What is in an Imperative? Sabine Iatridou 1 MIT 2010/2011

Transcription

1 What is in an Imperative? Sabine Iatridou 1 MIT 2010/2011 Section 1: What is in that functional projection? The form of the verb that we call the imperative has a host of interesting properties, some of which we have ideas about. By far the most common assumption about the imperative verb is that its semantic content is that of command. After all, even the name of the form is based on a verb that means command: Romance imperative from Latin impero, imperare, to command Greek prostaktiki from prostazo, to command Turkish emir kipi, command (noun) Slovenian velelnik from veleti, to command Hebrew civuy to command Albanian urdherore, from me urdheru to command Arabic fi'l?amr to command There is a common syntactic proposal that goes hand-in-hand with the above semantics, namely, that there is an imperative functional projection that ends up merging with the verb. For example, Rivero and Terzi (1995) postulate an imperative mood feature which can appear at different heights of the tree in different languages and which attracts the verb. This analysis is meant to capture, among other phenomena, the fact that the Greek imperative verb always precedes the clitic, even though the tensed verb normally follows the clitic. 1. i Maria to dhiavazi the Mary it read-prog Mary is reading it 1 Reflects ongoing discussions with Kai von Fintel. This presentation is based on a keynote address to the 2 nd Mediterranean Syntax Meeting in Athens, Greece, on October 8-10,

2 2a. dhiavase to! read-imp it! Read it! 3. b. *to dhiavase! (* as an imperative) It read-imp In other words, the hope and assumption is that a proper understanding of the functional material associated with the imperative form will make us understand all the concomitant properties. Several researchers postulate a universal deontic modal as part of the lexical entry of the imperative and this modal is part of the functional projection that ends up merged with the verb (Han, Schwager, many others). For Schwager 2006, for example (4) is the same as (5), with the modal performatively 2 used: 4. Open the door! 5. You must open the door! Note that if you don t have a modal but some sort of mood feature, you are in the same camp: either way, you believe that there is a functional projection with a certain semantics associated with it. But is the common assumption that an imperative modal contains a necessity modal or modal of command correct? Let us ask this question for 13 languages that have a Mediterranean coast. 2 When a universal deontic modal is used performatively, it bestows an obligation on somebody. When it is used descriptively, it asserts that somebody has an obligation. Similarly, when an existential deontic modal is used performatively, it gives permission to somebody to act. When it is used descriptively, it asserts that somebody has permission to act. The modal in the imperative is always and only performative, according to Schwager. 2

3 Here is an imperative form in each of these languages 3 : Table 1: Greek Anikse tin porta! Open the door! Turkish Çok çalis! much work Study hard! Hebrew 4 xabek ot-o! Hug him! Palestinian Arabic Iftah el-baab! open the door! Moroccan Arabic ftah l-bab! Open the door! Spanish Come lo! Eat it Catalan Obre la porta! Open the door! French Ouvres la porte! Open the door! Italian Apri la porta! Open the door! Slovenian Pojej to Eat this! Serbian Otvori vrata Open door! Croatian Otvori vrata! Open door! Albanian Hap deren! Open door-the There are uses of the imperative form that are not commands. One such use has been called permission and it arises when the hearer is assumed to have the desire to bring about the action in action (Sperber and Wilson 1988): 6. A: May I open the door? B: (Go ahead/sure.) Open the door. Of course, the desire does not have to be explicitly asserted. It is possible that A just looks at the window, or expresses the wish in some other non-linguistic way 5. 3 Throughout the paper, the sources of the languages are as follows: Turkish: Jaklin Kornfilt; Hebrew: Omer Preminger; Palestinian Arabic: Samer Al Khatib, Hadil Karawani; Moroccan Arabic: Jamal Ouhalla; Spanish: Maria Biezma, Eulalia Bonet, Joan Mascaro; Catalan: Eulalia Bonet and Joan Mascaro; French: Guillaume Thomas; Italian: Maria Giavazzi and Jacopo Romoli; Slovenian: Luka Crnic; Serbian: Miloje Despic; Croatian: Martina Gracanin; Albanian: Frederik Caku and Enkeleida Kapia 4 The imperative form is in the process of disappearing from spoken Hebrew and is gradually being replaced by the future form (Omer Preminger p.c.) 5 The desire has to be present; it is not sufficient for the hearer to merely be considering undertaking an action: i. A: Should I apply for that job in northern Alaska? 3

4 The imperatives of all the languages in Table I can be used as permissions. That is, any of the forms listed will function as permission-giving in the context in (6). Caveat: intonation may well play an important role that still needs to be investigated. We definitely see the impact of intonation on other covertly modalized sentences. For example, intonation determines whether the following is interpreted as containing an existential or a universal modal: 7. Na to valo edho? INFL it put here Should I put it here? or May I put it here? So, does this mean that the content of the imperative functional projection is sometimes command and sometimes permission? It has, in fact, been suggested that an imperative sometimes contains a necessity modal and sometimes an existential modal 6. However, this position seems hard to defend. On the one hand, it is suspicious if a lexical or functional item has the same ambiguity in language after language. In addition, in all the languages discussed here, overt modals are lexically specified for a quantificational force 7. It would be implausible then that there is a covert modal that can be either existential or universal. Maybe one could argue that the imperative feature is a covert universal modal which under special conditions can be coerced into an existential interpretation. This is implausible because overt universal modals cannot be coerced to a permission reading: 8. A: May I open the door? B: (Go ahead/sure.) #You must open the door B. (Go ahead/sure.) Apply. In (i), it is clear that A is not asking for permission to apply to the job in question but for advice about whether s/he should. B s response is not understood as permission, but as advice. 6 Throughout most of Schwager 2006 a necessity modal is proposed. Towards the end, she overhauls the theory and uses an existential modal. Grosz 2009 proposes that the imperative is ambiguous between containing a universal and existential modal. 7 This seems to hold more broadly crosslinguistically. A notable exception is some Salish language(s), in which certain modals can be either universal or existential (Davis, Mathewson & Davis) 4

5 What about the reverse? What if we said that the basic content/feature of the imperative is an existential modal, and analyzed the command reading as derivable? This also seems implausible, as existentially modalized sentences cannot normally 8 be used as commands: 9. (mother to child:) You may sit still! So we still do not have an obvious candidate for the imperative functional projection. Section 2: Enter the I(mperative) and D(eclarative) While it is already difficult enough to decide whether the imperative feature is a universal or existential modal, the picture becomes even more complicated by the fact that the imperative verb can appear in sentences where it is neither a command nor a permission. We will adopt Schwager s term IaD (for Imperative and Declarative ) for this construction: 10a. Study hard and you will pass the class. b. Ignore your homework and you will fail the class. The sentence in (10a) is used to instigate the addressee to study hard and for this reason, sentences like (10a) can be and have been analyzed as containing an imperative (Schwager 2006, Russell. 2007). However, notice that (10b), unlike (10a) or the command or permission use of the imperative form, is used to get the addressee to not do what the first conjunct says. That is, (10b) intends to instigate the addressee to not ignore the homework by pointing out its undesirable consequence in the second conjunct. And for this use of 8 Though under certain conditions, this does appear possible: i. (General to soldier:) You may leave now. What are these conditions? It seems that if it is presupposed by both speaker and hearer that the hearer has the obligation to undergo a certain action, a sentence like (i) can function as command for immediate action. Hadil Karawani points out that it may be sufficient if it is presupposed that the hearer is obliged to do everything the speaker says, even if some particular action is not presupposed: ii. (The queen to a servant:) You may bring me salmon in almond crust for dinner tomorrow. 5

6 the imperative verb, neither the command nor the permission view/function of the imperative verb are helpful. Quite the contrary, in fact! If you believe that (10b) feels like a conditional, you might be right. We will return to this: 10. Ignore your homework and you will fail the class = If you ignore your homework you will fail the class. Let us call the IaDs that contain the meaning of directive and that are amenable to the paraphrase containing a command, as in (10a), Type I IaDs. We will call the IaDs with the undesirable consequent, which cannot be paraphrased as containing a command, as in (10b), Type II IaDs. That is: IaDs with undesirable second conjuncts are always Type II IaDs. Type II also includes neutral cases where the speaker is neither trying to get the hearer to undergo the action of the imperative verb, as in (10a), nor to dissuade the hearer from that action, as in (10b). The reason we group these under Type II is that they cannot be paraphrased as containing a command, the way Type I can. So here are some more examples of Type II IaDs with undesirable second conjuncts: 11a. Ignore your homework and you will fail the class b. Insult him and he will get you fired. c. Don t study hard and you will fail the exam. d. Continue this way and you will be dead before you are 20. e. Eat that and your cholesterol will go through the roof. And some Type IIs with neutral second conjuncts: f. Open the paper and you will find 5 mistakes on every page 9 g. Touch this button and a light will go on Russell 2006 claims that the bare verb in the first conjunct in a Type II IaD is not an imperative but some sort of bare infinitive. However, regardless of the merits of this proposal for English, in other languages, the verb is clearly imperative in form. Of our 13 languages, 12 have IaDs. Turkish is the one that does not; we will come back to why this is. 9 Examples adapted from Bolinger 1967, Franke 2005, Russell

7 Table 2: Greek Type I Kane ta mathimata su ke ola tha pane kale do the lessons your and all will go well Type II Fae ena apo afta ke tha pethanis mesa se 24 ores Eat.IMP one from these and FUT die within 24 hours Eat one of these and you will die within 24 hours Anikse tin efimeridha ke tha vris 5 lathi se kathe selidha Open the paper and FUT find 5 mistakes on every page Turkish Type I??/* Cok CalIS ve basari-li ol-ur -sun! much work (imp.) and success-with be-aor.-2.sg Study hard and you ll succeed Type II.??/*Ev Odev-in -i unut ve basari -siz ol-ur-sun! home work-2.sg.poss.-acc. forget (imp.) and success-without be-aor.-2.sg Ignore your homework and you will fail??/* Bu gazete -yi aç ve her sayfa-da 10 baski hata -si bul-ur-sun. this newspaper open and each page-loc. 10 mistake-compound find Open that newspaper and you will find 10 typos on every page Hebrew Type I shev be-sheket ve-(ata) te-kabel pras IMP.sit(sg.M) in-quiet and-you(sg.m) FUT.2-receive(sg.M) prize Sit down quietly and you will get a prize Type II shte et ze ve-(ata) ta-mut tox sha?at-aim IMP-drink(sg.M) ACC this and-you(sg.m) FUT.2-die(sg.M) within hour-dual Drink this and you will die within 2 hours ftax/ptax?iton ve-(ata) ti-mca xamesh ta?uy-ot o yoter IMP.open newspaper and(sg.m) FUT-find five mistake-pl.f or more Open the newspaper and you will find five or more mistakes Palestinian Arabic Type I idros mneeh w b-tenjah study.imp well and b.pass.2sgm Study hard and you will pass Type II 7

8 Ilmis-ha w b-tindam tool omr-ak touchimp-it and b-regret.2sgm all life-your Touch it and you will regret it the rest of your life iftah el-malaf w bi-tlaa i khames ghaltaat b-kol open.imp the-file and b-find.2sgm five mistakes in-every safha page.fem Open the file and you will find five mistakes on every Moroccan Arabic Type I qra mzyan w gha t-njah study well and will you-succeed Type II hml l-xdma d l-mdrasa w gha t-sqt ignore the-work of the-school and will you-fail Ignore the homework and you will fail ftah l-jarid w gha t-jbar fi-ha bezzaf d l-?axta? open the-newpaper and will you-find in-it lot of the-mistakes Open the newspaper and you will find many mistakes Spanish Type I Come esto y tú sentirás major Eat this and you will feel better Type II Come esto y tú estarás muerto en dos horas. Eat this and you will be dead in two hours. Abre el periódico y tú encontrarás tres errores en cada página Open the newspaper and you will find three errors on each page Catalan Type I Dorm i demà et trobaràs millor Sleep and tomorrow you will-feel better Type II Menja això i d'aquí vint-i-quatre hores seràs mort Eat this and within 24 hours will-be.2sg dead eat this and within 24 hours you will be dead Obre el diari i trobaràs tres errors a cada pàgina open the newspaper and will-find.2sg three errors at each page French Type I travaille dur et tu réussiras work hard and you succeed-fut Type II ignore tes devoirs et tu échoueras ignore your homework and you fail-fut 8

9 ouvre le journal et tu trouveras cinq erreurs à chaque page open the newspaper and you find-fut 5 mistakes on every page Italian Type I mangialo e ti sentirai meglio eat this and you'll feel better Type II mangialo e morirai tra tre ore eat this and you'll be dead in 3 hours apri il giornale e troverai tre errori in ogni pagina open the newspaper and you will find three errors on every page Slovenian Type I pojej to in pocutil se bos bolje eat this and feel.prt self AUX better eat this and you will feel better Type II pojej to in umrl bos v 24 urah eat this and die.prt AUX.FUT in 24 hours eat this and you will die within 24 hours odpri časnik in našel boš pet napak na vsaki strani open newspaper and find 5 mistakes on each page Serbian Type I pojedi ovo i osećaćeš se bolje eat-imp this and feel fut refl-cl better Eat this and you will feel better Type II. pojedi ovo i umrećes za 24 sata eat-imp this and die fut for 24 hours Eat this and you will die in 24 hours otvori novine i naći ćes pet grešaka na svakoj strani open-imp paper and find fut 5 errors on every page Open the paper and you will find 5 errors on every page Croatian Type I pojedi ovo i osjećat ćes se bolje eat-imp this and feel fut refl-cl better Eat this and you will feel better Type II pojedi ovo i umrijet ćes unutar 24 sata eat-imp this and die fut within 24 hours Eat this and you will die within 24 hours otvori novine i naći ćes pet grešaka na svakoj strani open-imp paper and find fut 5 errors on every page Albanian Type I 9

10 haje kete dhe do te ndjehesh me mire eat this and you will feel better Type II haje kete dhe do te vdesesh brenda 24 oresh. eat this and you will die within 24 hours hap gazeten dhe do te gjesh pese gabime tipografike ne cdo faqe. open the newspaper and you will find 5 errots typographical on every page Open the newspaper and you will find 5 typographical errors on every page So now, we have three things that we want the imperative verb to do: command, permission, and first conjunct of (Type II) IaDs, which is neither command or permission. So if it is neither a command nor a permission, what is an imperative doing in an IaD? So let us try to understand IaDs better. Section 3: What are the ingredients of IaDs? 3.1 Type I IaDs The researchers who have worked on this topic (with the exception of Han 2000) assume that Type I and Type II IaDs have different derivations. The reason is that Type I can be paraphrased as containing a command, but Type II cannot. For this reason, Type I is considered easier, because one does not need to make the command reading of the imperative disappear: 12. Study hard and you will succeed = Study hard! You will succeed. 13. Ignore your homework and you will fail Ignore your homework! You will fail. Schwager 2006 and Russell 2007 have similar accounts for Type I IaDs, which consist of basically two ingredients: Type I IaDs are conjunctions of speech acts. (Krifka ) Type I IaDs contain modal subordination (Roberts ) 10 Krifka (2001) is credited with the idea that speech acts can be conjoined. Speech act and conjoins two sentences that already have force and returns another speech act. The result is as if the two speech acts applied in succession. Because it is a speech act conjunction, the meaning of the imperative is not buried in the conjunction, i.e. the meaning of the formally imperative verb is retained in its full glory, more specifically the command reading. 11 In particular, modal subordination of the following type (Roberts 1989, p. 699): The approach I suggest, which I will call the accommodation of the missing 10

11 However, we will see that this analysis of IaDs faces problems and moreover, that it is incompatible with the view that the imperative verb contains a covert modal, a view held by Schwager herself. So let s go into some details of these accounts of Type I IaDs.. The first conjunct of the IaD expresses a command. Subsequently, Modal Subordination takes us to the worlds in which the command is satisfied. 14. Study hard and you will pass the class. = 15. Study hard! In the worlds in which you study hard you pass the class. In effect, after the imperative speech act has been uttered, a conditional is created in which the modal will is restricted by an antecedent like if you study hard and the consequent of this conditional is the second conjunct. By its very semantics (fn 10), speech act conjunction is taken to be the same as speech act sequencing, i.e. two separate sentences in the discourse. We will represent sequencing as I.D s. In other words, in these accounts, (14/15) are claimed to be identical to (16): I!D: 16. Study hard! You will pass the exam. Certainly such an account of Type I IaDs seems to intuitively capture what we feel Type I IaDs to be saying. This is probably the reason why this idea is common to many accounts. In addition, IaDs provide the anaphora which is a well-known trademark of modal subordination. The pronoun it can only refer in the worlds in which the fig tree exists: 17. Plant a fig tree and it will give you plenty of good shade. Unfortunately, thuogh, on closer inspection, there are some problems that have to be dealt with (for any such account but we will focus on those accounts of the imperative form that postulate a covert modal). Problem 1: It is predicted that sentences with universal deontic modals should be amenable to a similar derivation. That is, MaDs (Modal and Declarative) should be able to be able to contain speech-act conjunction followed by modal subordination. But this is not borne out. Compare Type I IaDs with MaDs. antecedent approach to modal subordination, is the pragmatic accommodation of a contextually given hypothetical common ground to be the antecedent of the modally subordinated clause. 11

12 Type I IaDs don t behave like MaDs: 18 a. Invest in this company and you will become rich. b. */?? 12 You have to/must 13 /should invest in this company and you will become rich. 19 a. Speak to them in French and they will hire you immediately. b. */??You have to/must/should speak to them in French and they will hire you immediately. On the other hand, in speech act sequencing (as opposed to speech act conjunction), I.Ds and M.Ds do behave the same: 20 a. Invest in this company! You will become rich. = b. You must /have to/ should invest in this company. You will become rich. 21 a. Speak to them in French. They will hire you. = b. You must speak to them in French. They will hire you. In other words: In sequencing: Modal Subordination is fine with modalized sentences and imperatives alike: M.D I.D In conjunction: Modal Subordination is fine with imperatives, but not with modalized sentences: * M and D I and D In short: MaD IaD And since the conjunctor and is by assumption the same in the two cases, it follows that I M. That is, an imperative is different form an overtly modalized sentence. Of course, if we give up the idea that there is a covert modal in an imperative verb, then the fact that Type I IaDs differ from MaDs is in itself not a problem. We can say then that there is no modal subordination in conjunction --though we would have to investigate 12 The judgment is for the reading of a sentence where the second conjunct is modally subordinated to the first. The sentences are fine as plain conjunctions, of course. 13 The judgment holds even for must, which is taken to have performative uses (Ninan 2005) 12

13 why 14. And the ability of the imperative to form Type I IaDs would still have to be investigated. At any rate, it doesn t seem that a modal account is advantageous with respect to this first problem. Problem 2: A second problem that arises if we take Type I IaDs to contain a command followed by modal subordination regards the phenomenon of polarity switch. Polarity switch is a known possibility for modal subordination. Partee 1972 has this example 15 : 22. John won t buy a car because he wouldn t have space for it in his garage =John won t buy a car because in the worlds in which he does buy a car he does not have space for it in his garage. An imperative verb in sequencing (I.Ds) can involve modal subordination with a polarity switch: 23 a. Don t park there. You will be towed. = Don t park there. If you do, you will be towed. b. Conserve your energy. You will run out of breath. = Conserve your energy. If you don t, you will run out of breath. What the above shows us is that modal subordination in I.Ds can take us to the worlds in which the first clause is not satisfied (polarity switch). 14 Txurruka 2003 is a possible beginning for such an investigation. 15 At first glance, it might seem that polarity switch is somewhat easier from negative to positive than from positive to negative: (i) Don t park on the even side of the street today! You will get towed. = Don t park on the even side of the street today! If you park on the even side, you will get towed. (ii) Park on the odd side of the street today! You will get towed. Park on the odd side of the street today! If you park on the even side (i.e. if you don t park on the odd side) you will get towed. That is, it seems easier to subtract a negation rather than add one. However, example (23b) from the text and (iii) below show that it is possible: (iii) Be careful! You will fall. = Be careful! If you aren t, you will fall. It is unclear what the difference between (ii) and (23b)/(iii) is due to. 13

14 However, such a polarity switch is not possible in IaDs: 24 a. Don t park there and you will be towed. Don t park there. If you do, you will be towed. b. Conserve your energy and you will run out of breath. Conserve your energy. If you don t, you will run out of breath The only meaning that (24) can have is the one where the polarity is maintained, i.e., where you will be towed if you don t park there and where you will run out of breath if you conserve your energy. On the other hand, polarity switch is possible in sequencing not just with I.Ds but also with M.Ds 16 : 25. You must not park on the even side on the street. You will get towed. = You must not park on the even side of the street. If you do, you will get towed. 26. You should conserve your energy. You will run out of breath. = You should conserve your energy. If you don t, you will run out of breath. It suffices to say that if IaDs are reduced to speech act conjunction and modal subordination, there is no reason whatsoever why they should not be able to have a polarity switch. And yet, polarity switch is out for IaDs, while it is fine for sequencing in both I.Ds and M.Ds. This means that again, postulating a covert modal as the imperative content gets us into trouble, if we want to include speech act conjunction and modal subordination in the account.. Problem3: A third property of IaDs that seems to set them apart from other known cases of modal subordination is that typically, modal subordination permits a choice between will and would, even when the first clause is an imperative: 27. Read that book by Max. You will like it. 28. Read that book by Max. You would like it. 29. You have to/must/should read that book by Max. You will like it. 30. You have to/must/should read that book by Max. You would like it. Possibly this choice reflects the choice in the accommodated antecedent: if you do.. versus If you did,.. However, in IaDs, the choice can be only will: 16 MaDs, with modal subordination are not grammatical and thus we cannot test their behavior with respect to polarity switch and thereby compare IaDs to MaDs. 14

15 31. Study hard and you will pass the class. 32. *Study hard and you would pass the class. So this is another place where IaDs behave differently from the known cases of modal subordination. In summary, while the speech act conjunction plus modal subordination account of Type I IaDs seems intuitively appealing, it does face a few difficulties, especially for those who believe that imperatives contain a covert modal. So, as was said earlier, it can t be that both the covert modal theory of imperatives AND the analysis of Type I IaDs as containing modal subordination are correct. In short: We do not have a very good theory of Type I IaDs. What about Type II IaDs? We already saw that the not-so-good theory of Type I IaDs cannot be extended to Type II, as the latter do not contain the meaning of a command: 12. Study hard and you will succeed = Study hard! You will succeed. 13. Ignore your homework and you will fail Ignore your homework! You will fail. Are there other differences between the two Types that would make us believe that they should be analyzed differently from each other? Let s look at that next. 3.2 Type II IaDs: Same or different from Type I? Han 2000, Schwager 2006, Russell 2007 address this question. While they agree that Conjunct1 of Type II IaDs does not receive the command reading, they differ on its exact nature. For Han, it is a stripped imperative, that is, an imperative that has lost its [+directive] feature. For Russell, the first conjunct of a Type II IaD is not an imperative atvall but some sort of truncated, infinitival form. For Schwager it is still an imperative, but it gets reinterpreted in context in a way that we will come to later. Here is a quick summary as preview: Table 3: Type I IaD Type II IaD Han Conjunct 1 is not an imperative but infinitive-like. It refers to hypothetically possible worlds/becomes a conditional antecedent. Conjunct2 is the consequent. Russell Conjunct 1 is an Conjunct 1 is 15

16 Schwager imperative. Conjunct2 is modally subordinated Conjunct 1 is an imperative. Conjunct2 is modally subordinated infinitive-like. It becomes a conditional antecedent. Conjunct2 is the consequent. Conjunct1 is an imperative. It becomes a conditional antecedent. Conjunct 2 is the consequent As you can see in Table 3, all three propose that Type II IaDs should be analyzed as transforming into a conditional, with the first conjunct turning into the antecedent. Type II for Han, Schwager, Russell The Conditional Semantics Derivation: 33. Ignore your homework and you will fail = If you ignore your homework, you will fail This should not be confused with the Type I derivation of Schwager and Russell. Both contain a conditional but in (33), the antecedent of the conditional is Conjunct1, as indicated by the box. In the Schwager/Russell Type I account, Conjunct1 does not become an antecedent. The antecedent is provided by the operation of modal subordination: The Modal Subordination Derivation 17 : 34. Work hard and you will succeed = Work hard! If you work hard, you will succeed For Han, Type I and II undergo the same Conditional Semantics Derivation, namely the one in (33). Han says we think that some IaDs contain a directive for action (the ones we call Type I) but that is not the result of them containing a command. It is the result of 17 These terms are not ideal as conditionals semantics contains elements of modal subordination, but hopefully, in context it is clear what is meant here. 16

17 Conjunct2 being desirable, from which we draw the conclusion that we have an incentive to do what Conjunct1 describes. So for Han, (35) does not contain a command any more than (36) does: 35. Come closer and I will give you five dollars. 36. Come closer and I ll shoot. A unified account certainly seems appealing and it also provides a Gordian knot solution to the problems we saw exist for the speech act conjunction plus modal subordination account. If we take modal subordination out of the analysis, some problems fall away. Wrt Problem 1: We noticed that IaDs behave differently from MaD, which are basically ungrammatical on the relevant reading. On the other hand, I.Ds and M.Ds are both fine. If we say that modal subordination is not possible in conjunction but only in sequencing 18, we correctly predict the acceptability of I.Ds and M.Ds, as well as the unacceptability of MaDs. But we would still need an account of Type I IaDs of course. Wrt Problem 2: We noticed that Type I IaDs do not permit the polarity switch that is possible in modal subordination, as with I.Ds and M.Ds. If Type I IaDs is not the result of the Modal Subordination Derivation, the lack of polarity switch is not noteworthy. Moreover, if (Type I) IaDs undergo the Conditional Semantics Derivation as in (33), we expect the absence of a polarity switch, as these never happen with conditionals: 37. *If you conserve your energy you will run out of breath ( If you don t conserve your energy you will run out of breath) Basically, we have an argument that the conditional semantics is not arrived at in a contextually roundabout way but is fairly directly encoded, thus not offering an opening for a polarity switch. Wrt Problem 3: We noticed that unlike known cases of modal subordination, namely, I.Ds and M.Ds, which permit a choice between will and would in the second clause, IaDs do not. We consider the choice between will and would a characteristic of modal subordination and due to the fact that the covert antecedent which is responsible for modal subordination can be as in (38) or as in (39) and similarly for the M.Ds in (40-41) 38. Read that book by Max. If you do, you will like it. 39. Read that book by Max. If you did you would like it. 40. You must/should/ought to/have to read that book by Max. If you do you will like it. 18 Though we do not know why this should be the case. 17

18 41. You must/should/ought to/have to read that book by Max. If you did you would like it. On the other hand, with IaDs, there is no such covert antecedent, as there is no modal subordination. If IaDs undergo the Conditional Semantics Derivation, the CF morphology in the consequent is possible only with matching CF morphology in the first conjunct/ antecedent (specifically past morphology), and that is clearly lacking in IaDs. Finally, recall that we had given as argument in favor of the modal subordination account of Type I IaDs, the fact that we observe the anaphoric relationship associated with modal subordination: 42. Plant a fig tree and it will give you plenty of good shade. However, the acceptability of the anaphoric relationship between it and a fig tree would be just as expected under the Conditional Semantics Derivation for this sentence. That is, under the Modal Subordination Derivation, Sentence (42) would be derived as in (43), while under the Conditional Semantics Derivation, it would be derived as in (44): 43. Plant a fig tree and if you plant a fig tree, it will give you plenty of good shade. 44. If you plant a fig tree, it will give you plenty of shade In other words, while the acceptable anaphora in (42) is certainly consistent with a derivation that involves modal subordination, it does not argue in favor of it over an account that assigns a direct conditional semantics to the sentence. In short, there are several advantages to a unified approach. But, of course, there are also difficulties. Russell 2006 pointed out that there are several differences between Type I and Type II in English. Scontras and Gibson 2010 verified some of these with a quantitative investigation. Here are some examples that seem solid differences on the basis of the Scontras and Gibson verification 19 : Difference 1: Type I can contain emphatic do, Type II cannot: 45. Do study hard and you will succeed. 46. *Do ignore your homework and you will fail 19 To be more precise, what Scontras and Gibson found with respect to Difference 1 is that emphatic do degrades both Type I and Type II IaDs, but it degrades Type II significantly more. 18

19 Difference 2: Type I can contain an overt subject, as imperatives can in English in general: 47. a. Nobody move! b. Everybody/ each student pick up your/his pen! 47. Nobody goof off and all will go well 48. *Nobody work hard and all will be lost The above differences seem difficult to incorporate into a unified account of Type I and Type II IaDs. On the other hand, here is a difference that is possible to accommodate into a unified Conditional Semantics Derivation: Russell claims that only some IaDs can contain NPIs, namely Type II ones 20. Russell concludes this by putting an NPI in IaDs that have Type I properties for him, like emphatic do and overt subjects: 49. *Do eat any raw pork and you ll contract trichinosis. 50. *Anyone turn out the light and I ll show you my slides. 51. *Someone lift a finger to help and we ll finish building the model today. 52. Lift a finger to help her and you ll be sorry. 53. Say one word to anyone about this and I ll never forgive you. 54. Drink any more beer and you ll puke. Here is how a unified Conditional Semantics Derivation of both types could deal with this difference: Conditionals support NPIs in their antecedent, but not always. If we manage to show that the IaDs that do not support NPIs in the first conjunct correspond to conditionals that do not support NPIs in their antecedent, then the conditional account of both types will not be endangered by the variable acceptability of NPIs in IaDs. We find that this is indeed the case. Conditionals with a desirable consequent do not license NPIs in the antecedent: 55. a. Lift a finger to help him and I will never speak to you again b. If you lift a finger to help him I will never speak to you again 56. a. *Lift a finger to help him and he will finish the building in time b. *If you lift a finger to help him he will finish the building in time 57. a. *Someone lift a finger to help and we ll finish building the model today 20 That NPI can appear in IaDs was known since Bolinger

20 b. *If someones lift a finger to help, we ll finish building the model today. As we see from the above, NPI licensing in IaDs does seem to track the acceptability of an NPI in the corresponding conditional 21. So the difference in NPI-behavior between the two types does not necessarily endanger a unified Conditional Semantics Derivation. Even so, the fact that there are SOME differences between English Type I and Type II IaDs is a problem for a unified account. On the basis of these differences, Russell claims that Conjunct1 in a Type II IaD is not an imperative. If this were true, we would not need to worry about where the command reading has disappeared to, a problem which one faces if one thinks that Conjunct1 is an imperative. He says that the verb in Conjunct1 of a Type II IaD is a bare infinitive, the nature of which he does not investigate much 22 and he acknowledges that he really has no account for it. So what about the languages that we are interested in? For most of them it is very clear from the morphology that the verb in Conjunct1 is an imperative 23. So Russell s claim 21 Possibly the difference is expressible in terms of R. Lakoff s and Linebarger s Negative Implicatum. 22 Though Russell is very explicit about not committing himself to the exact nature of this bare VP, he speculates that they may be simple present declaratives with deletion of subject you. The existence of such deletion is motivated by sentences like the following: (i) (ii) Want a cracker? Know what I did today? These have in common with the base VP of a Type II IaD that the missing subject can only be you and no other person is possible (though in Type II IaDs, missing impersonal you is also possible). 23 Another piece of evidence consistent (though mysteriously so) with the position that Conjunct 1 is an imperative is that the whole IaD embeds like an imperative: i. He said (*that) call him at home (Crnic and Trinh 2008) ii. *He thinks/believes/etc that call him iii. He said (*that) eat this and you will die within 24 hours iv. a. *He thinks (that) eat one of these and you will die within 24 hours b. He thinks (that) if you eat one of those you will die within 24 hours Note that the inability of the IaD to be embedded is not a function of LSand (see section 3.3) but of the imperative in Conjunct1, as there appears to be nothing wrong with the following: v. John thinks that Mary frowns and Fred shies away in fear. 20

21 that Conjunct1 of an English Type II IaD is an infinitive does not carry over. Unfortunately, most other differences do not carry over either. There is nothing equivalent to the emphatic do-test (only English Type I IaD can). As for the overt subject test (only English Type I IaD can) in many of our languages, once you have an overt subject, you cannot use the imperative verb form anymore. For example, in Greek you need to move to the subjunctive with non-2 nd person subjects: 58. a. *Kathe fititis klise to tetradio tu each student close-imp the notebook his When the subject is 2 nd person, the verb can be imperative and there is a difference in acceptability: Type I b. Esi kane ta mathimata su ke ola tha pane kale You do the lessons your and all will go well Type II c.??esi fae ena apo afta ke tha pethanis mesa se 24 ores??you Eat.IMP one from these and FUT die within 24 hours However, if the subject is placed in a different position, Type II seems better again: 59. fae esi ena apo afta ke tha pethanis mesa se 24 ores Eat.IMP you one from these and FUT die within 24 hours In short, the subject test seems inconclusive, at least we do not quite understand it yet. We will, therefore, conclude that, though a unified account seems desirable (and we will soon see an additional argument from crosslinguistic facts in favor of it), there are difficulties with it. So now let s go to Type II. 3.3 Type II IaDs: Conditional Semantics So the idea is that a Type II IaD is basically a conditional. But how do you turn a conjunction into a conditional? Han is not very explicit in this; Russell a bit more, and Schwager even more than Russell. Both Russell and Schwager claim that IaDs contain what Cullicover and Jackendoff (1999) called Left Subordinating and ( LS and). Here is a primer on LS and: 21

22 Cullicover and Jackendoff (1999) (CJ) discuss a variety of points at which simple coordinating conjunction (and C ) differs from what they call Left Subordinating Conjunction ( LS and), which permits a conditional paraphrase 24. A. LS and permits an anaphor in conjunct1 bound by an antecedent in conjunct2, while and C does not. 60. a. Another picture of himself (appears) in the paper LS and Susan thinks that John will definitely go out and get a lawyer b. *Another picture of himself has appeared in the paper and C Susan thinks that John will definitely go out and get a lawyer B. LS and permits a pronoun in the first conjunct to covary with a quantifier in the second conjunct but and C does not. 61. a. You give him enough opportunity LS and every senator, no matter how honest, will succumb to corruption. b. *We gave him enough opportunity and C every senator, no matter how honest, succumbed to corruption. C. Both LS and and and C can be embedded but LS and permits only IP-conjunction, not CPconjunction. 62. a. You know, of course, that you drink one more beer and you get kicked out. (= that if you drink one more beer you get kicked out). b.you know, of course, that you drink one more beer and that you get kicked out. (=/= that if you drink one more beer you get kicked out). D. LS and does not permit VP-conjunction, unlike and C. 63. *Big Louie sees you with the loot and puts out a contract on you. (=/= If Big Louie sees you with the loot, he puts out a contract on you) CJ have more differences between LS and and and C and they argue that the two differ in ways which follow from the fact that with and C conjunction is symmetrical, while with LSand the two conjuncts are not on equal standing, so to speak. (though they warn against treating all asymmetrical coordinations as LS and) 24 As a safe- guard that we are dealing wth LSand, CJ use the conditional paraphrase, as well as keeping the Tense/Aspect contents of the two conjuncts what they would be in a conditional (see CJ for more details on the latter). 22

23 CJ argue that LS and is syntactically a coordination yet that there is a level where the first conjunct is (semantically) subordinated to the second and the whole sentence is interpreted as a conditional. For CJ, an important part of their paper (possibly the most important part) is to argue that the level at which LS and is interpreted as a conditional is not LF. The reason they give is that there are no syntactic transformations that will change a coordination into a subordination and the derivation from S-structure to LF can only be done with syntactic transformations. They conclude that LS and is a case of syntax-semantics mismatch for the Chomskyan Paradigm and there is a need for what they call Conceptual Structure, that is not derived syntactically from S-structure. While we think that the discovery and investigation by CJ of LS and is extremely valuable, we disagree with their larger conclusion. The reasoning is the following. When CJ claim that we are dealing with a case of a coordination that turns into a semantic subordination, what they in effect do is to say that a coordination turns into the syntax of an if-clause. That is, they compare the syntax of coordination of LS and to the syntax of an adjunct ifclause. And indeed, this is an impossible syntactic derivation. But the syntax of an ifclause is not the same as semantic subordination or even conditional semantics. It is just one of the syntactic structures that can end up with conditional semantics. In order to prove a syntax-semantics mismatch, they would need to give a semantics for conditionals for the semantic side of the mismatch. Instead, they give syntactic structures for both sides of the mismatch (that of LS and and that of an if-clause). To prove a mismatch, one would need to first assume a certain semantics of conditionals, which they don t do. Let s assume Kratzer s semantics, where one clause restricts a modal/quantifier over worlds (the restrictor) and another clause is a predicate of those worlds (the scope). What we need from the syntax is an indication as to which clause is the restrictor and which clause is the scope. One such indication can be seen in the syntax of if-then. But why should that be the only possible flag? We have another indication with LS and, at least in that the first conjunct is not postposable 25. Note that if-clauses are postposable (since we already have a sufficient flag for which clause is the restrictor). The inability to postpose the first conjunct in LS and may be exactly because we would then lose the clue as to which clause is the restrictor. In short, LS and and if-then structures contain the same amount of information that a conditional semantics needs, at least for the identification of the restrictor and scope of the modal. 25 And maybe the TMA specifications that CJ talk about is also such a flag, though we will see that IaDs defeat this point. 23

24 In addition to the full clauses (Declarative LS and Declarative) that are the main focus of CJ s discussion, they mention IaDs in a footnote and also show that a DP can function as Conjunct1 26 : 64. One more beer and you are fired 27 So hopefully now we have some idea of what LS and is. How is it computed into the interpretation of an IaD (TypeII)? Russell says nothing more about it, since he is not clear about what Conjunct1 is, other than to say that it is a bare infinitive in English. Schwager has an explicit account. Unfortunately, it has problems. Here is a brief summary of Schwager s proposal and what ails it: Unlike Russell, who thinks that Conjunct1 of a Type II IaD is no imperative at all, Schwager thinks that Conjunct1 is an honest-to-goodness imperative. So how does LS and turn something that looks like conjunction with an imperative as Conjunct1 into a conditional? In conditionals, the antecedent restricts a modal, and the consequent is the scope of that modal. In LS and environments, Conjunct1 becomes the restrictor of a modal and Conjunct2 becomes the scope. This much is intuitively clear. What Schwager argues for is that the modal whose restrictor and scope we are talking about is contained in Conjunct1. Schematically: 65. [Modal (α)] Conjunct1 LSand [(β)] Conjunct2 Modal [(α)] restrictor [(β)] scope This means that the first conjunct of LS and should always contain a modal. This is the modal that gets restricted by (the prejacent of the modal in) Conjunct1. What is the source of the modal? For the basic CJ cases, she says that present tense marks the presence of a generic operator, which plays the role of the modal in (65): 66. Big Louie looks at him and he shies away in fear 26 Conjunct1 is a DP like one more beer, it is interpreted roughly as If the addressee drinks one more beer, he is fired. 27 In fact, this is among the first cases to be noticed by linguists (Cullicover) 24

25 Imperatives can be the first conjuncts of LS and because they too are modalized in her account. This modal does not get applied to the prejacent material in Conjunct1 in the way it does in a regular imperative. Instead, the prejacent material in Conjunct1 gets mapped into the restrictor of that modal and Conjunct2 becomes the modal scope. Scwager argues that LS and comes with a special intonation contour that triggers mapping of the entire proposition embedded under a modal operator in the first conjunct into the restrictor of the modal operator (p.258) Where does the special intonation of the initial quote come into play? Schwager, following Halliday 1967, talks about how sentences like those (67) is in principle ambiguous: 67 a. (sign on an escalator:) Dogs must be carried. b. (sign outside a seaside restaurant:) Shirts must be worn Sentence (67a) has the following two readings: 68. In all the worlds compatible with the law in which there is an event involving this escalator and a dog, the dog is carried. ( w ƒ(w)) [ x e (dog w (x) on-this-escalator (x)(e))][carried w (x)(e)] and 69. In all the worlds compatible with the law in which there is an event involving this escalator, there is a dog that is carried. ( w ƒ(w)) [ e ( on-this-escalator (e))] [ x [dog (x) & carried w (x)(e)] In the first reading, dogs is in the restrictor of the universal modal. In the second reading it is not. In the first reading, dogs is deaccented. In the second reading it is accented. Deaccenting of dogs is more natural for (67a), but stressing shirts is more natural in (67b). In other words, deaccenting is a mark of being mapped in the restrictor of the modal. 25

26 With LS and, we see that Conjunct1 must be deaccented. That is, Conjunct1 must get mapped into the restrictor. Otherwise, the properties characteristic of IaDs are not possible. What are the challenges for Schwager s account? Here are some: Question1: According to Schwager, the modal in the imperative is necessarily performative (which is why FIV are always performatives and never descriptive statements) What happens to the performativity of the modal in Type II IaDs? Q2: The syntax-semantics mapping seems suboptimal: Schwager has left downward movement of the complement of the imperative into the restriction of the modal and then allows insertion of an abstractor capturing the trace left behind by the complement, and application of that abstract to the second conjunct. The operations get the right result but they are custom-made for this construction. As an attempt to fit the transformation of IaDs into a conditional by means of independently known operations, it fails. Q3: Schwager points out that IaDs can never be epistemic: 70. Feel warm and you will have caught something. 71. If you feel warm you will have caught something. She attributes this to the fact that the covert modal of the FIV is not an epistemic modal. However, as she herself notes, she wrongly predicts that other necessity operators in Conjunct1 of LS and should also work but this is not true (point attributed to Manfred Krifka). The following fails as a conditional: 72. You must come in time and you ll get a seat. 73.It must rain and you take an umbrella Q4: As also noted by Schwager, it is unclear how/why sufficiency modals would work in the LS and construction (von Fintel and Iatridou 2007): 74. You only have to look at him and he shies away in fear. Q5: Another challenge that Schwager herself notices has to do with an observation in Bolinger 1967 that she cannot account for. Bolinger 1967 points out that constructions like the ones we have been referring to as IaDs have the property that the second conjunct must be an intrinsic consequence of the first conjunct 28. Here are some examples from Bolinger, involving statives, that show the intrinsic consequent reading in action 28 See also von Fintel and Iatridou s 2007 automatic result. 26