REVIEWS
293
Groenn, A.: 2004, The Semantics and Pragmatics of the Russian Factual Imperfective (= Series of Dissertations Submitted to the Faculty of Arts), Oslo,
University of Oslo, 288 pp.
Atle Groenn’s book deals with Russian aspectology, in particular with Imperfective (Ipf) forms with factual meaning. Aspectology is one of the most
controversial areas in Slavic linguistics and the factual Ipf, though much discussed, is still poorly understood. The author has set an ambitious goal – to
combine the data concerning the Russian factual Ipf with the methods of formal
semantics, such as Hans Kamp’s Discourse Representation Theory. As a result,
we have a fundamental piece of research, which is of great use both for specialists
in the compositional semantics of aspect and for those who are interested in the
grammatical side of the question.
There are two main problems with the factual Ipf in Russian.
The first problem is how to differentiate semantically the factual Ipf from the
Perfective (Pf). In other words, what is the reason that makes the speaker say, for
example, Аня мыла (Ipf) пол, instead of Аня вымыла (Pf) пол.
The second problem is how to unite the factual meaning of the Ipf with its other
meanings, such as the Progressive and Habitualis.
Both problems are challenging, and the scientific narrative of the book develops
as a detective story. As is often the case, not all of the expectations are fully
realized, but the attentive reader will undoubtedly be rewarded for his/her labour.
It is generally accepted that the aspectual forms of a Russian verb possess a
variety of particular meanings – presumably engendered or at least motivated by
the context. It would have been impossible to embrace all of them at one go, and
the author ingeniously limited his task. A contemporary approach distinguishes
two kinds of context for aspectual forms: lexical context – on the one hand, and
the context of utterance, which creates the so called view point aspect, – on the
other. It is this second kind of aspect that is the main object of analysis in Groenn’s
work. In fact, lexical problems connected with aspect are reduced to a minimum:
only telic verbs are taken into consideration. So many issues connected, for
example, with stative or momentary verbs are put aside, and what we have is
a very natural class of occurrences when the meaning of the Ipf includes the
resultative component (e. g., подвозил implicates подвез); these Ipfs are called
factual.
Chapter 1 is devoted to telicity. In fact, it is rightly claimed that the factual
meaning appears only in the context of telic verbs. But what is telicity? Ordinary
syntactic tests for (a)telicity are examined, i. e. co-occurrence with за (Engl. in)
and в течение (Engl. for):
Russian Linguistics (2006): 293 – 302
DOI 10.1007/s11185-006-0703-9
294
(7)
(8)
REVIEWS
Ваня прочитал (Pf) книгу за два часа / *в течение двух часов
(telic verb);
Ваня играл (Ipf) в футбол *за два часа / в течение двух часов
(atelic verb).
In Russian these tests do not yield any classification of verbs, because one and
the same verb has different combinability in different aspects – for example, the
same читать in the Ipf, as in (9), behaves like играть:
(9)
Ваня читал [Ipf] книгу *за два часа / в течение двух часов.
Atelicity is grounded in the lexical semantics of a verb. Such verbs as висеть
or искать are atelic – and they will never acquire the factual (i. e. the resultative)
meaning. On the other hand, telicity may be context dependent: пересекать
<канал>, as a verb of movement, is always telic, because пересекать, has no
absolutive use; but читать may be both telic and atelic. I should say that in
the Russian tradition telicity is conceived as a property of a lexeme and not of a
perfective verb; пересекать, for example, is a verb of action; it is telic, and it has
two forms, one of them presents the action as an event, the other – as a process,
but the verb is said to be telic in both uses.
The compositional telicity of читать is seen in examples (a)–(c):
(a)
(b)
(c)
читать – activity, atelic;
читать книгу – action, telic;
читать книги <indefinite books> – activity, atelic.
The term telic, in the Western tradition, embraces both verbs of accomplishment (such as купить – покупать) and momentary verbs (such as проиграть
– проигрывать). But these verbs behave differently with respect to factuality:
(a)
Кто в прошлый раз покупал билеты, Ваня? – (b) *Кто в
прошлый раз проигрывал, Ваня?
This is why the term telic should not be translated into Russian as предельный. For translation I would rather use its synonym bounded, which can
be rendered in Russian as терминативный.
Thus, only those verbs may hope to get the meaning of the factual Ipf that enter
an aspectual pair (it has not been examined whether such statives as понимать
may acquire the factual meaning).
Later on in Chapter 1 the author enumerates the properties of the factual Ipf:
–
–
–
–
reference to complete events;
telic predicates;
singularity of event;
location in the past.
REVIEWS
295
With this list of properties at hand the author is able to classify each Ipf as being
either factual or not.
On p. 34 it is stated that the meaning of the Ipf is completely determined
by the context: this observation goes at least to Forsyth (1970). Multifarious
relationships between the meaning of the Ipf and its context are meticulously
studied.
The basic – and crucial – observation of the author is that the factual Ipf
can be either existential or presuppositional, see section 1.5.2. Up till now the
existential Ipf was given prominent attention, while the presuppositional one
(mostly agentive) rather passed unnoticed. Atle Groenn fills the gap, and the
presuppositional Ipf becomes the main object of his analysis. Presuppositions
are treated in the frame of Discourse Representation Theory and explicated as
anaphoric relation between events. This is an important contribution to Slavic
aspectology, where the notion of presupposition is widely used but lacks any strict
definition.
Existential Ipfs are divided into three groups.
1. Experiential Ipfs:
(26) Ты что-нибудь делал , чтобы ему помешать?
2. Cyclic Ipfs:
(29) Ты сегодня покупал газету?
3. Bydirectional Ipfs (presupposing an event of the opposite direction):
(31) Ты открывал окно?
As for example (25) (Я и раньше слышал о таких случаях), to my
mind it would be better to treat such verbs as слышать (analyzed at length
by Ju. Apresjan (Апресян 1986)) separately. In fact, they have factual but no
synchronous progressive meaning.
If we agree that only telic verbs can acquire the factual Ipf then the meaning of
a given verb always consists of the two components, source state and target state.
So the question of communicative status of these components arises. For Pfs (at
least for conatives) these may be presupposition and assertion; a reasonable pair
for the Ipf is <assertion, implicature>.
Using example (54) (Я не вызвал такси), the author denies that there is a
presupposition in the semantics of the Pf: no presupposition, merely a pragmatic
phenomenon. Somewhat earlier, considering example (39), Аня убирала квартиру, the author comes to the conclusion that it makes sense to relativize the
entailments to a particular reading of the Ipf. Thus, about the factual Ipf in (39)
it can be said that it ‘entails’ Pf. In other words, the ‘illegal’ resultative component in the semantics of the Ipf obtains the status of entailment. This comes as a
consequence if we accept that factual Ipfs assert or presuppose a complete event.
296
REVIEWS
Chapter 2 introduces or defines some important notions, indispensable in the
semantic analysis of tense and aspect. One of them is asserted time (cf. Klein
1995): the temporal part of the situation, which is made visible by the camera, is
asserted by the speaker.
Another valuable notion is that of viewpoint aspect. The epithet is due to
Carlota Smith, and there is an important idea behind it, for in order to understand
the grammatical meaning of aspect it is necessary to destinguish it from all
enrichments and augmentations generated by the semantic class and context of
the verb.
Evaluation time is approximately what was called the point of reference by
Reichenbach: a temporal perspective point.
Only in Chapter 3 is the reader given the opportunity to understand why factual
Ipfs are called this way: the puzzling label is deciphered. The author rightly
confesses that this label is a pure tribute to tradition (and, strictly speaking,
a mistake). Appealing to Arutjunova (Арутюнова 1976) and the verb помнить,
which is the best arena for the fact-event opposition, the author acknowledges
that factual Ipfs denote events, not facts. Indefiniteness of time is not a necessary
attribute of the factual Ipf. But the very ability to be localized in time is
characteristic of events, not facts.
Different parameters of contexts are discussed which contribute to “factual”
meaning: temporal adverbials, such as когда-нибудь, раньше; singularity
markers. Among them the particle уже is the most astonishing – (a) is ambiguous;
in (b) the meaning of the Ipf is unambiguously factive:
(a) Я убирал квартиру; (b) Я уже убирал квартиру.
The exact etiology of this crucial influence of уже upon the reading of the Ipf
is still to be discovered. In example (109) –
(109) A. Ты убирал квартиру? B. Убирал, но не убрал. –
interlocutor A suffers a communicative failure: B uses the Ipf in a different
meaning.
At the end of Chapter 3 the author returns to the semantic definition of factual
Ipf in (25), adding a new point:
The factual Ipf is not embedded under negation.
This sounds a bit strange – in fact, the context of negation is constitutional for
presuppositions, and factual Ipfs are, at least partially, presuppositional.
The factual Ipf should not be confused with the durative. The durative resembles the factual Ipf (Past Tense) in that both are retrospective. But here the
similarity ends.
REVIEWS
297
Chapter 4 gives a short overview of the history and state of the art of the
factual Ipf. The exposition is concise, exact and convincing. This is a brilliant
introduction into the Slavistic literature on the subject. I shall touch upon only
one point – that of the experiential Perfect.
Some uses of the factual Ipf conform to the view that it expresses the
experiential Perfect, for example – a dialogue, widely discussed:
– Хотите борща? – Спасибо, я обедал.
According to grammars, the Present Perfect in English expresses an action
the result of which is valid at the moment of speech (constituting somebody’s
experience, hence the term experiential Perfect). At first sight, this is what we
have in the example above. But what is, then, the difference between the factual
Ipf and the Pf, which expresses the idea that the resulting state (target state) of the
event is valid at the moment of speech? Now, the difference is clear: the Russian
Perfective aspect can be used to express the fact that the state contiguous to the
event still continues at the present moment. While with the experiential Perfect the
state in question is some consequence of the event in the past, but not its direct
continuation.
The author rightly concludes that the distinction between the resultative and the
experiential perfect in languages like English is partly grammaticalised in Russian
through the opposition Pf–Ipf (p. 243).
Chapter 5 is of special importance. Viewpoint aspect is conceived as a relation
between assertion time and evaluation time. One of the principal aims of the
author in this chapter is to substantiate the claim that the factual Ipf is restricted
to the Past Tense. The explanation is straightforward: “if the speaker either
presupposes or asserts the existence of a complete event the event is most likely
to be located in the past”. One way or an other, it is true that the analysis
of the factual Ipf in terms of the presupposition or assertion “of a complete
event” does not transpose to the future, see example (195) with the prospective
Future:
(195) A. Говорят, этот собор будут восстанавливать.
B. Кто его будет восстанавливать?
Only the beginning of the future action is meant both by A and by B. Thus,
Past and Future are not identical: The Past is oriented to the end, the future is
centered on the beginning. Perhaps, they can said to be symmetrical. In fact, they
are opposed to one another.
It is emphasized that in the Future Tense there is no result, no annulment of
result, and bidirectional verbs do not acquire their characteristic meaning: there is
no presupposition of an event with the opposite direction, cf.
298
REVIEWS
(a) Машина выезжала из гаража.
(b) Машина будет выезжать из гаража.
The question of why this is so was not addressed in the literature – or at best
taken for granted. For me it is one of the most interesting places of the author’s
work.
Let us go on with the notion of evaluation time in connection with viewpoint
aspect. Example (204) deserves special attention in this respect. In the Russian
aspectological literature evaluation time is interpreted in the frame of secondary
time deixis. The figure of Observer introduced by Apresjan (Апресян 1986) for
spatial deixis was used in Paducheva (Падучева 1996) to interpret Reichenbach’s
point of reference: the point of reference (and, consequently, the evaluation time)
can be understood as the present tense of the Observer. Now, if there is a thematic
adverbial of place in the sentence – as is the case in (204) – then the Observer
should localize himself in that place and, consequently, in that time:
(204) В этом домике Пушкин когда-то читал друзьям ‘‘Бориса
Годунова’’.
If this is so then the Ipf читал in (204) is not necessarily a factual Ipf. It
may also denote a process developing before the eyes of a synchronous observer.
In fact, the adverbial of place в этом домике, which imposes synchronicity,
overrides the adverbial of time когда-то, which otherwise would have invoked
a retrospective point of view.
Let us consider further temporal adverbials and their interaction with tense
and aspect. About such adverbs as однажды, когда-то it is rightly said that
their semantic contribution “must be treated as vacuous or redundant”, but they
are necessary “in order to maintain the appealing picture of viewpoint aspects as
relations between event time and assertion time”. To my mind, what is important
about когда-то is its evaluation time.
Combinability of the Ipf with temporal adverbials was studied earlier. It is
known from M. Glovinskaja’s book of 1982 (Гловинская 1982) that Ipfs tend
to be temporally indeterminate and, in a sense, to avoid time modifiers. Still there
are some adverbs – such as когда-то, раньше, сегодня, вчера – which do
combine with the Ipf. It was discovered recently that давно, which semantically
corresponds fo the general idea of the factual Ipf, cannot occupy the thematic
position in a sentence. One may think, on the basis of the examples adduced, that
когда-то is a thematic allolex (in the sense of A. Wierzbicka) of давно.
Of special importance in Chapter 5 are examples (214)–(216), where evaluation
time for a factual Ipf verb is determined by some event in its context.
Past Perfect uses of the factual Ipf are extensively studied, and this is an
important achievement. It is known that speakers of European languages sharply
REVIEWS
299
feel the lack of the Past Perfect in Russian; in fact, it creates difficulties in
understanding Russian texts. Thus, this part of the author’s study is of direct
practical importance. About the evaluation time parameter it is rightly said that
it can be either deictic or anaphoric.
Past Perfect uses of the factual Ipf give rise to one of the major theoretical
issues connected with the topic – namely the opposition of narrative and discourse
registers of interpretation of tense forms. How can we reconcile a Past Perfect
use of a factual Ipf form with its preferably discourse register of interpretation,
that is, with the fact that it is rarely used in narrative contexts? Retrospection is
the common denominator for the two registers: Past Perfect interpretation arises
when evaluation time gets shifted to some moment in the past.
Chapter 6 is highly technical and I shall comment on it only briefly. It concerns
mainly the presuppositional Ipf (the complete event is presupposed) and presents
a theory of presuppositions embedded in the dynamic framework of DRT.
I would like to dwell upon example (253) (from Mehlig (1997): – Надо
выключить свет. – *Я его уже выключал). Mehlig’s point is that the
sentence is bad with a complete event reading. He finds this surprising since he
expects that the “presuppositional” выключал should be anaphoric with respect
to выключить. The author, unlike Mehlig, can explain why выключал cannot
be anaphoric and presuppositional here. The idea is that a complete event of the
type выключить is not asserted, since it is embedded under “надо”. This means
that there is no available event antecedent for выключал.
Example (253) attracts our attention to the special interaction between the
factual Ipf and уже, which was noticed as early as in Forsyth (1970), see also
Paducheva (Падучева 1996, 60). The fact is that the asterisk in B’s reaction is
not correct: the sentence is fully grammatical, it only means that the attempt was
made but the agent failed to achieve the desired result.
It was noticed earlier that while уже with the Pf denotes a unique action, уже
with the Ipf introduces the implication that the action can be accomplished several
times. Now, the author gives an explanation for this observation: with the Pf the
target state, the result, is before our eyes, while with the Ipf the result disappears,
hence the possibility of repetition of the action (p. 274).
In Chapter 7, sections 7.2.3 and 7.2.4 devoted to bidirectional verbs are of
primary importance.
The existential Ipf is compatible with reversible target state predicates, the
so-called bidirectional verbs. The irrelevance of the target state is clearly an
implicature arising from the competition of the Ipf with the Pf. The annulment
of the target state is not excluded then. And this is claimed to be the semantic
basis for the bidirectional reading.
Not surprisingly, the bidirectional reading combines with на-adverbials, see
examples (302)–(304), such as приезжал на два дня. (Defining the class
300
REVIEWS
of verbs co-occurring with на-adverbials – “обстоятельство намеченного
срока” – is an interesting goal in itself.)
The event responsible for canceling the target state may have the same Agent
as that of the initial event. But this is not necessarily the case: the result of the first
event can be completely or non-completely converse.
Semantic classes of bidirectional verbs are presented, and an insightful analysis
of bidirectional verbs really constitutes the culminating point of the book. In some
respects, bidirectional verbs are the crucial argument in favor of the whole edifice
and the theoretical apparatus employed in the analysis of factual Ipfs.
One theoretical point should be touched upon. Contemporary methods and
instruments of semantic analysis put forward the idea of a connection between
the different contextual meanings both of a word and a grammeme. The factual
Ipf is undoubtedly a derivative use of the Russian Ipf. No matter how deeply
we become immersed in the meaning of the factual Ipf, we are obliged not to
forget about other meanings of the Ipf and to think about how factual meaning is
connected with the basic meaning of the Ipf, which is displayed in the contexts
more natural for the Imperfective, namely, in synchronous contexts. In this respect
something still remains for the future.
The author tends to reduce the distinction between processual and factual
readings of the Ipf to the size of the time interval: “A small assertion time
triggers a processual reading, while a factual Ipf arises with a big assertion time”
(p. 274). I would rather say that a big assertion time generates a retrospective (or
prospective) viewpoint on the situation.
Not casting any doubts on the analysis proposed I would like to suggest some
ways of developing it. The meanings of aspects are presented by the author in
temporal terms. But grammatically, tense is the time of the speaker, the speaker
being the orientation point for deictic reference. In the non-deictic register of
interpretation, for example, in narrative, there are substitutes of the speaker (such
as the narrator and the characters).
There are reasons to suppose that aspect, though also being a deictic category,
introduces another figure into the context of discourse interpretation – that of
an Observer. The temporal position of the Observer, his evaluation time, may
coincide with that of the speaker. But the Observer has at his disposal some
possibilities of movement in time not licensed for the speaker. The Observer is
the Origo for secondary deixis (according to Apresjan), with different substitutes
in a narrative text.
The author arrives at the conclusion that the Ipf conveys the pragmatic
implicature that the target state has been canceled. In fact, this is the basis for
the proposed analysis of bidirectional reading. Still there are contexts where I
would rather say that the implicature about the target state having been reached
REVIEWS
301
did not even arise: in fact, the border between factual and processual reading of
the Ipf is in many cases very subtle. Example (204)
(204) В этом домике Пушкин когда-то читал друзьям “Бориса
Годунова”.
is crucial in this respect: if we do not know the continuation we cannot decide
between the two readings, compare (204 ), with an unambiguously factual Ipf:
(204 ) Я читал “Бориса Годунова”.
The factual Ipf has two constitutive components – components which make it
different from the basic meaning of the Ipf, namely, the progressive:
I. the target state is reached (and hence the meaning of complete event);
II. the target state is canceled or is not relevant at the evaluation time (and
hence the difference from the Pf, which also denotes a complete event).
How are we to explain these two meaning components of the factual Ipf starting
from the semantics of Ipf in its basic progressive meaning: where do they come
from?
Component II (which the author ingeniously uses in order to explain the
bidirectional readings of Ipfs) is rightfully identified by the author as a pragmatic
implicature engendered by competition with the Pf.
But component I, also non-inherent for the nature of the Ipf, can also be
contextually caused. It might be explained by a combined influence of the two
factors: a) retrospection, which is due to the Past Tense; and b) telicity in the
lexical semantics of the verb. What is definitely inherent in the Ipf and must be
attributed to its own properties is the ability to denote a situation in development
not only in synchronous perspective but also in retrospect. And this is what
differentiates the Slavic Ipf aspect from, for instance, the Progressive in English.
More than that, the readiness of non-synchronous use distinguishes Russian from
other Slavic languages as well, for example, from Chech.
Thus, both components are contextually explained and, as to their genesis,
neither is assertive. The difference between them is that component II is not
essential in the semantics of the factual Ipf: the meaning of an Ipf can be factual
independent of whether we are able to find component II in its semantics or not;
while without component I we will not call the Ipf in question factual. It is in
this sense that component I may be ascribed the status of semantic entailment (it
follows from the meaning of the Ipf being identified as factual), while component
II has a lower status.
According to W. Klein (1995), Past Tense cannot be ascribed the component
‘at present situation P doesn’t hold’ – it may go on (cf. Я Вас любил).
Thus, component I arises as an implicature – at least in the context of telic-
302
REVIEWS
accomplishment verbs. The result is rather present in the context (of existing
houses, novels, etc.) than in the meaning of the aspectual form. (The same is
true of retrospection, however.)
One important consequence of the analysis of bidirectional verbs is that they do
not display the same effect in the future, cf. Я брал эту книгу в библиотеке
(= ‘and returned’) and Я буду брать эту книгу в библиотеке, which can
be interpreted, if at all, only in the iterative meaning. In general, approaches to the
analysis of the future tense must change after the Groenn’s research.
One feature of the book that causes real delight is its examples. To begin with,
they are mostly new and, which is even more important, one hundred percent
Russian. Many examples are taken from the Internet, but the Internet is not a
panacea: texts used in the Internet are contemporary but not necessarily standard
Russian. Atle Groenn unmistakably finds the golden mean. The examples are
highly stimulating, sometimes even outside the point they are supposed to
demonstrate (as, e. g., (258)). The only example that sounds odd is (224) (from
Schoorlemmer, who is notorious for her poor examples):
(224) Когда я вернулся, бандиты убили Ваню.
Atle Groenn’s work is one of the most interesting works on Russian aspectology I have come across in recent years. Slavic aspectology is a much studied field,
yet the author definitely managed to make an original contribution.
REFERENCES
Апресян, Ю. Д.: 1986, ‘Дейксис в лексике и грамматике и наивная модель мира’,
Семиотика и информатика, вып. 28, 5–33.
Арутюнова, Н. Д.: 1976, Предложение и его смысл: логико-семантические проблемы,
Москва.
Гловинская, М. Я.: 1982, Семантические типы видовых противопоставлений русского
глагола, Москва.
Падучева, Е. В.: 1996, Семантические исследования: Семантика времени и вида в
русском языке. Семантика нарратива, Москва.
Forsyth, J.: 1970, A Grammar of Aspect: Usage and Meaning in the Russian Verb, Cambridge
University Press.
Klein, W.: 1995, ‘A time-relational analysis of Russian aspect’, Language 71, 669–695.
Mehlig, H. R.: 1997, ‘Некоторые замечания по поводу описания категории вида в русском
языке’, Russian Linguistics 21, 177–193.
Всероссийский институт
научной и технической
информации РАН (ВИНИТИ)
paducheva@mtu-net.ru
ЕЛЕНА ВИКТОРОВНА
ПАДУЧЕВА