George N. Clements String-Vacuous Rule

Working Papers in Scandinavian Syntax, 2016

The goal of this paper is to provide a systematic overview and analysis of the syntax of Icelandic adverbial clauses in terms of the whether they do or do not allow so-called main clause phenomena. The classification of adverbial clauses follows the typology of Haegeman (2012) where adverbial clauses are divided into two classes: central adverbial clauses that resist main clause phenomena and peripheral adverbial clauses that may permit such phenomena (XP-fronting etc.). It turns out that fronting is possible in a subset of adverbial clauses exactly as predicted by Haegeman's typology and such examples are found in both in judgement data and written sources. Further, this initial work shows that there appears to be a distinction between argument fronting (less free) and adjunct fronting (more free) in Icelandic and this is a distinction that has not previously been systematically examined. It has long been observed that adverbial clauses exhibit variable word order. In English for example, some adverbial clauses allow argument and adjunct topicalization whereas other resist such fronting. Here, we are concerned with similar word order variation in adverbial clauses in Icelandic primarily and data from other Scandinavian languages is presented briefly for comparative purposes. This paper is largely descriptive, however it can be taken as a first step towards a typology of adverbial clauses in Scandinvian more generally. The framework adopted here is the typology of adverbial clauses set out in Haegeman (2012, and much previous work) where adverbial clauses are divided into two groups: those that allow main clause phenomena and those that do not. Further, Haegeman distinguishes between adverbial clauses that she terms 'peripheral' or 'central' to capture the degree of integration of adverbial clauses with respect to the clause that they modify. For Scandianvian verb second languages then, we might expect that adverbial clauses that are 'peripheral' in Haegeman's sense may allow main clause word order whereas such orders are resisted in central adverbial clauses. The second part of the paper presents an overview of Haegeman's typology as applied to adverbial clauses in English. The third part of the paper dicusses data from both written and spoken Icelandic. In the fourth section, we provide comparative data from other Scandinivian languages showing that adverbial clauses can be analyzed with respect to the possibility of main clause phenomena as predicted by Haegeman's typology. This is followed by a short discussion.

Working Papers in Scandinavian Syntax, 2008

This paper presents a novel observation on Swedish subordinate clause structure. I show that in embedded V2 clauses, negation may only precede quantified subjects in the position following the complementizer; this is the very same limitation as found in the first position of main clauses. This correlation I take to provide a strong argument for assuming V-to-C movement in embedded V2 clauses. This claim is of course not new in itself, but my approach to V-to-C movement is. Thus I argue that by focusing on the position immediately to the right of the complementizer, we are offered a new tool for distinguishing the structural properties of different subordinate clause types in Swedish.

Chicago Linguistic Society, 2011

"Beginning with Geis (1970), several authors have provided syntactic, semantic, and etymological arguments for a derivation of adverbial (subordinate) clauses that involves movement of an (often null) operator (see Haegeman 2010a for a review). This paper provides morphological evidence for this view while arguing for separate extraction sites for the moved elements in temporal and conditional clauses. The Bantu language Akɔɔse (Hedinger 2008) exhibits wh-agreement (see Reintges, LeSourd, & Chung 2006 for a typology); that is, it marks its verbs with respect to whether an element has been extracted to the left periphery. This extraction marking occurs not only in canonical wh-movement contexts (Chomsky 1977), such as constituent questions, relative clauses, cleft questions, and topicalization, but also in temporal and conditional adverbial clauses. Crucially, Akɔɔse wh-agreement encodes whether the extracted element originated above or below v. The distribution of wh-agreement morphology shows that the operator in central conditional clauses is extracted from a position above v (supporting Haegeman's (2010b) claim), the moved element in central temporal clauses originates below v (supporting Larson's (1987, 1990) position), and there is no extraction in peripheral adverbial clauses (as argued in Haegeman 2007). Wh-agreement provides compelling evidence for a movement analysis for both temporal and conditional central adverbial clauses. Due to its sensitivity to height of extraction, Akɔɔse lends insight into the question of where the moved elements originate, unlike languages like Irish (McCloskey 2001) where wh-agreement only registers the presence of movement."

Lingua, 2014

Argument structure in morphology and syntax: An introduction 1. Where do arguments come from?

The Journal of Comparative Germanic Linguistics, 2010

In this article we investigate the status of two different types of movement in subordinate clauses in Faroese: the movement of the finite verb to a position below the subject but above negation and medial adverbs (V-to-I), and the movement of some XP and the finite verb to positions above the subject (V2). The exact status of these phenomena in contemporary Faroese, a language that has been argued to be undergoing syntactic change, is a matter of dispute; we attempt to clarify this using the methodology of Magnitude Estimation (ME). We extend what is known by presenting the results of a systematic comparison of judgment data from Icelandic (where the finite verb obligatorily moves to a high position within the clause, and embedded V2 has been claimed to be possible quite generally), Danish (where the finite verb obligatorily remains in a low position, and embedded V2 has been claimed to be restricted), and Faroese (where the status of verb movement is precisely at issue, and the availability of embedded V2 has been little explored).

Journal of Languages and Culture, 2016

The theory of universal grammar relies predominantly on the biolinguistic concept of natural endowment and innate knowledge of the general principles of language. It postulates that all humans are naturally endowed with the general rules and configurations of language and to this extent, all natural languages have similar structural features. The theory of universal grammar as hypothesized by Chomsky and propagated by other linguists not only recognizes the universality of the general principles of language but also the existence of language-specific idiosyncratic features that constitute parametric variations among languages. These are the parameters of universal grammar. The most prominent parameters that create distinctions between languages are head directionality, pro-drop or null-subject and wh-parameters. This paper reviews the null-subject parameter in English and juxtaposes its occurrence or non-occurrence in the Ịzọn language. The aim of the paper is to characterize the parametric choices by English and Ịzọn languages in the derivation of grammatically convergent sentences with null-subject constituents. The study is competence-based and used data from tokens of sentences in conversation among competent native speakers of Ịzọn language. Data from each language were translated into the other via a gloss and comparatively analysed. The study reveals that null-subject constituent is not a characteristic feature of English syntax but a feature of Ịzọn syntax. The study is significant because it contributes fresh linguistic data for the principles and parameter theory.

Working papers in Scandinavian syntax, 2008

This paper presents a novel observation on Swedish subordinate clause structure. I show that in embedded V2 clauses, negation may only precede quantified subjects in the position following the complementizer; this is the very same limitation as found in the first position of main clauses. This correlation I take to provide a strong argument for assuming V-to-C movement in embedded V2 clauses. This claim is of course not new in itself, but my approach to V-to-C movement is. Thus I argue that by focusing on the position immediately to the right of the complementizer, we are offered a new tool for distinguishing the structural properties of different subordinate clause types in Swedish.

2019

In his article Three clause-final particles and the syntax of clausal complementation in Dravidian, K. A. Jayaseelan discusses the role of the question particle-oo, the complementizer ennǝ, and the relativizer-a, which occur in a fixed order in Malayalam, in case they co-occur. He argues that this order can be generated only if we postulate that the complementizer, which is a quotative element derived from the verb 'say', still retains its verbal syntax and projects its own clause. The relativizer-a can then be in the C domain of the clause projected by ennǝ, and the question particle-oo can be in the C domain of the CP complement of ennǝ. A surprising consequence of this analysis is that every embedded finite clause in Dravidian-the 'ennǝ + clause' structure-is in fact bi-clausal. Rahul Balusu's article Fine tuning the Dravidian left periphery: The three 'complementizers' in Telugu picks up on this, now with a focus on the related Dravidian language Telugu. He investigates in detail three left-peripheral morphemes that have been considered at various places in the previous literature as instances of complementizers. According to Balusu, none of these morphemes are typical complementizers. The linearly first left-peripheral morpheme-aa has all the signature properties of a polar question particle and is in many respects similar to its Hindi counterpart kyaa. The second left-peripheral morpheme,-oo, delimits the scope of questions in Telugu. This he attributes to its location in the Spec of CP, where it is basegenerated, and to its semantics, which is essential for interrogative semantics, thus explaining scope delimitations. The third left-peripheral morpheme, the quotative complementizer ani, is analysed as being syntactically and semantically true to its source, a verbum dicendi, the verb say, and its complementizer nature as arising only due to its not putting forth its extended projection (in the spirit of Grimshaw 2005) and instead being merged into the matrix clausal spine at various levels. The third contribution in this section turns to yet another Dravidian language, namely Tamil. In their article Discourse-driven scrambling to the peripheries in child Tamil, R. Amritavalli and Annu Kurian Mathew argue that the SOV-language Tamil has a pre-verbal focus and postverbal topic position. A subject wh-word must occur in focus, and not in a topic or in a canonical S(ubject) position. This leads to the distribution: *SwhOV, POSwhV, *OVSwh. Utterances from Josef Bayer & Yvonne Viesel 3 children 26-29 months of age are shown to obey these word order restrictions. The authors argue that child scrambling in Tamil moves arguments to criterial positions to check topic/focus features. A possible generalization with Japanese is suggested. A non-focus account of wh-is briefly critiqued. The second section, on Indo-Aryan, continues with the article Clause particles and cleft sentences in Bangla: Some preliminary generalizations by Probal Dasgupta. Intimacy-oriented discourse particles (DiPs), called Modul[ator]s in the Bangla syntax literature normally follow a finite verb or a compact wh-phrase. In his article, Dasgupta surveys interactions between a Modul and Zero Copula Construction (ZCC) in three subtypes of ZCC. He extends the discussion to other contexts now diagnosable as ZCCs-sentences in which a post-verbal constituent hosts either a Modul or some other DiP. He argues that certain sentences with these properties instantiate cleft constructions whose properties are explored here in the context of the study of DiP elements. Some preliminary generalizations are proposed. Section 3 contains two contributions on Japanese. The phenomenon of DiPs, which was introduced in Dasgupta's article, plays a role in the first article here, as well as in Sergio Monforte's article in Section 6, which concludes this volume. Yoshio Endo's article Exploring right/left peripheries: Expressive meanings in questions discusses non-standard questions in Japanese such as rhetorical, surprise, disapproval, exclamative, etc. (Obenauer 2006, Bayer and Obenauer 2011, Bayer 2018) within the framework of the cartography of syntactic structures. After introducing the basic ideas of the cartographic approach, Endo first examines the expressive meanings of some wh-expressions asking for reasons such as what…for, how come, etc. familiar from languages such as English, German, etc. He then turns to the main topic of examining various sentence final particles in the right periphery of the Japanese sentence to show how they contribute to creating expressive meanings in questions. Methodologically, he does this by looking at translations of Peanuts comics. Endo draws comparisons with German, where corresponding particles are placed in clause-medial position, and he speculates about the absence of similar particles in English. The article by Norio Nasu, Adverb-predicate agreement in Japanese and structural reduction, turns to the related topic of sentence adverbs (S-adverbs). In cartographic work, S-adverbs have a high position in the adverb hierarchy. Nasu shows that in Japanese, S-adverbs occur with a particular inflectional form of a predicate. He argues that this phenomenon is a manifestation of the agree relation between the adverb and a functional head. An agree-based analysis correctly predicts that an S-adverb can occur in more than one position as long as it is able to c-command the functional head it agrees with. It also accounts for restrictions on the cooccurrence of more than one S-adverb in a single clause. In Japanese, an epistemic adverb cannot precede an evidential adverb. The illegitimacy of this order is reduced to an intervention effect arising from agree. Nasu's analysis predicts that some S-adverbs in Japanese can occur at the edge of more than one functional projection as long as they enter an agree relation with the appropriate functional head. In this respect, the distribution of Japanese S-adverbs presents a departure from a principal assumption of the cartographic approach, i.e. a constituent appearing on the clausal left periphery is in a one-to-one spec-head relation with the appropriate functional head. The two contributions that appear in Section 4 discuss mainly the head-final language Turkish but also draw comparisons with the partially head-final language German. The article by Tamer Akan and Katharina Hartmann, SOV-X: Syntactic and pragmatic constraints of the postverbal domain in Turkish, sets out to develop a novel syntactic account for the postverbal domain in Turkish, which establishes a tight connection between syntactic and information-structural (IS) properties of the language. The authors first analyze the properties of the Turkish postnominal domain in comparison to the SOV-language German. Turkish is much less restricted than * I wish to thank Katalin Kiss for helpful comments on this paper. I also wish to thank the audience at the conference for an insightful discussion. 'He said (he) would come.' he (Nom.) come-1stP.Sg. say-3rdP.Sg. K. A. Jayaseelan 9 In (5), the complement of ennǝ is just a representation of a sound; there is no C domain here to generate ennǝ in. Even the noun complement construction can have a simple nominal as the complement of ennǝ, cf. (6) "kaakka" enn-a waakkǝ '(the) word "crow"' crow QUOT-REL word What such data show is that ennǝ is still a 'say'-verb, which can take as its complement anything that can be 'said', i.e. uttered; e.g. a sound ('Say "Boo!"'), or a word ('Say "crow"'), or a clause ('Say "Mary is pregnant"'). Though bleached in meaning-in (5), e.g., the machine doesn't 'say' anythingennǝ retains its verbal syntax. 2, 3 5 Clausal complementation in Dravidian What we have said has serious implications for the syntax of clausal complementation in Dravidian. When 'say' takes an object complement-irrespective of whether it is a sound, word, or clause-it goes without saying that it is outside that complement. Now consider a sentence where ennǝ takes a finite clause as its complement: (7) John [ Mary wannu ennǝ ] paRaññu 'John said that Mary has come.' John Mary came QUOT said We can now see that the correct analysis of (7) is that ennǝ is outside its CP complement; it is not in the C domain of the embedded clause at all. The 'say'-verb projects its own clause, which is nonfinite but can have its own C domain. The structure we postulate for (7) is (8) (abstracting away from word order): 4 2 Do we wish to entertain a "squishy" account of ennǝ, saying that it has been reanalyzed as a complementizer when it takes a clausal complement, but that it is still a 'say'-verb when it takes a nominal expression as its complement? Such a "two ennǝ's" analysis would be unsatisfactory for several reasons. First of all, note that ennǝ occurs indifferently with assertive and interrogative matrix verbs, showing an insensitiveness to the matrix predicate which is unexpected in a complement but is quite in keeping with an adjunct: (i) John [ Mary wannu ennǝ ] paRaññu 'John said that Mary has come.' John Mary came QUOT said (ii) John [ Mary wannu-oo ennǝ ] coodiccu 'John asked whether Mary has come.' John Mary came-Q QUOT asked Again, where do we generate ennǝ in the C domain? Suppose we generate it as the head of Finiteness Phrase. Then, in a sentence like (ii) above (or like (4)), the question particle-oo-and by implication ForceP-will have to be below the Finiteness Phrase; and a "low ForceP" will make Dravidian a typological oddity. 3 Readers unfamiliar with Dravidian languages might ask: Is ennǝ confined to the complements of 'verbs of saying'? It is not. The matrix verb can be any verb that takes a clausal complement, cf. (i) Mary [ John kaLLan aaNǝ ennǝ ] wiśwasiccu/ samśayiccu/ aaroopiccu Mary John thief is QUOT believed/ suspected/ alleged 'Mary believed/ suspected/ alleged that John is a thief.' But there is one restriction that needs to be noted on what ennǝ can take as its...