1 Introduction
All Russian verbal prefixes (elements added before the verb stem to modify the meaning of the verb) have more than one meaning. Traditionally, the meaning of the prefix is defined by comparing the meanings of the unprefixed and prefixed verb, of different prefixed verbs with the same stem, or of different verb stems with the same prefix. The result of this comparison is a list of meanings expressed by the prefix. Such lists face serious reliability issues. The definition of what is the same meaning and what is a different one is difficult, and it is not always based on any other criteria than the language instinct of the classifier and, in many cases, the categories proposed earlier for the same unit. A mere list of meanings neither explains in any detail the way the different meanings are related, nor accounts for the context on any level; nor can it explain when and why a certain meaning does not apply. Another common problem with previous approaches to prefixal polysemy is that the descriptions they offer are either too general to be able to predict any specific meanings (even if the meaning would fit the definitions given), or they describe meanings on such a detailed level that generalizations become impossible.
In this paper I offer a description of how the LCCM Theory (Evans, Reference Evans2006, Reference Evans2009) can be used to describe the polysemy of Russian pod-. Applied to the study of prefixes, the approach that I have chosen can be used to describe the processes that are involved in the cooperation between the verb stem and the prefix in determining the meaning of a prefixed verb. This happens as a result of adding the prefix and using the verb in context; thus, the semantic potential of any given prefixed verb is realized only in actual use.
The advantage of LCCM theory as compared to other recent approaches to polysemy such as Principled Polysemy theory (Evans, Reference Evans2004; Tyler & Evans, Reference Tyler and Evans2003) or prototype theory and the theory of radial categories (starting with Lakoff, Reference Lakoff1987), is in its attempt to explain polysemy in the entire context of meaning construction as well as providing a methodology for distinguishing between different meanings.
1.1 prefixation in russian
In Slavic languages such as Russian, prefixation is an important way of modifying the meaning of the verb. Adding a prefix can give, among other things, precise information on the direction of the movement (for example ehat’ ‘drive’ as opposed to pod’’ehat’ ‘drive close to’), the type of action (as in prygnut’ ‘jump’ and podprygnut’ ‘jump a little’), or the motivation of the actor involved (skazat’ ‘tell’ and podskazat’ ‘tell in secret’). The majority of Russian unprefixed verb stems can be combined with many of the approximately twenty verbal prefixes, and even the same prefixed verb often has several meanings. To make things even more complicated, not all stems that occur in prefixed words exist in modern Russian without prefixes and, related to this, the meaning of the stem can sometimes be defined only with the help of the prefixes combined with it. The combinatorial possibilities of prefixes and verb stems have not been described – indeed, they are probably indescribableFootnote 1 (Dobrušina, Mellina, & Paillard, Reference Dobrušina, Mellina and Paillard2001; Krongauz, Reference Krongauz1998). In accordance with the interactional view on prefixes, the meaning of a prefixed verb is not simply the meaning of the prefix plus the meaning of the verb stem, but both the meaning of the prefix and the choice of the stem can be dependent on one another (Dobrušina et al., Reference Dobrušina, Mellina and Paillard2001).
In Russian, verbal prefixes have not only semantic functions (adding information on the exact form of action or state) but also grammatical and word formational functions, because prefixation is one of the means of forming the perfective aspect verb from the imperfective.Footnote 2 At the same time, as noted by Krongauz (1998, p. 83), perfectiveness should not be interpreted as a feature of the prefixes, since the same prefixes also appear in secondary imperfective forms of the same verbs and their use is thus not limited to the perfective aspect only.
Adding a prefix may bring in redundant information that would be clear even without the prefix, on the basis of the prepositions and cases used. This, together with the grammatical role that prefixes play in perfectivization, has led to the interpretation that some prefixes or some uses of prefixes are semantically empty, i.e., they do not bring anything new to the verb. However, the existence of empty prefixes has been refuted by several contemporary scholars on the basis of the inevitable slight difference in meaning between primary and secondary imperfectives, i.e., the imperfective verb forms without and with a prefix (see, e.g., Janda, Reference Janda1986; Krongauz, Reference Krongauz1998). Psycholinguistic data (Sokolova, Lyashevskaya, & Janda, Reference Sokolova, Lyashevskaya, Janda, Divjak and Gries2010) and corpus data (Janda et al., Reference Janda, Endresen, Kuznetsova, Lyashvskaya, Makarova, Nesset and Sokolova2013) have also been used in order to prove that aspectual prefixes are not semantically empty.
The traditional way of describing the meanings of Slavic prefixes involves verbal descriptions that use the words of natural languages for defining meanings. Besides that, prefixal meanings have been described with various forms of metalanguage such as formal semantics (Dobrušina et al., Reference Dobrušina, Mellina and Paillard2001), structuralist formal models (Flier, Reference Flier1975, Reference Flier, Flier and Brecht1985; Gallant, Reference Gallant1979), and cognitive schemata (Janda, Reference Janda1986, Reference Janda and Rudzka-Ostyn1988). Krongauz (1998, pp. 55−98) provides a detailed survey of prefix studies for Russian up to the late 1990s. After that, prefixation in Slavic languages (mostly Russian) has been discussed from various methodological and theoretical viewpoints, for example in Krongauz and Paillard (Reference Krongauz and Paillard1997), Krongauz (Reference Krongauz1998), Dobrušina et al. (Reference Dobrušina, Mellina and Paillard2001), Plungjan (Reference Plungjan2001), a special issue of the linguistics journal Nordlyd (Nordlyd, 2004), Gehrke (Reference Gehrke2008), and in Janda et al. (Reference Janda, Endresen, Kuznetsova, Lyashvskaya, Makarova, Nesset and Sokolova2013).
The relationship between the different meanings of a single prefix has been described, first, from the standpoint of prototype theory (and radial categories), which assumes that each verbal prefix has one prototypical meaning from which the other meanings can be derived. The second option is the invariant theory, which sees the relationship as a relation between different variants of the same meaning invariant. The third option would be to describe the meanings as separate, which is usual in dictionaries and language teaching.
The study of prefixes has much to do with the study of prepositions, which has been very popular in cognitive semantics. Prefixes are both historically and semantically connected to prepositions: in Russian, prepositions and prefixes have common ancestors, are similar in form, prefixes are often used with prepositions, and are semantically related to them.
1.2 the prefixpod-
The Russian prefix pod- has several different meanings that can be seen as either unrelated to one another or as having a very vague connection. The most overt meanings of pod- are spatial: ‘under’ or ‘closer’. Even these purely spatial meanings have numerous manifestations that have seemingly little to do with one another. Besides concrete spatial meanings, pod- has a range of less concrete and less obviously spatial meanings. Pod- is often (but not always) used together with the preposition pod that has several meanings, the most concrete of which is ‘under’ (on its other meanings see Plungjan and Rahilina, Reference Plungjan, Rahilina, Paillard and Seliverstova2000).
The locational meanings of pod- are ‘being under’, ‘being down’, ‘going under’, ‘from under’, ‘upwards’, or ‘downwards’, depending on the verb that the prefix is connected to and on the context in which it is used. Some examples of these meanings: podbrosit’ ‘toss up’, ‘throw under’, ‘throw in’, ‘throw on’, as opposed to brosit’ ‘throw’; podprygnut’ ‘bob up and down (close to the original surface)’, ‘jump up’, as opposed to prygnut’ ‘jump’; podvesit’ ‘hang up’, ‘suspend’, as opposed to vesit’ ‘weigh’; podvintit’ ‘screw up’, ‘tighten’, as opposed to vintit’ ‘screw’. With verbs of motionFootnote 3 the prefix pod- takes (primarily or additionally) the meaning ‘closer’, for example podbežat’ ‘run up to’, ‘come running up to’, as opposed to bežat’ ‘run’; podletet’ ‘fly close’, ‘fly up’, ‘fly up to’, as opposed to letet’ ‘fly’. The different spatial meanings of pod- can even seem mutually contradictory, for example when different directions having to do with the downside are considered, as in podbrosit’, which was mentioned above. In this case the way of throwing can be specified with the single prefix in ways that have different directions − ‘to downwards’ or ‘from downwards’. The direction of the movement is, nevertheless, probably always clear from the context; at least I have not found evidence to the contrary.Footnote 4 The spatial meanings of pod- concentrate on indicating the connection of being under or down, rather than giving a certain direction. This is further supported by the fact that many of the verbs involved are also able to connect with prefixes that have a concrete directional meaning, for example letet’ ‘fly’ can be used with different prefixes that result in forms such as sletet’ ‘fly down, fall down’, naletet’ ‘fall on, fly on’, otletet’ ‘fly off’, and so on.
The meanings of pod- that are not purely locational are not quite as easy to describe: while all the purely locational meanings involve different ways of being under and down, the other meanings do not have a single unifying feature. The most detailed analyses of the meanings of pod- are Volohina and Popova (1993, pp. 83−89) and Plungjan (Reference Plungjan2001). Plungjan (Reference Plungjan2001) uses the term ‘non-spatial meanings’ for both non-spatial and not purely or solely spatial meanings. He divides the non-spatial meanings of pod- into three categories: describing supplementary action (additional, simultaneous, or lesser action), causing harm or damage or doing something in secret, and getting abstractly closer (modifying the action according to someone else’s needs or to fit something; imitating something). An example of supplementary action would be podbrit’ ‘shave additionally’, podložit’ ‘add’, or podlečit’sja ‘take some medical treatment’. The meaning of causing harm or damage is exemplified by podmočit’ ‘wet slightly’, ‘damage by exposing to damp’, and podžeč’ ‘set on fire’, and the meaning of doing something in secret, without anyone knowing, by podslušat’ ‘overhear’, ‘eavesdrop on’. The different cases of getting abstractly closerFootnote 5 can be seen, for example, in podkupit’ ‘bribe’, podstykovat’ ‘join together’, and podščëlkivat’ ‘click in tune with something’. Some of these verbs are actually polysemous: they can express several of the meanings described, e.g., podžeč’ could also imply ‘set on fire in secret’ and only context can determine the meaning at hand.
This traditional kind of description of different ‘meanings’ of pod- describes them as if they were accidental.Footnote 6 The description is aimed at explaining the meaning in terms of what the language user wishes to convey with the particular unit and, accordingly, how the units are understood. This is closely related to the way that dictionaries classify different meanings. Furthermore, the ability of various phraseological units to convey meaning has not been taken into account in such descriptions. In actual fact, meaning is not determined solely by adding a certain prefix, but also with the help of a wider phraseological context.
There is also a theoretical reason for wishing to find new means of description for the polysemy in prefixes. The different meanings of a single prefix have so far been described mostly in relation to spatial concepts. Meanings are primarily described as either concretely spatial or deriving from concrete spatial meanings. This kind of description requires, first, very strong generalizations about the centrality of spatial notions in language and, because of this assumption, either describes the meanings involved in very general terms (as some kind of invariants) or is able to describe the meaning involved only from the dictionary-like viewpoint, i.e., in very concrete, everyday terms.
This paper aims at a description of the different meanings of pod- that takes into account both the dynamic character of meaning and the special nature of prefixation as both a word-formational and a meaning-producing mechanism.
2 Theoretical background
2.1 discussing lexical ambiguity
Lexical ambiguity as an inherent feature of linguistic units has been discussed in terms of homonymy and polysemy. In homonymy, the lexical items look and sound the same but have different and unrelated meanings, i.e., the correspondences in form are accidental, while polysemy assumes a connection between the different meanings of the same linguistic unit. Different interpretations of lexical ambiguity in general and polysemy in particular can be found, e.g., in Cuyckens, Dirven, and Taylor (Reference Cuyckens, Dirven and Taylor2003), Evans (Reference Evans2009), Rakova (Reference Rakova2003), and Rakova, Pethő, and Rákosi (2007). Yet another interpretation sees linguistic units as inherently monosemic and their different uses as deriving from one meaning (the prototype and invariant theories discussed above with regard to the prefix studies). In the following I will use LCCM Theory (Evans, Reference Evans2009) to discuss lexical ambiguity.
2.2 theory of lexical concepts and cognitive models in discussing lexical ambiguity
In the Theory of Lexical Concepts and Cognitive Models (LCCM; Evans, Reference Evans2006, Reference Evans2009, Reference Evans2010) meaning is seen as a dynamic process − it is born in language use when the use of a word or other lexical unit activates lexical concepts related to it. LCCM theory includes both a model of how meaning is activated in language use (this is called ‘lexical representation’) and how meanings are organized (‘lexical composition’). Interpretation occurs not only on the basis of the presumed meanings that words and other lexical units have, but also with the help of all the knowledge that the speakers have of the functioning of the world. In LCCM theory, knowledge of language consists of symbolic units and cognitive models. Symbolic units consist of the phonological form of the word or other lexical unit (how this unit is pronounced) and also what is called a ‘lexical concept’, i.e., the linguistic information included in this form. In this way, a certain word or part of a word in its pronounced form is not seen as direct evidence for brain functions or different processes of conceptualization by speakers of different languages. Instead, non-linguistic knowledge is accessed through linguistic information encoded in lexical concepts.
A lexical concept is in LCCM theory a unit of schematic linguistic knowledge. It constitutes the semantic side of the phonological representation of a word or other linguistic unit. The linguistic knowledge included in a lexical concept is a ‘bundle’ of information since it comes from different sources and is a generalization that works as a result of the human tendency to see schematic similarities. In using a concrete linguistic unit, several lexical concepts can be activated at the same time. Lexical concepts are language-specific, and each language has a unique set of lexical concepts. This means that different languages do not have an identical semantic structure. Lexical concepts encode different kinds of linguistic content (information available to a language user). A lexical concept plus its phonological vehicle form a symbolic unit that is connected to one or several cognitive models. Cognitive models consist of a frame or several related frames.
LCCM theory distinguishes between conceptual structure and semantic structure. According to this theory, lexical concepts offer access to cognitive models that include schematic knowledge in the form of frames. The theory does not assume that the forms or structures of language could be considered direct evidence for the way that our conceptualization works. This assumption is different from the well-known and much-used Conceptual Metaphor Theory (e.g., Lakoff & Johnson, Reference Lakoff and Johnson1980, Reference Lakoff and Johnson1999), which assumes that linguistic metaphors, i.e., concrete expressions and structural features in a given language, can be used as direct evidence for the existence of a certain conceptual metaphor, a habitual way of thinking in which people use the terms of one conceptual domain when talking about another. This means that using a certain word in a way that differs from its literal sense – for example calling a person a creampuff (a celebrated example in the literature) – would, according to Conceptual Metaphor Theory, be a piece of evidence for the existence of a conceptual model (metaphor) in which a human being is referred to with a name that belongs to the conceptual domain of pastry and thus makes a connection between these two domains. LCCM theory supposes that if language examples include figurative language use, this means that they involve a clash between the primary meaning and the figurative meaning − the concrete use is in contradiction with the primary meaning and this necessitates interpreting it figuratively. Using the same example, when creampuff is used in a context in which the pastry interpretation does not make sense, it is interpreted in another, figurative meaning.
The existence of a given lexical concept can be proved with the help of semantic and/or formal selectional tendencies. This means that either their grammatical characteristics are different (= formal selectional tendencies), or that they select lexical items with different meanings (= semantic selectional tendencies). An example of different grammatical characteristics that could serve as formal selectional tendencies could be that certain verbs cannot take a direct object, i.e., are intransitive (like to die, to hover), or that they are used in a certain grammatical construction, for example with the preposition in. As for semantic selectional tendencies, an example is a semantic connection to money. The notion of selectional tendencies is used in order to justify the classification of lexical concepts on the basis of language data: they form the lexical profile of the lexical concept. Selectional tendencies also have to do with the cognitive credibility of the theory, describing human categorization tendencies for which there is evidence in language.
LCCM theory also uses the notion of lexical concepts in its approach to polysemy. Polysemy occurs when the same phonological unit (word as pronounced) refers to several inter-related lexical concepts. LCCM theory attempts to describe their mutual relationships, which may involve varying degrees of semantic closeness. Hence, lexical concepts are used in LCCM theory to describe the meaning potential that a given linguistic unit has and also to illustrate semantic connections between different lexical units. As I see it, the main idea of the notion of lexical concept and the whole LCCM theory is to show how different languages can have a different semantic structure. This means that languages differ not only in using different words for the same thing or having different syntactic structures, but also when it comes to the way they organize the world.
3 Material and methods
The starting points for this study were the prefixed verbs with pod- found in Bol’šoj Tolkovyj Slovar’ (Kuznecov, Reference Kuznecov1998), the most comprehensive single-volume dictionary of contemporary Russian. The dictionary includes 505 verb infinitives with pod-, both as separate entries and as examples in the entry for the prefix. It is assumed that the dictionary provides an almost complete listing of the verbs that exist in contemporary Russian. As a point of comparison, Plungjan’s study (2001) of the meanings of the prefix pod- was made on the basis of about 350 verbs that were included in the Grammatical Dictionary of Russian (Zaliznjak, Reference Zaliznjak1977).
Studying the meaning potential of a prefix with the help of dictionary material starts from the assumption that the different uses and (thus) meanings can be studied on the basis of a list of verb infinitives, provided that the list is comprehensive. Since the meanings of the prefix pod- include certain recurring tendencies, the essence of these meanings can be grasped with the help of the linguistic units (in this case verbs stems) that the prefix can be combined with. The combination of the prefix with the verb stem tells us about the way in which the prefix works in language, and what this combination process reveals about the meanings that can be expressed with this prefix.
The use of dictionary data has been restricted to finding the verbs involved, in which I find a comprehensive dictionary to be very useful. If a prefixed verb is established enough to be included in a comprehensive dictionary, this can be seen as proof of its existence. The assumed full list of verbs that can be used with the prefix pod- helped to ensure that no meanings realizable with the prefix pod- were missed. The working method for the study was one of saturation – the 505 verbs were all classified for their lexical concepts in their different meanings (on the basis of Plungjan, Reference Plungjan2001). In this way, the list of lexical concepts activated in different uses of pod- was based on dictionary data.
In the next step of analysis, twelve polysemous verbs from among the 505 were chosen for closer analysis. Further analysis of the verbs was not based on their meanings as described in the dictionary, but on examples of actual use from the spoken corpus of the Russian National Corpus (RNC, useable online at <www.ruscorpora.ru>). The Corpus of Spoken Russian includes recordings of public and spontaneous spoken Russian and transcripts of Russian films (1930−2007), a total of 10 million words. The corpus was used to find authentic examples, which were then used to evaluate the claims made. Equal consideration was given to examples confirming the hypothesis and those working against it. The Corpus of Spoken Russian was selected for this study because a corpus of spoken language comes closest to the kind of utterances that are seen as appropriate material for language study based on LCCM theory.
I will list here twelve verbs from my material that will be used as examples in this paper. I will return to them in various sections of the paper and give examples of their use from the spoken corpus of RNCFootnote 7 (retrieved on 4 April 2013).
(1) подвергать
pod-verg-a-t’
under-turn-ipfv-inf
‘put under, expose to’
(2) подбросить
pod-bros-i-t’
under-throw-prf-inf
‘throw under, throw a little, throw upwards’
(3) подпрыгнуть
pod-pryg-nu-t’
under-jump-once.prf-inf
‘jump a little’
(4) подбавить
pod-bav-i-t’
under-add-prf-inf
‘add some’
(5) подковырнуть
pod-kovyr-nu-t’
under-dig-once.prf-inf
‘pick a sore; catch out’
(6) подстилать
pod-stil-a-t’
under-spread-ipfv-inf
‘lay under, stretch under’
(7) подсказать
pod-skaz-a-t’
under-say-prf-inf
‘say secretly’
(8) подкараулить
pod-karaul-i-t’
under-watch (on guard)-prf-inf
‘lie in wait for’
(9) подбежать
pod-bež-a-t’
under-run-prf-inf
‘run close to’
(10) подладиться
pod-lad-i-t’-sja
under-succeed-prf-inf-refl
‘get accommodated to’
(11) подтереть
pod-ter-e-t’
under-wipe-prf-inf
‘wipe up’
(12) подцепить
pod-cep-it’
under-chain-prf-inf
‘hook on, couple on’
The use of examples with these particular verbs aims to illustrate the way that different lexical concepts are activated in specific uses and how they are related to certain cognitive models. The analysis of the language data meant in this case that the examples from the spoken corpus of the RNC were each looked at individually as evidence for conceptualization processes. The meaning that the prefix added to the verb stem in each case was assumed to tell us about the meaning potential of the prefix.
4 Results
In the following I will present the lexical concepts that are sufficient to explain the meaning potential of the 505 verbs with pod- in the material. In order to come up with this list and to decide that they really are lexical concepts, I have followed the principles for distinguishing between different lexical concepts discussed earlier. As for formal selectional tendencies, these different lexical concepts have different grammatical characteristics in that adding the prefix may or may not change the argument structure of the unprefixed verb.
The semantic selectional tendencies that give ground for distinguishing between different lexical concepts influence the range of meanings that the prefix conveys in conjunction with the verb stem. The resulting lexical concepts are open-class, i.e., besides encoding linguistic content they offer access to conceptual content. Each of the lexical concepts involves linguistic knowledge in the form of reduced parameters than may overlap. I will explain the content of each lexical concept in accordance with these parameters.
All the 505 verbs with pod- in my material are explainable with one or several of the following lexical concepts that will be dealt with in detail in the following:
[UNDER]Footnote 8
[VERTICAL MOVEMENT]
[CLOSE]
[CONTACT]
These lexical concepts represent the meaning potential that the prefix pod- has. In a particular use several lexical concepts can be activated at the same time. Lexical concepts have the phonological vehicle /pod/ (before voiced consonants and vowels) or /pot/ (before voiceless consonants), or /podo/ (in some special cases). The different phonological variants of the unit are counted as one since they follow regular pronunciation rules of Russian (assimilation of voiced and voiceless consonants). Evidence for the lexical concepts that sanction the use of pod- has been obtained by analyzing language material. This analysis has been done on the basis of how the linguistic unit in question behaves formally and semantically in different uses.
4.1 [under]
The lexical concept [UNDER] can be described as a statement of something being under or below something else. When described through the notions of landmark (LM, the spatially bound object that serves as the base for comparison when movement or situation is described) and trajector (TR, the subject being talked about that is not spatially bound),Footnote 9 [UNDER] is a state in which TR is vertically lower than LM. Of the twelve verbs introduced above, (1) podvergat’, (2) podbrosit’, (3) podprygnut’, (4) podbavit’, (7) podskazat’, and (8) podkaraulit’ represent, in some of their uses, this lexical concept. As polysemous units they represent other lexical concepts in other contexts. Apart from certain uses of the prefix pod- the lexical concept [UNDER] sanctions (makes possible) the use of various other linguistic forms in Russian, such as prepositional phrases indicating spatial relations.
The lexical concept [UNDER] does not follow the usual tendency for Russian to be very specific about the means and direction of movement, expressed with verbs of motion, different prefixes, cases, and prepositions. It is indifferent as to the direction of movement, merely conveying the meaning of position. For example, in (2) podbrosit’ ‘throw under’, the direction of the movement is downwards (to) under something else, or upwards (with the meaning of throwing something slightly up), while in (1) podvergat’ ‘make subject to’, literally ‘under-turn’, the turning under (influence, critique, etc.) comes with the figure of surrounding by revealing one’s face (upwards). In (3) podprygnut’ ‘jump up’, lit. ‘under-jump’, the direction of the movement can only be upwards.
The lexical profile of the lexical concept [UNDER] includes the factors that make it a separate lexical concept, the semantic and formal selectional tendencies. As for its semantic selectional tendencies, the lexical concept [UNDER] can be described as a state in which TR is vertically lower than LM, for example throwing down to the street from a balcony, where the balcony is the landmark that stays in one place and the thing being thrown is the trajector. If there is movement involved, this movement happens within the lower positioning of TR as compared to LM. This means that the movement does not change the fact that TR is lower vertically than LM. The point of comparison, in this case the landmark balcony, remains unchanged. The direction of the possible movement involved can be different, in addition to the concrete object towards which it is directed.
The grammatical pattern associated with [UNDER], showing the formal selectional tendency for [UNDER] (as for the prefixed verbs with pod-), is that transitive verbs with pod- are used with a direct object in the accusative without a preposition as in (18) and (19), or with the preposition pod ‘under’. In other words, the verb indicates ‘under-do what’. Besides purely transitive verbs, the reflexive verbs formed with the postfix -sja from transitive verbs can also appear, because they include the object of the action on the semantic level – even though it cannot be concretely stated because the intransitive verb cannot take a direct object, i.e., it is impossible to put an object after such a verb because grammatically the postfix -sja includes the notion of ‘to oneself’ – although this meaning is not always included in these verbs.Footnote 10 What is interesting here is that the additional information that pod- gives does not affect the concrete directional information of the verb, i.e., it does not mean ‘to under’, but merely indicates the location where the event takes place.
As for its schematic linguistic content, [UNDER] includes the category of spatial relation in topological reference between two objects (LM, TR). Of key importance in this specific case are parameters of vertical level, Lower and Higher. Other parameters of spatial relation here include Under and Movement within Range (the range here being location under a certain landmark). The precise difference in vertical level is not encoded in language, only the relative level of the two objects matters. In this way it is a case of deictic reference. Only this distinction is encoded in language use when the lexical concept [UNDER] enables us to use the prefix pod- in a certain way. In contrast, conceptual content includes finer distinctions and is accessed via open-class lexical concepts such as [UNDER], but is not directly encoded in language. This conceptual content includes the cognitive models associated with the lexical concept in the form of frames. In the case of [UNDER] these cognitive models include notions of lesser importance and secrecy. This will be discussed below with illustrative examples.
Let us now look at some examples of use from the spoken corpus of the RNC. Podvergat’ (1) (literally ‘turn under’ or ‘turn from under’) means exposing. Here the meaning components ‘under’ and ‘turning’ form together the meaning of turning in a certain way (having to do with the notion of being under). In (13) the impossibility of defining something is expressed as something ‘not being exposable’ to definition, and in (14) people are exposed to risk.
(13) Так устроена жизнь / что предвыборная агитация не подвергается точному определению / и все дело в интерпретации.
That.way built life / that before-elections-adj agitation not under.turn-pass.prs close-datdefinition-dat / and all thing in interpretation-loc.
’That’s the way life goes / it’s impossible to define electioneering in detail / and it is all a matter of interpretation.’
(14) Остановитесь. Вы подвергаете опасности жизни близких людей.
Stop-imp.2pl You under.turndanger-dat lives-acc close-gen people-gen.
‘Stop. You are risking the lives of your close ones.’
In (13) the agitation cannot be ‘turned under’ a definition. This turning, exposing a certain side, means being definable. In (14) ‘under-turning’ the lives of the close ones to danger means causing a situation in which they may be exposed to danger. These examples show how the notion of being under, of being exposed, can have various kinds of concrete consequences. In this example the lexical concept [UNDER] is on a linguistic level encoded with the parameters Movement within Range and Lower. These parameters explain the linguistic distinction that adding pod- to the verb stem vergnut’ involves. The lexical concept also gives access to cognitive models in the form of frames showing what happens to us or to objects as a result of turning. The way that the verb is used tells us about the way that the prefix and the turning create meaning in a certain context. While being definable (‘being exposable to definition’) (13) and being exposed to danger (14) are very different things, they have in common the notion of being left without protection. The protection is taken off by turning the weak side (as dogs do when they surrender by exposing their stomachs).
A very illustrative verb (2) podbrosit’ ‘pod- + throw’ can, besides concrete throwing under, down, or up (15), (16), refer to giving a lift by car (17),Footnote 11 or carelessly (casually, secretly) giving something away, even abandoning a child (18) or an idea (19). Podbrosit’ can also mean pouring money into something, giving money for a purpose that the speaker sees as unnecessary or arrogant, as in (20). In any case the ‘under-throwing’ has an object and the action itself is characterized by the meaning that arises from interaction between the prefix and the verb stem. In its concrete meaning (throwing under something) it can also appear with the preposition pod ‘under’.Footnote 12
(15) И если встать на вершину… ии… подбросить листок бумаги / то он полетит вверх и далеко.
And if climb.up-inf onto top-acc… and…under.throw-inf piece paper-gen/ then it fly-prf.fut.3sg up and far.
‘And if you climb to the top…and…throw down a piece of paper / it will fly up and far away.’
(16) Сжигаешь / а потом подбрасываешь в землю для витаминов?
Burn-2sg /and then under.throw-prs.2sg into ground-acc for vitamins-gen?
So you burn it / and then throw it to the ground for vitamins?
(17) Он меня щас до дому подбросит.
He I-acc now up.to home-genunder.throw-fut.3sg
‘He will soon give me a lift home.’
(18) Ну / она здесь сделала / конечно / огромную глупость. Подбросила своегo ребёночка.
Well / she here do-pst.f / of.course / great -acc stupid.thing-acc
Under.throw-pst.f own-acc child-acc
‘Well / what she did here was / of course / very stupid. She abandoned her own child.’
(19) Штирлиц подбросил эту идею Шелленбергу и начал рассказывать смешные истории.
Štirlic under.throw-pst.m this idea Šellenberg-dat and begin-pst.m
tell-ipfv.inf funny stories.
‘Štirlic pitched this idea to Šellenberg and started telling funny stories.’
(20) Если будет страна разваливаться / они там немножко долларов подбросят / а чтоб пожертвовать…
If will country fall.apart / they there some dollars
under.throw-3pl / and in.order.to donate-inf…
‘If the country starts falling apart / they will just throw in some dollars / and in order to donate…’
In these examples, whatever the ‘throwing under or down’ means in each concrete case, the prefix pod- defines the throwing in each case. The movement of the object that is being thrown, obviously involved in any kind of throwing, is connected both to a certain kind of movement and a movement in a given direction. Concretely throwing under or down can have the symbolical meaning of throwing away.
[UNDER] is an open-class lexical concept that also provides access for cognitive models. The concept of being under, going under, etc. can mean several things. Besides concrete notions of the placement of different objects (as in examples (15) and (16)), it can also relate to hierarchical positions (in (18), (19), (20)) and subordination (in (13), (14)). Pod- is the only prefix in Russian that can express completely opposite meanings when it comes to the direction of movement, as in the notions of throwing upwards or under something (compare examples (15) and (16)). The lexical concept [UNDER], having to do with the position under or below, is thus related to cognitive models that can include metaphorical or metonymical transfers. According to LCCM theory, this happens when there is a contradiction between the literal interpretation of the utterance and the situation. The cognitive models that [UNDER] is related to include the frames of the encyclopaedic knowledge that we have of being under or below, i.e., all the extralinguistic factors involved in this meaning.
The lexical concept [UNDER] gives access to a cognitive model that involves a purely metaphorical conceptualization. This metaphor includes such notions as dominationFootnote 13 and control. Thus, being under means being subjected to the influence or dominance of something, being subordinated to something. [UNDER] is also related to cognitive models that involve secrecy in action. Examples (18) and (20) show actions that can be interpreted as secret ones (abandoning a child and giving away dollars); in Section 4.2 there will be more examples of related uses. [UNDER] is also related to a cognitive model which sees things that are under as being less important, secondary (this corresponds to Plungjan’s, Reference Plungjan2001, category of ‘secondary action’). One such case is (4) podbavit ‘add some’. Two examples of its use in the spoken corpus of RNC:
(21) Да / да / да. Подбавьте яду / подбавьте / подбавьте.
Yes / yes / yes. Under-add-imp.pl poison-gen / under-add -imp / under-add-imp
‘Yes / yes /yes. Add poison / add / add.’
(22) А ну-ка я вам щас пару-то подбавлю!
And let’s I you-dat now couple under-add-1sg
’Now then, I’ll add you some now!’
In (21) and (22) podbavit’ means ‘to add some’ as compared to pribavit’ ‘add’. The prefix pod- conveys a notion of lesser degree or incompleteness.
As we have seen here, [UNDER] is a lexical concept that includes certain grammatical patterns and is realized by indicating the position of a TR as compared to a certain LM. In the case of the verbal prefix pod-, what is ‘under’ in these contexts is the action in its totality. The parameters involving the schematic linguistic content of [UNDER] include Higher, Lower, Under, and Movement within Range, and they represent the category of spatial relations. [UNDER] provides access to cognitive models linked to hierarchies, importance, and secrecy.
4.2 [vertical movement]
Of the twelve verbs introduced earlier, (5) podkovyrnut’ and (6) podstilat’ can represent the lexical concept [VERTICAL MOVEMENT]. The lexical profile of the lexical concept [VERTICAL MOVEMENT] is different from [UNDER] in that the downward movement is defined in terms of a change in the position of the subject, for example a leaf flying down. In this way the original position serves as a landmark in relation to which the new position is described; it does not require another landmark. The semantic selectional criterion for the lexical concept [VERTICAL MOVEMENT] is that it describes a downward movement in which the point of comparison is the former position of the same object. The formal selectional tendency that makes [VERTICAL MOVEMENT] different from [UNDER] is that it involves only the prefixed verb with pod- + accusative. Furthermore, when there is an object for action, the object is concretely put under something, i.e., it moves in this action.
The change in position may involve both downward and upward movement. The movement can also involve picking, as in (5) podkovyrnut’ ‘pick a sore; catch out’, so that its direction can go first downwards and then up. In (6) podstilat’ ‘lay under, stretch under’, downward movement with the notion of laying, stretching, or slipping under something is involved.
The three examples of use for (6) podstilat’ in the spoken corpus of RNC include:
(23) Они остынут / скоро начнут из тебя тепло тянуть. Шинельку подстилай.
They get.cold-3pl / soon start-fut.3pl from you warmly pull-inf Jacket-accunder.lay-imp.2sg
‘They will get cold / soon they will start getting the heat out of you. Put a jacket under yourself.’
In (23) the jacket is the direct object of the verb so the meaning of vertical movement, putting the jacket under oneself, comes only from the prefix.
Two examples of use for (5) podkovyrnut’ from the Spoken corpus of RNC:
(24) Я люблю посмеяться / я люблю пошутить / я люблю кого-нибудь подковырнуть / и с мужиками так разговариваю.
I like-1sg laugh / I like-1sg joking / I like some-
one under.dig / and with men-instr that.way speak-prs.1sg
’I like having a good laugh / I like joking / I like taunting someone / and that’s the way I speak with men.’
(25) Шутки для девочек поберегите / а то так подковырну!
Jokes for girls-gen save-imp.pl/or so.badly under-dig-1sg
’Save your jokes for girls / or I’ll show you!’
In (24) and (25) the meaning of picking something up (figuratively as irritating, teasing someone) shows a cognitive model in action. This cognitive model includes using the word for concrete action with reference to mental irritation. This metaphor is based on the similarity of the feelings that physical and mental irritation cause. The frames representing this in the conceptual system include information on how a person acts when exposed to physical or mental irritation.
What, then, is the relationship between the lexical concepts [UNDER] and [VERTICAL MOVEMENT]? [UNDER] is expressed with constructions that involve stating the whereabouts of an object as compared to other objects (expressed in various languages with prepositions, postpositions, cases, etc.) while [VERTICAL MOVEMENT] states the direction of movement. In many uses without pod- the two lexical concepts are separate. For example, when [UNDER] is expressed with verbs without pod-, the preposition pod is used either with the instrumental case indicating position or with the accusative case indicating movement. Expressing [VERTICAL MOVEMENT] without pod- does not, for its part, include the preposition pod. These formal selectional tendencies prove [UNDER] and [VERTICAL MOVEMENT] separate. There is also a difference in how the meanings of the prefix and the verb stem contribute to the use. In the case of [UNDER] they are equal in determining the meaning potential, while in [VERTICAL MOVEMENT] the prefix merely modifies the meaning of the stem. This is well illustrated in examples (13) and (24) – the meaning of exposing in (13) is based on the cooperation of the prefix and the verb stem meaning turning, while in (24) teasing slightly is a kind of teasing.
The lexical concepts [VERTICAL MOVEMENT] and [UNDER] are, nevertheless, clearly related in Russian. In some particular cases, the formal and semantic selectional tendencies given earlier seem quite vague in differentiating between the two. I noted earlier that the difference between [VERTICAL MOVEMENT] and [UNDER] is in the way that [UNDER] involves a landmark under which something is or goes. In many cases, however, the thing under which something hides exists only implicitly − i.e., we know from our everyday experience that such an obstacle to seeing must exist, but it is not mentioned. This comes about especially often when metaphorical or metonymical cognitive models enable pod- to be used in non-literal meanings. On the basis of what we know about human behaviour, we can imagine the ways that this secret action can be accomplished and the real-life equivalents of the notions of [UNDER] and [VERTICAL MOVEMENT].
As already mentioned, [VERTICAL MOVEMENT] is an open-class lexical concept and as such it provides access to cognitive models. The interpretation of something being under is based more on our knowledge of the world than on anything found at the level of language. Combining pod- with a verb that is not a verb of motion results in the meaning of something happening in secret. This meaning is clearly a consequence of a particular cognitive model, a metaphorical conceptualization that the lexical concept [UNDER] is related to. This metaphor can be called ‘Doing Under Is Doing in Secret’. This kind of conceptualization sees secret action as something that is done in hiding. Pod- cannot add the meaning of doing in secret to all verbs. It is, nevertheless, productive in the sense that if pod- is added to a verb with which its other meanings do not make sense, this is the meaning that seems to be evoked. This involves a clash with the primary meaning. Thus, pod- with this meaning is available for certain verbs, and these verbs mostly do not take the prefix pod- with any other meanings (with very few exceptions). This example illustrates how the relationship between lexical concepts and metaphors and metonymies is seen in LCCM theory. The existence of a separate lexical concept for doing in secret is denied because the uses of pod- in such contexts are not grammatically separate but follow the same rules as with the lexical concepts [UNDER] and [VERTICAL MOVEMENT]. As for other forms of expressing secrecy, this notion can mostly be expressed in Russian by certain adverbs such as sekretno ‘secretly’, taino ‘secretly’, and nezametno ‘discreetly’.
The metaphorical use of pod- is illustrated with some examples of the verbs (8) podkaraulit’ and (7) podskazat’ in the spoken corpus of the RNC:
(26) Тем более я не ту шайбу взял / меня могут подкараулить и забить.
That-instr more I not that-acc puck-acc took-m / I-acc can-3.plunder.watch-inf and score-inf
‘Especially since I took the wrong puck / they can ambush me and score.’
(27) «Как бы мне хотелось / чтобы вас подкараулили в тёмном переулке и трахнули кирпичом по голове»!
How cond I-dat want-pass.pst.n /so.that you-accunder.watch-pst.3pl in dark-loc alley-loc and bang-pst.3pl brick-instr on head-dat
‘I would really like to / see you being ambushed in a dark street and being hit on the head with a brick!’
(28) Обратитесь к Абрамовичу / он подскажет.
Turn.to-imp.3pl towards Abramovič-dat / he under.tell-fut.3sg
‘Ask Abramovič / he will tell you.’
(29) А с какой должности вы увольнялись / не подскажете?
And from which-gen position-gen you resign-pst.3pl / not under.tell-2pl
‘From which position did you resign / won’t you tell me?’
(30) А какая лучше / подскажите?
and which better / under.tell-imp.2pl
‘And which is better / please tell me?
(31) Немножечко могу подсказать?
little can-1sgunder.tell
‘May I hint a little?’
These uses (26)−(31) show how the metaphorical notion of secrecy is realized with the prefix pod-. Podkaraulit’ refers to waiting in guard for someone in order to attack. The verb is colloquial in style. In (26) the ambush is seen as a prerequisite for the (unwanted for the speaker) scoring and in (27) the speaker hopes that the possibility of being followed will fall upon another person (probably as a revenge). The slightly different uses of podskazat’ in (28)−(31) include both telling in secret and hinting, and the cognitive model of secondary, less important, action is also involved. This cognitive model conceptualizes things that are under as being less important. What cannot be said out loud can sometimes be said either secretly (only to someone), or selectively (only some parts of it). Both of these options, by comparison with the most prototypical form of telling or saying, have in common the notion of secrecy, of not revealing something publicly. In accordance with this, podskazat’ is also used in polite requests as if ‘saying a little’ would cause less inconvenience than saying.
The parameters of [VERTICAL MOVEMENT] in the category of Movement include Movement, Vertical Level, Under, and Down. [VERTICAL MOVEMENT] provides access to such cognitive models as the Dominance metaphor, Secret is Under metaphor, and Less Important is Down metaphor.
4.3 [close]
Of the twelve verbs introduced earlier, (9) podbežat’ and (10) podladit’sja can represent the lexical concept [CLOSE]. The lexical concept [CLOSE] in Russian relates to meanings that have in common the notion of approaching or being close. The semantic selectional criterion for [CLOSE] is the notion of LM approaching, coming closer to TR. The formal selectional criterion for verbs of motion is the prefix pod- accompanied by the prepositional construction k + dative.
Some examples of use for (9) podbežat’ from the spoken corpus of the RNC:
(32) Подбежала / смотрю / дышит / я растерялась / потрясла ее / сбегала за водой.
Under.run-pst.f / look-prs.1s / breath-prs.3s / I get.worried-pst.1s.f / shake -pst.1s.f her / run-pst.1s.f for water-instr
‘I ran to herFootnote 14 / look / she’s breathing / I got worried / shook her a little / ran to get some water.’
(33) Павел / но мы должны всё-таки написать. Почему вы подбежали к этой машине?
Pavel / but we must-adj.pl anyway write. Why you under.run-pst.3pl to this-dat car-dat?
‘Pavel / but we do have to write .Why did you run up to that car?’
(32) and (33) are very concrete in meaning – they mean running in such a way that one approaches, goes towards the object mentioned (or omitted, as in (37)).
The only example of (10) podladit’sja in the spoken corpus of the RNC [4.4.2013] is (34). Here the approaching is abstract and conveys the meaning of adjustment.
(34) И так он старался немножко / чуть-чуть подладиться / подладиться чуть-чуть.
And so he try-pst.3sg.m little / tiny.bit under.adjust / under.adjust tiny.bit.
‘So he tried a little / just a little to adjust / just to adjust a little.’
As can be seen from the examples above, the lexical concept [CLOSE] is an open-class lexical concept. Besides concrete approaching it is also related to ‘abstract approaching’.Footnote 15 Abstract approaching means things that have to do with changing one’s opinion, acting similarly, or adjusting one’s actions in order to please or feel psychologically closer to someone else. This metaphorical conceptualization and the frames involved are based on the features that physical and psychological closeness have in common. This notion is both metaphorical and metonymical, since physical and psychological (mental) closeness have shared features, influence one another, and are part of the same conceptual domain (metonymy), but at the same time they also involve transitions from one domain to another (metaphor). For example, we can think about the way that flattering is seen as possibly leading to a state in which the flatterer ‘gets under’ the one being flattered and gains some benefit as a result. The notions of getting close (under) and influencing are different notions (metaphor).
The parameters of the reduced linguistic content for the lexical concept [CLOSE] belong to the category of spatial relations and include Proximity and Movement. The cognitive models that [CLOSE] gives access to are the metonymy of abstract approaching and the Similar Is Close metaphor.
4.4 [contact]
Yet another lexical concept that makes it possible to use pod- with different meanings is [CONTACT]. This means that adding the prefix pod- adds the notion of getting into contact, touching, finding common ground. Among the twelve verbs this lexical concept is represented in (11) podteret’ and (12) podcepit’.
Some examples for (11) podteret’ from the spoken corpus of the RNC (4 occurrences altogether) [4.4.2013]:
(35) То там она не подтерла / то там она что-то не сделала / все скажет.
That there she not under.wipe-pst.f / that there she something not do-prf.pst.f / all say-fut.3s
‘Either she did not wipe it completely / or she did not do something else / she’ll say anything.’
(36) Ты ещё / рыбалка! Нос подотри!
You still / fishing! Nose under.wipe!
‘What are you talking about / fishing! Wipe your nose!’
In (35) and (36) podteret’ means wiping completely off, whether with reference to incomplete cleaning (35) or wiping one’s nose (36).
The semantic selectional criterion for [CONTACT] is the meaning of manipulating a surface in order to get something out of it, whether on purpose or not. The formal selectional criterion for this lexical concept is that the verb takes a direct object that has to bear the consequences of this contact − most notably the meaning of the unwanted substance disappearing completely, as in (35) and (36). Syntactically, the notion of contact is realized with pod- added to a verb. The verbs involved tend to have a meaning that is compatible with the idea of influencing the surface in order to get everything out of it. Besides contact, they activate other lexical concepts as well, namely [UNDER] and [VERTICAL MOVEMENT] and have through them a connection to the cognitive models of less important action and secrecy. Other consequences of contact with a surface include wearing out, being beaten out, or being damaged as a result of contact, establishing a connection, or fastening with something. In this way, the contact with a surface results in the surface becoming completely clean, wearing (out), pushing into something, filling holes, or establishing some kind of connection.
Examples (37)−(41) of (12) podcepit’ include establishing a kind of contact. Coming into contact, hooking into something, is sanctioned by the lexical concept [CONTACT] and is explainable by the cognitive models accessible through this lexical concept. These cognitive models include frames for getting something (such as a word in (23)) from somewhere, or taking someone away with oneself (as in (24)). The establishing of a connection at the formal level is metonymically connected to these actions, which explains this sanctioning. In this case, the first such model would be the metonymical conceptualization of seeing something that is under something else as its (coherent) part, as is the case when action directed at the rear side of something means action that is secondary in importance. It also includes the notion of contact. An example of such conceptualization would be (12) podcepit’ ‘hook, catch’,Footnote 16 in which the hooking is directed to the bottom (invisible) side of something. This verb can refer to making a concrete connection as in (37), where the hooking means putting something (another car or trailer) behind a Volga automobile, or catching with a hook when fishing (38). These kinds of concrete uses are rare in the spoken corpus of the RNC; the verb is more often used in abstract contexts. Grammatically, the construction can omit the part to which the thing being hooked is attached, as in (38), which does not mention the hook onto which the fish is caught (nor does the Russian verb include a stem meaning a hook).
(37) Слушай / друг / давай подцепим к твоей «Волге»!
Listen / friend / let.us under.attach-prf.1pl to your-dat “Volga”-dat!
‘Hear me / friend / let’s attach it to your “Volga”!’
(38) Это я подцепил эту форельку пять лет тому назад.
It I under.attach this-acc trout-acc five years that-dat back.
‘It was I who caught that trout five years ago.’
Here a metonymical conceptualization is further related to a metaphorical cognitive model that refers to what catching or connecting into the peripheral side can mean. It can be, for example, starting to use a new word (39), finding a boyfriend (40), or catching a disease – both in a concrete sense and in the case of computer viruses (41).
(39) Фу / где ты это слово дурацкое подцепила / а?
Ugh / where you this word stupid under.attach-pst.f /what?
Shame on you / where did you get that stupid word from / tell me?
(40) Ладно / так.уж и быть. Я вчера такого потрясного парня подцепила!
OK / so and be-inf I yesterday such-acc brilliant-acc boy-accunder.attach!
OK / so let it be. Yesterday I caught myself such a brilliant boy!
(41) Боишься включать / потому.что кроме вируса ничего не подцепишь.
Be.afraid-2sg turn.on-inf / because without virus-gen nothing-gen not under.attach-2sg
‘You are afraid of turning it on / because all you will get is a virus.’
In (40), the fact that a boy is used as an object that is ‘hooked’ or ‘pinned’ to a person reveals a metaphorical conceptualization where a human being is seen as something that can be attached to something. The love, closeness, and other benefits of having caught the boy are conceptually compared to having something attached to oneself and being thus available for use. The parameters of compressed linguistic content for [CONTACT], associated with the category of spatial relations are Contact, Touching, and Not Touching. [CONTACT] gives access to the cognitive models that involve the Under Is Less Important Metaphor and Part−whole metonymy.
5 Discussion
Table 1 offers a description of the semantic and formal selectional tendencies, linguistic parameters, and related cognitive models of the lexical concepts [UNDER], [VERTICAL MOVEMENT], [CLOSE], and [CONTACT].
My aim has been to use the notion of lexical concept as an interface between meanings realized in context and the cognitive models involved. This leads to a number of new observations. The different meanings of pod- (reported in Plungjan, Reference Plungjan2001) are related to cognitive models in at least the following ways:
1. The purely spatial meanings of pod- (‘being under’, ‘being down’, ‘going under’, ‘from under’, ‘upwards’, and ‘downwards’) correspond to the lexical concepts [UNDER] and [VERTICAL MOVEMENT], depending on whether they include the downward movement only or combine it with a spatial relationship that includes a landmark.
2. Of the non-spatial meanings, supplementary action (additional, simultaneous, or lesser action) is described in terms of the lexical concept [UNDER] and is connected to the metaphorical model that sees less important or non-dominant things as being down.
3. The meaning of causing harm or damage is described in more general terms as sanctioned by the lexical concept [CONTACT]. The cognitive model explaining this is the metonymy whereby the results of the contact are represented by the contact itself.
4. Doing something in secret has to do with a metaphorical and metonymical extension of the lexical concept [UNDER], including the notion of hiding under something.
5. Finally, getting abstractly closer is a metaphorical extension of [CLOSE].
Thus, my analysis of the uses of pod- is different from the grouping of its meanings by Plungjan in that using the notion of lexical concept helps us to be more precise in describing the formal and semantic differences between the various meanings. Describing them with a mere dictionary-type list of presumably stable meanings cannot tell the whole truth, especially since several lexical concepts can actually be combined in a single instance of use. The whole question is stated differently, and can be formulated as follows: What are the factors that enable a lexical item to be used in different meanings and contexts? In a specific case the same question can be put even more simply: What do different meanings have in common?
My material was collected from the standpoint of concrete linguistic forms including the prefix pod-. This makes it different from material collected for a particular lexical concept or in connection to cognitive models. It is also interesting that the linguistic unit in question is not a word but a prefix, which as a linguistic unit has its own meaning, but also works in cooperation with the verb stems to make up a complex semantic unit in which both parts influence one another. Thus the meaning of the prefix pod- is influenced in many cases by the meaning of the verb that it is combined with, its grammatical characteristics, and other features.
The different lexical concepts that sanction different uses of pod- are not in contradiction with the possible inter-relations that Dobrušina et al. (Reference Dobrušina, Mellina and Paillard2001) propose for explaining the roles of the prefix and the word stem in the process of creating meaning. According to them, the possible configurations for pod- are: (A) co-existence, where the prefix adds to the meaning of the word stem a meaning that is not foreign to it and that characterizes the action without changing the grammatical features of the verb; (B) re-arranging, where meaning components from the prefix supersede meaning components of the verb stem and result in change in grammatical behaviour; and (C) crossing, where the meanings of the prefix and stem work at the same time. Comparing this with the analysis just given, the difference between the various lexical concepts corresponds exactly to the difference between these configurations. As stated earlier, the difference between [UNDER] and [VERTICAL MOVEMENT] is that in [UNDER] the prefix and the stem are equal in determining the meaning potential, while in [VERTICAL MOVEMENT] the prefix merely modifies the meaning of the stem. This would indicate that [UNDER] offers for pod- configuration A, while [VERTICAL MOVEMENT] offers configuration B. As for the other lexical concepts discussed, [CLOSE] could be interpreted as yielding for pod- configuration B and [CONTACT] configuration C. This means that the relationship between the prefix and the verb when creating meaning can be one factor in determining the processes of meaning production with Russian prefixes, and this is not in contradiction with the analysis provided here.
6 Conclusion
The material reflecting different uses of pod- was explained with reference to four lexical concepts. The different meanings recorded for this prefix in previous research are in no contradiction with these lexical concepts. The strength of this kind of description comes from the understanding that a word-formation process such as prefixation can be seen as dynamic. The combinations of a certain prefix with different verb stems are the first dynamic step in this process, and the use of the prefixed verb in context the final one. The most interesting question is how common cognitive models and their combinations can explain the relationship between these meanings.
The lexical concept [UNDER] and the cognitive models to which it provides access explain many uses of the Russian prefix pod-. The cognitive models that were mentioned in this paper in relation to the prefix pod- were the frames for subordination, time coverage, domination, control, doing in secret (under cover), and mental closeness. These cognitive models consist of different sorts of frame. Besides proving right the meaning categories for pod- that previous research has suggested, the analysis also confirmed certain assumptions about the nature of polysemy in the uses of pod-. Each prefixed verb with pod- can be used in several meanings. In a particular context, the meaning involved is almost always clear, and polysemy in the verbs can cause difficulties in comprehension only when this is the intention of the speaker or writer (in fiction and puns). In this way, these verbs are usually polysemous only out of context.
For LCCM theory, polysemy is logical. The different meanings of a word or another linguistic unit are seen rather as a norm, since the context is emphasized and the meaning or semantic content of different uses of the unit are always different. LCCM theory is able to illustrate the components of meaning through the interface of lexical concepts, and also suggests a solution to the problem of figurative meanings for a lexical unit. For the Russian prefix pod-, the level of lexical concepts suggests a new approach for grouping its different meanings in a way that is both cognitively plausible and enables us to interpret the relationship between the meanings.
