Russian sentence adverbials: classification, orientation and representation

Abstract

In this article sentence adverbials (SA) in Russian are analyzed in their totality, i.e. from a lexical, syntactic, semantic and pragmatic point of view. They are classified according to Hare’s three utterance components, which yields (1) neustic, (2) tropic and (3) phrastic SAs. These components are used to represent semantic paraphrases of Russian SAs in utterances from various types of discourse in order to show their exact contribution to the meaning conveyed by the entire utterance. They are further subdivided according to their function: (1) into connectives and non-connectives; (2) into attitudinal and modal SAs; and (3) into temporal and areal SAs. It is demonstrated that many adverbials consist of micro-groups involving three and only three members, in which one is first person oriented, the other is second person oriented and the last one is third person oriented. This indicates that when making utterances Russians have to base them on the previous discourse in one way or the other. This paper focusses on a specificity of the Russian language—the system of modal adverbials that reflect a division between external and internal reality. We end the examination by discussing the function of word order in connection with more than one SA in an utterance.

Аннотация

В настоящей статье предлагается системное описание русских сентенциальных адвербиалов (СА), учитывающее их лексический, синтаксический, семантический и прагматический статус, а также их классификация на основе концепции Р. Хэара о трех компонентах высказывания—неустическом, тропическом и фрастическом. Схема Р. Хэара положена в основу используемых в статье семантических перифраз русских СА, употребляемых в разных типах дискурса, с целью определения доли их непосредственного участия в формировании значения всего высказывания. Выделяются три основных класса СА: (1) неустические СА, (2) тропические СА и (3) фрастические СА, которые, в свою очередь, в зависимости от функции подразделяются на (1) коннективы и неконнективы, (2) СА установки и модальные СА и (3) темпоральные и пространственные СА. Данное исследование показывает, что многие СА образуют минигруппы, состоящие строго из трех членов, каждый из которых соотносится с первым, вторым или третьим лицом. Этот факт свидетельствует о том, что любое высказывание в русском языке должно опираться на предшествующий дискурс. Особое внимание уделяется специфике русского языка, обусловленной разграничением внешней и внутренней действительности, что находит свое отражение в системе модальных СА. В заключение рассматривается роль порядка слов при употреблении нескольких СА в рамках одного высказывания.

1 Introduction

There are quite a few lexical and lexico-grammatical descriptions of Russian sentence adverbials (SAs) in the literature on the subject—in grammars (cf. Vinogradov 2001[1947], 1952; Barnetová et al. 1979; Švedova 1980; Švedova and Lopatin 1989; Zolotova et al. 1998; Sičinava 2011; Wade 2011; and others), dictionaries (cf. SSRLJa 1948–1965; Evgen’eva 1981–1984; Lopatin and Lopatina 1994; Ožegov and Švedova 1997; Efremova 2000; Kuznecov 2001; Rogožnikova 2003; Gerd et al. 2004–2012; Burceva 2005) and monographs (cf. Filipenko 2003; Pankov 2009; Savelova 2009 and the extensive bibliography therein). Moreover, it is relatively easy to find more elaborate analyses of some specific SAs and other items that are more or less synonymous to them (cf. Bulygina and Šmelev 1987, 1993; Rudnickaja 1994; Boguslavskij 1996; Burkhardt 1999; Janko 2001; Apresjan 2004; Grenoble 2004; Pen’kovskij 2004; Letučij 2008; Padučeva 2010, and references therein and below). But one can hardly find an integrated examination of all possible SAs in Russian involving an analysis of their lexical or lexico-grammatical meaning as well as their syntactic, semantic and pragmatic functions. The most frequent approach is to take a functional perspective and isolate a single group of SAs that share a specific function without taking their place in the overall structure of the entire system into consideration (cf. Anikin 1956; Evtjuxin 1979; Buglak 1990; Filipenko 1998; Apresjan 1999; Ovsjannikova 2000; Mukovozova 2002; Vinokurova 2004; Belova 2011). Ostensible examples of this kind of inquiry are the numerous studies devoted to evidentiality and evidential markers in which there is no specific account of their place and status in the system of the Russian language (cf. Kjul’moja 1997; Kozinceva 2000, 2007; Mordasova 2005; Wiemer 2008, 2010; Šestuxina 2009; Levontina 2010, 2011; Grigorenko 2011; Padučeva 2012¹). We take a holistic approach to SAs in which we introduce their classification into major and minor groups based on an analysis of their syntactic, semantic and pragmatic functions. To achieve all this, we use paraphrases structured according to a three-componential utterance model, which is identical to what we call the declarative superstructure. Another purpose of this study is to show the unique facets of the Russian language, i.e. how it differs from other languages in several ways or how it differs from the traditional view of language as such. This goal can only be achieved by comparing and contrasting Russian with a typologically completely different language, in this case English.

1.1 On identifying and classifying SAs in Russian

When Russians say Filipp bolen ‘Filipp is ill’, they, in fact, convey more than the utterance alone suggests. All Russian utterances with a declarative sentence form imply two other components besides the overt utterance part, viz. ‘I hereby say’ and ‘It is true’. These three components together constitute what we call the ‘Russian declarative superstructure’:

• ‘I hereby say’ relates to the speech act type ‘declarative’ and implies a corresponding declarative sentence form.

• ‘It is true’ relates to what is traditionally called propositional attitude, coinciding with the mood of an utterance.

• ‘Filipp is ill’ corresponds to the overt utterance component—traditionally called its propositional content (alternatively, state of affairs).

To be able to refer to the three components, we shall use the terms proposed by the English philosopher R. Hare (cf. Hare 1949, 1970). He coined the term ‘neustic part’ of an utterance to define its speech act type, ‘tropic part’ of an utterance to define its propositional attitude, and the term ‘phrastic part’ of an utterance to define its propositional content (see Table 1). They will be used in our classification of SAs below.

Table 1 Russian declarative superstructure

The three utterance components yield three types of SAs:

• neustic adverbials,

• tropic adverbials, and

• phrastic adverbials.

Neustic adverbials have the greatest possible scope in that they are outside the superstructure itself and thereby have the entire superstructure within their scope. Tropic adverbials have a medium scope in that they have two of the three components (i.e. tropic and phrastic) within their scope. Phrastic adverbials have the least possible scope in that they do not range over the two other utterance components (i.e. neustic and tropic), but only over the last one, i.e. only the propositional structures themselves are within the scope of phrastic adverbials and function as their possible domain. In short, neustic adverbials like vpročem ‘incidentally’ are long-range adverbials, tropic adverbials like konečno ‘of course’ medium-range adverbials, and phrastic adverbials like teper’ ‘now’ are short-range adverbials within the class of SAs, which is normally more or less iconically reflected by position, cf.:

• (1) Vpročem ₁, konečno ₂, teper’₃ trudno sudit’, kto byl prav.

  ‘Incidentally ₁, right now ₃ it is of course ₂ difficult to judge who was right.’^{2 , 3}

The notion of scope is thus syntagmatically defined, and the three-way distinction introduced above is a possible division of a scope continuum.

In our description of SAs, we will begin with adverbials that have the greatest possible scope and end with those that have the least possible scope. It should be noted that in principle all Russian neustic and tropic adverbials are either followed by a comma or encircled by commas. There are, as one could imagine, some exceptions to this general rule, e.g., vse že ‘after all’ and javno ‘evidently’, and furthermore the untranslatable -de and -to which are clitics and therefore natural exceptions (see Sect. 1.2.2). Although this punctuation rule is not followed every time as will become apparent from some of the examples presented in the following paragraphs, it is followed in the vast majority of cases, which makes it quite simple for the foreign reader to discern these two subclasses of SAs in written discourse, at least. As to the members of the phrastic class, their comparatively broad scope is reflected by their front position in the sentence—a factor that makes them relatively easy to identify and distinguish from the same lexical units as non-SAs.

It should, however, be emphasized that in all other respects SAs present difficulties for foreign language learners, especially for whom English is their first language. Many of the Russian adverbials that are examined below either do not have an exact English counterpart or completely lack one. The reason for this is the specific grammatical system in English where modal verbs like must have some of the functions of Russian adverbials, like, for instance, dolžno byt’ (lit. ‘must be’). This is yet another motive for providing the reader with exact paraphrases as will be provided below—the main motive is to overcome the structural differences between the two languages that do not appear just from viewing an English translation. If an adverbial completely lacks an English counterpart, we shall mark it ‘Ø’; if it has no exact counterpart, we shall propose an English adverbial that is close in meaning to the Russian one and mark it in small capitals (for instance, pravda ‘true’). It should be noted, however, that even if we are able to introduce an English adverbial as a lexical equivalent to a Russian adverbial, this does not necessarily mean that there is a one-to-one semantic and functional correspondence between them. That is why translations of some examples presented below may be odd-sounding to the English reader in those cases when we preferred a translation which was adequate from the point of view of the content, to an idiomatic well-formed English counterpart.

1.2 Introducing the tropic component

1.2.1 On the importance of ‘It is true’

Normally, one would think that languages possess the same kind of declarative superstructure, but this need not be the case. It can be argued that different types of languages possess different kinds of declarative superstructures. According to Durst-Andersen (1995a, 2011) Russian is a so-called ‘reality-oriented language’, the grammar of which reflects situations in reality and their internal structure. That Russian belongs to this communicative supertype can be traced back to the distinction between the perfective and imperfective aspects which present an action as an event and as a process, respectively, i.e. they represent two distinct types of situations. The two aspectual forms manage to refer to different types of situations by involving different truth value assignments to the propositional content involved. By choosing the aspectual form the Russian speaker continuously commits himself tacitly to the truth or falsity of the propositional contents involved. It can be claimed that the tropic component of Russian, i.e. ‘it is true’, has adapted itself to the aspectual structures within the phrastic component, i.e. ‘what is true’. This means that the tropic component plays an extremely important role.

The consequence is that every time a Russian speaker wants to produce an utterance, he must make sure that the propositional content of the utterance (be that a state, activity, process or event description) is in agreement with ‘I hereby say it is true’, which functions as a kind of frame for all declarative utterances in the indicative mood.⁴ All this not only has an impact on the linguistic system and internal structures of Russian, for instance, on the Russian case and mood systems (cf. Durst-Andersen 1996, 2005), but also on the Russian way of communicating (cf. Durst-Andersen 2011; Lorentcen 2012).

1.2.2 When ‘It is true’ is cancelled

Apart from having specific means to stipulate the truth of a statement (for instance, dopustim ‘let us assume’, predpoložim ‘let us suppose’, and others) Russian has at its disposal four particles that in fact cancel the assignment of truth value to the propositional content involved, viz. mol, deskat’, -de and jakoby. They have no exact counterparts in English or in any other language, and it is because of this that we have chosen to pay special attention to them. These particles are apparently more frequent in modern Russian than it is normally assumed. Apart from being used in stylized vernacular language, they seem now to be fairly common in oral speech, modern fiction, social and political essays, blogs and online diaries, all of which are characterized by a casual speech style (cf. Arutjunova 2000; Plungjan 2008; for a different view see Markasova 2009).

All four particles denote ‘represented speech and thought’ (cf. Banfield 1982) and are used by the speaker to introduce another person’s speech or thought within a clausal utterance—‘another person’ may as well be the speaker himself, see, e.g., (2):

• (2) A ja emu govorju: ja, mol, ee tam ne videl.

  ‘And then I say to him: I didn’t see her there, I said.’

• (3) On obidelsja i stal menja uverjat’, čto ja, mol / deskat’ / -de / jakoby ne tak ego ponjal.

  ‘He felt insulted and tried to convince me that I, so he said / claimed, had misunderstood him.’

Utterances containing these particles—cf., e.g., (3)—thus inform the hearer that the speaker cannot commit himself to the truth of the statement encompassed in the sentence and signal a represented speech or thought situation. Therefore we get a specific superstructure, where ‘I hereby say it is true’ is replaced by ‘I hereby say that X said, thought, meant, suggested, claimed… / it was said, thought, meant, suggested, claimed…’. It is for this reason that we call them ‘basic tropic modifiers’. The differences in meaning and in use between the particles mol, deskat’, -de and jakoby are subtle and not easy to grasp (cf. Vinogradov 2001[1947]; Otin 1966; Kolodeznev 1969; Arutjunova 1990, 2000; Perfil’eva 1991, 1992; Kamju 1992; Baranov 1994; Paillard 1994; Rakhilina 1996; Levin-Steinmann 1997; Grenoble 1998; Plungjan 2008; Daiber 2010; Padučeva 2011; Kopotev 2014; and others).

We have taken the fact that they may co-occur with each other as a starting point, see, e.g., (4):

• (4) […] i mnogie togda govorili za spinoj i v glaza, čto ja, mol, deskat’, podmazyvajus’ v druz’ja.⁵

  ‘[…] at the time many people said behind my back and directly to me that I was thought to be a kind of person who forces my friendship upon others.’

Based on this we suggest that the particles in question enter into two subcategories of represented speech markers that relate differently to what is said or thought by another person.

1.2.3 On the two subcategories of represented speech

The first subcategory is made up of mol and deskat’ which are used by the current speaker and do not involve his or her attitude to what is being said or thought by the represented speaker. Mol solely denotes the ‘wording’ of the represented speech or thought event, i.e. what was uttered. Deskat’ denotes the ‘content’ of the represented speech or thought event, i.e. what was stated. This explains why mol may co-occur with deskat’, cf. example (4). In this example, the former mol renders the words being uttered by the represented speaker, while the latter deskat’ points to the content of the utterance. In other words, the function of mol is more or less equivalent to what Austin (1965, p. 100 ff.) calls a locutionary act, whereas the function of deskat’ is more or less equivalent to what he calls an illocutionary act.

The second subcategory is made up of -de and jakoby which are used by the current speaker to express his or her attitude towards the represented speech or thought event. The particle -de is used to express the current speaker’s sceptical attitude towards the represented speaker, see, e.g., (5):

• (5) […] akademik Bextereva, v konce 80-x godov zajavila, čto ee, mol, ded posle medicinskogo osmotra Stalina nazval ego paranoikom, za čto i byl-de otravlen.⁶

  ‘[…] in the late 80’s academy member Bextereva said that, after a medical examination, her grandfather [so she said] had diagnosed Stalin as paranoid for which he was poisoned, she claimed.’

The particle jakoby is used to express the current speaker’s disagreement with the content of the represented speech or thought, see, e.g., (6):

• (6) Samaja grandioznaja illjuzija, peredavaemaja iz pokolenija v pokolenie,—ėto to, čto čelovek jakoby znaet sam sebja.⁷

  ‘The greatest illusion passed down from one generation to another is that people know themselves’ (so it is said, but I do not believe in it).

Our analysis of the four particles may thus explain why we have utterances containing chains of particles which are completely untranslatable:

• (7) I on, učitel’ literatury po professii, nagovoril mnogo čego. A Voloxovu, mol, deskat’, jakoby, dali ėto posmotret’ po bol’šomu blatu, i on p’esu-to i napisal.⁸

  ‘And having a degree in literature he [serial killer Čikatilo] managed to tell a lot of stories about his life and tape them. There is a rumour that using his connections Voloxov had seen all this stuff and had written the play on that basis (i.e. it was said so, but the speaker did not believe in it).’

It should be mentioned that jakoby seems to have gained a new function,⁹ cf. (8):

• (8) SMI: Basmannyj sud arestoval jakoby čast’ imuščestva Naval’nogo. Basmannyj sud Moskvy arestoval imuščestvo oppozicionnogo politika Alekseja Naval’nogo. Takim obrazom, bylo udovletvoreno xodatajstvo Sledstvennogo komiteta.¹⁰

  ‘Media: Basmannyj court has allegedly seized some of Naval’nyj’s property.

  Moscow’s Basmannyj court has seized the property of opposition figure Aleksej Naval’nyj. The petition of the Russian Investigative Committee was thus granted.’

The context for this specific use of jakoby reveals why in this case it means ‘Somebody says so’ and not ‘Somebody says so, but according to me it is not true’. In this type of contexts the normal role of what we call the current speaker is cancelled, and the function of jakoby becomes more or less identical to the function of mol. Therefore jakoby can be combined with deskat’, see, e.g., (9):

• (9) I pojavilas’ uže informacija, čto jakoby, deskat’, predprinimajutsja usilija, čtoby v kačestve kandidata v vice-prezidenty SŠA debjutirovala Kondoliza Rais.¹¹

  ‘We got the information that efforts, so it was said, were made to promote Condoleezza Rice as vice-president of the USA.’

2 Neustic adverbials

Neustic adverbials operate on the level of the speech act component ‘I hereby say’ in utterances that have a declarative sentence form and on the component ‘I hereby ask’ in connection with utterances that have a non-declarative sentence form, i.e. an interrogative or an imperative sentence form. This subclass of SAs naturally divide into two groups, namely into connectives, i.e. adverbials that relate to previous discourse and thereby connect the utterance to the preceding oral or written discourse unit, and non-connectives, i.e. adverbials that lack this quality and only point forward in the text.

2.1 Connectives

In a way, connectives, also called discourse markers or conjuncts by others, are a sort of hyper-SAs because they operate on the text level. This type of adverbials is composed of two distinct subtypes, viz. concessives and non-concessives.

2.1.1 Concessives

Concessives are used to correct somebody’s belief or knowledge by admitting that something was the case or is the case. The subgroup of concessives consists of adverbials such as vse-taki ‘after all’, vse že ‘after all’, vse ravno ‘after all’, kak-nikak ‘after all’, tem ne menee ‘nevertheless’, odnako ‘however’, naprotiv ‘on the contrary’, naoborot ‘quite the contrary’, nesmotrja na ėto ‘in spite of that’, pravda ‘true’, and others. Let us first look at a couple of examples which involve a state description so as not to complicate things more than necessary:

• (10) Ja vse-taki ne durak.

  ‘After all, I am not an idiot.’

The utterance in (10) has the following paraphrase (10′):

• (10′) I hereby admit that there were grounds for one’s believing in the truth of the statement ‘I am an idiot’, but I hereby say that it is true that I am not an idiot.

The meaning of vse-taki is rather complex, and due to this complexity it takes many years for Russian children to start using it. Note that the concession applies to the preceding discourse which presupposes a discussion about one’s being a fool. Let us now compare (10) with (11) and (12), which embed other concessives in the same context:

• (11) Ja vse že ne durak.

  ‘I am not an idiot, after all.’

• (11′) I hereby admit that there were grounds for my believing in the truth of the statement ‘I am an idiot’, but I hereby say that it is true that I am not an idiot.

• (12) Ja vse ravno ne durak.

  ‘After all, I am not an idiot.’

• (12′) I hereby admit that there were grounds for your believing in the truth of the statement ‘I am an idiot’, but I hereby say that it is true that I am not an idiot.

The three adverbials have a common element that is responsible for their belonging to a specific subgroup of concessives, i.e. the concession applies to the preceding discourse. However, they also differ from one another. (11) involves the speaker’s belief, (12) involves the hearer’s belief, while (10) is indifferent with respect to the speaker and the hearer and therefore can be used to point to the general belief, i.e. it may reflect the voice of society. It does not seem to be a mere coincidence that the micro-system is made up of exactly three members which point to the speaker (cf. (11)), the hearer (cf. (12)) and the reality or the world as such (cf. (10)), respectively, i.e. the three obligatory participants in a communication situation.

It appears that this three-way distinction within SAs is a characteristic feature of Russian. Therefore, we have to create terms for this distinction. We shall use ‘first person adverbials’ for adverbials like vse že, ‘second person adverbials’ for adverbials like vse ravno and ‘third person adverbials’ for adverbials like vse-taki which is unmarked and which may act as a substitute for the two others, but not vice versa, cf.:

• (13) Ja vse-taki / ^∗ vse ravno / ^∗ vse že ne durak, čtoby ne zametit’, kak on izmenilsja.

  ‘After all, I am not a fool not to notice how he has changed.’

The consequence of this structure is, for instance, that vse ravno is used when the speaker enters into discussion with the hearer by being in opposition to his or her belief. Vse že does not include this element of discussion with the hearer, but instead an internal discussion with the speaker himself. Note, for instance, that the sentence in (11) can be said by the speaker with a smile on his lips, whereas this is certainly not the case for (12). Thus it seems that just as one can conjugate a verb according to number and person because of the natural constitution of a situation referred to, in which a first, a second and a third person may be agent, patient or recipient, some SAs are used to inflect an utterance according to the communication situation in which first, second and third person function as speaker, hearer and reality, respectively. If they do and are alike in all other respects, they form a lexical micro-system (a paradigm). In Russian tradition the lexical micro-system is often called a synonymous group without any consistent differentiation between the individual members and with the implicit assumption that they exclude one another. But this view is challenged by the fact that they may occur together in the same utterance, e.g.:

• (14) Ja vse-taki ₁ vse ravno ₂ ostajus’ patriotom.

  ‘Nevertheless ₁, I am still ₂ a patriot.’

The SA kak-nikak, which is also translated as ‘after all’, differs from the adverbials just mentioned by referring not to belief, but to common knowledge, cf.:

• (15) No i moim mneniem ne sleduet prenebregat’ – ved’ ja, kak-nikak, professor, nejroxirurg.¹²

  ‘You shouldn’t ignore my opinion either; as you know, I’m a professor of neurosurgery after all.’

Kak-nikak may also be combined with any of the three adverbials examined above, cf.:

• (16) Kak-nikak ₁ vse ravno ₂ ėto stress.

  ‘In any case ₂, this is stressful, you know ₁.’

In contrast to all the adverbials mentioned above the concession involved in pravda ‘true’ does not serve as background for making an assertion, but serves as foreground in which the speaker admits something against the background of the previously made assertion:

• (17) Ėto očen’ xorošie tufli, pravda, očen’ dorogie.

  ‘These shoes are very nice, though very expensive.’

The paraphrase shows that pravda is multiform (i.e. it covers all three persons) and converse to vse že, vse ravno, vse-taki (cf. Nørgård-Sørensen 1992, p. 221; also see the discussion in Pajar 2003):

• (17′) I have just said (what you or somebody else said, i.e.) that it is true that the shoes are very nice, but I hereby admit that they are very expensive.

Pravda can also be used in a combined function in which it points backwards and forwards in the text, i.e. points to an utterance preceding it and to one that immediately follows it. In this combined function, it is always used together with adversative connectives, e.g., no, concessive connectives, e.g., xotja, and / or concessive adverbials, e.g., vse-taki; see, e.g., (18) and (19):

• (18) No poka Galja ničego ne znala o predstojaščem sokrašenii. Potom, pravda ₁, ona vse-taki ₂ uznala […] No ₃ k ėtomu vremeni Galja uže sdelala odnu bol’šuju ošibku […]¹³

  ‘But at that point Galja knew nothing about the reduction that was coming. Later on, though ₁, she learnt about it, after all ₂ […] But ₃ by then Galja had already made a big mistake […]’

• (19) Vmesto togo čtoby prjamikom otpravljat’sja na ee ssudnyj sčet, oni uxodili na čužie sčeta, xotja ₁ potom, pravda ₂, vse-taki ₃ vozvraščalis’.¹⁴

  ‘Instead of being sent directly to her own credit account, the money was sent to accounts belonging to others, although ₁ it was returned later, after all ₃, I should say ₂.’

Since concessive adverbials refer to the belief or common knowledge of the hearer / reader or society as such in one way or another, they can be used to play on his or her possible reservations about the speaker’s / author’s own descriptions. This is not true of non-concessives, which will be examined in Sect. 2.1.2.

2.1.2 Non-concessives

Non-concessives are used to correct or extend the author’s own descriptions. They include vpročem ‘incidentally’, meždu pročim ‘incidentally’, kstati ‘by the way’, ved’ ‘you know’, krome togo ‘apart from this’, k tomu že ‘add to this, moreover’, pričem ‘in addition’, pritom ‘besides’, vdobavok ‘in addition’, v pridaču ‘into the bargain’, sverx togo ‘furthermore’, pomimo togo ‘besides’, and others. Let us analyze the first three mentioned adverbials, which do not all seem to have exact English equivalents:

• (20) Vpročem, on mne ne nravitsja.

  ‘Incidentally, I don’t like him.’

• (21′) I hereby say something that just occurred to me and that I consider to be relevant to our discourse topic, namely that it is true that I don’t like him.

• (21) Meždu pročim, on mne ne nravitsja.

  ‘Incidentally, I don’t like him.’

• (21′) I hereby say something that just occurred to me and that you may consider to be relevant, namely that it is true that I don’t like him.

• (22) Kstati, on mne ne nravitsja.

  ‘By the way, I don’t like him.’

• (22′) I hereby say something that just occurred to me because of the previous discourse, namely that it is true that I don’t like him.

The paraphrases show what they have in common and what distinguishes them from one another. All three adverbials include the component ‘I hereby say something that just occurred to me’, which is the reason why they comprise a small subgroup of so-called parenthetical adverbials. As we can see, vpročem implies that the speaker himself finds the following information relevant to the discourse topic; meždu pročim implies that the speaker thinks that the hearer will find it relevant, while kstati is devoid of this type of information—that is why it is used in connection with pure associations. We shall argue that vpročem is a first person adverbial, meždu pročim is a second person adverbial, and kstati is a third person adverbial. The former two are marked, whereas the last is unmarked. That is why an utterance like the one in (23) feels odd, whereas an utterance like the one in (24) feels perfectly natural:

• (23) Meždu pročim, ėto dlja tebja ne važno.

• (24) Vpročem, ėto dlja tebja ne važno.

It is interesting to note that Dostojevskij made extensive use of neustic adverbials, cf. the following example:

• (25) My uže skazali sejčas, čto sam general xotja byl čelovek ne obrazovannyj, a, naprotiv, kak on sam vyražalsja o sebe, ‘čelovek samoučnyj’, no byl, odnako že, opytnym suprugom i lovkim otcom. Meždu pročim, […]¹⁵

  ‘I just said that although the general was not too well educated – on the contrary, he refers to himself as ‘a self-made man’ – he was, nevertheless, an experienced husband and a smart father. Incidentally […]’

The consequence for his readers is that they are forced to change text level all the time and to adapt themselves to different ‘voices’ (one of his favourite adverbials, viz. daže ‘even’, which can be paraphrased as ‘one would not expect’, will, for instance, be the voice of society) in line with the constant shift in first, second and third person adverbials from the subgroups of concessives and non-concessives. In this way, the reader will soon feel that he is placed in a whirlwind of thoughts and reservations made by odnako, naprotiv, etc. Having just begun a novel, the reader will feel just as confused as its hero. Seen from this perspective it is understandable that the works of Dostojevskij served as the basis for Baxtin’s theory of mnogogolosie ‘polyphony’: the author’s voice, the reader’s voice, the voice of society (see also Ducrot 1989 who built a whole theory on the basis of Baxtin’s observations).

2.2 Non-connectives

We now proceed to leave the so-called hyper-SAs behind in favour of the first group of proper SAs which consists of bukval’no govorja ‘literally speaking’, vernee govorja ‘correctly speaking’, voobšče govorja ‘generally speaking’, govorja drugimi slovami ‘in other words’, govorja inymi slovami ‘in other words’, drugimi slovami ‘in other words’, grubo govorja ‘roughly speaking’, inače govorja ‘in other words’, inymi slovami ‘in other words’, korotko govorja ‘in short’, koroče govorja ‘in short’, kstati govorja ‘by the way’, mjagko govorja ‘putting it mildly’, odnim slovom ‘in one word’, otkrovenno govorja ‘frankly speaking’, po pravde govorja ‘truly speaking’, poprostu govorja ‘simply speaking’, po sovesti govorja ‘speaking in all good conscience’, po spravedlivosti govorja ‘speaking in all fairness’, po suti govorja ‘essentially’, po suščestvu govorja ‘essentially’, po česti govorja ‘speaking quite honestly’, pravdu govorja ‘truly speaking’, pravil’nee govorja ‘correctly speaking’, prošče govorja ‘simply speaking’, ser’ezno govorja ‘seriously speaking’, sobstvenno govorja ‘strictly speaking’, strogo govorja ‘strictly speaking’, točnee govorja ‘speaking more precisely’, čestno govorja ‘honestly speaking’, bukval’no skazat’ ‘to put it literally’, vernee skazat’ ‘to be more accurate’, inače skazat’ ‘to put it in another way’, k slovu skazat’ ‘by the way’, kstati skazat’ ‘by the way’, lučše skazat’ ‘to say / put it in a better way’, otkrovenno skazat’ ‘to be frank’, po pravde skazat’ ‘to say / speak the truth’, poprostu skazat’ ‘to put it in a simple way / simply stated’, po sovesti skazat’ ‘in all good conscience’, po suti skazat’ ‘essentially’, po česti skazat’ ‘to be quite honest’, pravdu skazat’ ‘to tell / speak the truth’, pravil’nee skazat’ ‘to be more correct’, prošče skazat’ ‘to put it in a more simple way’, skoree skazat’ ‘to be more accurate / more accurately’, točnee skazat’ ‘to be more precise / more precisely’, čestno skazat’ ‘to be honest / honestly’, and others. They all operate directly on the speech act and are also called ‘speech act adverbials’ (or ‘style disjuncts’ by Greenbaum 1969).

The vast majority of the adverbials mentioned above can be divided into two subgroups. The first subgroup consists of all adverbials involving govorja. The govorja-adverbials all denote a mental state for which the speaker vouches when producing the utterance in question, cf. the following utterance having a declarative sentence form:

• (26) Otkrovenno govorja, on mne ne nravitsja.

  ‘Frankly speaking, I don’t like him.’

• (26′) I am frank as I hereby say that it is true that I don’t like him.

However, if used without govorja some of these adverbials may function as non-SAs in a performative utterance, cf. (27). Note that (27) may also be used non-performatively as a characterization of oneself:

• (27) Ja otkrovenno priznajus’ v svoix ošibkax.

  ‘I frankly admit my mistakes.’

• (27′) I am frank as I hereby make you know that I am committed to the truth of the following statement: ‘I am at fault’.

The second subgroup consists of adverbials involving skazat’ ‘to say’. They are used to modify the way in which a previous statement is expressed without violating its truth:

• (28) Otkrovenno skazat’, on mne ne nravitsja.

  ‘To be frank, I don’t like him.’

• (28′) I hereby say frankly to you that it is true that I don’t like him.

The skazat’-adverbials are more restricted in their usage and can only occur in connection with utterances that have a declarative sentence form in contrast to the govorja-adverbials, cf. (29) and (30):

• (29) Koroče govorja, sjad’ i uspokojsja!

  ‘In short, sit down and relax!’

• (30) Voobšče govorja, vy s ėtoj teoriej znakomy?

  ‘Are you in general acquainted with this theory?’

Both govorja-adverbials and skazat’-adverbials are often used to correct single elements of an utterance, see (31):

• (31) My poznakomilis’ s sosedjami, ili, vernee, s sosedkami.

  ‘We have met the neighbours, or rather the female ones.’

This example is ambiguous and can be interpreted as ‘if I have to be accurate’ in the case of vernee govorja, but as ‘to be more accurate’ in the case of vernee skazat’.

3 Tropic adverbials

The second subclass of SAs operates on the level of the tropic component, i.e. ‘It is true’, in reality-oriented languages like Russian. Tropic adverbials divide into three distinct subtypes, viz. attitudinal adverbials, modal adverbials and what we subsume under the heading ‘other tropic adverbials’. Not surprisingly, it is exactly at the tropic part of the utterance that the differences between Russian and English are most clearly displayed. This is one of two reasons why we chose to devote more space to the following paragraphs. The other reason has to do with grammar and lexicon, which from our point of view are intimately connected. In this respect, modal adverbials may function as a model example (cf. Sect. 3.2).

3.1 Attitudinal adverbials

All members of this subgroup of tropic adverbials denote the speaker’s attitude towards some propositional content that is presupposed to be true. In short, none of them has reservations about the truth of the statement involved (this does not exclude that the speaker might believe in the opposite of what is stated, cf. (33) below). The subgroup includes, for instance, k sčast’ju ‘fortunately’, k sožaleniju ‘unfortunately’, k moemu udivleniju ‘to my great surprise’, k vseobščemu vosxiščeniju ‘to everybody’s delight’, k obščemu izumleniju ‘to everybody’s astonishment’, slava Bogu ‘thanks God’, javno ‘obviously’, navernjaka ‘certainly’, točno ‘certainly’, nesomnenno ‘undoubtedly’, bessporno ‘indisputably’, bezuslovno ‘absolutely’, estestvenno ‘naturally’, konečno ‘of course’, (samo soboj) razumeetsja ‘it goes without saying’, and some others. Because the truth of the statement is presupposed, none of the above adverbials can be used in interrogative or imperative utterances.

However, in this subgroup we also include neuželi ‘really’ and razve ‘really’ both of which are used in utterances that have an interrogative sentence form, but these utterances function as rhetorical questions or questions posed to oneself as part of an internal monologue, and not as proper questions:

• (32) Neuželi ona sdala ėkzamen?

  ‘Did she really pass the exam?’

• (33) Razve ona sdala ėkzamen?

  ‘Did she really pass the exam?’

Although the English translations of (32) and (33) are identical, the Russian utterances are quite different in meaning and use:

• (32′) I hereby ask you to reconfirm the truth of the statement ‘She passed the exam’, the truth of which I do not dare to believe in.

• (33′) I hereby ask you to reconfirm the truth of the statement ‘She passed the exam’, the truth of which I do not believe in.

The paraphrases in (32′) and (33′) show that neuželi denotes a positive attitude, whereas razve denotes a negative attitude towards a statement that is presupposed to be true because of the preceding discourse (for a more detailed analysis, see Maier 1994, pp. 52–58).

Apart from these two adverbials, most attitudinal adverbials do not seem to be problematic for foreign speakers of Russian. Therefore, we restrict ourselves to some illustrative examples, cf. (34)–(35):

• (34) K sožaleniju, ona ne stala professorom.

  ‘Unfortunately, she didn’t become a full professor.’

• (34′) I hereby say it is unfortunately true that she is not a full professor now.

• (35) K moemu udivleniju, on stal professorom.

  ‘To my great surprise, he became a full professor.’

• (35′) I hereby say that to my great surprise it is true that he is a full professor now.

Some may, however, cause problems for foreign speakers due to their lexico-grammatical meanings. Thus, the three adverbials javno ‘obviously’, navernjaka ‘certainly’, točno ‘certainly’ are all relatively new members of the subclass of SAs. They form a specific micro-group of adverbials, which take their point of departure from the truth of a statement, but, simultaneously, strengthen it by showing the hearer where it comes from.

The adverbial javno ‘obviously’ marks that the speaker is confident about the truth of a statement, because he or she has direct evidence of what he or she is informing the hearer about, see (36):

• (36) Ona javno ne sdelala domašnee zadanie.

  ‘She obviously didn’t do her homework.’

It is stated that the speaker has direct evidence for what he or she is stating, i.e. she performed so badly in class that it becomes clear that she has not done her homework. The truth of the statement is based on the speaker’s direct visual experience of the person involved, which does not exclude the possibility of using other senses.

The adverb navernjaka ‘certainly’ points to the speaker’s personal background knowledge of the agent involved in the utterance:

• (37) Ona navernjaka ne sdelala domašnee zadanie.

  ‘She certainly didn’t do her homework.’

In other words, the speaker’s knowledge of the agent is so profound that he or she does not hesitate to 100 % (or close to 100 %) commit himself or herself to the truth of the statement.

The adverb točno ‘certainly’ is similarly felt to strengthen the truth of a statement, but neither due to the high reliability of visual perception as in (36), nor to the trustworthiness of the speaker’s student as in (37). (38) is the result of the speaker’s unsuccessful attempt to falsify the statement: Ona ne sdelala domašnee zadanie. That is why the hearer feels that the speaker is more than sure about his or her assertion.

• (38) Ona točno ne sdelala domašnee zadanie.

  ‘She didn’t do her homework for sure.’

The following three adverbials, viz. nesomnenno ‘undoubtedly’, bessporno ‘indisputably’, bezuslovno ‘absolutely’ constitute another micro-group. Although they seem to be used more or less freely and are more or less mutually interchangeable in the vast majority of contexts, they differ with respect to whether they include the speaker alone, cf. (39), the speaker and the hearer, cf. (40), or all other aspects, i.e. all what is included in the notion of reality, cf. (41):

• (39) On, nesomnenno, lučšij ispolnitel’ Baxa v mire.

  ‘He is undoubtedly the world’s best performer of Bach.’

• (40) On, bessporno, lučšij ispolnitel’ Baxa v mire.

  ‘He is indisputably the world’s best performer of Bach.’

• (41) On, bezuslovno, lučšij ispolnitel’ Baxa v mire.

  ‘He is absolutely the world’s best performer of Bach.’

In that way we can argue that nesomnenno ‘undoubtedly’ is a first person adverbial, bessporno ‘indisputably’ a second person adverbial, and bezuslovno ‘absolutely’ a third person adverbial.

The differences between estestvenno ‘naturally’, konečno ‘of course’ and (samo soboj) razumeetsja ‘it goes without saying’ seem to be founded on the same grounds. Although they all express the speaker’s attitude towards the presupposed truth of the statement, estestvenno ‘naturally’ does not involve the hearer or reality in contrast to konečno ‘of course’ which involves the hearer, cf. (42), (i.e. I knew it and you knew it), and in contrast to (samo soboj) razumeetsja which involves reality in the sense that the ‘naturalness’ of the truth is treated as being pure logic. In other words, estestvenno is a first person adverbial, konečno a second person adverbial, and (samo soboj) razumeetsja a third person adverbial:

• (42) Poezd, konečno, opozdal.

  ‘The train was, of course, delayed.’

Note that nesomnenno, bessporno, bezuslovno, estestvenno, konečno and razumeetsja can all be used instead of Da ‘Yes’ when answering a question positively (cf. Apresjan 1995, p. 139), see, e.g., (43):

• (43) Ona s ėtim spravitsja? – Nesomnenno! / Bessporno! / Bezuslovno! / Estestvenno! / Konečno! / Razumeetsja!

  ‘Will she be able to cope with it? – Undoubtedly! / Indisputably! / Absolutely! / Naturally! / Of course! / It goes without saying!’

3.2 Modal adverbials

The subgroup of modal adverbials, also called epistemic adverbials, includes, for instance, čaj ‘ø’, avos’ ‘ø’, nebos’ ‘ø’, vrjad li ‘ø’, požaluj ‘ø’, možet (byt’) ‘perhaps’ lit. ‘may be’, dolžno byt’ ‘ø’ lit. ‘must be’, verojatno ‘probably’, očevidno ‘obviously’, po-vidimomu ‘obviously’, kažetsja ‘Ø’, vidno ‘obviously’, vidimo ‘obviously’, skoree vsego ‘(the) most likely’ (among several possibilities), navernoe ‘most likely’ (I suppose), čego dobrogo ‘most likely’ (I am afraid), and others. As should be obvious from the list of Russian modal adverbials and their English translations or, perhaps more correctly, lack of translations, the two languages seem to be incommensurate on this point. The reason for this can be traced back to their belonging to different supertypes.

As a reality-oriented language (cf. Durst-Andersen 2011), Russian sharply distinguishes epistemic from non-epistemic modality or, in other words, belief from knowledge. Non-epistemic modality divides into (a) alethic modality which deals with knowledge of nature (e.g., what is possible or impossible in nature) and (b) deontic modality which is concerned with knowledge of laws of society (e.g., what is permitted or prohibited in society). Speaking of single actions, all alethic modes are expressed by the infinitive or imperative forms of the perfective aspect of action verbs, whereas all deontic modes are expressed by the infinitive or imperative forms of the imperfective aspect of action verbs, cf. Durst-Andersen (1992, 1995b); simplex verbs will always be ambiguous due to their simple propositional structures. In short, non-epistemic modality is part of the grammatical system of Russian in contrast to epistemic modality, which deals with belief or the ‘black box’, if you like. In Russian, all epistemic modes are expressed by lexical means by choosing among the members of the list of modal adverbials presented above.

In English, on the contrary, epistemic modality is part of the grammatical system. By using modal verbs the English speaker expresses what he or she considers to be highly probable, e.g., (44) or just probable, e.g., (45):

• (44) He must have left the house by two o’clock tomorrow.

• (45) He may have left the house by two o’clock tomorrow.

However, the grammatical means do not suffice to express all possible epistemic modes or values. In order to be able to do that, the English speaker must make use of modal adverbs, e.g., (46), so-called expressive verbs such as assume, e.g., (47), a combination of these and modal verbs, e.g., (48) or a combination of modal verbs and modal adverbs, e.g., (49):

• (46) Obviously, he is tired.

• (47) I assume that he is tired.

• (48) I assume that he must have been tired.

• (49) Obviously, he must have been tired.

All this suggests that Russian must have a relatively high number of modal adverbials at its disposal and explains why they cannot be properly translated into English lexical items. Even at the utterance level, it seems almost impossible to uncover the differences between the various Russian modal adverbials. That is to say, it is almost impossible to transfer the nuances of a Russian sentence with a modal adverbial into a normal English sentence.

Since Russian has ‘It is true’ as its tropic component, it goes without saying that the use of a modal adverbial can be taken as a sign of the speaker’s reservation about the truth of the propositional content involved:

• (50)

  

Due to the fact that the perfective aspect presupposes the activity description and asserts the state description of the action verb (50a) has, in fact, the status of a promise. By inserting navernoe in (50b) the speaker explicitly shows the hearer that something prevents him from committing himself a hundred percent to the statement involved. In (50b) the speaker thus refrains from giving a promise. In using navernoe he or she simultaneously points to the cause of his or her reservations, viz. he or she lacks some pieces of information. In other words, this adverbial has something to do with the speaker’s lack of knowledge. The notion of lack of certain knowledge together with its various sources is in fact the key to the understanding of the various modal adverbials in Russian. It appears that the use of modal adverbials in Russian is related to statements that express uncertain knowledge, i.e. beliefs that the speaker is not absolutely confident with. If he or she has no reservations at all, he or she just uses the indicative mood without inserting any modal adverbial. Let us take a look at some illustrative examples:

• (51)

  

Both examples involve statements, which cannot be classified as (certain) knowledge. This means that the speaker has no direct evidence for it, for instance, that Sergej looks tired. In Russian, direct visual experience is certain knowledge by definition (cf. (36)). In (51a) and (51b), we are dealing with indirect evidence. In (51b) the speaker expresses either his or her subjective judgement of Sergej’s behaviour, which gave him the impression that he was tired, or his own conclusion drawn on the basis of what he has heard from a witness who told him about Sergej’s behaviour or performance in general. His or her subjective conclusion is not based on a comparison and does not relate to other alternatives—that is why Russians often feel that vidimo denotes the only possible conclusion, i.e. the natural one under the given circumstances. When the speaker jumps to a conclusion without applying any rule, this can be considered to be a qualified guess. (51a) is stronger than (51b) because it is not a subjective judgement or a conclusion drawn from hearsay. The statement involved in (51a) is based on ‘deductive inference’, i.e. the speaker concludes ‘Sergej is tired’ by applying the case ‘Sergej is behaving in Y-way’ to the rule ‘If X behaves in Y-way, X is tired’. Thus, (51a) has the status of an objective conclusion, whereas (51b) represents a subjective one. They need not involve visual experience alone as becomes apparent in the following two examples:

• (52)

  

Both examples involve auditory experiences: (52a) is an objectively based conclusion (no other conclusions could be arrived at), while (52b) is a subjectively based conclusion, i.e. it is not based on any rule, only on the speaker’s impression. Due to the fact that both represent conclusions we shall name them ‘conclusives’. The reason for doing that is that there are two other adverbials which are used to explain or legitimate what is known to be a fact. These are očevidno and po-vidimomu which we have chosen to call ‘explicatives’. For illustrative reasons let us use the same example as we did for vidno and vidimo, see (53a,b):

• (53)

  

Both examples can be used in a context where somebody told the speaker in (53) that Sergej failed an exam. He or she has difficulty in accepting this and tries to look for an explanation that can account for the result, or, to put it in more logical terms, the premise that makes the conclusion valid. (53a) is based on abductive inference. In this example the speaker knows the result ‘Sergej performed below normal’ and is aware of the rule ‘if X is tired, X performs below normal’, and by applying the result to the rule, he or she arrives at an explanation as to why Sergej performed less well than normally, i.e. why he failed the exam: he was tired. (53b) also has the status of an explanation, but it does not involve any kind of rule: it is the speaker’s qualified guess. In other words, they both point backwards and are used as explicatives—not as conclusives. This also becomes evident in the following authentic examples:

• (54)

  

In (54a) the speaker tries to explain why the venetian blinds are closed, i.e. the statement ‘they are all asleep in the house’ is not a conclusion arrived at by looking at the windows. In the same way, ‘doctors’ in (54b) is not a conclusion arrived at by looking at white coats, but what seems to be a reasonable explanation as to why the two persons are wearing white coats. Again, we observe a difference in strength.

Thus, the adverbials vidno, vidimo, očevidno and po-vidimomu constitute a micro-system, which can be summarized in Fig. 1.¹⁶

Fig. 1

Forms of reality and forms of arguments

The adverbials dolžno byt’ and možet (byt’) form a micro-group of epistemic adverbials, which correspond to the English modal verbs when used to express epistemic necessity and epistemic possibility, i.e. what the speaker believes must be the case or may be the case, cf. (55)–(56′):

• (55) Dolžno byt’, oni sejčas rabotajut v laboratorii.

  ‘They must be working in the lab now.’

• (55′) I hereby say it must be true that they produce working activity in the lab now.

• (56) Možet (byt’), oni sejčas rabotajut v laboratorii.

  ‘They may be working in the lab now.’

• (56′) I hereby say it may be true that they produce working activity in the lab now.

The remaining modal adverbials are even more peculiar from the point of view of the English language since they also involve the speaker’s affirmative or negative assessment of the statement involved. This applies to požaluj, vrjad li / navrjad li / navrjad ¹⁷ and edva li, cf. (57)–(58′):

• (57) On, požaluj, sdast ėtot ėksamen.

  ‘He will pass the exam, I think.’

• (57′) I hereby say that if you ask me, I will say it is true that he produces an activity that is sufficient for the fact that he has the exam.

• (58) Vrjad li / edva li on pridet.

  ‘He hardly comes.’

• (58′) I hereby say that if you ask me, I will say on objective grounds / subjective grounds it is not true that he produces an activity, which is sufficient for the fact that he is here.

The difference between vrjad li and edva li seems to be similar to the distinction between vidno and očevidno, on the one hand, and vidimo and po-vidimomu, on the other, i.e. the difference between objective grounds and subjective grounds. In other words, both particles denote the speaker’s strong doubt about the realizability of a situation, most often in the future, but in addition to that edva li stresses the speaker’s lack of objective evidence.

Adverbials like čego dobrogo, ne daj Bog, ne roven čas etc. are all concerned with the speaker’s fear or anxiety and are so specific that their meaning can only be paraphrazed, cf., e.g., (59) and (59′):

• (59) Ona, čego dobrogo, zaboleet i zavtra ne pridet.

  ‘She may well get sick and won’t turn up tomorrow.’

• (59′) I hereby say that I am afraid that it is most probably true that some activity will be sufficient for the fact that she is sick and is not present at her job tomorrow.

The adverbial skoree vsego is used when, among several possibilities, the speaker has chosen the one most likely to happen or most likely to be the case. This component seems also to be involved in avos’ and in nebos’, which are used when speaking about future events. They differ from skoree vsego by expressing simultaneously the speaker’s attitude towards the desirability of the occurrence of the event:

• (60) On, skoree vsego, segodnja ne pridet.

  ‘He will most probably not turn up today.’

• (61)

  

The remaining adverbials (e.g., verojatno ‘probably’, vozmožno ‘possibly’, objazatel’no ‘definitely’, etc.) seem less problematic and have therefore not been dealt with.

3.3 Other tropic adverbials

The last group of adverbials is in fact a collection of adverbials that do not fit into the two other groups, but, nevertheless, belongs to the subclass of tropic adverbials. It consists of two subgroups: The first subgroup includes so-called evaluative adverbials that express the speaker’s evaluation of the truth of the statement involved, e.g., nakonec ‘finally’, v konce koncov ‘at last’, v konečnom sčete ‘in the last resort’, kakim-to strannym obrazom ‘oddly’, kak ni stranno ‘how strange it may be’, and a few others. It is important to stress that while English has a productive use of evaluative adverbs such as ironically, strangely, tragically, surprisingly, interesting (enough), etc., it is impossible to derive adverbs of this type in Russian. The reason is purely grammatical: Russian uses so-called predicatives with a following that-clause as the subject, cf. (62):

• (62) Stranno / interesno / žal’, čto ona ne prišla.

  ‘[It is] strange / interesting / a pity that she didn’t turn up.’

Let us illustrate the function of this subgroup with the following example:

• (63) V konce koncov on prodal staruju mašinu.

  ‘He finally succeeded in selling the old car.’

• (63′) I hereby say it is finally true that he produced an activity which was sufficient for the fact that the old car is gone.

The second subgroup includes the so-called point-of-view adverbials that mark the viewpoint that is adopted by the speaker when committing himself implicitly to the truth of the propositional content involved. It consists of a small number of adverbials, viz. v obščem ‘in general’, v celom ‘on the whole’, v obščem i celom ‘all in all’, po-moemu ‘I think’, po-tvoemu ‘you think’, po moemu / tvoemu / vašemu / ee / ego / ix mneniju ‘in my / your / her / his / their opinion’, s moej / tvoej / vašej / ee / ego / ix točki zrenija, ‘from my / your / her / his / their point of view’, v obščej složnosti ‘in sum’, v itoge ‘as a result’, po sluxam ‘from hearsay’, kak izvestno ‘as is well known’, and a few others. Here it must also be pointed out that English adverbs like mathematically / logically / sociologically speaking (they are treated as the seriously speaking type) have no exact counterparts in Russian. They may be translated into Russian by means of either s točki zrenija, e.g., s matematičeskoj točki zrenija / s točki zrenija matematiki ‘from a mathematical point of view’, or s pozicii, e.g., s pozicii matematiki ‘from the standpoint of mathematics’, but they completely lack the naturalness of the English way of expressing it. Let us have a look at some examples, which at the same time illustrate the difference between two so called synonyms, cf. (64)–(65):

• (64) Po-moemu / ^∗ po moemu mneniju, on tebe ne nravitsja.

  ‘I think you don’t like him.’

• (64′) I hereby say that if you ask me it is true that you do not like him.

• (65) Po moemu mneniju, ona vedet sebja pravil’no.

  ‘In my opinion she is behaving correctly.’

• (65′) I hereby say that in my opinion it is true that she behaves correctly.

In view of the fact that Russian and English belong to different communicative supertypes (cf. Durst-Andersen 2011), it does not seem to be a coincidence that Russian has a strong emphasis on modal adverbials in contradistinction to the English emphasis on evaluative adverbials or point-of-view adverbials.

4 Phrastic adverbials

This subclass of SAs has the least possible scope and ‘only’ has the propositional structure within its scope. They relate the truth of the entire statement to a particular point or period in time or to a certain place or area. In that way it can be said that they divide into two natural classes.

4.1 Temporal adverbials

This subgroup consists of adverbials such as teper’ ‘now’, zavtra ‘tomorrow’, v dva časa ‘at two o’clock’ etc., inogda ‘sometimes’, vsegda ‘all the time’, postojanno ‘permanently’, často ‘often’, vdrug ‘suddenly’, neožidanno ‘unexpectedly’, vnezapno ‘suddenly’, etc. Due to their scope, they partly correlate with the choice of propositional structure in the sense that some of them require the perfective aspect, whereas others require the imperfective aspect. Let us take a closer look at (66):

• (66) V dva časa Ivan napisal poslednjuju sms.

  ‘At two o’clock Ivan wrote his final text message.’

• (66′) I hereby say it is true that at two o’clock he produced an activity which was sufficient for the fact that the text message exists on world-location.

That the temporal adverbials operate on the the level of the phrastic component and have the entire sentence within their scope clearly becomes apparent from the following example, which is unacceptable to Russian speakers, cf. (67):

• (67) ^† V dva časa Ivan napisal stat’ju.¹⁸ ‘At two o’clock Ivan wrote an article.’

• (67′) ^†I hereby say it is true that at two o’clock he produced an activity which was sufficientfor the fact that an article exists on world-location.

The utterance is unacceptable (but not ungrammatical) in Russian because of the transitive nature of v dva časa: (67) means that Ivan not only succeeded in producing some article-writing activity at two o’clock, but also succeeded in finishing this article at two o’clock. That is considered to be impossible and therefore (67) is not accepted. Once again, we realize that an English translation cannot be used as the point of departure for our understanding of Russian. The English language has its own statement models, and these models have their own implicatures built in. The transitive nature of the Russian temporal adverbials is also evident in the following example:

• (68) Byvalo, Andrjuša zaxodil k nam posle zanjatij.¹⁹

  ‘There was a time when Andrjuša would drop by after classes were over.’

• (68′) I hereby say it is true that in a past period of time it was customary that …

Although byvalo involves habitual use, it does not exclude the possibility of using the perfective aspect. This would, however, not be possible in our next example, cf. (69):

• (69) Na lekcijax ona vsegda vnimatel’no slušaet prepodavatelja.

  ‘When attending lectures she always listens carefully to the professor.’

• (69′) I hereby say it is true that whenever she is at lectures she is attentive as she produces an activity that is intended to be sufficient for the fact that the professor’s talk exists for her as an auditory experience.

Apart from the aspectual consequences of using members of this subgroup of SAs, they are unproblematic from the point of view of foreign language learning. Note, however, that in Russian they must be used as front place elements, cf. (70) and (71):

• (70) Skoro ja vernu tebe knigu.

  ‘Soon I will return the book to you.’

• (71) Ja vernu tebe knigu (očen’) skoro.

  ‘I will return the book to you (very) soon.’

(70) involves a SA that functions as topic. (71) can only be understood as a non-SA, i.e. only the state description ‘you will soon have the book’ will be within the scope of skoro.

4.2 Areal adverbials

This subgroup consists of a whole range of adverbials such as doma ‘at home’, za obedom ‘sitting around the dining table’, v Pariže ‘in Paris’, etc. which all have to be used as front place elements in order to fulfill the function of a SA. Compare the SA doma in (72a) which gives a situation description that involves a certain location and the non-SA doma in (72b) which gives a characterization of a person:

• (72)

  

Moreover, we find phrastic adverbials which have a more abstract areal semantics, say, for instance, in the intellectual area, cf. (73)–(75):

• (73) Umstvenno Marija Stjuart načala razvivat’sja neobyčno rano.

  ‘Intellectually, Mary Stewart started to develop uncommonly early.’

• (74) Territorial’no korallovye rify raspoloženy v tropikax.

  ‘Territorially, coral reefs are found in the tropics.’

• (75) Formal’no oni mogli by ujti, no oni ostalis’.

  ‘Formally, they could have left, but they stayed.’

Note that their status as SAs is not marked by a comma. Only their front position indicates their scope.

It seems to be the case that adverbials such as v dejstvitel’nosti ‘in fact’, na samom dele ‘as a matter of fact’, dejstvitel’no ‘really’ and v samom dele ‘indeed’, should also be classified as phrastic adverbials, perhaps in a separate subgroup because of their meaning, see (76)–(77):

• (76) – On vernulsja, kak i obeščal, čerez god? – Da, on dejstvitel’no / v samom dele vernulsja čerez god.

  ‘Did he return, as promised, after one year? – Yes, he did indeed.’

• (77) – On vernulsja, kak i obeščal, čerez god? – Net, on xotel vozvratit’sja čerez god, a na samom dele / v dejstvitel’nosti probyl za granicej pjat’ let.

  ‘Did he return, as promised, after one year? – No, he wanted to, but in fact he stayed away for five years.’

Although they seem to express the speaker’s attitude towards the truth of the statement involved, they do not. It has to be emphasized that as a reality-oriented language Russian distinguishes the real world from the imagined world by having a sharp distinction between the indicative and the subjunctive mood. But speaking of the real world the speaker may, of course, specify that the situation referred to in the real world was not part of one’s imagination (cf. (76)) or one’s belief (cf. (77)).

5 On the importance of internal ordering

We have just examined the three subclasses of SAs and their various subdivisions. Previously it was pointed out that SAs belonging to the same subclass or even subgroup may co-occur in the same utterance, but nothing was said about specific word order combinations and their functions. We will try to do that below considering various types of combinations, including combinations compared to their mirror images. Special attention will be paid to SAs in contact position. Note that our analysis does not pretend to be exhaustive at all. It is merely an attempt to take the first steps in creating a discourse about a new, interesting subject with its own problems and by doing so try to suggest some solutions to the problems within our framework.

Let us first take a look at (78) which combines two connectives, one belonging to the group of concessive adverbials, the other to the group of non-concessives:

• (78) V statičnom že mire net mesta dlja pribyli: predprinimatelja net, ego zamenjaet upravljajuščij, polučajuščij platu za upravlenie. On ne neset ubytkov i ne polučaet pribyli. Pravda ₁, vpročem ₂, v uslovijax statičnoj ékonomiki est’ vozmožnost’ izvlečenija monopol’noj pribyli.²⁰

  ‘In the static world there is no such thing as normal profit: Businesses don’t have owners, they have managers who are paid for their work. The manager does not suffer any losses and he won’t receive any profits himself. Admittedly ₁, however ₂, in a static economy you can derive a profit from a monopoly.’²¹

The specific ordering is not accidental. Pravda is placed at the front of the utterance, because it carries the paraphrase itself, whereas vpročem is placed afterwards, because the paraphrase created by it fits into the paraphrase structure created by pravda as a possible extension, see (78′):

• (78′) […] Having said this I hereby simultaneously admit that I find it relevant to say something that just occurred to me, namely that …

In that way it is possible to say that the mutual ordering of the two connectives marks their internal ranking. While both types of connectives are paraphrase creating, non-connectives lack this quality. Due to this they just enter into paraphrases created by a connective non-concessive adverbial as in (79) or by a connective concessive adverbial as in (80). Their position in the utterance cannot change that:

• (79) A, meždu pročim ₁, otkrovenno govorja ₂, esli b ja točno znal, kakie imenno sredstva poėtičeskoj vyrazitel’nosti samye perspektivnye, i umel by ix podčinit’ sebe,—ja byl by velikim poėtom.²²

  ‘By the way ₁, if I honestly ₂ knew for sure which poetic devices are the most powerful, and if I were able to master them, I would be a great poet.’

• (79′) I am frank as I hereby say something that just occurred to me and that you may consider to be relevant, namely, that …

• (80) Pervyj raz I. P. Uborevič pribyl v diviziju ešče v 1934 godu. Pozdorovavšis’ so mnoj, on skazal, čto priexal posmotret’, kak učitsja divizija. Ja otvetil, čto očen’ rad, xotja, otkrovenno govorja ₁, vse že ₂ volnovalsja.²³

  ‘I. P. Uborevič came to the division for the first time as far back as in 1934. After greetings were exchanged he said that he had come to inspect the training regime of the division personnel. I said I was fine with that, although in all honesty ₁ I was still ₂ nervous.’

• (80′) […] I said I was fine with that, but although I admit that I said that, I am frank as I hereby say that it is true that I was nervous.

It is interesting to note that two parenthetical adverbials may co-occur in the same utterance, see (81):

• (81) Ladno, ja postarajus’ ne razočarovat’ vas i nadejus’, vam dostavit udovol’stvie poseščenie moego sajta. Kstati ₁, meždu pročim ₂, menja zovut Dima.²⁴

  ‘All right, I will try not to disappoint you, and I hope you will enjoy visiting my home page. By the way ₁, you might want to know ₂ that my name is Dima.’

This is possible because they are oriented towards different persons. Kstati is oriented towards the third person, whereas meždu pročim is oriented towards the second person. This means that kstati is used by the author of the message as an excuse for producing a new utterance (the preceding utterance marked an end of conversation), while meždu pročim motivates it by saying that its content is relevant to the reader in a given situation.

The reverse order is possible, too (see (82)). In this case it is stated that the relevance of the information to the hearer just occurred to the speaker. Pragmatically speaking meždu pročim still motivates the speaker’s utterance (cf. above), but his excuse conveyed by kstati does not apply to the preceding text. It affects the hearer and is read as a reproach:

• (82) Okazalos’, čto u nego est’ dogovor […] na texničeskoe obsluživanie domofona. […] Dogovor, kstati, meždu pročim, 5 let nazad sostavlen. I, meždu pročim ₁, kstati ₂, nikto iz žil’cov o nem ne znal […].²⁵

  ‘It turned out that he had a service contract for the door phone system. […] By the way, it was actually drawn up five years ago. But ₁, apropos ₂, none of the tenants were told about it.’

Looking at text corpora it is striking to see how often attitudinal and modal adverbials are used side by side in the same utterance with different mutual ordering. Sometimes the attitudinal adverbial precedes the modal one, as in (83) and (84); sometimes it follows the modal one, as in (85) and (86):

• (83) K sožaleniju ₁, vidimo ₂, te ljudi, kotorye segodnja za ėto delo otvečajut, absoljutno ne spravljajutsja i ne xotjat spravljat’sja […]²⁶

  ‘Unfortunately ₁, the people who are in charge of this area are apparently ₂ not able to cope with the problem and are not keen to deal with it either.’

• (84) K sožaleniju ₁, vidno ₂, on prav.²⁷

  ‘Unfortunately ₁, he seems ₂ to be right.’

• (85) Ja čeloveku ob”jasnil uže, čto ėto ser’eznaja programma i čto ėti momenty na forume ne obsuždajutsja. Ona, vidno ₁, k sožaleniju ₂, ne ponjala.²⁸

  ‘I tried to tell her that it’s a serious program and that we don’t usually discuss such questions in online forums. Obviously ₁, she got it wrong, I’m afraid ₂.’

• (86) Perexod na zapadnuju sxemu, po-vidimomu ₁, k sožaleniju ₂, neizbežen, no avtory dolžny byli podčerknut’ vozrastajuščuju pri ėtom stepen’ riska, na kotoryj idut predprinimateli, namerevajuščiesja investirovat’ svoj kapital v osvoenie novyx mestoroždenij […]²⁹

  ‘The Western way seems ₁, regrettably ₂, to be unavoidable, but in this connection the authors ought to have emphasized the risk entrepreneurs may run investing in new mineral extractions […]’

In (83) and (84) the speaker expresses his regretful attitude towards the truth of the propositional content conveyed by the respective utterances, but by placing a conclusive immediately after the attitudinal adverbial he signals that his own attitude also includes his own conclusions. In (85) and (86) the speaker is not expressing his regretful attitude towards the truth of anything. He is drawing a conclusion in (85) and giving an explanation in (86), but, at the same time, he is giving the additional information that one has to regret it. In short, when k sožaleniju precedes a modal conclusive or explicative, the utterance is first person oriented. When it follows them, k sožaleniju loses its first person orientation and becomes third person oriented and therefore it also gets more vague in meaning.

Let us look at further examples in which a shift in the internal ordering of an attitudinal and a modal adverbial marks a shift in voice:

• (87) No fil’m javno ₁, vidimo ₂, polučilsja ne dlja vsex, t.e. talantlivyj, no ne genial’nyj, sudja po Vašej reakcii. Nado budet posmotret’ objazatel’no.³⁰

  ‘But the film was clearly ₁ not for everyone. It was a talented piece of work, but it wasn’t brilliant, I concluded ₂ from your reaction. I’d need to see it for myself.’

• (88) Vot tak i rabotajut u nas dlja blaga naroda. Tol’ko kto dlja nix est’ narod, vot ėto mne bol’še vsego interesno. Vidimo ₁, javno ₂ ne žiteli Simferopolja.³¹

  ‘That’s the way they serve their people. The interesting question though in my view is who these people are. It seems ₁ to be clear ₂ that it’s not the people of Simferopol’.’

In (87) the speaker is so confident with his own impression that he draws a conclusion, whereas in (88) he makes a subjective conclusion the truth of which is evident. All this is shown by the internal ordering.

In other words, there is no general rule saying that members of the same subclass cannot follow one another or appear in the same clause or utterance. This in particular applies to phrastic adverbials. They certainly do not exclude one another, but rather seem to complement each other, cf., e.g., (89):

• (89) Territorial’no korallovye rify vsegda / na samom dele raspoloženy v tropikax.

  ‘Territorially, coral reefs are always / as a matter of fact found in the tropics.’

All this does not, however, affect what was said at the beginning of this section about relative position, namely that the scope of SAs is more or less iconically reflected by position, cf. (90):

• (90) Vpročem ₁, konečno ₂, teper’ ₃ trudno sudit’, kto byl prav.

  ‘Incidentally ₁, right now ₃ it is of course ₂ difficult to judge who was right.’

That is to say, first comes the neustic adverbial (cf. vpročem) because it has a long range, then the tropic one (cf. konečno) which has a medium range, and finally the phrastic one (cf. teper’) which has a short range.

The three subclasses of SAs and their respective groups and subgroups, which have just been examined above are, for illustrative reasons, put into Table 2.

Table 2 Classes, subclasses and groups of Russian SAs

6 Conclusion and further remarks

In this paper the SAs in Russian were for the first time analyzed in their totality and compared with English SAs in order to define the essential features of Russian viewed from a classificational, orientational and representational point of view. In accordance with Hare’s three utterance components, the Russian SAs were classified into neustic, tropic and phrastic SAs. It was emphasized that the tropic component is the most important component of an utterance due to the fact that it incorporates the specific communicative supertype membership of a language. As a reality-oriented language which speaks about reality through the situation being common to the speaker as well as the hearer, the Russian declarative superstructure contains ‘It is true’. It was demonstrated throughout the paper that any Russian speaker must consider that the propositional content of the utterance that belongs to the phrastic component must be in accordance with ‘I hereby say it is true’ which thus functions as an obligatory frame of reference for all Russian utterances. In order to cancel ‘I hereby say it is true’ Russian has four important discourse particles, viz. mol, deskat’, -de and jakoby, at its disposal. They all represent another person’s speech or thought without involving the speaker’s assignment of true or false to the propositional content involved. These four particles are the specific outcome of the fact that Russian belongs to a specific communicative supertype, and there are no equivalents in English, because it belongs to a different supertype, viz. hearer-oriented languages, which speak about reality through the hearer’s experience or memory of the situation referred to.

Another important feature of Russian SAs was found in the neustic component in which the speaker orients the utterance towards previous discourse. In some languages it is not obligatory to do so. This is, for instance, the case in English, which in return orients the present discourse towards the hearer. In Russian it seems to be an obligatory part. It was revealed that many so-called micro-groups of SAs consist of precisely three members. It turned out that this is so, because a Russian utterance is inflected in first person, second person and third person—not to be confused with verb conjugation in which the obligatory choice among three persons is rooted in the Verb Phrase and refers to one of the three participants of the situation referred to. To say that the Russian utterance is inflected with respect to person is quite different. In this case we are dealing with a communication situation that consists of three obligatory participants, viz. the speaker, the hearer and reality itself. It is indeed a characteristic feature of Russian that when referring to previous discourse Russian speakers must orient the utterance towards either the speaker himself, the hearer or reality before they assign true or false to the propositional content of the utterance. In short, Russians seem to be forced to take their starting point in the previous discourse separating new information from old information.

Within the so-called group of modal adverbials it was demonstrated that Russian sharply distinguishes external and internal reality when referring to situations. The most interesting thing about this, however, is that the two groups are divided similarly, namely into conclusives that are based on deductive inference and explicatives that are grounded in abductive inference—the specific Russian contribution to epistemic modality, which has no counterpart in English.

Last, but not least, the important question of co-occurrence of SAs and their internal ordering was raised—a question that has been ignored so far. It was demonstrated that Russian allows many SAs to appear in one and the same utterance—a circumstance that is problematic for any theory. We solved this problem by arguing that they either enter into different utterance components or they orient themselves towards different participants of the communication situation. It was also shown that the specific word order marks their mutual ranking, although it proved to be the case that so-called paraphrase-creating SAs outrange this feature.