Atlas Introduction Index of Sources

ANGUS McINTOSH, University of Edinburgh
M.L. SAMUELS, University of Glasgow
MICHAEL BENSKIN, University of Oslo
with the assistance of Margaret Laing and Keith Williamson
University of Edinburgh

INTRODUCTION TO INDEX OF SOURCES

1. Introductory

The Index of Sources includes brief notices of all the writings in Middle English known to the present authors and which have proved relevant to the compilation of the Atlas. The word ‘writings’ rather than ‘manuscripts’ is used here, for two reasons. Firstly, a single manuscript may contain linguistically diverse sections, whether the work of several different hands or of just one. Since the focus of inquiry is here the language rather than the codex, a single manuscript may require several independent notices. Secondly, many writings, albeit a small proportion of the whole, are known now only from variously antique printed reports, or from post-mediaeval transcripts: the original manuscripts, whether or not they survive, are lost to sight.

It should be noted that the Index is not a complete inventory even of the material known to us, let alone an attempt to list all the extant writings in later Middle English. The circumstances in which the Atlas was compiled precluded the sustained and systematic bibliographical inquiries that might be expected as its prelude: we are aware of the piecemeal character of the work that has been done, of the continuing interaction between the results of mapping and the ensuing selective searches for additional documentation. Had it been known at the outset that the project would take nearly thirty-five years to complete, then the investment represented by assiduous scrutiny of the volumes of the Historical Manuscripts Commission and the National Register of Archives, and the following of every lead they offered, might reasonably have been proposed; yet no less reasonably may it be observed that the outcome of such labour could as well have been an enviable list of manuscripts containing Middle English, but no Atlas founded upon them. Many texts have been brought to our notice by intending editors who wished to know something of the linguistic provenance of their manuscripts. From time to time such materials have proved to be highly relevant for our own purposes; some, indeed, even at a late stage of compilation, have occasioned local revision and incorporation in the Atlas maps themselves.

Were Middle English as well served by catalogues as is Old English, these remarks would be unnecessary. The list of manuscripts and editions in the Bibliography (1954) for the Ann Arbor Middle English Dictionary was a valuable aid during the early stages of the Atlas project, and it has since been updated (1984); but there is as yet nothing to compare with Ker’s Catalogue of Manuscripts Containing Anglo-Saxon, and the Atlas Index will inevitably serve as a partial substitute. Whatever its shortcomings, the Atlas Index is the largest and most comprehensive list of manuscripts containing Middle English yet published.

The Index is printed in two formats. In the first, the order is the familiar one of repository and manuscript. (Writings known from printed reports only are listed in a separate section of the repository index.) In the second version, the same notices appear, but they are ordered by county or otherwise specified region of origin for the language of the writings to which they refer. It is here possible to find, in a single section of the list, all of the writings entered in the Repository List which, on the basis of their dialectal characteristics or other evidence, can be associated, tentatively or otherwise, with some given county or otherwise specified region of the British Isles. The following account treats firstly the scope and contents of the Index, secondly the organisation of the repository and county lists. The first part is extensive, and is intended mainly as a general introduction to the local documents. The second is brief and utilitarian, a summary key to the arrangement of the Index proper.

2. Scope and content of the Index

It is convenient to distinguish ‘literary texts’ from documents. By ‘documents’ we mean legal instruments, administrative writings, and personal letters: the type of material that is calendared by historians, likely to be of known date and local origins. The ‘literary texts’ include imaginative and discursive writings, regardless of their quality as literature: so, for example, Piers Plowman beside The Prick of Conscience, Castleford’s Chronicle and Dan Michel’s Ayenbite. Less obviously, perhaps, they number such things as translations of the Bible, glosses, paraphrases of the Psalms, medical recipes, charms. For a given text, there may be many different manuscripts to be considered, or just one: the anonymous Prick of Conscience is known from 117 copies, the Ayenbite of Inwyt only from Dan Michel’s holograph. The two categories, of course, documents and literary sources, in some cases overlap. Robert Pilkington’s account of his conflict with the Ainsworths for the possession of Mellor (see North Yorkshire County Record Office ZDV XI) is a discursive chronicle no less than it is an administrative record; commonplace books typically combine texts of a documentary sort, especially personal and financial memoranda, with literary works, often devotional or medical. The distinction between ‘document’ and ‘literary text’ is not classification for its own sake; the material falls naturally into two classes, and they call for different kinds of treatment.

For present purposes, the main difficulties presented by literary manuscripts relate to the linguistic integrity of copies at perhaps many removes from an original composition, and to the means by which such unlocalised material can usefully be incorporated in a linguistic atlas. These matters are complex, and are treated at length in (respectively) chapters 3 and 2 of the General Introduction in this volume. The notices found in the present Index are by contrast fairly straightforward, and require little further comment. The mere description of documents, however, involves terminology and considerations of legal process seldom found in philological works. The documents of interest here, moreover, usually refer to named places; and central to the inquiry is whether a given document is a genuinely local product, its language reflecting that of the place to which, ostensibly, it belongs. Such associations cannot well be evaluated without reference to the legal and historical character of the document in question––in short, without knowledge of the diplomatic. There is further complication in the emergence of a national written standard. Over England as a whole, it spread much more rapidly in legal and administrative writings than in literary works; long before printing began to standardise the literary language, a genuinely local legal document might be anything but local in its forms. These matters, language and diplomatic, are closely connected with one another, and it has seemed best to treat them all in the same volume. Because the notices in the Index will themselves cause readers to refer to the account, the Index volume is probably the most useful place for it. The notices should in turn provide convenient illustration for some of the broad themes of the account.

3. Literary manuscripts

Manuscripts of the period ca. 1350 to ca. 1450 are included as a matter of course if they are written in a form of language that can be localised, and were known to the compilers by early 1984. The rationale for the historical limits of survey is discussed in the General Introduction above, 1.1.1–2, p. 3.

What counts as ‘a form of language that can be localised’ is of course debatable. By about 1960, the existence of a Central Midland regional standard had been recognised; it was localised in the sense that an area of origin could be defined–––‘the Central Midland counties, especially Northamptonshire, Huntingdonshire and Bedfordshire’ (Samuels 1963, pp. 84–7). The language of a given manuscript in Central Midland Standard, however, was another matter: local origins within the Urheimat of the standard could not be more closely defined, and it appeared in any case that the standard was adopted by writers from areas far beyond the Central Midlands. In both senses, writings in such language could well be considered unlocalised: the standard was in fairly widespread use, and no one place within the central Midlands could be called its home. These conclusions rested on analysis of a large number of manuscripts, and it appeared that little would be contributed to an understanding of the regional pattern by investigating yet more material of the same type. Accordingly, such writings were commonly passed over in the work that followed. As a census of literary manuscripts, therefore, the Index is distorted; even allowing that it is a very substantial sample of the total surviving corpus, its contents are biased by considerations of language.

Additionally, a good many items have been included on the grounds that, even though their language is not properly localisable, something can be said about its dialectal character; and there are many other notices that can be justified only as contributions to some future catalogue of manuscripts containing Middle English. The aim throughout has been to make accessible such information as may be useful.

4. Documents

Documents written in markedly local language, and dating from about 1350 until the end of the reign of Henry VI (1460/1), have been entered as a matter of course. In this category, an omission reflects oversight, or ignorance of the document’s existence. Additionally, much material in language of partly or even predominantly standard type has been included for this period; the linguistic character of documents in this class has been duly indicated in their descriptions. For the northern counties of England, non-standard material to ca. 1500 has regularly been noted; some items, in language virtually uninfluenced by the written standard, have been included from a decade or so later. Such Middle Scots material as is noticed provides only a necessary minimum for comparative purposes; but to the present sample of fourteenth- and fifteenth-century documents, one or two from the early sixteenth century have been added; cartographic expedience has been the chief arbiter for their inclusion. The Mediaeval Hiberno-English material, by contrast, is fairly comprehensively recorded. All of the known documents surviving in manuscripts from before 1500 are listed, with the exception of those written in Standard English. The earlier sixteenth-century documents, save those in strongly Hibernicised language, are not systematically recorded, and only exceptionally are items in hands later than 1540 noticed; they call for separate treatment elsewhere.

The limits for the main period of survey, ca. 1350 to 1460/1, should be further considered. From the point of view of diplomatic and the use of English for legal and administrative purposes, both the earlier limit and the later are arbitrary. In practice, the earlier poses no difficulties, because documents in fourteenth-century English are very rare until the last quarter; even then they are by no means numerous. There is only one marginal exclusion entailed by starting the inventory at 1350, namely the petition of John Drayton and his wife to Edward III, 1344 (PRO Ancient Petitions 192/9580; Chambers and Daunt 1931, p. 272); this has been added to the list. Documents in English are hardly commonplace until the second quarter of the fifteenth century. Even then, with the exception of certain classes relating to the Court of Chancery, legal and administrative writings continued for the most part to be in Latin. Preoccupation with the status of English relative to French during this period, and with the content of such legislation as the Statute of Pleading of 1362, has tended to obscure the real state of affairs; histories of the language imply too readily that English had become the regular language of law by this time, and so of the written instruments (cf. Legge 1941, p. 167). Extensive searches through archives in many different parts of England suggest that, in natural accumulations of deeds attaching to property, English writings rarely account for as much as one per cent of the material from before 1450. The figure is impressionistic, but it may serve to establish a perspective for the present inventory, lengthy though this be.

The later limit, 1460/1, coincides with the end of a long reign, that of Henry VI, and also with a change of dynasty. Although, obviously, no one date can be set beyond which it is pointless to seek for documents in local language, in so far as a general limit has to be declared, 1460/1 is a reasonable cut-off. Since documents are normally dated by regnal year (see below), it has also a certain practical convenience. From the point of view of the non-peripheral southern and Midland counties, however, it is excessively generous: in these areas, the spread of standard English for legal and administrative purposes was much more rapid than elsewhere, and relatively few of the documents even from the second quarter of the fifteenth century are of prime value for dialect mapping. In part this reflects the contemporary social geography, proximity to London and to the centre of government. Yet the linguistic character of the emerging written standard must itself be considered: it was much less of a foreign language to people from these parts than to writers from the northern counties or, say, Norfolk. The practical result is that texts in semi-standardised language of the Midlands and the south are much less informative for dialect mapping than are their northerly equivalents. The local elements in a semi-standardised northern text are for the most part distinguishable just because northern language is in general so unlike that of London. By contrast, in a text from, say, Northampton, having a proportion of adopted forms no higher than in the northern text, the distinctively local elements are much more elusive: the linguistic overlap between the two contributory sources leaves a relatively large component of indeterminate origins.

The search for local documents has accordingly been less intensive in the southern area of survey, for which semi-standard writings have been recorded much less assiduously, and only exceptionally after the reign of Henry VI. It is unfortunate that in the north of England local record offices were developed generally later than in the south, and access to some of the most important northern material could therefore be gained only after the Atlas project was already far advanced. It has not been possible to examine all of the archives which may be expected to contain documents in Middle English, and there are unlisted collections, not yet open to the public, even in archives that have been searched. The Atlas lists had to be closed in early 1984, but continued inquiry has already brought additional material to light; in due time, a supplement will of course be necessary.

No attempt has been made, beyond what is expedient, to catalogue the individual documents of large published collections. Here, the Index is a guide to the sources, but not an enumeration. So, for example, reference to Professor Norman Davis’s edition of the Paston letters (Davis 1971 and 1976) has disposed, in a few brief entries, of what would otherwise have been a monumental cataloguing task in itself. It has been thought sufficient merely to note the existence of most of this material, regardless of the time that has been spent on analysis and assessment of its suitability for dialect-mapping. Not all of the items that appear in the Mittelenglische Originalurkunden of Morsbach (1923) and Flasdieck (1926) are individually recorded; the material is sufficiently accessible, and notice has been confined, in the main, to documents in local language. Likewise, though for different reasons, there are omissions from Chambers and Daunt’s Book of London English (1931); it has been thought sufficient to provide entries for texts of special concern, and let the general reference suffice for the rest.

In the early stages of this work, the collection of documentary material depended almost entirely on published items: texts printed in county histories, in local historical and archaeological journals, in calendars of documents, among the records of municipal corporations, and so on. The fidelity of such printed versions to the original documents is, as may be expected, very variable. This is not just a matter of date of publication: some of the best and most careful work is to be found in editions that by now are themselves antique, and some of the least satisfactory efforts are modern. The shortcomings of ‘éditions de textes données par des historiens qui ne sont pas philologues’ are of course well-known (Legge 1941, pp. 163–4); but even where the substantive accuracy of the text is not in doubt, it is rare for a historian to have reported the original in such detail as to provide an adequate substitute for philological purposes (cf. Benskin 1977, pp. 505–7). If documents known from printed versions were to be relied upon for the compilation of a linguistic atlas, therefore, it was in most cases essential that the originals be traced.

For the muniments of civic and ecclesiastical corporations, it is in general not difficult to run such originals to earth; re-classification, and occasional losses, are the only familiar hazards. The great bulk of the material published in the later nineteenth and earlier twentieth centuries, reports originals that were then in private hands; and since then, the custody of such private muniments has changed considerably. Merely to reflect upon the development of local record offices and the character of their contents is to become aware of far-reaching changes in custody, public access, and ownership. These changes are publicly recorded, in the annual National Register of Archives Accessions, and in the Guide to the Location of Collections described in the Reports and Calendars Series, 1870–1880 issued by the Royal Commission on Historical Manuscripts in 1982 (too late, unfortunately, to benefit the present work). The private collections of antiquarians, by contrast, and private disposals of natural accumulations, are another matter: documents known from printed works do from time to time emerge in the sale catalogues, but probably most transfers of this kind are unpublicised. Even when notice of sale is given, individual items may not be catalogued (cf. Benskin 1982b, pp. 27–8); subsequent ownership need not, of course, be divulged. It is hence no surprise that the originals of a good many documents noticed in the Index remain untraced.

As may be expected, the search for new material in the archives themselves has brought to light a good many items known previously from print, often by accident. It is not always easy to recognise such documents. Occasionally, a printed version is so inaccurate that identifying it with even a purported original has given pause. Moreover, material of a sort only marginally useful for the present work has in general been scrutinised less thoroughly than documents of real linguistic value, and in this class especially there have no doubt been bibliographical oversights. In the course of compiling the Index, it emerged that several items had been unwittingly entered twice, once as a document in a repository, once as a printed source. There may well be other such identities that have been missed and the compilers would be duly grateful for notice of them. It is here intended that, wherever possible, documents should be entered by repository only, with references to printed versions if known; the Index is not designed to be a finding list for present locations, even though the Repository List will serve largely to that end.

When linguistic integrity and local origins are at issue, descriptions like ‘Document relating to N–––town, 24 June 1428’ are of little value: Does the document concern property at N–––town? Or persons belonging to N––– town? Was it written, or sealed, or dated, at N–––town? If the document is a deed, were any or all of the parties to it persons of N–––town? Was the writer himself a man from N–––town? There is no prima facie reason for supposing that the answer to any one of these questions need be the same as the answer to the next. Men travelled, and so did their language, including the emerging written standard: it is always possible that the language of a document does not belong to the place with which, on all other counts, the document itself is firmly associated. When, however, the language of a document so placed conforms to general expectations for the area in question, then there is at least a reasonable basis for regarding that place, rather than some other, as the dialectal locus. A coherent dialectal pattern, worked out in detail from a corpus of documents no less firmly localised, may of course indicate that the local origins of the document and its language do not precisely tally, even though they are fairly close; but that is a matter of fine-tuning––of detailed adjustment to a placing that is already right in its essentials, and does not affect the overall reconstruction.

Accordingly, the summary notices in the Index include whatever information bearing on local origins was available to the compilers; and for documents which have played any significant part in the construction of the maps, the effort to obtain such information has not been spared. It is a matter for regret that there remain many notices of the ‘Document relating to –––’ type, relics from the early days of the Atlas project before the relevance of diplomatic and documentary form was fully understood. It has seemed better to include them, even though some may turn out to be products of counties other than those with which they are presently associated, rather than fail to record their existence altogether. The descriptive formula itself will serve as sufficient warning.

As to the content and legal form of documents, it should be noted that general problems of description, quite apart from the multiple authorship of the Index, make consistency of usage hard if not impossible to attain. For example, a deed may declare itself to be a grant, but its effect may be that of a gift: should its opening formula determine the description, or should the record be of its legal effect? It is not necessary to dwell on such problems here, beyond noting that they are real; a succinct and revealing account will be found in R.B. Pugh’s introduction to the Calendar of Antrobus Deeds before 1625 (Wiltshire Archaeological Society, 1947). In general, the aim has been to indicate the physical character of a document (indenture, deed poll, roll, etc.), and then its legal form. The principal types, except for those generally familiar or whose names are self-explanatory, are described in the next section. Fine distinctions have not been attempted. Where place-names appear, they are duly cited, usually in their documentary form, and the modern identifications given if possible. Dates are usually given in their mediaeval form (see Dates of documents below). Seals are not recorded: the reasons are set out as Documents: seals below. When a document is a contract, the parties to the contract are distinguished as (i), (ii) and, if need be, (iii): in each party there may be several persons, and these are named individually (except that long lists may be cut short with ‘et al.’).

Dates are reported as given in the documents themselves, though in abbreviated form: MS ‘þe xiiij day of þe moneth of Nouember In þe yere of þe reigne of kyng Edward þe fourth after þe conquest of yngland þe xvij yere’ is rendered ‘14 Nov 17 Edward IV’. In English documents, years are normally counted from the accession of the reigning monarch, not from the birth of Christ. Kings are numbered from the Norman Conquest: the Edward named in the example above is the fourth king of that name since the conquest of England; no account is taken of the several kings called Edward who reigned in Anglo-Saxon England. For conversion to dating anno domini, a table of accession-dates is necessary, and for convenience a full list of regnal years is provided. Scottish documents are regularly dated anno domini; no list of Scottish regnal years is therefore required. Hiberno-English documents follow the English style, as also do English documents from Wales.

Days are commonly identified as feasts in the liturgical calendar, instead of by counting from the beginning of the month in which they fall. So, for example, ‘the Feast of the Decollation of St John the Baptist’, or ‘Saturday, the Feast of St Nicholas’. Reckoning to or from the nearest feast-day produces formulations like ‘Thursday next before the Feast of St Michael’, and ‘the fourth Friday next before the Feast of the Nativity of Our Lord’. Only rarely are these converted in the Index to day-of-the-month equivalences, and in general recourse must be had to the standard reference works. (See, e.g. Cheney 1947.)

The following notes are intended as no more than a guide for those unused to working with documents, among whom may perhaps be numbered the majority of students of mediaeval English language; archivists and mediaeval historians will find them merely elementary.

CARTULARY. A cartulary is a special form of register, a collection of those deeds (charters, c(h)artae) which relate to its possessor’s rights in landed property, whether of ownership or otherwise. By the fifteenth century, the title to a single property might well rest on an accumulation of deeds spanning three hundred years or more; for the great landowner, whether secular lord or ecclesiastical or other corporation, the cartulary is an organised reference work that eases administration. The original documents themselves may be incorporated, so that the cartulary is an album in whole or in part. Usually, however, the texts of the deeds are copied into the volume as a single act of compilation, after which the original documents were returned to the archives. Cartulary copies, it should be noted, had no status as legal records in themselves; title depended on the original documents from which they were transcribed. Accordingly, the very large proportion of cartulary texts for which no originals are now extant reflects hazard rather than deliberate destruction. From this, some further inferences about survival rates are possible (cf. INDENTURE).

For linguistic purposes, cartulary copies of deeds are subject to the same reservations as apply to other register copies, and to enrolments.

DEED POLL. ‘Deed poll’, like ‘indenture’, describes the physical form of a document, not the text that it contains. It is a deed on a single sheet trimmed straight––‘polled’ ––at the top. The form is appropriate for a deed when no counterpart or certified copy is required, as is commonly the case with depositions (affidavits), declarations of record, and wills.

Some few examples of indentures tripartite appear. These are normally recognised either by the opening formula ‘This indenture tripartite bears witness’, or by their distinctive physical form.

REGISTER. Registers are books of formal record, in which the separate entries are typically copies of independent documents (e.g. letters, wills, deeds), or fair copies made from drafts. Register copies, like other mediaeval copies of vernacular texts, need not be expected to preserve the language of their originals: the wills in the York probate registers at the Borthwick Institute, for example, cannot be trusted to represent the language of the testators of their localities, but in varying degrees attest to the habitual usages of the registrars (who may or may not be of York itself).

ROLLS. Rolls consist of separate membranes of vellum or parchment (rarely, sheets of paper) sewn together, usually end-to-end; their combined length may be anything from eighteen inches or so to forty feet or more. References to the text of a roll are by membrane (abbreviated ‘m.’, plural ‘mm.’); the sides of a membrane are distinguished as recto (‘r’) and dorso (‘d’). The indefinite extensibility of a roll makes it especially suitable for records of unpredictable length; it is the regular documentary form for the records of parliament (one roll for each separate parliament), and is commonly used for the cumulative records of the proceedings of other courts and councils.

The text of a roll may be continuous, or it may consist of related but independent items. Some rolls contain what look like the first writings of the texts in question, even rough drafts; more usually, however, they contain copies of other documents, or fair copies made from notes and drafts. Copies––enrolments––must be recognised for what they are: their local origins and linguistic form need have no bearing on those of the items which they report.

AWARD (arbitration). The settlement of disputes in mediaeval England was commonly effected not by the courts, but by privately agreed arbitrators or umpires (ME noumpere, OFr non per). Their written judgements were delivered in the form of an indenture, part to be retained by each of the parties to the dispute. Such written judgement is an award. Compliance with the award might be secured by making one or both of the disputants party to a bond, an instrument by which a party is bound to perform or refrain from specified actions, on penalty of forfeit (cf. DEFEASANCE OF A BOND below).

DEFEASANCE OF A BOND. A bond is an instrument whereby a party is bound to do certain things, or bound not to do them. The defeasance of a bond is an attached deed made at the same time, which sets out the conditions under which the bond is defeated, that is, made void. Conditions of obligation amount to much the same thing; though the English examples are generally additions to the Latin text of the bond itself, written beneath it or on the dorse, rather than separate attachments. Their form is summary: ‘Condition of this obligation is such that if N––– bound within do... [whatever is required], then this obligation is void; otherwise it is to apply with full force’. The two expressions, ‘defeasance of a bond’ and ‘condition of obligation’, are virtually interchangeable in the Index; but where the latter appears, it can be assumed to reflect the wording of the original.

ENFEOFFMENT TO USE. An enfeoffment to use is an instrument whereby property is transferred not directly to an intended beneficiary, but to a third party as trustee. It was a device for evading the burden of fines and feudal incidents attaching to landed property; it worked by establishing a fictional separation of the legal owner from the beneficiary. The trustee became the legal owner: in mediaeval language, he was the feoffee. He was enfeoffed in the property to the use of the intended real beneficiary, the cestui que use; the use is the revenue or other benefit that derived from the property. The avoidance of, in effect, death duties, was one of the main advantages; but crucially, an enfeoffment to use enabled an owner of landed property to leave it, after his death, to someone outside the feudal succession.

As feoffees––there was commonly more than one in such an arrangement––priests were a favoured class. They were celibate, and so could be presumed (though not always wisely) to lack ambition for worldly goods that might descend to progeny of their own. Trustees, whether priests or seculars, in any age may break their trust; enfeoffments to uses naturally produced their share of litigation. For the perversion of an intended outcome there was of course no remedy at the common law, because in common law the defaulting feoffees were the real owners. Such matters of conscience, however, lay precisely within the equitable jurisdiction of the chancellor––hence the mass of petitions relating to enfeoffments to uses among the Early Chancery Proceedings (see below).

EXTENT. Legally, an extent(us) is a writ ordering a valuation of both landed and movable property. In ordinary usage, it may refer to the survey and valuation, rather than to the writ, and the term is so used here.

GRANT. A grant is a transfer of property by means of a written instrument, without the delivery of possession. The property may be of any kind, landed or, like a right to a certain status or power of decision, incorporeal. Since documents issued by the crown and state offices are in general not noticed in the Index, it may be assumed that grants here are in the form of an indenture. Grants for a term of years at a money rent are commonly, but by no means systematically, described as leases; the usage is ordinary, not subject to the legal qualification that the interest created by the lease be less than that of the lessor’s own interest in the property. (When the qualification is exceeded, the contract is properly an assignment.)

INSPEXIMUS. An inspeximus (‘we have examined’) is a certified copy of a document, such as may be required for evidence in a legal dispute.

MARRIAGE ARTICLES (marriage contracts, marriage settlements). A contract between two parties, for the marriage of a person for whom one of the parties is responsible, to a person for whom the other party is responsible. The intended husband could himself be party to the contract, according as he was independent of his father or legal guardian. The intended wife’s interest was represented nearly always by father, uncle, brother or guardian. The chief intention to which the contract gave effect was very often an alliance between two families. Certainly, the sensibilities of the bridal pair can have played no very great part in settlements which contain provisos like the following: ‘and if E––– the said daughter of W––– die before she be of age (as heaven forfend), then H––– the said son of R––– shall take to wife the eldest daughter of W––– still surviving as soon as she be of age...’

The disposition of property was usually an important part of the contract. The landed property described as ‘in jointure’ is that which is settled on the pair, to descend to their issue and to provide dower in the event of widowhood. The jointure may derive from either party to the contract, or both of them, depending on (among other things) their particular wealth and status.

PERAMBULATION. A perambulation is a walking of boundaries, and the written record of the walking.

PRESENTMENT OF A JURY. The functions of the medi- aeval English jury were not confined to deliberating the innocence or guilt of an accused: in certain courts, it was the duty of the jury to bring to notice matters requiring legal action, and so to bring indictments. The presentments of a jury are evidences and indictments so brought.

RENTAL. A rental is an inventory of landed property, its tenants, and the rents due from them. (A rent roll is a roll used for recording such information.)

TERRIER. A terrier is a survey of lands, which includes such information as acreage, land use, tenants, and rent values.

For present purposes, a seal is a piece of wax bearing the impression of a personal or institutional emblem or heraldic device, attached to a document as a guarantee that it is authentic. In size, such seals vary from lumps as big as a hazel-nut to discs of four inches or more across. For deeds, the usual means of attachment is a seal strap. The strap may be a piece of vellum or parchment, or a cord, inserted through a slit cut in the turned-up foot of the document; or, a strip may be cut back from the foot of the document itself, to hang at the left edge. The wax of the seal is affixed in molten state, and the impression made with a heated metal stamp (also called a seal). Some writers have imagined that documents lacking their seals cannot be presumed authentic, and that they are therefore untrustworthy for philological purposes. Conversely, it has been tacitly assumed that a document with its seals intact can be taken at face value. The following, however, should be noted.

Firstly, the presence of an intact seal on a document written five hundred years ago is frequently a matter of chance: wax seals are very easily damaged, as anyone who has handled large quantities of mediaeval documents well knows, and the heap of crumbled fragments at the bottom of a deed box is all too often augmented by the mere act of removing documents for inspection. Disintegration, however, is not the only hazard: the systematic depredations of collectors also take their toll. Accordingly, the presence of seal straps, or slits for seal straps, may be the only remaining physical indications that the document was in fact sealed. (That it was intended to be sealed commonly appears from the sealing clause that concludes the legal text.)

Secondly, even when seals are present and intact, their identity, without the evidence of the document itself, may be obscure. If the criterion for authenticity and philological value were the status of their seals, then many of the private deeds executed by persons of humble rank would have to be excluded––and these are precisely the local documents that provide the best evidence for non-standard usage.

Thirdly, forgery may be in question, and it lies hardly within the competence of persons other than expert sigillographers to detect it merely from the seals; moreover, the necessary inquiry could not well be confined to vernacular documents. Even when the seal is genuine, however, forgery cannot be ruled out: seals––in the sense of the metal device for impressing the wax––can be misappropriated by unscrupulous persons and applied to documents of which their owners have no knowledge. One such perversion appears from the texts printed as numbers xvii and xviii in Morsbach’s Mittelenglische Originalurkunden (Morsbach 1923; cf. also numbers xv, xix and xx).

Fourthly, assessment of a document’s authenticity and of its value for philological purposes must in any case rest on a scrutiny of the diplomatic; that is, of the type of document, of the transmission of the text, and of the circumstances in which it was produced. Even then, linguistic considerations may rule out the use of a document as witness to local usage, impeccable though its diplomatic claims may be. Men travelled: what is recognisably the language of the south-west Midlands is not turned into the language of Durham by its appearance in what are undoubtedly Durham documents (cf. Durham, Prior’s Kitchen, Locellus XXV.58 & 68, and Misc. Charter 6367). Linguistic analysis may itself be an important adjunct to the established canon of diplomatic criticism.

Lastly, it must be emphasised that although considerable effort has been made to ensure that the use of documents in the Atlas rests on firm foundations, it is not to be supposed that individual evaluations are impeccable. Such work can be immensely time-consuming, the range of material that needs to be known is almost without limit, and the final result may still be a series of imponderables that admits only a linguistic resolution. In these matters, seals are usually the least of considerations.

The evidence for the place of origin of local documents is normally provided in their dating clauses: a document is dated––‘given’––on a named day and from a named place. So, for example, ‘Gyfen at the namptwyche the ix day of Octobre The yere of the regne of kyng henry the sext after the conquest the xxxiiijti’ (Morsbach 1923, no. xv). Frequently, the word used is not ‘given’ but ‘written’ or ‘made’; sometimes the place at which the document was sealed is stated instead. In perhaps most cases, these expressions amount to the same thing, but they need not do so. If a document presents itself as ‘written’ or ‘made’ at a certain place, then, forgery apart, the local origins of the final copy are assured. Even here, however, it cannot be assumed that the text was composed at the place in question: it may depend on drafts produced elsewhere, incorporated virtually unaltered into a final version. If the document is merely ‘given’ at the place stated, then it is possible that it was brought ready-made for mutual sealing and exchange of copies. The character of each document must be assessed individually.

Even when a document is stated to have been made or sealed or given at some particular place, it cannot be assumed that its language belongs to that place as a matter of course. It is merely reasonable to accept as a provisional hypothesis, in default of evidence to the contrary, that its language does indeed belong there. ‘Evidence to the contrary’ includes, among other things, linguistic incongruence with various other documents likewise dated from the place in question, and having equally good claims prima facie to represent the local language. More generally, linguistic incongruence with the surrounding dialectal configuration–––badness of ‘fit’–––would also pro- vide grounds for rejecting the stated local origins of a document as evidence for its linguistic provenance. Occasionally, non-linguistic evidence is available to show that a document is the work of a man who has left home. So, for example, enough is known of the life and circumstances of William Somerwell, registrar to the archbishop of Armagh from ca. 1429 to 1458, to explain the appearance of Bristol dialect in documents written in Ireland. In a linguistic atlas, these are rightly treated as source material for Bristol, regardless of their diplomatic origins in Co. Louth. Even were Somerwell’s biography not known, however, the incongruence of his language with the language of the other Mediaeval Hiberno-English documents, whether from Co. Louth or elsewhere in Ireland, would be enough to exclude them from the Irish material. Conversely, their agreement with the language of the Bristol monuments would provide strong evidence for the place of Somerwell’s upbringing.

Documents written in competent law hands and dated from very small villages or hamlets are intrinsically unlikely to represent the language of precisely the places to which their texts belong. Such places could not of themselves support an independent notary, and a man with proper legal edu- cation in such parts must have sought employment in the service of a substantial landowner, if indeed he were not from a like family himself. The retinues of the great might well be drawn from several counties, and clerks found acting on their behalf cannot be assumed, without close inquiry, to belong even to the region wherein their activities lay. By contrast, substantial but still local families like the Radcliffes (see below) can be supposed to have drawn their retainers from their own neighbourhood, at least for the most part.

Documents written in scripts that cannot be identified as law hands, particularly those scripts that are uplandish, and documents that are drafted in eccentric fashion, are by contrast likely to be genuinely local products. It must be remembered that legal practice during the later middle ages, especially in the provinces, was by no means a professional monopoly; the confidence of the client was the chief criterion for employment. A man’s competence in legal form might on occasion derive from amateur scrutiny of such legal instruments as came his way, rather than from a formal education or pupillage. Formal qualification as a necessary preliminary to practice belonged to a later age; attendance at an Inn of Chancery or Inn of Court was no commitment to graduation.

What part the lower clergy may have had in the production of legal documents is wholly uncertain, but the explicit testimony of Symonde Wagstaffe, vicar of Glossop, is worth recording. Manchester University, John Rylands Library: Misc. deeds relating to Cheshire, bundle 1 no. 868, is a declaration in the names of John of Oldom, priest, John of Glossop, monk of Basingwerk, and of various gentry of the Peak District. They met at Gamesley in the Peak, an isolated hamlet in the moorlands of north-west Derbyshire; there, Richard þo Masse of Sale showed them the deed by which he had enfeoffed John þo Masse, parson of Ashton, and Hew of Barlawe, in lands devised for the benefit of Richard and his heirs (on such enfeoffments to uses, see Documents: categories and terminology above). The vicar of Glossop testified that he had written that deed with his own hand, but at Godley, in east Cheshire. The deed had been drafted, however, by another priest, namely the parson of Ashton, who was one of the feoffees; and it was at his request that the vicar of Glossop had made the fair copy. The present declaration (which is to the effect that the feof-fees have given back Richard’s lands) is of obvious interest as evidence for the social and geographical complexities that may underlie the production of a single document: was the language of the enfeoffment Symonde’s, as fair copyist, or John’s, as draftsman? Either way, did that language belong to Ashton, where John was parson, or to Glossop, where Symonde was vicar, or to Godley, where he wrote it, or to none of these? As to the authorship of legal documents generally, it should be noted that priests were a favoured class among feoffees to uses; Symonde and John need not have been the first to provide the literacy and acquaintance with legal form that the deed itself requires.

Some of the difficulties which beset precise definition for the local origins of a document’s language may be illustrated from Lancashire County Record Office DDIb Culcheth 1381. This document is a memorandum of evidences shown by commandment of ‘my lord of Durem’ in support of Richard of Radclyf’s disputed claim to certain lands beside Culcheth, a village in south Lancashire. The Radcliffes themselves were not from Culcheth, but from Radcliffe some ten miles to the north-east; Smithhills, four miles west of Radcliffe, became the family’s chief seat. The evidences set down in the document are the testimonies of various aged inhabitants of Culcheth, delivered presumably viva voce at Culcheth. The writer can hardly have been other than a clerk in the Radcliffes’ employ, and his language is such as to leave no doubt that he belonged to their part of the country. Yet we cannot be sure that he was from Radcliffe itself. Indeed, the involvement of ‘my lord of Durem’, as it happens, rests on an earlier recruitment of this sort from Langley in Middleton just five miles away: Thomas Langley, whose career culminated in the offices of chancellor of England and bishop of Durham, first made his way as a clerk in the Radcliffes’ service (Storey 1969, pp. 2–3). In the present state of knowledge, the language of the memorandum is best entered at Radcliffe, even though it may belong rather to some neighbouring village. For all we know, the clerk’s presumed visit to Culcheth was a return home, and the gathering of testimonies drew upon his own local connections; but he may equally have belonged to somewhere a dozen miles the other side of Radcliffe. In assigning the language of his text to Radcliffe, the administrative focus is made a linguistic focus as well. The ensuing map is necessarily an approximation, therefore, even in terms of what must be treated as an ‘anchor text’; but it is an approximation within fairly close limits.

Arbitrations are commonly difficult to assign to one place rather than another. Magnates frequently appear as arbitrators well beyond the bounds of their manors, and some men ranged over several counties in this capacity. Thomas Haryngton, for example, served as justice of the peace for Cumberland (1448), for Lancashire during the 1440s (in which county he held other office for life), and for the West Riding of Yorkshire through the 1440s and 1450s (Roskell 1937, pp. 179–86). Even when an arbitrator’s activities are confined to his own neighbourhood, as were Sir John Stanley’s to Cheshire and south Lancashire, it is always a question whether the disputants went to the arbitrator for judgement, or whether the judgement was given locally. We need to know whether the deed of arbitration is the work of a man who belonged to the place where the disputants lived, or whether he came from elsewhere–– perhaps from the arbitrator’s own caput, to which he may or may not have been recruited from some other place. Mostly we have to settle for approximations, and some arbitrations involve so many different localities that for dia- lect mapping they are not usable. The value of a knowledge of topography and local history in such matters may be illustrated by the following example.

Cheshire Record Office: Cholmondely Collection C.139 is an arbitration in Cheshire language, dated 24 Dec 29 Henry VI, but from no named place. The arbitrators are Rondull of Brerton the elder, and John of Dutton of Hatton. Brereton is three miles north-east of Sandbach. Hatton, less readily identifiable, is possibly Hatton five miles south-east of Chester, some twenty miles west and slightly south of Brereton; or it may be Hatton near Runcorn, some sixteen miles north-west of Brereton. Since Dutton is but two miles south of this second Hatton, a presumption in favour of Hatton near Runcorn might be urged. The disputants are Hugh of Stretton et uxor, and John Ketull et uxor. Stretton is two and a half miles east of Hatton near Runcorn, but John Ketull’s local origins are wholly obscure. Thus far, there is some coherence: a disputant from Stretton–– though it is not certain that Hugh still lived there–– might well propose a man from neighbouring Hatton as an acceptable arbitrator. We might then conjecture that John Ketull lived in the neighbourhood of Brereton, and that Rondull of Brerton was involved for similar reasons. From that it may seem to follow that the language of the arbitration belongs to one or other of these localities, and that, since there is no means of choosing between them, it must be regarded as ‘mid-Cheshire, unlocalised’. As it happens, however, the local origins of the document can be determined, and it belongs to neither Hatton nor Brereton. The properties disputed belonged formerly to Patrik of Larketon; they lie in Larkton (MS Larketon), Bickerton (MS Bykerton), Edge (MS Egge), Macefen (so MS), and Royall Moore (MS Ruylesmore), all towns within the ecclesiastical parish of Malpas. (EPNS 47 Cheshire iv, pp. 5–6.) It is to this third neighbourhood that the document can be assigned. Firstly, places called ‘Hatton’ and ‘Stretton’ are both traceable in Malpas––as would have been readily discovered had the index volume to EPNS Cheshire been published. Secondly, Rondull of Brerton was of the Malpas family, as anyone familiar with the local history of mediaeval Cheshire would know, but as, on first acquaintance with the document, we did not. (Relevant notices appear in Ormerod’s Cheshire ii, pp. 686–7, and iii, p. 88; and in Record Society of Lancashire and Cheshire 116, p. 104.) Thirdly, on establishing the Malpas connection, close comparison with other documents from that neighbourhood followed as a matter of course. These are four indentures to which one Richard Cholmeley was party, spanning the years 1450–56; he was of the Cholmondeley of Chorley family, in whose muniments these indentures were formerly preserved, and Chorley is a township in Malpas. It emerged that the indentures and the present arbitration are all in the same hand.

It is a reasonable inference that the writer of these texts belonged to the neighbourhood of Malpas, even though on present evidence we cannot be confident that he was native to Malpas itself, or that his habits of written language were acquired there: the problem is in principle the same as that posed by the Culcheth evidences considered earlier. The local origins of the text are what is determined by historical and diplomatic criteria––though it must be said that in many cases the necessary investigations have been less thorough than we would like, simply for lack of historical reference books: by no means all counties are as well served as is Cheshire. Again, the place to which the extra-linguistic evidence points must be taken as a linguistic focus, not as a guaranteed linguistic origin.

It may then be asked what is the special value of the local documents, given that textual or diplomatic provenance are in principle independent of linguistic provenance; are not the distortions introduced by plotting an indeterminate number of texts on our maps at some (again indeterminate) distance from their real linguistic provenances an invalidation of the whole atlas? There are two considerations which suggest not. Firstly, such distortion as there is must be local: the misplacing by twenty miles of a document in Warwickshire has no bearing on the placings for, say, Cheshire. The distortion is contained, even though a precise limit cannot be set to its extent. (If it could, then we would have to do with approximation within defined limits, a very different matter from unidentified distortion.) Secondly, the demand that the maps reconstructed for the past display the same qualities as those which inhere in maps for modern dialect continua can be brought to bear. Since, for the most part, regional dialect changes in an orderly way over space, the placing of a document that disrupted the linguistic pattern already established, and did so for a large number of its features, would be inherently suspect. No matter that the document was written in that place: its language could not be assigned there, since it failed to cohere with the pattern established by a consensus of other documents, each of which had no less good a claim (in terms of dating clauses or other diplomatic evidence) to be considered as representing the language of that area. That people moved about is in no way contested: it is of course to be expected that some, perhaps many, local documents were written in places far removed from their writers’ home ground. The question to be answered is ‘Which are the likely candidates?’; and the dialect map itself is a powerful tool for identifying them.

As noted above, many of the documents noticed in the Index are written not in local language, but in the emerging London-based standard or something approximating to it. Various descriptive formulae have been used. ‘Chancery Standard’ is reserved for that type of London language so distinguished by Samuels (1963, pp. 88–9: ‘Type IV’). The criteria for its recognition are linguistic, and have nothing to do with the institution from which a document originated. It is a form of London English attested regularly from the 1430s onwards; as the fifteenth century progressed, legal and administrative language increasingly converged upon it. It is neither the monopoly of Chancery nor the only London English to be found among its writings. (For some pertinent observations on recent claims, see Davis 1983.)

Chancery Standard did not displace the older London language overnight. Some of the features characteristic in the usage of Chaucer and Hoccleve’s generations gave ground very rapidly to what became the standard forms; others persisted, in city and government writings, even into the sixteenth century. Moreover, the Central Midland element in Chancery Standard, which goes far to account for its differences from older London usage, had itself not stabilised: Central Midland forms such as found no lasting place in the written standard are attested commonly, albeit irregularly, in fifteenth- and sixteenth-century documents from city and state bureaucracies alike. In these circumstances, and with a view to preserving the linguistic content of the label ‘Chancery Standard’, terms like ‘state English’, ‘government English’, ‘standard’ and ‘near-standard’, have been freely applied.

The terms ‘near-standard’ and ‘semi-standard’ are also used to describe the language of provincial writings in which local usage is discernible, but which is accommodated to the written standard. Similarly, ‘dilute’ indicates a strong admixture of standard forms; ‘colourless’ describes language which, though it is not standard in the sense of conforming to London usage (whether Chancery Standard or otherwise), is yet devoid of markedly local forms. It should be emphasised that the application of these terms is inevitably somewhat subjective; they are indicators only, and not to be taken as the categories of a formal schema.

The term ‘colourless regional standard’ (Samuels 1981, pp. 43–4) is also to be found. Its characteristics are negatively defined, and reflect process rather than conformity to a particular model. It comes about when a writer replaces some or all of his distinctively local forms by equivalents which, although still native to the local or neighbouring dialects, are common currency over a wider area. The result is not a series of well-defined, regional standards like that of the Central Midlands, but a continuum in which the local element is muted, and one type shifts almost imperceptibly into another. There are degrees of neutrality, and the extent to which any particular specimen may be judged colourless depends on its regional context: what is colourless in York, say, would be conspicuously alien in Bristol. Such selection may lead to the co-existence in a single community of two distinct types of written language, one a local register, the other a neutral or upper register––a state of affairs which from the modern spoken language is entirely familiar. So it is that the language of writings for local use, like commonplace books and literary texts, may differ markedly from the language of those administrative and legal writings intended for a wider or more exalted public. This dichotomy is especially striking in the Norfolk material, and since it appears there well before English had become the regular language of government, it can hardly be ascribed to the spread of Standard English as ordinarily understood. It is, indeed, arguable that colourless regional standard emerged as a vernacular replacement for Anglo- Norman, a new lingua franca, and that its existence prepared the way for the otherwise surprisingly rapid adoption of Chancery Standard (with which it variously interacted) later in the fifteenth century. These matters will be treated more fully in a forthcoming monograph (Benskin and Sandved).

Documents in English produced by the chancery, the exchequer, the privy seal office, and the signet office, have in general been excluded from the Index. Of these, most by far are written in the language of London or something approximating to it, predominantly Samuels’ Type III in the early decades of the fifteenth century, but converging increasingly on Type IV––‘Chancery Standard’––as that century wears on (cf. Samuels 1963, pp. 87–93; 1972, 8.5). A handlist for the material to 1425 was prepared by M.M. Weale, as a supplement to Chambers and Daunt’s Book of London English 1384–1425 (1931, pp. 269–317); it is confined, however, to documents preserved in the Public Record Office. For the signet letters, which from August 1417 to the death of Henry V (September 1422) are written only in English, Kirby’s Calendar of Signet Letters of Henry IV and Henry V (1978) should be consulted; it includes items now among the Cotton manuscripts at the British Library (though there are some omissions), as well as the PRO documents. Other instruments of government written in English are likewise to be found in Sir Robert Cotton’s collections, and they appear also, whether as originals or contemporary copies, in many provincial archives. A great deal of this material has been examined in the course of work on the Atlas, and would have appeared in the Index of Sources had a more comprehensive publication been possible. Some further account will appear in a forthcoming work on the beginnings of Standard English (Benskin and Sandved: cf. Sandved 1980 and 1981).

Of the government documents not in London English, certain points may usefully be noted here. Firstly, several items in mixed language have come to notice and in some cases such documents may well reflect the habitual usage of clerks in government service. So, for example, the language of Henry V’s will dated 21 July 1417 (PRO E 23/2) bears a distinctively north-west Midland stamp: faure ‘four’, laide ‘lady’, -et ((-it)) wk. ppl.; cf. also ar ‘are’, (beside been, be), agayn ‘again’, geuen ‘given’, probably of similar origin. That these forms are to be assigned to the copyist rather than to an exemplar is strongly indicated by the absolutely regular y for ‘þ’: þ for ‘þ’ is never written. This is a markedly regional trait which as sustained usage is unlikely to be copied at all, let alone accurately; it cannot be regarded as London practice at this date (Benskin 1982a, esp. pp. 17 and 25). Other forms in the will, however, cannot be assigned to the relevant parts of the north-west Midlands at all, but are typical in the London language of the time: nat ‘not’ (beside noght), y- ppl. prefix, maad ‘made’, hir ‘their’ (beside yair), regular -eth 3 sg. pres. ind. Such mixed usage may very well represent a genuine état de langue, such as might be expected of an incomer from Cheshire or (more probably) Lancashire who had only partially accommodated to the written language of the capital. The seal of the document is missing, and with it proof that the scrivener was a clerk in a government department; but the hand and language rule out the clerks to the signet––their writings have been analysed exhaustively––and there is a strong presumption of origins in the exchequer (king’s remembrancer) or the privy seal office. The essential point here is that clerks at Westminster in government or royal service need not be expected to produce documents only in London language. The change of dynasty in 1399 brought various Lancastrian talent into government service, including clerks from the duchy’s chancery. In the absence of any national written standard, Lancastrian English was no less eligible than London English for such records of state as were not in Latin or French.

The will of Henry V is an original document, even though it may be a fair copy from a draft; it is possible that one scrivener was responsible for all stages of its production. Enrolled versions of documents, by contrast, are self-evidently copies of texts that exist in an independent final version, and the likelihood of linguistic contamination must be reckoned with, as in the case of a literary text: the forms of an exemplar may be transmitted faithfully, by a literatim-copyist, displaced sporadically or frequently by a more or less close copyist, or regularly converted into the scrivener’s habitual usage. In the case of PRO C 54/270 mm. 12d–11d, a Chancery enrolment of a royal proclamation of peace with France (14 June 1420), the north-westerly forms are probably, like those of Henry V’s Will, the imprint of the hand that wrote it, for ‘þ’ is again regularly written y. But even if the y for ‘þ’ is copied literatim, language having a strongly north-western component is still established within the Westminster bureaucracy, for whereas uncertainty attaches to the scrivener of the will, there can be no question of private origins for the original proclamation and writ. (The writ is in Latin, not English.) That the language of the original proclamation was itself mixed is indicated by -eth and -yth (beside Midland -en) in the pl. pres. ind. This is not to be taken as the southerly usage, but is a common hypercorrection of the northern pl. in -es, on the analogy of the northern 3sg. -es with the Midland (and southern) 3sg. -eth (Benskin and Laing 1981, 7.3.3–4; cf. McIntosh 1983). In this, and in some other features, the language of the proclamation appears to be more northerly than that of Henry V’s will.

Merely from these examples, it will be seen that the linguistic analysis of state documents may be complex. Even when the regionally-distinctive components of a two-layer text have been recognised, only such general designations as ‘NW Midland’ may be possible. The position here is very different from that presented by local documents in language similarly mixed (or, later in the fifteenth century, semi-standardised). If the provincial component in the local document be more or less what is expected from the place at which it is dated, then it may be reasonable to suppose that the provincial spellings do indeed derive from that neighbourhood; in some cases, of course, the assemblage of local elements is highly distinctive. With the state documents, by contrast, there is usually no explicit indication at all of the place to which the non-metropolitan forms belong. Although local origins might be determined, even fairly closely, by the linguistic methods outlined in the General Introduction (2.3.3–7, on the ‘fit’ technique) such ascriptions would be the goal of analysis, not a starting point. For regional dialect survey, as opposed to sociolinguistic inquiry, little if anything is to be gained from such an exercise: these writings cannot be used as primary sources or controls, and if they can be localised at all it is only because better material is already available. Accordingly, no attempt has been made to record them here, even though the Index contains a few strays of this type.

There remains a third class of state documents, that could not well be omitted from the Index. These are the documents written in some distinctively local language from outside London and its environs. Here, a literatim-copy counts no less than an original. The parliament rolls (PRO C 65) yield various examples. The petition presented to the parliament of 1436 on behalf of Isabell, widow of John Botiller of Beause, knight, is item 15 on the roll C 65/94. ‘Beause’ is Bewsey in south-west Lancashire, and the petition is in good Lancashire language. It need not be supposed that Isabell, or any of her family, wrote it; but composition by a lawyer in local practice, even a clerk of the Beause household, is very reasonably assumed. The Chancery clerk who copied it onto the parliament roll cannot possibly be held responsible for its Lancashire forms unless it be supposed that he was himself a Lancastrian, and additionally that he had cause to transpose the language of his original. Rather the clerk merely copied, more or less letter by letter, whatever was set in front of him. The preceding item on the roll, unquestionably in the same hand, is in the incipient standard language familiar from any number of mid-fifteenth-century state documents.

The composition of the parliament rolls has been succinctly stated by H.G. Richardson (1937): ‘The enrolling clerk... had a good many loose documents before him, all connected with the parliament, from which, it is to be supposed, he selected under the more or less precise instructions of the clerk of the parliament’ (p. 41)... ‘no medieval parliament roll is a full day to day account of the proceedings...it was written at leisure from other documents’ (p. 42). For a detailed account of both the administrative procedures and the diplomatic, reference may be had to Myers’ work on the parliamentary petitions (Myers 1937).

In some cases, it is possible to establish in detail the enrolling clerk’s treatment of his copy-text, because the original petition itself is extant. A petition concerning the non-residence of parsons and vicars, presented to the parliament of July 1425, survives as PRO SC 8/135/6716; the enrolled version is PRO C 65/86 m. 13, no. 38. The north-west Midland language of the original is faithfully rendered, in a virtually literatim-copy: the enrolled version even includes the form of the abbreviation in monñs ‘man’s’, and the probable error in agays for ‘against’; fidelity perhaps is exceeded in ‘qwat moner mon’ for ‘qwat maner mon’ of the original, with west Midland on in place of familiar an anticipated once too often.

The other English texts enrolled by this clerk are in various types of language, and even had the original of the petition concerning non-residence not been extant, the fact of close copying would still have emerged, just as for the Beause petition considered above.

As the legal and administrative language conforms increasingly to a national written standard, the extent of variation in the original petitions of course diminishes, and the English entries on the parliament rolls, even when copies literatim, become linguistically much more homogeneous. Even in rolls of the later fifteenth century, however, literatim-copying can still at times be detected from text-conditioned variation in the detail of spelling.

The class of documents called ‘Early Chancery Proceedings’ (Public Record Office, London, classification ‘C 1’) needs special note. The series begins late in Richard II’s reign, and numbers about ten thousand items in all; the mediaeval proportion is large but undetermined. Their language is regularly English throughout the reigns of Henry VI (1422–61) and his successors; before that, items in English scarcely appear. (English, and from the reign of Henry V, are C 1/1/14, C 1/1/18, and C 1/1/21.) The documents preserved in this class are mainly formal petitions addressed to the king’s chancellor of England, seeking redress for wrongs that the petitioners believed could not be righted by the ordinary processes of the common law. Depositions–– memoranda of written evidence––are also to be found. Often preserved with a petition is the corresponding replication, that is, the written answer given to the Court of Chancery by the person(s) complained of in the original petition.

Subsequent examination of the plaintiff and defendant was conducted viva voce before the chancellor. No formal record was kept either of the examination or of its outcome; a summary note in Latin, written at the foot of the petition or on its dorse, may record a decision, or the date on which a decision was reached, or the date on which a hearing was to proceed or had proceeded, but such notes are irregular. Most petitions, in the absence of independent testimony, can be dated only approximately: the usual form of address is ‘To the right worshipful and reverend father in God, N–––, bishop of B––– and chancellor of England’, or some variant of it; the episcopal tenure of the chancellorship so stated sets the limits of date for the petition.

Most petitions follow a common form, and are written in the third person. They show no signs of having been sent into the Chancery: there are no seals, seal straps, or slits for seal straps; they are addressed not in fashion of a letter, on the dorse, but at the head of the text of the petition, and they show no signs of folding. These considerations led to the view that they were not the original petitions at all, but paraphrases made by Chancery clerks (so Benskin 1977, p. 508; repeated in Laing 1978, p. 46). That view, however, is mistaken. The petitions in Early Chancery Proceedings, it now appears, are very largely the work of common lawyers, that is, of the petitioners’ own counsel. The reason that they show none of the ordinary signs of being missives is that they were delivered by hand into the Chancery, by the professionals responsible for drafting them. (‘Handing in’ was maintained until Elizabethan times, albeit by then as a fiction: see further, Jones 1967.) No signatures appear on the petitions; the practice of having counsel sign them began apparently in the reign of Edward VI (Baildon 1896, p. xxiv). At the foot of the petition, however, there normally appear the names of the two ‘plegii de prosequendo’, that is, of the guarantors of the action.

The extent to which a lawyer’s activities might be split between Westminster and the provinces should also be noted. It was nothing unusual for a Westminster lawyer to maintain an extensive provincial practice, on which he depended financially. Of the lawyers trained in the Inns of Court, over half went into provincial practice anyway, and the proportion from the Inns of Chancery was even higher (Ives 1968). Whether the professional developed his practice at Westminster or in the provinces, he was regularly the link between petitioners from the shires and the courts at Westminster. In so far as he had been educated in the Westminster law schools, he would have been exposed to London English, and his written language is likely to reflect it––whether as wholesale adoption or merely partial accommodation.

These considerations are of the first importance in assessing the language of any of the items among the Early Chancery Proceedings. A petition is associated with a given place in the first instance by the petitioner’s own person: ‘Meekly beseecheth your poor petitioner A––– B––– of N–––town...’ In the substance of the petition, places other than N–––town may be named: the locale of disputed property, perhaps, or of indignities inflicted on the said poor petitioner. Now, if the language of the document is standard––that is, London administrative language regardless of whether it is precisely Chancery Standard–– then there is obviously no reason to associate it with any of the places named in the petition, and the document is not of the slightest value for dialect mapping. Its language may be standard because it was drawn up by a local lawyer who had yet been educated at the Westminster law schools, or because the petitioner had consulted a lawyer in Westminster practice. If, however, the petition is written in clearly non-standard language, then it is all too easy to assume that it represents the dialect of the petitioner’s own neighbourhood.

Non-standard language in a petition, however, need have nothing to do with the local origins of the petitioner. The petitioner’s counsel, the man responsible for drawing up the petition, might well belong to a different part of England; and it is his language, not the petitioner’s, that appears in the document. So, for example, PRO C 1/9/39, in the name of John Matresmaker ‘of North Witham in the shire of Lincoln’. The language is largely standard, but it contains a few forms that are decidedly provincial: -uth 3sg. pres. ind., -ode wk. pt., onswere vb. ‘answer’. These forms, however, are assuredly not of North Witham, or even of anywhere else in Lincolnshire: as an assemblage, they are decisively western, the forms of a man whose written language, in spite of legal education, yet betrays his earlier upbringing. The circumstances that brought him as a client a man from Lincolnshire can only be guessed at: he may have been established in Westminster practice, and there consulted by John Matresmaker; or perhaps he had set up in Lincolnshire, after leaving one of the Inns of Court or Inns of Chancery in Westminster. Either way, this document is a warning that such petitions cannot be taken on trust as evidence for the language of a petitioner’s near neighbourhood.

The caveat applies with no less force to petitions in language of a sort that conforms, more or less, to the usage expected from the petitioner’s locale. A successful provincial practice involved in representations at Westminster might draw its clients across half a county or so, reflecting not so much a geographical focus, perhaps, as family and social connections. As likely as not, the principal of such a practice was himself from somewhere within that area. Were his written language non-standard, then the chances are that it would have much in common with the usage of almost any of the places to which his clients belonged. Because the texts preserved among the Early Chancery Proceedings are in general short, and because their vocabularies are commonly repetitive, a single specimen rarely affords a wide range of linguistic evidence; and the narrower the range of linguistic evidence, the less likely it is that a mis-match will be detected, even when there are other, localised, documents for comparison. Without consideration of these circum- stances, therefore, a philologist may be naturally predisposed to treat the petitioner’s dwelling as the place of linguistic origin. Only such blatant mis-matching as appears, for example, in John Matresmaker’s petition, would call in question the belief that the language of such documents is securely localised. The result, for dialect mapping, would be considerable local distortion at the least; and depending on a priori notions of what is plausibly the language of this or that part of the country, even severe dislocations might not be recognised for what they are.

The documents in Early Chancery Proceedings––whether petitions or replications or depositions––must therefore be considered on their individual merits as potential sources for dialect mapping. Productions in regular law hands are in principle less likely to reflect the written language of the person in whose name they appear than are texts which display the script and drafting of an amateur. Civic corporations especially, and some ecclesiastical institutions, maintained within their own ranks persons with the competence necessary for legal drafting; in so far as their recruitment was local, documents in their name, when in non-standard language, are more likely to reflect local usage than are similarly non-standard documents in the name of ordinary private persons. In all such cases, however, the language of a document must be assessed in the light of other, localised material; and even then, addition to a consensus, rather than establishment of independent testimony, is usually the best that can be expected. Sometimes, however, the collective testimony of several documents from the Early Chancery Proceedings alone is sufficient to establish what could not be taken on trust from any one of them individually: so, for example, when diverse petitions by different hands agree closely in their language, all of them are in the names of persons from the same one place, and their language, although distinctive, meshes closely with that of localised material from the surrounding area. Such a case is presented by C 1/9/426, C 1/17/81 A and B (two hands), and C 1/19/330, 331 and 332 (all one hand). These all belong to Towcester in Northamptonshire, and though they contribute too little to be worth entering on the Atlas maps themselves, they do enable the localisation of richer material in an area that is dialectally not well differentiated. In general, such a supporting role has been the part of Early Chancery Proceedings in this work, and they have been used, it is hoped, with due caution.

The existence of the Harleian Society’s index to persons named in Early Chancery Proceedings (Walmisley 1927 and 1928) renders the present list partly redundant; and only a sample of these documents has been examined. Yet the Harleian Society’s index does not classify its entries by county, though the counties of origin are duly recorded; and we have thought it well to provide such linguistic information as is available for these writings, even when, as so often, ‘in standard language’ is the report.

5. Organisation of the Repository and County Lists

Note. For a single manuscript there may be several successive entries, each constituting an independent article in the Index.

The categories of information contained by an article in the Index are printed in order as follows.

(1) The repository and name of the manuscript. The first word is normally the town or city in which the repository lies, followed by the name of the repository and then the name of the manuscript. So, for example, ‘Oxford, Bodleian Library, Douce 104’, ‘Cambridge University Library Gg.I.1’. Famous institutions in towns that are not well-known are, however, listed under the name of the institution: thus the manuscripts of the Henry E. Huntington Library at San Marino, California, are listed under ‘Huntington’, not ‘San Marino’. The aim has been to place entries where the user would probably first think of looking, rather than to impose an irksome consistency. Cross-references have been freely inserted.

Not all of the writings here considered are known directly from the manuscripts: for some, whether by reason of destruction or loss, printed versions are the only available source. Accordingly, there is a section of the Repository Index styled ‘Printed:’, in which such items are listed under the work in which they appear. The section itself is alphabetically ordered, by authors’ and editors’ surnames; works which are not assignable to an individual, as for example periodicals, are entered by title in the same list. Works that were known in recent times from manuscript repositories, but which have since been lost or destroyed, are entered under their former repository in the Repository List, with a note of the loss and reference to such printed versions, transcripts or facsimiles as may be available. The material destroyed by explosion and fire in the Public Record Office of Ireland in 1922 accounts for nearly all such entries.

(2) The hand that is represented by the attached LP. If the manuscript is the work of only one scribe, the fact is duly noted. Manuscripts that are the work of just one scribe are designated as ‘in one hand’; a scribe responsible for the bulk of a manuscript is designated ‘main hand’. If the hand represented by the LP is one of two or more making substantial contributions, normally the sigla ‘A, B, C,...’ are used to distinguish them.

Sometimes, however, hands are identified by the texts for which they are responsible, as, for example, in ‘Hand of the Speculum Christiani’; sometimes the folios or pages for which the hand is responsible are the only indication (so, for example, ‘Hand of ff. 27r–41v line 12’). Manuscripts that are the work of many hands are more conveniently treated in this fashion than by attempting to identify scribes by sigla: to establish that the hand of a few folios in English is, say, ‘H’ rather than ‘E’ or ‘K’, may involve detailed comparison of the hands of Latin works preceding, and hence labour irrelevant for present purposes.

It is not always possible to decide whether a text is the work of more than one hand. If the language of such a text is more or less uniform throughout, then it presents no operational difficulties for incorporation in the Atlas, though the fact of such uncertainty is duly noted in the summary description. Where what is probably a single hand contributes more than one type of language, then a formula such as ‘Hand B, language 2’ is invoked. Such bias as we have been able to detect in our work is towards over-differentiation of hands: only rarely have we treated as the work of a single hand what at a later stage we came to believe was written by two or more hands. This is perhaps to be expected in a study where language is the focus of interest; conflation of different varieties may prove seriously misleading, whereas error in the opposite direction does not materially affect the outcome. (It may be noted that in one or two cases the authors themselves differ in their views as to the number of hands represented. Oxford, New College 95, is a case in point: the division of the text into different linguistic sections is not in question, and differences of script as between one such section and another are readily observable; but there are also some striking continuities, and it is debatable whether there is more than one hand at work.)

(3) The identifying folios. The stretch of the manuscript for which the present scribe is responsible, is usually identified by folio or page number. For manuscripts that are unfoliated, reference is made sometimes by text, sometimes by counting the frames on the microfilm from which the manuscript was examined.

(4) Codicological or other relevant information, and references to printed editions. This category includes some very miscellaneous material, and from one manuscript to another the coverage is very uneven. Consistency of annotation seems here no virtue at all: if information relevant to the dating or provenance or descent of the manuscript was available then it was included in the summary description, along with anything else that might have a bearing on the language or hand. Printed versions have been cited in so far as they have been consulted, but no attempt has been made to supply a bibliography of editions. Works commonly referred to are cited in abbreviated form, though in principle abbreviations have been used only sparingly; a key is printed on pp. 000–00 of this volume.

(5) The sample of text represented by the analysis, if any. Reference is normally by page or folio number, sometimes by text or microfilm frame. If the analysis depends on a printed edition, the page numbers (and line references) of the printed version may be used. Where an edition is cited, but the analysis has been made from either the manuscript or a facsimile of it, the formula ‘Analysis from original’ is used. If no edition is cited, the analysis can be assumed to have been made from the original or a photographic copy.

(6) The key-number of the linguistic profile (LP), if a profile has been incorporated in the Atlas.

(7) The grid reference at which the LP, if incorporated, has been entered on the maps. Note that some LPs lack a following grid reference: most these belong to the northern group which cannot at present be localised, but which is incorporated in the County Dictionary (‘NME’).

(8) The county or other specified area of origin. The county sections are ordered alphabetically by county name; first printed are the counties of England, then of Wales, then of Ireland, finally of Scotland. The counties are the modern descendants of the mediaeval ones, substantially as they were just before the local government reforms of 1974 (England and Wales) and 1975 (Scotland); but Ely, the Soke of Peterborough, Middlesex and London are separately treated. The ridings of Yorkshire are assigned separate sections, and so also is the City of York. Sections relating to areas that cannot be described by county names are printed after the county sections for the relevant country of origin (London and York are among the county sections): so, in the English part, ‘East Anglia’, ‘Midland’ and ‘Northern’ appear after the last county, ‘Yorkshire, West Riding’.

By ‘London’ here is meant strictly the mediaeval city and Westminster: it is with that small area that the texts have been associated. However, largely for reasons of space required to accommodate the data, the LP locations (defined by grid reference) have been placed more widely, and London is represented on the maps in volume II by the boundaries of the old County of London. (For discussion of the problems of fitting texts associated with cities and cultural centres, see the General Introduction, 2.3.7, in this volume, p. 12.)

The Irish counties are the modern ones, which, for the southern and eastern parts, correspond roughly to the mediaeval divisions (see the map opposite p. 408 in A.J. Otway-Ruthven, A History of Medieval Ireland (London: Ernest Benn, 1968; 2nd edn. 1980)). The designation ‘Pale’ has been used when the language of a Hiberno-English manuscript appears to belong somewhere within the ‘obedient shires’ of Louth, Meath, Dublin and Kildare, but at present cannot be localised more precisely than that. (The Pale as defined by Poynings Act of 1494 is more confined, though it is the core of the area here so described.)

For many of the writings here considered, the probable area of linguistic origin extends across one or more county boundaries. Except for those writings in this class that have been entered on the maps––and the necessary reservations are duly entered in the summary descriptions–– multiple entries with appropriate cross-references have been made. Thus, in the repository form of the Index of Sources the probable area of origin is stated as two or more counties; in the county form of the Index, a separate entry is made in the sections devoted to each of the eligible counties, with cross-references from each such entry to the others.

The categories of information and their ordering follow those for the literary manuscripts, subject to limits imposed by the character of the documents themselves, and the circumstances of their preservation.

(1) Repositories. Most documents noticed here are deeds on single sheets of parchment or vellum. In a professionally organised archive, each deed normally has its own, uniquely defining, reference mark: identification is as easy as for a codex in a manuscript library. In private muniments, things are often somewhat different. Receptacles ranging from tin trunks to hatboxes are the familiar storage, and beyond the property name on the lid––if any––guidance as to the whereabouts of any particular document in the archive is commonly lacking. (Pending classification, some of the private deposits in local record offices are similarly preserved.) In cases like these, when documents are unnumbered or stored unsystematically, identification of the container affords the only means of delimiting a search. Hence such whimsical (and doubtless temporary) labels as ‘Henry Wilson cigar case’ (Yorkshire Archaeological Society, Leeds) and ‘F.S. Cleaver Prize Medal Toilet Soaps Box’ (Levens Hall), besides the more pedestrian ‘Winfell ––Old Deeds packet’ and ‘Bundle ‘Kendal & Kirkby ––3parcels of Ancient Deeds’’.

(2), (3) Hands, etc. In relatively few cases is a document, as the term is used here, the work of more than one hand. Accordingly, the sometimes elaborate accounts necessary for literary manuscripts are out of place, and a single hand is to be assumed unless otherwise stated.

(4) Contents. The contents of literary manuscripts have been only irregularly noticed. For documents, however, content is an important means of identification, particularly in view of the archival and custodial vicissitudes to which they are subject. Moreover, content bears directly on local origins, in a way that seldom obtains with literary productions. The descriptions are hence fairly detailed as a matter of principle, even though their quality is somewhat variable and they fall far short of calendaring. (See further, Descriptions of documents: introductory, above.)

(5) Sample analysed. For documents, it can be generally assumed that analyses are by intention exhaustive. Most texts are fairly brief anyway, and that they usually afford explicit evidence for their local origins is of itself sufficient reason for thorough scrutiny.

(6) County of origin. For most literary manuscripts, ‘county of origin’ is to be understood as ‘county of linguistic origin’; only rarely is there anything explicit to link a manuscript with its place of production. Documents, by contrast, are normally of stated local origins, and so fall naturally into the county sections of the present Index. Certain reservations, however, are still necessary. Firstly, documents written in standard English or something approximating to it do not naturally find a place among the county sections of a linguistically organised inventory. In so far as they are entered at all, it is on the basis of their diplomatic credentials, not their language. Such documents as are dated or made or sealed or written at a named place have at least a prima facie claim for inclusion in the relevant county section; but a document that is merely in the name of someone from a known place has an altogether more tenuous connection, and need never have been anywhere near the county in question. So, especially the Early Proceedings in Chancery, q.v. Secondly, a document in markedly local language may yet be the product of a man from outside the neighbourhood, even the county, to which the document itself relates. When such cases are recognised, they are duly reported. If necessary, a double entry appears in the County List, one for the county of testimonial origin, one for the county of apparent linguistic origin. Likewise, documents that relate equally to persons and places in more than one county are entered in each of the relevant county-sections.

It sometimes happens that a series of documents which from a linguistic point of view can be regarded as a single oeuvre, makes up a consecutive run of numbers in the reference system of a single repository. In such a case, there appears but a single entry in the Index of Sources. It is therefore necessary to read the entry preceding a supposed lacuna before concluding that a document is absent from the list. A breach in the continuity of the reference numbers requires an additional entry, even though none of the intervening items is recorded in the Index. For example, in the PRO series:

SC 1/46/50, SC 1/46/51, SC 1/46/52, SC 1/46/54, 55 and 56, SC 1/46/58, SC 1/46/64 and 65, SC 1/46/70

separate text blocks are required for 50, 51, 52, 54–6, 58, 64–5, 70, regardless of the fact that numbers 53, 57, 59, 60, 61, 62, 63, 66, 67, 68, 69 find no place at all in the Atlas inventory.

Within each county or other regional division of the list, three classes of entry are distinguished. These are organised as separate sub-divisions, and appear in the following order.

‘(1) Sources Mapped’: documents and manuscripts from which the linguistic profiles in volume III are derived. Most of this material is entered on the maps, but included also are the unlocalised sources from the northern counties of England. All of these profiles are incorporated in the County Dictionary, in Volume IV. The order of citation in this sub-division is by linguistic profile (LP) number. LPs conflated as an expedient for mapping are separately entered, with cross-references to their constituent LPs; the constituents appear independently, at their proper places in the numerical list. Items noticed in this first section are, self-evidently, those for which the area of linguistic origin is least in doubt.

‘(2) Local Documents’: items having explicit association with the county or region in question. These are mostly the local documents. Notices are printed in repository order; items known only from published versions appear under ‘Printed:’.

‘(3) Associated Literary Manuscripts’: items which are associated with the county or region solely on the evidence of their language. These are for the most part literary manuscripts, though some documents also appear. It should be noted that there may be several successive entries in the list for the same one manuscript, according as it contains the work of different scribes, or different kinds of language. The order of citation is by repository, as in the preceding sub-division.

Items recorded in categories (2) and (3) contribute neither to the linguistic profiles in volume III nor to the County Dictionary in volume IV.

The order of the sub-divisions for counties and regions is the same as that used in the linguistic profile volume (Volume III) and the County Dictionary (volume IV), but also included are some additional counties and divisions for which there are no printed LPs.

Acknowledgement

The index of documents and manuscripts (Index of Sources) could not have been compiled without the forbearance and good offices of their custodians––archivists, librarians, private owners. They are too many to thank individually, but to every repository here named, there is a debt of gratitude––and not least to the innumerable people who have fetched and carried and so helped with the search for materials on which this atlas now rests. To all of them we offer our sincere thanks.

Bibliography of Abbreviated Titles cited in the Index of Sources

Baildon, W.P. (1896) Select Cases in Chancery, A.D. 1364– 1471. Selden Society, vol. x. London: Bernard Quaritch.

Benskin, M. (1977) ‘ Local archives and Middle English dialects ’ , Journal of the Society of Archivists 5, 500–514.

Benskin, M. (1982a) ‘ The letters <þ> and <y> in later Middle English, and some related matters ’ , Journal of the Society of Archivists 7, 13–30.

Benskin, M. (1982b) ‘ Marian verses from a Hedon manuscript: some new materials for the Middle English dialectology of the East Riding ’ , Revista Canaria de Estudios Ingleses 5, pp. 27–58.

Benskin, M. and Laing, M. (1981) Translations and Mischsprachen. In Benskin and Samuels 1981, pp. 55–106.

Benskin, M. and Samuels, M.L., ed. (1981) So meny people longages and tonges. Philological essays in Scots and mediaeval English presented to Angus McIntosh. Edinburgh: the Editors.

Calendars of the Proceedings in Chancery in the Reign of Queen Elizabeth. London: House of Commons, vol. i, 1827. (Prints early proceedings for the reigns of Richard II, Henry V, and Henry VI.)

Chambers, R.W. and Daunt, M., ed. (1931) A Book of London English 1384–1425. With an appendix on English documents in the Record Office, by M.M. Weale. Oxford: Clarendon Press (repr. 1967).

Cheney, C.R., ed. (1947) A Handbook of Dates for Students of English History. London: Royal Historical Society.

Davis, N. (1952) ‘ A Paston hand ’ , Review of English Studies NS 3, pp. 209–21.

Davis, N. (1954) ‘ The language of the Pastons ’ , Proceedings of the British Academy 40, pp. 119–44.

Davis, N. (Part I, 1971; Part II, 1976) Paston Letters and Papers of the Fifteenth Century. Oxford: Clarendon Press.

Davis, N. (1983) ‘ The language of two brothers in the fifteenth century'. In Stanley and Gray (1983), pp. 23–8

Flasdieck, H.M., ed. (1926) Mittelenglische Originalurkunden, 1405–1430. Mit Einleitung und Anmerkungen hrsg. von H.M.F. Alt.- u. Mitteleng. Texte, 11. Heidelberg: Carl Winter.

Gray, D. and Stanley, E.G., ed. (1983) Middle English Studies presented to Norman Davis in honour of his seventieth birthday. Oxford: Clarendon Press.

Guide to the Location of Collections described in the Reports and Calendars Series, 1870–1980. (Royal Commission on Historical Manuscripts. Guides to Sources for British History, based on the National Register of Archives, 3. London: H.M.S.O., 1982).

Historic Society of Lancashire and Cheshire: Proceedings and Papers (1849–54). Liverpool. Continued as Transactions of the Historic Society of Lancashire and Cheshire. Liverpool and London.

Ives, E.W. (1968) ‘ The common lawyers in pre-Reformation England ’ , Transactions of the Royal Historical Society 5th ser. pp. 145–73.

Johansson, S. and Tysdahl, B., ed. (1981) Papers from the First Nordic Conference for English Studies (Oslo, 17–19 September 1980). University of Oslo: Institute of English Studies.

Ker, N.R. (1957) Catalogue of Manuscripts containing Anglo-Saxon. Oxford, Clarendon Press.

Kirby, J.L., ed. (1978) Calendar of Signet Letters of Henry IV and Henry V. London: H.M.S.O.

Kurath, H. and Kuhn, S.M. et al., ed. Middle English Dictionary. Ann Arbor: University of Michigan Press, 1956–. Plan and Bibliography, 1954; updated, 1984.

Laing, M. (1978) Studies in the Dialect Material of Mediaeval Lincolnshire. Diss. Ph.D., University of Edinburgh; 2 vols., unpubl.

Legge, M. Dominica (1941) ‘ Anglo-Norman and the Historian ’ , History 26, pp. 163–175.

McIntosh, A. (1983) ‘ Present indicative plural forms in the later Middle English of the north Midlands'. In Gray and Stanley 1983, pp. 235–44.

Moody, T.W., Martin, F.X. and Byrne, F.J., ed. (1984) A New History of Ireland vol. IX: Maps, Genealogies, Lists (being Part II of A Companion to Irish History). Oxford: Clarendon Press.

Morsbach, L., ed. (1923) Mittelenglische Originalurkunden von der Chaucer-Zeit bus zur Mitte des XV. Jahrhunderts, in der grossen Mehrzahl zum Erstenmal veröffentlich von L.M. Alt.- u. Mitteleng. Texte, 10. Heidelberg: Carl Winter.

Myers, A.R. (1937) ‘ Parliamentary petitions in the fifteenth century ’ , English Historical Review 52, pp. 385–406, 590–613.

Ormerod, G. (1882) The History of the County Palatine and City of Chester, 2nd. edn. rev. Thomas Helsby. London: George Routledge, 3 vols.

Otway-Ruthven, A.J. (1968; 2nd edn. 1980) A History of Medieval Ireland. London: Ernest Benn.

Powicke, F.M. and Fryde, E.B., ed. (1961) A Handbook of British Chronology. London: Royal Historical Society, 2nd edn.

Pugh, R.B., ed. (1947) Calendar of Antrobus Deeds before 1652. Devizes: Wiltshire Archaeological and Natural History Society.

Record Society of Lancashire and Cheshire 116. Letters and Accounts of William Brereton of Malpas, ed. E.W. Ives. Chester, 1976.

Richardson, H.G. (1937) ‘ Richard II's last parliament ’ , English Historical Review 52, pp. 39–47.

Roskell, J.S. (1937) The Knights of the Shire for the County Palatine of Lancaster (1377–1460). Manchester: Chetham Society, Remains Historical and Literary connected with the Palatine Counties of Lancaster and Chester, NS 96.

Samuels, M.L. (1963) ‘ Some new applications of Middle English dialectology ’ , English Studies 44, pp. 81–94.

Samuels, M.L. (1972) Linguistic Evolution, with special reference to English. Cambridge: Cambridge University Press.

Samuels, M.L. (1981) ‘ Spelling and dialect in the late and post-Middle English periods'. In Benskin and Samuels 1981, pp. 43–54.

Sandved, A.O. (1980) ‘ The rise of Standard English'. In Johansson and Tysdahl 1981, pp. 398–404.

Sandved, A.O. (1981) ‘ Prolegomena to a renewed study of the rise of Standard English'. In Benskin and Samuels 1981, pp. 31–42.

Stanley, E.G. and Gray, D., ed. (1983) Five Hundred Years of Words and Sounds. A Festschrift for Eric Dobson. Cambridge: D.S. Brewer.

Storey, R.L. (1969) Thomas Langley and the Bishopric of Durham. London: S.P.C.K.

Walmisley, C.A. (1927, 1928) Index of Persons Named in Early Chancery Proceedings, Richard II, 1385, to Edward IV, 1467, preserved in the Public Record Office, London. London: Harleian Society, publications lxxviii and lxxix.

Counties and regions used in the Index of Sources
Bedfordshire	Middlesex	Yorkshire, West Riding	Limerick
Berkshire	Norfolk	East Anglia	Louth
Buckinghamshire	Northamptonshire	Midland	Meath
Cambridgeshire	Northumberland	Northern†	Pale
Cheshire	Nottinghamshire	Isle of Man	Tipperary
Cornwall	Oxfordshire	Wales	Waterford
Cumberland	Rutland	Cardigan	Wexford
Derbyshire	Shropshire	Carmarthen†	Scotland
Devon	Soke of Peterborough	Carnarvon	Angus
Dorset	Somerset	Denbighshire	Ayrshire†
Durham	Staffordshire	Glamorgan	Berwickshire
Isle of Ely	Suffolk	Merioneth	Dumfriesshire
Essex	Surrey	Monmouth	Dunbartonshire
Gloucestershire	Sussex	Montgomery	East Lothian
Hampshire	Warwickshire	Pembrokeshire	Lanarkshire†
Herefordshire	Westmorland	Ireland	Midlothian
Hertfordshire	Wiltshire	Clare	Peeblesshire
Huntingdon	Worcestershire	Cork	Perthshire†
Kent	City of York	Down	Roxburghshire
Lancashire	Yorkshire†	Dublin	Selkirkshire
Leicestershire	Yorkshire, North West†	Galway	Stirlingshire
Lincolnshire	Yorkshire, North Riding†	Kildare	Wigtownshire†
London	Yorkshire, East Riding	Kilkenny	Unlocalised Soures


An italicised name indicates that the items noticed under this head do not contribute to the linguistic volumes III (LPs and IV (County Dictionary).		† indicates that none of the LPs noticed under this head contribute to the maps in volume II.