Mind the Gap! Null Modals (and Other Functional Verbs) in Finite Complementation in Italo-Greek
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsReview of ‘Mind the gap! Null modals in finite complementation in Italo-Greek’, an article submitted for publication in the journal Languages.
My recommendation to the editors of Languages is that the article entitled ‘Mind the gap! Null modals in finite complementation in Italo-Greek’ should be accepted subject to revisions which I will classify as minor because they are cosmetic – albeit essential – rather than substantive. Building on Ledgeway’s (forthc.) claim that silent modals are traceable in infinitival indirect interrogatives/free relatives and negative imperatives in Italo-Greek varieties of Southern Italy, the author suggests that another syntactic context is characterized by silent or absent modals in Greko and, more generally, all the Italo-Greek languages, namely clauses introduced by the complementizer na. This hypothesis is corroborated by the fact that the complementizer is irrealis and by comparison with other na clauses exhibiting MUST. The idea is that the spread of finite complementation in these varieties resulted in the reduction of the distribution of those modal verbs that can only occur in restructuring (i.e., monoclausal) contexts, where they are followed by an infinitive.
The author’s proposal seems to me to be falsifiable and worth disseminating. I therefore recommend publication. That said, the article is currently difficult to read and requires quite a bit of cosmetic work. The following revisions are recommended.
First, the reader may not know why Greek varieties are spoken in Southern Italy or indeed that any such varieties are found there. Therefore, the article should begin with a brief introduction to the varieties under examination and their history. A map would enhance the quality of the presentation.
After this introduction, the article should introduce the notion of modality gap, which in the current version is instead only introduced on p. 11. A brief summary of Ledgeway’s (forthc.) claim is in order since this is the starting point of the author’s analysis.
At this point, the author should explain, in the body of the article, which corpora they drew their data from. These are currently described in note 5, but it is essential that the reader should know from the beginning what data the analysis is based upon and how the data were selected from the relevant corpora.
In the course of the analysis, the author should clarify the following key point. Is their claim that the modals are absent altogether from syntactic contexts with na clauses or is it the case that there is a silent modal at some underlying level of syntactic representation? A graphic representation of the syntax of such clauses is needed. This can be given in one or a couple of syntactic trees.
The author is clearly not a native English speaker and, as a consequence, the narrative is somewhat obscure. I have suggested a number of minor editorial changes on the PDF of the article, which I attach to this review. In addition to the changes requested on the PDF, the author should make the minor revisions listed here:
P 2, l. 77: The author should explain what is meant by restructuring, with reference to Rizzi’s original definition of it.
P 2, l. 83: The author should explain here, with reference to the definition of restructuring given above, how a na clause can be part of a restructuring construction, which should by definition be monoclausal. The arboreal representations should probably be given here.
P. 3, ll. 112-113: The author should make explicit reference to examples (7)-(9), flagging the alternation of the infinitive with an irrealis complementizer. Otherwise, this will remain unnoticed and the reader will not see much evidence for the claim about a silent modal.
P. 3, l. 129: Surely, this is not the translation of example (9). Please check and replace. In fact, all the idiomatic translations of the example should be checked again.
P. 4, l. 182: Provide an example of the NEG + Infinitive structure (the negative imperative), which the reader may not be familiar with.
P. 5: As far as I can tell, the synchronic analysis proposed on p. 5 is not in contradiction with the established or traditional diachronic analysis, since instead of disappearing altogether the modal may simply have become silent. Why does the author present the two analyses as competing ones? What am I missing from their explanation? The author should clarify this point.
P. 7, l. 343: The author should check the idiomatic translation of example 20b.
P.8: I can see how 21c can be said to include a motion verb (fly), but the same cannot be said of 21d. Why is this example given here? The text does not explain this point. Please note that traditional treatments of bring up and get do not treat these as motion verbs. I would suggest replacing these examples with others that unequivocally include motion verbs.
P. 9, l. 415: ta ‘them’ presumably refers to ‘money’, which should be glossed as ‘plural’ to clarify the co-reference with the plural pronoun.
P. 10, l. 486: rephrase ‘this confirms the need to’ as ‘this piece of evidence lends further support to our analysis, which involves a silent modal…’
P. 10, l. 501: for the coding of the causee in causatives, the author should quote Comrie (1974), which is the original and most authoritative treatment of this issue.
References
Comrie, B. 1974. Causatives and Universal Grammar. Transactions of the Philological Society 73(1):1-32.
Comments for author File: 
Comments.pdf
I have commented on this in the review and I have suggested edits on the PDF of the article.
Author Response
I have taken on board some (not all) of the suggestions of the reviewer. I had English reviewed by a native speaker.
My answer concerns the following comment: "In the course of the analysis, the author should clarify the following key point. Is their claim that the modals are absent altogether from syntactic contexts with na clauses or is it the case that there is a silent modal at some underlying level of syntactic representation? A graphic representation of the syntax of such clauses is needed. This can be given in one or a couple of syntactic trees". For the moment, I decided to delete the entire paragraph, because a problem arises in my analysis regards to the na-finite clauses. Indeed, following Ledgeway, a silent modal can be reconstructed in monoclausal constructions, such as infinitival relatives or the imperative negative codified by the negator followed by the infinitive. On the contrary, na-clauses headed by a (covert or overt) modal are not considered monoclausal structures, as is shown by the ban of clitic climbing to the main predicate.
Reviewer 2 Report
Comments and Suggestions for AuthorsComments for author File: Comments.pdf
The paper requires heavy editing in English.
Author Response
I thank the reviewer for his/her insightful comments. I had English checked by a native speaker and I rewrote entire parts of the paper. I think that in some points the reviewer could have been disoriented by my inappropriate use of English. However, I continue to find my hypothesis possible (and, honestly, interesting). In a nutshell, I reformulated my hypothesis as follows: the constraint – though anything but categorical – of functional verbs on selecting na-clauses probably represents the last remnant of a diachronic stage in which some predicates (in my corpus, mainly modal and causative verbs) began to leave their previous restructuring contexts. As a result, monoclausality is lost, giving rise to several phenomena all pointing to a a decreased level of dependency between the matrix and embedded clauses. This may have led modal and other functional predicates to acquire a greater lexical status, partially freeing themselves from their otherwise (exclusively) functional role. Should this be the case, their resistance to surfacing in finite structures could represent a step in the degrammaticalization process which leads them to progressively realize more lexical features, and at the same time reduces their ability to act (exclusively) as functional heads. Their (albeit sporadic) absence in selecting finite complements could be a signal of functional instability, in which they still oscillate between a purely functional role and a lexical one.
Lines 446-448: “The replacement of CAN with MUST could follow from the ban on licensing CAN ‒ the true stronghold of the infinitival complementation ‒ as an overt functional auxiliary in finite complementation” – why is there such a ban on CAN?
Nobody knows the answer, that is the adversion of CAN to licence a finite complement. In an unpublished paper, I argued that the spreading of dependent finite clauses instead of the infinitive started from syntactically factual contexts (that is, positive sentences, the unmarked contexts), later by expanding to non factual contexts (that is, negative and interrogative sentences, the marked contexts, where infinitive complements are generally more preserved that in all the other contexts). One may wonder if the ban of the verb ‘can’ for encoding dependent finite clauses can be traced back to the same principle: within the scope of irrealis sentences, ‘can’ — if compared with ‘want’ (the verb which codes subordinate finite clauses more frequently than all other ones) —, can be conceived as provided with a lower degree of factuality, insofar it lacks of some features such as [volitionality], [agentivity] etc. (Bybee et al. 1994). As such, it represents the stronghold of the infinitive: the change started from verbs which express an agent-oriented event (whence a higher degree of potential factuality), spreading into verbs which code just the potentiality, with no role of the subject.
Reviewer 3 Report
Comments and Suggestions for AuthorsThe paper identifies and discusses a set of interesting data from Italo-Greek varieties which are shown to receive some modal interpretation while (arguably) lacking an overt modal marker. At the same time, the data allow the author(s) to speculate on the interaction of the loss of the infinitive in (as attested in many varieties of Greek) and the development of modality in finite structures (mostly involving finite verbs accompanied by na). The discussion refers to earlier literature discussing similar data and provides interesting insights regarding both the syntax and semantics of covert modals in general and the description of the verb in Italo-Greek.
The discussion is straightforward and quite convincing, although the use of English appears to be rather idiosyncratic at points (especially the longer paragraphs on pages 10 and 11 need to be revised to ensure clarity of argumentation). The author(s) might also consider revising some of the transitions between sections (e.g. section 2 ends in a bracketed structure; section 4 begins with “To sum up:”).
The following points might also be addressed in improving the presentation:
· The translation in example (9) does not correspond to the sentence presented and glossed (which should be something like ‘I have nothing to give you’)
· The data presented in examples (23)-(25) all involve modal combinations: there is always a modal element in the matrix clause (specifically, an obligation modal in (23) and (24) and an imperative in (25)) followed by exo na in the subordinate clause in all three cases. The discussion does not seem to take this into account – though of course the presence of modality in the higher clause may be a factor affecting the interpretation of the modal in the lower clause.
· The overall covert modal analysis put forward by the author(s) need not be the only possible account of the data discussed; the “opinion” expressed on p. 11 that na-clauses “constitute headless clauses with a silent modal” is only one of many options. The author(s) may also wish to address the idea of a specifically modal contribution on the part of na. An analysis of na itself as a modality marker would then relate to questions of both the historical source and the directionality of a possible grammaticalization path (along the lines of the discussion in Tsangalidis 2004: 198-200, already mentioned on p. 1 of the manuscript). The synchronic status of na as a functional head and its diachronic development has been discussed extensively in the literature – much of which would be relevant to the present discussion. At the same time, the paper under review may be seen as a contribution addressing these issues from the perspective of Italo-Greek.
Comments on the Quality of English LanguageAs also mentioned above, the use of English appears to be rather idiosyncratic at points (especially the longer paragraphs on pages 10 and 11 need to be revised to ensure clarity of argumentation).
Author Response
I have taken on board the suggestions of the reviewer and I had English reviewed by a native speaker.
My answer concerns the following comment: " the “opinion” expressed on p. 11 that na-clauses “constitute headless clauses with a silent modal” is only one of many options. The author(s) may also wish to address the idea of a specifically modal contribution on the part of na".
I accepted the suggestion of the reviewer and I decided to delete the entire paragraph concerning na-clauses, because a problem arises in my analysis. Indeed, following Ledgeway, a silent modal can be reconstructed in monoclausal constructions, such as infinitival relatives or the imperative negative codified by the negator followed by the infinitive. On the contrary, na-clauses headed by a (covert or overt) modal are not considered monoclausal structures, as is shown by the ban of clitic climbing to the main predicate. In these sentences, as the reviewer suggests, the modal meaning could be originated from the "specifically modal contribution on the part of na", as the literature on this topic generally claims.
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsComments for author File: Comments.pdf
Author Response
Thank you very much to the reviewer!
I reply below to some of your remarks
Before the discussion starts, it may be a good idea for A. to provide the verbal forms that correspond to the infinitive (is there a special inflection?). This will be useful to those readers who are not familiar with Greko or the relevant literature.
I add some informations about the infinitive in Greko.
Examples (6) and (7): these constructions have also been discussed for Modern (mainland) Greek. References were provided in my previous review, and I think they are worth being mentioned at least in a note here (Agouraki 2005, Daskalaki 2020)
Thank you for the references; however, I decided not to include these bibliographical references in my paper. I think some problems raised by these papers (especially Daskalaki 2020) concern the selection of wh-elements in these free-relative infinitival clauses, which in Modern Greek are coded by the interrogative pronoun (pion), instead of the free relative pronoun opion. This point is very interesting in comparison with Greko, which in the same context select only the relative pronoun (pou or ti < hoti). However, this issue is not the focus of my paper.
Line 150 ff: “Indeed, in these sentences too, …. is licensed.” The deontic reading is associated with the verb ‘have’ (see for example, Bjorkman & Cowper 2016). [I think it’s necessary to add the relevant reference]
I’m not agree. The deontic reading concerns not the main clause (where ‘have’ is coded), rather the subordinate one. I agree with Ledgeway in hypothesizing a silent modal in these indirect interrogative sentences codified by infinitival complements. A same modal operator can be licensed also in other sentences where the main verb is not ‘have’, as e.g. the sentence in (11) in my paper, where the main verb is ‘to know’:
(11) δen izzéro ti grázzi
neg know.1sg what write.inf
‘I don't know what to write (= I have/I can)’ (Rohlfs 1977, p. 191)
Lines 218-219: is this constraint a selectional one? Some clarification is required.
I’m sorry, I didn’t understand what you mean by “selectional”.
Line 530: where exactly does the VP move? Spec,TP? What about the subject? The structure requires some further clarification.
About this issue, I merely quoted Cruschina and Ledgeway 2016, p. 560: “Not only can the Infl domain be lexicalized by distinct auxiliaries (cf. epistemic use of Catalan auxiliary deure ‘must’ to express supposition in (25a)) but, in the absence of the latter, may be overtly filled by the raised lexical verb where its finite inflectional features can be licensed, as with the epistemic use of the future in substandard Catalan (Badia i Margarit 1962, I:391) to express supposition in (25b).
(25) a. [Infl Deu [VP tenir raó]].
(Cat.) he.must have.INF reason
- [Infl Tindrá [VP tindrá raó]].
(coll. Cat.) he.will.have reason ‘He must be right.’
Lines 633-639: I found the statement about nominative case rather misleading. Judging from Greek, the point is that although nominative case is possible (due to finiteness) the embedded subject emerges in accusative (as in ECM). In this respect, ton in (23a) is NOT cliticclimbing, but the embedded subject realized as an accusative. The same holds for (23b). The argument that this is clitic-climbing arguing against a biclausal structure is wrong. Example (23c) is of a different nature: the clitic to is doubled by to in the embedded clause. This is possible in Greek with the verb thelo: ta thelo na ta fao (them-cl want-1s prt them-cl eat-1s = lit. I want them to eat them). The interesting point has to do with the genitive (oblique) expression of the causee which is closer to the Romance causative. Incidentally, there is a paper by Manzini & Roussou (2024) which was just published in Isogloss that discusses this kind of causatives.
Thank you very much for this observation. I decided to remove these examples, because they are not central to the discussion. However, in my opinion, in 23.b, differently from 23.a, the clitic pronoun could be ambiguous between an interpretation as ECM and a clitic climbing. With intransitives it could be a formal overlap with the ECM causative constructions, since the pronoun is marked accusative in both constructions.
