2

Grand Meeting 2007 at Lake Mývatn

Abstracts

Artemis Alexiadou (University of Stuttgart):
On argument supporting nominals and the mass vs. count noun distinction
According to Grimshaw (1990), argument supporting nominals differ from non-argument supporting ones in that the former cannot pluralize and cannot occur with indefinite determiners. However, it has been noted that argument supporting (AS) nominals can occur both with plural morphology and indefinite deteminers, if they are telic. Since this seems to be a valid cross-linguistic generalization, in this talk I will investigate the link beetween AS nominals and the mass vs. count noun distinction.


Bennis, Hans (Meertens Institute):
Principles of Paradigmatic Levelling: Variation in Verbal Inflection in Varieties of Dutch
Within the context of the Syntactic Atlas of the Dutch Dialects (SAND) inflectional data of more than 250 dialects have been collected. A large amount of variation can be observed in the realization of present tense verbal inflection. In this paper I will present an overview of this variation and show that the occurring paradigms can be explained by linguistic principles exclusively. In order to do so I will relate synchronic geographic variation to diachronic variation. It will be argued that general principles of economy belonging to the linguistic component determine the variable process of deflection that is the dominant feature of verbal inflection in varieties of Dutch.


Barbiers, Sjef, Olaf Koeneman & Marika Lekakou (Meertens Institute):
Perfect Doubling
Perfect tense in Standard Dutch is expressed by an analytic construction consisting of an auxiliary (HAVE or BE) and a past participle. In southern Dutch dialects (i.e., Brabantish), a perfective auxiliary and participial morphology can be doubled:

(1)	Ik	heb	vandaag	nog	niet	gerookt	gehad				Brabantish
	I	have	today	yet	not	smoked	had

This has also been reported for Bavarian German (Poletto 2006), Franco-Provençal (Carruthers 1994) and Northern-Italian (Poletto 2006). The phenomenon of perfect doubling raises the following issues: (i) Does perfect doubling have semantic or pragmatic import, or is it redundant? (ii) Why don’t all languages with a comparable analytic perfect tense have perfect doubling? (iii) How should perfect doubling be analyzed syntactically? (iv) Does the difference between doubling and non-doubling varieties follow from the syntactic analysis? This talk offers answers to these question on the basis of a systematic investigation of 55 southern Dutch dialects. Two central findings of general relevance are that perfect doubling provides new evidence for the split tense hypothesis as advanced in Vikner 1985, Zagona 1988, Giorgi and Pianesi 1991 and Cinque 1999, a.o.) and that the absence of perfect doubling in Standard Dutch is the result of an haplology rule (cf. Neeleman and Van der Koot 2006) applying to morphosyntactic features at PF and subject to prosodic conditions (Selkirk 1986) and recoverability of deletion (Chomsky 1981).


Kristin Melum Eide (NTNU, Trondheim):
Finiteness
There exist numerous approaches to the notion of finiteness. In my talk, I will argue that the relevant distinction in the Germanic languages is the one between absolute and relative tense forms, and this is morphologically encoded as +/- finiteness. Whereas the erosion of this distinction in the verbal paradigm leads to a loss of main verb raising, as in English, the reverse situation is also possible; certain Creole languages being a case in point. Moreover, I will argue that certain Northern Norwegian dialects are likely to undergo a change similar to English, since the productive verb classes lack the morphological distinction between finite and non-finite forms.


David Lightfoot (Georgetown University & NSF):
Parameters as cues
Under a cue-based approach to language acquisition, a child grows her I-language in response to structures expressed in the ambient E-language. These structures are the cues designated in UG and they are expressed in sentences that a child hears which can only be analyzed, given everything else the child knows, if a particular cue is utilized. This provides strong predictions about the learning path and a discovery procedure for language acquisition instead of the usual procedure of evaluating grammars against a corpus of sentences. Furthermore, it changes the terms of recent debates about the nature of parameters.


Frederick Newmeyer (University of Washington):
In defense of the autonomy of syntax
Throughout most of its history, what has distinguished generative syntax from virtually all other approaches to grammar is the hypothesis of the autonomy of syntax (AS), namely the idea that the syntactic rules and principles of a language make no reference to meaning, discourse, or language use. However, AS has become progressively weakened in the past two decades, as is illustrated by the theta-criterion, the idea that c-selection can be derived from s-selection, UTAH and the Hale-Keyser approach, the cartography program, the wh-criterion and the Neg-criterion, the reintroduction of lexical decomposition, and much more. The purpose of this talk is to reassert the ‘classic’ view of AS, which was predominant among generative syntacticians until the 1980s.

I begin with two examples from English: the modal auxiliaries and derived nominalizations. There are profound formal generalizations involving both, which were uncovered many decades ago. But much AS-rejecting current work, in attempting to represent structurally subtle semantic generalizations regarding these phenomena, has all but abandoned any attempt to account for their well-known formal properties. I then turn to negation. For any language, the default assumption is now that where we have semantic negation, we have a Neg Phrase projection. I argue that this projection, whose primary motivation is semantic, obscures the formal similarities that we find (differently for different languages) between negative elements and other elements in the particular language. I go on to show that there are surface generalizations regarding the distribution of formal elements that current theory is in principle incapable of capturing, given the focus on capturing semantic generalizations structurally.

I conclude with two general remarks. First, I argue that current practice leads to redundancy, by representing in the syntax many semantic generalizations that need to be represented in conceptual structure anyway (the Cinque hierarchy is an example). Second, I note that the arguments for the poverty of the stimulus leading to an innate UG are based entirely on the correctness of AS. The rebuttal from cognitive linguists, AI people, etc. is ‘The arguments for UG fall through because Chomsky et al. do not realize how isomorphic syntax is to semantics’. Thus the semanticization of syntax poses a potential threat for the entire ‘Chomskyan’ program.


Cecilia Poletto (Universities of Padua and Venice):
On negation splitting and doubling
In this work I intend to examine the distribution of various negative markers in Northern Italian dialects and will start from Zanuttini’s observation that there are various negative markers located in different positions inside the sentence structure.

The structure of the clausal negative markers occurring in the NIDS is the following:

(1) [NegP1 non [TP2 [NegP2 mia [ TP2 [NegP3 nen [ Asp perf. [Asp gen/progr. [NegP4 no ...]

Each negative marker corresponds to an "etymological type", and negative doubling (or discontinuous negation) and tripling are widely attested. However, only some combinations of the various negative markers can be found: the combination between NegP3 and NegP4 is never attested.

Moreover, in dialects with optional doubling and tripling each negative marker provides a different semantic contribution to the sentence: while the ‘mia’ type and the no type are related to different presuppositions (as originally noted by Cinque (1977)), the ‘nen’ (nothing) type is related to Focus.

I will put forth an analysis in terms of a complex NegP containing several FPs, from which the negative markers are extracted and inserted into different FPs in the clausal structure, an analysis which can handle a) the distribution of the negative markers in the clause b) their semantics c) the impossible combinations.

At the end I will make some methodological considerations concerning the relation between etymology and syntactic analysis.


Luigi Rizzi (University of Siena):
On Delimiting Movement
A comprehensive formal theory of movement must include

1.     locality principles, determining the maximal structural space

In this talk I will give e general overview of the issues, and then will focus on delimiting principles, with special reference to the cases which force a movement chain to stop and pass the representation on to the interpretive systems.

In the perspective of “movement as last resort” guidelines, at least certain core cases of movement chains look like devices for associating two kinds of interpretive properties to linguistic expressions: properties of argumental semantics (thematic roles), and properties of what Chomsky calls in recent work scope-discourse semantics: topicality, focus, scope of different kinds of operators. I will argue that such properties are structurally assigned to elements through dedicated syntactic positions, adopting, in particular, the “criterial” approach to scope-discourse interface properties, a view congenial to the cartographic endeavour, which studies the configurational maps of such interpretively dedicated positions.

Argumental and criterial positions are natural “delimiting” points for movement. I will look in some detail at the effects of a particular kind of delimiting principle, Criterial Freezing, terminating a chain as soon as a criterial position is reached. A system based on the criterial freezing idea will be illustrated, and will be used to revisit two classical topics of Government-Binding syntax: the Extended Projection Principle, requiring that clauses have subjects, and the Empty Category Principle, originally introduced to capture subject-object asymmetries in extraction processes. It will be shown that if the obligatoriness of subjects is expressed in criterial terms, the difficulty of moving subjects may be derived from Criterial Freezing. Various kinds of strategies that natural languages use for circumventing the freezing effect, hence for extracting subjects, will be discussed and illustrated.

After establishing the empirical motivations for delimiting principles, I will discuss in the final part of the talk some the possible functional motivations for the existence of such principles in natural language syntax.


Henrik Rosenkvist (Lund University):
Multiple subjects in Övdalian
In his dissertation on Övdalian, Levander mentions that pronominal doubling is "very frequent" (ytterst vanlig) in Övdalian (1909:109). Two of his examples are: Read the rest of the abstract (pdf-file)


Peter Svenonius (University of Tromsø):
Parameters, Microparameters, and Things that go Bump in the Night
There are rather few serious candidates now for the kind of parameter that was envisioned 25 years ago, a switch setting that determines a host of properties that characterize a language in a typological way (such as pro-drop, head-initial, dependent-marking, non-configurational, etc.). The study of dialect syntax shows that the same sorts of properties which do vary across language families also vary across dialects (null elements, word order, morphological marking, optional movements, etc.), but at a finer degree of resolution, suggesting that parametric variation must involve numerous microparameters. But microparameters do not shoulder the explanatory burden that parameters were introduced to bear; in order to be explanatory, a parameter must have effects beyond the evidence that causes it to be set the way it is. Therefore, the search for Universal Grammar has been most fruitful when it has focused on what were referred to as principles in Chomsky (1981), including universal constraints on phrase structure such as Kayne's LCA, Cinque's postulation of a richly structured functional sequence, Baker and Brody's Mirror Theory of morphosyntax, and Chomsky's phase theory. None of these postulates introduces a parameter, in the 1981 sense of the term. Parametric variation is now commonly modeled in terms of properties of individual lexical items, in particular the featural specifications on them. A theory is needed of how lexical items may vary from each other, and how their variant properties can be learned. In this talk I hope to shed a little light on what is needed.


Höskuldur Þráinsson and Ásgrímur Angantýsson (University of Iceland):
The Rich Morphology Hypothesis Reconsidered
The so-called Rich Morphology Hypothesis (RMH) maintains that there is a correlation between rich verbal morphology and verb movement to the Infl-position. There are two main versions of the Rich Morphology Hypothesis (RMH) and they could be stated informally as follows (leaving aside for the moment what ‘rich verbal morphology’ means):

(1)  a.  The strong RMH:
           A language has V-to-I if-and-only-if it has rich verbal morphology.           
      b.  The weak RMH:   
           If a language has rich verbal inflection, then it has V-to-I.

Several linguists have argued against the RMH, especially its strong variant (1a). Much of the evidence presented against it has come from Scandinavian languages and dialects, such as the Tromsø-dialect (or Regional Northern Norwegian – a dialect that has poor verbal morphology but appears to have V-to-I) and, more recently, Älvdalsmålet (a Swedish dialect that appears to have rich verbal morphology but numerous exceptions to V-to-I). It has even been maintained that Icelandic may not have V-to-I after all.

In this paper we want to reconsider the basic evidence for and against the RMH. We will briefly consider different variants of the RMH and then we will show that one particular version of it, namely that of Bobaljik and Thráinsson (1998), in fact predicts that it should be possible to find apparent exceptions to the RMH of the kind that have been illustrated in recent literature on Scandinavian. We will then offer some speculations as to when and how such apparent exceptions might arise and what they might be related to.

References


Øystein Alexander Vangsnes (University of Tromsø):
Pinning down fluctuating grammars:
On main clause wh-syntax in a Western Norwegian dialect area
The lack of Verb Second (V2) in root wh-questions in Norwegian dialects is one of the few syntactic points of variation well-known from the traditional dialectological literature. The phenomenon has nevertheless not been exhaustively studied from a microcomparative perspective, i.e. with full, but hopefully the ScanDiaSyn investigations will bring about a more refined picture of the phenomenon from a cross-dialectal point of view. This paper will present some results from a sosiolinguistically inspired case study of lack of Verb Second in main clause wh-questions in dialectal Norwegian. The study had the following design.

1) Main point of variation: Whether lack of Verb Second is restricted to monosyllabic wh-constituents only (as e.g. in the Tromsø dialect) or not (as in e.g. the Nordmøre dialect).
2) Locations: Two municipalities in the inner part of the Sognefjord in western Norway, one (Sogndal) being the regional center (center), the other (Gaupne) a local center (periphery).
3) Working hypotheses: (i) There is an ongoing change whereby the dialect(s) in question are becoming less restrictive with respect to the form of the wh-constituent when allowing non-V2, (ii) the regional center should be leading on in the development whereas the other location should be more “archaic”.
4) Subjects: Two groups of eight females at each location were subject to study, i.e. four groups and 32 individuals in total. The subjects were either around 18 or around 40 years of age.
5) Methods: (i) Recodings were made where subjects were paired together and asked to interrogate each other with instructions that facilitated the production of wh-questions. (ii) Each subject was individually asked to judge sentences in an orally presented questionnaire according to a seven point scale.

This paper focuses on the questionnaire part of the investigation, and some general findings are the following:

– Acceptance of complex wh-constituents with non-V2 was obtained for each of the four groups: the older peripheral group was the least permissive one whereas the young peripheral and older central groups were the most permissive. (Spontaneously produced examples are present in the corpus for all groups as well.) This both supports and disconfirms the working hypotheses in interesting ways.
– Different kinds of complex wh-constituents are judged differently in non-V2 contexts and seem to be ordered along a scale where the wh-adverbs ketí ‘what-time’ and kelais ‘how’ get mean scores for each group that fall (more or less) within the ‘good’ part of the scale whereas degree questions obtain a lower score throughout. This suggests that complexity need not be just a matter of mono- vs. polysyllabicity in the dialects.


Workshop for PhD-students and younger researchers

Magnus Brenner (University of Helsinki)
S-forms of verbs in Finland Swedish dialects
The s-form of the verb in Swedish (dikten lästes upp 'the poem was read out') is not only used for the passive but also for other purposes. Some of the uses found in dialects in Finland do not occur in standard Swedish (for example Freudenthal 1870:81, Holm 1952, Huldén 1959). I am currently collecting data from dialects in order to pinpoint this variation. At the workshop I will present the results from my empirical investigation.

Freudenthal, A. O. (1870): Om svenska allmogemålet i Nyland. Finska littertur-sällskapet. Helsingfors.
Holm, Gösta (1952): Om s-passivum i svenskan, företrädesvis folkmålen och den äldre fornsvenskan. Lundastudier i nordisk språkvetenskap 9. CWK Gleerup. Lund.
Huldén, Lars (1959): Verbböjningen i Österbottens svenska folkmål II. Studier i nordisk filologi 48. Svenska littertursällskapet i Finland. Helsingfors.


Federica Cognola (University of Padua)
The loss of OV patterns in the German dialect spoken in the Fersina valley
The German dialect spoken in the Fersina Valley (from here on Mòcheno) has lost many syntactical properties typical of the German varieties it belongs to, sharing a well known pattern for the languages spoken in the so called “linguistic islands” (Grewendorf/Poletto 2005). In this paper I will concentrate on the variety of Palù, the most conservative one, which shows a very interesting shift from OV to VO syntax.
In main declarative clauses the unmarked word order is VO (1a), whereas the German syntax (Satzklammer) is not excluded, but conveys a focus interpretation to the XPs appearing in front of the past participle (1b). Read the whole abstract (pdf)...


Piotr Garbacz (Lund University)
What determines V°-to-I° movement in Oevdalian?
Oevdalian, an East-Scandinavian language spoken in Dalarna in Sweden by approximately 3000 people (Sapir 2005), was claimed in the beginning of the 20th century to exhibit obligatory verb movement in subordinated clauses (V°-to-I° movement) (Levander 1909:124), but the later data presented in Rosenkvist (1994) show that this verb movement no longer is obligatory. Trying to explain what determines V°-to-I° movement in Oevdalian, Rosenkvist (1994:22-25) claims that verb movement is obligatory in clauses with null subject (1), but optional in clauses with overt subject (2): Read the whole abstract (pdf)...

(2)    Hva tror du (at) kommer i morgen?
        what think you (that) comes in tomorrow
        'What do you think comes tomorrow?

Under the general freezing theory, direct objects are expected to be frozen in situ (assuming the Vergnaud-approach), which fits badly with the empirical terrain. I will propose a solution to this puzzle, which will be reminiscent of Rizzi's (2006) intuition that there is a special 'criterion' for subjects, although I will argue that we need a more principled explanation than he provides. I will follow Boeckx' decomposition of EPP and argue that it provides a natural explanation for the freezing asymmetry related to subjects and direct objects. This has interesting implications for the understanding of narrow syntax and its relationship to the interfaces, as it shows that it is possible to develop a theory of locality that relies on the hypothesis in (5) (Lohndal 2007c).

(5) Islands emerge due to independent principles of the computational system

This, I will argue, is ideally the most minimalist theory of islands avaliable. This talk will take some steps towards establising (5), but only future research will show whether it is attainable.


Einar Freyr Sigurðsson (University of Iceland)
Possessive hjá-construction in Icelandic
In Icelandic, the preposition "hjá" 'with, at', usually denotes a location:

(1)    Jón býr hjá mér þessa viku.
        John stays with me this week
        'John is staying with me this week.'