Review of Cross-linguistic Semantics of Tense, Aspect, and Modality
| EDITORS: Hogeweg, Lotte; Hoop, Helen de; Malchukov, Andrej
TITLE: Cross-linguistic Semantics of Tense, Aspect, and Modality
SERIES TITLE: Linguistik Aktuell/Linguistics Today 148
PUBLISHER: John Benjamins
D. Terence Langendoen, Department of Linguistics, University of Arizona
This book results from the TAM TAM: Cross-linguistic semantics of Tense, Aspect,
and Modality workshop held in Nijmegen in November 2006, which brought together
semanticists with a cross-linguistic perspective and typologists interested in
semantic theory to focus on cross-linguistic variation in systems of tense,
aspect and mood/modality (TAM). It contains 15 papers and an index. Each paper
contains its own set of references.
1. ''The semantics of tense, aspect and modality,'' by the editors, pp. 1-12,
contains a brief introduction that characterizes the domain of interest and the
goal of the book as follows: ''The time, the nature, and the factuality of
eventualities [events, states and processes] can be marked on ...verbs, by means
of tense, aspect, and mood or modality marking.... This book aims to give
greater prominence to the semantic richness of tense, aspect, modality, and
their interactions, in the languages of the world.'' (p. 1) The even briefer
conclusion expresses the ''hope that the present volume contributes to the
emergent field of semantic typology''. (p. 11) Between the introduction and
conclusion are four sections that describe the issues addressed by the remaining
14 papers. Section 1 ''Interactions between tense and aspect'' covers papers 2-6,
Section 2 ''Modality and factuality'' papers 7-9, Section 3 ''Different approaches
to modality'' papers 10-13, and Section 4 ''Case and modality'' papers 14-15.
2. ''Incompatible categories: Resolving the 'present perfective paradox''', by
Andrej Malchukov, pp. 13-32, surveys the study of constraints on cooccurrence of
grammatical categories, and examines one such combination in detail, present
tense and perfective aspect. The ''paradox'' arises from the incompatibility of
the meaning of the perfective aspect with that of the present tense (p. 18), and
its resolution is handled by constraint ordering in Optimality Theory.
3. ''The perfective/imperfective distinction: Coercion or aspectual operators?,''
by Corien Bary, pp. 33-54, argues for an 'aspectual operator' analysis of the
perfective/imperfective distinction for Ancient Greek and modern French that
also holds for any language in which the distinction is not restricted to past
tense, and that is superior to the 'coercion' analysis of the distinction in
French in de Swart (1998). The author reports (p. 35, n. 2) that her recently
completed PhD dissertation (Bary 2009) proposes a better theory to deal with the
distinction in Ancient Greek.
4. ''Lexical and compositional factors in the aspectual system of Adyghe,'' by
Peter M. Arkadiev, pp. 55-82, places Adyghe predicates into 'actional classes',
accounting for the distribution of temporal adverbials. These in turn can shift
the actional characteristic of the predicate by coercion in the sense of de
5. ''Event structure of non-culminating accomplishments,'' by Sergei Tatevosov &
Mikhail Ivanov, pp. 83-130, examines 'failed attempt' and 'partial success'
interpretations of accomplishment verbs cross-linguistically, and extends
Rothstein's (2004) theory of accomplishments to account for the diversity of
readings along these dimensions.
6. ''The grammaticalised use of the Burmese verbs la 'come' and θwà 'go','' by
Nicoletta Romeo, pp. 131-154, describes the metaphorical relation between the
Burmese verbs in the title, and their verbal 'marker' counterparts, the latter
being interpreted in different ways depending on the properties of the verb they
modify, the way in which events are represented in the clause, and the context
in which they appear. It also compares the function of the marker -θwà with that
of -laiʔ derived from the verb laiʔ 'follow' with respect to the relative
salience of the Agent and the Undergoer.
7. ''Irrealis in Yurakaré and other languages: On the cross-linguistic
consistency of an elusive category,'' by Rik van Gijn & Sonja Gipper, pp.
155-178, notes that the terms 'realis' and 'irrealis' lack ''a well defined
semantic content'', which has led some researchers such as Bybee et al. (1994) to
question the usefulness of the distinction for cross-linguistic research,
whereas others such as Mithun (1995) maintain that it is indeed useful for such
research. The paper defends the latter position, arguing that languages differ
systematically along an implicational scale with respect to the concepts
subsumable by the irrealis concept (future tense, imperative mood, negation and
8. ''On the selection of mood in complement clauses,'' by Rui Marques, pp.
179-204, proposes that the selection of indicative vs. subjunctive mood in
complement clauses in Romance languages depends on the 'expressed attitude' of
the main predicate; knowledge or belief selecting indicative, otherwise subjunctive.
9. '''Out of control' marking as circumstantial modality in St'át'imcets,'' by
Henry Davis, Lisa Matthewson & Hotze Rullmann, pp. 205-244, ''provides a unified
semantic analysis'' of the St'át'imcets circumfix ka- ... -a, as indicating
'circumstantial modality', covering the notions 'be able to', 'manage to',
'suddenly', 'accidentally' and 'uncontrollably', but does not encode
'quantificational strength' (effectively the distinction between necessity and
possibility). The paper contrasts this property of ka- ... -a and other
St'át'imcets modals with English modals, which lexically encode quantificational
strength but not the distinction between epistemic, deontic and the various
10. ''Modal geometry: Remarks on the structure of a modal map,'' by Kees de
Schepper & Joost Zwarts, pp. 245-270, analyzes the 'geometric' structure of the
semantic map of modality (van der Auwera & Plungian 1998) by decomposing the
various modalities into ''more basic modal features'', showing where deontic
modality fits and what role of connectivity plays in providing for
polyfunctionality. It also correlates a distinction on the map with the
grammatical distinction of raising/control.
11. ''Acquisitive modals,'' by Johan van der Auwera, Petar Kehayov & Alice
Vittrant, pp. 271-302 ''explores the fact that 'get' etymons may acquire modal
meanings'', particularly in languages of northern Europe and southeast Asia, and
extends the semantic map for modality to incorporate their findings.
12. ''Conflicting constraints on the interpretation of modal auxiliaries,'' by Ad
Foolen & Helen de Hoop, pp. 303-316, examines the interpretations of the Dutch
modal auxiliaries kunnen 'can' and moeten 'must' in a variety of sentential
contexts, and models the factors that lead to particular interpretations as
Optimality Theoretic constraints.
13. ''Modality and context dependency,'' by Fabrice Nauze, pp. 317-340 describes
several problems with the analysis of modality in terms of generalized
quantifiers, and uses the analysis of St'át'imcets modals as in paper 9 to show
that modality should not be analyzed as the interaction of a 'neutral' operator
with an external intensional context. A solution is sketched in the 'update
14. ''Verbal semantic shifts under negation, intensionality, and imperfectivity:
Russian genitive objects,'' by Barbara H. Partee & Vladimir Borschev, pp.
341-364, is a ''prolegomenon to a fuller study of shifts in semantics and in
fine-grained argument structure of verbs under negation and under the influence
of intensionality, modality, and imperfective aspect'' that examines ''the
relationships between negation and intensionality' on the one hand and between
''partitivity and imperfectivity'' on the other.
15. ''The Estonian partitive evidential: Some notes on the semantic parallels
between aspect and evidential categories,'' by Anne Tamm, pp. 365-402, examines
the use of partitive case on direct objects to express imperfect aspect and on
present participial forms to express evidentiality in Estonian, pointing out
that 'incompleteness' is part of the interpretation in both cases.
The book is well edited; the few typos and infelicities that I found are not
unduly distracting or confusing. The structure of the volume that the editors
outline in paper 1 is a useful guide to the flow of its contents, which proceeds
from an examination of tense and aspect, primarily the latter, to mood and
modality, again primarily the latter. It would have been helpful, however, if
the editors had addressed the distinction between mood and modality. They
implicitly treat realis and irrealis as modals, but paper 7 avoids the issue by
calling them 'markers'. In the literature, one can find them described both as
modals (Roberts 1990) and as moods (Deen & Hyams 2002). The discussion in paper
7 leads me to conclude that they are, if anything, moods.
Although tense, mood and aspect are treated as equal partners in the title,
tense is not the main focus of any of the papers in the book itself. Paper 2
deals with the interaction of present tense and perfective aspect in a variety
of languages, especially Russian, but the conclusion that the ''choice between
different OT approaches to model syntagmatic interaction between categories is a
matter of future research'' is disappointingly inconclusive. The paper notes in
passing that in languages that ''lack a [morphosyntactic] category of tense
altogether'', aspectual forms are used to render tense distinctions (pp. 21-22,
citing work on Maltese and Lango; see also Matthewson 2006 on St'át'imcets,
referred to in paper 9), but apparently the favor is not returned -- there are
no languages that lack a morphosyntactic category of aspect altogether and use
tense forms to render aspectual distinctions.
Paper 3 deals with a related problem in Ancient Greek having to do with the
interaction of tense and aspect, but its focus is on aspect, not tense. Paper 7
includes a short section on future tense in relation to the realis/irrealis mood
distinction, and brief mention of the 'legendary past'.
The analysis on perfective and imperfective aspect in papers 2 and 3 is part of
the study of 'outer aspectuality', or viewpoint. Paper 4 develops the theory of
'actionality' of Tatevosov (2002) for the combination of outer with 'inner'
(lexical or situational) aspectuality, noting that this combination is not
sufficient for a complete analysis of the aspectual domain, since
'quantificational aspect', including such notions as iterativity, is also
needed. The theory of actionality extends Vendler's (1967) classification of
eventualities into states, activities, achievements and accomplishments, of
which eleven may be called 'cross-linguistic actional classes' (p. 59) on the
basis of their recurrence in a large number of languages; these are listed in
Table 1, p. 60. The actional system of Adyghe consists of nine of these
cross-linguistic classes and no others (of which two are exemplified by only one
predicate each), and so ''seems to be rather straightforward'' (p. 65). The paper
concludes with a methodological caution about the use of temporal adverbials for
tests of aspectuality: ''it is justified only when there is independent evidence
that adverbials do not shift the lexically encoded actional meanings of
predicates as they do in Adyghe''. (p. 78)
Paper 5 probes the occurrence of two distinct aspectual interpretations of
accomplishment predicates cross-linguistically involving non-culmination:
failure and partial success. It is the most thorough of the papers in this
volume that explore semantic distinctions cross-linguistically; it concludes
with a call for ''a systematic cross-linguistic study of eventuality type'' (p.
126) to push the agenda forward. In the meantime, the first author has published
another paper on the topic of non-culminative interpretations of accomplishment
predicates (Tatevosov 2008).
Paper 6 examines the contributions of three verbal affixes in Burmese that are
related to independent verb stems (effectively 'light verbs') to the aspectual
and voice interpretations of sentences containing them. The analysis of the
contrasts between the marker related to 'go' and those related to 'come' and
'follow' is not convincing. For example, following an analysis of the come/go
contrast as involving an aspectual difference involving process, a pair of
examples ((72) and (73), p. 150) in which each marker is used with the same verb
(translated 'capture') are analyzed as manifesting a different contrast having
to do with the relative salience of the agent and patient. Also, the analysis of
the material in the brief section 4.2.5 ''Other uses of -θwà 'go''', pp. 151-152,
seems ad hoc.
Paper 7 is primarily concerned with the range of grammatical constructions in
which markers understood as manifesting the realis/irrealis contrast appear in
Yurakaré ('joint systems' in the terminology of Palmer 2001) and other languages
(including those with 'non-joint systems' such as Yimas). In Yurakaré, the
realis/irrealis contrast appears in the marking of same-subject status of
dependent clauses and on the repeated verb in the 'emphatic predicate'
construction; interestingly, the relevant markers are glossed SS 'same subject'
and NE 'non-experienced', a designation of evidentiality, rather than as
'realis' and 'irrealis'. The contexts in which the irrealis marker occurs in
Yurakaré include intentional, potential, desiderative, obligative, jussive,
imperative, prohibitive, future tense, and habitual. In other languages,
different contexts 'trigger' the use of irrealis, but in such a way that they
can be graphed as ''a continuum of distinct but interrelated aspects'' (p. 173; I
presume the word 'aspects' here is intended in its nontechnical sense) that can
also be represented by an implicational hierarchy (p. 176). However, the paper
notes three specific problem areas for which additional data or clarification of
existing analyses is needed: the interaction of irrealis with modality in Yimas,
the possibility of non-connectedness in the designation of irrealis habitual
aspect in certain dialects of Bininj Gun-Wok (covering past and future
habituality, but not present), and the scope of negation relative to the
irrealis marker in Caddo, Yurakaré and Central Pomo.
Paper 8 reaches the following elegant conclusion about indicative and
subjunctive mood in Romance (p. 201):
(i) In all Romance languages, indicative mood is selected for a proposition p
only if the instruction is given to consider an epistemic model ... where p is
(ii) In Romanian, (i) can be strengthened to if and only if; subjunctive mood is
selected if and only if otherwise.
(iii) In all other Romance languages, indicative mood is selected if no
information other than (i) is provided; otherwise subjunctive mood is.
The paper however only considers complement clauses, so does not directly answer
the question raised in paper 14 concerning the conditions for selecting
indicative vs. subjunctive in relative clauses as in the Spanish example 'María
busca a un profesor que enseña/enseñe griego' ''Mary is looking for a professor
who teaches Greek'' (p. 344), but perhaps the analysis can be extended to handle
Paper 9 offers ''a radical reanalysis of the St'át'imcets 'out of control'
circumfix 'ka- ... -a' as a circumstantial modal'' (p. 240), encompassing ability
(circumstantial possibility, or what the authors call the 'existential
circumstantial reading') and no-choice (circumstantial necessity, or what the
authors call the 'universal circumstantial reading'). It is the most thorough of
the papers in the volume that explore semantic contrasts in a single language in
depth. The analysis is radical only in the sense that it is offered as an
alternative that is very different from previous analyses of the same data.
Presenting the no-choice reading as the circumstantial necessity counterpart to
the ability reading as circumstantial possibility is compelling if for no other
reason than that the no-choice circumstantial necessity operator, which may be
represented as Nc, is interdefinable with the circumstantial possibility
operator Pc via negation in the usual way for modal operators; i.e. that Nc(p) =
~Pc~(p) and Pc(p) = ~Nc~(p). In English, which lacks a general-purpose no-choice
modal auxiliary ('have to', as in 'I have to sneeze' does the job, but only in
limited contexts), the no-choice operator can generally be expressed by 'unable
not', as in 'I am unable not to eat the whole box of chocolates', or more
idiomatically as 'I can't help eating the whole box of chocolates'. Going the
other way also works, but less idiomatically.
Paper 10 reanalyzes the semantic map of modality by treating the concepts
epistemic, deontic, and participant-internal as binary feature specifications
that can be combined to yield the following five distinct fully-specified modal
notions and an additional nine underspecified ones.
* [+propositional, -internal, -deontic] = epistemic
* [-propositional, -internal, -deontic] = non-deontic participant-external (what
paper 9 calls 'impersonal modality')
* [-propositional, +internal, -deontic] = non-deontic participant-internal (what
paper 9 calls personal or dispositional modality)
* [-propositional, +internal, +deontic] = deontic participant-internal
* [-propositional, -internal, +deontic] = deontic participant-external
The paper identifies the 14 classes defined by this feature system as linguistic
elements that are ''predicted to exist'' (p. 265). This is not quite the proper
interpretation; these classes (if correctly defined) may be said to constitute
the 'natural classes' of modal interpretations, analogous to the way in which a
phonological feature system characterizes the natural classes of segments,
without making the stronger claim that there are individual morphemes that
manifest each of these classes in the world's languages, including the class of
all possible modalities (analogous to the class of all phonological segments).
Much of the paper is taken up with an argument that the [+internal] possibility
modal verbs in Dutch behave syntactically like control predicates and the
[-internal] ones like raising predicates (cf. Lechner 2005 for a more general
claim to this effect, and Wurmbrand (1999), who argues that English, German and
Icelandic modal verbs are all raising predicates). The control structure of
[+internal] modal verbs makes them out to be two-place relations between
individuals and propositions, which conflicts with their modal semantics, in
which they must be treated as one-place operators on propositions. This can be
resolved by assuming that only the propositional complement of such verbs is
within the scope of the modal operator. For example, assuming that the English
deontic necessity modal expression 'The committee must nominate John',
understood either as ''The committee is obligated to nominate John'' or as ''The
only way for John to be nominated is for the committee to nominate him'', with
the former analyzed as containing a one-place [-internal] raising modal verb and
the latter as a two-place [+internal] control one, the deontic necessity
operator in the latter case would apply to its complement 'PRO nominate John'
only. But then, it would appear that nothing prevents us from analyzing the
two-place predicate 'certain' in 'John is certain that he is sick' as containing
an epistemic modal operator on its complement, and thus treating it as (I
presume) a [+internal] epistemic necessity modal (but not as a control
predicate, since 'John is certain to be sick' is understood only as containing a
[-internal] epistemic necessity predicate that undergoes raising).
Paper 11 also reanalyzes the semantic map of modality by considering modal uses
of verbs that derive from a non-modal acquisition meaning such as English 'get'
as in 'I get to watch TV tonight', in which 'get' is said to be understood as a
participant-external (= [-internal] in the notation of paper 10) deontic
possibility operator. Such uses are found throughout the languages of North
Europe and Southeast Asia, and extend to all non-epistemic (deontic and
circumstantial) possibility modal interpretations except that no such verb
expresses participant-internal modality only. (However, I don't know how to
interpret the entry for Faroese in Table 2, p. 286; no illustrative examples are
given for this language.) In all the examples illustrating 'acquisitive modals',
the verb is inflected for present tense or is uninflected, which I presume is no
accident, since in English, at least, 'I got to watch TV last night' does not
express deontic possibility. If it did, it would neither imply nor be implied by
'I watched TV last night', but in fact it implies the latter (cf. 'I had
permission to watch TV last night', which has the desired property for deontic
possibility). Finally, modal 'get' in English, being participant-external, is a
raising verb, including in sentences like 'I get John to help me', with
raising-to-object (or exceptional case marking), in which the external subject
is not in the scope of the modal, but must have a coreferent that is. (Note that
'I get John to help Mary' is not understood modally.)
Paper 12 analyzes the Dutch modal auxiliary verbs 'kunnen' and 'moeten' as each
having a default interpretation, participant-internal for the former and
participant-external for the latter, but non-default interpretations can arise
depending on the whether the activity of the main verb is controllable by its
subject, whether progressive aspect is present, and the person of the subject.
The choice of interpretation is modeled as a set of OT constraints operating on
sentences as input and interpretations as output. The paper argues for the
definition of modality in Narrog (2005): ''The expression of a state of affairs
is modalized if it is marked for being undetermined with respect to its factual
status ...''. (quoted on p. 311) On the face of it, this would appear to prevent
natural languages from expressing any of the modal systems such as T and S4,
including epistemic modality, in which N(p) implies p and p implies P(p), where
N is a necessity and P its corresponding possibility modal, a very curious
consequence, since '''necessarily true' is the notion usually used to introduce
modal theory itself'' (Koslow 1992: 267). The paper illustrates the claim using
examples and analysis from Kaufmann et al. (2006) according to which (a) 'it
must have rained overnight' does not entail (b) 'it rained overnight'. If
correct the modal must presumably be understood like a deontic modal in which
N(p), p and P(p) are logically independent. I find the general claim and the
arguments made for it implausible. Are there really circumstances in which (a)
is true and (b) false, and for the corollary in which (b) is true and (c) 'it
might have rained overnight' false? I don't think so.
Paper 13 examines standard analyses of modality involving generalized
quantification over possible worlds and finds them wanting for a variety of
reasons, including one due to the authors of paper 9 (paper 13 cites earlier
work of theirs). It concludes with ''a sketch of an update semantics analysis of
epistemic and deontic modality'' (p. 333) based on the author's PhD dissertation
(Nauze 2008), which also includes a comparable account for circumstantial
modality, involving in one way or another the notion of a to-do list.
Paper 14 opens by exploring the possibility that morphological marking of noun
phrases is related to degree of referentiality across languages, the central
case being the phenomenon of genitive case marking under negation in Russian. It
continues by considering the role of 'decreased referentiality' in the use of
partitive case in Finnish, and the Russian imperfective and genitive case, and
develops a 'family resemblance' model of intensionality and negation to account
for the commonalities of their morphological marking. This is an ambitious
program within the overall scope of this volume, particularly when extended to
inquire as to why certain distinctions are made across languages and not others,
and constraints on the variety of ways in which they are made.
Finally, paper 15 provides a detailed analysis of the interpretation of the
evidential partitive case in Estonian, ''which combines the two categories
[epistemic and evidential] in one morpheme''. (p. 366) The evidential partitive
attaches to nonfinite verbal forms, whereas the aspectual partitive (as also
discussed above in paper 14) attaches to direct objects, and the two are argued
to have parallel interpretations ''in terms of a comparison between the
expectation that the speaker holds about the event and the actual event'' (p.
367). The author provides three hypotheses about the semantics of the Estonian
partitive evidential, the indirect, the epistemic modality, and the partitive
(p. 375). The first takes it to have the semantics of reportative evidentials;
the second of epistemic modality, specifically to mark the speaker's ''degree of
confidence, certainty or belief in the proposition''; and the third of ''the
scalar semantics that is characteristic of the aspectual domain in Estonian and
remotely related to part-whole relationships''. I did not find in the subsequent
discussion a clear evaluation of these hypotheses; from the summary paragraph
(pp. 397-398), I get the impression that all three provide some insight into the
semantics of the construction.
In addition, it should be noted that not all indicators of the speaker's ''degree
of confidence, certainty or belief in the proposition'' are modal operators; 'be
certain' and 'be possible' are, but 'be likely' is not, contrary to a
consequence of a remark in paper 7 (p. 175). A necessity modal like 'be certain'
preserves implication no matter how many premises there are, i.e. if P1, ... Pn
==> Q, then N(P1), ..., N(Pn) ==> N(Q), and a possibility modal like 'be
possible' does so under the dual implication relation. (Koslow 1992) However,
'be likely' preserves implication only if there is one premise, e.g. (letting L
stand for 'be likely'), if P ==> Q, then L(P) ==> L(Q), but in general if P1,
..., Pn ==> Q, then L(P1), ..., L(Pn) =/=> L(Q). For example, let P be 'John
bought apples' and Q be 'John bought fruit'. Then both P ==> Q and L(P) ==>
L(Q), since in every model in which P is true, Q is true, and similarly for L(P)
and L(Q). Next, let P1 = 'John bought apples', P2 = 'Mary bought oranges', and Q
= P1 & P2. Then in every model in which P1 and P2 are true, Q is true, but there
is a model in which L(P1) and L(P2) are true, but L(Q) is false.
Overall, the book definitely achieves both the goal (quoted at the beginning of
this review) ''to give greater prominence to the semantic richness of tense,
aspect, modality, and their interactions, in the languages of the world'' and the
hope that it ''contributes to the emergent field of semantic typology''.
van der Auwera, Johan & Vladimir Plungian. 1998. Modality's semantic map.
Linguistic Typology 2:79-124.
Bary, Corien. 2009. Aspect in Ancient Greek: A semantic analysis of the Aorist
and Imperfective. PhD dissertation, Radboud University Nijmegen.
http://repository.ubn.ru.nl/bitstream/2066/74432/1/74432.pdf (accessed 2010-04-05).
Bender, Emily & D. Terence Langendoen. 2010. Computational linguistics in
support of linguistic Theory. Linguistic Issues in Language Technology 3(2):1-31.
Bybee, Joan, Revere Perkins & William Pagliuca. 1994. The Evolution of Grammar:
Tense, Aspect and Modality in the Languages of the World. Chicago: University of
Deen, Kamil Ud & Nina Hyams. 2002. The form and interpretation of finite and
non-finite verbs in Swahili. Proceedings of the Boston University Conference on
Child Language Development 26, ed. by Barbara Skarabela, Sarah Fish, and Anna
H.-J. Do, pp. 130-141. Somerville, MA: Cascadilla Press.
Kaufmann, Stephan, Cleo Condorvadi & Valentina Harizanov. 2006. Formal
approaches to modality. The Expression of Modality, ed. by William Frawley, pp.
71-106. Berlin: de Gruyter.
Koslow, Arnold. 1992. A Structuralist Theory of Logic. Cambridge: Cambridge
Langendoen, D. Terence. 2010. Just how big are natural languages? Recursion and
Human Languages, ed. by Harry van der Hulst, pp. 139-146. Berlin: de Gruyter.
Lechner, Winfried. 2005. Semantic effects of head movement. Paper given at XXXI
Incontro di Grammatica Generativa, Università degli Studi Roma Tre. Handout
available at http://vivaldi.sfs.nphil.uni-tuebingen.de/~nnsle01/IGG%202005.pdf
The Leipzig glossing rules: Conventions for interlinear morpheme-by-morpheme
glosses. 2008. Max Planck Institute for Evolutionary Anthropology Department of
Linguistics. http://www.eva.mpg.de/lingua/resources/glossing-rules.php (accessed
Matthewson, Lisa. 2006. Temporal semantics in a supposedly tenseless language.
Linguistics and Philosophy 29:673-713.
Mithun, Marianne. 1995. On the relativity of irreality. Modality in Grammar and
Discourse, ed. by Joan Bybee & Suzanne Fleischman, 367-388. Amsterdam: John
Narrog, Heiko. 2005. On defining modality again. Language Sciences 27.165-192.
Nauze, Fabrice. 2008. Modality in Typological Perspective. PhD dissertation,
University of Amsterdam. http://dare.uva.nl/document/112927?fid=112927 (accessed
Palmer, Frank R. 2001. Mood and Modality, 2nd ed. Cambridge: Cambridge
Roberts, John R. 1990. Modality in Amele and other Papuan languages. Journal of
Rothstein, Susan. 2004. Structuring Events: A Study in the Semantics of Lexical
Aspect. Oxford: Blackwell.
de Swart, Henriëtte. 1998. Aspect shift and coercion. Natural Language and
Linguistic Theory 16.347-385.
Tatevosov, Sergei. 2002. The parameter of actuality. Linguistic Typology 6.317-401.
Tatevosov, Sergei. 2008. Subevental structure and non-culmination. Empirical
Issues in Syntax and Semantics 7, ed. by Olivier Bonami & Patricia Cabredo
Hofherr, pp. 393-422. http://www.cssp.cnrs.fr/eiss7/tatevosov-eiss7.pdf
Wurmbrand, Susi. 1999. Modal verbs must be raising verbs. Proceedings of The
West Coast Conference on Formal Linguistics 18, ed. by Sonya Bird, Andrew
Carnie, Jason Haugen & Peter Norquest, pp. 599-612.
ABOUT THE REVIEWER
| ABOUT THE REVIEWER:
Terry Langendoen is Professor Emeritus of Linguistics at the University of
Arizona and Expert in the Robust Intelligence Program at the National
Science Foundation. He received his PhD in linguistics in 1964, and
previously held positions in the Department of Linguistics at The Ohio
State University, the PhD Program in Linguistics at the City University of
New York Graduate Center and the Department of English at Brooklyn College.
He also served as a Program Director in Linguistics at NSF from 2006 to
2008. His most recent publications are Langendoen (2010) and Bender &
Langendoen (2010), listed above.