‘Bad data’ is one of the methodological difficulties that confront the practitioner of historical (socio)linguistics. This is due to the fragmentary nature of the available material – mainly written, formal texts (which excludes spoken and colloquial language) by members of privileged, literate communities (which, for most historical periods, excludes females and lower-class speakers) and isolated from social and communicative contexts ‒ so that many registers, styles and the socio-demographic background of informants are absent (Hernández-Campoy & Schilling Reference Hernández-Campoy, Schilling, Hernández-Campoy and Camilo Conde-Silvestre2012: 65‒73). For obvious reasons, this is in a direct proportional relationship to the impending weight of time depth: the further back in time the object of analysis, the greater the bad-data problem. Thus, sociolinguistically oriented research on the history of English has privileged the early and late modern periods, with a relative engagement with Middle English texts and a general absence of studies on Old English (henceforth OE). This volume by Olga Timofeeva ‒ Professor of English historical linguistics at the University of Zurich ‒ intends to revert this state of affairs. And she does so, first, by reviewing previous ‘Sociolinguistic approaches to the study of Old English’ (chapter 1, pp. 1–26): from sociologically oriented analyses of language contact and diglossia (OE–Latin–Celtic–Old Norse), or studies on the development of normative systems, especially in connection with supracultural religious movements like the late tenth-century Benedictine reform, to some attempts at correlational approaches, in the social variationist tradition, like Thomas Toon's (Reference Toon1983) pioneer analysis of the lexical diffusion of vowel changes from text (glosses) in politically dominant late eighth-century Mercia, to early ninth-century charters in dominated Kent, or Ursula Lenker's (Reference Lenker2000) correlation of late West Saxon standardised vocabulary with the typically strong, close-knit networks (a norm-enforcing mechanism) in which monastic communities at Winchester were organised in late Anglo-Saxon England.

The negative overtones in the bad-data tag have lately been exposed and replaced with the possibilistic reliance on ‘informational maximalism’. The objective is to make use of ‘all reasonable means to extend our knowledge of what might have been going on in the past, even though it is not directly observable … gain[ing] a maximum of information from a maximum of potential sources’ (Janda & Joseph Reference Janda, Joseph, Joseph and Janda2003: 37). In this vein, the absence of direct socio-demographic information is compensated by the close analysis of non-canonical texts from different genres, registers and text types which have been made available by the recent spectacular developments of corpus linguistics. Timofeeva implicitly adheres to this methodological tenet when she describes the volume as ‘essentially about … smaller units: genres and registers of OE in relation to the communities that used them’ (p. 4). In parallel with this aim, she describes in detail the sociolinguistic potential of the categories available in the OE surviving corpus: inscriptions (runic and non-runic), glosses, glossaries, verse and prose. As regards corpus linguistics, digital collections of OE texts are fundamental for Timofeeva's project. They range from the all-encompassing Dictionary of Old English Web Corpus, containing at least a copy of every surviving OE canonical text, to a number of specialised corpora, some of them still in progress, covering mainly administrative and legal documents: the Anglo-Saxon Charters project (ASChart) ‒ an XML edition of charters written in England before AD 900 ‒ which sensibly expands Peter Sawyer's classical Anglo-Saxon Charters: An Annotated List and Bibliography (Reference Sawyer1968), now available electronically (The Electronic Sawyer: Online Catalogue of Anglo-Saxon Charters), as well as Documents of Early England Data Set (DEEDS), comprising also medieval Latin records. These databases are available online and allow searches based on different parameters ‒ region, date, archive, charter typology, internal structure ‒ including the retrieval of the names of the real people involved. Personal information like this can be checked against the Prosopography of Anglo-Saxon England (PASE): a fundamental database containing the reconstructed profile of the individuals mentioned in OE and Latin texts from 597 to 1100. In this way, even though the fragmentary nature of linguistic and extralinguistic data remains, researchers can now ‘make the best of it’ and reach the kind of ‘informational maximalism’ that makes historical sociolinguistic research feasible.

It must be admitted that first-wave analyses of OE linguistic variation ‒ quantitative studies involving essentialist extralinguistic categories like age, class or gender (Eckert Reference Eckert2012) ‒ are generally doomed to fail because these variables are often impossible to reconstruct, even with the help of PASE. But it is also true that the external variables that correlate with sociolinguistic processes in OE usually transcend the individual and are punctuated by ‘major developments in the political and religious spheres, as the most prolific periods coincide with reform movements and increased involvement of the political and cultural elites in the vernacular (as well as Latin) written culture’ (pp. 19‒20). Despite the anonymity of many of the participants in these cultural communal events ‒ like the revival of learning undertaken by King Alfred (871‒99) or the Benedictine reform promoted by bishop Æthelwold of Winchester (963‒84) and others ‒ relevant constructs within the second and third waves can be reconstructed with a focus on ethnographic and anthropological macrosociological categories, like social networks, discourse communities (henceforth DC) and communities of practice (henceforth CoP). Indeed, social networks have proved to be useful for the interpretation of historical linguistic variation and change, even if their individual participants are not identified and ego-based networks cannot be reconstructed: as long as groups and the circumstances favouring the establishment of different links and ties are envisaged, the construct becomes useful. As regards the third wave, Timofeeva (Reference Timofeeva, Kopaczyk and Jucker2013) is one of the pioneers in its application to the history of early English and this volume tackles the social meanings of linguistic variation from this perspective, by exploring both larger groups, like the Anglo-Saxon clergy in charge of state administration (a likely DC), and smaller ones, like monks engaged in joint monastic enterprises and sharing the resources to achieve them (possibly a CoP).

Chapter 2, on ‘Social networks at the court of King Alfred’ (pp. 27–48), describes synchronically the circle of scholars involved in King Alfred's cultural enterprise in the late ninth century: a social network of both foreign and Mercian people with close, although intermittent, ties, participating in the organisation of the West-Saxon proto-Chancery, established at Winchester in charge of the royal charters, as well as launching the Anglo-Saxon Chronicle project. This body of civil servants avant la lettre could sociolinguistically be defined as a CoP in view of their daily contacts, their involvement in joint enterprises and their use of shared norms. Interaction between its members may have resulted in processes of micro-accommodation leading to focusing and the formation of ‘a mixed and possibly somewhat levelled language that was adopted as a new norm …, suitable for the political and cultural agenda of the kingdom’ (p. 33). Diachronically, once this group of scholars disseminated, the joint enterprise and the shared norms became diffused and the CoP dissolved into a larger DC with extensive influence all over the realm. In the second part of this chapter, Timofeeva explores the diachronic spread of some of the lexical norms produced by the CoP into the DC. By concentrating on the semantic changes affecting the words Angelcynn (from ‘Anglia’ into ‘English people’) and here (from ‘band’ into ‘criminal troop’) in diverse texts over time, like the Laws of Alfred, the Anglo-Saxon Chronicle, ninth-century charters and later Ælfrician documents, she examines the progressive appropriation of the new meanings by the cultural elites to express acts of political identity: the in-group belonging with West-Saxon legitimate rule versus the out-group of Scandinavian invaders.

Acts of identity among the Anglo-Saxon social elite, both lay and religious, were most clearly embedded in written legal documents transmitting the social meanings of sovereignty, power and property relations. In chapter 3, ‘Legal Old English and its communities’ (pp. 49–64), Timofeeva highlights the relevance of the Church as a DC participating in all stages of legal compilation in writing, in the form of both laws promoted by the royal power and charters that recorded gifts of land and privileges, most often to monastic establishments themselves. Charters are considered especially valuable for historical sociolinguistic research ‒ ‘proxy genres’ as the author calls them (p. 52) ‒ not only because they demanded an intense participation of members of the Church, as ‘scribes who wrote the laws down, witnesses who legitimated their enunciation and archives where … documentation was preserved’ (p. 51), but also in view of their reliance on oral practices, through the public announcement of their contents, and of the internal references to participants, including producers, addressees and lists of witnesses. In fact, the following chapters in Timofeeva's book focus on charters and a full introductory typology of the text types within this genre is extensively discussed here, including diplomas, writs, writ-charters, charters proper and wills.

Chapter 4, ‘Diplomas: Members of the assemblies and variation by archive’ (pp. 65–86), focuses on diplomas: records of a transfer of land, property or rights, mainly written in Latin. Timofeeva traces their composition to the royal writing office (proto-Chancery), already described as a CoP in chapter 2; she emphasises the role of their members, not only in drafting the written documents, but also in announcing them orally at meetings of the royal assembly and in the acts of witnessing them. Incidentally, these meetings reinforced the face-to-face interaction between the members of the CoP. The author is also attentive to local and regional differences between Wessex, Kent and Mercia, especially as regards the composition of witness lists which give clues to the structure of the social networks involved in the production and promulgation of diplomas and their respective influence: local, kingdom-internal or kingdom-external. Regional differences correlate sociolinguistically with the range of finite forms of the Latin dispositive verbs found in diplomas from each of the three areas, whose use can be related to the structure (close-knit vs loose-knit) of groups of professional scribes at local, monastic or cathedral scriptoria.

Unlike Latin diplomas, writs had mainly a performative function: written for the most part in English, they were issued by a person in power ‒ king, ealdorman, archbishop, bishop ‒ who required the addressees (private individuals in a subordinate position, but also institutions like royal offices at shires and local courts) to take (or not to take) some course of action. In chapter 5, ‘Writs: Conservatism and linguistic variation’ (pp. 87–108), the pragmatic function of some formulaic elements in this text type, like the salutation-notification template, are closely analysed. Special attention is given to the diffusion of these standardised features out of the proto-Chancery, as described in chapter 2, into other texts like the Preface to the Pastoral Care at King Alfred's court (c. 890–7), Ælfric's late tenth-century letters to lay correspondents, and in royal-writs or proclamations issued at the courts of Cnut (1020 and 1027), Edward the Confessor (c. 1060–5) and Harold II (1066). Chronologically, variation in some adverbial formulas is shown to correlate sociolinguistically with the status of senders and addressees: freondlice (‘in a friendly manner’) and luflice (‘lovingly’) are used in letters sent from higher to lower ranks, and eadmodlice (‘in a humble manner’) in the other direction.

Wills are dealt with in chapter 6, ‘Wills: Variation by archive and gender’ (pp. 109–40). Despite their scarcity in proper OE manuscripts and their preservation in later medieval cartularies ‒ liable to suspicion of forgery (incidentally, another source for the indeterminacy of the available data) ‒ wills are an outstanding text category with possibilities of sociolinguistic analysis, both because many of them were commissioned by women and on account of their obvious connection to orality, through their spoken performance and the act of witnessing they involved. The extant corpus of OE wills is submitted to a threefold study. In the first place, attention is given to variation in their dispositive construction by subperiod, before and after 900, and archive: Christ-Church, Canterbury, Winchester, Bury, etc. Several linguistic variables are examined: the alternation between the first and the third persons (as a remnant of the oral stages of composition), the dispositive verbs used ‒ with the gradual standardisation of (ge)unnan (‘grant’) possibly due to the influence of the proto-Chancery in Winchester after 900 ‒ as well as the prepositional phrase expressing the testament's fulfilment: ‘æfter POSS dæge DAT’ (‘after my day’), based on the traditional Latin-based construction ‘post-diem POSS’, and the innovation ‘ofer POSS dæg ACC’, derived from spoken formulas and widely used in wills by female donors. As such, this reminds us of present-day sociolinguistic changes-from-below and of the active leadership of women in their diffusion, through the likely weak ties that they could have established with both ‘the professional members (scribes) and the speech community at large’ (p. 139). In the second case study, Timofeeva adopts a socio-pragmatic methodology, correlating variation in different sections of wills with gender as the independent variable. For instance, in soliciting royal (or other patrons’) protection, females tended to use more elaborate mitigation and politeness strategies than males, possibly reflecting the necessity to defend themselves from ‘kindred pressure and predation, particularly from male kin, as land was typically passed in the male line’ (p. 126). In the same vein, females also used stronger and more elaborated and emotional language than males in the curses and maledictions within the anathema that was added at the end of wills to ensure their enforcement, again hinting at social and economic vulnerability. Finally, the author tackles two special wills, whose authors behaved as sociolinguistic outliers when, for different reasons, they downplayed the standard structure of the text type, avoiding some of its formulaic constructions.

Chapter 7, ‘Mixed-language practice of William I's chancery: Contact and innovation’ (pp. 141–74), turns again to Latin charters, now in the post-conquest period. In particular, Timofeeva focuses on the extant legal documents issued in England between 1070 and 1087, when Latin had replaced English at the royal chancery. Sociolinguistically, the CoP of royal clerks and the wider legal DCs in England now became multilingual institutions with language contact leading to extensive borrowing of loan terminology in professional, specialised fields. The new terms incorporated into chancery Anglo-Latin came from both OE and Norman French. The former refer mainly to franchises, the rights and privileges of religious houses, as well as to administrative and land units. Timofeeva briefly engages in the risky business of pigeonholing the status of these items as either code-switches or established loanwords, pondering aspects like morphological integration, frequency of use and analysability in the target language as well as the necessity of culture-specific terms. In the end, no clear solution to the terminological puzzle is envisaged and, in this way, Timofeeva adheres to current trends that avoid applying the code-switch label to one-word items (Poplack Reference Poplack2018); her decision also points to the possible interpretation of these franchise terms as mixed-language vocabulary which scribes used invariably in English, Latin or French, just like the abbreviations in late Middle English business texts studied by Laura Wright (Reference Wright and Kastovsky1994; Reference Wright and Trotter2000). The author also explores possible ways for the diffusion of Anglo-Norman vocabulary via Anglo-Latin documents into English, mainly occupational terms that filled gaps in the target language; in this context, she highlights the role of oral court sessions where familiarity with different codes must have been the norm. In a way, it was the pragmatic attitude of Anglo-Latin scribes which fostered the transfer. They had to generate technical vocabulary that would have been understood in exactly the same way by all parties involved: the beneficiaries of charters, the clerks who drafted them, the witnesses of transactions and the members of local-courts. Thus, terminology that could be easily translated into equivalent Latin formulas and direct borrowings was privileged. The process must necessarily have been connected to the promotion of a closed-knit community of administrators, following the appointment of Lanfranc as Archbishop of Canterbury in 1070, with a similar age and educational background (many of them were trained at Bec Abbey in Normandy), which settled the normative quality of the new Latin linguistic practices and spread loanwords through their weak ties to members of other DCs throughout the country.

In a closing ‘Epilogue’ (chapter 8, pp. 175–8), the main methodological innovations and findings are highlighted. Especially relevant are, on the one hand, the potential of surviving administrative documents for the sociolinguistic analysis of OE variation, in so far as they include personal information and can be accurately dated, and, on the other, the necessity to transcend the first-wave approach, based on individual informants, to focus on social networks, discourse communities and communities of practice, within the second and third waves. This unveils the existence of close connections between communal linguistic processes observed in legal and political registers and text types ‒ focusing and normative convergence, diffusion of innovations, especially in writ templates, and in the dispositive verbs of diplomas and wills ‒ with political and cultural changes, like centralisation during King Alfred's reign or after the establishment of Norman rule. Notwithstanding some individual contributions like Ælfric's, it is the recourse to social networks and third-wave constructs ‒ CoPs synchronically and DCs diachronically ‒ that has helped Timofeeva obtain relevant results, showing the feasibility of the historical sociolinguistic analysis of OE variation.