Aspect Architecture in Bulgarian: Morphology and Semantics

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Review of "Aspect Architecture in Bulgarian: Morphology and Semantics"

This paper provides a thorough and insightful analysis of the aspectual system in Modern Bulgarian, focusing on viewpoint aspect (perfective vs. imperfective), temporal aspect (aorist vs. imperfect), and perfect aspect. The study synthesizes existing theoretical accounts within a compositional interval-relational framework, offering a holistic perspective on Bulgarian aspect.

The paper engages with an extensive range of literature and theoretical perspectives, providing a well-rounded discussion. Rather than introducing an entirely new theoretical model, the paper integrates existing proposals in a way that clarifies the intricate relations between aspectual categories. This is particularly valuable given the complexity of Bulgarian 

While the paper draws from many frameworks, some sections (e.g., the discussion of interval structures and assertion time) could benefit from clearer exposition, particularly for readers unfamiliar with Klein’s (1994) interval-relational approach. More diagrams or illustrative examples could help in this regard.

There are also some things in the manuscript which should be improved. I list them below: 

Line 98: perfective may also encode inception with inceptive prefixes or boundedness with delimitative prefixes - not only culmination

Line 103: perfective verbs are not necessarily telic for example pofective verbs are perfective but atelic 

Line 202: the perfect locates AT in the post-time of ST - not necessarily - it is true only for current relevance / resultative perfect but not necessarily for existential/experiential perfect and universal perfect. 

Lines 312-318: Please, refer and discuss this study briefly: Two types of secondary imperfectives Evidence from Polish and Bulgarian, Dorota Klimek-Jankowska, Vesela Simeonova: https://langsci-press.org/catalog/book/481

Lines 473-474: Please, add that this is a very tentative hypothesis. In order to say that aorist is like English simple past one would have to carry on a very well-desgined comparative study ideally with the use of parallel corpora. Unlike simple past aorist has some evidential semantics plus the author would have to check if in all the contexts in which simple past is used in English aorist can be used in Bulgarian. Without such empirical support this claim remains to be stipulative and it should be honestly stated in the manuscript.

Importantly, the author uses the term relative past while talking about the l-participle in the Bulgarian perfect. I cannot see the reference to Gronn and Perfect (2017) who clearly formalize the term relative past and they should be properly referenced https://www.hf.uio.no/ilos/personer/vit/atleg/perfect_gronn_stechow_oct2017.pdf

Conclusion:

This is a valuable contribution to the study of Bulgarian aspect. It successfully synthesizes multiple perspectives into a unified model. It can be published after my above concerns are carefully addressed. 

Author Response

Comment 1: "While the paper draws from many frameworks, some sections (e.g., the discussion of interval structures and assertion time) could benefit from clearer exposition, particularly for readers unfamiliar with Klein’s (1994) interval-relational approach. More diagrams or illustrative examples could help in this regard."
Response 1: I added some introductory sentences in the beginning of §2.1, below ex. (2) as well as an additional explanation (in brackets) in footnote 3.

Comment 2: "Line 98: perfective may also encode inception with inceptive prefixes or boundedness with delimitative prefixes - not only culmination."
Response 2: I added a paragraph (lines 94-102) and the schemata in (2) to account for inceptives and delimitatives, arguing that they contain one or two situational boundaries which can, in parallel to culminating situations, become included within AT.

Comment 3: "Line 103: perfective verbs are not necessarily telic for example pofective verbs are perfective but atelic"
Response 3: I deleted the word "only" to allow for the option of atelic perfecitive verbs.

Comment 4: "Line 202: the perfect locates AT in the post-time of ST - not necessarily - it is true only for current relevance / resultative perfect but not necessarily for existential/experiential perfect and universal perfect."
Response 4: I added footnote 12 which refers to Iatridou et al. (2001)'s claim of an overall Perfect Time Span which does not necessarily include a boundary between the two intervals.

Comment 5: "Lines 312-318: Please, refer and discuss this study briefly: Two types of secondary imperfectives Evidence from Polish and Bulgarian, Dorota Klimek-Jankowska, Vesela Simeonova: https://langsci-press.org/catalog/book/481"
Response 5: I added footnote 22.

Comment 6: "Lines 473-474: Please, add that this is a very tentative hypothesis. In order to say that aorist is like English simple past one would have to carry on a very well-desgined comparative study ideally with the use of parallel corpora. Unlike simple past aorist has some evidential semantics plus the author would have to check if in all the contexts in which simple past is used in English aorist can be used in Bulgarian. Without such empirical support this claim remains to be stipulative and it should be honestly stated in the manuscript."
Response 6: The reviewer is utterly right in their criticism. I deleted the relevant statement altogether.

Comment 7: "Importantly, the author uses the term relative past while talking about the l-participle in the Bulgarian perfect. I cannot see the reference to Gronn and Perfect (2017) who clearly formalize the term relative past and they should be properly referenced https://www.hf.uio.no/ilos/personer/vit/atleg/perfect_gronn_stechow_oct2017.pdf"
Response 7: I added the reference just above ex. (17) (the PTS semantics).

I wish to express my deep gratitude for the reviewer's attentive reading of my paper and their truly insightful and helpful comments. The paper has substantially benefited from them, and I am happy to have received both constructive criticism and support from an obvious expert.

Reviewer 2 Report

Comments and Suggestions for Authors

Some of the controversial statements concerning orist and evidentials in Bulgarian need to be substantiated and qualified.

Review of the article Aspect architecture in Bulgarian: Morphology and semantics

The paper discusses a long-lasting debate in contemporary Bulgarian – aspectual interpretation and its morphological marking. In accounting for the nature of aspect in Bulgarian and its marking, the paper offers a commentary on related issues such as the tense system in the language, the existence or not of renarrative/evidential moods or verb phrase, the place of the aspectual marker in the template of a verbal form, the conjugation classes in the language, etc. The paper operates with an interval and schemata-based understanding of time, with the latter capturing basically a distinction between telicity and atelicity (grouping together Vendler’s achievements and accomplishments as opposed to states and activities).

As admitted by the author(s) the merit of the paper is not novelty per se but a novel combination of theoretical constructs and interpretative frameworks, the result of which is a coherent and internally consistent picture of the temporal and aspectual systems in Bulgarian and their exponence (both in terms of morphological marking and in periphrastic expressions).

The only unsubstantiated claim is the contention that there are no evidential moods in Bulgarian. Admittedly, such a view is a matter of theoretical beliefs and epistemic commitments, and as such cannot be criticized, but is so dismissively mentioned in the paper that a bit of elaboration and substantiation might be in order.

The text is well structured marked for outstanding clarity and precision.

The chosen topic (and its treatment) fits ideally the scope of the journal.

The article ranks high on novelty on a number of counts: 1) excellent contextualization of the research in debates and extant interpretations of the temporal and aspectual systems in Bulgarian; 3) internal consistency and compatibly amalgamated existing analyses; 4) readability; 5) style and expression.

The statements made and the conclusions drawn are coherent and supported by the provided evidence and the presented arguments.

Two weaknesses:

• None of the cited references is recent publications (within the last 5 years);
• Some bold statements which are not directly substantiated: a) lack of evidentials in Bulgarian and b) aorist is “a delusional entity”. Neither is the modifier appropriate, nor is the claim convincing outside the idiosyncratic amalgamation of interpretations offered here.

A minor recommendation is the qualification of some of the statements and clear recognition that some of these are valid within the combination of various views, not all of which are necessarily compatible in their totality. Only the fact that certain portions of interpretative frameworks have been cherry-picked and combined can make these statements comprehensible, not necessarily valid.

1. 96 “perfective aspect marks the overlap of the AT-interval with the both the preparatory phase”

This abbreviation has to be explicated here, not below.

1. 264 “is empty by default.” (Manova 2007, 24) Thus,” Sould be “is empty by default” (Manova 2007, 24). Thus,”
2. 270 “while the rest is periphrastic (P)” – are
3. 647 “temporal tspect.” - aspect

 

Author Response

Comment 1: "None of the cited references is recent publications (within the last 5 years);"
Response 1: I added several publications starting with 2020 such that now we have a total of 10 recent publications.

Comment 2: "Some bold statements which are not directly substantiated: a) lack of evidentials in Bulgarian"
Response 2: I clearly see the reviewer's point and admit that my formulations were too imperative. I reformulated several passages so as to reduce this tone. Thus, in footnote 35, I removed the statement that it is mostly traditional grammars that claim that special evidential moods exist. Instead, I confine to list some authors who do not and state that I share their perspective. Additionally, I indicate in footnote 38 - where I say that there are more l-periphrases which some take to be special evidental mood forms - that I propose one possible and general source of the relevant interpretations. Moreover, below ex. (53), I did some slight reformulations to be less imperative, saying that my proposal has the mere potential to explain the ambiguity in temporal reference in non-confirmative (or neutral) l-periphrases with imperfect participles.

Comment 3: "Some bold statements which are not directly substantiated: b) aorist is “a delusional entity”. Neither is the modifier appropriate, nor is the claim convincing outside the idiosyncratic amalgamation of interpretations offered here."
Response 3: The reviewer is utterly right, and I am grateful for their criticism. I changed the sentence as follows: "Actually, the aorist lives up to its name (Ancient Greek \textit{aóristos} `undefined') in that it represents the default interpretation of past situations (finished, i.e., closed AT interval) and lacks additional aspectual marking."

Comment 4: "A minor recommendation is the qualification of some of the statements and clear recognition that some of these are valid within the combination of various views, not all of which are necessarily compatible in their totality. Only the fact that certain portions of interpretative frameworks have been cherry-picked and combined can make these statements comprehensible, not necessarily valid."
Response 4: The reviewer is completely right. Following a comment by another reviewer, I reformulated, among other things, the paragraph below the literature list in §1 to sound more humble, which I see is indeed necessary and advisable. Thanks to the reviewer for this critique!

Comment 5: "line 96 “perfective aspect marks the overlap of the AT-interval with the both the preparatory phase”. This abbreviation has to be explicated here, not below."
Response 5: I reformulated the relevant sentence so as to avoid mentioning AT already at this place. The definition is left where it was, but now it occurs at the first place of mention of AT. Additionally, I paraphrase "assertion time" in footnote 3 which precedes its definition in the main text.

Comment 6: "264 “is empty by default.” (Manova 2007, 24) Thus,” Sould be “is empty by default” (Manova 2007, 24). Thus,”
Response 6: I corrected the typo.

Comment 7: "270 “while the rest is periphrastic (P)” – are"
Response 7: I corrected the mistake.

Comment 8: "647 “temporal tspect.” - aspect"
Response 8: I corrected the typo.

I wish to express my deep gratitude for the reviewer's attentive reading of my paper and their truly insightful and helpful comments. The paper has substantially benefited from them, and I am happy to have received both constructive criticism and support from an obvious expert.

Reviewer 3 Report

Comments and Suggestions for Authors

The paper analyzes aspect morphology and semantics in Bulgarian considering three types of aspectual categories: grammatical (viewpoint) aspect, temporal aspect, and perfect aspect. To describe the aspectual system in Bulgarian, a schematization with bullets is used which combines Vendler’s classification of predicates (the so-called lexical or situation aspect) and the opposition between perfective and imperfective aspect. To account for the Bulgarian data, an additional aspectual category is introduced, namely the temporal aspect, following Sonnenhauser’s approach to analyze the imperfect and the aorist in Bulgarian. The perfect is viewed as a third truly aspectual category whose analysis in the paper is based on Iatrudou’s approach. To analyze the morphological structure of Bulgarian verb forms, the paper applies Manova’s model by dividing the word into morphological slots which may host different types of affixes with some modifications: PREF– BASE– DSUFF– ASUFF– TM– ISUFF (Manova 2007) > PREF– BASE– ASP– TNS– AGR (the derivational suffix is not considered as in verbal morphology it is usually an empty slot, the thematic marker is omitted, and the inflectional suffix is replaced by two slots – tense and agreement). Indeed, the aorist and imperfect inflectional suffixes have two components indicating tense and agreement, but this holds only for those two tenses.

I believe that in general the paper has an excellent structure, good argumentation and contributes to a better understanding of the Bulgarian aspectual system. Still, I have some observations that the author may decide to consider. I will present my remarks in the order the respective phenomena appear in the text.

1. 1, 6 “…and demonstrates how an interval-relational analysis of the perfect can solve some puzzles associated with the so-called evidential moods that many grammars and analyses claim to exist in Bulgarian.”

To solve the puzzles of Bulgarian evidentiality, the author should consider its temporal paradigm as a whole. Its is available in Nicolova 2017.

1. 1, 17 pìsàhme vs. napìsàhme

There is a big debate in Bulgarian linguistics whether basic imperfectives and prefixed perfectives form an aspectual “pair”. Opinions differ, but the prevailing one is that an aspectual pair should exclude the basic imperfective and should contain the prefixed perfective and its suffixed counterpart (secondary imperfective), i.e. napisha – napisvam. Maslov (1984) proposes aspectual “triads”: basic imperfective – prefixed perfective – secondary imperfective.

1. 1, 17 napìsàhme

The prefixed verbs do not move their stress, the only possible form is napìsahme (napisàhme occurs in some western dialects but is not accepted in the standard pronunciation).

1. 2, 40 “Needless to say, all these analyses are controversial. More crucially, they focus on peculiar grammatical phenomena and use different theoretical frameworks, which makes it difficult to evaluate their mutual compatibility in a possible overall picture. With the present paper, I intend to make a step towards the latter.”

In fact, all papers or books quoted here pretend to present the overall picture, and many others too. On the other hand, the present paper does not cover the future and the future perfect tenses, so an overall picture is difficult to be achieved only considering the past tenses. I think the author should reformulate this statement somehow by making it less pretentious.

1. 6 Perfect aspect

This is a general consideration I have regarding the analysis of the perfect aspect in Bulgarian. I think the author rightly opposes Klein’s point of view for the perfect as a third aspect based on English and German data. Indeed, Bulgarian data do not support such analysis as the participle of the full verb is already marked for perfective or imperfective aspect. In my opinion, this makes the perfect aspect similar to the temporal aspect analyzed earlier in the paper.

1. 12, 360

The verb pijna ‘sip’ is perfective and derives from the verb pija ‘drink’ by adding the perfectivizing suffix -n-, so its morphological structure is pij-n-a. For the analysis here it can be replaced by pija ‘drink’ (also an e-conjugation verb) which is the basic imperfective.

1. 17 Perfect(oid) periphrases

In historical perspective, the imperfect participle is believed to have emerged in connection with the category of evidentiality in Bulgarian and has been adapted to express evidential present and imperfect. This claim is also supported by the fact that Bulgarian is the only Slavic language that has both grammaticalized evidentiality and imperfect participle. The perfect is formed only with the aorist participle, whereas all the perfectoid forms containing the imperfect participle have evidential reading. However, some examples of perfect forms containing imperfect participles may be found. In my opinion, the reason is the general trend towards mixing of the imperfect and the aorist in contemporary Bulgarian (there are many experiments described in linguistic literature about this phenomenon, for example Fetvadzhieva 2018 discusses some mistakes in the recent use of the aorist and the imperfect, it is available here: https://journals.uni-vt.bg/poc/bul/vol11/iss1/art20). One of the effects of this mixing is the attempt to “reinforce” the imperfective semantics by adding the imperfect affix to the imperfective stem in the participle, i.e. using the imperfect participle instead of the aorist one.

1. 18, 520 “…vinagi se e govorel balgarski ezik”

Hauge’s sentence is a typical example of wrongly used imperfect participle (instead of the aorist one). In fact, this is a very common mistake when the only formal difference between the aorist and the imperfect is the thematic vowel, specifically -i- vs. -e-.

1. 20, 502 “Ivàn e strojàl pjàsacna kùla.”

The sentence has strictly evidential (inferential imperfect) interpretation ‘(I infer that) Ivan was building a sandcastle.’

1. 20, 614

Sentence (53) is ungrammatical. The only way to make it acceptable is to omit the auxiliary and thus make the verb form evidential (reportative imperfect), for example: Kogato Ivan postrojal pjasachna kula, vjatarat ja butal. ‘Whenever Ivan built a sandcastle, the wind destroyed it.’

1. 21, 617 “Many authors observe that this combination occurs almost exclusively in adverbial clauses being part of complex sentences whose main-clause verb is an imperfective imperfect, and that its interpretation is iterative or habitual (a.o., Comrie 1976; Gvozdanovi´c 2012; Maslov 1959; Pašov 1999; Rivero et al. 2017). In other words, perfective imperfects denote, multiple instances of the same situation.”

This observation is indeed correct but if it should be applied on perfect(oid) forms containing an imperfect participle, this is only possible in evidential contexts (reportative or inferential). I think this must be clearly stated in the paper, because in the current version it is only hinted in the conclusion.

To sum up, I think the paper offers a very good and original approach to the Bulgarian (past) aspectual system and it will gain if the author reconsiders some parts of the analysis of the perfect.

Author Response

Comment 1: "1, 6 “…and demonstrates how an interval-relational analysis of the perfect can solve some puzzles associated with the so-called evidential moods that many grammars and analyses claim to exist in Bulgarian.” - To solve the puzzles of Bulgarian evidentiality, the author should consider its temporal paradigm as a whole. Its is available in Nicolova 2017"
Response 1: I agree with the reviewer that a comprehensive analysis of BG evidentiality must address its complete (temporal) paradigm. However, my paper merely touches the issue, offering one possible way to explain why evidentially interpreted periphrases with imperfect l-participles are ambiguous between present and past temporal reference. In following another reviewer's criticism and advice, I reformulated several passages so as to reduce the hitherto over-imperative tone as concerns the existence of special evidential moods in BG. For instance, I removed in footnote 35 and in the last sentence of the abstract the statement that it is only traditional grammars that claim that special evidential moods exist. Instead, I confine myself in footnote 35 to list some authors who do not make that claim and add that I share that very view. Additionally, I indicate in footnote 38 - where I say that there are more l-periphrases which some take to be special evidental mood forms - that I propose one possible and general source of the relevant interpretations. Finally, below ex. (53), I did some slight reformulations to sound less imperative, stating that my proposal has the mere potential to explain the ambiguity in temporal reference in non-confirmative (or neutral) l-periphrases with imperfect participles.

Comment 2: "1, 17 pìsàhme vs. napìsàhme - There is a big debate in Bulgarian linguistics whether basic imperfectives and prefixed perfectives form an aspectual “pair”. Opinions differ, but the prevailing one is that an aspectual pair should exclude the basic imperfective and should contain the prefixed perfective and its suffixed counterpart (secondary imperfective), i.e. napisha – napisvam. Maslov (1984) proposes aspectual “triads”: basic imperfective – prefixed perfective – secondary imperfective."
Response 2: I added footnote 21.

Comment 3: "napìsàhme - The prefixed verbs do not move their stress, the only possible form is napìsahme (napisàhme occurs in some western dialects but is not accepted in the standard pronunciation)."
Response 3: I added a remark in footnote 1. I would like to keep the stress alternative also in prefixed forms since I believe that speakers (even if only of some dialects) use it for disambiguation if need be. I think that the amendment in footnote 1 avoids readers from overgeneralizing ste stress shift option.

Comment 4: "2, 40 “Needless to say, all these analyses are controversial. More crucially, they focus on peculiar grammatical phenomena and use different theoretical frameworks, which makes it difficult to evaluate their mutual compatibility in a possible overall picture. With the present paper, I intend to make a step towards the latter.” - In fact, all papers or books quoted here pretend to present the overall picture, and many others too. On the other hand, the present paper does not cover the future and the future perfect tenses, so an overall picture is difficult to be achieved only considering the past tenses. I think the author should reformulate this statement somehow by making it less pretentious."
Response 4: I see the reviewer's point and reformulated my statement as follows: "Needless to say, all these analyses are at least partially controversial. Moreover, most of the formal analyses focus on peculiar grammatical phenomena using different theoretical frameworks, which makes it difficult to evaluate their mutual compatibility in a possible overall picture. In the present paper, I combine selected formal analyses that address specific portions of the BG aspecto-temporal system. It goes without saying that the results of my proposal can only be one step towards a possible overall picture of the BG aspect and tense system."

Comment 5: "Perfect aspect - This is a general consideration I have regarding the analysis of the perfect aspect in Bulgarian. I thik the author rightly opposes Klein’s point of view for the perfect as a third aspect based on English and German data. Indeed, Bulgarian data do not support such analysis as the participle of the full verb is already marked for perfective or imperfective aspect. In my opinion, this makes the perfect aspect similar to the temporal aspect analyzed earlier in the paper."
Response 5: I am not sure if this comment requires changes in the paper. Rather, it seems to me that it supports my statements concerning the necessity to modify Klein's theory when speaking about BG. However, I do not see that Perfect Aspect (the way I define it in my paper) is similar to what I call temporal aspect. Admittedly, it builds on viewpoint aspect just like the temporal aspect does. It does not, however, concern the quality of the AT interval(s) as such - as do aorist and imperfect - but rather adds a second AT interval. Actually, I would say that the BG perfect is a second-order aspect. And if you asked me what sort of second-order aspect. But indeed, it's hard to decide if this second-order aspect is more like viewpoint or temporal. I'd say, it can be both, but in any instance it is second-order viewpoint (but simply because viewpoint is ubiquitous in BG, whereas temporal is restricted to the past domain).

Comment 6: "12, 360 - The verb pijna ‘sip’ is perfective and derives from the verb pija ‘drink’ by adding the perfectivizing suffix -n-, so its morphological structure is pij-n-a. For the analysis here it can be replaced by pija ‘drink’ (also an e-conjugation verb) which is the basic imperfective"
Response 6: I added footnote 24 where I explain that pijna is a derivation based on pija with -n- occupying the DSUFF-slot in Manova's (2007) full template, and that I treat this stem, which is really a derived stem, as a base in the subsequent morphological analysis.

Comment 7: "Perfect(oid) periphrases - In historical perspective, the imperfect participle is believed to have emerged in connection with the category of evidentiality in Bulgarian and has been adapted to express evidential present and imperfect. This claim is also supported by the fact that Bulgarian is the only Slavic language that has both grammaticalized evidentiality and imperfect participle. The perfect is formed only with the aorist participle, whereas all the perfectoid forms containing the imperfect participle have evidential reading. However, some examples of perfect forms containing imperfect participles may be found. In my opinion, the reason is the general trend towards mixing of the imperfect and the aorist in contemporary Bulgarian (there are many experiments described in linguistic literature about this phenomenon, for example Fetvadzhieva 2018 discusses some mistakes in the recent use of the aorist and the imperfect, it is available here: https://journals.uni-vt.bg/poc/bul/vol11/iss1/art20). One of the effects of this mixing is the attempt to “reinforce” the imperfective semantics by adding the imperfect affix to the imperfective stem in the participle, i.e. using the imperfect participle instead of the aorist one."
Response 7: This is a very valuable comment, and I added a rather long passage in footnote 37 (which comments on the "perfect progressive" example in (42)) to facilitate it, including the reference to Fetvadzhieva. Nonetheless, I express my own reservation as I say that it is at least conceivable that the evidential forms may have come to be used thanks to the emergence of the imperfect participle, which in turn may have been due to the persistance of the aorist/imperfect distinction in BG (notwithstanding present-day mixing up, which is, of course, a topic in itself). All in all, I think it's (still) a chicken-or-egg question.

Comment 8: "18, 520 “…vinagi se e govorel balgarski ezik” - Hauge’s sentence is a typical example of wrongly used imperfect participle (instead of the aorist one). In fact, this is a very common mistake when the only formal difference between the aorist and the imperfect is the thematic vowel, specifically -i- vs. -e-."
Response 8: Thanks for this comment. I believe that my amendment to footnote 37 concedes this option of judging the example without directly attacking Hauge's judgment. As to the marking, I am surprised that the difference in the theme vowel seems to be too feeble for some today's speakers of BG. I mean, many linguistic forms differ in only one segment. But maybe the point is that conjugational class III with its general non-marking of imperfect forms gives rise to doing the same in all comjugation (as Fetvadzieva indicates in her conclusion).

Comment 9: "20, 502 “Ivàn e strojàl pjàsacna kùla.” - The sentence has strictly evidential (inferential imperfect) interpretation ‘(I infer that) Ivan was building a sandcastle.’"
Response 9: I added "[I infer]" to the translation of the example. Additionally, I made a amendment in footnote 41 (it noted the 'conclusivity' of the "e strojàl"-periphrasis already before, but now it covers the remaining non-confirmative examples, too.

Comment 10: "20, 614 - Sentence (54) is ungrammatical. The only way to make it acceptable is to omit the auxiliary and thus make the verb form evidential (reportative imperfect), for example: Kogato Ivan postrojal pjasachna kula, vjatarat ja butal. ‘Whenever Ivan built a sandcastle, the wind destroyed it.’"
Response 10: Thanks so much for this correction! I have removed the auxiliary and changed the description in footnote 41 accordingly.

Comment 11: "21, 617 “Many authors observe that this combination occurs almost exclusively in adverbial clauses being part of complex sentences whose main-clause verb is an imperfective imperfect, and that its interpretation is iterative or habitual (a.o., Comrie 1976; Gvozdanovi´c 2012; Maslov 1959; Pašov 1999; Rivero et al. 2017). In other words, perfective imperfects denote, multiple instances of the same situation.” - This observation is indeed correct but if it should be applied on perfect(oid) forms containing an imperfect participle, this is only possible in evidential contexts (reportative or inferential). I think this must be clearly stated in the paper, because in the current version it is only hinted in the conclusion."
Response 11: Footnote 42 said already that "[t]he chain of instances may well continue in and after AT-2, which allows a non-confirmative (evidential) present-tense reading". I now added the following sentence to make the reviewer's point clear: "Irrespective of whether the reading is present- or past-tense, it will always be non-confirmative if the verb is an imperfect participle."

I wish to express my deep gratitude for the reviewer's attentive reading of my paper and their truly insightful and helpful comments. The paper has substantially benefited from them, and I am happy to have received both constructive criticism and support from an obvious expert.