperfect a
cally quite distinct. This, however, by no means enta
are polysemous. Of course, to the extent that the
semantic properties not predicted by Klein's theory
meanings cannot quite be those Klein proposes. An
here does not provide for monosemous expressions th
But this does not mean that there is no possible mon
out in some yet-to-be-developed third theory that acc
A polysemous analysis of the non-Present Perfect t
these forms are identifiable as polysemous using sta
discrete deniability of the two senses. Expressions t
senses in the mental lexicon or in the grammar of t
deniable under one sense while simultaneous assertion of the truth conditions of the
other in the same situation, with respect to the same context and the same set of
possible worlds, does not result in contradiction. This is illustrated by (17)-(20):17
(17) That's not a cow; that's a bull.

(18) #That's not a horse; that's a mare.
(19) Floyd isn't a doctor; he has a PhD in linguistics.
(20) #Sally isn't a monarch; she's a queen.
The tests suggest that cow is polysemous between the senses 'bovine' and 'fe-
male bovine' and that doctor is polysemous between 'physician' and 'person with a
16The German Preterit and the French Passé Simple appear to be largely restricted to narrative discourses.
In conversation, the Present Perfect/Passé Composé are used for past time reference. In contrast, in En-
glish, the Present Perfect underextends the domain of perfect aspect, being replaced by the Simple Past
in many contexts for reasons that are not clearly understood. For example, as Kratzer (1998) observes, if
somebody asks you out of the blue Who built this church? and you respond Borromini built this church ,
the Simple Past is robustly preferred over the Present Perfect in both question and answer, even though
you are discussing the result state of a past building event that holds at utterance time (my thanks to an
anonymous reviewer for pointing me to this observation).
17Most of these are adapted from (Cruse 1986:59-61). A different type of ambiguity test capitalizes on
anomaly resulting from coordination of different senses when ellipsis or VP anaphora is involved. Arche
(2013) applies a test of this kind in support of the analysis that stative clauses in the simple past are
ambiguous between perfective and imperfective interpretations in English.
ß Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
938 J. Bohnemeyer
doctorate',while horse
this diagnostic is mos
Cruse 1986:58-74), I se
words and inflections,
tween polysemy in th
would seem without m
However, (21 )- (22) sh
terpretations of the Pl
(21) #When I got ther

and waited for me t
(22) ?When I got ther

and left again befor
The examples try to d

the truth conditions o
first attempt clearly r
then he can also not h
sentence in the Pluper
tence under the Past-in
reading is assumed to
this is the time of the
The entailment relatio
in-the-Past reading do
event of a certain kin
that a suitable result st
the polysemy hypoth
the following set of ex
(23) ?I went to visit F

the day before, but

the day before, but

in the mail the day

before, but then lef
Two of my consultan
interpretations, wher
a possible exception ow
language speaker, seem
18 Strictly speaking, functi

that the triplets of phonolo
be learned item by item exa
Ô Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
conditions of the negated Pluperfect apply, when we

state at topic time against the prior occurrence of a
evidence.
Data such as the above lead me to believe that the Pluperfect and Future Per-
fect may well be monosemous, as Klein suggests. The fact that these meanings are
expressed by distinct constructions in the languages discussed in the following sec-
tions of course by no means precludes this. Just because English has distinct words
for 'arm' and 'hand' does not mean that Yucatec, like many other languages, cannot
have a single monosemous word covering both meanings (k' ab). Just because Russian
has unrelated monomorphemic words for 'light blue' (sinij) and 'dark blue' ( goloboj )
does not mean that English blue is polysemous. However, if the English Perfect tenses
really are monosemous, then there is no current theory of temporal semantics I am
aware of that can spell out their monosemous meanings and simultaneously account
for all their semantic properties discussed above.
5 Pure perfect aspects
By "pure" perfect aspects, I am referring to markers that do not conflate any tense
component, be it deictic or anaphoric. Exhibit A, and my main source of evidence and
focus of attention in this section, comes from my own field research on Yucatec, the
Mayan language spoken widely across the Yucatan peninsula in Mexico and Belize.
As argued extensively in Bohnemeyer (1998, 2002, 2009), Yucatec is a tenseless
language. Thus, a clause with the perfect aspect marker ts'o'k, customarily labeled
'terminative' by Yucatecanists due to its relation to the homophonous phase verb
meaning 'end', is freely compatible with topic times in the present, past, and future
of utterance time, as illustrated in (27): 19
(27) Ts'o'k in=mèet-ik le=nah=o'.

TERM AlSG=do:APP-INC(B3SG) DET=house=D2
'I (will) have/had built the house.'
The following example illustrates the use of ts'o'k with a topic time determined
by a preceding clause:
19Key to abbreviations in morpheme glosses: 1/2/3 - lst/2nd/3rd person; A - set-A (ergative/possessor)

bound pronominal clitic; ACC - accusative; ANT - anterior tense; APP - applicative derivation; ASP -
aspect; B - set-B (absolutive) bound pronominal suffix; CAUS - causative derivation; CAUSE -
'becausative' clause-linkage form; CL - classifier; CMP - completive; CON - connective particle; D2 -
anaphoric/distal clause-final particle; D3 - text-deictic clause-final particle; DA - locative/negative clause-
final particle; DEC - declarative; DET - determiner; DUB - dubitative; EMPH - emphatic (negation);
ERG - ergative; EXP - expected; GEN - genitive; IN - inanimate; INC - incompletive status; LOC -
locative; MOD - 'modalis' (case); ND - indicative; NEG - negation; NOM - nominative; PAST - past;
PERF - perfect; IPL - plural; PLUPERF - PluperfectPOST - posterior tense; PREP - generic (semanti-
cally empty) preposition; PRES - present; PRV - perfective; SG - singular; SUBJ - subjunctive status;
TE - converb clause-linkage form; TERM - terminative (perfect) aspect;TOP - topic marker; UP - upper
bound.
& Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
940 J. Bohnemeyer
(28) Káa=h-tàal-ech way h-ts'o'k ka'-p'éel ha'b=e',

CON=PRV-come-B2SG here PRV-end(B3SG) two-CL.IN y
ts'o'k in=mèet-ik le=nah=o'.
TERM AlSG=do:APP-INC(B3SG) DET=house=D2
'(When) you came here two years ago, I had built the house.'
The event time of the ts 'o 'k clause - the time of building the house - is understood
as preceding the event time of the first clause, the time of the addressee's earlier visit.
On the aspectual analysis, this follows from ts'o'k placing the topic time projec-
tion range inside a time at which a result state (the precise nature of which may be
pragmatically determined) of the building event obtains.20 On an alternative anterior
tense analysis, ts'o'k expresses anteriority with respect to a perspective time, which
in conversation may be interpreted to be the utterance time, as conversations are dis-
courses that are referentially grounded in the speech situation. However, under an
anterior tense analysis, it should be possible for ts'o'k clauses to accept event time
adverbiais - contrary to fact. Event descriptions with event time adverbiais referring
to the deictic or anaphoric past must be marked for perfective aspect, as illustrated in
(29a). Attempts at using ts'o'k instead are rejected by speakers (29b).
(29) a. T-aw=il-ah in=suku'n ho'lheak,

PRV-A2 =see-CMP(B3SG) AlSG=elder.brother yes
he'bix t-a=tukul-ah=e'?
like PRV-A2 =think-CMP(B3SG)=D3
'Did you meet my brother yesterday, as you had planned?'
b. #Ts'o'k aw=il-ik in=suku'n ho'lheak?
TERM A2 =see-CMP(B3SG) AlSG=elder.brother
(intended: 'Have you met my brother yesterday?')
The only possible interpretation of the adverbiais ho'lhe

time adverbial, which pragmatically makes little sense ('W
state of having met my brother?'). Moreover, topic time ad
ferred to be left-dislocated. In the same way, (30c) is unac
the question in (30a), which again forces an event time inter
With event time specifications in the (absolute or relative)
in (30b) is the only available option.
(30) a. Ba'x turneen hach ka'n-a'n in=pakat-ik-ech?

what CAUSE really tire-RES(B3SG) AlSG=look-INC-B2SG
'Why do you look so tired?' (lit. What reason (is it) that it is so tired
that I look at you?)
20 An anonymous reviewer wonders how result states are projected with stative and process verbs. There
are no stative verbs in Yucatec. Process and activity verbs occur with ts'o'k mostly under experien-
tial/existential interpretations and to express the state of the relevant process or activity having been com-
pleted. Cf. Bohnemeyer (2002) for detailed discussion and Moens (1987) on the (limited) compatibility of
activities and processes with the English Perfect.
Ê Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
b. Eskeh h-ah-en las kwàatroh

it.is.because PRV-awake-BlSG four.o'clock
te=ha'tskab+k'iin behe'la,-ak=a'
PREP:DEF=early+sun today-CAL=Dl
'It's because I woke up at four this morning.'
c. #Eskeh ts'o'k inw-ah-al las kwàatroh
it.is.because TERM AlSG-awake-INC four.o'clock
te=ha'tskab+k'iin behe'la'-ak=a'.
PREP:DEF=early+sun today-CAL=Dl
(intended: 'It's because I have woken up at four this morning.')
Similarly, ts'o'k is unacceptable in time-positional content questions and

clauses that modify heads understood to refer to their event times. For inst
dependent question in (31a) is ill formed. To repair it, ts'o'k must be replac
verbal core inflected for subjunctive status, as in (31b):
(31) a. Mix inw=ohel *ba'x òora ts'o'k u=hàan-al.

EMPH.NEG Al =know(B3SG) what hour TERM A3 =eat
(intended: 'I have no idea at what time he had eaten.')
b. Mix inw=ohel ba'x òora hàan-ak= i.
EMPH.NEG Al =know(B3SG) what hour eat-SU
'I have no idea at what time he had eaten.'
Ts'o'k clauses do not in fact entail any information about their event times other
than that they precede their topic times. This is shown by the lengthy textlet in (32),
in which a speaker asserts a ts'o'k clause (32i) with respect to a topic time set as the
time of a previously described event (32h), while simultaneously stating that he does
not know when the described event happened (32j):
(32) a. Pedro=e' h-hàan las sèeys.

Pedro=TOP PRV-eat(B3SG) six.o'clock
'Pedro, he ate at six.'
b. Chéen dyèes minùuto-s t-u=bis-ah.
only ten minute-PL PRV-A3 =go:CAUS-CMP(B3SG)
'It took him just ten minutes.'
c. Las syèete káa= h máan Pablo,
seven.o'clock CON=PRV pass(B3SG) Pablo
'(At) seven, Pablo came by,'
d. káa=t-uy=ohel-t-ah ts'-u=hàan-al
CON=PRV-A3 =know-APP-CMP(B3SG) TERM-A3 =eat
Pedro.
Pedro.
'(and then) he learned that Pedro (already) had eaten.'
e. Chenba'l=e', ma' t-uy=ohel-t-ah
only thing=TOP NEG PRV-A3 =know-APP-CMP(B3SG)
'However, he did not come to know'
& Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
942 J. Bohnemeyer
f. ba'x òora káa=h-hàan Pedro=i'.

what hour CON=PRV-eat(B3SG) Pedro=D4
'at what time Pedro had eaten.'
g. Las òocho=e' t-uy=a'l-ah Pablo ti'
eight.o' clock- TOP PRV-A3 =say-CMP(B3SG) Pablo PR
Juan=e':
Juan=TOP
'At eight, Pablo said to Juan:'
h. "Káa=h-máan-en t-uy=iknal Pedro las syèete=e'
CON=PRV-pass-B 1SG PREP-A3 =at Pedro seven.o'clock=TOP
"'(When) I went by Pedro's at seven,'
i. ts'o'k u=hàan-al leti'; chen ba'l=e',
TERM A3 =eat-INC it only thing=TOP
'he had (already) eaten; only,'
j. mix inw=ohel ba'x òora hàan-ak=i".
EMPH.NEG Al =know(B3SG) what hour eat-SUBJ(B3SG)=D4
'I have no idea at what time he had eaten.'"
Thus, ts 'o 'k clauses are neither compatible with event time adverbiais nor even
pragmatically understood as providing information about the time of the described
eventuality. An analysis that underspecifies the distinction between perfect aspect and
anterior tense cannot seem to explain this behavior. The analysis I wish to propose
accounts for it by placing the topic time projection range inside the runtime of a state
caused by the described eventuality. On this analysis, the semantic predicate evalu-
ated (asserted, questioned, etc.) with respect to topic time is not an event description,
but a (resultant) state description. My assumption is that only this state description is
accessible to temporal modification.21 Figure 7 illustrates this analysis for (27) fol-
lowing the conventions introduced in Sects. 2 and 4. For the sake of convenience,
Fig. 7 treats ts'o'k and the incompletive 'status' suffix -ik obligatorily triggered by
it as entering the semantic composition together as an unanalyzed package.22
One more piece of independent evidence in support of the analysis in Fig. 7 is
the fact that ts'o'k, just like the English Present Perfect, is anomalous or infelicitous
21 An anonymous reviewer suggests that the availability of the event as opposed to the result state for
anaphoric reference in subsequent discourse might be an additional diagnostic for the distinction between
perfect aspects and anterior tenses. This remains to be investigated, although I am skeptical. Based on my
experience with Yucatec discourse, I would be very surprised if it turned out not to be possible to continue
(27), for example, saying Tu bisahen ka'p'éel semàana 'It took me two weeks'. The reason for this would
appear to be that even though it is only the result state that is asserted or questioned to hold using ts'o'k ,
this state is still described as a result of an event of the relevant kind having occurred. A ts'o'k clause
in fact entails the occurrence of an event of the kind described by the verb. It would not be truthful, for
example, to assert (27) if the house had come into existence some way other than by the speaker building
it.
22The status suffixes combine viewpoint aspect and mood meanings (Bohnemeyer 1998, 2002, 2009,
2012). The incompletive expresses unmarked mood and, when governed by lexical matrix predicates,
imperfective aspect. When selected by a preverbal aspect marker such as ts'o'k , however, -ik does not
compositionally contribute to the truth conditions of the utterance. The occurrence of -ik with ts'o'k is
presumably a reflex of the diachronic relation between ts 'o 'k and the homophonous aspectual verb meaning
'end'.
Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
[Ts'o'k in=mèet-ik le=nah=o']c

TERM A 1 SG=do:APP-INC(B3SG) DET=house=D2
i (will) have/had built the house'
3s3e. ttopcCx(s) <£ e»s & build' (e) (house ')(spkr)
[ts'o'k ...-ik]c in=mèet le^ah^

TERM . . .-INC A 1 SG=do:APP(B3SG) DET=house=D2
XP.3j?.3ď. ttopJZx(s) & e»s & P(e) 'I build the house'
Xe. build '(e) (house ')(spkr)
Fig. 7 Simplified semantic composition for (27)
(depending on whether the effect is assumed to be semantic or pra

entirely clear which one it is) when an individual involved in the resu
exists at topic time. Thus, as (33b) shows, speakers reject ts'o'k as a
the perfective in (33a), because the speaker states that his father h
This is readily explainable under a result state analysis, because attrib
to an individual at a given topic time presumably gives rise to the pr
the individual exists during that topic time.
(33) a. T-a=k'ahóol-t-ah in=tàatah,

PRV-A2 =acquaintance-APP-CMP(B3SG) Al
le=máax h-kim te=ha'b h-maan=o'?
DET=who PRV-die(B3SG) PREP:DET=year
Miin chen hun-teen-ili' t-inw=il-ah.
DUB only one-time-UPPRV-AlSG=see-CMP(B3SG)
'Did you get to know my father who died last year? - 1 think I only met
him once.'
b. #TsVk a=k'ahóol-t-ik in=tàatah,
TERM A2 =acquaintance-APP-INC(B3SG) AlSG=
le- máax h-kim te=ha'b h-máan=o'?
DET=who PRV-die(B3SG) PREP:DET=year P
#Miin chen hun-téen-ili' ts'o'k inw=il-ik.
DUB only one-time-UP TERM AlSG=see-INC(B3SG)
This is entirely parallel to the famous (33):
(34) a. (Uttered in 1971) #Einstein has visited Princeton.

b. (Uttered in 1971) ?Princeton has been visited by Einstein.23
The existence of pure perfect aspects does not seem to be restricted to Yucatec. The
properties of ts'o'k discussed above are shared by the Kalaallisut (West-Greenlandic)
marker - sima -, according to Fortescue's (1984) description. Kalaallisut is one of three
23 Thus Chomsky (1971:212-213) and McCawley (1971:106-108); Comrie (1976:59 [fh. 4]) rejects (34b) .
as well.
& Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
944 J. Bohnemeyer
dialects of the Eskimo

language (Bittner 2005
mined by a preceding
(35) a. Angirla-rama allakkat atuar-sima-vai.

come.home-lSG.CAUSE letter(PL) read-PERF-3SG.3PL.IND
'When I came home he had read the letters.' (Fortescue 1984: 274)
(36) a. Aqagu siku-mi sivisuu-mik aallar-sima-ssa-(p)u-nga.

tomorrow ice-LOC long-MOD leave-PERF-EXP-DEC-lSG
'I will (lit. expect/am expected to) be gone out on the ice a long time
tomorrow/ (Bittner 2009:15)
Fortescue notes that -sima- can occur in the context of (37), but not in that of (
(37) Nuum-miis-sima-vunga.
Nuuk-be.in-PERF- 1 SG.IND
'I have been to Nuuk.' (Fortescue 1984:272)
(38) Juuli-p aappa-ani Nuum-miip- 0-punga.

July-ERG second-LOC Nuuk-be.in- ASP- 1. SG.IND
T was in Nuuk on the second of July.' (Fortescue 1984:273)
Thus, according to Fortescue, - sima -, like ts'o'k , appears to be incompatible with

event time specifications. However, Trondhjem (2009) presents an example of - sima -
in combination with such an adverbial. She suggests that this marker may have been
extended to perfective past time reference under the influence of contact with Danish.
If true, this would leave the hypothesis of - sima - as a pure perfect aspect marker intact
for contemporary or historic varieties of the language not affected by this change.
Having seen evidence of pure perfect aspects, the question now arises whether
the second interpretation of the complex perfect forms of English - the relative tense
interpretation - likewise occurs in isolation, i.e., in expressions that do not have per-
fect aspect uses. This question is the topic of the next section.
6 Pure anterior tenses
A pure anterior tense constrains topic time vis-à-vis some perspective time given in
the discourse context rather than vis-à-vis utterance time. That the anterior relation
holds between the perspective time and topic time, not the time of the described
eventuality, follows from the fact that anaphoric tenses, like deictic tenses, may ex-
press exclusively tense relations and overtly combine with separate viewpoint aspect
markers. Alternatively, the anaphoric tense marker may also conflate a viewpoint as
pect meaning in addition to its tense meaning, either by semantically expressing both
meaning components or by semantically expressing the anaphoric tense meaning and
associating with an aspectual interpretation via a conversational implicature. A case
in point is Japanese - ta on Ogihara's (1996, 1999) account. Example (39) shows -ta
in an embedded clause expressing anteriority relative to the matrix clause, which is
understood to describe a future eventuality:
Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
(39) Taroo-wa [terebi-o mi-ta ato-de] benkyoo-suru.

Taro-TOP TV-ACC watch-ANT after-LOC study-PRES
Taro will study after watching TV.' (Ogihara 1999:329)
In the absence of appropriate contextual specifications, the default perspective

time for matrix clauses is utterance time, as illustrated in (36):
(40) Taroo-wa kinoo hon-o yon-da.

Taro-TOP yesterday book-ACC read- ANT
1) Taro (had) read the book yesterday.'
2) '*As of yesterday, Taro had read the book.' (Ogihara 1999:330)
When not accompanied by an aspect marker, - ta clauses are interpreted perfec-

tively, i.e., to the effect of inclusion of the runtime of the described eventuality in the
topic time of the - ta form. Thus, the time adverbial in (40) is understood to spec-
ify the event time, not the topic time or perspective time. As Ogihara points out, a
perfect interpretation under which the time adverbial would specify a topic time fol-
lowing the time of the reading event is unavailable in (40). However, to obtain this
interpretation, -ta can be combined with the -te iru construction, as illustrated in
(41):
(41) Taro-wa kinoo-no jiten-de sudeni

Taro-TOP yesterday-GEN timepoint-LOC already
sono hon-o yon-de ita.
that book-ACC read-TE be:ANT
1) 'As of yesterday, Taro had already read the book.'
2) 'As of yesterday, Taro was already reading the book.'
(Sotaro Kita, p. c.; Mitsuaki Shimojo, p. c.)
The -te iru form (bolded in (41)) has both perfect and imperfective/progressive
interpretations. In either case, it maps an eventuality description into a related state
(Nishiyama and Koenig 2010). Figure 8 diagrams an analysis of (40) in line with
the ideas developed above. The adverbial kinoo 'yesterday' is interpreted as an event
time adverbial ignoring linear order. The function antday : D¡ -> Dj maps time in-
tervals into the calendar day preceding the day containing them. The perspective time
variable trc, like the topic time variable ttopc, is interpreted indexically, i.e., its value
is a function of the context.
Figure 8 describes -ta as semantically perfective. However, the compatibility of
-ta with the -te iru form (cf. (41)) suggests that perfectivity may actually not be part
of the semantics of the marker, but may be merely a conversational implicature based
on Grice's second Maxim of Quantity, which assigns simple, common expressions
with stereotypical interpretations (Atlas and Levinson 1981).
The pragmatic analysis assumes that event descriptions are necessarily interpreted
for viewpoint aspect, but that this interpretation may not be a part of the semantic
meaning of the utterance, but merely pragmatically generated and thus defeasible.24
Future research will have to clarify this.
24 See Bohnemeyer and Swift (2004) on implicature-based aspectual interpretation.
â Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
946 J. Bohnemeyer
[Taroo-wa kinoo hon-o yon-da]c

Taro-TOP yesterday book-ACC read-ANT
Taro read the book*
3*- 'k & x(e)Q^_&x(e^ '(e)(book ')(taro ')

[Taroo-wa kinoo hon-o yon-]c -tac
Taro-TOP yesterday book-ACC read- ANT
Taro read the book' XQ3e. ti0fK< tn & x(e) Q ttopc & Q(e)
Xe. x(e) C antday(tuJ & read '(e) (book ')(taro ')
kinooc Taroo-wa hon-o yon-

yesterday Taro-TOP book-ACC read-
XP. Xe.x(e) C ant day (tu J & P(e) Taro read the book'
Xe. read' (e) (book ')(taro ')
A set of properties strikingly similar to that of - ta has been

-á(k)a in the Bantu language Kituba, spoken in the Democrat
As (42) illustrates, -á(k)a encodes anteriority relative to s
irrespective of whether the latter is in the past, present, or
time.
(42) a. Ntángu ya María kwis-á(k)a, muna béto méne di-áka.

time CMP María come- ANT then we PERF eat- ANT
4 When Maria came, we had already eaten [a long time / quite some time
ago].'
b. Ntángu ya María ata kwis, muna béto méne di-áka.
time CMP Maria POST come then we PERF eat- ANT
'When María comes, we will have already eaten [a long time / quite
some time ago].' (Mufwene 1990:99-100)
The examples in (42) involve topic times that do not include the runtime of t
described eventualities, but instead follow them, indicating perfect aspect. Howev
it turns out that this is due to the presence of the perfect auxiliary mé(ne ), wh
expresses perfect independently of -á(k)a.Mé{ne) clauses admit topic time specif
cations, but not event time specifications, in the absence of - a(k)a (cf. (43)), wh
- á(k)a clauses without mé(ne ) allow for event time specifications, but require t
event time to be included in the topic time, i.e., are interpreted perfectively (cf. (
I tentatively conclude that - á{k)a , like Japanese -ta, may not express aspect, b
rather implicate perfectivity when not accompanied by an overt aspect marker.
Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
(43) Béto mé(ne) dia / búbu yáyi / mazóno.

we PERF eat /day this /yesterday
1) *'We ate today /yesterday.'
2) 'As for today/yesterday, we are/were in the s
(44) Beto di-á(k)a / búbu yáyi / mazóno.

we eat- ANT / day this / yesterday
1) 'We ate today/yesterday.'
2) *'As for today/yesterday, we are/were in the
(Mufwene 1990:101)
It has been established that both perfect aspects lac

tions and anterior tenses lacking perfect aspect inter
of the world. Deictic-anaphoric hybrid tenses such as
in other, unrelated languages. And where they do, th
interpretations, unlike their English counterparts.
7 Deictic-anaphoric hybrid tenses
Deictic-anaphoric hybrids are tenses that combine a

tween topic time and perspective time with a deicti
time and utterance time. The English Future Perfec
aspectual readings are cases in point. An example of
in an unrelated language is the Korean Pluperfect m
differs from Japanese -ta and Kituba -á(k)a in that
on its anterior tense interpretation, combines the e
topic time and an anaphorically tracked perspective
tween this perspective time and utterance time. At
with -ta and - á(k)a , but not with the English Plup
a perfect interpretation. - Essess might be consider
deictic-anaphoric 'hybrid' tense since it is etymolog
tion of the simple past marker -ess. Except for the
of -ess has anaphoric rather than deictic reference,
compositionally, one token expressing anteriority of
ance time and one anteriority of topic time vis-à-vis
pair of examples contrasts the two markers:
(45) a. Mina-ka cip-ey kaass-ta.

Mina-NOM home-LOC go:PAST-DEC
'Mina went home.'
b. Mina-ka cip-ey kassess-ta.
Mina-NOM home-LOC go:PLUPERF-DEC
'Mina had gone home (but has come back now).' (Lee 2010:770)
-Essess does not express perfect aspect. It is anomalous in (46) under the interpre-
tation of the subordinate clause as a topic time rather than an event time specificatio
(in other words, on the interpretation that the train left before Suni's arrival):
fi Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
948 J. Bohnemeyer
[Mina-ka cip-ey kassess-ta]c

Mina-NOM home-LOC go:PLUPERF-DEC
'Mina had gone home*
3e3hJtopc< tK & tirc< tuc & i(e) Q ttofX. & go '(e)(h)(mina ') & goal(h)(home ')
Mina-ka cip-ey ka- -essessc

Mina-NOM home-LOC go- 'Mina go home* PLUPERF
Xe3h.go '(e)(h)(mina ') ã goal(h)(home ') XP3e. ttofK< tn. & ttn<
(46) #Suni-ka yek-ey tochakhay-(e)ss-ul ttay

Suni-NOM station-LOC arrive-PAST-when
kicha-nun ttenassess-ta.
train-TOP leave:PLUPERF-DEC
Intended: 'When Suni arrived at the station, the train had (already) left.'
(Lee 2010:770)
The expression of perfect aspect in Korean is the construction -e iss bolded in

(47):
(47) John-un ahop-si-ey konghang-ey tochakhayss-ta.

John-TOP nine-o'clock-LOC airport-LOC arrive:PAST-DEC
'John arrived at the airport at nine.'
Mary-nun pelsse konghang-ey tochakhayiss-ess-ta.
Mary-TOP already airport-LOC arrive:PERF-PAST-DEC
'Mary had already arrived at the airport.'
John-un Mary-lul po-ko uws-ess-ta.
John-TOP Mary-ACC see-and smile-PAST-DEC
'John smiled at Mary.' (Lee 2010: 770)
Figure 9 diagrams an analysis of (45b). Like the analysis of Japanese - ta provided

in Fig. 8, the analysis in Fig. 9 assumes that - essess is semantically perfective due to
its compatibility with event time specifications. However, just as in the case of - ta ,
the perfective interpretation might in fact be a conversational implicature instead -
this is not clear at present and remains to be investigated. Figure 9 also ignores the
contribution of the declarative mood. Assuming that it is an unmarked mood for asser-
tions, it should not affect the truth conditions of the formula the example is translated
into.25
This precludes the presentation of typological evidence. I now turn to the discus-
sion of the implications of this evidence for the theory of temporal semantics.
25 The motion verb 'go' is treated in Fig. 9 as introducing an existentially bound path variable h and a
function goal that assigns it an endpoint location. I refrain from amending the model theory to reflect these
additions since they are not relevant to the topic at hand.
& Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
Table 4 A mereological theory of viewpoint aspect (Bohnemeye
Notional aspect Part of the causal Semantics Example

marker chain selected
Prospective Pre-state XP3e3s.s » e & Sally was going to

hope C r(s) & P(e) write a paper on
aspect
Ingressive Initial boundary X.P 3e.ini (e) ç tt0pC Sally started writing
& P(e) a paper on aspect
Progressive/ Central part kP3e.ttopc C r(e) & Sally was writing a

imperfective P(e) paper on aspect
Egressive Terminal boundary XP3e.fin(e) £tÎOpC Sally finished

& P(e) writinga paper on
aspect
Perfective Entire event kP3e.r(e) ç ttopc & Sally wrote a paper

P(e) on aspect
Perfect Post-state 'P3e3s.e s Sally had written a

& hope C r(s) & paper on aspect
8 I
Th
asp
wo
ten
ten
an
sub
A
Fo
th
vie
can
pec
th
ref
ov
ins
top
ch
me
Th
ari
asp
& Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
950 J. Bohnemeyer
Event time z(e) Top
Perspectiv
Viewpoint aspect Tense
Fig. 10 Expanding the tens
European languages, but participate in the functional category system o

aspect in other languages.
Turning to the tense component of the theory, the expansions necessita
analyses proposed in this article are captured by Fig. 10.
Anaphoric tenses such as the anterior pasts of Japanese and Kituba cons
topic time variable, not vis-à-vis utterance time, like deictic tenses, but vi
perspective time given in discourse. However, anaphoric tenses seem to
ophoric interpretations in reference to the speech situation as well, and th
fact to be their default use in matrix clauses in conversation. In hybrid te
those of English and Korean, the deictic tense component does not relat
time to topic time, but to perspective time. The latter adjustment appears
radical once perspective times are understood as a special kind of topic tim
hybrid tenses. Both the existence of utterances with multiple topic times
reference times (in a broad sense) in addition to topic times are already
in Klein (1994:218-221) in connection with the phenomenon of tense mar
subordinate clauses. A comparison of the dialogues in (48) and (49) illustr
discourse perspective times are topic times as well:
(48) Q: When you entered Sally's office, what did she do?
A: She took a notepad from a drawer in her desk, jotted down a pho
ber, tore off the note, and handed it to me.
(49) Q: When you met with Floyd, what did you find out about his book
A: He had signed the contract with the publisher, revised the out
written a draft of the introductory chapter.
The first dialogue is a standard example of narrative progression. The qu

an initial topic time, which the clauses of the response successively adva
second dialogue, the question defines the perspective time for the Pluperf
response. Across the Pluperfects, there is again referential shift, suggestin
forms are interpreted perfectively and therefore as anterior tenses. At the
these clauses together provide the requested information concerning the
the question asks about - and this time frame is precisely the perspective
Pluperfects. Thus, the Pluperfects each have their own topic situation and
situations of the Pluperfects are all parts of a larger topic situation introd
question. In other words, the distinction between perspective time and to
hybrid tenses introduces a hierarchical structure of times and situations
the utterance makes an assertion or asks a question, etc. So the finite tense
of hybrid tenses has a single unified meaning that covers both the purely
of the tense morpheme and its occurrence in the hybrid tense construc
& Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
Aspect vs. relative tense: the case reopened 95 1
unified meaning relates topic time to utterance time

multiple topic times, the finite tense component rel
the one defined in the surrounding discourse context
utterance time.
9 Summary
Progress in science is the result of optimizing the trade-off between the parsimony
of theories and the range of phenomena, or the set of data, they can account for.
A landmark theoretical innovation - the reinterpretation of reference times as topic
times - allowed Klein (1992, 1994, 1995) and subsequent work to revise the tradi-
tional theory of tense nowadays associated with Reichenbach (1947), whose roots go
back to antiquity. The result is a unified theory of tense and viewpoint aspect, which
is able to capture both kinds of phenomena in a highly parsimonious and elegant fash-
ion and which is applicable to tensed and tenseless languages alike and also explains
the relation between tense, aspect, and finiteness.
The present article has confronted this simple, elegant theory with additional
data from a variety of sources. Pure perfect aspects such as those of Yucatec Maya
and some (synchronic or diachronic) varieties of Kalaallisut are semantically sta-
tive and incompatible with event time adverbiais. Assuming the English Present Per-
fect expresses exclusively perfect aspect, this suggests that stativity, rather than p-
definiteness due to the Present tense component, may account for the Present Perfect
Puzzle (Klein 1992), i.e., the incompatibility of the Present Perfect with event time
specifications. Pure anterior tenses such as those of Japanese and Kituba, on the other
hand, are compatible with expressions of various viewpoint aspects and are inter-
preted perfectively in their absence (at least with lexical event descriptors of a req-
uisite type). Perfectivity also explains why the English Pluperfect and Future Perfect
exhibit referential shift on their anterior tense interpretations, but not on their perfect
aspect interpretations. In contrast, the persistence of some result state at topic time
can be denied under the anterior tense reading of the complex Perfect tenses, but not
under the perfect aspect reading.
All of the above evidence points toward the conclusion that perfect aspects and an-
terior tenses are semantically quite different beasts and should be treated differently
in a theory of temporal semantics that is valid crosslinguistically. The additions to the
theory proposed in Klein (1994) necessary so it can accommodate the new evidence
are considerable: a drastically expanded semantic ontology and treatment of situation
aspect; a revised viewpoint aspect component that distinguishes not only relations be-
tween topic and event time, but also relations between topic time and the runtimes of
states preceding and following the event in a causal chain; and an updated tense mod-
ule that distinguishes, in addition to relations between topic time and utterance time,
relations between topic time and perspective times. These additions significantly re-
duce the elegant simplicity of the theory. Such is the nature of scientific progress.
However, the theory in its original formulation remains entirely valid for the set of
data for which it was originally proposed. One can think of the two versions of the
theory as two perspectives on the same idea, a grand vista viewed from on high and
the more detailed and fine-grained perspective developed in the present article.
& Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
952 J. Bohnemeyer
Acknowledgements A ver
represented Languages of
thank the audience, the me
for Psycholinguistics, and t
comments and suggestions.
Levinson, Aron Marvel, Do
Yucatec consultants who con
J. Arche, for extensive com
anonymous reviewers prov
much stronger. As a matter
sponsibility alone. The resea
Society.
References
Arche, María J. 2013. The construction of viewpoint aspect: the imperfective revisited. Natural Language
& Linguistic Theory, doi: 10.1 007/s 1 1 049-0 1 3-9209-5 (this issue)
Atlas, Jay D., and Stephen C. Levinson. 1981. It-clefts, informativeness and logical form: Radical prag-
matics (revised standard version). In Radical pragmatics , ed. Peter Cole, 1-62. New York: Academic
Press.
Austin, John Langshaw. 1950. Truth. Proceedings of the Aristotelian Society 24(1): 1 1 1-129.
Bach, Emmon. 1981. On time, tense, and aspect: An essay in English metaphysics. In Radical pragmatics,
ed. Peter Cole, 63-82. New York: Academic Press.
Bartsch, Renate. 1986. On aspectual properties of Dutch and German nominalizations. In Temporal struc-
ture in sentence and discourse , eds. Vincenzo Lo Caseio, and Co Vet, 7-39. Dordrecht: Foris.
Bartsch, Renate. 1995. Situations, tense, and aspect. Berlin: Mouton de Gruyter.
Bittner, Maria. 2005. Future discourse in a tenseless language. Journal of Semantics 22: 339-387.
Bittner, Maria. 2008. Aspectual universais of temporal anaphora. In Theoretical and crosslinguistic ap-
proaches to the semantics of aspect, ed. Susan Rothstein, 349-387. Amsterdam: Benjamins.
Bittner, Maria. 2009. Tense, mood, and centering. Rutgers University manuscript, http://www.rci.rutgers.
edu/~mbittner/bittner09_tmc.pdf. Accessed 08/01/2013.
Bohnemeyer, Jürgen. 1998. Time relations in discourse: Evidence from Yukatek Maya. PhD thesis, Tilburg
University.
Bohnemeyer, Jürgen. 2002. The grammar of time reference in Yukatek Maya. Munich: LINCOM.
Bohnemeyer, Jurgen. 2UU7. Aspect, temporal anaphora, and tenseless languages: A new oncean account.
Presented at SULA 4: The semantics of under-represented languages in the Americas. Universidade
de São Paulo.
Bohnemeyer, Jürgen. 2009. Temporal anaphora in a tenseless language. In The expression of time in lan-
guage , eds. Wolfgang Klein, and Ping Li, 83-128. Berlin: Mouton de Gruyter.
Bohnemeyer, Jürgen. 2012. In the mood for status: Subjunctive and irrealis in Yucatec. Presented at SULA
7: The semantics of under-represented languages in the Americas. Cornell University.
Bohnemeyer, Jürgen, and Mary D. Swift. 2004. Event realization and default aspect. Linguistics and Phi-
losophy 27(3): 263-296.
Breu, Walter. 1985. Handlungsgrenzen als Grundlage der Verbklassifikation [Action boundaries as the
basis of verb classification]. In Slavistische Linguistik J 984 [Slavistic linguistics 1984 /, ed. Werner
Lehfeld, 9-34. Munich: Otto Sagner.
Breu, Walter. 1994. Interactions between lexical, temporal and aspectual meanings. Studies in Language
18:23-44.
Carlson, Greg N. 1977. Reference to kinds in English. Doctoral dissertation, University of Massachusetts,
Amherst.
Chomsky, Noam A. 1971. Deep structure, surface structure, and semantic interpretation. In Semantics , ed
Danny D. Steinberg, and Leon A. Jakobovits, 183-216. Cambridge: Cambridge University Press.
Chung, Sandra, and Alan Timberlake. 1985. Tense, aspect, and mood. In Language typology and syntac
tic description. Vol. 3 of Grammatical categories and the lexicon , ed. Timothy Shopen, 202-258.
Cambridge: Cambridge University Press.
Comrie, Bernard. 1976. Aspect. Cambridge: Cambridge University Press.
Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
Comrie, Bernard. 1981. On Reichenbach's approach to tense. In

meeting , eds. Roberta A. Hendrick, Carrie S. Masek, and Ma
Chicago Linguistic Society.
Comrie, Bernard. 1985. Tense. Cambridge: Cambridge University
Copestake, Ann, Dan Rickinger, Carl Pollard, and Ivan A. Sag. 20
troduction. Research on Language and Computation 3: 281-33
Cover, Rebecca, and Judith Tonhauser. 2014. Semantic theory in th
ence. In Methodologies in semantic fieldwork , eds. Ryan Boc
Oxford University Press.
Cruse, D. Alan. 1986. Lexical semantics. Cambridge: Cambridge U
Declerck, Renaat. 1991. Tense in English. London: Routledge.
Demirdache, Hamida, and Myriam Uribe-Etxebarria. 2004. The sy
time , eds. Jacqueline Guéron, and Alexander Lecarme, 217-2
Demirdache, Hamida, and Myriam Uribe-Etxebarria. 2007. The
330-366.
Dowty, David R. 1979. Word meaning and Montague Grammar. Dordrecht: Reidel.
Dowty, David R. 1986. The effects of aspectual class on the temporal structure of discourse: Semantics or
pragmatics? Linguistics and Philosophy 9(1): 37-61.
Fortescue, Michael. 1984. West Greenlandic. London: Croom Helm.
Hinrichs, Erhard. 1986. Temporal Anaphora in discourses of English. Linguistics and Philosophy 9: 63-
82.
Hornstein, Norbert. 1990. As time goes by. Cambridge: MIT Press.
Jespersen, Otto. 1924. The philosophy of grammar. London: George Allen and Unwin.
Kamp, Hans. 1979. Events, instants, and temporal reference. In Semantics from different points of view,
eds. Rainer Bäuerle, Urs Egli, and Arnim von Stechow, 376-418. Berlin: Springer.
Kamp, Hans, and Uwe Reyle. 1993. From discourse to logic. Dordrecht: Kluwer.
Kamp, Hans, and Christian Rohrer. 1983. Tense in texts. In Meaning, use , and interpretation of language,
eds. Rainer Bäuerle, Christoph Schwartze, and Arnim von Stechow, 250-269. Berlin: de Gruyter.
Kamp, Hans, Josef van Genabith, and Uwe Reyle. 2011. Discourse representation theory. In Handbook
of philosophical logic , eds. Dov M. Gabbay, and Franz Guenthner. Vol. 15, 125-349. Dordrecht:
Springer.
Klein, Wolfgang. 1992. The Present Perfect puzzle. Language 68: 525-552.
Klein, Wolfgang. 1994. Time in language. London: Routledge.
Klein, Wolfgang. 1995. A time-relational analysis of Russian aspect. Language 71: 669-695.
Klein, Wolfgang. 1998. Assertion and finiteness. In Issues in the theory of language acquisition : Essays in
honor of Jürgen Weissenborn , eds. Norbert Dittmar, and Zwi Penner, 225-245. Bern: Peter Lang.
Klein, Wolfgang. 2006. On finiteness. In Semantics in acquisition , ed. Veerle Van Geenhoven, 245-272.
Dordrecht: Springer.
Klein, Wolfgang. 2009. Finiteness, universal grammar, and the language faculty. In Crosslinguistic ap-
proaches to the psychology of language: Research in the tradition of Dan Isaac Slobin , eds. Jiansheng
Guo, Elena Lieven, Nancy Budwig, Susan Ervin-Tripp, Keiko Nakamura, and Seyda Özcaliskan,
333-344. New York: Psychology Press.
Klein, Ewan, and Ivan A. Sag. 1985. Type-driven translation. Linguistics and Philosophy 8: 163-202.
Klein, Wolfgang, Ping Li, and Henriette Hendriks. 2000. Aspect and assertion in mandarin Chinese. Nat-
ural Language & Linguistic Theory 18: 723-770.
Kratzer, Angelika. 1998. More structural analogies between pronouns and tenses. In Proceedings of se-
mantics and linguistic theory VIII , eds. Devon Strolovitch, and Aaron Lawson, 92-1 10. Ithaca: CLC.
Kratzer, Angelika. 201 1. Situations in natural language semantics. In The Stanford encyclopedia of philos-
ophy, ed. Edward. N. Zalta. http://plato.stanford.edu/archives/fall2011/entries/situations-semantics/.
Last accessed 08/01/2013.
Lee, EunHee. 2010. Pluperfects in Korean and English discourse. Journal of Pragmatics 42: 766-780.
Li, Charles N., and Sandra A. Thompson. 1981. Mandarin Chinese : A functional reference grammar.
Berkeley: University of California Press.
Lin, Jo- Wang. 2010. A tenseless analysis of mandarin Chinese revisited: A response to Sybesma. Linguistic
Inquiry 41(2): 305-329. 2007.
McCawley, James D. 1971. Tense and time reference in English. In Studies in linguistic semantics, eds.
Charles J. Fillmore, and D. Terence Langendoen, 96-1 13. New York: Holt, Rinehart, and Winston.
McCoard, Robert W. 1978. The English perfect: Tense choice and pragmatic inferences. Amsterdam:
North Holland.
& Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC
954 J. Bohnemeyer
Moens, Mark. 1987. Tense, a

Mufwene, Salikoko S. 1990
systems , ed. John V. Sing
Nishiyama, Atsuko, and Je
Ogihara, Toshiyuki. 1996. T
Ogihara, Toshiyuki. 1999. T
jimura, 326-348. Oxford:
Partee, Barbara. 1984. Nom
Reichenbach, Hans. 1947. E
Smith, Carlota S. 1991. The
Smith, Carlota S., Ellavina
interpretation. Internation
Stowell, Tim. 2007. The sy
Swift, Mary D. 2004. Time i
Sybesma, Rint. 2007. Whet
ter Meulen, Alice G. B. 199
Tonhauser, Judith. 2007. N
83(4): 831-869.
Trondhjem, N. Frederikke.
tions on poly synthesis: Th
182. Amsterdam: Benjami
Vendler, Zeno. 1957. Verbs
£) Springer

46.217.116.115 on Mon, 27 Sep 2021 04:49:31 UTC

This Content Downloaded From 46.217.116.115 On Mon, 27 Sep 2021 04:49:31 UTC

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

This Content Downloaded From 46.217.116.115 On Mon, 27 Sep 2021 04:49:31 UTC

Uploaded by

Copyright:

Available Formats

Aspect vs.

relative tense: the case reopened

Stable URL: https://www.jstor.org/stable/43697744

This content downloaded from

Aspect vs. relative tense: the case reopened

Keywords Temporal semantics • Relative tense • Viewpoint aspect • Perfect •

This content downloaded from

partially bound the tim

"I think that relative

Although Klein's Conje

throughout this article, I a

This content downloaded from

semantics of the Pluperfect, attributing the differen

(1) Bill had arrived at six o'clock. I arrived at six

In this article, I present evidence from English, J

This content downloaded from

Whether the English

This content downloaded from

First, this article addresses a theory of the semantics

This content downloaded from

Kleinian variable Situation time TSit Topic time TT Utterance time TU

Related by (in Klein's theory) Viewpoint aspect Tense

Fig. 1 Building blocks of K

This content downloaded from

So relative tense plays a similar role in the neo-Rei

(3) a. Sally wrote a letter to Floyd.

This content downloaded from

Table 1 Klein's (1994) analys

Aspect Relation Tense Relation

U op < ht tu C ttop tu < ttop

Perfective Simple Past Present Simple Future

(4) a. Juan entered. Sally wrote a letter to Floyd.

The topic situations of adjacent sentences in a coherent connected discourse will

This content downloaded from

be interpreted as extending the situation containing

This content downloaded from

What are the advance

(5) a. What did you

This content downloaded from

[Sallywas going to write a letter to Floyd]c

3e. ttopc<^ ') (sally ')

[BE going]c Sally write a letter to Floyd

Fig. 2 Simplified semantic composition for (3c) based on Klein

It could be said that Klein moves to reinterpret refe

This content downloaded from

description. The analy

Let me begin with a m

(6) Klein's Conjectu

The motivation behind

Under the above inte

This content downloaded from

Table 2 Relative tense categories and corresponding viewpoint

Traditional relative tense category Corresponding aspectual category in

anterior (relative past) E < R Perfect z(e) < ttop

other languages.7 The following semantic p

• True perfect aspects (i.e., expressions of p

7 In a somewhat reminiscent fashion, Arche (2013) po

8It was argued in Sect. 2 that one of the advantages o

9The very large body of literature on perfect aspects di

This content downloaded from

• True anterior tense

4 The English Perfect tenses revisited

This content downloaded from