Humboldt-Universität zu Berlin
The paper deals with the use of bare nouns in a non-coordination-based context. In the first part,
the communicational function of the bare noun construction is analysed as part of a more
general use of bare nouns for labelling evoked semantic frames. The specific subset of bare
nouns depends on one or more propositions in the left context and provides a speculative
categorisation that is usually associated with a negative attitude. Although the negative attitude
is not generally confirmed, the bare nouns serve to frame an event in a way that renders it
unimportant, illogical, untrue, or similar. A very common extension of this construction is the
use of the bare nouns slučajnost’ and sovpadenie, which insinuate a causal connection between
intuitively unrelated events by employing the function of the bare noun construction. In the
second part, the status of the bare noun construction as potential ellipsis is discussed. I conclude
that a non-propositional account as part of a discourse or action pattern (Handlungsmuster) is
the most appropriate explanation.
Sovpadenie? Frame manipulation with bare
Ellipsis is a topic in linguistics that has received much interest. This study is about a subtype of
ellipsis, which includes only bare nouns without any syntactic antecedent. The missing syntactic
embedding renders this type of ellipsis to be of the situational type. In contrast to syntactic
ellipses, situational ellipsis is said to be interpretable only with regards to the situation, context
knowledge or world knowledge (Cheng 2009, 143; Heine 2011, 58; Knobloch 2013, 28;
Rickheit and Sichelschmidt 2013, 161). As such, bare nouns haven’t received as much interest.
This seems to be connected to the widespread – though not uncontested – interpretation that
ellipsis represents a friction between the signifying and the signified (Дудникова 2013, 186).
The interpretation holds that ellipses have the same meaning as equivalent full sentences but
without the form of the latter. Research on ellipsis thus concentrates to a significant extent on
the question whether and how the representation of the meaning can be modelled in the case of
ellipsis. Some authors assume that there must be a representation of the missing syntactic
structure somewhere in the production process. Others hold that this is not necessary (Imo, 139).
However, the very assumption that ellipsis shows an incomplete structure has to be proven
empirically, which in turn presupposes that some missing elements must be intuitively be there
(Knobloch 2013, 31). This is rather difficult to show for bare nouns without antecedent but
much more convincing for phenomena like VP ellipsis, Gapping, Pseudo-Gapping, N’ ellipsis,
Right Nose Raising, Sluicing and the like – mostly adjacency and coordination ellipses
(Knobloch 2013, 31). This strict canon of ellipses the literature focuses on is, however, a stark
simplification, as there are many other forms of “incomplete” sentences which are difficult to
explain with traditional syntactic approaches (Heine 2011, 60). Thus, following the literature,
the bare noun ellipsis may also be identified as fragments or minimal sentences. In the
following text, I will use the term “bare noun construction” as a neutral term. This, however,
does not imply that the bare nouns correspond to a construction in the sense of Construction
In the following, I first introduce the bare noun construction and some of its core features. Then,
I analyse the function of the bare noun based on a corpus study in the Russian National Corpus
and show one example of extending the construction. In the last part, I discuss the status of the
construction as ellipsis and conclude in section 5, that a non-propositional approach is a more
All examples have been extracted from the Russian National Corpus on 08.10.2018 with the
search term S bques first. Saving a random number of the query results gave a number of
3028 examples. I deleted all examples without any or too short left context, all occurrences with
antecedent as well as answers on questions like čto 'what', kto 'who', gde 'where' or kogda
'when'. Among the result where also names, question words or verb forms, which have been
deleted. The resulting number of 486 examples has been coded for different variables. Since
bare nouns without syntactic antecedent occur in other constructions, too (see e.g. examples (22)
and (23) below), the resulting number of examples for which the analysis holds, amount to only
The phenomenon under investigation displays a noun in nominative or uninflected form. The
bare noun is followed by a question mark, which in the case of transcribed spoken language is
likely to correspond to a rising intonation. The construction can be found in written as well as in
spoken language, although the metadata of the RNC do not allow to clearly differentiate
between the two, since many instances which are taken from written sources are indeed
embedded in representations of spoken language.
The bare nouns I am investigating here, have several structural (or textual) features that set them
apart from other similar constructions. Crucially for the following analysis, the bare noun
obligatorily has a left context. Different from “real” ellipses, this context contains no syntactic
antecedent, question or similar elements to which the bare noun would refer. The context
usually contains assertions, not questions or imperatives. Neither does the left context contain
syntactic structures which would explain the use of the bare noun. Compare:
(1) Сердюченко с нескрываемым возмущением пишет: «по Е. Берковичу, не все
фашисты плохие!». Парадокс?
‘Serdjučenko writes with undisguised indignation: “According to E. Berkovič, not all
fascists are bad!”. Paradox?’
(Людмила Палисад. Попытка трицания отрицания, или Опыт антирецензии. 2004)
(2) Тот же консилиум находил его "всегдашним сумасшедшим" ― так называли
‘The very same concilium found him to be an “constant looney” – this is how they
called chronists. Error?’ (Юрий Давыдов. Синие тюльпаны. 1988-1989)
The bare nouns are often followed by an answer, which is also the case for other bare noun
questions. However, in distinction to the “coffee-construction” (Heine 2011; Wasserscheidt in
print), the answer is given by the same writer/speaker who poses the elliptical question. In the
coffee-construction and other constructions that are situated in non-linguistic actions, it is the
interlocutor who provides the answer. Thus, the bare nouns under investigation are not part of a
dialogical construction. Compare example (3) and examples (4-6):
(3) Чай? ― Ничего не буду… ― Негатив умел отказываться так, что больше не
‘Tea? – I want nothing … ― Negativ could refuse in a way that he wouldn’t be offered
anything again.’ (Захар Прилепин. Санькя. 2006)
(4) А данный законопрект как раз и направлен на то, чтобы детей не помещать в
приют, оставлять в семье, и контролировать семью. Вмешательство? Да.
‘The given draft law exactly aims at not hosting children at a caretaker, leaving them in
the families and controlling the families. Intervention? Yes.’
(Форум: В жижечке очередная истерия по поводу ЮЮ... Обсуждение. 2012)
(5) Но ежели специалист захочет дать рекомендацию, то он обязан представиться «по
всей форме». Нонсенс? Безусловно.’
‘But when an expert wants to give recommendation, he is obliged to himself introduce
“in due form”. Nonsense? Absolutely.’
(Красота, здоровье, отдых: Медицина и здоровье (форум). 2005)
(6) Что, если взять себе за железное правило: ни при каких обстоятельствах не
предлагать денег людям в милицейской форме? Утопия? Возможно.
‘What if one introduces the golden rule: Don’t offer money to people in police uniform
under no circumstances. Utopia? Maybe.’
(Анатолий Кучерена. Не надо, друзья, бояться... (2003)
However, the bare nouns under investigation do not strictly require a right context. This sets
them apart from yet another type of bare noun construction (see examples 23 and 24). Thus, the
construction can end with the question mark (or rising intonation).
3.1 The function of the bare noun
The bare nouns under investigation functionally seem to categorise a given predication in some
way. This makes them similar to other bare nouns and to nouns in general, which also
categorise. Compare the following examples:
(7) Я маленькая собачка женского пола. ― Сучка? ― весело сказала Надя. ― Вот-
‘I am a little dog of female sex. ― A bitch? ― said Nadja cheerfully. ―That’s it,
exactly.’ (Василий Гроссман. Жизнь и судьба, часть 3. 1960)
(8) Двое ловко сбили «напарников» с ног, а третий бросился вниз, к Вадиму.
Бандиты?! Конкурирующие наркоторговцы?!
‘Two neatly rushed the „mates“ off their feet, the third one rushed down, to Vadim.
Gangsters? Competing drug dealers?!’
(Даниил Корецкий. Менты не ангелы, но… 2011)
These bare nouns refer to one or several preceding assertions and label them. The predications
can have a form similar to an enumeration of relevant features of a category, as in example (7).
They often also refer to an action and the bare noun is a suggestion who could be the agent of
that action (8). The actions thus represent a typical action for the suggested agent. Those kinds
of bare nouns refer to entities in the real world which have a more or less concrete reality and a
conventionalized set of features or stereotypic behaviour. The categorisation seems to be based
on frame-like knowledge (Fillmore 1982; Ziem 2014). If the context contains lexical units
which relate to frame elements of one or several frames, the speaker may be prompted to label
that frame with the bare noun. As labelling is only necessary if the category is not already
mentioned, this kind of bare noun is used after propositions with an unknown or semantically
unspecified subject. The type of frames may be restricted to conceptual frames, which combine
an object with its stereotypical features, actions or states (Busse 2009). The relation between the
preceding context and the bare noun may be depicted as shown in Figure 1:
Figure 1: Categorisation of predicates against conceptual semantic frames (Type 1)
The use of bare nouns for the categorisation of a perceived entity is very common and not only
restricted to linguistic contexts. (Wittgenstein 2017) describes their use as one of the basics of
the language game. Thus, we find bare noun of this kind also in labels, titles, signposts and so
on. As the description may be incomplete – for example because the author of the description
has no access to the whole picture – those bare nouns are very often guesses. In these cases, it is
not unusual to find sequences of bare nouns – each of them matching the incomplete preceding
description, as in examples (8) above and (9) below:
(9) Из рваной раны на животе сочилась прозрачная масляная жидкость ― лимфа?
‘From the lacerated wound ran a transparent, oily liquid – lymph? juice? blood?’
(Иван Наумов. Обмен заложниками. 2007)
(10) А у меня ничегошеньки нету. ― Студентка? ― Да.
‘But I have really nothing. ― Student? ― Yes.’ (Нонна Мордюкова. Казачка (2005)
(11) Любофф! Кусно! ― Чеченец? ― Ага, чечен. Вайнах.
‘Love! Taisty! ― Chechen? ― Aha, Chechen. Wars.’
(Михаил Гиголашвили. Красный озноб Тингитаны: Записки о Марокко. 2006)
(12) … чего-то мутного, шокирующего, страшного метнулся перед внутренним
‘… something opaque, shocking, terrible cropped up before my mind’s eye. Pain?’
(Татьяна Соломатина. Отойти в сторону и посмотреть. 2011)
Guessing is also a typical use in spoken interaction, where one speaker makes assumptions
about the profession, ethnicity or religion based on statements (10) or behaviour (11) of the
interlocutor. Other bare nouns refer to descriptions that do not lend themselves to an intuitive
labelling. A case in point are feelings, which can be of a kind that does not fit any lexicalized
category (12). Here, the bare nouns are not guesses with a potential “true” answer any more, but
mere suggestions. These kinds of bare nouns represent a transitional type to the bare nouns
The bare nouns under investigation also categorise a preceding predication. However, they do
not propose a more or less intuitive label to an action or entity. Neither do they refer to some
characteristics that have been predicated to some object. Rather, they refer to the proposition as
such. And as such, propositions are not categorisable “naturally”. What bare nouns in this
position offer, is an assessment of the proposition. Thus, in example (13), the assessment
regards the cited assumption нас должно быть на два миллиарда больше, which is
categorised as possible error.
(13) Посмотрим на демографические прогнозы середины прошлого века. Согласно
некоторым из них, нас сейчас на Земле должно было бы быть чуть ли не на два
миллиарда больше. Ошибка?
‘Let’s look at the demographic prognoses from the mid of the last century. According to
some of them, we should be by now on Earth almost by two billion more. Error?’
(Коллективный. Сквозь призму Чернобыля: диалоги о мире и войне. 2006)
(14) И Алла Сергеевна и Андрей Андреевич прошли всю войну и остались живы.
‘And Alla Sergeevna and Andrej Andreevič went through the whole war and stayed
alive. Miracle?’ (Маша Трауб. Нам выходить на следующей. 2011)
Most often, the bare nouns don’t assess the predication as such. The categorisation of an event
as “miracle” in (14) does not naturally follow from the features of that specific description.
Rather, it is the clash between contextual information or world knowledge and the proposition
that prompts the categorisation. Thus, only if we know about a war and the fact that people are
killed in wars, the proposition that someone stayed alive can be interpreted as unexpected and
highly infrequent (which is what the noun miracle expresses somehow). Hence, the predicate
stayed alive can be regarded a miracle. The same holds for example (1), where it is not the
predication не все фашисты плохие itself that can be labelled as paradox, but rather our world
knowledge that renders the proposition irrational.
This type can be depicted as follows: A bare noun categorises the action or property expressed
in the preceding proposition against the background of (his/her or common) world knowledge:
Figure 2: Categorisation of a proposition against world knowledge (Type 2)
Other examples do not assess the proposition directly, neither relate them to the world
knowledge, but rather assess the events that caused the situation described in the proposition:
(15) Строится одноподъездная башня, скажем на 128 квартир. […]. Но вместо 128 или
хотя бы 64 машиномест вокруг дома вы насчитаете от силы 20-30. Ошибка?
‘A tower with one staircase is being built, let’s say with 128 apartments …. But instead
of counting 128 or at least 64 parking lots around the house you count at most 20-30.
Error?’ (Александр Будкин, Михаил Колодочкин. Дары волхвов. 2004)
Obviously, it is not an error, if one counts 20-30 parking lots, as mentioned in the example.
Rather, the bare noun proposes that the construction of 20-30 parking lots may be an error if
the building has 128 housing units. In many cases the assessment does not refer to only one
proposition, but to multiple propositions, usually two:
(16) В том-то и фишка, что правила не запрещают сторонним людям давать сколь
угодно безграмотные, вредные и даже опасные для здоровья (а может, и жизни)
советы, ничем не нарушая своей анонимности. Но ежели специалист захочет дать
рекомендацию, то он обязан представиться «по всей форме». Нонсенс?
‘Exactly this is the tricky part, that the rules don’t forbid strangers to give as many
incompetent, harmful and even dangerous to health (and maybe life) advises as they
want, without violating their own anonymity. But when an expert wants to give a
recommendation, he is obliged to introduce himself “in due form”. Nonsense?’
(Красота, здоровье, отдых: Медицина и здоровье (форум). 2005)
In this example, the categorisation as nonsense cannot be read as referring to он обязан
представиться directly. Rather it is the juxtapposition of the two propositions правила не
запрещают сторонним людям давать советы, ничем не нарушая своей анонимности and
специалист обязан представиться which is rendered unreasonable. But again, it is not the
assertions themselves, but rather the comparison of the rules of the online forum represented in
the propositions, which yields this interpretation. This holds for the majority of examples,
compare the following:
(17) За две тысячи долларов лицензию на сбор и переработку отходов получить легко,
― говорит Геннадий Хорошев. ― Но это неофициально. А вот оформление
лицензии по всем правилам обойдется уже в 4000 долларов. Парадокс?
‘For two thousand dollars you’ll easily get a license for collecting and processing litter
― Gennadij Xorošev says. ― But this is unofficial. However, issuing a license
according to the rules costs you already 4000 Dollars. Paradox?’
(Екатерина Чинарова. Бумажная проблема.
Можно ли заработать на макулатуре? 2004)
(18) Описание фактов его биографии допускается, но рассказы о его делах ― ни в
коем случае. По большому счету, весь позитивный материал об адвокате можно
посчитать вводящим в заблуждение его потенциальных доверителей. Абсурд?
‘The description of facts from his biography is allowed, but stories about his cases by
no means. By and large, all positive material about a lawyer can count as misinforming
potential mandators. Absurd?’
(В. Залманов. Информация об адвокате и реклама. 2004)
The number of lexemes that occur in this function is restricted. In my sample, there are only 29
different types that account for 63 tokens. Strikingly, these lexemes are very similar in their type
of categorisation as most of them refer to entities whose existence is questionable, unexpected
or simply false:
• marginal маргинал ‘fringe’, мелочь ‘peanuts’, пустяк ‘bagatelle’
• false ложь ‘lie’, неправда ‘untruth’, правда ‘truth’
• nonsense глупость ‘stupidity’, маразм ‘insanity’, нонсенс ‘nonsense’,
ошибка ‘error’, чушь ‘junk’
• unreal галлюцинация ‘hallucination’, утопия ‘utopia’, фантастика ‘fiction’,
мистика ‘mystic’, чудо ‘miracle’
• made-up пропаганда ‚propaganda’, риторика ‘rhetoric’, сказка ‚fairy tale’
• illogical абсурд ‘absurd’, парадокс ‘paradox’
The connotation of these lexemes is mostly negative, with the exception maybe of the group of
unreal entities or events. They all refer to two preceding propositions and categorise their co-
occurrence as being marginal, false, nonsense, unreal, made-up or illogical against a
presupposed world knowledge. So we can depict this third type as in Figure 3:
Figure 3: Categorisation of two propositions against world knowledge (Type 3)
As already mentioned, these bare nouns are often answered by the same author/speaker in the
following sequence. While answers of this type are provided in roughly half of the examples
(52,4%), it is interesting that the answers themselves are mostly (31,7%) negative. That is, the
negative evaluation that the bare noun evokes, is immediately negated afterwards. One eighth
(12,7%) of the answers are positive but are immediately followed by a “but”, while other eight
percent are negated indirectly in the following text. There is no unambiguously positive answer
that confirms the negative attribution inherent in the bare noun. If the negative connotation is
often negated, the question remains, why it is mentioned in the first place?
If we look at the examples, it seems that the bare noun has the function to anticipate a reaction
that the author suspects on the side of the recipient. This is most obvious for the categorisation
as marginal, compare the following example:
(19) Локтем травмированной руки можно работать на защиту, наносить удары и
выводить противника из равновесия (фото N 6). Но самый главный итог этих
действий состоит в том, что время до потери сознания увеличилось в 2-3 раза.
Мелочь? В борьбе за выживание мелочей не бывает.
‘The elbow of the injured arm can be used as protection, to deal out blows and throw
the enemy out of balance (picture No. 6). But the crucial outcome of these actions
entails that the time until losing consciousness has been extended by 2-3 times.
Peanuts? In the battle for survival there are no peanuts.’
(Если вас «зацепили» ножом... «Солдат удачи». 2004)
The author anticipates one possible reaction on a given advice and the proposed impact of it.
This is a discursive strategy which seems to aim at the strengthening of the own argument or the
own narrative. The same seems to be the case with other bare nouns of type 3: They take up a
perceived or at least allegedly anticipated mismatch between the world knowledge and the
preceding propositions, speak it out – and deny it. But by mentioning the category, the author
also opens a new angle from which the situation can be perceived – a perspective, which the
reader may not have chosen on her/his own. This puts the use of the bare nouns in the realms of
framing mechanisms (Goffman 1974).
3.2 Extending the schema
Although most of the bare nouns of type 3 in the given context contain some suspicion towards
the previous predication, there is one important exception. This concerns mainly two words,
namely slučajnost’ and sovpadenie, both translatable as ‘coincidence’. Both nouns do not
express an inherently wrong, illogical or unreal event. In fact, they do not represent a
categorisation of any event. Rather, they refer to the causal link between two events, or – more
precisely – the absence of causal links.
(20) Фотографы и камеры уже готовились запечатлеть историческое событие, как
внезапно хлынул проливной дождь. И мероприятие сорвалось. Случайность?
‘The photographers and cameras were already preparing to capture the historical event,
as all of a sudden cloudburst began. And the event fell through. Coincidence?’
(Виктор Лихачев. Поговорим о политике. 2003)
(21) Машину задержали минут на десять, и несколько раз за это время один из
чеченцев связывался с кем-то по рации. Потом оперов отпустили и через
километр машина наехала на мину… Совпадение?
‘The detained the car for about ten minutes and during that time, one Chechen several
times contacted someone by radio. Then they released the agents and after one
kilometre the car went on a mine … Coincidence?’
(Подполковник милиции «Гром». Кадыры решают всё. 2003)
However, the constructions clearly belong to the family of categorising bare nouns, since they
share most of the characteristics of the latter. Just as the rest of the family, they refer to a
preceding context and can be followed by an answer from the same author. Like type 3 of the
bare noun construction, they do not refer to one propositions, but to the interplay of two
The function of the two nouns differs from the previous examples, though. While the
categorising nouns of the first type refer to some kind of perceivable category features and the
second and third type take up a conceptual mismatch than can be expected to occur in the reader
based on stereotypic world knowledge, these two bare nouns refer to a connection that does not
exist. That is, there is no world knowledge and no semantic frame that contains both
propositions. Thus, the reader or listener is not likely to connect the preceding assertions in any
way on her/his own. As for example (20), it seems to belong to normal reasoning that rain
generally cannot be causally linked to any human action. As such, the allusion to such a
connection seems itself rather paradox. In example (21), the rational assessment of the relation
between the first event (A Chechen calling someone by radio) and the second (car drives on a
mine) is impossible. Any connection apart from temporal succession is a mere supposition.
And indeed, the semantics of the two nouns seem to deny the possibility of a causal link.
However, as part of the family of categorising bare nouns, they make use of the same
interpretational tendencies of denying the proposed categorisation. This way, the suggested
absence of a causal link is – directly or indirectly – denied. The constructions can be thus used
to invoke the possibility that the two propositions mentioned in the preceding context are indeed
Figure 4: Constructing a causal link between two propositions (Type 4)
It is interesting that nouns like rezul’tat ‘outcome’ or pričina ‘reason’ which could be used to
express causality directly are only used in constructions of the type shown in examples (22) and
(23). They open a new topic, but do not question the causality of previously mentioned events.
(22) Долгое время «независимые» СМИ поливали армию грязью и дерьмом со
страниц практически всех газет и журналов. Результат? Деморализация армии
‘The „independent“ press for a long time threw mud and shit on the army from the
pages of literally all newspapers and journals. Result? The demoralization of the army
…’ (Нет перспективы, нет смысла в службе «Солдат удачи». 2004)
(23) Но фирма […] разослала в свои 36 филиалов циркуляр, […]
предлагающий выбирать автомобиль или общественный транспорт. Причина?
Насыщенное автомобильное движение делает поездки на велосипеде по Лондону
‘But the company … sent a round mail to its 36 subsidiaries … recommending to
choose cars or local public transport. Reason? The intense car traffic renders bicycle
rides in London too dangerous.’
(Обобщенный. Кунсткамера. Вот это да! «Наука и жизнь». 2008)
The inherent negation of the proposed categorisation is not the only feature the two bare nouns
slučajnost’ and sovpadenie inherit from the broader construction type. They also seem to carry
the negative assessment of the other bare nouns. The use of both nouns thus gives birth to the
feeling that a causal relation between the two predications is something wrong, unwanted. This
is very obvious in example (21), where the writer alleges Chechens to have coordinated their
actions with the aim to blow the car. This is also clearly the case in the following example:
(24) В комнате, где спал Леонардо, висели две его картины ― «Иоанн Креститель» и
«Джоконда». По странному совпадению, лица обоих персонажей схожи.
Случайность? Давно родилось подозрение, что «Мона Лиза» была вовсе не
супругой Франческо дель Джокондо, а… переодетым любовником Леонардо […].
‘In the room where Leonardo slept, were two paintings: „John the Baptist” and “Mona
Lisa”. By a strange match, the faces of both figures are similar. Coincidence? Long
since suspicion raised that “Mona Lisa” was not at all the wife of Francesco dell
Gioconda, but … the disguised lover of Leonardo …’
(Александр Голяндин. Леонардо: человек и кодю 2006)
Here, the writer insinuates that Leonardo da Vinci has on purpose labelled the mentioned
paintings with other names than those who are really portraited. The argument provided is thus
aiming at uncovering hidden homosexuality. A negative assessment is clearly implicit, although
from the fragment it remains unclear whether it is the deceit or the homosexuality (both, of
course, only alleged) that is aimed at.
Both nouns are thus even more candidates for a clear framing technique. They evoke a causal
perspective on two events which may not be the first choice of the recipient. The nouns thus
don’t refer to a mismatch between world knowledge and two propositions, but rather create
knowledge or connections between parts of knowledge the reader is likely not to have had
before. Moreover, as member of the family of bare noun constructions, they also invoke a
negative assessment of the indirectly alleged causal link. As such, it is no wonder that this kind
of bare noun figures prominently in arguments which rather belong to the sphere of conspiracy
4 Status of the bare noun construction
Whether bare nouns are ellipses is principally a question that can be answered only within the
realm of a given grammatical theory. Both ellipsis and non-ellipsis accounts have to collect
arguments for their stance and for their view of language and grammar. It is not my intention to
answer this question here in general, not even for the bare nouns. But I want to discuss the
status of the bare noun construction in the light of the given functional analysis.
Ellipses have been assumed to have a specific role within a context. Either they make use of a
given situation, they take up a preceding linguistic structure or they at least complete a meaning
structure that is somehow inherent in the context. But not every linguistic structure is
traditionally recognized as ellipsis. For the following discussion, I thus refer to a definition of
ellipsis, which is very clear and basically theory-independent, namely the one proposed by
(Kindt 2013). He defines as necessary elements of a definition: the (elliptic) part of an utterance
(U), its meaning (M), and the given context (C). As premises, only those parts of utterances can
be regarded to be elliptical which can be expanded to a larger construction U’. Additionally,
there must be a partial context C’ (e.g. a zero context) in which the elliptical U does not have
the meaning M. Only those parts of utterances U are ellipses, whose expansion U’ has the same
meaning M in both context C and C’, but which itself do not bear the meaning M in the
shortened context C’ (Kindt 2013, 44).
In order to analyse the given construction within this definition, it is first necessary to assess
possible structural expansions of the bare nouns. In a second step, these have to be analysed
regarding their meaning in the full context C as well as in the shortened context C’. As the
critical (full) context for the given bare noun construction is the preceding proposition(s), it thus
needs to be examined whether the two contexts affect the meaning of the bare noun and its
possible expansion. Since Kindt’s definition holds only for structural ellipses, I will as a third
step discuss whether the bare noun construction can be counted as situational ellipsis.
4.1 Possible extensions
In order to identify the full construction the bare noun could be an ellipsis of, we certainly need
to look for syntactic environments in which the noun usually occurs. Since space is limited, I
will do this exemplarily for one of the most frequent nouns in the bare noun ellipsis, slučajnost’
‘coincidence’. The search for the word form in nominative in RNC gave 4.003 examples.
Saving a random sample of these and excluding plural forms and forms that are accusative left a
sample of 703 sentences. In these examples, the following structural templates with at least 10
occurrences were identified:
И это не случайность, это позиция.
‘And this is no coincidence, this is a position.’
(Р. Н. Аджубей. О науке, о жизни и о себе. 2009)
X – slučajnost’
С этих позиций распад империи ― случайность,
противоестественное течение событий, […]
‘From these positions the breakdown of an empire is a
coincidence, an unnatural flow of events, …’
(Егор Гайдар. Гибель империи. 2006)
Случайность может перевернуть всю жизнь […]
A coincidence can change the whole life …’
(Леонид Утесов. Спасибо, сердце! (1982)
Женская хромосома «умнее» мужской. Случайность,
сделавшая человека человеком
‘The female chromosome is „smarter” than the male. A
coincidence which turned humans into humans.’
(Рафаил Нудельман. Загадочная связь. 2005)
Вот везет, так везет. Чистая случайность, а как
‘What a luck. It was pure coincidence, but how
convenient.’ (Алексей Рыбин. Последняя игра. 2000)
Slučajnost’ – X
Ведь случайность ― одна из важнейших черт
реальной жизни […]
‘As a matter of fact, coincidence is one of the most
important features of real life.’
(Геннадий Горелик. Науки естественные и
Чистая случайность, что я проходил мимо.
‘It was sheer coincidence that I was passing by.’
(Юрий Трифонов. Предварительные итоги. 1970)
Esli by ne
Разговор она так и не завела бы, если бы не
‘She would have never started this discussion, if it
weren’t for a coincidence’
(Маша Трауб. Замочная скважина. 2012)
Table 1: Sentence patterns with the noun slučajnost'
Obviously, since the abstract noun can only metaphorically serve as agent, structures with a
nominal predicate form the majority of uses of the noun. These structures mostly identify
another mentioned entity or situation as being accidently. But there is also a significant amount
of cases in which the noun takes the role of a subject in a verbal construction. And –
interestingly – the use as bare noun with an attribute makes up nearly 10% of all uses of the
noun. Which structure is thus the origin of the ellipsis? A first guess would be maybe, that the
most frequent structure – [Eto slučajnost’] – is the original sentence. However, the sentence Eto
slučajnost’(.|?|!) occurs only seven times in the RNC. Together with future and past tense – and
thus overt copula – we only find 12 occurrences. It does not seem very plausible to assume that
the ellipsis is the shortened version of a sentence that is used only rarely. It could be even
argued that the full sentence is an extension of the bare noun (Wittgenstein 2017, 21). Indeed,
some researcher have underlined that full structures are sometimes less acceptable (if at all) than
the elliptic form (Kindt 2013, 46; Knobloch 2013, 31). The only thing we gain by expanding
ellipses is the correspondence of the full constructions with ideas about syntactic integrity.
But maybe the full sentence includes more elements that just the additional eto ‘this’. One could
think of structures like:
(25) Možet eto slučajnost’
‘Maybe this is coincidence’
(26) Eto prosto slučajnost’
‘This is only coincidence’
However, none of them occurs more often than the ellipsis itself. Another guess would be, that
the structure [X - slučajnost’] is the original or full form of the ellipsis. In this case, the
orthographic separation of the bare noun through punctuation rather than hyphenation may be
regarded a stylistic variation. It should then be possible to join the ellipsis with the preceding
sentence. But as example (20), repeated here as (27) shows, this produces questionable
(27) Фотографы и камеры уже готовились запечатлеть историческое событие, как
внезапно хлынул проливной дождь. И мероприятие сорвалось. Случайность?
?Фотографы и камеры уже готовились запечатлеть историческое событие, как
внезапно хлынул проливной дождь. И мероприятие сорвалось – cлучайность?
Without having asked native speakers for their acceptability judgement, I judge the version
given below to be orthographically rather ill-formed. When we compare these structures, we see
that they are indeed different to the bare noun constructions under investigation. Namely, the
non-elliptical structure takes a nominal element (or NP) as the left context and thus constitutes a
simple nominal predicate (examples (28) and (29)). The ellipsis, on the other hand, always
expects one or even two whole propositions since it evaluates predications, not single objects.
So, the structure [X - slučajnost’] is even less a candidate for an expanded construction than the
structure [Eto slučajnost’].
(28) Любые совпадения с реальностью ― случайность.
‘All matches with reality are coincidence.’ (Человек (форум). 2007)
(29) По-видимому, это молчание не совсем случайность […]
‘Obviously, this silence was not a coincidence …’
(Игорь Андреев. «Грабли» истории. 2012)
The structure with the noun as subject of a verbal predicate is, of course, ruled out as origin of
the ellipsis if we have in mind the categorising function of the bare noun construction. All other
constructions are less frequent than the use of slučajnost’ as bare noun and I doubt that they are
plausible candidates. Kindt defends the view that ellipses and their extension do not have to be
functionally or semantically identical and assumes even reformulations as valid method for
establishing an extended counterpart (Kindt 2013: 46). However, this understanding is
problematic as there is no evidence that ellipses and these kinds of “forced” extensions are
really related or that the relation between ellipsis and extension may not be the other way round,
that is, that ellipsis is basic and the extension secondary (Wittgenstein 2017, 21).
To sum up, the functionally most similar structure is [Eto slučajnost’], which, however, occurs
to be much less prominent in the given context. And although the bare noun needs a left context,
it does not refer to its syntactic structure in any way. I would thus hesitate to confirm that the
bare noun construction is an ellipsis in a structural understanding.
4.2 Meaning of the bare noun construction
Although it is structurally questionable whether the bare nouns under investigation are short
forms of full sentences, let´s assume for the sake of argument that the structure [Eto slučajnost’]
can be regarded the corresponding full form of the ellipsis. The definition of ellipsis provided
by (Kindt 2013) assumes that the ellipsis U must have the same meaning M as the whole
sentence U’ in the given context C. Indeed, there is little doubt that the bare noun has the same
meaning as the full sentence if preceded by the same predication. Compare example (6) again:
(30) a.Что, если взять себе за железное правило: ни при каких обстоятельствах не
предлагать денег людям в милицейской форме? Утопия?
b. Что, если взять себе за железное правило: ни при каких обстоятельствах не
предлагать денег людям в милицейской форме? Это утопия?
‘What if one introduces the golden rule: Don’t offer money to people in police uniform
under no circumstances. (Is this a) Utopia?’
But what is the meaning M that is supposed to be the same? One difference between (30a) and
(30b) is the anaphoric relation to something either preceding the utterance or surrounding the
interlocutors, which is referred to by the demonstrative pronoun eto ‘this’. The pronoun can
refer to some event or object mentioned before, either in the same sentence or in a preceding
sentence. Indeed, the bare noun itself bears no information about the directionality of its
interpretation. An anaphoric interpretation is not likely to be the default interpretation, as there
are other kinds of bare noun constructions which do not point backwards, but forwards (more
precisely both back- and forwards):
(31) В ожидании расстрела занимался самонаблюдением. Внешность? Эффектная.
‘While expecting the shot, he went in for self-observation. Appearance? Spectacular.’
(Даниил Гранин. Зубр. 1987)
(32) Но согласитесь, что мы плохо играем, бразильцы лучше. Хоккей? Пожалуйста!
‘But you have to agree that we play badly, the Brazilians are better. Hockey? Anytime!’
(Коллективный. Мне 75. И я еще нужен! «Психология на каждый день». 2010)
However, referring only locates the noun within the text, it does not supply any meaning by
which it would be clear how to interpret the single noun. The pronoun is only the placeholder
for the preceding proposition. This becomes even more visible, when we analyse the possible
ellipsis and the full construction in the shortened context C’:
(33) a. Утопия?
b. Это утопия?
‘(Is this a) Utopia?’
Both questions are incomplete, both questions would most likely evoke the question What? if
uttered in a situation of direct communication. In written texts, however, the bare noun could
also precede a textual sequence, for example in the role of a title. This possibility does not exist
for the full sentence, since the pronoun needs an antecedent. Otherwise, the meaning does not
seem to change. Hence if pointing to a preceding context is meaning, then the bare noun
construction (U) and the full sentence (U’) differ in their meaning M.
It is also possible to argue that the term meaning refers to propositionality. Whether there is
such a thing as propositions and what form they should take is, of course, a controversial
question. However, let us proceed from the probably undisputed assumption, that a proposition
is the sentence meaning and that it also has a truth value (McGrath and Devin 2018). The first
question regarding the bare noun construction is then whether this construction can have a
proposition at all. Diagnostic tests for propositions such as that-clauses or propositional
anaphora seem to indicate that the question Eto utopia? can indeed be regarded a proposition
(34), while the bare noun is not likely to be one (35).
(34) Eto utopia? Ne dumaju čto [eto utopia].
Is this an utopia? I don’t think that [this is an utopia].
(35) Utopia? ?Ne dumaju čto [utopia].
Utopia? I don’t think that [utopia]
Thus, the extension indeed refers to a sentence meaning, which the potential ellipse does not do.
It is interesting, however, that one can nevertheless refer to the elliptical question with a that-
clause if one includes the context. Accordingly, the that-clause with the bare noun in (36) seems
far more acceptable:
(36) Čto esli vzjat' sebe za železnoe pravilo: ni pri kakich obstojatel'stvach ne predlagat'
deneg ljudjam v milicejskoj forme? Utopia? Ne dumaju čto [utopia].
‘What if one introduces the golden rule: Don’t offer money to people in police uniform
under no circumstances. Utopia? I don’t think that [utopia].
The answers given in the right context of the bare noun construction using adverbials (vozmožno
‘possibly’, konečno ‘of course’) also seem to indicate that the bare noun is linked to a
propositional content. But it is a matter of dispute whether questions have propositions, because
although there seems to be a kind of sentence meaning, it obviously has no truth value. Instead,
one could say that questions are speech acts that ask if the components of a proposition are
related in some way. In our case, the question would be whether the proposition(s) in the left
context and the ellipse are somehow connected. To make this question possible, however, an
anaphoric reference is needed, which, of course, is missing in the case of bare nouns.
Since reference to the preceding or surrounding context seems to be crucial, it must be pointed
out that bare nouns can fulfil this function even outside of any linguistic context. For example,
inscriptions (see Wittgenstein 2017) or signs (Knobloch 2013) refer to the surrounding context
just in the same way as propositions with deictic pronouns. For such references, no modern
approach still assumes that a silent proposition is necessary for their interpretation (Knobloch
2013, 20). If both structures - the proposition (33b) and the bare noun (33a) - fulfil the same
function, then logically it cannot be concluded that the bare noun must have a hidden
proposition as well. This interpretation seems rather to be the result of a "cult of the
proposition" (Knobloch 2013, 27).
Some researchers proposed that the extension of the meaning of an ellipsis (M) does not refer to
a proposition, but is based on inference strategies (Kindt 2013, 79; Otal Campo et al. 2007,
118). (Culicover and Jackendoff 2012, 332) proposed, that the typical ellipsis makes use of the
same-different-analogy: An ellipsis presupposes a meaning already communicated in the left
context and adds only those elements which are different. However, the bare noun construction
cannot be explained through inference, as it does not presuppose any meaning structure. There
is no “same but different” relation here, as there is no “same”. Neither is the categorisation
provided by the ellipsis given in contrast to another categorisation in the context nor does the
ellipsis take up a previously used anaphoric reference. It stands for itself.
In the light of the functional analysis conducted above, the meaning and the function of the bare
noun construction can be described quite concretely. A full representation of the meaning M of
types 2, 3 and 4 would be something along the lines of the following:
(37) Can the aforementioned combination of propositions be categorised as X in the light of
our world knowledge?
This meaning is by no means represented in the “full” construction [Eto utopia?] although it
cannot be excluded that the proposition would yield the same interpretation in this context.
However, as has been shown, the bare noun construction does not only posit a categorisation but
is also mostly restricted to negative categorisation. This extra meaning is more likely to be
deleted when the full construction is used.
To sum up, there are some arguments to describe the bare noun construction under investigation
as ellipsis: There is a potential expanded construction U’ with a partially similar meaning M
which retains this however defined meaning M also without the full context C. The problem
with this analysis, however, is that the scope of the meaning M is not at all clear and crucially
depends on the semantic theory we chose to work in. As has been shown, the bare noun
construction misses the function of anaphoric reference and propositionality (and
predicativeness, to that end). On the other hand, it has very concrete meaning components,
which have presumably been developed through conventionalization in this specific kind of
context with this selected group of nouns. This meaning M is not likely to be preserved in the
4.3 Is the bare noun construction a situational ellipsis?
The discussion in the previous section has shown that reference to the preceding context is the
decisive feature of the bare noun construction. Exclusive contextuality, in turn, is the crucial
characteristics of situational ellipsis. A situational ellipsis is embedded in a common ground, a
joint activity or an object of joint attention. These provide a frame, a shared social world, which
allows the interlocutors to resort to the shortest form possible (Thorne and Lantolf 2006, 257).
Often, this context is in some sense self-explaining (Rickheit and Sichelschmidt 2013, 172). It
could also be argued that it is the context or the situation which is elliptic. That means, that in a
given situation, only one short linguistic cue is missing (Rickheit and Sichelschmidt 2013, 163).
Thus, it can be argued that in a situation, where guests are entering a household or office, it is
only necessary to pronounce the bare noun “Coffee?” to prompt both politeness and an
invitation (Heine 2011; Wasserscheidt in print). However, the bare nouns under investigation
are not embedded in such a kind of situation. There is no environment with participants, props,
joint actions or the like. Hence there is no elliptical context which needs the insertion of a
linguistic cue. The preceding sequence of predications is likewise not incomplete and can be
interpreted without the following bare noun. Only if the interpretational process of the reader
itself (the incremental processing of the preceding predications) is seen as a situation does it
make sense to speak of a situational ellipsis. However, most linguistic units make use of the
incremental reception of some preceding material. This way, every ellipsis and also most
propositions could be described as situational (which is maybe a useful generalisation if we
want to understand language as action). So, although the bare noun construction is interpretable
only within the given context, it is not a situational ellipsis in the sense used in the literature.
5 Conclusion: A non-propositional explanation
The bare noun construction presented here is clearly part of a more general use of bare nouns
that seem to have a general function: categorisation. The specific subset of bare nouns depends
on one or more propositions in the left context and provides a speculative categorisation that is
usually associated with a negative attitude. Since the negative attitude is not generally
confirmed, bare nouns seem to have the function of taking up possible objections to the own
argumentation or to frame an event in a way that renders it unimportant, illogical, untrue, or
The question of whether this construction is an ellipse, in my view cannot be answered. In the
literature, mainly two approaches are discussed, namely the so-called ellipsis approach (=
syntactic expansion) and the direct interpretation approach (= semantic expansion) (Merchant
2004, 662). As shown above, a syntactic extension of the ellipsis is only conditionally plausible
since there are no corresponding structures with the same salience in actual usage. An
argumentation in favour of a syntactic extension approach, it seems, can only be put forward if
one takes the ellipsity of the bare noun construction for granted. A semantic extension approach
has to answer the question, how to define the target meaning to which the meaning of an ellipsis
should be extended – and how to include the specific meaning that is connected to the bare noun
in the given context. It may therefore be more appropriate to describe the bare nouns as a
fragments or interactive units without sentence character (Knobloch 2013, 28).
However, if one does not focus on imputed syntactic or semantic well-formedness, but on the
function of linguistic utterances, it seems that the bare noun construction is better understood as
discourse (Nir 2015) or activity patterns (Handlungsmuster) (Busse 1991, 10; Thorne and
Lantolf 2006). For what unites sentences, ellipses, fragments and interactive units is their
function: to tell the hearer/listener what she/he should think (Rickheit and Sichelschmidt 2013,
172). The bare noun construction calls on the listener to evoke a particular frame and to
interpret the preceding propositions in relation to that frame. Of course, this is not a physical
activity, and so the bare nouns described here differ from situational ellipses that are part of
action patterns such as those described by (Wittgenstein 2017, 21) or the "coffee construction"
(Heine 2011, 74; Wasserscheidt in print). However, as described above, bare nouns do not
merely categorise propositions and thus tell the speaker how he/she should think about them.
The propositions are also categorised in the context of a presupposed world knowledge. This
basically means that the function in question is not purely linguistic (at least not in the narrow
sense used in most grammatical theories). Rather, the goal is above all the manipulation of
cognitive structures. Thus, the discourse pattern in which the bare nouns are integrated and
through which it is instantiated is a specific manipulation of cognitive structures during the
perception of a given text. Just as bare nouns in our environment (e.g. inscriptions) are always
interpreted within the realms of a given activity (Knobloch 2013, 28), these bare nouns are
interpreted within the activity of reading and adding propositions to the knowledge base (or
mental situational model, compare Rickheit and Sichelschmidt 2013, 166).
Crucial for linguistic considerations is that an integration of the bare noun construction into (or
interpretation as) an action pattern renders it no longer necessary to search for a hidden
meaning. First, it must be recalled that it is not sentences or any other linguistic elements that
carry meaning, but people create meanings by interpreting linguistic inputs. This insight is
widely used in studies on ellipses (Imo 2013, 287; Rickheit and Sichelschmidt 2013, 179;
Thorne and Lantolf 2006, 249). If the bare noun is integrated into an action pattern, there is no
need to ascribe certain functions to it. Rather, for example, anaphoric reference and negative
assessment can be understood as part of the action pattern, as well as possibly the functions
usually ascribed to propositions. The meaning of the bare noun, on the other hand, need not be
expanded, but can remain the same that the noun has in other contexts (e.g. the other contexts of
the noun slučajnost' above).
The choice of explaining the bare noun construction is therefore not only between a so-called
ellipsis approach (= syntactic expansion) and a direct interpretation approach (= semantic
expansion) (Merchant 2004, 662), but also between a sentential/propositional and a non-
sentential/non-propositional approach. If we only assume the structure and the meaning of the
bare noun (without additional syntactic or semantic elements), we do not have to complicate the
theory by changing the syntax-semantics mapping or the syntax by proposing an unpronounced
structure, as suggested by (Merchant 2004, 674). Rather, we should describe the specific
function of the construction or discourse pattern. This type of function - a specific manipulation
of the relation between situational frames provided in the context - has not been described to the
best of my knowledge by any existing grammatical theory or communication model.
Busse, D.: 1991. Wortbedeutung und sprachliches Handeln. Überlegungen zu den Grundlagen
Busse, D.: 2009, ‘Textbedeutung und Textverstehen aus Sicht einer linguistischen
Epistemologie’, in Bachmann-Stein, A. (ed.), Perspektiven auf Wort, Satz und Text.
Semantisierungsprozesse auf unterschiedlichen Ebenen des Sprachsystems; Festschrift für
Inge Pohl (Koblenz-Landauer Studien zu Geistes-, Kultur- und Bildungswissenschaften, Bd.
3), Trier, pp. 45–56.
Cheng, W.: 2009, ‘Ellipsis’, in Cummings, L. (ed.), The Routledge Pragmatics Encyclopedia,
Culicover, P. and Jackendoff, R.: 2012, ‘Same-except: A domain-general cognitive relation and
how language expresses it’, Language 88, 305–340.
Fillmore, C. J.: 1982, ‘Frame Semantics’, in The Linguistic Society of Korea (ed.), Linguistics
in the morning calm, Seoul, pp. 111–137.
Goffman, E.: 1974, Frame analysis. An essay on the organization of experience, Cambridge,
Heine, L.: 2011, ‘Non-coordination-based ellipsis from a Construction Grammar perspective:
The case of the coffee construction’, Cognitive Linguistics 22, 55–80.
Imo, W., ‘Elliptical structures as dialogical resources for the management of understanding’,
Imo, W.: 2013, ‘Ellipsen und Verstehen in der Interaktion’, in Hennig, M. (ed.), Die Ellipse.
Neue Perspektiven auf ein altes Problem, Berlin, pp. 281–320.
Kindt, W.: 2013, ‘Theoretische und methodische Grundlagen der Ellipsenmodellierung’, in
Hennig, M. (ed.), Die Ellipse. Neue Perspektiven auf ein altes Problem, Berlin, pp. 39–106.
Knobloch, C.: 2013, ‘„Ein Teil, das fehlt, geht nie kaputt“ – Ellipsen in Grammatik und
Kommunikation’, in Hennig, M. (ed.), Die Ellipse. Neue Perspektiven auf ein altes Problem,
Berlin, pp. 19–38.
McGrath, M. and Devin, F.: 2018, ‘Propositions’, in Zalta, E. N. (ed.), The Stanford
Encyclopedia of Philosophy.
Merchant, J.: 2004, ‘Fragments and Ellipsis’, Linguistics and Philosophy 27, 661–738.
Nir, B.: 2015, ‘Frames for clause combining. Schematicity and formulaicity in discourse
patterns’, Constructions and Frames 7, 348–379.
Otal Campo, J. L., Ruiz de Mendoza and Francisco: 2007, ‘Modelling thought in language use:
At the crossroads between discourse, pragmatics, and cognition’, Jezikoslovlje 8, 115–167.
Rickheit, G. and Sichelschmidt, L.: 2013, ‘Verstehen von Ellipsen – ein holistischer Ansatz’, in
Hennig, M. (ed.), Die Ellipse. Neue Perspektiven auf ein altes Problem, Berlin, pp. 159–182.
Thorne, S. L. and Lantolf, J. P.: 2006, ‘A Linguistics of Communicative Activity’, in Makoni,
S. and A. Pennycook (eds.), Disinventing and (re)constituting languages, Clevedon, pp. 245–
Wasserscheidt, P.: in print, ‘Situational ellipsis and its role in frames and cultural routines’,
Wittgenstein, L.: 2017, Philosophische Untersuchungen, Frankfurt am Main.
Ziem, A.: 2014, Frames of Understanding in Text and Discourse. Theoretical foundations and
descriptive applications, Amsterdam / Philadelphia.
Дудникова, Л. В.: 2013, ‘Эллиптические конструкции в составе бессоюзных структур в
русском и французком языках’, Филология, 186–194.