The Heroic Age

A Journal of Early Medieval Northwestern Europe

Issue 10—Saints and Sanctity (May 2007)   |   Issue Editors: Celia Chazelle & Deanna Forsman

Founded 1998   |   ISSN 1526-1867

Issue Navigation

Issue Homepage

Relics, Authority and Domestic Space

Miracle Stories

Recapitation in Irish Hagiography

Boniface's Booklife

Cult of St. Guthlac

P-Celtic Place-Names

Roman Imports

Revelatio Ecclesiae

Forum—Historicity and Historiography of Arthur

Electronic Medievalia

Continental Business

History by Biography—St. Æthelthreda

History by Biography—St. Elisabeth



Ashley, Mike. 2005. The Mammoth Book of King Arthur. London: Robinson. 670 pages. 184119249X.

As the title suggests, this book is indeed "mammoth." That alone can make it intimidating for the average reader. Ashley compiles an impressive array of material, which he divides into discussions about the historical Arthur, the Arthurian legend, and the modern Arthurian novels and films. Tables, charts and maps are sprinkled liberally throughout the text. The book is very useful as a resource for the variety of types of information available on various topics; however, this recommendation comes with the caveat that the reader needs to already know enough about the topic in order to determine which sources are considered solid and which are flights of modern fancy. Ashley provides minimal source criticism for the reader when he provides any at all, and works of debatable scholarly value are given the same weight as solid academic resources.

The first section of the book is the most problematic, which is not surprising, since in it Ashley attempts to cover all of the arguments about every historical figure who has ever been presented as having an impact on the Arthur of legend. An enormous topic, simply assembling and presenting the data is a great service to scholars who work in this field. Ashley's presentation, though, runs into several problems that make his effort of questionable assistance to readers who are not already familiar with the various candidates for the historical Arthur. Ashley presents with equal weight candidates who have solid historical evidence behind their presence on the list, candidates who may be fictional, and candidates whose presence on the list results from forged evidence or other invalid sources of information. Ashley's basic assumption, presented in the Preface (xiv), is that the only time period in which a valid historical source for the Arthurian tradition can exist is "between the end of the Roman administration of Britain, a date usually assigned as 410 A.D., and the emergence of the Saxon kingdoms, which were taking a strong hold by the start of the seventh century." Although Ashley does give some coverage to historical figures who fall outside this arbitrary time frame, he dismisses these figures as unlikely or problematic at best because they do not meet the artificial criterion that he imposed. This leads to all sorts of odd reasoning, such as, "if Arthur really was a king, then he must have ruled one of" the fifth-century British kingdoms (xvii). First, early sources, such as Nennius, make a point of saying that Arthur was not a king. Second, there is absolutely nothing that prevents Arthur from belonging to a period outside the fifth century except Ashley's decision to limit him to that stretch of time. Ashley's methodology becomes extremely questionable when he addresses the genealogies and king lists. Although he does start with the "huge caveat that of all the sources covered in the next few chapters, these are amongst the most unreliable" (47), he then goes on to assign figures on these lists dates at roughly 20 year intervals—except when he needs a shorter or longer span to make his preferred figures fall within the span of time he wants them to have lived. Ashley blithely includes mythological figures, such as the deity Beli, on these lists along with figures who certainly existed, without explaining his reasoning for doing so. Anyone who studies history knows that more than one ruler can reign in the same year (Rome in 69 and 193 C.E. leaps to mind) or that a single ruler can reign well beyond twenty years (Queen Victoria is one of many possible examples). Yet Ashley constructs his hypothetical time line of rulers and then accepts or dismisses candidates for the historical Arthur on the basis of whether or not they are compatible with this rather fanciful guideline he has manufactured. As a result, Ashley winds up presenting an unattested "Arthur of Badon" as his candidate for the historical King Arthur. Ashley would have been better off simply presenting the cases for and against the various candidates who have already been advanced—something that he periodically tries to do in this section.

The survey of texts that forms the central portion of the book is comprehensive and fairly well laid out. Ashley occasionally goes into detail about why he has assigned certain dates to various manuscripts, but sometimes his reasoning appears to be something of a mystery unless the reader already happens to know which scholar he is following. The same inconsistent presentation of reasoning plagues other aspects of this section, with the question of why he assigned authorship of some manuscripts the way he did or why he chose to present a certain configuration of the relationship among the manuscripts often remaining unclear. Additionally, Ashley occasionally dismisses some schools of thought, such as the one that proposes that Guiot de Provins was Wolfram von Eschenbach's "Kyot" (352), without explaining why he has chosen not to accept a particular argument. This section also contains the disturbing misstatement that the legendary warriors of Ossetian lore were the Nartyamonga (304). These warriors are actually called "Narts", and the "Nartyamonga" is the "Cup of the Narts," the Ossetian parallel to the Arthurian Holy Grail. Such sloppiness in one place makes me wonder about the accuracy of Ashley's statements elsewhere in the text and is another reason that this book should be approached with extreme caution.

The most approachable part of the book is the third section, which surveys novels, films and other Arthurian-inspired texts of the post-Victorian period through the present day. The only real problem I noted in this section is that some of the information presented is incomplete. For example, Barbara Ferry Johnson's novel is listed simply as Lionors rather than as Lionors, King Arthur's Uncrowned Queen, which is the full title of the book. The list of Arthurian characters and thumbnail biographies of each is fairly comprehensive, and the gazetteer of Arthurian sites provides a quick reference for most of the sites mentioned in the medieval texts. The listing of websites and societies is a bit incomplete and the format is inconsistent. For instance, some websites are listed with "http://" while others start with "www." I was also surprised to see only eight pages of bibliography for a book that covers so much data.

Problematic though the book is, I find myself consulting Ashley's work frequently when I want a quick reference to everything anyone has ever said on a topic, such as where Badon may have been fought or who the historical contenders are for a fifth-century or sixth-century or even seventh-century Arthur. I recommend the text as an addition to the library of anyone who is already well versed in multiple aspects of Arthurian scholarship, but I fear that a novice to the field will come away with numerous inaccurate impressions about a number of topics.

Linda A. Malcor
Lake Forest, CA

Bredehoft, Thomas A. 2005. Early English Metre. Toronto Old English Series. University of Toronto Press. viii + 183 + ii. 080203831X.

Modern study of Old English metre begins with the seminal work of Eduard Sievers in the mid 1880s. Working inductively and comparatively, Sievers proposed a metrical system consisting of five basic "types" or stress-patterns that he argued could be used to account for the metre of almost all lines in the corpus of the earlier Old English poetry. These metrical types, identified by letter in descending order of their frequency, are well known to beginning students of Old English literature through the—slightly problematic—mnemonic reproduced in early editions of Bruce Mitchell and Fred C. Robinson's very widely used textbook, A Guide to Old English (in the following, / represents a full stress, \ a "half-stress", and x "lack of stress") :

Type Mnemonic Stress Pattern
Type A Anna angry / x / x
Type B and Byrhtnoth bold x / x /
Type C in keen conflict x / / x
Type D ding down foemen / \ / \
Type E each one with edge / \ x /

Sievers's system was by no means perfect. While, as the last hundred and twenty years and numerous experts have demonstrated, his types do seem to account for most apparently legal patterns in conservative Old English verse, they do less well in explaining the underlying rules actually used by the poets to produce acceptable Old English metre. Indeed distinctions between legal and illegal stress patterns in Sievers's system can seem somewhat arbitrary from a theoretical perspective. Certain specific kinds of words or syllables are assigned stress—or not—depending on grounds little removed from pure expedience. Some inflectional syllables—the tense marker -od- on weak preterite verb forms, for example—are counted as stressed when required to make up the metre, and considered unstressed when not. Finite verbs early in the clause, similarly, are usually considered unstressed—unless required to be stressed by alliteration or metrical necessity: thus, where grette goldhroden (Beowulf 614a) would be scanned as a Type D line ( / x / \ x ) because the finite verb grette alliterates with goldhroden, eode goldhroden (Beowulf 640b) must be understood as a Type C ( x x / \ x ). In some cases—such the famous type A-3 lines found at the beginning of many Old English poems—well attested metrical patterns seem to defy the rules altogether: while a basic tenet of Sievers's formalism is that Old English metrical types consist of two primary stressed syllables, lines like the opening verse of Cædmon's Hymn (nu sculon hergan) shows what other tenets of the same theory would suggest is but a single stressed syllable (on hergan).

Since Sievers's ground-breaking work, the field of Old English metrical studies has been well-ploughed. Significant new books on Old English metre have been published at least every decade since the 1940s—and for most of this time at a considerably faster clip. Some of these approaches, such as the work of Bliss in the 1950s and early 1960s (Bliss 1958), improved on Sievers's system by regularising some of its theoretical and descriptive anomalies. More commonly, scholars have sought to replace Sievers's account with formalisms based on often completely different theoretical principles (e.g. Pope 1942/1966, Cable 1974, Russom 1987).

The often indifferent success this research has had in gaining widespread acceptance among Anglo-Saxon scholars calls to mind the parable of the sower in Matthew 13:3-23: many seeds have been sown, but only a few have stuck in ground fertile enough to allow a successful harvest. Bliss's refinement of Sievers's work has become the de facto standard in most traditional accounts of Old English metre. Russom's work on foot patterns is probably the only true alternative to Sievers's approach to gain anything like widespread acceptance among non-metricists. Most other accounts seem to have fallen on the rocks, or among the thickets—if not devoured by the fowls awaiting at the side of the road.

Thomas A. Bredehoft's new book, Early English Metre, represents the latest effort to entice fruit from this field. Offering what he claims is a new metrical formalism, Bredehoft attempts in relatively small space (120 pages, with 50 [!] pages of notes) both to address a number of famous theoretical deficiencies in Sievers's metrical system (as reformulated by Bliss) and account for several cultural, literary, and historical aspects of Old English metrical history that he argues are given short shrift in many modern studies.

The book is well-written within the bounds of its subject matter and, while not nearly as theoretically comprehensive as its introduction suggests, does propose a number of significant improvements to our understanding of both the theoretical underpinnings of Anglo-Saxon metre and its later history and use.

Perhaps the most successful and immediately convincing aspect of the book's argument concerns the "new metrical formalism" it proposes to account for the construction of classical Old English verse. In actual fact, this "new formalism" is based heavily on the foot-pattern theory popularised primarily by Russom. With the exception of one major formal improvement (suggested ultimately by Kendall's work on alliteration of finite verbs), scansions proposed in this book for non-hypermetric lines of classical poetry largely agree with those found in Russom's 1987 work Old English Meter and Linguistic Theory—a work to which Bredehoft gives repeated and appropriate acknowledgement. The major formal improvement involves the treatment of finite non-auxiliary verbs. In Bredehoft's approach, such forms are assigned to a special, "s-foot" class. Words in this class are stressed but participate only secondarily in alliteration, if at all. This allows us to explain the syntactically parallel examples from Beowulf line 614a and 640b using an identical markup: in Bredehoft's terms, both would be analysed sx/Ssx. Building on this approach, Bredehoft goes on to use s-feet convincingly to account for problematic aspects of anacrusis.

Other chapters in the book look at hypermetric verse, the role of rhyme and alliterative effects like "crossed" or internal alliteration in classical and later Old English verse, "metrical prose," and later developments in Old English and early Middle English alliterative metre. These chapters are interspersed with chapters on "poetics" in which the use of metre for literary effect is examined in several works of each genre and period.

Bredehoft's attention to the development of Old English metre throughout the period and his insistence that late Old English metre is composed on different principles rather than through a simple debasement of the classical norm are useful correctives to approaches that tend to consider late poetry as being simply bad. His book is too short for these sections to offer more than a sketch of the principles used in this late metre and metrical prose but nevertheless represents a useful companion for students of non-classical verse—to be placed on the same shelf as Momma's Composition of Old English Poetry.

The chapters on "poetics" are in my view less successful—both because of the intrinsic difficulty of relating metrical form to the production of specific literary effects and because, in the case of the classical poetry in particular, the case argued rests on relatively unusual passages from a limited and unusual choice of texts (Judith and the Ruin). While the points raised are intriguing, this chapter in particular would be far more convincing were it either tied to larger critical treatments of the poems in question, or drawn from a larger corpus of samples. As they stand, the chapters are valuable and suggestive companions to the specific works in question rather than as a theoretical primer introducing readers to a useful approach to formal criticism.

Early English Metre is an attractive, well-written, thought-provoking, and, in the case of the "s-feet" in particular, quite convincing book. It provides a useful improvement to word-foot approaches to Old English poetry such as that proposed by Russom. And it is a useful correction to those—such as, I must confess, myself—who have ignored late poetry in general studies of the corpus because of its "poor metrical quality" (O'Donnell 1996). Readers might want to take some of the claims made in the introduction with a grain of salt—the approach is neither as fundamentally unique nor as simple and intuitive as the introduction suggests. But the book as a whole has clearly been planted in good soil.

Works cited

Bliss, A. J., Jr. 1958. The Metre of Beowulf. Oxford: Blackwell.

Cable, T. 1974. The Metre and Melody of Beowulf. Urbana: U of Illinois P.

Momma, H. 1997. The Composition of Old English Poetry. Cambridge Studies in Anglo-Saxon England. 20. Cambridge: Cambridge UP.

O'Donnell, D. P. 1996. "Manuscript Variation in Multiple Recension Old English Poetic Texts: The Technical Problem and Poetical Art". Unpublished PhD dissertation. Yale U. Also available on-line:

Pope, J. C. 1942/1966. The Rhythm of Beowulf. New Haven: Yale UP.

Russom, G. 1987. Old English Metre and Linguistic Theory. Cambridge: Cambridge UP.

Daniel O'Donnell
University of Lethbridge

Fjalldal, Magnús. 2005. Anglo-Saxon England in Icelandic Medieval Texts. Toronto: University of Toronto Press. 200 pages. 0802038379.

Magnús Fjalldal surveys the depiction of Anglo-Saxon England (c. 500-1066 AD) in the literature of medieval Iceland written during the twelfth through fourteenth centuries. His objective is to ascertain two things: (1) the historical accuracy of information about Old English language and culture, as well as about certain Anglo-Saxon leaders and events, supplied in Icelandic sources composed a century or more after the Norman Conquest, and (2) the purpose and character of the image of Anglo-Saxon England offered in these various Old Norse texts.

In the first two chapters, Fjalldal considers the mutual intelligibility of Old English and Old Norse as represented in Gunnlaugs saga Ormstungu 'The Saga of Gunnlaugr Serpent-tongue,' Fyrsta málfrœfiritgerfin 'The First Grammatical Treatise,' Hauksbók (an early fourteenth-century anthology), and other texts. He finds that the evidence of these sources is ambiguous, even rather contradictory. Sometimes they depict the unlikely comprehension of very difficult, often opaque Norse skaldic poetry by Anglo-Saxon monarchs; at others they build their story around the inability of Norse- and English-speakers to communicate. Fjalldal cites the legal text Grágás of the twelfth century, which treats very early Middle English as a foreign language, to suggest that medieval Icelanders were simply not in a position to know much about the subject one way or the other. He concludes that when Icelandic authors speculate on linguistic relations, they are driven by their own narrative needs, most often by the desire to depict a particular native hero in warm and friendly communication with a distinguished English ruler.

In chapter three, Fjalldal searches for whatever "General Knowledge and Attitudes about Anglo-Saxon England and Its Customs" he can discover in medieval Icelandic literature, noting in his Index the surprisingly short list of "English books postulated to have existed in Iceland"—only about ten or so (160). He devotes four full chapters to the reliability of information about Norse relations with the Anglo-Saxon world in Snorri's Heimskringla 'The Orb of the World,' Ágrip af Nóregskonunga sögum 'A Summary of the Histories of the Kings of Norway,' Fagrskinna 'Pretty Vellum,' Kn´ytlinga saga 'The Saga of the Sons of Knútr,' Morkinskinna 'Rotten Vellum,' Egil's saga Skallagrímssonar 'The Saga of Egill Skallagrímsson,' Breta sögur 'The Sagas of the Britons' (a retelling of Geoffrey of Monmouth's Historia Regum Britanniae), Saga Ósvalds kónungs hins helga 'The Saga of St. Oswald the King,' Dunstanus saga 'The Saga of St. Dunstan,' and Jatvarfar saga 'The Saga of Edward the Confessor.' As with the question of linguistic comprehension, Fjalldal finds that "there is little reason to believe that Icelanders ever had much, if any, first-hand historical knowledge of events in Anglo-Saxon England" and that "[m]ost if not all the adventures of the fornaldnarsögur ['legendary sagas'] and the ~ættir ['short prose narratives'] that are supposed to take place on English soil appear to be pure fantasy" (123).

Nevertheless, thirteenth-century Icelanders did know one important thing about England, that it was comparatively "wealthy and populous, and they also knew it was a great place for buying all kinds of quality merchandise not readily available in Scandinavia" (124). Since English merchants did not trade with Iceland directly, this particular kind of knowledge "probably came from Icelanders who had stayed in Norway, which had flourishing trade relations with England" (124). Icelanders thus projected the comparative prosperity of the two countries back upon an earlier period when a similar differential in wealth and population also happened to exist.

To help explain the interest of medieval Icelanders in Anglo-Saxon England, Fjalldal sees Egil's Saga as providing a strong clue in its alternative model of monarchy against which the Norwegian crown is implicitly and invidiously compared. "There was much about Norway that thirteenth-century Icelanders did not like, such as the tyranny of its rulers, their meddling in Icelandic affairs, and the lack of respect that was often felt by Icelanders" (122). These attitudes are especially clear in the characterization of King Haraldr Fair-hair and his sons. Fjalldal thus very plausibly reads the Vínheifr episode of Egils saga, which describes the battle known in Old English as Brunanburh, as a "thinly veiled" critique of the royal family of Norway whose treatment of that hero and his kinsmen is sharply contrasted with the English King Athelstan's appreciation for one of the most distinguished figures in Icelandic national history. "The Vínheifr episode is the climax of Egill's life," writes Fjalldal, without which his "achievement would not amount to much more than having opposed a Norwegian king on several occasions and lived to tell the tale. The honours Egill receives in King Athelstan's court also may well be intended to [offer] an imaginary vision of an English court where justice and generosity prevails" (81-82), and Icelanders in particular are properly valued for their intelligence, courage, and inherent worth.

Geraldine Barnes argued in 1992 that earlier works of history in Icelandic presented better-informed descriptions of Anglo-Saxon England than later ones, especially the legendary or "lying" sagas of the fourteenth century. Fjalldal finds no evidence for such an evolution in the medieval Icelandic literary tradition, regardless of period or genre. "There is no noticeable difference between the way England is represented in the histories of the Norwegian and Danish kings, the family sagas, the ~ættir, and the fornaldnarsögur," he concludes (124). In all of these texts, Anglo-Saxon England is an imaginary realm where Icelandic heroes can achieve the fame, fortune, and recognition they have been denied elsewhere. On this simple point, Fjalldal brings to a close his succinct but thorough review of sources, quelling any lingering hope that sagas like Egil's may have retained some semblance of historical memory, if not in their precise narrative details, at least in their general depiction of life in Anglo-Saxon England during the Viking age. After Fjalldal's study, the strength of this memory in Iceland appears to have been very tenuous indeed. Rather than provoking disappointment, however, Fjalldal's findings have impressed this reader with an even higher regard for the very effective historical verisimilitude that many writers of Icelandic saga prose managed to achieve.

Craig R. Davis
Smith College, Northampton, MA

Larrington, Carolyne. 1995. Women and Writing in Medieval Europe: A Sourcebook. London and New York: Routledge. 296 pages. 0415106850.

Although its title would suggest a focus on medieval women writers, in fact this book is a compendium of writings by and about women arranged by themes central to medieval women's lives. Furthermore, although subtitled a sourcebook, it is equally a survey of women's status in the Middle Ages as reflected in their most important roles and activities. In short, Larrington's book offers an abundance of useful information for teaching the Middle Ages through the lives of women.

The scope of the book is broad, covering the period 500-1500 from Wales to Byzantium, and from Iceland to Italy, as Larrington puts it. She assembles evidence from a variety of literary and historical materials to indicate the range of customs, beliefs, and practices regarding women in medieval Europe. The organizing themes, each comprising a chapter, are marriage; love, sex and friendship; motherhood and work; women and Christianity; women and power; education and knowledge; and women and the arts.

Perhaps the topics that interest American undergraduates the most in courses on the Middle Ages are marriage and love. These work particularly well in Larrington's format. The introduction to the readings on marriage covers such issues as the institution of marriage, marital property, widows and second marriages, and the role of wife. The accompanying selections from primary sources take up these issues in turn: the Old English maxims, Marie de France's Laüstic, the "Erec and Enide" of Chrétien de Troyes, Njal's saga, Boccaccio's The Decameron, Christine de Pizan's The Treasure of the City of Ladies, The Book of Margery Kempe, and so forth. All of the excerpts, including those composed in Middle English, are translated. Additionally, the treatment of courtly love from the perspectives of men, women, the "authorities," and the poets in the chapter on love, sex and friendship offers a concise background useful, in particular, for the study of romance. Here selections include "Wulf and Eadwacer," a letter from Heloise to Abelard, various love lyrics, an excerpt from Andreas Capellanus, several erotic items but curiously, no excerpts from romance.

As a teacher of a course on women writers in the Middle Ages, I found the chapters on education and knowledge, and on women and the arts, very helpful. Although these topics were addressed to a degree by the women themselves as well as by their mentors, to have so many sources brought together for comparison is a boon, especially as Larrington offers some surprises. Who knew that a play drawing upon the Faust legend with a female protagonist was written by a woman in the Netherlands in the fourteenth or fifteenth century?

In her introductions to the various topics, as well as the headnotes to the excerpts, Larrington draws widely upon primary and secondary sources, including debates on critical issues. In so doing she is able to offer the ideal combination of history and literature which beginning students need in order to approach the period. This book provides an interesting but sound perspective on the Middle Ages, and, when supplemented with the full texts of selected works such as the Middle English "Pearl," could anchor a course on medieval literature and culture.

Mary P. Richards
University of Delaware

Pryor, Francis. 2004. Britain AD: A Quest for Arthur, England and the Anglo-Saxons. Great Britain: Harper Collins. 268 pages. 0007181868.

This book was a great disappointment. Pryor is a part-time archaeologist who has a bee in his bonnet about the way archaeology has, in the past, played second fiddle to the study of written sources for post-Roman Britain. In the two centuries after the end of Roman rule in 410 the written record in Britain almost dries up, so archaeology is of course immensely valuable. This was the time that an historical Arthur may have lived, and the time during which the Anglo-Saxons came to dominate most of England, hence Pryor's title.

Drawing on his experience of archaeological studies of prehistoric cultural change, Pryor advocates applying the same methodology to this period. That is, he ignores the written records and tries to reconstruct post-Roman history from the archaeological record alone. The results are ludicrous. Based largely on evidence for continuity in farm use, Pryor concludes that there was never an Anglo-Saxon invasion of (or even peaceful migration to) Britain. He explains the change, from a literate, Christian, Brittonic- and Latin-speaking society with coinage, masonry, and manufactured goods, to an illiterate, pagan, Old-English speaking society with none of these, as a deliberate cultural choice. He says in all seriousness that the people in eastern Britain changed "for their own reasons, which we must allow them to have ... presumably they admired the way things were done across the North Sea" (214).

Pryor says that he has "chosen to set the literary accounts aside ... to redress a historical imbalance" (215). One would think from this that Pryor has read the literary accounts before setting them aside, but this is evidently not the case. First, he seems to think that Gildas' On the Ruin of Britain (ca. 540) is the only contemporary source to mention fighting against Saxon invaders of Britain. Thus he tries to discredit Gildas, but fails miserably. Just to take one example of many, Pryor says that Gildas "greatly exaggerated the severity of Anglo-Saxon piracy" (143) when in fact Gildas never mentions Saxon pirates or sea-raiders at all. It is also apparent that Pryor has not read Bede, despite referring to him several times. For example, Pryor assumes that the British clergy who met St. Augustine in 603 were from the south-east, when obviously they were from the west because Augustine travelled to the Severn to meet them (174).

The other contemporary sources for Saxon invaders in Britain, which Pryor completely ignores, are continental: the Gallic chronicles and the Life of St. Germanus. But this ignorance pales in comparison to Pryor's disregard for a huge body of Roman history when he states that "the very idea of a 'Migration Period' is absurd. Why should people suddenly decide to move around in this peculiar and hyperactive fashion?" (176). It seems Pryor cannot believe that riches, power, and a better climate were incentives enough for Germanic armies and their hangers-on to invade the Roman Empire. Presumably he believes that all over the Empire citizens suddenly decided they liked "the way things were done" on the other side of the Rhine or Danube, became Germans, and started pillaging their own cities (while never moving home of course). Laughably, he asks "why should the social disruption brought about by the end of the Western Roman Empire cause people to wander aimlessly about?" (148), when it was the invasions (which were far from aimless) that were the immediate cause of the end of the Empire.

Returning to Britain, Pryor contrasts his views with the long-discredited idea that the Anglo-Saxon invaders drove out all the Britons from eastern Britain, or swamped them by their numbers. It is now generally thought that the Anglo-Saxon invaders numbered in the tens of thousands at most, and formed a military elite. In this scenario, it is not surprising that farms continued to be used as before. It can even be imagined that some of the cultural change in post-Roman Britain was a reversion to pre-Roman practices, as Pryor argues. Most of the peasants would have stayed put, and indeed would have chosen (over the course of some generations) to identify themselves as Anglo-Saxons in order to improve their social prospects. But even in this scenario, they changed because there was an Anglo-Saxon invasion. Pryor in fact shoots himself in the foot by explaining the spread of Anglo-Saxon culture in Britain by analogy with the spread of the Spanish language and religion in Peru, where the native population always far outnumbered the Spanish. But to imply that the process whereby the conquistadors destroyed the Inca Empire and massacred tens of thousands of its citizens is a model for "acculturation" (240) is simply insulting.

Britain AD is published with a "Post-Script" including an interview with the author. In it, he is asked how he starts to writes a book. He replies: "With an outline which I agree [sic.] with my agent and publisher at great length; I then abandon it when I start to write." Unfortunately Britain AD reads exactly as if had been written in this way. It comes across as a collection of arguments the author wants to get off his chest, interspersed with anecdotes about the profession of archaeology and rural life in the Fens. The material in each chapter flows to some extent, but does not seem to be focussed on anything in particular. For example, the chapter "Arthurian Britain" begins with a discussion of Roman towns in general, concentrates for some time on Wroxeter, then moves to towns in the Anglo-Saxon east, to Christianity and continental trade there, to continental trade in the south-west and finally, to inscribed stones and the Age of Saints. Let me finish, however, on a positive note by saying that it is plentifully illustrated with colour plates, and the maps and diagrams are first-rate.

Howard M. Wiseman
School of Science, Griffith University, Australia