Expressing Ingressivity in Slavic : The Contextually - Conditioned Imperfective Past vs . the Phase Verb stat ' and Procedural za - * Stephen

This article discusses different modes of expressing ingressivity in the Slavic languages—the grammatical expression of ingressivity (by means of imperfective verb forms) and its lexical expression (by means of the use of stat' as an ingressive phase verb or perfective procedural verbs prefixed with za-)—and relates them to one another as two competing systems. It is shown that these phenomena are in complementary distribution: languages that employ the contextually-conditioned imperfective past to a high degree only employ stat' and zato express ingressivity to a very low degree or not at all, and vice-versa. More specifically, the contextually-conditioned imperfective past is characteristic of the extreme western end of Slavic (Czech, Slovak, Sorbian, Slovene), whereas stat' and zaare characteristic of an eastern group of languages (Russian, Ukrainian, Belarusian, Bulgarian); two languages (Polish and Serbo-Croatian) occupy a “transitional” position between the two groups. Finally, the respective modes of expressing ingressivity are discussed within the theory of Slavic aspect developed in Dickey 1997. 0. Introduction and Theoretical Background In Dickey 1997, I examined Slavic aspectual usage in several categories of discourse context—repeated events, the general-factual, the historical present, the scenic present and other kinds of instructions, performatives and other cases of coincidence, the use of imperfectives in sequences of events, and also the effect of aspect semantics on the derivation of verbal nouns—and presented abundant evidence that Slavic aspect is not a monolithic phenomenon. Rather, the Slavic languages pattern into two distinct groups: an eastern group (Ru, Uk, Br, Bg) and a western group (Cz, Sk, Sor, Sn); Pol and SC are transitional zones between them. * I would like to thank Susan Kresin, Gary Toops, and Charles Townsend for valuable comments on a conference presentation of this paper. I am also grateful to Lilia Burganova (Russian), Boris Kovatchev (Bulgarian) and Dariusz To∏czyk (Polish) for production of and/or comment on examples. Two anonymous JSL reviewers also gave many helpful criticisms. Of course, none of the above are responsible for any errors or inaccuracies contained in this paper. 1 The following abbreviations are used in this article: ipf imperfective Cz Czech Sk Slovak pf perfective Pol Polish Sn Slovene Bg Bulgarian Ru Russian Sor Sorbian Br Belarusian SC Serbo-Croatian Uk Ukrainian 12 STEPHEN M. DICKEY On the basis of details of aspectual usage in the different languages, I offered a theory of the semantic meaning of aspect in the respective groups. In the western group, the pf aspect views a situation as an integral, complete whole (totality, the meaning often ascribed to the pf in all of Slavic), and the ipf aspect carries a meaning of the assignability of a situation to more than one point in time (quantitative temporal indefiniteness). In the eastern group, the pf aspect views a situation as a complete whole occupying a single, unique point in time relative to other situations in the fact structure of a discourse (temporal definiteness), and the ipf aspect expresses the contradictory opposite, the inability of a situation to be assigned to a single, unique point in time (qualitative temporal indefiniteness). For details, the reader is referred to Dickey 1997. This discussion will relate one of the parameters I discussed, the (primarily ingressive) use of ipf verbs in sequences of events, and compare it to the use of stat' as an ingressive phase verb and ingressive procedural verbs in za-. The theory of language employed in this discussion is that of Cognitive Grammar (CG; for details, see Langacker 1987). In CG, meaning is what motivates all aspects of language and its use. The meanings of linguistic units are categories, which have internal structure. A category’s internal structure is centered around a prototype, a term which has already acquired different meanings in different analyses; this discussion employs Taylor’s (1990: 529) slight modification of Langacker’s (1987: 371) definition: a prototype is a “mental representation (possibly, one quite rich in specific detail) of a typical instance of a category, such that entities become assimilated to the category on the basis of perceived similarity to the prototype” [my emphasis—SMD]. Dickey 1997 argues that the prototype of the eastern pf is temporal definiteness, as defined above, and that the prototype of the eastern ipf is qualitative temporal indefiniteness. Further, in the western languages, the prototype of the pf is totality; the prototype of the western ipf is quantitative temporal indefiniteness. This discussion will not hinge on the issues motivating such a prototype approach to Slavic aspect (the interested reader is referred to Dickey 1997). What is important here is that these categories of totality and temporal definiteness form the central semantic mechanisms of the respective aspectual categories in the various languages, motivating different strategies in expressing ingressivity. The relationship of aspect to lexico-semantic types of predicates will be relevant in section 2 and elsewhere. Since Vendler 1957, verbs have generally been divided into four classes: states, activities, accomplishments, and achievements. States are “nondynamic situations without EXPRESSING INGRESSIVITY IN SLAVIC 13 natural conclusions” (e.g., know, hate); activities are “dynamic processes where any part is of the same nature as the whole” (e.g., dance, read); accomplishments are “goal directed situations [...] characterized by the presence of an activity preceding the end-point” (e.g., read a book, walk a mile); achievements are “instantaneous leaps from one state into another without an accompanying activity” (e.g., notice, begin). 1. The Contextually-Conditioned Imperfective Past In his 1961 study, Svetomir Ivanãev described for the first time in detail the widespread use of ipf verbs in Cz to denote events occurring in sequence. He labeled this phenomenon the “contextually conditioned ingressive use of imperfective verbs in Czech” (kontekstovo obuslovena ingresivna upotreba na glagolite ot nesvâr‰en vid v ãe‰kija ezik; henceforth contextually-conditioned ipf past or CCIP). Ivanãev’s study is in fact more than a treatment of Cz: he also discusses the phenomenon in a majority of the modern Slavic languages, as well as in Old Church Slavic (Old Bulgarian), Old East Slavic (Old Russian) and Old Czech. Outside of Cz and Sk the CCIP is less common and reaches a minimum in the east (Ru and Bg). On the basis of cross-Slavic data for the CCIP, Ivanãev divides Slavic into a western group (consisting of Cz, Sk, and Sor) and an eastern group (consisting of East Slavic and Bg) and considers Pol, Sn and SC to be transitional zones in the north and south (respectively), even using that very term (prehodna zona; Ivanãev 1961: 49). In the following sections, the CCIP is described largely on the basis of Ivanãev’s excellent collection of data, examined and then explained in terms of the theory of aspect offered in Dickey 1997. Then, an examination is made of stat' as an ingressive phase verb and ingressive procedural verbs in zain the various Slavic languages. The distributions of the CCIP on the one hand, and of the special ingressive verbs on the other are found to be complementary, and the division is then incorporated into the theory of aspect offered in Dickey 1997. 1.1. The Contextually-Conditioned Ipf Past in Czech In its broadest definition, the Cz contextually-conditioned ipf past is the use of an ipf past-tense form in narrated sequences of events, where some 2 The descriptions here are Brecht’s (1984: 10–11) concise paraphrases of Vendler 1957. 3 The reason for omitting the word “ingressive” of the original is that, as pointed out below, ingressivity is not in fact the core meaning of the form, but rather one contextual interpretation. I specify past, since this discussion is limited to past-tense narratives. 14 STEPHEN M. DICKEY other Slavic languages (Ru and Bg, for instance) generally require a pf form or strongly prefer it. Ivanãev (1961: 5) observes: In its purest and simple form this phenomenon occurs in complex clauses, in which the predicates of individual simple clauses, joined by the conjunction a [‘and’], are in the past tense (preterite), have one and the same subject (either singular or plural), and the first is a verb of the perfective aspect which denotes the completion of an action, whereas the one immediately following the conjunction a is a verb of the imperfective aspect and denotes an ingressive action, the beginning of the action that is expressed by the verb. The following example is representative: (1) Zvedl se tedy a ‰el 5 k vychodu. ‘Then he got up and went to the exit.’ [Cz; Ivanãev 1961: 11; Drda, Mûsteãko na dlani] As noted in a number of studies (e.g., Ivanãev 1961; Galton 1976; Stunová 1993), motion verbs account for a large number of occurrences; however, verbs of other types are quite common: (2) Sedl si a psal. ‘He sat down and started writing.’ [Cz; Ivanãev 1961: 6] The examples to be considered here are largely limited to Ivanãev’s basic structure, which involves a “pf verb... a + ipf verb”, although what is really important is the presence of a predicate coded as ipf in a sequence of foregrounded events. Ivanãev (1961: 5–6) emphasizes that, in his opinion, the expression of ingressive actions in sequences of events in Cz is more commonly expressed by means of ipf forms than by the combinations of phase verb + infinitive or other kinds of ingressive procedural verbs exemplified in (3): (3) a. Sedl si a zaãal psáti. ‘He sat down and began to write.’ 4 Unless otherwise indicated, all translations and glosses are my own. 5 Czech specialists almost unanimously consider Cz jít ‘go’ to be ipf. For details, the reader is referred to Bondarko 1961 and Kopeãn ̆ 1961. EXPRESSING INGRESSIVITY IN SLAVIC 15 (3) b. Sedl si a rozepsal se. ‘He sat down and got started writing.’ [Cz; Ivanãev 1961: 6] KfiíÏková 1963: 287 seconds his opinion. Although the CCIP often occurs in contexts where ingressivity is an appropriate interpretation, several studies (Ivanãev 1961: 83; KfiíÏková 1963: 287; Galton 1976: 70; Stunová 1993: 107) point out that ingressivity is not necessarily expressed, i.e. it is not an entailment. Other aspectual nuances are possible, even, according to Stunová, “terminativity”. In the context of a narrative sequence containing pf forms, the presentation of an action from within its midst via an ipf form leaves the immediately prior beginning of that action as an inference. In this regard, KfiíÏková’s (1963: 287) observations on the connection to ingressivity are particularly helpful: In cases when it is impossible in Russian or especially in Bulgarian to leave ingressivity unexpressed, in the Czech linguistic consciousness [ingressivity] does not come to the fore. Only on the basis of a logical analysis and comparison with other languages does a Czech determine that he is actually expressing an action which in another Slavic language is conceptualized as in its beginning. What a speaker of Russian or Bulgarian conceptualizes as a beginning action is felt by a speaker of Czech to be neutral with respect to ingressivity, lacking an explicit meaning of the beginning of an action. Between perfective actions, which are projected onto the time axis as points, imperfective verbs appear as [linear stretches]; the speaker actually stylizes the [action] in such way that it is impossible to be conceived synoptically with its beginning and end, we are, as it were, in the middle of it. [emphasis mine— SMD] The first sentence of KfiíÏková’s remark means that, in cases where Cz uses an ipf form which is often interpreted as expressing ingressivity, Ru and Bg necessarily view these actions as ingressive, and must express this by means of pf phase verbs (e.g., Ru naãat', stat' ‘begin’) or special ingressive procedural verbs (e.g., Ru zaigrat' ‘start playing’, pojti ‘set out [on foot]’). This will be illustrated in examples in 1.2. The idea of viewing a situation as from within its midst is particularly important; Stunová (1993: 126–27) gives this concept the label in medias res (the term adopted here). Another important effect of the CCIP 16 STEPHEN M. DICKEY on the textual level is what Galton (1976: 70) calls the continuity of action: “the ipf. verb makes the action join on immediately to the preceding one, almost without a seam”. Ivanãev (1961: 82) describes the CCIP in compatible terms: “Above all, this mode of expression enables actions occurring successively to be presented without clearly defined contours, whereas their concrete aspectual semantics are only alluded to...” [emphasis mine—SMD]. In my view, these are the most salient and relevant properties of the CCIP; although each scholar characterizes the phenomenon in his or her own way, their intuitions are complementary and often overlap, expressing the same idea in different terms. Before examining concrete instances, however, it should be pointed out that some of the adduced examples may be explained effectively in other ways. This is Galton’s (1976: 69) opinion: “Ivanãev has collected many excellent examples to illustrate his ‘contextually-conditioned’ function of the ipf. past, but different interpreters will, needless to say, explain some of his illustrations differently, e.g., verba dicendi and related ones need not belong here, others are clearly conative, etc.” Yet, as Galton goes on to say, given the abundance of examples there can be no doubt that this is a real phenomenon in Cz. One case which I do think is worth separating from more canonical instances is one which Stunová includes in the category of in medias res; in my opinion it belongs more properly to backgrounding than to sequences of foregrounded events. Stunová includes many examples in which the ipf past is introduced by kdyÏ ‘when’. Consider (4): (4) Ale kdyÏ uÏ jsem sedûl ve vagónû a vlak se hnul, já jsem se, pane, dal do breku jako mal ̆ kluk... ‘When I was already sitting in the car and the train lurched forward, I began sobbing, sir, like a little boy...’ [Cz; Stunová 1993: 142; âapek] In this case it is doubtful that the ipf form denotes a foreground event sandwiched between other foregrounded pf forms; in my opinion this ipf verb (as well as the pf se hnul) sets up a (new) static background against which other sequential (perfective) events will be foregrounded. The presence of kdyÏ ‘when’ suggests that sedûl is simply functioning to form the background of an episode. Such cases can also be found in Ru narratives, even within Stunová’s discussion: (5) Kogda ja uÏe sidel v vagone, poezd tronulsja, i tut ja zaplakal... [Ru; = (4)] EXPRESSING INGRESSIVITY IN SLAVIC 17 Ivanãev does not include cases of stative ipf past forms introduced by kdyÏ ‘when’ in his data (note that they do not conform to his original definition), and I think that they are best treated separately as cases of backgrounding. This discussion will be limited to examples that appear to be most representative of the phenomenon, leaving out verba dicendi and other arguably marginal types. Let us now examine some further examples of the CCIP. Although the CCIP usually occurs after a pv past-tense verb within a single sentence, Stunová gives a rather interesting example of three ipf forms denoting foregrounded events: (6) KdyÏ me vidûl, porouãel se té paniãce a ‰el ke mû. ‘When he saw me, he took leave from the lady and came up to me.’ [Cz; Stunová 1993: 112; âapek] Stunová observes that the “interpretation of a sequence is preferred here”. This is certainly true: no one would claim that the events depicted are strictly simultaneous. However, my informants confirmed that the events are easily interpreted as “partially overlapping”, and in fact preferred this interpretation. Accordingly, the interpretation is that the man began taking his leave while still looking at the narrator. Informants also confirm Galton’s idea of the seamless continuity of the actions, i.e. the smooth transition from one action to the next, as opposed to a clear articulation of events occurring in strict sequentiality. The latter must be coded as pf in Cz as well: (7) KdyÏ me uvidûl, odporouãel se té paniãce a pfiistoupil ke mû. [Cz; Stunová 1993: 112 = (6)] Galton gives another example of the continuity of action: 6 It should be kept in mind that narratives are not all neatly divisible into two simple categories of backgrounded and foregrounded actions. Thelin (1990) subdivides foreground into actual foreground and contextual foreground, and background into actual background and proper background (for details, see Thelin 1990: 22–29). In addition, narratives often consist not a single episode, but of several constituent, selfcontained episodes (Koschmieder’s (1979: 143) term is Knoten ‘knots’ of events), each of which contains its own background and foregrounded events. In examples such as (4) and (5), the ipf form is part of the introduction of a new episode, forming a background against which new foregrounded events will appear. 7 Ivanãev himself (1961: 79) suggests that verba dicendi should probably be treated separately. 18 STEPHEN M. DICKEY (8) Hasiãi pfiestali stfiíkat, zatroubili a jeli domu. ‘The firemen stopped spraying water, sounded the horn and went home.’ [Cz; Galton 1976: 70; âapek] Here Galton observes that there is no “joint” between the signal and the fire brigade’s incipient ride home—they gave the signal as they were setting into motion. One may consider the effect of in medias res in the following manner: by placing the reader’s perspective in the midst of an action which directly follows another, the beginning boundary of the second action is ignored or defocused, not only yielding the smoothness of transition mentioned by Galton, but also allowing the related interpretation of partial overlap of the actions. 1.2 The Contextually-Conditioned Ipf Past in Other Languages Ivanãev (1961: 40–48) provides abundant data from other West and South Slavic languages. According to Ivanãev, the CCIP is most characteristic of Cz, but is well-represented in the other West Slavic languages. Sk examples are given in (9): (9) Chytil ma za ruku a t'ahal. ‘He grabbed me by the arm and started pulling.’ [Sk; Ivanãev 1961: 44; Ondrejov, ·ibeniãné pole] However, Ivanãev notes his impression that Sk employs the CCIP somewhat less than Cz, occasionally opting for a pf verb or a pf phase verb + infinitive: (10) a. Ujo Gajdo‰ik ma schytil okolo hrdla a pohli sme sa na mesto. ‘Uncle Gajdo‰ik grabbed me by the throat and we left for the city.’ [Sk; Ivanãev 1961: 45; Ondrejov, ·ibeniãné pole] b. Za‰vihol som prútom a zaãal som zaháÀat' pochab ̆ k⁄dlik na mesto. ‘I lashed with the switch and began to drive the raucous little herd into the town.’ [Sk; Ivanãev 1961: 45; Ondrejov, ·ibeniãné pole] The phenomenon is also represented in Upper and Lower Sor. Examples are given in (11) and (12): EXPRESSING INGRESSIVITY IN SLAVIC 19 (11) Zeƒdêechmoj na ∏ubju a ãakachmoj z nowa. ‘We went down to the deck and started waiting again.’ [Upper Sor; Ivanãev 1961: 46; ̧aras, Zastojãki] (12) ·wjela zejmje ‰apku a bjatowa‰o “WoÊe na‰”. ‘·wjela took off his hat and started praying the Pater noster.’ [Lower Sor; Ivanãev 1961: 46; Chrestomatija dolnoserbskego pismowstwa] In (9, 11–12) we see that in the other languages of the western group the CCIP is fairly well-represented. Ivanãev (1961: 48) considers Sn to be the beginning of a transitional zone that includes SC. He implies that the CCIP is less frequent in Sn than Cz, but points out that he did not have access to extensive data. Thus, a precise description of Sn in this respect still needs to be made. Representative Sn examples are given in (13): (13) a. Marijan je zgodaj ustal, poklical psa s seboj in ‰el ãez travnik. ‘Marijan got up early, called his dog to him and went through the meadow.’ [Sn; Bajec et al. 1971: 202] b. Îupan je naslonil pleãati hrbet na klop in je gledal mrko. ‘The mayor leaned his broad back on a bench and watched darkly.’ [Sn; Ivanãev 1961: 48; Cankar] Indicative of the relative rarity of the CCIP in Sn is the fact that it apparently occurs to a greater degree with the verb of motion iti ‘go’, and not as frequently with other types of verbs. A complicating factor is that Sn iti appears to be biaspectual (as is the SC equivalent içi), so that the only clear examples of the phenomenon are those with unambiguously ipf verbs, such as (13b). On the basis of Pol translations of Cz originals as well as Pol originals, Ivanãev (1961: 41) determines that the CCIP “is not completely unknown to Polish, but [...] it is relatively very rare and consequently [...] an element of the style of individual writers” (the emphasis on the individual styles of Pol writers was seconded by my informant). He observes that the dialect background of an individual writer determines the extent of his predilection for the CCIP. He also notes that in Pol the CCIP occurs primarily with verbs of motion. The examples in (14) are representative of motion verbs: 20 STEPHEN M. DICKEY (14) a. ... i wnet wysun ́∏a si ́ kobiecina niska, okryta szarà p∏achtà i sz∏a po ∏awie, p∏acz ́cy za wod ́... ‘... and soon a low woman emerged, wrapped in a grey sheet, and went along the bench, crying, for water...’ [Pol; Ivanãev 1961: 41; Orkan, Komornicy] b. Poszed∏ w las, a Ma∏wa wybieg∏ na drog ́ i jak szalony p ́dzi∏ ku Morzelanom. ‘He went into the forest, but Ma∏wa ran out onto the road and rushed as if mad toward Morzelany. [Pol; Ivanãev 1961: 41; Dygasiƒski, Zajàc] Ivanãev also gives examples from other types of verbs, e.g.: (15) a. Jakub usiad∏ na ∏ó ̋ku i szlocha∏. ‘Jakub sat down on the bed and started sobbing.’ [Pol; Ivanãev 1961: 42; Na∏kowska, Niecierpliwi] b. Czerwony krzy ̋ w Lublinie wy∏ama∏ nareszcie furtk ́ na pole um ́czonych wi ́ ̋niów, dostarcza∏ chorym po ̋ywnych zup i... ‘The Red Cross in Lublin finally broke down the wicker gate to the field of exhausted prisoners, started delivering nutritious soups to the ill and...’ [Pol; Ivanãev 1961: 42; Orzeszkowa, Meir Erofowicz] Ivanãev (1961: 47) notes that CCIP occurs much more rarely in SC than in the western group. In fact, he gives no good examples, only cases where the Cz ipf past is translated into SC either with a pf phase verb + infinitive (16) or with a pf verb (17): (16) a. Nûkolik dní to pomáhalo, Gierke se drobet uklidnil, ale pak zase pobíhal od okna k oknu a lomcoval tûmi mfiíÏemi, aby se pfiesvûdãil... ‘That helped for a few days, Gierke calmed down a little, but then he started running from window to window again and shaking the grating to make sure...’ [Cz; âapek] b. To je nekoliko dana pomoglo, Girke se malo smirio, ali naskoro je opet poãeo da trãkara od prozora do prozora, drmajuçi pri tom re‰etkama da se uveri... [SC; Ivanãev 1961: 47] EXPRESSING INGRESSIVITY IN SLAVIC 21 (17) a. V tu chvíli tam vrazil Oberhuber, fialov ̆ vztekem, a kfiiãel uÏ ve dvefiich. ‘At that moment Oberhuber burst in, livid with rage, and screamed from the doorway.’ [Cz; âapek] b. U taj ãas uleteo je unutra Oberhuber, crven i modar od besa, i veç s vrata povikao. [SC; Ivanãev 1961: 47] Ivanãev’s opinion that such ipf usage occurs in SC much less than in the languages of the western group is accurate. However, infrequently one does find similar usage of the ipf in SC. One very good colloquial example which I have found is (18): (18) Nakon ‰to je do‰lo do poga∂anja ovog traktora kao ‰to sam opisao, nastala je velika panika i ljudi su beÏali, a svi ovi ljudi i Ïene koji su bili u mom traktoru iskoãili su iz prikolice [...] ‘After this tractor was hit, as I described, a great panic ensued and the people started running, and all the men and women who were in my tractor jumped out of the trailer [...]’ [SC] Consider also (19), which, while not representative of Ivanãev’s phenomenon in a narrow sense, is a good example of in medias res, the placing of one’s perspective in the midst of a situation in a sequential relationship with other situations: (19) Me∂utim prolazili su meseci pa i godine a novac nije stizao. Momci su napu‰tali posao. ‘However, months passed and even years, but the money did not arrive. The workers started leaving their jobs.’ [SC; Andriç, Na Drini çuprija] Informants, when asked exactly what the ipf napu‰tali means in the context, readily paraphrase it as momci su poãeli da napu‰taju posao ‘the workers began to leave their stations’, but point out that the ingressive meaning of (20) is not as distinct as that expressed by a phase verb; a phase verb pinpoints the time when the action started and foregrounds it. Ivanãev 1961: 48 observes that farther eastward, in Ru and Bg, occurrences of the CCIP reach a minimum. (He gives no data for Uk or Br but suggests that the situation in those languages is identical to that in Ru. Although this is very likely the case, I have not yet had the opportunity to confirm this with Uk and Br informants, and therefore must refrain from 22 STEPHEN M. DICKEY including them in the parameter for the CCIP.) He gives numerous examples where a Cz ipf past form is translated into Ru with a pf phase verb (often stat' ‘begin’) + ipf infinitive: (20) a. ... zvolna si sedl vedle mne a Joseka, poloÏil hlavu do dlanû a díval se na mne. ‘... he sat down quietly beside Josef and me, put his head in his hands and looked at me.’ [Cz; Ivanãev 1961: 36; Nûmcová] b. ... on tixo sel vozle menja i Iosefa, sklonil golovu na ruki i stal smotret' na menja. [Ru] (21) a. ... pak vyprovodila Bára El‰ku domÛ a od té doby chodívaly zase k sobû. ‘... then Bára accompanied El‰ka home and from that time they began to go to one another’s homes again.’ [Cz; Ivanãev 1961: 37; Nûmcová] b. ... Bara provodila Èl‰ku domoj. I s tex por oni opjat' stali xodit' drug k drugu, kak i preÏde. [Ru] Particularly illustrative of the eastern extreme are examples in which Cz and Pol have ipf past forms, but the Ru translation has a pf verb/phase verb: (22) a. Jednou k nám pfii‰el a já breãela a Ïalovala si mu na tu chudobu. ‘Once he came to our place and I started sniveling and complaining about that poverty.’ [Cz; Ivanãev 1961: 39; Nûmcová] b. Raz przyszed∏ do nas, a ja becza∏am i skar ̋y∏am mu si ́ na t ́ bied ́. [Pol] c. OdnaÏdy pri‰el on k nam. Ja stala xnykat' i Ïalovat'sja na ni‰ãetu... [Ru] 8 Since the writing of this article, however, I have had an opportunity to question two Uk informants, both of whom rejected ipf verbs in ingressive contexts, as shown in the following examples: (i) Vin siv ta *pysavi/ stavp pysaty/poãavp pysaty. ‘He sat down and started writing.’ [Uk; = (2)] (ii) Jakos' ja vzjav klarnet ta *hravi/zahravp. ‘Once I picked up the clarinet and started playing.’ [Uk; = (27)] This informant data confirms Ivanãev’s opinion, and Uk will therefore be included in the eastern group for this parameter. EXPRESSING INGRESSIVITY IN SLAVIC 23 However, Ivanãev gives just as many examples in which Pol patterns like Ru, as in (23): (23) a. Jednou vzal jsem klarinet a pískal; uãitel to sly‰el... ‘Once I picked up the clarinet and started playing; my teacher listened...’ [Cz; Ivanãev 1961: 38; Nûmcová] b. Raz wzia∏em klarnet i zaigra∏em... [Pol] c. Raz kak-to vzjal klarnet i zaigral... [Ru] In the last example, the Ru equivalent employs an ingressive procedural verb where Cz has a simple ipf past; occasionally one runs into an ordinary pf verb in the Ru equivalent, as exemplified in (24): (24) a. ... vykfiikla na plno, aÏ hoch uskoãil leknutím, a chodci se zastavovali... ‘... she shouted so loud that the boy jumped up in fear, and the passers-by started stopping...’ [Cz; Ivanãev 1961: 39; Pujmanová, Na kfiiÏovatce] b. ... zakriãala ona tak gromko, ãto Ondrej vzdrognul, a proxoÏie ostanovilis'... [Ru] Ivanãev does note, however, that even in Ru one can find sporadic instances of similar usage. Consider the examples in (25): (25) a. Vse raboãie, ostanoviv svoi raboty, podnjav topory, dolota prekratili stukotnju i smotreli v oÏidanii. ‘All the workers, having stopped their work and raised their hatchets and chisels, stopped their hammering and watched in expectation.’ [Ru; Ivanãev 1961: 43; Gogol', Taras Bul'ba] b. Potom ona vdrug obratilas' k knjazju i, grozno naxmuriv brovi, pristal'no ego razgljadyvala. ‘Then she suddenly turned to the prince and, having frowned threateningly, began to examine him closely.’ [Ru; Ivanãev 1961: 43; Dostoevskij, Idiot] Such examples, however, do not seem to be entirely parallel to those from Cz and the other western languages; there is a tendency for the ipf past form in Ru to occur with certain kinds of adverbials focusing on the 24 STEPHEN M. DICKEY manner of action or slowing it down, or to occur in sequence with pf verbal adverbs (25b), which are not generally employed in the denotation of foregrounded events in an episode. In any case, examples that might qualify as the CCIP do not occur nearly as frequently in Ru as in the languages of the western group. This is confirmed by Stunová’s (1993: 112) frequency counts for Ru and Cz predicates in sequences of events: out of a total of 140, Ru had 135 pf forms (96%) and 5 ipf forms (4%), whereas Cz had 75 pf forms (54%) and 65 ipf forms (46%). Ivanãev observes that examples of the CCIP are very rare in Bg. Examples with the ipf aorist are, according to Ivanãev (1961: 35), isolated, and occur with only a very few lexical items. He gives examples with ãakam ‘wait’ and plaãa ‘cry’: (26) a. Bojão se prituli pak zad edin kamâk i ãaka da vidi Marijka, kato trâgne... ‘Bojão hid again behind a rock and waited to see Marijka leave...’ [Bg; Ivanãev 1961: 35; Vazov] b. Stremski koleniãi i plaka. ‘Stremski knelt and began crying.’ [Bg; Ivanãev 1961: 35; Vazov] Ivanãev also gives a few examples of ipf imperfects, such as the following: (27) ‘Sega e minutata da izfirjasa, bez da go usetjat’—misle‰e si Kovaãev, kato xvârli bârz, znamenatelen pogled na Levski, kojto stana ot stola si i si popravja‰e vratovrâzkata pred ogledaloto. ‘“This is the moment for him to disappear, without them noticing him”—Kovaãev thought to himself as he threw a quick, meaningful glance at Levski, who got up and began straightening his necktie in front of the mirror.’ [Bg; Ivanãev 1961: 35; Vazov] Again, Ivanãev notes that such cases occur only in isolation. Moreover, he points out that the context allows for a past-perfect interpretation, i.e. kojto be‰e stanal ot stola si i si popravja‰e vratovrâzkata pred ogledaloto ‘who had gotten up from the table and was straightening his tie’, so that the ipf form is not really denoting an event in an articulated sequence of foregrounded events. (It should be noted that some of his Ru examples can also be explained in the same manner.) 9 Stunová 1993 Chapter 3 is an excellent data-oriented study of aspect usage in sequences of events in narratives, based on Cz and Ru parallel texts. EXPRESSING INGRESSIVITY IN SLAVIC 25 Regardless of whether one considers the Ru and Bg examples (25–27) to be exceptions or explains them in another manner, Ivanãev’s observation holds: where Cz and the western languages are likely to employ ipf verbs in sequences of events, the eastern languages still prefer a pf phase verb + infinitive or some other kind of pf verb in the vast majority of cases. Figure 1 gives a broad impression of the geographic distribution of the CCIP: Figure 1. “Geography” of the CCIP in Slavic West + Transitional (+) East –