A different perspective on embedded Verb Second

2020 ◽  
pp. 297-322
Author(s):  
Rebecca Woods

This chapter compares embedded verb movement phenomena in English with embedded Verb Second clauses in German and Swedish. Close examination of the syntactic—but more particularly the semantic and pragmatic—properties of these phenomena reveals striking similarities, and the claim is made that these phenomena exhibit independent illocutionary force in the sense that the perspective holder for the embedded proposition or question is disambiguated—a departure from the claim that embedded verb movement structures are asserted (cf. Julien 2015 and Chapter 11 of this volume). It is proposed, following recent innovations in speech act syntax (Wiltschko and Heim 2016; Woods 2016) that these structures are dependent, as the ‘embedded’ clause contains less structure than full a root clause, yet is still structurally larger than a typical embedded clause. However, they are not selected and are instead in an apposition relation with a (usually covert) nominal complement to the matrix verb.

2020 ◽  
pp. 265-280
Author(s):  
Marit Julien

This chapter addresses the assertion analysis of Mainland Scandinavian embedded declarative V2 clauses. These clauses are identified by having the finite verb preceding all sentence adverbials and/or having a non-subject in initial position. Whereas this word order is mainly found in that-clauses embedded under certain predicates in modern Mainland Scandinavian, it was more generally allowed in Old Scandinavian. The Old Scandinavian word order arguably involved verb movement to the inflectional domain. In modern Mainland Scandinavian it necessarily involves movement to the C-domain, which means that embedded V2 clauses have root properties. On this analysis, they are asserted, and both the illocutionary force and the V2 order are consequences of the presence of a Force head. For most speakers of modern Mainland Scandinavian, direct or indirect assertions can be embedded whenever they are compatible with the semantics of the matrix clause.


Linguistics ◽  
2016 ◽  
Vol 54 (4) ◽  
Author(s):  
Timur Maisak

AbstractA crosslinguistically unusual case of morphological fusion, in which two clauses fuse morphologically in the absence of preceding syntactic fusion or clause union, is found in the East Caucasian language Agul. This phenomenon involves a set of “verificative” verbal forms (forms that seek ‘to find out the truth value or the value of an unknown variable’). The verificatives are completely morphologically bound, but manifest clear biclausal properties: in particular, the introduction of a new agentive argument by the verificative (the ergative “verifier”) causes no change in the argument structure of the embedded clause. This article argues that the Agul verificative has grammaticalized from the matrix verb ‘see’ plus an indirect question complement in the conditional form: over time, the two verbal heads have fused into one form. Partial parallels to this development can be found in the related languages Archi and Lezgian, where a semantic shift from ‘see’ to ‘check, find out’ is attested, together with a change in subject encoding from typically experiential (dative) to canonically agentive (ergative). Still, the complete morphologization of the verificative structure in Agul dialects remains exceptional given its comparatively recent origin, the infrequency of the construction, and the general absence of observed cases in which matrix verbs become fused with their complements.


Author(s):  
Jila Ghomeshi

AbstractIn this article it is shown that Persian has core control constructions in which the obligatorily empty subject of an embedded clause takes its reference from an antecedent in the next higher clause. Evidence is provided that these embedded clauses are relatively transparent for scrambling and lack independent tense. It is therefore argued that core control verbs in Persian take complements that lack CP, TP, and a Case position for their subjects. Control complements do manifest subject agreement, however, suggesting that agreement checking takes place within vP. The implications of this view are explored with respect to the periphrastic progressive construction, in which both the auxiliary and the main verb bear subject agreement, and raising constructions, in which preposed subjects do not trigger agreement on the matrix verb. The relevant contrast is presented in minimalist terms as the idea that agreement in Persian is checked within a strong phase (CP or vP).


2016 ◽  
Vol 7 ◽  
pp. 9-52 ◽  
Author(s):  
Axel Holvoet

The article deals with the ‘permissive middle’, a permissive construction characterized by the coincidence of the permittor and the embedded clause patient (as in They allowed themselves to be cheated) and belonging to the middle voice in the sense that its formal markers, though originating as reflexive pronouns, have lost their original reflexive function. Such permissive middles can be clearly set apart from permissive reflexives in those languages which have a formal differentiation of reflexive pronouns proper and originally reflexive markers that have shifted to middle or mediopassive function. The data of the Baltic languages are used in the article to illustrate the formal properties of permissive middles (a characteristic feature is the oscillation between reflexive marking on the matrix verb and on the embedded verb) and the tendencies in their development. Permissive middles are also shown to be attested outside Baltic, e.g. in East Slavonic. The second part of the article is devoted to a discussion of the place of the permissive middle on the semantic map of the middle voice, and in particular to its relationship to the ‘curative’ middle (the ‘causative-reflexive’).


Author(s):  
Christine Meklenborg Salvesen ◽  
George Walkden

Old English (OE) and Old French (OF) both display verb-second (V2) word order in main declarative clauses. Different models may account for V2: (a) the finite verb must move to a head in the CP field; (b) it must remain in the IP field; or (c) it moves to the left periphery only when the preceding XP is not a subject. While the IP-model should allow free embedded V2, the two others would either exclude completely or strongly limit the possibilty of having embedded V2. We select embedded that-clauses and analyse the word order with respect to the matrix verb: embedded V2 is possible in both OE and OF, although the availability of this structure is restricted. OE has very few occurrences of embedded V2, whereas OF seems to permit this construction more freely. We link this difference to the site of first Merge of complementizers in the two languages.


2020 ◽  
Vol 132 (3) ◽  
pp. 287-306
Author(s):  
Lorena Núñez Pinero

This paper offers a pragmatic analysis of a rarely used construction in Classical Spanish: an emphatic comparison of equality with optative illocution A comparative sentence such as Así me ayude Dios como fue buena mi intención (’May God help me just as my intention was good‘) is used for emphasizing the assertion fue buena mi intención (’my intention was good‘) This construction is probably a Latinism It occurs in Latin, especially in Plautus and Terence, and is mostly attested in Spanish in humanistic comedy and in the Celestinesque tradition of the sixteenth and seventeenth centuries The first member of the construction is interpreted at the pragmatic level as a reinforcer of the illocutionary force of the comparative construction as a whole, which expresses an indirect assertive speech act Speakers perform this type of act by satisfying its sincerity condition: they believe that the event of the second member is true, because if it were not, they would run a risk, i.e. the optative would entail a curse for themselves By contrast, when the event is true, the optative entails a good wish for themselves This paper also analyzes how the pragmatic properties of the construction are reflected in its semantic and morphosyntactic properties


Author(s):  
Craige Roberts

This essay sketches an approach to speech acts in which mood does not semantically determine illocutionary force. The conventional content of mood determines the semantic type of the clause in which it occurs, and, given the nature of discourse, that type most naturally lends itself to a particular type of speech act, i.e. one of the three basic types of language game moves—making an assertion (declarative), posing a question (interrogative), or proposing to one’s addressee(s) the adoption of a goal (imperative). There is relative consensus about the semantics of two of these, the declarative and interrogative; and this consensus view is entirely compatible with the present proposal about the relationship between the semantics and pragmatics of grammatical mood. Hence, the proposal is illustrated with the more controversial imperative.


Author(s):  
Sarah E. Murray ◽  
William B. Starr

This essay sketches an approach to speech acts in which mood does not semantically determine illocutionary force. The conventional content of mood determines the semantic type of the clause in which it occurs, and, given the nature of discourse, that type most naturally lends itself to serving as a particular type of speech act, that is, to serving as one of the three basic types of language game moves-making an assertion (declarative); posing a question (interrogative); or proposing to one’s addressee(s) the adoption of a goal (imperative). This type of semantics for grammatical mood is illustrated with the imperative.


Author(s):  
Sam Wolfe

This book provides the first book-length study of the controversial subject of Verb Second and related properties in a range of Medieval Romance languages. Both qualitative and quantitative data are examined and analysed from Old French, Occitan, Sicilian, Venetian, Spanish, and Sardinian to assess whether the languages were indeed Verb Second languages. The book argues that unlike most modern Romance varieties, V-to-C movement is a point of continuity across all the medieval varieties, but that there are rich patterns of synchronic and diachronic variation in the medieval period which have not been noted before. These include differences in the syntax–pragmatics mapping, the locus of verb movement, the behaviour of clitic pronouns, the syntax of subject positions, matrix/embedded asymmetries, and the null argument properties of the languages in question. The book outlines a detailed formal cartographic analysis both of both the synchronic patterns attested and of the diachronic evolution of Romance clausal structure.


1982 ◽  
Vol 12 (3) ◽  
pp. 509-526 ◽  
Author(s):  
Thomas Hurka

John Searle has charged R.M. Hare's prescriptivist analysis of the meaning of ‘good,’ ‘ought’ and the other evaluative words with committing what he calls the ‘speech act fallacy.’ This is a fallacy which Searle thinks is committed not only by Hare's analysis, but by any analysis which attributes to a word the function of indicating that a particular speech act is being performed, or that an utterance has a particular illocutionary force. ‘There is a condition of adequacy which any analysis of the meaning of a word must meet,’ Searle writes, ‘and which the speech act analysis fails to meet. Any analysis of the meaning of a word must be consistent with the fact that the same word (or morpheme) can mean the same thing in all the different kinds of sentences in which it can occur.' Hare maintains that the word ‘good’ is used to indicate the speech act of prescribing. He maintains that one of the principal functions of this word is to indicate that utterances of sentences containing it have prescriptive illocutionary force, and that an analysis of its meaning must make explicit and ineliminable reference to this force-indicating function. But ‘good’ regularly occurs in sentences utterances of which appear to have no prescriptive illocutionary force.


Sign in / Sign up

Export Citation Format

Share Document