Archbishop Wulfstan enjoyed a high reputation as a stylist amongst his contemporaries; when he was still bishop of London (996–1002) one correspondent spoke of the ‘very sweet wisdom of [his] eloquence and the richness of [his] composition fittingly organised’, whilst the wide dissemination of his sermons and their susceptibility to imitation bear dual witness to his popularity throughout the eleventh century.

Early English medicine, including that of Anglo-Saxon England, has been frequently described as a backwater in which superstition flourished until the mainstream of more rational and advanced Salernitan practices flowed into the country in late medieval times. This view has been strongly countered by C. H. Talbot, who has shown, convincingly I think, that works usually attributed to Salernitan writers were in use in England at least as early as they were at Salerno, and that, far from being backward, English medicine in Anglo-Saxon times may have been in the forefront of contemporary practice.

This discussion of Anglo-Saxon coinage attempts to look beyond the detail of numismatic classification in order to consider the relationship between the underlying variations and the economic life of the times. Those parts of it which deal with the classification of the coinage and analyse the observed metrology are intended to be a critical summary of the numismatic research carried out in the past thirty years. Other parts, in which I seek to relate the metrology to such documentary evidence as is known to me – and thus trespass across the vague dividing line between numismatics, of which I have some knowledge, and economic history, of which I have little – are aimed at stimulating awareness and discussion of the problems involved. Finally, a section is devoted to numismatic methods because it is important that their use and limitations be generally understood.

It is a matter of general recognition that Wulfstan, bishop of London from 996 until 1002 and archbishop of York from 1002 until his death in 1023, was responsible for the formulation of the series of law codes issued by Æthelred II, most likely between the years 1008 and 1014. For many years, however, the prevailing opinion among scholars was that the code of Cnut, which borrowed extensively from the 1008 code of Æthelred, was not produced until after Wulfstan died. This opinion was based on arguments set out by Liebermann in his article on relations between Wulfstan and Cnut, which he later reproduced in abbreviated form in the Gesetze: Liebermann believed that Cnut's code must have followed his proclamation of 1027.4 Such a dating would, of course, preclude the direct involvement of Wulfstan in its compilation.

It has long been recognized that the early sections of the so-called Historia Regum, a work attributed to Symeon of Durham (ob. c. 1130) and preserved uniquely in Cambridge, Corpus Christi College 139 (written c. 1164 at Sawley, Lancashire) at 5 3v–130v, originally constituted a separate work, probably composed in the pre-Conquest period and subsequently incorporated into the Historia Regum. Thomas Arnold, who edited the Historia Regum for the Rolls Series in 1885, was persuaded ‘that the more attentively any experienced person may study the curious document between pages 14 and 94 [of the edition], the more firmly will he be convinced that it is a composition of the tenth…century’. His conclusions were based on the Latin style of the work, which he regarded as ‘pretentious and bombastical on the one hand, obscure and ineffectual on the other’ and which affiliated the work, in his opinion, with other Anglo-Latin works of the tenth century. Because he believed that certain passages in the work betrayed an origin in the congregation of St Cuthbert (then at Chester-le-Street), Arnold referred to the compiler of the early sections of the Historia Regum as the ‘Cuthbertine’. His conclusions appear to have been accepted by later historians; for example, W. H. Stevenson (who referred to the early sections of the work as SD 1) wrote as follows: ‘we may readily grant that SD 1 was an older compilation, but the evidence that it was drawn up in the tenth century is, in the absence of a MS of that period, necessarily hypothetical’. No such manuscript has yet come to light, but in recent times Arnold's postulation of a tenth-century origin for the early sections has been accurately and comprehensively reinvestigated by Peter Hunter Blair. By a series of detailed stylistic arguments Hunter Blair has been able to show that the first five sections of the Historia Regum (occupying pp. 3–91 of Arnold's edition) may reasonably be regarded as the work of oneauthor. These five sections are as follows: (1) Kentish legends, particularly pertaining to the Kentish martyrs Æthelberht and Æthelred (pp. 3–13); (2) lists of Northumbrian kings (pp. 13–15); (3)material derived from Bede, particularly the Historia Abbatum (pp. 15–30); (4) a chronicle from 732 to 802 (pp. 30–68); (5) a chronicle from 849 to 887, based mainly on Asser (pp. 69–91). Hunter Blair also recognized that two passages had been interpolated at a later date into the material of these first five sections: one concerning the relics of Acca of Hexham (pp. 32–8), the other concerning those of Alchmund, also a bishop of Hexham (pp. 47–50); he reasonably suggested that these interpolations were added at Hexham in the early twelfth century. As to the date of compilation of the five early sections Hunter Blair was able to affirm, albeit cautiously, Arnold's suggestion of a tenth-century date, but he concluded that ‘in the end judgement will perhaps rest upon opinions about [their] latinity’.

Murder by fellow Christians for secular motives may seem to us an improbable qualification for sanctity. To the Anglo-Saxons, however, the matter evidently appeared differently, for several members of pre-Conquest royal families seem to have been regarded as saints chiefly because of their violent ends at the hands of Christian assassins. The lives and cults of some of these murdered royal saints are well attested in pre-Conquest sources. Others, however, are known chiefly from post-Conquest texts which are mainly hagiographical in character and which were written considerably later than the events which they describe. For these reasons these latter texts have not generally been taken seriously as historical sources and the saints who figure in them have been largely neglected by historians. The hagiographical character of these texts, however, should not exclude them from historical consideration. They may preserve details of the saints’ careers which can be accepted as fact, and the accounts of miracles and wonders which they contain, although not to be treated as factual narratives, provide evidence bearing on the veneration of these saints. As for the date of composition, there is always a possibility that late medieval hagiographical texts represent stylistic and literary modifications of much more ancient versions, the former existence of which may in some cases be revealed by close analysis of the extant texts. The evidence relating to the whole group of murdered royal saints said to have been venerated in pre-Conquest England must therefore be evaluated before conclusions about the significance of this type of sanctity can be drawn. In what follows the saints in question will be considered in order of the apparent validity of the evidence pertaining to them, so that those whose existence and veneration are less certain can be examined in comparison with more firmly documented examples.

Scribal errors involving proper names appear throughout the manuscript of Beowulf. The scribes misrecognized names, converted them into common nouns of similar appearance, and spaced them into units that reflect incomprehension. This study assesses the implications of these errors for the engagement and awareness of the scribes, then considers their bearing on the dating and editing of Beowulf. The proper name errors reveal that the scribes were unfamiliar with the heroic-legendary traditions constituting Beowulf. The collective presence of these errors supports the probability that the extant manuscript of Beowulf is a copy of a centuries-old poem, not of a recent composition.

Cambridge, University Library, Gg. 5.35 is known to scholars principally for the contents of fols. 432–41, the lyric anthology known (somewhat mis-leadingly) as the ‘Cambridge Songs’. Important though this group of poems is for the history of the Latin lyric, it has diverted attention from the contents of the manuscript as a whole, which presents a remarkable range of texts, mainly poetic, from the Early Christian, Carolingian and Anglo-Latin periods. Physical description, moreover, has generally been confined to the section containing the ‘Cambridge Songs’ and consequently the method of compilation of the whole codex has been neglected. The Cambridge University Library catalogue, for instance, gives the impression of forty-four works entered sequentially in the manuscript, and leaves unexplained such curiosities as no. 41 ‘Prose treatise on medicine’ and no. 44 ‘Certain medical prescriptions’, which appear to sandwich between them the ‘Cambridge Songs’ and another group of ‘hymns’. This article describes the compilation as a whole, its physical appearance, its genesis, and its contents in detail.

The range of cultural assumptions underlying Old English poetry and the sentiments and sources which inspired it, not to mention the methods of composition – oral or written – are subjects of vigorous dispute. Much that was once considered indigenous and Germanic is now being revalued in the light of Latin analogues and works accepted for a century as secular are being treated by many as Christian doctrinal discourses. Since the intellectual milieu is so uncertain a basis for interpreting controversial poems, other approaches are needed. One that suggests itself is the exploration of idea-complexes. When a theme or situation recurs in a number of poems, in widely differing contexts, patterns of association can be isolated and analysed. The theme of exile, for example, is a centre of a cluster of ideas. Another is the concept of the hall: what is looked to for safety and what is feared as a threat to that security make apt points of departure for a study of a culture's major assumptions. Moreover, because the hall is the focus for conflicting attitudes, the array of associations proves useful to a more general understanding of the nature of Old English poetry.

In recent years there has been a great increase in both excavation and research in the field of pagan Anglo-Saxon archaeology. Yet much of the literature remains so detailed and specific for a non-specialist it resembles a maze without obvious clues. A recent book, The Archaeology of Anglo-Saxon England, dealt with many aspects of later Anglo-Saxon archaeology but did not cover some of the topics which have been central to study of the pagan period in recent years. This article is an attempt at an outline of some of those topics. The subject falls into two interrelated parts. First there is the course, date and nature of the Anglo-Saxon settlement of England, involving consideration of the continental background, both late Roman and Germanic, and assessment of the significance of the earliest Germanic finds in England. The second part is concerned with the material culture of early Anglo-Saxon England, chiefly as reflected in the cemeteries of the period. Although I discuss settlement patterns in general, I do not examine individual settlements or house types in detail, because this subject has been dealt with twice recently, once by Addyman in this publication and once by Rahtz. I refer only occasionally to documentary records, since my intention is to present the archaeological evidence to non-archaeologists.

The main part of this paper will deal with the typescene of the beasts of battle, and will describe its content and use as rigorously as possible in formulaic terms. In particular, the relationship between the conventional and the original will be examined in order to show that the scene is more traditional than has been thought and that many of the variations occur within conventional parameters. The term ‘typescene’ has been used with too much flexibility by Anglo-Saxonists, and will here be used with the meaning outlined by G. S. Kirk:

In failing to abide by the chief rule of their genre, the Enigmata of Aldhelm have themselves become something of a riddle: for why should their author give, in the form of a title word or phrase, the solution to each of the one hundred riddles? By violating from the start the essential quality of any riddle – a seeming impossibility of solution – Aldhelm would seem to neglect the point made in this Swedish riddle about riddles: ‘When one does not know what it is, then it is something; but when one does know what it is, then it is nothing.’ Certainly the work of Symphosius provided Aldhelm with a precedent for including solutions to his riddles; but this fact in itself does not explain very much about Aldhelm's practice, for it ignores the pleasure he took in exploring the possibilities of the riddle as a genre. Since his work as a whole demonstrates a taste for the difficult and esoteric, it is not surprising that his Enigmata should pose this riddle of their own. His practice in this regard is all the more intriguing because very few of the enigmata are difficult to solve, even without their titles. They differ quite markedly from the riddles of the Exeter Book, many of which are insoluble. In the paradoxical fashion so beloved of riddlers, the mystery of the enigma is also the important clue. That is, the title word or phrase offers in the obvious sense a solution; but in a more esoteric, linguistic sense it may also present a riddle of its own.

In a recent article, Michael Lapidge has cautioned the scholarly world not to assume too readily that any glossed medieval manuscript was necessarily used in the schoolroom. Identical glosses, he points out, often appear in several manuscripts, forcing us to conclude that glosses were either copied along with the text or from a commentary. The glosses naturally show greater variety than the texts because the scribe could at will omit them, change them, or add to them. Nonetheless, the fact that the same glosses appear in several manuscripts indicates that they were not the spontaneous reactions of a teacher or student.

More than half of the extant manuscripts from Anglo-Saxon England, both vernacular and Latin, contain Greek. How much Greek did the early English know? M. L. W. Laistner accepted only a handful of early authors, Bede among them, as ‘competent Hellenists’. Bernhard Bischoff, too, noted that among the numerous witnesses to Greek writing in the medieval West, only a few show knowledge of the language itself, and the majority in their corrupt state suggest just the opposite; moreover, he points out, their function is very often liturgical. By the same token, a recent survey of the rich Greek materials from Sankt Gallen makes the general observation that ‘few medievals possessed an ability to read Greek prose, an ability based on, at least, an acquaintance with the elementary principles of grammar’. For a number of years, I have been compiling a catalogue of Anglo-Saxon manuscripts containing Greek, and on the basis of what I have seen, these various assumptions – that much of the Greek was badly copied, that its vocabulary was largely ecclesiastical or liturgical, that such a vocabulary would necessarily repeat itself, yielding therefore perhaps no more than some 500 to 800 Greek words, and that knowledge of Greek grammer (declensions, inflexions and so forth) was minimal – need major modification. In what follows, I shall examine these various assumptions in turn.

Over a hundred years ago, T. O. Cockayne published his Leechdoms, Wortcunning and Starcraft of Early England, in which he edited the texts of all the medical writings in Old English he could find. It is massive in its scope, and no modern scholar is ever likely to produce its equal. Yet we, metaphorically standing on Cockaynes's shoulders, and equipped with aids provided by more recent research, are able to examine more closely than he could some of the special features of the field which he revealed to us. Its English substrata have been comparatively neglected, however, and therefore I propose in this paper to examine closely the relationships of the hundred or so medical remedies in Old English which have been preserved – usually in different manuscripts – in two or three versions so close that it is obvious, even on a superficial view, that they either derive from the same English original, or are copied the one from the other. These remedies usually begin by specifying the ailment for which they are recommended, and then go on to set out the ingredients and method of making the appropriate herbal concoction. Nearly all the repeated remedies are found at least once in the Leechbook manuscript, now London, British Library, Royal 12. D. xvii, and so I will begin by describing it, and use it as the basis of the argument. Then I will describe briefly in turn the other manuscripts in which the remedies are found, discussing as I proceed those with minor parallels to Bald's Leechbook; and then, separately and in detail, the important duplications in the two final manuscripts under consideration. It may thereafter be possible to draw some conclusions about the method of compilation of Bald's Leechbook.

Eduard Norden's great book Die antike Kunstprosa is grounded on first-hand acquaintance with an astonishingly wide range of literature, both from classical antiquity and from the Middle Ages. But at the authors of Anglo-Saxon England Norden does seem to have drawn the line. ‘The two great writers, Aldhelm and Bede’, he says, ‘write, like all Anglo-Saxons, a stylistically uncultivated (verwildertes) though grammatically correct Latin.’ There is no need to labour the point that Aldhelm and Bede are not to be mentioned thus cavalierly in the same stylistic breath: we are all familiar today with the distinction between the ‘hermeneutic’ Latin of the one and the ‘classical’ Latin of the other. But at least Norden could not fall victim to another widely accepted doctrine that purports to explain the origin of that distinction: the doctrine that Aldhelm's style was influenced by Ireland, Bede's by the continent of Europe. I doubt if this is true even of Bede. But my present business is with Aldhelm; I shall try to show that his literary origins are not to be found in Ireland. At the same time I shall be challenging Norden's claim that his Latin was uncultivated. I shall suggest, indeed, that its cultivation was of a kind that Norden himself would have been uniquely qualified to analyse.

The theories of oral-formulaic composition advanced by Albert B. Lord, his mentor and collaborator Milman Parry, and their later twentieth-century followers have been adduced frequently in studies of Old English verse, elements of whose language must go back ultimately to an oral tradition. After decades of research, however, scholars have yet to find conclusive answers to some basic questions: did literate Anglo-Saxons continue to practise techniques of extemporaneous versification? If so, did they continue to develop the mnemonic skills attributed to oral poets? It is clear that the monuments of Old English verse reveal many examples of formulaic language (for example, se mæra maga Healfdenes, se mæra mago Healfdenes and se mæra maga Ecgðeowes); but should we regard this language as a reliable witness to oral-formulaic versification or, perhaps, as a hybrid, ‘literary-formulaic’ idiom? Finally, if we accept the synchronic (or achronic) models of the formulaic ‘word-hoard’ that inform many Old English studies, is it pointless even to speculate about poetic influence, direction of borrowing and similar concerns? If so, how should we regard, say, two parallel uses of the unusual phrase enge anpaðas, occurring verbatim in Beowulf and the poetic Exodus but nowhere else among the surviving monuments? Must we view these parallels as isolated outcroppings in the trackless expanse of the Old English poetic corpus? Largely as a result of the scarcity of verse preserved in multiple copies, such questions have remained debatable into the present century.

The Taunton Fragment (now Taunton, Somerset, Somerset County Record Office, DD/SAS C/1193/77) consists of four leaves containing portions of brief expositions or homilies on the pericopes for four successive Sundays after Pentecost. In the Fragment, brief passages in Latin regularly alternate with the Old English translations of these passages. The manuscript to which the four leaves once belonged was written probably at some point around or after the middle of the eleventh century in an unknown (presumably minor) centre in Anglo-Saxon England. Until recently, the existence of the Taunton leaves had escaped the notice of Anglo-Saxonists; the texts which they contain are printed here for the first time. It will be obvious that eight pages, half of which are in Old English prose, add in no negligible way to the corpus of Old English. Through analysis of the texts in the second part of this article, I hope to show that their contribution to our knowledge of various kinds of literary activity in Anglo-Saxon England is significant indeed, and that the linguistic evidence they present has no parallel elsewhere in the corpus of Old English.

Slaves were an integral and numerically important part of English society in the Anglo-Saxon period. They appear in the earliest English law code promulgated between 597 and 616 by Æthelberht of Kent; nearly half a millennium later at the beginning of the Norman age their continued widespread presence in English society is attested by Domesday Book. Yet they do not seem to have excited much attention from scholars. The longest treatment in print remains that by Kemble, which was written over a century ago. Stenton in his magisterial survey of Anglo-Saxon England made only four references to them. Some other recent histories, however, have discussed slavery in more detail. Professor Whitelock rightly included slaves in her analysis of the social classes of England up to the time of the Norman Conquest. H. P. R. Finberg took this further in his agrarian history of Anglo-Saxon England by dividing the society into three chronological periods and examining the regional variations within England during those periods. Both works mention the slave trade. This receives a more detailed discussion in H. R. Loyn's economic and social history. But the evidence on slavery in England is mostly fragmentary and in widely scattered sources. Inevitably general histories of the period but skim the surface. Only by patiently assembling all the evidence, as Professor Verlinden has been doing for many years in his studies on slavery in continental Europe, can knowledge about this significant element in English society be advanced.

‘The Anglo-Saxon glosses are part of the Anglo-Saxon literary heritage.’ This oracular pronouncement of Professor Stanley's has been elucidated by Michael Lapidge's recent article on the school of Canterbury, to which mine is in some ways complementary. Lapidge makes a strong case for his view that the ‘original English collection’ of glossae collectae in the Leiden Glossary and other continental glossaries was compiled in Canterbury under Archbishop Theodore (669–90) and Abbot Hadrian (671–709 × 10) and transmitted to continental centres of Anglo-Saxon missionary activity during the eighth century, and he emphasizes its importance as ‘a wonderful treasury of evidence for the books which were known and studied in early England’. One piece of evidence that he adduces for the English origin of this collection is the fact that batches of glosses derived from it are found in the Epinal–Erfurt and Corpus glossaries, which establishes that the collection was already in existence and in England when the Epinal–Erfurt Glossary was compiled between c. 675 and the end of the seventh century, and was still there intact when the compiler of the Corpus Glossary used it a century or more later.