There have been several claims about negative contraction in Old English dialects. According to Levin (1958) contraction is much less regular in Anglian dialects than in West Saxon, and this conclusion has been widely accepted. More recently, Hogg (2004a) has argued that uncontracted forms were also frequent in at least one variety of West Saxon. Both of these claims rely crucially on evidence taken from word-by-word glosses of Latin. However, no assessment of the evidence was done in either case to ensure that the observed patterns could not be due wholly or in part to the influence of Latin. This paper looks in detail at the data from glosses to determine to what extent they can actually be used as evidence for the behaviour of negative contraction in the varieties of Old English concerned. It will be seen that, despite indications of various degrees of Latin influence, the data from the Anglian glosses give sufficient grounds to support Levin's conclusion that uncontracted forms are more frequent in Anglian than in West Saxon, although some refinements are needed in relation to particular verb forms. In the case of Hogg's claim, on the other hand, it turns out that the observed pattern can be attributed entirely to the effect of Latin influence and/or copying from an exemplar.

This paper reconsiders the role of the Subset Principle (Berwick 1985) in the context of language acquisition and change. "Traditional" parameters (e.g. the null subject parameter) define intersective languages, with the result that it is not clear how superset versus subset relations of the kind entailed by the Subset Principle are to be defined. Here, we argue that, once the role of true formal optionality ("free variation") is fully taken into consideration, it becomes possible to define grammars which generate languages that are in inclusion, rather than intersection relations. This, we propose, in turn facilitates a natural explanation of certain diachronic changes, the key idea being that the absence of sufficiently robust evidence for the grammar which generates the large language leads acquirers to "default" to a grammar generating a smaller language. We develop this idea in relation to two types of change attested in the history of English: one where a previously available pied-piping option is lost in favour of obligatory stranding, and the other what we call "restriction of function" in which an operation becomes restricted to a subset of the contexts in which it formerly applied.

This paper examines the use of right dislocation in a corpus of speech from York, England. A number of different right dislocation structures are considered, including one which is currently said to be restricted to Northern varieties of English (He's a nice lad is Leon). This variant is used alongside two variants found throughout the English-speaking world (He's a nice lad Leon and he's a nice lad Leon is). Results show that the supposedly Northern variant is used in York and, more importantly, that it is the most frequently selected form in all age groups and both genders. Tokens in which the dislocated subject is a pronoun (He's a nice lad, he is) are the only exceptions to this, in that they overwhelmingly favour the third variant. These findings highlight the importance of examining the three right dislocation forms together and also that the supralocal variants are not increasing in the youngest generation of speakers.

This paper develops an account of do-support in verb focalisation constructions in Central and Western Basque (CWB) dialects. In particular, this paper argues that CWB dialects, along with Korean, form a class of do-support languages whose dummy verb insertion mechanism differs slightly from that in English (Chomsky 1957, Pollock 1989) and Monnese (Benincà and Poletto 2004). In all four of these languages, the dummy verb values a strong feature in a position that is, in marked environments, inaccessible to the verb. However, in Korean and CWB, unlikely in English and Monnese, the verb's inability to raise to value this feature is not due to its inflectional poverty, but rather because it must bear a nominalising infinitival affix for independent reasons; this nominal infinitive may not bear aspectual morphology, and a dummy verb is merged to do so instead. Moreover, Basque do-support is not a last-resort phenomenon as in Chomsky's classic analysis of English do-support (Chomsky 1957). That is, the unavailability of do-support in non-verb focalisation constructions is not due to competition with a more economical alternative, but rather is independently excluded. This approach, then, avoids a violation of the Inclusiveness Condition (Chomsky 2000) inherent in economy-based approaches to do-support that generate the dummy verb in the computational component (Chomsky 1995).

The loss of Negative Concord (NC) has long been attributed to external factors. This study readdresses this issue and provides evidence for the failure of external factors to account for the ultimate disappearance of NC in Standard English. A detailed study of negation in Late Middle English and Early Modern English reveals that the process of decline of NC was a case of a natural change, preceded by a period of variation. A close study of n-indefinites in negative contexts and their ultimate replacement with Negative Polarity Items (NPIs) in a number of grammatical environments shows that the decline of NC follows the same pattern across contexts, which indicates that the loss of NC is a natural process. The decline of NC took place at the same rate across contexts indicating that the loss of NC is triggered by a change in a single underlying parameter setting. Accordingly, a theory-internal explanation is suggested. N-words underwent a lexical reanalysis whereby they acquired a new grammatical feature [+Neg] and were thus reinterpreted as NQs, rather than NPIs. This lexical reanalysis was triggered by the ambiguous status of n-words between [±Neg] and thus between single and double negative meanings. The observed change, involved in this process of disambiguation, is treated as a case of parameter resetting as this lexical reanalysis affected a whole set of lexical items and can thus economically account for the different observed surface changes.

In this paper, I wish to present an account of the historical development of the subjective function of the progressive in combination with always-type adverbials (i.e. adverbials with a meaning similar to always). In present-day English, the combination is connected with negative speaker-attitude. The present paper aims to explain the development of this feature and to relate it to general insights about subjectification processes. It is structured as follows: after some introductory comments (section 1), a definition of subjectivity will be provided and a brief sketch of the subjective functions of the progressive today will be presented (section 2). The main part (section 3) will look at the semantic development of the combination of progressive + always-type adverbial, as well as at the development of such adverbials and of the progressive independently from one another. I shall then discuss the question in how far this development can be described as an instance of subjectification in (section 4). Finally, the conclusion summarises the most important results (section 5).

This paper presents an innovative method of data elicitation for use in both large-scale regional language variation studies, and for use in studies of socially-stratified variation in a particular variety. The methodology combines the elicitation of data for various levels of linguistic analysis simultaneously as well as allowing for the collection of metalinguistic comment and attitudinal data. It was originally devised and designed to fit the requirements of a proposed national collaborative venture, the Survey of Regional English (SuRE).¹ It was then expanded for use in a sociolinguistic study of Middlesbrough English undertaken by the author. It has since been used in adapted form for various studies of language variation and change in British English of both the in-depth single variety type and the large-scale survey type.

A standard observation concerning basic constituent order in Old English (OE) is that the position of finite verbs varies by clause type. In root clauses, the finite verb tends to occur towards the beginning of the clause and we frequently find Verb Second (V2) order. In contrast, in subordinate clauses finite verbs generally occur towards the end of the clause, and these clauses are frequently verb-final. In this article we challenge the traditional assumption that verb-final orders and, hence, the occurrence of the finite verb in a head-final structural position are rare in OE root clauses. We present new data demonstrating that the frequency of head-final structure in OE root clauses is much higher than previously acknowledged. We then explore some of the implications of this finding for the general structural analysis of OE.

Despite technological advances, 'taking the history' remains the doctor's most important tool in the diagnosis of seizures. In this paper we focus on the differential diagnosis of epileptic and psychogenic non-epileptic seizures (NES). Misdiagnosis rates are high, and diagnostic errors often results from poor history taking, including suboptimal questioning and inaccurate interpretation of the patient's responses. We describe a novel procedure for history taking and a method of analysis which appears to be superior to the conventional fact-oriented approach in the differential diagnosis of patients with seizures. The procedure is designed to leave the patient free to direct the course of the conversation, as opposed to the doctor asking a series of questions. The analytic method takes cues from linguistics and Conversation Analysis, and focuses as much on how patients talk about their seizures as on what precise symptoms they list. We illustrate the procedure and method through a case comparison of two of our patients, one of whom has epilepsy and one of whom has NES.

This paper examines three different constructions in Japanese (relative clauses, cleft constructions, and right dislocation). It is shown that these constructions do not display subjacency when the head phrase lacks a Case-marker, but show full subjacency effects when the head is Case-marked. After listing empirical data, we show that even those sentences that do not display subjacency effects derive through movement. We then point out that these constructions display properties of topicalised sentences, and are derived from topicalised sentences.

Our research investigates variation within the use of two dependent variables (dh) and (th) by employees of the Inland Revenue in Kingston-Upon-Hull, East Yorkshire. We report on the occurrence of the non-standard variant [v] for the variable (dh) and the non-standard variant [f] for the variable (th): variants referred to collectively as 'th-fronting'. With observations of the geographical spread of th-fronting over the last two decades, we make use of both real- and apparent-time methods in order to investigate the change in usage within Hull, comparing our results with those of Williams and Kerswill (1999). We focus specifically on the effects of age and gender, showing (i) the possible retention of th-fronting beyond adolescence, and (ii) an apparent shift in usage across genders. We also observe speakers fronting (dh) and (th) to a similar degree unlike Williams and Kerswill (1999) where (dh) tended to be fronted more than (th); thus providing potential evidence towards a change in the overall distribution of fronted variants across (dh) and (th).