Vernacular Verse Histories in Early Medieval England and Francia: The Bard and the Rag-Picker 9780367459710, 9780367711085, 9781003149309

In a provocative take on Germanic heroic poetry, Taranu reads texts like Beowulf, Maldon, and the Waltharius as particip

398 106 4MB

English Pages [256] Year 2021

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Vernacular Verse Histories in Early Medieval England and Francia: The Bard and the Rag-Picker
 9780367459710, 9780367711085, 9781003149309

Table of contents :
Cover
Endorsement Page
Half Title
Series Page
Title Page
Copyright Page
Table of Contents
Permissions
Acknowledgments
Abbreviations
Introduction
Chapter 1: Beyond Germanic Heroic Poetry: Poets, Historians, and the Gods of Our Fathers
‘And the Words That Are Used/ for to Get the Ship Confused’: Academic Terminology and Underlying Assumptions
Past Approaches and the Division of Academic Labour
Bridging the Sundering Seas: Carolingian and Early English Vernacular Verse
The ‘Germanic Heroic’ and ‘Anglo-Saxon’ Paradigms: A Critique
What Is ‘History’, Anyway?
Alternative Modes of History and Vernacular Theories of Narrative Representation
The Bard, the Rag-Picker, and the Nostalgic Scribe
The Bard and the Rag-Picker at Work
Notes
Bibliography
Primary Sources
Secondary Sources
Chapter 2: What We Talk About When We Talk About History: The Old English Vocabulary of Narrative and Historical Representation
Notes
Bibliography
Primary Sources
Secondary Sources
Chapter 3: ‘Truth Is the Trickiest’: Vernacular Theories of Truth in Early Medieval Culture
Notes
Bibliography
Primary Sources
Secondary Sources
Chapter 4: Why Is There Something Rather Than Nothing?: The Social Logic of Frankish Verse Histories
Introduction
Carolingian Germanic Poetry – A Marginal Constellation
‘The Savage Words of an Uncultivated Language’: Linguistic Community and Alienation
Residual Orality and the Politics of Poetic Form
Carolingian Ethnicities (I): Teutoni, Theodisci, or Germani?
Carolingian Ethnicities (II): ‘True Franks’ in the East and West
The Interplay of Ethnicities in the Waltharius
How to Be a Man, a Noble, and a Son in Carolingian Society
Tying the Threads Together: The Background to the Figure in the Carpet
Notes
Bibliography
Primary Sources
Secondary Sources
Chapter 5: Beowulf in Times of Anxiety: The Archaeology of Emotions in Old English Verse History
Angling for Anxieties: A Method
Nervous Laughter: Wrestling with Monsters
Beowulf the Wrecca : The Monster Within
Ellorgæstas : Guests in Their Homeland
Crushing the Laðan Cynnes : On ‘Anglo-Saxon’ Race, Again
Giant Women and Haunting Danes: Race and Gender around Beowulf
Notes
Bibliography
Primary Sources
Secondary Sources
Index

Citation preview

The de facto pigeon-holing of Old English poetry as ‘literature’ rather than ‘history’ has impeded our understanding of early medieval north-west European culture for a century, and although Old English poetry has been studied comparatively with Old Norse and Latin literature to great effect, it is far too seldom read in relation to the rest of the early medieval West Germanic poetic corpus. This study boldly tackles both these problems, producing an array of fascinating insights on points of detail alongside new thinking on how history works. —Alaric Hall, University of Leeds This study offers a completely novel way of approaching the function of heroic poetry within Germanic societies, and enabled me to read Beowulf with fresh eyes. Taranu’s ground-breaking and innovative work is likely to remain essential reading for many years to come. —Ralph O’Connor, University of Aberdeen This book is a worthy contribution to a vital and ongoing conversation about the socio-cultural significance of Old English vernacular poetics and a welcome intervention into very topical debates surrounding nation, masculinity, and race. —Manish Sharma, Concordia University

Vernacular Verse Histories in Early Medieval England and Francia

In a provocative take on Germanic heroic poetry, Taranu reads texts like Beowulf, Maldon, and the Waltharius as participating in alternative modes of history-writing that functioned in a larger ecology of narrative forms, including Latinate Christian history and the biblical epic. These modes employed the conceit of their participating in a tradition of oral verse for a variety of purposes: from political propaganda to constructing origin myths for early medieval nationhood or heroic masculinity and sometimes for challenging these paradigms. The more complex of these historical visions actively meditated on their own relationship to truthfulness and fictionality while also performing sophisticated (and often subversive) cultural and socio-emotional work for its audiences. By rethinking canonical categories of historiographical discourse from within medieval textual productions, Vernacular Verse Histories in Early Medieval England and Francia: The Bard and the Rag-Picker aims to recover a part of the wide array of narrative poetic forms through which medieval communities made sense of their past and structured their socio-emotional experience. Catalin Taranu is a literary-historical scholar working on the vernacular poems and cultures of early medieval England and Francia. He has taught medieval literature and Old English and has shared his research on Beowulf, medieval rhizomes of narratives, and vernacular theories of truth and history in talks and publications. Catalin is currently a postdoctoral researcher at New Europe College, Bucharest, where he studies the socio-emotional economy of shame and honour in medieval heroic poetry. An essay collection titled Vera Lex Historiae?: Constructions of Truth in Medieval Historical Narrative (co-edited with Michael Kelly and forthcoming from Punctum’s Gracchi Books) explores strategies of constructing, authorizing, and assessing truth in medieval historical narrative.

Routledge Studies in Medieval Literature and Culture

From Medievalism to Early-Modernism Adapting the English Past Edited by Marina Gerzic and Aidan Norrie Avid Ears Medieval Gossips and the Art of Listening Christine M. Neufeld Zöopedagogies Creatures as Teachers in Middle English Romance Bonnie J. Erwin Before Emotion: The Language of Feeling, 400–1800 Juanita Feros Ruys Forging Boethius in Medieval Intellectual Fantasies Brooke Hunter Sanctity and Female Authorship Birgitta of Sweden & Catherine of Siena Edited by Unn Falkeid and Maria H. Oen Tracing the Trails in the Medieval World Epistemological Explorations, Orientation, and Mapping in Medieval Literature Albrecht Classen Vernacular Verse Histories in Early Medieval England and Francia The Bard and the Rag-picker Catalin Taranu For more information about this series, please visit: h t t p s : / / w w w. r o u t l e d g e . c o m / R o u t l e d g e - S t u d i e s - i n - M e d i e v a l Literature-and-Culture/book-series/RSML

Vernacular Verse Histories in Early Medieval England and Francia The Bard and the Rag-Picker

Catalin Taranu

First published 2021 by Routledge 52 Vanderbilt Avenue, New York, NY 10017 and by Routledge 2 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN Routledge is an imprint of the Taylor & Francis Group, an informa business © 2021 Taylor & Francis The right of Catalin Taranu to be identified as author of this work has been asserted by him in accordance with sections 77 and 78 of the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging-in-Publication Data A catalog record for this book has been requested ISBN: 978-0-367-45971-0 (hbk) ISBN: 978-0-367-71108-5 (pbk) ISBN: 978-1-003-14930-9 (ebk) Typeset in Sabon by SPi Global, India

Contents

Permissions viii Acknowledgments ix Abbreviations xi Introduction xii 1 Beyond Germanic Heroic Poetry: Poets, Historians, and the Gods of Our Fathers

1

2 What We Talk About When We Talk About History: The Old English Vocabulary of Narrative and Historical Representation

64

3 ‘Truth Is the Trickiest’: Vernacular Theories of Truth in Early Medieval Culture

90

4 Why Is There Something Rather Than Nothing?: The Social Logic of Frankish Verse Histories

127

5 Beowulf in Times of Anxiety: The Archaeology of Emotions in Old English Verse History

198

Index 235

Permissions

Earlier versions of three of the chapters included here have appeared (or are scheduled to appear) elsewhere. Part of Chapter 2 appeared in an early form as an article: ‘Senses of the Past: The Old English Vocabulary of History’, Florilegium, 29 (2012 [2014]), 65–88. This material is republished by permission of Florilegium and the Canadian Society of Medievalists / Société canadienne des médiévistes, for which I am grateful. Chapter 3 will appear as a chapter in a collective volume co-edited by me, ‘“Truth Is the Trickiest”: Vernacular Theories of Truth and Strategies of Truth-making in Old English Verse’, in Vera Lex Historiae?: Constructions of Truth in Medieval Historical Narrative, ed. by Catalin Taranu and Michael Kelly (New York: Punctum/Gracchi Books, forthcoming). Chapter 5 appeared as ‘Men Into Monsters: Troubling Race and Masculinity in Beowulf’ in Dating Beowulf: Studies in Intimacy, ed. by Daniel C. Remein and Erica Weaver (Manchester: Manchester University Press, 2019). The last two chapters have been published under the Creative Commons license, which allows me to retain the copyright and freely reprint material from my own previously published work.

Acknowledgments

A postdoctoral appointment at the University of Bucharest’s Institute for Research in the Humanities (IRH-ICUB) in the year 2017–2018 enabled me to develop new work for this volume. I have become increasingly aware that my own contribution to the work I am doing is very small. Whatever I may have accomplished is the result of many encounters with people and ideas. It is with great joy that I now express my gratitude to these people (although many will remain unnamed and unknown even to myself). The idea behind this book and much of the work going into it started at the University of Leeds as a doctoral dissertation. I would like to thank my supervisors, Alaric Hall and Catherine Karkov, for their years of guidance, constant support and encouragement, and unwavering kindness while at Leeds and beyond, during my wandering scholar years. I am also grateful to Ian Wood for his scholarly and personal generosity – his weekly doctoral meetings have been a congenial space for open discussion about all things early medieval and for starting many of my Leeds friendships listed next. My thanks also to Clare Lees for providing muchneeded advice and support during my defence and following years. I cannot leave Leeds without thanking the staff and colleagues at the Institute for Medieval Studies and the Schools of History, English, and Fine Art, especially Emilia Jamroziak and William Flynn for their unflagging support and ever-useful suggestions. The writing of the thesis would have been a much less cheerful and inspiring experience without the friendship and consummate scholarship of Michael Kelly, Hervin Fernández-Aceves, Mike Burrows, Otávio Luiz Vieira Pinto, Sami Kalliosaari, Clarck Drieshen, Ricky Broome, Mark Tizzoni, Sarah Lynch, Hope Williard, Paul Gorton, John Latham, James Harland, and Pierre Gaite. May the hairs on your toes never fall out! Bucharest’s IRH-ICUB provided a vibrant intellectual environment for doing much subsequent work on this book. I am grateful to the Institute’s Iulia Nițescu and Mihnea Dobre, as well as to the friends I found during that time: Viktor Ilievski, Marian Coman, Ionuț Epurescu-Pascovici,

x Acknowledgments Matthew Dentith – in particular to Oana Cojocaru, who has been a light in times of darkness. During the elaboration of this book, I have benefited from the sage advice and kind encouragement of a number of scholars who have generously shared their work and expertise with me. My thanks to Ralph O’Connor, Eileen Joy, Irina Dumitrescu, Manish Sharma, Stephen Harris, Erica Weaver, and Dan Remein. I am also grateful to the many colleagues who have provided helpful anonymous peer-review on submitted pieces of writing and who have made comments at the conferences and workshops where I presented my research – also to my students at Leeds and Iași who also taught me so much. Going back in time, I would like to thank the teachers who have guided and inspired me: Gabriela Lopez (who taught me English and helped me find my voice), Maud Pérez-Simon, Veronica Popescu, Adrian Poruciuc, and the late Dumitru Dorobăț. I have also been privileged to call Adrian Haret, Ioana Baciu, Lavinia Ludovic, Andrei Alecsa, and Roxana and Dumitrel Ghiba friends for a long time. I am also indebted to Mitchell Manners and Jennifer Abbott for their patience and help with the editorial process. I dedicate this book to my parents, Gabriela and Octavian, without whom none of this would have been possible: I thank them for their love and support, for fretting about my setbacks and rejoicing in my successes more than I ever have, and for being the home I always go back to. Lastly and firstly, nu sculon herigean heofonrices weard | meotodes meahte and his modgeþanc.

Abbreviations

ASC  EETS   GHP  HE  JEGP  MGH  OE  OHG  ON  OS  PBA  PL 

Anglo-Saxon Chronicle Early English Text Society Germanic Heroic Poetry Historia Ecclesiastica Gentis Anglorum Journal of English and Germanic Philology Monumenta Germaniae Historica Old English Old High German Old Norse Old Saxon Proceedings of the British Academy Patrologia Latina

Introduction

This book approaches Germanic heroic poetry (GHP) as a form of history-writing. It may seem perverse to read Beowulf – a mainstay in the canon of English literature, a text that features a fire-breathing dragon and a hero fighting naked with a humanoid monster – as history. But this is because we think we know that texts like Beowulf or the Waltharius are supposed to be ‘myths’, ‘legends’, ‘epics’, or ‘GHP’ and thus incompatible with ‘history’ – all of them scholarly categories that would have been alien to their early medieval creators and audiences. This book goes against the grain of the received wisdom embodied by these labels and argues that the peculiar ways in which such sources function as texts and as cultural products in their societies can be explained by the fact that they work as several distinct modes in a larger ecology of genres of historical representation which include the biblical epic (the Heliand or Judith) and more canonical historiography (Bede’s Ecclesiastical History of the English People or the Anglo-Saxon Chronicle [ASC]). Another scholarly temptation rarely resisted has been to take the formal conservatism and relative metrical homogeneity of the corpus as a reason to treat texts as different as the Hildebrandslied, The Finnsburg Fragment, or Widsith as quasi-interchangeable manifestations of some ‘Germanic heroic’ essence they all presumably share. Yet this paradigm obscures the variety of modalities in which different texts narrativize different pasts and perform different types of cultural and emotional work. The triumphant propaganda of The Battle of Brunanburh, the mournful tragedy of Maldon, the fractal workings of inter- and intra-tribal conflict in Beowulf’s brooding poetic voice, and Widsith’s dizzying panorama of long-gone tribes and heroes all share, besides obvious formal qualities, a deep engagement with the past, from the very distant to the very recent. Yet the ways in which they turn these pasts into historical narrative are very much different, as are the ways in which they functioned for different audiences throughout time. Accordingly, this book proposes a more fine-grained model for understanding ‘GHP’ based on providing a conceptual road map to rethinking

Introduction  xiii traditional categories of modern historiographic discourse from within medieval textual productions by tracing the vernacular theories of narrative authority, truthfulness, and historical representation lying implicit in these texts (in the first three chapters) and revealing the socio-cultural and emotional roles these texts fulfilled for different communities in early medieval England and Francia (in the last two chapters). In doing so, the present study is invested in wresting ‘GHP’ from two common paradigms that obscure its socio-cultural, emotional, ideological, and narrative meaning: one that refuses it value as history-writing since it looks nothing like what is usually considered history-writing in the Middle Ages and the other one, which flattens its versatility, occasional experimental spirit, and diversity of voice and meaning by presuming it to be an essentially homogeneous corpus in its conservative ideological thrust, looking back with longing to the image of a ‘heroic age’ preserved in a pre-literate ‘Germanic tradition’ presumably shared by all Germanic-speaking peoples of North-Western Europe. Now, a road map for this book. This brief introduction shares some of the functions of a classical introduction together with Chapter 1, which foregrounds all the red threads going through the book while also making detailed arguments about all of them (historicity, the ‘Germanic heroic’, and ‘Anglo-Saxon’ paradigms, presenting my bard-rag-picker model, showing how it works on actual texts). The first chapter, then, traces previous approaches to the historicity of GHP, argues that this corpus is fundamentally historiographical in its production and reception, and proposes a paradigm of three modes of historical representation in which individual texts from this corpus participate in different ways: (1) a ‘bardic’ mode (e.g., The Finnsburg Fragment, the Hildebrandslied: self-contained tragic episodes from the distant ‘Germanic heroic’ past, narrated linearly with no discernible propagandistic purposes), (2) a ‘neo-bardic’ mode (e.g., The Battle of Brunanburh and other ASC poems, the Ludwigslied: triumphalist narrativizations of recent events for clear propagandistic purposes), and (3) a rag-picker’ mode (e.g., Beowulf, Widsith, Deor, the Waltharius – often longer, complex, polyvocal, self-referential texts, with often subversive or at least ambivalent ideological thrust). The second chapter provides a wide-ranging, lexical-semantic study of the Old English vocabulary of history and narrative in their various contexts in glosses, translations, and vernacular prose and verse. By using approaches drawing on cognitive linguistics, it traces the implicit vernacular theories or cultural schemas underlying the way these words and phrases are used, thereby offering a redefinition of what counts as history from within an alternative cultural horizon distinct from that in which classical categories of narrative and history drawing on Isidore and Cicero made sense.

xiv Introduction The third chapter takes a similar approach to the issues of truthfulness, veracity, and realism. It starts from the oft-remarked upon presence of apparently incongruously fabulous elements in otherwise ‘realistic’ medieval narratives, which it explains by proposing the existence of a variety of underlying theories of truth and narrative representation governed by different expectations, patterns of eventuality, and social functions that are at work in different types of discourse and modes of historical production. Focusing on a variety of Old English sources, ranging from heroic verse to elegies and maxims, the chapter traces three main truth procedures throughout this corpus: auctoritas (the dominant mode in theological and later medieval narratives, whereby truth is based on a written record and on the religious authority and/or narrative prestige of a named author), traditio (truth is procedural, guaranteed by following a certain traditional process, such as composing oral verse in the right way or fulfilling the formalities of the judicial process), and collaboratio (truth is in abeyance, waiting to be assessed and pieced together by a knowing audience, thus establishing it as a collective enterprise which often has an agonistic, polemical dimension). The fourth chapter explores vernacular verse histories, such as the Waltharius, the Hildebrandslied, and the Heliand, as cultural products for elite communities in Carolingian society in a context of shifting ethnic, gender, and social identities. Much more than propaganda or entertainment, these texts provided a space where the tensions and anxieties of the Carolingian elites could be called into question, toyed with, and taken to the breaking point. These tensions were the results of major redefinitions of identities underway in the crisis of the second-third of the ninth century in Carolingian Francia. By approaching these poems as cultural products responding to these stringent social needs, this chapter aims to move the extant scholarly conversation beyond the unhelpful disputes focusing on whether individual texts are more Germanic or more Latinate, more heroic or more pious, more critical or more laudatory. A more fruitful pathway is here considered to be the exploration of the social, cultural, and political forces to which these poems both responded and gave voice. The fifth and last chapter is a socio-emotionally inflected exploration of how rag-picker verse histories, such as Beowulf, performed important work as dream-screens on which different audiences could project their idealization of ‘heroic’ honour-based masculinity, anxieties about the destructive potential of this gender construct and about racial constructions of an Angelcynn in relation to the native Britons and, later, the invading Danes, and even perform collective mourning in times of turmoil. More specifically, it looks at how different textual communities in early medieval England might have related socio-emotionally to Beowulf: first by reading it in relation to Vita Guthlaci and other sources

Introduction  xv pertaining to the relationships between Anglo-Saxons and native Britons and then in the larger picture of its containing manuscript (particularly in relation to the killing of the giant women by Alexander the Great in Wonders of the East) and how contemporary fears about the Danish raids and guilt for St. Brice’s Day massacre might have played into it.

r.?\ Taylor & Francis � Taylor & Francis Group http://taylora ndfra ncis.com

1 Beyond Germanic Heroic Poetry Poets, Historians, and the Gods of Our Fathers

In his Historiae adversus paganos, Orosius cites the historian Pompeius Trogus and his epitomizer Justin as his sources for Joseph’s prediction of the famine in Egypt. Justin and Pompeius were certainly considered to be historians by the standards of classical antiquity and thus also of the early medieval English elite for whom a translation was produced around 900 as part of King Alfred’s programme of vernacular learning.1 In the Old English (OE) translation, however, this standard establishment of textual authority is rendered rather unexpectedly as from ðæm Iosepe, Sompeius (sic!) se hæþena scop ond his cniht Iustinus wæran ðus singende: ‘the pagan poet [historicus in the original Latin text] Pompeius and his apprentice [breviator in the Latin] Justin were thus singing about Joseph’.2 In the same translation, Homer (poeta in the original Latin) is also called a scop, which coincides with the usual meaning of the word – ‘poet’.3 This detail has attracted little comment from scholars writing on the naturalization of this text for the early medieval English cultural landscape.4 But the greater import of this detail should not escape us: an elite, literate, early medieval English individual who was familiar enough with classical learning and the Latinate tradition of historiography to be able to translate Orosius did not think it improper to turn two sober historians from the classical past into a bard and his apprentice singing rather than writing their histories, just like the poet Homer presumably sang his epics. Are we simply seeing the translator nodding – is it an unthinking anachronism or perhaps an inside joke? Or is it rather an intentional attempt at naturalizing the persona of the historian from classical and classicizing history-writing by turning it into a vernacular figure with a specific socio-cultural function and narrative authority that pertained to the domains of both poetry and history-writing in the cultural horizon of early medieval England? It is the latter possibility that illustrates my approach throughout this book, which consists of teasing out vernacular theories of historical representation, truthfulness, and narrative authority lying implicit

2  Beyond Germanic Heroic Poetry in Germanic heroic verse written down in early medieval England and Francia and revealing how they performed cultural and emotional work for its different audiences. For, regardless of the actual intent behind this peculiar translation in Orosius, the mental image that it must have conjured for both the translator and audiences needs to be accounted for since it is certainly not a singular occurrence. The underlying conceptualization of narrating the past conforms to neither modern nor classical conceptual schemata of historical production: a historian is a particular type of poet, and history is verse made to be performed. We may surmise its grip on the cultural imagination of early medieval England (and of at least some communities in Francia) from the prevalence of this way of thinking about history-writing in texts such as Beowulf, The Battle of Maldon in England, or the Waltharius in Francia – all of them narrative representations of past events, however different both from each other and from more canonical history-writing, such as the Anglo-Saxon Chronicle (ASC) or Bede’s HE. The Orosius translation is remarkable because it projects this conceptualization on the space of Latinate antiquity, even while the history-writing mode remains that of Orosius’s salvation-focussed classical historiography, albeit with plenty of changes meant to make the text perform ideological work for the newfound national entity of the Angelcynn in post-Alfredian England.5 For this reason, if we want to see this vernacular verse-centred mode of historical representation actually at work, we have to examine the corpus usually labelled ‘Germanic heroic poetry’ (henceforth referred to as ‘GHP’). Hence, in what follows, I argue that most GHP can be more productively situated within an ecology of cultural products that attempt to make sense of the past. This extended genre of history-writing includes, besides GHP, Latinate Christian histories (such as Bede’s or Gregory of Tours’s works), chronicles (the ASC, as well as Æthelweard’s Latinate adaptation of it), and biblical vernacular epics (such as the OE Genesis or Judith and the OS Heliand), without privileging any of them over history-writing in ‘heroic’ vernacular verse. Still, my book does not aim to provide an indepth study for all of these texts and modes of historicity but only for some of the secular vernacular verse commonly known as GHP (or, in the case of the Waltharius, Latin verse, although heavily influenced by the vernacular heroic tradition). This is a much less controversial proposition than it may appear. All GHP is obviously telling stories about the past, regardless of our belief in the accuracy or truthfulness of these stories. As such, the reasons for which we tend not to regard it as history have much more to do with scholarly habitual practices and mental shortcuts than with what the texts themselves say – especially in light of their sophisticated sense of the past, the important cultural work that they accomplished in society, and the emotional needs they fulfilled for its audiences. Revealing

Beyond Germanic Heroic Poetry  3 these aspects of GHP calls for a two-pronged strategy: closely examining the narrative and poetic workings of vernacular heroic verse sources as they participated in a form of history-writing, all the while making sense of how these texts functioned in a socio-cultural and emotional web of meaning that is distinct (although not disconnected) from that in which more canonical medieval history-writing worked. But before launching into my exploration of these aspects, I first proceed to unpack the accreted layers of interpretative practices that obscure the historical nature of GHP – the first half of this introductory chapter is dedicated to this task. After examining previous approaches to the issue of historicity in GHP, I make an argument for considering it as historical representation. Then I extricate this corpus from the habits of interpretation we bring to it, embodied by labels such as ‘poetry’, ‘legendary’, ‘heroic’, and ‘Germanic’ (mostly modern categories which would not have made much sense to early medieval people), which hinder our ability to perceive the seriousness of their historical investment and the diversity of its modes of historicity. With this preliminary work done, I then move on to explain the interpretive model I propose, encapsulated in the title’s bard and ragpicker. This model accounts for both the GHP’s ‘obsession with history’ and the variations in the historical consciousness at work in each of these texts, which I attribute to the differences between the modes of historicity of which they partake (and thus between their respective narrative strategies, genre expectations, socio-cultural functions, etc.).6

‘And the Words That Are Used/ for to Get the Ship Confused’: Academic Terminology and Underlying Assumptions GHP has been the subject of a monumental amount of scholarship over the past two centuries. Despite the variety of its approaches and the complexity of its conclusions, several parameters have remained mostly unchanged throughout this corpus of research: Germanic heroic poems are considered to be legendary/fictional/fantastic narratives whose occasional resemblance to historical narrative is a by-product of their artistry; they are ‘Germanic’ through their rootedness in an oral tradition of alliterative verse in which similar stories were preserved, retold, and shared by early medieval Germanic cultures; they are ‘heroic’ in their adherence to a heroic ethos based on extolling martial prowess, on preserving honour in the face of great adversity, and on the special relationships between the lord of the warband and his thegns. Yet texts like Beowulf and the Waltharius do not sit easily within the genre boundaries that scholars have constructed for early medieval textual production.7 Beowulf is as much a story of monster-extermination as it is a tribal history of the spiralling feuds between Scandinavian tribes and warlords. Widsith is as much a poetic tour de force as a primer of

4  Beyond Germanic Heroic Poetry heroic history and geography. Deor’s central conceit (a fallen court poet’s complaint) hinges on its framework of references to ‘that period of northern European history that was so much the creation of heroic legend that modern scholars have called it the Heroic Age’.8 The Waltharius is as much mock-heroic as historical romance. At the same time, albeit many of them refer to similar-sounding sets of characters and an interest in a set of themes and types of situations, none of these texts show any sign of awareness of their participating in a pan-Germanic culture at the time they were composed. One would not conclude from the commonalities of types situations, characters, turns of phrase, and emotional style shared by Arthurian romances everywhere that they, therefore, all shared in an ‘Arthurian culture’. Furthermore, the very heroic ethos that these texts are presumably celebrating is usually the subject for deep and intense anxiety (as in the Hildebrandslied, where it causes the tragic duel between father and son, or in Beowulf and The Finnsburg Fragment, where it dictates spiralling cycles of revenge), when they are not outright mockery (in the Waltharius). Although most of the stories and events alluded to in these sources need a considerable scholarly apparatus for modern readers to make sense of, this was probably not always the case. Even if one argues that the allusions to historical figures and events in these poems were just as mysterious for early medieval audiences as they are for us, the stories they are based on cannot have been invented wholesale out of thin air.9 At some point in their evolution, these narratives referred to past events that were perceived as real, even if the larger context may have been mostly forgotten by the time these poems were read (although in many cases, they were probably not composed) in tenth-century England. That these allusions and narratives are most often characterized as ‘legendary’ (or ‘mythical’ or ‘fantastic’) rather than ‘historical’ is not due to what early medieval English or Frankish people thought about them but rather to the systemic forces of habitude that determined the gradual separation of medievalist scholarship between historians, literary scholars, and philologists in the nineteenth century. Canonical medieval history became the province of the first camp and medieval verse that of the second and the third (although their respective focuses and methods of exploration also came to move away from each other, despite their enduring interdependence). Hence, although early medieval sources such as Beowulf or the Waltharius are effectively narratives about the past, they are never read as ‘history’ simply because they do not fit into what modern scholarship regards as history in the early Middle Ages – namely, texts written in Latin authored by (usually) named authors, such as Bede or Gregory of Tours, who write in the recognizable narrative matrix of classical history made Christian. But it is not at all clear that most people in early medieval England and Francia shared this view. The vast majority of them did not possess

Beyond Germanic Heroic Poetry  5 the full package of cultural schemas and habits of the mind, literary references, and linguistic knowledge (and the socio-political conditions making them possible) that would have given them the ability to appreciate the kind of texts written by these authors. Indeed, much of this book presents evidence to the contrary and aims to understand the types of historical representation at work in GHP on its own terms rather than according to expectations set by such canonical histories (which often do have historia in their title). We (i.e., modern scholars) focus on these texts because this is what has survived, and this is what looks like history to us, but in doing so, we risk looking for the proverbial keys under the proverbial street light rather than where we (or rather the early medieval people whose keys we are looking for) may have dropped them. To adapt Nicholas Howe’s description of the governing assumption of the mid-twentieth-century wave of patristic-theological Anglo-Saxonist criticism, such a narrowly historicist approach assumes ‘that the text must be read within the accepted or, from another perspective, the hegemonic ideas and forms of its culture’.10 In this, it ascribes to ‘a masternarrative about medieval culture that is at once modern and fabricated: the assumption that all of its texts have the same didactic intention and can be opened with the same key of…exegesis’.11 This master-narrative, for historians working on canonical history-writing, is that of Christian classicizing historiography (even when it is written in the vernacular and even when it takes the recognizable shape of chronicles, charters, hagiographies), while for philologists and oral formulaicists, it is that of an ideal Germanic oral tradition forever out of reach and thus in perpetual need of being created out of the only written sources it engendered – namely, GHP. But, as Howe explains, such historicisms distort the relations between text and context by creating a single context for all texts and thus, at least potentially, reducing them all to a single text’, and in doing so, ‘often betray…the past by ignoring its competing strands of belief, practice, expression, of all that circulates within and makes for a messy, living culture.12 The aim of the present book is precisely to reclaim this variety, at least with respect to the modalities of historical representation and to recover GHP as a space where sophisticated thinking about history, fiction, and the representation of the past took place. In this, I seek to bypass both the modernist and the medievalist biases that Leslie Lockett reveals in previous scholarship on the psychology manifested in OE texts.13 Adapting her observations to the concerns of my study, the medievalist bias is the present-day reader’s assumption that the classical conceptualizations of history – such as those espoused by Cicero, Isidore, and Bede, among others (as we shall see in the next chapter) – were ‘well-known

6  Beyond Germanic Heroic Poetry and fully assimilated’ wherever Latinate literacy took hold in early medieval England. Two basic misapprehensions underlie this bias: (1) that opinions on what exactly was considered to be history-writing among Western European Latinate intellectuals were monolithically dominated by the oft-quoted Isidorian (actually Ciceronian) tripartite model of historia (the things that happened), argumentum (the things that could happen, although did not), and fabula (the things that neither happened nor could happen) and (2) that direct and indirect knowledge of such discourses were ‘readily available in most Anglo-Saxon centres of learning’.14 As Monika Otter suggested, this classification is ‘something of a blind alley’ since ‘[n]o Roman or medieval theorist…was able to go any further with it than merely to repeat it’, nor can it provide ‘a satisfactory category for thoroughly canonical and admired but “untrue” narrative masterpieces, such as the Aeneid…let alone…Arthurian romance’.15 The modernist bias, on the other hand, consists of the present-day reader’s inclination to interpret history and literature, fiction and fact, narrative invention and report as ‘wholly distinct and even opposed to one another, without taking into account whether a given text emanates from a culture in which…dualism is as pervasive as it is in popular present-day Western thought’.16 This need not result in re-embracing the ‘old-fashioned anthropological primitivism’ underscoring the long tradition of placing ‘orally derived’ OE heroic verse next to Ancient Greek Homeric corpus or to living oral traditions, such as Eastern African or Southern Slavic ones as presumably issuing from the same universal ‘heroic’ stage of society and thus understandable resulting from similar pre-Enlightenment cultural horizons.17 What vernacular theories of historical representation and modes of reception in early medieval England and Francia have in common with similar conceptualizations encountered in non-industrial or ‘traditional’ cultures is not that they are ‘primitive’ but rather that they developed to be distinct (although not necessarily in isolation) from classical theories of history.18 Meanwhile, Stephen Harris has exposed another set of biases underlying the philological nationalism in which the ‘Germanic heroic’ paradigm is rooted (on which more below), which ‘paints a vivid and romantic picture of Anglo-Saxon poetry as the literary inheritor of a long Germanic tradition descended from the ancient scops, who stored the history of their people in metrical form in their capacious and inventive memories’.19 The main problem with this paradigm that still dominates, however subtly, the way these texts are read, interpreted, and taught, is the assumption that ‘an author is speaking widely and representatively of a culture in all his or her idiosyncrasies’, as if ‘a Geist infuses all social subjects (authors, bishops, farmhands, criminals) equally’, as if ‘culture exists beyond its textual articulation,…a Platonic form that occasionally took

Beyond Germanic Heroic Poetry  7 terrestrial shape in oral or written communication’.20 On the other hand, another (directly related) problematic assumption is that the audience automatically shares the text’s ideological, ethical, or social claims. Yet the relation between text and audience (that is, the production of textual culture) is modal or possible – not yet affirmed, but possibly so. A literary text offers an audience a possible world, ‘other possible states of affairs‘, which may or may not be acceptable to an audience.21 It is this more capacious understanding of a text and its potential audiences that guides my endeavour. The biases revealed by Howe, Harris, and Lockett lie behind the implicit assumptions of much of the scholarship referred to in the following. Here, I sought to clarify in advance what I see as the problematic implicit conceptions underlying many of them. To return to the Orosius translator, the odd rendering of both historicus and poeta as scop needs to be accounted for in terms that would do justice to the specific cultural horizon in early England in which they made sense. This horizon was distinct (although not socially isolated) from that of, say, Alcuin or Bede, who functioned within the Orthodox conception of truth championed particularly (although not exclusively) by Carolingian intellectuals, for whom Christianity was ‘a total system of thought’ that encompassed all human knowledge so that it became ‘a political theory about empire, a Ptolemaic astronomy, a Plinian natural history, an Augustinian totalizing history of the world and its varied civilizations’.22 Thus the cultural horizon of the poet-historians was likely shared by the more secularly inclined clerics whom the epistles of Alcuin and Bede deplored for listening to fabulae and stories of Ingeld and among whom Patrick Wormald saw the probable audience for Beowulf and other texts with a similar worldly penchant.23 Alcuin’s oft-quoted rebuke ‘quid Hinieldus cum Christo?’, ‘what has Ingeld to do with Christ?’ to bishop Unwona of Leicester (and not to the abbot of Lindisfarne, as previously thought) about vernacular songs (carmina gentilium, ‘the songs of pagans’) being unacceptable in monastic spaces makes it clear that this type of discourse was unsuitable for the Christians.24 Yet as such glimpses (including the ban on English priests reciting the liturgy ‘tragico sono’ in the manner of secular poets at the 747 Council of Clovesho) show, this form of discourse was, although frowned upon, still central to the cultural horizon of early medieval England and Francia.25 I come back to this simultaneously marginal and culturally powerful status of vernacular verse when I explain my model next, as well as in Chapter 4. For now, we should keep in mind, as Emily Thornbury warns, that ‘Anglo-Saxon history is haunted by the remnants

8  Beyond Germanic Heroic Poetry of [textual] communities that must once have had distinct identities’, ‘lost groups’, some ‘isolated on the borders of the Anglo-Saxon world’, whose experience we cannot recover from any records outside of the texts they produced.26 It is not that there were no words for ‘historian’ in OE. I will look at some of them (like staerwritere or wyrdwritere) more closely in the following chapter. However, as I show there, most words that translate concepts from classical cultural horizon without naturalizing them are either Latinate borrowings or OE calques on Latin words. Even if people in early medieval England certainly had the concept of a writer of history, no cultural schema may have underpinned it as emotionally compelling as that of the scop-historian. At the same time, such an abstract concept may not have even been available outside of a narrow circle of intelligentsia moulded into Christian Latin learning from their childhood, absorbing its totalizing, monolithic conception of narrative truth and the power of its literate, authorial forms of expression. Vernacular terms for the kind of discourse embodied in Beowulf or the Hildebrandslied (including for the process of production and the people producing it) existed, but they are usually translated into modern English using terms that function within the dominant critical paradigm, according to which GHP is legendary verse rather than historical production. Pressed to describe the process of composition of one of these texts, an early medieval English person not inclined to dismiss such discourse outright (as Alcuin and other like-minded ecclesiastic did) might have said that Beowulf, for instance, is a giedd (‘song’ but more accurately ‘heightened speech’ or ‘versified narrative’) or a soðspell (‘true story’), that it was wordcræftum wefen (‘woven with word-craft’, adapting Elene, l. 1237) or soðe gebunden (‘properly/truly bound together/composed’, Beowulf, l. 871) by a scop (‘poet’ but possibly also ‘verse historian’, even ‘bard’, employing the Romantic associations of the modern term to render the aura commanded by the OE term for at least some early medieval English communities), that it was gehyrde (‘heard’, usually a shared experience, even if actually ‘read’) by an audience and that, of course, it was soð (‘true’ but a special kind of ‘true’ at work within the bounds of the type of discourse and not necessarily ‘factual’).27 But these OE words do not give us access by themselves to the vernacular theories of history, truth, and narrative authority and poetics with which the people using them operated; they do not by themselves reveal the way these texts functioned in their socio-emotional economy or cultural horizon. Hence the need for a lexicological and semantic archaeology of the terms people in early medieval England and Francia would have used for these notions, understood in the constellation of discourses and, more broadly, within the ‘total cultural system’ of meaning GHP was part of – which this chapter and particularly the next one aim to do.28

Beyond Germanic Heroic Poetry  9

Past Approaches and the Division of Academic Labour Previous approaches to historicity in GHP can be roughly categorized in three chronological periods: the first, the pre-Tolkien age, dominated by the search for the seldom glitter of hard historical facts among the chaff of the ‘fictional’ material GHP was seen to be mostly made up of; second, post-Tolkien, a general rejection of any historical quality except that of artistic make-believe to Beowulf (and more broadly, all GHP); third, and finally, riding on the wave of the linguistic turn, the recent reappraisal of the historical qualities of GHP not as a source for reconstructing history, but as its own, sui generis, form of narrative about the past. Nineteenth-century scholars, such as Nikolai Grundtvig, were more interested in what Beowulf had to say about the bits of historical trivia confirmed by ‘real’ historiography (such as Hygelac in the poem being identified as a Chlochilaicus in Gregory of Tours’s Histories).29 Yet the poems themselves continued to be regarded as muddled affairs in which the rare fragment of authentic history was drowned under a torrent of fantasy containing monsters, superhuman heroes, and the occasional romance. J. R. R. Tolkien famously overturned this tide but caused subsequent scholarship to throw the baby of history out with the bathwater of our notion of history – thus Beowulf became ‘literature’. His rejection of the use of Beowulf as a mere source for the study of Germanic antiquities by emphasizing the essentially artistic (and thus presumably ahistorical) nature of the poem was a product of its time. It echoed the rise of ‘literary criticism’ in the modern sense as ‘a respectable academic activity’ and (whether by design or by osmosis) adhered to the formalist tenets of New Criticism, chief among which was ‘the rejection of historicism and other extraliterary considerations’, alongside the value placed on “universal” literary virtues like a balanced structure and organicism.30 Around the same time, in the first half of the twentieth century, two separate institutional evolutions were taking place among scholars of early medieval England and medievalists in general. For one, historical research proper had begun to branch off from the multidisciplinary ‘heroic age’ approach to GHP, which saw it as essentially equivalent to the Homeric corpus or Serbian oral poetry – all of them the inevitable fruits organically growing out of the ‘heroic age’ of peoples across the world and throughout history. Studies of history and literature were becoming specialized professional fields ‘dividing fact and fiction, Latin and Old English between them’, while at the same time, literary studies themselves started ‘rejecting philology and a preoccupation with the ancient Germanic past’, instead focussing on the Christian background of medieval literature and claimed patristic literature as the key to understanding OE texts.31 Between the ahistorical study of GHP as the province of philology and the narrow historicism of patristic literary study,

10  Beyond Germanic Heroic Poetry there was little room for recognizing the historiographical qualities of the intentions of these sources. There was even less room for this in the ‘proper’ historical study of the early Middle Ages, which focussed on either Latinate works or on OE sources whose functions were more easily squared with the disciplinary bounds of interests of modern historical study – charters, chronicles, genealogies. Yes, as Nicholas Howe argues, Tolkien’s approach to the historicity of GHP was in fact more nuanced than it is often given credit for and was responsible for the first reappraisals of the sense of history in Beowulf after these evolutions in scholarship. He saw Beowulf as ‘an historical poem about the pagan past’ and its poet as having an ‘instinctive historical sense’, even while ‘literal historical fidelity founded on modern research was, of course, not attempted’.32 There was a tension between Tolkien’s reading of Beowulf as ‘historical in spirit’ and his distrust of historians being able to perceive this spirit as more than a ‘moodily elegiac evocation of a lost time and place, of a pagan golden age’.33 But, as Howe remarks, Tolkien was right in that a proper assessment of the poem as ‘a sustained meditation on the past’s richly subtle senses of its own past’ required ‘historians or historicist critics of a less positivistic variety than Tolkien attacked’.34 This ‘meta-historicism’ that Tolkien perceived in the poem occasioned the earliest appraisals of the sense of history in the poem on its own terms: Robert W. Hanning’s ‘Beowulf as Heroic History’ and Roberta Frank’s ‘The Beowulf Poet’s Sense of History’. Frank’s essay, despite containing insightful, albeit incidental comments about the sense of history inside the poem, is more invested in placing it in the political order of early medieval England. She sees in it an attempt at historical verisimilitude on ideological bases: the poet is interested in these early Danes because by the age of Alfred, they were taken to be the common ancestors both of the Wessex royal line and of its Danish subjects.35 Other early attempts at understanding the sense of history in Beowulf see it as dominated by a deeply nostalgic, backward-looking desire to recreate a lost heroic age: ‘The poet makes mythology into history by creating and emphasizing the sense of the past’.36 While Frank remains wary of assigning a genre category to Beowulf’s historicity, Hanning classifies it as one of a series of ‘heroic histories’ going back to classical antiquity, defined as ‘intensely imagined chronicles of an intersection between a “chosen” individual, uncompromisingly dedicated to his destiny and endowed with powers and virtues that make him unique, and a world living in time’.37 These definitions simply describe the poem in terms that Hanning and Tennenhouse abstract from canonical classifications of historical narrative and myth or heroic epic, thus inscribing it in a framework of narrative genres sharply separating ‘history’, ‘imagination’, and ‘mythology’. In this, they echo an oft-repeated description of Beowulf and other GHP as legend mixed with history, existing in a murky no man’s land between

Beyond Germanic Heroic Poetry  11 fact and fiction. However, this assessment assumes that early medieval creators of verse operated in a matrix of genres in which this was a bug and not a feature. For, as John Niles pointed out about Beowulf – a remark that is just as valid for all GHP – ‘most of the debates about historicity that have dominated prior scholarship are posed in terms foreign to the conceptual world of the Anglo-Saxons’.38 In other words, much of the scholarship examining these texts are simply seeking information that they were never designed to provide or are asking the wrong questions of them. And if, as Renée Trilling remarks, ‘historians…have perceived Beowulf’s historical priorities to be confused…that is perhaps because the poem’s historical methodology differs so markedly from that of traditional medieval historiography’.39 These early essays are some of the very few studies taking GHP’s historicity seriously, even though they almost never explicitly categorize it as history-writing. Albeit perceptive and path-opening, they embody what I see as the three main limitations of much of the subsequent scholarship on the historical nature of GHP: (1) the almost exclusive focus on Beowulf and the consequent assumption of its centrality for early English culture (although perfectly understandable – there is simply no other early medieval text, ‘historical’ or ‘literary’, attempting a similarly complex and sophisticated construction of the past); (2) the justification of the historicity of these otherwise ‘literary’ sources by appealing to features derived from both Christian Latinate classicizing historical production and modern institutionalized history-writing as standards for what counts as history; and (3) a set of assumptions on the ideological and cultural work that these texts performed as histories, usually explained as catering for a type of proto-national identity or (more broadly) as based on an essentially nostalgic, conservative impulse to resurrect or at least preserve a heroic past. Even more recent studies are limited by these assumptions to varying degrees. For instance, Andrew Scheil’s insightful piece about ‘a fundamentally historiographic imperative’ at work in Beowulf feels the need to frame his argument by making recourse to the features of normative medieval history-writing (as found in Jordanes, Orosius, Polybius, Paul the Deacon).40 Along similar lines, Shami Ghosh’s book-length study of early medieval ‘barbarian’ national histories takes the bold (and long overdue) step of considering Beowulf and the Waltharius together with such canonical Latinate histories but still considers the former two as ‘secular epics’, albeit embodying an alternative ‘early medieval historical consciousness’ comparable with (although distinct from) that manifested by the former.41 Although I am in general agreement with Scheil’s and Ghosh’s arguments, and my study owes much to their framing GHP as historical in intent and sensibility, I suggest that their underlying assumption (namely,

12  Beyond Germanic Heroic Poetry that GHP needs to be considered in the framework of canonical Latinate salvation-centred historiography) participates in the continuing exclusion of GHP from the category of ‘history-writing’. Rather, there is still much work to be done in order to provide ‘a more comprehensive understanding of how early medieval people thought about their past(s)’, which Ghosh himself acknowledges his book was not intended to be.42 This work would approach different genres and individual sources on their own terms rather than by reference to what usually counts as historiography. And while the book you are reading aims to do just that for GHP, there are two important precursors to my endeavour that fully embrace this approach. Roy Liuzza’s meditation on Beowulf’s historical imagination remains unsurpassed and establishes that the poem is ‘a work of history’, composed by a poet and for an audience fascinated with the old heroic world that ‘was slipping into oblivion, disconnected from the world of written texts and Christian learning, its very memory in danger of being lost’.43 Yet it remains focussed on the one poem which, despite (or perhaps precisely because of) its sophisticated ambiguity and complexity, may not have been as representative as is often assumed, even for what could be done within the ‘Germanic heroic’ tradition, let alone canonical for the broader early medieval English cultural horizon.44 On the other hand, Renée Trilling’s The Aesthetics of Nostalgia is the latest and most thorough exploration of the issue of historicity in GHP on its own terms, and as such, it has provided much of the crucial groundwork that the present study can take for granted. Yet despite being deeply indebted to Trilling, I aim to both refine her main argument and take some of her questions and conclusions in quite different directions. Trilling avoids two of the aforementioned limitations of most studies of historicity in GHP: Beowulf is decentred from its usually assumed position of foremost representative of vernacular history (indeed, it is almost entirely absent from her study), and the entire argument starts from the premise that vernacular verse works in a framework of historicity dominated by nostalgia and past-centred, that is very different from that of Latinate historiography (dominated by hope for salvation and future-centred). However, the most problematic aspect of Trilling’s work stems from its main virtue: in attempting to provide a unitary account of a vernacular historical consciousness that is distinct from that manifested by canonical historiography, Trilling over-extends the reach of the one paradigm of Benjaminian creative nostalgia, which she sees at work everywhere in the corpus of heroic verse and even in some of the biblical OE poems. While I certainly agree with many of Trilling’s conclusions, nostalgia can exhaustively explain neither the full gamut of ideological and cultural work GHP fulfilled nor the variety of engagement with the past at work in different sources gathered under this label. My book thus aims to provide a more fine-grained model that accounts for – besides

Beyond Germanic Heroic Poetry  13 nostalgia – the playfulness, agonistic polyvocality, irony, and even subversiveness of the vernacular theories of narrative and history and ideological engagements present in many of these texts. As other studies have shown, not all GHP offers ‘nostalgia as a refuge’ or looks ‘to the past as the way forward’.45 Elizabeth Tyler has made a convincing argument for an ironic reading of the treatment of the ‘heroic’ vocabulary (and the ‘heroic ‘ethos’ it stands in for) in The Battle of Maldon: rather than ‘naively representing Byrhtnoth and his men as warriors from a fondly imagined heroic past’, the poem juxtaposes this nostalgic potentiality with the grim realities of Viking invasion and thus undercuts the poetic tradition it invokes through its formal conservatism.46 Hence, ‘although the poet is drawn to the nostalgia offered by poetry, he does not find in it solutions for England’s present troubles’.47 In the two final chapters of this book, I extend this ambivalent reading of Maldon to other GHP since reverence for the heroic tradition does not preclude the irony with which this tradition is employed in many of the poems we have. Indeed, once we bypass the usual assumptions of reverence for the ‘heroic’ behaviour showcased in these sources (an issue I tackle next), we are able to perceive that one of the main points of much GHP, from Beowulf to the Waltharius, is precisely to show the socio-emotional economy of honour and heroism as admirable in theory but destructive in practice. Looked at from a broader theoretical perspective, reading GHP as predominantly nostalgic in its ideological thrust has other problems, too. In a provocative overture, Manish Sharma sees this argument as based on ‘an overly schematic opposition between vernacular and Latin, heroic and providential corpora’, whose ‘narrow focus on the formal qualities of Old English verse’ results in a discussion framed around a series of ‘exclusive disjunctions’ resulting in something analogous to the essentialistic understanding of GHP through formulaic theory.48 More colloquially put, when nostalgia (or, more damningly, the ‘Germanic heroic’ paradigm, as we shall see next) is the hammer, every problem or question starts to look like a nail. This is not to say that the nostalgia ideological investment of much GHP is not a powerful explanatory model but that any one paradigm, however useful, cannot capture the whole reality of the often deeply ambivalent vectors of emotion and thought driving the production of these texts, which I seek to do with my own bard–rag-picker spectrum. In a similar vein, Kathleen Davis critiques the recent nostalgic and conservatist readings of GHP as unwittingly encouraging ‘the reinscription of the deeply engrained image of Old English literature as looking mournfully back to an ancient Germanic past’.49 Hence, Davis offers a non-nostalgic reading of Alfred’s preface to Gregory the Great’s Pastoral Care: rather than looking to the past as a way to chart the future, ‘the past society that Alfred invokes is described as Christian and proficient in Latin learning, far from the image of a mythic or distant “Germanic” past – the very idea of which is a function of…philological nationalism’.50

14  Beyond Germanic Heroic Poetry Indeed, as I aim to show in the following, not all segments of a society are invested in the same socio-emotional dynamics when it comes to their past, and even when they appear to be, many conflicting or simply disjunctive drives are at work at all times within any given individual, let alone within any community. In Sharma’s formulation, ‘we need to insist that the past was not just an object of nostalgic fantasy or for stoic Christian repudiation, but a living plane of consistency, a fluid and volatile arrangement of differential tensors’.51 These tensors, forces, or vectors (be they cultural, social, political, emotional, economical – not that one can always differentiate their nature so sharply) need to be accounted for and engaged with just as much as the traditional or conservative forms used in the discourses that voice them. To use Elizabeth Tyler’s terms, the ‘largely ahistorical character’ of the style of much OE poetry should not make us lose sight of its ‘obsession with history’ (especially GHP, in OE but also other vernaculars).52 This very paradox needs to be historicized and accounted for, which Tyler aimed to do by focussing on the enduring formal conservatism of OE verse, which she interprets as a consciously adopted mode of discourse employed in the service of various topical ideological positions rather than preserved by sheer inertia as a default from the ‘Germanic heroic age’ – an important point that my more socially embedded account of the different modes of historical representation in GHP builds on. On that point, John Niles’s work provides crucial grounding for OE heroic verse in the cultural matrix of early medieval England understood not as a remnant of heroic Germania but as a society fully integrated within the broader political life of Europe, looking back to that mythical time for guidance in building cultural narratives aimed to make sense of its ongoing social and political changes and challenges. Niles argues for an understanding of GHP as ‘mythistorical’, a type of discourse that provides a narrative about the shared past of a community (hence historical) that provides a culturally authoritative grounding for a state of affairs that is recent and in need of such a legitimating story (hence mythical).53 Although an important inspiration for my argument in this book, I would suggest that his method is much more productive than its application by Niles himself. In other words, using his insights on how GHP performs cultural work need not result in agreeing with his conclusions on what cultural work it performed – namely, contributing to the early tenth-century ideological impulse of nation-building in England, to which he attributes virtually all OE heroic verse in the form that survives. At the same time, Niles is much more invested in the mythical (understood ideologically rather than anthropologically) rather than the historical aspect of GHP. Although providing illuminating explorations of the ‘social order’ that produced OE texts such as Beowulf, Maldon, and Widsith, Niles still considers them ‘pseudo-historical writings’.54 And while I have derived much pleasure and wisdom from Niles’s work, here I aim to

Beyond Germanic Heroic Poetry  15 look at the cultural work that GHP fulfilled in a broader emotional and social context, avoiding tying down the texts to a specific dating, which Niles does by pinpointing specifically to the political context in which, he argues, they were produced but which may have just as likely been one medium for reception among several. Specifically, Niles makes an otherwise compelling argument that the tenth century would have been a propitious environment for Beowulf and Widsith to be read within a cultural horizon dominated by a newfound national ideology. Yet this does not necessarily mean that they were composed in that environment or at that time. Also, neither one of these two points need follow from the other. The texts themselves, as Howe reminds us, ‘never unambiguously locate the value of that past for an English audience’ since they all feature events distant in time and space: indeed, ‘Beowulf poses its own riddle by not making all the overt claims about national identity that historical epics are usually said to make’.55 As such, as Niles himself recognizes, such a poem is best approached ‘as a polyphonic work whose messages are contingent and sometimes contrary’, rather than as simply ‘reflecting the static conditions of a single or simple age’.56 I suggest then that we can envisage an early Beowulf and Widsith being composed sometime in the early eighth century with other ideological investments in play (for instance, anxiety about race and masculinity or cultural mourning, as I suggest in the final chapter of this book). Meanwhile, a tenth-century audience may well have read these texts within a then-coalescing national historical myth as part of a constellation of discourses on genealogy and geography unavailable or uninteresting to the original audience and composers. At the same time, one can envisage an even later readership (potentially responsible for the very manuscript in which Beowulf survives) that may have read them with a nostalgia for a mythical heroic past in whose terms the early eleventhcentury social and political turmoil could be narrativized and thus understood. I develop this approach to reading GHP through successive layers of early medieval reception in the final chapter of this book. Here I have merely pointed to a few possible alternatives to the nostalgic-nationalistic ideological complex as the dominant explanation for the cultural work GHP fulfilled.

Bridging the Sundering Seas: Carolingian and Early English Vernacular Verse While commentators have raked over OE poetry ever more thoroughly and sustained a lively dialogue between English and later Norse literature, scholarship has paid strikingly little attention to Frankish verse both contemporary with and comparable to the OE material. This oversight is more remarkable because the political contexts of Carolingian

16  Beyond Germanic Heroic Poetry vernacular poetry are markedly better known than those of the OE material. Meanwhile, the kind of approaches OE secular vernacular verse has enjoyed (informed by literary theory and socio-political and cultural studies) have rarely been used on similar continental material. There is much to be gained – as this monograph aims to show – by considering GHP as a broader North-Western European phenomenon not restricted to early medieval England and its socio-political context while resisting the other usual narrative of reducing it to the offshoot of an ahistorical ‘Germanic heroic’ culture. This will enable an understanding of the modes of verse history explored in this chapter as experiments with historical representation using the traditional mode of discourse textualized in the form of GHP and responding to different socio-cultural, emotional, and ideological-political forces while also shaping them. That being said, it must be acknowledged that much of the discussion in the first three chapters revolves around England and OE material. The main reason for this fact is that continental vernacular verse has been (1) rarely considered together with the OE material and are (2) lacking in the sophisticated engagement with historicity that OE GHP has enjoyed – both in the sense of the poems’ own sense of history and of their existence as cultural products one can historicize. On point (1), when OS and OHG poems are considered together with OE GHP, it is usually in the terms of the assumption that they participated in a pan-Germanic cultural horizon, all critical efforts being directed at showing the commonalities between these texts and explaining them by recourse to oral/mythical/pre-Christian reconstructions back-engineered from these same texts – virtually the same ‘Germanic heroic’ paradigm used by early twentieth-century scholars of the ‘heroic age’, such as H. M. Chadwick or Jan de Vries (a point I discuss further later in this chapter).57 The exceptions to this trend have either studied the connections between continental and English vernacular heroic biblical verse or, more interestingly for my approach here, approached GHP as a product of the wave of ninth-century Carolingian ‘Gothicism’ (a newfound elite interest in all things ‘Gothic’, Scandinavian, or ‘Teutonic’), which soon found admirers in England, where it was deployed to support a growing sense of national identity– more on the latter approach in Chapter 4.58 On point (2), the recent work that does grapple with the cultural work that continental GHP fulfilled points to medieval nationalism (such as Jennifer Neville) or falls back on the framework of Christian morality as dominating even the irreverent exercise in irony that the Waltharius clearly is (Rachel Stone) or reads it as political commentary on the dissolution of the Carolingian Empire (Alice Rio). Much of the work on Carolingian secular vernacular verse tends to embody both Lockett’s medievalist bias (all medieval cultural manifestations are traceable and attributable to specific Christian or classical authors) and Howe’s masternarrative bias (all texts produced in one culture have the same didactic

Beyond Germanic Heroic Poetry  17 intention and ideological investment and are explainable by recourse to the same exegetical key – in this case those of classically inspired Christian learning). There are good reasons for this state of affairs outside mere scholarly biases: vernacular culture and language was indeed much more marginal in Francia both before and after Charlemagne than in England, where OE was much more prestigious, becoming the language of ecclesiastical and political institutions during Alfred the Great’s time.59 For these reasons, most of my critical references in this chapter and the next two continue to be related to OE material, albeit reference will be made to continental GHP. Ultimately, as I aim to show in Chapter 4, the conclusions of work on early medieval England can be productively used to illuminate aspects of Carolingian GHP that have been rarely approached on their own terms – such as the cultural work they did, the socio-emotional needs they fulfilled, and the ideological investment they participated in. Meanwhile, as I aim to show there, the work of Carolingianists on the specific political contexts to which the production of continental GHP can be connected proves very useful in bringing to light the possible points of inflection in the socio-political and cultural life of early medieval England at which different OE heroic poems may have been produced or read in specific ways (much harder to do for most of the English material, considering its lack of temporal and obvious political anchorage). We can learn much from approaching both corpora with the same questions and methods, even while recognizing the fundamental differences between the socio-cultural and political contexts to which they belong.

The ‘Germanic Heroic’ and ‘Anglo-Saxon’ Paradigms: A Critique So far, I have traced a brief sketch of the more recent attempts to approach the historicity of GHP on its own terms, which, for all the limitations I have emphasized, have been instrumental in my own endeavour to come  to grips with this issue. Two paradigms I have hinted at are much more dominant in the way GHP is still being read and taught and are much less conducive to understanding its complex relationship with history-writing. : One of them is that of canonical historiography, which refuses the very quality of historical production to GHP, and the other is that of Germanic heroic philology, which sees GHP as the surviving remnant of a Germanic heroic oral tradition and thus only explainable through recourse to this ideal object of study. As Carl Edlund Anderson has remarked, despite an increasing reservation in the claims medievalists make, there are still scholars who ‘aggressively downplay the historical value of a literary source such as Beowulf, but paradoxically might also defend with equal vigour the aboriginal origins of its material, thereby implicitly connecting such materials with historical processes in Scandinavia’.60 Despite periodical conceptual

18  Beyond Germanic Heroic Poetry overhauls in the theoretical approaches to medieval literature and culture, which usually make their way more slowly into the study of GHP, the very notion that the phrase ‘Germanic heroic culture’ refers to a wellestablished historical reality has proven surprisingly enduring. I wish to invoke Stephen Harris’s apt metaphor referring to the ‘strata of competing identities, sometimes contradictory, sometimes consonant’ always being created, reshaped, and passed on to new generations that make up the landscape of ethnic narratives in early medieval England (or anywhere else): ‘As each new generation of Anglo-Saxons put the gods of their fathers into the cupboard, this struggle [“to understand who they were in relation to who they ought to be”] was continually renewed’.61 This may be said about each generation of medievalists and academia generally – scholarly conversations are always joined by new interlocutors, and the topics of interest shift in ever new directions as we try to understand the medieval people whose work we study and, through it, understand our own age and condition. Few students of GHP nowadays would approach these texts as survivors of Germanic paganism, as theologically infused homilies, as the result of a purely oral-formulaic impulse. Yet by using the very phrase ‘Germanic heroic poetry’, we cling to the gods of our fathers and implicitly perpetuate the paradigm it subtends. There has been plenty of work revealing its roots in nineteenth-century masculinism and Romantic nationalism. Here I will briefly refer to some of this research to provide reasons why the ‘GHP’ paradigm may be very useful for a number of purposes but unproductive for understanding the historiographical dimension of the corpus it purports to describe. The ‘Germanic heroic’ paradigm, uncritically perpetuated in even the more perceptive studies of GHP, flattens the variety of the texts gathered under this label and of the socio-cultural and political contexts from which they emerge into an ahistorical model. The texts lumped together under this label are thus tacitly assumed to be traceable to the oral poetic reservoir of ‘Germanic antiquities’, to participate in the same ‘heroic ethos’, to have the same social functions, and to fulfil similar cultural needs.62 Trilling perhaps best sums up the main point of my critique: the oral heroic tradition these poems presumably stem from ‘is a product not of generations of Germanic storytelling but of poems committed to manuscript in the tenth and eleventh centuries and providing our only evidence of such a tradition’, and this narrative construction is created by the fact that ‘these poems lay claim to the representation of history by situating themselves alongside the legendary tales of the Anglo-Saxon pagan ancestors’, even though ‘the heroic world of pre-Migration Germania bears little resemblance to the world of late Anglo-Saxon England. The Germanic lord was as anachronistic then as he is now’.63 This is not to say that the technique of alliterative verse and its vocabulary of formulae, the socio-emotional conventions shaping the world of

Beyond Germanic Heroic Poetry  19 these texts and their story-world do not predate the writing down of the texts we label ‘GHP’ or were invented out of thin air rather than going back to some earlier tradition of oral verse. But none of these likely facts supports some of the claims often derived from them: (1) that there was ever a coherent tapestry of stories, a ‘matter of Germania’ which was shared by all Germanic-speaking ethnic groups at some point; (2) that the ‘heroic’ socio-cultural values and structures in this fictional world were fully shared or even idealized by their intended audiences; (3) that the creators of the corpus of GHP that survives believed that in doing so they were participating in a pan-Germanic ‘common civilization of those who spoke the Germanic languages…to which they clung tenaciously through the centuries’ (which Eric Stanley was already critiquing in 1975);64 and (4) that in composing these texts, they sought to faithfully record already existent oral ‘poems’ out of an antiquarian impulse.65 The very notion of a Germanic culture shared by the people speaking early medieval Germanic languages is derived from nineteenth-century philological nationalism, whereby language is an organic expression of culture, and therefore the people speaking related Germanic languages shared a larger Germanic culture, expressed in ‘heroic poems’: a concept unlikely to have been shared by the early medieval English or, indeed, as Shami Ghosh argues, by any other early medieval ethnic group.66 As Roberta Frank argued, ‘The literary category we call “Germanic legend” is ours, not theirs, and it is not so much a description as an explanation… an interpretation of the evidence that has the potential to mislead’.67 Indeed, in the following, I aim to follow Niles’s approach: ‘The word “Germania” will be almost as absent from my discussion as either that noun or any synonym of it is absent from the vocabulary with which most Anglo-Saxons talked about their Continental origins’, outside of my use of the acronym GHP, which seeks to distance the corpus it describes (a useful category, albeit overly broad) from the paradigm underlying its descriptors ‘Germanic’, ‘heroic’, and even ‘poetry’ (unhelpful for my purposes here).68 It is not only that these notions are conceptually fuzzy and overly capacious, engulfing a range of implied meanings ranging across a dozen disciplines (archaeology, philology, legal history, etc.) in each of which ‘Germanic’ and ‘heroic’ mean very different things: from the poetical technique of GHP, the story-world it refers to, the non-Roman system of laws, to the feuding and war-loving late antique immigrants to the British Isles.69 It is also their linkage that is problematic, rooted as it is in the nineteenth-century ‘rise of scientific philology, with its emphasis on the Germanic origins of the English language’, which ‘went hand in hand with political pan-Germanism’, a political construction allowing ‘people in an increasingly militarized Europe [to] glamorize such sentiments as “love of the feud”’ and ‘to ascribe such sentiments to the Anglo-Saxons, drawing on Tacitus for authority in that regard’.70

20  Beyond Germanic Heroic Poetry Yet, as Walter Goffart pointedly reminds us, to proffer the Nazi uses of the ideology of Germanentum as an ‘all-purpose bogeyman’ is to obscure the long tradition of linking the Romantic idea of Germanic antiquities to present-day Germans (or that of ‘Anglo-Saxon England’ to present-day English or white Americans) that began ever since Tacitus’s Germania became known to humanists in the fifteenth century (or, as Allen Frantzen argued for ‘Anglo-Saxon studies’ since the first recoveries of OE texts began as part of the English Reformation): National Socialist ideologues were petty players in a game whose stars have included the cream of German erudition, scholars whose memory commands esteem and admiration. Far from being a disowned side road of Nazi lunacy – how convenient to think so – the illusory connection of Tacitus’s Germania with modern Germany remains fundamental to German self-awareness.71 Not that this is a singularly German problem – different flavours of philological nationalism abound in the Anglo-Saxonism cultivated in nineteenth-century England (often in a context of superiority to the Irish) and in the United States from the Founding Fathers to later imperialistic ambitions.72 In this, as Stephen Harris, John Niles, and Allen Frantzen have shown, Germanicism is deeply linked to Anglo-Saxonism, which comes from a similar ‘desire for origins’: the belief that ‘[t]he origins of English literature…reflect fundamental characteristics of an English race’.73 Indeed, such beliefs have never quite disappeared from the field, as a number of scholars have recently revealed. Already in 1997, John Niles warned how hard it was to imagine the vehemence with which some professional AngloSaxonists resist this suggestion [that ‘Anglo-Saxon England’ is ‘a creation of language, a rhetorical trope with no other corresponding content than the mental associations scholarship and political use of the phrase have imprinted in us and our respective society’], not to mention some other people, most of them no part of the academic world, who take their Anglo-Saxon racial origins seriously, and seek ‘to confirm the tensile strength of such lines they appeal to genealogies, genetics, archaeology, the statements of reliable authorities’.74 This resistance is alive at the time of this writing, manifested in the controversy around repeated calls within the field for discontinuing the use of the phrase ‘Anglo-Saxon’ as a racist term almost never used by the early medieval people, territory, or corpus of texts it is usually used to describe. A number of recent studies of the term painstakingly trace the intellectual history of the word and the discipline of Anglo-Saxon studies, both grounded in nineteenth-century British colonialism and American

Beyond Germanic Heroic Poetry  21 white supremacy.75 Catherine Karkov sums up the problem with refusing to acknowledge this history: To value an outdated term – ‘Anglo-Saxon’ – over the lives and careers of colleagues is both absurd and horrific. It is a clinging to the past that remains stubbornly blind to the racism of the past and its direct connections with the racism of the present.76 Yet apart from the heavy baggage of racism and ethno-nationalism they carry, the terms ‘Anglo-Saxon’, and ‘Anglo-Saxon England’ are fundamentally unhelpful as they flatten into one (culturally, ethnically, racially) homogeneous monolith what was a heterogeneous assemblage of spaces, temporalities, polities, ethno-racial communities, socio-cultural groups, and modes of discourse. As Susan Reynolds has shown, ‘the inhabitants of what would come to be called England were…using the simple word “English” (Angli, Anglici) to refer to themselves’, and ‘except when they were making vague but grandiloquent claims to supremacy over the whole island of Britain, the normal title of all kings from the later tenth century on seems to have been rex Anglorum’.77 Meanwhile, temporally, the ‘Anglo-Saxon period’, Reynolds argues, ‘is too long and contained too much change to form a significant unit for study – a coherent “period”’ since until the tenth century any broadly ‘Anglo-Saxon’ ‘manifest destiny of political unity’ was constantly undermined by political conflict.78 At the same time, this notion erases the different ethnical communities that lived in early medieval England – which I will discuss more in my final chapter.79 At the same time, if we are to take poems such as Widsith as expressing ‘an awareness of shared origins and intertwined histories’, a sense of Anglo-Saxon or Germanic identity, we will have to concede that early medieval English people thought of the Hebrews, Babylonians, Ancient Greeks, and Romans as their ancestors as much as the tribes of Germania since they all feature in the poem together.80 Accordingly, we need to distinguish between ‘speakers of Old English’ (not all of whom would have been part of early medieval polities ruled by English leaders), ‘Anglo-Saxons’ as the ideal subjects of post-Alfredian Wessex kings (not all of whom would have been speakers of OE or even if they were, that they identified as such rather than as Mercians or East-Saxons), and the early medieval people living on the territory of modern-day England (many of whom would have been neither speakers of OE, nor belonging to any English, Saxon, or ‘Anglo-Saxon’ polity).81 The consequences for the literature of the periods and places we gather under the portmanteau ‘Anglo-Saxon England’ should not be understated. As Elizabeth Tyler reminds us, while the conventionality of the poetic tradition…encourages us to think of Old English verse as the product of an Anglo-Saxon England which

22  Beyond Germanic Heroic Poetry changed little from the fifth to the eleventh centuries, this composite Anglo-Saxon England is a fiction which impedes our ability to understand Old English verse.82 Tyler provides an instructive comparison which brings into light ‘[t]he distorting effect of the periodization of Anglo-Saxon history’ as a single era extending from the fifth to the eleventh century by remarking that ‘a similar time span in the history of the area of Gaul which was to become modern France is divided up into five different periods: Late Antiquity, Merovingian, Carolingian, post-Carolingian and Capetian’, which ‘indicate much more fully the major social, political, economic and cultural changes which took place in the early medieval period throughout Europe, than does the notion of Anglo-Saxon England’.83 During this six-century period, OE poetry maintained its style while adapting to and assimilating tremendous religious, political, and social change. But there is a deeper misapprehension at work here, which Stephen Harris has revealed: it is not that OE or GHP is an expression of a pre-existent ethnic, racial, or national identity (or of culture, more generally) but that culture is the perpetual creation of the textual corpus through the conceit of an oral tradition the latter purports to be consubstantial with – ‘the Volkslieder may not actually be a vehicle of cultural memory, but a static cultural fiction maintained to declare the borders of race at any given time’.84 For all of these reasons, I use the term ‘Anglo-Saxon’ only when quoting and referring to previous scholarship using the term or when discussing the field of ‘Anglo-Saxon studies’ or ‘Anglo-Saxonists’ to signal its philological nationalist heritage. Otherwise, I instead refer to early medieval England and early medieval English people, artefacts, or texts. And rather than understanding GHP as an ahistorical tradition equal to itself, suspended above cultural, social, and political change, I seek to reveal the points of inflection and vectors of change that are often manifested even in the shape of the texts themselves (outside the constancy of the metrical form) and in the constellations of texts and discourses they are contemporary with, next to which they are compiled in manuscripts or which show similarities of subject and treatment. Stephen Harris sums up the main problem with the favourite source (or rather, ‘origin myth’) of these various strands of philological nationalism: Tacitus’s Germania ‘is inappropriate evidence for an accurate description of an Anglo-Saxon society nearly a millennium its senior’, as part of a long tradition of classical accounts of an ‘idealized barbarian society’ that ‘often served as implicit correctives to Roman mores, as perhaps idealized versions of Anglo-Saxon society painted in Old English literature may have served as correctives to Anglo-Saxon mores’.85 This takes us to the other two terms of the triple conundrum of the GHP paradigm: ‘heroic’ and ‘poetry’ (often in tandem more or less covertly

Beyond Germanic Heroic Poetry  23 with ‘orality’). Unsurprisingly, Tacitus’s picture of the ancestral Germanic society is also the main source/origin myth for the paradigms that subtend them – namely, that these texts idealize a ‘heroic code’ of conduct, shared by both the fictional society that they depict and the historical societies that produce them and, secondly, that their manuscript, textual embodiment is merely a vehicle for a fundamentally oral reality, an urtext to which reference can then be made as if it were a real entity. Much has been written disabusing us of these notions, yet they remain strong. The ‘heroic code’ that all GHP presumably idealized and that its audiences uncritically shared is ‘a literary creation rather than a lived ideology – no one poem or text defines it; rather, these values are a constellation of ideas found throughout the wider corpus of Anglo-Saxon [or indeed, ‘Germanic heroic’] literature’.86 Yet although the fact that the comitatus was a literary rather than a lived ideal has been firmly established’, the assumption that the corpus of GHP idealizes heroes and heroic behaviour is still strong.87 Indeed, I will not rehearse the compelling arguments of Niles, for instance, against a long-standing assumption of ‘a kind of collective lemming-like impulse towards self-destruction’ of the English warriors in The Battle of Maldon, a comitatus ethos they presumably shared with the fictional warriors of older GHP and the Germani of Tacitus – in fact a nineteenth-century creation of Romantic martial hyper-­masculinism or, in Niles’s terms, of the ‘post-Romantic cult of suicidal devotion’.88 As David Clark and a number of scholars have shown, ‘the putative link between Tacitus and Maldon has a long and dubious history of association with fascist and patriarchal ideologies.89 Meanwhile, the very notions of ‘hero’ and ‘heroic’ used as a shorthand for a number of behaviours, characteristics, and states of mind but which usually revolve around (usually) male violent behaviour attempting to boost one’s honour, are (once again) creations of modern scholarship. They have more to do with a mixture of nineteenth-century Victorian hero-worship and the roughly contemporary evolution of Germanic philology as it tried to construct its own Heroic Age by analogy with the one sung by Homeric Greek epic.90 There, the hērōes hagnoi (‘holy heroes’), the olbioi hērōes (‘blessed heroes’), were actual political-religious figures around which cults grew hand in hand with poetic celebration within a cultural horizon that was very different from that of Tacitean Germania, let alone those of the later Germanic-speaking ethnies.91 Meanwhile, Germanic languages have no word for ‘hero’ – although they do for ‘warrior’, ‘young man’, ‘haughty exile’, none of which quite corresponds to the notion of hero modern readers are used to reading into GHP, at least.92 What is clear is, as Clare Lees has shown, is that ‘the patriarchal project of Beowulf [or indeed, GHP] criticism’ has created ‘a reading of the poem that naturalizes gender and thereby promotes masculinism’, even while in the text itself ‘there is nothing sentimental about its ambiguous and ambivalent gaze on men’.93

24  Beyond Germanic Heroic Poetry I return to the deep undercurrent of anxiety about masculinity, honour, and the ever-present spectre of shame in GHP in the last chapter. Here, I merely wish to point out that GHP itself offers up the ‘heroic life’ (a life in pursuit of honour, gained by deeds of more or less socially condoned violence) as a subject of deep meditation and a focus for projecting a complicated mix of admiration and apprehension. It is commonplace of the more recent scholarship on Beowulf, for instance, that the poem neither unambiguously celebrates the dom of a heroic role model [nor] offers a reassuring guide to the present order; as each story falls short of its role as an exemplar, the reader of Beowulf is drawn into a more critical position toward the life of Beowulf himself, more attuned to its own ambiguities and insufficiencies.94 Indeed, in his posthumous eulogy, Wiglaf referred to Beowulf’s deeds as dada dollicra (‘foolish/audacious deeds’, Beowulf 2646) and later took ‘on the voice of the poet, marveling at a heroism that he cannot quite condone’ by suggesting that the Geats suffer because of Beowulf’s stubborn will (Beowulf: 3077–84a).95 Meanwhile, the Waltharius poet ‘carefully dismisses the concept of a heroic age’ diminishing or ridiculing ‘the traditional grandeur of leading figures’ and of heroic gestures, such as fighting for treasure or haranguing one’s man to battle, turning ‘[h]eroic legends, by means of a highly elaborate Latin literacy…to a divertimento of connoisseurs’.96 But in even the most straightforward and unsubversive GHP, ‘protagonists display, through their heroic acts, also some form of monstrous behavior which severely undermines the whole notion of heroism’.97 And, as Ralph O’Connor shows, a wide range of corpora of heroic literature (Old Norse, Old Irish, Ancient Greek, alongside GHP) ‘did not always invite their audiences to identify with the dehumanized warrior in any simple manner. Sometimes forthright disapproval is implied’.98 To sum up, the heroic culture inside GHP cannot simply be taken to mirror the cultures of its various audiences, nor can it be understood as an idealized model of behaviour. Even if many in the audience may have lived lives revolving around war-making, honour-based culture was not the only habitus available but clashed and overlapped and interacted with other modes and ideals of behaviour. At the same time, GHP (especially the more complex and self-referential texts) problematizes heroic culture and allows participants in it to work out their social, cultural, emotional, and moral anxieties against a historical dream-screen. But even the notion of ‘poetry’ applied to these texts is an anachronism. Niles has argued that the need to distinguish between poetic genres was probably never felt by most early medieval people in England, where the main vernacular term for poetry, gied(d), was used ‘with reference to

Beyond Germanic Heroic Poetry  25 many things, from poetry, to prophecy, to healing charms, to riddles, to heightened speech’, while other words such as leot ‘poem, verse’, song ‘song’, and spell ‘story’ or ‘performance’, ‘are almost impossible to pin down in terms of generic distinctions’.99 I explore these terms in Chapter 2. The main implications here is that this type of discourse is set apart from the language used in everyday life by its heightened diction, metre, special lexicon, and intricate figures of speech: giedd ‘proclaims itself as a form of ritualized discourse’, ‘a vehicle of wisdom, not just knowledge’.100 Meanwhile, modern English ‘poetry’, when used in a general sense, evokes the conceptual schema of the post-Romantic individual genius seeking to express oneself through the medium of verse, not seen as a treasury of formulae and figures of speech but as a personal form of expression resulting from the struggle to develop a deeply original style. By contrast, as Niles evocatively explains, When Hrothgar speaks a long cautionary gid or sententious speech to Beowulf (Beowulf 1723b), when the unnamed woman of The Wife’s Lament delivers a giedd that is powerfully laden with personal grief and anger (1a), when the speaker of The Seafarer speaks a soðgied ‘song of truth’ (1b) that expresses hard Christian doctrine in quasi-allegorical terms, just as when Widsith, a man who is gydda gleaw ‘knowledgeable in songs’ (139a), delivers a formal address that incorporates a great deal of ancient lore, then the members of a speech community are invited to enlarge their personal consciousness and deepen their understanding of the world by absorbing words of more than ordinary power.101 The fundamentally social, collective dimension and formally traditional nature of vernacular verse in early medieval England has been amply documented: unlike modern poetry, such ‘Highly conventional poetry… relies on its audiences’ familiarity with tradition’, which therefore ‘is not located in a poet alone, but in a wider community’.102 Indeed, a common trope among scholars working on OE poetry (especially those doing so from a critical theory perspective) is that this corpus works within the logic of an accretive, reproductive authorship (a very different concept from the modern one), whereby the texts are always open to reworking on subsequent reading and copying. I explore this notion at greater length in Chapter 3. Here, I only wish to point out that while this is a powerful idea that has increased our understanding of medieval textual production, it runs the risk of cementing a vision of this corpus as fundamentally traditional, not only in form but also in the cultural and ideological work it is tasked with. Tyler warned that the tradition of oral-formulaic theory that underlies these more sophisticated arguments tends to be more aware of ‘the social, rather than historical, aspects of tradition’: ‘As traditional, the poetry

26  Beyond Germanic Heroic Poetry is seen to perform roles which are conservative, socially cohesive and rooted in communal rather than individual experience’.103 Yet this need not be the case, as Tyler’s work shows. It is one of the purposes of this book to show that the traditional mode of expression of much GHP was put to sometimes very untraditional uses and could become the vehicle (via authorship) or focus of projection (via readership) for a wide variety of ideologies and forms of emotional investment. Coming back to the main thread of the argument, not all GHP is alike, and we should differentiate between several categories with respect to how they historicize different pasts. As the work of Liz Tyler and John Niles demonstrates, the apparently ahistorical formal qualities of vernacular verse obscure the often innovative and subversive cultural work that it accomplishes in different political contexts. Here, I have provided reasons why notions such as ‘GHP’ or ‘Anglo-Saxon England’ flatten this variety and are thus unproductive models for understanding the different cultural, political, and emotional work that these poems fulfil.

What Is ‘History’, Anyway? I now turn to the issue of what counts as history and history-writing, pointing to reasons why GHP can be considered a form of historical production while emphasizing the necessity of moving past the post-Enlightenment framework of genre categories it is a part of (another set of ‘gods of our fathers’) if we are to make sense of the different ways in which this form of discourse narrativizes the past. As we have seen, GHP has often been considered of historical value, as a source either for reconstructing the events of the Germanic heroic age or, more productively, ‘for reading the ethnography of cultural and geographical belief in Anglo-Saxon England’ and Carolingian Francia.104 But it is important to recognize that GHP not only has historical value but also historical ‘formal allegiances and sensibilities’ and that it is possessed by ‘a fundamentally historiographic imperative’.105 This has not been the case outside the handful of scholars (virtually all working on OE verse) surveyed earlier. Any general survey of medieval historiography only recognizes genres, such as universal history, national history, institutional history, and biography/hagiography, as historical in nature and intent.106 Other genres of writing about the past claiming veracity, such as epics, romances, or heroic verse, are generally excluded from the category of history-writing by definition.107 Interestingly, there has been a wave of interest in twelfth- and thirteenth-century texts standing on the divide of ‘history’ and ‘literature’ such as Wace’s Roman de Brut as histories in vernacular verse rather than literary experiments and more generally in the way they use fictionality and historicity in complex ways and for novel cultural-political purposes.108 But this approach is also based on ‘an irreconcilable difference in implied veracity between the two kinds of narrative’.109

Beyond Germanic Heroic Poetry  27 This development in the scholarship on high medieval history-writing and literary production has been framed in the terms of an evolution of self-conscious fictionality (a recently contested notion) and a parallel growing professionalization of history-writing at this time in the Middle Ages.110 As such, there has been very little cross-pollination with the work of medieval Anglicists on the historicity of OE heroic verse, which has mostly built on the Tolkienian framework of historicity as a function of artistry and later as an exercise in nostalgia.111 This was in some ways for the best since the unhelpful model of ‘fictionality’ as opposed to ‘historicity’ that dominates the scholarship on both high medieval history-writing and ‘literature’ (romance, epic, saga) was thus mostly avoided by students of early medieval English canonical historiographic and poetic production who had to come to grips with the relationship of their object of study to the obviously non-classical story-world it is based on.112 Yet the unwarranted implicit assumption that ‘the more made-up a narrative is, the more fictional it is’ persists, and ‘literary’ techniques in a medieval historiographical source are often pointed to as arguments that the work is not ‘historical’, either in its intention or in its reception.113 At the same time, scholars of OE verse have generally not been interested (with notable exceptions, such as Elizabeth Tyler and Catherine Cubitt) in entering conversations opened up by work on high medieval narrative demonstrating that formal or functional distinctions do not help us place such texts in either category: ‘Historical narrative in the Middle Ages was any narrative willingly believed, and the limits of belief were set by the resources and attitudes of each audience’.114 I discuss the issue of veracity at length in Chapter 3. For now, I only wish to trace the general lines along which the conversation of these textual genre boundaries has been carried out. Overlooking the fact that texts so obsessed not only with the past but also with the meaning of history as Beowulf or Deor participate in historical production is not explicable just by pointing to the different evolutions of these fields of medievalist inquiry. It is also true that Germanic heroic poems are ‘unruly poem[s]’, which do not stay ‘within the framework of our generic expectations’.115 A more significant distinction for early medieval English textual communities was not that between ‘literature’ or ‘fiction’ and ‘history’ but between prose and verse, ‘the latter marked by its elevated diction and artificial conventions, as well as by metrical forms’, that required some form of training (not necessarily formal, at least for the vernacular) to be able to be properly enjoyed, let alone produced.116 This distinction seems to traverse other boundaries that may seem more significant today, such as those between Latin and the vernacular or the religious and the secular: ‘Even the unlikeliest material could be versified, including a calendar of saints’ feasts (The Menologium), the preface to a rule for canons (Vainglory), and the philosophical ruminations on God’s foreknowledge and human free will in

28  Beyond Germanic Heroic Poetry Boethius’ Consolatio philosophiae’.117 It is not unlikely, then, that at least some heroic verse may have intended to be historiographical and read as such, regardless of any modern assumptions about poetic genres as improper from history-writing. For what we call historiography and fiction are ‘genres of writing, not bundles of fact or nonfact in verbal shape’, and what makes them are ‘the rules of the writing game, namely, to the premises, conventions, and undertakings that attach to the discourse as an affair between writer and audience’.118 Thus, according to Meir Sternberg’s argument that biblical narrative was seen as a kind of historical discourse, there are more important questions than are usually posed to these texts: ‘What kind of contract binds them [author and audience] together?’ And, ‘What do both sides expect historiography (or fiction) to be and do?’119 Genre is ‘a communicative transaction’, making these questions never answerable a priori but only when looking at the entire cultural system of meaning in which they function, and as such in any culture, there is a multiplicity of narrative genres, only some of which may cleave to the rather recent ‘linkage of history writing to documentation’, which would deny heroic verse or biblical narrative ‘the name of history telling but not the claim (and effect) of truth telling’.120 In these terms, even framing GHP as history-writing is an exercise in imposing post-Enlightenment genre categories on discourses that predate them and eschew their strictures. Eileen Joy and Mary Ramsey (echoing Jean Luc Nancy) come, perhaps, the closest to the mark in redefining what it means for these texts to be historical: ‘When we understand history…not as “some presence hidden behind the representations” but as “the coming into presence, as event”’.121 In other words, history is not the ‘content’ of a text, separable from the textual formalisms it is encoded in but dwells in the very act of selecting, hierarchizing, and giving narrative shape to the thousands of events happening in the world at all times and finding order and causality in their Brownian motion and inextricable interdependences. Hence it is the method (implicit or explicit, consciously innovative or simply ‘traditional’, or anything in between) governing these acts of selection, storytelling, and drawing causality that makes a particular discourse a historical production, not the quantity of factual information it contains. Hayden White’s oft-quoted (but rarely understood in its critical context) work is instrumental in bringing to light this understanding. His main point (shared by many philosophers of history like Frank Ankersmit) is that ‘the historical work is a verbal artifact, a narrative prose discourse, the content of which is as much invented – or as much imagined – as found’.122 For we do not actually have access to any real past but only to the traces it left behind, such as stories, documents, artefacts about which we construct stories that would explain their existence in terms that satisfy the human predilection for narrative structure and closure, encoded

Beyond Germanic Heroic Poetry  29 differently according to various sets of culturally embedded ‘patterns of eventuality dramatized’ and social contexts.123 This is ‘not to say that “facts” do not exist, rather that the most they can offer…are the elements that can be made into a story by emplotment’.124 Another one of White’s insights lies in dismantling the age-old belief underpinning historiography that history consists of congeries of lived stories, individual and collective, and that the principal task of the historian is to discover these stories and retell them in narrative form, the truth/accuracy of which would reside in the degree of correspondence of the story told to the story lived.125 The problem is that, of course, people do not actually live stories ‘either individually (at the level of “real-life” stories) or collectively (at the level of, say, metanarratives which give purpose and meaning to the past as, for example, in Marxist or Whig theories of history)’.126 To see the past as a story or series of stories (‘of great men, of wars and treaties, of the rise of labour, the emancipation of women’, etc.) is ‘to give to it an imaginary series of narrative structures and coherences it actually never had’ – in other words, it is to invent a fiction.127 This insight rubs many historians the wrong way,128 but Hayden White’s point is not that nothing is knowable about the past or that there is no point to history. Rather, he merely observes that history (including its current institutionalized scholarly form) is a genre of narrative whose formal requirements do not make it a ‘truer’ or ‘more realistic’ mode of historical representation than, say, heroic verse, or Arthurian romance. Modern historiography’s ‘demand for scientization…represents only the statement of a preference for a specific modality of historical conceptualization, the grounds of which are either moral or aesthetic, but the epistemological justification of which still remains to be established’.129 This understanding of narrative and history opens up some very interesting possibilities. Since ‘the possible modes of historiography are the same as the possible modes of speculative philosophy of history’, White sees these modes as ‘formalizations of poetic insights that analytically precede them and that sanction the particular theories used to give historical accounts the aspect of an ‘explanation’.130 Thus going beyond the myriad scholarly conversations based on the ‘fact versus fiction’ paradigm, there emerges the possibility to see as many visions of history as there are genres of narrative that are historiographical according to the contract between their creators and their audiences rather than according to the formalisms of one mode of history-writing (whether that is classical Latinate or modern post-Rankean). Thus GHP can be treated not as ‘imperfect histories’ but rather as ‘particular products of possible conceptions of historical reality, conceptions

30  Beyond Germanic Heroic Poetry that are alternatives to, rather than failed anticipations of, the fully realized historical discourse that modern history is supposed to embody’.131 In John Niles’s words, a poem like Beowulf is ‘the expression of an ars poetica that is radically unlike that of most modern verse, just as it practises an ars historica that is incomparable with modern standards of historiography’.132 It is these artes historicae, these formalizations of poetic insights, these theories of history and narrative truth and meaning informing the shape of different kinds of GHP that are the red thread guiding this entire book. And, as we shall see, while not all GHP is quite so subversive, it certainly belies the genre categories we have imposed on it. My task in what follows, then, is to make explicit the alternative conceptions of historical reality underlying the specific modes of historical production at work in the different heroic poems I discuss here while being careful not to flatten them into one historical consciousness that would explain all of them by reference to one totalizing vision. The model I propose next is not meant to propose a new set of genre categories in which to pigeonhole the various texts gathered under the portmanteau GHP but rather to trace the underlying alternative theories of history and narrative representation with which these texts worked, alongside the way they functioned in various socio-cultural-political-emotional economies of meaning, suspended between their creators and their audiences.

Alternative Modes of History and Vernacular Theories of Narrative Representation To take the usual subject of studies of historicity in GHP, even approached in terms of ‘fact versus fiction’, Beowulf is much less fantastic than it may seem: it gives Grendel and his mother human characteristics and ‘a historical human plausibility, hence their descent from Cain’.133 The ‘digressions’ of the poem into dynastic history serve to embed these ‘monsters’ (our terms, not the poet’s) into a historical context. In a nutshell, the poem was more likely read as a type of history than as a fantastic tale by at least some of its early audiences. The division of the poem into ‘“historical” and ‘fabulous’ elements is…based on a simplified assumption about “what people really believe”’, as Derek Pearsall remarked, ‘one that we can hardly make even about the materials of our own experience, in a sceptical and empirical age’.134 I am in full agreement with the arguments of Nicholas Howe and Andrew Scheil that, rather than a liber monstrorum, Cotton Vitellius A.xv (the Beowulf manuscript) was more likely conceived and read as a historiographical miscellany, a liber de diversis historiis, anglice (Scheil’s formulation, echoing Kenneth Sisam’s original one): ‘Would anyone have felt the need to see this manuscript as a book of monsters if it did not

Beyond Germanic Heroic Poetry  31 contain Beowulf?’ And, indeed, with regards to its often assumed canonicity, ‘[c]ould it be that the Anglo-Saxons thought these prose texts [The Letters of Alexander, The Wonders of the East] comparable to Beowulf as entertainment, or even preferred them?’.135 More provocatively, the Beowulf manuscript could have been ‘a kind of laboratory for more extreme experiments with inclusive disjunctions’, such as the ones Manish Sharma reveals dwelling in Judith and Beowulf, both of whose heroes are ‘disconcerting concatenations of affects’, being ‘at times monstrous, at times Christlike, resisting assimilation to both secular-heroic and Christian models’. 136 Beowulf’s dragon, Grendel, and the sea monsters fought by the protagonist while swimming (or rowing) for heroic lengths of time are recounted in the same narrative breath as verifiably historical events such as Hygelac’s raid on Frisia or the Yngling Swedish royal dynasty. There is no sign that such other-worldly deeds or beings are to be taken figuratively. Thus one can envisage at least some textual communities in early medieval England for whom the Scandinavian tribal history presented in the former text was no less true and accurate than Bede’s account of English origins, although each of these texts functioned in different ‘horizons of expectation’ and systems of socio-cultural meaning, which need not have been disjoined social milieus.137 One can imagine individuals and, indeed, textual communities conversant in both of these horizons that would have been able to enjoy both Beowulf (or similar complex GHP) and Bede’s Ecclesiastical History. Indeed, the very existence of a text like the Waltharius implies such a community as a site for its production and reception. Such claims to veracity are not unprecedented in medieval narrative, but they do not need to be ascribed to ‘credulity’ but to a different conception of what history composed in this heightened verse form can contain or is supposed to accomplish (narratively, socio-culturally, emotionally). For instance, Erich Poppe has argued that medieval Irish narrative prose in the heroic ‘cattle raids’ and ‘destructions’ genres were considered by their contemporaries as historical to the same extent legal tracts more serious stoir (the term itself derived from historia).138 Elizabeth Tonkin’s work on West African oral history is a source of fresh insights on these issues: realism, she explains, is a ‘culture-bound judgment of likelihood’, which involves an audience’s assessment of ‘whether the linguistic and genre patterns, as well as the content of the discourse, are appropriate for its representation’ of a reality.139 Thus truth (‘that elusive historical goal’) ‘can also lie in the intersection of narrator and discourse, where we have to see how accounts are authorised’, an endeavour that ‘varies generically and politically and culturally, as does the kind of truth claimed, expected or accepted’ – indeed, ‘[a] genre of discourse can carry with it a claim to a particular kind of truth’.140

32  Beyond Germanic Heroic Poetry Tonkin stresses the deeply interpersonal nature of all genres, written or oral: they are ‘social practices’ and ‘active mediators of structuring representation’.141 Since genres depend on an implicit ‘agreement between writer or speaker and reader or listener on what sort of interpretation is to be made’, they are ‘dynamic’ and ‘situational’ processes that are ‘not explicable by form alone, even in written literature’.142 This recognition of genre as an essentially social and situational reality rather than a codified category of discourse is important for the approach I take here, which aims to extricate GHP from the purely semantic and formal dimensions of oral-formulaic and philological criticism, as well as from the classical or post-Enlightenment assumptions of narrative shape, realism, and veracity in canonical historiography. In arguing that ‘Germanic heroic’ poems are in fact part of several genres of history-writing, I do not claim that these were formally codified categories of texts, with clear rules on narrative structure and the extent of invention and veracity. Rather, I suggest that these texts were originally produced and read within horizons of expectations that are more fundamentally akin to historiography than to anything we might term ‘fiction’, ‘legend’, or ‘myth’ or to what medieval Latinate ecclesiastics with Cicero/Isidore on their minds might have derogatively called ‘fabulae’. At the same time, such an account of the genres of ‘Germanic heroic’ verse would have to be based on and feed into a more thorough understanding of how these texts functioned socially, culturally, and emotionally (and not just politically, as has usually been the case). In Tonkin’s terms (adapting Karin Barber), such an understanding of what she calls ‘representations of pastness – a cumbersome phrase, but more exact here than “history”’ has to be triadic: To grasp their historical intent we need to view [‘histories’] as literature; to grasp their literary mode we need to view them as part of social action; to grasp their role as social action we need to understand their historical intent.143 And while the work of John Niles has contributed greatly to the first two aspects, and while there is a growing wave of research on the history of emotions focussing on early medieval (mostly OE) poetry, there is still much to do, as this book shows. The present chapter focuses on the first and particularly the third aspect of the triad (the literary form and the historical intent, respectively) and the last two chapters aim to reveal the role of GHP as social (hence also political, cultural, and emotional) action. Yet, as Tonkin (via Barber) suggests, these concerns are inextricable from each other throughout this monograph and, indeed, in this case, are based on the lexicological, semantic, philological (and more broadly, theoretical) work of the second and third chapters.

Beyond Germanic Heroic Poetry  33

The Bard, the Rag-Picker, and the Nostalgic Scribe Here I briefly explain the tripartite model of genres of history-writing gathered under the label GHP promised from the book’s title, followed in the next section by a lengthier demonstration of their explanatory power harnessed to answer the questions I pose. I have so far established that not all texts described as GHP are alike and that we, therefore, need a system of categories that would account for these differences with respect to how they historicize different pasts – even if there are no vernacular words to describe them. Although both The Battle of Brunanburh and that in Beowulf, for instance, are seriously invested in constructing stories about the past by using the conceit of their being part of an ancestral oral tradition, the ways in which they do so and the shape of the historical representations they are constructing are very different. Based on these differences, I propose three modes of historical representation that I see at work in various texts commonly read as ‘GHP’, although not as separate portions of a spectrum but rather as modalities (or even cultural and narrative toolkits) whose elements were often used in conjunction. (1) The bardic mode (e.g., the OE Finnsburg Fragment and the OHG Hildebrandslied) produces brief episodes understood to be part of a broader story-world but that are never aimed at telling any other story outside the scope of the one they are telling – their ideological thrust is unobtrusive outside of a broader ‘antiquarian’ impulse. (2) In the neo-bardic mode (e.g., The Battle of Brunanburh and other ASC poems but also the OHG Ludwigslied), a repurposing of the formal qualities of the bardic mode in the particular cultural-political context of tenthcentury England produces texts whose ideological investment is clear – political propaganda for a growing sense of national identity. (3) The rag-picker mode (using Walter Benjamin’s figure of the ideal materialist historian as a collector-saviour of the refuse of the past who arranges its fragments in new constellations imbued with political power) constitutes the main focus of this book. This latter mode of historical representation exemplified by Beowulf but also the shorter Widsith and Deor, The Battle of Maldon, and the Latin Waltharius in Francia, is characterized by a diffractive structure, temporality, and narrative flow. Their ideological investment is ambivalent, at times subversive, at times hegemonic and fundamentally opened up towards the audience’s assessment. It aims to save the fragments of a ‘Germanic heroic’ tradition by assembling them into an ad hoc quilt of bits and pieces offered up as a metonym for this tradition which, to the eyes of someone like Alcuin, may well have appeared as a heap of rags, detritus – for in the words of a Carolingian cleric, even studying Virgil was like looking for gold in a pile of excrement – more on this in Chapter 4.144 This model is aimed to map only a portion of the broader spectrum of modes of historical representation available to early medieval people

34  Beyond Germanic Heroic Poetry in England and Francia. A closely related mode would be that of biblical heroic verse, which marries the orally derived traditional form of GHP (alliterative verse, its repertoire of particular formulae, figures of speech, and poetic words) and biblical subject matter. One could term it the Cædmonian mode by analogy with Bede’s well-known origin story of this form of discourse, which identifies Cædmon the shepherd as the first to perform this naturalization of the biblical story-world to the early medieval English cultural horizon and patterns of eventuality.145 Charles Wright has made a convincing argument for Genesis A belonging to the genre of universal history or world chronicle: the text is both GHP and deeply indebted to Latinate Christian histories, such as the Chronicle of Eusebius in Jerome’s Latin version.146 Similarly, Genesis B, Exodus, and Daniel in England, as well as the OS Heliand and the OHG Evangelienbuch in the Carolingian Empire, are based on the fundamentally historiographical impulse to emplot the biblical past through culturally powerful narrative patterns and stylistic forms that would make sense to their intended audiences, which cannot be separated in any meaningful way from their didactic and spiritual intentions(at least for these early medieval people). Thus these biblical heroic poems also belong to the historiographical spectrum described at the start of this chapter, which further extends to more canonical modes, such as Bede’s ecclesiastical historiographical mode, the ASC’s annalistic form (shared by Æthelweard’s more experimental Chronicon), or the panegyric mode of the later Encomium Emmae. Yet these modes are not the main concern of this book (although I do explore Frankish Cædmonian poems in Chapter 4), in part because they have been successfully investigated as history by previous research.147 What is more, neither are the first and second mode principal to my study, although I do discuss them, especially in the present chapter. I am primarily interested in exploring the rag-picker mode and the vernacular theories of narrative representation it embodies, the peculiar formal features it developed, and the way it functioned in different socio-cultural and emotional economies of meaning since I see it as probably one of the most misunderstood modes of pre-modern historical production, as well as a rich source of potential insights into the central questions of the philosophy of history. I will illustrate all the details of this brief account of my model in the following section, and over the next chapters I investigate the sources of the differences between these modes of historical representation: in Chapters 2 and 3, I focus on the vernacular vocabulary (and underpinning theories) of history, narrative, and truth, and in Chapters 4 and 5, I trace the ways in which these modes work in society. In the rest of this section, I provide reasons for choosing these specific terms by tracing their history and pointing to ways in which they conceivably echo some

Beyond Germanic Heroic Poetry  35 of the conceptualizations of the early medieval people involved in the production and reception of these texts. The bard and the rag-picker in this book’s title are meant to work on two levels: as a pair of heuristic figures meant to describe the relationship between the two modes of GHP and as symbolic of two types of relationships that scholarship has had with GHP. The bard points to the figure of the Germanic oral poet, usually in its incarnation as the scop, surrounded by the almost sacral aura of a bearer of the tradition handed down from the Germanic heroic age. This figure (aura included) originates in eighteenth-century Romantic Celticist (later adopted by Germanist) philological nationalism but still informs present-day scholarly discussions and popular perceptions, even though we know very little about the role and, indeed, the very existence of such individuals in early medieval societies.148 Roberta Frank, in particular, has traced an archaeology of this concept of oral poet, showing it to be an emotionally compelling myth rather than a historical reality recorded in GHP.149 Yet, as John Niles deftly shows, long before modern Romanticism, the figure of the oral poet already ‘loomed large in the imagination of the Anglo-Saxons themselves’, and precisely because representations of the scop were so culturally central to them, we need to take this myth seriously, ‘if not for [its] factual content then for what [it] contribute[s] to the history of mentalities’.150 And even if we cannot thereby reconstruct an accurate image of the actual early medieval practice of oral verse (however removed temporally and culturally from its representation in GHP), we can understand much about how early medieval English authors may have thought about their production of vernacular secular verse – namely, ‘as grounded in the art of ancient Germanic singers of tales’.151 The presence of figures of oral poets precisely in rag-picker verse histories, then, may play a significant role: as I suggest in the next section, a culturally significant image embodying the oral tradition these poems present themselves as faithfully continuing or a marker of the authenticity of what are ‘textual simulacr[a] of an oral poetics’.152 Hence, I use the word ‘bard’ precisely because of its cultural baggage, which I deploy to approximate in modern terms the powerful yet conceptually fuzzy aura that the figure of the scop must have held for these literate, cultured early medieval English people looking back with nostalgia to the oral verse tradition and the heroic society associated with it. At the same time, its Romantic, larger-than-life associations caution us to resist the equivalence these poems want us to make between a preliterate, pre-historical oral tradition and their use of the verse technique they employ for novel purposes. As Mark Amodio argued, although the people involved in the production of these poems were not ‘consciously attempting to recreate oral poetry but…they were using the only idiom at their disposal for the articulation of poetry’; this poetic tradition’s ‘role in

36  Beyond Germanic Heroic Poetry shaping how our extant Old English poetry means does not circumscribe what the poetry means or how poets can construct their poems’.153 This is why I take seriously the idea of oral poetry [that] served the people of pre-Conquest England as a cultural myth: that is, that the people of that time prized the image of the ‘bard of ancient days’ as a counterpart to the actual learned authors and scribes whose role in society was becoming ever more important.154 In this, I follow Niles’s evocative account of the important cultural and emotional work that this cultural schema was doing: it fulfilled an acutely felt need ‘for a soothing application of bardic balm’ in a ‘turbulent period of reform and reaction and, evidently, of “history wars” based on rival conceptions of the past’.155 This book is not concerned with the poets or authors of such texts (Emily Thornbury has recently published a comprehensive study on this topic, albeit focussed more heavily on the Anglo-Latin corpus) but with the historical visions involved in the production of secular vernacular verse and its social bearings. However, it is important here to provide an example of one way these people might have conceived the work they were doing. Niles suggests that, for instance, the end of Deor invites its educated and probably ecclesiastical audience to imagine itself in the presence of a larger-than-life figure of the heroic age, a ‘bold conceit’ underpinned by an ‘act of historical imagination…not as routine as it may seem’: ‘a poem that so deliberately invokes, without the intervention of a narrative frame, an ancient Germanic era of feuding heroes, master metal-smiths, passionate lovers, reckless tyrants, and clear-throated bards’: Even as you were sitting in the cathedral library reading this very manuscript that we call the Exeter Book, turning over its neatly written pages, you would be invited to imagine yourself listening to an ancient bard named Deor who was speaking to you directly, over a gap of six centuries or so, from his Continental homeland.156 In this, he is following Wilhelm Busse, who argued that the recording of ‘heroic and legendary poetry’ was a reaction to the stronger claims to the authority of the written word of the tenth-century Benedictine reform: ‘Under the fictitious names of Widsith and Deor’, members of the secular aristocracy with ties to the Church ‘offer themselves as connoisseurs of such tradition’, by which ‘they want to keep alive the social function of the scop as historicus, opposing the monastic claim to a better history’.157 I remain agnostic about our ability to precisely pinpoint the creation of GHP to such a specific circumstance since the emergence of ASC poems

Beyond Germanic Heroic Poetry  37 (the ‘neo-bardic’ mode) around this time shows a very different type of ‘heroic verse’ used for a very different function. In this evolution, GHP changes its focus away from the centrality of the ancient ‘heroic age’ and its women (in texts such as Beowulf and the Finnsburg Fragment) to contemporary events (in all ASC poems), also drawing back from its often mournful tone and its predilection for tragic stories to upbeat propaganda.158 Of course, these different modes of GHP may have arisen in tandem in different communities – yet it is hard to argue for or against either temporal positioning. However, Niles and Busse’s remain powerful descriptions of the sociocultural and emotional mechanics behind the emergence of this corpus, as they show the simultaneously marginal and culturally and emotionally powerful status of such poetry. As I suggest throughout this book, but particularly in the last two chapters, bardic and rag-picker verse histories arose against the mainstream of Orthodox opinion and thrived at least in certain communities not because they were of direct political usefulness (as the propagandistic neo-bardic mode proved to be) but because they fulfilled deeply felt (though more ideologically diffuse) cultural-emotional needs: the need for a common origin myth or for a simpler age less dominated by the institutionalization of hierarchies (as Niles suggests) but also the need for ludic spaces in which social panics, cultural anxieties, and personal hardships could be worked through, expressed, and questioned in a culturally significant type of discourse.159 This takes me to the second term of my model, the rag-picker. I suggest that while the creators of most GHP may have thought of themselves as bards, there were different dimensions to their craft. Some of them were satisfied to record brief tragic accounts from the distant heroic past in a shape that may have borne some relation to pre-existing oral performances circulating at some point (in the bardic mode). Meanwhile, other authors were more sophisticated poet-historians who had an acute awareness of both vernacular poetic tradition and canonical historiographical modes and its classical Latinate and biblical story-worlds and who wove together fragments from all of these to make new assemblages using a type of discourse in a process that may well have appeared to someone like Alcuin as stitching together rags. The rag-picker is a figure I repurpose from Walter Benjamin, a philosopher close to the hearts of many early medieval Anglicists for his ability to provide us with strikingly prescient ways of formulating ‘the heterogenous, the antiteleological, the fragmentary’ nature of much OE verse.160 In this, I am inspired by Trilling’s powerful use of other Benjaminian figures (most centrally the constellation) for reconceptualizing historical representation in OE verse. The rag-picker is an extended metaphor repurposed from Walter Benjamin’s allegory of the poetic method of Baudelaire, which he saw as also the ideal approach of the materialist historian. Historically, the rag-pickers (les chiffonniers) were those on the

38  Beyond Germanic Heroic Poetry bottom rung of nineteenth-century society, living in shanty towns on the edge of cities where they brought the old rags, shards, and other refuse to recycle.161 Benjamin draws inspiration from Baudelaire’s essay Du vin et du haschisch, which ‘likens the chiffonnier to an archivist or collector, attributing to him a creative rather than a destructive power’: ‘With the same devotion and attention to detail as the archivist, the ragpicker collects, organizes and catalogues the growing mountain of superfluous waste material produced by a society obsessed with the cult of the new’.162 But unlike an archivist, the rag-picker finds ‘his material and inspiration not in the officially sanctioned sites of a cultural text but in the refuse and debris that has been overlooked, repressed, or marginalized’.163 Benjamin sees this as the perfect reflection of the method of materialist historians, which he describes as a form of ‘literary montage’, ‘making use’ of ‘the rags, the refuse’, which is then assembled to gain previously inconceivable historical perspectives that hold subversive political and even messianic potential: ‘Through a strategic montage, in which the neglected debris of history is put into a new grammatical constellation, a true revolutionary image emerges’.164 Similarly, I suggest that the rag-picker mode of historical production is responsible for texts such as Beowulf and the Waltharius by self-consciously cobbling together fragments of several story-worlds, historical visions, and discursive practices to create assemblages that often coil back upon themselves in meditating on the nature of historical representation, narrative authority, and truthfulness. As Trilling argues about Widsith, the poem’s speaker becomes an allegory for an Anglo-Saxon culture that collects fragments from the myths and legends of the past, both to keep them from vanishing forever and to complete its own image of itself. This is the task of the Anglo-Saxon scop and Benjamin’s critic alike.165 I follow Trilling in making a broader argument that ‘[h]eroic history does not create a world out of whole cloth, but stitches together a patchwork of ideas, myths and relics of the past to create a coherent picture of historicity in the present’, albeit I see this historical vision at work only in some embodiments of ‘heroic history’ – in Beowulf but not in Brunanburh, in Deor but not in the Finnsburg Fragment.166 Thus I do not simply argue for the replacement of the dominant heuristic figure in GHP scholarship (that of the bard, which still shapes the reception of this corpus of texts) with a presumably better one (that of the rag-picker). Rather, the phrase in the title encapsulates the spectrum of poetical modes of expression available to early medieval English and Frankish verse historians stretching from a ‘bardic’, through the propaganda of the ‘neo-bardic’, to the ‘rag-picking’ mode, which arises at the confluence of biblical, classical, and Germanic poetic traditions and

Beyond Germanic Heroic Poetry  39 story-worlds through the sorting out and cobbling together by rag-picker poet-historians skilful in the traditional technique of alliterative verse and knowledgeable in its ‘heroic’ word-hoard. The following section demonstrates how this model works on the ground by tracing the workings of these different modes on several levels: the structural and formal one, the levels of the socio-cultural functions they fulfilled, and, finally, the level of reception and audience expectations. These multiple differences point to a recognition of these categories of texts as working in distinct modes of GHP, even by the people involved in their production and reception. Thus while the categories I use here are not meant to completely enclose these sources in a system of classification, I do suggest that they correspond to actual structures of thought and feeling of early medieval people, if not always to their linguistic representations thereof. As such, while I have used the term ‘genre’ loosely, more according to Tonkin’s socially situated paradigm than to the usual formally determined category of texts, an even better concept for these modes of historical representation might be ‘cultural attractors’ (an anthropological concept meant to describe the structure and dynamics of cultural change in relation to certain idea-vectors).167 My point is that these are not modern impositions on pre-modern realities but that ‘the mode of historical consciousness articulated in Benjamin’s work found expression in the art of an earlier age and can illuminate our understanding of its aesthetic function’.168 For instance, composing verse and carrying out a dispute or a conversation are often expressed by textile metaphors in OE sources, and as such, the idea of weaving together a text from several sources would have not appeared strange in this cultural horizon.169 Kathleen Davis has also argued that while much early medieval English textual production ‘insist[s] upon the relentless composition of an archive…the representations within that archive are not positioned as objects of mourning’ and are rather ‘gathered as a resource: an acquired hoard that when learned, recollected, and meditated upon becomes the generative experience yielding subject and text’.170 Davis showcases this underlying conceptualization with close discussions of Alfredian texts, such as the preface to the OE version of Augustine’s Soliloquies or Gregory the Great’s Pastoral Care, where the metaphor of the word-hoard is understood as ‘not simply a learned accumulation of stories and poetic formulae; rather, the structured archive of “useful” thought is the very composition of the mind that writes or recites poetry’.171 In these sources, Davis shows the construction of the archive to be the basis for a process of construction of the self and of a community, enabled by a ‘“strong” individual gather[ing] and secur[ing] material for an archive, or word-hoard’.172 These examples glimpse at a powerful deep-level conceptualization of mental representation and textual production in early medieval England

40  Beyond Germanic Heroic Poetry concurrent with that of the ‘bardic’ scop as a Benjaminian rag-picker and quilt-weaver: that of the word-hoard that needs to be employed to generate new assemblages out of the old materials for it to bear fruit. I suggest this is the type of cultural schema accessed in the rag-picker mode of historical representation. As Niles remarks about Widsith, such poetry is not a ‘neutral “poetic encyclopedia of early medieval Germania”’, or merely ‘mnemonic’ verse encoding a list of figures of the heroic age but ‘an arrangement of information into a sequence that embodies an underlying ideology’, which thus ‘serves the function of what Michel Foucault has called an archive, a storehouse where knowledge about the past is too potent an intellectual commodity to be neutrally or evenly distributed’.173 As I suggest in the last two chapters, this function can also be ideologically subversive or altogether more emotionally – than politically – inflected.

The Bard and the Rag-Picker at Work Having established the terms guiding my discussion of the modes of historical representation at work in GHP, I now turn to showcasing their productivity by interpreting primary sources with their help. I begin with the more complex and self-referential among them, the ‘rag-picker’ poems, moving as it were from the periphery of poetic and ideological experiment to the more canonical centre or at least to the more structurally conventional and ideologically straightforward discourse. In taking these conventions to the breaking point (or at least to a point where they are allowed to contend with each other and reflect upon themselves), the rag-picker mode provides more insights into these very conventions than those gained from exploring by themselves the deceptively ahistorical and formally conservative types of discourse the other two modes embody. As has been pointed out before, it is often parodies that ‘most clearly display the identifying features of a genre’ since ‘the humour of parody is…that it draws attention precisely to those features of style, content and form which the source of the parody presents as unexceptional’.174 Thus the parodic Waltharius and the ambivalent Beowulf may be much more useful for parsing the features of the supra-genre GHP than the most authentic putative oral ‘Germanic lay’ back-engineered from the textual record. The more self-consciously meta-historical GHP, such as Beowulf, Widsith, and Deor go beyond claims to veracity and explore the very process of making history by emplotment. These three texts are obsessed with the act of poetic composition and with the ways in which they represent the past. Significantly, all three feature scops as subjects of contemplation and as embodiments of a tradition that the creators of these poems put to quite different uses. Beowulf is particularly obsessed with the capability of verse to emplot the past in a meaningful and ‘true’ way, as well as with how conflicting accounts of the same events challenge (or

Beyond Germanic Heroic Poetry  41 at least pose in question) the possibility of historicization. I discuss these topics at length in the next two chapters. As Roy Liuzza reminds us, ‘Beowulf is not only about a particular history of certain Danes and Geats, but about History itself: the meaning of the past, the means by which it is recorded, and its remaining value to the present’.175 The poem is ‘a deeply conflicted reflection on the uses of history’ and ‘the inaccessibility of the past’.176 In a similar line of thought, Janet Thormann argued, ‘The insistent repetitions of episodes from the Frisian and Swedish wars; their fragmentary, segmented, often allusive structure; and the absence of chronological order in their delivery are symptoms of the narration’s effort to construct a history’.177 The unconventional shape of the emplotment in the poem is neither accidental nor the product of confusion but a purposeful construction placing in doubt the very possibility of telling a linear, unitary history of the Geats or of the greater panorama of tribal warfare: ‘Instead of the past “as it really was,” there were many “pasts,” all of them “real,” all of them relative’.178 In the apt words of Eileen Joy and Mary Ramsey, Beowulf does not make sense of the past so much as it calls the supposed coherence of the past (and of conventional narrative) into question. It also calls into question and unsettles how we (and the Anglo-Saxons as well) have traditionally defined and understood the relationships between past and present…and history and memory.179 As such, ‘in the absence of an evident ordering, the separate segments of a Beowulf’s and Wiglaf’s personal memories, together with the messenger’s retellings and anticipation, add up to a history of the Geats that must be constructed in the minds of the poem’s audiences through a purposeful connection and ordering of the separate segments of language’.180 This unconventionally complex (at least within a medieval context) vision may be the characteristic of a deeply unique text, which need not have been representative for early medieval England or even for GHP.181 Yet this propensity to pose in question the very making of history by fronting figures of poets and tellers of history seems to be a characteristic of at least several texts usually labelled as GHP. Indeed, although Beowulf has been recognized as ‘a monumental exercise of the historical imagination, poetically re-creating a past which is itself multi-layered and temporally complex’, its very centrality has cast a shadow on recognizing all GHP as exercises in historical production in their own right (albeit perhaps not quite as monumental).182 Even outside its historical dimension, Beowulf has been convincingly argued to be fundamentally experimental in its use of narrative strategies for which ‘there is no satisfactory model in antecedent Western literature’.183 Its non-linear narrative structure is ‘eccentric and unprecedented’,

42  Beyond Germanic Heroic Poetry its use of ‘multiple internal focalization’ anticipates modernist and postmodern novel techniques, and its deliberate presentation of conflicting accounts of the same events shows its poet to be preoccupied with epistemology and narrative representation.184 In Lapidge’s assessment, this is ‘the work of a highly literate and meditative poet and not the product of impromptu composition by a scop or the result of stitching together originally separate lays’.185 This is not to say that the poet did not work with already existing material, from narrative lines to a word-hoard of formulae and poetic terms – just that the use of such traditional material and of a formally conservative mode of discourse could very well lend itself to such innovative and uncanonical uses. At the same time, while Beowulf cannot be taken to be representative of GHP at large, it does showcase what the rag-picker mode could do when taken to its limits and/or put in the service of a vision that was not quite as ideologically conservative as its formal features (verse form, poetic vocabulary) and its obsession with the past might lead one to believe. And since Beowulf did not spring out of the void, we can see some of its novel or experimental features at work on a smaller scale in other GHP. Widsith’s kaleidoscope of world history and geography, while similar in shape with Old Norse mnemonic þula verse, is unique in its bringing together the Germanic, biblical, and classical past.186 At the same time, Deor’s self-consoling refrain entwines the scop’s misfortunes with episodes in the ‘Germanic’ past, not only making the latter illuminate the former but also showing ‘the hope of a recurring pattern in history’.187 While the Waltharius does not share their structural complexity and self-referentiality, it is nonetheless self-conscious in its ironic employment of the narrative tropes and characters of the ‘heroic tradition’, which it knows deeply – more on this in Chapter 4. So despite its being part of a different cultural horizon, and even if does not use the vernacular alliterative verse form (though there are Latin-Germanic puns), I categorize the Waltharius as participating in the rag-picker mode. Deor especially, as much akin to wisdom poetry as to GHP (the usual label of ‘elegy’ is mostly useful for shaping modern readers’ horizon of expectation), invites the audience ‘to evaluate past and present, transposing and mentally extending a general truth into an understanding of their own experience’.188 The riddling quality of such texts, accompanied by their weaving together of strands of material coming from very different traditions (Germanic, biblical, and classical) is echoed by early medieval English material artefacts like the Franks Casket carvings, which Leslie Webster qualifies as ‘contrapuntal’, going far beyond a simple compilation of historical material.189 In the Franks Casket, just as in Deor, Widsith, and Beowulf, the reader is invited to consider this very act of weaving together apparently incongruous materials and to pay attention to how different scenes, figures, and themes echo each other, clash or diffract against each other. The attention is always to past events, yet their

Beyond Germanic Heroic Poetry  43 meaning is not decided, being rather offered up for the audience to give meaning to by reference to their own experiences and needs. I shall focus on this mode of historical production in Chapter 3, especially on the particular truth procedure underpinning it, which works by pointing to the audience as the ultimate determiner of the truth and meaning of the (sometimes mutually contradictory or at least disjunct) narrative strands it weaves together. For now, I wish to point out that this appears to be another feature of the rag-picker mode of historicity – namely, leaving the reader/listener to piece together the meaning and ultimate truth of conflicting accounts (such as Unferth’s and Beowulf’s accounts of the swimming contest or the latter’s narration of his own adventures – which I explore in Chapter 3). With Beowulf, Lapidge argues that the reader ‘is induced to reflect on what has proceeded, so that the text becomes the object of progressive discovery, of a dynamic perception which is constantly changing’.190 This is also true in Deor and Widsith, albeit in less obtrusive ways. Temporally, Widsith works by threading together a kaleidoscope of different temporalities stretching from Old Testament times (the Hebrews and the Philistines), through classical antiquity (Caesar and Alexander the Great) on to a vast number of Germanic peoples and figures through the temporality of the poet’s own life, who has not only ‘heard of’ them but also ‘been with’ all of them. Widsith the scop interacts with the historical and legendary figures he meets: he receives rings and other treasures from Ealhhild, from Guthhere the Burgundian, which he gives to his lord Eadgils, and witnesses the deeds of a number of other heroes. Any educated early English audience would have been aware that it was impossible for these events to have spanned a lifetime. But the point is that all these cameos point to histories and temporalities beyond them, which haunt the poem even as it superficially looks like a collection of disparate elements. The key to interpretation lies with the audience. Joyce Hill warns us about assigning too much previous knowledge to all the poem’s audiences – some of them might have been as bewildered as us by the list of names and events or just as ignorant of the larger stories behind them, and as such, the sought effect may have been that of ‘a litany of Germanic poetry’.191 Hill’s reading depends on her identification of Widsith as a tenth-century composition, an argument which has been countered.192 My point is, however, not related to the dating of the poem: Hill may well be right that a tenth-century audience read the poem for the Tolkienesque frisson of a litany of half-comprehensible names, their very obscurity evoking a quasi-mythical past, but there is a space for seeing an early Widsith produced by and for a textual community knowledgeable in all these different pasts, for which the references would have pointed to wellknown and – understood histories, threaded together through the voice of the poet-historian Widsith.

44  Beyond Germanic Heroic Poetry Similarly to Beowulf, then, Widsith also invites its audiences to construct a history according to their different experiences and levels of knowledge of the narrative traditions (biblical, classical, Germanic). The production of historical discourse thus also takes place in the minds of the readers – the object of the poem ceases to be ‘Germanic legend’ but the simultaneous acts of reception and composition that each reader accomplishes when faced with these poems. Indeed, as I argue here and elsewhere, this is what the rag-picker mode of historical GHP is invested in: providing an open rhizome of narrative possibilities that are reconfigured according to the needs of its different audiences – poetic history is always in the making.193 Similarly, Deor brings a sadder and wiser perspective, standing as a sober answer to Widsith: heroes, victims, kings are listed, their stories threaded together, but here they stand in the service of a fragile consolation.194 The refrain þæs ofereode, þisses swa mæg (‘that passed away, so may this’) is less straightforwardly consolatory than it may seem, for the first five times þæs is repeated, it refers to the long-past misery felt by others, but the sixth time it refers to recent prosperity lost by the speaker – the most extreme example of what Tom Shippey sees as ‘the Old English liking for expressing violent semantic contrast by minimal phonetic change’.195 As such, the refrain creates a new temporality that does not merely transcend grief and happiness but that integrates them into one weave in which they stand together yet apart. Once again, the poet-historian’s voice threads together through his personal history Weland’s misery, Beadohild’s sorrows, Mæðhilde’s hopeless love, Theodric’s long rule, and Eormanric’s cruelty. Deor is often assumed to be about the mutability of fate,196 although the text itself does not offer any pious closure of the kind The Seafarer and other elegiac or especially gnomic poetry ends on. It is neither pessimistically fatalistic nor optimistically trusting in the mercy of God; instead, it is providing a historicized account of the contingency of the human experience and the ability to narrativize it as a means of gaining a sense of perspective, if not agency. The strength of Deor the scop lies in his ability to diffract his own story against several different threads of history through the vantage point of rag-picker verse.197 Yet the truth and ultimate meaning of the entire narrative construction of history whose hub is the refrain is opened up to the readers/listeners, who can repeat Deor’s operation and diffract their own experience through this kaleidoscope of history. Indeed, I will refer to Beowulf again and again throughout this book, not out of an assumption that it can be used as a key to explaining other GHP but rather because all these poems were part of related cultural systems of meaning. Specifically, I suggest Widisth or Deor are neither its precursors nor its epigonic descendants but parallel experiments in the same rag-picker mode of the historical imagination, functioning according to

Beyond Germanic Heroic Poetry  45 the same implicit theories and strategies of historical representations that are at work in Beowulf, where they are pushed to their limits and taken to breaking point, allowed to contend with each other or pondered on and offered up to the audience as judge of its truth and ultimate meaning. Yet this type of historical consciousness does not seem to underlie other GHP, like The Battle of Brunanburh, The Finnsburg Fragment, or the Hildebrandslied. I thus briefly turn to the bardic mode in order to show how differently it frames the distant ‘heroic’ past from which both it and the rag-picker mode take their emplotment. The OE Finnsburg Fragment (reprised in Beowulf at greater length in a self-standing episode presented as a scop’s recitation) and the OHG Hildebrandslied each present one tragic episode. The former tells of a conflict between Danes and Frisians complicated by kinship ties, set in the same, more distant, Scandinavian and continental past Beowulf is so invested in. The latter presents an ill-fated duel between father and son against the background of the legendary Theodoric’s campaigns. The OE Waldere fragments seem to point to a longer poem, although the surviving remnants are very similar to the blow-by-blow battle accounts accompanied by bouts of canny hero-speak the other two poems feature. There is nothing resembling the self-referentiality of Beowulf or the multi-level ironies of the Waltharius (which I explore in Chapter 4). These poems have long been thought to bear the closest resemblance to the oral ‘lays’ that formed the ever-elusive Germanic heroic oral tradition.198 Nor is this improbable – although ultimately impossible to prove. What is significant is that they, too, are interested in history, although usually only as the context enhancing the tragic personal histories unfolding in the foreground. Here the conceit of these texts being the work of actual oral poets comes closest to plausibility – the structure is linear, the narrative single-focussed and clear, and the ideological thrust is indeed aimed towards the past, although neither propagandistic nor necessarily nostalgic: the heroic past it depicts, albeit attractive for its aura of authenticity, seems to be one to which ‘one does not, of course, want to return’.199 Interestingly, in Beowulf, as Howe remarks, the extremely disjointed and closure-less narrative of the Swedish-Geatish feud ‘is harder to understand than the Finnsburg episode precisely because it is not offered as a set piece performed by the scop in the hall’ and because ‘it has yet to reach its conclusion’, for both the intra-textual hero and for the extra-textual audience.200 It is thus possible that the difference I am suggesting between the ‘rag-picker’ mode of historical representation (disjointed, fragmented, self-reflective, always pointing outside itself to a grander panorama that has to be pieced together by the audience) and the ‘bardic’ (relatively linear and self-contained, aimed at conveying either a unitary episode or a straightforward political message) is hinted at and recognized within Beowulf itself, which may point to an awareness of this distinction by at least some of its audiences.

46  Beyond Germanic Heroic Poetry These are all poems narrating different past events and are thus historiographical, albeit in a clearly different way when compared to the three texts discussed earlier. Yet they are similar in their comparatively linear structure and clear ideological investment, as well as their use of the same orally derived poetics and formal qualities as other GHP to narrativize a portion of the past. The figure of the scop is missing, but he lives in the poetic voice, as well as in the collective audience invoked by the ‘we have heard’ formulae. Clearly, they are not quite as interested in meditating on the craft of historiographical rag-picking and quilt-weaving, nor are they invested in invoking different historical traditions. Returning to the three rag-picker texts, another element they have in common needs to be explored: the prominent figures of imagined scops, oral poets, which they foreground as the central focus that holds together the strands of the past. Beowulf features several scenes of poetic recitation and composition (to be explored at length in the next two chapters). It is unlikely that these common elements are unrelated to each other: the figure of the scop, plucked from the quasi-mythical past that these poems purport to represent to the audience, is what produces this rag-picker mode of poetic history by bringing together strands of the past into a novel whole. In these poems, the oral verse tradition, embodied in the scop, becomes the vehicle for an innovative mode of history-writing that fulfils the specific needs of a later society while presenting itself as consubstantial with this tradition. These poems do not simply narrate action-packed episodes out of the ‘heroic past’ but bring together complex textures of specific bits of the past meant for consolation and meditation on the ever-changing fate (in Deor), for situating the English past within world history (in Widsith), or for a number of purposes including mourning the pre-Christian, pre-institutional past, understanding the rise and fall of peoples and heroes, meditating on violence, kinship ties, revenge, and the very meaning of history (in Beowulf).201 Hence the need for the presence of the scop as an embodiment of the ‘world of primary orality’ invoked in texts ‘that were surely meant for reading’: in replicating ‘the functions of oral literature in writing’, these texts ‘create links to an ancestral past that was inaccessible to the AngloSaxons except through storytelling’.202 But while all GHP invoke this world of oral tradition, not all of them are as interested in subjecting the very process of poetic composition (and thus, that of narrating history into existence) to the scrutiny and assessment of the audience – which is accomplished by foregrounding the scop figures. Other ‘Germanic heroic’ poems simply take on the voice of a scop in their use of the heightened mode of inscribed alliterative verse posing as oral giedd, with its poetic vocabulary and its repertoire of orally derived formulae and ‘heroic’ themes and figures, as if they were direct communications from oral poets unmediated by textuality. Unlike Deor,

Beyond Germanic Heroic Poetry  47 Widsith, and Beowulf (and, in different forms, The Battle of Maldon and the Waltharius), which delight in pondering on how these scops stitch history together, most GHP seeks to make invisible the main conceit on which they are based – that they themselves are presumably composed by oral poets and meant to function in the oral context of a tribal ‘heroic’ society based on interpersonal ties. Indeed, as Carol Braun Pasternack has argued, ‘the textuality of Old English poetry does not employ an idea of the author but rather an idea of tradition’, a fact proven by the very absence of both named authors and the first-person singular poetic voice that poses as the author.203 Throughout the corpus of OE GHP, ‘[f]ormulaic echoes and patterns that are frequently used to express an idea function as a code that readers can interpret as “tradition”’, whatever the particular uses of this tradition.204 Yet the poet is not absent from all of these texts – as we have seen, fictional scops are very much in the foreground of these three poems, presented as the makers of the poems we are reading or, in Beowulf’s case, as the creators of verse which the homodiegetical audience is listening to. In doubling down on this perceived lack, Pasternak points to a possible reason why this is so: the absence of the author-poets ‘characterizes this verse almost as much as the poet’s physical presence characterizes oral productions’, and so inscribed GHP is a tertium quid, something that is neither the uninscribed oral verse whose formal qualities we may assume it preserves, nor the later medieval written-authorial verse that fully assumes a fictional narrator implying ‘a particular subjectivity’, without necessarily being a missing link between them, a teleological buffer zone.205 And while many creators of GHP found this conceit (the voice of the oral poet not as an author but as a bearer of tradition) perfectly suitable for their needs, those involved in the production of the texts historicizing in the rag-picker mode (Beowulf, Widsith, Deor) were moved to cast this conceit into the limelight, drawing attention to its artificiality. Deor, Widsith, the oral poets at Heorot, are obviously not the recorded voices of real scops but are the creations of fundamentally literate textual productions, and their stitching together of variegated strands of discourse becomes a topic of meditation on the underpinning and limits of narrative (and thus historical) representation. This central (though covert) conceit of bardic GHP is harder to sustain in the only sub-corpus of GHP that can be pinpointed to specific political contexts without any doubts. The neo-bardic poems included in the ASC employ the voice of the scop, and thus of the oral tradition it embodies, to historicize particular events that are clearly not taking place in any tribal ‘heroic’ social setting. There, the narrative structure is linear and clearly situated in time, the ideology is transparent, and the message quite straightforward. The OE Battle of Brunanburh, the bestknown Chronicle poem, presents a recent victory in triumphal terms, praises King Athelstan, and presents a nationalistic narrative of history,

48  Beyond Germanic Heroic Poetry whose protagonist is the English nation led by the Wessex royal dynasty. In this, it is comparable to the OHG Ludwigslied, which presents the victory of the Frankish army, led by Louis III, over Danish raiders – albeit in a verse form that is not alliterative but which in its vernacularity (the text describes itself as a ‘Teutonic rhythm’) is meant to evoke the same world of heroic verse, whose formal shape had been lost, at least by this textual community’s cultural memory – more on this in Chapter 4. But, as Janet Thormann argues, while the ASC poems use the conceit of being part of the same genre as traditional heroic poetry to ‘celebrate, enhance, and justify power in the present by establishing it as a continuation of an ideal past, embodied in those very traditional formulas’,206 the trappings of the genre are sometimes out of place in an overtly politicised, propagandistic context, such as in the poem celebrating Edgar’s consecration in 973, where rhythms are mechanical and lifeless; the formulas are inappropriate, since priests don’t come in troops; the signal of oral transmission of legend conveyed by mine gefrege cannot refer to the contemporary witness of 973, who could not have heard of them even from afar. Their inappropriate use turns the metonyms of tradition into archaisms of a moribund sort, dead in the present they are supposed to exalt.207 Another neo-bardic poem, The Death of Edgar, has also been assessed as lacking in sophistication due to its ‘overaccumulation of…older poetic formulas with little regard for specificity of meaning’.208 Recent reappraisals of these poems argue that despite being often maligned by earlier scholars, the ASC poems may have been received just as (if not more) favourably by the audiences they were meant for as texts such as Beowulf that are often assumed to be representative of OE alliterative verse,– after all, the ASC verse corpus ‘conforms metrically to the phraseological rules of Old English poetic articulation’.209 Still, the point remains that the ASC poems, even though metrically not much different from Beowulf and other rag-picker poems, employ the cultural capital that this mode of verse evidently still possessed in tenthcentury England for ideological purposes that are perfectly transparent: the uplifting of the Wessex dynasty as the natural ruler of England by historicizing them in the mould of heroic figures from the more distant past on which ‘rag-picker’ poems focus, without all the ambivalence and selfreferentiality of the latter. And in some of these tenth-century creations, the political work takes precedence over maintaining the illusion of continuity with an oral tradition – the latter’s signifiers suffice as a vehicle for the more pressing ideological matter they convey. The historicity of the neo-bardic mode attempts (and succeeds only occasionally) to sustain the conceit of these events taking place in the heroic age that is implied by

Beyond Germanic Heroic Poetry  49 the traditionality of the verse, even while the events they emplot belong to a very different context, closer to the sophisticated political-religious machine of the Carolingian Empire than the mythical bands of brothers roving across primeval Germania. Meanwhile, even though Brunanburh performs very similar ideological work, Thormann sees it as aesthetically much more successful, ‘cast[ing] the present as an exuberant repetition of the past’, whereby ‘Athelstan embodies and exemplifies his genealogy by performing in traditional terms as a heroic warrior’.210 Martin Irvine sees the entire Parker manuscript of the Chronicle where the poems can be found as participating in ‘the ideology of an English nation’ and Brunanburh specifically as serving ‘the West Saxon dynasty by evoking a “heroic past” for its kings and representing Athelstan as the leader of a consolidated English force’.211 The straightforward propagandistic purposes of these poems are uncontroversial, but Trilling interprets their use of the verse tradition as an innovation in historical representation. Brunanburh ‘does not simply memorialize a king’s victory; it removes the battle from the public domain of current events’, as it would be as a simple Chronicle entry, and ‘translates it to the more rarefied sphere of History’, wherein ‘heroic values lose any sense of being anachronistic and Athelstan and Edmnund become larger than life’.212 Employing the cultural capital of the voice of heroic oral tradition to historicize recent events rather than to provide a narrative quilting of distant pasts filtered through the haze of nostalgia is indeed an innovative use of the historicizing force of GHP. Yet not all texts produced in this mould were created equal. Elizabeth Tyler has convincingly argued that The Battle of Maldon can be seen partly as a response to the triumphalist propagandistic thrust of the ASC poems: while the latter ‘used heroic poetry to legitimize West Saxon hegemony by presenting it as an extension of the traditional’, the former ‘critiques that use of poetry suggesting that longing for the imagined England of poetry was not the future’.213 While both share the nostalgia for the stability of traditional social structures implied by the ‘heroic’ verse tradition and use it to emplot recent events, Maldon subtly uses the conceit of the oral tradition to subvert the expectations of closure and stability associated with it: ‘heroic’ values are admirable but are not a successful defence against the Vikings, and the old world in which they made sense is forever gone. Also, Emily Thornbury has shown how subtly Maldon invites the assessment of the audience in terms similar to Beowulf: ‘the lack of closure – of assurance as to Byrhtnoth’s ultimate fate – returns the power of judgment back upon the audience’, which is thus invested ‘with a sense of immense power’. 214 As I discuss in Chapter 5, in both of them, the protagonist is simultaneously offered as a focus of projective idealization and of resentment for the negative consequences for their underlings of the heroic excess they embody. This complex ideological investment makes Maldon more akin to the polyvocal, occasionally subversive

50  Beyond Germanic Heroic Poetry rag-picker poems than to neo-bardic poems such as Brunanburh, despite formally being more like the latter. The reason for this cannot be provided by the formal qualities of these poems but by the socio-emotional and cultural work they are tasked with (since, as I have argued, genre is mostly determined by the social setting providing the occasion for the discourse). In this case, unlike the triumphalist historicization of the Brunanburh victory or the eulogistic mourning of King Edgar, Maldon contends with socio-political questions that are both broader and more complex. (Is it worth fighting the Danes rather than pacifying them via tribute? Is traditional heroism the solution?) Moreover, as I argue in an article in progress, Maldon is also a means for performing mourning and for collectively working through what must have been a traumatic event in a way that would be just as meaningful as a liturgical commemoration, although for a different audience and in a different setting. While liturgy is local, meant for the family of the deceased, it is not an appropriate ideological vehicle for an entire nation to deal with a traumatic event, anxiety, and mourning over political instability and the unravelling of the social order. Thus, Niles suggests, the purpose of texts like Maldon cannot be seen as narrowly ‘propagandistic, but rather psychological…helping to preserve a people’s sense of selfworth in the midst of very trying circumstances’ by ‘vicariously atoning’ for the ‘“sins” and shameful deeds that the English were attributing to themselves during this disastrous period of their history’.215 Or, as I suggest, Maldon could have provided the focus for members of its audience to perform collective mourning for both a beloved leader and an already unravelling socio-political order. All these needs are left unfulfilled by the other narrativizations of the Maldon event and of Byrhtnoth’s death (of which there are several, including an ASC entry and a hagiography, as well as post-Conquest Classicizing legendarized accounts, each aimed at different audiences).216 Hence the opportunity for using the vehicle of the orally derived alliterative heroic verse to emplot these traumatic events in a shape that would make cultural and emotional sense to the broader audience needing it. I suggest Beowulf does the same, although the issues it contends with are even broader since they are not localized to one event. To use James Earl’s argument about Beowulf, these are works of cultural mourning, understood as the process through which early medieval English society attempted – even if only as the initiative of a small textual community – to integrate its lost past (oral, pagan, kinship-based), to ‘internalize’ and ‘transform’ it in ‘healthy cultural growth’.217 In conclusion, unlike both neo-bardic verse histories and more canonical history-writing (Bede’s HE, the ASC), GHP in the rag-picker mode does not serve any overt political purposes since it makes no effort to ‘provide an authoritative version of a past cherished by the ruling class as a means of establishing a sense of origins and continuity with a distant

Beyond Germanic Heroic Poetry  51 past (possibly with some sort of legitimising function)’, although it can be used for such purposes in later appropriations.218 Rather, it opens up a space of play and reflection, where profound thinking about the meaning of history and the possibility of narrative representation can take place and where the anxieties and desires of the textual communities around them are expressed, exorcised, and taken to the breaking point.

Notes 1 For the date, see Bately (2003). 2 OE Orosius (1980: I.5, 23). 3 For discussions of the semantics and pragmatics of OE scop, see Thornbury (2014: 11–35) and Opland (1980: 230–56). 4 See, for instance, Godden (2011). The important exception is Opland (1980: 236 and 239), where he briefly makes a similar suggestion to the one pursued here. 5 Godden (2002). 6 Tyler (2006b: 226). 7 On the impossibility of drawing genre distinctions in the Old English corpus, see Niles (2016: 223). 8 Niles (2007: 87). 9 For this interpretation, see Hill (1994). 10 Howe (1997: 83). 11 Howe (1997: 84). 12 Howe (1997: 84–5). 13 Lockett (2011: 111, 151, 177). 14 Lockett (2011: 177). 15 Otter (2005: 113). 16 Lockett (2011: 111). 17 Lockett (2011: 151). 18 Idem. 19 Trilling (2009: 127). 20 Harris (2003: 31, 32). 21 Harris (2003: 32). 22 Koziol (2007: 93). 23 Wormald (2006: 51). 24 Alcuini Epistolae (1895: 124, p. 183). For the identification of the recipient as Unwona, see Bullough (1993). 25 Garizpanov (2008: 33). Councils and Ecclesiastical Documents (1879: 366). 26 Thornbury (2014: 200). 27 Elene (1932: 99). 28 Niles (2007: 108). 29 Shippey and Haarder (1998: 18, 31, 152) et passim and Frantzen (1990: 190–200). 30 Fulk and Cain (2003: 204). Tolkien (1936). 31 Cubitt (2006: 219). 32 Howe (1997: 87). 33 Howe (1997: 88). 34 Idem. 35 Frank (1982: 63). 36 Tennenhouse (1971: 145). See also Greenfield (1963). 37 Hanning (1974: 85–6).

52  Beyond Germanic Heroic Poetry 38 Niles (1998b: 224). 39 Trilling (2009: 15). 40 Scheil (2008: 282). 41 Ghosh (2016: 4). 42 Ghosh (2016: 9). 43 Liuzza (2005: 100–01). 44 For the intriguing argument that Beowulf was culturally a dead end, never managing to achieve the wide cultural resonance and currency of other long complex epics in other cultures, such as the Aeneid or the Epic of Ghilgamesh, see Davis (2007). For a compelling argument for the uniqueness of Beowulf’s narrative technique among all European pre-modern verse, foreshadowing post-modern strategies, see Lapidge (2001). 45 Tyler (2006b: 250). 46 Tyler (2006b: 242). 47 Tyler (2006b: 247). 48 Sharma (2014: 320). 49 Davis (2012: 216). 50 Davis (2012: 217). 51 Sharma (2014: 323). 52 Tyler (2006b: 226). 53 Niles (2007: 80). McNeill (1986). Niles (1998a: 7). 54 Niles (2007: 77). 55 Howe (1997: 89). 56 Niles (2007: 15). 57 Chadwick (1912); de Vries (1963). 58 On the OS influence on Genesis B and four other OE poems, see Bredehoft (2011); ‘Old. On Gothicism, see Frank (1991); Niles (2007: 13–115); Taranu (forthcoming). 59 Niles (2007: 25–7). 60 Edlund Anderson (1999: 93). 61 Harris (2003: 43). 62 For some examples, see Murdoch (1996); Murdoch (2004); Magennis (2010). 63 Trilling (2009: 3). 64 Stanley (1975: 91). 65 For a helpful selection of the highlights of the gigantic amount of scholarship abiding by the GHP paradigm, see Ghosh (2016: 7). For a thorough discussion of the issue and a new conceptual model, see Taranu (2015). On a ‘Germanic’ heroic age, see the classic Heusler (1905) and the later Andersson (1988). For critiques of the concept, see Frank (1991) and Goffart (1995a). 66 Ghosh (2016: 8). See his note 15 for another helpful trove of critiques of the concept of an early ‘Germanic culture’. See especially Goffart (1995b), Goffart (2006: 187–229), and Halsall (2007: 22–4, 118–31). 67 Frank (1991: 92). 68 Niles (2007: 79). 69 This is my argument in Taranu (2020). For ‘fuzzy concepts’, see Lakoff (1973). 70 Niles (1997: 211); Niles (2015: 199). 71 Goffart (1995b: 17; Taranu (2020: 102). 72 MacDougall (1982: 77–98. For American Anglo-Saxonism, see Frantzen and Niles (1997). 73 Harris (2003: 18). 74 Niles (1997: 208–09).

Beyond Germanic Heroic Poetry  53 75 Miyashiro (2019); Ellard (2019). 76 Karkov (2020: 21–2). 77 Reynolds (1985: 398). 78 Reynolds (1985: 414). 79 Harris (2003: 37). 80 Neidorf (2013: 174–5). For a compelling critique of Neidorf’s latter-day Germanicism, see Ghosh (2016: 252 ff). 81 I am grateful to Alaric Hall for many insightful conversations out of which emerged these often disregarded distinctions. 82 Tyler (2006b: 236). 83 Tyler (2006a: 1). 84 Harris (2003: 36). 85 Harris (2003: 25, 22). 86 Trilling (2009: 128). 87 Trilling (2009: 156). 88 Niles (2007: 228, 231). 89 Clark (2011: 475), Fanning (1997), and Toswell (1996). 90 Stanley (1997: 73). Ousby (1982: 157). Atkinson (2010: 47). 91 Nagy (1999: 17–20). 92 Bremmer (2005: 75). Green (1998: 67–83). 93 Lees (1994: 146). 94 Liuzza (2005: 106). 95 Ghosh (2016: 210). Orchard (2003a: 263). 96 Wolf (1991: 74–5). 97 Classen (2003: 296). 98 O’Connor (2016: 181). 99 Niles (2016: 223). See also Reichl (1992) and Niles (1999: 16–30). 100 Niles (2007: 105). 101 Idem. 102 Tyler (2006a: 4). See also O’Brien O’Keeffe (1990) and Pasternack (1995). 103 Tyler (2006a: 3). 104 Howe (1997: 94). 105 Scheil (2008: 282). 106 Deliyannis (2003). Kempshall (2011). 107 Taranu and O’Connor (forthcoming). 108 Ainsworth (2003: 403). See also Stein (2006); Partner (1977); Otter (1996). 109 Taranu and O’Connor (forthcoming: adapting Spiegel [1993: 63]). 110 Taranu and O’Connor (forthcoming). 111 For a more thorough discussion of the scholarship on this issue, see Taranu and O’Connor (forthcoming). I am deeply indebted to Ralph O’Connor for providing insightful reflections on the enormous amount of literature on these issues. 112 For the notion of ‘story-world’, and the conscious rejection of the mainstream cache of origin stories circulating in Continental Europe, mostly based on the Trojan myth and more generally the classical story-world, see Tyler (2013: 4, 6, 19). 113 Taranu and O’Connor (forthcoming). 114 Idem. 115 Liuzza (2000: 16). 116 Fulk and Cain (2003: 38). 117 Idem. 118 Sternberg (1984: 26). 119 Idem.

54  Beyond Germanic Heroic Poetry 20 Sternberg (1984: 26, 32). 1 121 Joy (2005) quoting Nancy (1990: 166). I am deeply indebted to Eileen Joy for enlightening conversations on these issues. 122 Jenkins (1995: p. 19). White (1978: p. 82, 1973: 2). 123 Davis (2007: 32). 124 Jenkins (1995: 25). 125 Jenkins (1995: 20). 126 Idem. 127 Idem. 128 For a survey of (mostly negative) reactions to Keith Jenkins, Hayden White’s most representative continuator, as well as potential responses, see Macfie (n.d.). 129 Jenkins (1995: 20) . White (1973: xi–xii). 130 Jenkins (1995: 147). 131 White (1980: p. 10). 132 Niles (2007: 108). 133 Scheil (2008: 285): Grendel is called rinc (man, Beowulf 720b), healdegn (hall-thane, 142a), wonsceli wer (unhappy man, 105a), gromheort guma (hostile-hearted man, 1682a), and feasceaft guma (wretched man, 973a); he walks on weres wæstmum (in the shape of a man, 1352a. Grendel’s mother is a wifunhyre (monstrous woman, 2120b) bearing idese onlicnces (the shape of a woman, 1351a). 134 Pearsall (1977: 8–9). 135 Howe (1997: 93, 95). Scheil (2008: 302). Pace Orchard (2003b). Sisam (1953: 96). 136 Sharma (2014: 322). 137 For ‘horizon of expectation’, as well as the classical account of medieval genres, see Jauss (1982). 138 Øverby (2009: 33). Poppe (2007). 139 Tonkin (1990: 8). 140 Tonkin (1990: 8). 141 Tonkin (1990: 105). 142 Tonkin (1990: 51, 50). 143 Tonkin (1990: 3), adapting Barber (1989: 15). 144 Ermanrich (1899: 563). 145 Bede (1969: 414–21). 146 I would like to thank Manish Sharma for pointing this out, as well as for the thoughtful comments on my conceptual model. Wright (2012). 147 For biblical epic, see Trilling (2009: 64–124). For Encomium Emmae, see Tyler (2005). For Æthelweard, see Jezierski (2005). For ASC, see Bredehoft (2001). 148 For a more detailed history of the term ‘bard’ and its associations, see Niles (2007: 195–7). 149 Frank (1993). 150 Niles (2007: 148). 151 Niles (2007: 150). 152 Liuzza (2005: 106). 153 Amodio (2004: 30). 154 Niles (2016: 125). 155 Niles (2007: 152). 156 Niles (2007: 174). 157 Busse (1987: 36), quoted in Niles (2007: 174). 158 Tyler (2006b: 235).

Beyond Germanic Heroic Poetry  55 59 Cf. Trilling (2009: 124). 1 160 Trilling (2009: 3). For previous uses of Benjamin by early medievalists, see Trilling (2009: 31). 161 Le Roy (2017: 129). 162 Le Roy (2017: 130). Benjamin (2002: 349–350). 163 Richter (2006: 233). 164 Benjamin (2002: 460). Richter (2006: 233). 165 Trilling (2009: 61). 166 Trilling (2009: 4). 167 Buskell (2017). 168 Trilling (2009: 32). 169 Clegg Hyer (1998). 170 Davis (2012: 231). 171 Davis (2012: 220). 172 Davis (2012: 231). 173 Niles (2007: 78). Foucault (1972: 128–31). 174 Davenport (2004: 27), quoting Butterfield (1990: 186). 175 Liuzza (2005: 105). 176 Liuzza (2005: 106). 177 Thormann (2006: 302). 178 Frantzen (1990: 129). 179 Joy and Ramsey (2006: xxxiv). 180 Thormann (2006: 302). 181 Trilling (2009: 219). 182 Liuzza (2000: p. 16). 183 Lapidge (2001: 76). See also Lapidge (1993). 184 Lapidge (2001: 88). 185 Idem. 186 Harris (1991: 241). 187 Kopár (2010: 205). 188 Webster (1999: 228). 189 Idem. 190 Lapidge (2001: 67). 191 Hill (1994). 192 Neidorf (2013). 193 Taranu (2015). 194 For a powerful discussion of Deor, see Trilling (2009: 42–9). 195 Shippey (2010: 263). 196 Fell (1991). 197 I am grateful to Catherine Karkov for introducing me to the possibilities opened by diffraction theory. For an introduction to diffraction theory, which explores the effects that Heisenberg’s principle of uncertainty and Bohr’s principle of indeterminacy have for epistemology and ontology, as well as to how it helps illuminate literary discourses of this type, see Barad (2014). 198 For more on previous readings of Finnsburg, see Niles (2016: 141). 199 Niles (2007: 63). 200 Howe (1989: 165). 201 For comments on the figure of the oral poets in these three poems, see Harris (1991: 241). 202 Niles (2007: 108). 203 Pasternack (2006: 531). 204 Pasternack (2006: 532).

56  Beyond Germanic Heroic Poetry 05 Pasternack (2006: 528). 2 206 Thormann (2010: 215). 207 Thormann (2010: 216). 208 Greenfield and Calder (1986: 248). 209 Amodio (2004: 54). 210 Thormann (2010: 216). See also Scragg (2003). 211 Irvine (1994: 451–3). 212 Trilling (2009: 199, 196). 213 Tyler (2006b: 249). 214 Thornbury (2014: 116). 215 Niles (2007: 242, 246). 216 For a comprehensive range of studies discussing the various accounts of the event, see the essays in Scragg (1991). 217 Earl (1994: 132). 218 Ghosh (2016: 219).

Bibliography Primary Sources Alcuini Epistolae 124, in MGH Epistolae, vol. 4, Epistolae Karolini Aevi, vol. 2, ed. by Ernst Dümmler (Berlin: Weidmann, 1895). Bede, Historia Ecclesiastica gentis anglorum, ed. and trans. by Bertram Colgrave and R. A. B. Mynors (Oxford: Clarendon Press, 1969). Beowulf = Klaeber’s Beowulf and The Fight at Finnsburg, ed. by Robert D. Fulk, Robert H. Bjork and John D. Niles, fourth edition (Toronto: University of Toronto Press, 2008). Councils and Ecclesiastical Documents Relating to Great Britain and Ireland, ed. by Arthur W. Haddan and William Stubbs, 3 vols (Oxford, 1869–79), III (1879). Elene, in Krapp, George Philip, ed., Vercelli Book, the Anglo-Saxon Poetic Records (New York: Columbia University Press, 1932), pp. 67–102, pp. 132–52. Ermanrich, Epistola ad Grimaldum abbatem, in MGH Epistolae, vol. 5, Epistolae Karolini Aevi, vol. 3, ed. by Ernst Dümmler (Berlin: Weidmann, 1899), pp. 534–79. The Old English Orosius, ed. by Janet Bately, EETS Supplementary Series, 6 (London, Oxford University Press, 1980).

Secondary Sources Ainsworth, Peter, ‘Legendary History: Historia and Fabula’ in Historiography in the Middle Ages, ed. Deborah M. Deliyannis (Leiden: Brill, 2003), pp. 387–416. Amodio, Mark, Writing the Oral Tradition: Oral Poetics and Literate Culture in Medieval England (Notre Dame, IN: University of Notre Dame Press, 2004). Andersson, Theodore M., ‘Walter Haug’s Heldensagenmodell’, in Germania: Comparative Studies in the Old Germanic Languages and Literatures, ed. by Daniel G. Calder and T. Craig Christy (Woodbridge: D.S. Brewer, 1988), pp. 127–41. Atkinson, Juliette, Victorian Biography Reconsidered: A Study of NineteenthCentury ‘Hidden’ Lives (Oxford: Oxford University Press, 2010).

Beyond Germanic Heroic Poetry  57 Barad, Karen, ‘Diffracting Diffraction: Cutting Together-Apart’, Parallax, 20 (2014), 168–87. Barber, Karin, ‘Interpreting Oríkì as History and as Literature’, in Discourse and Its Disguises: The Interpretation of African Oral Texts, ed. by Karin Barber and Paulo de Moraes Farias (Birmingham: Centre of West African Studies, 1989), pp. 13–23. Bately, Janet, ‘The Alfredian Canon Revisited: One Hundred Years On’, in Alfred the Great: Papers from the Eleventh-Centenary Conferences, ed. by Tim Reuter (Aldershot: Ashgate, 2003), pp. 107–20. Benjamin, Walter, The Arcades Project, trans. by Howard Eiland and Kevin McLaughlin (London: Belknap Press, 2002). Bredehoft, Thomas A., Textual Histories: Readings in the Anglo-Saxon Chronicle (Toronto: University of Toronto Press, 2001). Bredehoft, Thomas A., ‘Old Saxon Influence on Old English Verse: Four New Cases’, in Anglo-Saxon England and the Continent, ed. by Hans Sauer and Joanna Story (Tempe: ACMRS Press, 2011), pp. 83–111. Bremmer, Rolf, ‘Old English Heroic Literature’, in Readings in Medieval Texts: Interpreting Old and Middle English Literature, ed. by Elaine Treharne and David Johnson (Oxford: Oxford University Press, 2005), pp. 75–90. Bullough, Donald, ‘What Has Ingeld to Do with Lindisfarne?’, Anglo-Saxon England, 22 (1993), 93–125. Buskell, Andrew, ‘What Are Cultural Attractors?’, Biology and Philosophy, 32 (2017), 377–94. Busse, Wilhelm G., ‘Boceras: Written and Oral Traditions in the Late Tenth Century’, in Mündlichkeit und Schriftlichkeit im Englischen Mittelalter, ed. by Willi Erzgräber and Sabine Volk (Tübingen: Gunter Narr, 1987), pp. 27–37. Butterfield, Ardis, ‘Medieval Genres and Modern Genre Theory’, Paragraph, 13 (1990), 184–201. Chadwick, Hector Munro, The Heroic Age (Cambridge: Cambridge University Press, 1912). Clark, David, ‘Notes on the Medieval Ideal of Dying with One’s Lord’, Notes and Queries, 58 (2011), 475–84. Classen, Albrecht, ‘The Downfall of a Hero: Siegfried’s Self-Destruction and the End of Heroism in the “Nibelungenlied”’, German Studies Review, 26 (2003), 295–314. Clegg Hyer, Maren, ‘Textiles and Textile Imagery in Old English Literature’ (unpublished doctoral thesis, University of Toronto, 1998) . Cubitt, Catherine, ‘Folklore and Historiography: Oral Stories and the Writing of Anglo-Saxon History’, in Narrative and History in the Early Medieval West, ed. by Elizabeth M. Tyler and Ross Balzaretti (Turnhout: Brepols, 2006), pp. 189–221. Davenport, Tony, Medieval Narrative: An Introduction (Oxford: Oxford University Press, 2004). Davis, Craig R., ‘Theories of History in Traditional Plot’, in Myth in Early Northwest Europe, ed. by Stephen Glosecki (Tempe: ACMRS and Brepols, 2007), pp. 31–45. Davis, Kathleen, ‘Time’, in A Handbook of Anglo-Saxon Studies, ed. by Jacqueline Stodnick and Renée Trilling (Oxford: Blackwell, 2012), pp. 215–34.

58  Beyond Germanic Heroic Poetry de Vries, Jan, Heroic Song and Heroic Legend, trans. by B.J. Timmer (Oxford: Oxford University Press, 1963). Deliyannis, Deborah M., ‘Introduction’, in Historiography in the Middle Ages, ed. by Deborah M. Deliyannis (Leiden: Brill, 2003), pp. 1–13. Earl, James W., Thinking About Beowulf (Stanford: Stanford University Press, 1994). Edlund Anderson, Carl, Formation and Resolution of Ideological Contrast in the Early History of Scandinavia (unpublished doctoral thesis, University of Cambridge, 1999). Ellard, Donna Beth, Anglo-Saxon(ist) Pasts, postSaxon Futures (New York: Punctum Books, 2019). Fanning, Steven, ‘Tacitus, Beowulf, and the Comitatus’, Haskins Society Journal, 9 (1997), 17–38. Fell, Christine, ‘Perceptions of Transience’, in The Cambridge Companion to Old English Literature, ed. by Malcolm Godden and Michael Lapidge (Cambridge: Cambridge University Press, 1991), pp. 172–89. Foucault, Michel, The Archaeology of Knowledge, trans. by A. M. Sheridan Smith (New York: Routledge, 2002 [1969]). Frank, Roberta, ‘The Beowulf Poet’s Sense of History’, in The Wisdom of Poetry: Essays in Early English Literature in Honor of Morton W. Bloomfield, ed. by Larry D. Benson and Siegfried Wenzel (Kalamazoo, MI: Medieval Institute, Western Michigan University Press, 1982), pp. 53–65. Frank, Roberta, ‘Germanic Legend in Old English Literature’, in The Cambridge Companion to Old English Literature, ed. by Malcolm Godden and Michael Lapidge (Cambridge: Cambridge University Press, 1991), pp. 88–106 Frank, Roberta, ‘The Search for the Anglo-Saxon Oral Poet’, Bulletin of the John Rylands Library, 75 (1993), 11–36. Frantzen, Allen, Desire for origins: New Language, Old English, and Teaching the Tradition (New Brunswick: Rutgers University Press, 1990). Frantzen, Allen J., and John D. Niles, eds, Anglo-Saxonism and the Construction of Social Identity (Gainesville: University Press of Florida, 1997). Fulk, Richard D., and Christopher Cain, A History of Old English Literature (Oxford: Blackwell, 2003). Garizpanov, Ildar H., The Symbolic Language of Authority in the Carolingian World: c. 751–877 (Leiden: Brill, 2008). Ghosh, Shami, Writing the Barbarian Past: Studies in Early Medieval Historical Narrative (Leiden: Brill, 2016). Godden, Malcom R., ‘The Anglo-Saxons and the Goths: Rewriting the Sack of Rome’, Anglo-Saxon England, 31 (2002), 47–68. Godden, Malcom R., ‘The Old English Orosius and Its Sources’, Anglia, 129 (2011), 297–320. Goffart, Walter, ‘Conspicuous by Absence: Heroism in the Early Frankish Era (6th–7th Cent.)’, in La Funzione dell’eroe germanico: Storicità, metafora, paradigma, ed. by Teresa Pàroli (Rome: Il Calamo, 1995a), pp. 41–56. Goffart, Walter, ‘Two Notes on Germanic Antiquity Today’, Traditio, 50 (1995b), 9–30. Goffart, Walter, Barbarian Tides: The Migration Age and the Later Roman Empire (Philadelphia: University of Pennsylvania Press, 2006). Green, Dennis, Language and History in the Early Germanic World (Cambridge: Cambridge University Press, 1998).

Beyond Germanic Heroic Poetry  59 Greenfield, Stanley B., ‘Geatish History: Poetic Art and Epic Quality in Beowulf’, Neophilologus, 47 (1963), 211–17. Greenfield, Stanley and Daniel Calder, eds., A New Critical History of Old English Literature (New York: New York University Press, 1986). Halsall, Guy, Barbarian Migrations and the Roman West, 376–568 (Cambridge: Cambridge University Press, 2007). Hanning, Robert W., ‘Beowulf as Heroic History’, Medievalia et Humanistica, 5 (1974), 77–102. Harris, Joseph, ‘Beowulf in Literary History’ in Interpretations of Beowulf: A Critical Anthology, ed. by Robert D. Fulk (Bloomington: Indiana University Press, 1991), pp. 235–41. Harris, Stephen, Race and Ethnicity in Anglo-Saxon Literature (London: Routledge, 2003). Heusler, Andreas, Lied und Epos in germanischer Sagendichtung (Dortmund: Ruhfus, 1905). Hill, Joyce, ‘Widsið and the Tenth Century’, in Old English Shorter Poems: Basic Readings, ed. by Katherine O’Brien O’Keefe (New York: Garland, 1994 [1984]), pp. 319–33. Howe, Nicholas, Migration and Mythmaking in Anglo-Saxon England (New Haven: Yale University Press, 1989). Howe, Nicholas, ‘Historicist Approaches’, in Reading Old English Texts, ed. by Katherine O’Brien O’Keefe (Cambridge: Cambridge University Press, 1997), pp. 79–100. Irvine, Martin, The Making of Textual Culture: ‘Grammatica’ and Literary Theory: 350–1100 (Cambridge: Cambridge University Press, 1994) Jauss, Hans Robert, ‘Theory of Genres and Medieval Literature’, in Towards an Aesthetic of Reception, trans. by Timothy Bahti (Brighton: Harvester, 1982 [1970]). Jenkins, Keith, On ‘What Is History?’ (London: Routledge, 1995). Jezierski, Wojtek, ‘Æthelweardus redivivus’, Early Medieval Europe, 13 (2005), 159–78. Joy, Eileen A., ‘James W. Earl’s Thinking About Beowulf: Ten Years Later’, in The Heroic Age, 8 (2005) [accessed 28 January 2020]. Joy, Eileen A., and Mary K. Ramsey, ‘Introduction: Liquid Beowulf’, in The Postmodern Beowulf: A Critical Casebook, ed. by Eileen A. Joy and Mary K. Ramsey (Morgantown: West Virginia University Press, 2006), pp. xxix–lxvii. Karkov, Catherine E., Anna Kłosowska, and Vincent W.J. van Gerven Oei, ‘Introduction: Disturbance’, in Disturbing Times Medieval Pasts, Reimagined Futures, ed. by Karkov et al. (New York: Punctum Books, 2020), pp. 21–22. Kempshall, Matthew S., Rhetoric and the Writing of History: 400–1500 (Manchester: Manchester University Press, 2011). Kopár, Lilla, ‘Spatial Understanding of Time in Early Germanic Cultures: The Evidence of Old English Time Words and Norse Mythology’ in Interfaces between Language and Culture in Medieval England: A Festschrift for Matti Kilpiö, ed. by Alaric Hall et al. (Leiden: Brill, 2010), pp. 203–30. Koziol, Geoffrey, ‘Truth and Its Consequences: Why Carolingianists Don’t Speak of Myth’, in Myth in Early Northwest Europe, ed. by Stephen Glosecki (Tempe: ACMRS and Brepols, 2007), pp. 71–103. Lakoff, George, ‘Hedges: A Study in Meaning Criteria and the Logic of Fuzzy Concepts’, Journal of Philosophical Logic, 2 (1973), 458–508.

60  Beyond Germanic Heroic Poetry Lapidge, Michael, ‘Beowulf and the Psychology of Terror’, in Heroic Poetry in the Anglo-Saxon Period: Studies in Honor of Jess B. Bessinger, Jr., ed. by Helen Damico and John Leyerle (Kalamazoo: Medieval Institute, 1993), 373–402. Lapidge, Michael, ‘Beowulf and Perception’, PBA, 111 (2001), 61–97. Lees, Clare, ‘Men and Beowulf’, in Medieval Masculinities: Regarding Men in the Middle Ages, ed. by Clare Lees (Minneapolis: University of Minnesota Press, 1994), pp. 129–48. Le Roy, Frederik, ‘Ragpickers and Leftover Performances: Walter Benjamin’s Philosophy of the Historical Leftover, Performance Research, 22 (2017), 127–34. Liuzza, Roy, ‘Introduction’, in Beowulf: A New Verse Translation, trans. by Roy Liuzza (Ontario: Broadview Press, 2000). Liuzza, Roy, ‘Beowulf: Monuments, Memory, History’, in Readings in Medieval Texts: Interpreting Old and Middle English Literature, ed. by Elaine Treharne and David Johnson (Oxford: Oxford University Press, 2005), pp. 90–108. Lockett, Leslie, Anglo-Saxon Psychologies in the Vernacular and Latin Traditions (Toronto: University of Toronto Press, 2011). MacDougall, Hugh A., Racial Myth in English History: Trojans, Teutons, and Anglo-Saxons (Hanover: University Press of New England, 1982). Macfie, Alexander, ‘Keith Jenkins Retrospective’, in Reviews in History (n.d.) [accessed 4 May 2020]. Magennis, Hugh, ‘Germanic Legend and Old English Heroic Poetry’, in A Companion to Medieval Poetry, ed. by Corinne Saunders (Oxford: WileyBlackwell, 2010), pp. 85–100. McNeill, William H., Mythistory and Other Essays (Chicago: University of Chicago Press, 1986). Miyashiro, Adam, ‘Our Deeper Past: Race, Settler Colonialism, and Medieval Heritage Politics’, Literature Compass, 16 (2019), 1–11. Murdoch, Brian, The Germanic Hero: Politics and Pragmatism in Early Medieval Poetry (Cambridge: Cambridge University Press, 1996). Murdoch, Brian, ed., German Literature in the Middle Ages (Woodbridge: Camden House, 2004). Nagy, Gregory, The Best of the Achaeans: Concepts of the Hero in Archaic Greek Poetry (Baltimore: Johns Hopkins University Press, 1999). Nancy, Jean-Luc, ‘Finite History’, in The States of Theory, ed. by David Carroll (New York: Columbia University Press, 1990) . Neidorf, Leonard, ‘The Dating of Widsið and the Study of Germanic Antiquity’, Neophilologus, 97 (2013), 165–83. Niles, John D., ‘Appropriations: A Concept of Culture’ in Anglo-Saxonism and the Construction of Social Identity, ed. by Allen J. Frantzen and John D. Niles (Gainesville: University Press of Florida, 1997), pp. 202–28. Niles, John D., ‘Introduction’, in A Beowulf Handbook, ed. by Robert E. Bjork and John D. Niles (Lincoln, Nebraska: University of Nebraska Press, 1998a), pp. 1–12. Niles, John D., ‘Myth and History’, in A Beowulf Handbook, ed. by Robert E. Bjork and John D. Niles (Lincoln: University of Nebraska Press, 1998b), pp. 213–33. Niles, John D., Homo Narrans: The Poetics and Anthropology of Oral Literature (Philadelphia: University of Pennsylvania Press, 1999).

Beyond Germanic Heroic Poetry  61 Niles, John D., Old English Heroic Poems and the Social Life of Texts (Turnhout: Brepols, 2007), p. 87. Niles, John D., ‘The Myth of the Feud in Anglo-Saxon England’, JEGP, 114 (2015), 163–200. Niles, John D., Old English Literature: A Guide to Criticism with Selected Readings (Malden, MA: Wiley-Blackwell, 2016). O’Brien O’Keeffe, Katherine, Visible Song: Transitional Literacy in Old English Verse (Cambridge: Cambridge University Press, 1990). O’Connor, Ralph, ‘Monsters of the Tribe: Berserk Fury, Shapeshifting and Social Dysfunction in Táin Bó Cúailnge, Egils saga and Hrólfs saga kraka’, in Kings and Warriors in Early North-West Europe ed. by Jan Erik Rekdaland Charles Doherty (Dublin: Four Courts Press, 2016), pp. 180–236. Opland, Jeff, Anglo-Saxon Oral Poetry: A Study of the Traditions (New Haven: Yale University Press, 1980). Orchard, Andy, Critical Companion to ‘Beowulf’ (Cambridge: Boydell & Brewer, 2003a). Orchard, Andy, Pride and Prodigies: Studies in the Monsters of the Beowulf Manuscript (Toronto: University of Toronto Press, 2003b) Otter, Monika, Inventiones: Fiction and Referentiality in Twelfth-Century English Historical Writing (Chapel Hill: University of North Carolina Press, 1996). Otter, Monika, ‘Functions of Fiction in Historical Writing’, in Writing Medieval History, ed. by Nancy Partner (London: Hodder Arnold, 2005), pp. 109–30. Ousby, Ian ‘Carlyle, Thackeray, and Victorian Heroism’, The Yearbook of English Studies, 12 (1982), 152–68. Øverby, Thomas, Breuddwyd Rhonabwy: A Historical Narrative? (unpublished MA thesis, University of Oslo, 2009). Partner, Nancy, Serious Entertainments: The Writing of History in TwelfthCentury England (Chicago: University of Chicago Press, 1977). Pasternack, Carol Braun, The Textuality of Old English Poetry (Cambridge: Cambridge University Press, 1995). Pasternack, Carol Braun, ‘The Textuality of Old English Poetry’, in The Postmodern Beowulf: A Critical Casebook, ed. by Eileen A. Joy and Mary K. Ramsey (Morgantown: West Virginia University Press, 2006), pp. 519–46. Pearsall, Derek, Old English and Middle English Poetry (London: Routledge, 1977). Poppe, Erich, ‘Literature as History / History as Literature: A View from Medieval Ireland’, in Literature as History / History as Literature, ed. by S. Fielitz (Frankfurt am Main: Peter Lang, 2007), pp. 13–24. Reichl, Karl, ‘Old English giedd, Middle English yedding as Genre Terms’, in Words, Texts and Manuscripts: Studies in Anglo-Saxon Culture Presented to Helmut Gneuss on the Occasion of His Sixty-Fifth Birthday, ed. by Michael Korhammer, Karl Reichl, and Hans Sauer (Woodbridge: D.S. Brewer, 1992), pp. 349–70. Reynolds, Susan, ‘What Do We Mean by “Anglo-Saxon” and “Anglo-Saxons”?’, Journal of British Studies, 24 (1985), 395–414. Richter, Gerhard, ‘Acts of Self-portraiture: Benjamin’s Confessional and Literary Writings’, in The Cambridge Companion to Walter Benjamin, ed. by David Ferris (Cambridge: Cambridge University Press, 2006), pp. 221–23.

62  Beyond Germanic Heroic Poetry Scheil, Andrew, ‘The Historiographic Dimensions of Beowulf’, JEGP, 107 (2008), 281–302. Scragg, Donald, The Battle of Maldon AD 991 (Oxford: Blackwell, 1991). Scragg, Donald, ed., ‘A Reading of Brunanburh’, in Unlocking the Wordhord: Anglo-Saxon Studies in Memory of Edward B. Irving, Jr, ed. by Mark C. Amodio and Katherine O’Brien O’Keeffe (Toronto: University of Toronto Press, 2003), pp. 109–22. Sharma, Manish, ‘Beyond Nostalgia: Formula and Novelty in Old English Literature’, Exemplaria, 26 (2014), 303–27. Shippey, Tom, ‘“The Fall of King Hæthcyn”: Or, Mimesis 4a, the Chapter Auerbach Never Wrote’, in On the Aesthetics of Beowulf and Other Old English Poems, ed. by John M. Hill (Toronto: University of Toronto Press, 2010), pp. 247–65. Shippey, Tom A., and Andreas Haarder, Beowulf: The Critical Heritage (London: Routledge, 1998). Sisam, Kenneth, Studies in the History of Old English Literature (Oxford: Clarendon Press, 1953). Spiegel, Gabrielle, Romancing the Past: The Rise of Vernacular Prose Historiography in Thirteenth-century France (Berkeley: University of California Press, 1993). Stanley, Eric G., The Search for Anglo-Saxon Paganism (Cambridge: Cambridge University Press, 1975). Stanley, Eric G., ‘The Early Middle Ages = The Dark Ages = The Heroic Age of England and in English’, in The Middle Ages After the Middle Ages in the English Speaking World, ed. by Marie-Françoise Alamichel and Derek Brewer (Cambridge: D. S. Brewer, 1997), pp. 43–77. Stein, Robert, Reality Fictions: Romance, History, and Governmental Authority, 1025–1180 (Notre Dame: University of Notre Dame Press, 2006). Sternberg, Meir, The Poetics of Biblical Narrative: Ideological Literature and the Drama of Reading (Bloomington: Indiana University Press, 1984). Taranu, Catalin, ‘Goths, Geatas, Gaut: The Invention of an Anglo-Saxon Tradition’, in Transforming the Early Medieval World: Studies in Honour of Ian N. Wood, ed. by Kivilcim Yavuz and Ricky Broome (Leeds: Kismet Press, forthcoming). Taranu, Catalin, ‘Who Was the Original Dragon-Slayer of the Nibelung Cycle?’, Viator, 46 (2015), 23–40. Taranu, Catalin, ‘The Balloon that Wouldn’t Burst: A Genealogy of “Germanic”’, in Interrogating the Germanic, ed. by James Harland and Matthias Friedrich (Berlin: De Gruyter, 2020), pp. 99–123. Taranu, Catalin and Ralph O’Connor, ‘Introduction’, in Vera Lex Historiae?: Constructions of Truth in Medieval Historical Narrative, ed. by Catalin Taranu and Michael Kelly (New York: Punctum & Gracchi Books, forthcoming). Tennenhouse, Leonard, ‘Beowulf and the Sense of History’, Bucknell Review, 19 (1971), 137–46. Thormann, Janet, ‘Enjoyment of Violence and Desire for History in Beowulf’, in The Postmodern Beowulf: A Critical Casebook, ed. by Eileen A. Joy and Mary K. Ramsey (Morgantown: West Virginia University Press, 2006), pp. 287–318. Thormann, Janet, ‘The Subject of Language: A Psychoanalytic Approach to the Aesthetic of Old English Poetry’, in On the Aesthetics of Beowulf and Other

Beyond Germanic Heroic Poetry  63 Old English Poems, ed. by John M. Hill (Toronto: Toronto University Press, 2010), pp. 209–26. Thornbury, Emily, Becoming a Poet in Anglo-Saxon England (Cambridge: Cambridge University Press, 2014). Tolkien, J.R.R., ‘Beowulf: The Monsters and the Critics’, PBA, 22 (1936), 245–95. Tonkin, Elizabeth, Narrating Our Pasts: The Social Construction of Oral History (Cambridge: Cambridge University Press, 1990). Toswell, M. J., ‘Tacitus, Old English Heroic Poetry, and Ethnographic Preconceptions,’ in Studies in Old English Language and Literature: ‘Doubt Wisely’. Papers in Honour of E.G. Stanley, ed. by M. J. Toswell and E. M. Tyler (London: Routledge, 1996), pp. 493–507. Trilling, Renée, The Aesthetics of Nostalgia: Historical Representation in Old English Verse (Toronto: Toronto University Press, 2009). Tyler, Elizabeth, ‘Fictions of Family: The Encomium Emmae Reginae and Virgil’s Aeneid’, Viator, 36 (2005), 149–79. Tyler, Elizabeth, Old English Poetics: The Aesthetics of the Familiar in AngloSaxon England (Woodbridge: Boydell & Brewer, 2006a) Tyler, Elizabeth, ‘Poetics and the Past: Making History with Old English Poetry’, in Narrative and History in the Early Medieval West, ed. by Elizabeth M. Tyler and Ross Balzaretti (Turnhout: Brepols, 2006b), pp. 225–50. Tyler, Elizabeth, ‘Trojans in Anglo-Saxon England: Precedent Without Descent’, The Review of English Studies, 64 (2013), 1–20. Webster, Leslie, ‘The Iconographic Programme of the Franks Casket’, in Northumbria’s Golden Age, ed. by Jane Hawkes and Susan Mills (Stroud: Sutton, 1999), pp. 227–46. White, Hayden, Metahistory: The Historical Imagination in Nineteenth-Century Europe (Baltimore: The Johns Hopkins University Press, 1973). White, Hayden, Tropics of Discourse: Essays in Cultural Criticism (Baltimore: The Johns Hopkins University Press, 1978). White, Hayden, ‘The Value of Narrativity in the Representation of Reality’, Critical Inquiry, 7 (1980), 5–27. Wolf, Alois, ‘Medieval Heroic Traditions and Their Transitions from Orality to Literacy’, in Vox intexta: Orality and Textuality in the Middle Ages, ed. by A.N. Doane and Carole Braun Pasternack (Madison, WI: University of Wisconsin Press, 1991), pp. 67–88. Wormald, Patrick, ‘Bede, Beowulf and the Anglo-Saxon Aristocracy’, in Times of Bede: Studies in Early English Christian Society and its Historian, ed. by Stephen Baxter (Oxford: Blackwell, 2006 [1978]), pp. 30–105. Wright, Charles D., ‘Genesis A ad litteram’, in Old English Literature and the Old Testament, ed. by Manish Sharma and Michael Fox (Toronto: University of Toronto Press, 2012), pp. 121–71.

2 What We Talk About When We Talk About History The Old English Vocabulary of Narrative and Historical Representation Early medieval people had a wide spectrum of sophisticated senses of history interacting in intricate ways throughout a broad range of discourses with historiographical intents and affinities. The previous chapter argued for this position, building on previous work in an attempt to understand the exact nature of this sense of the past. Such an understanding has so far remained elusive and contradictory, based on insightful comments on individual texts or on particular passages from them. There has been no attempt to provide a unified account for the historical consciousness at work in vernacular verse discourses such as GHP or biblical epic  – with the one significant exception of Renée Trilling’s The Aesthetics of Nostalgia. Yet Trilling’s book and many other briefer studies dedicated to the issue – as I argued in the preceding chapter – are mostly focused on the stylistic aspects of the texts discussed, drawing conclusions abstracted from textual features on the verse and text level. There has yet been no attempt at rethinking traditional categories of narrative and historiographic discourse from within medieval textual productions by focusing on the words and phrases the early medieval people involved in the production and reception of GHP and other historiographical modes would have used for these concepts. This chapter aims to provide a conceptual tool kit for this undertaking, opening up the possibility to trace vernacular theories of historical and narrative production and representation underlying the vocabulary of history and narrative used by the early medieval English. The following chapter will do the same with conceptualizations and theories of truth and veracity. Thus this chapter and the next one form a diptych whose panels cast new light on each other, and I invite the reader to treat them as such. In this endeavour, I am inspired by Nicholas Howe’s mostly overlooked suggestion that since students of early medieval England lack the thick documentary resources available for later periods of English (or, indeed, Frankish) history, they ‘must turn to the language itself and perform a kind of linguistic ethnography or archaeology, a reading of the culture through its words and its grammar’ – precisely my approach, which aims to be a ‘historicist criticism of Old English texts’ that ‘grounds itself in

What We Talk About When We Talk About History  65 a sense of the language and yet also ventures beyond the self-imposed proscriptions of traditional philology’.1 The introductory chapter traced the limitations of previous approaches to the issue of historicity of GHP. We have already seen how even the more sophisticated studies tend to cast the underlying historical consciousness of much vernacular textual production as fundamentally nostalgic, and in doing so, they oppose it to the sense of history at work in canonical historiography, such as Bede’s Historia Ecclesiastica or the Anglo-Saxon Chronicle.2 However, if one thinks about the terms of this dichotomy as possible pathways in a wider rhizome of conceptualizations of history, narrative, and veracity, the disjunction is surpassed, but the need emerges for a comprehensive, integrative study of this whole rhizome.3 Such an all-encompassing project would involve looking at the entire corpus of OE, OS, Old High German, and Latin writing with a historiographical claim, as well as the Old Irish, Middle Welsh, and Old Norse corpora, and the scope could always be enlarged in such an attempt to catalogue all the pathways of the rhizome, all the conceptualizations of history almost never explicitly theorized but always discreetly implicit in the very shape and intertextual echoes of these texts. What I propose in this chapter instead is a way to access this variety of vernacular theories of historical and narrative representation by means of examining all words that would roughly correspond to the notion of ‘history’ (itself a quasi-insurmountable issue, as we shall see next) spread throughout the OE corpus and charting their meanings. After all, individual words are the nodes of meaning which together construct and convey ‘a sense of the past’ – this sense is not a Platonic essence, floating beyond its textual embodiments and accessible only to the initiated. Thus by mapping the semantic fields of ‘history’ and ‘narrative’ (and ‘truth’ in Chapter 3) in OE, we may be able to get closer to understanding the ways in which early medieval people thought about writing/performing history and conceptualized the past and the possibilities of historical representation and narrative production.4 I restrict myself to the English space for reasons that will become apparent in Chapter 4: there is simply much less material directly relevant to these concepts in the continental vernacular verse corpus, while the Latin Waltharius is much less preoccupied with these concepts than its English rag-picker mode counterparts, as we have previously seen. However, many of the conclusions I reach in these two chapters are illuminating for a deeper understanding of these issues in continental GHP, too. This chapter, then, is a lexical and semantic study of the OE vocabulary of history. Here, I sketch a model of cultural conceptualizations of history which has the advantage of encompassing all the areas of the aforesaid spectrum of modes of historicity, integrating the different ways of conceptualizing history present in (and across) the early medieval cultural horizons which have so far mainly been seen in disjunction.

66  What We Talk About When We Talk About History What words did the early medieval English use for the concept of ‘history’? In a nutshell, they both translated Latin historiographical notions and adapted native words, transferring them from day-to-day speech to the sphere of scholarship. These two strategies produced a wide array of complementary yet quite different history-words covering a variety of shades of meaning, all of which are invaluable in uncovering early medieval English ideas about history. Hence, I will first examine prose attestations and then compare them with evidence of a different nature found in verse in order to ascertain the mental deep structures underlying the different ways of encoding the notion of ‘history’ (be they words, phrases, or figures of speech – in the former case, or associative patterns – ­emotions and social roles and situations – in the latter). One of the main obstacles in investigating the semantics of OE terms becomes apparent even when doing a simple search in the Thesaurus of Old English (TOE).5 The form in which TOE lists its words is what could be called ‘notional trees’, semantic hierarchies going from 18 very general categories (‘mental faculties’, ‘emotion’, ‘life and death’, etc.) to subdivisions of ever narrower semantic spheres. Several such notional trees point to ‘history-words’: 02. Life and death 02. Creation 02.03. Humankind 02.03.02. Family/household 02.03.02.03. Ancestry, descent framcynn, woruldgebyrd, cynnreccenes, folctalu, mæggewrit, mægracu 06. Mental faculties 06. Spirit, soul, heart 06.01. The head (as seat of thought) 06.01.04. Faculty of memory gemynd, fyrngemynd, efengemynd, læran Also stemming from ‘06.01. The head’: 06.01.07. Truth, conformity with absolute standard 06.01.07.03. Truth of speech or thought, veracity soþcwed 09. Language and communication 09. Speech, vocal utterance 09.03. A language 09.03.07. Writing 09.03.07.07. A book 09.03.07.07.03. Composition, arrangement, writing

What We Talk About When We Talk About History  67 09.03.07.07.03.04. Chronicle, annals, history cranic, woruldgewritu, gewyrdelic The results of the search all belong to the wider semantic field of ‘history’, although each word instantiates slightly different aspects of the central concept of ‘history’ by translating different mental images into different verbal forms. For instance, heroic verse such as Beowulf often refers to the ancestry, descent, or origin (framcynn, cynnreccenes, etc.) of people but also objects – what is meant by this is a personal history; hence, according to the Thesaurus classification, it is listed as type ‘02.03.02.03. Ancestry, descent’ under the ‘02. Life and Death’ general category. In the same type of cultural horizon, collective, oral, memorial forms of history are at other times instantiated as ‘old memories’, ‘old lore’ (all the gemynd composites), hence ‘06.01.04. Faculty of memory’ under ‘06. Mental Faculties’), and the truthfulness of these memorial narratives is emphasized – soþcwed (‘06.01.07.03. Truth of speech or thought, veracity’, also under ‘06. Mental faculties’). In the cultural horizon of Bede, Alfred, and Latinate culture, history often takes the form of chronicles and annals (‘09.03.07.07.03.04. Chronicle, annals, history’ under ‘09. Language and communication’). All these OE words codified different shades of meaning of history for the speakers of the language, yet they can be translated into Modern English only by means of the single word ‘history’. It is true that we also have account, recollection, memory, narrative, anecdote, treatise, etc., for different forms of historical discourse, yet we still cannot capture the semantic subtleties of the OE terms in translation simply because the two series of semantically related words belong to two different cultural horizons and thus there is no simple one-to-one correspondence between Old and Modern English lexemes. The modern words pertaining to the semantic spheres of history and, more generally, narrative date back to the Norman-French wave of lexical innovation and therefore do not have OE etymons. Although both Modern English history and OE istoria/stær share the classical Latin etymon historia, the former does not directly descend from the latter, being subsequently reborrowed from Anglo-Norman and Old French istorie~estoire.6 Like ‘story’, the other more specialized terms for historical narrative (‘chronicle/chronique’, ‘res gestae’, ‘annals’, ‘archive’, even ‘memory’) are borrowings from Norman French or Latin. Undoubtedly, the Thesaurus is an invaluable tool for any student of OE. Yet was this the way early medieval English categorized their concepts?7 It is true that the Thesaurus does not claim its notional trees are faithful images of early medieval English hierarchies of meaning – they are simply useful ways of organizing data for the use of modern scholars. And yet what might genuine early medieval English notional trees have looked like? Or, indeed, under what other forms did they mentally organize such

68  What We Talk About When We Talk About History conceptualizations? Beyond the multiple history-words encountered in early medieval English sources, there lie several dominant cultural conceptualizations of the past and its recording in writing or oral traditions which also shed light on the larger issue of accessing medieval mentalities, which are quite different from the way we think about history today. But what do words indicate about the mental patterns and implicit theories about the world underlying them? Can one even make such a distinction between a verbal vehicle and its corresponding notional ‘contents’? I have found the vocabulary and methodologies of cognitive linguistics useful to glean mentalities and worldviews from discourse and language by using the texts that survive as windows to the past to reach the deep structures of mentality.8 Cognitive linguistics fulfils a need felt by many anthropologists of present or past societies in their attempt to describe the mechanics of how such imponderable notions as ‘mentalities’ or mental models of the world emerge in communities of people (‘cultures’). Thus, Jan Vansina argues that it is difficult for outsiders to a culture to grasp its worldview since within the community this worldview is obvious, being known and understood by everyone – at the same time, nowhere does a worldview exist as ‘a single system worked out in a noncontradictory way which accounts for everything in relation to everything else’.9 Meanwhile, ‘favored and gifted informants can create a much more systematic and broader worldview over time than existed hitherto’, which Vansina deplores as being ‘their own idiosyncratic culture and no longer the collective representation of the community’.10 The work on cultural conceptualizations I use in the following has the virtue of bypassing these dilemmas since it sees such a ‘worldview’ as composed of a number of cultural schemas, which, however, need neither be shared by all the members of a culture nor do all of its elements need to be recognized as equally essential. Rather, cultural schemas are distributed and emergent, being the result of a myriad interpersonal negotiations, which accounts for both their stability and their ever-changing nature. Yet despite the recent wave of popularity cognitive approaches enjoy in OE studies (see Note 8) and in other fields in the humanities and sciences not directly involved in the still ongoing debates within the philosophy of mind, it is worth remembering that these approaches are no more value-free, ‘empirical’, or ‘scientific’ than, say, psychoanalytical theories or Deleuze and Guattari’s dazzling (and potentially mystifying) concepts, which I use elsewhere to describe the ecology of narratives growing around the ‘Germanic heroes’ Siegfried and Siegmund.11 While cognitivism (broadly understood as the computational model of the mind whereby human behaviour and mental states are reducible to physical or neurophysiological processes) has now achieved broad recognition as the explanatory framework for how the mind works, it is still only one in a variety of philosophies of cognition.12 Cognitive linguistics sees language not so much as a means of communication but rather as an instrument for organizing and processing

What We Talk About When We Talk About History  69 knowledge which directly reflects the nature and structure of thoughts and mental patterns.13 A recent contribution to the field, Farzad Sharifian’s Cultural Conceptualizations and Language, takes this approach one step further to decode the conceptualizations of entire cultures which are encoded in the very fabric of its language.14 Sharifian posits that the lexicon of a language is ‘the most direct link with cultural conceptualizations in the sense that lexical items largely act as labels, and hence “memory banks”, for conceptualizations that are culturally constructed’.15 While linguists access the ‘collective memory banks’ of living communities (in Sharifian’s case, Persian speakers and Australian Aborigines) by way of interviews and anthropological and ethnographic approaches, the only way to access early medieval English conceptualizations is via the texts they have produced that are available to us. For ‘cultural conceptualizations may be instantiated and reflected in cultural artefacts such as painting, rituals, language, and even in silence’ and, more importantly for us, ‘in stories’, but the privileged cultural artefact in the present case is, of course, discourse as ‘a vehicle for the representation of cultural conceptualizations’.16 I centred my methodology on the notion of ‘conceptualization’ to refer to modes of mental representation which are also elements of mentality, preferring it to ‘concept’, as the former better emphasizes the dynamic nature of such cognitive phenomena. Conceptualizations are not ‘things’ in the mind but ‘patterns which emerge from knowledge which is represented in a distributed fashion across the network [of people sharing a certain culture]’. 17 Cultural conceptualizations are the products of human collective cognition, which means they are developed through ‘interactions between the members of a cultural group and enable them to think as if in one mind, somehow more or less in a similar fashion’.18 This does not exclude the fact that specific conceptualizations may belong to only one or several individuals (as unique, original thoughts), but the important thing is that at least some cultural conceptualizations (which can be understood to be the building blocks that make a culture) can be recovered from the linguistic artefacts produced by that culture. These conceptualizations are negotiated and renegotiated through time, across generations. In this case, ‘cultural groups are formed not just by the physical proximity of individuals but also by the relative participation of individuals in each other’s conceptual world’.19 Cultural conceptualizations are not only collective but also distributed differently across the minds constituting a cultural group, meaning that they are not homogenously accepted and adhered to by all the members of a cultural group.20 Conceptualizations usually develop into complex, dynamic systems of knowledge, which are not totally and equally shared by the members of the target cultural group. What makes them representative of a specific culture is that over time, ‘such dynamic systems may

70  What We Talk About When We Talk About History act as major anchor points for people’s thought and behaviour and may even constitute a worldview’ (i.e., ‘a group-level cognitive system’).21 Yet ‘the minds that constitute the cultural network do not equally share all the elements of the schema, nor does each mind contain all the elements of the schema’.22 This pattern of collective and distributed representation ‘clearly accounts for ‘fuzzy’ understandings that characterize the reality of our communications. In reality, people draw more or less on a schema; they do not necessarily share all the components associated with a given schema’.23 These realities will be amply demonstrated by early medieval conceptualizations of history instantiated in words used in different contexts. Unsurprisingly, most of the individual history-words I analyse come mainly from prose translations and glosses. This is not very surprising since they are usually used to translate Latin concepts belonging to the semantic field of ‘history’, usually historia or derivatives thereof. All the extant writing in OE was penned by Christian clerics who had been educated in Latin, whose education, cultural horizon, and intentions need not have been representative of society at large.24 Since in this type of cultural milieu, the activity of history-writing was at least declaratively rooted in the late antique Christian tradition of historiography, the early medieval English glossators and translators had to invent new words for this foreign conceptualization of history. For this purpose, they borrowed Latin words (historia > OE stær, chronica > OE cranic), they coined novel compounds (gerecednys, spellcwide, etc.), or they simply used extant words in new, more specialized ways (spell and sægen, words already having a wide semantic sphere which could be extended to accommodate new senses). However, the apparently skewed distribution of OE history-related words favouring those belonging to the world of early medieval English Latinate learning against those found in verse is not the result of a lack of a sense of history in oral vernacular mentalities but is the natural consequence of a difference in conceptualizing history. Thus the translation and glossing of Latin works and words took place in the context of early medieval English cultural milieus that are trying to adopt classically inspired notions of history into an early medieval English context. Rooted in the tradition of Isidore and Orosius (and hence, Augustine) but also drawing on pre-Christian Greek and Roman authors, this way of doing historiography is naturally based on Aristotelian and Ciceronian taxonomies, as we have seen in the previous section. Hence, the terms pertaining to these cultural endeavours and communities are more likely to be based on categorical (taxonomical) conceptualizations – which include concepts that enter into an ‘x is a kind of y’ association (‘a chronicle is a type of history’). On the other hand, history-related words feature abundantly in more traditional vernacular forms of writing, such as ‘heroic’ verse, but they belong to schematic types of conceptualization. As such,

What We Talk About When We Talk About History  71 they are more likely to be associated with specific situations (oral recitation by scops, declaring one’s lineage to a friend or a foe are forms of history-making and commemoration) or emotions (nostalgia, the dream in the mead-hall) rather than being clearly set in taxonomic relationships to each other. Due to these differences in the way the conceptualizations of history are embedded in the text, I use somewhat different methodologies (although both circumscribed by cognitive linguistics) for each of these modes of thought. The first part of this section traces the etymology and history of the use of specific words encountered in prose texts and glosses to Latin sources by comparing their semantic sphere to that of Latin terms or other OE words. The second part of the section examines entire passages (rather than individual words) drawn from verse texts, deriving insights from the part they play in the economy of the text (mainly Beowulf) as a whole. The following section tracks the vocabulary of history found in prose texts and glosses, using those of Bosworth-Toller’s definitions, which are borne out by the context in which the particular words are found.25 Of the two Latin loans, the first occurs very frequently in early medieval English writing. The preferred Latinate word (with 23 occurrences throughout the corpus) for a historiographic account is stær, used by translators into OE of Bede, Orosius, and the glossator of Aldhelm to translate the Latin historia. As is the case with historia, stær refers not to historical events but to their narration, and it is used accordingly. The translator of Bede’s Historia Ecclesiastica calls his history ‘þis ure stær’ (‘this history of ours’), or ‘cyriclecan stær (‘church history’). His use of the word conceivably shows its usual semantic value in learned milieux.26 Its two derivatives, denoting the professionals of the discipline, are stærleornere, ‘a historical scholar’, and stærwritere, ‘a writer of history, a historian’ (‘Swa soþsagol stærwritere ða þing ðe be him oþþe ðurh hine gewordene wæron ic awrat’ [like a truthful historian I wrote about what was done by him or through him]).27 Interestingly, stær is an indirect loan from Latin, part of the small number of loans coming from Goidelic Celtic associated with Christianity, thus it was apparently borrowed from Irish missionaries (OE dry ‘magician’ < Old Irish drui, ancor ‘hermit’ < OI anchara, cros (which only appears in place names – the usual OE is rōd) and, finally, stær < OI stoir).28 The historical context could have been the influx of Irish learning starting with Aidan’s mission of 635 to Northumbria at the behest of King Oswald. The very fact of borrowing a term for ‘history’ is a result of the conceptual foreignness previously discussed, and the Irish cultural world happened to be the first source for this specific conception of history (rooted in the classical tradition) with which early medieval English cultural elites became acquainted. Later on, when the opening towards Latin learning via Rome superseded the ties with Ireland, the term historia was borrowed again, this time directly, as istoria. The greatest concentration of

72  What We Talk About When We Talk About History stær can be found in the OE translation of Bede’s Historia Ecclesiastica and the glosses on Aldhelm’s prose De laudibus virginitatis (but also twice in the OE Orosius in the compound stærwritere).29 However, (h) istoria is a favourite of Ælfric’s – his homilies and Grammar account for more than half of all the occurrences of the word.30 Conceivably, this difference between the early Alfredian and the late Ælfrician West Saxon use of historia-derivatives is due to the early cultural influence of Irish learning on the former and of later cultural ties to Rome on the latter. The other Latin loan was a more specialized term for historical narrative, cranic, ‘chronicle’ (‘swa Hieronimus sæde on his cranice’ [as Jerome said in his chronicle]),31 with its derivative, cranicwritera, ‘chroniclers’ (glossing ‘chronographorum’).32 This word seems to be especially popular in late OE homilies.33 Turning now to the many native OE compounds, the main difficulty in correctly identifying history-words lies in separating those which mean simply ‘narrative in general’ and those more specialized to denote specifically ‘historical narrative accounts’. In fact, as shall be seen, this separation would have appeared as somewhat meaningless to early medieval English thought: history and narrative (even in the sense of fiction, fabulation) are so greatly overlapping semantic spheres as to be almost indistinguishable. In this, they are remarkably close to recent understandings of history, whereby the only thing historians can do is tell stories about the past within a fictional matrix.34 There seem to be, however, slightly more specialized terms that would correspond to a relative differentiation of history in the narrative spectrum: ealdspræc, ‘an old speech, history’;35 ealdwritere, ‘a writer on ancient history’, and ealdspell, ‘old story, history’ (‘Ælfred us ealdspell reahte’ [Alfred told us an old story]).36 The conceptualization underlying this series of terms built around eald seems to be that history is the narration of things past, of old, ancient events. For a different perspective, there is endebyrdnes, meaning primarily ‘a row, series, rank’ but also ‘succession in place or time’ and, hence, ‘with reference to narrative or statement in which circumstances are stated in proper order’ (‘He him sæde his sið be ændebyrdnysse’ [he recounted him his journey in order]),37 a regular narrative, a series of statements (‘We habbað nu miccle maran endebyrdnysse þære Cristes bec gesæd þonne ðis dægðerlice godspel behæfð’ [we have mentioned many more stories than are contained in the gospel for the day]).38 For a similar understanding of history there is getæl, ‘a number, series, reckoning, computation’ (‘Seo Abbudisse het hine [Cædmon] læran þæt getæl þæs halgan stæres and spelles’, translating the Latin original ‘Abbatissa jussit illum [Cædmonem] seriem sacræ historiæ doceri’ [the abbess (Hild) commanded (them) to teach him (Cædmon) the series of the holy story and narrative]).39 The latter term is, of course, derived from talu, ‘a tale, talk, story, account’ (‘Þa spræcon hi betwux him…and seo modor sæt hlystende hire tale’ [then they spoke between them and the mother sat

What We Talk About When We Talk About History  73 listening to their talk/account]).40 In both getæl and endebyrdnes, the underlying conceptualization is that of order, series of events narrated in the proper order. In this, it is akin to the Old Norse telja, ‘to (re)count, reckon, enumerate’, cognate with tal, ‘conversation’, but also ‘list’, which preserves both ideas of ‘number’ and ‘story’.41 As can already be seen, the semantic field of OE history-words cannot be easily separated from that of generic ‘narrative’. There is a small but diverse lexical family centred on reccing or rece(d)ness~rece(d)nyss, ‘a story, narrative, history’ (glossing ‘historiae’ and ‘ecclesiasticae liber’):42 cynnrecceniss, ‘a reckoning of relationship, a genealogy’ (in one manuscript, it glosses the Latin incipit of the Matthew Gospel ‘de generatione Iesu Christi’) – more on genealogy as one of the dominant early medieval English conceptualizations of history further down43 – also gerecednes (glossing expositionibus, narrationibus, relatione), ‘a narration, history, report’.44 Here, history is conceptualized as a reckoning, the unfolding of a tale in due order (but also the enumeration of generations in a genealogy). The senses of reccan extend to expounding a meaning, unravelling a riddle, and correcting; hence, history as a narration, a succession of events, thus potentially a good story but also a narrative with a meaning to be discerned, a moral and perhaps even an eschatological meaning. An atypical gloss for historia used several times is gewyrd. The Bosworth-Toller definition for gewyrd is ‘event, fate, destiny, condition’, but in Aldhelm’s De laudibus virginitatis, it is used to gloss a variety of partially overlapping meanings: ‘fortune’ (fortunae),45 ‘one of the Fates’ (parcarum),46 ‘fate’ (fatus, ta, tum),47 ‘historically’ (gewyrdelice historialiter),48 and the specialized term ‘historiography’ (wyrdwritere glosses ‘historiografhus [sic], historiam conscribens’).49 An interesting quotation is taken from a liturgical manuscript: the OE interlinear version of ‘quam pro conditione carnis migrasse cognovimus’ reads ‘fore giwyrd lichomes foerde we ongeton’ [from the condition of mortality imposed upon the flesh we know it has departed]).50 Obviously, the range of meanings of the term was ample. In prose contexts, the ‘fate’ translation could be a modern reading exoticizing the more mundane meaning of ‘event’. Still, wyrd certainly belongs at least etymologically to the semantic area of ‘fate’.51 What exactly this entails in terms of what the people using the word actually thought is open to interpretation: one could conclude that the underlying conceptualization is that of ‘event, series of events’, hence ‘history’. Nevertheless, since fate was seen to have an important role in everyday life, history could be seen as the result (or embodiment) of fatality or, in a Christian interpretation, history as foreordained events.52 The transition to a more theological key of interpretation could have been easily made in this case: (ge)wyrd as history could have been underpinned (for the select few) by a Christian Augustinian view of history as a foreordained series of events whose deeper sense remains unknown to people – unless

74  What We Talk About When We Talk About History enlightened by God’s revelation, with the historian (wyrdwritere) being the one who can see, understand, and record the true course of history. Among less contentious OE words, another term for history is soþsagu (compare ON sannsaga ‘a true tale’), ‘true speech, a history’ (in one manuscript of the Gospels it glosses historiae),53 emphasizing the ‘truthful narration’ aspect of history, although, of course, ‘true’ does not necessarily mean ‘factual’ – more about narrative truthfulness further down. The compound soþsagu takes us to the more general term (with some 70 occurrences) sægen, which usually denotes ‘a saying, statement, assertion’ but also ‘tradition, report, story’ (‘Ic wolde gewitan hweþer sio segen soð wære þe me mon be þon sægde’ [I wanted to know whether the story I had been told about it was true]);54 hence, history is also an account handed down orally from one person to another, as well as from one generation to the next. Here, history as a form of oral memorial tradition is acknowledged (which conceivably led to adopting the term) in the sphere of clerical culture. The other very general term that means both ‘narrative’ and ‘history’ is spell (also a very widely used word, with hundreds of occurrences in the corpus and an enormous lexical family), usually meaning ‘a story, narrative, account, relation’ but also ‘a historical narrative, history’. Hence the Latin ‘historiam abbatum monasterii hujus in libellis duobus descripsi’ [I wrote the history and narrative of that monastery’s abbot in the second book] is interestingly translated as ‘Þara abbuda stær and spell þisses mynstres on twam bocum ic awrat’, which supplies a double gloss for the original historia: stær and spel, and thus ‘I wrote the history and narrative (or perhaps: ‘words, sayings’) of the abbots of that monastery in two books’.55 The term spel(l) has an extremely wide semantic range: it can mean ‘a false or foolish story, a fable’ (‘ealdra cwena spell’ [old wives’ tale]),56 ‘an instructive talk, a philosophical argument’, and even ‘a sermon, homily’ (‘Ðæt nis to spelle ac elles to rædenne’ [it is not to be taken as a sermon but to be read otherwise]).57 It has many derivatives, some of which are expressly used as history-words: ealdspell (see the aforementioned), soþspell, ‘a true story, history’ (glossing ‘historia’).58 Some forms of history are felt to be more truthful than the others (‘ciricalicra saegde soðspell’ [narrated the ecclesiastical (true-) history]),59 especially since this use of the term features in the context of heresy and fighting for truth, whereas in other similar contexts, the more neutral spell, not soðspell, is used. An even more interesting compound is spellcwide, ‘historical narrative, the language of history’ (‘ic wolde gesecgan cwæð Orosius, hu Creca gewinn [ongonn], þe of Læ[ce]demonia ðære byrg ærest onsteled wæs, mid spellcwidum gemearcian’ [I intended to tell, says Orosius, how the strife of the Greeks was first initiated from the city of the Lacedemonians, and describe it in the language of history]),60 being used here as a description for the specialized type of discourse proper to the historian and thus set apart from simple spell, mere ‘narration’.

What We Talk About When We Talk About History  75 A very wide-spread compound from the spell family is bi(g)spell (65 occurrences in the Corpus of Old English, especially in Boethian and Biblical paraphrases and translations), glossed by Bosworth- Toller as ‘a by-history, a parable, fable, example, proverb, story’. It is ambivalent in connotation in that it usually translates parabola (Solomon and Jesus both narrate bispelles), but it can also mean ‘frivolous fables’ (‘idele byspelln forbuh’ for the Latin ‘inanes fabulas devita’ [avoided frivolous stories]).61 Building on this negative connotation of narrative there are two other related words: leasspell, ‘a false story, fiction, fable’ (as a gloss for ‘figmenta’, or in the OE Bede: ‘be swylcum menn leasspell secgaþ’ [as are spoken in fabulous stories]),62 and even more negatively charged, leasungspell, ‘a false or foolish story, a fable’ (‘Ða hæfdon monige unwise menn him to worde and to leasungspelle ðæt sio hæte nære for hiora synnum ac sædon ðæt hio wære for Fetontis forscapunge’ [In accounting for this (phenomenon), some people would not concede to God His own power but sought worthless, petty explanations and invented the ridiculous story of Phaethon]).63 Interestingly, this extreme semantic range of spell and its compounds parallels that of the Latin fabula as used in early medieval historiography: Bede and Isidore, like many others, warn against the dangers of listening to idle confabulations, although they make use of classical fables.64 However, for Gregory of Tours and Fredegar, being ‘iocundus in fabolis’ was the attribute of a likeable man and a competent politician.65 Misleading advice is usually called fabula ficta, but the parables of Christ are occasionally called fabulae too. Interestingly, the early medieval English did not see anything wrong with equating fabula and historia (the two terms which are opposite, at least in theory, in the classical and early medieval classically rooted historiographical tradition). The Latin fabula is glossed, just like historia, either with spellung (‘Sægdon me þa unrihtan spellunge ah nalys swa æ þin drihtyn’ ‘Narrauerunt mihi iniqui fabulationes set non ita ut lex tua domine’ [The wicked have told me unjust fables, but not as thy law, Lord])66 or raca (‘stolidas fabulas’ glossed as ‘stunte raca’ [foolish stories])67. The sense is in most cases, negative, yet spellung seems to accommodate this variety of senses and moral values very well (‘stultiloquium 7 otiosas fabulas stuntspæce 7 idele spellunga’ [babbling and vain stories]).68 On the surface, all the terms I have surveyed do not have much connection to one another apart from having ‘history’ as one of their meanings. Yet this array of lexemes is the closest possible approximation to an early medieval English taxonomic system of history-words.69 Admittedly, they are not ordered in proper Aristotelian or Linnean taxonomies. Instead, this is the way in which these words might have been categorized in the minds of OE speakers: ealdspell, soþspell, etc., are types of historiographical spell (a category described in some contexts by stær), which in its turn is a type of spell as ‘generic narrative’. This type of taxonomy is

76  What We Talk About When We Talk About History paralleled in the cases of recedness, talu, etc., which, as folk taxonomies tend to do, present vertical polysemy (whereby the same word is used for both the superordinate category – ‘generic narrative’, and for the subordinate – ‘historical narrative’).70 Each of them also has subordinate categories which refer to specific nuances of the central concept of ‘historical narrative’ (hence ‘true history’, ‘ancestry history’, ‘old history’). These subcategories (usually compounds) are organized according to a binary pattern made up of a first element, semantically significant (eald, cyn, soþ) and a second lexically significant one (spell, sagu, talu, recednes). In other words, the first term of the compound belongs to a set of recurring conceptualizations associated with narrative and history across the early medieval English spectrum of mentalities (‘ancientness’, ‘kin/ancestry’, ‘truthfulness’), while the second element indicates the type of narrative these concepts are attached to (‘story’, ‘saying/oral tradition’, ‘tale(s)’, ‘reckoning/series of events’). The words in the first series of terms convey the larger ideas and are embodied by one of the elements in the second series. Early medieval English authors, glossators, and translators seem to combine terms from the two series of elements rather freely in order to attain the right semantic nuance. Of course, these words belong to the realm of early medieval English categorical/taxonomical conceptualization. Yet as will become apparent next, the same recurring conceptualizations of narrative and history (‘ancientness’, ‘kin/ancestry’, ‘truthfulness’) are present in oral-derived texts such as Beowulf but in these cases according to schematic patterns of conceptualization. This argument implies the substantial changes involved in the adaptation of classical Isidorian notions of history and fiction to early medieval English mentalities, even as the Latinate notions sometimes seem to be preserved in letter form. I now turn to representations of history in early medieval English poetic sources. The main point about them is that they are not taxonomic conceptualizations, like the previous ones, but schematic and, therefore, ‘fuzzy’, less hierarchically ordered than taxonomies, yet based on ordering principles which are meaningful even if not straightforwardly logical.71 For example, in an early medieval English cultural setting, the notions of ‘scop’ and ‘historical account’ are related in lexical memory because they label categories of objects that have been functionally connected and thus experienced together in the same event; hence, they form a schematic conceptualization, with scop being a culturally authoritative role schema in early medieval England, as we have seen in the preceding chapter. Taxonomically organized representations, on the other hand, are based on similarities among the units being represented – that is, on shared meanings.72 For example, ealdspræc and stær are related in memory because they refer to categories of historical narrative. Schemas are more characteristic of orally derived and collective memorial traditions, although, of course, we should not take the literary conceit of orality in Beowulf, for

What We Talk About When We Talk About History  77 instance, for a pure, untroubled window onto the actual world of oral memorial history, even if the illusion is strong enough even today.73 We have no unmediated access to any early medieval English oral culture – we can imagine this world only through its traces in written texts, such as Beowulf (very aptly designated ‘a pastoral of pre-textuality’), which outline ‘an imaginary cultural space of their own’.74 And yet by the very fact that it celebrates this oral tradition, Beowulf (and other such texts, conceivably less self-conscious about it) provides the only possible (for us) point of access to this tradition.75 There are many types of schematic conceptualizations, three of which are particularly relevant here: role, event, and emotion schemas. Role schemas are knowledge structures that people have about specific social, political, and/or cultural roles in a group (‘secretary’, ‘actor’, ‘CEO’), which include sets of behaviours that are expected of people in these positions.76 In the early medieval English oral culture, the figure of the poet, the scop, was the central role schema related to the preservation and commemoration of the past. A scop (according to Beowulf 867–76) is familiar with old stories (gidda gemyndig) and filled with eloquence (guma gilphlæden). His memory is emphasized (ealdgesegena worn gemunde [he had in mind a great multitude of ancient stories/traditions]), yet he does not merely imitate old stories but composes new verse (word oþer fand) bound in truth (soðe gebunden).77 Both verbs are equally significant: gemunde ‘remembered’ and fand ‘found/invented’ – the oral tradition is woven by the concerted actions of the warp of remembered narratives and formulas and the weft of innovative performance and incremental creation of new text.78 This passage also proves that the fundamental conceptualizations of history instantiated in prose evidence (truthfulness, old traditions, history as artful narrative) are present in Beowulf, too, even if less overtly. But schematic conceptualizations of history are not limited to heroic verse, as we have seen at the beginning of the previous chapter in Orosius’s History, where the role schema of ‘historiographer’ partly overlaps with that of ‘poet’. As in the case of spell, the semantic spheres of historiography and fictional narrative/poetic text clash again in this strangely permissive word, scop, showing a sophisticated understanding of history which embraced poetic manipulation and fictional distortion.79 Event schemas are another type of conceptualization abstracted from the speakers’ experience of certain events (for instance, the event schema ‘wedding’, which in Western cultures is associated with an array of heterogeneous elements: ‘love’, ‘altar’, ‘wedding ring’, ‘white dress’, etc. – its very conventionality signals the fact that it is the cultural schema for this event, even if not all the members of the culture adhere to it). In the case of early medieval English literary representations of history inspired by oral culture, the event schemata of ‘oral recitation’ and ‘recounting one’s ancestry’ are recurrent in texts like Beowulf. An important element in

78  What We Talk About When We Talk About History oral recitation is, as we have seen, the scop, the main agent of the event. But there is more to this schema. Texts such as Beowulf or Widsith never speak of ‘history’ as such. When they do refer to it, the sense of history is inextricably linked to the act of telling stories and the act of remembering. To take just one example, Beowulf 2105–17 describes the telling of tales and remembering ancient deeds. Hrothgar (gomela Scylding) remembered many things (worn gemunde), he had heard of many things from long ago narrated (fela fricgende feorran rehte) – this is precisely history being recalled and narrated. Even if the retelling of the events is a pleasant occasion in itself (andlangne dæg niode naman [throughout the day (they) took pleasure]), it is no mere diversion – the narrative/song (gied) has to be true and tragic (soð and sarlic), and even strange tales (syllic spell) are related correctly (rehte æfter rihte). I come back to this passage and the previous one to explore its underlying truth procedures in the following chapter. This conceptualization of history is in fact close to that shared by historians of the antiquity: for them, history was only what one had seen and could recall.80 This is, after all, Hrothgar’s personal history, and in this cultural frame (that of the society in the poem, not necessarily that of the society that read it or heard it), this is the chief way history could be soð. Perhaps not the only way since scops can also narrate truthful (hi)stories, but their authority comes from validation by way of memory and poetic skill – thus they can narrate true things about events they did not witness.81 In other words, ‘instead of implying an author, Old English verse implied tradition’.82 This seems to be a commonly held belief in oral vernacular memorial traditions: the audiences expected the ‘truth’ from their poet-historians. In the case of Old Norse literature, for instance, people indeed thought saga authors could reproduce true events from the Saga Age because they assumed that the story had been handed down in an unbroken transmission from the time of the events – thus tradition was maintained as a reservoir of truth.83 In Beowulf, this is marked by the phrase we gefrunon, ‘we have heard tell, we have learned by asking’ (oft-repeated from the first lines), which places the poem in an ancient and thus authoritative context – we have heard it from people who had heard it in their turn, etc.; therefore, it must be true. In this case, the underlying conceptualization of history is something one can hear and which is validated by its having been heard, just like in the case of sægen and the cwide compounds which are evidently still rooted conceptually in the oral memorial tradition, even when they are embedded in prose contexts which clearly refer to written history. This is yet another example of the strong conceptual ties across genres and cultural codes permeating early medieval English thinking about history. Another word which conceptualized history in ways that are related to oral memorial traditions (although not usually thought of as such)

What We Talk About When We Talk About History  79 is dom. The term means much more than the ‘fame’ it is usually translated to. Its connotation of ‘judgement’ is equally as important as that of ‘glory’. Indeed, all strive for dom in the world of Beowulf, and it is gained by doing extraordinary things, but it has to be granted by the judgement of the community that witnesses, remembers, and recounts those things. In this, dom is understood to be not so much the glory of the hero but rather the memory of the hero as a social and fictional (although no less true for it) construct. Thus dom is a form of history. People have to tell stories that recount heroic deeds for dom to work. In Beowulf, ‘dom and the stories that bestow, preserve and transmit it are depicted as the threads that stitch together the fabric of society’.84 Thus dom, usually thought of in terms of abstract ‘glory’, appears as an instantiation of a complex schematic conceptualization of history which encompasses the event schema of oral recitation, the role schemas of scop and hæleþ, and the emotion schemas of nostalgia, sadness, and awe. Another event schema prevalent in verse is ‘recounting one’s genealogy’. In fact, genealogy, as personal (but also ethnic and political) history is one of the privileged cultural patterns by which even distant history can be accessed throughout the early medieval English cultural world. Beowulf is full of genealogies, where they can sometimes be taken as a byword for history. Thus when Beowulf arrives on the Danish shore, he is asked to recount his lineage (frumcyn) before making any move: Nu ic eower sceal frumcyn witan|aer ge fyr heonan [no I must learn your lineage before you can go further] (Beowulf 251–2), to which Beowulf replies by narrating his personal history – that is, his ethnic ties and lineage. Beowulf is welcomed when it is made known that the leader of the East-Danes knows his noble descent: eow het secgan…aldor East-Dena þæt he eower æþelu can [the elder of the East-Danes says about you that he knows your noble-birth] (Beowulf 391–2), which is a way of saying that he knows and acknowledges the hero’s history. Yet even in the sphere of more official, politically relevant historical narrative, genealogy is paramount to early medieval elite culture in England and, conceivably, everywhere else. Here the event schema of genealogy performance is connected to the role schema of the poet-historian. For scops also played a much more immediately political role – they legitimized royal power. Before the ecclesiastical monopoly on learning, the scops would have been the ones composing and performing the early royal genealogies, thereby effectively legitimizing kings – they were ‘the link between the king and his divine source of power’, acting as chroniclers and historians, as well as mediators between the ruler and the ruled.85 Thus the scop acted as both the preserver and reenactor of history. This dialectic of storing the past in traditional form, and performing it is seen at work in the episode previously analysed (Beowulf 867–76), where the king’s scop recounts the events that had just happened, praising Beowulf’s killing of Grendel. Here history is being written on the spot

80  What We Talk About When We Talk About History and immediately re-enacted. Truthfulness is still a requisite – what the scop is doing here is not fictionalizing but ordering events in a narrative form acceptable for this type of literary tradition. In prose and glosses, ‘genealogy’ (cynnrecennis) is used to translate the generic historia, as previously seen, which shows the persistence of this schema in the minds of the glossators (‘genealogy is history’) – more on this in Chapter 3.86 The act of recollection is laden with nostalgia: as Hrothgar begins to recount his personal history, his heart wells up inside him (hreðer inne weoll) as he is mourning his youth (gioguðe cwiðan). Certainly, nostalgia is deeply embedded in early medieval English conceptualizations of history.87 In cognitive terms, this is an emotion schema. Theoretically, these types of conceptualizations are collective and cognitive since people relate these emotions to the situations in which they occur. Hence, the event schema ‘oral recitation’ can be connected to emotion schemas of ‘nostalgia’ but also ‘merriment, collective enjoyment (the dream of Heorot)’. In Beowulf, nostalgia mediates between a longing for communion with ancient heroes and the recognition that their antiquity sets them apart; hence, it comes from the fact that the past is seen as something that is simultaneously separate from and embodied in the present.88 Nostalgia as an emotion embedded in history and memory is also present in some of the Exeter Book elegies, especially in The Wanderer and The Ruin.89 Hwær cwom mearg? Hwær cwom mago? Hwær cwom maþþumgyfa? [Where is the horse gone? Where the rider? Where the giver of treasure?] (The Wanderer 92 ff.). Here, the wanderer’s complaint is more than a personal lamentation: he commemorates all the archetypal figures of heroic oral traditions, mourning not only their physical passing but also perhaps their waning as literary figures, as the memorial tradition they are part of is slowly being forgotten and replaced. In the same text (The Wanderer 74 ff.), but more so in The Ruin, nostalgia is combined with awe at the remains of a past that is not early medieval English (the Roman ruins) but which the literary tradition adopts through a shared attitude towards the Roman past combining admiration, wonder, and sadness. An alternative way of conceptualizing history is in fact also an alternative way of reading. In an oft-commented episode, Hrothgar looks at the hilt of the ancient sword with which Beowulf had killed Grendel’s mother and sees that on ðæm wæs or written/ fyrngewinnes [on it was written the origin of the ancient strife] (Beowulf 1688–89), referring to the biblical story of the flood.90 Just how this narrative was ‘written’ is not explained – whether in script (runes) or engraved, but the striking fact is that this type of writing is yet another type of conceptualizing the past prevalent in the early medieval English world: engraving images or writing on important objects (such as swords, armour) that could be handed down from one generation to another along with the piece of history represented on it. Of course, this type of preserving the past is sui generis, being part of neither literate elite tradition (although the actual messages

What We Talk About When We Talk About History  81 are sometimes written), nor the oral one (though the messages are handed down personally and in ways that resemble the way stories are passed on by word of mouth). A helpful model for this type of conceptualization is that of ‘visualized voice’.91 Since writing and visual representations were integrated phenomena in early medieval English England, often ‘writing visualizes voice, and the early medieval English used inscription to embed voice into objects’.92 In this way, the act of reading transcends the boundaries between literacy and orality through art. These examples indicate a convergence of mental patterns and, ultimately, a coherent and unitary set of conceptualizations that, despite the heterogeneity of the forms in which early medieval English expressed the notion of history, remains roughly the same across the ‘great divide’ between orality and literacy, folk and elite, and verse and prose. The cultural horizons of the translators of Bede into OE and of the author(s) and audience of Beowulf are thus separated only by different ways of conceptualizing the world. Thus the ways of thinking about history and narrative are mostly the same across the spectrum, and they differ significantly from those shared by Bede or Isidore: history narrates old events (ealdspræc), and it has to be truthful (soþ); although history is a narrative (spell), sometimes even a poetical account, recited by an oral poet (scop), the very fact that it belongs to a memorial tradition (gemund) and was handed down orally (we gefrunon) as ‘sayings’ (spræc) guarantees its truthfulness; the sequence of events is preserved as far as possible, for history is a reckoning (reccennys) and an ordering (getæl) of events in the fictional matrix of the historians, be they scops or eald-/cranicwritere; genealogies are powerful ways of ordering the past and shaping identities  – they are personal histories but become relevant for entire communities when kings want to legitimize their rule by having scops or chroniclers compile their cynreccennys. But what is the locus of mentality expression? Is it the very text which contains the specific history-word? Is it its pretextual history of usage? In other words, do these words uncover pre-existing mentalities, or are they created to give shape to new conceptualizations? Both ways of looking at it are essential and probably impossible to extricate from one another: a scribe/author might have used a word they had heard to express a new meaning that is akin to and yet different from the original meaning of the word. In such a scenario, gewyrd may have had the mundane meaning of ‘event’ in current spoken OE, but a glossator may have used it to channel connotations from the semantic sphere of wyrd as ‘fate/will-of-God’. Thus emerges the usefulness of using the cognitive-inspired lexical and semantic methodology I have used in this chapter to uncover the implicit conceptualizations of history in early medieval English words and texts. Also, it has enabled me to account for both the stability and the mutable and fragmentary nature of thought patterns pertaining to them, which would not be otherwise visible across different genres, types of discourse,

82  What We Talk About When We Talk About History and literary conventions. The cultural world of Bede and Alcuin and that of Beowulf and Widsith are thus subtended by different conceptions of history and narrative, yet early medieval English glossators or translators share some conceptualizations with the latter. The conceptualizations originating in the matrix of classical historiography are usually (but not always, as we have seen) categorical/taxonomic, those rooted in oral tradition are mostly schematic. Meanwhile, the glossators and translators are trying to adapt the conceptualizations of a foreign culture – that of classical learning – to early medieval English cultural models of thinking about history, and the result is far from being strictly Aristotelian or taxonomical but is an original synthesis of classical and early medieval vernacular conceptualizations of history; the latter is schematic in that it is not a result of trying to organize mental concepts into categories and notional hierarchies but stems from associations with specific early medieval vernacular cultural schemas, as I have argued. Thus the proposition schema ‘history is sung by poets’ describes a mode of historiographical writing, but it is rooted in an actual action performed over and over again in different milieus and contexts at one point in early medieval English history which became fixed into a cultural schema, that of the scop as singer of heroic poetry and royal genealogies (as previously argued). The schema continued to be relevant even after the actual actions associated with it had fallen out of everyday use. Originally, it was one of the fundamental ways of experiencing the making and rehearsing of history (as can be seen in Beowulf), and it remained one of the conceptualizations of history that would have been available (although not necessarily unanimously adhered to) to many early medieval English. If the locus of mentality is the community, however small, what are the types of communities which might have shared these word-embodied conceptualizations? Textual communities could be composed of both literate readers of texts and illiterate hearers of texts read or performed aloud, but oral communities could, of course, encompass the literate as well.93 Even Bede was familiar with traditional verse (he even appears as an oral performer in his Death Song, according to Cuthbert’s account) and was part of a lively oral community conversant with folk stories.94 Thus some of these communities of text and thought may have been as broad as encompassing all the speakers of late West Saxon OE (in the case of spell), while others could have been as limited as the literati of a small monastic community (cranic). The former would have been used in most social and cultural milieus, instantiated in various schematic (but also categorical) representations; the latter would have been based on taxonomical conceptualizations (‘cranic is a type of historia’). Yet these are extreme cases – most mentalities and communities must have been along the spectrum: some schemas would have been reshaped and adapted to Christian ecclesiastical/monastic contexts, as in Bede’s reshaping of the role schema of scop and the event schema of oral recitation in

What We Talk About When We Talk About History  83 the Cædmon episode, turning a secular diversion belonging to vernacular oral tradition into a religious literate genre. On the other hand, certain secular (or even ecclesiastical, as Alcuin’s Ingeld reprimand seems to indicate) aristocratic circles would have enjoyed vernacular epic verse and adopted it into elite literary culture. In such a narrative (e.g., Beowulf), the schema of oral recitation is reshaped and becomes culturally relevant again, albeit as a literary fiction. Conceptualization-wise, this marks a transition from schema to category, as oral recitation might have been a long-gone tradition for these people and thus not a schema in any original sense. Educated in a clerical learning environment, they would not have had the first-hand experience of actual oral delivery of verse. Their use of oral verse was itself a literary fiction, confined to the folios of a parchment. This is what the Beowulf poet(s) might have been trying to preserve and to revive: the forms of history and memory which preserved and constructed the heroic past, which was slipping into oblivion, ‘disconnected from the world of written texts and Christian learning, its very memory in danger of being lost’.95 For these reasons, the canonical mode of historiography espoused by Bede and the Chronicle and those on the bardic-rag-picker spectrum are not two separate horizons of expectation and narrative eventuality, even in their rhetorically different ways of understanding history and the past. Instead, these modes of historical production are part of a wider rhizome of implicit vernacular theories of narrative representation, temporality, and truthfulness implicit in the different shapes of verse histories possible. In the following chapter, I turn to the latter issue, employing a similar approach to reveal underlying theories and practices of constructing, recognizing, and certifying truth and verisimilitude along the broad historiographical spectrum of early medieval England and Francia.

Notes 1 Howe, 1997, p. 89. 2 Bede (2006). Trilling (2009: 20–1) argued against this antagonism, but on pp. 22–3, she discreetly restated the terms of the dichotomy, laying the foundation for her subsequent study of the aesthetics of nostalgia (more characteristic of ‘a Beowulfian aesthetic’ than of Bede’s ‘inexorably forward’ vision of history); for other such endeavours, see the essays in Hen and Innes (2000); see also Katherine O’Brien O’Keeffe (1990: 13 et passim). 3 For a rhizomatic approach to ecologies of heroic narrative, see Taranu (2015). 4 For a similar approach, see Hall (2007: 2 ff). 5 Roberts and Kay (2000). 6 ‘History’ in Oxford English Dictionary (2000). 7 For similar questions and helpful answers regarding the validity of the way TOE organizes its entries and semantic hierarchies, see Hall (2007: 9–11) and Anderson (2003: 20–65).

84  What We Talk About When We Talk About History 8 For arguments pleading for cognitivist approaches to OE texts, see Lockett (2011: 3–16) and Harbus (2012: 1–23). 9 Vansina (1985: 133). 10 Idem. 11 See Taranu (2015). 12 For critiques of cognitivism coming from such diverse fields as psychology, philosophy, and sociology, see Leidlmair (2009), Descombes (2001), and Woolgar (2012). 13 Geeraerts and Cuyckens (2007: 1 ff). 14 Sharifian (2011). 15 Sharifian (2011: 39). 16 Sharifian (2011: 12). 17 Sharifian (2011: 4). 18 Sharifian (2011: 5). 19 Sharifian (2011: 4). 20 Sharifian (2011: 3). 21 Sharifian (2011: 26). 22 Sharifian (2011: 6). 23 Sharifian (2011: 7). 24 Opland (1980: 231). 25 Bosworth and Toller (n.d.). 26 OE Bede (1890–98: IV.7, vol. 2, p. 282 and IV.30, vol. 2, p. 378). 27 Bouterwek (1853: 545). 28 Lass (1994: 189). 29 Dictionary of Old English Corpus (n.d.). 30 Idem. 31 Ælfric, ‘Homily’, in Assmann (1889: 79). 32 Napier (1900: no. 24, p. 156). 33 Cf. DOE Corpus (n.d.). 34 White (1978: 30). 35 Leo (1877: no. 149). 36 Tupper (1864: 2). 37 Thorpe (1844–46: vol. 2, p. 486). 38 Thorpe (1844–46: vol.1, p. 220). 39 OE Bede (1890–98: IV.24, vol. 2, pp. 344–46). 40 Skeat (1881–1900: vol. 2, p. 210). 41 Slocum and Krause (n.d.). 42 Napier (1900: nos. 2272–73, p. 61 and no. 2900, p. 78). 43 Skeat (1871–87: vol. 1, p. 25). 44 Goossens (1974: no. 1135, p. 222; no. 1591, p. 250). 45 Napier (1900: no. 2628, p. 71). 46 Napier (1900: no. 5480, p. 138). 47 Napier (1900: no. 32, p. 187). 48 Goossens (1974: no. 4141, p. 417). 49 Napier (1900: no. 1971, p. 53). 50 Stevenson (1839: 47). 51 For a recent assessment of the semantics of wyrd, see Pollack (2006). 52 Green (1998: 374–91). 53 Kemble (1858: 7 and 9). 54 Cockayne (1861: 24). 55 OE Bede (1890–98: V.24, vol. 2, p. 484). 56 Wright (1857: 55). 57 Cockayne (1864–66: 232).

What We Talk About When We Talk About History  85 58 Kemble (1858: 9). 59 Skeat (1871–87: vol 1, pp. 5–8). 60 OE Orosius (1980: III.1, p. 55). 61 Rhodes (1889: 213). 62 OE Bede (1890–98: IV.22, p. 591). 63 OE Orosius (1980: I.7 p. 26). 64 Bede (2013: 4 and 17); Isidore (2006: 1.44, p. 67). 65 Wood (2015). I am grateful to Ian Wood for allowing me to read a preprint version of this paper, which inspired me to nuance several positions in this chapter. Gregory of Tours (1937: II.32); Fredegar (2007: III 23 and IV 28). 66 Psalms 118.85, in Wildhagen (1910). 67 Napier (1900: no. 87, p. 224). 68 Förster (1914: 332). 69 For such taxonomical systems, see Anderson (2003). 70 Anderson (2003: 25–6). 71 For the notion of ‘fuzzy concept’, see Lakoff (1973). 72 Sharifian (2011: 8–11). 73 Liuzza (2005: 106). 74 Liuzza (2005: 105). 75 Caie (2008). 76 Nishida (1999: 758). 77 All subsequent quotations are from Beowulf (2008), and all translations are mine, unless specified otherwise. 78 Cf. Caie (2008: 115). 79 Opland (1980: 239). 80 Croke and Emmett (1983). 81 For authority in early medieval English oral contexts, see Bredehoft (2009: 7–14). For authentication/validation, see Greenfield (1976). 82 Pasternack (2006: 531). 83 Andersson (2008: 9). 84 Liuzza (2005: 95). 85 Opland (1980: 265). 86 The importance of royal genealogies (and their constructed, largely fictional nature) even later in early medieval England has been thoroughly explored in Dumville (1977); for the connection of oral tradition to textual genealogical lists, see Moisl (1981: 215–48). 87 Trilling (2009: 4–14 et passim). 88 Trilling (2009: 12). 89 Klinck (1992: 75–8 and 103–5). 90 For interpretations of the episode, see McNelis (1996) and Lerer (2006). 91 Karkov (2013). 92 Karkov (2013: 73). 93 Stock (1983: 1–10 and 30–87). 94 Bredehoft (2009: 20–4); McNamara (1994); Higham (2006). 95 Liuzza (2005: 101).

Bibliography Primary Sources Assmann, Bruno, ed., Angelsächsische Homilien und Heiligenleben, (Kassel: Georg H. Wigand, 1889).

86  What We Talk About When We Talk About History Bede, Historia ecclesiastica gentis anglorum, ed. and trans. by Bertram Colgrave and Roger Mynors (Oxford: Clarendon Press, 1969). Bede, Epistola ad Ecgbertum episcopum, ed. by Chris Grocock and Ian Wood (Oxford: Oxford University Press, 2013). Beowulf = Klaeber’s Beowulf and the Fight at Finnsburg, ed. by Robert D. Fulk, Robert H. Bjork and John D. Niles, fourth edition (Toronto: University of Toronto Press, 2008). Bosworth, Joseph, and Thomas Toller, eds, An Anglo-Saxon Dictionary (n.d.) [accessed 20 January 2020]. Bouterwek, Karl, ‘Angelsächsische Glossen: Die angelsächsischen Glossen in dem Brüsseler Codex von Aldhelms Schrift De virginitate’, Zeitschrift für deutsches Altertum, 9 (1853), 401–530. Cockayne, Oswald, ed., Narratiunculæ Anglice conscriptæ: De pergamenis exscribebat notis illustrabat eruditis copiam (London: John R. Smith, 1861), pp. 1–33. Cockayne, Oswald, ed., Leechdoms, Wortcunning, and Starcraft of Early England, 3 vols (London: Longman, Roberts & Green, 1864–66). Dictionary of Old English Web Corpus (n.d.) [accessed 5 April 2020]. Förster, Max, ‘Die altenglischen Beigaben des Lambeth-Psalters’, Archiv, 132 (1914), 328–35. Fredegar, Die Fredegar-Chroniken, ed. by Roger Collins, MGH Studien und Texte 44 (Hannover: Hahnsche Buchhandlung, 2007). Goossens, Louis, ed., The Old English Glosses of MS Brussels, Royal Library 1650, Verhandelingen van de koninklijke Academie voor Wetenschappen, Letteren en schone Kunsten van België, Klasse der Letteren 36 (Brussels: Paleis der Academiën, 1974). Greenfield, Stanley, ‘The Authenticating Voice in “Beowulf”, Anglo-Saxon England (1976), 51–62. Gregory of Tours, Decem Libri Historiarum, ed. by Bruno Krusch and Wilhelm Levison, MGH Scriptores rerum Merovingicarum, vol. I, 1 (Hanover: Hahn, 1937). Isidore of Seville, Etymologies, ed. and trans. by Stephen Barney et al. (Cambridge: Cambridge University Press, 2006). Kemble, John, ed., The Gospel according to Saint Matthew in Anglo-Saxon and Northumbrian Versions (Cambridge: Cambridge University Press, 1858). Klinck, Anne, ed., The Old English Elegies: A Critical Edition and Genre Study, ed. and commentary by (Montreal: McGill-Queen’s University Press, 1992). Leo, Heinrich, Angelsächsisches Glossar (Halle: Verlag der Buchhandlung des Waisenhauses, 1877). Napier, Arthur S., ed., Old English Glosses, Chiefly Unpublished (Oxford: Clarendon Press, 1900). OE Bede = The Old English Version of Bede’s Ecclesiastical History of the English People, ed. and trans. by Thomas Miller, 4 vols, EETS Original Series 95, 96, 110, 111 (London: Trübner, 1890–98). OE Orosius = The Old English Orosius, ed. by Janet Bately, EETS Supplementary Series, 6 (London, Oxford University Press, 1980). Oxford English Dictionary, third edition online (Oxford: Oxford University Press, 2000) [accessed 23 June 2020].

What We Talk About When We Talk About History  87 Pasternack, Carol B., ‘The Texuality of Old English Poetry,’ in The Postmodern Beowulf: A Critical Casebook, ed. By Eileen A. Joy and Mary K. Ramsey (Morgantown: West Virginia University Press, 2006), pp. 519–46. Rhodes, E.W., ed., Defensor’s Liber Scintillarum with an Interlinear AngloSaxon Version Made Early in the Eleventh Century, EETS Original Series, 93 (London: Trübner, 1889). Roberts, Jane, and Christian Kay, with Lynne Grundy. A Thesaurus of Old English in Two Vols (Amsterdam: Rodopi, 2000) [accessed 3 May 2020]. Skeat, Walter W., ed. The Holy Gospels in Anglo-Saxon, Northumbrian, and Old Mercian Versions: Matthew (Cambridge, Corpus Christi College, MS. 149), 4 vols (Cambridge: Cambridge University Press, 1871–87). Skeat, Walter W., ed. Ælfric’s Lives of Saints, 4 vols, EETS Original Series 76, 82, 94, 114 (London: Trübner, 1881–1900). Slocum, Jonathan, and Todd B. Krause, Old Norse Online: Base Form Dictionary, Linguistics Research Center, University of Texas at Austin (n.d.) [accessed 20 May 2020]. Stevenson, Joseph, ed., Rituale Ecclesiae Dunelmensis: With Interlinear Version Into the Saxon Language (London: Nichols, 1839). Thorpe, Benjamin, ed. and trans., The Homilies of Ælfric, 2 vols (London: Ælfric Society, 1844–46). Tupper, Martin F., ed. and trans., The Anglo-Saxon Version of the Metres of Boethius, in King Alfred’s Anglo-Saxon Version of Boethius De Consolatione Philosophiae, ed. and trans. by Fox, Samuel (London: H.G. Bohn, 1864), pp. 263–352. Wildhagen, Karl, ed., Der Cambridger Psalter (Hamburg: Henri Grand, 1910). Wright, Thomas, ed., A Volume of Vocabularies (n.p.: privately printed, 1857).

Secondary Sources Anderson, Earl, Folk-Taxonomies in Early English (Madison: Fairleigh Dickinson University Press, 2003). Andersson, Theodore M., ‘From Tradition to Literature in the Sagas’, in Oral Art Forms and Their Passage into Writing, ed. by Else Mundal and Jonas Wellendorf (Copenhagen: Museum Tusculanum Press, 2008), pp. 7–17. Bredehoft, Thomas A., Authors, Audiences, and Old English Verse (Toronto: University of Toronto Press, 2009). Caie, Graham D., ‘Ealdgesegena worn: What the Old English Beowulf Tells Us about Oral Forms’, in Oral Art Forms and Their Passage into Writing, ed. by Else Mundal and Jonas Wellendorf (Copenhagen: Museum Tusculanum Press, 2008), pp. 109–20. Croke, Brian, and Alanna M. Emmett, ‘Historiography in Late Antiquity: An Overview’, in History and Historians in Late Antiquity, ed. by Brian Croke and Alanna M. Emmett (Sydney: Pergamon Press, 1983), pp. 1–12. Descombes, Vincent, The Mind’s Provisions: A Critique of Cognitivism, trans. Stephen Adam Schwartz (Princeton: Princeton University Press, 2001). Dumville, David N., ‘Kingship, Genealogies and Regnal Lists’, in Early Medieval Kingship, ed. by P. H. Sawyer and I. N. Wood (Leeds: School of History, University of Leeds, 1977), pp. 72–104.

88  What We Talk About When We Talk About History Geeraerts, Dirk, and Hubert Cuyckens, ‘Introducing Cognitive Linguistics’, in The Oxford Handbook of Cognitive Linguistics, ed. by Dirk Geeraerts and Hubert Cuyckens (Oxford: Oxford University Press, 2007), pp. 3–21. Goetz, Hans-Werner, ‘Historical Consciousness and Institutional Concern in European Medieval Historiography (Eleventh and Twelfth Centuries)’, in Making Sense of Global History, ed. by Sølvi Sogner (Oslo: Universitetsforlaget, 2001), pp. 350–65. Green, Dennis H., Language and History in the Early Germanic World (Cambridge: Cambridge University Press, 1998). Hall, Alaric, Elves in Anglo-Saxon England: Matters of Belief, Health, Gender and Identity (Woodbridge: Boydell Press, 2007). Hanning, Robert, ‘Beowulf as Heroic History’, Medievalia et Humanistica, 5 (1974), 77–102. Harbus, Antonina, Cognitive Approaches to Old English Poetry (Cambridge: D. S. Brewer, 2012). Hen, Yitzhak, and Matthew Innes, eds, The Uses of the Past in the Early Middle Ages (Cambridge: Cambridge University Press, 2000). Higham, Nicholas J., (Re-)Reading Bede: The ‘Ecclesiastical History’ in Context (London: Routledge, 2006). Howe, Nicholas, ‘Historicist Approaches’, in Reading Old English Texts, ed. by Katherine O’Brien O’Keeffe (Cambridge: Cambridge University Press, 1997), pp. 79–100. Karkov, Catherine E., ‘Art and Writing: Voice, Image, Object’, in The Cambridge History of Anglo-Saxon Literature, ed. by Clare E. Lees (Cambridge: Cambridge University Press, 2013), pp. 73–98. Lakoff, George, ‘Hedges: A Study in Meaning Criteria and the Logic of Fuzzy Concepts’, Journal of Philosophical Logic, 2 (1973), 458–508. Lass, Roger, Old English: A Historical Linguistic Companion (Cambridge: Cambridge University Press, 1994). Leidlmair, Karl, ed., After Cognitivism: A Reassessment of Cognitive Science and Philosophy (New York: Springer, 2009). Lerer, Seth, ‘Hrothgar’s Hilt and the Reader in Beowulf’, in The Postmodern Beowulf: A Critical Casebook, ed. by Eileen A. Joy and Mary K. Ramsey (Morgantown: West Virginia University Press, 2006), pp. 587–628. Liuzza, Roy M., ‘Beowulf: Monuments, Memory, History’, in Readings in Medieval Texts: Interpreting Old and Middle English Literature, ed. by David F. Johnson and Elaine Treharne (Oxford: Oxford University Press, 2005), pp. 91–108. Lockett, Leslie, Anglo-Saxon Psychologies in the Vernacular and Latin Traditions (Toronto: University of Toronto Press, 2011). McNamara, John, ‘Bede’s Role in Circulating Legend in the Historia Ecclesiastica’, in Anglo-Saxon Studies in Archaeology and History, vol. 7, ed. by William Filmer-Sankey and David Griffiths (Oxford: Oxford University Committee for Archaeology, 1994), pp. 61–69. McNelis, James I., ‘The Sword Mightier than the Pen? Hrothgar’s Hilt, Theory, and Philology’, in Studies in English Language and Literature: ‘Doubt Wisely’: Papers in Honour of E.G. Stanley, ed. by M. J. Toswell and E. M. Tyler (London: Routledge, 1996), pp. 175–85.

What We Talk About When We Talk About History  89 Moisl, Hermann, ‘Anglo-Saxon Royal Genealogies and Germanic Oral Tradition’, Journal of Medieval History, 7 (1981), 215–48. Nishida, Hiroko, ‘A Cognitive Approach to Intercultural Communication Based on Schema Theory’, International Journal of Intercultural Relations, 23 (1999), 753–77. O’Brien O’Keeffe, Katherine, Visible Song: Transitional Literacy in Old English Verse (Cambridge: Cambridge University Press, 1990). Opland, Jeff, Anglo-Saxon Oral Poetry: A Study of the Traditions (New Haven: Yale University Press, 1980). Pollack, Sean, ‘Engendering Wyrd: Notional Gender Encoded in the Old English Poetic and Philosophical Vocabulary’, Neophilologus, 90, no. 4 (2006), 643–61. Sharifian, Farzad, Cultural Conceptualizations and Language: Theoretical Framework and Applications (Amsterdam: John Benjamins, 2011). Stock, Brian, The Implications of Literacy: Written Language and Models of Interpretation in the Eleventh and Twelfth Centuries (Princeton: Princeton University Press, 1983). Taranu, Catalin, ‘Who Was the Original Dragon-Slayer of the Nibelung Cycle?’, Viator, 46 (2015), 23–40. Trilling, Renée, The Aesthetics of Nostalgia: Historical Representation in Old English Verse (Toronto: University of Toronto Press, 2009). Vansina, Jan, Oral Tradition as History (Madison: University of Wisconsin Press, 1985). White, Hayden, Tropics of Discourse: Essays in Cultural Criticism (Baltimore: Johns Hopkins University Press, 1978). Wood, Ian, ‘Iocundus in fabolis et strenuus in consiliis: Roman Trickery and Frankish Mythmaking in the Chronicle of Fredegar’, in History, Frankish Identity and the Framing of Western Ethnicity, 550–850, ed. by Helmut Reimitz (Cambridge: Cambridge University Press, 2015), pp. 166–239. Woolgar, Steve, ‘Reconstructing Man and Machine: A Note on Sociological Critiques of Cognitivism’, in The Social Construction of Technological Systems: New Directions in the Sociology and History of Technology, ed. by Wiebe E. Bijker, Thomas P. Hughes, and Trevor Pinch (Cambridge: The MIT Press, 2012 [1987]), pp. 303–20.

3 ‘Truth Is the Trickiest’ Vernacular Theories of Truth in Early Medieval Culture

One day in the early ninth century in the kingdom of Mercia, a learned individual from the king’s entourage sat down and inserted Julius Caesar in the genealogy of the royal line of East Anglia as the son of the pagan god Woden. This piece of knowledge was then spread to other centres of power in early medieval England through authoritative channels of information.1 Judging by the cultural-political context and prestige of manuscripts containing this piece of information, this act of creative history-writing was by no means singular in early medieval England. Nor can we assume it was a mere slip of the quill or a fanciful instance of private creativity. While for Bede, Woden was the starting point for the genealogy of many English royal dynasties, by the late eighth century, competitive genealogists in the service of the different royal houses had begun to extend the pedigrees back in time beyond Woden.2 The pedigrees of Deira, Bernicia, Mercia, Kent, and East Anglia were all pushed a step beyond Woden to Frealaf. The Lindsey genealogy was the first to trace its dynasty back five generations beyond Woden to Geat: Woden – Frealaf – Friodulf – Finn – Godulf – Geat, and soon the other genealogists followed suit, eager to imbue their patrons with the cultural capital afforded by these ancestral back-formations. The insertion of Caesar was thus no giant leap for a royal genealogist but only an additional small step in a tradition of creative (and sometimes competing) history-writing. It is unclear who exactly all these mythistorical figures were (the name Geat, for instance, is particularly mysterious) or rather who the early English writers and readers of such information believed they were.3 Still, we are quite confident of what their addition to the genealogies was meant to achieve: previous research on early English royal genealogies and regnal lists has copiously shown that such fanciful rewritings of the past were very deliberate political acts, meant to bestow prestige and cultural capital on the kings who commissioned them.4 In Craig Davis’s words, royal pedigrees were ‘an ideological workshop’, a place to assemble and reconstruct the different traditions important to the culture of early English courts.5

‘Truth Is the Trickiest’  91 Yet nobody ever asks what these people believed they were doing.6 I mean this quite literally: how did the early medieval people involved in the making and reception of such repeated acts of obvious fabrication regard the truth of the fictions they were producing, consuming, and perpetuating with no obvious qualms? The question of whether serious-minded, Christian, early medieval English intellectuals believed that Julius Caesar was truly the son of Woden and that both of them were indeed the ancestors of their kings has never been asked despite the wealth of scholarship revealing the complex socio-cultural and political work such fictions fulfilled. A negative response is by no means obvious. For explaining the functions of these narratives in no way solves the issue of their truth status: were such claims meant to be believed literally? Were they some sort of figure of speech whose meaning and rhetorical mechanics we have lost? Indeed, scholars of early England (and medievalists, more generally) virtually never discuss whether, for instance, the writers and readers of the Anglo-Saxon Chronicle really believed that ‘fiery dragons were seen flying in the air’ in Northumbria in 793 or whether the people reading or listening to Beowulf believed (and were expected to believe by the poet – these are two separate, although related, issues) that the protagonist existed as a historical figure and truly killed all those monsters – let alone whether reports of dog-headed saints, monstrous people in the Orient, or supernatural beings such as elves were really believed. Such questions might appear as either imperceptive (‘It does not matter whether they believed these things – only what they did with them matters’) or may be read as veiled Gibbonesque indictments of medieval credulity. We medievalists are fond of the real people we sense behind the sources we study and thus would rather not address the cognitive dissonance between our view of reality and their belief in things and events that appear impossible to us. Yet these questions are pertinent if we are to understand why these features are present in many sources alongside more everyday occurrences which require no suspension of disbelief on our part. The texts I mentioned (the royal genealogies, the Anglo-Saxon Chronicle, Beowulf) never seem to signal in any way that their most outlandish claims are no more than tall tales. Their authors never wink at their reader, Pynchonstyle, to draw attention to their stories’ fictionality. On the contrary, these texts are thoroughly invested in presenting their narratives as true: as explained earlier, the genealogies were seriously political acts, while the Chronicle includes the dragon episode in a series of forebeacna, ‘foretelling signs’ of the disasters that would befall Northumbria. As such, the dragons are tacnum, ‘omens’, mentioned in relation to their very palpable historical fulfilment in the guise of a severe famine and the Viking depredation of Lindisfarne.7 Meanwhile, Beowulf’s dragon and Grendelkin are recounted in the same narrative breath as verifiably historical events, such as Hygelac’s raid on Frisia or the Yngling Swedish royal dynasty.

92  ‘Truth Is the Trickiest’ There is no sign that the dragons in the Northumbrian or Geatish skies and Caesar as the son of Woden are to be taken figuratively. We saw in Chapter 1 that Beowulf and other similarly ‘fictional’ texts are much less fantastic than they may seem. Indeed, it is one of the arguments of this chapter that such narratives are shaped according to truthmaking strategies precisely so that they may be believed. The concept of truth-making strategy is meant to encapsulate the means by which a narrative invites and compels its audience’s belief in its truth while taking into account both the larger social acts which the narration is part of, as well as the even larger cultural web of meaning in which are inscribed both the signals towards veracity within the story and the story’s social life. Since these signals are not cultural universals and are often implicit, buried in the texture of the account, they must be excavated, abstracted, and made explicit from the narrative, poetic, rhetorical, and emotional traces they leave in the text. These strategies can vary so significantly that the narrative truth they conjure becomes unrecognizable across socio-cultural/textual communities. Thus a particular type of truth may become identified as a blatant lie when encountered by a community that abides by different truthmaking standards (hence the denunciation of myths and fables as sinful by Christian theologians or of Christian theology as fabulous by Enlightenment apologists). More to the point, I argue that the narratives arising in early medieval England and Francia explored here were constructed and taken by at least some part of their audiences as true, but the even more fruitful part of this argument lies in making explicit the strategies of truth-making that describe the implicit contract between narrators of apparently impossible events and their audiences. For this remains a dead angle in many otherwise perceptive and highly valuable scholarly discussions of the fictionality of otherwise factuallyoriented medieval historical writings.8 For instance, Elizabeth Tyler’s groundbreaking work on historical narrative in early medieval England rightfully destabilizes received notions of fiction and history by pointing out that medieval narratives worked in milieux which did not recognize our category of ‘fiction’.9 Yet her account of the made-up status of certain parts of the historical narrative woven by the author of Encomium Emmae is based on the implicit assumption that these made-up stories were recognized as untrue by both their authors and audiences – it is simply that they held such texts to different standards of factuality, wherein inventio, fictional narrative, was not rejected as improper for historywriting but savoured as part of a tacit contract between the author and his audience that was based on a very particular notion of historia.10 This may well be the case for the Encomiast and his audience, but one cannot assume this very particular understanding of fiction to be representative of other texts narrating things that we understand to be fictional for the benefit of other audiences. I aim to go one step further from

‘Truth Is the Trickiest’  93 investigating the various ways in which medieval narratives play with fictionality and historicity to exploring the strategies of truth-making that enabled certain medieval narratives of (to us) impossible events to be actually believed. For, as I have argued elsewhere, much of the scholarship on medieval fictionality of the past decades still works within (and thus reinforces) the post-Enlightenment logic of truth familiar to us (i.e., that truth is a correspondence between an utterance or narrative and a state of facts of the external world), despite providing much-needed correctives to previous assumptions of credulity, confusion, or unsophistication as explanations for the bewildering multiplicity of configurations of fact and fiction in medieval history-writing.11 In a nutshell, then, this chapter is concerned with how such strategies of truth-making come to be processes, both cognitive and socio-cultural, by which truth is legitimized, enjoined, and sanctioned –and how they differ from what we, children of the Enlightenment, understand as ‘truth’. This constructivist approach to truth is inspired by Foucault’s notion of ‘regime of truth’: Each society has its regime of truth, its general politics of truth: that is the types of discourse which it accepts and makes function as true; the mechanisms and instances which enable one to distinguish true and false statements, the means by which each is sanctioned the techniques and procedures accorded value in the acquisition of truth, the status of those who are charged with saying what counts as true.12 But I contend that even this notion of a regime of truth is totalizing, based on the assumption that a hegemonic regime is sustainably enforceable in any society, especially in a medieval one lacking many of the modern state institutions involved in the production of knowledge and power on which Foucault focusses most of his work. My preoccupation lies specifically with vernacular theories of narrative truth – i.e., the heuristic reconstructions of the conceptualizations of truth produced via such truth-making strategies as they are implicitly present in OE poetic sources. Yet I by no means take them to be somehow unique to the loosely related set of cultural and political contexts that scholarship usually describes as ‘Anglo-Saxon England’ or as a manifestation of some particular Geist of ‘Anglo-Saxon’ culture.13 Rather, this corpus preserves strategies of truth-making that are widely encountered in other socio-cultural settings that are non-Western, pre-modern, or both (and that sometimes manifest themselves even in Western/modern/industrialized milieux) and which are distinct from the one prevalent in most early medieval narrative or historiographical sources rooted in the Latinate Christian or classical narrative tradition. Still, as we shall see, these strategies did not exist in separate socio-cultural spheres but can often be found at work within the same texts.

94  ‘Truth Is the Trickiest’ The peculiarity of configurations of truth in early medieval England (although, with the one exception mentioned next, not for the vernacular cultural horizons in Francia) has not gone unnoticed by previous scholars, and I will refer to their work throughout this chapter – here, I only briefly review their approaches and outline how my aims are different. Jeremy Downes is the first scholar of OE attempting to tease out the peculiar nature of truth in Beowulf via the concepts of verisimilar and legisimilar truth: the former born out of an impulse to account for every detail and accidental fact of a state of facts and the latter aiming to provide coherence and conformity to the laws of the speaker’s universe.14 John Niles’s insightful musings on the essentially social and personal nature of truth in early medieval English society are based on his work on storytelling and myth in oral traditions, which he sees as rooted in a particularly counterfactual notion of truth related to ‘thinking in the subjunctive mood’ that is very different from the objective, calculating, evidence-based path to truth we moderns like to think we pursue.15 Carolingianist Geoffrey Koziol eloquently opposes the pluralistic and context-dependent meaning of the OE term soð (usually translated as ‘truth’ but nowhere near what we understand of the word) to the Carolingian absolute faith in the possibility (indeed necessity) of separating true from false, orthodoxy from heresy, theology from idle fable.16 Yet one cannot assume all groups of Carolingian society shared in this conception with equal fervour or conviction – as I argue in the following chapter, there were spaces in Frankish society and history for dissenting opinions and, more generally, for alternative structures of thought and feeling. Tom Shippey and Craig Williamson have made valuable contributions, particularly regarding what they call the riddling nature of truth in OE sources. Particularly, Shippey focusses on the socio-cultural context of speaking truth in early medieval England where both audience and speaker are part of tightly knit communities in which loss of face is catastrophic, in which it was crucial to master the art of saying what you mean without really saying anything by using riddles, proverbs, and references to a corpus of verse and sayings encoded in highly traditional genres via tactful implications that may or may not be understood by the concerned party. These studies, however perceptive and inspiring, tend to explore one characteristic of the multifarious ways in which truth is claimed, assessed, and legitimized in early medieval England (all three aspects are different, although inextricably linked to each other) and tacitly assume that it describes the one early medieval English concept of truth (or at least the one specific to medieval or traditional oral cultures): Niles focusses on its social embeddedness, Koziol on its pluralism, Downes on its oral-cultural context, Williamson on the endlessly refractive and playful mechanics of truth. The only exception (also the most recent study and the most direct

‘Truth Is the Trickiest’  95 inspiration for the present chapter) errs in the opposite direction: Tom Shippey’s superb account of the multiplicity of truth sources identifies a potentially infinite number (although he stops at seven) of what he calls ‘polymorhps of truth’ that can be reducible, I suggest, to only three to four strategies of truth-making, which I aim to prove in the following discussion.17 Shippey puts to work Kurt Vonnegut’s concept of ice-nine (from his novel Cat’s Cradle, where it is a form of water that turns everything into ice) in order to make explicit some of the underlying concepts of truth lying implicit in OE verse, of which there are many, just as there are many polymorphs of water. To provide a brief rehearsal of his model, Truth One is used to label the biblical truth not open to debate; Truth Two is the metaphorically playful and potentially misleading truth of riddles; Truth Three is maxims expressing ‘cultural imperatives even in the face of a history or a reality which denies them’; Truth Four is proverbs ranging from the banal to the oracular; Truth Five is that of promises, vows, boasts, and other pronouncements that are in abeyance until they become true by being fulfilled through actions; Truth Six is the performative truth of charms; Truth Seven is that of allegories; and the series could go on indefinitely if Shippey hadn’t abandoned it for fear of making ‘one of the characteristic errors of the literate mind confronting the preliterate’ – namely, ‘to fix boundaries, make distinctions, reduce reality to bullet points’.18 This is a fair point, yet this is what Shippey accomplishes with nuance and gusto, and we are left much the wiser for it, and while his model was one of the starting points for this chapter, I suggest that what he reveals are not so much different theories of truth as context-dependent uses of strategies of truth-making. Thus what he calls Truths Two, Three, and Four can be envisaged as different genre-dependent manifestations of the same theory of truth based on a participative (or rather constellative) strategy of truth-making that provides disparate or cryptic utterances that express a truth awaiting an audience to piece it together based on cultural patterns of meaning and personal experience. At the same time, Truth Five, while very different in its social embeddedness and performative dimension, could be understood as being rooted in the very same theory of truth not as a quality already present in an utterance but as something to be established post factum, a virtuality that awaits fulfilment which is conceptually the same as the truth of a riddle or maxim that awaits to be brought to light by an audience or by its members making the truth lying in the maxim or riddle their own. The strategy of truth-making is thus conceptually the same, being based in each case on a performance that fulfils the truth of the oath, of the proverb, or of the riddle – performances that differ only in their socio-cultural and genre shapes (a heroic action in the former case, a mental piecing together in the latter two.

96  ‘Truth Is the Trickiest’ This chapter, then, aims to provide pathways to describe the several processes by which truth is identified, enjoined, and authorized in a variety of discourses from early medieval England (without assuming they are particular to this time and space). My focus is on the multi-level (cognitive, rhetorical, aesthetic, socio-cultural, political) processes of generating truth and not on any one early medieval English (or even vernacular medieval) concept of truth. In order to make these strategies of truth-making explicit, I will survey a number of different types of sources (individual words and glosses, proverbs, genealogies – as noted earlier), but as will become apparent, instances of storytelling and poetic composition in Beowulf are particularly rich venues for investigating implicit theories of truth. Indeed, I am not the first to notice Beowulf’s preoccupation with epistemology, with how knowledge of the truth is possible, although troublesome.19 Thus, I will focus on a number of scenes where the poem itself brings to the fore issues related to how the truth of a narrative is constructed (including correct and incorrect ways of recounting events and contradicting versions of the same events): the disputation with Unferth, verse-making in Heorot, the protagonist retelling his own adventures, and his adventures being made into verse inside the poem. But I leave these close readings of Beowulf scenes for the second half of the chapter, because in them, the different strategies of truth-making are closely entangled, allowed to contend with each other or taken to the breaking point, and my first aim is to make these strategies explicit as they are encountered more straightforwardly in other OE sources and only then read the Beowulf poet’s sophisticated use of truth procedures in light of this preliminary work of reconstruction. But before getting even there, I first need to defamiliarize the theory of truth widely current in the cultural horizon most readers of this piece probably share in order to open a space for alternative conceptions. For one cannot become aware of how truth is constructed via different strategies until one realizes that one’s own implicit theory of truth is neither natural nor self-evident but simply one among many. First, the usual methodological operation of focussing on the cultural, social, or political functions fulfilled by narratives of impossible events rather than on their truth status prevents us from understanding the variety of processes by which truth is judged, recognized, and legitimized as truth. Some of these processes (i.e., truth-making strategies) may appear wildly dissonant to the researchers working with pre-modern or nonWestern sources overwhelmingly belonging to WEIRD (Western, educated, industrialized, rich, and democratic) societies.20 Even the theories of truth elaborated by present-day academic philosophers are multifarious, although they can be roughly categorized into five families of theories: correspondence-based (truth is the correspondence between an utterance and a real fact in the world), pragmatist (truth

‘Truth Is the Trickiest’  97 is whatever is socially or psychologically useful to believe, a function of the ‘practices people engage in, and the commitments people make’), deflationary (only tautologies can be said to be true – to assert that a statement is true is just to assert the statement itself), pluralist (‘different statements can be all true without being true in the same way’), and coherence-based (the truth of an utterance derives from its coherence with the other utterances within a system of beliefs or representations).21 The implicit theories of truth I discover at work in OE sources will be shown to be more in tune with pragmatist, pluralist, and coherence theories of truth than with the correspondence-based one which is the default for most educated present-day people. In these alternative theories of truth (which may well appear counterintuitive to habits of the mind that considers an utterance as either true or false and nothing in between), truth is not a relationship existing a priori between things in the world and things in the mind but something made true by its being embedded in a narrative and poetic tradition (formally and stylistically) or an assessment reached post factum, a virtuality left in abeyance until a commitment is fulfilled or until a collective judgement is made. Looking at the lexicon of truth in the Germanic vernaculars, there seems to be no concept of truth as representing a fidelity to an abstract ‘reality’ or to a pre-existent corpus of divinely or socially sanctioned axioms. The cognates of ‘truth’ in all early Germanic languages (OS trûên, OHG trôsten, OE truwian) in fact denoted ‘loyalty, uprightness, and trustworthiness’.22 While OE soð did come to mean ‘truth’ when used to translate Latin veritas, it was mainly used in contexts of speaking oaths, citing customs and proverbs, telling the future, and telling stories.23 Indeed, like ON saga, OE soð basically means ‘saying, something said’ and is closely related to secgan ‘to say’.24 In Koziol’s formulation of the implicit theory of truth in the word, ‘soð was less a veridical proposition about the world than a capacity adhering to the man deemed credible to speak it or a quality of the truth being spoken’.25 As such, it is deeply rooted in a social context dominated by the spoken word and personal bonds based on values such as ‘trustworthiness, wisdom, standing, cunning’ in which an oath was a common means of juridical proof, ‘where interpreting dreams and healing illnesses needed canniness and command of lore; where truths were expressed by maxims, similes, riddles, and tales’.26 Koziol, in fact, suggests that the increasing appearance of soð in tenth-century early medieval English law codes and sermons and its gradual semantic evolution towards a closer correspondence with veritas is a result of trying to absorb Carolingian innovations in law and pastoral writing.27 Hence, it appears that the type of detailed cataloguing of all instances of the use of these terms that I performed in the OE corpus in the previous chapter might not take us far in this case. Nor can the IsidorianCiceronian tripartite categorization of narratives according to their

98  ‘Truth Is the Trickiest’ probability (historia narrates events that happened, argumentum – things that could have happened, fabula – things that are impossible by nature). As we have seen in Chapter 1, this approach is mostly a dead end since it only provided a useful soundbite for historians eager to legitimize their representational endeavour by appealing to a classical authority. Even when this theory became a standard historiographical reference (during the twelfth-century revival of the Latin tradition of historiography), this schema only provided an authoritative means to conceptualize and shape new ideas of narrative representation rather than a handbook method to be scrupulously followed.28 Indeed, as Justin Lake convincingly argued, the Isidorian view of history-writing had virtually no impact in the early Middle Ages (with the exception of a group of tenth-century authors who included Dudo and Richer), and that to a certain extent, ‘early medieval historians re-invented the genre of historiography as they went, picking and choosing between classical and late antique models, and in many cases puzzling modern readers with finished products that seem to defy strict classification’.29 Therefore, in order to see some of the previously mentioned truthmaking strategies in action, I will use the cognitive-semantic approach to cultural conceptualizations as they become manifest in cultural schemas established previously. To begin, I briefly rehearse here the argument I made elsewhere, where I unpacked the two truth-making strategies at work in Bede’s Historia Ecclesiastica that cause the author considerable uneasiness, which is evident in his preface.30 On the one hand, there is what Bede sees as the superior truth of theology and/or the authority of written sources (in both cases, a truth only discernible to a minority of elite readers), and on the other hand, there vera lex historiae, by which, as both Ray and Goffart point out, Bede did not mean any definitive and universal ‘law of history’ in an anachronistic Hegelian or even Rankean sense but instead a grudging concession to the role of fama vulgans, public opinion, however wrong, when writing a certain type of historical truth.31 Thus, ironically (more to us than to him), the phrase verax historicus Bede uses to describe himself means not that he holds himself to any quasi-modern standard of objective truth but rather the reverse – that in Historia Ecclesiastica he is compelled to write the narrative as it was known by a more public audience whose opinion already had a definitive shape.32 It is only in, for instance, The Life of Cuthbert that he can strip down the ‘(woefully) historical’ ‘picturesque irrelevancies’ of the Anonymous Life of St Cuthbert which he reworked to unravel the spiritual significance of the saint’s life via a theological mode of narration meant to reveal the superior truth (however unrealistic from a merely human perspective) for a different audience that was not as interested in the accidents of Cuthbert’s material-historical existence.33 Similarly, it is only in his Letter to Ecgbert that he can complain about the problematic

‘Truth Is the Trickiest’  99 proliferation of Northumbrian Eigenkirchen, whereas he appears content with this development in Historia Ecclesiastica.34 Even in the classicizing Latinate mode of history-writing Bede is employing, there are at least two truth-making strategies, which he describes in the preface to the Historia Ecclesiatica in very different terms. One is based on either trustworthy oral informants, some of which he names (such as Abbot Albinus of Canterbury for the southern material), or information handed down through a faithful line of bearers of an oral tradition (designated explicitly as traditio seniorum, traditio priorum, traditio majorum and only in the preface fama vulgans or implied with verbs and phrases, such as fertur, perhibentur, and ut aiunt), while the other rests on the spiritual authority of written sources (a priest of London named Nothelm, under the guidance of Albinus, had gone abroad to search the papal archives for Roman sources relevant to Augustine’s mission, and for the life of St. Cuthbert, Bede writes that he ‘partly took and faithfully copied’ from the authorized Lindisfarne vita).35 These two strategies are auctoritas (spiritual or simply factual truth via Scripture or trustworthy written sources such as canonical authors) and, in my formulation, traditio (community-dependent and socially useful truth expressed and transmitted via oral tradition). Clearly, none of them is based on a correspondence theory of truth but on different flavours of pragmatic truth: the former is legitimized by divine and/or institutional authority, the latter by the judgement of a community as to the true narrative of events. A full survey of the use of these concepts in Bede is unnecessary here and would require an entirely separate study, so I provide here only one example of where we can see them at work in tandem (and also in opposition). In book IV, Chapter 19 of the Historia Ecclesiatica, he feels compelled to assure his audience of the fact that queen Æthelthryth preserved her virginity even if she had been married for 12 years to Tondberct, king of the South Gyrwas, and then to Ecgfrith of Northumbria since some doubted this (sicut mihimet sciscitanti, cum hoc an ita esset, quibusdam venisset in dubium).36 He does this by quoting his oral informant, Bishop Wilfrid, as a undoubted witness of her virginity (testem integritatis ejus esse certissimum), but in order to strengthen this personal testimony since he feels some might still doubt that such things could happen in that age (nec diffidendum est nostra etiam aetate fieri potuisse), he states, ‘True histories tell us they happened several times in former ages’ (quod aevo praecedente aliquoties factum fideles historiae narrant) through God’s assistance.37 This alludes to other saintly virgins and presumably the archetype of them all, the Virgin Mary. Thus the ‘true histories’ he mentions can be no other than the hagiographies narrating the lives of these previous virgins and Scripture itself – all written sources containing divinely inspired narratives. They are the true histories that Bede credits more than even his most trustworthy

100  ‘Truth Is the Trickiest’ informants since in this passage he feels the need to appeal to their authority to establish the truth of an unlikely event for which he had already quoted the oral testimony of a bishop. Interestingly, here, too, we have a confirmation of the dynamics between the belief and scepticism Steven Justice describes in his study quoted in note 6: Bede acknowledges that Æthelthryth’s virginity is an improbable fact, and he twice admits that people expressed doubt, and even after his initial citation of testimony, some were still expected to doubt it, but, as an auctor himself, he guides his audience with a sure hand from natural scepticism to believing in a divinely facilitated truth that is improbable or even impossible for mere humanity. Here we see in nuce two truth-making strategies which can be called traditio (community-dependent and socially useful truth via oral tradition) and auctoritas (spiritual or simply factual truth via Scripture or trustworthy written sources, such as canonical authors). Bede clearly prefers the former theory of truth and is only compelled to use traditio by the scarcity of written reports and primarily by his writing a history meant for a secular elite audience (rather than a hagiography or private letter aimed at a theologically trained audience) for which the version of events known to public opinion was what mattered. Yet if we focus on Bede, we run the risk of forgetting that he was by no means representative of the large majority of people living at all times in early medieval England (and indeed in other places of the world). His intimate knowledge of Latin classical and Christian authors and consequently of the theories of truth with which they operated had little in common with not only the way common folk understood truth but also with the certainly elite communities from which texts like Beowulf emerged, monastic though they may have been.38 And, as I argued in the previous chapter, Bede and Beowulf do not belong to two separate worlds but were both part of ‘the living Anglo-Saxon world, dominated by talk and not texts, gossip not parchment’.39 Along similar lines, Alaric Hall has made the compelling argument that the common assertions of difference between us and early medieval people or ‘oral cultures’ may tell us more about our own need to construe ourselves as moderns than about pre-modern or non-Western rationality: ‘The way we write and think now is less radically different from people in highly oral medieval societies than the prevailing discourse in medieval studies would suggest’.40 I completely agree with Hall’s point that the configurations of truth I discover in early medieval texts are not fundamentally different from those at work in our culture. Indeed, I am not suggesting traditio is characteristic of ‘oral culture’, and auctoritas, of ‘literate culture’, as if any meaningful distinction can be made between them (and Hall argues convincingly against the usefulness of these concepts). Not only the Historia Ecclesiastica, but also Beowulf use both traditio and auctoritas – indeed, they also employ other truth-making strategies.

‘Truth Is the Trickiest’  101 Once we can recognize their similar mechanics beneath the variety of genre, narrative, and poetic conventions they underlie, we can see one or both of these two truth-making strategies at work in any text or discourse: it is traditio, for instance, that enables one’s great-grandfather’s wartime stories or conspiracy theories that are circulating on social media (both of which adhere to particular traditional patterns of narrative expectation) to be regarded as true, and it is auctoritas that allows us to see the work of a present-day professional historian writing on the same war as equally (although differently) true. While Beowulf predominantly relies on truth procedures other than auctoritas, the latter is indeed present in its inclusion of Christian narratives which many early scholars found so incongruous. The Grendelkin as descendants of Cain, the poetic commentary on the Danes’ slip into idolatry, the song of creation in Heorot using recognizable Genesis tropes, the story of the flood – all of these are meant for an audience not just acquainted with Christian lore but also implicitly with the truth procedure on which it is based – namely, auctoritas, the divine authority legitimizing scriptural truth that the poem’s audience certainly recognized and abided by. But the suspicious readings of these passages as later ‘interpolations’ – although wrong, as has been thoroughly demonstrated – attest to the tension existing between not simply different story worlds but primarily between different truth-making strategies at work in these different narrative traditions.41 For it is traditio that is the main truth-making strategy in Beowulf, as it is in many cultures that are predominantly oral (as is the heroic society imagined in the poem) or in which orality still informs the way people tell stories, relate to each other, or write verse (as was the case with many communities in early medieval England). The many instances of poetic composition in the poem give us valuable insight not only into oral poetic craft (at least at the time the poem was composed) but also into the truth procedure underlying it, which is different from that underlying scriptural culture. And though we should beware of assuming these scenes are anything more than fictional representations of purely oral composition imagined by early medieval English poet-historians composing in a textual mode of discourse constructing ‘a pastoral of pre-textuality’, there is much to be learned about this textually expressed but orally informed culture that the latter were a part of.42 Unlike with auctoritas, with traditio it is not the voice of ‘authors’ (poets) that makes the story true but rather their ability to remember and expand on the tradition they not only bear within themselves but in which they are also active participants – it is the process (of transmission and re-/production using culturally sanctioned forms), not the person, that authorizes the story. This is not only a function of orality but also, as Carol Braun Pasternack has demonstrated, of the orally informed textuality of OE poetry, which ‘does not employ an idea of the author but

102  ‘Truth Is the Trickiest’ rather an idea of tradition,’ and in which ‘the virtual absence of poets’ names signifies the author’s insignificance’.43 Yet this tradition is not envisioned as a reservoir of stories and linguistic patterns but as something which exists as part of a process in which ‘inscribed verse lays itself open to recomposition by subsequent poets and…in certain respects, scribes and readers could function as poets themselves’.44 I explore this collaborative nature of traditio at greater length next (indeed, its agonistic nature too – for often narrators contend with the tradition or with one another), but for now, I only wish to point out the peculiarity of OE verse, whose orally derived nature (open to recomposition and reworking by future scribes and readers) has been thoroughly explored and demonstrated.45 For there is a very different truth-making strategy underlying what Thomas Bredehoft regards as the productive (as opposed to the reproductive) ideology on which much of vernacular early medieval English textual culture is based (wherein an originary text is merely the raw material for subsequent reworkings rather than an archetype of which all subsequent reproductions aim to be faithful copies).46 While the construction of truth in a reproductive logic of textuality is based on auctoritas, in a productive mindset, traditio reigns. Much of the power of traditio in legitimizing truth is based on the stability of its formal features and of the linguistic patterns it employs. We can see the power to claim truth that a living oral verse tradition can deploy in Scandinavia and Iceland, which had a living formalized oral verse tradition late into the Middle Ages, enacted by skaldic poets for whose politically powerful audiences they constructed ‘a flattering and definitive version of the life and works of the king or chieftain being praised, securely enmeshed in the strict and complex forms of dróttkvætt which would ensure its enduring testimony’.47 Like so many times in the OE poetic corpus, skaldic poets, too, refer to hearsay (what Bede called fama vulgans, the narrative of events as known to public opinion) as the main source for their poetic presentation of the praise-worthy deeds of their addressees via the ‘I have heard’ formulae (ON frák, OE we gefrunon). Yet as Judith Jesch argues, skaldic verse does not merely ‘allude’ to ‘oral discourse and oral tradition’ but ‘is still very much a part of them’, for unlike the (fictional) early medieval English oral poets who ‘cannot assume the kind of stability in discourse and in the matter of discourse which the literate poet can’, skaldic poets legitimate the truth of their narrative through the strict forms of dróttkvætt, ‘designed precisely to ensure as much fixity in the text as possible in an oral culture, and which also ensured that the texts survived reasonably intact until they were written down’.48 The fidelity to the formal features of the tradition (epithets, oral formulae, verse patterns) and its productive logic, infinitely amenable to generating new truths, are inseparable, for as Elizabeth Tonkin’s work on still active epic traditions demonstrates, tradition ‘offered a means of evoking

‘Truth Is the Trickiest’  103 genuine emotions’, although ‘it would use an existing rhetoric to reformulate…unique, subjective responses’.49 With traditio, the teller ‘“codes” memories or reports of remembered events into existent stereot[y]pic forms’, although sometimes even members of WEIRD cultures feel that they can represent their experience ‘more truthfully by working through a well-standardised genre – and one that in many ways seems formulaic, and inimical to individual autobiography’.50 Thus when the participants in such traditio-based cultures try to proffer this experience into words, they will turn to known formulations, modes and genres to do so. This may mean that deeplyfelt experiences appear cliché-ridden, but even the most ‘original’ experience has to be represented through accepted rules of language and narrative production.51 It is the ‘presence and recognition of familiar plot structures’ that make stories ‘true’ to the participants in a tradition, and it is the intertexts, the formal features of versification or formulae, that ‘label the text as part of the community’s traditions in that it expresses similar thoughts in similar language, and thereby sanction it as “true”’.52 New truths (novel experiences, events, insights) have to be made understandable in the extant cultural horizon; hence, they have to be formulated in extant poetic or narrative forms. Rather than signifying stagnation, it is this insistence on faithfulness to traditional forms that ensures that new truths are understandable within mental frameworks already in place. In Katherine O’Brien O’Keeffe’s formulation, it is because ‘formulaic language appears in many places instead of being specific to one context’ that ‘people hearing the echoes may bring to their experiences diverse intertexts’ so that ‘though the language announces the conservativeness of the text, it does not require the reader to conform to a certain interpretation’ but rather through ‘its multiplicity of possible associations, [it] opens the text to varieties of interpretation’.53 Underlying the ‘aesthetics of familiarity’ that characterizes, as Elizabeth Tyler argues, much OE verse (whereby the impression of its belonging to an atemporal tradition masks its ability to express sharp political and social commentary), traditio as truth-making strategy is based on a theory of truth that is coherence-based (truth as what is formally coherent with a corpus of knowledge). This understanding of truth pervades many of the scenes of poetic production and storytelling in Beowulf. Here is the poet at Heorot turning Beowulf’s recent adventure (the defeat of Grendel) into song: Hwilum cyninges þegn guma gilphlæden gidda gemyndig se ðe ealfela ealdgesegena

104  ‘Truth Is the Trickiest’ worn gemunde word oþer fand soðe gebunden secg eft ongan sið Beowulfes snyttrum styrian ond on sped wrecan spel gerade, wordum wrixlan (Beowulf 867b–874a) This court poet’s authority (he is ‘the king’s retainer’) is established by his description as gilphlæden ‘full of grand stories’ (in Roy Liuzza’s translation) and gidda gemyndig ‘mindful of songs/ remembering songs’, signifying not only the quality of his memory, which is also praised (se ðe ealfela ealdgesegena| worn gemunde ‘he remembered much, many of the old stories’), but also his formal fidelity to the tradition. The other side of his skill lies in ‘finding other words’, or perhaps more idiomatically, ‘finding new words’ (word oþer fand). This is not a question of stylistic innovation but of appropriateness: the words need to be soðe gebunden, ‘bound in truth’, which testifies to the song being true while primarily referring to the technical skill of proper alliteration so that the phrase could be better translated as ‘rightly strung together’. The point is that formal appropriateness and moral truthfulness are equivalent here. There is no doubt to his poetic ability, which legitimates his ability to convey the truth: he snyttrum styrian ‘recites with skill’, and on sped wrecan spel gerade ‘adeptly tells an apt tale’ (per Liuzza), wordum wrixlan ‘interweaving his words’. Much has been made of the latter phrase, and after many decades, Leyerle’s argument about poetic interlace still stands. Indeed, this is what the poet does here on a narrative level, interweaving Beowulf’s tale with Sigemund’s and with the entire tradition to which the latter belongs, as well as on the verse level, where syntactic dislocation and interlace seems to have been a prized skill of poets composing OE verse.54 The point is that this technical poetic skill ensures the truthfulness of the new story: on the one hand, one has to keep in mind a tradition, which is both a canon of stories and themes and a style characterized by a particular technique of composition. Words and verses have to alliterate properly, but more generally, they (as well as the narrative they build) have to be woven together with skill. On the other hand, the tradition exists so that new events (such as Beowulf’s exploit) can become part of the community’s cultural memory and used in their turn. In light of the nature of traditio explained earlier, these two aspects are, of course, interconnected: the truth of the new narrative is guaranteed by its being interwoven with a canon of narratives about heroes such as Heremod or Sigemund, which in its turn requires a strict adherence to the stylistic and technical specificities of this poetic mode of narrative. Thus the tale of Beowulf is accepted as true because it is so skilfully told that it comes to resemble the stories about events that had happened many

‘Truth Is the Trickiest’  105 generations before him and that make up the tradition. In a sense, the narrative about current events has to be transported into the story-world of Sigemund or at least made formally homogeneous with the latter for it to become true.55 The epithet gilphlæden may signify the ‘lofty speech’ characteristic of gilp (often translated as ‘heroic boast’), although its usual association with heroic action may appear slightly incongruent when attributed to a poet. Yet, as Nolan and Bloomfield argued, the epithet points to the task of ‘the official story-teller’ of ‘determin[ing] that the hero’s gilp has been properly fulfilled and that his performance does indeed fit an a priori pattern of heroism’ and of ‘maintain[ing] the long tradition…and reiterating anew the moral values which distinguish every hero from his fellows’.56 Another passage of oral composition that also evinces this truth-making strategy comes later in the poem when Beowulf provides a lengthy account of his adventures to Hygelac, from which he omits many actionoriented details but in which he dedicates a full 16 lines (2101–17a) to describe and assess the verse-making going on at the feast in honour of his defeating Grendel. Interestingly, it is the same event that occasions these two meditations on poetic craft and its ability to encapsulate and convey truth: the previous one by the Beowulf poet, this one by Beowulf himself. These are rare occasions throughout the OE poetic corpus where anything resembling a theory of truth, storytelling, and poetic composition implicit throughout the corpus is put into quasi-explicit terms. What are the conditions of an authoritative narrator of the past and implicitly, of a truthful account in the protagonist’s view? The old Scylding-making verse (whether Hrothgar or not) had ‘heard tell of many things, from long ago narrated’, or, possibly, it was he who ‘told of far-off times’ (felafricgende feorran rehte, l. 2106). In either case, we encounter again the need for a tradition of time-proven tradition whose meeting point with the present revival of the past is the poet’s live verse-making. This characterization is reinforced further next when it is unequivocally the king himself who ‘remembers much’ (worn gemunde, l. 2114b) and can thus tell many tales: hwílum gyd awræc | soð ond sarlic hwílum syllic spell | rehte æfter rihte (2108b–2110a). There are two types of compositions discussed here: a song ‘true and tragic’, or ‘true and sorrowful’ (soð ond sarlic), and ‘strange tales’ (syllic spell), which are ‘recounted rightly’ (rehte æfter rihte). The former seems to describe ‘heroic’-themed verse histories (as we have seen in Chapter 1, where I discussed the particular heightened form of discourse described in OE as giedd or gyd), such as the Finnsburg narrative, whose popularity is indicated by its presence inside the poem, as well as independently, of Beowulf. Meanwhile, the latter (building on my argument in Chapter 2 regarding the semantic field of OE spell as encompassing all forms of narrative) might be something more akin to a ‘Wonders of the East’ type of tale, telling of strange faraway lands or perhaps more fantastical tales

106  ‘Truth Is the Trickiest’ of supernatural beings. In either case, much is made of their truth – in the former case, the tragic theme and mood of the poem is coupled with its truth, and in the latter, we encounter again the insistence on correctness of form, which ensures their truth.57 The strangeness of the tales does not lead them to be considered any less true than the songs from the former category; rather, it is the rightness of the telling that makes them true or not. We can envisage strange stories not being narrated correctly from a formal point of view (in terms of poetic craft or what Craig Davis called culturally determined ‘patterns of narrative eventuality’) and thus rejected by the audience, which Davis argued happened with Beowulf itself.58 It is true that the songs from the former category do not seem to need an authorizing strategy – they are already ‘true and sad’, yet the poet does dwell on the truth strategies involved in embedding Beowulf’s recent exploits into the tradition that includes Sigemund and Heremod. It is not that the mood and theme (sarlic) guarantee their truth but that the mood and theme involve their being composed in a certain mode which guarantees their truth. Thus sarlic is shorthand for the correct poetic technique – correct because it evokes the sar (sorrow), which is one of the emotion schemas characteristic of such songs (as we have seen in the previous chapter). This sorrow describes the emotional schema embedded in a certain genre of verse which thus constitutes its truth – although the range of emotional nuance and their textual expressions subsumed under the rather vague sar varied greatly from nostalgia through anxiety to sheer grief. This understanding is corroborated by the poet of another ‘sorrowful song’, the Exeter Book elegy The Seafarer, whose claim that he is ‘telling a true story’ (soðgiedd wrecan) – even though the story, of course, does not correspond to any factual reality and is in any case ‘less a story than a song built around an extended metaphor’ – can be understood in light of its belonging to a tradition which validates its moral and existential content, which makes it, in Geoffrey Koziol’s formulation, ‘a fragmentary utterance that speaks to the truth of the world’.59 These passages not only show us how the textual community from which Beowulf emerged imagined a poet of the preliterate world within the poem to have composed but also how this community may have envisaged Beowulf itself: as a composition aiming to be read as belonging to the tradition into which the Heorot poet inserted Beowulf’s recent exploits. If Beowulf was indeed a new hero, invented or magnified by the poet to the status of protagonist, with no pre-existing stories, such as the ones Ingeld or Sigemund had (as Larry Benson and Roberta Frank have argued), then passages such as these may be read as this new narrative being anchored into this Germanic story-world as a claim to its truthfulness.60 The Heorot poet’s skill in making Beowulf’s story true for his audience may echo the Beowulf poet’s desire for his own larger story about Beowulf to be perceived as ‘bound in truth’.

‘Truth Is the Trickiest’  107 But what happens when two or more narratives purporting to be true come to clash? And how does the collaborative and sometimes adversarial nature of producing truth work in the world of early medieval England? As we shall see, traditio and auctoritas only partly cover the complexities of truth in Beowulf. As at least some early medieval people recognized, truth gets trickier still – I now move to other regions of the OE spectrum of discourses. In the OE wisdom poem Cotton Maxims II, among other pithy utterances about appropriate or typical states of facts in the world (e.g., ‘a king shall rule his kingdom’, ‘a dragon shall lie on his hoard’), we find the statement soð bið swicolost (‘truth is the trickiest’, l. 10). In some editions and translations, this is emended to switolost (‘clearest’), although the manuscript itself (BL Cotton Tiberius B.i, f. 115r) clearly displays a ‘c’, not a ‘t’.61 Underlying this emendation might be a need felt by modern editors to correct the original so that it corresponds to the more expected dignified rhetoric of a wisdom poem – a certain clarity as to what exactly truth is and how one can find it. This is what, after all, one expects to find in a wisdom poem: wisdom, not further confusion. Or at least this is the expectation set by a post-Enlightenment theory of truth. However, as Craig Williamson deftly demonstrates, in the context of Maxims II, this statement makes perfect sense.62 This wisdom poem, as many other riddles and proverbs in the vernacular tradition of early medieval England (and indeed in many oral societies trust-based Gemeinschaft-type societies across the world and through time), is not meant to provide an uncomplicated image of ‘reality as the early medieval English saw it’.63 Take the first line of the poem: the apparently banal Cyning sceal rice healdan. Greenfield and Evert list the following possible translations, including the implications of each of them: ‘A king ought to rule/preserve a kingdom’ (i.e., a king ought to rule it rather than abuse or neglect it) or ‘a king shall rule a kingdom’ (i.e., it is in the nature of a king to rule a kingdom) or ‘a king must rule a kingdom’ (i.e. each kingdom must be ruled by some king rather than by an upstart or an usurper).64 Each of the maxims could be thus exploded into a variety of different meanings that are mutually contradictory when considered at once. This shows that the poem is not simply meant to be a static reservoir of wisdom and that in it the maxims are not simply quilted together but placed ‘against one another – colluding, colliding’ in a pattern that, as Williamson showed, ‘raises the question of perception’ and ‘defamiliarizes and deepens reality’.65 At any rate, any culture’s corpus of proverbs and maxims – despite (once again) modern expectations – is not meant to provide so much a coherent canon of axioms on the way the world works and on rules of proper behaviour (a vernacular Physics and Ethics) as a toolkit of culturally validated truths fitting for different occasions from which members of said culture can pick and choose depending on the message they

108  ‘Truth Is the Trickiest’ want to convey and on the social context in which they do so. The truth espoused by such maxims is not unitary and non-contradictory but rather a testament to an underlying belief that reality is not so much an external entity separate from its observer, as ‘a mosaic of man’s perception’ that can be apprehended only when the audience of a riddle or of such a riddling weave of proverbs restructures their perceptual categories when confronted with a constellation of possible objects to be perceived.66 In the case of the king proverb, ‘beneath the apparently straightforward gnomic half-lines, the poem points to a variety of possible kingly behaviours’ so that ‘what is slides into what should or might be [and] the possibility of “might not” always lurks beneath the surface [so that] the ideal is haunted by the shadow of real-world kingly faults and failures’.67 Still, I suggest that even Shippey’s and Williamson’s profoundly perceptive comments cast the underlying theory of truth in modern correspondence-truth terms: it is not ‘reality’ that the maxims and riddles are aimed at conveying but truth. We cannot assume that the people involved in the production and consumption of these discourses found the source of truth in any modern notion of realism – in other words, that they had a conception of ‘reality’ as separable from the thinking subject. In the statement truth is the trickiest, the very action of uttering the truth is called into question and, in the genres of riddle and riddling maxims (but on a grander level in Beowulf, as we shall see), is reconceptualized as a participative experience by an audience tasked with recasting the shards of language into a truth that is culturally and personally relevant. This activity of co-generating truth is often profoundly social and communal in the OE corpus, as we can see in two poems in the Exeter Book that provide us with a glimpse into one possible social context in which such riddlic-proverbial truths were exchanged and generated together by a community of speakers.68 Vainglory (15–18a) presents a scene in which warriors ‘sittaþ æt symble, soðgied wrecað | wordum wrixlað, witan fundiaþ | hwylc æscstede inne in ræcede | mid werum wunige’ (sit at feast, pronouncing true sayings, exchanging words, seeking to find out which battlefield might still dwell among men within the hall’). As Emily Thornbury argues, ‘The verb wrixlan “to interchange” and the creation of gied – here soðgied, doubly true sayings – links this passage with the challenge of Maxims I (1–4a)’:69 Frige mec frodum wordum! ne læt þinne ferð onhælne, degol þæt þu deopost cunne! Nelle ic þe min dyrne gesecgan, gif þu me þinne hygecræft hylest ond þine heortan geþohtas. Gleawe men sceolon gieddum wrixlan. (Question me with wise words. Do not keep your mind concealed, leaving hidden that which you know most deeply. I will not tell you my secrets

‘Truth Is the Trickiest’  109 if you hide from me your mind’s power and the intentions of your heart. Wise men ought to exchange sayings.) But rather than simply dialogue, Vainglory depicts a ‘many-sided conversation focused on shared memories’, while ‘the generic scope of gied allows for the possibility that some of these speeches took the form of poems’, which did not exclude ‘a competitive edge – later in the poem, arrogant words lead to bloodshed’, although at this point, ‘the exchange of words is chiefly a medium of social solidarity’.70 These passages imply a society where conversation and disputation are taken seriously, whether via the ritualized exchange of wisdom in the Exeter Maxims or the emphasis on soð in the warriors’ debate in Vainglory.71 We can see the competitive, agonistic side of this truth-making strategy in another OE source often characterized as a ‘wisdom poem’, the second of the dialogues of Solomon and Saturn, found in MSS Corpus Christi College Cambridge 422 and 41, which are some of the most complex OE texts to survive72 Here, Saturn is a ‘wandering scholar’ who seeks out Solomon to be taught wisdom and for ‘a contest of wits’: in Solomon and Saturn I, he wishes to understand the truth about the Pater Noster, while in Solomon and Saturn II, the two are presented as ‘sages engaging in a contest of wisdom’, wherein they’ test one another by asking riddles with mysterious, legendary answers – a four-headed lamenting bird, a Beowulf-like hero – or simple, quotidian solutions, like a book or deep water’.73 This is a prime example of truth-finding via dispute, and this social function of the truth procedure is present throughout the poems, especially in the latter, which shows ‘knowledge…gained and displayed’ in the ‘agonistic verbal performance characteristic of oral cultures’:74 Hwæt! Ic flitan gefrægn on fyrndagum modgleawe men, middangeardes ræswan, gewesan ymbe hira wisdom (Solomon and Saturn II 185a–87a) (‘Lo! I have learned through disputation in days gone by,| mind-perceiving men, counselors of middle-earth | working about their wisdom.’) Filistina witan, ðonne we on geflitum sæton, bocum tobræddon and on bearm legdon, meðelcwidas mengdon, moniges fengon, (Solomon and Saturn II 471a–73b) (‘the Philistine wise men, when we sat at disputation,| spreading out books and laying them on our laps,| mixing up discourse, taking up many of them’)

110  ‘Truth Is the Trickiest’ But despite the appearances, these are not judicial proceedings but congenial disputations generating truth by exchanging items of wisdom but also allowing them to contend with each other. As in Maxims I, where wise men should gieddum wrixlan (which in this context can be translated as ‘exchange sayings’, although in a more Beowulfian context, ‘interweaving songs’ would be more appropriate), this interweaving of truths is at once agonistic and collaborative, pointing to the tension between these separate individual truths which generate truth when allowed to compete – the ultimate judgement often belongs to the audience. More importantly, this competitive collaboration is not seen as a zero-sum game, for even the defeated parties come out of it with a renewed sense of truth: Hæfde ða se snotra sunu Dauides forcumen and forcyðed Caldea eorl. Hwæðre was on sælum se ðe of siðe cwom feorran gefered; næfre ær his ferhð ahlog. (Solomon and Saturn II 181–84) (‘Then the wise son of David had overcome| and rebuked the nobleman of Chaldea.| Nevertheless he was joyful, he who had come| on that journey, travelling from afar: | never before had his very soul laughed.’) What is remarkable about these passages is the co-existence of both an agonistic and a socially cohesive function to such exchanges, whose purpose is finding, observing, and/or arriving at truth. The strategy of truth-making at work in such contexts can be thus conceptualized as both a battle between opposing narratives and a communal exercise out of which truth can emerge, without the competitive side drowning out its collaborative basis – truth seen as a relational, intersubjective practice rather than a quality already dwelling in conceptual objects of perception. I suggest there is an overarching truth-making strategy at work in these texts, which I label collaboratio – an essentially constellative truth procedure, the result of a collective judgement of a community which can be either in-dwelling in the text or can be identified with the audience of a text. This type of truth can coagulate into a definitive sentence on a state of facts presented as such in the text after a community’s deliberation (as in Solomon and Saturn II), can be presented in a process-oriented fashion (as in the passage of collegial competition in Maxims I), or, finally, can be opened up towards the audience for a final extra-textual assessment (as I will argue with relation to Beowulf). In the latter embodiment of this truth strategy especially, truth can remain pluralistic, indeterminate, in abeyance – in which case, the audience is expected to piece together the truth of the matter depending on their personal experience and immediate needs. These characteristics define collaboratio as essentially constellative, an understanding which I derive from Renée Trilling’s compelling argument

‘Truth Is the Trickiest’  111 that much of OE poetry represents history in a constellative mode as distinct from the teleological ideology underlying salvation history. Trilling uses Walter Benjamin’s notion of the constellation to great effect in accounting for the peculiar nature of the vernacular theory of history present in the OE poetic corpus: A constellation takes shape from the relative position of the stars that form it, and is thus a function of the position of the stargazer, who sees a pattern and names it, thus giving it meaning, which the stars in themselves do not of course have, which is a very apt figure for grasping the way in which ‘concepts… appear to the critic in such a way that their relative arrangement is suddenly perceived as meaningful and becomes an image, or idea’.75 This understanding of constellative arrangements of concepts and narrative elements whose meaning is left open to the audience builds on the insights of scholars such as Carol Braun Pasternack and Katherine O’Brien O’Keeffe, who argue that OE poetry is a collaborative phenomenon and that scribes and readers play an active and creative role in textual production, which throws into question the issue of ‘authorial intention’ and thereby our ‘accustomed goals of interpretation’.76 In light of these practices of textuality, composition, and reading, collaboratio is merely a scholarly label describing a way of generating truth that must have appeared as natural to at least some audiences of OE verse. It is also used predominantly in the rag-picker mode of historical representation, with its disjunctive and self-referential narrative emplotment of the distant past. This distances it from the bardic and neo-bardic modes, where traditio lends its force to a more utilitarian form of historical representation. In other words, at least some communities of readers and listeners of texts such as Beowulf would have been expected and felt invited to deliberate, judge, and apply their own wits and experiences and thus find their truth in what modern readers have often regarded as a fundamentally inconclusive string of narrative moments and cryptic statements. As Elizabeth Tyler insists, medieval texts, as products of a world in which orality remained primary, are completed by a web of social and textual relations which call into question modern expectations that coherence relies on a single author’s vision, or that closure must be woven into the text rather than, for example, supplied by a shared understanding of the progress of time within salvation history, or by the social ritual in which a text played a part, or by the place of a poem within poetic tradition.77 In the light of this understanding (which is not unique to medieval – even our present-day society is fundamentally ‘oral’, despite a comparatively

112  ‘Truth Is the Trickiest’ greater reliance on inscription and various forms of textuality), apparently confusing passages such as the Unferth-Beowulf battle of narratives or the meta-narrative juxtaposition of the first roughly 2,000 lines of the poem with Beowulf recounting his own adventures appear strikingly different. These scenes have received the attention of generations of scholars, so I will neither recount them at length nor attempt to rehearse all the arguments that have been made about them – I am strictly focussing on the truthmaking strategies that come to light in these clashes of diverging narratives. Shortly after Beowulf arrives at the Danish court and meets Hrothgar, he is challenged abruptly by Unferth the þyle (variously translated as anything from ‘jester’ to ‘court spokesman’), concerning a swimming match between Beowulf and his friend Breca. In Unferth’s account of the events (lines 506–28), Breca won the contest because, after a week in the water, he was washed up among the Heatho-Ræmas from where he made it back home, thus apparently obtaining victory. In his reply (lines 530–606), Beowulf gives us a different version of the events: after five days’ swimming together, he and Breca did indeed become separated, after which Beowulf was busy killing sea monsters before being washed up in Lapland.78 A common interpretation of the episode is to take Beowulf’s story at face value, including his implicit devaluation of Unferth’s story by his assessment of the latter as drunk or sinister, which hints at his being a murderer of his kin. Unferth is often made out to be a malignant opponent of the protagonist, jealous of the latter’s heroic virtue.79 Yet as Michael Lapidge remarks, ‘Neither of these accounts is wholly true or demonstrably false: they simply report the incidents from differing perspectives’ and while ‘it is usually assumed that Beowulf “won” the contest because he presented a truer account of the events’ or demonstrated finer rhetorical skills, ‘there is nothing in the text to support this assumption’.80 Indeed, as Scott Gwara has demonstrated, there is much in the text to support a very different attitude towards the protagonist, detectable in what he calls ‘subaltern’ characters in the poem (such as Unferth, Wulfgar, the guard on the beach) voicing doubts about the consequences of Beowulf’s cavalier heroism.81 Gwara makes a powerful argument that there is an essential ambivalence built in the poem about the heroic ethic that is usually assumed to lie at its heart, embodied in the figure of the wrecca, a foreign or exiled warrior of superior strength and courage but also overconfident and reckless. Beowulf is implicitly characterized as one of them, being placed in a gallery of figures both heroic and sinister alongside Sigemund and Heremod. Reading the poem with eyes unclouded by a hypermasculine ideal of heroism that is more Victorian than medieval, it is clear that people in ‘subaltern’ positions in the heroic society at whose apex are both generous kings and arrogant wreccan express serious anxieties about Beowulf, who appears to be more the latter than the former.

‘Truth Is the Trickiest’  113 Read in this light, then, Unferth is not the heel to Beowulf’s babyface but a voice of reason at Heorot, expressing doubts that both some of Hrothgar’s retainers and some of the poem’s audience members may have initially had towards this newly arrived wrecca, who by all signs may be no more than an individualistic seeker of glory at all costs, potentially at the expense of the lives of men under his command. As Gwara argues, Unferth’s accusation against Beowulf has seemed ‘“mean-spirited” to many, but it highlights a common anxiety of the warband’, indeed he ‘expresses a majority opinion – not the view of cowards or rogues but of Hrothgar’s fighting men – that engaging Grendel is foolhardy’.82 While Beowulf mocks Unferth as ‘drunk with beer’ (beore druncen, l. 531a), the text makes it clear that Unferth is a highly respected member of Hrothgar’s household. He is not just a foolish drunkard, but an esteemed Dane sitting at the feet of Hrothgar (l. 500), whose ‘spirit everyone trusted’ (ll. 1165b-68a), who was known to have great courage, (l. 1465), and even Beowulf concedes that his ‘wit is clever’ (l. 589b). Michael J. Enright has challenged views of a negative Unferth by arguing that the þyle holds an important warband position as the king’s official spokesman.83 It is not clear that Beowulf’s attempts at discrediting Unferth are as well-received by either the intra- or extra-textual audiences as usually thought, including his accusation of kin-slaying (ll. 587–89) as deserving of ‘punishment in hell’ or ‘in the hall’ (according to Mitchell and Robinson’s more probable reading) – the fact that he hasn’t yet may point to the fact that at least Hrothgar doesn’t think Unferth deserves any such thing. Beowulf’s next accusation is purely counterfactual, and rather than involving Unferth’s worth, it could be read as more of a faux pas, the newcomer casting aspersion on the entire court, if not on all the Danes: Grendel could not have wreaked such havoc had Unferth’s heart been more battle fierce, indeed the monster found that the ‘Victory-Shieldings’ do not give him any trouble at all (ll. 591–97). The use of Sige-Scyldinga in this particularly un-victorious context may have come across as an irony. As Gwara noted, neither the Danes nor any other audience of the poem may have been too impressed with Beowulf’s boastful and rather arrogant attitude, at least at this point in the poem. Throughout his flyting with Unferth, Beowulf seems very preoccupied with being perceived as telling the truth: he repeatedly declares, soð ic talige (‘I tell the truth’, l. 532b) and secge ic þe to soðe (‘I say to you truly’, l. 590), which may be interpreted as no more than rhetorical markers or speech tics. On the other hand, this repeated reference to truth can be read as a more substantial proof of either the protagonist’s trustworthiness or, on the contrary, as testifying to anxiety over that which his story might be lacking most: truth. Both readings are possible, but my point is that Beowulf’s truth was not necessarily the truth for all audiences of the poem, some of whom may have been much more sympathetic to Unferth’s

114  ‘Truth Is the Trickiest’ truth. The intratextual subalterns’ anxieties about his motivations and ‘the potential for immoderation that he seems to express’ may have been echoed by at least some of the audience’s members, even while for others the protagonist may have been ‘an enigmatic figure whose incommensurate power they admire and fear’.84 It is not improbable, then, that some people reading or listening to the episode would have had misgivings about Beowulf’s claim to truth without necessarily taking Unferth’s at face value – for them, truth was thus left in abeyance, or both narratives may have been seen as potentially true in different contexts. Beowulf’s pre-emptive truth claims are different from most other instances of soð or soðe in the poem (ll. 524, 533, 590, 700) – they refer neither to a boast or promise that can be fulfilled with deeds, nor to an eternal truth, such as ‘that mighty God has always ruled humankind’ (ll.  700b-702a soð is gecyþed | þæt mihtig god manna cynnes| weold wídeferhð). So his pronouncements of speaking truth, while not necessarily coming across as dubious, call into question the very issue of the truth of these clashing accounts. At the end of the episode, the truth of the matter is left suspended, unresolved, for the audiences (both the Danes in the poem and the early medieval English hearing/reading it) to ruminate on and judge for themselves or collectively, as in the more collegial and wisdom-seeking flytings in Maxims I or Solomon and Saturn. Indeed, the use of soð in other instances may cast some light on the issue. Its peculiarities encapsulate many of the aspects of the procedures of generating truth discussed so far (their social dimension and their collaborative and agonistic logic in which the audience and the tradition bearer are involved in the production of truth and meaning). Their strangeness appears all the more poignant when the OE vocabulary of truth is set side by side with the Carolingian one. In a thought-provoking study of Carolingian myth (or rather of the reasons for its inexistence), Geoffrey Koziol argues that, in distinction to the language used by Carolingian intellectuals (all obsessed with issues of truth, authenticity, and orthodoxy, which they framed as dependent on their correspondence to scriptural truth and its various ecclesiastical forms of institutionalization), OE had no word for what we understand by ‘truth’.85 In his study of the ‘subjunctive mood’ type of truth encountered in myth and storytelling, John Niles poignantly corroborates Koziol’s argument through an account of early medieval English judicial proceedings in which ‘what mattered was not exactly the answer to the question, “What happened?”, for the people of that time did not necessarily assume the possibility of direct access to the truth’ but instead that which was determined by the implicit question, ‘Which of the two parties has the power of speaking a “true” story?’86 The procedure of arriving at the truth of a case was not forensic and evidence based but entirely socially determined, a matter of ‘trustworthiness and…powerful connections’: the defendant was legally aþwurþe, ‘oath-worthy’, if recognized as someone for whom

‘Truth Is the Trickiest’  115 ‘enough people of high rank were willing to offer…mundbyrd “personal protection”, thereby serving as sureties for his word’.87 The judicial clash of narratives was thus decided in the favour of the one with a higher degree of ‘oath-worthiness’, measured by the number or rank of such ‘sureties’, which were not exactly witnesses in the sense of being able to corroborate the narrative put forth by one side as much as attesting to the social standing of the narrator.88 Far from being relegated to the realm of storytelling, this pragmatic conception of truth as whatever produces social cohesion and as the judgement of a community held wide cultural currency in early medieval England. As Elizabeth Tonkin noted, not all legal systems ‘are geared to finding “the truth”, or to making “an impartial” decision’, but to generating ‘social solutions to perceived social breakdowns and support therefore the litigant who has mobilised the strongest support, which may be an equally rational (and honourably considered) decision’.89 With this understanding, we may surmise what many early medieval audiences of Beowulf would have thought when hearing both the flyting and the protagonist’s protestations of soð. Since Beowulf’s dispute with Unferth was not a judicial proceeding but a contest of truths, there was no need for Hrothgar (or the poet) to intervene or pronounce a final judgement. Preserving the social cohesion of the Danes was the foremost duty, and when expedient, this need required a collaborative and pluralistic theory of truth (whereby truth is left in abeyance or the possibility of more than one truth is entertained). Read in the light of the workings of collaboratio, such clashes of perspectives or narratives appear as not simply verbal contests with one clear victor, for they still generate truth according to the same constellative strategy as the more congenial exercises of collaboration analysed above. Rather, I would characterize them as points of inflection for truth procedures, where the issues of truth and of how it is generated, authorized, and assessed are brought to the fore to an extent not encountered in other early medieval texts. This willingness to allow divergent narratives to exist, the issue of their ultimate truth being left unresolved can also be seen on an even grander scale in the scene of Beowulf’s homecoming, in which he provides a ‘critical retelling of the poem to this point, the hero becom[ing] his own poet’, as Seth Lerer remarks.90 The episode has received close scrutiny from, among others, Lerer and Michael Lapidge, whose investigations provide a range of important insights into the poem’s highly sophisticated, even experimental, play with narrative authority and interpretative possibilities.91 Lapidge explains the striking discrepancies between Beowulf’s retelling of his adventures to Hygelac (2000–2151) and the roughly 1,700 lines of the poem preceding it as a point at which ‘the poet clearly expected the audience retroactively to compare’ his own account with the protagonist’s perception of events.92 Meanwhile, Lapidge explains the tension between these different narratives of the same events

116  ‘Truth Is the Trickiest’ as ‘a humorous critique of tale-telling, and another reassessment of the nature of narrative authority’.93 Both scholars poignantly bring home the point that this tension between the accounts is absolutely intentional and allows us a glimpse into a very different way of conceiving perception (Lapidge) and constructing narrative (Lerer), but at the same time, this is yet another point of inflection for truth procedures, one in which collaboratio is opened up to an even more radical extent towards the audience than in the Unferth episode Beowulf’s insistence on declaring that he speaks the truth in the Unferth episode is echoed here in the way he is ‘carefully manipulating’ his recreation of his own past in his account to Hygelac through ‘his emphases on the precision of detail and the correctness of his own and of the scop’s earlier performances’ (discussed with traditio).94 This savvy manufacturing of consensus around his narrative, this seduction towards his own truth, may have worked in this episode (more than in the Unferth flyting at least) in establishing ‘the audience’s trust in Beowulf’s narrative authority’ through both ‘his imposing bearing’ and ‘from the pervasive associations between the hero and the scop’.95 Yet, as in the case of the Unferth episode, it is not clear that all members of all audiences of Beowulf would have seen Beowulf’s account as true – indeed, the poet’s juxtaposing it with his own points to the problematic nature of assessing the truth. In both of these cases, the lack of commentary on the poet’s part may well confuse any audience, modern or medieval. What are we to make of these accounts? Which one are we to believe? The lack of overt, explicit commentary does not exclude more subtle cues, although different audience members can piece them together in different ways. Lapidge sees this feature of the poem as intentional and indeed ‘unprecedented’: his argument is that ‘the Beowulf-poet’s mental orientation was philosophical and epistemological’, which explains the poem’s ‘eccentric’ narrative structure.96 At moments of repetition such as these two (although they are not the only ones – indeed, the battle with Grendel is retold three times), the poet fully ‘intended the audience of the poem to reflect, retroactively, on the narrated events and their relationships, during the course of the telling’.97 In these inflection points, the audience is fully brought into the truth-making strategy of collaboratio, although this also happens to a lesser or greater extent in wisdom poems such as the ones explored earlier, indeed the entire tradition of OE riddling verse and gnomic discourse prepared Beowulf for this unprecedented flyting of the poem with its audience, understood as both collegial weaving together and clashing of narrative with traditional and personal truths. It is not that these people belonging to an otherwise highly traditional society were ambivalent about truth like some avant la lettre post-modernists (or, rather, like the vulgarized understanding of the latter). My point is that these texts show us that they understood very well that truth

‘Truth Is the Trickiest’  117 claims are often adversative – one person’s truth is another one’s lie – but for society to work, the final judgement must often be open-ended, the truth always awaiting an audience to piece it together via exchanges of wisdom, stories, and perspectives veering more towards the agonistic or the collaborative but at no point turning into either a zero-sum game or, at the other end of the spectrum, a choir singing in unison. Hence, Beowulf versus Unferth is not a trial but a momentary clash of narratives whose ultimate truth the Danes in the poem or the early medieval English outside it have to establish for themselves. The poem is interested not in establishing but allowing the procedures of generating truth to unfold, intermingle, and contend with each other and sometimes fail. In the Unferth-Beowulf flyting and Beowulf’s account of his adventures, we see collaboratio and that socially embedded notion of soð turn competitive and shrill while also being opened up to the audience. In the scenes of verse composition, we see traditio at work restating old truths and producing new truths out of recent events, and through its uses, we can glimpse at what the Beowulf poet is trying to achieve in telling these sad and strange tales. In Koziol’s assessment, in cultural horizons dominated by such vernacular theories of truth, a tradition (both as a story-world and as a set of formal requirements) can be the bearer of truth via traditio as truth procedure not because the events it conveys are factual since in it ‘there might be good stories and bad stories, useful stories and not-so-useful ones, stories that conformed to and supported tradition and ones that did not’, but because their adherence to the tradition enabled the criteria according to which their truth was assessed to be ‘highly specific to particular needs in particular situations (just as a particular god was good and useful in on situation but not another) – in any case, ‘the idea that a story was simply “true” as an abstract absolute – this is something we do not find [in early medieval England]’.98 But, despite Koziol’s statement to the contrary, such conceptions of truth were probably shared by at least some Frankish communities (as evidenced in the next chapter), just as the insistence on the all-encompassing quality of the truth of Christian doctrine was not exclusively ‘Carolingian’ but shared by communities such as the ones in which Bede, Aldhelm, or Alcuin felt at home – in which no song of Ingeld was to be heard, however loud it may have resounded in other halls. Here is where collaboratio comes into play: the task of finding truth rests upon the audience or, in a more adversarial context, on each other’s narrative antagonist. Several truths can coexist depending on the different needs of members or groups of the audience, but when there is a need for one truth, as in a judicial debate, that truth is as much a function of the social standing of the narrative agonists (in its turn a function of the performative truth an individual has enacted) as of the social cohesion that the better truth can foment (in its turn at least partly a function of the cultural and narrative-patterning coherence it can elicit).

118  ‘Truth Is the Trickiest’ Recast as based on a concept of truth that combined collaboratio and traditio, then, the Caesar-Woden additions to the early medieval English genealogies I discussed at the beginning of this chapter open up a space for narrating the past ruled equally by cultural-political symbolism and playful riddling, not by factual truth as representing a historical reality. The self-conscious insertion of such figures as Geat, Woden, or Caesar in the line of East Anglian kings can be understood only when considered from outside the paradigm set by correspondence-based truth and understood rather as a puzzle to be pondered by highly educated audience members who have the necessary background knowledge for its solution. Its truth exists not so much in the piece of information conveyed as in the negative space of the audience’s act of reading it and mulling it over – a narrative truth that dwells in virtuality, becoming actual insofar as this audience can piece it together from the names strung together by the genealogist. In the truth of genealogical tradition and the story-worlds it refers to (subtended by a culturally authoritative procedural cultural schema, as we have seen in the previous chapter), the constellative dynamics of collaboratio which awaits an audience to cause it to emerge is married to the conventionality of traditio underlying the canonical form of the genealogy that alone can be made to express new cultural-political truth – namely, the insertion of both the Roman past and the Germanic pre-Christian past into the new national ideology of Angelcynn.99 Both of these functions of the genealogy are based on underlying pragmatist theories of truth: in the former case, the truth is constellative, in abeyance, waiting to be fulfilled and pieced together by the narrative’s audience; in the latter, the truth of the line of descent, however improbable, is legitimized by the authority of the form of the genealogical tradition and, narratively, of the story worlds of Roman history and of Germanic pagan gods (however much euhemerized). These conceptualizations of truthfulness and narrative verisimilitude map onto the bardic-rag-picker model outlined in the first chapter in the following way. The constellative truth procedure of collaboratio is one of the central characteristics of what makes the rag-picker mode of historical representation. As evident from the range of discourses it features in, this is by no means to be found only in the vernacular ‘heroic’ verse I focus on in most of this book. Rather, I suggest the rag-picker mode should be seen in dialogue with the constellation of discourses surveyed in this chapter in which collaboratio is the predominant truth procedure, whose principles it uses on the grander scale of narrative emplotment of heroic verse history, offering up the possibility of constructing a meaningfully unitary and cohesive narrative history as a subject of meditation. Meanwhile, the bardic and neo-bardic mode predominantly employ traditio as the strategy of emplotment of its relatively linear, more ideologically transparent histories via the same vehicle of culturally authoritative

‘Truth Is the Trickiest’  119 vernacular heroic verse, with auctoritas remaining the dominant truth procedure in more canonical historiography. The Cædmonian mode of biblical heroic history, meanwhile, combines the traditio of vernacular heroic verse and the auctoritas of its scriptural story-world. In the following chapters, I move to a socio-emotionally inflected exploration of how the formal features and ideological investments analysed in Chapter 1, the implicit theories of narrative representation (Chapter 2), and the truth procedures (Chapter 3) underlying the different modes of historical representation at work in GHP function in society by responding to and influencing cultural and political conditions, as well as by providing a space in which its audiences could work through intense emotions.

Notes 1 ‘Caser Wodning/Uodning’ is found in London, British Library, MS Cotton Vespasian B.vi, 109v (probably written in Mercia in the early ninth century); London BL, MS Cotton Tiberius B.v, 23r (Sussex, the second quarter of the eleventh century); and Cambridge, Corpus Christi College, MS 183, 66v (Wessex, the first half of the tenth century). Documented in Dumville (1976: 24–6 [provenance and dating] and 31, 34, 37[genealogies]). 2 Davis (1992: 28). Bede (1969: I.xv, p. 73). 3 For Geat, see Taranu (forthcoming). For mythistory as the playful commingling of myth and history, see Niles (1998:218). 4 Dumville (1977) called these the ‘Anglian collection’ as opposed to Wessex traditions associated with the ASC. Dumville (1976, 1977). 5 Davis (1992: 28). 6 Except in a broader, functionalistic sense, against which we see the extremely insightful and thought-provoking Justice (2008: 11). For a comprehensive critique of the terms in which discussions on veracity have usually been carried out, see Taranu and O’Connor (forthcoming). 7 The 793 entry in Manuscript E of the Anglo-Saxon Chronicle (although not, for instance, in Manuscript C) reads: ‘Her wæron reðe forebecna cumene ofer Norþanhymbra land 7 þet folc earmlice (26r) bregdon: þet wæron ormete ligræscas, 7 wæron geseowene fyrene dracan on þam lyfte fleogende. Þam tacnum sona fyligde mycel hunger, 7 litel æfter þam þæs ilcan geares on .vi. idus Ianuarii earmlice heðenra manna hergung adiligode Godes cyrican in Lindisfarenaee þurh reaflac 7 mansleht’ (ASC 2004: 42). 8 For general surveys, see Partner (1977); Otter (1996); Kempshall (2011). 9 For narrative traditions of early medieval England, Elizabeth Tyler is the foremost authority – among others, see Tyler and Balzaretti (2006); Tyler (2006); Tyler (2017). 10 Tyler (2017: pp. 51–134, especially p. 105). 11 Taranu and O’Connor (forthcoming). 12 Foucault (1980: 131). 13 Harris (2003: 3). 14 Downes (1995: 130). 15 Niles (2007: 298). 16 Koziol (2007: 86, n. 39). 17 Shippey (2017: xxvi–xxxv). 18 Shippey (2017: p. xxxv). 19 Lapidge (2001: 88).

120  ‘Truth Is the Trickiest’ 20 For the landmark critique of the bias of most sociological research towards the individuals from WEIRD societies who are the majority of its subjects, see Henrich, Heine, and Norenzayan (2010). 21 Brief introductions to each in David (2016), Capps (2019), Pedersen and Wright (2018), Stoljar and Damnjanovic (2014), and Young (2018). 22 Koziol (2007: 84). Green (1965: 117–26). 23 Koziol (2007: 85). 24 De Vries (1962) ‘sanna’, Holthausen (1963) ‘soð’, Robert and Kay (1995: vol. 1, pp. 372–73), Liebermann (1903: vol. 2), ‘soð’, ‘soðfæst’, all quoted in Koziol (2007: 85). 25 Koziol (2007: 85). 26 Idem. 27 Koziol (2007: 86, n. 39). See also Wormald (1999: 122, 280, 283, 316). 28 Mehtonen (1996: 16). 29 Lake (forthcoming). See also Mehtonen (1996: 16). 30 Taranu and O’Connor (forthcoming). 31 Ray (1980: 11). Goffart (2005: 114). Taranu and O’Connor (forthcoming). 32 Ray (1980: 18). 33 Goffart (2005: 113). 34 Goffart (2005: 116). 35 Ray (1980: 17–18) for oral informants and page 12 for written sources. On the role of orality in Bede’s work, see Cubitt (2006: 209–11). 36 Bede (1969:19, p. 219). 37 Idem. 38 For the groundbreaking and compelling argument against usual assumptions of how representative or well-known Augustinian psychology was for the vast majority of people in early medieval England or Europe, which entertained very different conceptualizations regarding the soul, mind, and body, see Lockett (2011). For the classical argument that Beowulf probably originated in an aristocratic monastic foundation (Eigenkirche), see Wormald (1999). 39 Cubitt (2006: 221). 40 Hall (2008: 285). I am deeply grateful to Alaric for his thoughtful comments on this chapter in particular. 41 The best survey of the historiography on the poem’s essential unity remains Shippey (1998). 42 Liuzza (2005: 105). 43 Pasternack (2006: 531). 44 Pasternack (2006: 525). 45 Pasternack (2006); O’Brien O’Keeffe (2006); Amodio (2004); Bredehoft (2014). 46 Bredehoft (2014: 29). 47 Jesch (2006: 264). 48 Parks (1987: 47 and 51), quoted in Jesch (2006: 259). 49 Tonkin (1990: 60). 50 Idem. 51 Tonkin (1990: 87). 52 Davis (2007: 34). O’Brien O’Keeffe (2006: 20–1). 53 O’Brien O’Keeffe (2006: 20–1). 54 Leyerle (1967). 55 Taranu (2015). 56 Nolan and Bloomfield (1980: 510).

‘Truth Is the Trickiest’  121 57 For an insightful comment on rihte as it appears here and in Hrothgar ‘reading’ the magic sword hilt, see Lerer (2006). 58 Davis (2007: 32). 59 Koziol (2007: 85). 60 Scheil (2008: 287). Frank (1991: 98, 100–1); Benson (1970: 43). 61 The Exeter Anthology (2000). 62 Williamson (2011: 178–81). 63 For the origin of the concepts of Gemeinschaft (‘community’ based on personal social interactions) versus Gesellschaft (‘society’ constructed via indirect interactions, impersonal roles, formal values), see Tönnies (1955). For a recent appraisal of the concepts, see Hardt and Weeks (2000: 145). 64 Williamson (2011: 179). Greenfield and Evert (1975: 342). 65 Williamson (2011: 181). 66 Williamson (1977: 25), quoted in Shippey (2017: xxvii). 67 Williamson (2011: 179–80). 68 As quoted, punctuated, translated, and interpreted in Thornbury (2016: 101). 69 Idem. 70 Idem. 71 Idem. 72 The latest edition is The Old English Dialogues of Solomon and Saturn (2009). See also Hansen (1988). 73 Dumitrescu (2017). 74 O’Brien O’Keeffe (2006: 54). See also Waugh (1995: 218); Hermann (1989: 36). 75 Trilling (2009: 31). The origin of the image of the constellation is Benjamin (1998). 76 Pasternack (2006: 522). O’Brien O’Keeffe (2006: 193). 77 Tyler and Balzaretti (2006: 2). 78 Lapidge (2001: 68). The discrepancies between the two accounts of the swimming match are discussed by Robinson (1995: 86–92). 79 See, for instance, Downes (1995: 130). 80 Lapidge (2001: 69). 81 Gwara (2008: 2). 82 Gwara (2008: 129 and 131). 83 Enright (1998: 310), quoted in Gwara (2008: 129). 84 Gwara (2008: 13). 85 Koziol (2007: 84). 86 Niles (2007: 282). 87 Idem. 88 For different interpretations of these issues, see O’Brien O’Keeffe (1998) and Hyams (1981). 89 Tonkin (1990: 114). 90 Lerer (2006: 589). 91 Lerer (2006); Lapidge (2001). 92 Lapidge (2001). 93 Lerer (2006: 608). 94 Lerer (2006: 617). 95 Idem. 96 Lapidge (2001). 97 Idem. 98 Koziol (2007: 86). 99 See Taranu (forthcoming).

122  ‘Truth Is the Trickiest’

Bibliography Primary Sources ASC = The Anglo-Saxon Chronicle: A Collaborative Edition, vol. 7: MS E, ed. by Susan Irvine (Cambridge: D.S. Brewer, 2004). Bede, HE = The Ecclesiastical History of the English People, ed. by Bertram Colgrave and R.A.B. Mynors (Oxford: Oxford University Press, 1969). Liebermann, Felix, ed., Die Gesetze der Angelsachsen, 2 vols (Halle: Max Niemeyer, 1903). Lockett, Leslie, Anglo-Saxon Psychologies in the Vernacular and Latin Traditions (Toronto: University of Toronto Press, 2011). Robert, Jane, and Christian Kay, with Lynne Grundy, A Thesaurus of Old English, 2 vols (London: King’s College London, Centre for Late Antique and Medieval Studies, 1995). The Exeter Anthology of Old English Poetry: An Edition of Exeter Dean and Chapter MS 3501, ed. by Bernard J. Muir, 2 vols (Exeter: University of Exeter Press, 2000). The Old English Dialogues of Solomon and Saturn ed. and trans. by Daniel Anlezark (Cambridge: D.S. Brewer, 2009).

Secondary Sources Amodio, Mark, Writing the Oral Tradition: Oral Poetics and Literate Culture in Medieval England (Notre Dame, IN: University of Notre Dame Press, 2004). Benjamin, Walter, The Origin of German Tragic Drama, trans. by John Osborne (London: Verso, 1998). Benson, Larry, ‘The Originality of Beowulf’, in The Interpretation of Narrative: Theory and Practice, ed. by Morton W. Bloomfield (Cambridge: Harvard University Press, 1970), pp. 1–43. Bredehoft, Thomas A., The Visible Text: Textual Production and Reproduction from Beowulf to Maus (Oxford: Oxford University Press, 2014). Capps, John, ‘The Pragmatic Theory of Truth’, The Stanford Encyclopedia of Philosophy (Summer 2019 Edition), ed. by Edward N. Zalta (2019). [accessed 23 July 2020]. Cubitt, Catherine, ‘Folklore and Historiography: Oral Stories and the Writing of Anglo-Saxon History’, in Narrative and History in the Early Medieval West, ed. by Elizabeth M. Tyler and Ross Balzaretti (Turnhout: Brepols, 2006), pp. 189–221. David, Marian, ‘The Correspondence Theory of Truth’, in The Stanford Encyclopedia of Philosophy (Fall 2016 Edition), ed. by Edward N. Zalta, (2016). [accessed 23 July 2020]. Davis, Craig R., ‘Cultural Assimilation in the Anglo-Saxon Royal Genealogies’, Anglo-Saxon England, 21 (1992), 23–36. Davis, Craig R., ‘Theories of History in Traditional Plots’, in Myth in Early Northwest Europe, ed. by Stephen Glosecki (Tempe: ACMRS and Brepols, 2007), pp. 31–45.

‘Truth Is the Trickiest’  123 de Vries, Jan, Altnordisches etymologisches Wörterbuch (Leiden: Brill, 1962). Downes, Jeremy, ‘Or(e)ality: The Nature of Truth in Oral Settings’, in Oral Tradition in the Middle Ages, ed. by W.F.H. Nicolaisen (Binghamton: Medieval & Renaissance Texts & Studies, 1995), pp. 129–44. Dumitrescu, Irina A. ‘Solomon and Saturn’, in The Encyclopedia of Medieval Literature in Britain (Oxford: Wiley-Blackwell, 2017) [accessed 20 May 2020]. Dumville, David, ‘The Anglian Collection of Royal Genealogies and Regnal Lists’, Anglo-Saxon England, 5 (1976), 23–50. Dumville, David, ‘Kingship, Genealogies, and Regnal Lists’, in Early Medieval Kingship, ed. by Peter H. Sawyer and Ian N. Wood (Leeds: University of Leeds, 1977), pp. 72–104. Enright, Michael J., ‘The Warband Context of the Unferth Episode’, Speculum, 73 (1998), 297–337. Foucault, Michel, ‘Truth and Power’, in Power/Knowledge: Selected Interviews and Other Writings by Michel Foucault, 1972–1977, ed. by C. Gordon (Brighton: Havester, 1980), pp. 109–33. Frank, Roberta, ‘Germanic Legend in Old English Literature’, in The Cambridge Companion to Old English Literature, ed. by Malcolm Godden and Michael Lapidge (Cambridge: Cambridge University Press, 1991), pp. 88–106. Green, Dennis H., The Carolingian Lord: Semantic Studies on Four Old High: German Words: Balder, Frô, Truhtin, Hêrro (Cambridge: Cambridge University Press, 1965). Greenfield, Stanley B., and Richard Evert, ‘Maxims II: Gnome and Poem’, in Anglo-Saxon Poetry: Essays in Appreciation for John C. McGalliard, ed. by Lewis E. Nicholson and Dolores Warwick Frese (Notre Dame: University of Notre Dame Press, 1975), pp. 337–54. Goffart, Walter, ‘Bede’s uera lex historiae Explained’, Anglo-Saxon England, 34 (2005), 111–16. Gwara, Scott, Heroic Identity in the World of Beowulf (Leiden: Brill, 2008). Hall, Alaric, ‘The Orality of a Silent Age: The Place of Orality in Medieval Studies’, in Methods and the Medievalist: Current Approaches in Medieval Studies, ed. by Marko Lamberg et al (Newcastle upon Tyne: Cambridge Scholars Publishing, 2008), pp. 270–90. Hansen, Elaine Tuttle, The Solomon Complex: Reading Wisdom in Old English Poetry (Toronto: University of Toronto Press, 1988). Hardt, Michael, and Kathi Weeks, eds, The Jameson Reader (New York: WileyBlackwell, 2000). Harris, Stephen, Race and Ethnicity in Anglo-Saxon Literature (London: Routledge, 2003). Henrich, Joseph, Steven J. Heine and Ara Norenzayan. ‘The Weirdest People in the World?’, Behavioral and Brain Sciences, 33 (2010), 61–83. Hermann, John P., Allegories of War: Language and Violence in Old English Poetry (Ann Arbor: The University of Michigan Press, 1989). Holthausen, Ferdinand, Altenglisches etymologisches Wörterbuch (Heidelberg: Carl Winter, 1963). Hyams, Paul, Trial by Ordeal: The Key to Proof in the Early Common Law (Chapel Hill: University of North Carolina Press, 1981).

124  ‘Truth Is the Trickiest’ Jesch, Judith, ‘The ‘Meaning of the Narrative Moment’: Poets and History in the Late Viking Age’, in Narrative and History in the Early Medieval West, ed. by Elizabeth M. Tyler and Ross Balzaretti (Turnhout: Brepols, 2006), pp. 251–65. Justice, Steven, ‘Did the Middle Ages Believe in Their Miracles?’, Representations, 103 (2008), 1–29. Kempshall, M.S., Rhetoric and the Writing of History: 400–1500 (Manchester: Manchester University Press, 2011). Koziol, Geoffrey, ‘Truth and Its Consequences: Why Carolingianists Don’t Speak of Myth’, in Myth in Early Northwest Europe, ed. by Stephen Glosecki (Tempe: ACMRS and Brepols, 2007), pp. 71–103. Lake, Justin, ‘Narratio Probabilis in Early Medieval Historiography: A Reconsideration’, in Vera Lex Historiae?: Constructions of Truth in Medieval Historical Narrative, ed. by Catalin Taranu and Michael Kelly (New York: Punctum & Gracchi Books, forthcoming). Lapidge, Michael, ‘Beowulf and Perception’, PBA, 111 (2001), 61–97. Lerer, Seth, ‘Hrothgar’s Hilt and the Reader in Beowulf’, in The Postmodern Beowulf: A Critical Casebook, ed. by Eileen A. Joy and Mary K. Ramsey (Morgantown: West Virginia University Press, 2006), pp. 587–628. Leyerle, John, ‘The Interlace Structure of Beowulf’, University of Toronto Quarterly, 37 (1967), 1–17. Liuzza, Roy M., ‘Beowulf: Monuments, Memory, History’, in Readings in Medieval Texts, ed. by David Johnson and Elaine Treharne (Oxford: Oxford University Press, 2005), pp. 91–108. Mehtonen, Päivi, Old Concepts and New Poetics: Historia, Argumentum, and Fabula in the Twelfth and Early Thirteenth-Century Latin Poetics of Fiction (Helsinki: Societas Scientiarum Fennica, 1996). Niles, John D., ‘Myth and History’, in A Beowulf Handbook, ed. by Robert E. Bjork and John D. Niles (Lincoln: University of Nebraska Press, 1998), pp. 213–33. Niles, John D., ‘True Stories and Other Lies’, reprinted in John D. Niles, Old English Heroic Poems and the Social Life of Text (Turnhout: Brepols, 2007), pp. 279–307. Nolan, Barbara, and Morton Bloomfield, ‘Beotword, Gilpcwidas, and the Gilphlædan Scop of Beowulf’, JEGP, 79 (1980), 499–516. O’Brien O’Keeffe, Katherine, ‘Body and Law in Late Anglo-Saxon England’, Anglo-Saxon England, 27 (1998), 209–32. O’Brien O’Keeffe, Visible Song: Transitional Literacy in Old English Verse (Cambridge: Cambridge University Press, 2006). Otter, Monika, Inventiones: Fiction and Referentiality in Twelfth-Century English Historical Writing (Chapel Hill: University of North Carolina Press, 1996). Parks, Ward, ‘The Traditional Narrator and the “I Heard” Formula in Old English Poetry’, Anglo-Saxon England, 16 (1987), 45–66. Partner, Nancy, Serious Entertainments: The Writing of History in TwelfthCentury England (Chicago: University of Chicago Press, 1977). Pasternack, Carol Braun, ‘The Textuality of Old English Poetry’, in The Postmodern Beowulf: A Critical Casebook, ed. by Eileen A. Joy and Mary K. Ramsey (Morgantown: West Virginia University Press, 2006), pp. 519–46.

‘Truth Is the Trickiest’  125 Pedersen, Nikolaj Jang Lee Lindingand Cory Wright, ‘Pluralist Theories of Truth’, in The Stanford Encyclopedia of Philosophy (Winter 2018 Edition), ed. by Edward N. Zalta, (2018). [accessed 23 July 2020]. Ray, Roger, ‘Bede’s Vera Lex Historiae’, Speculum, 55 (1980), 1–21. Robinson, Fred C., ‘Elements of the Marvellous in the Characterization of Beowulf: A Reconsideration of the Textual Evidence’, in Beowulf: Basic Readings, ed. by Peter S. Baker (New York and London: Garland, 1995 [1974]), pp. 79–96. Scheil, Andrew, ‘The Historiographic Dimensions of Beowulf’, JEGP, 107 (2008), 281–302. Shippey, Tom, ‘Structure and Unity’, in A Beowulf Handbook, eds. Robert E. Bjork and John D. Niles (Lincoln: University of Nebraska Press, 1998), pp. 149–74. Shippey, Tom, ‘Introduction’, in The Complete Old English Poems, translated by Craig Williamson (Philadelphia: University of Pennsylvania Press, 2017), pp. xv–li. Stoljar, Daniel, and Nic Damnjanovic, ‘The Deflationary Theory of Truth’, in The Stanford Encyclopedia of Philosophy (Fall 2014 ed.), ed. by Edward N. Zalta, (2014). [accessed 23 July 2020]. Taranu, Catalin, ‘Goths, Geatas, Gaut: The Invention of an Anglo-Saxon Tradition’, in Transforming the Early Medieval World: Studies in Honour of Ian N. Wood, ed. by Kivilcim Yavuz and Ricky Broome (Leeds: Kismet Press, forthcoming). Taranu, Catalin, ‘Who Was the Original Dragon-slayer of the Nibelung Cycle?’, Viator, 46 (2015), 23–40. Taranu, Catalin and Ralph O’Connor, ‘Introduction’, in Vera Lex Historiae?: Constructions of Truth in Medieval Historical Narrative, ed. by Catalin Taranu and Michael Kelly (New York: Punctum & Gracchi Books, forthcoming). Thornbury, Emily, Becoming a Poet in Anglo-Saxon England (Cambridge: Cambridge University Press, 2016). Tonkin, Elizabeth, Narrating our Pasts: The Social Construction of Oral History (Cambridge: Cambridge University Press, 1990). Tönnies, Ferdinand, Community and Association, trans. by Charles P. Loomis (London: Routledge and Kegan Paul, 1955). Trilling, Renée, The Aesthetics of Nostalgia: Historical Representation in Old English Verse (Toronto: Toronto University Press, 2009). Tyler, Elizabeth M., ‘Poetics and the Past: Making History with Old English Poetry’, in Narrative and History in the Early Medieval West, ed. by Elizabeth M. Tyler and Ross Balzaretti (Turnhout: Brepols, 2006), pp. 225–50. Tyler, Elizabeth M., England in Europe: English Royal Women and Literary Patronage, c.1000–c.1150 (Toronto: University of Toronto Press, 2017). Tyler, Elizabeth M., and Ross Balzaretti, ‘Introduction’, in Narrative and History in the Early Medieval West, ed. by Elizabeth M. Tyler and Ross Balzaretti (Turnhout: Brepols, 2006), pp. 1–9. Waugh, Robin, ‘Competitive Narrators in the Homecoming Scene of Beowulf’, The Journal of Narrative Technique, 25 (1995), 202–22. Williamson, Craig, The Old English Riddles of the Exeter Book (Chapel Hill: University of North Carolina Press, 1977).

126  ‘Truth Is the Trickiest’ Williamson, Craig, Beowulf and Other Old English Poems (Philadelphia: University of Pennsylvania Press, 2011). Wormald, Patrick, The Making of English Law: King Alfred to the Twelfth Century (Oxford: Blackwell, 1999). Young, James O., ‘The Coherence Theory of Truth’, in The Stanford Encyclopedia of Philosophy (Fall 2018 Edition), ed. by Edward N. Zalta, (2018). [accessed 23 July 2020].

4 Why Is There Something Rather Than Nothing? The Social Logic of Frankish Verse Histories

The previous three chapters have opened up a theoretical space for the different modes of historical representation (and, underlying them, of historiographical thinking) in early medieval England by exploring a broad range of mostly OE sources from glosses to heroic poetry, revealing vernacular theories of constructing truth and narrating the past that have hitherto lain implicit in the larger web of meaning encompassing different early medieval subcultures. In the fourth and fifth chapters, I explore these texts as social acts and as cultural products fulfilling important ideological but also emotional work for specific communities in the societies in which they functioned. The present chapter is dedicated to reading continental GHP against the background of socio-cultural and political changes and anxieties characterizing the Carolingian and post-Carolingian eras in Frankish society. In this, it provides a necessary counterweight and an instructive complementary case study to how OE GHP functioned socio-emotionally in early medieval England. The Frankish material can be much more safely connected to specific moments in the social and political evolutions of the ninth century. By contrast, the English material, as we have seen in Chapter 1, is characterized by a remarkable formal stability which results in much greater difficulties when it comes to anchoring OE texts to specific socio-political contexts – the obvious exception being the neo-bardic Anglo-Saxon Chronicle verse. As scholars of OE verse often remark, by this very lack of historical mooring and their obsession with the past, such texts ‘generate [their] own historical context’ – ‘“Anglo-Saxon England” as a distinct socio-political culture… is a concept generated in important ways by Beowulf itself’.1 In these circumstances – as I have argued in Chapter 1 – it has often been hard to resist the temptation to treat the corpus of OE GHP as one ahistorical construction virtually consubstantial with the oral verse tradition it purports to represent. Vernacular Carolingian textual productions, meanwhile, usually contain detailed explanations of why they were written and what the authors

128  Why Is There Something Rather Than Nothing? meant to do with them, or at least, they can be pinpointed to specific locales and slices of time, thus lending themselves to readings dominated by the political-ideological work they were put to, which may obscure the larger socio-cultural and emotional vectors they responded to. Thus the final two chapters illuminate each other in that my analysis of Frankish GHP enables me to look at similar potential points of socio-political anchorage for the English material, while my exploration of Beowulf as a point of focus for the emotions of different audiences over time allows me to see the continental material outside of the immediate political needs it may have served. This chapter, then, focusses on a corpus of texts rarely placed side by side with the OE and Old Norse heroic material, though it is much closer geographically and temporally to the former. In its broadly socio-cultural and emotional approach, this chapter will provide a hitherto unattempted account of Carolingian vernacular or vernacular-derived sources (particularly the Waltharius and the Hildebrandslied) considered together as different manifestations of a poetic-historical sensibility that is distinct from the mainstream of Frankish Latinate historiography in its ideology, theoretical underpinnings, and socio-cultural functions.

Introduction Sometimes we discover the most curious things by asking the simplest of questions. There is now a large amount of scholarship on the Waltharius, the Hildebrandslied, and other examples of the earliest samples of what scholars usually call ‘Germanic heroic poetry’. Many important questions have been asked and sometimes answered (albeit conflictingly at times), advancing our knowledge of these sources and the society in which they were produced. Ingenious arguments have been put forward for a large number of ideas: these texts, it has been argued, reflect the evolution of the ‘heroic institution’ of lordship, embody the ideals of Carolingian masculinity, prove the survival of pagan belief or at least the influence of vernacular oral lays, are basically Carolingian propaganda, or are, in fact, pious sermons on the dangers of avarice and violence, and the list could go on for a long time.2 Yet perhaps we have not asked the simplest questions: why was there GHP in Carolingian Francia in the first place? Why does it emerge in writing for the first time in this space and at this time rather than in early medieval England or Scandinavia where this type of texts also flourished later? Even if we are to accept the unbroken oral transmission of heroic poetry since the Migration Age (itself a problematic view), there is nothing natural or inevitable in it being set down in writing in these specific historical circumstances.3 On the contrary, as I argue in the following, the social, political, and cultural context in the early Carolingian period was anything but welcoming of the discourse these sources embody. By

Why Is There Something Rather Than Nothing?  129 official accounts, the Waltharius or the Hildebrandslied should not have existed in the pious, Latinate atmosphere of the reigns of Charlemagne and his son Louis the Pious – nor should OE heroic verse have existed if it had been up to Bede or Alcuin. Hence these naïve but irreducible questions: why did someone go to the trouble of composing or writing them down? Why did someone feel that these texts needed to exist? What emotional want did this type of discourse fulfil? What cultural work did it accomplish?4 As this chapter and the next one aim to show, these simplest of questions can be very productive when it comes to understanding not only the political contexts out of which these texts grew but also the cultural work they did in the societies of early medieval England and Francia, for works of literature are often the answers to a question, pearls ‘formed around some irritating speck of cultural or philosophical grit’.5 This chapter attempts to see the questions which provoked answers like the Waltharius and the Hildebrandslied. In other words, I attempt to reconstruct the socio-cultural logic of these mostly vernacular and heroic (i.e., having to do with the martial prowess of male protagonists) sources to which I add the Muspilli, the Heliand, and Otfrid of Weissenburg’s Evangelienbuch, all of them emerging in writing in ninth-century Carolingian Francia. The Waltharius is, of course, written in Latin, and inspired by epic authors such as Virgil and Statius, but previous scholarship has proven that its author and audiences must have been not merely familiar but also fully conversant in both a Germanic vernacular language and in some form of oral poetic tradition, as attested by some of its puns, allusions, and pastiches of heroic tropes.6 As such, I hold that the Waltharius can well be called ‘Germanic’, as well as Latinate; it is an inheritor of both traditions, and it would be difficult to understand exactly what it does culturally and politically if one were to ignore any of its two roots. All of these texts are linked not simply by genre: indeed, the genres in which they were rooted were vastly different (apocryphal and vernacular apocalyptic traditions, Vergilian epic, oral poems, Gospel concordances), while the genres to which they belong were invented in the very making of these poems. Rather, I consider all these sources together because they gave shape to and were shaped by the same historical socio-cultural concerns specific to ninth-century Carolingian Francia (including the post-imperial rival polities emerging with the civil wars of the second third of the century). Thus the meanings of such works ‘can only be ascertained by reference to the total cultural system in which [they] occur’7. Hence I approach these texts as articulations of and responses to several phenomena: first, the linguistic alienation of elite Germanic speakers from the Latinate environment of the royal court (at least before the civil wars starting in the 830s) and the western part of the empire, where virtually everyone spoke one form or another of Latin.8 I then move to reading them as expressions of an emerging ethnic identity accompanied

130  Why Is There Something Rather Than Nothing? by strong social and political overtones encouraged by Louis the German among his nobles in the 850s and 860s (but possibly starting earlier). Finally, I look at how the texts under scrutiny problematize masculinity, heroism, and nobility, the definitions of which were subject to different kinds of tensions throughout the Carolingian period: earlier, in relation to a pious, almost monastic Christian ideal that became the norm among Frankish aristocracy, and later, in the context of the ever-shifting and conflicting loyalties of elite Carolingian families during the civil wars starting in the 830s. Along the way, I point to more theoretical ways of conceptualizing these forces to which Carolingian heroic poetry gave voice. My approach will be shown to provide a way out of several impasses in the scholarship on these texts. Echoing Clare Lees’s complaint about the ‘endless project of identifying the precise blend of Christian and Germanic cultures offered in a poem like Beowulf’, the scholarship on the Waltharius, for example, can also be said to be dominated by attempts to divine whether the text is bona-fide heroic or a critique of heroism, whether and to what extent it is ‘Germanic’ or ‘Vergilian’ or ‘Christian’.9 The problem with this long-established modus operandi of scholarly analysis is that it concentrates on the identification of sources or on the critical interpretation of individual texts. Fragments of the sources are quarried for quotes that strengthen a particular argument, while the social context is more or less taken for granted. Thus the main questions which these Carolingian texts are made to answer are limited to where exactly they are situated on a series of spectra between appraisal and critique of heroism, between ‘Germanicity’ and Latinity, between Christian piety and pagan survival/ Germanic heroism, between literacy and orality. While these cultural tensions were real and previous scholarship has done plenty to uncover them, the present chapter sees these vernacular and/or heroic sources as providing ludic spaces in which these anxieties could be explored, problematized, and interrogated rather than simply criticized or appraised – as can be seen in contemporary discourses with more propagandistic purposes (mainly voiced by historiographical and hagiographic sources).

Carolingian Germanic Poetry – A Marginal Constellation It is all the more fitting that all these texts should be considered together since another thing they have in common is that they articulated marginal discourses in their society which resisted (or at least reacted, in some form or another, to) the dominant discourses of the Frankish court during Charlemagne and Louis the Pious. They provided ‘a broad collective response to changes that affected a complex society during a period of major transformations’.10 Hence, when attempting to reveal what cultural work these texts were doing in the society to which they were important enough to be written down, ‘we should not speak of this work as being done by any one individual poem, but rather by its discourse, taking

Why Is There Something Rather Than Nothing?  131 that discourse as the sum total of poetic impulses of this kind, whether voiced aloud or written down’.11 Their very materiality testifies to their marginality: these texts are written on the spare front and back leaves of a manuscript (like the Hildebrandslied), ‘scrawled untidily’ in the margins and blank pages of an imposing Latin manuscript dedicated to Louis the German (the Muspilli), probably written down by someone with a form of dyslexia (the Georgslied, a poetic life of St. George), written upside down at the bottom of a page (the Lorsch Bee Charm) or conceivably written down by a Romance-speaking scribe, although the text itself is in a Germanic dialect (the Ludwigslied); generally, the Germanic vernacular ‘emerges timidly as interlinear or occasional glosses, much smaller than the imposing Caroline script of Latin’.12 Even the Waltharius, although found in a high-quality manuscript that seems to have been the commission of someone in a position of power, is the only surviving copy of the text, suggesting that while it did not teeter on the margins, it was in no way a representative of the dominant discourse – it conceivably was produced for the enjoyment of a coterie of intellectuals who would have been equally conversant both in late antique Latin epic poetry and the conventions of what would later be identified as GHP. Thus even in a physical sense, these texts are ‘sites where cultural issues of great magnitude and complexity are contested’.13 These manuscripts are a battleground for two clashing cultural impulses: on the one hand, the preservation and appropriation of an orally derived poetic tradition that was seen as inferior and even dangerous for the soul, and on the other hand, the internal (and certainly external) censure of this type of discourse in a devoutly Christian, literate, Latinate culture and society. We still have to explain why people went to the trouble and expense of committing to parchment what was, by strict devotional standards, a ‘useless’, ‘secular’ poem like the Hildebrandslied.14 To give a foretaste of my socially-situated argument, one explanation of the change of attitudes that allowed for this discourse to emerge is that ‘social groups most affected by changes in status tend to be most conscious of alternative modes of discursive behaviour’.15 One such social group to which I refer further down is the eastern Frankish lower aristocracy that, on the one hand, conceivably felt alienated linguistically in the Latin-speaking empire of Charlemagne, and on the other was encouraged to form an individual warrior-like and ‘German’ (not Germanic but theodisc, as will later be seen) identity during the reign of Louis the German. Yet in order to gain a sense of the effort it took for these texts to emerge against the grain of the Carolingian official discourse, we should have a clear representation of the ideological baseline in Charlemagne’s empire. As Alcuin and his contemporaries warn, neither clerics nor lay aristocrats were supposed to read the kind of texts that are here under scrutiny. Rather, this is the kind of literature that was fit for Carolingian kings and elites:

132  Why Is There Something Rather Than Nothing? Shortly before his death in 855, the Emperor Lothar I, Charlemagne’s grandson (…) commissioned a liturgical compendium for use on his travels, containing the readings for mass all year round, each accompanied by its own explanatory homily (expositio et omiliaticus sermo). The homilies were to be read aloud to the emperor during meals, to sustain his homo interior with the infinite riches of spiritual food while he sat down at the imperial table.16 This is often construed (or simply assumed) to have been the spiritual fare of the large majority of the aristocracy. Yet fractures in this dominant discourse will be brought to light in the following, and part of this chapter is a critique of past scholarship that was complicit in making the centre invisible, in Russell Ferguson’s expression. What he argues about modern cultures can be enlightening for understanding the dynamics of centre and periphery in the Carolingian one: The dominant discourse tries never to speak its own name. Its authority is based on absence. The absence is not just that of the various groups classified as ‘other’ although members of these groups are routinely denied power. It is also the lack of any overt acknowledgement of the specificity of the dominant culture which is simply assumed to be the all-encompassing norm. This is the basis of its power.17 In the case of Carolingian discourses, the main voices of the dominant discourse (figures like Alcuin and Einhard) are no more responsible than modern scholars for obscuring the centre by assuming this represented the Carolingian culture, while peripheral discourses did exist, as shall be seen in the following. Yet it is true that the dominant discourses (represented by ‘mirrors of morality’, such as Alcuin’s hugely successful De virtutibus et vitiis) were those of correctio and piety.18 Carolingian elites were (or represented themselves as being) deeply preoccupied with morality and proper behaviour, while various attempts at religious reform strived to align the laity to monastic ideals centred on self-control.19 Even on a less straightforwardly religious level, Peter Godman presents the full picture of a sophisticated, highly allusive, elitist courtly literature steeped in the late antique Latinate poetic tradition and available to only a small coterie of nobles at the Carolingian court.20 Rosamond McKitterick, a voice of authority representative of many other Carolingianists, draws the picture of an elite steeped in the ‘habitual use of literate modes in all aspects of everyday life’ which sustained a fervent ‘Carolingian religious sensibility’.21 Looking only at the dominant discourse embodied by these Carolingian intellectuals, this seems indeed to have been the full picture: Ermoldus Nigellus called Charlemagne’s son, Louis the Pious ‘Caesar et abba simul’ – the Carolingian leader was thus seen to be the monarchial

Why Is There Something Rather Than Nothing?  133 abbot of the realm.22 Upon taking control of his father’s empire, Louis surrounded himself with a circle of Aquitanian counsellors, including ecclesiastical reformers such as Theodulf and Jonas of Orleans, Ebbo of Reims, Benedict of Aniane, and introduced ‘a regime of almost Calvinistic morality’.23 This was a representative slice of the spiritual life at the top of Carolingian society. While one of the fundaments of the Carolingian Renaissance was the all-pervasive love and use of classical literature, those at the Carolingian court tended to value it not for its own sake but rather for its usefulness in understanding Scripture.24 Yet even the otherwise stern Alcuin, sending a letter to an abbot asking him to bring relics of saints back with him from Rome, saw no problem in including a quotation from Ovid’s Ars Amatoria – a behaviour conceivably representative of the social group he was part of.25 Still, as Ermanrich wrote to Grimald, studying Virgil was like looking for gold in a pile of excrement: ‘As dung gets the fields ready to bring forth better crops, so the works of pagan poets are like stinking manure. Although they are not true, they greatly aid in understanding divine eloquence’.26 If this was a Carolingian intellectual’s opinion of a form of literature they revered, it is not hard to imagine what they must have thought about vernacular songs. We actually know what they thought from the same Alcuin: vernacular songs were not only unacceptable in monastic spaces but also unsuitable for the enjoyment of Christians.27 Indeed, the vernacular language itself was ‘redolent of the stench of dung and the sweat of the warrior’, the contrast between the rude, uncultivated vernacular with its popular oral tradition of pagan songs and folktales and the Latin language and literature of learned Christianity being an often-used topos of Christian Latin writers of the period. 28 Yet this very insistence on the dangers of vernacular songs shows that they must have been popular enough to become a spiritual threat. And since ‘those matters that the actors of an age attempt to hide from one another [or from themselves] may be the most important ones of all’,29 this silencing of traditional poetic modes points the way to another aspect of Carolingian elite society that is acknowledged to a far lesser extent by scholarship. The great reform synods of 813 at Mainz, Rheims, Chalonsur-Saône, Tours, and Arles all condemn in the harshest terms the lay and ecclesiastical aristocrats who enjoy histrionum quoque turpium et obscenorum insolentias iocorum.30 All four canons seem to refer to the same cultural phenomenon, thus attesting its existence in all corners of the regnum Francorum, both in Germanic- and Romance-speaking parts. As Michael Richter shows, they are neither the first nor the last admonitions of their kind, which proves that their subject of condemnation was widespread and difficult to stamp out.31 That there was a tradition of oral performance that pervaded all spheres of Carolingian society has been thoroughly and irrefutably argued by Richter, and there is no need to

134  Why Is There Something Rather Than Nothing? rehearse all his extensive evidence – it is surprisingly wide-ranging, going from Alcuin’s complaints about one of Charlemagne’s favourite courtiers, Angilbert, nicknamed Homer for his skill in performing oral poetry to abbots being very much taken with citharisti (harp players).32 I will come back to this thriving scene of elite oral performance (that admittedly need not have been in any way ‘Germanic’ or ‘heroic’ – this assumption is what makes Richter’s thorough survey problematic). For now, let this be a reminder that Carolingian society was far from being equivalent to the pious image that most official sources are constructing. It is not that Carolingian court culture was (…) a fraud, but neither was it quite what it pretended to be, at least not at its core, for its energetic and pulsating heart was a sovereign who was not fully literate and yet around him gathered a phalanx of client-scholars who carefully constructed an image of a learned, wise, eloquent and all-knowing king.33 To adapt Dutton’s argument about Charlemagne’s literacy, it is not that the pious Latinate culture which Carolingian luminaries such as Alcuin were promoting was a mere veneer on the surface of a society that was busy doing the exact opposite. Rather, the same people who would be reading homilies trying to better themselves morally did at times indulge in oral performance and in reading or writing texts that were not supposed to have a place in the ideal Carolingian society, much as nowadays respectable members of society go to the opera (not least because of the social prestige associated with it) and still listen to bubblegum pop or hair metal tunes, reminding them of their youth while winding down or with intimate friends.34 Texts like the Hildebrandslied or the Waltharius, although not suitable for the enjoyment of Christians, were, however, clearly felt by certain members of the elite to be important enough to be written down. Yet the modern parallel is, although potentially illuminating, also misleading, since it fails to convey the sense of urgency and the effort needed to go to the trouble of recording such poems (or indeed composing them since, as I explain in the following section, it was not a simple matter of writing down stable, self-contained texts). Here is where an understanding of GHP as a socially-embedded cultural act is crucial to discerning the reasons why the texts under scrutiny did emerge after all. This is not about simply paying closer attention to the sociolinguistic situation in which such texts emerged but also the more radical realization that ‘the representations of the social world themselves are the constituents of social reality’.35 In Gabrielle Spiegel’s words, we should not favour the historical events as more real than the texts since what is real ‘are the semiotic codes that govern the representation of life both in writing and in incorporated social structures’.36 Such an approach considers texts as not only responding to deep cultural tensions and desires extant in a society or community, but also as articulating them.

Why Is There Something Rather Than Nothing?  135 What were the desires and tensions (and their underlying sources) in ninth-century Carolingian society which GHP answered?37 As I shall argue, they can be summarized in the following three constellations of complementary and competing forces:38 (1) How can one reconcile speaking a despised language and partaking in the morally objectionable poetic tradition associated with it while being a member of the regnum Francorum and of an ideal Christian Latinate Carolingian elite? (2) How can one be a Frank and yet speak a different language than that at the court? How can one be a Frank and yet remain a Saxon? From a wider perspective, how can we share the same ethnicity with our neighbours across the Rhine if we do things so differently here? (3) How can one square martial prowess, war, and plunder – all perceived to be necessary for the survival of the state and for the affirmation of one’s identity as a nobleman – with the Christian interdiction to kill or even commit violence? How can one avoid fighting against one’s own kin and yet keep the loyalty one owes to one’s lord? These are the social and emotional forces which the sources I am looking at shaped and gave voice to. It is against this background that in the first half of ninth-century Francia several texts were written down quite unlike anything else that a scribe had consigned to parchment before. The Waltharius (most probably dating to the first half of the ninth century) is a Latin mock-heroic poem whose author had detailed knowledge of the themes and characters of what would later emerge as written vernacular heroic poetry.39 The Hildebrandslied (written down at Fulda, conceivably in the 830s) is an OHG alliterative poem that tells the tragic encounter in battle between father and son caught up in bonds of conflicting loyalty.40 The mid-ninth century Muspilli is a puzzling OHG alliterative poem that narrates the fate of the soul after death and that of the world before the Last Judgement in a manner that seems to draw just as much on apocryphal writings as on non-Christian material that would emerge much later in Scandinavia.41 A bit later, in the second quarter of the ninth century, the Heliand is written down – a 6,000-line alliterative poem in OS that casts the life of Jesus into a cultural matrix for which the labels ‘Germanic’ and ‘heroic’ are not inappropriate. In the following, I attempt to reconstruct the larger context of which these texts were part. At its heart, this chapter is a sociolinguistic study of Carolingian vernacular heroic poetry. Yet the social aspect aims to be about more than class (I also look at masculinity, kinship, loyalty), and the linguistic side is not strictly looking at language and the Germanic dialects (I consider these to be markers of nascent ethnic identities, tightly meshed in with the social aspects). As the sociology of knowledge advocates, the interpretation of discourse needs to look beyond just the content to three contextual settings which contribute to the construal of meaning: 1. the structural context (the way language is used to convey

136  Why Is There Something Rather Than Nothing? meaning); 2. the wider discursive context (what is uttered before and after and in other discourses); 3. the social context (the power relations embodied in and realized by the discourse).42 In what follows, although I mainly make use of the second and third frameworks of interpretation, I also approach the structural context in a section on the politics of orality. While the texts mentioned earlier are my main focus, I will not have my sights on them most of the time. As one is more likely to perceive a faint star or nebula in the night sky if one uses one’s peripheral, rather than direct, vision, so is my chapter looking at the rich life structures in which these poems were embedded with the hope that we will be able to see the texts themselves more clearly.43 As can be already seen, I use a wide variety of sources, yet I do not mean to say that the Waltharius, the Heliand, and other sources I am looking at can be interchangeably used to make arguments about the whole socio-political context of a Carolingian ninth century frozen in time. Each of these is, of course, situated in its own political framework which unfortunately cannot always be securely reconstructed, although we can be certain that this framework evolved from one decade to the next, from one year of peace and unity to another one of war between royal pretenders and aristocratic families. The hope that fuels my endeavour is that we can more safely (or at least fruitfully) revive the social logic of the texts, from whence at times we can also glimpse the larger constellations of thought, emotion, gesture, and discourse they were originally part of.

‘The Savage Words of an Uncultivated Language’: Linguistic Community and Alienation I start with the first constellation, rising around the turn of the ninth century under the sign of what I would call ‘linguistic alienation’. Dominant figures among Carolingianists are always keen on stressing the degree to which Latin was the language of the empire par excellence, in which most Carolingian aristocrats thought, wrote, and ordered their political and religious lives.44 McKitterick has famously argued that literacy and Latinate culture and education suffused not only Carolingian aristocratic life but also other social actors in the life of the empire.45 Yet as previously seen, what may have been perfectly true for the centre of Carolingian society was not necessarily so in more marginal textual communities (not necessarily in a spatial sense). For while in the part of the empire situated roughly west of the Rhine, Latin was probably not a foreign or learned second language for the Franks but ‘their native tongue in its regularized and conventionalized written representation’,46 the large majority of the people living in the eastern territories were speaking Germanic vernaculars. In other words, in later Romance-speaking areas of Western Europe,

Why Is There Something Rather Than Nothing?  137 there probably was no difference between spoken and written Latin until the Carolingians created it, ‘not just with their emphasis on correct Latin, but by introducing new, and reviving old, rules for the pronunciation of written Latin. This made Latin virtually unintelligible to those used only to speaking it’.47 Thus seeing written Latin and spoken Romance as two separate languages coexisting before 800 is probably a modern scholarly construct. In this case, if west Frankish aristocrats conceivably grew up in a spoken Latin environment, they only had to adapt to writing and speaking an acrolect (a formal, spoken register and a certain literary, religious, and political vocabulary associated with higher status) of their native language.48 In contrast, east Frankish nobility (perhaps with the exception of families on the very highest rungs of the social ladder) would have grown up speaking their respective Germanic vernaculars, especially since wet nurses who were employed by so many aristocratic families might well be from lower vernacular-speaking social strata.49 While many among the higher aristocracy surely were conversant in several languages, Ernst Hellgardt has emphasized that the Carolingian Empire below the level of the greater aristocracy remained divided fundamentally by language so that even among the elite, excepting those who grew up in mixed linguistic zones, second and third languages had to be learned as foreign languages.50 Since Latin was so central to the Carolingian state and society, whoever wanted to play a part in the political or cultural scene had to know or learn Latin since Latin was the language of secular and sacred power – ‘could one pray to God in any other language?’51 Thus the Carolingian intellectuals, by their very attempt to rethink their world as a universe ordered according to the norma rectitudinis, the law of orderliness and righteousness, not only in religion of morals but also in culture and language, provoked ‘a massive upheaval in the cultural and linguistic spheres, generating conflicts and fluctuations’.52 On the individual level, for Germanic-speaking aristocrats, having to learn a language that was not related in any way to their native one could have conceivably induced feelings of alienation (or at least uneasiness) in at least some Germanicspeaking, mid-level cultural and political actors and/or communities. Of course, this is an attempt at source-based creative speculation, and I do not aim to fill the backdrop of this painting with brooding, tormented Franconian-speaking nobles conspiratorially writing heroic poetry on manuscript flyleaves. Still, I think linguistic alienation is a good concept to consider in this context. In any case, it need not have manifested dramatically – only through variable degrees of a conceivable uneasiness with Latinate court culture and a corresponding openness towards cultural products that would address these emotions, however obliquely. Indeed, emotions, especially when they are shared by a social or textual community, can be a powerful trigger for changes in identities, policies, or mentalities.53 And while it is true that such a language-triggered

138  Why Is There Something Rather Than Nothing? alienation cannot be directly traced in the historiographical or biblical exegetic discourses that are the privileged medium through which historians commonly attempt to understand Carolingian society, this linguistic disaffection can be discerned obliquely in, for instance, liturgical texts. Ildar Garizpanov has looked at the texts of royal masses in Carolingian sacramentaries, reconstructing what he calls the ‘liturgy of authority’, which represented ‘an ardent quest of its participants, first and foremost, the clerics, to define royal/imperial authority in its relation to God, universal divine order, and Christian believers’.54 The role of lay participants in this liturgy was quite participative when they spoke a vernacular of which Latin was still understood as a formal spoken register (i.e., Romance) and were thus able to grasp the main agenda of a mass. However, this was not the case in Germanic-speaking areas of the Carolingian realm. There the commoners hardly understood a word of the mass, although the priest in a church could have explained the main theme of a particular mass, as was required by some Carolingian capitularies. Thus, the difference in languages alienated some participants in a mass, although such alienation has itself a communicative function. It is not surprising, therefore, to find that in Germanic-speaking areas, especially to the east of the Rhine, the liturgy of royal/imperial authority developed to a lesser extent than in France.55 Such communities would conceivably have felt a similar detachment from other officially-sanctioned forms of cultural, religious, and political discourse because the language they were in could have been perceived as distant and disconnected from both everyday experience and their own cultural horizon, at least by some individuals. Another, more humorous, instance of alienation is discernible in the socalled Paris Conversations, probably dating from the later ninth century, which seem to project the Germanic speaker as the subaltern or at least as the subject of satirical wordplay. On the surface, it is a vocabulary/phrasebook meant to facilitate understanding between speakers of Latin and of Germanic vernacular. However, on closer inspection, the text is revealed to be more of a ‘dirty Germanic phrasebook’ than an actual glossary, a form of applied parody, as well as a means of teaching the essentials of conversational lingua theodisca. It seems to have come from a scholastic and almost certainly monastic milieu since it shows knowledge of a thirdcentury Greek-Latin school manual, the Hermeneumata pseudodositheana, but also because, like the Waltharius, the Conversations imply a monastic disapproval of the earthier aspects of secular life.56 In any case, this source provides a picture of the warrior elite which stands in sharp contrast to representations of laity found elsewhere in Carolingian literature, which usually advocated a fusion of the monastic and the lay.

Why Is There Something Rather Than Nothing?  139 It is hard to say whether the Conversations contain actual interchanges recorded on an actual journey or were intended mainly as parody. As Kershaw adeptly observes, ‘The humour stems not from the failure of communication to cross the language barrier, but rather from the way in which the barrier was breached’.57 The list includes terms for body parts, such as guanbe, ‘venter’ (stomach), but also fóllo guanbe, ‘plenus venter’ (full stomach), and questions such as, ‘Where do you come from?’, or injunctions like ‘give me my sword’, before returning to alimentary issues: ‘I want a drink’ (erro, e guile trenchen, ‘ego volo bibere’).58 The insistence on the earthly needs and appetites would probably have been perceived as humorous, and we get glimpses into more intimate aspects, as this exchange between senior and vasallus shows: ‘Why weren’t you at Matins?’ ‘I didn’t want to go’. ‘You were in bed with your wife’. More realistically useful phrases were provided, too, but the value of communication is subverted into parody, and its limits are made explicit in the phrase: ‘I don’t know this word’ (begot eh ne uitst nen hurt, ‘nullum uerbum scio de hoc’).59 Thus the Paris Conversations correlate linguistic anxiety with satire of warrior culture, voiced by the injunctions of a Latinate master to his Germanic-speaking subaltern. It is this same alienation – although certainly not seen in such a humorous perspective – that led Otfrid to write his Evangelienbuch in the 860s. The Evangelienbuch is a Germanic vernacular paraphrase and synthesis of the four gospels, integrated with homiletic and exegetical material. Unlike the Heliand, an earlier vernacular biblical paraphrase, it is not in alliterative verse but in an original internal rhyme metre. The entire project is very high status, as shall be seen further down. In his letter to Archbishop Liutbert, Otfrid explains his undertaking from the point of view of the Frankish readers who did not know Latin so that whoever is horrified by the difficulty of a foreign language in them (qui in illis alienae linguae difficultatem horrescit) might comprehend the most holy words here in his own language, and understanding the law of God in his own language, might shrink from deviating from it even a little through his own thinking. (20–8)60 Otfrid knows his audience very well, so this contemporary testimony is harder to brush aside than Garizpanov’s conclusions: both imply that some Germanic speakers indeed felt marginalized in a Latinate ecclesiastical culture. The opposite was also true (and probably a crucial factor in creating this alienation): the vernacular was outrightly despised, as we can

140  Why Is There Something Rather Than Nothing? clearly see from the energy which Otfrid expends in the preface to his work to justify his enterprise and excuse the barbarity of the Germanic vernacular. Indeed, the language is always described in unmitigated terms as barbaries, inculta, agrestis and indisciplinabilis.61 The embarrassment of the author is so great that he gives up quoting the vernacular to exemplify the difficulties encountered in his translation so as to avoid ‘the scoffings which the savage words of an uncultivated language will provoke with the erudites confronted with the smooth form of Latinity’.62 If this is the opinion of someone who wanted to make a literary language out of Germanic, one can imagine what the adversaries of such an idea must have thought about the vernacular. Evidently, the alienation was felt on both sides. In order to perceive how extreme these attitudes towards language could become in the Carolingian world, it would be useful to contrast them with the lack of such abasement in roughly contemporary insular sources. Thus in the OE corpus, the vernacular is highly appraised and used for all genres, while the early medieval Irish (self-consciously cheerful) view of the vernacular was that it had been made by God out of all the best bits of other languages.63 Still, as we have seen in Chapter 1, negative views of vernacular culture and especially of vernacular verse were not uncommon in England either, being espoused by elite members of the clergy espousing an ideal of perfect piety such as Bede and Alcuin – the latter of whom is one of the few personalities equally important for Carolingian and early medieval English Orthodox elites. What can happen in this situation of mutual alienation is either that a separate subculture emerges, one that would give voice to this marginalized community, or that the centre attempts to win the margins by framing its hegemonic discourse in a cultural matrix familiar to the latter. In the latter case, ‘the putative center welcomes selective inhabitants of the margin in order better to exclude the margin’.64 Of course, these two outcomes need not be mutually exclusive. In fact, we can see that the latter is Otfrid’s intended strategy. In the same letter to Liutbert, still justifying his undertaking, Otfrid expresses his desire that his vernacular versification of the gospels would replace the ‘obscene lay songs in the vernacular’ (laicorum cantus obscenus) and the ‘voices of secular plays/performances’ (ludum saecularium uocum). In fact, Otfrid merely spells out what earlier generations of Carolingian intellectuals had merely brushed under the carpet with generic interdictions against this obviously vigorous vernacular culture which they might have simply expected to disappear. His avowal is yet another testimony to the importance of secular and/or vernacular songs in eastern Frankish society. Another source that follows this model (although closer to a middle ground between the two outcomes) is the OS Heliand. Otfrid’s strategy was to invent almost from scratch a poetic idiom in which to express the gospel narrative without getting too close to that of the ‘obscene songs’. He exploits the traditional long line but innovates by using rhyme and

Why Is There Something Rather Than Nothing?  141 assonance at the caesura and the end of the line to link both halves.65 By contrast, the Heliand embraces an idiom that is indeed ‘Germanic heroic’, but, as Dennis H Green showed, its author(s) do(es) not use it for its own sake but as part of a larger, subtler strategy.66 Thus the text contains plenty of Germanic oral-formulaic patterns (rightly labelled ‘Germanic’ because some of them are encountered in similar form in OE and Old Norse): Christ as landes uuard (guardian of the land) and thiodo drohtin (lord of the peoples), Virgin Mary as adalcnosles uuif (woman of noble lineage), Herod as boggebo (giver of rings), and folccuning (king of the people). Note also the vernacularization of many episodes, such as the portrayal of the feasts, the omission of the ass from the entry in Jerusalem, and the replacement of the shepherds by ehuscalcos (horseherds).67 Yet the discourse it constructs can be better described as ‘the Christianization of Germanic rather than the usual Germanization of Christianity’.68 Thus in the Heliand, what on the surface seems to be a ‘heroic vocabulary’ is subverted, with traditional poetic terms being used for enemies of Christ and of Christianity (strídiga thioda, warlike band, erlos obarmuoda, proud men; Heliand 3990 ff.), the only heroic conceptualization thoroughly associated with Christ’s disciples being thegnly loyalty: Thomas is drohtines thegan (line 3994). Peter is referred to by heroic terms, such as snel suerthegan (brave sword-warrior, line 4866) and suíðo thrístmód thegan (very brave thegn, line 4870), and gibolgan, angry (line 4871) when he draws his sword in defence of Christ at the moment of the garden of Gethsemane arrest. Peter’s attitude is immediately rebuked, with Jesus condemning anger and rejecting warfare (line 4882 ff., 4895 ff.). The passage in which Peter is at his most ‘Germanic heroic’ (at least he is usually interpreted as such) presents in fact the occasion for the repudiation of this ideal: ‘The very accumulation of heroic terms and epithets renders their subsequent deflation all the more effective’.69 Passivity in the face of violence (Christ enjoins the disciples to be mildi, merciful, in their dealings with others, line 2491), rather than a display of heroic virtues in combat, is what all these passages in the Heliand convey.70 In these cases, a familiar cultural matrix was taken over only to be subtly undermined. Of course, the value judgement inherent in the word ‘undermined’ is ours, the original goal being to instil a deeper Christian morality in its audience. This integration into the dominant discourse by keeping the appearances of marginality is one possible response to feelings of cultural and linguistic alienation. Another one could have been the creation of a discourse that would be perceived as prestigious by the marginalized community, yet it would remain marginal to the cultural horizon of its society. Looking across the channel, at least some individual social actors took the issue into their own hands and solved the conundrum that Garizpanov deciphers in Frankish religious life by reciting the liturgy in the manner of secular poets. They are the early medieval English

142  Why Is There Something Rather Than Nothing? priests entertaining this practice that was perceived significant enough to be banned by the Council of Clovesho in 747.71 These social and political actors are obviously not the prime movers of culture and chief purveyors of the dominant discourse (nor do they need to be very numerous) but neither are they in any way ‘low class’ or socially marginal – they are part of the lower religious and cultural aristocracy. I would suggest that it is a similar social and ethnic grouping – namely, the Germanic-speaking lower aristocracy (both secular and religious), which is the locus for the emergence of vernacular and/or heroic verse in Carolingian society. The reason why I stress the role of the lower aristocracy here is that after the incorporation of their territories during Charlemagne’s reign, the most powerful aristocratic families of Bavarian, Alemannic, and even Saxon origin merged with the Frankish aristocracy to form a ruling elite that transcended borders.72 Meanwhile, the smaller aristocracy (the minores or even mediocres which the Capitulare episcoporum, distinguishes from the fortiores) might have been more likely to see itself as separate ethnically from the cosmopolitan upper rungs of the social ladder and thus to develop a more local (rather than imperial and trans-ethnic) identity.73 For, as I argue in the following section, different layers of Carolingian society constructed different ethnic identities. Guy Halsall has already convincingly shown that the Carolingian elite should not be seen as a single, uniform, and unified body but that there were distinctions between the upper echelons of the nobility and more lowly, locally based families ‘where the higher level of super-aristocrats may have related to the state in a different way from lesser families, though the latter may have become more entrenched in their local power bases’.74 In any case, Michel Banniard has astutely argued that the promotion of Germanic dialects (such as OS of the Heliand or the Franconian of the Evangelienbuch and the Muspilli – mixed with Bavarian in the case of the latter) to the rank of a literary language is due to the efforts of Germanophone intellectuals who, out of rivalry or imitation, responded to similar evolutions in the ranks of the Latinate eastern Frankish elite.75 In fact, Banniard sees the first Romance texts (such as the Sequence of St. Eulalia, written down on the same page as the ‘rithmus teutonicus’ Ludwigslied) as emerging in response to the first Germanic texts being written down, which ‘brought about the desire for an identical consecration for the Romance speech, hitherto excluded from this creative level to the advantage of Latin’.76 The great merit of his article is to show the sociolinguistic aspects of the emergence of the first vernacular ‘literary monuments’ and the complex interactions of the elites on both sides of the Rhine as they try to construct marginal prestigious discourses by creating a language fit for literature. In this context, texts like the Heliand, the Evangelienbuch or (in Old French) the Sequence of St. Eulalia were not destined to be ‘the people’s prayers’ but were part of a self-conscious effort to create a language fit for an elite with a growing sense of local

Why Is There Something Rather Than Nothing?  143 identity. Thus the passage into vernacular literacy was accomplished ‘not by the resigned acceptance of a scripta rustica (basilectal level), but on the contrary by the creation of a literary language (the construction of an acrolect)’.77 Yet this is a relatively late phenomenon, taking place after the middle of the ninth century, and an important factor in these evolutions is the political context (the division of the empire and the rivalry between Charles the Bald’s western kingdom and Louis the German’s eastern polity – more on this further down). There are earlier literary texts in Germanic, however, thus disproving Otfrid’s claims that vernacular poetry was simply non-existent before his attempt – ‘unless we take them only for his specific dialect of Rhenish Franconian, and if we grant that he may have been unaware of the Muspilli’ (which again points to the peripheral position of such attempts).78 How could Otfrid have developed such a bold undertaking in the absence of prior similar attempts and/or of an extant vernacular (oral but also orally derived) literary effervescence? Furthermore, looking at the issue from a socially situated perspective, what statement about language do texts such as the Hildebrandslied, the Wesobrunn Prayer, and the Muspilli make? What is clear is that they stand in stark contrast to the secular literature available to Carolingian aristocrats: Shrouded with learned references to classical literature and encoded not simply in Latin, but in a full-blown, flowery, elliptical, and deliberately clever Latin, these [Carolingian court] poems probably made little impact on anyone other than the poets themselves competing with each other for patronage. Vernacular poetry, on the other hand, might have helped to create a more widely-held identity.79 In these vernacular texts, I would argue, one can see at play an earlier stage of the other outcome of linguistic alienation: marginalized textual communities constructing an alternative to the dominant discourse (not necessarily contradicting it but aiming to provide a better expression of the values and/or anxieties relevant to that community). As the following sections of this chapter will argue in more detail, these values and anxieties (related to masculinity, nobility, ethnic, and linguistic identity) would have conceivably found expression in the thriving oral culture in which both clerical and lay nobles indulged at times.80

Residual Orality and the Politics of Poetic Form The rootedness of the Heliand, the Muspilli, the Hildebrandslied, and even of the Waltharius in orality has been amply demonstrated.81 Yet the emergence of these texts is not simply a matter of recording a ‘text’ circulating orally (since such a ‘text’, moulded in an oral-formulaic framework,

144  Why Is There Something Rather Than Nothing? would have been by its nature mutable and constantly in flux).82 In any case, orality and literacy are not mutually exclusive categories but rather opposite ends of a broad spectrum ‘that extends between two imaginary absolutes, the purely oral and the purely learned, taking “oral” and “learned” to include not just a mode of composition but also such factors as performance, audience, style, and rhetoric’.83 The passage from one to the other can encompass several stages but also different flavours of textualization.84 Thus Ursula Schaefer distinguishes Verschriftung (conceptual textualization, the simple codification of the oral production into a written one) from Verschriftlichung (cultural textualization, which involves a much deeper restructuring of thought and expression from those characteristics of an oral medium to those of a written one).85 Indeed, the texts here under scrutiny display ‘residual orality’ (the persistence of oral modes of thought and oral techniques of composition in texts that are of literary origin).86 They are also somewhere along the spectrum between Schaefer’s categories: while the Hildebrandslied could result from the simple transcription of an oral performance, the Heliand presupposes several steps of reflection characteristic of Verschriftlichung, whereby the theological, the scriptural narrative, and the ‘Germanization’ of picturesque details in the story were integrated so well. It follows that in the composition of at least some of these texts, there existed the option to renounce these oral features altogether. For instance, Otfrid invented a whole new verse form for his Evangelienbuch. Thus the question remains: why did the authors of most Carolingian vernacular poetry choose to preserve (or indeed, construct) this type of orality in their texts? It is necessary to see what this choice correlates to on the social and political level since the form and content of discourse are to be taken equally seriously as carriers of meaning.87 Indeed, even by looking at the Hildebrandslied, one can see that it is not simply a written representation of an oral poem. Rather, the copy we have now (written down probably at Fulda in the 830s) was copied from a text already written down since some words are repeated.88 Metrically, some lines seem impossible, although this might be due to the problematic language: it is not strictly in OHG but probably represents a Bavarian original, badly adapted into Low German.89 The juxtaposition of a wide range of orally derived poetic means with rather intricate theological concepts in the Heliand points to yet another convoluted story of transmission and composition. Not long ago, Harald Haferland argued that the Heliand was probably the result of a team of poets working together. According to his argument, at least some of its authors would have been illiterate and intimately familiar with oral heroic poetry, while other members of the author-team would have been quite sophisticated theologians.90 His argument (which need not be accepted as such) points to a fact that we can easily lose sight of when faced with the artistry with which the Heliand (like the Waltharius) melds two very different and seemingly incongruent poetic and thought

Why Is There Something Rather Than Nothing?  145 traditions. There is nothing natural or inevitable about the way in which such texts responded to and adopted the later Carolingian interest in theology and evangelization and the oral vernacular heroic poetic tradition. Rather, it was a conscious choice and a phenomenon unique to this space and time, and we need to account for the social and cultural forces that shaped it. Hence, when texts like these are marked by residual orality, this need not be an unintended consequence of a simple transition from orality to literacy but a strategy consciously carried out. The vernacular, oralformulaic format of the poem can be seen as a textual strategy and not necessarily as evidence that the poem existed in a broader community: by ‘representing oral production in written discourse, orality becomes a significatum of textuality, an element of written discourse’.91 Thus the poem need not ‘speak with the voice of the community; rather, it appropriates the authority associated with the act of speaking from the community’.92 As such, these texts (fixed and enshrined as ‘poems’ by their being written) mark the appropriation of an orally-derived poetic mode by particular textual communities for their own purposes. From a socially-embedded perspective, this appropriation could be envisaged as being part of an ideological initiative (self-conscious to various degrees) of a part of the Germanic-speaking Carolingian aristocracy that was affected by a growing sense of linguistic and cultural alienation from the Latinate pious royal centre during the reigns of Charlemagne and especially Louis the Pious, then by the internecine wars between various Carolingian royal heirs in the 830s to the 850s putting unprecedented stresses on kinship and loyalty ties, and, finally, by the emergence of an eastern Frankish sense of ethnic and cultural identity under the reign of Louis the German from the 850s onwards.93 Thus ‘these poems might have participated in a rhetorical strategy of calling upon the heroic past represented by the formulaic tradition to create a sense of the “people”, perhaps a sense of nation’.94 It is the same valuation of tradition (even if invented) that subtended other Carolingian practices and discourses. Thus when Louis the German visited Saint Gall, Notker wrote that the king was wearing the austere antiquorum omatus vel paratura Francorum, the ‘dress and equipment of the Old Franks’: a simple tunic, brown linen hose bound with long laces, boots with gold spurs, a sword belt, and a grey or blue cloak.95 As Goldberg argues, although this description of the king might have been an idealization by royal courtiers, it nevertheless reveals the value system and political culture which Louis as a Frankish nobleman would have shared with his elite. Thus, traditional attire (or at least one perceived as such) was one of the strategies through which a strong eastern Frankish masculinist aristocratic identity was constructed around the middle of the ninth century – more on this further below. But apart from the immediate political concerns, employing alliterative verse and poetic oral formulae which belong to a poetic tradition

146  Why Is There Something Rather Than Nothing? which was conceivably archaic by the time it was written down might have marked the creation of a ludic space. In such a space, deep tensions and serious issues within Frankish and English societies could be problematized and explored without the danger of being subjected to the usual moral concerns. In the definition of Mary Carruthers, ludic spaces ‘are not wholly separate worlds ruled by magic, yet they do occupy a privileged and protected space within actual social life wherein all sorts of matters and relationships can be explored’.96 By their being set in a legendary past and in a poetic mode both familiar (from oral performances or stories heard fragmentarily) and new (by its being written down and raised, at least aspirationally, to the level of a literary language), GHP provided a space wherein elite audiences could explore and express identities and psychosocial conundrums. I explore some of these conundrums in what follows. For the time being, a preliminary conclusion on the nature of these texts can be hopefully drawn: they were neither purely oral ‘texts’, preserved unchanged from the times of Tacitus, fished out from an ‘open sea of orality’ straight into the net of Frankish or English textuality (the authors conceived as avant-la-lettre ethnographers), nor were they inventions of scholars for political propaganda, scriptorium-bound fanfictions of Fredegar (more on this view in the following section).97 Rather, they are inventions only in a more etymological sense (in Latin, invenio, invenire, ‘to find, to come upon’ but also ‘to invent, to contrive’): they are things found but at the same time made up, things which a person or a community comes upon, takes over, and reinvents. In this sense, a vernacular heroic (in the case of most texts discussed here) or mythical (as in the Muspilli) story-world and its accompanying verse form is (re)discovered, or at least revaluated – in both acceptations of finding something out and creating something anew – by communities within early medieval Frankish or English elites as they construct and seek to advertise intersecting planes of emerging identity. After having discussed linguistic identity at length (although, of course, not exhaustively), in the following section, I turn to another of these intertwining planes of identity – namely, ethnicity – and to the ways in which heroic and/or vernacular texts helped shape and responded to the emergence of a sense of being ‘German’ (although not necessarily ‘Germanic’).98 At the least, I will argue that there was a space for the manifestation of this type of identity, which sources like the Waltharius or the Hildebrandslied filled or sometimes opened up.

Carolingian Ethnicities (I): Teutoni, Theodisci, or Germani? To talk about a Germanic, let alone a ‘German’ ethnic identity in the ninth century may seem dangerously close to falling into the trap of reading modern concepts of ethnicity and nationhood into early medieval texts and events. Indeed, after the Second World War scholars have been

Why Is There Something Rather Than Nothing?  147 careful either to avoid the issue altogether or to combat notions such as that, for example, the Treaty of Verdun of 843 (which divided the Carolingian Empire into three vertical strips, two of which resembled modern France and Germany) was ‘the birth-certificate of Europe’.99 Instead, a considerable amount of scholarship underlines that ‘there is no convincing evidence for recognisably French and German national identities before the eleventh century, until which time politics continued to be articulated in a resolutely Frankish idiom’.100 Yet this type of approach threatens to throw the baby of ethnic identity out with the bathwater of modern nationalism since it fails to acknowledge the importance that ethnicity played in the construction of early medieval identities.101 At the same time, although ethnic identity has been downplayed, it is widely agreed that Louis the German was the patron and probably initiator of an unprecedented cultivation of the Germanic vernacular in the written record.102 Yet scholars often fail to make the connection (out of the same wariness of reviving anachronistic notions of nationhood) to a sense of ‘German’ ethnicity which is, as we shall see, quite apparent in the texts under scrutiny here (among other textual and non-textual strategies).103 Hence we are told time and again that language does not equate to ethnicity: ‘Carolingian writers knew perfectly well that language did not make peoples’104 or that ‘the Carolingians of the ninth-century did not… envisage their lands as linguistic units and accordingly their attempts to divide them paid little or no attention to language’.105 Even if we ignore the fact that different conceptions of identity exist in different communities and strata of society, there is plenty of evidence that early medieval authors did envisage language as being one of the markers of a separate ethnicity (or, in Walter Pohl’s terms, ‘strategies of distinction’).106 Otfrid himself framed the Evangelienbuch in a decidedly ‘ethno-nationalist programme’ of vernacular literature (I discuss this at greater length next).107 Regino of Prüm, writing c. 900, offered four categories for classifying ethnic variation: ‘the various nations differ in descent, customs, language and law’.108 And although he placed genus rather than lingua as the first of his categories, ‘racial differences were generally considered less relevant in the formation of concepts of nationhood in the Middle Ages than cultural qualities such as customs, language and law’.109 In an earlier but not very distant context, Bede distinguishes the peoples of Britain (Britons, Picts, Irish, and English) according to the languages they spoke.110 And, as Sarah Foot argues, Alfred the Great’s active promotion of the vernacular Englisc to bind together his subjects as the Angelcynn, the English nation, shows how useful this strategy was ‘in overriding the inheritance of political and ancestral separateness in the creation of a new identity’.111 However, it is true that we should not mistake this sense of ethnicity for a purported proto-national identity. Nor should we ‘use terms like “imagined community” or “ethnicity” unquestioningly as a safe replacement for the term “nation”’.112 Even if, pace

148  Why Is There Something Rather Than Nothing? Neville, these identity-forming strategies were trying to build nations (as has been argued in the case of the Alfredian Angelcynn), it is important that these early medieval nationes are not to be seen teleologically as the acorn from which the modern nations arose.113 In that case, a better point of access to any sense of ethnic identity might be afforded by linguistic evidence – the names that people actually used to call the languages and dialects grouped under the modern label ‘Germanic vernaculars’ since ‘the collective names adopted by communities play a significant part in the process of the formation of their identity’.114 Thus how were these vernaculars actually called by the actors on the cultural and political stage of Carolingian society? The synods and capitularies of 813 enjoined the priests to translate their sermons into the Romance and Germanic vernaculars: in rusticam Romanam linguam aut Theotiscam.115 The alliance between Charles the Bald, Louis the German and Lothar II of the Middle Kingdom in 860 was sworn both in romana lingua and lingua theodisca, but the written record was made in Latin. It is not clear whether theodisca refers to a particular Germanic language or to all the Germanic vernaculars as a group of mutually understandable dialects (or perhaps, especially in the first case, to an undifferentiated notion of vernacular languages that are not Latin or Romance, irrespective of any relationships inside their linguistic family). The Strasbourg Oaths of 842, marking the alliance between Charles and Louis against Lothar, were sworn in romana lingua and in teudisca lingua: here the specific dialect defined by the latter is Rhenish Franconian (which, of course, is no guarantee that the term refers to the same dialect in the previous two instances or that it was understood in this limited sense by the people involved in the oaths).116 The adjective theodiscus is a Latinized version of an OHG *theodisk (from theoda, ‘folk, people’, meaning ‘of the people’) – the library of the Reichenau even had ‘poems for teaching the German (theodiscam) language’. 117 It was used at the Carolingian court from the late eighth century to refer to the vernacular, spoken language.118 As can be seen from the previous examples, it was a catch-all term, not meant to identify one dialect but a grouping of people, as in the phrase nationes theodiscae, ‘German-speaking peoples’ – ‘without implying any kind of panGermanic community’, as Reuter hastens to warn, although this is not entirely correct, as will be argued shortly.119 There is another word for the Germanic vernacular(s) that has a curious affinity to this one: in the ninth century, there is a revival of the classical word teutonicus (or the noun teutones), often used in conjunction with theodiscus. Thus, in 893, Archbishop Fulco of Reims, in a letter to Arnulf, king of the east Frankish kingdom, used the word ‘teutonic’ to describe the written sources (libris teutonicis) in which Arnulf could read about Ermanaric, the legendary Gothic leader who by this time must have been part of the same complex of narratives as Attila, Walter of

Why Is There Something Rather Than Nothing?  149 Aquitaine, and, why not, Sigfried.120 The fact remains that already by the end of the ninth century, some elite Carolingian individuals were connecting ‘Germanic heroic’ narrative (about the Goth Ermanaric no less – a significant clue, as shall be seen shortly), the east Frankish polity, and the word teutonicus. The latter conceivably connoted the respectability and even venerability conferred upon the Germanic vernacular through this classical Latinate framing.121 Matthew Innes traces the word back to an early Carolingian origin in the western part of the empire: at Tours around 830 the ‘Teutons’ in Virgil’s Aeneid, VII.741, were glossed as ‘the people of Germania… who speak the lingua theodisca’.122 It thus seems that teutonicus was used by Carolingian writers as a classical-sounding synonym for the Latinized Germanic word theodiscus. Yet this example does even more than equate the one with the other: it brings together all three names used in Carolingian sources for a ‘Teutonic’ ethnic group, the ‘theodisca’ language they speak, and the territory they occupy (‘Germania’). The fact that two of the three originate in Latin Antiquity does not prevent them from having been reemployed (precisely because of the prestige due to the classical echoes they evoked) by the Carolingian elites to name a newly emerging community of language, ethnicity, and territory. As Pohl shows, ‘The same biblical and classical [ethnographic] models could serve to establish new systems of perception and to legitimise the existence of the new ethnic kingdoms’.123 Tacitus’s Germania would have provided one of the sources for this model – its first certain use after 525, when it is cited in Cassiodorus’s Variae, occurs in the mid-ninth-century Translatio Sancti Alexandri by Rudolf of Fulda.124 Tacitus (and especially the Germania) was virtually unknown in this period, so why is this source used in this context? The reasons for this have everything to do with the search for German origins that had begun at this time in the Carolingian world. Rudolf begins his account of the translation of the relics of St. Alexander by going back to the origins of the Saxons: in a surprising reverse migration myth, he asserts that the Saxons sailed from Britain to the Continent. In this origin story, he makes heavy use of three chapters from Tacitus’s Germania by changing what the classical writer ascribed to the Germani to refer to the Saxons: ‘Tacitus’s noble barbarians have become Rudolf’s noble pagans’.125 The codicological context of this unique copy of Tacitus is also relevant and shows the same propensity for origin stories. The Germania is found in the Jesi manuscript (also known as Codex Aesinas, now at the National Library in Rome, catalogued as Cod. Vitt. Em. 1631), which also contains Tacitus’s Agricola and the Bellum Troianum by the spurious author Dictys Cretensis and which was probably written at Fulda in the second quarter of the ninth century, where Rudolf used it.126 It is well-known that Fulda was one of the crucial cultural centres of interest in the vernacular in the Carolingian Empire.127

150  Why Is There Something Rather Than Nothing? As shall be demonstrated next, Carolingian intellectuals like Freculph of Lisieux were also aware of both the Trojan origin story of the Franks (as instantiated in the Bellum Troianum in the Jesi manuscript) and a Scandinavian/‘Teutonic’ one (rooted in Jordanes rather than Tacitus, in Freculph’s case). Thus Rudolf was certainly not the only one keenly interested in narratives about the origins of the Germanic peoples of the empire. The Germania provided the necessary classical legitimation for this new preoccupation. While Tacitus provided the source for this conceptualization, the text itself need not have been the cause for this emergence at this time and place. Rather, since ‘a text is not a discrete work with its own essential meaning and aesthetic, but rather a cultural object situated within an array of other contemporary texts, events, and institutional practices’, it was the social, political, and emotional conditions which I am exploring that allowed for ancient sources such as the Germania to be rediscovered and employed in a new mental framework.128 This is something that can be easily overlooked when one is merely searching for textual antecedents and manuscript transmission. The sense of political community that developed in the east Frankish kingdom in the later ninth century created a fertile environment for the emergence of a linguistic self-consciousness, which Innes defines as ‘the recognition that the inhabitants of the kingdom shared a language which differed from that spoken in much of west Francia’.129 This is partly accurate: it is unlikely that people from Bavaria and Rhenish Franconia ‘shared’ a language, but it is probable that their dialects were mutually intelligible. Roughly, the dialects spoken in eastern Frankish territory can be grouped in high dialects, such as Franconian (or ‘Frankish’), Alemannian, Bavarian, and Thuringian and low dialects, such as Saxon and Frisian. The problems involved in cross-dialectal communication comes out nicely in the transmission history guessed in the language of the Hildebrandslied. Metrically, some of the lines in the poem seem impossible, and this has been argued to be due to the problematic linguistic transmission: it probably represents a Bavarian original badly adapted into Low German.130 This opens up a problem that linguists have even for dialects which are copiously documented: where did mutual intelligibility stop? Of course, languages are not separated in sharply bounded dialect-zones, as they are sometimes depicted on maps – linguistic variety manifests itself in dialectal continua gradually merging one dialect to another (these border areas indeed are rarely between just two dialects).131 Speakers of High German dialects probably understood each other, and even in the absence of mutual intelligibility, there must have been recognition of the fundamental relatedness of these languages. The different dialects spoken by the Franks during this period ‘surely showed no more differentiation than, say, the OE or the OHG dialects did, and perhaps less’.132 Franconian (which became the preferred dialect at Louis the German’s court and

Why Is There Something Rather Than Nothing?  151 among intellectuals connected to him such as Otfrid) was probably easily understandable to all east Franks, Bavarians, Alemans, and Thuringians but somewhat less so to the Saxons.133 And even, for instance, Saxon- and Bavarian-speakers must have understood at least some of each other’s words and might have got the gist of each other’s utterances much better than a native Latin speaker and a Saxon one. For comparison, Matthew Townend has forcefully argued for ‘a situation of adequate mutual intelligibility between speakers of Norse and English in the Viking Age’, which does not preclude bilingualism among both populations.134 Thus if there was dialectal congruity between speakers of two languages which had been isolated for 200–250 years, how much more can the same be said of the Germanic eastern Frankish dialects.135 Thus it is not inconceivable that all the speakers of Germanic dialects in eastern Frankish territories would have been considered by others (and seen themselves) as speakers of the same language. A look at the way the early Icelandic settlers understood their place in the larger Old Norse linguistic continuum can prove illuminating. Stephen Leonard argues that the term dönsk tunga was used in early Icelandic sources (tenth to twelfth centuries) to mean ‘Old Norse’ rather than for any Danish dialect: ‘Old Icelandic was recognised as part of this linguistic family, and dönsk tunga represented the speech of the wider Scandinavian community and not just that of the settlers’.136 In other words, Leonard shows that in the case of the Old Norse-speaking area, a linguistic continuum could be defined by speakers in both Iceland and Denmark as one language, even if there was some variation – what was necessary was some degree of mutual comprehensibility. Plenty of evidence suggests that Carolingian intellectuals did envisage larger families of Germanic languages. In fact, they conceivably developed at least two conceptualizations of superordinate linguistic orders that came to describe different levels of ethnic identity. The first one chronologically (which I have begun to explore) was inspired by both classical ethnography (Tacitus and Jordanes) and contemporary perceptions of relatedness and was instantiated by the constellation Germania–theodiscus–teutonicus. The second one was dictated by east Frankish politics but also represented and was shaped by the discourse of intellectuals, such as Otfrid. The latter concentrated, as I argue in the next section, on the Franconian adjective frankisgon, ‘Frankish’, used by Otfrid to describe both the Germanic vernacular into which he translates the gospels and the elite of Louis the German’s kingdom, the ‘true Franks’ – that is to say, the eastern ones (at least in Otfrid’s opinion). I have so far traced the former strategy and level of identity as it emerged out of a growing perception of related Germanic vernacular dialects but also due to the social and emotional forces explored in this chapter. I now continue on this path, looking at how it becomes a discourse of constructing a mythical ‘Germanic’ past and hence a fertile environment for the emergence

152  Why Is There Something Rather Than Nothing? of legendary heroic narratives, following this with Otfrid’s and Louis the German’s constructions of eastern Frankish ethnicity as Germanic in opposition to the western Latinate Frankish identity. Having drawn attention to the anachronistic nature of the adjective ‘Germanic’, I am now guilty of using it again. I do so with a clearer conscience since, pace the long line of scholars arguing that there was no pan-Germanic sentiment among Germanic speakers, it may have been Carolingian intellectuals after all who created the notion of Germanic supra-ethnic relatedness or, as Roberta Frank terms it, ‘Gothicism’.137 Of course, this notion was vastly different from the ‘pan-Germanic’ kinship imagined by nineteenth-century scholars – it responded to, expressed, and shaped very different social and political needs. The Carolingian ninth century was marked by a flowering of interest in vernaculars, which had led some intellectuals, like their early modern successors, into a search for common histories to explain linguistic similarities.138 This process is seen perhaps most clearly in Freculph of Lisieux’s world chronicle, written in the 820s and 830s and dedicated to the Empress Judith for the education of her son, the future king, Charles the Bald. Significantly, his history associates the well-established legend that the earliest Franks were the descendants of the Trojans with the unexpected foray into the Scandinavian and Gothic origins of the same people: ‘Other men insist that they [the Franks] had their origins on the isle of Scandza, the womb of nations, from which the Goths and the other nationes theodiscae came: what is more, the idiom of their languages shows this’.139 Here Freculph is paraphrasing Jordanes’s account of the origins of the Goths: ‘Long ago the Goths are said to have come from the isle of Scandza, as from a hive of races or a womb of nations, under their king Berig’.140 Of course, as Goffart argues, this origin story is probably purely literary creation, resting on ethnographical topoi, not a true account.141 But the more interesting point about the story and especially the context in which it is reemployed is the fact that we get a glimpse into a competing narrative, a very different account of the Frankish origins and language (two essential markers of self-conscious ethnic identity in both modern and early medieval accounts, as we have seen). Freculph, while not dismissing the Scandinavian-Gothic story, is clearly taking his distance from it: ‘alii vero affirmant’ (in Innes’s translation, ‘other men insist’). It is as if he is compelled to report this other well-known story because many believe it and circulate it, although, in fact, he does not give it the credence afforded to the prestigious Trojan one. In fact, while it is clear that Jordanes provided the inspiration for this story, Freculph does not report it as being read (as the aforementioned Fulco report of ‘Teutonic books’) but heard from ‘others’. For while his wording is very close to that in Jordanes, Freculph adds as a further proof for the plausibility of the Scandinavian origin ‘the idiom of their languages’. The same constellation of ideas that preoccupied many other Carolingian intellectuals can

Why Is There Something Rather Than Nothing?  153 be seen here at work: the linguistic similarities between the Germanic vernaculars conceived as proof of related ethnicities and the matter of common origins of Germanic peoples, which was resolved by other writers into the idea of Gothic origins as a shorthand for their ‘Teutonicity’. This testimony of an important Carolingian intellectual is also very precious for letting us in on one type of the competing discourses I am analysing here (although, of course, never in a very straightforward way). Thus it is not simply that Freculph himself, starting from linguistic links between Goths, Franks, and other peoples, posits ‘a shared history and a shared origin legend beginning in a shared Scandinavian homeland’.142 Rather, he reports one of the discourses extant in the Carolingian marketplace of ideas, one origin story which shows that at least some Franks thought of themselves as having Scandinavian roots and a kinship that is both ethnic and linguistic with other ‘Teutonic’ peoples, such as the Goths. Indeed the Franks were not the only ones claiming this narrative. While Jordanes was the first historian to record a Scandinavian origin myth for the Goths, his account was influential and imitated by Lombard and Burgundian discourses before the Frankish one, all depicting their own ethnic group as originating on ‘the isle of Scandza’.143 While narratives such as that in Freculph’s report are indeed rooted in classical ethnography and not in any Traditionskern tradition carried over from the Migration Age, their spread among Carolingian intellectuals may have opened the way for at least some textual communities to look favourably on the vernacular legendary narratives that must have circulated orally before they emerged in the written record as the Waltharius, the OE Waldere, the Hildebrandslied. As Shami Ghosh deftly argues, these traditional vernacular narratives need not have been part of a specific ethnic or social group’s political identity (in the sense of a Traditionskern tradition as envisaged by Wenksus).144 Instead, they were conceivably seen as a part of inherited tradition (and not a particularly prestigious tradition at that, at least for Carolingian nobles) or, to employ modern categories, as ‘folklore’. Ghosh is careful to distance himself from the Wenksus school of thought that asserted that these narratives were the tradition cores around which the elites of Germanic peoples built their identities. Yet it is hard to explain why the texts building on this narrative tradition emerged only in this specific timeframe and why they took this specific shape without acknowledging the role they played in shaping or expressing political or ethnic identities. Thus I would suggest that, while they did not fulfil a self-conscious political role (as ‘propaganda’) as the Heliand did, they provided a space where the very issue of ethnic identity (among others, such as gender identity) could be explored, toyed with, and expanded to the breaking point within the comparatively safe haven of satire and entertainment. What Carolingian intellectuals were doing with Tacitus and Jordanes, reframing them in their new emerging

154  Why Is There Something Rather Than Nothing? consciousness of relatedness between ‘Teutonic’ ethnic groups and their dialects, other parts of the elite were doing with legendary narratives about Goths, Burgundians, and Huns. As shall be seen in the following, the Hildebrandslied and the Waltharius allow us to see glimpses of the preoccupations of such aristocrats, who were perhaps more relaxed about both the monastic and classically learned ideals of Carolingian nobility than Alcuin or Freculph, not to the point of denying them but feeling safe in playing with them. Innes sees in Freculph’s dependence on Jordanes’s proof against any vernacular narrative tradition being manifested in the first ‘Germanic heroic’ texts.145 If by this contention he means the ‘Germanic warrior culture hypothesis’, I concur that the texts are not unproblematic survivors of a putative pan-Germanic oral tradition that was naturally ‘heroic’. Yet I disagree with his (echoing Goffart) conclusions that ‘the language of recitation or transmission did not determine the content of traditions’ or that ‘there is no indication that oral traditions about Theodoric either subverted, or originated in a different context from, written discussions of his reign’.146 The latter would imply that all Carolingian vernacular texts about men fighting are scholarly creations or royal propaganda. Yet not all stories featuring Goths mention Theodoric (the Hildebrandslied only mentions his name to set the scene), neither need they all have political import (as ‘propaganda’).147 Walahfrid Strabo can be intensely critical of Theodoric in the poem on his statue (De imagine Tetrici) and yet he is not averse to thinking about the Goths as the ancestors of the Franks – thus not all stories about the Goths are actually about Theodoric.148 It is true that poetic discourse can be a ‘vehicle for political work’ in a time when new political, cultural, and linguistic identities are emerging. But ‘political’ is too narrow a term for the work these texts were doing – rather, they are sites where cultural issues of great magnitude and complexity can be affirmed but also contested.149 Ghosh combats very effectively the notion that the production of vernacular poetry about Gothic kings in Francia or early medieval England is merely an effect of the knowledge of Jordanes and an awareness of linguistic kinship: Cases of vernacular borrowing from written, Latin material are exclusively confined, until the end of the Carolingian period at least, to religious or philosophical literature, and arise largely from conscious clerical efforts to disseminate the message of Christianity. It is impossible to establish direct textual relationships between any extant Germanic legend and any extant contemporary Latin text; this is not the case for the many vernacular religious works of this period. I would argue, therefore, that the existence of oral vernacular narratives stimulated an interest in a ‘Germanic’ past among writers of Latin, rather than, as Frank (1991) and Goffart (2002) propose, knowledge of Jordanes causing vernacular poets to compose the extant works.150

Why Is There Something Rather Than Nothing?  155 Indeed, Freculph is not alone in teasing out the Gothic and Scandinavian connections of Frankish ethnicity.151 Writing a commentary on Donatus’ Grammar c. 805, Abbot Smaragdus of St. Mihiel was struck by the similarities of Gothic and Frankish personal names and explained this by the existence of a historical link between the two peoples.152 Walahfrid Strabo, tutor to Charles the Bald (for whom Freculph’s work was intended), writing in the 840s, also linked the Gothic language, and thus their history, to that of the Franks: Goths ‘spoke our – that is, the Germanic – language (nostrum, id est Theotiscum, sermonem), and, as chroniclers show, late scholars of that people translated the Holy Scripture into their own language, of which today a few traces are found’.153 Walahfrid’s comment on the Gothic vernacular translation (and, apparently, transmission) of the Scripture is interesting if we think of the experimentation with vernacular biblical literature taking place around the middle of the ninth century. The ‘few traces’ of these translations in Walahfrid’s day may refer to the several copies of Ulfila’s Gothic translation, including the Codex Carolinus at Grimald’s monastery of Wissembourg.154 Indeed, the interest in Ulfila’s Gothic translation was not restricted to Walahfrid: the Salzburg teacher Baldo, who wrote a commentary on Gothic letters, copied excerpts from Ulfila’s Bible, tried to translate these excerpts into Frankish, and included phonetic marks for pronunciation.155 It seems that Baldo had close ties to the east Frankish court: Louis the German composed a poem thanking Baldo for his religious writings, and Baldo later became chancellor of Louis’s eldest son, Carloman.156 It thus appears that the interest in the Goths as a long-lost relation of the Franks that emerged around the early 800s grew into much more than this by the time Louis the German established his kingdom as a self-standing state. East Frankish intellectuals not only commissioned copies of the Gothic Bible and studied its language in-depth but also thought of themselves as more than related to the Goths – indeed, Walahfrid states, ‘They speak our language’, suggesting a virtual identification of ‘Frank’ with ‘Goth’. As I will suggest in the following section, this fascination with Goths and other Teutonic peoples is visible also in texts inspired by the vernacular poetic tradition, like the Waltharius. Indeed the honour of being ‘Teutonic’ was not confined to Franks but extended by Carolingian intellectuals to the Danes and the Saxons. The Saxons are a less surprising object of this inclusion: Gottschalk, the Saxon monk and controversial theologian, saw Germanic speakers as constituting a single people, the gens theodisca.157 Indeed, this is to be expected of the Saxons since all Carolingian kings since Charlemagne tried to secure the loyalty of the Saxon elite by the discourse of oneness with the Frankish aristocracy. Einhard wrote that the defeated Saxons ‘were joined to the Franks and made one people with them’.158 As Timothy Reuter observed, ‘It is not clear, incidentally, that the Saxons agreed. Tenth-century Saxon writers often also showed considerable hostility to

156  Why Is There Something Rather Than Nothing? the Franks’.159 Thus the narrative of the Saxons and Franks as one ‘theodisc’ people is understandably advantageous for both the Frankish hegemonic discourse and for the Saxon elites eager to integrate (though not unconditionally). Again, one must see this preoccupation with ethnicity as a socially bounded phenomenon.160 The Saxons had different conceptions of ethnicity dependent on social class. Rudolf of Fulda reports in his history of the Saxons (probably intended to be an independent work but now found together with his Translatio sancti Alexandri) that the Saxon nobles were a distinct race (genus) and emphasized their strict segregation from the other social castes (diferentiae) in Saxony.161 Rudolf also asserts that they were descended from the conquerors of the region, while those of lower birth were the offspring of the local indigenous peoples whom the Saxon conquerors had made into serfs.162 Rudolf also mentions the ethnic unity topos: after the conquest, they abandoned their pagan beliefs and were ‘united with the Franks…[the Saxons] were made one people with them’.163 The implication is that the only real Saxons were the aristocracy, thus conditioning ethnicity on social class. While Rudolf is the only Carolingian to spell this connection out, we can imagine that this socially bound ethnic identity was not unique among the Saxons. Indeed, one royal strategy to build and preserve Wirgefühl among the members of the same ethnic community was to summon assemblies of ‘all the Franks’, as the Annals of St-Bertin and the Annals of Fulda repeatedly report.164 Yet we should not imagine literally that all the Franks did attend them: If ‘all the Franks’ attended these assemblies, ‘Frank’ needs to be redefined: it appears to indicate a certain level of prestige and power rather than merely ethnic characteristics. The community imagined by these chroniclers is thus exclusive, not universal or ‘nation’-wide, and the identity created by annal-writers might have been the property not of the ethnic community of the ‘people’ as a whole, but rather of a particular class.165 Indeed, a legal definition of the word Franci is used in a context where the word is synonym with ‘free men’, as distinct from unfree members of the estate’s familia.166 Thus the need for constructing an ethnic identity was particularly stringent for a certain part of the aristocracy and was thus expressed in the discourses they allowed themselves to produce. As stated in the previous section, while the highest strata of Saxon aristocracy conceivably saw themselves as being part of a trans-ethnic Carolingian elite, the lower nobility (especially when Germanic speaking) might have felt a stronger need to claim a special descent (and thus ethnicity).167 Hence the entire constellation of ‘Teutonic’ language and ethnicity (common with the Frankish one in the earlier ninth century, specifically east Frankish in

Why Is There Something Rather Than Nothing?  157 Louis the German’s reign around the middle of the century) and vernacular orally derived narratives about the legendary ‘Teutonic’ past. This ‘Teutonic’ past (and thus the ethnic category embodied by it) seems to have been spacious enough to accommodate even groups which did not have political ties with the Franks as close as the Saxons or other eastern-Rhennish ethnic groups. Hrabanus Maurus, Gottschalk’s and Walahfrid’s teacher, as well as Freculph’s friend, identified the Danes as Germanic because of their language and argued that they descended from the Marcomanni (which are rooted in Roman ethnography).168 In a panegyric description of the baptism of the Danish king at Louis the Pious’s court in 826, Ermold the Black also made a link between the Franks and the Danes and claimed that the Franks originally came from Scandinavia.169 As Innes argues, the idea of a common Germanic linguistic heritage was probably discussed by Hrabanus, Freculph, Walahfrid, and others at court, which led to the idea of the Germanic-speaking peoples having a shared history.170 But there is another way to see this lively preoccupation with ethnicity in the Carolingian world other than as a debate belonging to the realms of scholarly speculation and royal propaganda. Why do some people start thinking of themselves as ‘Goths’, ‘Teutons’, or ‘true Franks’ in this time and place? Perhaps because ethnicity as a category begins to be problematic in a socio-political context in which fissures begin to appear in Charlemagne’s empire that made all its peoples honorary Franks: Carolingian royal heirs start to fight each other, local elites have to choose sides, Germanic-speaking nobles already feel estranged from Latinate official discourses. Thus the preoccupation with ethnicity (especially in its post-Charlemagne ‘Teutonic’ and, as shall soon be seen, ‘true Frankish’ phases) is characteristic of a moment when the category itself begins to be problematized in Carolingian discourses (some of which I have already explored). In any case, what the present chapter attempts to unravel is not so much an ahistorical category of ‘ethnicity’ universally applicable to any human society. Thus I am not arguing that a ‘Teutonic’ identity emerged against the older Frankish one or that people started to think of themselves as ‘Germans/Goths/Teutons’ in a more biological sense than previously. Rather, I bring to light the ways in which the historical socio-cultural concerns specific to ninth-century Carolingian Francia manifested as and were constructed by discourses that were concerned with ethnicity to a greater extent than previously, also that these ethnic discourses and practices are always gendered, socially localized, and use both vernacular languages and orally derived poetic traditions in complex ways. Thus this specific type of ethnic identity came to be problematized once the polities of the former empire began to contend with each other not only politically but also through competing discourses.

158  Why Is There Something Rather Than Nothing?

Carolingian Ethnicities (II): ‘True Franks’ in the East and West As Eric Goldberg thoroughly argues, in the most acute phase of east versus west Frankish strife, these conflicting discourses manifested themselves in anything from royal attire and court ceremony to biblical paraphrases or epic poetry.171 For instance, during the many diplomatic meetings between Charles the Bald and Louis the German, ‘nobles would have seen the contrast between Louis the German’s austere ‘dress and equipment of the Old Franks’ and Charles the Bald’s luxurious golden, purple, and silk clothing decorated with stars, griffins, elephants, and peacocks’.172 East Frankish aristocrats would have looked approvingly as Louis and Charles publicly exchanged staffs to demonstrate their fraternal accord in 849 and would have taken to heart the contrast between Louis’s ‘hard and threatening staff of apple wood’ and Charles’s ornate ‘scepter of gold and precious stones’.173 The opposing narratives that these gestures and artefacts were telling was that of an east Frankish pride in their king’s cherishing the traditional attire of his forefathers and on the other side that of a west Frankish elite asserting its imperial aura that favoured a luxurious orientalizing aesthetic over local customs and hoary heirlooms. Thus when Charles appeared in public wearing Byzantine clothing, the east Frankish author of the Annals of Fulda criticized him for ‘scorning the entire tradition of the Frankish kings’ and ‘holding the glories of the Greeks to be the best’.174 As Goldberg argues at length, through their respective public personae and gestures, one favoring the ‘equipment of battle’ and the ‘hardness of iron’ and the other the ‘splendor of banquets’ and the ‘glitter of gold’, Louis the German and Charles the Bald each offered his own distinctive commentary on the nature of power and political community in his nascent kingdom.175 This pageant is, however, not primarily about the two kings themselves but rather about the narratives they project on the aristocracy following each of them. As I will argue more at length in the following section, the discourses that Louis the German encouraged among his nobles connected an ‘old Frankish’ or ‘true Frankish’ ethnic identity with a very masculinist aristocratic identity, both expressed in a ‘Frankish’ vernacular raised to the status of literary and Gospel-worthy language. This takes us to the other strategy of ethnic identity construction that is characteristic of Louis’s east Frankish polity. After the constellation Germania–theodiscus–teutonicus was enlarged to contain not only Franks but also Goths, Saxons, Danes, and other groupings of people, the leader (but also the elite) of the emerging eastern Frankish polity might have felt the need for a more localized and less protean ethnic identity. It found its expression and was shaped by the discourse of people like

Why Is There Something Rather Than Nothing?  159 Otfrid – its label was the adjective frankisgon, ‘Frankish’, employed to describe the Rhenish Franconian dialect favoured at Louis’s court but also the ‘true Franks’ – that is to say, the eastern ones.176 The aforementioned clashing discourses give us a taste of what the qualities of a ‘true Frank’ might have been: they include, as will be argued in the following section, a warlike type of masculinity and nobility (since, as we have seen, not all Franks are really Franks), as well as a taste for vernacular (and conceivably ‘heroic’) poetry. For now, I turn to an expression of this new linguistic and ethnic identity in Otfrid’s Evangelienbuch. This source is often read simply as biblical paraphrase, intended for the spreading of the Gospel’s message to Germanic speakers. But there are other discourses that the Evangelienbuch is responding to without ever mentioning them. Parshia Lee-Stecum convincingly reads it in the context of the aforementioned east-west Frankish political rivalry.177 Otfrid’s near contemporary, John Scottus, only a decade or so before the Evangelienbuch wrote a poem called Hellinas Troasque in which he interwove royal praise with Christian doctrine, juxtaposing Christ as saviour of humanity with Charles the Bald as saviour of the west Frankish kingdom. More specifically, the poem partly commemorates Charles’s victory over Otfrid’s dedicatee, Louis the German in 858. Lee-Stecum convincingly argues that it is to this discourse and others of its kind that Otfrid is responding in the proem to the Evangelienbuch and his other justificatory writings (the letters to Liudbert). This makes sense when considering the fact that most poets writing in Latin at the court of Charles, as at the court of Charlemagne before him, were not ethnically Frankish – Scottus himself was probably Irish.178 This may or may not have been in Otfrid’s mind when he wrote these lines, but for a reader or listener who made the connection it would certainly add an extra sting to his claim that the Franks themselves have never celebrated their deeds or divine beneficence in verse.179 This is part of a very adept strategy at presenting the discourse originating in (and supporting) the west Frankish court and kingdom as being doubly ‘un-Frankish:’ they are authored by people neither writing in Frankish nor being ethnically Franks. While the negative goal of this strategy was to undermine competing narratives coming from west Francia, the positive purpose was to construct an ‘ethno-nationalist’ poetic programme as presented in the Evangelienbuch. Most of the 126-line-long verse preface of Otfrid’s poem is taken up by what is basically laus Francorum: Sie sint so sáma chuani, sélb so thie Románi;/ni thárf man thaz ouh rédinon, thaz Kríachi in thes giwídaron./Sie éigun in zi núzzi so sámalicho wizzi,/in félde joh in wálde so sint sie sámabalde

160  Why Is There Something Rather Than Nothing? ‘They [the Franks] are as brave as the Romans, and no one could claim that the Greeks are their equals in this. Similarly, they possess effective knowledge (they are equally bold in field and forest)’.180 The subtext here is that ‘just as the Franks equal or exceed all other peoples in force of arms and riches, so they should have a correspondingly exceptional native literature, and so Otfrid’s own poem presents itself to provide this literature’.181 For Otfrid, the language of the Franks, ­frénkisga zúngun (Evangelienbuch I. 1.119–26), is the Rhenish Franconian dialect Louis himself favoured.182 Yet language and ethnicity are not the only two markers of true Frankishness. The previously quoted lines project – similarly to the other constellation explored in this section, although less obliquely – a certain model of manliness in relation to the two other features. The Franks are brave (just as the Romans, and more than the Greeks – both bold claims in a context where the classical antiquity was the standard for most things, at least among the educated). They are also wise, although this latter quality is understood martially more than anything (‘effective knowledge’). The Franks are indeed bold, both in field and forest – again, the implied context is obviously martial. A constellation thus emerges in which the true Frank is conceptualized as male, noble (martial prowess and political involvement are manifestations of nobility, as will follow from the following section), ethnically ‘Frankish’ (although presumably from east of the Rhine), a speaker of Germanic (mainly Franconian), and, importantly, a fine warrior. The quintessential ‘true Frank’ can, of course, only be the king. Louis himself is offered as an example of Frankishness in the opening lines of Otfrid’s verse epistle to the king (named rex orientalium regnorum, ‘king of the eastern realms’ in the acrostics formed by both the initial and the final letter of each couplet): Lúdowig ther snéllo, thes wísduames fóllo,/ er óstarrichi ríhtit ál, so Fránkono kúning scal (‘Ludwig the brave, full of wisdom, he rules the entire eastern kingdom as a Frankish king should’).183 Again, the absent referent could be the west Frankish leader, who presumably does not behave like a true Frankish king. In any case, in the portrait of Louis, Frankish identity (and the focus on the Franks, their land and their Frankishness of Louis continues in the following lines: 3, 13–14, 17–18, 89–90) is bound up again with the eastern Frankish polity and its people, their vernacular language, and bravery accompanied by a martial sort of wisdom. As Lee-Stecum argues, Otfrid’s juxtaposition of the martial and cultural achievements of the Franks with the language in which he is writing ‘takes on a particular polemic edge at this time in the ninth century when the attribution of specific ethno-national labels and identities was problematic and the meaning and ownership of Frankish identity in particular were fluid and contested’.184 Thus he takes advantage of these ideological discontinuities by positing the eastern Franks as the Franks

Why Is There Something Rather Than Nothing?  161 par excellence. At the same time, by contrasting his own vernacular ethnic-poetic-linguistic programme with the Latin western Frankish poetic tradition, he effectively excludes those ‘western Franks’ from the category of ‘Franks’ altogether, thus breaking with the entire narrative of Frankish consensus and unity in kingship which had been the standard discourse for several centuries but which Otfrid himself explicitly draws on in the proem to the Evangelienbuch. Were the western Franks oblivious to this radical redefinition of the very notion of ‘Frank?’ Quite the opposite, for a symmetrical reorientation can be seen in western Frankish discourses, although there the preferred vehicle is more commonly historiography rather than vernacular poetry (fittingly for a court taking pride in its Latinity and continuity with classical tradition). Helmut Reimitz has convincingly proven that in rewriting historiographical sources such as the Royal Frankish Annals during the reign of Charles the Bald, a decidedly west Frankish identity as truly Frankish takes shape.185 He shows how the Austrasian (from the eastern region of the regnum Francorum) elements are simply erased from genealogies of kings but also from the versions of the Liber historiae Francorum copied under Charles, with Carolingian history being depicted ‘as an uninterrupted, western Frankish affair’. 186 Both these mutual erasures and exclusions from participation in Frankish identity are unprecedented. In order to see just how unusual this is, a brief review of the evolution of the notion of ‘Frank’ is necessary. Indeed, there is plenty of scholarship on the development of Frankish identity until the times of Charlemagne. My contribution to this is merely tracing in the previous pages the subsequent development of what it meant to be a Frank during and after the civil wars beginning in Louis the Pious’s reign. It is true that a sense of east/west Frankish rivalry (between the Frankish kingdoms of Austrasia and Neustria) existed from the seventh century. Yet Louis the German’s eastern Frankish polity and the nascent sense of ethnic identity included more than just the territory or people of the old Austrasia: the Saxons, the Bavarians, the eastern Franks, at least ideally, all became ‘Teutons’ or, as we have seen, ‘true Franks’. In any case, the occasional rivalry between the two polities did not prevent Frankish authors writing in the seventh and early eighth centuries from believing in the existence of a Frankish community – that is, ‘a group of nobles who shared descent from a common ancestry, who shared political and cultural concerns and who had a vested interest in the overall unity of their kingdom despite the existence of separate sub-kingdoms’.187 Richard Broome convincingly traces the evolution of what being a Frank was understood to entail, from a geographical/regional conception, on to a more political one (to which the present chapter adds the ethnic dimension more fully realized after Charlemagne).188 Broome shows that while each author wrote from a regional perspective, the emphasis was

162  Why Is There Something Rather Than Nothing? on Frankish community and consensus beyond and above any political differences. Thus, the term Austrasian, for example, was a way of referring to a Frank from the kingdom of Austrasia. These were specifically geographical distinctions for members of the gens Francorum and they remained regional identities; there never developed any concept of ethnic Austrasians or Neustrians, while the Burgundian ethnicity that had only ever been weakly developed in the pre-Merovingian period seems to have fallen out of use.189 In any case, by the time the Annals of Metz were written (c. 805), the Frankish sub-groups had all but disappeared from Frankish discourses, the simple identifier ‘Francus/Franci’ being used without qualifiers to apply to all the Franks.190 Broome also argues that around the same time, the focus of historiography moves from the internal politics of the Frankish heartland to the wars against peripheral peoples waged by the Franks as a whole. This marks the emergence of a political definition of ‘Frank’ which laid the foundation for an empire that included any ethnic group as long as it was Christian and faithful to the emperor but that needed a periphery of peoples against which to define itself (the root for the ‘border mentality’ of Louis the German’s elite). Janet Nelson argues that in Charlemagne’s time, the Frankish people were understood to be multi-ethnic, glued together by Christianity, ‘a Christian people of many peoples, rather than any actual gens’.191 In Broome’s terms, ‘membership of the community was no longer contingent on identifying as a Frank’, with the existence of the Frankish community relying upon its members (naturally aristocrats) swearing loyalty to the Carolingian dynasty and embracing a vision of community moulded at the Carolingian court that stressed orthodox Christianity.192 As Reimitz shows, the Carolingian constructions of Frankish identity indeed constituted a new phase in understanding the term, precisely because of ‘its establishment as a key term for political integration’.193 Thus the breaking of the Frankish narrative into symmetrical mutually excluding discourses in the western and eastern post-Carolingian kingdoms is due to this move from a regional/geographical conception to a political and religious one, which stretched the definition of being a Frank to its limits. As anxieties were growing about one’s place within the now-fragmented imperium, along the comparatively more fluid social ladder, and between conflicting loyalties, this translated into a greater preoccupation with ethnicity and language. Thus, paradoxically, Charlemagne’s move towards inclusiveness once vast new territories (especially across the Rhine) were incorporated into the empire is what led to a greater fragmentation and to the emergence of new models of ethnicity, at least among the aristocracy. These models went, as I have shown, from the earlier Teutonicism/

Why Is There Something Rather Than Nothing?  163 Gothicism to the later model of exclusive Frankishness adopted by both eastern and western elites. The former, on the one hand, integrated all peoples related by language into a Gothic or Teutonic kinship, while on the other hand, it gave the Franks a more Germanic (in Tacitus’s sense) or Scandinavian rather than Roman (as in the Trojan legend) origin. The latter provided different models of ethnicity for the two nascent kingdoms while preserving the central notion of Frankish domination. In any case, the two models not only stemmed from the same sociocultural anxieties but also were never mutually exclusive or localized to different eras or spaces. Otfrid’s Evangelienbuch provides a striking example of the correlation of the Teutonic and ‘true Frankish’ identities. While the prefatory vernacular texts keep talking about frénkisga zunga (the Frankish language – i.e., Rhenish Franconian), the title page of the Evangelienbuch has in large majuscules ‘quatuor euangelia theotisce uersa’, and theotisca is the usual way by which the Latin paratext (written in red majuscules) refers to the vernacular verses (black minuscules): ‘Incipit liber euangeliarum Domini gratia theotisce conscripsit’ (8r) or ‘scriptor hunc librum theotisce dictauerit’ (9r).194 Thus we come full circle to the lingua theodisca of earlier writers that could include Saxon and other Germanic languages.195 With Otfrid, as with plenty of other intellectuals of his time, we witness a tension between exclusion and inclusion which accounts for the particularities of Carolingian identities explored here. Towards the end of the century, the Ludwigslied, a martial panegyric vernacular poem praising the victory of a King Ludwig (probably the west Frankish Louis III) over the Vikings in 881, is introduced in the manuscript as a rithmus teutonicus.196 The verse is not alliterative but inner-rhymed, like with Otfrid, and it demonstrates the persistence of a nostalgia for a Teutonic vernacular poetry eulogizing royal martial prowess. Yet in the Ludwigslied, this ‘Teutonicity’ is constructed through an innovative verse form that had little to do with the actual Germanic metre but which is explicitly marked as teutonicus, code for ‘traditional’, ‘heroic’, ‘martial prowess’ – the same values we tend to associate with ‘Germanic’. Of course, this is the courtly fantasy of a ‘Teutonic’ masculine king, a figure that haunted the entire Carolingian ninth century.

The Interplay of Ethnicities in the Waltharius So far, I have analysed the emergence of ‘Teutonic’ and/or ‘Gothic’ discourses of identity in the context of the Carolingian multi-ethnic empire, followed by major reorientations in the meaning of ‘Frank’ in the second third of the ninth century as the empire disintegrated in at least two poles of power to the east and west. These discourses are mostly voiced by members of the Carolingian intellectual high life, and in some cases, they are used ideologically (especially with Otfrid). All of the texts scrutinized

164  Why Is There Something Rather Than Nothing? so far manifested a tension between what it means to belong to one ethnicity (Frankish or Saxon) and, on the other hand, to a supra-ethnic entity, whether it be Charlemagne’s multi-ethnic empire or the imagined community of ‘Teutonic’ peoples related by language (and invested with a putative common Gothic ancestry). This anxiety about ethnicity and language is undercut by anxieties about what makes a man and an aristocrat, as shall be seen in the following. These probings into the Carolingian socio-emotional mechanisms, however, are only meant to show the structures of thought and feeling that are problematized in subtler ways in rag-picker verse histories, such as the Waltharius. The Waltharius is usually looked at in terms of its classical epic and Germanic heroic roots (both have been shown to be equally strong), its biting humour (or lack thereof), or attitudes towards masculinity.197 The poem’s preoccupation with ethnicity, however, has gone virtually unnoticed by scholars so far. Yet the Franks as a people are present from the first lines. Indeed, in the Waltharius, ethnic labels are carefully assigned, and the reader is dutifully reminded of them throughout the text by oblique details. Weaponry is an indicator of ethnicity: ‘He swung a double-bladed axe straight at his face/ the Franks in those days had this kind of weapon’ (Waltharius 918–19).198 Legendary origins also have a strong identity-forming strategy, and Hagen is shown to be Frankish by being de germine Troiae (‘from Trojan stock’, line 28). Different ethnies are also said to have certain essential characteristics – thus the Burgundian king Hereric refers to the Trojans as ‘that bold tribe, to whom we cannot be compared’ (line 58). The textual community from which this text emerged was clearly highly sensitive to ethnicity and to the interlocking political systems based on them (in which the Franks seem to play a crucial role). Throughout the poem, ethnicities are not incidental qualities but move the narrative forward and determine the course of events. Thus, the first 90 lines describe the domino effect which the submission of the Frankish king to the Huns has on the other two peoples and polities, the Burgundians and the Aquitanians. Hildegund, the daughter of the Burgundian king Hereric, is promised to Walter. Gunther is, although a Frank, never said to be a Trojan, as Hagen is. Fittingly, Gunther appears as a ridiculous image of greed and cowardice, while Hagen is a noble warrior, although constrained to fight his childhood friend Walter. Echoing (not necessarily consciously) the historical transition from a unitary empire to warring post-Carolingian states, the Waltharius has the Franks, the Burgundians, and the Aquitanians living in close alliance and friendship at the start of the poem, only to become enemies with Walter’s escape from the Hunnic court. At this point, Gunther makes the decision to pursue Walter for his treasure. While the poem’s eponymous protagonist is Aquitanian, and his betrothed is Burgundian, his two adversaries are Frankish. The latter, accompanied by a small Frankish troop, unsuccessfully try to wrest the treasure and

Why Is There Something Rather Than Nothing?  165 Hildegund from Walter. In the end, the three warriors seem to be reconciled as they make ironic comments on each other’s disabilities inflicted during the climactic three-way duel. If those for whom this poem was intended (be they early ninth-century nobles or late ninth-century monks) thought of themselves as Franks, it may seem strange that the hero of the poem is not a Frank, while his antagonists are, one of them proving to be a thoroughly ridiculous individual.199 Then again, the pervasive humour in the text destabilizes any audience’s attempt at identification. In fact, the Franks are the object of all sorts of irony. Walter refers to a troop of Franks he recognizes in the distance with relief as Franci nebulones: ‘These are not Avars here, but Frankish Nibelungs’ (Waltharius 555). The phrase, as Jan Ziolkowski argues, plays on the interlingual meanings of nebulones – in Latin, it is a rare word meaning ‘bandits, scoundrels’, while in Germanic, it can be taken to refer to the Nibelung family which plays a crucial role in the Nibelung cycle (the Gibichungs become, by the time the Nibelungenlied was written down, the Nibelungs). Walter’s reaction to spotting Franks is a mixture of relief and disparagement that he and Hildegund are approached not by Huns but by mere Franks, and it belongs to a complex of at best ambivalent observations and insinuations about the Franks in the poem.200 The stage for the joke is set earlier since in line 505, Walter warns her to keep her eyes peeled for a cloud of dust (nebula).201 The audience would have known already that the Nibelung Franks were the real threat. And ‘to the same complex of Frank speaking – or frank ­punning – belongs the name of Eleuthir’, one of Walter’s adversaries: his name means ‘free’ in Greek, and as I mentioned earlier, one of the meanings of ‘Frank’ was ‘free man’.202 The Franks in the poem are thus ‘shadow people’ and ‘scoundrels’. But who would mock the Franks in a poem composed and read on Frankish territory? Could it be people to the east of the Rhine before they began to consider themselves the true Franks? Could it be east Frankish nobles with a sense of humour in the time of Louis the German or later? Indeed, self-deprecation can strengthen one’s belonging to the same group. In this case, the role of the Franks in the Waltharius would be that of an elaborate inside joke. One can also construct a sense of ethnic identity by laughing at one’s own group (perhaps this is the case with the Paris Conversations, too). Yet no one escapes the biting satire of the Waltharius poet. For, in Dennis Kratz’s opinion, the Waltharius is an epic that has no hero since ‘the poet has taken the traditional function of epic, the celebration of heroic excellence, and inverted it to emphasize instead the vitia which prevent Walter, Gunther and Hagen from being heroic’.203 But even if we take Rachel Stone’s much more straight-faced reading of heroism in the poem, Walter is still far from being unproblematic.204 After all, he too acts out of greed and an inflated sense of honour. In the denouement, he appears just as ridiculous as the other two ‘heroes’ with their significant

166  Why Is There Something Rather Than Nothing? reciprocal maimings.205 As we shall see in the following section, ethnicity is only one of the facets of identities problematized in the Waltharius: the machismo of the warrior elite is certainly another one. Yet this insistent irony is what has made many scholars miss the point this poem must have had for its original audience: One gets the sense that the poet is not entirely serious when using the commonplaces of epic narratives dealing with war; whether this sort of parodying is intended as critique, however, and what value we are to give such a critique, is far from clear.206 Indeed, it seems as if the Waltharius is expected to do one of two things: either to have been ‘propaganda’ and thus upheld an ethos (as Rachel Stone argues) or else to have critiqued it (Dennis Kratz, among others). Yet humour can both critique an ethos and affirm it at the same time, and even if ‘it is difficult to see Waltharius as contributing to any sort of political programme’, this does not mean it did not have socio-political significance.207 Instead, I would suggest that the Waltharius and other vernacular and/or Germanic poems provided a sandbox for playing with identities, not a political programme but a ludic space where constructions of identity could be brought into question and the tensions between and within them could be resolved or sublimated into laughter. Thus a better question than ‘is the poem ironic or heroic (or not)?’ would be ‘what are the issues that the poem is playing with and how do they relate to the tensions encountered in contiguous discourses and life structures?’ We do not need to localize the poem with precision to see that the same questions that Carolingian intellectuals were asking about ethnicity are asked and answered in different form in the Waltharius and the Hildebrandslied. The Waltharius projects the three ethnies into a legendary ‘Teutonic’ past. It is true that the text does not directly use ‘Teutons’, but it does talk about Huns, Franks, Aquitanians, Burgundians (Goths, in the case of the Hildebrandslied). The Waltharius itself is only the earliest attestation of a constellation of characters and narratives that have a long medieval career: Hagen and Gunther in the Siegfried legends, Walter himself in a very popular Romance epic cycle. Compared to all these other narratives, the Waltharius is remarkable in that no other version insists so much on the ethnicities of their characters and certain associations between the ethnies in the poem that are encountered nowhere else. Thus in the Germanic vernaculars, the Walter legend is invariably brought into some kind of contact with the Burgundian legend (though the reverse is not also true); in Romance traditions, this is not the case.208 For instance, Gunther is Burgundian in all the other attestations of the legend, including in the fifth-century chronicle by which the later ‘fall of the Burgundians’ narrative were inspired. Only in the Waltharius are

Why Is There Something Rather Than Nothing?  167 the Gibichungs Franks, not Burgundians.209 As Dronke shows, most of the other Walter legends do not contain two but only one pursuit and only one set of enemies (those from Attila’s court).210 In the OE Waldere, two fragments of a larger alliterative poem on the same matter, the Ostrogothic king Theodoric is mentioned (as in the Hildebrandslied), but not in the Waltharius. Could this be an attempt to steer clear of Theodoric (given the controversial nature of his memory in Carolingian circles) and talk about ‘the other Goths?’ While in the Waltharius, the protagonist is clearly identified as an Aquitainian, this can be merely another way of referring to a Visigoth. Thus, both Jordanes and Gregory of Tours inform us that the Visigoths (then ruling in Aquitaine) and the Franks fought the Huns under Attila in 451; significantly, Jordanes also has Burgundians in his list of Roman allies fighting the Huns.211 The three peoples’ confederations against the Huns thus coincide perfectly in the two records, with the exception of the Aquitanians of the Carolingian poem – identified as Visigoths in the earlier historiographical sources. As Ghosh explains, Walter could reflect the Visigothic king Vallia who ruled in Aquitaine for three years, although ‘beyond the first element of the name, there is no discernible connection between Waltharius and Vallia’.212 Be that as it may, the connections found in these early sources are not to be taken lightly since the information contained therein would have probably been part of the historiographical background of most of the poem’s educated audience. Being an Aquitainian need not have meant anything more than being from Aquitaine, while the Visigoths were a well-known people who had their own kingdom that included the latter territory for a time. As such, Walter could well have been simultaneously seen by the audience of the poem as an Aquitainian (in geographical terms) and a Visigoth (in ethnic terms). Walter being perceived as a sort of honorary Goth would mesh well with the fascination of Carolingian intellectuals for Gothic origins and narratives explored in the previous section. All this evidence points to the fact that the insistence on ethnicity is singular to the Waltharius – in other related narratives, the ethnicity of the characters is superfluous to the narrative as such. The complicated relationships between people of different ethnies in the poem could have been a way to problematize Charlemagne’s multi-ethnic empire and the emergence of regional identities that went hand in hand with the civil wars that followed. The tensions between the historiographical accounts, other potential versions of the legend, and the poem itself are consonant with the tensions between the emerging regional identities and the all-encompassing political rather than ethnical ‘Frankish’ identity under Charlemagne. Already, Freculph attests to conflicting origin stories about the Franks, as noted in the previous section. Vernacular or heroic poems such as the Waltharius, placed these tensions in a legendary Teutonic past, where they lost some of their urgent political contingencies.

168  Why Is There Something Rather Than Nothing? From the aforementioned, we can conclude that at least some elite Frankish textual communities included the Goths as a closely related people, their language as essentially the same Theotiscum sermonem that the Goths spoke, and their wide knowledge and dissemination of narratives concerning legendary Goths or Burgundians, such as Ermanaric, Hildebrand, and (probably) Walter. There are Huns in all of these narratives, perhaps playing the shadowy Other to warriors who are, if not Gothic, at least Teutonic (which, according to Walahfrid Strabo, is more or less the same thing). Thus in the Hildebrandslied, Hildebrand’s lord, Theodoric, is engaged in war with a vague ‘lord of the Huns’. More significantly, Hadubrand, not recognizing his father and thinking the old warrior wants to buy him with treasure, calls him insultingly ‘old Hun’ (Hildebrandslied, line 39). Meanwhile, in the Waltharius, the Huns are the natural enemies of all peoples west of the Rhine, while Attila, the Hun par excellence, is an apparently ferocious but actually effeminate leader made ridiculous by his complaints about a hangover. The Waltharius characteristically makes fun of what other discourses present as threatening or glorious. In this case, could the appearance of the Huns in both poems give voice to reactions towards the real-world pressure from the eastern neighbours of the Carolingian Empire (later of Louis the German’s kingdom)? A clue may be found in the identification of Huns with Avars in the Waltharius (the former ethnonym is used seven times, the latter twice). The Franks had indeed fought another people from Pannonia, the Avars, in the 560s, and more recently under Charlemagne. The identification of Avars as Huns was a very old one, occurring in Gregory of Tours, while the late-ninthcentury historical poem by an otherwise anonymous Poeta Saxo refers to the Avars as Huns, stating that they had also fought against the Franks under Attila.213 These heroic poems thus play with this anxiety about eastern neighbours, which the Frankish polities needed in order to justify the ‘border mentality’ that arose in the eastern Frankish territories (more about this in the following section). The Huns/Avars are thus emblematic of the border regions (marcha), zones of ethnic mixing and tension that ‘Tim Reuter called the Wild East, where Franks interacted with, and even shaded off into, assorted Others’.214 Like other Carolingian rulers before him, Louis the German waged campaigns almost every year in a continual effort to establish overlordship over neighbouring pagan ‘barbarians’: Danes, Obodrites, Linones, Sorbs, Dalaminzi, Bohemians, Moravians, and other peoples.215 The Waltharius, in which Attila and his retainers cut a rather ridiculous figure, could have been exorcising not only inter-regional identitarian tensions but also anxieties about eastern invaders. Waltharius is the perfect proof that a Carolingian intellectual could be as well-versed in Virgil and Statius as in the workings of the vernacular ‘heroic’ tradition, employing both in creating a much more oblique commentary on ethnicity, language, and masculinity than most of the

Why Is There Something Rather Than Nothing?  169 discourses explored so far in this chapter. Behind the preoccupation with ethnicity present in these texts, one can sense the contemporary negotiations of such categories and identities as I explore here. Thus the Latinate sources I have mostly dealt with in this section help frame these negotiations that must have been taking place within and between the textual communities in which texts such as the Hildebrandslied emerged. Indeed, heroes such as Walter, Hildebrand, or Ermanaric would not have been interesting enough for the social and cultural milieu associated with the written texts that contain them without the existence of the ethnic conceptions I have traced so far. The fact that they were all conceivably seen as belonging to a legendary ‘Teutonic’ past made them attractive to certain elite Germanic-speaking textual communities since they could problematize issues of ethnicity, language, and, as we shall see in the following, masculinity and nobility.

How to Be a Man, a Noble, and a Son in Carolingian Society The previous sections have hopefully at least partly unravelled the rich and complex texture of social, political, ethnic, and linguistic tensions from which emerged the constellation of verse histories under scrutiny in this book. In the following, I explore other threads of this texture, which are inextricably linked to the others: masculinity and nobility, setting the ground for similar explorations of OE material in the next chapter. While the subjects dealt with so far (the emergence of ethnicities in ninth-­century Frankish territories and its relations to new perceptions of language and new attitudes towards vernacular poetic traditions) have been comparatively understudied, Carolingian constructions of masculinity and nobility have been the subject of an impressive growth in scholarship in the past two decades.216 I do not aim to review all the literature on these subjects since I am interested in masculinity and nobility only insofar as they are facets of identity that were problematized in Carolingian vernacular and/or heroic poetry. In other words, I explore here only masculinity (and nobility, from which it is virtually indistinguishable in certain discourses, as will be suggested) as a locus of confrontation and anxiety which was given voice in (but also shaped by) texts like the Waltharius. In relation to the facets of identity explored so far, I thus argue here that in the context of the Carolingian ninth-century, ethnic discourses and practices are gendered, socially localized, and use both vernacular languages and orally-derived poetic traditions in complex ways. Previously, I have argued that Otfrid’s construction (among others) of eastern Frankish aristocratic identity is bound up not only with the Germanic vernacular but also a kind of aggressive martiality. Indeed, Eric Goldberg superbly demonstrates that the militaristic political culture at Louis the German’s court was a purposely built and arduously sustained construction of not merely kingly authority but of a new identity for the

170  Why Is There Something Rather Than Nothing? elite of a novel political community: the regnum orientalium Francorum, the ‘Kingdom of the East Franks’.217 From chronicles to martial ritual, from aristocratic pastimes to courtly ceremonial, all discourses and practices gave voice to a new kind of identity into which were woven ‘the military virtues of strength and bravery, the hereditary honor of bearing arms and riding war-horses and the legitimizing of warfare and violence in Christian terms’.218 But while many of the discourses Goldberg explores do refer primarily to Louis himself, they did not aim to set the king apart from his nobles but rather to foster ‘a shared aristocratic ruling ethos and gradations of internal hierarchy among them’.219 Described by chroniclers like Notker as the ‘dress and equipment of the Old Franks’, Louis’s attire emphasized ‘his membership to the Frankish warrior nobility at large, thus presenting him more as primus inter pares rather than an exalted theocratic monarch adorned with gold and precious stones’.220 Indeed, Louis’s magnates were referred to as amici regis, the ‘king’s friends’, for they gave Louis counsel and led their retinues with him into battle, and he bound them to him with ties of ‘friendship’ (amicitia) and oaths of fidelity.221 These oaths would often have been sworn in the Germanic vernacular, no doubt because a considerable part of the elite would have spoken it but also due to the politics of language that was promoted by the likes of Otfrid.222 In this picture of the novel elite identity emerging in the east Frankish kingdom can be discerned the convergence of a certain type of masculinity, the Germanic vernaculars, the valuation of martial prowess, and a new kind of nobility shared with the king, bound by mutual loyalty. This co-evolution of masculinity, nobility, and ethnic identity in the east Frankish kingdom, however, has roots that go deeper than the rivalry with Charles the Bald’s western polity. One point of origin can be ascertained in the civil wars of the previous decades when it was precisely the lack of such a clear model (or rather the presence of conflicting models) that lead to all-pervasive anxieties about what it meant to be a man, a warrior, a noble but also a loyal retainer faithful to both kin and lord. It is these social forces which are negotiated and problematized in texts such as the Waltharius and the Hildebrandslied in ways that could not be done in other genres or types of discourses available in that socio-cultural matrix. Thus this specific type of poetry emerged in this specific context because it was able to provide a space for thinking about these issues that did not exist in hagiographies, sermons, or chronicles. This fact will become even plainer in the next chapter, where I review some of the OE material from a similar perspective. It is true that these types of texts are just as marked by these issues as ‘heroic’ poetry, but the biting satire of warrior culture coexisting with an avid interest in bloody duels in the Waltharius or the tragic tension between saving face and preserving the life of kin in the Hildebrandslied simply did not have a place in other officially sanctioned discourses, although they were social and cultural

Why Is There Something Rather Than Nothing?  171 realities lived by a large number of Carolingian nobles. In the following, I briefly look at how such realities as conflicting loyalties; fatherhood issues; a martial culture based on honour, violence, and love of loot; and contradictory ideals of manliness were negotiated in vernacular heroic poetry in relation to other kinds of contemporary discourses. The novelty of the solution that was provided by the construction of a Germanic masculinist elite identity under Louis the German can be perceived more sharply when contrasted to the problems to which it responded. These problems are, on the one hand, the dissolution of the empire starting under Louis the Pious, and on the other, they are the crisis of masculinity that Janet Nelson has famously argued for. She explains that in times of social stress, such as the Carolingian mid-ninth century, gender identity comes under pressure and has to be rethought and redefined; hence, many male aristocrats began to perceive ‘existing social arrangements as dysfunctional, and…themselves as subjected to dissonant social messages and conflicting imperatives’.223 These dissonant messages were, on the one hand, that of the traditional construction of male nobility as determined by martial prowess and on the other that of the insistence of Carolingian discourses advocating a universal conformity to quasi-monastic practices and ideals but also an ‘imperial peace’ that was fitting for a unified empire that had subdued all its internal enemies.224 As to the dissolution of this short-lived unified empire, Timothy Reuter argues that it did not simply result from bad decisions regarding the distribution of the territories to Louis’s sons but that it had to do more with his perceived lack of support for his aristocracy.225 Louis, fittingly for his persona of a pious, almost monkish ruler, attempted to implement a set of reforms that involved a ban on further transfers of property from the church to his nobles and succeeded in antagonizing both the majority of the aristocracy and the smaller (but vocal and influential) party of clerics who wanted the reforms to be more radical. Thus the new identity available to east Frankish nobles under Louis the German provided an answer both to the anxiety deriving from opposing constructions of masculinity and to the breaking of the bonds between king and nobles. Yet these issues remained real problems throughout the ninth century for many others, as we shall see; east Frankish martial nobility provided one solution to these conundrums but caused others in its turn, and in any case, not all Frankish nobles could choose to take up this model. Thus I am not arguing for a precise dating of heroic poems to the reign of one Louis or the other in relation to the evolutions described here. Rather, I aim to bring to light the socio-cultural and political forces which did evolve and change during the Carolingian ninth-century but which remained real and topical enough throughout that era for texts such as the Waltharius to pick up on them without being confined to a specific point in time. Thus I do not dwell in this section on the extent to which

172  Why Is There Something Rather Than Nothing? Carolingian heroic poems reflected the real life of aristocracy: there is plenty of excellent scholarship on the essential violence and strenuous physicality of aristocratic life and education, on the Männerbund that the elite martial life engendered, and on the reality of boasting, honourseeking, and looting in Carolingian warfare.226 What interests me here is what is usually left unsaid by poets and preachers alike, not intentionally so, but because it is usually the people living under the action of sociocultural lines of force that are least capable of making them explicit. Coming back to the dissolution of the empire, we have seen that it was not an event following from a single decision of one or another political actor but from a range of social and political vectors that transcend any individual action. Thus Reuter argues that the portioning of the kingdom only ‘brought the underlying instability of Frankish politics into the open’, which stemmed not so much from ‘ideologically determined political groupings, as from their absence’.227 Thus the civil wars were not conflicts between clearly delineated factions, and as such, very few people ‘exposed themselves to such an extent that they could not change with the political wind’, which meant that there was no incentive for either a peaceful resolution or for stable loyalties – in Reuter’s words, ‘whenever a settlement appeared to have been reached, there were too many people with an interest in overturning it for it to become a point of equilibrium’.228 These shifting alliances and their contradictory underlying reasons are the social realities that Hagen embodies when he finally decides to fight Walter, his childhood friend. To Gunther, whom he grudgingly helps when the king flees the battle after Walter kills all his men, he explains with biting irony that ‘since I see that you are suffering more from shame/ than from grief at the carnage… my own distress yields to the honour/ of the king’ (Waltharius 1107–10).229 To himself, Hagen explains his decision as necessary for ‘his own reputation/ for courage, which perhaps might be cheapened as a result,/ if he should spare himself in any way in these matters’ (Waltharius 1094–96). When facing Walter, he attributes his decision to the duty to take revenge for his nephew’s death (Waltharius 1272–78). These contradictory narratives that Hagen offers within the space of 200 lines must have been directly relevant to the experience of many Frankish nobles during the civil wars: ‘A conflict between several different bonds of loyalty, all of which should, if all had been well with the world, have been mutually reinforcing, but now no longer were: namely, lordship, kinship, and friendship’.230 Cases such as that of Adalhard, a powerful magnate at the court of Louis the Pious who decided to support Charles, were common in such times. In his letter to Ermengard, the wife of Lothar I, Adalhard defends himself against accusations of disloyalty to Lothar by emphasizing his concern to defend his ‘rights’ (presumably his properties) but above all to defend his honour and reputation.231 Such abstractions provided a

Why Is There Something Rather Than Nothing?  173 culturally-appropriate rhetorical package for underlying tensions between competing social ideals. Stuart Airlie masterfully shows how Nithard’s portrayal of the disintegration of social structures (of royal authority and aristocracy) during the civil war is based on biting satire, caricatures of political actors, and hyperboles of their failings through which he ‘highlights what he sees as the glaring discrepancy between ideal and reality in the behaviours of the aristocracy’.232 Thus satire (which the Waltharius is very rich in) was indeed a fitting mode for voicing the breakdown of society and identities during this time. As Guy Halsall notes, ‘Very serious points can be made through satire, irony and ridicule. To say that a passage in the sources is satirical or ironic is not to denude it of serious content’.233 Hence the uneasiness of recent appraisals of the Waltharius to acknowledge its rich ironies is unwarranted since the text is not simply chastising traditional aristocratic masculinity but also problematizing the cultural tensions of the milieu from which it emerged. The ironies do not make it any more light-hearted than Nithard’s bleak account but, in fact, underline the sombreness of these social realities. Another such reality was that the borders between these ever-shifting warring factions often separated families so that at one point or another, due to the close interconnections within the nobility, the political rivals that some Frankish nobles needed to fight could also be their kin.234 This reality is problematized in the Waltharius under the form of the duels between Gunther’s men and the protagonist, all of which are driven by revenge for fallen relatives or comrades (Waltharius 686–701, 1239–61, 1268–78). During the fighting in the Vosges, each individual duel is triggered by the previous one, and the drive to preserve one’s honour and the attempt to be faithful to both kin and lord are presented as a fatally interlocking system of loyalties that cause the death of all but the three main characters. That hierarchizing one’s loyalties was a serious matter for male nobles can be gleaned from Dhuoda’s Liber Manualis, a handbook of advice written for her son, William. Dhuoda’s view was that William’s loyalty to his father came before that to the king.235 While this opinion was not one that rulers would have appreciated, it was what William ended up choosing in his own circumstances. For, the life of Dhuoda’s family was a real-life instantiation of the same conflicting loyalties, oathbreaking, and revenge across generations one would expect to read about in the Waltharius – a reminder that social reality and poetic accounts are just as substantially real since they share the same ‘semiotic codes that govern the representation of life both in writing and in incorporated social structures’.236 Bernard of Septimania, Dhuoda’s husband, was the chamberlain of Louis the Pious and one of the most powerful magnates of the realm.237 Bernard was caught in the family dispute between Louis and his sons, and his rumoured affair with Judith, the king’s second wife, and his personal warlike style made him hated by a number of political figures. After

174  Why Is There Something Rather Than Nothing? several spectacular turns of events, which had Bernard and his family exiled then coming back into power, he was forced to hand his teenage son William to Charles the Bald as a hostage. This is referred to in Dhuoda’s text in Book 3.4, where she asks William to hold fast to Charles. Appropriately (but also ironically in the light of what was to follow), this comes right after the section of the book dedicated to being faithful to one’s father. The bitter irony lay in the way in which these injunctions played out in the historical events in which William’s life was enmeshed early on. In 844, Bernard was found guilty of treason and executed on Charles’s orders. William probably witnessed his execution, and in any case, he was soon planning revenge. The bond to one’s lord was lifelong, but it could be broken if a lord tried to kill the vassal, strike him with a rod, violate his wife or commit adultery with her, fall on him with a raised sword, etc. 238 Within months, William and his followers ambushed Charles’s army on its way south. He won this battle, but he pursued his revenge further, going so far as to seek help from the emir Abd-er-Rahman II in 848, when William besieged Barcelona, the paternal stronghold. Soon, being betrayed, he was either killed in battle or executed by Charles the Bald. The search for revenge went on, however, with Dhuoda’s younger son, Bernard, who lived on to harass Carolingian authorities and even tried to ambush Charles in 864. Obviously, conflicting loyalties between kinship, friendship, and lordship make up the prime matter for poetry in the Waltharius. Yet the conflict is perhaps even more poignantly given voice in the Hildebrandslied, which pits the eponymous protagonist, one of King Theodoric’s best warriors, against his son, Hadubrand, which does not recognize him. They come to blows (only verbally at this point), and Hilderand tries to placate Hadubrand with gifts. When his son rejects these as trickeries, Hildebrand chooses to fight. This is not a decision he makes lightly, yet here saving face and preserving one’s honour are paramount. That one should face his own son in battle was the nightmarish possibility which many Frankish nobles in the turmoil of the 830s and 840s would have kept at the back of their minds. The Hildebrandslied exorcises this potential conflict between father and son (and generally between relatives) by bringing it into the limelight of a narrative that is ostensibly presented as an old traditional composition. What the Waltharius does through irony and humour, as well as through the choice of Latin epic as mode of narration, the Hildebrandslied achieves through the rhetoric of oral vernacular poetry – both strategies providing the necessary distance for the tensions to be acted upon in a manner non-threatening to the audience in a ludic space. Fittingly, the fragment that was preserved from the poem ends before the conclusion of the duel – the social tensions within Carolingian society could only be postponed, not avoided.

Why Is There Something Rather Than Nothing?  175 The Hildebrandslied opens up another conversation that seems to have been highly relevant in Carolingian society: father-son relationships. Indeed, the poem employs the hapax sunufatarungo, which literally means ‘father and son’, thus presenting a seemingly unbreakable unity between them (since they are combined in the same word), yet in fact placing them in opposition as contenders in a combat to the death. In any case, conflicted relationships between fathers and sons are recurrent in Carolingian aristocracy and royalty. One only needs to remember that the civil wars opposed father and son (Louis the Pious and Lothar) against the other two sons (Louis the German and Charles the Bald). An intergenerational conflict can be glimpsed at also in the Waltharius, between, on the one hand, the three Germanic kings, who, in proper Carolingian fashion, favour diplomacy and securing peace at any cost over useless heroism on the battlefield and on the other hand, the younger generation, who seem to embrace conflict on all occasions, with disastrous results.239 Could this detail be voicing an older generation’s commentary on a newfangled macho aristocratic identity, bound up with readings of ‘Teutonic books’ that youth were adopting? Gunther’s decision to recover the treasure and thus attack Walter is motivated by the desire to make up for his father’s failings (namely, giving it as tribute to the Huns). This obsession with making up for the past and especially for the mistakes of the father is what unleashes the disaster.240 Fatherhood weighs heavily on Gunther’s mind since he rejects Hagen’s advice not to attack Walter and humiliates him in front of his men in the worst possible way, by insulting his family honour: ‘I see you take after your father Hagathie./ He too carried a timorous heart under a chill breast,/ And with many words shrank from fighting’ (Waltharius 629–31). As Rio notes, ‘The accusation is all the more unfair as it had of course been Gunther’s own father who had shown cowardice by handing over the treasure to the Huns in the first place’.241 When Gunther flees the battlefield and calls for Hagen’s help, the latter ironically reminds him of his insult by reclaiming it as an impediment to fight: My line of shameful ancestors prevents me from warring, And a chill blood robs me of any heart for arms. For my father melted when he saw weapons, And, timid, with many words declined to fight. When you hurled these words, king, among your followers, Our help obviously seemed unworthy to you. (Waltharius 1067–72) Of course, it is an ironic rejoinder that wittily turns Gunther’s insult against him. But the implication here is that a cowardly son can only issue from a cowardly father; hence, perhaps Hagen insinuates that it is

176  Why Is There Something Rather Than Nothing? this relationship that intimately binds the king and his father. In any case, the anxiety about the relationship between father and son (whether it is about being a good son, rising to the father’s level or failing to do so, avenging one’s father, or at least making up for his sins) is an important but understudied component of the crisis of masculinity. In any case, the crisis of identity Janet Nelson masterfully argues for is especially visible in contemporary descriptions of the experience of adolescent boys – in her felicitous formula, ‘anxious young men’. Nelson discusses Gerald of Aurillac and Odo of Cluny, whose fathers changed their minds several times about whether to make them oblates or to educate them as young warrior nobles.242 But the most striking example of aristocratic masculinity in crisis is that of Charles, the youngest son of Louis the German. Welcomed as emperor in 883, ten years earlier, he had ‘performed an extraordinary act of role rejection’: at an assembly in his father’s palace of Frankfurt, Charles leaped up and said that he wished to abandon the world and would not touch his wife in carnal intercourse. He took his sword from its belt and let it fall to the ground. As he tried to undo his sword belt and take off his princely clothing, he began to shake violently.243 Nelson asks what made him suddenly, at 33, reject the life of a prince and points out that contemporary annalists suggested that he felt guilt at rebelling against his father, with was, as she wryly notes, ‘almost a necessary condition of being a Carolingian’s son’. 244 Evidently, from the cases presented, he was not the only one to feel the burden of being a son in Carolingian times. While Charles’s rejection of one dominant embodiment of Carolingian masculinity was perhaps the most spectacular (and the most visible in sources), the tension that made it possible was conceivably widespread throughout the aristocracy, even if it manifested outwardly only in indecision about whether to follow a monastic or a political-martial route, whether to boast about one’s deeds on the battlefield or to repent for killing people. The eponymous protagonist of the Waltharius is also haunted by indecision. His strange behaviour just before the battle with Gunther’s men is a case in point. After boasting that no Frank would take his treasure and live to tell his wife about it, he falls to the ground begging for forgiveness (Waltharius 564–65). It is the same tension between the need to repent and the need to kill that plays out in the scene where Walter reattaches the heads to the corpses and prays for his victims (Waltharius 1157–67). As Rio suggests, ‘Perhaps what he needed to apologize for was less his arrogance than the flippancy with which he had treated the prospective massacre of his enemies’.245 I agree with her that we do not need to choose between seeing this scene as the denunciation of ‘the little hypocrisies of

Why Is There Something Rather Than Nothing?  177 life in the world of Carolingian warfare’ or as a sympathizing account of warriors torn between conflicting socio-cultural forces – and not only because the text was open and different members of the audience could read the text very differently. The intentions of the poem are also hard to pin down because the pervading irony of the text creates a less-than-real space where the moral thrust of such actions dilutes against the background of bumbling heroics, which thus are in no immediate need to be approved or condemned. Rather, these ethical issues were allowed to play out all their consequences, even though they may appear to us both grotesquely hilarious and tragically serious. Yet, the effect of reading or listening to the Waltharius could have been neither bewildering nor simply didactic, but cathartic. Again, heroic poems and social reality prove to be elements of the same web of meaning that were only instantiated in different discourses and practices. I take this reading of the emotions foregrounded in (and projected on) these rag-picker histories to another level in the next chapter. It has become evident from the previous that vernacular (or vernacular tradition-inspired) heroic poetry provided a space for audiences to negotiate and problematize conflicting models of masculinity and nobility in Carolingian society. The discourses explored here emerged in a time of political and social crisis and responded to the burning questions that haunted male nobles throughout the empire (or warring polities): ‘How can one choose between conflicting loyalties between one’s relatives, friends, and lord?’ ‘How can one reconcile the traditional masculinist warrior identity of an aristocrat with the new constructions of lay nobility as being quasi-monastic?’ ‘Should one pay for the sins of one’s father?’ Thus texts like the Waltharius or the Hildebrandslied were neither exercises in antiquarianism, nor simple entertainment – nor were they moralizing sermons. They were cultural products responding to a set of stringent social anxieties through a novel type of discourse.

Tying the Threads Together: The Background to the Figure in the Carpet The present chapter has hopefully shown that the figure in the carpet (verse histories like the Waltharius or the Hildebrandslied) could not have existed without its material support, the weave contexted together from the warp of dominant discourses on masculinity, nobility, language, and ethnicity, as well as the weft of individual or localized practices and competing narratives and traditions. This is the reason why I have explored the figure (GHP) not formally or stylistically but in its interactions with the background of the carpet and with the entire social, political, and ideological loom that provided the mechanism for weaving. Of course, the metaphor breaks down when we take into account the interactions between discourses and social, political, and emotional

178  Why Is There Something Rather Than Nothing? realities – each is shaped by the others, with no discernible genealogies to be found. What I hope to have provided for vernacular and/or heroic discourses in the Carolingian ninth century is an understanding of the texts’ ‘social logic, [their] site within a highly particularized and local social environment’.246 To return to the question in the title, ‘Why is there something (vernacular verse histories labelled as GHP) rather than nothing?’ in the Carolingian era, I have suggested that it came about at the intersection of cultural and social demands because it provided a space for the expression, confrontation, and shaping of emerging identities. Texts such as the Heliand or the Hildebrandslied were composed not because of any inevitable organic evolution towards literarization of orally circulating ‘texts’. Writing down vernacular orally derived or inspired poetry is neither accidental nor ‘natural’ – it is a choice, for the availability of a language does not automatically entail the apparition of literature in that language, as the example of Italy given by Banniard shows (it did not have a vernacular literature until the thirteenth century, although it did have a vernacular language much earlier).247 The oral poetic traditions which probably existed at some point before these texts were composed provided a quarry of narrative material and poetic form which was adopted and adapted to various needs by various textual communities in search of a mode of discourse that would fit their needs and give voice to their concerns and preoccupations. The ‘Germanic’ story-world associated with this tradition of verse was not so much a vehicle for preserving the identity of a people or even a Traditionskern group but rather a rhizome of narratives which underlay all vernacular poetry of the period without it being preserved faithfully in any one text. Thus texts such as the Waltharius or the Heliand are part of the same textual constellation, not by virtue of their being part of the same genre (particularly the scholarly construction of GHP) or their being genetically linked to oral tradition but because they are sui generis responses to or articulations of the same deep-seated concerns in Carolingian society. As such, the texts explored here have been shown to be firmly grounded in the social texture of the time and space in which they were written down. The emergence of these texts can be linked to privileged groups representing a small minority of Carolingian and early medieval English society communities with access to both the technologies of writing and the leisure to write down these ‘useless’ (in the view of dominant discourses) texts, marked by a desire not merely to preserve them but also to make them accessible to other members of the same milieu. In performing these sustained actions, they acted much like rag-picker poet-historians, gathering the refuse of a maligned type of discourse and fashioned out of it various new assemblages in response to the various socio-­cultural, ideological, and emotional needs – verse histories, in their variety of modes explored in Chapter 1.

Why Is There Something Rather Than Nothing?  179 In the case of Carolingian Francia, these groups can be localized as Germanic-speaking aristocratic communities in the eastern Frankish territories in a time of shifting linguistic, ethnic, political, social, and gender identities. The first phase in these evolutions is marked by the civil wars of the 830s through to the 850s, with their ever-changing loyalties pitting kin against class through conflicting models of masculinity and nobility and by a growing awareness of linguistic differences between the vernaculars and Latin, as well as, on the other hand, a kinship between speakers of Germanic dialects. The vernacular heroic poetry written down even after this period would have given voice to the anxieties resulting from attempting to reconcile these opposing forces. Later, starting in the late 840s during the reign of Louis the German over an independent east Frankish polity (but continuing after the 860s), nobility was constructed through discourses and practices that emphasized martial prowess, speaking the Germanic vernacular, and belonging to an elite connected to the emperor in mutual ‘friendship’ and loyalty – a construction of gendered and socially localized ethnicity as ‘true Frankish’. Yet the forces acting in the previous phase did not cease to act upon individuals and mentalities – the same tensions could still have been voiced in vernacular poetry. Still, a new emphasis on ethnicity (both the Teutonic and the new Frankish mutually excluding constructions in the eastern and western kingdoms) is characteristic of this era. Thus the texts explored here could be seen not only as testimonies to the changes in how Carolingian textual communities perceived masculinity, heroism, language, and ethnic identity but also as forces that shaped these evolutions. My approach in exploring Frankish vernacular verse histories has involved looking at the less visible forces that act on individuals and their actions – less visible in the textual record because they are not named as such; they are, however, noticeable in silences, in absences, in an insistence on certain details unwarranted by the narrative flow, and in the use of certain poetic forms and certain words. Like dark matter in the immensity of space, the social, cultural, and emotional forces that acted in early medieval societies can be detected not by direct observation but through the pull they exerted on individuals and groups, manifested in discourse through the choice of genre, shape of texts, poetic form, and practice through political actions, rituals, gestures, and education. Yet these same forces were shaped by and through these discourses and practices. For it has become clear that indeed, ‘texts both mirror and generate social realities, are constituted by and constitute the social and discursive formations which they may sustain, resist, contest, or seek to transform’.248 This view, which has guided the present endeavour, also provides a way out of scholarly conundrums revolving around the ethos the texts are supposed to embody or the values they are supposed to transmit. As I have suggested, the debates on whether the Waltharius

180  Why Is There Something Rather Than Nothing? was meant to critique or praise heroism or the Heliand aimed to instil Christian piety or to rhapsodize about Germanic heroism are not very useful for understanding to what extent vernacular and/or heroic verse shaped Carolingian social and cultural life by the novel solutions it provided to distinct cultural and social needs. For this type of discourse was neither simply a mirror held up to society nor a vehicle for ‘propaganda’ and thus meant to transmit a specific message (whether certain individuals in groups of people were to be loved or hated or certain deeds were reprehensible or honourable). Questions of gender, morality, loyalty, ethnicity, and language were of great interest in Carolingian society, and many felt the need to reflect on and play with them. Texts such as the Waltharius provided a ludic space that could be found neither in the moralistic medium of the speculum moralitatis nor in the sophisticate Latinate courtly poetry, neither in sermons nor in chronicles – although all these genres were permeated by the same wider societal forces. Neither could the oral tradition qua oral tradition provide this space since, although many nobles must have enjoyed it, as Richter shows, it would probably have been perceived as divertissement or in any case as a genre unfit for meditation on conflicting models of masculinity, for instance. On the other hand, I have opted against too localized readings of individual texts. Plenty of excellent scholarship advocates such linking of poems and events (to mention only Alice Rio’s recent reading of Waltharius as problematizing the aftermath of the Battle of Fontenoy), yet I have found that the texts as such rarely allow themselves to be pinned down in such neat political configurations.249 For while one set of circumstances or historical events or one person’s political decisions could certainly have influenced them, the texts under scrutiny here transcend both individual agendas and precise correspondences to events – they are neither propaganda nor romans à clef. Quite the opposite, they might have been representative of disruptive discourses competing with or at least complementing the official ones. In any case, I hope to have brought to light the passive resistance offered by the dominant culture and its hegemonic discourses against which these texts emerged (a point rarely acknowledged by previous scholarship). In this respect, I have tried to sail a steady course between the Scylla of cultural materialist political optimism (Raymond Williams’s ‘structures of feeling’ constantly produce new discourses which would make social and political continuity impossible) and the Charybdis of new historicist political pessimism (Foucault’s ‘discursive practices’ are virtually always seen as reinforcing the dominant ideology).250 I have proposed that while indeed most surviving sources are the manifestation of cultural and social hegemony, there is always the possibility for competing discourse to arise, especially with the texts looked at here. Vernacular heroic poetry was all things to all men in Carolingian Francia: for the creators of Cædmonian

Why Is There Something Rather Than Nothing?  181 verse histories such as OS Heliand or the OE Genesis poems or Judith, it was a vehicle for spreading the word of the gospel; for Otfrid and Louis the German, it was an embodiment of a new top-down discourse of Frankish aristocratic identity as gendered ethnicity clothed in the repurposed rhetoric of Frankish domination; while the Muspilli expressed royal concerns about justice (the centre expressing anxieties about its efficacy) in a language and poetic form reclaimed from the peripheries (both those of folk belief and apocryphal tradition). Verse histories in the bardic mode, like the Hildebrandslied, meanwhile, provided a stern meditation on conflicted loyalties and on the impossibility of inhabiting traditional definitions of masculinity in a changing world. Finally, the expression of the rag-picker mode in the Waltharius could be even more subversive precisely because of its Latinity (an unexpected range of critiques offered under the cover of a familiar and prestigious genre); it would have had the potential to unsettle into meditation both those who shared in the masculinist warrior culture and those who saw fit that all poetry should express Christian doctrine, while, of course, providing entertainment to all those who had enough of a sense of humour to transcend both. It is hard to imagine how all these anxieties, tensions, jokes, conversations, and projections could be coherently given voice in another type of discourse other than the newly invented (out of the scraps of oral tradition) vernacular verse history.

Notes 1 Scheil (2008: 281). 2 For the evolution of lordship, see Green (1965); for Carolingian masculinity, see Stone (2011); for pagan belief see Murdoch (2004a); for vernacular oral lays, see Murdoch (2004b); for pious sermon and Carolingian propaganda, see Innes (2000). 3 All the more since, as I will explain next, the actual process of writing these texts surely entailed much more than simply ‘setting down in writing’ presumed ready-made oral versions that would have been dictated or brought to memory. For the complex interactions between oral, orally derived, and literate poetic modes/literary forms, see Schaefer (1996). 4 Niles (2007, 15). 5 Liuzza (2005, 100). 6 Ziolkowski (2001, 34–6). 7 Niles (2007: 108). 8 McKitterick (1988: 7–13). See also Wright (1982). 9 Lees (1999: 4). For bona-fide heroism, see Stone (2011), for a critique of heroism, see Kratz (1980: 18–40). For ‘Germanic’, see Ziolkowski (2001: 34–6); for ‘Vergilian’, see Kratz (1980: 24–6 and 52–3); for ‘Christian’, see Kratz (1977: 130–37). 10 Niles (2007, 15). 11 Niles (2007: 57). 12 Murdoch (2004c: 19). 13 Niles (2007: 57). 14 Niles (2007: 49).

182  Why Is There Something Rather Than Nothing? 15 Spiegel (1990: 82). 16 de Jong (2000: 191). 17 Ferguson (1990: 11). 18 Stone (2011: 3). 19 Innes (2002: 155). 20 Godman (1987). See also McKitterick (1988: 229–32). 21 McKitterick (1988: 270, 268). 22 Booker (2009: 226). Ermoldus Nigellus (1932: II 1248–49, p. 96). 23 Booker (2009: 223). Collins (1998: 203). 24 Goldberg (2006: 167). 25 Alcuini Epistolae (1895: 97, p. 141). 26 Ermanrich (1899: 563). Eric Goldberg’s translation 27 Alcuini Epistolae (1895: 124, p. 183). 28 Edwards (1994: 141), quoting Haubrichs (1988: 42). See also Lee-Stecum (2001: 94). 29 Dutton (2004: 130). 30 Concilium Turonense (1906: 287). See also Richter (1994: 125). 31 Richter (1994: 126–45). 32 Richter (1994: 129–31). 33 Dutton (2004: 91). 34 Anthropologically, the former can be said to embody social capital, while the latter is often the carrier of emotional capital by association with childhood memories or the like (not that the two are mutually exclusive. For the notion of cultural (and other forms of) capital, see Bourdieu (1986). 35 Chartier (1982), p. 41. 36 Spiegel (1990: 68). 37 Niles (2007: 56). 38 For the main inspiration for this approach, see Spiegel (1990). For the notion of constellation, see Benjamin (1999: N2a,3 (p. 462), N1,9 (p. 458), H°,16 (p. 845) et passim). 39 Kratz (1980: 16–17). For a very recent and wide-ranging survey on the scholarship on the poetics (also supporting a ninth-century dating) of the poem, see Stone (2013). 40 Murdoch (2004b: 123). 41 Wells (2004: 161, 164). 42 Dant (1991: 7–8). 43 What I call ‘life structure’ is meant to encompass a number of phenomena, such as constructions of masculinity, kinship, ethnicity, mentalities, social and political institutions, and what Pierre Bourdieu popularized under the name habitus. See Bourdieu (1990: 66–7 et passim). 44 See, especially, McKitterick (1988), McKitterick (2005), and de Jong (2005). 45 McKitterick (1988). 46 McKitterick (1988: 13). 47 McKitterick (1988: 9). See also Wright (1982) and Richter (1983). 48 For the notion of acrolect and its use in the context of the earliest Romance and Germanic vernacular texts, see Banniard (2003: 39–41). 49 Compare with a similar evolution in thirteenth-century English aristocracy: ‘The women who served as nurses and “rockers” for noble children were often Anglophone, and in their early years, noble children might well speak English, at least until they started communicating with a French-speaking parent or parents’, in Jambeck (2005: 167). For the social status of nurses, compare the use of female slaves (especially Muslim women) as wet nurses in thirteenth-century Aragon, see Winer (2008). For nursing and motherhood in Carolingian society, see Butt (2002: 60).

Why Is There Something Rather Than Nothing?  183 50 Hellgardt (1996). 51 de Jong (2005: 107). 52 Banniard (2003: 29): ‘Un bouleversement important dans le champ culturel et langagier, générateur de conflits et de fluctuations’. 53 Rosenwein (2006: 1–31). See also the papers in Nagy and Boquet (2009). 54 Garizpanov (2008: 33). 55 Garizpanov (2008: 33). 56 Kershaw (2002: 201). 57 Idem. 58 Idem. 59 Idem. 60 Lee-Stecum (2001: 95). Lee-Stecum’s translation. 61 Banniard (2003: 31). 62 Banniard (2003: 33): Dum agrestis linguae inculta uerba inseruntur latinitatis planitiei, cachinnum legentibus praebent. 63 For OE, see Hollis (2012). For Irish, see Charles-Edwards (1998: 76, 77). 64 Spivak (1990: 381). See also bell hooks (1990a, 1990b) and Mortensen (2006: 262–69). Mortensen shows how hagiographies written in peripheral cultures in the high Middle Ages were authenticated by being as faithful as possible to master narratives coming from the cultural centre. See also Geary (2006). 65 Wells (2004: 159). 66 Green (2003). 67 Edwards (1994: 153). 68 Green (2003: 258). 69 Green (2003: 259). 70 Green (2003: 260). 71 Garizpanov (2008: 33). Councils and Ecclesiastical Documents (1879: 366). 72 Reuter (2005: 192). 73 For the categorization of counts, see Capitulare episcoporum (1883: 52). For the resentment of smaller aristocrats towards ‘supermagnates’ in the later part of the timeline the present chapter is dealing with (the times of Charles the Fat), see MacLean (2003: 116). 74 Halsall (2003: 12). 75 Banniard (2003). 76 Banniard (2003: 25). 77 Banniard (2003: 39): ‘non par l’acceptation résignée d’une scripta rustica (niveau basilectal), mais bien au contraire par la création d’un langage littéraire (construction d’un acrolecte)’. 78 Lee-Stecum (2001: 102–03). 79 Neville (2001: 120). 80 For Alcuin’s reproaches to both bishops and courtiers for amusing themselves with oral performers and vernacular songs, see Richter (1983: 131 ff). 81 For overviews of the issue, see Renoir (1988); Haymes (1976); Wolf (1988). For the Heliand, see Suzuki (2011); Haferland (2002a; 2002b). For the Muspilli, see Krogmann (1953). For the Hildebrandslied, see Tyler (1992) and Renoir (1977). For the Waltharius, see Hennessey Olsen (1993). 82 The classical accounts are Lord (1960) and Parry (1971). Scholarship in the Parry-Lord tradition is elaborated on in Foley (1985 and 1988). 83 Niles (2007: 148). For the first proposition of the continuum model, see O’Brien O’Keeffe (1990: 13). 84 For some of the stages of this transition, see Dunphy (2004: 108–111). 85 Schaefer (1996: 21).

184  Why Is There Something Rather Than Nothing? 86 Niles (2007: 145–54). For the notion of residual orality, see Bredehoft (2009: 11–19). 87 Dant (1991: 7). 88 Murdoch (2004b: 122). 89 Murdoch (2004b: 123). See also Lühr (1982). 90 Haferland (2002a; 2002b). 91 Irvine (1991: 197). 92 Neville (2001: 121). 93 Compare Spiegel (1990: 82). 94 Neville (2001: 122). Pace Neville (who cautions that we should not see these strategies as failed attempts at nation building since a ‘nation’ in its modern sense is not what they sought to build), I do not think the term ‘nation’ is anachronistic with reference to the Middle Ages. See Davis’s (1998) seminal article. 95 Goldberg (1999: 51). Notker the Stammerer (1959: 1.34, pp. 46–7). Goldberg’s translation. 96 Carruthers (2013: 17). 97 The latter view has been espoused most famously by Goffart, Innes, and Frank in works quoted throughout the present chapter. While in theory one could invent a heroic age from scratch, it is hard to argue that one can invent an entire tradition of alliterative verse and poetic formulae that are otherwise found in many attestations of parallel poetic traditions in early medieval Germanic languages. Against this, Ghosh and Pohl have argued very successfully. See Ghosh (2007: 247) and Pohl (2002). 98 Of course, as will become clear shortly, I use ‘German’ only as a convenient shorthand for what will prove to be a complex of multiple discursive strategies of ethnic identification and differentiation. 99 Riché (1993), p. 168. For a discussion of the relationships between nationalism and Carolingianists, see MacLean (2003: 4–6). 100 MacLean (2003: 5). 101 For the importance of language in constructing ethnic identity, see, for instance, Pohl (1998b: 22–7). See also Halsall (2003: 11). 102 The classical account is in Geuenich (1983). See also Goldberg (2006: 178– 83). Haug (1997: 29–31 et passim). 103 For the importance of language in articulating medieval ethnic identities, see Bartlett (1993: 198–204). 104 Reuter (2005: 190). 105 Wallace-Hadrill (1985). 106 Pohl (1998a, 1998b). 107 Lee-Stecum (2001: 100). 108 Regino (1890: xix–xx): sicut diuersae nationes populorum inter se discrepant genere, moribus, lingua, legibus. Compare Smith (1986: 22–30), which suggests an ethnic community identifies itself by the possession of six primary features: (1) a collective name, (2) a common myth of descent, (3) a shared history, (4) a distinctive shared culture, (5) an association with a specific territory, and (6) a sense of solidarity. Language thus appears to be even more straightforwardly central in medieval theorizations (although presumably it is also included in Smith’s ‘shared culture’. 109 Foot (1996: 29). 110 Bede (1969:1, pp. 16–17). 111 Foot (1996: 29). See, for example, Alfred’s preface to the OE Regula pastoralis (1967: 5). 112 Neville (2001: 110). For ‘imagined communities’, see Anderson (1983: 6).

Why Is There Something Rather Than Nothing?  185 13 Davis (1998). 1 114 Foot (1996: 28). 115 Concilium Turonense (1906: 288). 116 Edwards (1994: 147). 117 Reuter (1991: 52 and 110). See also Lehmann (1918: 260). 118 Innes (2000: 227). 119 Reuter (1991: 52). 120 The letter is transmitted through the summary recorded by the tenth-century historian of Reims, Flodoard (1998: 4.5, p. 383). 121 Venerability since the Ermanaric story is given as a parable (albeit negative) next to one of Gregory the Great’s homilies. 122 Innes (2000: 232). 123 Pohl (1998b: 68). 124 Frank (1991: 104). Cassiodorus (1894: 143–44). Rudolf of Fulda (1933). 125 Krebs (2012: 63). 126 McKitterick (2004: 42). 127 Hummer (2005: 136–37). 128 Thompson Smith (2012: 70). 129 Innes (2000: 232). See also Thomas (1988, 1990). 130 Murdoch (2004b: 123). See also Lühr (1982). 131 For the concept of dialect continuum, see Robinson (1992) and Townend (2002). 132 Robinson (1992: 203 and 238–45). 133 Goldberg (2006: 179). 134 Townend (2002: 182). 135 Townend (2002: 41). 136 Pax Leonard (2012: 122). 137 Reuter (2005: 190); Frank (1991). 138 Innes (2000: 233). 139 Freculph of Lisieux, col. 967: Alii vero affirmant eos de Scanza insula, quae vagina gentium est, exordium habuisse, de qua Gothi et caeterae nationes Theotiscae exierunt: quod et idioma linguae eorum testatur. 140 Jordanes (1882: 60): Ex hac igitur Scandza insula quasi officina gentium aut certe velut vagina nationem cum rege suo nomine Berig Gothi quondam memorantur egressi. 141 Goffart (1988: 84–96). 142 Innes (2000: 233). 143 Innes (2000: 233). For Lombards: Pohl (2000). For Burgundians, see Goffart (1988: 114), in relation to the mid-eighth century Passio Sancti Sigismundi Regis. See also Ghosh (2009: 230). 144 Ghosh (2007: 246–49). Wenksus (1961). 145 Innes (2000: 246). 146 Innes (2000: 246 and 244). 147 Against ‘propaganda’ as a useful category in understanding early medieval discourses, see Nelson (2008: 74). 148 Walahfrid Strabo (1991). 149 Cf. Niles (2007: 57). 150 Ghosh (2009: 233). 151 For the inspiration for this entire paragraph, see Innes (2000: 234). 152 Smaragdus (1986: 20–4). 153 Walahfrid Strabo (1996: ch. 7, p. 72). 154 Heather and Matthews (1991: 155–58). Goldberg (1999: 78). 155 Goldberg (2006: 182).

186  Why Is There Something Rather Than Nothing? 56 Idem. Hludowicus Rex (1884). 1 157 Oeuvres théologiques et grammaticales de Godescalc d’Orbais (1945: 195): Gens Teudisca sic habet pene distinctos casus in lingua sua sicuti sunt in latina. 158 Einhard (1911: 9 and 11). 159 Reuter (2005: 190). 160 On the importance of social identity for understanding ethnic identity in the early Middle Ages, see Geary (2008: 45–9). 161 Goldberg (2006: 178). 162 Rudolf and Meginhard (1829: 674–81). 163 Rudolf and Meginhard (1829: 679). Goldberg (2006: 178). 164 Neville (2001: 119). 165 Idem. 166 Capitularia regum Francorum (1897: I, no. 32, De villis, c. 4). 167 Reuter (2005: 192). 168 Hrabanus (1579–83). For more on Hrabanus and the vernaculars, Haubrichs (1981); Innes (2000: 234). 169 Ermold the Black (1932: 144 [line 1899]). 170 Innes (2000: 234). 171 Goldberg (1999: 65–78). 172 Goldberg (1999: 78). Annales Fuldenses s.a. 876 (1891: 86). 173 Goldberg (1999: 78). 174 Annales Fuldenses s.a. 876 (1891: 86). 175 Goldberg (1999: 78). 176 For Franconian being the ‘literal lingua Franca’ of Louis’s polity, Goldberg (2006: 179). 177 Lee-Stecum (2001: 101–04). 178 For an overview of the politics of the vernacular in Carolingian Francia, see Hummer (2005: 133). 179 Lee-Stecum (2001: 102). 180 Otfrid von Weissenburg (2004: 1.1.59–62). Quoted in and translated by Lee-Stecum (2001: 103). 181 Lee-Stecum (2001: 104). 182 Goldberg (1999: 179). 183 Ad Ludowicum 1–2. Quoted in and translated by Lee-Stecum (2001: 104). 184 Lee-Stecum (2001: 104). ‘Francus, thus, seems to have been a word with many nuances, even within a single source’: Wood (1995: 48). See also Suzuki (1990). 185 Reimitz (2008: 64–7). 186 Reimitz (2008: 65). 187 Broome (2014: 36). I am grateful to Ricky Broome for our illuminating conversations and for kindly sharing his work with me. 188 Broome (2014: 29–88). 189 Broome (2014: 45). 190 Broome (2014: 83). Annales Mettenses priores s.a. (1905: 688, 714). 191 Nelson (2008: 75). 192 Broome (2014: 86). 193 Reimitz (2008: 58). 194 Heidelberg Universitätsbibliothek, MS Cod. Pal. Lat. 52. 195 Murdoch (2004b: 130–31). 196 Murdoch (2004b: 130–34). For the manuscript context, see also McKitterick (1988: 234).

Why Is There Something Rather Than Nothing?  187 197 For the equal rootedness in Germanic and classical traditions, see Ziolkowski (2001). Kratz (1980: 18–40) is the most thorough analysis supporting the satirical reading. For a more recent opinion (which, however, fails to see the irony in the text), see Stone (2013). 198 The translations from Waltharius in this section belong to Kratz (1977). 199 Although tenth-century dates have been proposed for the Waltharius, it was most probably written around the middle of the ninth century. Peter Dronke argued for an 820s date, although not everyone agrees, while the 890s seem to be the terminus ad quem in the more up-to-date appraisals. See Dronke (1977: 70). For a useful survey, see Kratz (1977: xiv). 200 Ziolkowski (2001: 45). 201 Ziolkowski (2001: 46). 202 Ziolkowski (2001: 46). 203 Kratz (1977: 130). 204 Stone (2011: 57 et passim). For the problematic nature of heroism, see, among others to be discussed in the following section, Townsend (2000). 205 For the biblical and patristic significance of the body parts lost by each warrior, see Kratz (1977: xxi–xxiii). 206 Ghosh (2009: 166). 207 Ghosh (2009: 164). 208 Ghosh (2009: 163). Dronke (1977) has provided a full survey of non-Germanic traditions regarding Walter. 209 Ghosh (2009: 156). 210 Dronke (1977: 29–34). 211 Ghosh (2009: 146). Jordanes (1882: 185–218) for the battle; page 191 for Franks and Burgundians; Gregory of Tours (1951: II.7). 212 Ghosh (2009: 147). 213 Ghosh (2009: 146). Gregory of Tours (1951: IV.23. Poeta Saxo, pp. 31–2). 214 Nelson (2008: 77). 215 Goldberg (1999: 55). 216 To mention only the work of Stone, Nelson, and Goldberg quoted throughout this chapter. 217 Goldberg (1999: 45). 218 Goldberg (1999: 45). 219 Goldberg (1999: 42). 220 Goldberg (1999: 55). 221 Goldberg (1999: 47). For amici regis, see Annales Fuldenses s.a. 849 (1891: 38). 222 Capitularia regum Francorum (1897: II, pp. 152–58; at Koblenz in 860); Annales Fuldenses s.a. 876 (1891: 89). Goldberg (1999: 66). 223 Nelson (1999: 123). See also Leja (2008). 224 For imperial peace, see Stone (2011: 78–82). For monastic ideals, see Nelson (1999: 123–29). 225 Reuter (1991: 48). 226 Halsall (2003: 30–5). Airlie (2005). 227 Reuter (1991: 48). 228 Reuter (1991: 48). 229 The translations from the Waltharius in this section belong to Rio (2015). 230 Rio (2015: 53). 231 Epistolae variorum (1899: 27, p. 344). Airlie (2011: 57). 232 Airlie (2011: 69). 233 Halsall (2002: 3).

188  Why Is There Something Rather Than Nothing? 34 Stone (2011: 210). Le Jan (1995: 87–8). 2 235 Dhuoda (1998: 3.2, p. 88). 236 Spiegel (1990: 68). 237 For the history of Dhuoda’s family, see Marcelle Thiébaux’s introduction to Dhuoda (1998: 13–21 and 34–7). The following account owes much to her presentation. 238 Thiébaux in Dhuoda (1998: 34–7). 239 See Stone (2011: 78–82). 240 Rio (2015: 50). 241 Idem. 242 Nelson (1999: 131). 243 Nelson (1999: 134). Hincmar of Rheims (1991: 182). 244 Nelson (1999: 135). 245 Rio (2015: 60). 246 Spiegel (1990: 78). 247 Banniard (2003: 26). 248 Spiegel (1990: 77). 249 Rio (2015). 250 For a comparative view, see Barry (2002: 184–87). The classical discussion can be found in the introduction to Dollimore and Sinfield (1994: 3).

Bibliography Primary Sources Alcuini Epistolae, in MGH Epistolae, vol. 4, Epistolae Karolini Aevi, vol. 2, ed. by Ernst Dümmler (Berlin: Weidmann, 1895). Annales Fuldenses, MGH Scriptores rerum Germanicarum, vol. 7, ed. by Georg H. Pertz (Hanover: Hahn, 1891). Annales Mettenses priores, MGH Scriptores rerum Germanicarum, vol. 10, ed. by B. de Simson (Hanover: Hahn, 1905). Bede, Historia Ecclesiastica gentis anglorum, ed. and trans. by Bertram Colgrave and R. A. B. Mynors (Oxford: Clarendon Press, 1969). Capitulare episcoporum, in MGH Capitularia regum Francorum, vol. 1, ed. by Alfred Boretius (Hannover: Hahn, 1883). Capitularia regum Francorum, MGH vol. 2, ed. by Alfred Boretius (Hannover: Hahn, 1897). Cassiodorus, Variae, in MGH Auctores antiquissimi, vol. 12, ed. by Theodor Mommsen (Berlin: Weidmann, 1894). Concilium Turonense, MGH Concilia aevi Karolini, vol. 1, ed. by Albert Werminghoff (Hanover: Hahn, 1906). Councils and Ecclesiastical Documents Relating to Great Britain and Ireland, ed. by Arthur W. Haddan and William Stubbs, 3 vols (Oxford: Clarendon Press, 1869–79), III (1879). Dhuoda, Handbook for Her Warrior Son: Liber Manualis, ed. and trans. by Marcelle Thiébaux (Cambridge: Cambridge University Press, 1998). Einhard, Vita Karoli Magni, MGH Scriptores rerum Germanicarum, vol. 25, ed. by O. Holder-Egger (Hanover: Hahn, 1911). Epistolae variorum, in MGH Epistolae, vol. 5, Epistolae Karolini Aevi, vol. 3, ed. by Ernst Dümmler (Berlin: Weidmann, 1899).

Why Is There Something Rather Than Nothing?  189 Ermanrich, Epistola ad Grimaldum abbatem, in MGH Epistolae, vol. 5, Epistolae Karolini Aevi, vol. 3, ed. by Ernst Dümmler (Berlin: Weidmann, 1899), pp. 534–79. Ermold the Black, In honorem Hludowici Pii, II 1248–49, ed. by Edmond Faral, Ermold le Noir – Poème sur Louis le Pieux (Paris: Champion, 1932). Flodoard, Historia Remensis Ecclesiae, ed. by Martina Stratmann, MGH Scriptores, vol. 36 (Hanover: Hahnsche Buchhandlung, 1998). Freculph of Lisieux, Chronicon, PL 106, cols. 957–1258 (n.d..). Gregory of Tours, Decem Libri Historiarum, MGH Scriptores rerum Merovingicarum, vol. 1, ed. by Bruno Krusch and Wilhelm Levison (Hanover: Hahn, 1951 [1884]). Heidelberg Universitätsbibliothek, MS Cod. Pal. Lat. 52. Hincmar of Rheims, The Annals of St. Bertin, trans. by Janet L. Nelson (Manchester: Manchester University Press, 1991). Hludowicus Rex, Ad Baldonem, in MGH Poeta Latini aevi Carolini, vol. 2, ed. by Ernst Dümmler (Berlin: Weidmann, 1884), pp. 643–44. Hrabanus Maurus, De Inventione Linguarum, PL 112, cols. 1579–83 (n.d.). Jordanes, Getica, MGH Auctores antiquissimi, vol. 5, ed. by Theodor Mommsen (Berlin: Weidmann, 1882). Notker the Stammerer, Gesta Karoli magni imperatoris 1.34, MGH Scriptores rerum Germanicarum, new series, vol. 12, ed. by Hans Haefele (Berlin: Weidmann, 1959). Oeuvres théologiques et grammaticales de Godescalc d’Orbais, ed. by Cyrille Lambot (Louvain: Peeters, 1945). Otfrid von Weissenburg, Evangelienbuch: Edition nach dem Wiener Codex 2687, ed. by Wolfgang Kleiber and Rita Heuser (Tübingen: Niemeyer, 2004). Poeta Saxo, Annalium de gestis Caroli Magni imperatoris, in MGH Poeta Latini aevi Carolini, vol. 4, ed. by Paul von Winterfeld (Berlin: Weidmann, 1899). Regino, Letter to Archbishop Hatto of Mainz, in Regionis Prumiensis Chronicon, in MGH Scriptores rerum Germanicarum, vol. 50, ed. by Friedrich Kurze (Hanover: Hahn, 1890). Regula pastoralis, in Sweet’s Anglo-Saxon Reader, rev. by Dorothy Whitelock (Oxford: Oxford University Press, 1967). Rudolf of Fulda, Translatio Sancti Alexandri, in Nachrichten von der Gesellschaft der Wissenschaften zu Göttingen, Philologisch-historische Klasse, ed. by Bruno Krusch (Göttingen: Weidmann, 1933). Rudolf and Meginhard, Translatio sancti Alexandri, MGH Scriptores, vol. 2, ed. by Georg H. Pertz (Hanover: Hahn, 1829). Smaragdus of Saint-Mihiel, Liber in Partibus Donati, ed. by Bengt Löfstedt, Louis Holtz and Adele Kibre, CCCM 68 (Turnhout: Brepols, 1986). Walahfrid Strabo, De imagine Tetrici, ed. and trans. by Michael W. Herren, The Journal of Medieval Latin, 1 (1991), 118–39. Walahfrid Strabo, Libellus De Exordiis et Incrementis quarundam in observationibus ecclesiasticis rerum, ed. and trans. by Alice Harting-Correa (Leiden: Brill, 1996). Waltharius = Waltharius and Ruodlieb, ed. and trans. by Dennis M. Kratz (New York: Garland, 1984).

190  Why Is There Something Rather Than Nothing?

Secondary Sources Airlie, Stuart, ‘Charlemagne and the Aristocracy: Captains and Kings’, in Charlemagne: Empire and Society, ed. by Joanna Storey (Manchester: Manchester University Press, 2005). Airlie, Stuart, ‘The World, the Text and the Carolingian: Royal, Aristocratic and Masculine Identities in Nithard’s Histories’, in Lay Intellectuals in the Carolingian World, ed. by Patrick Wormald and Janet L. Nelson (Cambridge: Cambridge University Press, 2011), pp. 51–76. Anderson, Benedict, Imagined Communities: Reflections on the Origin and Spread of Nationalism (London: Verso, 1983). Banniard Michel, ‘Latinophones, Romanophones, Germanophones: Interactions identitaires et construction langagière (VIIIe-Xe siècle)’, Médiévales, 45 (2003): Grammaires du vulgaire, 25–42. Barry, Peter, Beginning Theory (Manchester: Manchester University Press, 2002). Bartlett, Robert, The Making of Europe: Conquest, Colonization and Cultural Change, 950–1350 (London: Penguin, 1993). bell hooks, ‘Marginality as a Site of Resistance’, in Out There: Marginalization and Contemporary Cultures, ed. by Russell Ferguson et al. (Cambridge, MA: MIT Press, 1990a), pp. 241–43. bell hooks, ‘Talking Back’, in Out There: Marginalization and Contemporary Cultures, ed. by Russell Ferguson et al. (Cambridge, MA: MIT Press, 1990b), pp. 337–40. Benjamin, Walter, The Arcades Project, trans. by Howard Eiland and Kevin McLaughlin (Cambridge, MA: Harvard University Press, 1999). Booker, Courtney, Past Convictions: The Penance of Louis the Pious and the Decline of the Carolingians (Philadelphia: University of Pennsylvania Press, 2009). Bourdieu, Pierre, ‘The Forms of Capital’, in Handbook for Theory and Research for the Sociology of Education, ed. by John G. Richardson (New York: Greenwood, 1986), pp. 241–58. Bourdieu, Pierre, The Logic of Practice, trans. Richard Nice (Oxford: Polity Press, 1990). Bredehoft, Thomas A., Authors, Audiences and Old English Verse (Toronto: Toronto University Press, 2009). Broome, Richard, Approaches to Community and Otherness in the Late Merovingian and Early Carolingian Periods (unpublished doctoral thesis, University of Leeds, 2014). Butt, John J., Daily Life in the Age of Charlemagne (Westport, CT: Greenwood Press, 2002). Carruthers, Mary, The Experience of Beauty in the Middle Ages (Oxford: Oxford University Press, 2013). Charles-Edwards, Thomas, ‘The Context and Uses of Literacy in Early Christian Ireland’, in Literacy in Medieval Celtic Societies, ed. by Huw Pryce (Cambridge: Cambridge University Press, 1998), pp. 62–82. Chartier, Roger, ‘Intellectual History or Sociocultural History? The French Trajectories’, in Modern European Intellectual History: Reappraisals and New Perspectives, ed. by Dominick LaCapra and Steven L. Kaplan (Ithaca: Cornell University Press, 1982), pp. 13–46.

Why Is There Something Rather Than Nothing?  191 Collins, Roger, ‘Charlemagne and His Critics, 814–29’, in La royauté et les élites dans l’Europe carolingienne, ed. by Régine Le Jan (Lille: Villeneuve-d’Ascq, 1998), pp. 193–211. Dant, Tim, Knowledge, Ideology and Discourse: A Sociological Perspective (London: Routledge, 1991). Davis, Kathleen, ‘National Writing in the Ninth Century: A Reminder for Postcolonial Thinking about the Nation’, Journal of Medieval and Early Modern Studies, 28 (1998), 639–56. de Jong, Mayke, ‘The Empire as Ecclesia: Hrabanus Maurus and Biblical Historia for Rulers’, in Uses of the Past in the Early Middle Ages, ed. by Yitzak Hen and Matthew Innes (Cambridge: Cambridge University Press, 2000), pp. 191–226. de Jong, Mayke, ‘Charlemagne’s Church’, in Charlemagne: Empire and Society, ed. by Joanna Story (Manchester: Manchester University Press, 2005), pp. 103–35. Dollimore, Jonathan, and Alan Sinfield, eds, Political Shakespeare: New Essays in Cultural Materialism (Manchester: Manchester University Press, 1994). Dronke, Peter, ‘Waltharius-Gaiferos’, in Barbara et antiquissima carmina, ed. by Ursula and Peter Dronke (Barcelona: Universidad Autónoma, Faculdad de Letras, 1977), pp. 29–79. Dunphy, Graeme, ‘Orality’ in Early Germanic Literature and Culture, ed. by Brian Murdoch and Malcolm Read (Woodbridge: Camden House, 2004), pp. 103–18. Dutton, Paul E., Charlemagne’s Mustache and Other Cultural Clusters of a Dark Age (New York: Palgrave Macmillan, 2004). Edwards, Cyril, ‘German Vernacular Literature: A Survey’, in Carolingian Culture: Emulation and Innovation, ed. by Rosamond McKitterick (Cambridge: Cambridge University Press, 1994), pp. 141–70. Ferguson, Russell, ‘Introduction: Invisible Center’, in Out There: Marginalization and Contemporary Cultures, ed. by Russell Ferguson et al. (Cambridge, MA: MIT Press, 1990), pp. 9–14. Foley, John Miles, Oral-Formulaic Theory and Research: An Introduction and Annotated Bibliography (New York: Garland, 1985). Foley, John Miles, The Theory of Oral Composition: History and Methodology (Bloomington, IN: Indiana University Press, 1988). Foot, Sarah, ‘The Making of Angelcynn: English Identity before the Norman Conquest’, Transactions of the Royal Historical Society, Sixth Series, 6 (1996), 25–49. Frank, Roberta, ‘Germanic Legend in Old English Literature’, in The Cambridge Companion to Old English Literature, ed. by Malcolm Godden and Michael Lapidge (Cambridge: Cambridge University Press, 1991), pp. 88–106. Garizpanov, Ildar H., The Symbolic Language of Authority in the Carolingian World: c. 751–877 (Leiden: Brill, 2008). Geary, Patrick J., ‘Reflections on Historiography and the Holy: Center and Periphery’ in The Making of Christian Myths in the Periphery of Latin Christendom (c. 1000–1300), ed. by Lars Boje Mortensen (Copenhagen: Museum Tusculanum Press, 2006), pp. 323–29. Geary, Patrick J., ‘Ethnicity, Group Identity, and Social Status in the Migration Period’, in Franks, Northmen, and Slavs: Identities and State Formation in Early

192  Why Is There Something Rather Than Nothing? Medieval Europe, ed. by Ildar H. Garipzanov, Patrick J. Geary, Przemysław Urbańczyk (Turnhout: Brepols, 2008), pp. 17–49. Geuenich, Dieter, ‘Die volkssprachige Ueberlieferung der Karolingerzeit aus der Sicht des Historikers’, Deutsches Archiv, 39 (1983), 104–30. Ghosh, Shami, ‘On the Origins of Germanic Heroic Poetry: A Case Study of the Legend of the Burgundians’, Beiträge zur Geschichte der deutschen Sprache und Literatur, 129 (2007), 220–52. Ghosh, Shami, The Barbarian Past in Early Medieval Historical Narrative (unpublished doctoral thesis, University of Toronto, 2009). Godman, Peter, Poets and Emperors: Frankish Politics and Carolingian Poetry (Oxford: Clarendon Press, 1987). Goffart, Walter, The Narrators of Barbarian History (A.D. 550–800) (Princeton: Princeton University Press, 1988). Goldberg, Eric J., ‘“More Devoted to the Equipment of Battle Than the Splendor of Banquets”: Frontier Kingship, Martial Ritual, and Early Knighthood at the Court of Louis the German’, Viator, 30 (1999), 41–78. Goldberg, Eric J., Struggle for Empire: Kingship and Conflict under Louis the German: 817–76 (Ithaca: Cornell University Press, 2006). Green, Dennis H., The Carolingian Lord (Cambridge: Cambridge University Press, 1965). Green, Dennis H., ‘Three Aspects of the Old Saxon Biblical Epic, the Heliand’, in The Continental Saxons from the Migration to the Tenth Century: An Ethnographic Perspective, ed. by Dennis H. Green and Frank Siegmund (Woodbridge: Boydell, 2003), pp. 247–69. Haferland, Harald, ‘Mündliche Erzähltechnik im Heliand’, GermanischRomanische Monatsschrift, 52 (2002a), 237–59. Haferland, Harald, ‘War der Dichter des ‘Heliand’ illiterat?’, Zeitschrift für Deutsches Altertum und Deutsche Literatur, 131 (2002b), 20–48. Halsall, Guy, ‘Introduction: “Don’t Worry, I’ve Got the Key”’, in Humour, History and Politics in Late Antiquity and the Early Middle Ages, ed. by Guy Halsall (Cambridge: Cambridge University Press, 2002), pp. 1–21. Halsall, Guy, Warfare and Society in the Barbarian West: 450–900 (London: Routledge, 2003). Haubrichs, Wolfgang, ‘Althochdeutsch in Fulda und Weissenburg – Hrabanus Maurus und Otfried von Weissenburg’, in Hrabanus Maurus: Lehrer, Abt und Bischof, ed. by Raymund Kottje und Harald Zimmermann (Wiesbaden: Steiner, 1981), pp. 182–93. Haubrichs, Wolfgang, Geschichte der deutschen Literatur von den Anfängen bis zum Beginn der Neuzeit. Band I: Von den Anfängen bis zum hohen Mittelalter, in zwei Teilen. Teil 1: Die Anfänge: Versuche volkssprachiger Schriftlichkeit im frühen Mittelalter (ca. 700–1050) (Frankfurt-am-Main: Athenäum, 1988). Haug, Walter, Vernacular Literary Theory in the Middle Ages: The German Tradition, 800–1300, in its European Context, trans. Joanna M. Catling (Cambridge University Press, 1997 [1985]). Haymes, Edward R., ‘Oral Poetry and the Germanic Heldenlied’, Rice University Studies, 62 (1976), 47–54. Heather, Peter, and John Matthews, The Goths in the Fourth Century (Liverpool: Liverpool University Press, 1991).

Why Is There Something Rather Than Nothing?  193 Hellgardt, Ernst, ‘Zur Mehrsprachigkeit im Karolingerreich: Bemerkungen aus Anlass von Rosamond McKittericks Buch The Carolingians and the Written Word’, Beiträge zur Geschichte der deutschen Sprache und Literatur, 118 (1996), 1–48. Hennessey Olsen, Alexandra, ‘Formulaic Tradition and the Latin Waltharius’, in Heroic Poetry in the Anglo-Saxon Period: Studies in Honor of Jess B. Bessinger, Jr, ed. by Helen Damico and John Leyerle (Kalamazoo: Medieval Institute Publications, 1993), pp. 265–82. Hollis, Stephanie, ‘Anglo-Saxon Secular Learning and the Vernacular: An Overview’, Amsterdamer Beiträge zur älteren Germanistik, 69 (2012), 1–43. Hummer, Hans J., Politics and Power in Early Medieval Europe: Alsace and the Frankish Realm, 600–1000 (Cambridge: Cambridge University Press, 2005). Innes, Matthew, ‘Teutons or Trojans? The Carolingians and the Germanic Past’, in The Uses of the Past in the Early Middle Ages, ed. by Yitzak Hen and Matthew Innes (Cambridge: Cambridge University Press, 2000), pp. 227–49. Innes, Matthew, ‘Laughing Monks, Smiling Kings, Frowning Emperors: The Politics of Laughter in the Carolingian Empire’, in Humour, History and Politics in Late Antiquity and the Early Middle Ages, ed. by Guy Halsall (Cambridge: Cambridge University Press, 2002), pp. 131–56. Irvine, Martin, ‘Medieval Textuality and the Archeology of Textual Culture’, in Speaking Two Languages: Traditional Disciplines and Contemporary Theory in Medieval Studies, ed. by Allen J. Frantzen (Albany, NY: State University of New York Press, 1991), pp. 181–210 and 276–84. Jambeck, Karen, ‘The Tretiz of Walter of Bibbesworth: Cultivating the Vernacular’, in Childhood in the Middle Ages and the Renaissance: The Results of a Paradigm Shift in the History of Mentality, ed. by Albrecht Classen (Berlin: Walter De Gruyter, 2005), pp. 160–83. Kershaw, Paul, ‘Laughter After Babel’s Fall: Misunderstanding and Miscommunication in the Ninth-century West’, in Humour, History and Politics in Late Antiquity and the Early Middle Ages, ed. by Guy Halsall (Cambridge: Cambridge University Press, 2002), pp. 179–202. Kratz, Dennis M., ‘Quid Waltharius Ruodliebque cum Christo?’, in The Epic in Medieval Society: Aesthetic and Moral Values, ed. Harald Scholler (Tübingen: Max Niemeyer, 1977), pp. 126–49. Kratz, Dennis M., Mocking Epic: Waltharius, Alexandreis and the Problem of Christian Heroism (Madrid: José Porrua Turanzas, 1980). Krebs, Christopher B., A Most Dangerous Book: Tacitus’s Germania from the Roman Empire to the Third Reich (New York: W. W. Norton, 2012). Krogmann, Willy, ‘Muspilli und Muspellsheim’, Zeitschrift für Religions- und Geistesgeschichte, 5 (1953), 97–118. Le Jan, Régine, Famille et pouvoir dans le monde franc (VIIe-Xe siècle): Essai d'anthropologie sociale (Paris: Publications de la Sorbonne, 1995). Lees, Clare, Tradition and Belief: Religious Writing in Late Anglo-Saxon England (Minneapolis: University of Minnesota Press, 1999). Lee-Stecum, Parshia, ‘Otherwise the Same: Latin Models of Poetic SelfPresentation in the Evangelienbuch of Otfrid (1.1.1-126)’ in Germanic Texts and Latin Models Medieval Reconstructions, ed. by Karin Olsen, Antonina Harbus and Tette Hofstra (Leuven: Peeters, 2001), pp. 93–106.

194  Why Is There Something Rather Than Nothing? Lehmann, Paul, ed., Mittelalterliche Bibliothekskataloge Deutschlands und der Schweiz, vol. 1: Die Bistümer Konstanz und Chur (Munich: Oskar Beck, 1918). Leja, Meg, ‘Making of Men, Not Masters: Right Order and Lay Masculinity according to Dhuoda and Nithard’, Comitatus, 39 (2008), 1–40. Liuzza, Roy M., ‘Beowulf: Monuments, Memory, History’, in Readings in Medieval Texts, ed. by David Johnson and Elaine Treharne (Oxford: Oxford University Press, 2005), pp. 91–108. Lord, Albert, The Singer of Tales (Cambridge, MA: Harvard University Press 1960). Lühr, Rosemarie, Studien zur Sprache des Hildebrandsliedes (Frankfurt am Main: Lang, 1982). MacLean, Simon, Kingship and Politics in the Late Ninth Century: Charles the Fat and the End of the Carolingian Empire (Cambridge University Press, 2003). McKitterick, Rosamond, The Carolingians and the Written Word (Cambridge: Cambridge University Press, 1988). McKitterick, Rosamond, History and Memory in the Carolingian World (Cambridge: Cambridge University Press, 2004). McKitterick, Rosamond, ‘The Carolingian Renaissance of Culture and Learning’, in Charlemagne: Empire and Society, ed. by Joanna Story (Manchester: Manchester University Press, 2005), pp. 151–66. Mortensen, Lars Boje, ‘Sanctified Beginnings and Mythopoetic Moments. The First Wave of Writing on the Past in Norway, Denmark, and Hungary, c. 1000–1230’, in The Making of Christian Myths in the Periphery of Latin Christendom (c. 1000–1300), ed. by Lars Boje Mortensen (Copenhagen: Museum Tusculanum Press, 2006), pp. 247–73. Murdoch, Brian, ‘Charms, Recipes, and Prayers’, in German Literature in the Middle Ages, ed. by Brian Murdoch (Woodbridge: Camden House, 2004a), pp. 57–72. Murdoch, Brian, ‘Heroic Verse’, in German Literature in the Middle Ages, ed. by Brian Murdoch (Woodbridge: Camden House, 2004b), pp. 121–38. Murdoch, Brian, ‘Introduction’, in German Literature in the Middle Ages, ed. by Brian Murdoch (Woodbridge: Camden House, 2004c), pp. 1–32. Nagy, Piroska, and Damien Boquet, eds, Le sujet des émotions au Moyen Age (Paris: Beauchesne, 2009). Nelson, Janet L., ‘Monks, Secular Men and Masculinity, c. 900’, in Masculinity in Medieval Europe, ed. by Dawn Hadley (London: Longman, 1999), pp. 121–42. Nelson, Janet L., ‘Frankish Identity in Charlemagne’s Empire’, in Franks, Northmen, and Slavs: Identities and State Formation in Early Medieval Europe, ed. by Ildar H. Garipzanov, Patrick J. Geary, Przemysław Urbańczyk (Turnhout: Brepols, 2008), pp. 71–83. Neville, Jennifer, ‘History, Poetry, and “National” Identity in Anglo-Saxon England and the Carolingian Empire’, in Germanic Texts and Latin Models Medieval Reconstructions, ed. by Karin Olsen, Antonina Harbus and Tette Hofstra (Leuven: Peeters, 2001), pp. 107–26. Niles, John D., Old English Heroic Poems and the Social Life of Texts (Turnhout: Brepols, 2007). O’Brien O’Keeffe, Katherine, Visible Song: Transitional Literacy in Old English Verse (Cambridge: Cambridge University Press, 1990).

Why Is There Something Rather Than Nothing?  195 Parry, Milman, The Making of Homeric Verse, ed. by Adam Parry (Oxford: Oxford University Press, 1971). Pax Leonard, Stephen, Language, Society and Identity in Early Iceland (Chichester: Wiley-Blackwell, 2012). Pohl, Walter, ‘Introduction: Strategies of Distinction’, in Strategies of Distinction: The Construction of the Ethnic Communities, 300–800, ed. by Walter Pohl and Helmut Reimitz (Leiden: Brill, 1998a), pp. 1–15. Pohl, Walter, ‘Telling the Difference: Signs of Ethnic Identity’, in Strategies of Distinction: The Construction of the Ethnic Communities, 300–800, ed. by Walter Pohl and Helmut Reimitz (Leiden: Brill, 1998b), pp. 17–69. Pohl, Walter, ‘Memory, Identity and Power in Lombard Italy’, in The Uses of the Past in the Early Middle Ages, ed. by Yitzak Hen and Matthew Innes (Cambridge: Cambridge University Press, 2000), pp. 9–28. Pohl, Walter, ‘Ethnicity, Theory, and Tradition: A Response’, in Barbarian Identity: Critical Approaches to Ethnicity in the Early Middle Ages, ed. by Andrew Gillett (Turnhout: Brepols, 2002). Reimitz, Helmut, ‘Omnes Franci: Identifications and Identities of the Early Medieval Franks’, in Franks, Northmen, and Slavs: Identities and State Formation in Early Medieval Europe, ed. by Ildar H. Garipzanov, Patrick J. Geary, Przemysław Urbańczyk (Turnhout: Brepols, 2008), pp. 51–69. Renoir, Alain, ‘The Armor of the Hildesbrandslied: An Oral-formulaic Point of View’, Neuphilologische Mitteilungen, 78 (1977), 389–95. Renoir, Alain, ‘Oral-formulaic Tradition and the Affective Interpretation of Early Germanic Verse’, in Germania: Comparative Studies in the Old Germanic Languages and Literatures, ed. by Daniel G. Calder and T. Craig Christy (Woodbridge: Brewer, 1988), pp. 113–26. Reuter, Timothy, Germany in the Early Middle Ages: c. 800–1056 (London: Longman, 1991). Reuter, Timothy, ‘Charlemagne and the World Beyond the Rhine’, in Charlemagne: Empire and Society, ed. by Joanna Story (Manchester: Manchester University Press, 2005), pp. 183–94. Riché, Pierre, The Carolingians: A Family Who Forged Europe, trans. by Michael Idomir Allen (Philadelphia, PA: University of Pennsylvania Press, 1993). Richter, Michael, ‘À quelle époque a-t-on cessé de parler Latin en Gaule? À propos d’une question mal posée’, Annales E.S.C., 38 (1983), 439–48. Richter, Michael, The Formation of the Medieval West: Studies in the Oral Culture of the Barbarians (Dublin: Four Courts Press, 1994). Rio, Alice, ‘Waltharius at Fontenoy? Epic Heroism and Carolingian Political Thought’, Viator, 46 (2015), 41–64. Robinson, Orrin W., Old English and Its Closest Relatives: A Survey of the Earliest Germanic Languages (London: Routledge, 1992). Rosenwein, Barbara H., Emotional Communities in the Early Middle Ages (Ithaca: Cornell University Press, 2006). Schaefer, Ursula, ‘A Dialogue Between Orality and Literacy. Considerations on Linguistic Strategies in the Old English Bede’, in Dialogische Strukturen/ Dialogic Structures: Festschrift für Willi Erzgräber zum 70. Geburtstag, ed. by Thomas Kühn and Ursula Schaefer (Tübingen: Narr, 1996), pp. 17–33. Scheil, Andrew, ‘The Historiographic Dimensions of Beowulf’, JEGP, 107 (2008), 281–302.

196  Why Is There Something Rather Than Nothing? Smith, Anthony, The Ethnic Origins of Nations (Oxford: Blackwell, 1986). Spiegel, Gabrielle, ‘History, Historicism, and the Social Logic of the Text in the Middle Ages’, Speculum, 65 (1990), 59–86. Spivak, Gayatri, ‘Explanation and Culture: Marginalia’, in Out There: Marginalization and Contemporary Cultures, ed. by Russell Ferguson et al. (Cambridge, MA: MIT Press, 1990), pp. 377–93. Stone, Rachel, Morality and Masculinity in the Carolingian Empire (Cambridge: Cambridge University Press, 2011). Stone, Rachel, ‘Waltharius and Carolingian Morality: Satire and Lay Values’, Early Medieval Europe, 21 (2013), 50–70. Suzuki, Seiichi, ‘Frenkisk. Zur Geschichte von theodiscus und teutonicus im Frankenreich des 9. Jahrhunderts’, in Beiträge zur Geschichte des Regnum Francorum. Referate beim Wissenschaftlichen Colloquium zum 75. Geburtstag von Eugen Ewig am 28. Mai 1988, ed. by Rudolf Schieffer, (Sigmaringen: Thorbecke, 1990), pp. 67–95. Suzuki, Seiichi, ‘Kaluza’s Law in the Old Saxon Heliand’, Amsterdamer Beiträge zur älteren Germanistik, 68 (2011), 27–52. Thomas, Heinz, ‘Der Ursprung des Wortes Theodiscus’, Historische Zeitschrift, 247 (1988), 295–331. Thompson Smith, Scott, ‘Historicism’, in A Handbook of Old English Studies, ed. by Jacqueline Stodnick and Renée R. Trilling (Oxford: Blackwell, 2012), pp. 69–84. Townend, Matthew, Language and History in Viking Age England: Linguistic Relations between Speakers of Old Norse and Old English (Turnhout: Brepols, 2002). Townsend, David, ‘Ironic Intertextuality and the Reader’s Resistance to Heroic Masculinity in the Waltharius’, in Becoming Male in the Middle Ages, ed. by Jeffrey Jerome Cohen (London: Routledge, 2000), pp. 67–86. Tyler, Lee Edgar, ‘The Heroic Oath of Hildebrand’, in De Gustibus: Essays for Alain Renoir, ed. by John Miles Foley, Chris Womack and Whitney A. Womack (New York: Garland, 1992), pp. 551–85. Wallace-Hadrill, J. M., The Barbarian West, 400–1000, rev. edn. (Oxford: Oxford University Press, 1985). Wells, Christopher, ‘The Shorter German Verse Texts’, in German Literature in the Middle Ages, ed. by Brian Murdoch (Woodbridge: Camden House, 2004), pp. 157–200. Wenksus, Reinhard, Stammesbildung und Verrfassung: Das Werden der frühmittelalterlichen gentes (Cologne: Böhlau Verlag, 1961). Winer, Rebecca Lynn, ‘Conscripting the Breast: Lactation, Slavery and Salvation in the Realms of Aragon and Kingdom of Majorca, c. 1250–1300’, Journal of Medieval History, 34 (2008), 164–84. Wolf, Alois, ‘Die Verschriftlichung von europäischen Heldensagen als mittelalterliches Kulturproblem’, in Heldensage und Heldendichtung im Germanischen, ed. by Heinrich Beck (Berlin: De Gruyter, 1988), pp. 305–28. Wood, Ian, ‘Defining the Franks: Frankish Origins in Early Medieval Historiography’, in Concepts of National Identity in the Middle Ages, ed. by Simon Forde, Lesley Johnson and Alan V. Murray, Leeds Texts and Monographs, 14 (Leeds: Institute for Medieval Studies, 1995), pp. 47–57.

Why Is There Something Rather Than Nothing?  197 Wright, Roger, Late Latin and Early Romance in Spain and Carolingian France (Liverpool: Francis Cairns, 1982). Ziolkowski, Jan M., ‘Fighting Words: Wordplay and Swordplay in the Waltharius’, in Germanic Texts and Latin Models Medieval Reconstructions, ed. by Karin Olsen, Antonina Harbus, Tette Hofstra (Leuven: Peeters, 2001), pp. 29–51.

5 Beowulf in Times of Anxiety The Archaeology of Emotions in Old English Verse History

By bringing together the previous chapter’s insights into the socio-­ cultural functions of vernacular verse histories in Carolingian Francia and the understanding of its conceptual underpinnings and formal features gained in the first three chapters, the fifth and final chapter focuses on how OE verse histories functioned as a focus for powerful emotions, thereby performing important work in society. I found anxiety to be a useful point of entry to this task, as it subtends the variety of emotional and ideological investments of different audiences of OE GHP, especially in their collective and interpersonal aspects. Sometimes we find the deepest intimacy not in sex, friendship, communal joy, or grief but in shared anxiety. It is a subtler, though no less powerful, kind of togetherness, communed less overtly through sideway glances, heavy silences, and nervous laughter. As such, it subtends emotional communities that are harder to trace in texts such as Beowulf, which are notorious for how opaque their emotional language has become to us.1 Perceiving anxiety in others is difficult primarily because the people subjected to it can often be at pains to articulate it or may even be unaware that they are experiencing it. At times, the strongest clue of its existence is the effort people make to deny it. Anxiety is protean, too: it can take forms as mundane as being ashamed or embarrassed and as vivid as full-blown panic or inexplicable bursts of rage. It can coalesce into cultural anxieties or moral panics, or it can take the shape of a brief individual episode.2 Yet I suggest, apart from the joy of reading and listening, it is anxiety that gathered so many audiences around Beowulf for so long a time. The purpose of this chapter, then, is to probe some of the points in rag-picker verse histories that trouble certain audiences in order to understand the ways in which these communities function emotionally in relation to the text. While looking for anxieties in what was and still is an entertaining and aesthetically pleasing poem, we should not forget that enjoying a text need not be divorced from the anxiety it makes us feel, from the fears and troubling emotions inside us it gives a voice to. After Beowulf

Beowulf in Times of Anxiety  199 defeats Grendel, during a time of relaxation and gift-giving in Heorot, a scop sings of the strife of Finn and Hengest. His song, despite its unhappy theme, is described as a healgamen, ‘hall entertainment’ (Beowulf 1066a), and is followed by a new round of mirth and drinking (gamen eft astah, ‘merriment resumed’) (Beowulf 1160b). Later, as Beowulf recounts his adventures to Hygelac, his uncle, back in his Geatish homeland, telling of the gidd ond gleo (‘song and music’) (Beowulf 2105a) that arose in Heorot when Hrothgar took up the harp and sang, the gloomy content of some of these songs did not spoil the festive occasion.3 As I have argued in the first three chapters, especially rag-picker verse histories are characterized by a complicated relationship with modern concepts of truthfulness, realism, and history-writing. Meanwhile, their occasional fragmentariness, self-referentiality, and lack of interpretational closure and of any transparent ideological investment make them even harder to accept as a form of historical production that does not exclude things ‘that lived reality cannot admit’ and that, therefore, can be treated as simultaneously historiographical and as collective fantasies.4 As I argued in the previous chapter, rag-picker verse histories such as Beowulf or the Waltharius exist in a ludic emotional space in which the anxieties, beliefs, and desires of different textual and emotional communities are toyed with, allowed to measure against each other, and brought to their ultimate consequences within the space of a past always just out of reach. Emotionally, such verse histories can act as ‘dream-screens’ on which the anxieties and desires of the audiences were projected, worked through, and, thus, potentially exorcised.5 Adopting Derek Neal’s argument about high medieval romance, Beowulf can be seen as functioning like a collective dream that ‘solve[s] problems and deal[s] with conflicts that are too difficult for conscious life’ or, to use John Niles’s terms, as ‘a form of play, a mental theater’ that ‘not only gives voice to a given mentality or worldview, but is also in which issues of worldview are precisely what are at stake’.6 Beowulf did not function simply as a static repository of moral or political wisdom, nor as a mere historical record of tribal strife in Scandinavia but was ‘open to human experience and hence to conflicting perspectives…simultaneously inviting or demanding interpretation and resisting interpretive closure’.7 Aesthetic enjoyment, entertainment, historical representation, and anxiety are not mutually exclusive, and it is the latter reaction to the poem that has been so far understudied. Whether or not we fix a date on this poem, we still need to account for the emotional life of the poem as it circulated before, within, and beyond its early eleventh-century manuscript. Who would have read Beowulf at different points in time and space? How would they have related and reacted to it? What made successive textual and emotional communities come back to the same text again and again?

200  Beowulf in Times of Anxiety If reader-response theory has taught us anything, it is that different audiences do not read the same text, even while formally the text they interact with is the same.8 Thus even if Beowulf did remain a stable text for 300 years, it is unrealistic to presume that, say, a Mercian 720s audience and a 1030s Wessex one would have read the poem in the same ways, would have latched onto the same things in it, or would have experienced the same emotions at the same moments in the text. There is consistent evidence for these differences in the very way the poem has been preserved in writing. The scribal errors (particularly of proper names but also archaic poet words) in the manuscript of Beowulf point to the likelihood that by the eleventh century, ‘the traditions informing Beowulf were no longer active and productive, and no longer known to every member of the poem’s audience’ or, indeed, to the two scribes copying what was probably an earlier manuscript of the poem.9 As I suggested in Chapter 1, it is conceivable that Beowulf, as much as Widsith or any other verse history, may have been enjoyed very differently by different audiences. The early eighth-century audience they were probably composed for ‘could fill in the blanks and understand the subtleties of the briefest allusion to a heroic narrative – an audience containing members probably named after the heroes to which he alludes’.10 By contrast, for the tenth- and eleventh-century audiences for which the manuscripts that preserve them to this day, these hero names, poetic formulas, and entire story-world may have seemed strange and distant yet vaguely familiar, possibly read with a Tolkienesque frisson within a very different constellation of intertexts, discourses, and socio-political conditions than those influencing their earlier reception. Yet the very fact that such poems were preserved even after their original intertextual points of reference were no longer available (or no longer interesting or culturally authoritative to this late audience) needs to be accounted for, which I aim to do in this chapter. They evidently still preserved their historical representational force, which enabled new audiences to make sense of new events and evolutions by emplotting them in this older narrative framework and projecting their emotions on them. For instance, an early tenth-century audience in Wessex could have read Beowulf as a myth of origins at a time of nation-building when the concept of Angelcynn came to label a new ethnic group that would include all Christian Anglo-Saxons and Danes under the leadership of the West-Saxon royal dynasty and would have responded to cultural anxieties about the definition and borders of this identitarian configuration that was still being negotiated. Hence, this social context proposed by John Niles and Craig Davis as a benchmark for dating the text can be reinterpreted to point to a specific context in which the poem was read rather than to a terminus post quem.11 An early eleventh-century audience, however, would have read Beowulf, much like The Battle of Maldon, in a more sombre socio-political context related to inner instability and the

Beowulf in Times of Anxiety  201 external Viking threat. Rag-picker verse histories are particularly open to this accretive process of recontextualization, which is easily explained by their collaborative, constellative truth procedure outlined in Chapter 3. This last chapter is dedicated to unveiling some of these different potential readings discernible in English verse histories, particularly Beowulf and, in a final subsection, Maldon. Such an understanding supports my argument in Chapter 1 that the cultural horizons, the social and political circumstances, the textual expectations of these audiences differed greatly, and the label ‘Anglo-Saxon England’ used to encompass all of them and many others whose distinctiveness has not survived in the textual record elides these differences. As Stephen Harris reminds us, the “Anglo-Saxon” of the age of Bede is not the same as the “Anglo-Saxon” of the age of Alfred, and ‘whenever we speak of their stories, we are actually talking about two different theys’.12 Indeed, even the same individual might read the same text at different points of their lives and experience it as different theys, playing out on a smaller scale this process of recontextualization. A telling example can be found in Felix’s Life of St. Guthlac, which first presents us with Guthlac as a young member of seventh-century Mercian martial nobility eagerly listening to songs of the deeds of his ancestors (his father, Penwalh, was a descendent of the legendary Icel, part of the Anglian genealogies going back to the migration-era rulers of Angeln), who decided to become a warband leader when ‘he remembered the valiant deeds of heroes of old’ (valida pristinorum heroum facta reminiscens).13 However, later in life, Guthlac decided to abandon his warlike way of life and become a hermit: ‘When, with wakeful mind, he contemplated the wretched deaths and the shameful ends of the ancient kings of his race in the course of past ages’ (cum antiquorum regum stirpis suae per transacta retro saecula miserabiles exitus flagitioso vitae termino contemplaretur).14 Here we have a rare insight into the possible receptions of GHP by members of an audience it was meant for and their emotional reactions to it. Young Guthlac reads GHP as a source for his personal genealogy (we recall the force of this cultural schema for early medieval elites from Chapter 2) and his tribal history, as well as an idealized representation of the martial honour-based way of life. The older Guthlac, after taking part in real war and later internalizing Christian morality, is inclined to see the same GHP as an indictment of this heroic way of life and a rejection of the genealogy-based emplotment of his personal history. This dynamic conceivably played out in myriad ways with each of these texts and in each act of reception they enjoyed from multiple textual and emotional communities. Beyond the common discussions around dating the composition of these texts, we are thus left with the crucial task of recovering an emotional dating of Beowulf and other rag-picker verse histories – in other words, understanding the ways in which the same text would have been

202  Beowulf in Times of Anxiety experienced affectively and recontextualized by different audiences. In this final chapter, I aim to provide a starting point for this work of emotional archaeology by tracing anxieties around race, ethnicity, and masculinity that found their expression in Beowulf or that different audiences projected onto it throughout its early medieval existence. Methodologically, I rely on a theoretical paradigm combining readerresponse theory and the history of emotions while incorporating questions and approaches from gender studies, critical race theory (CRT), postcolonial criticism, and cognitive psychology in order to tease out anxiety in and around Beowulf, no matter how resistant it may be to revealing itself. This involves allowing oneself to become troubled by the poem and by the ways it was read, while perhaps troubling the very readers of this chapter in the process. I have found while dealing with these topics that gender, race, and ethnicity are too intimately entwined both in the text and its reception and, more surprisingly, that this conjunction is too severely understudied to focus on either one of these aspects in isolation. For, to take an identity category like race or ethnicity and treat it in isolation from other categories, such as gender, class, or sexuality, is ‘to risk universalizing one possible (usually privileged or dominant) experience’ of gender or race by ‘failing to consider the ways in which the meaning of “woman” or “man” differs as these other aspects of identity vary’ (or that of wealh or englisc or Geat).15 As Geraldine Heng remarks, ‘The ability of racial logic to stalk and merge with other hierarchical systems – such as class, gender, or sexuality’ allows race to function as class, ‘ethnicity’, religion, or sexuality.16 Unfortunately, most of the work done on such intersections of gender, race, and sexuality in the Middle Ages is relegated to post-Conquest England and Europe.17 The present chapter will contribute to these conversations by exploring joint constructions of race and masculinity in early medieval England. Thus in the following, I fix my gaze particularly on emotional communities made up of men, not because I assume that Beowulf is in any way fit reading for men only or that its audiences were or are made up of mostly men.18 Yet it is in the ways in which men are reading these poems through the ages that I find a fascinating and hitherto unexplored pervasive pattern of conjoined anxiety and aspirational projection, which in its turn provides significant clues to cultural and social change in the twenty-first as much as in the eighth or eleventh century, especially with regard to how gender, race, and ethnic belonging are constructed. Hence, the two main threads of my chapter are that (1) we can detect anxieties related to masculinity, race, and ethnicity in the ways men behave in and read Beowulf, and (2) these anxieties are almost always interconnected in complex ways so that our focus needs to be intersectional. This is why CRT and indigenous studies can help us see that Beowulf could be read as relating to both the racialized Britons and Danes at

Beowulf in Times of Anxiety  203 different historical moments, while social psychology and gender studies cast light on the troublesome dynamics of proving oneself to be a man in early medieval English society. Reconstructing the emotions felt in each of these cases provides a fuller understanding of intimacy in anxiety that I argue was an important, although largely overlooked, dimension of Beowulf in its early medieval socio-emotional context. In doing so, I hope to spark a conversation regarding the ways in which Beowulf and medieval ‘Germanic heroic’ masculinity have always been a site of both utopia and anxiety for communities of men desiring an ethnically pure, hypermasculine mythical origin and the dangers inherent in this project.

Angling for Anxieties: A Method One need not go very far to find anxiety in Beowulf. It is present throughout the text, expressed in OE words ranging from general terms from the semantic field of sadness or trouble, such as cearu (‘sorrow, anxiety’), sorh, (‘care, anxiety, sorrow, grief, affliction, trouble’), murnan, (‘to be sad, be anxious, to mourn’), and meornan (‘to care, feel anxiety, trouble oneself about anything’), to the more specific or contextually connoting anxiety words, such as bysgu (‘business, labour, care, toil, difficulty, trouble, affliction, anxiety’) and wea (‘woe, misery, evil, affliction, trouble, anxiety’): þæt ys sio fæhðo ond se feondscipe,/ wælnið wera, ðæs ðe ic [wen] hafo/ þe us seceað to sweona leoda (‘such is the adversity and the enmity,/ the slaughterous hate of men, which is why I am anxious/ that the Swedish folk will seek us’, Beowulf 2999–3001). Here Wiglaf (the lone faithful retainer who returns to Beowulf’s side during his fight with the dragon) provides an example of one of the great causes of anxiety in the poem and, more specifically, to the Geats that were the audience for the unfolding drama of Beowulf’s demise – tribal and personal enmity and their accompanying causes and effects: choosing loyalty to one’s kin over one’s lord, the compulsion towards vengeance and rash words, and the hot-headedness of young warriors’ unmanly (or too manly) behaviour, as well as monsters. Indeed, it seems that the general mood in the heroic world of the poem is that of anxiety interrupted by brief calms and timid celebrations of victory (during which, however, gloomy songs foreshadowing future anxiety are often sung). Nonetheless, this is not the kind of anxiety that is the primary object of this chapter. In the previous example, ‘anxiety’ is used to translate a range of words with loosely connected meanings, ranging from ‘distress’ to ‘trouble’, ‘woe’, and ‘fear’ proper. Yet just as ‘sadness’ is not equivalent to ‘depression’, although nowadays, it is often described as such in popular discourse, (localized) ‘fear’ or (generalized) ‘apprehension’ is not the same as ‘anxiety’. In psychology and psychiatry, anxiety has been defined as ‘an unpleasant state of inner turmoil, often accompanied by nervous behaviour, such as pacing back and forth, somatic complaints, and rumination’

204  Beowulf in Times of Anxiety or as ‘a feeling of uneasiness and worry, usually generalized and unfocused as an overreaction to a situation that is only subjectively seen as menacing’.19 Still, while I am neither interested in its pathological aspect nor claiming to diagnose anxiety disorders in OE texts, it is in this more precise sense of anxiety that I will use it henceforth. Furthermore, I am more interested in the anxieties experienced by the poem’s diverse audiences than by its characters, although the latter are important clues to the former. These emotions inhabit the space between the text and its readers, where, according to Wolfgang Iser, the work of literature actually comes into existence.20 These anxieties can be reconstructed from clues within Beowulf that are corroborated with the different socio-cultural horizons and constellations of texts in which the poem was read. Focussing on just the text would further the fiction of a poem suspended in time, doing the same cultural and emotional work for any reader anywhere at any time in early medieval England. But, as John Niles reminds us, why would a thegn’s understanding of this poem tally with a monk’s? A king with a ceorl’s? A man with a woman’s? A jaded old man’s with a hot-headed boy’s? A Dane’s with a Saxon’s? If the modern reception of Beowulf sometimes more closely resembles a kennel of yapping dogs than a massed chorale, then this lack of unison is understandable given that polyphony is an outstanding feature of the poem itself.21 And it is through this ‘interplay of individual textual voices generated by its external and internal readers’ that Beowulf generates meaning.22 Thus, following Allen Frantzen, I focus on the gaps or areas of indeterminacy in the narrative as ‘sites for reading and writing’, places where readers of the poem create meaning as they construct ‘their’ Beowulf.23 Indeed, there are numerous scenes of reading, composition, and recitation of other texts and songs which place the poem in layers of mise en abyme from a traditional poet in Heorot’s almost simultaneous putting into verse of Beowulf’s victory over Grendel (Beowulf 867–71) to the hero’s retelling of his own adventures as he gets back in his Geatish homeland (Beowulf 1999–2151), which in its turn contains a description of Danish king Hrothgar’s composition of poetry (Beowulf 2105–14). It is clear that the poem delights in bringing to the fore the issue of its own making.24 For all these acts of creation, there are internal (or homodiegetic) audiences, and, as Janet Thormann argues, ‘Whatever the audience of the poem was and is…it is imaged in the audience within the poem that receives the scop’s song with pleasure’ (or, as the case may be, anxiety).25 Throughout the poem, the narrator offers countless images of fictive or dramatic audiences that respond to the unfolding of events as surrogates for real readers or listeners.26 Through their responses to the

Beowulf in Times of Anxiety  205 action of the poem or sometimes simply through the direction of their gaze, these internal audiences guide the reactions of the external audience and thus serve as a functional model for the audience listening to the poem.27 By extrapolating from these in-text reactions, we may be able to reconstruct the responses of external (or heterodiegetic) audiences to the poem itself.28 For these purposes, reader-response theory provides a fruitful paradigm to understand the successive acts of reading a poem like Beowulf as so many acts of co-creation of the poem.29 In Iser’s view, through the experience of reading, readers respond to structures present in the text and act as co-creators of the work through a process of ‘concretization’ whereby the reader’s imagination is required as each reader fills in ‘gaps’ and areas of ‘indeterminacy’ present in the text.30 In this understanding of the reading process, both the text and the reader are endowed with power in the creation of meaning, as meaning ‘must clearly be the product of an interaction between the textual signals and the reader’s acts of comprehension’.31 Something that is not always emphasized when using this approach is that it is not a merely cognitive process but also (or primarily) an emotional one. Similarly, the emotions of the audience are conceivably anticipated in the very text: the way in which the homodiegetical audiences react to narratives inside the text is an invaluable resource for any attempt to reconstruct audience emotions to the framing text itself. I am interested in how this dynamic works not only within textual but also emotional communities (the two can but need not overlap). In Barbara Rosenwein’s terms, emotional communities are groups in which ‘people adhere to the same norms of emotional expression and value-or devalue-the same or related emotions’.32 They can be viewed as sharing what Foucault called a common ‘discourse’: ‘Shared vocabularies and ways of thinking that have a controlling function, a disciplining function’, or what Bourdieu called ‘habitus’: ‘internalized norms that determine how we think and act and that may be different in different groups’, or, again, what some sociologists call ‘group styles’: in which ‘implicit, culturally patterned styles of membership filter collective representations’ that may include ‘vocabularies, symbols, or codes’, without exactly corresponding to any of these theoretical concepts.33 The explanatory power of Rosenwein’s concept lies in its balancing the constructed and performative nature of emotions with their apparent idiosyncrasies and individual characteristics, as well as in its emphasis on the communal nature of emotions. Although intuitively they may seem to issue from our innermost individuality, the way we express them is part of a web of socio-cultural meaning. And it is in their expression to others or to ourselves (in which case the discursive shape it takes is still part of a field of socio-culturally determined possibilities) that they graduate from affect (the ‘biological portion of emotion’, the ‘unvarying physiological

206  Beowulf in Times of Anxiety mechanism’) to emotion, wherein an affect is placed within a script or a story.34 Thus emotions are ‘borrowed and ascribed, moving across and between individuals, and never coming absolutely to rest within them’.35 Recent work on emotions in OE literature shows the complex interplay of feelings found in texts like Beowulf, which ‘indicates an appreciation on the part of the poem and audience for the intricacies of emotional states’.36 Kristen Mills explores the intersecting complex emotions described in Hrothgar’s farewell to Beowulf (sorrow at the hero’s departure, fear they will not meet again, strong affection, prospective langað (‘longing’) for Beowulf, which he keeps confined in his breast), all of which the poet subtly connects to other moments of grief and sorrow. We can imagine this must have affected deeply at least some audiences, although certainly in different ways. Each member of the audience related to these emotions through similar ones they experienced, but they could have been as diverse as grief for a brother dead in a recent skirmish, longing for an old friend living abroad, or affection for a daughter who would have to leave the home one day. In each of these acts of reception, the reader (or listener) processes the emotions in the text through a double filter that makes this experience at once individual (by relating it to one’s own emotional life) and collective (by placing it within the spectrum of affect deemed socially and culturally acceptable by the textual-emotional community one is part of – in the terms used earlier, by integrating it in the latter’s ‘group style’, ‘habitus’, or ‘emotional discourse’). At this juncture lies the type of intimacy with and around Beowulf that I wish to explore: sharing an emotion with fellow humans immersed in the same socio-cultural matrix while preserving a portion of it that will always remain deeply personal, kept deep within one’s breostcofa (OE ‘heart, mind’, literally ‘breast chamber’).

Nervous Laughter: Wrestling with Monsters Yet some anxieties might have been less context-dependent and more pervasive for early medieval audiences of Beowulf. Rather than the straightforward anxiety occasioned by the doom and gloom prevalent in the poem, I will now consider a moment that is troubling in a more oblique fashion. During Beowulf’s fight with Grendel’s mother, an embarrassing episode occurs. After he has pulled her to the floor, and she has grabbed him, he falls on his back on the floor while Grendel’s mother sits astride him, having pulled her short sword: ‘Ofsæt þa þone selegyst, ond hyre seaxe geteah,/ brad ond brunecg, wolde hire bearn wrecan,/ angan eaferan’ (Beowulf 1545a–47a) ‘then she sat on the hall-guest and drew her short seax, broad and burnished, she wanted to avenge her son, her only child’. For Beowulf, this is a position that is at once dangerous and embarrassing, and for my purposes, it is one that makes not

Beowulf in Times of Anxiety  207 only him but also, as Fred Robinson noticed, scholars and students alike uncomfortable. In a 1994 article, Robinson asks the question, ‘Did Grendel’s Mother Sit on Beowulf?’ and argues that this was not the only way the OE of the passage could be construed. His motivation for reconsidering the translation is that like the students in our classes, the translators of the poem…are often uncomfortable with the meaning which the glossaries stipulate for ofsittan. To avoid the comic indignity of Beowulf’s being sat upon, they fudge the verb’s meaning in artful ways.37 Yet, as Dana Oswald argues, in order to use this weapon effectively, Grendel’s mother must be on the same level as Beowulf, and since he is on the floor, the choreographic logic of the scene requires her to be on top of him.38 What reactions would men in an early medieval English audience have had to this scene? Vicarious fear for the hero’s life is one possible response, comic embarrassment would have been another, or both emotions at the same time. Oswald suggests that Beowulf’s passive posture, however temporary, is alarming because of the resulting gender instability, which is what makes students and translators uncomfortable.39 Thus Beowulf being topped by Grendel’s mother is not so much comic, as Robinson suggests, but alarming, and the nervous laughter it can provoke in audiences (both modern and medieval) is a response to a deeper anxiety this situation brings out in men, especially men who define themselves according to the scripts of hegemonic masculinity.40 In Oswald’s terms, Beowulf is at the mercy of a phallic woman who does indeed exceed the boundaries of gender, and of sexuality, in this fight. In claiming and employing these phalli, she symbolically castrates Beowulf, removing from him the excess masculinity gained by his fight with and defeat of Grendel, even if her blade never pierces his body.41 Whether one agrees with the psychoanalytical description of the situation or not, it is clear from Robinson’s account of modern reactions to this disquieting situation (the first time in the poem when Beowulf’s life seems to be genuinely endangered) that it has been perceived at least by some audiences as uncomfortably comic. The scene certainly is narrated in such a way as to resemble sexual intercourse, as other scholars have previously remarked, and, as such, it ‘plays out anxieties about female sexuality’ while also illustrating anxieties about Beowulf’s own sexual identity.42

208  Beowulf in Times of Anxiety While rolling around on the floor of Grendel’s mother’s cave, Beowulf is described as ‘battle-hard’ (Beowulf 1539a), ‘beadwe heard’, and ‘þa he gebolgen wæs, feorhgeniðlan’ (Beowulf 1539b–40a) ‘swollen (or enraged) by the life-enemy’. Although these also work as combat metaphors, the whole episode has been described as ‘a lengthy erotic double entendre riddle fused into a longer narrative poem’, employed purposefully to paint Beowulf as full of a very masculine vigour but also as a sexually engaged combatant.43 As Oswald deftly argues, Beowulf’s response to Grendel’s mother is sexual: ‘She is an abjected figure, a taboo, who is simultaneously repulsive and fascinating, an uncanny embodiment of what the hero both rejects and desires’.44 But there is more at work here than discomfort about forbidden erotic impulses. Beowulf’s masculine authority in this sexually charged battle is called into question not only by his near defeat by Grendel’s mother but also by his problematic relationship with swords.45 The possession of weapons, especially of swords, is closely related to masculine identity in early medieval English culture.46 One of the main OE terms for ‘man’, wæpnedmann, and a variety of compounds signifying the male sex attested copiously from royal wills to vernacular poetry testify to the understanding of ‘weapon’ as metaphorical penis in the most pragmatic sense: wæpnedcild (male child), wæpnedhealf (male line), etc.47 Hence, taking up Stacy Klein’s question, in a world in which ‘masculinity hinges so crucially on martial exploits, what happens when a woman takes up arms and subsequently acquits herself with great élan?’48 For in Grendel’s mother taking up Grendel’s arms and other weapons, she is ‘appropriating and revising masculine identity and acting as phallic mother, which demands a response similarly laden with sexual overtones’.49 On the other hand, what happens to the masculinity of a man when his sword fails in a medieval honour-based culture constructed around warmaking? Anxiety ensues, especially when the hypermasculine hero has his weapons repeatedly break down – the last time is in his final fight with the dragon. The first instance is during his attempt to stab Grendel’s mother (although Grendel being impenetrable to a weapon can be counted as the first time his sword is useless), when his sword completely fails him – it is simply unable to penetrate her flesh. This forces Beowulf to take up ‘a phallus that belongs not to men, but to giants’, which Oswald sees as an ‘ephemeral and external excess that demonstrates his own profound impotence’ since the masculine authority by which he eventually manages to kill Grendel’s mother is not his own but prosthetic, ‘imbued in the sword, not the man’ so that ‘it is the sword’s virility, not his own, that grants him victory’, emphasized by the giants’ sword melting away immediately after his hard-gained victory.50 It is understandable why this scene in particular would have provoked the anxiety of male audiences, whether warriors or members of an aristocratic monastic foundation, and why it continues to trouble students and

Beowulf in Times of Anxiety  209 scholars alike to this day. The male protagonist (and in their identification with him, male audiences too) finds himself ‘at the mercy of a “phallic woman”, monstrous because overstepping the boundaries of gender and of sexuality’.51 The nervous laughter that it might have provoked may be seen as both a sign of deep-seated anxieties about masculine identity and an occasion for deeper male bonding in the intimacy of a shared anxiety. Recent research on the cognitive science of emotions argues that laughter is an essential behaviour for helping to de-escalate negative emotional experiences, as well as a social emotion that increases the willingness of people sharing a laugh to disclose intimate information.52 While the cause for this nervous laughter lies in the uncomfortable recognition of the eroticism of the encounter between the hero and the monster, the anxieties it points to are deeper than that. For this episode is not an isolated instance of monstrous bodies troubling its medieval male audiences beyond straightforward fear and disgust; the violent mutilation of Grendel and Grendel’s mother that Beowulf performs bears witness to this. As Grendel’s mother continues her attack, Beowulf decapitates her: ‘Yrringa sloh/ þæt hire wið halse heard grapode,/ banhringas bræc; bil eal ðurhwod/ fægne flæschoman’ (Beowulf 1565b–68a) ‘he angrily struck so that it cut hard through her neck, broke bone-rings; the sword went entirely through the body doomed to die’. But he does not stop here. He turns his attention and rage to the aldorleasne (‘lifeless’, Beowulf 1587a) body of Grendel and proceeds to decapitate it: ‘Hra wide sprong/ syþðam he æfter deaðe drepe þrowade/ heorosweng heardne, ond hine þa heafde becearf’ (Beowulf 1588b–90b) ‘the corpse burst wide open, when after death it suffered a blow, a hard sword stroke, and then he cut off his head’. The poet draws our attention to the fact that Grendel is already dead in this passage so that we understand the oddity of Beowulf’s action. As Oswald explains, since he came to the mere knowing that Grendel is already dead and he defeated his mother (the stated objective for this venture), this final action ‘seems dictated not just by rage, but perhaps by shame, and certainly by the desire to re-assume his enemy’s excessive power and masculinity’ since once he has removed the heads of his enemies, he can ‘rearticulate his mastery over their bodies’.53 What emotional reaction could early medieval English male audiences have had to this moment located at the very centre of the poem (speculative though this endeavour may be)? After a first half of the text organized in a see-saw pattern of rising and falling, of alternating hope and fear, success and failure, and a crescendo of greater threats and greater triumphs,54 the mutilation of the monstrous bodies appear as a cathartic peak to the tension accumulating for 1,500 lines. Now all threats are removed from the world, and a less complex heroic poem could have ended right there, possibly after a brief coda celebrating the hero’s success. In its audience, this moment could have conceivably occasioned not

210  Beowulf in Times of Anxiety merely relief but also a vicarious satisfaction in the utter destruction of the monsters that had threatened in so many ways not only Beowulf but also the entire culture of the world of the poem while making the listeners or readers squirm at the mixture of sexuality, monstrosity, and unnatural martial prowess present in both of the hero’s foes. In effect, what Beowulf performs and what his male audiences would have reveled in is a symbolic castration, a vehement act of erasure via dismemberment, the motivation for which is, according to Oswald, the very real threat that the sexed bodies of these monsters represent to humankind, as well as to masculinity.55 Contemporary scholars, especially Jeffrey Cohen and Asa Mittman, have demonstrated the importance of monsters and monstrosity in medieval literature.56 Dana Oswald, on whose work I have relied in this section, has argued that the monsters that are a part of the landscape of OE literature and culture embody ‘social and sociosexual concerns and anxieties’ while also representing ‘illicit desires and prohibited practices’.57 Yet as I argue further down, this reasoning can and should be extended to racial alterity – the Grendelkin were also conceivably the focus of anxieties about race and ethnicity, which connect to those about gender in troubling ways.

Beowulf the Wrecca: The Monster Within While the dynamics described earlier may seem a straightforward strategy to enhance the audience members’ negative reactions to the monsters and hence augment their identification with the protagonist, let us not forget that it is not just Grendel and his mother who are monstrous but also the hero himself. Beyond his burst of shame and anger that makes him mutilate the corpses, Beowulf has much in common with Grendel.58 They are each angenga or anhaga (‘solitary’, Beowulf 449a and 2368a), gebolgen (‘swollen up with rage’, Beowulf 723b and 1539b), healðegnes (‘hall-thanes’, Beowulf 142a and 719b), and wrecca (‘wanderers in foreign lands’); both killed 30 at once, so they share a co-extensive identity, being each other’s alter ego.59 Furthermore, deducing from the comments of the poet or some of the characters’ reactions to the protagonist, Beowulf may not have been perceived as an exemplary hero or a perfect leader, ‘nor did darkness lie exclusively outside of him’, and as such, even for the most sympathetic audience, he might have appeared as ‘a man who [was] not fully likable or understandable’.60 Indeed, Scott Gwara has argued very convincingly for the poem’s ambivalence about Beowulf as a feature of the text itself rather than as a post-factum projection stemming from a modern indecision as to which cultural paradigm (such as the ‘heroic code’ or Christian morality) the hero is judged against, when in fact, ‘this dual consciousness comprises the poet’s subject’.61 Far from being unconditionally in awe of the hero, not all the inhabitants of Beowulf’s warrior society approve of his confidence,

Beowulf in Times of Anxiety  211 and some worry about the consequences of his success. The problematic aspects of Beowulf’s actions or motivations that are explicitly (although sometimes ambiguously) touched on in the text are often elided in scholarly conversations because of the assumption that he is either presented as heroic (according to a putative ‘Germanic heroic code’) or sinful. Gwara’s reading proposes instead that characters in the poem debate Beowulf’s motivations and effectiveness, allowing for (and anticipating) a wide spectrum of reactions to the same issues in the audience outside the poem.62 In this understanding, what he terms ‘the subaltern perspective’ is the main source of conflicting opinions on the hero, and it consists of comments made by members of the male warrior community who would bear the brunt of the consequences of Beowulf’s actions.63 For instance, while Unferth’s accusation against Beowulf are usually read as ‘mean-spirited’, ‘envious’, or ‘cowardly’, it highlights a common anxiety of the warband, especially since, as Michael Enright has argued, the þyle holds an important warband position as the king’s official spokesman: he ‘expresses a majority opinion – not the view of cowards or rogues but of Hrothgar’s fighting men – that engaging Grendel is foolhardy’.64 Thus, Unferth speaks for the subordinate warrior when he suggests that ‘Beowulf overemphasizes his own accomplishments, embraces recklessness, and interferes in a situation that the Danish warband has resolved, however unhappily’ (by ruining the fragile but working compromise by which the Danes abandon the hall for Grendel’s use at night).65 As Gwara explains, from the subordinate perspective, ‘Beowulf’s boast to kill Grendel without weapons sounds more like arrogance than confidence’, and this is precisely what provokes anxieties about ‘the potential for immoderation that he seems to express’, which is consistent with the self-serving behaviour of a wrecca.66 For the core of Gwara’s argument about the essential ambivalence of the poem’s perspective on the hero is located in his powerful demonstration that Beowulf is perceived as a wrecca by most at Hrothgar’s court. This does not mean he is necessarily condemned to depart his homeland forever – he is, however, socially liminal, for ‘wreccan are exiled for the same ruthless ambition that motivates other foreign fighters seeking glory abroad’.67 So while Beowulf is not technically an exile, his appearance at Heorot prompts suspicions as to his reason for being there, and the most likely answer is that he comes as one of these two types of mercenary fighters. The unstated issue here is the anxiety that Beowulf could cross ‘the behavioral threshold separating wreccan from other adventurers’, an anxiety voiced each time he is compared to one of the three famous (but ‘sociopathic’) exiles: Sigemund, Hengest, and Heremod, all of which are ‘figures of supreme ability whose motivations remain arguably impulsive, solitary, and socially marginal’, thus casting Beowulf’s leadership as highly problematic.68 The OE term wlenco captures the ambivalent nature of Beowulf’s heroic (over)confidence. For although in an honour-based society, all

212  Beowulf in Times of Anxiety members of the warrior elite are motivated by the pursuit of glory, wreccan are ‘disposed to maximize their prestige’ at the expense of their subalterns in the warband, making their deeds look more foolhardy than brave, ‘especially when men of lower status, ambition, and skill suffer for their zeal’.69 Hence Wiglaf’s reproach of Beowulf is woven into his postmortem panegyric: ‘Oft sceall eorl monig anes willan/ wræc adreogan, swa us geworden is’ (‘often many men, because of the will of one, must endure distress, as has happened to us’, 3077a–78b). It is not hard to imagine the reaction of many early medieval English audiences, however eager in the pursuit of honour in everyday life, being congruent with Wiglaf’s. Still, anxiety towards the destructive potential of the socio-emotional currency of all honour-based societies – other people’s esteem – need not exclude admiration for a character who earns it as much as humanly possible and beyond. This is why reading Beowulf as a straightforward condemnation of heroic excess and/or of the heroic culture promoting it is a limiting perspective: the protagonist needs to be a locus of identification and aspiration (however phantasmatic) for the male audience in order for the destructive potential of the heroic excess desired by both the former and the latter desire to become manifest. This double perspective built in the poem works emotionally by provoking anxiety in its audiences or providing the space for anxieties to manifest themselves. As Clare Lees remarks, Beowulf ‘is as much preoccupied with the ways in which aristocratic warriors dominate other men as they do monstrous Others’ and ‘aggression is central to the maintenance of power in the ruling families and is formulated throughout the poem in terms of a heroic ideology, or code’.70 This is likely consonant with the lived experiences of the male audience members of Beowulf, who were most likely part of honour-based elites, a culture ripe for violence and anxieties, with honour being ‘an unstable commodity, always rising or falling in value’, always in need of being proven usually by symbolic or material deeds of violence that can escalate easily.71 Recent research demonstrates how restrictive and damaging for the mental and physical health of their members similar traditional masculine gender scripts are.72 By masculine scripts, psychologists refer to the internalization of rules and roles by which hegemonic masculinity (the dominant form of masculinity in an honour-based society) operates.73 While the precise shape of hegemonic masculinity is always elusive, all the members of the society upholding it recognize when its rules are broken. The irony is that ‘few, if any men, can actually achieve these overly restrictive roles, try as they may, due to powerful socializing forces which tell men what “real men” must do and be’.74 The consequences of not being able to achieve these roles are grouped under the category ‘gender role strain/stress’, defined as ‘the constant pressure and awareness that a man is always, to some degree or another, failing in his most important task – to be a man’.75

Beowulf in Times of Anxiety  213 As David Rosen argues, such a male, leading his life according to the early medieval warrior elite scripts of hegemonic masculinity, has been ‘prepared to apprehend aspects of himself as threatening, to project those aspects upon the world, where he is able to recognize and subdue these versions of what he would consider his own “monstrosity”’.76 This is the anxiety about masculinity that the male audiences of Beowulf may have brought to the poem and that they witnessed playing out, pushed to its ultimate consequences. Beowulf’s response to the bodies of both Grendel (‘the hypermasculine cannibal’) and his mother (‘the threateningly reproductive and gender-transgressive female’) – namely, ‘to erase them through acts of dismemberment’ – highlights his own ‘conflicted masculine identity, as well as his inability to participate in the reproductive economy’ (since he has no son to perpetuate his male lineage, a crucial concern for all aristocratic men in heroic poetry).77 For early medieval English male audiences, this moment could have been both cathartic (as I argued earlier) and possibly frightening. The final failure-in-triumph of Beowulf and the hegemonic masculinity he epitomized would have given them pause.

Ellorgæstas: Guests in Their Homeland The Beowulf poet calls Grendel and his mother mearcstapan (‘borderwanderers’, Beowulf 104a and 1,348a). While metaphorically they straddle the boundaries of gender and humanity, they also occupy a very concrete liminal space. They live in the fens (104a), moras healdan, ‘holding the moors/ damp wastelands’ (103b and 1348b). Of course, symbolically, they live in a locus horroris et vastae solitudinis (‘place of horrors and lonely wastelands’), the kind of space that hermits go to banish the demons of the place or to found monastic communities (which is what happens eventually, see the following discussion).78 Although from a pragmatically economic point of view, the fens were an important land resource in medieval East Anglia (albeit not immediately desirable to the continental settlers), in the poetic corpus – and, hence, one may surmise in the collective imagination of the early medieval English – they were constantly used as a topos of desolate isolation. The two monsters are also described as ellorgæstas (1349a), usually translated as ‘aliens ghosts’. In virtually all editions of Beowulf, the aesc on ellorgæstas has been interpreted as a long vowel, making gæst mean ‘ghost’ or ‘spirit’ rather than ‘guest’, as a short aesc (æ) would make it. Of course, there is no way of telling from the manuscript – the early medieval English did not use macrons to mark long vowels in vernacular manuscripts – nor is the context always helpful; sometimes the monstrous enemies are called, ironically, ‘guests’ (thus the water monster at 1441a or the dragon at 2074b). At the same time, gæst can also mean ‘stranger’, if not ‘enemy’ – a guest is, of course, a stranger and can easily

214  Beowulf in Times of Anxiety turn out to be an enemy, and the anxiety about this potential of guests to turn inimical is pervasive in Beowulf. It is significant that two of the three examples from the Bosworth-Toller entry showcasing this second (apparently contradictory) meaning of short-aesc gæst (basic meaning ‘guest’) are taken from Beowulf, where it is used to describe Grendel (102a) and the dragon (2312a). It is not hard to imagine that an oral performer could have played on the quasi-homonymy of gæst and gǣst, as well as the semantic sliding between the threatening and the benign senses of the former word to underscore the blurring of the lines between them in everyday life.79 My point is that Grendel and his mother would have been seen not necessarily as demonic but as troublesome guests populating the edges of the territory inhabited by the Danes.80 She is, after all, idese onlicnes (‘in the likeness of a woman’, 1351a), while he is on weres wæstmum (‘in the form of a man’, 1352a); she avenges her son as any human kinsman would, and they are the descendants of Cain, who is, however evil, a human. Thus ellorgæstas could be translated as ‘guests from elsewhere’. There is some irony in that they are acknowledged as indigenous to the land (‘landdwellers in days of old’ had seen them, 1354a–55a), and yet they are now dwelling on its edges, discursively pushed to the margins of humanity and gender in the process and described as ‘guests from elsewhere’.81 As Stephen Harris reminds us, in ‘literary negotiations of communities’, foreigners and outsiders are often depicted as monstrous or as ‘fearsome variations on existing creatures’, being somehow recognizable by the very ‘terror of fear’ which ‘they strike in humankind’ (quae maximum formidinis terrorem humano generi incutiunt, as the author of the eighth-century Liber monstrorum puts it).82 This can happen just as readily when it comes to a conquering population negotiating its relationship with indigenous people. More precisely, I suggest that Grendel and his mother could have been the focus for early medieval English audiences projecting anxieties about their relationship to the autochthonous Britons and, at a later stage, to the Danes that always intersected with anxieties about masculinity since these audiences would have identified with aristocratic men, placed in existential opposition to the gender-­defying abjected monsters. In her study of sexual humour in the Exeter Book riddles, Nina RulonMiller brought to light the constellation in which the few Welsh personae appearing in them are placed – it consists of oxen, the concepts of yoking or fettering, dark skin, and the borderland.83 For instance, in Riddle 72, an ox tended by a sweart hyrde (‘dark herdsman’ – compare the sweart ond saloneb, ‘black and dark-faced’, servant in Riddle 49) is bunden under beam (‘bound under yoke’) and treads the mearcpaþas Walas (‘the paths on the Welsh march’). In Riddle 12, swarthy Welsh men are tightly fettered with ox-leather bonds, and a dark Welsh woman works on an object made of ox-hide. Rulon-Miller proposes as the most likely

Beowulf in Times of Anxiety  215 solution to Riddle 52 ‘yoke of oxen led by a female slave’ since the riddle subjects are associated with binding (ræpingas…nearwum bendum/ gefeterade fæste togædre, ‘bound ones…with narrow bonds fettered fast together’), as well as with a female Welsh slave (Wale).84 The semantic linkage between the notions ‘native Briton’ and ‘slave’ is unsurprising for any student of early medieval England: both meanings are equally well exemplified in the Bosworth-Toller for the OE word wealh, and often, it is not clear at all that the early medieval English writing, reading, or hearing the word would have cared to make a distinction. And as John Tanke points out, ‘The violence which makes a slave out of a Welsh person parallels the violence in language which makes one say “slave” when one means “Welsh” and “Welsh” when one means “slave”’, thus making wealh a word ‘whose usage dramatizes its meaning: it is a word from whose otherness there is no escape’.85 The association with oxen is thus quite transparent: both Britons and oxen are perceived in the riddles as servile creatures, physically or socially fettered, even if not all wealhas are slaves. But their description as ‘dark-skinned’, coupled with their association to animality, point to their essential otherness, with the potential of sliding into something that is so different to the ideal audience of the riddles as to be almost inhuman.86 The derogatory connotations attached to wealh or the adjective wilisc are plentifully attested: ‘shameful person’, ‘bad servant’ (in the OE translation of Matthew 24:50, although the non-pejorative þeow was used to translate the Latin servus in every other instance), Ælfric equates weala win with crudum uinum (rough, inferior wine) and has a sinner speak wealode mid wordum (‘strangely’, ‘impudently’).87 The associations survive up to this day in present-day British slang words, such as ‘to welsh’ (to cheat) and ‘welsher’ (an untrustworthy person).88 David Pelteret demonstrates the large array of status terms available to early medieval English, a necessity for ‘a society deeply preoccupied with class and ethnicity’.89 Such discursive practices are not divorced from material realities and are often used to reinforce and sustain material inequity by creating ‘a reality in which it is reasonable for a few to control and to possess the material at the sacrifice of the well-being of others’.90 The first written occurrence of wealh is already a juridical one, appearing in the seventh-century Laws of Ine, where the ‘inferior social position’ of the Britons in early medieval England was made law, which is to say made official policy a distribution of power already in place.91 In Ine’s laws that deal with wergeld, the free wealh is ‘accorded only half the value of his English counterpart’, while in laws concerning oaths, ‘a man charged with stealing or harbouring stolen cattle had to produce an oath of sixty hides if he were accused by a wealh, whereas if the accuser were English the oath required was doubled’.92 What this meant in real life is amply demonstrated by two studies, one from the perspective of genetics and demography and the other from

216  Beowulf in Times of Anxiety that of socio-economics.93 Using computer simulations, genetic information, legal proscriptions of intermarriage in other European conquered areas, and early medieval English law codes, Thomas, Stumpf, and Härke propose an ‘apartheid-like’ society as the best explanation for the disproportion between the prevalence of “Germanic stock” Y-chromosomes in modern Britain and the probable number of Continental invaders.94 In this case, being an early medieval English male thus meant a greater chance at reproductive success. This stirred some controversy among scholars of early medieval England, but this model is certainly not new, and the use of the term ‘apartheid’ is a reasonably factual description of what likely happened with the Britons in very early medieval England.95 Alexander Woolf showed that this strategy (outlined by the Laws of Ine) is very effective since in the short term, the system would protect most individual Britons, while ‘the erosion of their economic base would generally be so gradual as to be barely perceived on the basis of individual experience’.96 Yet in the long run ‘individual British households would, one by one, become bankrupt and break down, with children being sold into slavery or sent to live with relatives as prospect-less ­hangers-on’.97 In what Woolf describes as a ‘long drawn-out process of economic decline’, individual Britons would have ‘found themselves drifting into Anglo-Saxon households, as slaves, hangers-on, brides and so forth, but they would have come into these communities as one among many’.98 This explains their very limited power to impact the cultural or linguistic identity of the early medieval English communities they became part of, which would have become ‘ethnic sausage machines, recycling stray biological material in such a way that it would not carry its ethnicity with it into the next generation’.99 Against the assumption that the Welsh and English ethnic affiliations would have been fluid, Woolf explains that if firm legal distinctions existed between Britons and early medieval English, ‘individuals and communities will not have slipped from one identity to another with ease’ since there were strong economic incentives to preserve this segregation from an English perspective.100 In this society, the assimilating Britons may well have been seen as ‘guests from elsewhere’ (while obviously indigenous) within the English households or communities that integrated them as people belonging to a different category – socially and ethnically but also perhaps racially. Before discussing the implications of understanding the early medieval English perspective of Britons as racialized, I wish to draw attention to what can be understood as a symbolic narrativization of the early medieval English experience of colonizing the Britons. As Jeffrey Jerome Cohen argues, both Felix’s Vita Guthlaci and Guthlac A are ‘texts obsessed by the annexation of new land and its conversion into secure possession’ – the demons hate Guthlac not so much for his holiness as for his seizing their dwelling place, their long-held territory (Guthlac A, 205–13).101

Beowulf in Times of Anxiety  217 On the one hand, ‘Guthlac’s colonization of the demons’ cherished home’ (the Fenlands, much like the fen-dwelling Grendelkin) in the hagiographies ‘re-enacts in miniature the dispossession of that very territory’ by the Germanic tribes settling there in the fifth century.102 On the other hand, according to Cohen, it enacts a border war taking place on the other side of the kingdom in which Guthlac himself had been involved as a young warrior. When he composed the Vita Guthlaci, Felix probably had the ongoing wars against the Brittones in mind since he describes them as ‘the implacable enemies of the Saxon race’ [infesti hostes Saxonici generis], troubling the nation of the English [Anglorum gentem] with their attacks, their pillaging and their devastations of the people’.103 Thus, Felix ‘was able to construct a flattened, pan-racial “Anglo-Saxon” identity for the island (Saxonici and Anglorum appear to be synonyms here)’ by providing a common enemy threatening the cohesion of this gens, natio, þeod against whom Guthlac fights, at first as warrior and later as saint.104 In a chapter entitled in the OE translation Hu þa deofla on brytisc spræcon, ‘How the devils spoke in Brittonic’, the Vita Guthlaci narrates that just as the Welsh are invading Mercia from the west during the reign of King Coenred, in the fens, a crowd of demons ‘impersonates a band of British marauders and sets fire to Guthlac’s dwelling, attacking him with spears’.105 Guthlac chants a psalm and the demon Britons vanish velut fumus, like smoke, potentially a symbolic parallel to the ‘powerful and attractive group fantasy’ of a Mercian ‘manifest destiny’ entitling it to conquer British lands while their inhabitants simply vanished.106 It is not too great of a stretch of the imagination to think of early audiences of Beowulf perceiving the interactions of its protagonist and the Grendelkin in very similar terms. Felix reports that Guthlac himself listened to what was probably heroic poetry in seventh-century Mercia, so, adapting a thought experiment of James Earl, if we imagine him as a reader of Beowulf, the othering of the Britons as demons in his hagiographies can be understood as participating in an already extant discourse in which the indigenous population was represented symbolically as monstrous fen dwellers.107 By the time the Exeter Book riddles were inscribed in their extant forms, when the Welsh kingdoms no longer posed an immediate military threat, and the Britons inside English territories had been integrated as slaves, servants, or brides, this discourse could have developed into a condescending (rather than outright monsterizing) linkage of the wealhas with animality, bondage, and dark skin. Of course, Guthlac probably did not actually read Beowulf, but his story realistically was one of the trajectories an Anglo-Saxon nobleman’s life could take. Elite English in 730s Mercia could (and conceivably some of them did) read or listen to both Beowulf-like GHP and martial hagiographies like Vita Guthlaci, and their understanding of (and emotional

218  Beowulf in Times of Anxiety reaction to) one of them would have been coloured by their reading the other.108 If, as Cohen argues, both lives of Guthlac are ‘suffused with colonial desires, displacing into religious history a version of the engagement that was then occurring as martial history’,109 then so is Beowulf, only here the colonial desire is projected on a poetic narrative that probably began its life as folktale conjoined to legendary histories of pre-migration time rather than on hagiography. Similarly to how Felix offers up the Britons as a common enemy against which he defines a presumably homogeneous early medieval English race, the Beowulf poet creates a sense of superior moral solidarity among its readers and listeners who share the political values of Hrothgar and Beowulf, regardless of ethnicity or regional sense of identity, as long as they oppose the descendants of Cain, who are radically dehumanized. As Craig Davis argues, in Beowulf ‘there is neither Geat nor Dane when it comes to opposing evil monsters who embody values destructive of national kinship’, and the poet joins the Danes, their subjects, allies, and even their human enemies ‘into a kind of ethnicity of salvageable heroes intended by God for the blessings of national kinship, the temporal equivalent of spiritual salvation’.110 Thus, Davis remarks, the primal ethnic dichotomy in the poem is not ‘between Dane and Heathobard or Geat and Swede, but between royalist and renegade, human and monster, Sethite and Cainite’.111 I suggest that this is more than about ethnic difference: while the poet conveys a keen sense of ethnic differences among tribes which to us are mere names, there is an uncrossable gulf between all of these ethnicities taken together (however inimical to each other) and the cyn of the Cain-descended ghost-guests. Thus early medieval English audiences of Beowulf, whether in early eighth-century Mercia or tenth-century Wessex, would have conceivably read this monsterization of the indigenous dwellers of the Danish fens as an opportunity for transcending ethnic and regional differences through their own cultural experience (however dim) of othering the Britons as inhuman enemies against which a pan-racial early medieval English identity could emerge. This need not have been a self-conscious judgement but an affective response consonant with an entire discourse that was itself part of a network of power relations legally expressed in the apartheid instated by the Laws of Ine. In this respect, Beowulf fulfilled a need for not just a search for origins but also a desire for a trans-ethnic or even (as I argue in the next section) racial sense of identity. As Susan Reynolds demonstrated, the road to an ‘Anglo-Saxon’ sense of unity was long and troubled: ‘Until the tenth century their sense of unity was frequently undermined by political conflict, and we cannot assume that they had a manifest destiny of political unity’.112 Those whom we are accustomed to call ‘Anglo-Saxons’ (itself an anachronistic concept levelling a five-century period and a region with significant local specificities into an ahistorical notion) were definable

Beowulf in Times of Anxiety  219 primarily by their military allegiance and by regional identity, and early medieval English sources themselves seem to be uncertain of how to define ethnic belonging.113 Until the emergence of early West-Saxon as a literary language (most probably in the decades around AD 700), even among the Germanic-speaking groups, ‘there was no cohesive sense of shared English identity and significant dialect variation’.114 And, as Stephen Harris argues, ethnic disunity sometimes produces (or invites) an ideal and imagined unity, ‘an abstract community expressed as, but in rhetoric at odds with, political reality’, which was certainly the case for the various early medieval English kingdoms until Alfred the Great and his successors which started referring to themselves as kings of the Angli Saxones, Angolsaxones, Anglosaxones, or Angulsaxones.115 The only position afforded to the Britons in this uncertain but wished-for sense of trans-regional and trans-ethnic identity was that of abject Other. Reactions to Beowulf and Vita Guthlaci would have fed on this uneasiness about ethnic and racial identity. These anxieties about indigeneity and establishing an opposing transethnic (almost racial) sense of identity always intersected with anxieties about masculinity: after all, Beowulf is not a poem about all Danes, Geats, or, obliquely, early medieval English or even Mercians but about a particular class of men, associated by their aristocratic rank and their lords. What unifies the male warbands of different ethnicities is ‘a certain ethic of warrior behaviour’ or, in other terms, a hegemonic masculinity.116 Hence, regional or ethnic difference is not an issue if you are a male aristocrat fighting the monsterized ancient dwellers of the land. Thus the previously discussed erasure enforced by Beowulf on the monstrous bodies would have been doubly satisfying to its audiences – as an annihilation of a threat to a pure masculine and ‘Anglo-Saxon’/Mercian/Englisc identity. And yet, just as Guthlac’s recognition of the brytisc language and his haunting by the Welsh-speaking demons point to troubling internal differences, so too does Beowulf’s participation in monstrosity haunt the poem so that his final mutilation of the bodies of the Grendelkin brings to mind a history of violence against the Britons. At least some members of the early medieval English male audiences of Beowulf would have shared these conjoined anxieties about gender, race, and ethnicity in the intimacy of listening to the poem together. Their emotional reactions could have included a sense of cathartic jubilation at the erasure of the monstrous bodies and an anxiety which they were able to exorcize by projecting it on figures of abjection such as the Grendelkin or the demonic Britons.

Crushing the Laðan Cynnes: On ‘Anglo-Saxon’ Race, Again When the dragon comes spewing flames over the land of the Geats, at the end of the poem, the poet remembers how Beowulf ‘æt guðe forgrap Grendeles mægum,/ laðan cynnes’ (‘in battle he crushed Grendel’s

220  Beowulf in Times of Anxiety kin, the hated race’, 2353a–54a). As Dana Oswald remarks, this use of the plural recalls Beowulf’s use of the plural feondum (‘foes’) in his first telling of the fight and that not only is Grendel exterminated but also his whole ‘race’: ‘The feud can only be ended when the audience is absolutely assured that no Grendelkin remain’.117 I have so far used the term ‘race’ to describe both the Britons in the imaginary of early medieval English ethnic difference and the Grendelkin, but in this section, I focus on the issue of how productive the use of this word is, especially considering its problematic baggage, as well as its connection to gender.118 In his doctoral thesis investigating the OE vocabulary of ethnic and racial belonging, Christopher Roberts finds two different concepts of identity: one based on physical or generic similarities (usually denoted with the lexemes cynn ‘sort, kind’ and mægð ‘family’) and one based on social categories (consisting of the terms leode ‘people, tribe’ and þeod ‘people, nation’, each of which are frequently related to social concepts like place, authority, and collective name).119 Leode seems to have originally connoted ‘a smaller group tied to an abstract sense of place and generic leadership’, while þeod probably connoted ‘a group under the power of a larger authority in control of a named territory’. 120 Both are used in Beowulf for naming all the tribal or ethnical groups (for instance, Geata leode at 1213b, or fremde þeod at 1691b). Cyn is reserved for families (cynnes Wægmundinga, 2813b–14a), the species of animals that God created (‘lif ac gesceop/cynna gewhwylcum’, 97b–98a), humanity (moncynn, 164b et passim) and the grouping made up of Grendel, his mother, and Cain (107a). When the Danish coastguard asks Beowulf for his group’s identity (Beowulf 244a–57b), he responds by saying first that they are ‘of man-kind’ (‘we synt gumcynnes’, 260a), secondly that they are of the tribe or people of the Geats (‘Geata leode’, 260b), and only then that they are Hygelac’s hearthmates and that his father is Ecgþeow (261a–63b). There is no clear taxonomy here, and at times, metrical and alliterative reasons could have led the poet to use one of them rather the other. Roberts’s model still applies, but the translation of cyn in the case of Cain’s cyn, of which Grendel and his mother are the last survivors, needs to be problematized. It is not conceived simply as a family or a lineage, although that too is part of the concept. They are certainly not a leod or þeod – the separation between them and any other tribal or ethnical grouping in the poem is greater than between any difference among the latter. Still, they are not simply unthinking beasts, like the water monsters that Beowulf fights on his way to Grendel’s hall. They are human, descended from Cain, yet almost as if they are part of another species of human, although not one that God created but one which separated itself from humanity because of its sinfulness. Still, the Grendelkin look like humans (she is idese onlicnes, at 1351a, he is on

Beowulf in Times of Anxiety  221 weres wæstmum, at 1352a), are intelligent, possess emotions human enough to feel rage at being left out of the human community of Heorot, have humanlike social structures that predicate revenge for one of their own, and live in a hall (an inversion of Heorot though it may be). For Beowulf, cyn is also a way to describe the community of all men – his masculinity is ever at the forefront, which is significant since, as we have seen, both Grendel and his mother tread the boundaries of gender, so by defining himself and his troop as part of the race of men (rather than humans), he predicates belonging to humanity on the quality of maleness. Next, I turn to this intersection of gender and race or even species. Thus, if race is usually used to describe ‘a group whose boundaries are relatively difficult to cross’, and ethnicity, ‘a group with relatively porous boundaries’, then both the Grendelkin and the Britons of the early medieval English imaginary of Beowulf and the lives of Guthlac form a race.121 The use of the term ‘race’ in any premodern context has been criticized for early medieval England in particular.122 Asa Mittman has critiqued its use for the monstrous people in The Wonders of the East by arguing that it places modern concepts on medieval texts and images, while also unwittingly suggesting that ‘notions of race are somehow “natural” or inevitable, rather than historically situated phenomena’.123 But while in The Wonders of the East, the monstrous peoples are looked upon with a spectrum of emotions ranging from wonder and respect to disgust,124 the Grendelkin are monsterized from the beginning, and at the same time, in the audience’s projection of anxieties about the indigenous populations, racialized. And while in the Wonders, the author’s interest in these peoples is ‘soteriological: the individual marvels are arranged to show the hierarchical spectrum of those people to whom God offers grace’, the Grendelkin is godes andsacan (‘God’s enemy’, 786b), bearing God’s fury (godes yrre bær, 711b), already in hell (feond on helle, 101b), and, as such, surely beyond salvation.125 Of course, modern terms such as ‘race’ and ‘ethnicity’ do not correspond exactly to the variety of Latin and OE words they are used to translate. But in the case at hand, to translate the cynn to which Grendel, his mother, and Cain belong to as either ‘family’/’kin’ or ‘tribe’ would be disingenuous because it would erase the possibility that is certainly present in the text for early medieval English audiences to relate to the Grendelkin as their culture did to the Britons. As Geraldine Heng argues, the refusal to use ‘race’ when discussing medieval phenomena de-stigmatizes the impacts and consequences of certain laws, acts, practices, and institutions in the medieval period, so that we cannot name them for what they are, nor can we bear adequate witness to the full meaning of the manifestations and phenomena they install. The unavailability of race thus often colludes in relegating such manifestations to an epiphenomenal status.126

222  Beowulf in Times of Anxiety Used in this understanding, ‘race’ does not have to be equated to any presumably biological difference between groups of people (in the way it is popularly used). Rather, in Heng’s definition, ‘race’ is ‘a structural relationship for the articulation and management of human differences, rather than a substantive content’, or in other words, ‘a repeating tendency, of the gravest import, to demarcate human beings through differences among humans that are selectively essentialized as absolute and fundamental, in order to distribute positions and powers differentially to human groups’.127 According to this definition, the Britons are clearly racialized in the early medieval English imaginary, belonging to a system of racial categorization in which several categories intersected: bodily distinctions (Britons described repeatedly as sweart, dark), language (Welsh as demonic speech in Guthlac or as incomprehensible gibberish in Ælfric), and social status (according to the Laws of Ine and the economic consequences thereof, also evident in the very term wealh and its entire lexical field of derogatory connotations). While I am not claiming that the Grendelkin are present in the poem due to a self-conscious choice of the poet to represent the indigenous people which the Germanic tribes found in England, I suggest that some early medieval English audiences would have perceived them (emotionally, if not discursively) as monstrous echoes of their real-life relationships with both the Britons and, later, the Danes. CRT, from which I derived my frame of interpretation here, was originally developed to address the specific needs of the African American community, along a ‘black-white’ binary, which, while not similar to the relationship between the Britons and the invading Germanic tribes, is useful for understanding how groups are constructed discursively ‘within a social space and held there by institutional practice’.128 As other race theorists have recently proposed, the dynamics of racialization are protean and work differently, for instance, for Native Americans than for African Americans. Thus, a framework such as TribalCrit, developed by Bryan M. J. Brayboy is meant to emphasize indigenous peoples’ racialized experience of colonization, which ‘is not just an experience of racial oppression’ but also ‘primarily an experience of territorial oppression’.129 In the light of the devolution of the socio-political and economic status of the Britons as described by the apartheid model of Woolf and Härke et al., this seems a particularly relevant model through which to understand the indigenous experience of the colonized wealhas. Ryan Craig and Victoria Davis argue that ‘the practices to bring Indigenous peoples into the fold of Whiteness’ (being sent to boarding schools, converted to Christianity, forced to switch from collective to individualized forms of land ownership) were in fact strategies for acquiring their land and resources.130 While no similar concerted effort to turn the Britons into English took place in early medieval England, this seems to have happened in time due to the racialization instated

Beowulf in Times of Anxiety  223 legally, economically, discursively, and symbolically through such texts as Beowulf and Vita Guthlaci.

Giant Women and Haunting Danes: Race and Gender around Beowulf As I have suggested throughout this chapter, novel anxieties could be projected onto what was already a focus for racialized abjection – namely, Grendel, and especially his mother, in new socio-political (and manuscript) contexts. Brian McFadden describes the social context in which the Nowell Codex was assembled as rife with ‘anxieties caused by tenth and early eleventh-century Viking Invasions, the Benedictine Reform, and eschatological concerns provoked by the coming millennium’ such that the OE works gathered together in this compilation could be understood as different forms of ‘resistance of foreign others to containment in either a social or narrative order’.131 As we have seen, connections to socio-political developments in early medieval England may serve as clues to layers of emotional responses to the poem. Consider Helen Damico’s interpretation of Fitt II of Beowulf (characterized by Klaeber as ‘Grendel’s Reign of Terror’) as a poetic rendering of the Danish series of attacks on England in the early eleventhcentury, paralleling the more poetic account of the same events in the C-text of the Anglo-Saxon Chronicle.132 Setting other kinds of dating aside, Damico’s argument details one possible act of reception, wherein an element of the poem (Grendel as a terrifying character rampaging the Danish countryside) could provide a dream screen on which early eleventh-century audiences projected their own anxieties about Danish attacks on England. If Grendel and his mother could become a focus for anxieties about Welsh indigeneity, they could certainly fulfil a similar role for Danish invasion. What is more, Kathrin Powell argues that the Beowulf manuscript as we have it, probably written during the latter part of the reign of Æthelred the Unready, displays a new preoccupation with the tension between (not always good) rulers and foreigners, which would have been particularly relevant in the context of the Viking raids when ‘the ability of those in power to keep that aggression at bay was such an important issue for the English’.133 Thus, just as I have suggested that it would have been difficult to read Beowulf in early eighth-century Mercia while ignoring the intertextual context of the growing cult of Guthlac and his experience with the real-life and ghostly Britons and the other way around, Powell explains that ‘it would have been difficult to read the works in this manuscript without reflecting on one’s own rulers and their foreign policies’.134 Indeed, for an early eleventh-century English audience, Beowulf was now a part of a novel textual constellation. Many of the clashes with foreigners depicted in the other works in the manuscript would have

224  Beowulf in Times of Anxiety been reminiscent of recent events in England and would have provided a different context for the emotional reception of Beowulf. Powell deftly argues that in the Letter of Alexander to Aristotle, Alexander the Great’s accumulation of wealth attracting monstrous foes might mirror ‘in an exaggerated way, the wealth that attracted the Danes to England at this time and about which the Anglo-Saxons seem to have felt a certain moral unease’, while Alexander’s slaughter of a race of monstrous women in The Wonders of the East could possibly recall the St. Brice’s Day massacre of 1002.135 This is when, according to the Anglo-Saxon Chronicle, ‘the king [Æthelred] commanded that all the Danish men who were among the English be slain (…) because the king was informed that they wished to ensnare his life, and afterwards all of his counselors, and afterwards have this kingdom’.136 Yet, as Eileen Joy remarks, this was not an isolated event, the racialized violence of the attacks being part of a larger discourse seeking to instate ‘bodily purity through the elimination of supposedly impure elements’.137 A charter of 1004 from the monastery of St. Frideswide at Oxford records that a group of Danes who had ‘sprung up’ in England like ‘cockle amongst the wheat,’ had been forced to flee to the barred church, the doors and bolts of which they broke by force to get inside, and once securely settled there, an angry mob of their neighbors set fire to the church, apparently burning the Danes inside, along with ‘its ornaments and books’.138 Yet who were these monstrous women? The Wonders of the East reports that ‘for heora micelnesse hy gefylde wæeron from þæm miclan macedoniscan Alexandre’ (‘because of their great size, they were killed by Alexander the Great of Macedon’).139 Æthelred’s order, like Alexander’s action, ‘appears to target an entire group of foreigners indiscriminately for the acts of some of its members’ and, as such, Powell suggests, would not have reflected positively on him.140 But early medieval English audiences reading these texts with recent events in mind might have reacted differently to Beowulf, too. Grendel is described as shaped like a man (‘on weres wæstmum’, 1352a), but ‘mara þonne ænig man oðer’ (‘greater than any other man’, 1353), so since his mother is idese onlicnes (‘in the likeness of a woman’, 1351a), we can deduce that she, too, is a monstrously large woman. For a reader or auditor of Beowulf in the context of the Nowell Codex, it would be difficult to ignore the connection to the giant women killed by Alexander the Great in The Wonders of the East. Thus, in this new constellation made up of racialized Danes, giant monstrous women, and bad kings, the different texts collected in a new compilation based on the theme of tension between foreigners and rulers, the emotional reaction of early medieval English male audiences to the Grendels, and, especially, to

Beowulf in Times of Anxiety  225 Grendel’s mother, would have mingled anxiety (possibly including guilt), jubilation, and disgust for these doubly transgressing bodies. In a striking parallel with what we have seen happening with Beowulf and Grendel’s mother, the readers of the OE Wonders, Eileen Joy argues, May very well have been drawn to the image of these women as both frightening and attractive, leading to feelings of both sexual desire (or sexual astonishment), followed by feelings of violent revulsion, the relief of which (through dark enjoyment) might have been provided by Alexander’s decisive act of execution.141 Corroborated with the considerable divergence between accompanying images in the Vitellius and Tiberius manuscripts, the text seems to have greatly disturbed its audiences.142 In this scenario, echoing (and providing a parallel to) the violent erasure of the monstrous bodies in Beowulf, ‘Alexander’s murder of the animal-human women would be the natural outcome of Alexander’s (and the author’s) sudden recognition of the fragility of the subject’s “own and clean self,” which would need to be purified by some violent means’.143 This is not a theoretical framework imposed on a medieval text – as Joy following Powell argues, an early medieval English readership could have entertained such a notion since their behaviour on St. Brice’s Day ‘suggests that the Danes had become for the English a homogeneous Other who existed solely to deprive them of their every enjoyment – life, land, wealth, and power – and who were unworthy of human sympathy’.144 In the senses proposed earlier, the Danes became racialized, if not as part of a long history of structural oppression, at least in the early medieval English imaginary at the turn of the eleventh century. In reading and listening to this new Beowulf, early medieval English audiences would have celebrated (and been haunted by) the erasure of the racialized abjected bodies of the race of Grendel, who once again became the focus for shared anxieties about race, ethnicity, and masculinity.

Notes 1 McCormack, Wilcox, Jorgensen (2015). For emotional community, see Rosenwein (2006: 2 and 24–5). 2 For discussions on the ways in which anxiety recombines with shame and on how crucial shame is to the experience of homosociality, see Frank and Sedgwick (1995: 6 and 147–60) and Sedgwick (2003: 35–121). I am grateful to Erica Weaver and Daniel Remein for pointing me to this illuminating work and for their thoughtful comments on the essay on which this chapter is based. 3 Niles (1998b:149). All quotations from Beowulf are from Beowulf (2008), and all translations are mine unless stated otherwise. 4 Neal (2008: 10). 5 James Earl sees Beowulf as a dream-screen in Earl (1994: 129–36).

226  Beowulf in Times of Anxiety

6 Neal (2008: 10). Niles (1998b: 146). 7 Hansen (1988: 176–77). 8 Iser (1989). 9 Neidorf (2014: 123). See Neidorf (2013, 2017). For the earlier archetype of the poem, see Lapidge (2000). 10 Neidorf (2014: 123). 11 Niles (2007: 66–92). Davis (2006a: 117). 12 Harris (2003: 35). 13 Neidorf (2014: 133). Guthlac (1956: 80–1). 14 Neidorf (2014: 134). Guthlac (1956: 82–3). 15 Spelman (1988: 186), quoted in Kruger (1997: 158). 16 Heng (2011: 262). 17 Kruger (1997). For research on this conjunction in modernity, see also Nagel (2003). 18 For a classic critique of this masculinist approach to Beowulf, see Lees (1994). 19 Seligman, Walker, and Rosenhan (2001: 275). Stavrakaki and Lunsky (2007: 113). 20 Iser (1974: 274–5). 21 Niles (1998a: 10). 22 Frantzen (1990: 181). 23 Frantzen (1990: 182). Kisor (2010: 237). 24 Oswald (2010: 68). Creed (1963). 25 Thormann (2010: 223). 26 Niles (1998b: 157). 27 Kisor (2010: 241). 28 For the terminology of narratological theory see Genette (1980). 29 Iser (1974: 274–5). 30 Kisor (2010: 237). 31 Iser (1978: 9), quoted in Kisor (2010: 240). 32 Rosenwein (2006: 2). 33 Rosenwein (2006: 25). Eliasoph and Lichterman (2003: 735), quoted in Rosenwein (2006: 25). 34 Nathanson (1992: 49 and 50). 35 Connor (2013: 14). 36 Mills (2015: 175). 37 Robinson (1994: 2). 38 Oswald (2010: 95). 39 Oswald (2010: 96). 40 See also Tomkins’s definition of shame as an incomplete lowering of interest (hence its characteristic expression in the downturned face) – we would perhaps like to look away, but there is enough interest that we are still motivated to bashfully or embarrassingly peek at the scene; see Frank and Sedgwick (1995: 135–36). 41 Oswald (2010: 96). 42 Chance (1986: 102). Horner (2006: 485). Oswald (2010: 94). 43 Davis (2001: 50). Oswald (2010: 94). 44 Oswald (2010: 94). 45 Oswald (2010: 95). 46 See also Gilian Overing’s chapter on swords and signs in Overing (1990: 33–67). 47 Christie (2004: 152). 48 Klein (2012: 41).

Beowulf in Times of Anxiety  227 49 Oswald (2010: 93). 50 Oswald (2010: 84 and 99). 51 Oswald (2010: 96). 52 Scott (2014). Aragón (2015). 53 Oswald (2010: 99). 54 Andersson (1991: 96). 55 Oswald (2010: 25). 56 Cohen (1996, 1999); Mittman (2006). 57 Oswald (2010: 67). 58 Cohen (1999: 1–28). See also Orchard (2003: 30–4); Greenfield (1982); Dragland (1977); Chadwick (1959). 59 Gwara (2008: 17). Köberl (2002: 98). 60 Dragland (1977: 606). Rosen (1993: 1–25). 61 Gwara (2008: 13). 62 Gwara (2008: 13). 63 Gwara (2008: 362). 64 Gwara (2008: 129). See also Enright (1998: 310). 65 Gwara (2008: 131). Gwara’s use of ‘subaltern’ is problematic since it refers to an aristocratic warrior class and is incompatible in the way it is most often used in the postcolonial criticism, as in the classical Spivak (1988). 66 Gwara (2008: 12 and 131). 67 Gwara (2008: 16). 68 Gwara (2008: 21 and 17). 69 Gwara (2008: 25). 70 Lees (1994: 142). 71 Baker (2013: 18). 72 Saucier et al. (2016); Cohen and Nisbett (1994). 73 Furman and Dill (2012: 103); Mahalik, Good, and Englar-Carlson (2003); Connell and Messerschmidt (2005). 74 Furman and Dill (2012: 103). 75 Furman and Dill (2012: 103). For gender role strain, see Pleck (1981). For a recent review of the literature, see Baugher and Gazmararian (2015). 76 Rosen (1993: 5). 77 Oswald (2010: 25). For the importance of perpetuating one’s male line, see Lees (1994: 141). 78 Deuteronomy 32:10. 79 Carolyn Anderson argues for the semantic indeterminacy of the two words, showing how editors interpreted the word according to whether they saw the Grendelkin as demonic – see Anderson (2001). 80 Since the publication of the essay on which this chapter is based, I have been informed that Adam Miyashiro had been independently working on a parallel idea. I regret not being acquainted with his work on the colonial dimensions of the poem’s presentation of the Grendelkin, albeit I have since found his study of the racial politics of Anglo-Saxonism particularly illuminating: Miyashiro (2019). See also Abram (2019). 81 See also Siewers (2003). Abram (2019). 82 Harris (2003: 12). Orchard (2003: 87–8). 83 Rulon-Miller (2000: 117). 84 Idem. 85 Tanke (1994: 24). 86 Cf. Heng’s (2018) new book on medieval race, The Invention of Race. 87 Rulon-Miller (2000: 115–16). 88 Rulon-Miller (2000: 114). Banham (1994: 155).

228  Beowulf in Times of Anxiety 89 Pelteret (1995: 3). Rulon-Miller (2000: 117). 90 Craig and Davis (2015: 100). 91 Faull (1975: 20–1) quoted in Rulon-Miller (2000: 114). 92 Faull (1975: 20–1). 93 But see Brady (2017), especially pages 3–16, where she demonstrates that there was space for different types of interactions between Britons and Anglo-Saxons, although as her book makes clear, this was an exceptional situation, being possible on the borderlands between the territories occupied by the two polities. 94 Thomas, Stumpf, and Härke (2006). 95 For an inspired discussion of the heated debate on ANSAXNET around the Apartheid article, see Joy (2006). 96 Woolf (2007: 128). 97 Woolf (2007: 129). 98 Idem. 99 Idem. Although, of course, the biological material itself has nothing to do with ethnicity – a minor imprecision in a compellingly argued essay. 100 Woolf (2007: 127). 101 Cohen (2003: 141). 102 Cohen (2003: 142). 103 Cohen (2003: 143). Guthlac (1956: 108–09). 104 Cohen (2003: 143). 105 Idem. 106 Cohen (2003: 144). Guthlac (1956: 109–10). 107 Earl (1994: 175–87). 108 Colgrave, in Guthlac (1956: 19) argues for a date c. 730–40 for the text. 109 Cohen (2003: 144). 110 Davis (2001). 111 Davis (2001). 112 Reynolds (1985: 414). 113 Reynolds (1985: 403–04). 114 Woolf (2007: 127). See also Hines (1994). 115 Harris (2003: 10). Reynolds (1985: 398). 116 Cf. Lees (1994: 140). 117 Oswald (2010: 111). 118 See Whitaker (2015), where he argues that we need to take for granted that the Middle Ages were raced in complex ways and to start unravelling them. 119 Roberts (2013: 119). 120 Roberts (2013: 120). 121 Harris (2003: 1). 122 Jordan (2001). For other critiques of the use of race for the medieval period, see Appiah (1990). 123 Mittman (2015: 39). 124 Austin (2002: 51). 125 Austin (2002: 28). 126 Heng (2011: 265). 127 Heng (2011: 268 and 267). 128 Craig and Davis (2015: 99). 129 Idem. For TribalCrit, see Brayboy (2005, 2013). For CRT in general, see Delgado and Stefancic (2012). 130 Craig and Davis (2015, 99). 131 McFadden (2001: 91). 132 Damico (2011: 150).

Beowulf in Times of Anxiety  229 33 Powell (2006: 6). 1 134 Idem. 135 Powell (2006: 9). 136 Rositzke (1940: 55–6) quoted in Joy (2008: 222). A necessary caveat is that the order was probably directed only against the Danes who had recently settled in various parts of England, whether as traders, as mercenaries, not against those with Danish ancestry – see Keynes (1980: 204). 137 Joy (2008: 222). 138 Whitelock (1979: 591), quoted in Joy (2008: 222). 139 Powell (2006: 10). 140 Idem. 141 Joy (2008: 220). 142 Idem. For the different images of these giant women, see London, British Library, MS Cotton Vitellius A XV, ff. 105v–106r and London, British Library, MS Cotton Tiberius B V/1, f. 85r. 143 Joy (2008: 222). For the ‘own clean self’, see Kristeva (1982: 53). 144 Powell (2002: 157).

Bibliography Primary Sources Beowulf = Klaeber’s Beowulf and The Fight at Finnsburg, ed. by Robert D. Fulk, Robert H. Bjork and John D. Niles, fourth edition (Toronto: University of Toronto Press, fourth edition, 2008). Guthlac = Felix’s Life of St. Guthlac, ed. and trans. by Bertram Colgrave (Cambridge: Cambridge University Press, 1956). Rositzke, Harry August, ed., The C-Text of the Old English Chronicles (BochumLangendreer: Heinrich Pöppinghaus, 1940). Whitelock Dorothy ed., English Historical Documents I, c. 500–1042, second edition (Oxford: Oxford University Press, 1979). London, British Library, MS Cotton Tiberius B V/1, f. 85r. London, British Library, MS Cotton Vitellius A XV, ff. 105v–106r.

Secondary Sources Abram, Christopher, ‘At Home in the Fens with the Grendelkin’, in Dating Beowulf: Studies in Intimacy, ed. by Erica Weaver and Daniel Remein (Manchester: Manchester University Press, 2019), pp. 120–44. Anderson, Carolyn, ‘Gæst, Gender, and Kin in Beowulf: Consumption of the Boundaries’, The Heroic Age, 5 (2001) [accessed 24 January 2020]. Andersson, Theodore M., ‘Tradition and Design in Beowulf’, in Interpretations of Beowulf: A Critical Anthology, ed. by Robert D. Fulk (Bloomington: Indiana University Press, 1991), pp. 90–106. Appiah, Kwame Anthony, ‘Race’, in Critical Terms for Literary Study, ed. by Frank Lentricchia and Thomas McLaughlin (Chicago: University of Chicago Press, 1990), pp. 274–87. Aragón, Oriana, ‘Dimorphous Expressions of Positive Emotion’, Psychological Science, 26 (2015), 59–273.

230  Beowulf in Times of Anxiety Austin, Greta, ‘Marvelous Peoples or Marvelous Races? Race and the AngloSaxon Wonders of the East’, in Marvels, Monsters, and Miracles: Studies in the Medieval and Early Modern Imaginations, ed. by Timothy S. Jones and David A. Sprunger (Kalamazoo: Medieval Institute Publications, 2002), pp. 25–51. Baker, Peter S., Honour, Exchange and Violence in Beowulf (Cambridge: D.S. Brewer, 2013). Banham, Debby, ‘Anglo-Saxon Attitudes: In Search of the Origins of English Racism’, European Review of History, 2 (1994), 143–56. Baugher, Amy R., and Julie A. Gazmararian, ‘Masculine Gender Role Stress and Violence: A Literature Review and Future Directions’, Aggression and Violent Behavior, 24 (2015), 107–12. Brady, Lindy, Writing the Welsh Borderlands in Anglo-Saxon England (Manchester: Manchester University Press, 2017). Brayboy, Bryan McKinley Jones, ‘Toward a Tribal Critical Race Theory in Education’, The Urban Review, 37 (2005), 425–46. Brayboy, Bryan McKinley Jones, ‘Tribal Critical Race Theory: An Origin Story and Future Directions’, in Handbook of Critical Race Theory in Education, ed. by Marvin Lynn, Adrienne D. Dixson (New York: Routledge, 2013), pp. 88–100. Chadwick, Nora, ‘The Monster and Beowulf’, in The Anglo-Saxons, ed. by Peter Clemoes (London: Bowes, 1959), pp. 171–203. Chance, Jane, Woman as Hero in Old English Poetry (Syracuse: Syracuse University Press, 1986). Christie, Edward, ‘Self-Mastery and Submission: Holiness and Masculinity in the Lives of Anglo-Saxon Martyr-Kings’, in Holiness and Masculinity in the Middle Ages, ed. by P. H. Cullum and Katherine J. Lewis (Cardiff: University of Wales Press, 2004), pp. 143–57. Cohen, Dov, and Richard E. Nisbett, ‘Self-Protection and the Culture of Honor: Explaining Southern Violence’, Personality and Social Psychology Bulletin, 20 (1994), 551–67. Cohen, Jeffrey Jerome, ‘Monster Culture (Seven Theses)’, in Monster Theory: Reading Culture, ed. by Jeffrey Jerome Cohen (Minneapolis: University of Minnesota Press, 1996), pp. 3–25. Cohen, Jeffrey Jerome, Of Giants: Sex, Monsters, and the Middle Ages (Minneapolis: University of Minnesota Press, 1999). Cohen, Jeffrey Jerome, Medieval Identity Machines (Minneapolis: University of Minnesota Press, 2003). Connell, R. W., and J. W. Messerschmidt, ‘Hegemonic Masculinity: Rethinking the Concept’, Gender and Society, 19 (2005), 829–59. Connor, Steven, Collective Emotions: Reasons to Feel Doubtful, The History of Emotions Annual Lecture, Queen Mary University of London, 9 October 2013 [accessed 22 February 2020]. Craig, Ryan, and Victoria Davis, ‘“The Only Way They Knew How to Solve Their Disagreements was to Fight”: A Textual Analysis of Native Americans Before, During, and After the Civil Rights Movement’, in Doing Race in Social Studies: Critical Perspectives, ed. by Prentice T. Chandler (Charlotte: Information Age Publishing, 2015), pp. 89–125. Creed, Robert, ‘The Singer Looks at His Sources’, in Studies in Old English Literature in Honor of Arthur G. Brodeur, ed. by Stanley B. Greenfield (Eugene: University of Oregon Press, 1963), pp. 44–52.

Beowulf in Times of Anxiety  231 Damico, Helen, ‘Grendel’s Reign of Terror: From History to Vernacular Epic’, in Myths, Legends and Heroes: Essays on Old Norse and Old English Literature in Honour of John McKinnell, ed. by Daniel Anlezark (Toronto: University of Toronto Press, 2011), pp. 148–66. Davis, Craig R., ‘Redundant Ethnogenesis in Beowulf’, The Heroic Age, 5 (2001) [accessed 13 February 2020]. Davis, Craig R., ‘An Ethnic Dating of Beowulf’, Anglo-Saxon England, 35 (2006a), 111–29. Davis, Glenn, ‘The Exeter Book Riddles and Sexual Idiom’, in Medieval Obscenities, ed. by Nicola McDonald (York: York Medieval Press, 2006b), pp. 39–54. Delgado, Richard, and Jean Stefancic, Critical Race Theory: An Introduction, second edition (New York: New York University Press, 2012). Dragland, S. L., ‘Monster-Man in Beowulf’, Neophilologus, 61 (1977), 606–18. Earl, James, Thinking About Beowulf (Stanford: Stanford University Press, 1994). Eliasoph, Nina, and Paul Lichterman, ‘Culture in Interaction’, American Journal of Sociology, 108 (2003), 735–94. Enright, Michael J., ‘The Warband Context of the Unferth Episode’, Speculum, 73 (1998), 297–337. Faull, Margaret Lindsay, ‘The Semantic Development of Old English Wealh’, Leeds Studies in English, 8 (1975), 20–44. Frank, Adam, and Eve Kosofsky Sedgwick, eds, Shame and Its Sisters: A Silvan Tomkins Reader (London: Durham University Press, 1995). Frantzen, Allen, Desire for Origins: New Languages, Old English, and Teaching and Tradition (New Brunswick: Rutgers University Press, 1990). Furman, Rich, and LeConté Dill, ‘Poetry Therapy, Men and Masculinities’, The Arts in Psychotherapy, 39 (2012), 102–06. Genette, Gérard, Narrative Discourse: An Essay in Method, trans. Jane E. Lewin (Oxford: Blackwell, 1980 [1972]). Greenfield, Stanley B., ‘A Touch of the Monstrous in the Hero, or Beowulf Re-Marvellized’, English Studies, 63 (1982), 294–300. Gwara, Scott, Heroic Identity in the World of Beowulf (Leiden: Brill, 2008). Hansen, Elaine Tuttle, The Solomon Complex: Reading Wisdom in Old English Poetry (Toronto: University of Toronto Press, 1988). Harris, Stephen, Race and Ethnicity in Anglo-Saxon Literature (London: Routledge, 2003). Heng, Geraldine, ‘The Invention of Race in the European Middle Ages I: Race Studies, Modernity, and the Middle Ages’, Literature Compass, 8 (2011), 258–74. Heng, Geraldine, The Invention of Race in the European Middle Ages (Cambridge, Cambridge University Press, 2018). Hines, John, ‘The Becoming of the English: Identity, Material Culture and Language in Early Anglo-Saxon England’, Anglo-Saxon Studies in Archaeology and History, 7 (1994), 49–60. Horner, Shari, ‘Voices from the Margins’ in The Postmodern Beowulf: A Critical Casebook, ed. by Eileen A. Joy and Mary Ramsay (Morgantown: West Virginia University Press, 2006), pp. 467–500. Iser, Wolfgang, ‘The Reading Process: A Phenomenological Approach’, in The Implied Reader, ed. by Wolfgang Iser (Baltimore: Johns Hopkins University Press, 1974), pp. 274–94.

232  Beowulf in Times of Anxiety Iser, Wolfgang, The Act of Reading: A Theory of Aesthetic Response (Baltimore: Johns Hopkins University Press, 1978). Iser, Wolfgang, From Reader Response to Literary Anthropology (Baltimore: Johns Hopkins University Press, 1989). Jordan, William C., ‘Why “Race”?’, Journal of Medieval and Early Modern Studies, 31 (2001), 165–74. Joy, Eileen A., ‘Anglo-Saxons Were Apartheid Racists!’ (2006) [accessed 3 February 2020]. Joy, Eileen A., ‘The Signs and Location of a Flight (or Return?) of Time: The Old English Wonders of the East and the Gujarat Massacre’, in Cultural Diversity in the British Middle Ages: Archipelago, Island, England, ed. by Jeffrey Jerome Cohen (New York: Palgrave Macmillan, 2008), pp. 209–29. Keynes, Simon, The Diplomas of King Æthelred the Unready, 978–1016: A Study in Their Use as Historical Evidence (Cambridge: Cambridge University Press, 1980). Kisor, Yvette ‘The Aesthetics of Beowulf: Structure, Perception, and Desire’, in On the Aesthetics of Beowulf and Other Old English Poems, ed. by John M. Hill (Toronto: Toronto University Press, 2010), pp. 227–46. Klein, Stacy S., ‘Gender’, in A Handbook of Anglo-Saxon Studies, ed. by Jacqueline Stodnick and Renée Trilling (Oxford: Wiley-Blackwell, 2012), pp. 39–54. Köberl, Johann, The Indeterminacy of Beowulf (Lanham: University Press of America, 2002). Kristeva, Julia, Powers of Horror: An Essay on Abjection, trans. Leon S. Roudiez (New York: Columbia University Press, 1982). Kruger, Steven F., ‘Conversion and Medieval Sexual, Religious, and Racial Categories’, in Constructing Medieval Sexuality, ed. by Karma Lochrie, Peggy McCracken, and James A. Schultz (Minneapolis: University of Minnesota Press, 1997), pp. 158–79. Lapidge, Michael, ‘The Archetype of Beowulf’, Anglo-Saxon England, 29 (2000), 5–41. Lees, Clare, ‘Men and Beowulf’, in Medieval Masculinities: Regarding Men in the Middle Ages, ed. by Clare Lees (Minneapolis: University of Minnesota Press, 1994), pp. 129–48. Mahalik, J. R., G. E. Good, and M. Englar-Carlson, ‘Masculinity Scripts, Presenting Concerns, and Helping Seeking: Implications for Practice and Training’, Professional Psychology: Research and Practice, 34 (2003), 123–31. McCormack, Frances, Jonathan Wilcox, and Alice Jorgensen, eds, Anglo-Saxon Emotions: Reading the Heart in Old English Language, Literature, and Culture (Surrey: Ashgate, 2015). McFadden, Brian, ‘The Social Context of Narrative Disruption in The Letter of Alexander to Aristotle’, Anglo-Saxon England, 30 (2001), 91–114. Mills, Kristen, ‘Emotion and Gesture in Hroðgar’s Farewell to Beowulf’, in AngloSaxon Emotions: Reading the Heart in Old English Language, Literature, and Culture, ed. by Frances McCormack, Jonathan Wilcox, and Alice Jorgensen (Surrey: Ashgate, 2015). Mittman, Asa Simon, Maps and Monsters in Medieval England (New York: Routledge, 2006). Mittman, Asa Simon, ‘Are the “Monstrous Races”, Races?’, Postmedieval, 6 (2015), 36–51.

Beowulf in Times of Anxiety  233 Miyashiro, Adam, ‘Our Deeper Past: Race, Settler Colonialism, and Medieval Heritage Politics’, Literature Compass, 16 (2019), 1–11. Nagel, Race, Ethnicity, and Sexuality: Intimate Intersections, Forbidden Frontier (Oxford: Oxford University Press, 2003). Nathanson, Donald L., Shame and Pride: Affect, Sex, and the Birth of the Self (New York: W. W. Norton, 1992). Neal, Derek, The Masculine Self in Late Medieval England (Chicago: University of Chicago Press, 2008). Neidorf, Leonard, ‘Scribal Errors of Proper Names in the Beowulf Manuscript’, Anglo-Saxon England, 42 (2013), 249–69. Neidorf, Leonard, The Origins of Beowulf: Studies in Textual Criticism and Literary History (unpublished doctoral thesis, Harvard University, 2014). Neidorf, Leonard, ‘Germanic Legend, Scribal Errors, and Cultural Change’, in The Dating of Beowulf: A Reassessment, ed. by Leonard Neidorf (Cambridge: D.S. Brewer, 2017), pp. 37–57. Niles, John D., ‘Introduction’, in A Beowulf Handbook, ed. by Robert E. Bjork and John D. Niles (Lincoln, Nebraska: University of Nebraska Press, 1998a), pp. 1–12. Niles, John D., ‘Reconceiving Beowulf: Poetry as Social Praxis’, College English, 61 (1998b), 143–66. Niles, John D., Old English Heroic Poems and the Social Life of Texts (Turnhout: Brepols, 2007). Orchard, Andy, Pride and Prodigies: Studies in the Monsters of the Beowulf Manuscript (Toronto: University of Toronto Press, 2003). Oswald, Dana, Monsters, Gender and Sexuality in Medieval English Literature (Woodbridge: D.S. Brewer, 2010). Overing, Gilian, Language Sign, and Gender in Beowulf (Carbondale: Southern Illinois University Press, 1990). Pelteret, David A.E., Slavery in Early Mediaeval England: From the Reign of Alfred until the Twelfth Century (Woodbridge: Boydell & Brewer, 1995). Pleck, J. H., The Myth of Masculinity (Cambridge: MIT Press, 1981). Powell, Kathryn, ‘The Anglo-Saxon Imaginary of the East: A Psychoanalytic Exploration of the Image of the East in Old English Literature’ (unpublished doctoral thesis, University of Notre Dame, 2002). Powell, Kathryn, ‘Meditating on Men and Monsters: A Reconsideration of the Thematic Unity of the Beowulf Manuscript’, The Review of English Studies, 57 (2006), 1–15. Reynolds, Susan, ‘What Do We Mean by “Anglo-Saxon” and “Anglo-Saxons”?’, Journal of British Studies, 24 (1985), 395–414. Roberts, Christopher M., The Form, Aspect, and Definition of Anglo-Saxon Identity: A Study of Medieval British Words, Deeds, and Things (unpublished doctoral thesis, Arizona State University, 2013). Robinson, Fred C., ‘Did Grendel’s Mother Sit on Beowulf?’, in From AngloSaxon to Early Middle English, ed. by Malcolm Godden, Douglas Gray, and Terry Hoad (Oxford: Clarendon, 1994), pp. 1–7. Rosen, David, ‘The Armor of the Man-Monster in Beowulf’, in The Changing Fictions of Masculinity, ed. by David Rosen (Urbana: University of Illinois Press, 1993), pp. 1–25.

234  Beowulf in Times of Anxiety Rosenwein, Barbara H., Emotional Communities in the Early Middle Ages (Ithaca: Cornell University Press, 2006). Rulon-Miller, Nina, ‘Sexual Humor and Fettered Desire in Exeter Book Riddle 12’, in Humor in Anglo-Saxon Literature, ed. by Jonathan Wilcox (Cambridge: Cambridge University Press, 2000), pp. 99–126. Saucier, Donald A., et al., ‘Masculine Honor Beliefs: Measurement and Correlates’, Personality and Individual Differences, 94 (2016), 7–15. Scott, Sophie, ‘The Social Life of Laughter’, Trends in the Cognitive Sciences, 18 (2014), 618–20. Sedgwick, Eve Kosofsky, Touching Feeling: Affect, Pedagogy, Performativity (London: Durham University Press, 2003), pp. 35–121. Seligman, Martin E. P., Elaine F. Walker, and David L. Rosenhan, Abnormal Psychology (New York: W. W. Norton, fourth edition, 2001). Siewers, Alfred K., ‘Landscapes of Conversion: Guthlac’s Mound and Grendel’s Mere as Expressions of Anglo-Saxon Nation-Building’, Viator, 34 (2003), 1–39. Spelman, Elizabeth V., Inessential Woman: Problems of Exclusion in Feminist Thought (Boston: Beacon Press, 1988). Spivak, Gayatri Chakravorty, ‘Can the Subaltern Speak?’, in Marxism and the Interpretation of Culture, ed. by Cary Nelson and Lawrence Grossberg (Basingstoke: Macmillan, 1988), pp. 271–313. Stavrakaki, Chrissoula, and Yona Lunsky, ‘Depression, Anxiety and Adjustment Disorders in People with Intellectual Disabilities’, in Psychiatric and Behavioural Disorders in Intellectual and Developmental Disabilities, ed. by Nick Bouras and Geraldine Holt (Cambridge: Cambridge University Press, second edition, 2007), pp. 113–30. Tanke, John W., ‘Wonfeax wale: Ideology and Figuration in the Sexual Riddles of the Exeter Book’, in Class and Gender in Early English Literature: Intersections, ed. by Britton J. Harwood and Gillian R. Overing (Bloomington: Indiana University Press, 1994), pp. 21–39. Thomas, Mark G., Michael P.H. Stumpf, and Heinrich Härke, ‘Evidence for an Apartheid-Like Social Structure in Early Anglo-Saxon England’, Proceedings of the Royal Society B: Biological Sciences, 273:1601 (2006), 2651–57. Thormann, Janet, ‘The Subject of Language: A Psychoanalytic Approach to the Aesthetic of Old English Poetry’, in On the Aesthetics of Beowulf and Other Old English Poems, ed. by John M. Hill (Toronto: Toronto University Press, 2010), pp. 209–26. Whitaker, Cord, ‘Race-ing the Dragon: The Middle Ages, Race and Trippin’ into the Future, postmedieval, 6 (2015), 3–11. Woolf, Alexander, ‘Apartheid and Economics in Anglo-Saxon England’, in Britons in Anglo-Saxon England, ed. by Nick Higham (Woodbridge: Boydell & Brewer, 2007), pp. 115–29.

Index

Ælfric 72, 215, 222 Æthelred II (the Unready) 223, 224 Æthelweard 2, 34 Alcuin 7, 8, 33, 37, 82, 83, 117, 129, 131–34, 140, 154 Aldhelm 71–73, 117 Alfred the Great 1, 10, 13, 17, 39, 67, 72, 147, 148, 201, 214, 219 Alienation 129, 131, 136–45, 157 Ambiguity, xiv, 12, 13, 23, 24, 31, 33, 40, 44–46, 49, 75, 97, 110–16, 165, 173, 199, 210–13 Angelcynn xiv, 2, 48, 49, 118, 147, 148, 200 Anglo-Latin 36 Anglo-Saxon: concept of xii, 17–23, 26, 217, 218; critiques of the concept xiii, 18–22, 26; elite 1, 146, 217; England 8, 18, 20–22, 26, 93, 100, 127, 201; identity 21, 93, 200, 201, 218, 219 Anglo-Saxon Chronicle, xii, 2, 12, 13, 33, 36, 37, 47–50, 65, 91, 127, 223, 224 see Neo-bardic verse history Anglo-Saxonists 5, 20, 22, 27, 37 Anxiety 4, 15, 24, 50, 106, 113, 139, 162–69, 171–77, 179–81, 198, 202–209, 211–14, 219, 225; see also Alienation Apartheid 216, 218, 222 Aquitainian 167 Archive 39, 40, 42, 67, 99 Athelstan 47, 49 Auctoritas xiii, xiv, 1, 8, 19, 36, 38, 78, 98–102, 104, 107, 115–19, 132, 138, 145

Baldo of Salzburg 155 Bard see Oral poet Bardic verse history xiii, 4, 33, 37, 38, 45, 47, 83, 105, 111, 118, 128, 129, 131, 135, 143, 144, 146, 150, 154, 166–70, 174–78, 181 Battle of Brunanburh 33, 38, 45, 47, 49, 50 Battle of Maldon xii, 2, 13, 14, 23, 33, 47, 49, 50, 200, 201 Bavarian 142, 144, 150, 151 Bede 2 4–7, 67, 71, 75, 81–83, 90, 98–100, 102, 117, 129, 140, 147, 201; Historia ecclesiastica gentis Anglorum 65, 71, 72, 98–100 Beowulf 8–12, 15, 23, 24, 30, 31, 33, 40–50, 71, 76–83, 90–92, 94, 96, 100–107, 110–17, 127, 128, 130, 198–214, 217–25 Bible 99–101, 114, 119, 133, 144, 155 Biblical epic, xii, xiv, 2, 31, 34, 64, 119, 129, 135–44, 153, 178, 180, 181 Britons see Wealhas Burgundian 43, 153, 154, 162, 164–68 Byrhtnoth 13, 49, 50 Caesar 43, 90–92, 118 Cain 30, 101, 214, 218–21 Canonical history-writing, xii, 1–13, 16, 17, 26–30, 32, 34, 37, 50, 65, 67, 70–76, 81, 83, 93, 97, 99, 100, 111, 119, 128–30, 138

236 Index Carolingian 148: dissolution 129, 130, 158–63, 167, 170–79; elite xiv, 132–36, 142–45, 149, 156, 175; Empire xiv, 16, 26, 34, 49, 128, 129, 137, 147, 149, 157, 168, 179, 180, 198; intellectuals 152, 153, 156, 157; society 94, 133–38, 142, 147, 148, 169, 174–78, 180 Cassiodorus 149 Charlemagne 17, 129, 130–34, 142, 145, 155, 157, 159, 161, 162, 164, 167, 168 Charles the Bald 143, 148, 152, 155, 158, 159, 161, 170, 172, 174, 175 Cicero xiii, 5, 6, 32, 70, 97 Classical antiquity 1, 2, 10, 22, 43, 70, 78, 80, 118, 133, 149, 157, 160, 163 Cognitive approach xiii, 68–71, 80, 81, 98, 202, 205, 209; see also Schema Cognitive schema see Schema Collaboratio xiv, 31, 33, 43, 44, 46, 49, 97, 102, 107, 110, 111, 114–18, 201 Comitatus see Heroic ethos Community: emotional 205, 206, 211; political 14, 147–50, 156, 158, 161, 162, 164, 170, 219; socio-cultural 14, 25, 31, 68, 79, 82, 92, 99, 100, 108, 137, 140, 141, 143, 145, 148; speech 25, 108, 136–41, 151; textual 25, 31, 43, 50, 51, 106, 137, 145, 164, 206, 214 Conservatism xiii, 11–14, 25, 26, 40, 42, 97 Council of Clovesho, 7, 142 Critical race theory 202, 222 Cultural milieu see Socio-cultural context Cultural work see Socio-cultural role of texts Cuthbert 82, 98, 99 Danes xv, 10, 14, 41, 45, 50, 79, 101, 112–17, 151, 155, 157, 158, 168, 200, 202, 204, 211, 214, 218–25 Death of Edgar 48, 50 Deor xiii, 4, 27, 33, 36, 38, 40, 42–47 Dhuoda 173, 174 Dream-screen 14, 24, 37, 119, 130, 146, 166, 174, 177–78, 180, 199, 223

East Anglia 90, 118, 213 Einhard 132, 155 Elene 8 Emotional response 164, 178, 209, 213, 217–19, 223–25; see also Anxiety; Shame Emotions, history of 3, 4, 8, 13–18, 24, 26, 30, 32, 34–39, 40, 50, 92, 94, 119, 127–28, 137, 164, 177, 180, 198–99, 201–206, 212; see also Alienation; Anxiety; Nostalgia; Shame Emplotment see Narrativization Encomium Emmae 34, 92 English nation see Angelcynn Entertainment, xiv, 24, 31, 153, 177, 180, 181, 199 Ermanaric 148, 149, 168, 169 Ermanrich 133 Ermold the Black 157 Evangelienbuch see Otfrid of Weissenburg Exeter Book Riddles see Riddles Fatherhood, 4, 45, 135, 168, 171–77 Fictionality 3, 9, 19, 23, 26, 27, 47, 72, 77, 79–81, 91–93, 101, 102 Finnsburg Fragment (The Fight at Finnsburg) 4, 12, 13, 33, 37, 38, 45, 105 Franconian 137, 142, 143, 147–52, 159, 160, 163 Frankish: identity 145, 151, 152, 155–63, 167, 170, 171; language see Franconian Freculph of Lisieux 150–55, 157, 167 Fredegar 75, 146 Geats 24, 41, 45, 92, 199, 203, 204, 219, 220 Gender 29, 37, 153, 171, 179, 180, 202, 203, 207–214, 219–25; see also Masculinity Genealogy, 10, 15, 20, 49, 73, 79, 80–82, 90, 91, 96, 118, 161, 201 Genesis OE poems (A and B) 2, 34, 181 Germania 14, 18, 19–23, 40, 49, 146, 149–51, 153, 158

Index  237 Germanic : concept of xii, 2–5, 9, 16–22, 130, 152; critiques of the concept xii, xiii, 2–6, 13–22, 26, 35; culture 3, 4, 16, 19, 23, 130, 148, 152–154; identity 4, 16, 19, 20, 148, 152–54; languages, 19, 23, 97, 129, 131, 135–43, 147–54, 160, 163, 166, 169, 170, 179 Giedd 8, 24, 25, 46, 78, 105, 108–110 Gothicism 16, 152, 163 Goths 16, 148, 149, 152–58, 163, 164, 167, 168 Gottschalk 155, 157 Gregory of Tours 2, 4, 9, 13, 39, 75, 167, 168 Grendelkin 30, 31, 79, 103, 105, 113, 116, 199, 204, 207–214, 220–25 Guthlac 201, 216–18, 221–23 Guthlac A see Guthlac Heliand xii, xiv, 2, 34, 129, 135–44, 153, 178, 180, 181 Heroic : age, xiii, 4, 9, 10–18, 23–26, 35–37, 40, 42–49, 83, 128, 145, 146, 151, 153, 201, 218; concept of xii–xiv, 3–6, 9, 13, 128–30, 179, 180; critiques of the concept xii–xiv, 2–6, 9, 13–24, 26, 49, 112, 113, 130, 141, 164–66, 173–77, 203, 210–13; ethos, 3, 4, 13, 18, 23, 24, 49, 50, 79, 105, 112, 113, 128–30, 135, 138–41, 154, 159, 160, 163–81, 201–203, 207–214, 217, 219 Hildebrandslied 4, 8, 33, 45, 128, 129, 131, 135, 143, 144, 146, 150, 154, 166–70, 174–78, 181 Historical conception, 2–5, 10–12, 16, 17, 26–30, 34–39, 41, 45, 51, 64, 65, 70, 71, 78–80, 111, 128 Homer 1, 6, 9, 23, 134 Honour culture, xiv, 3, 4, 13, 19, 23, 24, 36, 45, 46, 105, 113, 165, 171–76, 201, 203, 206, 208, 211, 212, 214, 221 Hrabanus Maurus 157 Hrothgar 25, 78, 80, 105, 112–15 Humour see Laughter Huns 154, 164–68, 175 Hygelac 9, 31, 91, 105, 115, 116, 199, 220

Ideology 15, 20, 23, 40, 47, 49, 102, 111, 118, 128, 180, 212; see also Nationalism; Nostalgia; Subversiveness Interpretive closure see Ambiguity Irish 20, 24, 31, 65, 71, 72, 140, 147, 159 Isidore of Seville 5, 6, 13, 32, 70, 75, 76, 81, 97 Jordanes 11, 150–54, 167 Judith (OE Biblical poem) xii, 2, 31, 181 Kent 90 Kinship 45, 46, 50, 76, 112, 113, 135, 145, 152–54, 163, 170–73, 174, 179, 203, 214, 221 Latin poetry 2, 33, 129, 132, 143, 168, 180 Laughter, 1, 16, 24, 40, 110, 113, 138–39, 153, 164–68, 170–74, 181, 198, 206–209, 214 Literacy 1, 6, 24, 32–36, 42, 47, 77, 80–83, 95, 98–102, 130–37, 142–49, 154, 178 Louis the German 130, 131, 145, 147, 148, 150, 155–61, 168–71, 175, 176, 179, 181 Louis the Pious 129–32, 145, 148, 157, 161, 171–75 Ludic space see Dream screen Ludwigslied xiii, 33, 48, 131, 142, 163 Manuscripts: Heidelberg Universitätsbibliothek, Codex Palatinus Latinus 52 163; London, British Library, Cotton Tiberius B V/1 107, 225; London, British Library, Cotton Vitellius A XV (the Nowell Codex, the Beowulf MS) 30, 31, 200, 223–25; Rome, National Library, Codex Vittorio Emanuele 1631 (Codex Aesinas) 149 Marginality 7, 17, 37, 38, 40, 130–36, 139–43, 162, 181, 211, 213

238 Index Masculinity, 3, 15, 18, 23, 24, 128–30, 135, 143, 145, 158–60, 163, 164, 168–81, 201–203, 207–214, 217–25 Maxims I (Exeter Book Maxims) 108–110, 114 Maxims II 107 Memory 6, 12, 20, 22, 41, 48, 49, 66–69, 74–83, 101, 103–105, 109, 167, 175, 201, 219 Mercia 21, 90, 200, 201, 217–19, 223 Metre, xii, 3, 5, 6, 13, 18, 22, 25, 27, 34, 39, 42, 46, 48, 50, 77, 102–104, 135, 139, 141–45, 150, 163, 167, 200, 220 Migration Age see Heroic age Monstrosity xii, 3, 9, 24, 30, 31, 91, 112, 113, 203, 206–213, 217–221, 223–25 Mourning 12–15, 37, 39, 46, 50, 80, 203 Muspilli 129, 131, 135, 142, 143, 146, 181 Myth 4, 10, 13–15, 19–22, 23, 31–37, 43, 46, 49, 50, 90, 94, 114, 146, 149–153, 164, 167, 200, 203, 218 Narrative authority see Auctoritas Narrativization xii, xiv, 13, 15, 26, 29, 30, 33–36, 40–50, 83, 103, 106, 111, 118, 200, 201, 204, 205, 209, 216 Nationalism: medieval 2, 11, 15, 16, 33, 47, 118, 147–160, 200, 218; see also Angelcynn; modern, 6, 13, 18–22, 35, 147 Neo-bardic verse, xiii, 33, 36–38, 45, 47–50, 111, 118, 127 Nithard 173 Northumbria 71, 91, 92, 99 Nostalgia 10–15, 27, 33, 35, 46, 49, 65, 71, 79, 80, 106, 163 Notker the Stammerer 145, 170 O Old Norse 15, 24, 42, 65, 73, 78, 128, 141, 151 Orality 9, 18, 23, 34–36, 46–48, 76–81, 94, 97, 100–102, 109–111, 129, 134–38, 143–46, 178

Oral poet 1–3, 8, 33–38, 40, 45–47, 76–82, 102, 116, 199 Origin story see Myth Orosius 1, 2, 7, 11, 70–74, 77 Otfrid of Weissenburg 34, 129, 139–43, 144, 147, 151, 159–63, 170, 181 Paris Conversations 138–39 Periphery see Marginality Philosophy of history see Historical conception Polyphony 12, 13, 15, 33, 41–45, 49, 112, 115, 116, 204, 210–13 Propaganda xii–xiv, 33, 37, 38, 45, 48–50, 128, 130, 146, 153, 154, 157, 166, 180 Race 20–22, 147, 156, 201, 202, 210, 217–25 Rag-picker xiii, xiv, 3, 13, 33–50, 65, 79, 111, 118, 164, 177–81, 198–201 Reader response 130, 141, 200, 205–208, 213, 218 Reception 6, 13, 15, 27, 31, 35, 38, 39, 44, 64, 91, 167, 200–206, 223, 224 Regino of Prüm 147 Rhizome 44, 65, 83, 178 Riddles, 15, 42, 73, 94, 95, 108, 109, 116, 118, 208, 214–17 Romance vernacular 67, 131, 133, 136–38, 142, 147, 148, 154, 166 Romanticism 6, 8, 18, 20, 23, 25, 35 Rudolf of Fulda see Translatio Sancti Alexandri Ruin 80 Saxon (Continental) 149, 151, 155–58, 161 Scandinavia 3, 16, 17, 31, 45, 102, 128, 135, 150–55, 157, 163, 199 Scandza 152, 153 Schema 2, 8, 25, 36, 40, 70, 76–83, 98, 118, 201; of emotion 66, 80, 106; of event 77, 79, 80, 82; of role 76, 77, 79, 82 Scop see Oral poet Scripture see Bible Seafarer 25, 44, 106

Index  239 Self-referentiality, xiii, 24, 38, 40–42, 45, 48, 111, 115, 199, 204 Sense of the past see Historical conception Shame 24, 50, 172, 175, 201, 209–215 Slaves 215–17 Smaragdus of Saint-Mihiel 155 Social class 156; see also Anglo-Saxon elites; Carolingian elites; Slaves Socio-cultural context, 1, 3, 7, 8, 12–23, 30–34, 39, 42, 67, 70, 71, 81, 82, 91–96, 103, 111, 117, 127–30, 134–36, 138, 141, 145, 157, 164, 166, 169–73, 177–81, 201, 204–206, 212–25 Socio-cultural role of texts, xiii, 1–3, 11–18, 26, 30–40, 43–47, 50, 91, 110, 117, 127–30, 135, 150–53, 178–81, 198, 200, 204, 205, 209, 210, 213, 214, 223–25 Sociolinguistics 134–36, 142, 148, 150, 164 Solomon and Saturn 109, 110, 114 Story-world 19, 27, 33, 34, 37–39, 101, 105, 106, 117–19, 146, 178, 200 St. Brice’s Day massacre xv, 224, 225 Subaltern 112, 114, 138, 139, 211, 212 Subversiveness, 13, 24, 26, 30, 33, 38, 40–42, 49, 116, 139, 141, 154, 166, 168, 173, 180, 181, 203, 209, 212 Swedes 31, 41, 45, 91, 203, 218 Tacitus see Germania Teuton 16 48, 146–58, 161–69, 175, 179; see also Theodisci Textuality, 16, 22–28, 31, 35, 39, 40, 43, 46–51, 64, 65, 77, 101, 102, 106, 111, 112, 144–46 Theoderic the Great 45, 154, 167, 168, 174 Theodisci 138, 148–52, 155–58, 163, 168; see also Teuton Traditio 35, 74, 78–81, 99–107, 111, 116–19

Traditionality see Conservatism Translatio Sancti Alexandri 149 150, 156 TribalCrit 222 Trojans 150–152, 163, 164 Truthfulness 1, 2, 10, 26, 27, 31, 32, 38, 40, 64–67, 71, 74–83, 90–119, 199 Ulfila 155 Unferth 96, 112–17, 211 Vainglory 27, 108, 109 Vernacular theory xiii, 1, 6, 8, 13, 30–37, 64, 65, 83, 90–93, 111, 117, 127 Viking 13, 49, 91, 151, 163, 201, 223 Violence, xv, 23, 24, 46, 112, 128, 135, 141, 170–72, 174, 176, 209–212, 215, 219, 224, 225 Virgil 33, 129, 133, 149, 168 Visigoths 167; see also Goths Vita Guthlaci see Guthlac Walahfrid Strabo 154, 155, 157 Waldere 45, 153, 167 Waltharius 2–4, 13, 16, 24, 31, 33, 38, 40, 42, 45, 47, 65, 128–31, 134, 143–46, 153, 163–81 Wanderer 80 Wealhas xiv, 147, 202, 214–23 Web of meaning see Socio-cultural context Welsh see Wealhas Wessex 10, 21, 48, 49, 72, 82, 200, 218, 219 Widsith 3, 14, 15, 21, 25, 33–44, 46–48, 78, 82, 200 Wiglaf 24, 41, 203, 212 Woden 90–92, 118 Wonders of the East 31, 105, 221, 224 Word-hoard see Archive Wrecca 112, 113, 210–12