A note on epistemic and effective meanings of the Polish perfective and imperfective.

The paper offers an analysis of selected uses of the Polish perfective and imperfective in the non-past indicative and in the imperative construction. In uses under consideration, both the perfective and the imperfective refer to a single complete occurrence of a telic process and, hence, the semantic contrast between them is not a matter of distinctions such as boundedness/unboundedness, completion/non-completion, telicity/atelicity, punctuality/durativity, etc. The paper presents a qualitative analysis of selected corpus examples which is aimed at elucidating the nature of the relevant contrast. The claim advocated in the course of the discussion is that the perfective/imperfective contrast may play a role in the system of clausal grounding in Polish, as it may convey the idea of, respectively, epistemic and/or effective non-immediacy/immediacy of the profiled process relative to the ground.


Introduction
The aim of the present paper is to offer an analysis of selected uses of the Polish perfective and imperfective. In these uses, the two aspectual variants do not exhibit their prototypical aspectual meanings characterizable in terms of contrasts such as e.g. boundedness/unboundedness, completion/non-completion, telicity/atelicity, punctuality/durativity, etc. (cf. e.g. Comrie 1976Comrie [1985, ch. 1; Forsyth 1970, 8;Holvoet 1991;Jakobson 1957; see also a discussion in Dickey 2000, 15-27 or in Janda 2004. Instead, both the perfec-tive and the imperfective refer to a single complete occurrence of a telic event involving an individual participant or individual participants. Two types of uses of this kind are exemplified in (1) and (2)  The examples in (1) involve the use of the perfective (1a) and the imperfective (1b) in the non-past indicative construction, while the example in (2)the use of the two aspectual variants in the imperative construction. In the case of both (1) and (2) the perfective and the imperfective refer to the same kind of objective situation, a single occurrence of a telic event. The semantic contrast between the two aspectual variants is not a matter of the properties of the profiled event, but rather of something else. The aim of the paper is to elucidate the nature of this contrast. The specific claim advocated in the course of the discussion is that verbal aspect plays a role in the system of clausal grounding in Polish, where the idea of grounding in general and clausal grounding in particular is understood in the sense of Langacker (cf. e.g. 1990, 12;2002, 207;2008, 474, 296-309;2009, 164). The main idea of the paper is that the perfective and the imperfective involve two different ways of grounding the process profiled by a non-past indicative or an imperative clause.
The overall considerations will start -by way of providing some theoretical background to the following analysis -with a brief general discussion pertaining to the phenomenon of grounding. This will be the topic of section 2.
Then, in section 3, the import of the perfective and the imperfective aspect in non-past indicative sentences will be considered. Section 4 will discuss the role of the perfective/imperfective contrast in the imperative construction. In both sections, the discussion will focus on a qualitative analysis of selected corpus examples. Finally, section 5 will offer a summary of the overall analysis, as well as some general conclusions.

Grounding
Grounding, as characterized by Langacker (cf. 1990, 12), involves establishing some relationship between the profiled entity (the profiled process in the case of clausal grounding, which is the immediate concern in this paper) and the ground, where the ground is understood as the speaker, the hearer, their interaction, and its immediate circumstances (cf. e.g. Langacker 1991, 53;2008, 259). The relevant relationship between the profiled entity and the ground is the relationship pertaining to "such fundamental issues as reality, existence, and speaker/hearer knowledge" (Langacker 1990, 12). As observed by Langacker (2008, 259), grounding establishes a basic connection between the interlocutors and the content evoked by a nominal or a finite clause. If left ungrounded, this content has no discernible position in their mental universe and cannot be brought to bear on their situation. It simply floats unattached as an object of idle contemplation. Langacker (cf. e.g. 2008, 474) distinguishes grounding at two levels: the epistemic level and the effective level. In the case of clausal grounding, grounding at the epistemic level has to do with epistemic control, that is, with situating the profiled process in relation to the conceptualizer's idea of reality (cf. Langacker 2008, 296-309; see also Langacker 2002, 207). In turn, clausal grounding at the effective level has to do with effective control, that is, with "the effort to influence what happens" (Langacker 2009, 164;original emphasis). In other words, effective grounding is a matter of specifying what the speaker wishes the hearer to do with the profiled content, that is, for example, whether she 1 wants the hearer to take it as a suggestion, a request, an order, etc. (see the discussion Langacker 2009, 153-158). Grounding at any of the two levels should not be thought of as excluding grounding at the other level. Thus, an expression such as You leave! profiles an instance of the process you leave, which is grounded at the effective level as an action that the speaker wishes the hearer to perform and that is to be performed in virtue of the fact that the speaker has some authority over the hearer. At the same time, the process in question is also grounded at the epistemic level as an occurrence which may only be considered as belonging to the future potentiality, rather than to reality as it currently is.
With these ideas in mind, let us now turn to an analysis of selected uses of the perfective and the imperfective aspect in non-past indicative sentences, as well as in imperatives. The focus will be on the question of whether at all andif so -how the aspectual contrast in question contributes to the grounding of the profiled process in cases under analysis.
3 The role of the perfective/imperfective contrast in non-past indicative sentences The present discussion of the perfective/imperfective contrast in non-past indicative sentences follows the lines of the analysis proposed in Kochańska (2002). The conclusion that the aspect of the main verb in a non-past indicative clause may play a role in the grounding of that clause is perhaps unavoidable in view of the data such as the examples in (3) below: (3) a. Sam pojadę 1 sg. non-past indicat. perf. jutro na lotnisko. (NCP) 'I will drive to the airport tomorrow by myself.' b. Basiu voc. , jadę 1 sg. non-past indicat. imperf. teraz do klubu. (NCP) 'Basia, I'm driving to the club now.' context: the speaker is driving through the streets of Warsaw and talking to the hearer through a mobile phone As illustrated by (3), Polish does not have two separate grammatical tenses for locating the profiled process temporally either in the present or in the future. 2 In addition to the past tense, it has just a single non-past tense. The function of locating the profiled process temporally in the present or in the future is fulfilled by aspect: the non-past indicative perfective profiles an event which is conceptualized as located in the future, while the non-past indicative imperfective profiles an event which is conceptualized as being in progress at the time of speaking, that is, as located in the present.
It may thus be argued that in cases comparable to those in (3) the perfective/imperfective contrast has to do with two different ways of grounding the profiled process: while the perfective grounds this process by locating it in the future, that is, grounds it as temporally non-immediate to the conceptualizer, the effect of employing the non-past imperfective is that the profiled process is grounded as located in the present, that is, as temporally immediate to the conceptualizer. Temporal immediacy/non-immediacy goes together with epistemic immediacy/non-immediacy: a process which is located in the future can only be, by its very nature, an aspect of potential reality, which may turn into reality only as a matter of reality's future evolution, while an event which is unfolding "in front of the conceptualizer's eyes" at present is already being included into the conceptualizer's conception of reality, that is, is already entering the conceptualizer's dominion of epistemic control (cf. Langacker 2002).
In addition, it may be noted at this point that -as suggested in Kochańska (2002, 354-360) -the above two different ways of grounding the profiled event, effected, respectively, by the perfective and the imperfective, may be viewed as related to the basic meanings of these two aspectual variants. In its basic meaning, the perfective profiles a complete process instance together with its endpoints. In turn, in its role of a grounding predication, the non-past indicative perfective profiles a future process, that is, a process which is sufficiently removed from the conceptualizer's immediate present to make the processual endpoints conceptually visible to the viewer. On the other hand, the basic meaning of the imperfective involves the removal of the processual endpoints from the region of maximal conceptual acuity. In turn, in its grounding capacity, the imperfective locates the profiled process in the present andvariant of the future form of the verb być 'be' + V inf . However, with the exception of the verb być, the verbal paradigm in Polish does not contain separate present-and future-tense forms of verbs -the present-vs. future-time interpretation is signaled by the aspectual contrast. I would like to thank Peter Arkadiev for bringing to my attention the necessity of making this qualification. as argued in Kochańska (2002, 359) -"viewing an event from a zero distance in the temporal dimension results in the exclusion of its endpoints from (…) the region of [the conceptualizer's -A. K.] maximal conceptual acuity". However, as illustrated by (1) above, which -for the reader's convenience -is repeated as (4) below, non-past indicative imperfectives are not only employed to designate present-time occurrences, but also have uses which make reference to future processes, just like their perfective counterparts: (4a) above seems to evoke a conceptualization of two actual future events in a conditional relation, in which the potential future occurrence of the event of solving some case is viewed as a precondition for the occurrence of the event of going into retirement. Since the future occurrence of the event of going into retirement is specifically portrayed as dependent upon the prior occurrence of the event of solving the case, which itself is viewed as far from guaranteed, the event of going into retirement, which is profiled by the perfective clause, is conceptualized as a potentiality which may, but does not have to turn into reality. It is thus construed as an event whose occurrence is epistemically uncertain.
In turn, in (4b), in which the imperfective is used, the future event of going into retirement is conceptualized as an occurrence which is a matter of "how things are supposed to work" -for instance, a matter of the legal regulations in the relevant domain. This understanding of "how things are supposed to work" provides the conceptualizer with a measure of epistemic control over the relevant future event, which -although located in the future -is nevertheless conceptualized as almost bound to occur, a near certainty, given how the world is supposed to work.
The claim made in Kochańska (2002, 362) is that the contrast between the perfective and the imperfective in examples analogous to (4a) and (b) above should be characterized in the following way: while the perfective profiles simply a future actual event, the conceptualization evoked by the imperfective crucially involves the idea of some mental script for future occurrences, which specifies how these occurrences are going to unfold. 3 The relevant mental script may be a plan for future actions, a schedule, some idea of how things generally tend to work in the world, etc. Once established, such a script may be accessed by the conceptualizer at any time, offering her a measure of epistemic control over the occurrences involved. The imperfective in cases analogous to (4b) is thus claimed to involve a conceptualization in which the relevant mental script is accessed at the speech time and a particular record within this script is selected for profiling.
This use of the non-past indicative imperfective is quite similar to its basic use, exemplified in (3b) above -also in this case, what is profiled is a process coextensive with the speech event. The only difference is that while in the prototypical case the profiled process is located at the actual plane, in the case under analysis the profiled event is virtual in nature and resides in reading off, at the speech time, an appropriate record from the relevant mental script. Hence, also in cases analogous to (4a) and (4b) the contrast between the perfective and the imperfective may be characterized in terms of non-immediacy vs. immediacy, which this time is purely epistemic, rather than temporal: while the event profiled by the perfective is conceptualized as epistemically uncertain and, thus, non-immediate, the event profiled by the imperfective is construed as epistemically immediate in the sense that the conceptualizer may exercise a measure of epistemic control over it in virtue of the existence of some mental script predetermining to some extent the future occurrence of the event in question.
The proposed analysis does not preclude the possibility of the imperfective occurring in conditional sentences, provided that the conceived scene involves the idea of a mental script which comes into force, once certain preconditions are realized. A context of this kind is exemplified in (5)  In (5), the speaker, who is in a position of considerable power dominance with respect to the hearer, presents the hearer with a particular script for the future evolution of reality, which is going to be realized once the hearer behaves in the way the speaker wants him to behave.
Of course, the imperfective in examples comparable to (4b) or (5) may always be replaced with its perfective counterpart. The claim made in the present analysis is that this will result in a different construal of the conceived scene, with the profile shifted from a record within some mental script to an actual future occurrence corresponding to that record. On the other hand, a prediction of the proposed analysis is that there are contexts in which the perfective is not easily replaceable with its imperfective counterpart. The contexts in question are contexts in which -for one reason or another -there is no mental script for future occurrences that the speaker could consult at the time of the utterance. An example of this kind is offered in (6)  'I will also go! Ok? I will go! -Well… -Granddad replied without conviction.' In (6), the speaker, who is in a position of subordination with respect to the hearer's authority, wants to get the hearer's permission for the proposed course of action. In this case, it may be assumed that at the time of speaking no mental script with respect to the relevant future event is available, because establishing such a script is dependent on the hearer's acceptance, which has not, as yet, been granted. In consequence, the use of the imperfective in contexts analogous to the one in (6) does not seem very natural.
4 On the grounding role of perfectives and imperfectives in the imperative construction After the above analysis of the grounding import of perfectives and imperfectives in the non-past indicative construction let us now, in this section, turn to an analysis of how the perfective and the imperfective aspect may contribute to clausal grounding in the imperative construction (see also the discussion in Kochańska 2015, 69-75 and2018, 5-8). Since the imperative construction is specifically designed for directive uses, the following analysis will be primarily concerned with the question of how the perfective/imperfective contrast contributes to grounding at the effective level.
As argued elsewhere (cf. Kochańska 2015, 69-70;2018, 5), the imperative itself is a grounding predication: it profiles the process designated by a verb stem and grounds this process as figuring within a directive scenario in the role of the action that the speaker wishes the hearer to engage in. A directive scenario involves, further, the idea of some psychological and/or social force that the speaker invokes in a directive speech act as a means of "pushing" the hearer into compliance. The imperative itself does not specify the nature or degree of this force -this remains implicit and subjectively construed. In different contexts, forces of different natures and degrees may be invoked in a directive speech act, resulting in different kinds of effective grounding. This goes together with grounding at the epistemic level: the profiled process, which is conceptualized as not yet effected, is also understood as not yet real. Depending on the particular kind and degree of the psychological and/or social force invoked by the speaker in a directive speech act, the future occurrence of the profiled process is conceptualized as characterized by differing degrees of epistemic certainty. The claim made in the following part of this section will be that, just like in the case of the non-past indicative construction, the aspectual contrast between the perfective and the imperfective in the imperative construction may also have to do with two different ways of grounding the profiled process: the perfective grounds it as non-immediate, while the imperfective -as immediate to the conceptualizer. The only difference is that in the case of the imperative the profiled process is grounded primarily at the effective level, while epistemic grounding has secondary status.
To substantiate this claim, let us first consider two cases in which the speaker invites the hearer to perform some action which is understood as beneficial to the hearer (cf. Kochańska 2018, 5-6): (7) Usiądź 2 sg.perf. imp. -elf, nie odrywając wzroku od malowidła, ruchem pędzla wskazał Geraltowi głaz. (PWN) 'Have a seat' -the elf, without turning his gaze from the painting, showed the stone to Geralt through a move of his brush.' Apparently, the effect of employing the perfective imperative in (7) is a neutral invitation or even an invitation which may be interpreted as extended with certain reservations, as the speaker's attention is not turned whole-heartedly to his interlocutor.
The situation is different in the case of (8)  The speaker's selection of the imperfective imperative in (8) seems to convey the meanings of the speaker's enthusiasm, emotional warmth, encouragement, or friendly emotional involvement. The warm hand shake accompanying the utterance, as well as the iconic repetition of the imperfective imperative verb both convey the idea of the speaker's strong desire that the hearer follow the invitation.
The claim made in Kochańska (2018, 7-8) is that when the speaker decides to use the perfective imperative, this conveys the idea that the psychological and/ or social force invoked in the speech act to "push" the hearer towards performing the profiled process is just the standard or default force applicable in the relevant circumstances. In turn, the use of the imperfective imperative is a signal that the speaker invokes some extra psychological and/or social force with the aim of "pushing" the hearer more strongly towards the performance of the relevant action. This translates into two different ways of grounding the profiled process.
Specifically, the perfective imperative grounds the profiled process as non-immediate to the conceptualizer, both at the epistemic and at the effective level. This kind of grounding is the default grounding for a process figuring within a directive scenario in the role of the process the speaker wants the hearer to perform, since a future occurrence of this process is a matter of the hearer's decision whether or not to comply with the directive. The hearer is a free agent of his own actions and the speaker may only try to influence his future behavior via invoking in the directive speech act a particular kind of psychological and/or social force that is supposed to "push" the hearer towards the desired behavior. Thus, perfective imperatives ground the profiled process as non-immediate to the conceptualizer in the sense that this process is straightforwardly construed as located outside the speaker's dominion of epistemic and effective control, that is, as a future process whose occurrence cannot be epistemically certain and effectively guaranteed.
On the other hand, the effect of invoking some extra psychological and/ or social force via employing the imperfective imperative is that the profiled process is grounded as immediate to the conceptualizer: although it is still outside the speaker's dominion of epistemic and effective control (by the very nature of a process figuring within the directive scenario), the overall conceived scene involves also the idea of a high degree of epistemic and effective tension and striving aimed at gaining full epistemic and effective control over the profiled action (cf. Langacker 2002 for a discussion of the concept of the control cycle). At the effective level, as mentioned above, the use of the imperfective imperative conveys the idea that the speaker invokes in her directive speech act some extra psychological and/or social force, thereby nearly guaranteeing the hearer's subsequent compliance with her wish and in this way gaining the highest degree of effective control over the profiled occurrence that may be achieved in a directive speech act. At the epistemic level, the relevant tension and striving amount to conceiving of the profiled process as a matter of the most immediate and/or certain future, which is just about to turn into the present reality.
As illustrated by the examples in (7) and (8) above, in the case of invitations the use of the imperfective imperative may be motivated by the fact the speaker invokes in her utterance the extra force of her whole-hearted warm emotional involvement and encouragement, which are supposed to counteract the hearer's potential reservations and thereby nearly guarantee his future compliance with the relevant directive. Other kinds of extra forces may be involved in imperative utterances produced in different directive speech acts.
For instance, another characteristic context for the use of the imperfective imperative are very categorical order-like utterances in the case of which the speaker is actually ready to back up the directive with using physical violence. An example is offered in (9) below: (9) Oddawaj 2 sg. imperf. imp. mi dziecko -krzyknął agent, grożąc Darlympe'owi zabezpieczonym karabinem. (PWN) 'Give the child back to me -shouted the agent, threatening Darlympe with his locked rifle.' In contexts of this kind, the extra force invoked by the speaker to "push" the hearer more strongly towards compliance is the psychological force following from the threat of resorting to physical violence. As a result, utterances comparable to those in (9) are aggressive power show-offs and -not infrequentlyintroductions to actual acts of physical violence.
Imperfective imperatives may also be used in order-or request-like directives in contexts in which the speaker has no chances of resorting to physical violence as a means of ensuring the hearer's compliance. An example of this kind is (2) above, which -for the reader's convenience -is repeated as (10)  The context in (10) is such that the speaker is a teenage girl and the hearer -a teenage boy. Hence, the speaker is physically weaker than the hearer and has no socially sanctioned power over the hearer. In the exchange in (10), the speaker starts with the perfective variant of the imperative. When the hearer refuses to comply with the speaker's demand, she repeats it two more times, getting progressively more and more impatient, indignant, and exasperated -and both utterances involve the imperfective variant of the imperative. It may be noted at this point that in uses comparable to (10) the selection of the imperfective, rather than the perfective, apparently needs to be sanctioned by the speaker's belief that she has a right to make the relevant demand on the hearer and that the hearer has a moral obligation of some kind to follow that demand -as is the case, for instance, in a situation, when the speaker makes a demand that the hearer return to her what is her rightful possession. 4 In (10), this is reflected in the speaker's threat of appealing to her interlocutor's mother -apparently, the speaker's firm conviction is that the mother will support her demand as a demand that is morally justified. The claim, thus, is that in uses similar to (10), the extra force invoked by the speaker is an appeal to certain moral principles which should be at work in the relevant context -in her helpless exasperation and indignation, invoking the idea of what should be the hearer's moral obligation is the speaker's last resort attempt at "pushing" the hearer into compliance.
The semantic nature of the perfective/imperfective contrast in the imperative in Slavic languages has recently been the topic of a number of studies which are directly relevant to the present considerations, in particular Benacchio (2002;2010;2013; see also the discussion in Kochańska 2018, 8), Janda (2004 see also the discussion in Kochańska 2015, fn. 10 on p.75), and Dickey (2020). The analysis offered in this paper appears to be consonant with certain insights made in each of these works. At the same time, the present analysis accounts for uses in which the selection of the imperfective results in a harsher, rather than a gentler directive (cf. example (9) above), which are not covered by the analysis offered in Janda (2004). It also copes with uses such as (10), which seem problematic for both Benacchio (2002;2010;2013) and for Dickey (2020). They are problematic for Benacchio's account, as they do not convey the meaning of rudeness, but rather the meaning of the speaker's exasperation and rightful indignation. They are also problematic for Dickey's analysis, as in cases such as (10) it is quite obvious that the hearer has not already made the decision to carry out the profiled action nor is the speaker in a position to suspend the hearer's decision making role in this respect.

Concluding remarks
The present paper has been concerned with investigating the role that the perfective and the imperfective aspect supposedly play in the system of clausal grounding in Polish. The first area of Polish clausal grammar under investigation, which has been considered in section 3, is the non-past indicative construction. Here, the grounding role of the aspectual contrast appears quite obvious, at least in prototypical cases, as it is the combination of the non-past tense with the imperfective aspect which locates the profiled event in the present, while the combination of the non-past tense with the perfective places the profiled process in the future. This basic grounding function has to do with locating the profiled processes in time: the contrast between the perfective and the imperfective is the contrast between, respectively, temporal non-immediacy (futurity) and temporal immediacy (presentness).
Non-past indicative imperfectives may also be used -just like non-past indicative perfectives -with the future-time meaning. In uses of this kind, grounding effected by selecting either the perfective or the imperfective is no longer a matter of temporal non-immediacy vs. immediacy. Instead, it takes a more abstract shape: the perfective grounds the profiled process as epistemically non-immediate, that is, as located beyond the conceptualizer's epistemic control (which is the default grounding for a future process), while the imperfective effects the grounding of the profiled occurrence as epistemically immediate, that is, as an occurrence over which the conceptualizer may exercise a measure of epistemic control, by virtue of the existence of some mental script specifying how reality is going to evolve in the relevant respects.
In section 4 the role of the perfective/imperfective contrast in the imperative construction has been analyzed. It has, again, been argued, that while the perfective imperative brings about the default construal of the profiled event as epistemically uncertain and effectively not guaranteed (that is, epistemically and effectively non-immediate), the imperfective conveys the idea of a stronger degree of striving, on the part of the conceptualizer, to achieve epistemic and effective control over the profiled event. In other words, it conveys the idea of the profiled event which is epistemically and effectively as immediate as a future event figuring within a directive scenario may only be. The specific idea advocated in the course of the analysis has been that the selection of the imperfective variant of the imperative signals the speaker's invocation of some extra psychological and/or social force which is supposed to "push" the hearer more strongly towards compliance with the issued directive. This force may have very different characters in different contexts: it may be the force of the speaker's whole-hearted encouragement and concern with the hearer's well-being, the force of the speaker's exasperation brought about by the hearer's preceding non-compliance with the speaker's rightful demand, or the psychological force associated with the speaker's readiness to employ physical violence to enforce compliance, etc. In any case, the idea has been that the invocation of the relevant extra psychological and/or social force is supposed to result in nearly bringing the profiled process under the conceptualizer's epistemic and effective control, thereby making it epistemically and effectively immediate.
The proposed analysis seems to work quite well at least for the data under consideration, both in the case of the non-past indicative and the imperative construction. In fact, in the case of the imperative it seems to be more comprehensive than some alternative accounts that have been put forward recently. Apparently, the analysis accounts for certain subtle differences in the distribution of the two aspectual variants in different contexts, as well as for equally subtle pragmatic effects that these two variants may produce. Its further advantage is that the respective grounding meanings proposed for the perfective and the imperfective in the non-past indicative and in the imperative constructions may be viewed as related not only to each other, but also, in each case, to the prototypical meaning of the relevant aspectual variant. Hence, the proposed account does not seem to have an ad hoc, but a motivated character.