Developments in English historical morpho-syntax, edited by Claudia Claridge and Birte Bös, consists of thirteen individual studies on various topics in English historical linguistics. The studies are briefly discussed and classified according to their topics in an introductory chapter written by the editors.

Claridge and Bös point out that although all the periods in the history of English are represented in the studies, the main weight is on Old English (OE) and (Late) Modern English (LModE). They recognise one common denominator for all the studies: they are concerned with profound typological changes in the long diachrony of English, affecting morphophonology on the one hand and syntax and word order on the other. Jointly, these changes have resulted in a large-scale typological restructuring of English, and each study included in the volume contributes to our understanding of these macro-level changes. All the studies strike their own balance between different degrees of empiricism and theory. One noteworthy feature characterising the present volume is that in many of the chapters the authors build on their previous research. This, however, should not be regarded as a drawback: on the contrary, these studies have benefited from extended, or differently focused, databases and sharpened theoretical acumen.

The first study in this volume,‘The dynamics of changes in the early English inflection: Evidence from the Old English nominal system’, is by Elżbieta Adamczyk, who discusses the remodelling of OE nominal inflection in the light of data derived from the Dictionary of Old English Electronic Corpus (DOEC). The changes seen during the OE period turn out to be the result of the interplay of several factors: the frequency of occurrence, the morphophonological salience of inflectional exponents, and the formal inflectional overlap across the paradigms. Analogical pressures towards making the system transparent and functional are the driving force behind the changes. At a more detailed level, frequency is analysed as consisting of four distinct planes: frequency of the paradigm form, relative frequency, lemma frequency and frequency of the paradigm/scheme. One of Adamczyk's findings is that, in the plural paradigms of OE minor stems, the more extensive the spread of analogical inflection, the less salient the original inflectional exponent. Another factor discussed is the correlation between neutral forms and the degree of innovation in a minor paradigm. Generally speaking, the more neutral forms two paradigms share, the higher the level of innovation. The chapter concludes with a discussion of the interaction between the salience of inflectional exponents and frequency of occurrence.

In ‘“Subsumed under the dative”? The status of the Old English instrumental’, Kirsten Middeke explores the question of whether the instrumental case in OE should be regarded as a case in its own right or only as vestigial and non-productive. Her data come from the York–Toronto–Helsinki Corpus of Old English Prose (YCOE). Middeke sees the dative case as a merger of Germanic instrumental and dative; the merger is non-reciprocal as the dative adopts functions of the instrumental but not vice versa. The forms studied include þy, which is the instrumental form of se/þæt, and instrumental-case adjectives. With instrumental-case determiners, the typical environments are adverbials of time and manner. The same is true of instrumental-case adjectives, such as ælc, sum, oþer and the possessive min.

Middeke makes use of collocational and collexeme analysis. The former relies on the POS annotation in the YCOE, and in at least one case, the tagging has led the author to a wrong conclusion: according to table 4, mæssepreosthad is a collocate of þy, although þy is in fact linked with the ellipted dative singular gere ‘year’.

In his chapter, ‘“Thone vpon thother”: On pronouns one and other with initial th- and t- in Middle English’, Jerzy Nykiel discusses the Middle English (ME) pronouns thone, thother, tone and tother. The first two are straightforward combinations of the reduced definite article th’ and the pronouns one and other. The origin of the corresponding forms with initial t- is more complex: the mechanism behind them can be either metanalysis or the reduction of the definite article. The pronouns with initial t- are often preceded by a full form of the definite article (e.g. þe tone, þe toþer). In such cases, definiteness is marked twice. Nykiel provides a detailed analysis of the uses and features of the four pronouns in ME. Most of the uses are compatible with those of one and other in Present-day English (PDE).

The diachronic development of the reduced th’ is examined in connection with the four pronouns in ME and Early Modern English (EModE). The pronouns show a marked rise in frequency in late ME with a peak reached in M4 (1420–1500). With nouns, there are two peak periods, M2 (1250–1350) in ME and E1 (1500–70) in EModE. Nykiel's data come from three corpora: the Penn–Helsinki Parsed Corpus of Middle English (PPCME2), the Penn–Helsinki Parsed Corpus of Early Modern English (PPCEME) and the Corpus of Middle English Prose and Verse (CMEMPV). The last-mentioned corpus, although more unwieldy than structured corpora, was included because it is considerably larger than PPCME2.

The main question discussed in Rebecca Colleran's chapter, ‘Leveraging grammaticalization: The origins of Old Frisian and Old English’, is whether the similarities attested between Old Frisian (OFris) and OE are due to ‘genealogical’ relatedness or to geographical proximity. Colleran addresses the problem by studying two test cases where she sees shared grammaticalisation processes between the two languages.

The first test case is the development of the lexical verbs OFris aga and its OE cognate agan, which both mean ‘have, own’, into an auxiliary ‘have to’, a process only attested in OFris and OE among the Germanic languages. This grammaticalisation process does indeed provide impressive evidence for shared inheritance. The second test case is one where Colleran finds evidence for participle-based complements of the type I saw John crossing the street in the two languages. She sees this as a possible case of grammaticalisation, but the evidence is somewhat problematic. OFris has here a ‘perception infinitive’ in -n, a form derived from the present participle according to one theory. OE does have the present participle in -ende in corresponding constructions; these, however, are according to Timofeeva (Reference Timofeeva2010) mostly found in translations from Latin, and regarding them as instances of grammaticalisation is problematic.

In her chapter, ‘Old English wolde and sceolde: A semantic and syntactic analysis’, Ilse Wischer examines the semantics and syntax of the past forms wolde and sceolde of OE pre-modals willan and *sculan. Unlike the present tense forms, the past forms have so far received scanty attention in the literature. Wischer's contribution takes an important step towards filling this desideratum. The data for the study consist of the poetry texts in DOEC. Within the chosen data there is one omission for which no grounds are given: the fused negated forms of nolde with 48 attestations have been left out of the discussion while the 33 non-fused forms of the type ne wolde remain part of the data.

Wischer draws a basic distinction between event modality and propositional modality. The former is ‘directive’ and subject-related, the latter ‘assumptive’ and speaker-related. Event modality branches into dynamic and deontic, and propositional modality into epistemic and evidential. Both verbs undergo grammaticalisation, as a result of which they are predominantly used as auxiliaries. The bulk of the study is centred on an analysis of the uses of wolde and sceolde. It turns out, among other things, that wolde typically represents dynamic modality, while sceolde is basically deontic. Epistemic modality and non-past-time reference are attested with both verbs. It is noteworthy that with both sceolde and particularly with wolde the modal meaning can be bleached, and the verbs become markers of future in the past.

Sofía Bemposta-Rivas addresses the semantic and structural changes seen in the verbs dare, tharf and need from ME to EModE in her chapter, ‘A corpus-based study on the development of dare in Middle English and Early Modern English’. The topic is complex as the semantic and syntactic developments of these verbs are interconnected. PPCME2, PPCEME and the York–Helsinki Parsed Corpus of Early English Correspondence provide the data. Out of the many semantic and structural developments examined, a few can be mentioned here. In OE, dare (*durran) and tharf (þurfan) are two separate verbs. In ME, they come to be confused partly because of their orthographical and phonological similarity, partly due to the semantic and syntactic influence of need on tharf. As a result, tharf becomes obsolete, while need becomes increasingly frequent in the transition from M4 to E1. From M3, need begins to be used to express meanings of ‘force’, and in LME need starts to occur with ‘fear’ verbs. Due to the influence of need, dare acquires a number of lexical verb features.

In contrast with the closely argued analysis, the presentation of the examples is less accurate. In some dozen cases the PDE translation of a citation should be revised. Further, one citation is given twice: examples (17a) and (17b) point to the same example which has been taken from two different sources.

In ‘Counterfactuality and aktionsart: Predictors for be vs have +past participle in Middle English’, Judith Huber discusses the choice between be vs have in perfect periphrases of eight ME manner of motion verbs: climb, creep, leap, run, ride, sell, swim and walk. The data come from CMEMPV. The variation between be and have as auxiliaries in perfect constructions has been intensively studied. Huber, thoroughly familiar with the previous work on this auxiliary alternation, has chosen an underexamined area which offers rich material for the evaluation of different factors influencing auxiliary selection. She discusses a number of contexts for the increasing share of have at the expense of be: combinations with modals, past perfects, perfect infinitives, counterfactual statements, iterative and durative contexts and contexts highlighting the process of an event. What complicates the situation is that these factors often overlap.

The corpus study offers many clearly profiled findings. The share of be decreases with manner of motion verbs throughout ME. When the semantics is counterfactual or when the emphasis is on the process of moving, have is the predominant auxiliary. In contexts highlighting change of location be prevails. Finally, Huber applies mixed-effect logistic regression analysis to the non-counterfactual manner of motion verbs data. The results show that next to aktionsart, the form of the auxiliary counts as a significant predictor, and in the past tense there are higher odds for have.

The chapter by Nuria Calvo Cortes (‘Conservatism or the influence of the semantics of motion situation in the choice of perfect auxiliaries in Jane Austen's letters and novels’) focuses on the language of Jane Austen, investigating the oft-repeated claim that she was a relatively conservative language user. Like Huber, Calvo Cortes is interested in the variation between be- and have-perfects in motion verb constructions, which she studies both in Austen's novels and in her private letters. The choice of the items studied is appropriate, as Austen was writing at a time when have gradually superseded the older auxiliary be in the description of motion events. Calvo Cortes hypothesises that be might be used more often in Austen's private correspondence than in her published novels either because the older pattern caught the eye of Austen's editor or because Austen herself was mindful of the contemporary usage trend and avoided using be perfects in her published works.

Calvo Cortes’ data support her hypothesis, showing that be is indeed more often coupled with motion verbs in Austen's personal letters, although there is some verb-specific variation. In addition to the potential effect of editorial interference and audience design, Calvo Cortes shows that the semantics of the motion event could also affect the choice of the auxiliary; however, she admits that this result should be corroborated with larger datasets. Studying the precise meanings and syntactic realisations of the Figure, Ground and Path, Calvo Cortes finds that the use of have was supported by the overt expression of a physical Ground, while be was chosen more often if the Ground or the Path were not syntactically realised or if the Figure referred to something that could be carried (e.g. a letter is arrived today).

In ‘Signs of grammaticalization: Tracking the get-passive through COHA’, Sarah Schwarz traces the grammaticalisation of the get-passive in the Corpus of Historical American English (COHA). Schwarz focuses on three parameters in the grammaticalisation process: (i) situation type, (ii) subject type and (iii) the past participle used in the construction. The data are examined at forty-year intervals from the 1870s to the 1990s by drawing 200 random tokens of get- and be-passives from each decade studied. In addition to the grammatical factors affecting the choice of the passive, Schwarz studies her data in terms of register variation, examining whether the spread of the get-passive could be interpreted in terms of colloquialisation.

Schwarz shows that get-passives are increasingly used as central passives in the period studied, while be-passives show a substantial decrease. These frequency changes are also mirrored in Schwarz's data on semi-passives. From a register perspective, get-passives are initially most frequent in fiction, and their frequency increases considerably in popular magazines and newspapers in the twentieth century. Interestingly, Schwarz shows that the participles used in get-passives do not show significant overlap with those used in be-passives. Consequently, she suggests that the decline of the be-passive cannot be explained by the frequency increase of the get-passive alone. The participles with get-passives have also become more varied over time, which Schwarz takes as evidence of increased morphosyntactic generalisation and grammaticalisation.

Susanne Chrambach's study, ‘From time-before-place to place-before-time in the history of English: A corpus-based analysis of adverbial clusters’, focuses on the ordering of adverbials in ME and EModE with data from the parsed corpora of the Penn–Helsinki family. By comparing her results to an earlier study on Old English, Chrambach shows that the change in the ordering of adverbial clusters from time-before-place to place-before-time has been gradual, with the most substantial changes taking place in the final ME subperiod (1420–1500).

Chrambach investigates a number of factors that have been argued to influence adverbial ordering in English, including obligatoriness (complement vs adjunct), complexity, end-weight, information structure (given vs new) and syntactic form (adverb phrase, prepositional phrase, clause). She carries out a regression analysis of the data, identifying the most important factors that govern the ordering of adverbial clusters in ME and EModE and concluding that these factors are often interrelated. For example, time adverbials are often formally realised as single adverbs, and they are also used to organise discourse on a textual level, which makes them likely to occur cluster-initially. In OE, for example, þa typically comes before the more complex adverbials of place in adverbial clusters. Chrambach also examines the potentially confounding effect of genre and discusses the various factors affecting adverbial placement clearly and concisely.

The chapter by Ole Schützler, ‘Variation and change at the interface of syntax and semantics: Concessive clauses in American English’, provides a study of concessive adverbials headed by three conjunctions: although, though and even though. The data come from COHA, and the results reveal differences in the use of the individual connectives, on the one hand, and evidence of diachronic change, on the other. A multifactorial analysis brings further information about the factors affecting adverbial placement in American English from the late nineteenth to the late twentieth century.

Schützler shows that even though patterns differently from although and though in many respects. For example, while even though is mainly used to encode content concessives, although and though are mainly used in speech act concessives. Furthermore, the complements of even though increase substantially in length over time, while the other two connectives show little change.

Schützler's data also show that the proportion of speech act concessives increases with all the conjunctions studied, which could be taken as suggestive of ongoing subjectification. Finite clauses have also gained ground as complements at the cost of non-finite (participial or verbless) clauses. Schützler concludes by conducting a multifactorial analysis, which shows that adverbial placement (final vs non-final) is affected both by the conjunction (adverbials headed by even though are most likely to occur sentence-finally) and by complement length (longer complements favour sentence-final position). There is also a modest tendency for the concessive clauses to occur in a non-final position in the most recent dataset.

In his chapter, ‘Further explorations in the grammar of intensifier marking in Modern English’, Günter Rohdenburg compares the use of morphologically marked intensifiers (e.g. exceedingly, extremely) to unmarked ones (e.g. exceeding, extreme) in EModE and LModE. Adopting a gradient view of ‘verbality’, Rohdenburg's study complements and enriches earlier analyses according to which the -ly suffix was more likely to be used with verbal and participial than with adjectival and adverb heads in EModE. Importantly, Rohdenburg shows that intensifier marking is affected by adjective position: intensifiers used with predicative adjectives are more often marked with -ly than those modifying attributive adjectives, and the probability of morphological marking is also greater for past participles (which are almost exclusively used after be) than for present participles (which often occur prenominally). Furthermore, the presence of a complement (e.g. angry with his servants) increases the likelihood of morphological marking. All these results agree with the proposed cline of ‘verbality’, but there is one category that does not conform to expectations: manner adverbs are more often used with marked intensifiers than predicative adjectives, even though the verbality cline predicts the contrary. Rohdenburg's study offers valuable information about the history of intensifiers and the role of context in language change. The results can also be taken to provide empirical support for a gradient view of categorisation, which makes the study even more interesting from a theoretical perspective.

In the final chapter, ‘The rivalry between far from being+predicative item and its counterpart omitting the copula in Modern English’, Uwe Vosberg and Günter Rohdenburg discuss the conditions under which being is used with far from (e.g. far from being satisfied vs far from satisfied). The authors hypothesise that the occurrence of being can be explained by the Complexity Principle: the processing load of grammatical constructions correlates with explicit morphosyntactic marking. To test this hypothesis, Vosberg and Rohdenburg carefully analyse various factors that correlate with the use of being in their data, such as the form of the complement adjective (adjectives in -y like happy vs other adjectives), complement type (AP, NP, PP), complement length (mono- vs di- vs trisyllabic adjectives), and information structure (cataphoric vs anaphoric reference of the complement). The care exercised in the categorisation allows the authors to show convincingly that the use of being is generally tied to cognitively complex environments as predicted by the Complexity Principle.

Vosberg and Rohdenburg also study their data from a diachronic perspective, comparing the variation between the use of being and the zero variant in both British and American English from 1650 to the early 2000s. In general, both varieties show an increasing preference for the zero variant in the Late Modern period. The trend is initially spearheaded by British English, but American English catches up and takes the lead in the early nineteenth century.

Overall, the volume presents an interesting selection of studies that cover a wide range of topics. The individual chapters give food for thought for researchers working on different areas of English grammar and different periods of the history of English, and most of the chapters are based on widely used historical corpora, which helps the reader contextualise and evaluate the results. The volume includes contributions by both well-established scholars and early-career researchers, and while the results of some of the studies are clearly more wide-ranging than others, the overall quality of the research included in the volume is commendable. Having said that, the thematic unity of the volume is not as good as it could be, and the division of topics as outlined in the introduction seems more like a post hoc classification of the chapters than an intended outcome. Indeed, with a few exceptions the volume reads more like a collection of unrelated articles than a truly coherent whole where each chapter would contribute to a general understanding of a specific theme. This, of course, is not uncommon in our field, and it does not take anything away from the fact that, on the whole, the volume provides an enjoyable and interesting read.