In order to account for the absence of the verb-second (V2) phenomenon in the presence of a complementizer, early work on Dutch and German proposed that the finite verb in V2 clauses moves to the position of the complementizer. 1 More recent work, however, has exposed considerable crosslinguistic variation in the availability of V2 in

Abstract: This paper takes a closer look at the attraction properties of T. It highlights an empirically attested distinction between rich agreement inflection, exhibited by null-subject languages, and rich tense inflection, found in Romance, but not Germanic, and argues for the syntactic relevance of this distinction. We propose a novel typology of the ways in which T’s featural requirements can be satisfied, and, focusing on tense requirements, show how the typology enables us to understand the verb-movement behaviour of the Romance languages vis-à-vis their Germanic counterparts. We also show how the proposed analysis facilitates a new understanding of relevant aspects of the modern English verbal system and its diachrony.

Introduction Clahsen (1991) and Clahsen & Penke (1992) have argued using data from the 2-year-old German children Mathias, Daniel, Julia and Simone that verb placement stabilizes only when the second person singular (2SG) suffix on main verbs has been acquired. In contrast, Poeppel & Wexler (1993) argue based on Andreas' data (age 2;1) that verb placement is already correct at the earliest stage of syntactic acquisition. In this paper, we present new evidence in support of early verb raising based on Katrin and Nicole who are several months younger than Andreas and have an impoverished inflectional paradigm lacking the 2SG affix. On the other hand, the lack of CP-related constructions suggest that Katrin and Nicole maximally project an IP in finite clauses, contrary to Poeppel & Wexler's proposal. Moreover, the preponderance of empty subjects in nonfinite clauses suggests that such utterances reflect a bare VP-structure, following Roeper & Rohrbacher (1994). The data discussed

Abstract not found

distinction'. It contains a detailed discussion of the analysis of Germanic verb second phenomena presented in Zwart (1993a) (henceforth referred to as DS) and earlier work. Scattered throughout Gärtner and Steinbach's paper are a number of interesting problems for the analysis of DS. Some of these are problems for any analysis of verb second, and all of them present excellent material to work on if we wish to further our understanding of the intricacies of Germanic syntax. Gärtner and Steinbach, however, do not appear to have had such constructive goals in mind when writing their paper. Their single aim appears to have been to show that the analysis in DS is ‘objectionable on conceptual as well as on empirical grounds'. In trying to do so, they fail to recognize that an analysis that generates problems is not necessarily wrong. Their is no doubt in my mind that the analysis in DS will turn out to be wrong in many respects. This reply, however, is the place to defend that it is both consistent within the theoretical framework adopted and conceptually pleasing. Gärtner and Steinbach are wrong both in denying the former and in ignoring the latter. 1. The Analysis of DS

The article proposes to describe agreement as a relation between sisters, i.e. as a function of the structure building operation Merge. Subject-verb agreement, then, is not mediated by a functional head in a specifier-head configuration, or triggered by a probe-goal mechanism, but the expression of a direct dependency relation between the subject and its sister. Supporting evidence is found in cases where agreement has multiple controllers, multiple targets, or finds alternative realizations within the predicate. 1.

As has been observed many times, complex linguistic forms can be divided into a lexical and a functional part. Roughly, the lexical part tells what the word means, and the functional part tells what syntactic properties the word has. For

In a number of Continental West Germanic dialects, complementizers may be inflected for person and/or number agreement with the subject. 1 (1) a...da-n-k ik goa-n West Flemish that 1SG subject_clitic I go 1SG

Language as perceived by the ear is one-dimensional: a string of sounds. But no language utterance is understandable if the sound bites are not grouped together as morphemes, words, and constituents. For language to make sense, it needs this second dimension: structure. Comparative grammar addresses the two-dimensional structure of sentences rather than

I consider some of the claims that have been made for and against the nature of the INPUT in OT syntax as developed within the assumptions of the Minimalist Program, leading to suggestions for further specification of the architecture of this approach. Comparing with the role of faithfulness in the OT approach developed from Lexical-Functional Grammar, I argue that specific linguistic analyses crucially involve reference to faithfulness constraints (MAX and DEP in correspondence-based OT) which apply across different parts of the output structures, but do not need to refer to the INPUT. I conclude that while OT syntax does not need INPUTs per se, it does need faithfulness constraints.