While it is undoubtedly true that historical data do not lend themselves well to the reproduction of experimental findings, the availability of increasingly extensive data sets has brought some experimenting within practical reach. This means that certain predictions based on a combination of synchronic observations and uniformitarian thinking are now testable. Synchronic evidence shows a negative correlation between analysability in morphologically complex words and various measures of frequency. It is therefore expected that when the frequency of morphologically complex items changes, their analysability will change along with this. If analysability decreases, this should in turn be reflected in decreasing sensitivity to priming by items with analogous composition. The latter prediction is in principle testable on diachronic data, offering a way of verifying the diachronic effect of frequency change on analysability. In this spirit, the present article examines the relation between changing frequency and priming sensitivity, as a proxy to analysability. This is done for a sample of 250 English ly-adverbs, such as roughly, blindly, publicly, etc. over the period 1950–2005, using data from the Hansard Corpus. Some of the expected relations between frequency and analysability can be shown to hold, albeit with great variation across lexical items. At the same time, much of the variation in our measure of analysability cannot be accounted for by frequency or frequency change alone.

This article examines the use of the semi-modals have to, have got to and need to in the Diachronic Electronic Corpus of Tyneside English (DECTE), a corpus of spoken Northeastern English dating from the late 1960s to the present day. It will be shown that the semi-modals have, in many contexts, replaced the historically older must as markers of obligation and necessity in this variety. Moreover, the two most frequent variants in the corpus, have to and have got to, will be examined in the light of current theories of grammaticalisation. Internal and external constraints, which have been shown in the literature on root modality to have played an important role in the distribution of variants in other regional varieties of British and North American English, will be tested in DECTE. The article will also examine the rise of need to in this northeastern variety, as the most recent addition to the group of variants marking obligation and necessity.

This article analyses the formal and stylistic features of phrasal verbs in the Old Bailey Corpus and compares them with results obtained from the speech-related genres of ARCHER (A Representative Corpus of Historical English Registers), namely diaries, drama, letters and sermons. Phrasal verbs tend to be associated with spoken colloquial registers both in PDE and previous stages of the language, although this statement has been challenged by Thim (2006a, 2012), who argues that in EModE the (non-)occurrence of phrasal verbs in a particular text seems motivated by its contents, rather than its degree of formality. The results in this study show that phrasal verbs were a feature of the spoken language already in LModE, since their frequency is remarkably higher in trial proceedings, arguably the closest representation of the spoken word of the past. However, in contrast to other speech-related text types, the frequency of phrasal verbs in trial proceedings decreases over time, which seems to point towards a decolloquialisation of this genre over time. The study also shows differences at the formal level across genres, demonstrating that sermons display more formal and archaic characteristics, whereas trial proceedings contain features which bring them closer to the spoken language.

Susan Hunston, Corpora in applied linguistics. Cambridge: Cambridge University Press, 2002. Hardback £40.00, ISBN 0 521 80171 0. Paperback £14.95, ISBN 0 521 80583 X. Pp. xii + 241.

Charles F. Meyer, English corpus linguistics: an introduction. Cambridge: Cambridge University Press, 2002. Hardback £45.00, ISBN 0 521 80879 0. Paperback £16.95, ISBN 0 521 00490 X. Pp. xvi + 168.

Gerald Nelson, Sean Wallis, and Bas Aarts, Exploring natural language: working with the British component of the International Corpus of English. Amsterdam: John Benjamins, 2002. Hardback £95.00, ISBN 90 272 4888 5. Paperback £55.00, ISBN 90 272 4889 3. Pp. xviii + 342.

Pre-R Dentalisation (PreRD), the dental pronunciation of /t/ and /d/ before /r/ and /ər/, is a well-known feature of English varieties throughout Ireland. PreRD is often accompanied by an /r/-Realisation Effect (RRE), whereby /r/ is pronounced as a tap after the dentalised consonant, and a Morpheme Boundary Constraint (MBC), such that PreRD is blocked by Class 2 morpheme boundaries. Although an Irish origin for PreRD has been suggested, the presence of PreRD, the RRE and the MBC in northern English dialects in a form nearly identical to what is found in Ireland suggests that the origins of PreRD lie instead in English in Britain. The possible existence of PreRD in Scotland is suspected, but definitive evidence for PreRD, the RRE and the MBC there has never been published. In this article, I provide the first detailed analysis of these features in Scotland, using unpublished data collected as part of the Linguistic Survey of Scotland. It will be seen that there is substantial evidence for PreRD, the RRE and the MBC in Scots dialects. The presence of these features in Scotland has important consequences for their history in Britain, and confirms the British origin of PreRD in Ireland.

This study is a corpus-based diachronic analysis of English reporting parentheticals, i.e. clauses introducing direct speech, placed after or in the middle of the reported message. The aim of the investigation is to trace the development of the construction throughout the history of English, establishing the main factors influencing the choice between VS and SV patterns (i.e. with and without quotative inversion respectively), showing how various reporting verbs were increasingly attracted to the construction, and demonstrating the gradual morphological reduction of the main reporting verbs: quoth and say. The study is based on syntactically annotated corpora of Old, Middle, Early Modern and Late Modern English, and uses other corpora to illustrate more recent changes. The study reveals that reporting clauses do not show regular quotative inversion with all subject types until the Early Modern English period and links this development to the emergence of the comment clause with say. It is also claimed that quotative inversion is not directly derived from the V-2 rule and that parenthetical reporting clauses have functioned as a separate construction since the Old English period.

This article approaches the study of pragmatic meaning in English modal verbs like can, may, and shall in early eighteenth-century English prose using the techniques of corpus linguistic analysis, politeness theory, and social network analysis. The study is based on a quantitative survey of modal expressions across registers in a corpus of fourteen early eighteenth-century authors, which yields relative frequencies of occurrence according to genre and to author. Together with the sociohistorical information provided by the strategic coalitions constructed to develop, nurture and maintain relationships between clients and literary and political patrons, this survey provides the context for a pragmatic study of politeness strategies. The subsequent qualitative examination of letters in the oeuvres of selected authors yields a rich picture of individual pragmatic variation. The study reveals that although the corpus as a whole exhibits trends for different usages in different genres, the letters exhibit a predominance of modal expressions that may be used to index the facework involved in building and maintaining relationships of patronage. Within this genre, modal choice is influenced by a writer's sense of the relationship with the patron as well as the degree of risk and cost involved in the communication.

This study tests the applicability of the Complexity Principle (Rohdenburg 1996) and the Uniform Information Density Principle (Jaeger 2010) on adjectival data as regards the variation between retaining and omitting the complementizer that in English adjectival complementation constructions. More specifically, the study tests the effect of different factors of potential importance on this variation across extraposed (e.g. It was inevitable (that) he should be nicknamed ‘the Ferret’) and post-predicate clauses (e.g. I'm happy (that) we are married). While both the factors concerned with the Complexity Principle and the Uniform Information Density Principle are found to have an effect on post-predicate clauses, less clear effects are found concerning extraposed clauses. I attribute these findings to the difference between the two constructions in terms of their frequency of co-occurrence with different matrix subject types and with different adjectives.

The be like quotative emerged rapidly around the English-speaking world and has quickly saturated the quotative systems of young speakers in multiple countries. We study be like (and its covariants) in two communities – Toronto, Canada, and York, United Kingdom – in apparent time and at two separate points in real time. We trace the apparent-time trajectory of be like and its covariants from inception to saturation. We take advantage of the prodigious size of our dataset to examine understudied aspects of the linguistic factors that condition quotative variation. Building on earlier suggestions (Cukor-Avila 2002; Durham et al.2012) that be like might show patterning over time consistent with the Constant Rate Effect (or CRE, Kroch 1989), we argue that the CRE does indeed apply to the rise of be like, but needs to be handled with care. Logistic modelling assumes that the top of the S-curve is located at 100 per cent of a given variable context. In the case of be like, the saturation point is nearer 75–85 per cent, with minor variants retaining small semantic footholds in the system. In conjunction with our analysis, we suggest how to adapt the predictions of the CRE to changes likely to lead to saturation but not categorical use.

This paper deals with the metrical behavior of a class of verbs in Old English whose descendants became the syntactically distinct auxiliaries of the modern period (have, be, may, will, shall, and associated forms). Contrasting two poems from the Old English period (Beowulf and The Battle of Maldon), I show that while the verbs in question show consistently stressed metrical placement in Beowulf, in Maldon they show a pronounced tendency to be placed in unstressed metrical positions, while verbs outside the eventual class of auxiliaries differ indiscriminately. In this way, the poetry suggests a phonological difference between pre-auxiliaries and other verbs perhaps centuries before corresponding morphological and syntactic differences fully emerged in the Middle and early Modern English periods.

This article discusses the niche role that the oblique genitive of the type the friend of John’s occupies in the context of genitive variation. The article shows that the oblique genitive should be considered an independent construction which competes marginally in two syntactic contexts with the s-genitive (as in John's friend) and the of-genitive (as in the friend of John). The first context is one in which all three constructions function as the predicative complement of the clause (e.g. He is a friend of John's / John's friend / a friend of John). Note that in this context the definiteness effect of the s-genitive is downplayed, so that competition is possible with indefiniteness of the other two constructions. The second context is one where the oblique genitive and the of-genitive are introduced by the determiner the. Contrary to the claim that oblique genitive constructions introduced by the definite article must receive restrictive modification of the head (see e.g. Barker 1998; Lyons 1986), the quantitative data presented in this article reveal that oblique genitives introduced by the determiner the are not confined to pre- or postmodification of the head but can occur, albeit rarely, without any modification as in the example the executor of Sir Ralph’s.

The article further compares the oblique genitive, s-genitive and of-genitive with respect to the following five features: noun-headed vs pronoun dependent; animacy of the dependent; length of the noun-headed dependent; determiner of the head; and the semantic relations that can hold between head (e.g. friend) and dependent (e.g. John). The most intriguing theoretical conclusion is that the semantic relations available to head and dependent in the oblique genitive are a subset of those found in the s-genitive, which, again constitute a subset of those that exist in the of-genitive. This means that variation between all three constructions is not only restricted to the two syntactic contexts outlined above but also to a shared set of semantic relations.

This article presents an analysis of the phonotactic structures of English presented in The Cambridge English Pronouncing Dictionary, paying attention to morphological boundaries, the difference between stressed and unstressed syllables, the difference between native and non-native, and considering the distribution of vowels as well as consonants. The phonotactic status of names turns out to be unlike the status of other morphologically unanalysable words, and some new observations are made on consonant clusters as well as vowel sequences, which have previously been overlooked.

Ninth-century Kentish Old English <b>/<B>, where it represents a reflex of Germanic [ß], has hitherto been dismissed as an orthographic ‘archaism’. Here, however, it is argued that the graph is a genuine reflection of ninth-century Kentish phonology. The linguistic argument, which invokes the concept of neutralization in reconstructing the Old English obstruent system, is placed in the context of reconstructions of the external history of Kent. Data are drawn from forms of moneyers' names on early Anglo-Saxon coins, discussed in the context of interpretations of manuscript data.

In this article I present extensive quantitative evidence showing that it is possible to distinguish different Middle English varieties on the basis of the treatment of word-initial h-, and that it is necessary to postulate that, in some varieties, word-initial h- fails to surface in given contexts, though being present as a consonantal phoneme in the underlying representation. This casts a new light on the old problem of whether word-initial h- was lost in Middle English and restored at a later stage: the data presented here suggest that h- loss was never generalized, though h-less forms did surface as contextual variants of h-ful forms.

This article examines grammatical conditioning of the use of nonstandard -s in expletive there sentences. The discussion is based on data from a Midlands variety of English (UK). In negative constructions, speakers tend to allow for nonstandard -s with no and constituent not, but not to use it with n't or clausal not. Further, nonstandard -s is admitted in declarative existentials, but disfavoured in interrogatives. By comparison, occurrences of nonstandard -s in nonexistential structures did not seem to demonstrate these constraints. Investigating such constraints is instructive in several respects. First, they help to identify the nature of nonstandard -s. Second, they provide new evidence for the generative inquiry into issues relating to subject–verb agreement. Third, they delimit variable contexts for variationist analyses. In this way generative and variationist objectives meet. I will argue that researching syntactic variation necessitates combining generative and variationist methodology.

This article presents new evidence for the early history of the Northern Subject Rule in the form of an exhaustive corpus study of plural present-tense indicative verb forms in Northern and Northern Midlands early Middle English, analysed in relation to their syntactic context, including subject type and subject–verb adjacency. We show that variation between -∅/e/n and -s endings was conditioned by both subject type and adjacency in a core area around Yorkshire, whereas in more peripheral areas, the adjacency condition was weaker and often absent.

We present an analysis of these facts in relation to the presence of multiple subject positions in early English, which we show contra earlier literature to be relevant for Northern English as well, We view -∅/e/n endings as ‘true’ agreement, which in the relevant dialects is limited to contexts with pronominal subjects in a high subject position, Spec,AgrSP; other forms of agreement (-s or -th) represent default inflection occurring elsewhere. This analysis supports the hypothesis that the NSR arose when the extant morphological variation in Northern Old English was reanalysed as an effect of pre-existing multiple subject positions.

Based on the occasional presence of personal pronoun objects and particles to the right of the nonfinite verb, Pintzuk (1999) argues that we must allow VO phrase structure in Old English as well as OV phrase structure. This article shows that personal pronoun objects and particles follow the nonfinite verb often enough to indeed assume VO as a genuine option. During the Old English period there is a significant increase in absolute as well as relative frequency. A quantitative analysis reveals that the relevant cases are found particularly often in coordinate contexts (pronouns) and ACI constructions (particles).

My contribution explores the potential of Standard Average European (SAE) as a methodological yardstick for the assessment of the Celticity of Standard English (SE), i.e. the degree to which SE may have been influenced by Insular Celtic languages. SAE allows a quantification of the extent to which SE differs from other SAE languages and at the same time shows similarities with Insular Celtic languages which are generally thought to deviate from SAE. The concept of SAE will be introduced and a complex test case will be analyzed: the rise of identical reflexives and intensifiers and the frequency of ‘labile’ verbs, e.g. to break both intransitive and transitive, in English. Reflexives and intensifiers are identical in Insular Celtic, but separate categories in SAE. According to Haspelmath (1993), labile verbs are unusually frequent in English by SAE standards, and they are also very frequent in Insular Celtic. It will be shown that the frequency of labile verbs in SE and Insular Celtic results from the identity of reflexives and intensifiers in these languages. Since the frequency of labile verbs thus can be shown to be a typologically secondary phenomenon, it is the rise of the new system of identical reflexives and intensifiers in SE which is methodologically the relevant linguistic subsystem within the complex test case for which Insular Celtic influence can be proposed and which can be adduced as an argument in favor of the Celtic hypothesis. For the assessment of the Celtic hypothesis, it will therefore be rewarding to complement the initial comparison of SE and Insular Celtic with further data from SAE and general typology whenever the linguistic subsystems under scrutiny allows the inclusion of such data.

This article offers a syntactic analysis of the construction [be done NP], e.g. I am done dinner, I am finished my homework, as found in Canadian English and some US dialects. After situating this construction in the context of a productive transitive be perfect in Scots/English dialects, [be done NP] will be distinguished from a set of its conceptual and structural relatives, and ultimately be shown not to be reducible to a surface realization of another underlying structure. From the perspective of syntactic theory, the article problematizes the parsimony of the mainstream generative approach (most recently in MacFadden & Alexiadou 2010) in accounting for the facts of [be done NP] on strictly compositional grounds, as well as the mainstream view of lexical items as projecting theta grids and subcategorization frames (as e.g. in Grimshaw 1979; Emonds 2000). Following Fillmore et al. (1988), Goldberg (1995, 2005) and others, what will be suggested instead is a construction grammar approach to [be done NP], under which a construction holistically licenses its argument structure. Along these lines [be done NP] will be characterized as an abstract construction with some fixed material.