Journal of Medieval Military History. Volume XIII [13] 1783270578, 9781783270576

Citation preview

spine 24mm

The Journal’s hallmark is broad chronological, geographic and thematic coverage of the subject, bearing witness to ‘the range and richness of scholarship on medieval warfare, military institutions, and cultures of conflict that characterises the field’.  HISTORY CONTENTS DAVID BACHRACH THOMAS K. HEEBØLL-HOLM

Feudalism, Romanticism, and Source Criticism: Writing the Military History of Salian Germany When the Lamb Attacked the Lion: A Danish Attack on England in 1138?

MICHAEL S. FULTON

Development of Prefabricated Artillery during the Crusades

RABEI G. KHAMISY

Some Notes on Ayyu ¯bid and Mamluk Military Terms

OREN FALK

Helgastaðir, 1220: A Battle of No Significance?

NICOLÁS AGRAIT

Por la guarda de la mar: Castile and the Struggle for the Sea in the Thirteenth and Fourteenth Centuries

MICHAEL LIVINGSTON

The Battle of Hyddgen, 1401: Owain Glyndw ˆ r’s Victory Reconsidered

DAN SPENCER

The Provision of Artillery for the 1428 Expedition to France

DEVIN FIELDS

1471: The Year of Three Battles and English Gunpowder Artillery

L. J. ANDREW VILLALON

TONIO ANDRADE

“Cardinal Sins” and “Cardinal Virtues” of “El Tercer Rey,” Pedro González de Mendoza: The Many Faces of a Warrior Churchman in Late Medieval Europe

Cover: Coll. Bibliothèques municipales de Chambéry, MS13, fo. 105r (circa 1170–1190). The Italian artist who illuminated this copy of Gratian’s Decretum beautifully captured the deadly seriousness of medieval warfare – for townsmen as well as knights. Image used by generous permission of the Bibliothèques municipales de Chambéry. an imprint of Boydell & Brewer Ltd PO Box 9, Woodbridge IP12 3DF (GB) and 668 Mt Hope Ave, Rochester NY 14620–2731 (US) www.boydellandbrewer.com

FRANCE, DEVRIES, ROGERS (editors)

Late Medieval Divergences: Comparative Perspectives on Early Gunpowder Warfare in Europe and China

JOURNAL OF MEDIEVAL MILITARY HISTORY VIII JOURNAL OF MEDIEVAL MILITARY HISTORY

XIII

J ournal of

MEDIEVAL MILITARY HISTORY

·  X III · Edited by JOHN FRANCE, KELLY DEVRIES and CLIFFORD J. ROGERS

JOURNAL OF

Medieval Military History Volume XIII

JOURNAL OF MEDIEVAL MILITARY HISTORY

Editors Clifford J. Rogers Kelly DeVries John France ISSN 1477–545X

The Journal, an annual publication of De re militari: The Society for Medieval Military History, covers medieval warfare in the broadest possible terms, both chronologically and thematically. It aims to encompass topics ranging from traditional studies of the strategic and tactical conduct of war, to explorations of the martial aspects of chivalric culture and mentalité, examinations of the development of military technology, and prosopographical treatments of the composition of medieval armies. Editions of previously unpublished documents of significance to the field are included. The Journal also seeks to foster debate on key disputed aspects of medieval military history. The editors welcome submissions to the Journal, which should be formatted in accordance with the style-sheet provided on De re militari’s website (www.­ deremilitari.org), and sent electronically to the editor specified there.

JOURNAL OF

Medieval Military History Volume XIII Edited by John France CLIFFORD J. ROGERS KELLY DeVRIES

THE BOYDELL PRESS

© Contributors 2015 All Rights Reserved. Except as permitted under current legislation no part of this work may be photocopied, stored in a retrieval system, published, performed in public, adapted, broadcast, transmitted, recorded or reproduced in any form or by any means, without the prior permission of the copyright owner First published 2015 The Boydell Press, Woodbridge ISBN 978–1–78327–057–6 The Boydell Press is an imprint of Boydell & Brewer Ltd PO Box 9, Woodbridge, Suffolk IP12 3DF, UK and of Boydell & Brewer Inc. 668 Mt Hope Avenue, Rochester, NY 14620–2731, USA website: www.boydellandbrewer.com

A CIP catalogue record for this book is available from the British Library The publisher has no responsibility for the continued existence or accuracy of URLs for external or third-party internet websites referred to in this book, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate This publication is printed on acid-free paper

Contents

List of Illustrations and Table

vi

1. Feudalism, Romanticism, and Source Criticism: Writing the Military History of Salian Germany David Bachrach 2. When the Lamb Attacked the Lion: A Danish Attack on England in 1138? Thomas K. Heebøll-Holm 3. Development of Prefabricated Artillery during the Crusades Michael S. Fulton 4. Some Notes on Ayyūbid and Mamluk Military Terms Rabei G. Khamisy 5. Helgastaðir, 1220: A Battle of No Significance? Oren Falk 6. Por la guarda de la mar: Castile and the Struggle for the Sea in the Thirteenth and Fourteenth Centuries Nicolás Agrait 7 The Battle of Hyddgen, 1401: Owain Glyndŵr’s Victory Reconsidered Michael Livingston 8. The Provision of Artillery for the 1428 Expedition to France Dan Spencer 9. 1471: The Year of Three Battles and English Gunpowder Artillery Devin Fields 10. “Cardinal Sins” and “Cardinal Virtues” of “El Tercer Rey,” Pedro González de Mendoza: The Many Faces of a Warrior Churchman in Late Medieval Europe L. J. Andrew Villalon 11. Late Medieval Divergences: Comparative Perspectives on Early Gunpowder Warfare in Europe and China Tonio Andrade

1

List of Contributors

27 51 73 93 139 167 179 193 213

247

277

Illustrations and Table

When the Lamb Attacked the Lion: A Danish Attack on England in 1138? Figure 1. Genealogy of the Kings of Denmark

49

Helgastaðir, 1220: A Battle of No Significance? Figure 1. Map of Iceland. Adapted by the author from outline map 94 available at http://geography.about.com/library/blank/blxiceland.htm. Figure 2. Map of Reykjadalr, drafted in 1931. Excerpted from Geodætisk 103 institut, Blað 72: Húsavík [map], 1:100,000 (Uppdráttur Íslands; Copenhagen: Geodetic Institute, 1945). Figure 3. Cemetery (formerly church) at Helgastaðir. Photo by Dov 108 Levitte, 25 August 2013. Figure 4. View to the south from Helgastaðir. Photo by Dov Levitte, 109 25 August 2013. Figure 5. View to the west from Helgastaðir. Photo by Dov Levitte, 114 25 August 2013. Figure 6. Map of medieval Helgastaðir and its immediate environs. 115 Adapted by the author (not to scale) from the map in Figure 2. Figure 7. Western façade of Trondheim Cathedral. Photo by Erik A. 124 Drabløs, 23 August 2006 (from Wikimedia Commons). The Battle of Hyddgen, 1401: Owain Glyndŵr’s Victory Reconsidered Figure 1. The site of the battle of Hyddgen 1471: The Year of Three Battles and English Gunpowder Artillery Figure 1. Diagram of the battle of Barnet based on the work of Burne and Hammond

176 198

Late Medieval Divergences: Comparative Perspectives on Early Gunpowder Warfare in Europe and China Table 1. Ratio of length to bore, Chinese guns, 1288–1423 271 Figure 1. Trends in the development of Chinese guns, 1288–1423 273

1 Feudalism, Romanticism, and Source Criticism: Writing the Military History of Salian Germany1 David Bachrach

The military history of the German kingdom in the eleventh century remains to be written. The most recent monographic treatment that considers the Salian kings at war, Bruno Scherff’s 1985 dissertation “Studien zum Heer der Ottonen und der ersten Salier (919–1056),” ends with the death of King Henry III (1039–56) and is notable for the topics that it explicitly avoids, including military equipment, logistics, numbers, military levies, fortifications, and above all, sieges.2 There are no monographic studies of the conquest of the Burgundian kingdom by King Conrad II of Germany (1024–39) in his two-year campaign of 1033– 34.3 Since the end of the nineteenth century, scholars have assiduously avoided investigating the wars of Conrad II, Henry III, and Henry IV (1056–1106) in Poland and Hungary.4 Even the civil wars within Germany during the final quarter of the eleventh century and the first decade of the twelfth, which have received considerable attention from a political perspective, have inspired only limited commentary regarding military matters.5 At an even more basic level, 1

2 3

4

5

This paper was originally delivered as the De re militari Lecture at the medieval studies conference at Western Michigan University in May 2014. I thank all of the participants for their helpful questions and insights. Bruno Scherff, “Studien zum Heer der Ottonen und der ersten Salier (919–1056)” (Ph.D. dissertation, University of Bonn, 1985), pp. 1–5. The very dated dissertation by Otto Blümcke, “Burgund unter Rudolf III. und der Heimfall der burgundischen Krone an Kaiser Konrad III” (Ph.D. dissertation, University of Greifswald, 1869) focuses on the transition in power between the last Burgundian ruler and the Salian Conrad II, but does not do so from a military perspective. Julius Reinhard Dieterich, Die Polenkriege Konrads II. und der Friede von Merseburg (Giessen, 1895); and Emil Friedrich Kuemmel, Die zwei letzten Heereszüge Kaiser Heinrichs III. nach Ungarn (1051 und 1052) mit Rüchsichtsnahme auf die bairisch-kärtnische Empörung (Strassnitz, 1879). A general treatment of German–Hungarian relations can be found in the studies by Friedrich Schuster, Ungarns Beziehungen zu Deutschland von 1056–1108 (Hermannstadt, 1899); and Otto Rademacher, Ungarn und das deutsche Reich unter Heinrich IV. (Merseburg, 1885). Two notable exceptions to the otherwise spare historiographical tradition are the articles by Leopold Auer, “Die Schlacht bei Mailberg am 12. Mai 1082,” Militärhistorische Schriftenreihe 31 (1975), 1–31; and John Gillingham, “An Age of Expansion, c. 1020–1204,” in Medieval Warfare: A History, ed. Maurice Keen (Oxford, 1999), pp. 59–88. Several of the major battles of the Saxon wars also were outlined by Hans Delbrück, Geschichte der Kriegskunst in Rahmen

2

David Bachrach

the enormous volume of source materials pertaining to warfare in the Salian century, including narrative texts, letters, charters, and archaeological evidence, has not received the kind of detailed critical analysis that has become the norm in Anglo-Saxon, Carolingian, and increasingly Ottonian military history.6 The lack of scholarly attention to the military history of the German kingdom generally, and to the Salian period, in particular, is due to several factors. First, until very recently non-German-speaking scholars, including military historians, generally avoided the history of Germany altogether.7 Second, German-speaking historians largely have avoided topics in military history, especially since the end of the Second World War.8 Finally, the one consistent exception to this aversion to military matters, namely the focus on social and economic questions regarding the relationship among military service, nobility, and feudalism, has actually served to hinder a thorough investigation of the nature and conduct of war, and the concomitant examination of its institutional, material, and human elements. Even before the 1940s, many topics that are currently understood as the province of military history were largely ignored in the German scholarly tradition. These include, but are not limited to, logistics, siege warfare, military technology, military organization, recruitment, military institutions such as local and expeditionary levies, pay, morale, education, and military training. Instead, scholars devoted their energies to explaining the supposed transition from a

6 7

8

der politischen Geschichte, vol. 3, third edition (Berlin, 1923, originally published 1907), trans. by Walter J. Renfroe as History of the Art of War, Volume III: Medieval Warfare (Lincoln, NE, 1982, repr. 1990), pp. 131–45. David S. Bachrach, “Milites and Warfare in Pre-Crusade Germany,” forthcoming in War in History, is an attempt to redress this imbalance, particularly in regard to narrative texts. Fortunately, there has been a recent blossoming of English-language scholarship on the German kingdom. An early precursor of this tradition was John W. Bernhardt, Itinerant Kingship and Royal Monasteries in Early Medieval Germany c. 936–1075 (Cambridge, 1993). Now also see, for example, Jonathan Lyon, Princely Brothers and Sisters: The Sibling Bond in German Politics, 1100–1250 (Ithaca, NY, 2013); and John Eldevik, Episcopal Power and Ecclesiastical Reform in the German Empire: Tithes, Lordship, and Community, 950–1150 (Cambridge, UK, 2012). From a military perspective, see Charles R. Bowlus, The Battle of Lechfeld and its Aftermath, August 955: The End of the Age of Migrations in the West (Aldershot, 2006). Concerning the lack of scholarship on German military history in the period after the Second World War, see Hans-Hennig Kortüm, “Der Krieg im Mittelalter als Gegenstand der historischen Kultur-Wissenschaften: Einer Annäherung,” in Krieg im Mittelalter, ed. idem (Berlin, 2001), pp. 13–43; and Scherff, Studien zum Heer der Ottonen und der ersten Salier, pp. 2–4. One notable exception has been the Austrian scholar Leopold Auer, whose numerous publications on military matters include “Der Kriegsdienst des Klerus unter den sächsischen Kaisern,” Mitteilungen des Instituts für Österreichische Geschichtsforschung part 1, 79 (1971), 316–407; and part 2 in ibid., 80 (1972), 48–70; “Zum Kriegswesen unter den früheren Babenbergern,” Jahrbuch für Landeskunde von Niederösterreich 42 (1976), 9–25; “Formen des Krieges im abendländischen Mittelalter,” in Formen des Krieges: Vom Mittelalter zum “Low-IntensityConflict,” ed. Manfried Rauchensteiner and Erwin A. Schmid (Graz, 1991), pp. 17–43; and “Mittelalterliches Kriegswesen im Zeichen des Rittertums,” in Krieg im mittelalterlichen Abendland, ed. Andreas Obenau and Christoph Kaindel (Vienna, 2010), pp. 65–79.



Military History of Salian Germany

3

putative “nation in arms” among the early Germanic peoples to small retinues of “feudal knights” who ostensibly dominated both warfare and politics through their monopolistic mastery of equestrian combat. A brief overview of this historiographical tradition will help to illuminate the current (limited) state of understanding of warfare in Germany in the pre-crusade period. From Levies to Feudal Knights In his classic study, first published in 1887, Heinrich Brunner made the case for Charles Martel’s rapid transformation of Frankish military organization from large armies of men on foot to much smaller armies of mounted warriors.9 Brunner argued that Charles Martel, having defeated an army of mounted Muslims with his infantry phalanx, now sought to develop a cavalry army that could face the Muslims in battle. The implicit paradox in this argument was never fully explained by Brunner. In order to achieve this new army, Charles is depicted as having “borrowed” large numbers of estates from the church, and granting these as beneficia to his loyal supporters so that they could purchase and maintain horses that were suitable for mounted combat. Brunner averred that because these mounted troops were so much more militarily effective than the foot soldiers whom they replaced Charles Martel simply stopped mobilizing the latter and relied on his new cavalry army. The ultimate consequence of this change in military organization of the Frankish kingdom was that only those men who could equip themselves and their dependents as mounted warriors remained militarily viable. The bulk of the population, which now was excluded from military service, declined in social, economic, and political status. The warrior aristocracy and their beneficed retainers thus took their place alongside the king as the major figures in the new “feudal” order. Military historians in the late nineteenth and early twentieth centuries generally did not accept Brunner’s model of a rapid transition of Frankish military organization towards an army that was composed largely of mounted warriors – not least because Carolingian military legislation from the ninth century clearly indicated that foot soldiers continued to be mobilized for combat operations.10 9

10

The information in the following paragraph is drawn from Heinrich Brunner, “Der Reiterdienst und die Anfänge des Lehnwesens,” Zeitschrift der Savigny-Stiftung für Rechtsgeschichte, Germanistische Abtheilung 8 (1887), 1–38, reprinted in Brunner, Forschungen zur Geschichte des deutschen und französischen Rechts (Stuttgart, 1894), pp. 39–74. In this context, Delbrück, History of the Art of War, 3:16–18, took the view that the transition was very rapid, while numerous other scholars argued for a much more gradual approach. In his review essay of the third volume of Delbrück’s massive study on the history of war, in which Delbrück argued in favor of Brunner’s rapid transition to an army of mounted fighting men, W. Erben, “Zur Geschichte des karolingischen Kriegswesens,” Historische Zeitschrift 101 (1908), 321–36, drew attention to a series of errors that Delbrück had made, and specifically noted Delbrück’s failure to account for the capitulary evidence that discussed the continued mobilization of foot soldiers for campaign duty. Many other scholars called into question the rapid transition to a mounted army on the basis of references to foot soldiers in narrative

4

David Bachrach

However, Brunner’s argument that mounted military service provides the key to understanding both the establishment of a warrior nobility and the development of feudalism did win widespread approval from scholars in the early twentieth century. Thus, for example, in his study of the constitution of the state and the army in early medieval Europe, published in 1906, Otto Hintze started from the premise that there was a transition in Frankish military organization from large armies of foot soldiers to smaller armies of mounted troops. Hintze argued that this transformation was accompanied by a professionalization of the army, with the concomitant requirement that mounted fighting men should have sufficient wealth not only to equip themselves with horses and arms, but also the leisure time that was necessary to train to fight on horseback. The difference in wealth combined with the difference in status as warriors resulted, according to Hintze, in the development of a knightly class (Ritterstand). These knights were socially, as well as economically, superior to the bulk of the population that did not have sufficient wealth to participate in mounted combat.11 The distinction between those who fight and those who farm was given an ostensible legal foundation by Hans Fehr in his 1914 study.12 Fehr compared the discussion of fighting men in Carolingian capitularies of the ninth century with the texts of royal commands for territorial peace (Landfrieden) issued by Salian and Staufen kings from the eleventh through the thirteenth centuries. Fehr observed that although the latter documents clearly permit all men to engage in self-defense, the Landfrieden do not make provisions for Bauern to participate in expeditionary campaigns. He concluded on this basis that farmers no longer had the right to bear arms (Waffenrecht) and consequently suffered a social decline as a class.13 It is noteworthy that Fehr did not utilize any narrative sources in his study. Nor did he find an adequate solution to the problem that as late as 1179 the Landfrieden issued by Frederick Barbarossa banned farmers from carrying swords while in their home villages, but did not ban them from carrying swords while they went on journeys.14 In the interwar period, many scholars adopted the theme that men of the lower social orders, variously identified as farmers or simply as free men (Freie), were demilitarized, and that warfare came to be dominated by a knightly class of

11

12 13 14

sources. See, for example, Erich Sander, “Die Heeresorganisation Heinrichs I.,” Historisches Jahrbuch 59 (1939), 1–26; Hans von Mangoldt-Gaudlitz, Die Reiterei in den germanischen und fränkischen Heeren bis zum Ausgang der deutschen Karolinger (Berlin, 1922); Franz Beyerle, “Zur Wehrverfassung des Hochmittelalters,” in Festschrift Ernst Mayer zum 70. Geburtstag (Weimar, 1932); and Paul Schmitthenner, “Lehnskriegswesen und Söldnertum im abendländischen Imperium des Mittelalters,” Historische Zeitschrift 150 (1934), 228–67. Otto Hintze, Staatsverfassung und Heeresverfassung (Dresden, 1906); and reprinted in idem Staat und Verfassung. Gesammelte Abhandlungen zur allgemeinen Verfassungsgeschichte, ed. Gerhard Oestreich, 3rd ed. (Göttingen, 1970), pp. 52–83. Hans Fehr, “Das Waffenrecht der Bauern im Mittelalter,” Zeitschrift der Savigny-Stiftung für Rechtsgeschichte. Germanistische Abteilung 35 (1914), 111–211. Ibid., pp. 158–61. Ibid., p. 139.



Military History of Salian Germany

5

mounted warriors. Franz Beyerle, for example, argued: “from the later Carolingian period, the army increasingly was dominated by vassalic contingents.”15 In support of this argument, Beyerle drew attention to Regino of Prüm’s discussion of the defeat of the tenants of his monastery at the hands of the Vikings in the late ninth century, which Beyerle used as evidence of the poor fighting skills of local levies that resulted in their increasing exclusion from military affairs.16 In this context, Beyerle did not draw attention to Regino’s specific observation that the men in question were untrained, i.e. disciplina militari nudatum.17 Eugen von Frauenholz also depicted the ninth century as a period of transition between armies of foot soldiers and smaller armies of mounted warriors. Like Hintze, von Frauenholz initially depicted this development as bringing about a concomitant professionalization of the army. However, this professional force quickly developed, in the view of von Frauenholz, into a new hereditary social class (Geburtsstand) with rules governing not only military life, but all aspects of newly developing knighthood (Ritterschaft).18 In the decades after the Second World War, scholars continued to adhere to the intertwined argument that there was a transition from armies of men on foot to armies composed predominantly of mounted warriors, and that this transition played a crucial role in the development of both a warrior aristocracy and feudalism. In his study of the participation of German ecclesiastical magnates in the wars of the Ottonian kings, published in 1971–72, the Austrian scholar Leopold Auer asserted: “Undoubtedly, vassalic contingents gained considerably in importance because of the efforts of the Carolingians, and the number of free men in the army, who had been quite numerous up to this point, now shrank.”19 A few years later, Auer returned to this idea in a study focused on the military organization of the early Babenberger in Austria, and argued with regard to vassalic contingents that: “we are dealing here with a fundamental issue of medieval military history, namely the central question of the relationship between war and social structure, or put another way, the relationship between status and the obligation to undertake military service.”20 He added that: “heavily armored mounted men increasingly comprised the core of the army from the time of 15 16 17 18 19

20

Franz Beyerle, “Zur Wehrverfassung des Hochmittelalters,” in Festschrift Ernst Mayer zum 70. Geburtstag (Weimar, 1932), pp. 31–91, here 31. Ibid., p. 40. Regino of Prüm, Regionis abbatis Prumensis chronicon cum continuatione Treverensi, ed. F. Kurze, MGH SRG (Hanover, 1890), p. 118. Eugen von Frauenholz, Das Heerwesen der germanischen Frühzeit, des Frankenreiches und des ritterlichen Zeitalters (Munich, 1935), p. 60. Leopold Auer, “Der Kriegsdienst des Klerus unter den sächsischen Kaisern,” Mitteilungen des Instituts für Österreichische Geschichtsforschung part 1, 79 (1971), 316–407; and part 2 in ibid., 80 (1972), 48–70. Auer, “Kriegsdienst,” part 1, 320; and idem, part 2, 50, “Zweifellos hat aber das vasallitische Aufgebot durch die Maßnahmen der Karolinger … stark an Bedeutung gewonnen und die Zahl der Freien im Heer, die in dieser Zeit noch recht zahlreich vertreten waren, eingeschränkt.” Leopold Auer, “Zum Kriegswesen unter den früheren Babenbergern,” Jahrbuch für Landeskunde von Niederösterreich 42 (1976), 9–25, here 16.

6

David Bachrach

Pippin III onwards, while foot soldiers retreated in importance. The material demands that resulted from this situation (above all the possession of a horse and a byrnie) were too high for the great mass of the free men, particularly in light of the increasing number of wars. The result was that participation in offensive campaigns, that is expeditionary duty, increasingly was limited to the wealthier strata of the population.”21 However, Auer did accept that members of the lower social orders continued to be mobilized for territorial defense (Landfolge) into the eleventh century in periods of emergency.22 By contrast, Auer claimed, those members of the lower social orders found on campaign were there merely as servants or attendants in the baggage train.23 Auer’s views regarding the changes in military organization under the Carolingians were echoed by Josef Fleckenstein in his study on the intertwined development of the nobility and the nature of military organization in the Frankish empire. Fleckenstein focused his attention on Carolingian capitularies, and identified a three-stage process in which the position of the free fighting man was weakened relative to the vassals of the great magnates. The first phase, Fleckenstein argued, came with Charlemagne’s 808 capitulary that called for a mobilization of the entire army, with men coming to the host either with their senior or with the count. Fleckenstein interpreted this to mean that the army was now divided into two elements. The first consisted of vassals led by their feudal lords, and the second consisted of non-vassals who were led to war by the count in whose pagus they resided. Fleckenstein argued that the next stage of development was marked by the 847 capitulary of Meersen that summoned all men to come to the host with their seniores, without making any mention of men coming to the host with their counts. Fleckenstein interpreted this to mean that non-vassalic forces no longer participated on campaign. The third stage, Fleckenstein argued, came with the increasing number of nobles who appear as vassals and the concomitant numerical domination of vassalic contingents by men of high social status who held benefices from their superior lords.24 The arguments regarding the transformation of Frankish military organization and the concomitant domination of warfare by vassalic contingents of noble mounted warriors were further developed in the context of Ottonian history by Hagen Keller and Karl Leyser. The former observed in his study on the foundations of Ottonian royal rule that: “The transition from the Carolingian period, namely the accentuation of the personal character of the structure of lordship, had a particularly clear effect in the change in military organization. The Ottonians abandoned the territorially-based mobilization of the free – aside from

21 22 23 24

Ibid. Ibid., p. 17. Ibid., p. 21. Josef Fleckenstein, “Adel und Kriegertum und ihre Wandlung im Karolingereich,” Settimane di Studio del Centro Italiano di Studi sull’alto Medioevo 27 (1987), 2 vols., 1:67–94, here 89–92.



Military History of Salian Germany

7

territorial defense – and relied entirely on the vassals of the bishops and secular magnates.”25 In a series of essays published from the 1960s to the 1980s, Karl Leyser also developed the theme that warfare in the Ottonian period was dominated by nobles, and their noble vassals, who were distinguished by their mastery of equestrian combat.26 Unlike Keller, however, Leyser looked for this change in the tenth rather than the ninth century. He argued that the Saxons lagged behind their Frankish neighbors in military development, and so the first Saxon king, Henry I (919–936), was compelled to bring his people into modernity by developing a force of heavily armed mounted warriors in order to deal with the Magyar threat. Although he did not cite Heinrich Brunner’s study on mounted military service and the origins of feudalism, Leyser’s argument paralleled Brunner’s model concerning Charles Martel, namely that the creation of an army of mounted vassals led to the creation of a knightly class.27 The concomitant effect on military organization was to eliminate the need for foot soldiers, other than in support roles, such as maintaining the wagon train.28 Leyser was willing to concede that, “The ignobile vulgus was by no means yet without weapons.” However, he added that: “the honour of arms had become the hallmark and was already the hereditary possession primarily of nobles.”29 The views of scholars such as Fleckenstein and Leyser have had an enormous influence on subsequent scholarship. With just a few notable exceptions, over the past forty years the transition to armies consisting of vassalic contingents of mounted knights with the concomitant disappearance of Carolingian-style expeditionary levies has been taken for granted by specialists in the history of Ottonian and Salian Germany.30 I have argued elsewhere that this model is 25

26

27 28 29 30

Hagen Keller, “Grundlagen ottonischer Königsherrschaft,” in Reich und Kirche vor dem Investiturstreit: Vorträge beim wissenschaftlichen Kolloquium aus Anlaß des achtzigsten Geburtstag von Gerd Tellenbach, ed. Karl Schmid (Sigmaringen, 1985), pp. 17–34, here 21, n. 14. See, for example, Karl Leyser, “The Battle at the Lech, 955: A Study in Tenth-Century Warfare,” History 50 (1965), 1–25 and reprinted in K. J. Leyser, Medieval Germany and its Neighbours 900–1250 (London, 1982), pp. 43–67; idem, “Henry I and the Beginnings of the Saxon Empire,” The English Historical Review 83 (1968), 1–32 and reprinted in Leyser, Medieval Germany, pp. 11–42; and idem, “Early Medieval Canon Law and the Beginnings of Knighthood,” in Instutionen, Kultur und Gesellschaft im Mittelalter: Festschrift für Josef Fleckenstein zum 65. Geburtstag (Sigmaringen, 1984), pp. 549–66, repr. in Communications and Power in Medieval Europe: The Carolingian and Ottonian Centuries, ed. Timothy Reuter (London, 1994), pp. 51–71. Leyser, “Henry I,” pp. 40–41. Leyser, “Henry I,” p. 21. Leyser, “Early Medieval Canon Law,” pp. 70–71. Numerous studies can be cited in this context. See, for example, Heinrich Koller, “Das mittelalterliche Stadtmauer als Grundlage städtischen Selbstbewußtseins,” in Stadt und Krieg, ed. Bernhard Kirchgässner and Günter Scholz (Sigmaringen, 1989), 9–25, here 17; Charles R. Bowlus, “The Early Kaiserreich in Recent German Historiography,” Central European History 23.4 (1990), 349–67, here 357; Franz-Reiner Erkens, “Militia und Ritterschaft: Reflexionen über die Entstehung des Rittertums,” Historische Zeitschrift 258 (1994), 623–59, here 629; Otto

8

David Bachrach

ill-suited to describe warfare under the Ottonians.31 My focus in the remainder of this essay will be on ascertaining whether this consensus among scholars is justified with regard to the eleventh century. Was warfare in the Salian period truly dominated by small groups of knightly vassals who fought cavalry battles against other knightly vassals?32 Was it really the case, as one scholar averred, that: “the grotesque situation was in force that the nobility built fortifications that could only be defended on foot, but continued to hold fast to the principle of only fighting on horseback”?33 In sum, is it true that military forces, which might be characterized as non-noble levies, were not deployed for military operations in eleventh-century Germany? In the remainder of this paper I will address two issues that have the potential to illuminate these questions, and the broader military organization of the German kingdom under the Salian dynasty. First, how does the parti pris of the narrative sources from the eleventh century affect their depiction of warfare? Second, what do these same authors have to say about the role played by nonmilites in the conduct of war? Before turning to these questions, however, it is useful to make a few observations about the milites who populate the pages of eleventh-century narrative works. In general, it is these milites who have been identified by many scholars either as feudal knights or as beneficed vassals of lords who dominated warfare through their monopoly on mounted combat.34 However, the identification of milites as either knights or vassals cannot be sustained for eleventh-century Germany. First, the term miles/milites retained its traditional meaning as a

31 32

33 34

Ackermann, “Comites und Milites: Grafen und Krieger im Hochmittelalter,” Werdenberger Jahrbuch 7 (1994), 11–20, here 11; Michael Toch, “The Medieval German City under Siege,” The Medieval City Under Siege, ed. Ivy A. Corfis and Michael Wolf (Woodbridge, 1995), pp. 35–48, here 35–36; and Malte Prietzel, Kriegführung im Mittelalter: Handlungen, Erinnerungen, Bedeutungen (Paderborn, 2006) 21–23, and passim; and idem, Krieg im Mittelalter (Darmstadt, 2006), passim. See, for example, David S. Bachrach, Warfare in Tenth-Century Germany (Woodbridge, 2012). A number of scholars have called into question whether there was a knightly class in Germany during the eleventh century. See the discussion by Johann Johrendt, Milites und Militia im 11. Jahrhundert. Untersuchung zur Frühgeschichte des Rittertums in Frankreich und Deutschland (Nüremberg, 1971); idem, “‘Milites’ und ‘Militia’ im 11. Jahrhundert in Deutschland,” Das Rittertum im Mittelalter, ed. Arno Borst (Darmstadt, 1976), pp. 419–36; and Joachim Bumke, Studien zum Ritterbegriff im 12. und 13. Jahrhundert mit einem Anhang zum Stand der Ritterforschung 1976 (Heidelberg, 1977), and translated into English as The Concept of Knighthood in the Middle Ages (New York, 1982), who made the case that there is no evidence of a knightly class in Germany during the eleventh century, and the term miles cannot properly be translated as knight in this period. For a broader commentary on the problem of scholars using the term inappropriately before 1100, see the discussion by Jean Flori, “Knightly Society,” in The New Cambridge Medieval History: c. 1024–c. 1198, vol. 4, ed. Rosamond McKitterick (Cambridge, 2004), pp. 148–84, particularly 149. Koller, “Das mittelalterliche Stadtmauer,” 17. For a discussion of the historiography treating the meaning of the term miles in pre-crusade Germany and a fuller discussion of the points made in the remainder of this paragraph, see Bachrach, “Milites and Warfare,” forthcoming.



Military History of Salian Germany

9

fighting man, and more particularly as a professional soldier, throughout the eleventh century. Second, the sources for this period provide no basis for concluding that men denoted as milites possessed a legal or social status that differentiated them from other laymen on the basis of their identity as milites. The term designates a professional rather than a legal or social status. Nor is there any basis in the sources for concluding that most or even many milites were men with high social status. Rather the opposite is true. Most milites were men of low social status and could be killed with impunity in sieges and on the battlefield, as illuminated by the fact that contemporary observers felt no need to identify them by name. In addition, it is clear that although some milites did possess beneficia, many did not and instead received wages, i.e. stipendia, for their military service. Finally, although some milites can be shown to have possessed horses, and even to have ridden them in combat, most milites appear in the narrative sources of the eleventh century as garrison troops in fortifications, or in armies besieging fortifications where horses were of limited utility. Authorial Bias and Non-Milites at War in Salian Germany In turning back to problem of the non-milites who served in warfare, it must be emphasized that much of the argument for the transformation of the military organization of Frankish empire and its successor states, whether under the Carolingians or the Ottonians, depends upon the reading of legal texts, such as capitularies and Landfrieden, in isolation from narrative sources.35 Those scholars who adduce evidence from narrative sources to support the contention that military levies gradually disappeared (whether in the ninth or the tenth century) tend to use examples that are not treated in context, particularly with regard to the role that these passages were intended to play in the author’s overall account.36 From a methodological perspective, selecting isolated or unrepresentative passages from narrative works in order to suit a scholar’s chosen conclusion is particularly problematic. It is well understood that medieval authors, in general, and eleventh-century authors, in particular, were not engaged in a transparent effort to reveal the past “as it really happened.”37 Rather, authors in the pre35 36

37

See above for the use of legal sources by scholars including Hans Fehr and Josef Fleckenstein. With regard to the selective use of narrative texts that leads to a misrepresentation of the text as a whole, see the discussion by Bernard S. Bachrach and David S. Bachrach, “Saxon Military Revolution, 912–973? Myth and Reality,” Early Medieval Europe 15(2007), 186–222. The crucial observations by Walter Goffart, The Narrators of Barbarian History: Jordanes, Gregory of Tours, Bede, and Paul the Deacon (Princeton, 1988) with regard to the highly politicized approach to the past by early medieval writers certainly are apropos to the discussion of the authors of eleventh-century texts. For an overview of the main narrative sources for the Salian period and a discussion of their biases with a focus on what they have to say about warfare and violence, see Simon M. Karzel, Nihil crudelius a barbaris perpeti potuissent: Die Darstellung von Krieg und Gewalt in den historiographischen Quellen zur Zeit Heinrichs IV. (Marburg, 2008).

10

David Bachrach

crusade period, most of whom were aristocratic clerics, were heavily biased not only in favor of their particular patrons, but also toward members of their own families, and more broadly toward individuals characterized by an elevated social and economic status.38 In the context of writing military history these biases are manifested in the considerable attention given to relatively small numbers of aristocrats engaged in combat operations, but also, on occasion, in the denigration of members of the lower social orders for rhetorical purposes. As a consequence, when medieval writers note only the deaths of aristocrats in combat, this cannot be understood ipso facto as evidence that non-aristocrats were absent from the battlefield. Similarly, when these same authors write disparagingly about men with low social status on military campaign, it is incumbent upon the historian to determine whether this passage is reflective of reality, or whether the author is indulging his biases for rhetorical purposes. A couple of examples will help to illuminate this problem. Bruno of Merseburg, a Saxon aristocrat and cleric who was employed by both the archbishop of Magdeburg and the bishop of Merseburg during the course of the 1070s and 1080s, devoted considerable attention throughout his history of the civil war in Germany, the Saxonicum Bellum, to lauding and highlighting the positive contribution of nobles and the crude, violent, and uncivilized behavior of peasants and others in the lower strata of society.39 Bruno’s hostility toward the lower social orders manifests itself on several occasions in military contexts. Bruno claims, for example, that during the siege of Würzburg in August 1077, the anti-king Rudolf of Rheinfelden was reluctant to capture the city by storm because he feared that he would not be able to keep the lower-class elements in the army in check, and stop them from looting churches and stealing clerical property once they overwhelmed the defenders.40 Consequently, according to Bruno, Rudolf preferred to deploy his troops in a lengthy and expensive siege rather than risk the dishonor that would result from allowing a successful assault

38

39

40

See, for example, David S. Bachrach, “The Military Organization of Ottonian Germany, c. 900–1018: The Views of Bishop Thietmar of Merseburg,” Journal of Military History 72 (2008), 1061–88, here 1067–68. With regard to authorial focus on aristocrats in narrative texts so as to memorialize the author’s own family and friends, see Lutz E. von Padberg, “Geschichtsschreibung und kulterelles Gedächtnis: Formen der Vergangenheitswahrnehmung in der hochmittelalterlichen Historiographie am Beispiel von Thietmar von Merseburg, Adam von Bremen und Helmold von Bosau,” Zeitschrift für Kirchengeschichte 105 (1994), 156–77. The best edition of Bruno’s Saxonicum Bellum is now to be found in Quellen zur Geschichte Kaiser Heinrichs IV.: Die Briefe Heinrichs IV., Das Lied vom Sachsenkrieg, Brunos Sachsenkrieg, Das Leben Kaiser Heinrichs IV., ed. and trans. by Franz-Josef Schmale and Irene Schmale-Ott (Darmstadt, 1974), 192–405 with facing-page German translation, hereafter Saxonicum Bellum. See the discussion of Bruno’s background and employment by David S. Bachrach and Bernard S. Bachrach, “Bruno of Merseburg and his Historical Method c. 1085,” Journal of Medieval History 40 (2014), 381–98. With regard to Bruno’s tendency to highlight the deaths of aristocrats and to downplay the deaths of members of the lower social orders in combat, see Saxonicum Bellum, chs. 46, 98, 102 and 117. Bruno, Saxonicum Bellum, ch. 94.



Military History of Salian Germany

11

on the city by the “vulgar crowd,” an assault that he believed would lead to the destruction of sacred sites.41 An evaluation of this passage suggests that it was unlikely that Rudolf refrained from launching an assault on Würzburg simply because he feared what the undisciplined vulgus would do once inside the walls. It is rather more probable that he feared to sustain the heavy casualties that would result from attempting to storm this well-defended fortress city. What is clear, however, is that Bruno wanted his audience to believe that the vulgus was responsible for Rudolf’s failure to capture the city, and that he expected his readers to find it plausible not only that the vulgus was present as part of the besieging army, but that these lower-class men would play an important role in any assault against the city. Another author who sometimes indulges in gratuitous attacks on those whom he viewed as social inferiors is Lampert of Hersfeld. For example, in his account of the rebellion by the citizens of Cologne against their archbishop in 1074, Lampert castigates the former for their soft manner of life and their inability to understand the true rigors of military life.42 Lampert claims: “it is not difficult to move men of this kind in any way that you might wish, just like a leaf is seized by the wind, because they are raised from their youth with the delights of the city, and never have any experience in war. After selling their wares, they are accustomed to discuss military affairs over wine and food, and think that anything that might occur to them is no sooner said than done. They have no idea how this will end.”43 Lampert was a strong partisan of the Thuringian and Saxon nobility, who were engaged in a lengthy rebellion against King Henry IV. In addition, he demonstrates throughout his Annales an acute sense of the rights of ecclesiastical magnates.44 So it is hardly surprising that he chooses to denigrate townsmen who not only opposed their bishop, but also were strong supporters of Henry IV. Even when keeping these biases in mind, however, when one considers this passage in isolation, the reader is left with the image of fat burghers boasting around the dinner table about the great feats of military prowess that they planned to achieve. Certainly such men could be of little value on the battlefield. However, reading further, one finds that Lampert paints a very different picture of the actual military effectiveness of the population of Cologne. He presents the populace marching to the archiepiscopal palace, equipped with helmets and

41 42 43

44

Ibid. Lamperti Annales, hereafter Lampert, Annales, in Lamperti Monachi Herfeldensis Opera, ed. Oswald Holder-Egger (Hanover, 1894, repr. 1956), p. 187. Ibid., “Nec difficile fuit id hominum genus in omne quod velles tamquam folium quod vento rapitur transformare, quippe qui ab ineunte aetate inter urbanas delicias educati nullam in bellicis rebus experientiam habebant, quique post venditas merces inter vina et epulas de re militari disputari soliti omnia quae animo occurrissent tam facilia factu quam dictu putabant, exitus rerum metiri nesciebant.” See the discussion by Karzel, Nihil crudelius, pp. 41–44.

12

David Bachrach

armor, as the light of flaming torches gleamed off the points of their swords.45 After a brief fight, they drove the archbishop and his men from the city.46 As these examples suggest, the biases of the narrative sources are toward the depiction of the kind of aristocratic warrior nobility identified by scholars such as Auer, Fleckenstein, and Keller. In addition, the tendency of aristocratic authors to use images of their social inferiors to make rhetorical points often has been interpreted, as in the case of Beyerle and Leyser, as indicative of the lack of military effectiveness of non-noble combatants. However, these scholars have not taken into account the biases of the aristocratic authors of narrative sources, authors who were focused on the exploits of their aristocratic audiences and frequently used men of lower social and economic status to make rhetorical points. In light of these pervasive biases, it is clear that arguments from silence regarding the absence of men of low social status on campaign have a rather high hurdle to clear. It is methodologically unsound to conclude that such men were not on campaign simply because authors who were not interested in them did not mention them. Concomitantly, when authors do discuss the presence of fighting men from the lower orders on campaign, this must be understood as going against their natural tendencies, and therefore having rather more significance as a result. Similarly, when aristocratic authors mention the effective military performance of men with low social and economic status, this too must be taken seriously as going against their general tendencies and biases. When the information presented in eleventh-century narrative sources is evaluated, several other factors also require attention. First, authors of the main narrative texts for eleventh-century Germany, such as Berthold of Reichenau, Lampert of Hersfeld, Bruno of Merseburg, Bernold of Constance, and the anonymous author of the Vita Heinrici, were very careful in their discussion of military matters to distinguish between milites, who were the members of the military households of secular and ecclesiastical magnates, and other types of fighting men.47 Second, these authors also were inclined to see themselves as operating in the Isidorean tradition of history, in which the narrator was responsible for discussing events that actually happened in the past, in so far as this reality served the rhetorical objectives of the text.48 45 46 47 48

Lampert, Annales, p. 187, “galeatum, loricatum, igneo mucrone terribiliter fulgurantem …” Ibid. This point is treated in depth by Bachrach, “Milites and Warfare,” forthcoming. Isidore of Seville, Etymologiarum sive originum libri xx, ed. W. M. Lindsay (Oxford, 1911, repr. 1957), 1.40. “Historiae est narratio rei gestae, per quam ea, quae I praeterito fact sunt, dinoscuntur.” Also see the translation of the work, The Etymologies of Isidore of Seville, trans. Stephen Barney, J. W. Lewis, J. A. Beach, and Oliver Berghof (Cambridge, 2006). Regarding the widespread acceptance of Isidore’s dictum see Benoît Lacroix, L’Historien au moyen âge (Paris, 1971), particularly ch. 1, and the review article of this work by Robert W. Hanning, History and Theory 12 (1973), 419–34; A. J. Minnis, Medieval Theory of Authorship: Scholastic Literary Attitudes in the Later Middle Ages (London, 1984), particularly pp. 1–20; D. H. Green, Medieval Listening and Reading: The Primary Reception of German Literature 800–1300 (Cambridge, 1994), pp. 246–48; Gabrielle M. Spiegel, The Past as Text: The Theory



Military History of Salian Germany

13

Finally, these authors also appear to have accepted the imperatives of the Ciceronian tradition of historical writing that required the presentation of a plausible framework in which to tell a story to an audience that was familiar with the broader topic, but not necessarily with the details of particular events.49 As a consequence, when the authors indicate that the men in a particular military force were not milites, the proper conclusions to draw are, first, that such troops were not members of the military households of magnates, and second, that the audiences for these narrative texts found it plausible that such non-professional fighting men served in military operations. In this context, the authors under discussion here tend to treat the participation of non-milites in offensive military operations in two ways: (1) the use of the presence of such men in order to make a rhetorical point; (2) the formal distinction between milites and other troops who are participating in a battle or siege. Rhetorical Depiction of Non-Milites on Campaign As I noted above, Bruno of Merseburg chose to blame the failure of Rudolf of Rheinfelden to capture Würzburg in 1077 on his ostensible fear that the vulgus in his army would get out of hand when the city was taken. This use of men from the lower social orders as a rhetorical foil in support of the author’s parti pris is a common topos in the narrative texts of this period.50 Bruno himself utilized this ploy again when discussing the outbreak of a widespread rebellion against Henry IV and the junction of rebel Saxon and Swabian armies in the Rhineland in the autumn of 1076. Although they were now allies, the Saxon and Swabian princes feared to bring their armies into a single camp, according to Bruno, because they were worried about the potential for violence between their men. The two armies had fought on opposite sides the previous year at the battle of the Unstrut (9 June 1075) and many men in both the Swabian and Saxon armies were nursing grievances. Bruno specifically comments that it was the men of the lower social orders (viles personae) who posed the greatest source of

49

50

and Practice of Medieval Historiography (Baltimore, 1997), pp. 88–90; Almut Sauerbaum, “Accessus ad auctores: Autorkonzeption in mittelalterlichen Kommentartexten,” in Autor und Autorschaft im Mittelalter, ed. Elizabeth Andersen, Jens Haustein, Anne Simon, and Peter Strohschneider (Tübingen, 1998), pp. 29–37; and Sebastian Coxon, “Zur Form und Funktion einiger Modelle der Autorenselbstdarstellung, ‘Wolfdietrich’ und ‘Dietrichs Flucht’,” in ibid., pp. 148–62, particularly 151 and 162. A full study of the approach toward history by German writers in the Salian period is a desideratum. For an introduction to this issue and a discussion of the approach taken by Bruno of Merseburg, see Bachrach and Bachrach, “Bruno of Merseburg,” 381–98. With regard to the impact of Cicero’s ideas about the role of the historian on the writing of historical works in the period after c. 1000, see Justin C. Lake, “Truth, Plausibility, and the Virtues of Narrative at the Millennium,” Journal of Medieval History 35 (2009), 221–38; and Matthew Kempshall, Rhetoric and the Writing of History, 400–1500 (Manchester, 2011), here 255. Concerning the depiction of violence by members of the lower social orders as well as city dwellers, see Karzel, Nihil crudelius, pp. 124–34.

14

David Bachrach

concern for the commanders.51 As it turned out, there was no violence between the Saxon and Swabian armies, and, as a result of their unity, Henry IV was forced to sue for peace and promise to seek absolution from Pope Gregory VII (1073–85) for his sins. However, Bruno was able to use this scene to pursue his agenda of demonstrating the moral superiority of aristocrats as contrasted with the vulgus. Bruno again points to the difficulties of employing men of the lower social orders on campaign in his discussion of the battle between Rudolf of Rheinfelden and King Henry on the Elster river (14 October 1080). Before the battle began, the Saxon army was divided into two parts, with the intention of catching the royalist forces in a pincer. Mounted troops under the direct command of Rudolf approached the royal camp from the west, but were met and defeated by Henry IV’s own mounted forces. The Saxon foot soldiers, under the command of Otto of Northeim, approached the royal camp from the north, cutting through a densely wooded area that Henry and his advisors had believed was impassable. When the Saxon foot soldiers launched their attack on Henry IV’s camp, the surprise was complete and the Saxons drove the king’s men into the waters of the White Elster river, where many of them are reported to have drowned.52 Clearly, the foot soldiers of the Saxon levy acquitted themselves quite well in this battle. However, Bruno again permitted his bias against men from the lower social orders to come to the fore when describing the aftermath of this victory. According to Bruno, once the royalists had been driven off, the entire focus of the Saxon foot soldiers was on looting King Henry’s camp, so that they could satisfy their greed and keep all the booty for themselves. Bruno asserts that they were restrained from indulging their foolish desires only by the good sense of the nobleman Otto of Northeim, who warned the foot soldiers to stay away from the royal camp until it was clear that all of the enemy troops had been defeated.53 The Saxon foot soldiers did obey Otto’s commands, which indicates that they could follow orders even when these were contrary to their perceived interests. Lampert of Hersfeld, despite his hostility toward the urban levy of Cologne, noted above, was less inclined than his contemporary Bruno to engage in the wholesale denigration of men from the lower social orders in military matters. This was particularly the case in regard to those who opposed King Henry IV, the anti-hero of Lampert’s Annales. However, when discussing military matters Lampert did play on what would seem to have been the anti-peasant biases of 51 52 53

Bruno of Merseburg, Saxonicum Bellum, ch. 88. Ibid., ch. 122. Ibid. As a post-script to the battle on the Elster river, Bruno drew attention to the many calamities that were suffered by King Henry’s troops. Among these were the subsequent deaths of many powerful men at the hands of peasants defending their own homes, who were armed with axes and clubs. Yet other magnates were captured and held for ransom by local inhabitants, whom Bruno denotes as low-born people (personae viles). By contrast, Bruno claimed: “if any of the enemy were brought to one of our upright men, they were healed if they were wounded, and were sent back to their fatherland free of charge, well outfitted with clothing and arms.” See ch. 123.



Military History of Salian Germany

15

many aristocrats. For example, in his account of the days leading up to the battle of the Unstrut, Lampert took pains to explain why King Henry and his advisors were so confident about engaging their Saxon opponents. Henry had dispatched scouts (exploratores) to determine the size and disposition of the Saxon army. The scouts reported back that although the Saxons equaled the royal army in numbers and were equipped with roughly the same armament, the king’s troops were otherwise superior in every other way for battle.54 In discussing whether to advance against the Saxon camp, the king’s advisors, according to Lampert, stressed that they had a picked force (miles lectissimus). By contrast, the majority of the Saxons were peasants (rustici) and comprised, therefore, an inept rabble (ineptus vulgus), who were more accustomed to agricultural labor than to military service.55 The king’s advisors added that the Saxons lacked the military spirit that was necessary for war, and had been mobilized because of their fear of their princes. Consequently, these peasants could not be expected to stand and fight once the battle began and sword was raised against sword. In fact, they would certainly flee before the battle even began, terrified by the noise as the two armies approached one another.56 In discussing the battle itself, Lampert takes considerable pains to show that this assumption by the king’s advisors was ill-founded. The Saxon troops, including the vulgus, initially gave a very good account of themselves in the fighting, before ultimately being driven from the field.57 While not so subtly mocking the assumptions by Henry and his advisors about the lack of military skill of the Saxon peasantry, Lampert, himself, indulges in this very same rhetorical theme in his description of the aftermath of the battle with regard to the men of the lower social orders who were serving in the royal army. As Lampert makes clear, following a lengthy battle, the Saxon line was shattered by simultaneous attacks on both flanks by King Henry’s troops.58 In discussing the subsequent rout of the Saxons, Lampert, drawing upon Sallust’s description of the retreat of an army in the Jugurthine war, observed “as is always the case when the enemy flees, the most cowardly and the most courageous are equally brave [in pursuit].”59 Lampert then made the severe observation that all of the royal troops went off in pursuit of the fleeing Saxons, “even the plebeians and the peasants who usually hung around the camp doing servile work.”60 Lampert’s contention here regarding the plebei ac rustici could, if taken 54 55 56

57 58 59 60

Lampert, Annales, p. 216. Ibid., “Illinc vulgus esse ineptum, agriculturae pocius quam miliciae assuetum.” Ibid., “et ideo non expectaturum, ut commisso certamine comminus stricto ense feriret et feriretur, sed ante commissum prelium solo concurrentis exercitus strepitu et clamore terrendum, fugandum et fundendum esse.” Ibid., pp. 217–21. Ibid., p. 220. Ibid., “ut semper in fuga hostium ignavissimis et fortissimis par solet esse audacia.” Here, Lampert draws on Sallust, Jugurtha, c. 53. Lampert, Annales, p. 220, “omnes in exercitu regis legiones confusis ordinibus, omnes etiam plebei ac rustici, qui castrorum usibus servilem operam dependebant …”

16

David Bachrach

out of context by scholars seeking “evidence” for the military marginalization of the rustici, be utilized to show that men of this status had only ancillary roles in military campaigns as servants, cooks, attendants for horses, and the like. However, as we will see below, this is not accurate.61 In addition to using the rustici and plebei as a rhetorical foil in his discussion of the pursuit of the Saxons at the battle of the Unstrut, Lampert also addresses Henry IV’s mobilization of non-milites in an effort to show how little support the king had from the magnates for his war against the Saxons. In the context of the initial Saxon and Thuringian rebellion in July 1073, Henry tried to divert an army that he had mobilized for operations in Poland to suppress the widespread revolt against his rule in Saxony and Thuringia. However, Henry’s secular and ecclesiastical magnates refused to consider this plan.62 Consequently, Henry dismissed the army but sent messengers to all of the principes in the kingdom, announcing that he would be mobilizing a new army in early October that was to join him at the villa of Bredingen near Hersfeld. The specific purpose of this army was to undertake operations in Saxony.63 Lampert takes pains to stress that on this occasion the royal messengers were dispatched not only to the princes but to the people as well (non solum principibus sed et popularibus). As Lampert makes clear, Henry’s decision to mobilize the populus was inspired, in no small measure, by his fear that the magnates either would not come to the muster at Bredingen or would not fight if they did go on campaign.64 It turns out that King Henry was right to be worried. The German princes refused to fight, and instead compelled him to make a peace agreement with the Saxons and Thuringians at the villa of Gerstungen in the autumn of 1073.65 With his options limited in regard to Saxony, Henry turned his attention to the potentially advantageous request by his brother-in-law Salomo to help him regain the Hungarian throne. In return for his aid, Henry would receive a series of frontier fortifications in Hungary that would help to ensure that the Hungarians would no longer be able to raid into Bavaria.66 Henry therefore issued summonses in July 1074 to his principes ordering them to mobilize for a campaign in Hungary. But, according to Lampert, the princes refused, citing the costs that they had just undertaken for the expedition to Saxony the previous year that had led to the peace of Gerstungen. As a consequence, Lampert avers, Henry was forced to begin his Hungarian campaign with a very limited force comprising just

61 62 63 64

65 66

As noted above, both Leopold Auer and Karl Leyser presented men with lower social and economic status serving only menial tasks on campaign. Lampert, Annales, p. 157. Lampert, Annales, p. 158. Ultimately, the German princes refused to fight the Saxons, and Henry IV was forced to make concessions to the Saxons in an agreement that is commonly known as the peace of Gerstungen. See Lampert, Annales, p. 163. Ibid. See the discussion by Lampert, Annales, pp. 197–98.



Military History of Salian Germany

17

the royal military household (miles privatus) and a force of non-professional fighting men, whom Lampert denotes as the common herd (gregarius).67 Milites and other Troops In addition to the mention of rustici, plebei, viles personae, gregarii, and the vulgus in military campaigns for the purpose of making a rhetorical point, the authors of narrative sources also routinely distinguish between milites and other types of troops when discussing the mobilization and deployment of fighting men for military operations. The Swabian author Berthold of Reichenau, for example, emphasized that when Henry IV returned to Germany from Canossa in June 1077, he immediately mobilized a large army in order to crush the rebels who were opposed to his rule.68 Berthold distinguished among these various forces, noting the presence of Henry’s own milites as well as the troops provided by the Swabian bishops of Basel and Strassburg and the secular magnate Count Palatine Hermann. These household troops are contrasted with a substantial auxiliary force (non modica auxiliariorum militia) of Bavarians and Carinthians, almost the entire strength (virtus) of Burgundy, and substantial force (non modica pars) of Franconians.69 Berthold, who was very precise in his use of military terminology, reserved alternate terms to denote the Bavarians, Carinthians, Burgundians, and Franconians. This was his way of signaling to his audience that these forces included non-milites in their ranks. As we will see below, many of these men were, in fact, rustici. Lampert of Hersfeld also frequently used the technique of distinguishing between milites and other types of troops and used a variety of terms to denote the latter, including exercitus and copiae.70 In his discussion of the initial phases of the battle on the Unstrut, noted above, Lampert conspicuously does not draw the reader’s attention to the milites in the royal army, but rather observes that the exercitus of the Swabians and the exercitus of the Bavarians retreated after suffering considerable losses at the hands of Otto of Northeim’s Saxon troops.71 As noted above, these Saxons included men of all social statuses, including rustici. Lampert then makes clear that the royal position on the battlefield was salvaged when the milites from the military households of the count of Gliesburg and the bishop of Bamberg struck the Saxon line on the two flanks.72 As noted earlier, Bruno of Merseburg, writing independently of Lampert, observed that the Swabian forces at the battle of the Unstrut included viles

67 68 69 70 71 72

Ibid., p. 198. Die Chroniken Bertholds von Reichenau und Bernolds von Konstanz, ed. I. S. Robinson, MGH SS rerum germanicarum nova series 14 (Hanover, 2003), hereafter Chronicon, here p. 276. Berthold, Chronicon, p. 276. See, for example, Lampert, Annales, pp. 105, 107, 119, 123, 142, 148, 161, 188, and 214. Lampert, Annales, p. 220. Ibid. The successful assault by these milites was then followed up by the copiae of the Bohemians and the Lotharingians, including a substantial number of mounted troops (equis).

18

David Bachrach

personae, who were still angry about the losses that they had suffered in battle at the hands of the Saxons. The presence of such men in the Swabian battle line certainly helps to explain Lampert’s decision to denote the entire Swabian force as the exercitus of the duchy rather than as a congeries of military households of Swabian magnates, i.e. milites, since the Swabians included a mixed group of professional and non-professional fighting men. Lampert similarly distinguished between the mobilization of a large force consisting of both non-milites and professional forces in his discussion of a campaign undertaken by Margrave Ekbert of Meissen against Henry IV’s Bohemian allies in August 1076.73 In the aftermath of his resounding victory at the battle on the Unstrut, Henry had handed over a series of fortifications in the march of Meissen to Duke Vratislav II of Bohemia (1061–92) as a reward for his aid in this campaign. Ekbert, joined by “the Saxons,” as Lampert has it, recaptured all of these strongholds. After the Saxons withdrew and returned to their home, Ekbert installed garrisons of his own milites, to keep the region secure.74 In a similar manner, Lampert distinguishes between the Thuringians who besieged royal fortifications in the summer of 1073 and the royal milites who defended these strongholds.75 Lampert observes that a multitude (multitudo) of Thuringians from the nearby area (ex vicinis locis) undertook a siege of the fortress of Heimenburg. Clearly this was not a small force of heavy cavalry. Rather than attempting to starve the garrison into submission, they assaulted the castellum, and captured it.76 The Thuringians, whom Lampert now denotes as an exercitus, then began operations at the fortress of Asenburg. But the strong natural defenses of the fortress as well as the toughness of the garrison, whom Lampert denotes as milites regis, were sufficient to thwart the efforts of the Thuringians, who maintained their siege from August through December 1073.77 The anonymous author of the Carmen de bello Saxonico, a panegyric written on behalf of King Henry just after his victory on the Unstrut in June 1075, similarly distinguishes between milites and other types of troops in his discussion of the mobilization of the Saxons to defend their region in early 1074.78 He claims that as word went throughout Saxony to prepare for war, the rustici cast aside their farming implements, the shepherds left their flocks, and the merchants abandoned their wares. These men, from every station and every profession, girded themselves for war. The poet then declared that this varied

73 74 75 76

77 78

Lampert, Annales, p. 273. Ibid. Lampert, Annales, p. 161. Ibid. The author of the Carmen de bello Saxonico claims that some three thousand men participated in the attack on Heimenburg . See Quellen zur Geschichte Kaiser Heinrichs IV.: Die Briefe Heinrichs IV., Das Lied vom Sachsenkrieg, Brunos Sachsenkrieg, Das Leben Kaiser Heinrichs IV., ed. and trans. by Franz-Josef Schmale and Irene Schmale-Ott (Darmstadt, 1974), hereafter Carmen, bk 1, lines 87–124. Lampert, Annales, pp. 161 and 172. Carmen, bk 2, lines 130–44.



Military History of Salian Germany

19

force of farmers, shepherds and merchants, equipped with whatever arms they had available, joined with the milites and rushed to war.79 Many of the non-milites considered up to this point are identified in the sources as rustici, that is, men from the countryside. However, urban levies also played an important role in military operations. As I noted earlier, Lampert of Hersfeld drew attention to the urban levy of Cologne, and their successful effort to drive the archbishop’s household troops from the city. The author of the Carmen de bello Saxonico, in the passage discussed above, mentions that mercatores were to be found in the Saxon host in early 1074. Earlier in the poem, he had observed that cobblers, smiths, bakers, and butchers had joined with milites from Goslar to participate in the Saxon siege of the royal fortress of Harzburg.80 Bruno of Merseburg, for his part, scornfully observed that Henry IV’s army in the summer of 1077 was exceptionally weak, and that most of his men were mercatores.81 The anonymous author the Vita Heinrici, which was written shortly after the emperor’s death in 1106, presents a much more positive image of urban levies in his discussion of the conflict between Henry IV and his son Henry V.82 After the elder Henry’s escape from captivity in early 1106, he made his way to Liège, where he was warmly received by both Bishop Otbert (1091–19) and Duke Henry of Lower Lotharingia (1101–06). According to the Vita, when Henry V began making preparations to march against his father, the Lotharingian duke along with the citizens of Cologne and Liège prepared their arms, gathered their forces, and improved the defenses of their cities. In addition, they promised Henry IV their full support in the coming struggle.83 The Vita goes on to make clear that the citizens of Cologne, Liège, and other cities did not intend simply to defend their walls against Henry V, they also planned to take action against his forces directly in the field. When Henry V began his siege of Cologne, his initial assaults were driven back by the citizens. At the same time, the citizens of other cities prepared a fleet and blockaded the Rhine river so that the younger Henry’s army could not receive supplies by water. Then, when parties of foragers were sent out from Henry V’s siege camp, the citizens of Cologne and others, who knew the local terrain, attacked the foraging parties.84

79 80 81 82

83

84

Ibid. Carmen, bk 1, lines 198–99. Bruno of Merseburg, Saxonicum Bellum, ch. 95, “nam maxima pars eius ex mercatoribus erat.” For the text of the Vita Heinrici see Quellen zur Geschichte Kaiser Heinrichs IV.: Die Briefe Heinrichs IV., Das Lied vom Sachsenkrieg, Brunos Sachsenkrieg, Das Leben Kaiser Heinrichs IV., ed. and trans. by Franz-Josef Schmale and Irene Schmale-Ott (Darmstadt, 1974), hereafter Vita Heinrici, c. 13. Ibid., “Cum igitur audisset H dux et Colonienses cum Leodicensibus, quod super se rex exercitum ducere vellet, arma parabant, copias colligebant, urbes firmabant et ad resistendum pari voto studioque se accingebant.” Ibid., 462–64.

20

David Bachrach

Military Organization of Non-Milites The examples treated thus far indicate that non-milites, that is, men outside of the military households of magnates, played a substantial role in combat operations. One important question, however, is the extent to which the mobilization of such men can be understood as routine rather than as ad hoc. Two authors working independently from each other and in different regions of the German kingdom, Berthold of Reichenau and Lampert of Hersfeld, give indications that the mobilization of such men was not an ad hoc response to the exigencies of a particular campaign. Rather, non-milites expected to serve on campaign when called upon to do so by leaders with the legal jurisdiction to summon them. Moreover, both authors indicate that this jurisdiction was territorially based. For example, Berthold of Reichenau observes that in the summer of 1078, Henry IV dispatched bishops Burchard of Basel and Thiepald of Strassburg, noted above, to deploy their forces to ravage the lands of Duke Berthold of Carinthia.85 The chronicler comments that the military forces of the two bishops were comprised of two separate elements. The first of these consisted of their household troops, that is, their milites. The second component consisted of rustici. These rustici, Berthold of Reichenau explains, were mobilized on the basis of the counties (comitatus) in which they lived. Moreover, these men were required to serve in this campaign because of the oaths that they had sworn to the bishops, i.e they were adiurati.86 The transfer of comital authority to German bishops, with the concomitant responsibility to mobilize the levies of the comitatus for military operations, dates back to the reign of Otto III (983–1002).87 The bishops of Strassburg had enjoyed comital rights and undertaken comital obligations since the late tenth century.88 The bishops of Basel had received specific comital authority in Augstgau and Sisgau in 1041 from King Henry III.89 Consequently, the mobilization of these rustici is consistent with the ongoing practice of both Ottonian and Salian kings to utilize their bishops to undertake comital duties on their behalf. Berthold again draws a clear connection between territorially based mobilization of non-milites and the taking of oaths in his discussion of the mobilization of Franconian rustici, who fought just a short time later against Swabian, Carin-

85 86 87

88

89

Berthold of Reichenau, Chronicon, p. 332. Ibid. See the discussion of the utilization of bishops to serve comital functions by Hartmut Hoffmann, “Grafschaften in Bischofshand,” Deutsches Archiv für Erforschung des Mittelalters 46 (1990): 375–480. Die Urkunden Otto des II und Otto des III., ed. Theodor Sickel (Hanover, 1888), DO II, nos. 72 and 73; Die Urkunden Heinrichs II. und Arduins, ed. Harry Bresslau Robert Holtzmann and Hermann Reincke-Bloch (Hanover, 1900–3), no. 367; and Die Urkunden der deutschen Könige und Kaiser: Die Urkunden Heinrichs III., ed. Harry Bresslau and Paul Kehr (Berlin, 1931), no. 220. Die Urkunden Heinrichs III, no. 77.



Military History of Salian Germany

21

thian, and Bavarian forces under the command of Duke Welf of Bavaria and Duke Berthold of Carinthia.90 The latter were marching to join the Saxon army under the command of Rudolf of Rheinfelden, who was advancing into eastern Franconia. Henry IV was attempting to keep his Saxon and southern opponents from joining their forces, with the plan of facing and defeating them seriatim. Henry therefore marched into eastern Franconia to face the Saxons, whom he faced at the battle of Mellrichstadt (7 August 1078). Concomitant with his own military operations, the king ordered western Franconian rustici to mobilize in an effort to delay Duke Welf and Duke Berthold. According to Berthold of Reichenau, these Franconians, whom the author denotes as both comprovinciales and rustici, were mobilized from their centenariae, i.e. their local sub-comital districts. Moreover, he observes that they were bound together for military service by their oaths (coniurati). Not incidentally, Berthold also notes that these rustici were equipped with weapons of war (arma militaria), likely in contrast to farm implements such as scythes and pitchforks.91 This final observation permits the inference that these were wealthier rustici, who could afford specifically military equipment. The purchase and possession of arma militaria, moreover, would seem to be prima facie evidence of the expectation on the part of these men that they could be called upon to do military service. Lampert of Hersfeld, for his part, illuminates the legal and jurisdictional basis for the mobilization of men in Lotharingia and in Saxony for military service. In discussing the army that King Henry mobilized for operations in Saxony in the autumn of 1075, Lampert notes that Duke Theoderic II of Upper Lotharingia (1075–1115) and Duke Godfrey IV of Lower Lotharingia (1069–76) brought enormous forces to join the royal army. In fact, Lampert mocks the dukes for attempting to show off by standing out among the other princes on the basis of the size of their armies.92 In commenting on these troops, Lampert observed that the Lotharingians were very well equipped. The parallel with the Franconian rustici is quite clear. Lampert then explains that the two dukes had insured the high quality of their military forces by imposing very strict selection criteria on the men whom they mobilized, that is, they picked only the best men for campaign duty. Lampert explicitly states, moreover, that these men were mobilized from the regions over which the two dukes had jurisdiction.93 Lampert’s comments in this regard illustrate that Theoderic and Godfrey had the authority to mobilize troops because of their ducal offices. In addition to being territorially based, this authority to mobilize fighting men had an intrinsic selection mechanism that allowed the two dukes to mobilize the men who were best equipped and best suited for 90 91 92 93

Berthold of Reichenau, Chronicon, pp. 332–33. Ibid. Lampert, Annales, p. 234. Ibid., “ita militaribus armis instructas, ita de tota cui preerat regione severissimo delectu habito exquisitas …”

22

David Bachrach

service on military campaign. In light of Lampert’s consistent effort to mention when particular magnates and milites were mobilized for service, his failure to do so in this case indicates that the copiae in question included large numbers of non-milites. In one final example, Lampert describes the mobilization of a vast Saxon army in January 1074 to repel what the Saxons feared would be an invasion of their region by King Henry.94 Almost certainly exaggerating, Lampert claims that as many as 40,000 Saxons, of all ranks, came to Werra river.95 This is the same force regarding which the author of the Carmen de bello Saxonico, noted above, had claimed the mobilization of farmers, shepherds, and merchants alongside the milites. As it became clear that there would be no immediate invasion, the Saxons sent a substantial part of their army home. In discussing the demobilization of part of the Saxon force, Lampert specifies that 12,000 of the men of lower social and economic status (plebei) were released. Lampert explains that these men, in particular, had been summoned with so little notice that they had not brought food with them, and had to go home.96 The clear implication of Lampert’s comment is that even men of “plebian” status not only had the expectation of going on campaign outside their home districts to fight, but under normal circumstances would bring sufficient supplies with them to be able to remain in the field for a substantial period of time. Conclusions As the many examples of plebei, rustici, vulgus, gregarii, viles personae, mercatores, and just plain cives treated here make clear, warfare in the eleventh century, including combat operations, was not monopolized by milites, whether or not one wishes to construe such men as mounted knights. In fact, non-milites played a central role both in battles in the field and in sieges. In fact, they likely were the numerically preponderant element in any large army. It is notable that all of the combat operations that I have discussed took place during Henry IV’s reign, and most of these in the context of the civil wars of the 1070s and 1080s. One might conjecture on this basis that the mobilization of the vulgus was the consequence of the enormous demands of the lengthy military conflict, and therefore a return to the type of military organization that typified the Frankish empire before the ostensible disappearance of military levies. However, such a conjecture would be incorrect. First, none of the authors discussed here gives any indication that the mobilization of non-milites for combat operations marked a new or surprising turn of events. Rather the 94 95

96

Lampert, Annnales, p. 176. The vast size of the Saxon army is confirmed by both the anti-Henry Bruno of Merseburg, and the pro-Henry author of the Carmen de bello Saxonico . See Bruno of Merseburg, Saxonicum bellum, c. 32; and Carmen, bk 2, lines 130–60. Lampert, Annales, pp. 176–77, “quoniam subito clamore in expeditionem evocata cibos secum non sumpsissent, in domos suas, tamquam minus sibi necessaria remitterent.”



Military History of Salian Germany

23

opposite is the case, in that the mobilization of rustici is depicted as normal and well-developed practice. Second, the vulgus and rustici as well as mercatores appear in both Saxon and royal armies right from the beginning of the civil wars in 1073, long before there could have been any question of the exhaustion and depletion of magnate military forces. The weight of the evidence for the deployment of non-milites during Henry IV’s reign is largely an artifact of the production of a large number of lengthy historical works, which devoted enormous attention to military matters because they were inspired by the dramatic events during the final three decades of the eleventh century. Finally, it should be observed that the main basis for the explanation of the elimination of levies as an important element in early medieval military organization is the claim that battles in the field, and moreover battles between mounted forces, came to dominate warfare. As virtually all specialists in military history are well aware, this was never true. Sieges dominated warfare both in the Carolingian empire and in its various successor states.97 Henry IV’s reign was actually quite conspicuous for the relatively large number of battles that were fought, with at least five and perhaps six between 1075 and 1086.98 Nevertheless, these battles were outnumbered by a factor of at least four to one by sieges, including operations against fortress cities such as Würzburg and powerful fortresses such as the Harzburg. In order to undertake these numerous siege operations, military commanders required far more troops than could be mobilized from among the milites employed by magnates. Thus, the rustici and other non-milites of the eleventh century continued to play the same role as their forbears had under the Ottonian kings, and before them the Carolingians and Merovingians as well.

97

98

The scholarship in this area is too vast to summarize here. However, for valuable entry points, see Jim Bradbury, The Medieval Siege (Woodbridge, 1992); and Peter Purton, A History of the Early Medieval Siege, c. 450–1200 (Woodbridge, 2010), and the enormous literature cited there. Cf. Guy Halsall, Warfare and Society in the Barbarian West 450–900 (London, 2003), pp. 215–27. For the relatively large number of battles fought during Henry IV’s reign in the context of the Saxon wars, see Gillingham, “An Age of Expansion, c. 1020–1204,” pp. 73–76. By comparison, there were at least twenty-seven significant sieges during this same period. See the appendix below.

24

David Bachrach

Appendix Sieges during Henry IV’s Saxon Wars, 1072–88 1072 1073 1073 1073 1073 1073 1073 1074 1074 1077 1077 1077 1077 1077 1078 1078 99 100 101 102 103 104 105 106 107 108

109 110 111 112 113 114

Siege of Lüneburg by Saxon forces commanded by Hermann Billung99 Siege of Heimenburg by Thuringian forces opposing Henry IV100 Siege of Asenburg by Thuringian forces opposing Henry IV101 Siege of Harzburg by Saxon forces102 Siege of Wiganstein by Saxon forces103 Siege of Moseburg by Saxon forces104 Siege of Sassenstein by Saxon forces105 Siege of Spatenberg by Thuringian forces opposing Henry IV106 Siege of Vokenrot by Thuringian forces opposing Henry IV107 Siege of Sigmaringen by Rudolf of Rheinfelden108 Siege of fortification in Burgundy by bishops of Basel and Lausanne on behalf of Henry IV109 Swabians loyal to Henry IV siege fortifications of Swabian magnates opposed to Henry IV110 Siege of Würzburg by Rudolf of Rheinfelden111 Siege of the fortifications of Count Eckbert I of Formbach in Bavaria by Henry IV112 Siege of a fortification in Bavaria by Henry IV113 Siege of Tübingen by Henry IV114

Lampert, Annales, pp. 160–61. Lampert, Annales, pp. 159 and 161; Carmen, bk 1, lines 87–130. Lampert, Annales, pp. 159 and 161. Bruno of Merseburg, Bellum Saxonicum, chs. 28–29; Lampert, Annales, p. 159; Carmen, bk 1, pp. 139–237 and bk 2, pp. 74–114. Lampert, Annales, pp. 159. Lampert, Annales, p. 159. Lampert, Annales, p. 159. Lampert, Annales, pp. 159 and 174. Lampert, Annales, pp. 159 and 174. Berthold of Reichenau, Chronicon, p. 277; Bernold of Constance, Chronicon, p. 413; Casuum Sancti Galli continuation II, ed. G. H. Pertz MGH SS 2 (Hanover, 1829), p. 157; and Casus monasterii Petrishusensis, ed. Otto Abel and Ludwig Weiland MGH SS 20 (Hanover, 1869), p. 646. Berthold of Reichenau, Chronicon, p. 289; and Gesta episcoporum Lausannensium, ed. I. Heller MGH SS 24 (Hanover, 1879), pp. 799–800. Berthold of Reichenau, Chronicon, p. 290. Bruno of Merseburg, Bellum Saxonicum, ch. 94; Berthold of Reichenau, Chronicon, pp. 292–93; Bernold of Constance, Chronicon, pp. 458–; Vita Heinrici, p. 422. Berthold of Reichenau, Chronicon, p. 301. Berthold of Reichenau, Chronicon, p. 313. Annales Zwifaltenses, ed. D. Otto Abel MGH SS 10 (Hanover, 1852), p. 54.



Military History of Salian Germany

25

1078 Siege of a fortification in eastern Franconia by Duke Welf of Bavaria and Duke Berthold of Carinthia115 1079 Storming of a fortification held by Count Otto II of Buchhorn in Raetia by Duke Welf116 1079 Siege of a fortification near Ulm by Duke Welf of Bavaria117 1080 Siege of Saxon fortifications by Margrave Eckbert on behalf of Henry IV118 1084 Siege of Augsburg by Duke Welf of Bavaria119 1084 Siege of Burgdorf in Burgundy by forces loyal to Henry IV120 1086 Siege of Regensburg by Duke Welf and Swabian magnates121 1086 Siege of Würzburg by Swabian, Saxon, and Bavarian magnates122 1087 Siege of a fortification in Bavaria by Henry IV123 1088 Siege by Henry IV of the fortification of Gleichen in Thuringia, southwest of Erfurt124 1088 Siege of Quedlinburg by Margrave Eckbert II, Margrave of Meißen125

115 116 117 118 119 120 121 122 123 124 125

Berthold of Reichenau, Chronicon, p. 335; Bernold of Constance, Chronicon, p. 420. Berthold of Reichenau, Chronicon, pp. 348–39. Berthold of Reichenau, Chronicon, p. 360. Berthold of Reichenau, Chronicon, p. 379. Bernold of Constance, Chronicon, p. 438. Bernold of Constance, Chronicon, p. 444. Bernold of Constance, Chronicon, pp. 457–58. Bernold of Constance, Chronicon, pp. 458–60; Vita Heinrici, p. 422. Bernold of Constance, Chronicon, p. 462. Bernold of Constance, Chronicon, p. 473. Vita Heinrici, p. 426.

2 When the Lamb Attacked the Lion: A Danish Attack on England in 1138?1 Thomas K. Heebøll-Holm

In the continuation of Sigebert of Gembloux’s Chronicon sive Chronographia the anonymous continuator noted on the year 1138: When the king of the Danes heard that the king of the English was dead he gathered many ships and an army of knights and footmen and ravaged the borders of England in a most cruel manner, saying that due to his ancestors’ hereditary right and close connection via the sea between them he had a greater claim to the throne of England than king Stephen and the Normans who had acquired it through William the Bastard’s invasion. Considering that it would be dangerous to meet such a fierce enemy head on in an unequal battle the king of the English held back and waited for an advantageous moment and when they had dispersed their forces far and wide gaping at the easy plunder, he defeated them in a battle, killed and captured many of them and forced the rest to go home in disgrace.2

This attack is not mentioned in any English or Danish contemporary sources, and to my knowledge only three modern historians, Frederik Schiern writing in 1860, John Beeler in 1966 and finally Peter Sawyer in 1995, have taken note of it. Schiern lauded the attack, whereas Beeler mentioned it in a footnote and mostly as a curiosity and Sawyer – also in a footnote – flatly refused that the attack had taken place, since it was not mentioned in any other source.3 Apart 1

2

3

The research for this article has generously been funded by the Danish Research Council for Culture and Communication. I would also like to thank Michael H. Gelting and Tiffany D. Vann Sprecher for their suggestions for improvement. “Rex Danorum, audita morte regis Anglorum, cum multo navium apparatu, cum militari et pedestri exercitu, fines Angliae devastabat omni crudelitatis genere; dicens, antecessorum hereditario iure et collimitanei maris vicinitate sibi magis deberi regnum Angliae, quam Stephano regi et Normannis ex Wilhelmi bastardi pervasione. Rex Anglorum reputans periculosum primo impetu cum tam efferato hoste non ex equo confligere, habita dilatione et inspecta oportunitate, dispersos longe lateque et inhiantes predae facili superavit congressione, multisque eorum interfectis aut captis, reliquos conpulit cum dedecore ad propria redire.” Sigeberti Gemblacensis Chronographia, ed. Ludwig Conrad Bethmann, MGH SS 6 (Hannover, 1844), p. 386. Frédéric Schiern, Descente en Angleterre projetée par le roi de Danemark Valdemar Atterdag de réunion avec les Français (Copenhagen, 1860), pp. 11–12; John Beeler, Warfare in England 1066–1189 (Ithaca, NY, 1966), p. 99; Peter Sawyer, Scandinavians and the English in the Viking Age (Cambridge, 1995), pp. 14–15, n. 14.

28

Thomas K. Heebøll-Holm

from these I have found no mention of this attack in either English or Danish modern historiography.4 The Stand der Forschung on Anglo-Danish relations in the twelfth century can therefore be summarized by the following quotation: The last expedition against England was prepared in 1085 and came to nothing when the Danish King Canute IV was murdered in 1086. After that, Danish military activity concentrated upon the Baltic area. A border became fixed between Denmark and England. It is difficult to understand solely in political terms. England after the conquest in 1066 became a strong, centralized state, but it was not so in the early twelfth century – at times during Henry I’s reign, and especially under King Stephen – when [the] Danish kings expanded militarily elsewhere, but seem to have given up [on] England. This shift in orientation is contemporary with and might perhaps best be explained by the initiation of the crusading movement. The conquest of Jerusalem in 1099 and the common Christian fight against [the] infidels must have made it more difficult for [the] Danes to wage war against Christian England when they actually lived next to [the] heathen Wends in the Baltic. Whatever the explanation is, the picture is clear: from 1100 Denmark was oriented only towards the East.5

The passage in the Continuation challenges this view severely, but if the Danes did indeed attack England in 1138 it significantly changes our understanding of Danish and perhaps also English twelfth-century history. The aims of this article are therefore two-fold. First, I discuss whether the Continuation could be right in asserting a Danish attack in 1138. Second, I discuss why we have only one source on the attack and suggest that the reason for this is the result of a deliberate Angevin and Valdemarian suppression of the memory of the event in the contemporary historiography.6

4

5

6

For a sample of recent works on the period 1130–40s in England and Denmark, respectively, see: H. A. Cronne, The Reign of Stephen 1135–54. Anarchy in England (London, 1970); David Crouch, The Reign of King Stephen 1135–1154 (Harlow, 2000); Paul Dalton and Graeme J. White, eds., King Stephen’s Reign (1135–1154) (Woodbridge, 2008); Johannes C. H. R. Steenstrup et al., Danmarks riges historie, 6 vols. (Copenhagen, 1896–1907); Niels SkyumNielsen, Kvinde og Slave (Copenhagen, 1971); Helge Paludan et al., Danmarks historie. Tiden indtil 1340 (Copenhagen, 1978), Nils Hybel, Danmark i Europa 750–1300 (Copenhagen, 2003). Kurt Villads Jensen, “The Blue Baltic Border of Denmark in the High Middle Ages: Danes, Wends and Saxo Grammaticus,” in Medieval Frontiers: Concepts and Practices, eds. David Abulafia and Nora Behrend (Aldershot, 2002), p. 177. John Edward Lloyd argued in 1860 that the Welsh in 1138 in their siege of Cardigan Castle were aided by fifteen Danish ships from Dublin. While he seems to be unaware of the mention of a Danish attack the same year in the continuation of Sigebert’s chronicle, I found it worthwhile to make sure that the chronicler had not conflated this event with a Hiberno-Danish attack. Lloyd’s source for this piece of information is the version C of the Annales Cambriæ. In this passage the people supplying naval reinforcements to the Welsh princes are merely called gentiles, not Dani. Thus there is no evidence of any Danish or Scandinavian participation in the Welsh actions against King Stephen. John Edward Lloyd A History of Wales, 2 vols. (New York, 1911), 2:476; Annales Cambriæ, ed., J. Williams ab Ithel (London, 1860), p. 41. However, there seems to have been a Norwegian raid under the command of Eystein Haraldsson against Yorkshire in the reign of King Stephen. This raid took place in the 1150s and thus the attack



When the Lamb Attacked the Lion

29

The Chronicle, the Continuation and the Franco-Flemish Context Sigebert’s chronicle and the continuations were widely copied, especially in France during the twelfth and thirteenth centuries. In 1844 Ludwig Conrad Bethmann, the editor of the Monumenta Germanica Historica edition of Sigebert’s chronicle and its continuations, counted the chronicle to have existed in sixty-three manuscripts.7 Today, forty-two manuscripts dating from between the twelfth and sixteenth centuries, including Sigebert’s own manuscript currently kept in the Bibliothèque royale de Bruxelles, still exist. Mireille Chazan asserts that there must have been at least sixty-six manuscripts, but many have been lost and the number of manuscripts made in the Middle Ages may have been even higher.8 The Gembloux manuscript (i.e. Sigebert’s chronicle and the continuations) were produced during the twelfth century and, especially after 1148, were copied and continued in a large number of historiographical works written in abbeys in the Flanders region and northern France, most notably Anchin, Affligem and Marchiennes.9 The passage on the Danish attack was written by the second anonymous continuator of Sigebert’s chronicle. This was a universal chronicle stretching from 381 to 1111, written by Sigebert in the Benedictine abbey of Gembloux in the diocese of Liège.10 When Sigebert died in 1112, the chronicle was continued by abbot Anselm of Gembloux, and his writings cover the years from 1111 until his death in 1135. While following the annalistic structure established by Sigebert, Anselm was more given to stylistic exercises and had a weakness for reporting catastrophes and wonders. He mostly treated events concerning Liège, Lotharingia, the Empire and the Holy Land, but reported little on France and nothing on England.11 After Anselm’s death, the two monks of Gembloux continued the work; the first briefly recorded events for 1136 and 1137, and the second wrote on the years 1138–48. This second continuator began his writings with a brief prologue in which he stated his intent to produce a terse account and to stick to the most important events, perhaps – as Chazan remarks – as a critique of Anselm’s more sensationalist style. Though the last three years, from 1146 to 1148, were devoted completely to describing the preparation of the Second Crusade, the entries covering the years 1138–45 conformed to the conventions of chronicle writing. They were remarkably well informed and were probably

7

8 9 10 11

mentioned by the Continuation cannot be the Norwegian raid, since the Continuation stopped in 1148. Sawyer, Scandinavians and the English, p. 4. MGH, Sigeberti Gemblacensis Chronographia, p. 284; Mireille Schmidt-Chazan, “Sigebert de Gembloux, le Lotharingien,” Publications de la Section historique de l’Institut g.-d. de Luxembourg 106 (1991), 33. Mireille Chazan, L’empire et l’histoire universelle. De Sigebert de Gembloux à Jean de SaintVictor (XIIe–XIVe siècle) (Paris, 1999), pp. 311–12. Mireille Chazan, “La nécessité de l’Empire de Sigebert de Gembloux à Jean de Saint-Victor,” Le Moyen Âge 106 (2000), 13 ; Chazan, L’empire, pp. 319–31. Chazan, L’empire, p. 25. Ibid., pp. 316–17.

30

Thomas K. Heebøll-Holm

written soon after the occurrence of the events. The continuator maintained the focus on Liège, Lotharingia, the Empire and the Holy Land but, in contrast to Anselm, he also paid much attention to events in England and France.12 The information in the second Continuation may well have been derived from day-to-day news reaching the abbey. Gembloux, which lies north-west of Namur, was located at the crossroads of northern Europe, and connections between England and the region comprising northern-eastern France and Flanders were legion in the twelfth-century. Information and rumors must have flowed constantly between the two countries, carried by, for instance, merchants, mercenaries and nobles.13 It is thus no surprise that the continuator was fairly well informed about English events, and in comparison with Anselm’s continuation, the most remarkable thing is that he paid so much attention to England. Less well studied are the twelfth-century connections between the Flanders region and Denmark that are evident in the Continuation. Denmark had for centuries been fairly well connected to Flanders and the surrounding principalities by trade. Furthermore, by 1082 Count Robert I of Flanders had married his daughter Adela to the Danish King Cnut IV, probably as part of an alliance with the purpose of invading England.14 In 1085 Cnut IV’s planned invasion of England was disturbed by domestic unrest and in 1086 he was murdered in Odense. Adela fled to Flanders with her son, Charles, where he was raised as a Frankish knight. In 1119 he succeeded his cousin, Baldwin VII, as count. Charles (the Good) ruled for eight years and Danes may well have frequented the court, for Galbert of Bruges recounts that Charles had a treacherous knight called Eric in his service. This name suggests a Danish origin.15 In 1127 Charles was murdered and Flanders descended into a brief but bloody war of succession. Amongst the several contenders for the comital dignity was Charles’s nephew, Arnold of Denmark (Arnold de Dacia), who ensconced himself in Saint-Omer with the support of the citizens, but who was defeated by William Clito.16 It is possible that this Arnold is identical with a Danish magnate called Riber Ulf, a magnate who in Denmark served Sven Grathe in the Danish Wars of Succession (1131–34 and 1146–57) against his rivals Cnut Magnussen and Valdemar.17 There thus seems to have been a presence of Danish 12 13 14

15 16

17

Ibid., p. 319. Eljas Oksanen, Flanders and the Anglo-Norman World, 1066–1216 (Cambridge, UK, 2012), pp. 88–89, 91. Oksanen, Flanders, p. 18; Jarl Gallén, “Knut den helige och Adela av Flandern. Europeiske kontakter och genealogiske konsekvenser,” in Studier i äldre historia tillägnade Herman Schück, ed., Robert Sandberg (Stockholm, 1985), pp. 51–54. Galbert of Bruges, De multro, traditione, et occisione gloriosi Karoli comitis Flandriarum, ed., Jeff Rider (Turnhout, 1994), pp. 31, 135. Galbert of Bruges, De multro, pp. 141, 144–45, 146, 147, 157; Walter of Thérouanne, Vita Karoli comitis Flandrie et Vita domni Ioannis Morinensis Episcopi, ed., Jeff Rider (Turnhout, 2006), p. 70; Gallén, “Knut den helige,” pp. 64–65. Michael H. Gelting, “En bispekarriere. Helias af Ribe, en flamlænding i 1100-tallets Danmark,” in Festskrift til Troels Dahlerup, eds. Aage Andersen et al. (Aarhus, 1985), pp. 6–7.



When the Lamb Attacked the Lion

31

nobles at the court of Charles. This connection was not one way, either. It is extremely likely that the former dean of St. Donatian, a certain Helias, had to flee to Denmark in the wake of the Flemish war of succession. He first appears in Denmark as a canon at the cathedral of Roskilde between 1135 and 1142, but soon Helias embarked on an illustrious career in the Danish church – perhaps due to the patronage of Riber Ulf or, more likely, the Sealand magnate Peter Bodilsen. In 1142 at the latest, he became bishop of Ribe, the westernmost seaport of Denmark, and while initially he was allied with Sven Grathe, when Valdemar I defeated Sven in 1157, Helias submitted to Valdemar and continued to act as the de facto leader of the Danish college of bishops until his death in 1162.18 Finally, a fragment of a medieval text recently discovered in the Danish National Archives mentions Saint Aldegunde, who was (and still is) mainly venerated in the Hainaut area. The document, of which we only have a fragment, was brought to Denmark seemingly as part of the veneration of the saint by Flemish prelates such as Helias.19 Thus, the Abbey of Gembloux was well positioned to receive news from both these countries, meaning that the continuator had ample sources at his disposal. England around 1138 When King Henry I of England died in December 1135, Count Stephen of Boulogne quickly crossed the Channel and within a few weeks was crowned king of England. However, he was by no means the only candidate for the throne and his reign soon became contested, most notably by Henry’s daughter Matilda. This conflict was due to Henry I’s handling of the succession in 1126. Henry’s lack of a legitimate son to inherit England and Normandy raised the question of succession and inheritance practices.20 When he designated his daughter Matilda to succeed him, the decision split the English court into two factions.21 At the time of Henry’s death in 1135 he was embroiled in a conflict with Matilda’s husband, Geoffrey of Anjou, and this fact, combined with Stephen’s many supporters in England, led the English to accept Stephen as king.22 While things were relatively calm for Stephen in the first two years of his reign, by 1138 threats appeared from every corner of the Anglo-Norman world. In the south, Matilda and Geoffrey of Anjou moved against Normandy. In the west, Henry’s illegitimate son, Robert of Gloucester, rebelled (probably in alliance with Matilda) and in the north, King David of Scotland invaded England, penetrating as far

18 19

20 21 22

Ibid., pp. 1–15. Steffen Harpsøe, “Contacts between Denmark and Flanders in the Eleventh and Twelfth Centuries. The Tiniest of Evidence,” in Denmark and Europe in the Middle Ages, c. 1000–1525. Essays in Honour of Professor Michael H. Gelting, eds. Kerstin Hundahl et al. (Farnham, 2014), pp. 77–90. Heather J. Tanner, Families, Friends and Allies (Leiden, 2004), p. 204. Ibid., p. 183. Crouch, Reign of King Stephen, pp. 30–35, Tanner, Families, pp. 191–94.

32

Thomas K. Heebøll-Holm

south as York.23 If the Continuation is to be believed, Stephen’s reign was also challenged in the east with a Danish assault on England. At this point it should be mentioned that it is possible that the continuator confused or conflated the Danish attack with the Scottish invasion of England, which resulted in a Scottish defeat at the Battle of the Standard, 22 August 1138. However, in what follows I shall argue that since the continuator was rather accurate in his other passages concerning England, a Danish attack was probable, and perhaps even coordinated with the Scots. It is likely that the Danes, if they indeed attacked England, would have aimed for the Humber estuary, and ultimately eastern England around York. This was where King Sven Forkbeard landed in 1013, and in 1069 and 1075 the Danish armies who sought to assist the Anglo-Saxon uprisings against the Normans did the same. All these landings had the explicit purpose of securing York, the old capital of the Danelaw.24 At least in the eleventh century, the Danes received a warm welcome from the local magnates, and while by 1138 more than sixty years had passed since a Danish army had set foot on England, it was still likely to be the most receptive part of England to a Danish invasion. Based on the above-mentioned campaigns, the Danish attack is most likely to have occurred in the period from July to September, which would make the attack coincide with the Scottish invasion of Yorkshire in July and August. It is thus possible that the Danish attack was somehow coordinated with the Scots, and perhaps also with disgruntled northern Anglo-Norman magnates, some of whom had joined the Scottish invasion army.25 The notion of a Danish attack happening in the York area is further suggested by a discussion of the possible Danish sympathies in the old Danelaw areas of Yorkshire, East Anglia and Lincolnshire, as represented in the treatment of Danes in Geffrei Gaimar’s Estoire des Engleis. The patrons of the Estoire were Constance and Ralph fitz Gilbert, feeholders under the great Lincolnshire magnate Walter de Gant (Ghent), who had strong Flemish connections. According to Ian Short, the Estoire was written between March 1136 and April 1137 and the intended audience was the minor Anglo-Norman aristocracy of Lincolnshire, who carried the burden of ruling in a former part of the Danelaw with a presumably well-kept and somewhat positive historical memory of the Danish rule.26 Should Short be right in his conclusions this would mean that, in

23

24 25

26

Crouch, Reign of King Stephen, pp. 72–83; Paul Dalton, “The Date of Geoffrey Gaimar’s Estoire des Engleis, The Connections of his Patrons, and the Politics of Stephen’s Reign,” The Chaucer Review 42 (2007), 27. The Anglo-Saxon Chronicle, trans. G. N. Garmonsway (London, 1972), pp. 143, 204–5. Crouch, Reign of King Stephen, pp. 80–81; Beeler, Warfare in England, p. 85; Richard of Hexham, Historia Ricardi, prioris ecclesiæ Haugustaldensis, De gestis regis Stephani, in Chronicles of the Reigns of Stephen, Henry II, and Richard I, 4 vols., ed. Richard Howlett (Wiesbaden, 1964), 3: 152. Geffrei Gaimar, Geffrei Gaimar. Estoire des Engleis/History of the English (hereafter Gaimar), ed. and trans. Ian Short (Oxford, 2009), pp. xi, xi, xii.



When the Lamb Attacked the Lion

33

1138, Danish attackers would probably find this region the most welcoming in England to a Danish king. Short further argues that: the persistence of a Scandinavian cultural substratum in the area, even into the 1130s, can be assumed with a high degree of plausibility. The regional culture had been facilitated by the mutual intelligibility between the Norse and English languages that had long since been a feature of the area’s multiculturalism. A significant Norse influence on the language of the Middle English author Orm from Lincolnshire, for example, is one tangible argument in favour of some measure of Scandinavian cultural continuity even into the second half of the twelfth century. Another […] is the important and privileged place occupied by the Danes, and by narratives of Danish interest and origin, in Gaimar’s Estoire des Engleis, the provenance, patronage, and audience which can all be ascribed, beyond reasonable doubt, to Lincolnshire.27

Thus, “by enhancing his history’s Scandinavian dimension, he was targeting a specifically Lincolnshire audience.”28 However, Short’s conclusions are contested. Paul Dalton questioned Short’s dating of the work, and argues for a date circa 1141–50, based on external evidence.29 However, such a redating by no means rule out Danish sympathies. In fact, Dalton’s interpretation of Gaimar’s account of Buern Bucecarle’s rebellion against the felonious King Osbert as a parallel to Robert of Gloucester’s struggle with King Stephen could indicate that the Danish attack in 1138 was instigated by Gloucester through the patrons of Gaimar, the fitz Gilberts, as Buern brought in Danish troops to kill Osbert and seize York.30 Should he be correct in this parallel, the meaning could be taken further to indicate a clandestine invitation to England of the Danish king, whose attack was omitted in later sources because it was a failure, and in any case could be regarded as treason. According to Brian Levy, Gaimar’s portrayal of the Danes was decidedly hostile. Levy stresses the negative image of the Danes in the Estoire – for instance, that they are referred to as fel paien. Even when the tone changes in the treatment of Cnut the Great, who was a Christian, he is still (according to Levy) portrayed as a false messiah who is not about to bring reconciliation between the Danes and the English. Rather, under Cnut the Danes had made thralls of the English. After his death, the Danes are again cast as villains. William the Conqueror is praised for slaughtering Danes and Norwegians in York, and Harold Godwinson is portrayed a villain, due to the blending of Anglo-Saxon and Danish blood in his family. In contrast, Hereward the Wake is portrayed as a true English hero who is righteous in his actions against the Normans.31 27 28 29

30 31

Ibid., pp. ix–x. Ibid., p. xxix. Brian J. Levy, “The Image of the Viking in Anglo-Norman Literature,” in Scandinavia and Europe 800–1350. Contact, Conflict, and Coexistence, eds. Jonathan Adams and Katherine Holman (Turnhout, 2004), pp. 269–88; Dalton, “Date of Geoffrey Gaimar’s Estoire,” pp. 23–47. Ibid., pp. 36–37. Levy, “Image of the Viking,” pp. 274–78.

34

Thomas K. Heebøll-Holm

Short is aware of these tendencies in Gaimar’s chronicle. However, he explains them in a radically different manner. Short stresses that though the Danes are indeed sometimes described as evil, pagan raiders, they are just as often described as heroes, and Short rejects an inherent antagonism in Gaimar’s work between Anglo-Saxons and Danes. He writes: “Gaimar’s pro-Danish sympathies are deliberately, repeatedly, and consistently articulated from the start of his chronicle when his Haveloc saga ‘reworks the Danish invasions into a success story of intermarriage and international alliance’.”32 Rather, Short detects a pro-Danish tendency before the Norman Conquest, which after 1066 becomes a pro-English stance. He ascribes this stance to the expectations of Gaimar’s audience, who favor a conciliatory history of the fusion of Danes, Anglo-Saxons and Normans into an English people, triumphing over a vehement and enduring ethnic antagonism. In this regard, William Rufus figures as the first king of England who is English, rather than Norman, Anglo-Saxon or Danish. However, the story is not anti-Norman either.33 When it comes to persons like Hereward the Wake, then, in Short’s interpretation, this is less a story of an Anglo-Saxon/Danish patriot against Norman usurpers and more a story of the magnate’s right to oppose the tyrannical rule of kings. This story, Short surmises, must have appealed to the local Lincolnshire magnates, and indeed he asserts that the baronial class in the Estoire is always cast in a favorable light.34 Short’s interpretation of the audience and the focus in the Estoire is also affirmed by Scott Kleinmann, who, on the status of the Danes in the Estoire writes: “Geoffrey of Monmouth’s story of British sovereignty over Denmark [under King Arthur] is reversed, and by implication, the Anglo-Normans are the heirs of Danish rights. Such a transformation of the story may well have appealed to an Anglo-Norman audience trying to establish hereditary rights in an East Anglian regional context.”35 Henry Bainton also supports this line of interpretation. He argues that the Estoire inscribes itself into a domestic English struggle for historiographical hegemony between the West-Saxon bid for dominance over Anglo-Saxon history (as expressed in the Anglo-Saxon Chronicle, hereafter ASC) and other regions of England, such as Northumbria, with their own separate histories that incorporated Danes and Northumbrians against a (West-)Saxon dominance.36 In the Estoire, the Danes were portrayed as almost as indigenous to England as the Britons and Saxons, though the Danes here have a historical enmity towards the Britons, due to Arthur’s wars against them. Furthermore, according to Gaimar, the Danes settled England before the Saxons.

32 33 34 35 36

Gaimar, p. xlviii. Ibid., pp. xliv–xlv, xlix. Ibid., pp. xlvii–viii. Scott Kleinmann, “The Legend of Havelok the Dane and the Historiography of East Anglia,” Studies in Philology 100 (2003), 264. Henry Bainton, “Translating the ‘English’ Past: Cultural Identity in the Estoire des Engleis,” in Language and Culture in Medieval Britain. The French of England, c. 1100–c.1500, ed. Jocelyn Wogan-Browne et al. (York, 2009), pp. 186–87.



When the Lamb Attacked the Lion

35

Hence the Saxons had historically no more right to England than the Danes (or for that matter the Normans). According to Bainton: “the post-Conquest severing of the literary language and historical mythology of pre-Conquest England from its West Saxon guardians encouraged regional diversity of postConquest England to be articulated and explored.”37 Thus, argues Bainton, the Danes are not portrayed as God’s punishment of the Anglo-Saxons, as is usual in the ASC. Rather, their actions are part of regular seigneurial disputes. He writes that “by making the Danes allies of the Northumbrians rather than simply the enemies of the English, Gaimar gives both the Danes and the Northumbrian aristocracy a distinct role in the history of a region quite apart from the more familiar story in which the invasion of the Danes was merely a prelude to the West Saxons fighting them off for all of Christian England.”38 In this regard, the story of Haveloc, who is king of Lincolnshire, Norfolk and Denmark, mirrors Cnut the Great’s empire and makes the distinction between Britain and Denmark less sharp, thus affirming a distinct identity of the East Anglian/Lincolnshire area in contrast to the “official” West-Saxon history of England in the ASC.39 It thus seems clear that most historians consider the Estoire to be a story focused on a Lincolnshire audience. It is a conciliatory history of the different peoples of the region with a somewhat positive memory of the Danes even though the aristocracy of Lincolnshire in the 1130s must have been of at least largely – if not wholly – of French origin writ large (i.e. Normans, Bretons, Flemings etc.). Nevertheless, their independent-mindedness and adoption of the local history and legacy may well have made some of them sympathetic to a Danish king in opposition to the contested King Stephen. Indeed, as Mia Münster-Swendsen has pointed out, sometime before 1140, Hugh the Chanter expressed fear that the archbishopric of York – if it should become independent of Canterbury’s primacy – would make a Dane, a Norwegian or a Scot king of England.40 While this may well have been nothing more than unfounded jitters, it still signaled a doubt about the loyalty and proclivities of the northern English. Furthermore, if the king of Denmark took it seriously, it would have been a pertinent threat. Thus, to my knowledge there is nothing that proves that it did not take place. Denmark around 1138 Denmark in the 1130s, like England, was plagued by a war of succession which would not be settled until the 1150s. When Erik I (Evergood) died in 1103 while 37 38 39 40

Ibid., p. 181. Ibid., p. 183. Ibid., p. 187. Mia Münster-Swendsen, “Educating the Danes: Anglo-Danish Connections in the Formative Period of the Danish Church c. 1000–1150,” in Friendship and Social Networks in Scandinavia, c. 1000–1800, ed. Jon Vidar Sigurdsson and Thomas Småberg (Turnhout, 2013), p. 164.

36

Thomas K. Heebøll-Holm

on crusade to the Holy Land, his brother Nicolas (Niels) succeeded him in 1104, though Erik had several sons.41 Nicolas’s rule was seemingly uncontested in the first decades of the twelfth century, but by the late 1120s trouble and unrest were brewing over the succession. In 1131, Erik I’s son, Cnut Lavard (father of King Valdemar I), was murdered by Nicolas’s son Magnus. This caused the eruption of a bloody internecine war of succession in Denmark, where Nicolas’s nephew Erik led a rebellion against the king. This did not go well, however, and Erik was eventually pushed back to Scania. In 1134, Magnus and Nicolas were preparing to deliver the final blow against Erik, but while they were disembarking at Fotevik in Scania, they were surprised and slaughtered on the beach by Erik’s forces. In this battle, Magnus was killed, in addition to one Swedish and four Danish bishops. Although Nicolas managed to escape, he was murdered a few weeks later in Schleswig. Erik II (called the Memorable) was crowned the same year, but his reign was contested by his brother Harald Kesja, who had sided with Nicolas. Thus, in 1135 Erik murdered Harald and at least eight of his sons. Erik seemingly also mounted a successful raid against the pagan island of Rügen in the Baltic, where he forced the Rugians to convert, and in 1137 Erik, in support of his ally, Harald Gille – a contender to the throne of Norway – plundered Oslo.42 In 1136 or 1137 Bishop Eskil of Roskilde (ca. 1100–81/82) and a coalition of Sealand magnates led by Peter Bodilsen rebelled against Erik II. They were defeated, however, and Bishop Eskil bought peace while Peter Bodilsen was temporarily forced into exile.43 Nevertheless, dissatisfaction with Erik continued. In 1137 Erik was murdered by a magnate called Plouh or Plogus (i.e. Plough) at a thing in Ribe. Erik’s nephew, Erik III (Lamb), was chosen by the magnates as Erik II’s successor.44 Erik III’s first year as king saw trouble looming, but no open warfare erupted until 1139, when Oluf, Erik’s cousin and the last surviving son of Harald Kesja, rebelled. Oluf was finally defeated and killed, presumably in 1141. Erik ruled until 1146, when he abdicated and became a monk in St. Cnut’s Abbey in Odense, only to die the same year. This internecine war did not occur as an isolated Danish event. Both Emperor Lothar III of Supplinburg and the archbishopric of Hamburg-Bremen had strong interests in Denmark and interfered in Danish domestic affairs. Cnut Lavard was lord of Schleswig, but with German assistance he also came to rule the Slavonic 41

42 43

44

Lars Hermanson, “How to Legitimate Rebellion and Condemn Usurpation of the Crown: Discourses of Fidelity and Treason in the Gesta Danorum of Saxo Grammaticus,” in Disputing Strategies in Medieval Scandinavia, ed. Kim Esmark et al. (Leiden, 2013), p. 107. See Figure 1 for a genealogy of the kings of Denmark. Skyum-Nielsen, Kvinde, pp. 68–83. Michael H. Gelting, “Da Eskil ville være ærkebiskop af Roskilde,” in Ett annat 1100-tal: Individ, kollektiv och kulturella mönster i medeltidens Danmark, ed. Peter Carelli et al. (Gothenburg/ Stockholm, 2004), pp. 192–94. Gelting, “Da Eskil,” p. 195; Scriptores Minores Historiæ Danicæ (hereafter SM), 2 vols., ed. Martin C. Gertz (Copenhagen 1970/ 1917–18), I:31 (“Plouh” quod Latine “aratum” sonat); Saxo Grammaticus, Gesta Danorum- Danmarkshistorien, 2 vols., ed. et trans. Karsten FriisJensen and Peter Zeeberg (Copenhagen, 2005), 2:14.1.13 (“Plogus”).



When the Lamb Attacked the Lion

37

Obodrites. This German assistance entailed homage to Lothar. When Cnut was murdered in 1131, Lothar gathered an army and moved against Nicolas and Magnus to avenge his vassal and assist Lothar’s ally, Erik. Nicolas bought peace from the emperor and Magnus had to swear an oath of hominium to the king, who then abandoned the alliance with Erik.45 While Lothar was fighting in Italy, between 1131 and 1133 Nicolas and Magnus seem to have persecuted Germans in Denmark, but when Lothar emerged victorious from his Italian campaign, Magnus had to go to Halberstadt in Easter 1134. Here he bought peace, gave hostages, did homage to the emperor and made the future Danish kings subservient to the emperors. Lothar reciprocated by crowning Magnus as king. In 1135, Erik II in all likelihood did homage to the emperor on the same conditions as Magnus had done.46 Lothar’s attempts to gain dominance over Denmark went hand in hand with the ambitions of the archbishopric of Hamburg-Bremen to regain primacy over the Nordic church, which it had lost in 1103. After the papal schism of 1130, Adalbert of Hamburg-Bremen and Lothar eventually supported Innocent II against Anacletus II, but in return they managed to pressure Innocent into suspending Lund’s primacy over the Nordic churches.47 This caused much grief in the Danish church and an already existing contact with English churches was probably reinforced. Despite the Danish kings’ aggressions against England in the eleventh century – or perhaps because of their persistence in the claim to the English throne – the Danish church, and indeed kings, maintained an amicable relationship with some English abbeys. Thus Paul Gazzoli has proposed that in order for Erik I to have his brother Cnut IV canonized he visited Durham, and perhaps also Evesham, shortly after his accession to power in 1095, and that Erik was granted confraternity in Durham, an act which required his personal presence. Durham may also have had fond memories of the Danish rule, as Cnut the Great had made lavish donations to that community.48 Furthermore, monks from Evesham were sent to Odense (the centre of the cult of Cnut IV the Holy) to found a daughter house.49 Moreover, Evesham was connected to the Benedictine chapter in Worcester, and it was in confraternity with the Benedictines in Gloucester, Pershore, Chertsey, Bath, Whitby and St. Mary’s, York. It is possible that Odense was part of this network. In any case the confraternity between Odense and Evesham was reconfirmed between 1135 and 1139.50 Thus the cult of Cnut and its writings were supported by English monks, despite the 45 46 47 48 49

50

Gelting, “Da Eskil,” p. 185. Ibid., pp. 188–89. Skyum-Nielsen, Kvinde, pp. 74–75. Münster-Swendsen, “Educating the Danes,” p. 159. Paul Gazzoli, “Anglo-Danish Connections and the Origins of the Cult of Knud,” Journal of the North Atlantic 4 (2013), 72, Michael H. Gelting, “Two Early Twelfth-Century Views of Denmark’s Christian Past: Ailnoth and the Anonymous of Roskilde,” in Historical Narratives and Christian Identity on a European Periphery, ed. Ildar H. Garipzanov (Turnhout, 2011), p. 36. Münster-Swendsen, “Educating the Danes,” p. 161.

38

Thomas K. Heebøll-Holm

fact Cnut had planned to invade England. The connection between the cult of St. Cnut and England was further strengthened by the fact that Cnut IV’s Vita, the Gesta Swenomagni regis et filiorum eius et passio gloriossimi Canuti regis et martyris from around 1111/12 was written by the English monk Ailnoth, who was originally from Canterbury, but who from 1086–89 came to reside in Odense. Michael Gelting suggests that the manifestly anti-Norman Ailnoth, who may have been a Danish agent in England and collaborator in Cnut IV’s planned invasion of England, was a political refugee in Denmark. While this cannot be proven with any certainly, the exile of two high-ranking Anglo-Saxon ecclesiastics to Denmark after the failed Danish invasions of 1169–70 makes the notion at least plausible.51 Furthermore, Michael Gelting’s studies of the Chronicle of Roskilde, finished circa 1138, show that Roskilde must have been in possession of a very early copy of Henry of Huntingdon’s chronicle (finished in the version that began circulating circa 1133), which points to close connections between English and Danish monastic milieus.52 Finally, it would seem that until the mid-twelfth century many Danes seeking an education went to England, but unfortunately the source material does not permit more than an educated guess at the number of Danes in English monasteries.53 The German pressure thus seems to have rekindled Danish interests in England, and in 1137 and 1138 Denmark was freed from the German iron grip by two events. In December 1137 Lothar died and the Empire was plunged into a war of succession, and in January of the following year the anti-pope Anacletus II died. Innocent II, now sole pope and freed from imperial pressure, was favorably disposed towards reversing the initiative to put the archdiocese of Lund under Hamburg-Bremen, and by 1144 any claims to primacy by HamburgBremen were put to rest.54 It was in the first year of Erik’s reign that the attack would have taken place, and it would seem to have been a sensible thing after a lengthy domestic war. Such an attack on a foreign country could have served to gather and reconcile the various factions of royal contenders and their magnate allies against a common external foe. Thereby Erik III could create cohesion and support for himself as the new king by the dispersal of spoils to his supporters. Indeed, that seems to have been what Erik II had done by attacking Rügen shortly after his accession to power, and it was what Valdemar I did soon after his ascension to the throne in 1157, by attacking the Slav Wends in the western Baltics. It furthermore makes sense that Erik III, in light of the turmoil in England, 51 52

53 54

Michael H. Gelting, “The Kingdom of Denmark,” in Christianization and the Rise of Christian Monarchy, ed. Nora Berend (Cambridge, UK, 2007), p. 105. Gelting, “Da Eskil,” p. 182; Gelting, “Two Early Twelfth-Century Views,” pp. 39–41; Michael H. Gelting, “Henry of Huntingdon, the Chronicle of Roskilde, and the English Connection in Twelfth-Century Denmark,” in The Writing of History in Scandinavia and its European Context, 1000–1225, ed. Thomas K. Heebøll-Holm et al. (Durham, forthcoming); Gelting, “Da Eskil,” p. 198. Münster-Swendsen, “Educating the Danes,” pp. 159, 162–63. Gelting, “Da Eskil,” pp. 196–97, 217; Gelting, “Two Early Twelfth-Century Views,” pp. 50–51.



When the Lamb Attacked the Lion

39

would have turned his attention to the West. Even though Duke William had been crowned king of the English in 1066, the Danes still claimed that they had the greater right to throne. Thus, in 1069/70 and in 1075 the Normans had had to defeat Danish armies called to northern England to help the Anglo-Saxons, and Adam of Bremen reported that Edward the Confessor had chosen King Sven Estridsen as his successor. This piece of information could very well have derived from Sven himself, who was one of Adam’s sources. It thus signaled an understanding on the part of the Danes that they still had legitimate claims on England. Finally, in 1085 a Danish invasion of England was called off because of the defeat of the papal party in Germany. The anti-king Herman of Salm, the archbishop of Magdeburg and other magnates seemingly fled to Denmark, forcing Cnut to guard his southern boundary,55 but the claim was by no means forgotten, as will be shown. To sum up, conditions in Denmark were favorable for an assault as several rivals fought Stephen for the throne. Why should the Danish king not be one of them? The chronology of Danish history in the 1130s certainly permits of a failed attack on England in 1138, followed by a rebellion under the contender Oluf, bolstered by Erik’s defeat in England. Thus, Denmark had by no means ended contact and connections with England after Cnut IV’s planned invasion of England in 1085. In this light it would make good sense for a new king, eager to rally the warring magnate factions and to free his country from imperial dominance, to turn west and try his luck in the English war of succession. Angevin England In 1154 Henry II succeeded Stephen as king of England, and while historiography has often portrayed him as a most victorious king who aggressively defended and expanded the Angevin empire, the fact remains that his rule was insecure.56 Thus the 1160s were marked by the conflict with Thomas Becket; in 1173–74 Henry’s sons, his wife, Eleanor of Aquitaine, and a baronial coalition launched a massive rebellion; and in the year of his death, 1189, he was embroiled in conflict with his son Richard. Accordingly, he took various initiatives to legitimize and strengthen his rule. One was the coronation of his son, Henry the Younger, as king of England on 22 May 1170, and again in August 1172. This was an unusual practice in England, seemingly copied from Capetian France, and was somewhat controversial.57 Another action was to stress how he, 55

56 57

Adam of Bremen, Gesta Hammaburgensis Ecclesiae Pontificum, 3rd ed., ed. Bernhard Schmeidler MGH SS in usum scholarum 2 (Hannover, 1917), p. 152; Niels Lund, “Knuts des Heiligen beabsichtiger Zug nach England im Jahre 1085,” in Mare Balticum. Beiträge zur Geschichte des Ostseeraums in Mittelalter und Neuzeit, ed. Werner Paravicini (Sigmaringen, 1992), pp. 101–10; Skyum-Nielsen, Kvinde, pp. 10–11, Gelting, “Two Early Twelfth-Century Views,” p. 35. Charity Urbanski, Writing History for the King. Henry II and the Politics of Vernacular Historiography (Ithaca, NY, 2013), p. 40. W. L. Warren, Henry II (London, 1973), pp. 110–12.

40

Thomas K. Heebøll-Holm

Henry, in his person united the lineages of the Anglo-Saxon kings and the dukes of Normandy. The culmination of this effort was the canonization of Edward the Confessor in 1161. Henry had thus acquired a holy stirps regia.58 Another of his initiatives was to strengthen his claim through a program of historical writing aimed at exalting the royal lineages of the Anglo-Norman kings, and thereby his own rule as the descendant of these illustrious kings. In this effort, Henry II emphasized his maternal connections, i.e. to the Anglo-Saxon and Norman rulers, over his paternal connections to the counts of Anjou. Thus, in 1163 Aelred of Rievaulx wrote that Henry II was the cornerstone uniting the walls that were the English and the Norman peoples. Based on Edward the Confessor’s alleged prophesy, Henry was, according to Aelred, supposed to be the one uniting the split tree of Edward’s vision. Aelred’s genealogy of the kings of England showed how Henry descended from Saint Margaret of Scotland, the pious king Alfred the Great and Edward the Confessor. The Nun of Barking elaborated on this: since Henry was the son of the Empress Matilda, who in turn was the daughter of Henry I and Matilda of Scotland whose great grandfather was King Edmund (1016), half-brother of Edward the Confessor, Henry was descended from the Anglo-Saxon kings, and in him was united the quasi-holy ducal family of Normandy represented by Count Robert, Duke Richard, William the Conqueror, and Emma, the mother of Edward the Confessor. There was thus blatant propaganda in favor of Henry, which was much needed because his Angevin ancestry caused problems regarding his legitimacy to be on the throne, as was pointed out by Herbert of Bosham.59 As part of this effort, the Danes and the Dano-Angles such as the Godwinsons were airbrushed away or cast in a very unfavorable light. Such treatment of the Danes was not a novel, Angevin invention. In Osbert of Clare’s Life of King Edward from 1138, the role of Godwin and his sons in Edward’s government was simply rejected and the rights of the Norman dukes as Edward’s legitimate successors were emphasized.60 In 1163 Aelred rewrote Osbert’s Life of King Edward in the Historia regum Anglorum, and while he admitted a role for Godwin and his son in Edward the Confessor’s government, Aelred produced a very negative image of them.61 This was continued by Denis Piramus, who in his Vie de seint Edmound le rei from around 1170 produced an utterly negative image of the Danes.62 Another part of this program was Henry’s effort to have a history exalting his lineage produced in French. Thus he commissioned Wace (and later Benoît de Sainte-Maure) to write the glorious history of the Norman dukes and Anglo-

58 59 60 61 62

Urbanski, Writing History, pp. 72–73. Martin Aurell, L’empire des Plantagênet 1154–1224 (Paris, 2003), pp. 149–50; Urbanski, Writing History, pp. 75–77, 79. Aurell, L’Empire, p. 149; Frank Barlow, ed. and trans., The Life of King Edward, 2nd ed. (Oxford, 1992), p. xxv. Barlow, Life of King Edward, pp. xxxvii–xxxviii. Levy, “Image of the Viking,” pp. 279–80.



When the Lamb Attacked the Lion

41

Norman kings, the Roman de Rou.63 This work was probably intended to be read out at Henry’s court and was to explain how the Normans legitimately came to power in England, therefore confirming Henry II as a right and legitimate king of England. The reason for the composition of this work was probably as Glyn Burgess suggests: “Even in the 1160s, when Wace began his work, some English lords were still refusing to accept the right of William the Conqueror and his successors to the throne of England; in addition, some Norman lords, who had grown accustomed to the freedom afforded them by the lax rule of Henry’s predecessor, Stephen of Blois, were reluctant to offer Henry their full allegiance.”64 In the Roman de Rou Wace supported the view that the Danes had no right to the English throne. He wrote how duke Robert I had demanded that Cnut the Great surrender England to the rightful heirs, the sons of Æthelred, Alfred and Edward, who had taken refuge with the duke in Normandy.65 However, Wace’s portrait of the Danes (and other Scandinavians) is ambiguous. They are mighty in battle when they assist the Normans against the French, but felonious when they refuse the sons of Æthelred their birthright. Wace was thus in accordance with chroniclers such as Aelred. Yet, while Wace did support the somewhat anti-Danish discourse favored by the Anglo-Norman kings, he opposed Henry II’s manipulation of the Anglo-Norman history. Thus, Wace cast doubt on Duke William’s right to invade England. According to Wace, William, on his deathbed, stated that he had conquered England wrongfully and without right, and that he would beg the archbishop of Canterbury to grant his son William Rufus the English Crown, the king-duke himself having no right to invest his son with regal power.66 Thus both the Danes and the Normans were felonious conquerors and the record was not set straight until an English archbishop granted the crown to William Rufus. Wace’s work, which was begun in the early 1160s, was left unfinished in the middle of the 1170s because, as Wace noted, Benoît de Sainte-Maure had been commissioned instead of him to write a glorious history of the Norman dukes and kings of England up until Henry II. Several causes for Wace’s abandonment of his work have been suggested, but almost all of them assume that Wace’s history did not meet Henry’s political needs and expectations.67 For instance, Wace refused to exalt the sacrality of the Norman dukes. He also 63 64 65

66 67

Karen M. Broadhurst, “Henry II of England and Eleanor of Aquitaine: Patrons of Literature in French?” Viator 27 (1996), 58–59. Wace, The History of the Norman People. Wace’s Roman de Rou, trans. Glyn S. Burgess (Woodbridge, 2004), p. xi. Wace, Le Roman de Rou de Wace, 2 vols., ed. A. J. Holden (Paris, 1970), 1:262–63, ll. 2730–44: “Le duc out od sei Auvered/ e Ewart, les dous fiz Alred;/ d’Engleterre esteient jeté/ e a grant tort desherité./ Al rei Kenut, ki out lur mere/ e teneit l’erité lur pere,/ manda li ducs par sun message/ k’il lur rendist lur eritage;/ grant piece aveit la terre eüe,/ bien lur deüst estre rendue./ Kenut dist ke rien nen av(e)reient/ se par force nel conquerreient;/ e li ducs out grant marrement/ qu’il n’out respundu autrement.” Wace, Le Roman de Rou, 2:225–26, ll. 9133–54. Burgess, History of the Norman People, pp. xii–xiii.

42

Thomas K. Heebøll-Holm

was not an unrestrained admirer of William the Conqueror, whose right to the English throne over Harold Godwinson was left somewhat ambiguous in Wace’s work. Furthermore, Wace insinuated that William Curthose had more right to the English throne than had Henry I, Henry II’s grandfather.68 As Jean BlackerKnight writes: “Wace saw Robert [Curthose] as the rightful heir to the throne, according to the custom of primogeniture. Wace portrays Henry [I] as a greedy usurper who erred not only by setting himself upon the throne of England, but also by forcing his illegitimate claim to Norman lands.”69 And: This interpretation of the conflict between Henry and Robert might not have sat well with Wace’s patron for the following reasons: first, Henry I, grandfather of Henry II, is seen in a negative light; second, primogeniture was deemed paramount to the uniting of England and Normandy by a legitimate son already enthroned; and third, the issue of usurpation and land-rights was a particularly sensitive one at the time Wace was writing, because of Henry’s ongoing disputes with his own sons.70

Furthermore, Jean-Guy Gouttebroze has argued that Wace probably was a partisan of Thomas Becket’s and on grounds of religious conviction refused to recognize Henry’s and his lineage’s sacrality as rois thaumaturges. Wace thus opposed Henry’s effort to gain control over the English church through his sponsorship of his sacred ancestors, foremost amongst these Edward the Confessor, whom Wace portrayed as a strictly secular lord, a good but not especially pious or chaste man.71 Henry II did not order this propaganda just to legitimize his claim to England in the eyes of the French and Anglo-Norman nobility of his empire. The propaganda also served to delegitimize the Danish claim to the English throne. For the French rulers of England – be it Stephen or Henry II – with a hotly contested claim to the English throne, there was every reason to silence and suppress any Danish claims to the throne – just as had been attempted with Godwin and his son. Accordingly, if the attack in 1138 did take place, it is not surprising that no one in England wrote about it. Whatever antagonisms there were between Stephen and the Angevins, the French aristocrats ruling England could agree on one thing: nobody wanted a Danish interference in this French family matter. Valdemarian Denmark King Valdemar I (1154/57–1182) ascended the Danish throne in 1157 after twenty-six years of intermittent war of succession and, like Henry II’s, his rule was also contested by members of rival royal lineages and their aristocratic supporters. Thus Valdemar, his son Cnut VI and their allies in the aristocratic 68 69 70 71

Aurell, L’empire des Plantagênet, pp. 152–53. Jean Blacker-Knight, “Wace’s Craft and his Audience: Historical Truth, Bias, and the Patronage in the Roman de Rou,” Kentucky Romance Quarterly 31 (1984), 358. Ibid., p. 359. Jean-Guy Gouttebroze, “Pourquoi congédier un historiographe, Henri II Plantagenêt et Wace (1155–1174),” Romania 112 (1991), 299, 304–7.



When the Lamb Attacked the Lion

43

Hvide kin group (hereafter the Valdemarians) initiated a number of initiatives to stabilize and legitimize their rule. Like Henry II, Valdemar had his son crowned king while he was alive, and by having his father, Cnut Lavard, canonized he obtained a holy stirps regia. This was effected at a ceremony in the newly rebuilt church of St. Benedict in Ringsted at Whitsun 1170. Here Cnut Lavard’s sainthood was formally celebrated and the archbishop of Lund crowned Valdemar’s seven-year-old son, Cnut, to whom the magnates some years earlier had sworn oaths of fealty.72 As in England, this initiative inspired by foreign practice was somewhat controversial and it seems to have caused resentment amongst some Danish magnates. It thus indirectly gave the Valdemarians just cause for suspicions of treason and the eradication of these dissidents. This process had in fact been under way for some time by 1170.73 The more conspicuous move, however, was the Valdemarians’ slow but deliberate purge of royal contenders. In 1162, Valdemar crushed a rebellion led by Archbishop Eskil of Lund, leader of the powerful Trund kin group. This led Eskil to go into voluntary exile to Clairvaux in France, from where he returned in 1168 after he had been reconciled with Valdemar. Meanwhile, in 1167 Valdemar’s kinsman Buris Henriksen was imprisoned in Søborg Castle (Sealand) because he allegedly had conspired with the Norwegians to kill Valdemar, and, according to the Annals of Rüde, Buris was blinded and castrated.74 In 1176 Valdemar uncovered another conspiracy against him, led by Magnus Eriksen, son of Erik III (Lamb), the brothers Karl and Cnut Karlsen, descendants of Cnut IV (the Holy), and the Trund kin group under Archbishop Eskil. Magnus was pardoned in exchange for confessing his guilt and revealing the names of the conspirators. The conspirators were exiled and the Trund kin group’s powerbase was shattered. In 1177 Eskil resigned his office, most probably having been compelled to do so, and Valdemar’s right-hand man, Absalon, bishop of Roskilde, became archbishop. However, by 1178 a new conspiracy led by Magnus was uncovered; he stood trial and was sentenced to imprisonment in Søborg. The same year, Valdemar crushed a rebellion led by the Karlsen brothers; Cnut was imprisoned with Magnus in Søborg, while Karl was killed by low-born peasants.75 These measures cemented the power of Valdemar and the Hvide kin group, yet, when Valdemar died in 1182, Archbishop Absalon had to crush a Scanian rebellion led by the royal contender, Harald Skreng, allegedly a grandson of Harald Kesja,76 and ten years later, in 1192, bishop Valdemar of Schleswig (son of Cnut Magnussen) rebelled 72

73

74 75 76

Hermanson, “How to Legitimate Rebellion,” p. 124; Karsten Friis-Jensen, “In the Presence of the Dead. Saint Canute the Duke in Saxo Grammaticus’s Gesta Danorum,” in The Making of Christian Myths in the Periphery of Latin Christendom (c. 1000–1300), ed. Lars B. Mortensen (Copenhagen, 2006), p. 196. John Danstrup, “Træk af den politiske kamp 1131–82,” in Festskrift til Erik Arup, eds. Astrid Friis and Albert Olsen (Copenhagen, 1946), pp. 80–85; Hermanson, “How to Legitimate Rebellion,” pp. 108–9. Hermanson, “How to Legitimate Rebellion,” p. 126. Ibid., pp. 127–32. Saxo, Gesta, 16.1.3–8.

44

Thomas K. Heebøll-Holm

against Cnut VI. He was defeated, however, and was imprisoned in Søborg from 1192 until 1206.77 Thus the reign of the Valdemarians was continually challenged, but the seriousness of the revolts and the complicated measures taken by the Valdemarians were somewhat obscured by another of the Valdemarians’ initiatives that closely resembled the actions of Henry II, namely royal attempts to control the historiography. As Lars Hermanson puts it: “In order to avoid painting Valdemar as a bloodthirsty avenger and a traitor to his own kin, Saxo [Grammaticus] depicts the king’s pruning of the royal tree as a judicial process. Thus Valdemar’s blood-relatives are portrayed as usurpers of the crown who commit treason against the kingdom and treachery against a king who has divine protection.”78 In what follows I pay close attention to three aspects of this historiography, namely the portrayal of King Erik III, the “crusadification” of Danish warfare and the Danish claims to the English throne. The earliest source to treat Erik III (Lamb) was the Chronicle of Roskilde, finished circa 1138. It described Erik as a man more plain than what was fitting for a king, and in all things fickle and duplicitous. It also accused him of causing the destruction of the kingship and the priesthood.79 This negative characterization of Erik was probably due to a quarrel between the king and Eskil, at that time bishop of Roskilde, over who was to be archbishop of Lund and a possible transfer of the archbishopric from Lund to Roskilde.80 Neither the Chronicle of Roskilde nor any other later Danish sources (many of which drew on the Chronicle of Roskilde) mentions a Danish attack on England, but such an omission was not uncommon in Danish chronicles. Thus the Chronicle of Roskilde wrote that King Cnut IV (the Holy) was murdered because he imposed a hated tax, the nefgjald,81 but the same chronicle completely omits the planned and aborted invasion of England. There was thus a precedent for omitting the military failures of Danish kings. The next to write about Erik III was the German chronicler, Helmold of Bosau, whose work the Chronica Slavorum was finished around 1170. In the chronicle, Helmold reported several interesting pieces of information about Erik, namely that Erik was chosen as guardian of the realm until the maturity of the rightful heirs, and Helmold was the first to mention Erik’s cognomen, Spac (the Wise). Helmold described Erik as “a man of peace and that he placidly governed the kingdom but that he resisted too little the fury of [the] Wends. For at this time their depredations became unusually severe.”82 The Valdemarian court historian Saxo Grammaticus, who wrote some thirty to forty years later, followed Helmold by stating 77 78 79 80 81 82

Thomas K. Heebøll-Holm, “A Franco-Danish Marriage and the Plot against England,” Haskins Society Journal 26 (forthcoming 2015). Hermanson, “How to Legitimate Rebellion,” p. 139. SM, 1:32. Gelting, “Da Eskil,” pp. 199–202. SM, 1:24. Helmold of Bosau, Cronica Slavorum, 3rd ed., ed. Bernhard Schmeidler MGH SS in usum scholarum 32, (Hannover, 1937), p. 124; Helmold of Bosau, The Chronicle of the Slavs by Helmold, Priest of Bosau, trans. Francis Joseph Tschan (New York, 1966), p. 182. (I ch. 67).



When the Lamb Attacked the Lion

45

that Erik was merely the guardian of the realm until Valdemar reached maturity. However, Saxo omitted his cognomen, Spac, and while he characterized Erik as brave and pious, apart from that Erik was not endowed with any other positive qualities. He was, rather, characterized as imprudent and ineloquent, and so careless about his own life that when a battle was about begin his men had to restrain him in order to prevent him from attacking the enemy singlehandedly and getting killed. Furthermore, Saxo portrays him as superstitious, hesitant, wavering and easily swayed by the opinion of lesser men. He earned the scorn of Saxo, who believed that his weakness made the Wends bold and that their attacks on Denmark multiplied during his reign.83 This in essence made him unfit to rule, since a king’s primary duty, according to Saxo, was to defend the patria against foreign aggression and tyranny.84 Furthermore, Saxo refused altogether to recognize him as king. However, not all agreed with Saxo’s interpretation. The chronicler Sven Aggesen, nephew of Archbishop Eskil and a man whose family had consistently sided with the rebels against the Valdemarians,85 wrote a subtle history of the Danish kings, the Brevis historia regvm Dacie, finished after 1185. In it he exalted Valdemar, but also reprimanded him for being more cruel against his own than was just.86 He described Erik, whose son Magnus had been in league with Aggesen’s rebellious family, thus: “They called him the lamb (agnum) on account of his sweet and gentle nature, and in his days there was a plenteous abundance of everything.”87 Finally, the Knytlingasaga, finished around 1259 but drawing on a number of sources including Saxo Grammaticus, asserted that Erik was merely a warden of the kingdom until the maturity of the heirs. However the remainder of the portrait was somewhat positive. The saga thus notes that Erik Lamb [sic] was a wise man and much beloved by the people, which is why they called him spaki – the wise.88 After the (possible) attack and defeat of the Danes in England in 1138, the Danes directed their military attention eastwards. Thus, nine years later, in 1147, the Danes participated in the Second Crusade against the pagan Wends in the Baltic. During the reigns of Valdemar I and Cnut VI these wars were intensified, and by 1185 the Wends finally surrendered and swore allegiance to the 83 84 85

86 87 88

Saxo, Gesta, 2:14.2.2–15. Thomas Riis, Les institutions politiques centrales du Danemark 1100–1332 (Odense, 1977), pp. 86–101. Hermanson, “How to Legitimate Rebellion,” p. 130; Mia Münster-Swendsen, “The Making of the Danish Court Nobility: The ‘Lex castrensis sive curiae’ of Sven Aggesen Reconsidered,” in Statsutvikling i Skandinavia i middelalderen, ed. Sverre Bagge et al. (Oslo, 2012) p. 261. Ibid., pp. 267, 275. SM, 1:136, Sven Aggesen, The Works of Sven Aggesen. Twelfth-century Danish Historian, trans. Eric Christiansen, (London, 1992), p. 136. Bjarni Guđnason, ed., Danakonunga sogur (Reykjavik, 1982), pp. 268–72; Lars Boje Mortensen, “A Thirteenth-Century Reader of Saxo’s Gesta Danorum,” in The Creation of Medieval Northern Europe. Essays in Honour of Sverre Bagge, ed. Leidulf Melve and Sigbjørn Sønnesyn (Oslo, 2012), pp. 346–45.

46

Thomas K. Heebøll-Holm

Danish king. These wars were cast as crusades. Indeed, the canonization of Cnut Lavard in 1170 was likely a reward for the Danish conquest and conversion of the pagan island of Rügen in 1168, and Cnut Lavard was praised as a crusader saint.89 Furthermore, in the contemporary Danish historiography Valdemar’s wars against the Wends were cast as pious crusades, with the Danish king as a miles Christi on the frontier of Christendom. In that regard it seems that medieval historians followed a twelfth-century stance on warfare, as suggested by John Gillingham, where attacks on fellow Christians became increasingly difficult to defend and a dichotomy between acceptable acts of warfare against Christian and pagan enemies became prevalent.90 Saxo Grammaticus’ account of Valdemar I’s reign contains at least three passages in which the usually bellicose and indeed bloodthirsty Saxo stresses how the Danes under Valdemar were reluctant to attack fellow Christians.91 In 1158, relates Saxo, Valdemar suspected the inhabitants of the island of Falster of treasonous collaboration with the Wends and decided to conduct a punitive raid against them. However, God struck him with a deadly illness which receded only when Valdemar was cured by bishop Absalon’s intensive prayer. Accordingly, Valdemar realized the error of attacking fellow Christian countrymen and cancelled the raid. Saxo described how, in a skirmish between Danish and German warriors in 1170, Danish knights pursued the fleeing Germans. Some of the Danes, however, struck the Germans from their horses with the blunt ends of their lances so as not to imperil their souls by killing fellow Christians; and in the same year a Danish fleet conducting a raid against the pagan Estonians came to the Swedish island of Øland. Though the Danes were in conflict with Sweden, they refrained from plundering the islanders because they were fellow Christians. No such mercy was accorded the pagan Wends. For instance, in 1171 Absalon ordered the merciless slaughter of the Circipan Wends.92 Furthermore, by the time that Saxo was writing (probably between 1200 and 1219) the Valdemarians enjoyed a good relationship with the English. Not only does Valdemar I’s chancellor, bishop Radulf of Ribe (successor to Helias), seem to have been an Englishman,93 but Philip Augustus’ repudiation of Ingeborg of 89

90

91

92 93

Eric Christiansen, The Northern Crusades (London, 1997), pp. 50–72; Thomas K. HeebøllHolm, “Between Pagan Pirates and Glorious Sea-Warriors: The Portrayal of the Viking Pirate in Danish Twelfth-Century Latin Historiography,” Viking and Medieval Scandinavia 8 (2013), pp. 161–66; Friis-Jensen, “In the Presence of the Dead,” pp. 202, 213. See for instance John Gillingham, “Conquering the Barbarians: War and Chivalry in TwelfthCentury Britain and Ireland,” in John Gillingham, The English in the Twelfth-Century (Woodbridge, 2000), pp. 41–58, and “Women, Children and the Profits of War,” in Gender and Historiography. Studies in the Earlier Middle Ages in Honour of Pauline Stafford, eds Janet L. Nelson et al. (London, 2012), pp. 61–74. Stephen C. Jaeger, “Chrétien de Troyes and Saxo Grammaticus, Clerical Satirists,” in The Creation of Medieval Northern Europe. Essays in Honour of Sverre Bagge, ed. Leidulf Melve and Sigbjørn Sønnesyn (Oslo, 2012), p. 365. Saxo, Gesta, 2:14.22.2–6, 14.40.3, 14.45.6, 14.47.7 Michael H. Gelting, “Kansleren Radulfs to bispevielser,” (Dansk) Historisk Tidsskrift 14 (1980). pp. 325–36, Münster-Swendsen, “Educating the Danes,” p. 160.



When the Lamb Attacked the Lion

47

Denmark in 1193 drove the Danes straight into the arms of the Angevins and their Welf allies and kinsmen. Even after Philip had taken Ingeborg back in 1213, relations between the Danes and the French king were strained, though it seems that both sides were commencing a slow rapprochement.94 Nonetheless the idea of a Danish claim to England was kept alive in Denmark and abroad. In the treatise The Dialogue of the Exchequer, from the 1170s, Richard Fitz Nigel noted that the Danes claimed a hereditary right to rulership over England. In 1196–97 William of Newburgh wrote as a matter of fact that the reason for Philip Augustus to seek a marriage with princess Ingeborg of Denmark was that the French king desired – through the marriage – to obtain the Danish claim to England, and between 1198 and 1206 the Picard chronicler, Lambert of Ardres, remarked that at the time of his writing the Danes still desired to reclaim England.95 In Denmark the claim seemingly was also kept alive. In the 1180s Sven Aggesen wrote at length in his Lex Castrensis about Cnut the Great’s glorious rule over England,96 and during the negotiations over Philip’s repudiation of Ingeborg the Danes produced a genealogy of the Danish kings that showed a clear lineage from Ingeborg to Cnut the Great, thereby proving a Danish claim to the English throne.97 Finally, in the first of book of the Gesta Danorum, Saxo Grammaticus opened the narrative proper with a description of how the Danes were an autochthonous people and stated that the first two Danish kings were Dan and Angul. Dan named Denmark, whereas Angul gave name to England. Thus Saxo claimed that from time immemorial the English kings were descended from Danish royal stock. This notion probably served as a rebuttal to Geoffrey of Monmouth’s claim that before the Danish Viking invasions King Arthur had invaded and subjugated Denmark.98 Accordingly, there seems to have been a

94

95

96 97 98

Heebøll-Holm, “A Franco-Danish marriage,”; Thomas K. Heebøll-Holm, “Why was William of Æbelholt canonised? The Two Lives of Saint William,” in The Writing of History in Scandinavia and its European Context, 1000–1225, ed. Thomas K. Heebøll-Holm et al. (Durham, forthcoming). Richard Fitz Nigel, Dialogus de Scaccario, ed. and trans. Charles Johnson et al. (Oxford, 1983), p. 55; William of Newburgh, Historia Rerum Anglicarum, in Chronicles of the Reigns of Stephen, Henry II., and Richard I, 4 vols., ed. Richard Howlett (Wiesbaden, 1964), 1:368–70; Lambert of Ardres, Lamberti Ardensis historia comitum Ghisnensium, ed. Johann Heller, MGH SS 24, p. 557. Though a claim to the English throne through the marriage of a French king and a foreign queen with a tenuous claim to England may seem far fetched, it was in fact part of Louis VIII’s justification for invading England in 1216. Through his wife, Blanche of Castille, Louis claimed a right to the English throne. See Charles Petit-Dutaillis, Étude sur le règne de Louis VIII, 1187–1226 (Paris, 1894). pp. 84–85. I thank Frédérique Lachaud for making me aware of this work and argument. SM, 1:64–92, 120–24. Ibid., 1:176–85. Saxo, Gesta, 1.1.1. On Saxo’s knowledge of Geoffrey of Monmouth’s chronicle, see Lars Boje Mortensen, “Saxo og Geoffrey af Monmouth,” Renæssanceforum 3 (2007). http://www. renaessanceforum.dk/rf_3_2007.htm (accessed 11 June 2014); Niels Lukman, “Saxos kendskab til Galfred af Monmouth,” (Dansk) Historisk Tidsskrift 10 Rk., 6 (1942–44), 593–607; Thomas

48

Thomas K. Heebøll-Holm

generally accepted notion that the Danes had a legitimate claim to England, but since Erik III was not recognized by the Valdemarians as a true king, any claim to England on his behalf was written out. Simply put, a Danish attack led by a rival lineage to the Danish throne against a fellow Christian country and resulting in an ignoble defeat was not something that the Valdemarians wished to commemorate. Conclusion In 1138, England was hit by invasions and revolts led by contenders to the throne. In general the contenders are all portrayed as members of the AngloNorman aristocratic world, yet the Continuation of Sigebert’s chronicle shows that the Danes – the old contenders to the English throne – in all likelihood also tried to stake their claim. In Denmark conditions for an attack were ideal, as the new king, Erik III, had every interest in rallying the warring magnate forces against an external foe in order to stabilize his regime. The reason why no one wrote about it in either England or Denmark was that after the 1150s both countries had new regimes, eager to legitimize their rule. Thus, in Angevin England any notions of Danish claims were repressed, while in Denmark the Valdemarians sought to erase and pass over the reign of Erik III as a mere interim period of regency before the rightful heir to the throne reached maturity. Furthermore, for the Valdemarians, who wanted to present themselves as pious warriors on the Christian frontier, there was no need to be reminded of a military failure against a fellow Christian country. Thus, independently of each other, the Angevins and the Valdemarians repressed the history of the attack. However, while the memory of the attack was suppressed, the Danish claim to England was feared in England, and kept alive in Denmark – and in France, which was what caused Philip Augustus in 1193 to seek the hand of a Danish princess. This marriage was a carefully considered arrangement, carried out because the Danes still had a claim to the English throne and in recent times had shown a will to enforce it. Thus the threat of a Danish attack on England was on the minds of many northern European rulers throughout the twelfth century, even though the French aristocrats who ruled England fiercely tried to repress the memory of the attack in 1138 and of the Danish claim to England.

K. Heebøll-Holm, “Priscorum quippe curialum, qui et nunc militari censentur nomine. Riddere i Danmark i 1100-tallet,” (Dansk) Historisk Tidsskrift 109 (2009), 50–60.

Valdemar II 1202–41 ~Dagmar of Bohemia ~Berengaria of Portugal

Ingeborg Magnus Eriksen †1237/38 †1178 (?) ~Philip Augustus of France

Cnut Karlsen †1178 (?)

Cnut V 1147–52, 1154–57

Magnus †1134

Buris Henriksen †1167 (?)

Henry Svensen †1134

Nicholas Harald II Sven 1104–34 1074–80 †1104

Karl Karlsen †1178

Valdemar I Sven III Grathe Erik III Oluf Haraldsen Karl Eriksen 1154/57–82 1147–57 1137–46 †1141 (?) †?? ~Sophia of Minsk

Cnut VI 1182–1202 ~Gertrude of Saxony



Cecilia †??

Cnut IV the Holy 1080–86

Cnut Lavard Erik II Ragnhild Harald Kesja Charles the Good †1131 1134–37 †1135 Count of Flanders 1119–27

Erik I Olaf 1096–1103 1086–95

Cnut II the Great Estrid England 1016–35, ~ Ulf Jarl Denmark 1018–35 Sven II Estridsen 1047–74

Sven I Forkbeard Denmark 986–1014, England 1013–14

Figure 1.  Genealogy of the Kings of Denmark

3 Development of Prefabricated Artillery during the Crusades1 Michael S. Fulton

In 1291, the Mamluk sultan al-Ashraf Khalil brought the largest collection of artillery that had hitherto been witnessed in the theater of the crusades before the walls of Frankish Acre. These machines were immense in both scale and number.2 However, the progressive development of prefabricated artillery and the logistics that permitted its effective deployment have largely been taken for granted by modern scholarship. Most discussions of medieval mechanical stone-throwers have been consumed by debates over the source of their mechanical power and the amount of force that they were capable of generating. The progressive use and reuse of these engines, by incorporating them into siege trains, has largely been overlooked and the point at which it became more common to transport prefabricated engines to and from a siege, rather than building them anew each time, has remained unappreciated. Like the improvement of the engines themselves, this was the product of a developmental process that spanned beyond the two hundred years of Latin lordship in Syria-Palestine. At its core, this transition was a joint product of technological advancement in artillery science and the logistic capabilities of the period’s most powerful figures. It was left to those capable of financing such engines, costly enough to render their storage and movement with an army less expensive than to build them anew, to do so. Although an apparently simplistic balance of cost and value, the issue was made infinitely more complicated when elements of geography, which drove the cost of both construction and transportation up and down by region, political climate and even the changing seasons were factored in. The siege artillery of the twelfth and thirteenth centuries was of the trebuchet type: engines consisting of a rotating beam that was fixed to an axle at a point that created a disparity between the two arms. When a downward force was applied to the short arm of the beam this movement was augmented at the end of the long arm according to the principle of mechanical advantage. Attached to the tip of the long arm was a sling, which further augmented the velocity of the projectile couched within prior to release. Evidence for the use of classical 1 2

I would like to extend my thanks to Denys Pringle and Rabei Khamisy for reading an early draft of this article and providing helpful feedback. For artillery at the siege of Acre see below.

52

Michael S. Fulton

torsion engines in the Latin and Arab worlds is sparse by the seventh century, and by the ninth century both cultures had clearly adopted swing-beam engines as their dominant form of stone-throwing siege artillery, using a mix of classical and new medieval terminology to refer to them.3 These engines were still relatively light, relying on traction or “pulling” power to propel the short end of the beam downwards and, correspondingly, to raise the long arm, accelerating the couched projectile before its release. A technological leap was made when a new power source was developed, most likely during the twelfth century, replacing the traction energy source with a dead-weight counterpoise. This allowed machines to be constructed on a larger scale, no longer constrained by the distance an individual could physically pull a rope in one tug. The emergence of the counterweight trebuchet was as a complement to the traditional traction type: each had its own strengths and both remained in use beyond the end of the period under discussion. Most traditional assumptions regarding the use of prefabricated and transportable artillery have revolved around the presumption that only counterweight trebuchets were regarded as valuable enough to warrant this treatment. For the present purposes, this assumption will be disregarded and a more elemental search for the inception and development of this practice through the course of Frankish-Muslim interactions in the Levant will be explored. At the beginning of the twelfth century, nearly all Frankish siege engines were built anew at each siege.4 All of the engines used through the course of the First Crusade were built in situ. Even though the artillery described at the siege of Nicaea in 1097 seems to have been of exactly the same type as that used at the siege of Jerusalem in 1099, there is no evidence to suggest that any of the engines used at the former, or any other siege in between, were used again at a later one.5 Eighty years later, at Harim, little had changed. William of 3

4

5

For theories on this development and diffusion see Donald Hill, “Trebuchets,” Viator 4 (1973), 99–114; Carroll Gillmor, “The Introduction of the Traction Trebuchet into the Latin West,” Viator 12 (1981), 1–8; Paul Chevedden, “The Invention of the Counterweight Trebuchet: A Study in Cultural Diffusion,” Dumbarton Oaks Papers 54 (2000), 71–116; John France, Western Warfare in the Age of the Crusades (Ithaca, NY, 1999), pp. 118–23. See also Randal Rogers, Latin Siege Warfare in the Twelfth Century (Oxford, 1992), Appendix 3, pp. 254–73. For context immediately preceding the First Crusade see, for example, Robert of Normandy and Robert of Belleme’s engines at Courcy (1091) and Breval (1092), Orderic Vitalis, The Ecclesiastical History of Orderic Vitalis 8.16, 24, ed. and trans. Marjorie Chibnall, 6 vols. (Oxford, 1969–1980), 4:232–37, 288–95. For the use of artillery at Nicaea see Anonymous, Gesta Francorum et aliorum Hierosolimitanorum ed. and trans. Rosalind Hill (Edinburgh, 1962), pp. 14–15; Guibert of Nogent, Die gesta per Francos 3.6–10, ed. R. B. C. Huygens, Corpus Christianorum 127a (Turnhout, 1996), pp. 145–56, trans. Robert Levine, The Deeds of God Through the Franks (Woodbridge, 1997), pp. 62–66; Raymond of Aguilers, Historia Francorum qui ceperunt Iherusalem, in Recueil des Historiens des Croisades: Historiens Occidentaux (hereafter RHC Oc), vol. 3 (Paris, 1866), p. 239; trans. John Hill and Laurita Hill, Historia Francorum qui ceperunt Iherusalem (Philadelphia, 1968), p. 26; Fulcher of Chartres, Historia Hierosolymitana 1.10, ed. Heinrich Hagenmeyer, (Heidelberg, 1913), pp. 185–87; trans. Frances Rita Ryan, A



Prefabricated Artillery

53

Tyre explicitly states that the Franks, under Philip I of Flanders, Raymond III of Tripoli and Bohemond III of Antioch, built their engines after encamping in front of the stronghold in late 1177.6 The evident simplicity and limited value assigned to these machines can perhaps best be seen through the way in which they were discarded after many sieges. Destroying siege engines after the conclusion of a siege, successful or otherwise, was a common practice and reveals that it was less profitable to preserve or dismantle the engines for reuse at a later date or different site than to build new engines when required, eliminating the challenges of storage and transportation. In 1108, Albert of Aachen records that Baldwin II burnt his engines before withdrawing from Sidon after the garrison had been strengthened by the arrival of a Fatimid fleet.7 Similarly, the Anonymous Syriac Chronicle states that Bursuqi burnt his engines, with which he was besieging ‘Azaz, on the arrival of Frankish forces in 1125.8 Decades later, Amalric burnt his machines before departing from Alexandria for Cairo, and eventually Ascalon, in 1167.9 William of Tyre claims that the Franks repeated this practice in 1169 after giving up on their siege of Damietta, although Niketas Choniates asserts that it was the Byzantine component of this force that burnt its engines. In this instance, Baha’ al-Din and al-Maqrizi note that it was, rather, the citizens of Damietta who burnt these engines after the Franks departed for Palestine.10 If these Muslim sources are correct, they reveal how little value the garrison assigned to the Frankish artillery. This was the third successive year that the Franks had invaded Egypt, and for the garrison not to salvage and appropriate the abandoned engines for its own

6 7 8

9 10

History of the Expedition to Jerusalem, 1095–1127, ed. Harold S. Fink (Knoxville, 1969), p. 82; Albert of Aachen, Historia Ierosolimitana: History of the Journey to Jerusalem 2.29, 32, ed. and trans. Susan B. Edgington (Oxford, 2007), pp. 110–11, 114–17; Robert the Monk, Historia Iherosolimitana 3.3, in RHC Oc, vol. 3 (Paris, 1866), p. 756; William of Tyre, Cronique 3.6–9 (7–10), ed. R. B. C. Huygens, Corpus Christianorum 63, 2 vols. (London, 1986), 1:203–8; trans. Emily Babcock and A. C. Krey, A History of Deeds Done Beyond the Sea, 2 vols. (New York, 1976), 1:158–63. For the use of artillery at Jerusalem see Raymond of Aguilers, Historia Francorum 10, pp. 298–99, trans., pp. 125–26; Peter Tudebode, Historia de Hierosolymitano itinere 15.2, in RHC Oc, vol. 3 (Paris, 1866), p. 107; Albert of Aachen, Historia Ierosolimitana 6.2, 9, 14–15, 17–19, pp. 406–7, 414–17, 420–23, 424–29; Fulcher of Chartres, Historia Hierosolymitana 1.27, pp. 296–99, trans., pp. 120–21; Ralph of Caen, Gesta Tancredi 123–25, pp. 691–93; William of Tyre, Cronique 8.6, 8, 13, 1:392–93, 395–96, 426–29, trans., 1:351, 354, 361–63. William of Tyre, Cronique 21.18 (19), 2:986–87, trans., 2:425–26. Albert of Aachen, Historia Ierosolimitana 10.51, pp. 766–67. Anonymous Syriac Chronicle, trans. Arthur S. Tritton with notes by H. A. R. Gibb, “The First and Second Crusade from an Anonymous Syriac Chronicle,” Journal of the Royal Asiatic Society 65 (1933), 97. William of Tyre, Cronique 19.32, 2:909, trans., 2:343. Ibid. 20.17, 2:933, trans., 2:368; Niketas Choniates, O City of Byzantium, Annals of Niketas Choniates 1, trans. Harry J. Magoulias (Detroit, 1984), p. 95; Baha’ al-Din Ibn Shaddad, The Rare and Excellent History of Saladin, trans. D. S. Richards (Aldershot, 2001), p. 46; al-Maqrizi, A History of the Ayyubid Sultans of Egypt, trans. R. J. C. Broadhurst (Boston, 1980), p. 37.

54

Michael S. Fulton

use against a possible subsequent siege is revealing. Baha’ al-Din relates that a similar episode occurred in 1174, when Sicilian forces abandoned their engines when they lifted their siege of Alexandria ahead of Saladin’s approach. Once more, it was the inhabitants of the city who apparently burnt these machines.11 By the end of the twelfth century, artillery had reached a sufficiently high level of sophistication that certain notable figures found value in transporting their engines with them in sections that could be quickly reassembled. However, abandoning and burning artillery did not completely cease once the practice of transporting these engines to and from each siege became more widespread. For example, Imperial forces appear to have burnt their engines when the siege of Beirut was abandoned in 1232.12 The ungainly machines were yet another component of baggage to be laboriously removed if the assailing army came under pressure from an opposing force. To improve mobility while denying the enemy any gain, engines were frequently burnt with the rest of the cumbersome baggage. Although this practice is most commonly noted in connection with failed siege efforts, there is little evidence to suggest that artillery was normally preserved after successfully concluded sieges in the early twelfth century. Instead, these engines were subsequently harvested for their materials and put towards the repair efforts that necessarily followed most sieges. However, in the theatre of the crusades, the practice of removing engines from a siege, successful or not, has its roots in the twelfth century and was a common procedure by the middle of the thirteenth. Long before this, the transportation of artillery had been widely embraced by Roman armies. In AD 70, over a millennium before the First Crusaders arrived in Palestine, Titus had brought ballistic machinery to Jerusalem before investing the city.13 Twelfth-century sources within the sphere of the crusades imply that 11 12

13

Baha’ al-Din, History of Saladin, p. 50. Philip of Novara, Les Gestes des Chiprois 170, ed. Gaston Raynaud, Recueil de Chroniques Françaises (Paris, 1887), reprinted (Osnabrück, 1968), p. 89 ; trans. John L. La Monte, The Wars of Frederick II against the Ibelins in Syria and Cyprus (New York, 1936), p. 137; L’estoire de Eracles empereur et la conqueste de la terre d’Outremer (hereafter Eracles) 33.29, in RHC Oc 2, p. 396. Similarly, Philip II of France burnt the twenty-four perrariae that he had positioned around Rouen in 1193 and the Mongol force besieging al-Bira in 1265 burnt its artillery when Mamluk relief forces arrived, apparently repeating this practice in 1273. In 1313, Khar-Banda abandoned, rather than burnt, his engines when he gave up the siege of al-Rahba, an apparent mistake, as they were salvaged by the garrison and brought within the town. Roger of Hoveden, Chronica, ed. William Stubbs, Rolls Series 51, 4 vols. (London, 1868–71), 3:206–7; trans. Henry T. Riley, The Annals of Roger de Hoveden: Comprising the History of England and of Other Countries of Europe, 2 vols. (London, 1853), 2:289; Rigord, Gesta Philippi Augusti 94, in Oeuvres de Rigord et de Guillaume le Breton, ed. H. François Delaborde, vol. 1 (Paris, 1882), p. 126; Ibn ‘Abd al-Zahir, Baybars I of Egypt, ed. and trans. Syedah Fatima Sadeque (Dacca, 1956), p. 238; idem, ‘A Critical Edition of an Unknown Arabic Source for the Life of al-Malik al-Zahir Baibars’, trans. Abdul Aziz al-Khowayter, vol. 2 (unpublished Ph.D. thesis, S.O.A.S., 1960), 2:784; Abu’l-Fida’, The Memories of a Syrian Prince: Abu’l-Fida’, Sultan of Hamah, trans. P. M. Holt (Wiesbaden, 1983), p. 63. Josephus, A History of the Jewish Wars 5.2, in The Works of Flavius Josephus, trans. William Whiston (Hartford, CT, 1916), p. 798.



Prefabricated Artillery

55

this practice was employed by the Byzantines, to at least some degree, in the eleventh and twelfth centuries. For example, Matthew of Edessa records Tughrul Beg’s use of a particularly notable piece of artillery during his siege of Manzikert in the eleventh century. The engine was brought up from Baghash, where it had been built by the Byzantine emperor Basil for the defense of that city.14 Similarly, William of Tyre seems to imply that John Comnenus employed prefabricated artillery during his push against Cilicia and north-western Syria in 1138–39.15 Usama ibn Munqidh, whose family commanded the defence of Shayzar in 1138, also implies that the Byzantines brought their artillery with them.16 William of Tyre also suggests that Manual Comnenus brought siege engines on his visit to Antioch twenty years later and that ten to twenty Byzantine dromones carried engines of war (tormenta bellica), along with other arms and supplies, by sea to join the Frankish venture against Egypt in 1169.17 These machines appear to have been artillery, as Baha’ al-Din confirms that the Frankish-Byzantine force brought arms and artillery with it to Damietta.18 Although Niketas Choniates omits any direct mention of whether artillery was transported with the fleet, his account of its rapid deployment shortly after the Byzantines arrived at Damietta suggests that it was.19 The alignment of these distinct accounts strongly suggests that some form of siege equipment was transported to Egypt, most likely by the Byzantines, and the logistics of such imply that it was most likely artillery. Such engines would have been easy to move and all accounts note their prevalent use during the siege.20 These notions are echoed in a letter from Manuel Comnenus to Henry II of England in 1176, found in Roger of Hoveden’s chronicle, which recounts the transportation of mangana and other siege engines on campaign.21 When coupled with other Byzantine accounts, such as Anna Comnena’s rendition of the siege of Apollonias in the 1080s and the eastern campaigns of John Comnenus in the 1130s, as told by Niketas Choniates,22 it appears as though the Byzantines were accustomed to transporting artillery throughout the Comnenan

14 15 16

17 18 19 20

21 22

Matthew of Edessa, The Chronicle of Matthew of Edessa 2.3, trans. Ara Edmond Dostourian (Lanham, 1993), pp. 86–88. William of Tyre, Cronique 15.1, 3, 1:674, 677, trans., 2:94, 98. Usama ibn Munqidh, The Book of Contemplation: Islam and the Crusades, trans. Paul M. Cobb (London, 2008), p. 125. Although the use of artillery is noted by Niketas Choniates through the Cilician campaign as well as against Shayzar, he fails to specify if it was transported to Syria or built anew, Niketas Choniates, O City of Byzantium 1, pp. 14–18. William of Tyre, Cronique 18.25, 20.13, 2:847–49, 926–27, trans., 2:279–81, 361–62. Baha’ al-Din, History of Saladin, p. 45. Niketas Choniates, O City of Byzantium 2.5, pp. 91–96. Tarver found that each component of his reconstructed traction trebuchet could be carried by a single man and that six or seven could easily carry all of the machine’s various components, which could be reassembled within an hour, W. T. S. Tarver, “The Traction Trebuchet: A Reconstruction of an Early Medieval Siege Engine,” Technology and Culture 36 (1995), 156. Roger of Hoveden, Chronica, 2:102–4, trans., 1:419–23. Anna Comnena, The Alexiad 6.13, ed. and trans. Elizabeth A. Dawes (London, 1928), p. 163; Niketas Choniates, O City of Byzantium 1, pp. 12–18.

56

Michael S. Fulton

period. This practice would be steadily adopted, either directly or organically, by the Franks and Muslims of Syria-Palestine through the period of the crusades. There are a few indications in the Latin sources that some prefabrication may have occurred during the first half of the twelfth century. Writing in Europe, Albert of Aachen records that Baldwin I built siege engines, including mangenae and tormenta, at Acre in 1106, intending to use them against Sidon.23 Fulcher of Chartres claims that tormenta were brought by sea by the Fatimid force that attacked Jaffa in 1123.24 William of Tyre implies that Pons of Tripoli had prepared a siege train, which he intended to use against Raffaniya, by the time that he was joined by Baldwin II in 1126.25 Although such instances may seem to outwardly indicate practices of prefabrication, these are just three perspectives on three different sieges out of the innumerable sieges to be found in the sources across the first seven decades that the Franks occupied the Holy Land. Furthermore, Albert of Aachen never visited the East, William of Tyre recorded his account over half of a century later and Fulcher of Chartres, although the closest source to these events, like William of Tyre, was clearly not an expert on siege warfare and its implements. In Ibn al-Qalanisi’s concurrent chronicle of events from Damascus, there is a steady progression in the way artillery is dealt with. Initially these machines are treated largely as foreign weapons, predominantly ascribed to the Franks, Greeks and Zanki. In general, references to artillery in his chronicle are quite sparse until they dramatically increase from 1138.26 It is his account of preparatory measures through the mid-twelfth century that is most notable here. In the autumn of 1145, Ibn al-Qalanisi explicitly notes that Zanki was preparing men and machines, including artillery, for an upcoming campaign. In this instance, Zanki’s plans changed following a revolt in Edessa, compelling him to send his artillery back from Baalbek to Homs.27 Whether these engines were built at or transported to Baalbek is relatively insignificant, as Zanki evidently regarded them as having sufficient value to make arrangements for their transfer to Homs before he set off for northern Syria. Long before the First Crusade, the potentates of Damascus appear to have kept stores of timber within the city, possibly

23 24

25

26

27

Albert of Aachen, Historia Ierosolimitana 10.3–4, pp. 720–21. Fulcher of Chartres, Historia Hierosolymitana 3.17, pp. 661–63, trans., pp. 240–41. This is a development in itself, as a Fatimid force had transported ladders with them in a similar set of rapid assaults in 1115, Fulcher of Chartres, Historia Hierosolymitana 2.53, pp. 585–89, trans., pp. 212–13; William of Tyre, Cronique 11.24, 1:531–32, trans., 1:502–3. William of Tyre, Cronique 13.19, 1:610–11, trans., 2:30. Although Fulcher clearly indicates that the Franks employed artillery at this siege, he omits these preparations by Pons before departing Tripoli. Fulcher of Chartres, Historia Hierosolymitana 3.51, 53, pp. 793–97, 798–99, trans., pp. 292–93, 294. The impetus for this shift may relate to Zanki’s move on Damascus in 1139; through his account of which, Ibn al-Qalanisi gives his first description of artillery beyond a summary inclusion, Ibn al-Qalanisi, The Damascus Chronicle of the Crusades, trans. H. A. R. Gibb (London, 1967), pp. 254–56. Ibn al-Qalanisi, Damascus Chronicle, pp. 269–70.



Prefabricated Artillery

57

due to a limited supply of large trees that were supported along the course of the Barada River.28 This may have encouraged the storage of artillery components at an earlier date here than in other power centers which Following his initial relative silence, Ibn al-Qalanisi notes no fewer than six instances through the 1140s and 1150s when artillery was prepared in one locale of southern Syria for use in another.29 In each instance, except that of 1145, these engines were apparently stored in the arsenal of Damascus. It seems that William of Tyre corroborates at least the last of these, as he confirms that Nur al-Din brought his engines with him against Frankish Banyas in 1157.30 Although it is tempting to judge William’s reference on its own with skepticism, it does confirm Ibn al-Qalanisi’s assertion of the same. Likewise, there is no precedent for such a turn of phrase in the cleric’s writing. Accordingly, the textual evidence, supported by the natural geography and regional importance of Damascus as a power center, make it entirely likely that prefabricated artillery components were stored within the city before this became a common practice among the Franks. Although it appears unlikely that these engines existed through the first four decades of the Frankish presence, the aggression of Zanki, Nur al-Din and the Franks towards Damascus through the 1140s and 1150s likely contributed to the decision to stockpile such weapons, or at the very least the specific materials necessary for their construction, before Nur al-Din took power there in 1154. This tradition continued into the thirteenth century and it was artillery from Damascus that al-Salih Isma‘il brought against Beaufort in 1240, and likely from here that al-Salih Ayyub collected his artillery before besieging Homs in 1248.31 Under the Mamluks, artillery stored in Damascus was used to invest a number of Frankish strongholds through the 1260s and 1270s.32 If artillery was being stored by the reign of Nur al-Din, by that of Saladin there is strong textual evidence that artillery was being transported from siege 28

29

30 31

32

Such stores of timber were also found in Jerusalem upon the arrival of the First Crusaders. According to William of Tyre, this advantage meant that the defensive artillery of the garrison was superior to that of the assailing Franks. William of Tyre, Cronique 8.8, 13, 1:395–96, 403–4, trans., 1:354, 361–63. 1145, Zanki prepared engines in Homs to use against Damascus; 1146, Mu’in al-Din Unar prepared engines in Damascus to use against Baalbek; 1147, Mu’in al-Din Unar prepared engines in Damascus to use against Sarkhad and Bosra; 1151, Nur al-Din summoned engines from Damascus to use against Bosra; 1152, Mujir al-Din summoned engines from Damascus to use against Bosra; 1157, Nur al-Din summoned engines from Damascus to use against Banyas. Ibn al-Qalanisi, Damascus Chronicle, pp. 269–70, 272–73, 276, 310, 312–13, 333. William of Tyre, Cronique 18.12, 2:827–28, trans., 2:257–58. La continuation de Guillaume de Tyre de 1229 à 1261, Dite du manuscrit de Rothelin (hereafter Rothelin) 32, in RHC Oc, vol. 2 (Paris, 1859), p. 552; Abu Shama, Le livre des deux jardins, in Recueil des Historiens des Croisades: Historiens Orientaux (hereafter RHC Or), vols. 4–5 (Paris, 1898–1906), 5:205; Abu’l-Fida’, Resume des histoire des Croisades tire des annales d’Abou’l-Feda, in RHC Or, vol. 1 (Paris, 1872), p. 125; al-Maqrizi, Ayyubid Sultans of Egypt, p. 85. See below for examples.

58

Michael S. Fulton

to siege in some cases. Ballistic engines played a major part in Saladin’s campaigns against Kerak in the 1180s. However, despite excellent accounts of these campaigns and the obvious employment of artillery, there is little conclusive evidence to suggest where it was manufactured. Al-Maqrizi notes that artillery was sent to Saladin’s army as it prepared to march out of Egypt for Syria in 1182;33 but, as he was writing much later, this reference must be viewed with caution. William of Tyre and Ibn al-Athir both imply that Saladin brought his artillery with him to Kerak in 1183.34 Unfortunately, neither source was an eye-witness and the only surviving first-hand account of Saladin’s Transjordan campaigns, that of ‘Imad al-Din, does not identify the origin of these prominent engines, nor of those used against Kerak the following year.35 A much clearer picture emerges regarding this facet of siege warfare in the aftermath of the fateful battle of Hattin. In early July 1187 Saladin decisively and devastatingly crushed the Frankish army of Jerusalem at Hattin. Just days before, Saladin had laid siege to Tiberias in an effort to entice the Franks to take to the field against him. Although Michael the Syrian claims that Saladin prepared arms and machines before crossing the Jordan and Baha’ al-Din confirms the use of artillery at the siege of Toron later in July, ‘Imad al-Din is the only source to suggest that artillery might have been used against Tiberias.36 The events of early July 1187 are similar in some ways to Saladin’s attack on the castle at Jacob’s Ford in 1179. In this earlier instance, 33 34

35

36

Al-Maqrizi, Ayyubid Sultans of Egypt, p. 67. William of Tyre, Cronique 22.29 (28), 2:1055–56, trans., 2:499–501; Ibn al-Athir, The Chronicle of Ibn al-Athir for the Crusading Period from al-Kamil fi’l-Ta’rikh, trans. D. S. Richards, 3 vols. (Aldershot, 2008), 2:297–98. Ronnie Ellenblum has proposed that the brevity of the 1170 siege of Kerak and apparent use of only two stone-throwers are evidence of a shift in Muslim siege techniques to the use of heavy artillery. Ronnie Ellenblum, Crusader Castles and Modern Histories (Cambridge, 2007), p. 258. See also Abu Shama, Le livre des deux jardins, 4:153–54; Ibn al-Athir, Chronicle, 2:185. However, Ellenblum apparently fails to acknowledge that the use of small numbers of ballistic engines was the norm throughout the Latin East during the twelfth century and does not reconcile the subsequent use of between seven and nine engines during both the 1183 and 1184 sieges of the same castle. Likewise, he does not address the lack of any apparent damage to the Frankish masonry of Kerak’s town-facing northern front, testament to the relatively limited power of even late twelfth-century artillery. Accordingly, far weaker defences would have sufficed at this time to resist assailing artillery. Michael the Syrian, Extrait de la Chronique de Michel le Syrien, in Recueil des Historiens des Croisades: Documents Armeniens, vol. 1 (Paris, 1869), p. 396; ‘Imad al-Din al-Isfahani, Conquête de la Syrie et de la Palestine par Saladin, trans. Henri Massé (Paris, 1972), p. 24. Cf. Baha’ al-Din, History of Saladin, p. 73; ‘Imad al-Din, Conquête de la Syrie, p. 95; Ibn al-Athir, Chronicle, 2:320–21; Eracles 23.31, pp. 47–48. The debatable reference comes in the form of the Arabic term hajjarun, defined elsewhere by Reuven Amitai as “masons who cut the stones for the mangonels”, Reuven Amitai, “Foot Soldiers, Militiamen and Volunteers,” Texts, Documents and Artefacts, ed. Chase F. Robinson (Leiden, 2003), pp. 233–50, at p. 236. For a broader investigation of the term see also Rabei Khamisy, “Some Notes on Ayyubid and Mamluk Military Terms” in this volume. For the Arabic see Abu Shama, Le livre des deux jardins, 4:264. For the siege of Toron see ‘Imad al-Din, Conquête de la Syrie, pp. 37–38; Baha’ al-Din, History of Saladin, p. 76.



Prefabricated Artillery

59

Saladin had left his baggage train at Banyas until he had defeated the Frankish field forces that came against him. From ‘Imad al-Din’s statement that vinepoles and timber were collected from around Safed to construct defenses for his stone-throwers, this baggage presumably included artillery.37 The rapid fall of the castle at Jacob’s Ford, to mining and frontal assault, seems to have prevented the use of these engines, as neither the sources nor archaeology support the use of such at this siege.38 Saladin may have repeated this practice in 1183, keeping his artillery on the east side of the Jordan within the confines of a stronghold such as Banyas; however, his inability to draw out and defeat the army of Jerusalem in a pitched battle denied him any occasion to use it.39 The rapid movement of Saladin’s army in July 1187 and the established practice of employing prefabricated Damascene artillery indicate that the artillery used at Toron was probably part of a siege train, rather than constructed locally. Although the foliage of Galilee would support the construction of smaller engines, the repeated use of artillery during this campaign and ‘Imad al-Din’s references to its damaging impact at Toron suggest that it was prefabricated. If Saladin had crossed the Jordan with a significant siege train before defeating the Franks, rather than summoning it afterwards, it would certainly speak to his confidence, as the presence of this added baggage would naturally have hindered the mobility of his army. Although evidence is patchy through the course of July, there is little doubt that Saladin employed significant artillery against Ascalon and Jerusalem in the ensuing weeks. The sources almost unanimously emphasize Saladin’s use of artillery at Ascalon in September 1187.40 By this time Saladin’s brother, al-Adil, had arrived from Egypt to join the campaign of conquest. When considering the ambiguous references to artillery being moved out of Egypt earlier in the 1180s, 37

38

39

40

Abu Shama, Le livre des deux jardins, 4:198–206; William of Tyre, Cronique 21.27 (28)–29 (30), 2:1000–4, trans., 2:440–45; Ibn al-Athir, Chronicle, 2:264–65; al-Maqrizi, Ayyubid Sultans of Egypt, pp. 59–60. Despite claims that Saladin employed machinae at this siege by the contemporary, if distant, Robert of Toringi, no stone projectiles have thus far been uncovered through the course of excavations. Personal correspondence with excavators Adrian Boas and Kate Raphael, May 2013; Robert of Torigni, The Chronicle of Robert of Torigni, in Chronicles of the Reigns of Stephen, Henry II, and Richard I, ed. Richard Howlett, Rolls Series 82, vol. 4 (London, 1889), p. 288. See also William of Newburgh, Historia rerum anglicarum 3.11, in Chronicles of the Reigns of Stephen, Henry II, and Richard I, ed. Richard Howlett, Rolls Series 82, vols. 1 and 2 (London, 1884–85), 1:244. William of Tyre, Cronique 12.26 (25)–28 (27), 2:1048–54, trans., 2:492–98; Abu Shama, Le livre des deux jardins, 4:242–48; Baha’ al-Din, History of Saladin, pp. 60–61; Ibn al-Athir, Chronicle, 2:297; al-Maqrizi, Ayyubid Sultans of Egypt, p. 72. Artillery is not mentioned at any point through this episode in 1183, making this entirely conjectural. Baha’ al-Din, History of Saladin, pp. 76–77; ‘Imad al-Din, Conquête de la Syrie, pp. 97, 100; Abu Shama, Le livre des deux jardins, 4:312–13; Ibn al-Athir, Chronicle, 2:329–30; al– Maqrizi, Ayyubid Sultans of Egypt, p. 84; Itinerarium peregrinorum et gesta regis Ricardi 1.8, in Chronicles and Memorials of the Reign of Richard I, ed. William Stubbs, Rolls Series 38, vol. 1 (London, 1864), p. 20; trans. Helen J. Nicholson, Chronicle of the Third Crusade: A Translation of the Itinerarium Peregrinorum et Gesta Regis Ricardi (Aldershot, 1997), p. 37.

60

Michael S. Fulton

it is possible that stone-throwing machines, now resoundingly highlighted, were also brought to Palestine from Cairo, either overland or by sea. If prefabricated artillery was used at Ascalon, these engines were almost certainly the same ones that were subsequently used against the north wall of Jerusalem only weeks later.41 Despite their apparent value, the machines employed here in 1187 were not wall-breaching engines, akin to those found in the fourteenth century. Instead, like the ballistic engines of the twelfth century, they were used to inflict casualties on the garrison, or at most to chip away at the battlements that sheltered these men. More generally, they suppressed the defenders’ ability to mount an active defense from the parapet and provided cover for those assailants attempting to sap the besieged fortifications.42 Baha’ al-Din states that before laying siege to Tyre in late 1187, Saladin pitched his camp within sight of the city to await the arrival of his siege engines. Once these engines, including artillery, had arrived around nine days later, the siege was opened.43 Michael the Syrian also states that Saladin brought his engines to Tyre.44 Ironically, it is here, with what seems to be clear evidence that Saladin was moving his engines from siege to siege, that there was an apparent reversion to a much earlier practice. Saladin had initiated the siege in late November and by the end of December he was compelled to lift it. At this point the sources disagree concerning Saladin’s treatment of his siege artillery. The Eracles account, and those based on it, states that he burnt his engines.45 But Baha’ al-Din relates that Saladin disassembled his artillery and withdrew the machines with the army: it was only those materials that could not be transported that were burnt. Although it may be tempting to see the burning siege camp, and presumed artillery within, as indicative that this is reflective of a popular Frankish preference to build artillery anew before each siege, there is a more likely explanation. It is doubtful that the original Frankish source would have had a sufficient vantage point to witness the burning of the Muslim siege camp himself, and it would seem that, when informed of this, either he included Saladin’s siege engines for dramatic effect or he was genuinely mistaken. During 41

42

43

44 45

‘Imad al-Din, Conquête de la Syrie, pp. 44–46; Baha’ al-Din, History of Saladin, pp. 77–78; Abu Shama, Le livre des deux jardins, 4:326–30; Ibn al-Athir, Chronicle, 2:330–33; De expugnatione Terrae Sanctae libellus, ed. Joseph Stevenson, Rolls Series 66 (London, 1875), pp. 218–22; Itinerarium 1.9, pp. 20–21, trans., p. 38; Eracles 23.55–57, pp. 83–85. ‘Imad al-Din portrays the artillery at Jerusalem as more destructive than do the other sources, the nearest parallel being Marino Sanudo’s early fourteenth-century account. Marino Sanudo, Liber secretorum fidelium crucis 3.9.6, ed. J. Bongars (Hanover, 1611), reprinted with foreword by Joshua Prawer (Jerusalem, 1972), p. 192; trans. Peter Lock, The Book of the Secrets of the Faithful of the Cross – Liber Secretorum Fidelium Crucis (Farnham, 2011), p. 305. Baha’ al-Din, History of Saladin, pp. 78–79. Cf. ’Imad al-Din, Conquête de la Syrie, pp. 64–65; Eracles 24.2, pp. 104–5. See also Abu Shama, Le livre des deux jardins, 4:342–44; Ibn al-Athir, Chronicle, 2:335–38; Roger of Hoveden, Chronica, 2:346–47, trans., 2:90–91. Michael the Syrian, Extrait, p. 400. Eracles 24.4, p. 110. Cf. Itinerarium 1.10, p. 25, trans., p. 41; William of Newburgh, Historia rerum anglicarum 3.20, 1:264–65; Marino Sanudo, Liber secretorum 3.9.8, ed. Bongars, pp. 193–94, trans. Lock, pp. 307–8.



Prefabricated Artillery

61

Saladin’s rapid campaign through Syria in 1188, which included the sieges of Latakia, Saone, Shughr-Bakas, Bourzey, Trapessac and Baghras, there is little doubt that he travelled with a train of artillery. His rapid progression, the direct references to the use of significant stone-throwers, their inclusion at each siege and the suggestive, if not always conclusive, remarks that they were hauled from siege to siege all make this such a near certainty that it need not be further discussed here.46 So if some Muslim armies were campaigning through SyriaPalestine with prefabricated artillery in the second half of the twelfth century and supplies to allow for the prefabrication of artillery are evident from the mid-twelfth century, it remains to assess the degree to which the Franks had institutionalized the use of reusable artillery. The most-cited example supporting the Franks’ use of prefabricated artillery in the twelfth century is that of Richard I of England’s campaigning during the Third Crusade. Before he arrived at Acre, according to Roger of Hoveden, Richard had perariae, among other engines, prepared on Sicily to be taken with the fleet to assist his upcoming war efforts in the Holy Land. Apparently corroborating this, Baha al-Din notes that war machines were found within a number of captured ships that Richard had sent ahead of his departure for Acre.47 Given the general shortage of suitable timber in Palestine, exacerbated by the circumvallation of the Frankish forces then besieging Acre, these advanced measures were a near necessity. Previously, the earliest deliberate example of this practice would appear to be the importation of large timbers by the Venetian fleet that arrived in the Levant in 1123, the lumber being put to use soon after at the 1124 siege of Tyre.48 There seems little reason to doubt that Richard had at least the raw materials, if not certain prefabricated components, prepared to provide him with artillery before he left Sicily for Palestine.49 Furthermore, he may even have collected 46

47

48

49

See Baha’ al-Din, History of Saladin, pp. 31–32, 83–88; ‘Imad al-Din, Conquête de la Syrie, pp. 124–24; Ibn al-Athir, Chronicle, 2:345–55. Intriguingly, the sources omit any mention of artillery during the preceding sieges of coastal Tortosa and Jabala. Roger of Hoveden, Chronica, 3:72, trans., 2:174; Baha’ al-Din, History of Saladin, p. 147. William of Newburgh also states that Richard prepared engines (machinae bellicae), William of Newburgh, Historia rerum anglicarum 4.20, 1:350. Fulcher of Chartres, Historia Hierosolymitana 3.14, pp. 656–57, trans., 239. Cf. the movement of engines by Genoese fleets when they sailed against the Pisans in 1120 and Muslim Minorca in 1146, Caffaro, Annales, in Annali Genovesi di Caffaro e de’ suoi continuatori, ed. Luigi Tommaso Belgrano, vol. 1, Fonti per la storia d’Italia (Genoa, 1890), pp. 16, 33; trans. Martin Hall and Jonathan Phillips, in Caffaro, Genoa and the Twelfth-Century Crusades (Farnham, 2013), pp. 59, 69. These engines may even have been used on Cyprus, see Richard of Devizes, The Chronicle of Richard of Devizes, in Chronicles of the Reigns of Stephen, Henry II, and Richard I, ed. Richard Howlett, Rolls Series 82, vol. 3 (London, 1886), p. 425. Cf. Ambroise, L’estoire de la guerre saint ll. 1765–2064, ed. Gaston Paris (Paris, 1897), pp. 47–56; trans. Marianne Ailes, The History of the Holy War: Ambroise’s Estoire de la Guerre Saint, ed. Marianne Ailes and Malcolm Barber (Woodbridge, 2003), pp. 56–61; Itinerarium 2.38–41, pp. 199–204, trans., pp. 191–95.

62

Michael S. Fulton

ammunition for these engines before departing: the island’s hard volcanic rock would add a considerable advantage when cast against the soft kurkar fortifications of the Palestinian coast. One of these stones, fired by one of Richard’s perieres/petrariae, was supposedly collected and shown to Saladin after killing twelve of Acre’s defenders.50 Although the notion of importing projectiles may seem far fetched, there is evidence for a precedent at the Sicilian siege of Alexandria in 1174. In this oft-overlooked episode of the crusades, a Sicilian force, which apparently imported engines on board its fleet, attempted to invade Egypt in conjunction with a revolt against Saladin. Along with its three stone-throwing engines, this force brought conspicuous black stones that were thrown with apparent effect.51 Richard’s use of his fleet as a logistic advantage was not restricted to transporting his artillery to Palestine. Soon after Acre was taken in July 1191, Richard quickly prepared his artillery for transport.52 In May 1192, he used his fleet to move his artillery southwards down the coast, while he marched his army overland from Ascalon, to besiege Darum.53 The maneuver allowed Richard to conclude the siege successfully within four days. Richard seems to have replicated this practice at Nottingham in 1194 with equal effectiveness.54 Possibly influenced by this success, Saladin seems to have employed a similar tactic not long after. On 28 July 1192, Saladin’s forces arrived at Frankish Jaffa and on the following day Saladin ordered the assembly of artillery to facilitate siege efforts. Working through the night, his engineers made two engines serviceable by 30 July and another two on the following day.55 As in Richard’s siege of Darum, it was mining rather artillery that allowed the stronghold to be taken, but the

50

51 52 53 54

55

Ambroise, L’estoire de la guerre saint ll. 4791–4800, p. 128, trans., pp. 98–99; Itinerarium 3.7, p. 219, trans., p. 209. Ambroise notes that these were sea-stones, cf. Vegetius, Epitoma rei militaris 4.8, ed. Carl Lang (Leipzig, 1885), pp. 133–34. Abu Shama, Le livre des deux jardins, 4:164–66; al-Maqrizi, Ayyubid Sultans of Egypt, pp. 48–50. See also Baha’ al-Din, History of Saladin, p. 50; Ibn al-Athir, Chronicle, 2:229–30. Ambroise, L’estoire de la guerre saint ll. 5379–82, p. 144, trans., p. 106; Itinerarium 4.2, p. 240, trans. p. 227. Ambroise, L’estoire de la guerre saint ll. 9151–9373, pp. 245–51, trans., pp. 155–58; Itinerarium 5.39, pp. 352–55 trans., pp. 316–18. Roger of Hoveden, Chronica, 3:238–40, trans., 2:314–16; The Great Roll of the Pipe for the Sixth Year of the Reign of King Richard the First, Michaelmas 1194 (Pipe Roll 40), ed. Doris M. Stenton, Pipe Rolls Society Publications 43 (London, 1928), p. 43. These episodes bear similarities to the logistic effectiveness that characterizes Edward I’s 1304 siege of Stirling, see Peter of Langtoft, The Chronicle of Pierre de Langtoft, ed. Thomas Wright, Rolls Series 47, vol. 2 (London, 1868), pp. 357–59; Joseph Bain, ed., Calendar of Documents Relating to Scotland (Edinburgh, 1884), nos. 1491, 1498–1500, 1504, 1510, 1519, 1539, 1560, 1569, 1599, 1680, pp. 387, 388–90, 395, 400, 405, 406, 419–20, 452; A. Z. Freeman, “Wall–Breakers and River–Bridgers: Military Engineers in the Scottish Wars of Edward I,” Journal of British Studies 10 (1971), 13–15. Baha’ al-Din, History of Saladin, pp. 217–19; Itinerarium 6.13, pp. 400–3, trans., pp. 349–51; Ambroise, L’estoire de la guerre saint ll. 10807–46, pp. 289–90, trans., p. 176.



Prefabricated Artillery

63

geography and the rate at which these engines were brought into action suggest that Saladin was again employing prefabricated artillery. Thus, there is strong evidence that, at least in certain instances, powerful figures, both Christian and Muslim, were transporting and employing prefabricated artillery in the Holy Land by the end of the twelfth century. This roughly corresponds with the origin of this practice in Western Europe. In addition to Richard’s siege of Nottingham, Roger of Wendover notes the inclusion of wagons loaded with engines of war in Philip Augustus’ army in Flanders, just before the battle of Bovines in 1214, and as part of Louis IX’s invasion of Brittany in 1231.56 Matthew Paris, who rarely mentions siege engines in the original component of his chronicle, that beyond 1235, also notes the presence of wagons bearing engines of war in 1242, presumably those used by the French army against Fontenay.57 In his autobiographical chronicle, James I of Aragon states that he was transporting and employing prefabricated artillery throughout his reign (1213–76).58 Farther east, more than 300 perieres and mangoniax, as well as other war machines, were sensationally reported to have been transported by the fleet of the Fourth Crusade as it left Venice in 1202.59 Throughout the thirteenth century, evidence for the Frankish use of prefabricated machines in the East remains sporadic. Ibn al-Athir is alone among the sources, both Muslim and Frankish, to claim that the Franks used artillery against the Ayyubid stronghold of Mount Tabor in 1217.60 Soon after, likely in 56

57

58

59

60

Roger of Wendover, Flores historiarum, ed. Henry G. Hewlett, Rolls Series 84, 3 vols. (London, 1886–89), 2:106; 3:13; trans. J. A. Giles, Roger of Wendover’s Flowers of History, 2 vols. (London, 1849), 2:299, 541. In the latter case, the wagon-loaded engines fell victims to an ambush by Henry of Brittany and Ralph of Chester; the ambush was supposedly enough to drive Louis to the bargaining table the following year. Matthew Paris, Chronica majora, ed. Henry Richards Luard, Rolls Series 57, 7 vols. (London, 1872–83), 4:195, 202, 206, 412; trans. J. A. Giles, Matthew Paris’s English History: From the Year 1235 to 1273, 3 vols. (London, 1852–54), 1:408, 414, 417, 422. Philip Augustus may even have maintained stockpiles of prefabricated engines at Chinon and Falaise. J. F. Fino, “Machines de jet médiévales,” Gladius 10 (1972), 25–43, here 39–40. For an examination of artillery maintained by the English crown during the thirteenth century, based strictly on Chancery and Exchequer records, see David S. Bachrach, “English Artillery 1189–1307: The Implications of Terminology,” English Historical Review 121 (2006), 408–30. Instances of such include Albero and Lizana (1220), Ponzano and Lascellas (1227), Majorca (1229), Cullera and the towers of Montcada and Museros (1235), Valencia and Silla (1238), Lizana (1267), as well as by the bishop of Tarragona at Ibiza (1235) and a company with Don Ferdinand at Villena (1240), James I of Aragon, The Book of Deeds of James I of Aragon: A Translation of the Medieval Catalan Libre dels Fets 15, 28–29, 69, 192–95, 197, 202–3, 262–63, 461–63, 125–26, 311, trans. Damian Smith and Helena Buffery (Ashgate, 2003), pp. 30–31, 47–48, 93, 177–78, 180–81, 183–84, 219, 326–27, 134–35, 246. Like Baybars, James I appears to have frequently built artillery on location as well. Geoffrey of Villehardouin, Conquête de Constantinople, avec la continuation de Henri de Valenciennes, ed. and trans. M. Natalis de Wailly, 3rd ed. (Paris, 1882), p. 42; trans. M. R. B. Shaw, The Conquest of Constantinople, in Chronicles of the Crusades (London, 1963), p. 47. Ibn al-Athir, Chronicle, 3:175. More reliable sources state that the Franks used no artillery at all, Abu Shama, Le livre des deux jardins, 5:163–64; Eracles 31.12, p. 324; Oliver of Paderborn, Relatio magistri Oliveri Coloniensis scolastici de expeditione Jherosolimitana, in Die Schriften

64

Michael S. Fulton

1220, Mu‘azzam ‘Isa employed artillery against Frankish Caesarea and ‘Atlit, although where this was originally constructed is unclear. Three perrieres were employed during the successful attack on the former, according to the Eracles account, and Oliver of Paderborn states that no fewer than a trabuculum, three petrariae and four mangonelli were subsequently used against ‘Atlit.61 The number of machines and established precedent suggest that these engines were likely brought against their respective targets from Damascus or another stronghold, although there is no direct evidence to prove this.62 The same also appears true for the multiple engines used concurrently by the Franks against Damietta through 1218–19.63 While it seems sensible to propose that the Franks brought some engines with them to Egypt, it is at least as likely that others were built on location, possibly by harvesting the necessary materials from disassembled ships. In any case, the engines apparently had minimal impact on the fortifications of Damietta, which may support Ibn al-Athir’s claim that the Franks burnt their artillery, along with their baggage, at Mansura following their failed push south towards Cairo.64 There is evidence that engines were stockpiled in Acre by the second quarter of the thirteenth century. In a letter denouncing Frederick II, the patriarch of Jerusalem lists among the emperor’s transgressions that he had the “engines of war, which for a long time had been kept in Acre for the defense of the Holy Land, to be secretly carried into his vessels,” adding that some of these were even sent to his friend, al-Kamil of Egypt.65 It is possible that these engines had been specifically intended for the defence of Acre, rather than designed to be taken on campaign, though this is questionable in light of the construction of further engines in Acre during the 1280s, under Odo Poilechien, to be used

61

62

63

64

65

des Kölner Domscholasters, späteren Bischofs von Paderborn und Kardinal–Bischofs von S. Sabina , ed. Hoogeweg (Tubungen, 1894), pp. 288–90; al-Maqrizi, Ayyubid Sultans of Egypt, p. 165. Eracles 32.5, p. 334; Oliver of Paderborn, Historia Damiatina 53, in Die Schriften des Kölner Domsholasters, pp. 254–55. Although the Eracles account places these attacks in 1218, and there may have been smaller engagements in this year, the significant sieges of both most likely occurred in 1220. Cf. Abu Shama, Le livre des deux jardins, 5:178; Abu’l-Fida’, Resume des histoire, p. 94. Damascus is the most obvious candidate, given its use as an arsenal for such siege weapons by the mid-twelfth century; however, these engines may also have come from one of a number of strongholds which are known to have become important Mamluk arsenals, such as ‘Ajlun and Kerak. See Oliver of Paderborn, Historia Damiatina 10–32, in Die Schriften des Kölner Domsholasters, pp. 175–228; Eracles 31.14–32.14, pp. 326–46; Ibn al-Athir, Chronicle, 3:176–82; al-Maqrizi, Ayyubid Sultans of Egypt, pp. 166–81. Ibn al-Athir, Chronicle, 3:181. The episode foreshadows Louis IX’s similar withdrawal from Mansura in April 1250, cf. Ibn al-Furat, Ayyubids, Mamlukes and Crusaders: Selections from the Tarikh al-Duwal wa’l-Muluk, trans. U. and M. C. Lyons with introduction and notes by Jonathan Riley-Smith, vol. 2 (Cambridge, 1971), p. 27; al-Maqrizi, Ayyubid Sultans of Egypt, p. 307. Matthew Paris, Chronica majora, 3:183.



Prefabricated Artillery

65

by Charles of Anjou against Henry II of Cyprus.66 Although there are suggestive indications that the Franks used prefabricated artillery during the thirteenth century, especially during the Seventh Crusade,67 the very small number of significant sieges prosecuted by the Franks prevents any comprehensive picture of this. On the other hand, there is abundant evidence that this practice was embraced by the Mamluks. Even before the rise of Baybars, Damascus continued to service the Ayyubids with its arsenal of engines. In 1240, following his agreement with the Franks to surrender Safed and Beaufort, al-Salih Isma‘il of Damascus ordered engines to be brought up from Damascus to besiege the garrison of the latter, which refused to surrender the castle to the Franks.68 Although strongholds such as Kerak and ‘Ajlun were also used as arsenals during the Ayyubid period, the proliferation of this practice reached its pinnacle through the centralized organization and logistical strength of the Mamluks. In late February 1265, Baybars ordered a number of mangonels of the maghribi (‘western’) and firanji (‘Frankish’) types to be assembled at al-Auja.69 Over the course of two days, four large and other small ballistic engines were completed here while further artillery, workmen and masons were summoned. This arms build-up was then employed against Frankish Caesarea, which fell early in March. Although artillery played a relatively small part in the siege, despite its emphasis in the sources, additional engines continued to be summoned from the surrounding castles through the course of the siege.70 This account, which comes from Ibn ‘Abd al-Zahir and Ibn al-Furat, is interesting because it indicates that a degree of construction or assembly took place before the artillery components were moved to Caesarea. And although it is tempting to see the appeals for further machines as a rhetorical device, this seems not to have been the case, as Saif al-Din al-Zaini arrived with engines from Subayba after Caesarea had

66

67

68 69

70

Templar of Tyre, Cronica del Templare di Tiro 437–38, ed. Laura Minervini (Naples, 2000), pp. 326–28; trans. Paul Crawford, The ‘Templar of Tyre’: Part III of the Deeds of the Cypriots (Aldershot, 2003), p. 86. There are indications that Louis brought siege engines up the Nile with him and Joinville’s account of his negotiations with al-Mu‘azzam Turanshah and the Mamluks in 1250 suggest that siege engines were still being moved with the army, Rothelin 62, p. 596; Jean of Joinville, Vie de Saint Louis 301–3, 359, 370, ed. and trans. Jacques Monfrin (Paris, 1998), pp. 148–51, 176–77, 182–83. Rothelin 32, p. 552. These engines are now generally equated in the Arabic nomenclature to counterweight trebuchets, see Hill, “Trebuchets,” p. 104; Paul Chevedden, “King James I the Conqueror and the Artillery Revolution of the Middle Ages,” Institut d’estudis Catalans: memories de la seccio historico-arqueologica 91 (2011), 317; Peter Thorau, Sultan Baibars I. von Agypten (1987), trans. P. M. Holt, The Lion of Egypt: Sultan Baybars I and the Near East in the Thirteenth Century (Harlow, 1992), p. 160 n. 4. Ibn ‘Abd al-Zahir, A Critical Edition, pp. 554–57; Ibn al-Furat, Ayyubids, Mamlukes and Crusaders, p. 69. See also Badr al-Din al-Ayni, Extraits du livre intitule le collier de perles, in RHC Or, vol. 2a (Paris, 1887), pp. 219–20.

66

Michael S. Fulton

fallen.71 Three weeks after arriving at Caesarea, Baybars turned his force southwards against Hospitaller Arsuf. When considering prefabricated artillery, it is perhaps tempting to envisage quite large counterweight trebuchets, reminiscent of the sizable engines employed during the fourteenth century. Arsuf provides a clear reality check for these imaginative presumptions. Over 2,700 stone projectiles have been recorded during the excavations of Arsuf castle.72 Among the 800 stones that have been weighed and measured, over 80% have a mass of less than 20 kg, and there are five times as many stones weighing less than 3 kg as weighing more than 70 kg.73 The quantity and the disparity in weight suggest that there may have been only one counterweight engine present, while the remainder were traction trebuchets. This may also correspond with the accounts of Ibn ‘Abd al-Zahir and Ibn al-Furat following, which repeatedly reference the artillery being worked by pulling: the traction force applied to the ropes hanging from the short end of the beam that rotated the throwing arm.74 These, then, were not heavy breaching weapons but, rather, ones of an anti-personnel scale. However, their effect is still apparent on the pocked walls of Arsuf.75 Despite their necessarily low velocity and accompanying force of impact, nearly all of the projectiles were imported, sourced from limestone at the edge of the Samarian hills about 15 km to the east, and most of the measured sample were dressed.76 This logistical exercise is not mentioned in the sources but is, as with the arrival of additional engines, in this case from Damascus,77 testament to Baybars’ organization in this respect. As at Caesarea only weeks earlier, so as not to become reliant on any single source for his artillery, it seems that in addition to these imported engines, he had others built anew in situ. The number of projectiles with a mass of less 71 72 73

74 75

76

77

Ibn al-Furat, Ayyubids, Mamlukes and Crusaders, p. 72. This figure excludes the surrounding fosse, which, although unexcavated, displays countless projectiles in the surface layer alone. Kate Raphael and Yotam Tepper, “The Archaeological Evidence from the Mamluk Siege of Arsuf,” Mamluk Studies Review 9 (2005), Chart 1A, p. 99. Although it is possible that some of these stones were stockpiled by the Hospitallers to be used in defence, their number suggests that the vast majority were fired against the castle. Ibn ‘Abd al-Zahir, A Critical Edition, pp. 264, 266; Ibn al-Furat, Ayyubids, Mamlukes and Crusaders, pp. 74–76. Such damage is not to be found on the talus and surviving courses of vertical masonry at Caesarea, testament to the rapid fall of the town defences and possibly also the amount of preparatory time required to assemble even light machines. Subsequent damage to the citadel has made it difficult to assess impact signatures here, where artillery fire was most concentrated, according to the sources. Raphael and Tepper, “Mamluk Siege of Arsuf,” pp. 87–88. The greater effect of the hard limestone relative to the local kurkar needs no explanation. This is similar to the Sicilians’ use of volcanic stones, which they brought with them, at Alexandria in 1174 and Richard I’s use of the same at Acre in 1191. Furthermore, Baha’ al-Din’s note that stones were brought from distant places and riverbed to serve as the ammunition for Saladin’s engines at Jaffa in 1192 indicates that there was an established precedent for this in Palestine. Baha al-Din, History of Saladin, p. 219. Ibn al-Furat, Ayyubids, Mamlukes and Crusaders, p. 76.



Prefabricated Artillery

67

than 20 kg reveals both the importance of these locally constructed machines, possibly responsible for throwing those stones weighing less than 10 or 15 kg, and also the surprising number of stockpiled engines that the Mamluks had in the region. The heavier stones can likely be assigned to these engines. Counterweight trebuchets were almost certainly responsible for hurling those stones weighing in excess of about 30 kg.78 Baybars seems to have again undertaken a level of prefabrication in the field in 1266. While awaiting the return of raiding detachments, he had his force near Acre construct artillery, which was then distributed among his emirs for transport, while Saif al-Din al-Zaini was again ordered to bring up additional artillery, from Damascus this time.79 Intriguingly, the Damascene engines were supposedly transported on men’s shoulders, as camels were unable to carry them for some reason. Although some historians have rendered this to mean that the components were too heavy, if they were carried by men, rather than beasts, this seems just as likely to suggest the opposite. Likewise, the prolific archaeological evidence supporting the use of smaller engines at Arsuf also supports the less popular latter scenario. By comparison, Reynald of Chatillon seems to have had few problems using camels to transport a whole squadron’s-worth of disassembled ship components overland to the Red Sea in 1182–83.80 Thirty-eight stones have been identified as projectiles through the course of excavations at Safed; however, unlike Arsuf, Safed remained in use through the Mamluk period. Although most were discovered in Mamluk layers of stratigraphy, these cannot be firmly tied to Baybars’ siege of the castle in 1266. Barbé has divided these stones into two groups: ‘ampoule’ stones, which except for one are between 2 kg and 17 kg, being chisel-dressed and ovoid in shape with one flattened end; and spherical stones, 3 kg to 38 kg, being hammer-dressed.81 As a whole, well over 80% of the stones are less than 10 kg, the mean weight of spherical stones alone being 12.6 kg (13.6 kg excluding the exceptional small one). If the projectiles used against Safed were of a similar size, this would align 78

79 80

81

Among the largest 20% of the measured projectiles, ranging from around 25 kg to almost 80 kg, there are few notable concentrations throughout the weight range, despite around twice as many weighing under 50 kg as over. This implies that there was minimal calibration and that speed was valued in their production and shaping rather than a systematic standardization which would ensure any counterweight engines were capable of achieving a consistent range and high degree of accuracy. Ibn ‘Abd al–Zahir, A Critical Edition, pp. 587–88, 591; Ibn al-Furat, Ayyubids, Mamlukes and Crusaders, pp. 89, 91. Abu Shama, Le livre des deux jardins, 4:230–35; Ibn Jubayr, The Travels of Ibn Jubayr, trans. R. J. C. Broadhurst (New Delhi, 2001), pp. 51–52; Ibn al-Athir, Chronicle, 2:194, 289–90. See also Bernard Hamilton, The Leper King and His Heirs: Baldwin IV and the Crusader Kingdom of Jerusalem (Cambridge, 2000), pp. 179–85. Hervé Barbé, ‘Le château de Safed et son territoire a l’époque des croisades’ (unpublished PhD thesis, Hebrew University of Jerusalem, 2010), pp. 351–52, my thanks to Dr. Barbé for providing me with his study. The exception to the ‘ampoules’ being 17 kg and that of the spherical stones being 0.5 kg. The fine dressing of the ‘ampoule’ stones leaves their purpose open to question.

68

Michael S. Fulton

them quite closely with those measured at Arsuf. It likewise stands to reason that the manjaniq maghribi, noted in the sources at both Caesarea and Safed, may have been responsible for throwing the limited number of heavy stones found at Arsuf and Safed, while the majority of the artillery at both locations were smaller traction trebuchets.82 As a whole, the evidence strongly suggests that Baybars’ artillery through the 1260s was fairly light and was thus employed in a predominantly supportive role. Baybars’ use of Damascus as an arsenal continued throughout his life. According to Abu Shama, in 1267 he conveyed artillery out of Egypt, presumably built or stockpiled there, that was destined for Damascus.83 These engines, or at least the beams that would later compose them, may have been among those dispatched from Damascus to Beaufort ahead of siege operations there in 1268. These may have been the first two engines erected there, joined by some twenty-four others over the following days. This latter group, at least in part, likely reveals the continuation of the trend of building smaller engines on site to complement, presumably heavier, prefabricated artillery. In the case of Beaufort, if Baybars moved his artillery to the high ground of the Templar citadel-outwork after its abandonment, the implication follows that these were again reasonably light engines.84 Following the siege’s successful conclusion, it is likely that the transportable machines were dispatched, with the baggage, back to Damascus. Without this burden, Baybars was able to lead a lighter force over the Lebanon mountains and eventually take Antioch. The apparent absence of significant artillery at the 1268 siege of Antioch seems to confirm that by this point the sultan was reliant on his logistical network to provide him with prefabricated engines, the hostile terrain having prevented their transport in this instance, rather than building his most significant engines anew at each siege. In 1271, after concluding the famous siege of Crac des Chevaliers,85 Baybars used carts, on top of which he apparently rode, to transport the timbers of his mangonels to the nearby castle of ‘Akkar.86 This appears to be the first instance in which carts are explicitly noted as having been used to move Mamluk siege engines. In a subsequent letter sent by Baybars to Bohemond VI of Tripoli,

82 83 84 85

86

Ibn ‘Abd al-Zahir, A Critical Edition, pp. 555, 591; Ibn al-Furat, Ayyubids, Mamlukes and Crusaders, pp. 69, 91. Abu Shama, Le livre des deux jardins, 5:205. Ibn ‘Abd al-Zahir, A Critical Edition, pp. 640, 642–44; Ibn al-Furat, Ayyubids, Mamlukes and Crusaders, pp. 110–11. The siege of Crac des Chevaliers reveals little regarding prefabrication. The initial period of time required by the Mamluks to set up their artillery as well as that around its possible movement during a later phase of the siege offer only very ambiguous and hypothetical indications. For discussions of this siege see Paul Deschamps, Les Château des Croises en Terre-Sainte, 3 vols. (Paris, 1934, 1939, 1973), 1:132–34; D. J. Cathcart King, “The Taking of Le Krak des Chevaliers in 1271,” Antiquity 23 (1949), 83–92; Thorau, Lion of Egypt, Appendix 5, pp. 265–66. Ibn ‘Abd al-Zahir, A Critical Edition, pp. 748–49; Ibn al-Furat, Ayyubids, Mamlukes and Crusaders, p. 148. See also al-‘Ayni, Extraits, p. 242.



Prefabricated Artillery

69

and formerly of Antioch, half of its length is spent boasting about this feat of transportation.87 Although the distance between Crac and ‘Akkar is only about 27 km, the challenge was the admittedly rough terrain in between. What this reveals with certainty is that Baybars judged the potential value of using these engines against ‘Akkar, be it their physical or psychological power, to be sufficient to warrant the trouble associated with their laborious movement. Accordingly, his enthusiasm and use of carts indicate that these engines were probably larger than those employed in 1265 and 1266, and thus the largest engines yet transported by Baybars in Syria-Palestine. The archaeological evidence from the later siege of Montfort seems to confirm this. After taking Crac des Chevaliers and ‘Akkar, Baybars returned briefly to Damascus. Possibly collecting stockpiled artillery from his Syrian capital, he then moved to Safed, where he gathered its store of engines, and then made for Montfort, stronghold of the Teutonic Knights.88 The topography of Montfort permits only two positions from which artillery can be erected against the castle: one is along the spine of the spur to the east and the other is from the far side of the southern wadi. In each case the nearest suitable position to place artillery is in excess of 200 m away, though both enjoy a slight advantage in elevation.89 Although this distance could possibly have been reached by machines of similar size as those that assailed Arsuf, the projectiles found at Montfort are of an entirely different scale. Excavations in 1926 uncovered roughly shaped spherical stones that ranged from 22 cm to 37 cm in diameter.90 One of these larger stones, currently housed in the Metropolitan Museum in New York, has a mass of 66 kg.91 They are all hewn from what appears to be the local limestone. Although there is evidence for the use of projectiles of this size at Arsuf, there is no indication that they were fired from any kind of distance. To propel stones of this mass the required 200 m to reach Montfort, where they were found, would involve a considerable amount of force, much more than the material evidence supports was exerted at any previous siege in the Levant.92 87 88 89

90 91 92

Ibn ‘Abd al-Zahir, A Critical Edition, pp. 749–51 Ibn al-Furat, Ayyubids, Mamlukes and Crusaders, pp. 148–49. Ibn ‘Abd al-Zahir, A Critical Edition, pp. 755–56; Ibn al-Furat, Ayyubids, Mamlukes and Crusaders, p. 151. When the keep of Montfort was constructed c. 1230, its designers judged its spine-facing front to be the most vulnerable, which may account for the donjon’s rounded front, creating additional outward depth, similar to the towers anchoring the south end of Crac’s inner line of thirteenth-century defences. The comparatively thin exterior walls of the court that extends west of the donjon (less than 2 m thick, compared to the up to 7 m thick walls of the donjon) indicate that these were considered relatively safe from sappers, given the topography, and out of range of contemporary artillery. Bashford Dean, “The Exploration of a Crusaders’ Fortress (Montfort) in Palestine,” The Metropolitan Museum of Art Bulletin 22 (1927), pp. 23, 38. My thanks to Donald LaRocca at the MET for communicating the particulars of their Dean stone, then on display (2013). The architectural layout and practicalities of using stones of such size in defence almost certainly rule out the possibility that these stones were part of a defensive stockpile. Accordingly, the

70

Michael S. Fulton

When viewed as a whole, the practice of using prefabricated and deconstructable artillery developed from a luxury belonging to the twelfth century’s most wealthy, to being the dominant means of employing artillery by the end of Baybars’ reign. The wealth and authority of the Mamluk sultan allowed him to finance machines with sufficient value to warrant their reuse as well as to command the logistical network required to do so. While the practice of storing timber and other materials to be used in wartime was something that predated the First Crusade, it took the persona and wealth of great men to develop this. First evident with Zanki and Nur al-Din, then Saladin and Richard I of England, it was not until the Mamluk period that Baybars institutionalized the practice. Building upon his predecessors’ groundwork, Qalawun embraced this tactic. In 1285 he brought up artillery from Damascus and other castles of western Syria against Margat.93 Although they are noted as being carried on the shoulders of men, the inclusion of three of the ‘Frankish’ type, as well as at least seven other stone-throwers, and the evident impact signatures still discernable in the masonry of the donjon, imply that machines of significant size were employed. Qalawun was, furthermore, able to call on experts in siege warfare, revealing the degree to which Mamluk poliorcetics had been professionalized. After the siege’s successful conclusion the engines and other war materials were recommissioned to a defensive role within Margat, as had been done by Baybars at Safed in 1266. When al-Ashraf moved on Acre in 1291, the wealth, logistical capabilities and technology required to amass his impressive arrangement of artillery had already been bequeathed to him. When he dispatched his mustering orders, artillery was as much a part of these as was manpower. The sultan personally marched out of Egypt with engines and with men trained specifically to reassemble them. Across Syria his emirs collected the further artillery. Al-Muzaffar Mahmud of Hama requisitioned an engine from Crac which is said to have required one hundred cartloads to transport.94 Even if this number is exaggerated, the difficulties that were encountered in moving the carts from Crac to Damascus in the winter of early 1291, which Abu’l-Fida’ (who was present) claims took a month, are understandable. At Acre, at least four and as many as fifteen engines of the ‘Frankish’ (counterweight) type were erected against the city, supposedly capable of throwing projectiles that weighed a quintal.95

93 94

95

discovery of these stones firmly dictates the size and state of development that these Mamluk machines had reached by 1271. Abu’l-Fida’, Memoirs of a Syrian Prince, p. 12. Ibid., p. 16. This engine was apparently named al-Mansuri or ‘Victorious’. The engine may have been abstractly named or dubbed in honour of its financier. Although this was a popular cognomen for notable Muslims throughout this period, it may suggest that the engine was built during Qalawun’s reign. Abu’l-Marsin, selection trans. in Francesco Gabrieli, Storici Arabi delle Crociate (1957), trans. E. J. Costello, Arab Historians of the Crusades (New York, 1989), p. 347; Templar of Tyre, Les Gestes des Chiprois 490, ed. Gaston Raynaud (Paris, 1887, repr. Osnabruk, 1968), p. 243, trans. Crawford, pp. 105–6. This reference is provided by both the Arabic and French source. If



Prefabricated Artillery

71

In addition to the largest engines a number of smaller ones were also set up to assail the city’s defenders. Even by contemporary standards the quantity of artillery was viewed as exceptional by witnesses on both sides. In the late seventeenth century, stone projectiles 30 cm in diameter, attributed to the 1291 siege, could still be seen beyond the trace of Acre’s medieval walls.96 More recently, many stones of this type have been discovered through twentieth-century excavations.97 Unfortunately, neither the size nor the type of stone seems to have been recorded, or at least published. During further excavations in 2002, eighteen spherical stones 20 cm to 42 cm in diameter, presumed to be ballistic ammunition, were uncovered and a further fifty-two stones, all tied to the crusader period, were observed during surface inspections around the excavation area.98 In each case, the projectiles were all identified beyond the outer thirteenth-century city wall. As most of the excavated stones have been contextually tied to Frankish stratigraphy, they support the greater trend of developing Mamluk artillery. The quantity of finds is reminiscent of Arsuf, although the size of many of these projectiles bears more resemblance to those found at Montfort. Both the sources and the archaeology indicate that the Mamluks fired larger engines in greater numbers against Acre than against any other Frankish stronghold. The much slower rate of fire of the counterweight trebuchets was supplemented by that of smaller traction engines, which, despite their comparative weakness, remained a staple of siege warfare throughout the period of the crusades. Unlike Arsuf, Acre was reoccupied during the modern period and urban sprawl spread beyond the medieval-Ottoman city in the twentieth century, obscuring most of the spent Mamluk ammunition. Although the larger stones are those most evident and celebrated, firmly indicating the number and scale of certain significant engines, the presence of smaller engines and established Mamluk practice of building such on site should not be overlooked. By the basic principles of entropy, cross-seasonal hydrolyses and disinterest, these smaller projectiles, likely quite recognizable to a participant of the First Crusade, have been the first to be lost to obscurity. Thus, by 1291 the prefabricated artillery that al-Ashraf was able to call upon was likely the most significant yet employed in the Levant.99 Aided by

96 97 98 99

interpreted as one hundred units, using the Syrian ratl this is translatable to the immense figure of 185 kg, but a more modest 41 kg or 44 kg if the Iraqi or Egyptian ratl is used. The Cypriot qintar appears to be the largest measure, at 226 kg. For the Muslim measures, see Walther Hinz, Islamische Masse und Gewichte, trans. M. Ismail Marcinkowski, Measures and Weights in the Islamic World (Kuala Lumpur, 2003). For the Cypriot qintar, see Nicholas Coureas, Assizes of the Lusignan Kingdom of Cyprus (Nicosia, 2002), p. 58. Henry Mandrell, A Journey from Aleppo to Jerusalem in 1697 (Beirut, 1963), pp. 72–73. Benjamin Kedar, “The Outer Walls of Frankish Acre,” ′Atiqot 31 (1997), 160–61; Moshe Dothan, “′Akko, 1976,” Israel Exploration Journal 26 (1976), pp. 207–8. Leea Porat, “Akko, Dov Gruner Street: Final Report,” Hadashot Arkheologiyot 117 (2005), 243–46. Its effectiveness was not unappreciated, evident in the specific and extensive provisions for

72

Michael S. Fulton

his wealth and centralized position, he fully exploited the technological and logistical framework cultivated by Baybars. Despite his contributions, Baybars founded neither the practice of stockpiling prefabricated siege artillery nor that of effectively employing it. Damascus was the region’s most obvious center for stockpiling the long timbers required to construct swing-beam artillery, something recognized by both its independent Seljuk lords and its subsequent overlords. Testament to this preparedness, in 1140 the city was capable of furnishing lumber suitable for the framework of a siege tower which was constructed by the Franks against Banyas.100 Even if these stockpiles were originally intended for use in a defensive capacity, this practice predated the arrival of the Franks. In his account of the siege of Jerusalem in 1099, William of Tyre reflects that the defenders’ artillery was superior to that of the assailing Franks, as it had been constructed using timber collected before the crusaders arrived.101 Despite its ever-increasing power, artillery remained a largely supporting weapon throughout the period of the crusades.102 In the late twelfth century, great figures such as Saladin and Richard I were able to invest in engines valuable enough to be transported, but it was the Mamluks who institutionalized this practice. The Franks lacked the finances and cohesiveness to employ much of an offensive capacity through the thirteenth century and it is, rather, in Europe that the genesis of this process can be seen, through figures such as Philip II Augustus of France, as well as its fruition, manifest under figures like Edward I of England.

artillery recommended by Marino Sanudo in his suggestions for re-acquiring Palestine, Marino Sanudo, Liber secretorum 2.4.16, 19–22, 3.15.1, pp. 104, 127–36, 262, trans., pp. 104, 127–36, 418. 100 William of Tyre, Cronique 15.9–11, 2:686–91, trans., 2:108–12; Ibn al-Qalanisi, Damascus Chronicle, pp. 260–61. Both authors also emphasize the use of artillery during this siege. 101 William of Tyre, Cronique 8.8, 13, 1:395–96, 403–4, trans., 1:354, 361–63. 102 For a more general discussion of thirteenth-century siege warfare and artillery’s role within it see Christopher Marshall, Warfare in the Latin East, 1192–1291 (Cambridge, 1992), pp. 210–61.

4 Some Notes on Ayyūbid and Mamluk Military Terms1 Rabei G. Khamisy

Many of the Arabic terms that were used by Ayyūbid and Mamluk historians to describe military objects and actions – armies, wars, sieges, engines and so forth – have been discussed by scholars, sometimes in detail.2 However, several important terms which were extensively used in Ayyūbid and Mamluk sources have not been dealt with in depth; presumably scholars have assumed that their meanings are obvious. Yet some of these terms can have nuanced meanings, depending on context, and in fact have sometimes been translated imprecisely. Such terms merit a closer look. This study will focus on two such terms: ḥajjār (‫حجار‬, pl. ḥajjārīn/ḥajjārūn) and naqqāb (‫نقاب‬, pl. naqqābīn/naqqābūn). These terms generally refer to skilled workers who performed various construction and engineering tasks. Ḥajjār is derived from the noun ḥajar (‫ )حجر‬which means stone;3 thus, the ḥajjār is a person who works with stones.4 Naqqāb is derived from the root N.Q.B (‫)ن ق ب‬, 1

2

3

4

My profound thanks are extended to Yad Hanadiv for awarding me the Rothschild Fellowship, to the Council for Higher Education – Planning and Budgeting Committee for awarding me the VATAT Fellowship, to Haifa University for its generous grants, and to Cardiff University for its hospitality. These allowed me to carry out my post-doctorate under optimal research conditions. I would like to dedicate this article to my uncle, the late Faraj Khamīsy, who passed away while I was writing it. See for example Reuven Amitai, “Foot Soldiers, Militiamen and Volunteers in the Early Mamluk Army,” in Texts, Documents and Artefacts: Islamic Studies in Honour of D. S. Richards, ed. Chase F. Robinson (Leiden, 2003), pp. 233–49; David Ayalon, “Ḥarb – The Mamlūk Sultanate,” in The Encyclopedia of Islam, 2nd ed. (hereafter EI²), 3:184–90; idem., “Studies on the Structure of the Mamluk Army – I,” in Muslims, Mongols and Crusaders: an Anthology of Articles Published in the Bulletin of the School of Oriental and African Studies, ed. Geoffrey R. Hawting (London, 2005), pp. 42–67; idem, “Studies on the Structure of the Mamluk Army – II,” in ibid., pp. 68–96; idem, “Studies on the Structure of the Mamluk Army – III,” in ibid., pp. 97–130; Clifford E. Bosworth, “Mutaṭawwiʿa,” in EI², 7:776–77; Claude Cahen, “Aḥdāth,” in EI², 1:256; H. A. R. Gibb, “The Armies of Saladin,” in Studies on the Civilization of Islam by H. A. R. Gibb, eds. Stanford Shaw and William Polk (London, 1962), pp. 74–90. For the word ḥajar see M. Plessner, “Ḥadjar,” in EI², 3:29–30. See also Ibn Manẓūr, Lisān al-ʿarab, ed. ʿAbd Allāh al-Kabīr, Muḥammad Ḥasb Allāh and Hāshim al-Shādhilī, 6 vols. (Cairo, 1981), 2:781; Rohi Baalbaki, Al-Mawrid: a Modern Arabic-English Dictionary, 7th ed. (Beirut, 1995), 454. I was unable to find any mention of the word ḥajjār in contemporary dictionaries. However,

from DR 4 Khamisy.indd 73

16/07/2015 17:09

74

Rabei G. Khamisy

which means, inter alia, digging (making a hole), and in our case means undermining/sapping. Therefore, the naqqāb is a person whose job is to engage in undermining or sapping and the term may be translated as miner or sapper. These terms may appear to be quite clear cut, and Gibb claimed that the ḥajjārīn manned the mangonels and catapults, while Amitai suggested that the ḥajjārīn were the masons who cut stones for the mangonels.5 Moreover, the term ḥajjār has usually been imprecisely translated, as will be demonstrated in this study. In fact, the use of the two terms in the Ayyūbid and Mamluk sources under consideration here raises questions as to the exact nature of these men’s work. We will try to identify the specific roles of these men in various contexts, especially in times of war, and attempt to point out the differences between them.6 To do so, I shall refer to just a handful of basic contemporary sources.7 For the Ayyūbid period I will refer to ʿImād al-Dīn al-Iṣfahānī (1125–1201) and Bahāʾ al-Dīn Ibn Shaddād (1145–1234), both of whom joined Saladin for long periods and described many of his military acts. For the Mamluk period I will refer to Muḥyī al-Dīn Ibn ʿAbd al-Ẓāhir (1223–92), biographer of Baybars and Qalāwūn, and ʿIzz al-Dīn Ibn Shaddād (1217–85), who wrote a biography of Baybars. However, references to other historians – such as Ibn al-Athīr (1160–1233), Abū Shāma (1203–68), Ibn Wāṣil (1208–98), al-Yūnīnī (1242– 1326), Baybars al-Manṣūrī (1247–1325), al-Nuwayrī (1278–1332), Ibn al-Furāt (1334–1405), al-Maqrīzī (1364–1442), Ibn Taghrī (1410–70) and others – will be included when necessary. Ḥajjār and Naqqāb in Texts Relating to the Ayyūbid Period Bahāʾ al-Dīn Ibn Shaddād mentioned the term ḥajjār several times. When describing the destruction of Ascalon in 1191, he wrote, “‫وذكر بعض الحجارين‬ ‫ أن عرض البرج الذي ينقبون فيه مقدار رمح‬، ‫” للسلطان – رحمه هللا – وانا حاضر‬, which means: “Some ḥajjārīn mentioned to the Sultan – God rest his soul – when I was

5 6

7

modern dictionaries such as al-Mawrid translate it as stonecutter, stone-dresser, stonemason, quarrier and quarryman (Baalbaki, Mawrid, p. 453). We prefer to translate it as stone worker. Gibb, “Armies of Saladin,” p. 83; Amitai, “Foot Soldiers,” p. 236. According to Gibb’s suggestion, the catapults were called ʿarrāḍāt in Arabic. It should be noted that no mention of a ḥajjāra (pl. ḥajjārāt) or naqqāba (pl. naqqābāt) – the feminine forms of the terms – has been found, indicating that these jobs were monopolized by men, as were the majority of jobs that required physical strength. Usually there is some relationship between the different sources. In general, the later historians commonly cite the earlier ones. See for example Reuven Amitai-Preiss, “In the Aftermath of ʿAyn Jālūt: The Beginnings of the Mamlūk-Īlkhānid Cold War,” al-Masāq 3 (1990), 12–13; idem, “ʿAyn Jālūt Revisited,” Tārīḫ 2 (1992), 142, 148–150, “An Exchange of Letters in Arabic between Abaγa Īlkhān and Sultan Baybars (A.H. 667 / A.D. 1268–69),” Central Asiatic Journal 38 (1994), 13–15, 28; idem, Mongols and Mamluks: The Mamluk-Īlkhānid War 1260–1281 (Cambridge, 1995), pp. 4–6; and idem, “The Convention of Tegüder Ilkhanti Islam,” Jerusalem Studies in Arabic and Islam 25 (2001), 19. For other material see also Linda Northrup, From Slave to Sultan: The Career of al-Manṣūr Qalāwūn and the Consolidation of Mamluk Rule in Egypt and Syria (678–689 A.H./1279–1290 A.D.) (Stuttgart, 1998), pp. 25–58.

from DR 4 Khamisy.indd 74

16/07/2015 17:09



Ayyūbid and Mamluk Military Terms

75

present, that the thickness of the tower they were sapping was a spear’s length.”8 In this case it is clear that the ḥajjārīn were engaged in undermining the walls. This term was translated into eighteenth-century Latin as destructorum (destroyers),9 a word which provides only a general description, simply informing the reader that the men telling the story had worked on destroying the tower walls. In the French translation published in the Recueil des historiens des croisades, the word was literally translated as Pierre.10 Wilson translated it into English as stonemasons,11 which is similar to the French.12 In Richards’s more recent translation, the word is rendered as masons.13 It should be noted that Ibn Shaddād clearly stated that he was an eye-witness to this exchange, meaning that the men he mentions in this specific instance really are ḥajjārīn. As we shall see below, other terms – such as naqqāb – were not always used precisely, sometimes being applied to other groups of workers. In describing the same event – the destruction of Ascalon – Bahāʾ al-Dīn Ibn Shaddād said that the Sultan “‫عزم على الرحيل وعلى ان يخلف في عسقالن حجارين ومعهم‬ ‫خيل تحميهم مستقصون في الخراب‬,”14 meaning that “he intended15 to make a move

8

9 10 11 12

13 14

15

Bahāʾ al-Dīn Ibn Shaddād, al-Nawādir al-sulṭāniyya wa’l-maḥāsin al-yūsufiyya: sīrat Ṣalāḥ al-Dīn, ed. Jamāl al-Dīn al-Shayyāl, 2nd ed. (Cairo, 1994), p. 282; Bahāʾ al-Dīn Ibn Shaddād, Anecdotes et beaux traits de la vie du sultan Youssof (Salah ed-Din), in Recueil des historiens des croisades: historiens orientaux (hereafter RHC Or.), 5 vols. (Paris, 1872–1906), 3:266. This and other information were cited by Abū Shāma and al-ʿAynī (1360–1451). See Shihāb al-Dīn al-Dimashqī Abū Shāma, Kitāb al-rawḍatayn f ī akhbār al-dawlatayn al-nūriyya wa’lṣalāḥiyya, ed. Ibrāḥīm Shams al-Dīn, 5 vols. (Beirut, 2002), 4:163; Badr al-Dīn al-ʿAynī, ʿIqd al-jumān f ī ta’rīkh ahl al-zamān, ed. Maḥmūd Riziq Maḥmūd, 4 vols. (Cairo, 2010), 2:204. Albertus Schultens, Vita et Res Gestae Sultani Almalichi Alnasiri Saladini (Lugduni Batavorum, 1732), p. 201. This is the earliest edition of the manuscript. RHC Or. 3:266. In the French translation of Abū Shāma, it was translated to maçon (RHC Or. 5:42). Charles Wilson, Saladin; or What befell Sultan Yûsuf, The Life of Saladin, in Palestine Pilgrims’ Text Society (hereafter PPTS), 13 vols (London, 1890–97), 13:299. Wilson based his translation on the French edition (see D. S. Richards, The Rare and Excellent History of Saladin or al-Nawādir al-sulṭāniyya wa’l maḥāsin al-yūsufiyya by Bhāʾ al-Dīn Ibn Shaddād, Crusader Texts in Translation 7 (Aldershot, 2001), p. 8), and so it is closer to the French translation than to the Arabic original. Richards, Rare and Excellent History, p. 180. In all editions except the Latin one, the word “‫ ”بعض‬was translated as “one” instead of “some”. Ibn Shaddād, Nawādir, p. 282. See also Abū Shāma, Rawḍatayn, 4:164, in RHC Or. 5:43; Jamāl al-Dīn Ibn Wāṣil, Mufarrij al-kurūb f ī akhbār banī ayyūb, ed. Jamāl al-Dīn al-Shayyāl, 3 vols. (Cairo, 1953–72), 2:370; al-ʿAynī, ʿIqd, 2:204. The word “‫( ”عزم‬intended) was translated by Richards as “decided” (Richards, Rare and Excellent History, p. 180). However, the more correct Arabic word for decided is “‫”قرر‬ (qarrara). To “decide” to do something is one step after “intending” to do it. However, we know that Saladin had indeed later decided to leave Ascalon, and did so. For twelfth-century meanings see al-Ḥumayrī al-Yamānī, Shams al-ʿulūm wa dawāʾ kalām al-ʿarab min al-kulūm, ed. Ḥusayn al-ʿUmarī, Muṭahhar al-Iryānī and Yūsuf bin ʿAbd Allāh, 12 vols. (Damascus, 1999), 7:4519. Also the classical dictionary of Ibn Manẓūr gave the same meaning (Ibn Manẓūr, Lisān, 4:2932–33).

from DR 4 Khamisy.indd 75

16/07/2015 17:09

76

Rabei G. Khamisy

and to leave ḥajjārīn at Ascalon to continue16 the demolition while protected by cavalry.” Here the ḥajjārīn were working on destroying the city, most probably by undermining the walls. Again this word was translated as destructores in the Latin edition,17 while in the French it was translated as mineurs (miners).18 As these men really were engaged in undermining, mineurs seems apt here. Wilson also translated ḥajjārīn as miners,19 whereas Richards preferred stonemasons.20 Later on, Ibn Shaddād mentioned that the sultan employed ḥajjārīn for the purpose of destroying Ramla and Lydda that same year.21 This time, the word was translated into Latin as coementarios (masons, stonemasons, stonecutters).22 In the French edition it was translated as ces gens,23 “these people,” without explaining the nature of their job. The same is true of Wilson’s translation.24 Richards, on the other hand, preferred the word stonemasons.25 The last time that Ibn Shaddād mentioned ḥajjārīn, it appeared alongside naqqābīn. He wrote that after concluding the truce with the Franks in 1192, Saladin ordered Ascalon to be demolished, and sent a group of naqqābīn and ḥajjārīn to carry out the work.26 The Latin translation used the terms coniculariorum (miners) and caementariorum (masons, builders), respectively.27 In the French edition these terms were translated as mineurs and maçons (masons),28 which were indeed rendered into English as miners and masons by Wilson, while Richards preferred sappers and stonemasons.29 This specific reference to naqqābīn and ḥajjārīn together serves as conclusive evidence that while ḥajjārīn may have engaged in mining work, as mentioned above, they were not exactly miners; for that specific role there existed a more precise term, naqqābīn. This may lead us to ask what the exact occupation of the ḥajjārīn was, both here and in instances when they were mentioned alone. This will be discussed below. However, in order to do so, we must look into the precise meaning of naqqābīn and try to define these men’s actual work. The term naqqābīn was used several times by Bahāʾ al-Dīn Ibn Shaddād, first appearing when he wrote that in the destruction of al-Dārūm in 1192, King 16 17 18 19 20 21 22

23 24 25 26 27 28 29

The word “‫ ”مستقصون‬means “continuing”, rather than “finishing” as translated by Richards (see Richards, Rare and Excellent History, p. 180). Schultens, Vita, p. 201. RHC Or., 3:266. Maçons in the translation of Abū Shāma (RHC Or. 5:43). PPTS, 13:299. Richards, Rare and Excellent History, p. 180. Ibn Shaddād, Nawādir, p. 284; al-ʿAynī, ʿIqd, 2:205. Schultens, Vita, p. 202. The word used here, coementarios, appeared as caementarios later in this edition. The word could also refer to “someone who builds using cement,” in which case this builder would use cement and stone. RHC Or. 3:268. PPTS, 13:301. Richards, Rare and Excellent History, p. 181. Ibn Shaddād, Nawādir, p. 349. Schultens, Vita, p. 262. RHC Or. 3:348. PPTS, 13:387; Richards, Rare and Excellent History, p. 323.

from DR 4 Khamisy.indd 76

16/07/2015 17:09



Ayyūbid and Mamluk Military Terms

77

Richard employed naqqābīn from Aleppo whom he had captured during the crusader siege of Acre.30 These men worked on undermining the walls31 and both the title of their job and their actual work indicate that this term should be translated as sappers or miners. In fact, the word was translated into Latin as quassatores (destroyers), into French as sapeurs and into English as sappers.32 Later, in the descriptions of the Ayyūbid expedition against Jaffa that same year, naqqābīn/naqqābūn are mentioned on several occasions, always undermining walls.33 The term used in the Latin translation was usually cunicularios, whereas the French always used mineurs, which Wilson translated as miners. Richards, however, preferred the word sappers.34 Later on, Ibn Shaddād mentioned that naqqābīn were employed to destroy the walls of Bayt Dajan and Yāzūr.35 The same translations were used again in the various editions,36 as is true of the reference to a hundred naqqābīn sent to Ascalon by the sultan to demolish it after concluding the truce.37 Naqqābīn are last mentioned by Bahāʾ al-Dīn Ibn Shaddād in concurrence with the last mention of ḥajjārīn given above. It would appear from the context that the aforementioned “hundred naqqābīn” are none other than these naqqābīn and ḥajjārīn who were mentioned together as having been sent to demolish Ascalon. This could mean that ḥajjārīn were sometimes also designated as naqqābīn by the same historian, who had witnessed the events and whose detailed descriptions are reliable. Alternatively, it is possible that alongside those hundred naqqābīn, the sultan also dispatched a number of ḥajjārīn who were not mentioned by Ibn Shaddād in this instance; however, we 30

31 32

33

34 35 36 37

For these Aleppan sappers and the different stories of their capture, see Nicolas Prouteau, “‘Beneath the Battle’? Miners and Engineers as ‘Mercenaries’ in the Holy Land (XII–XIII Siècles),” in Mercenaries and Paid Men: The Mercenary Identity in the Middle Ages, ed. John France (Leiden, 2008), pp. 107, 108, 111. See discussion below in the conclusion. Ibn Shaddād, Nawādir, p. 313; Abū Shāma, Rawḍatayn, 4:176; idem, RHC Or. 5:54. Schultens, Vita, p. 227; RHC Or. 3:301; PPTS, 13:337; Richards, Rare and Excellent History, p. 203. The translator of Abū Shāma preferred the word mineurs (RHC Or. 5:54). The word used in the Latin translation, quassatores, is derived from quasso, meaning to shake, which was usually used when destroying walls by hitting them with a kind of machine. It is also used to describe the effect of an earthquake. Thus, it seems that the editor wanted to describe the naqqābūn as destroyers in this specific event. Ibn Shaddād, Nawādir, pp. 330–33; al-ʿAynī, ʿIqd, 2:241. Al-ʿAynī also quoted a passage from Mirʾāt al-zamān of Sibṭ Ibn Jawzī where he mentioned the naqqābūn as taking part in the siege (al-ʿAynī, ʿIqd, 2:239). Schultens, Vita, pp. 244–47; RHC Or. 3:324–25, 327; PPTS, 13:361–64; Richards, Rare and Excellent History, pp. 217–19. Ibn Shaddād, Nawādir, p. 338; Ibn Wāṣil, Mufarrij, 2:400. Schultens, Vita, p. 253; RHC Or. 3:335; PPTS, 13:373; Richards, Rare and Excellent History, p. 224. Ibn Shaddād, Nawādir, p. 348. Schultens, Vita, p. 261; RHC Or., 3:347; PPTS, 13:386; Richards, Rare and Excellent History, p. 231. The description of Ibn Shaddād was again cited by Abū Shāma and naqqābūn was translated as mineurs (Abū Shāma, Rawḍatayn, 4:193; ibid, RHC Or. 5:80). This event was also mentioned by other historians. See Ibn Wāṣil, Mufarrij, 2:408; Shihāb al-Dīn al-Nuwayrī, Nihāyat al-arab f ī funūn al-adab, vols. 28–31, ed. Najīb Fawwāz and Ḥikmat Fawwāz (Beirut, 2004), 28:293; al-ʿAynī, ʿIqd, 2:245.

from DR 4 Khamisy.indd 77

16/07/2015 17:09

78

Rabei G. Khamisy

find the first suggestion to be more probable, because the ḥajjārīn were sometimes called naqqābīn, as we shall presently demonstrate. As is clear from these descriptions, naqqābīn had a specific role which entailed undermining walls, but apparently this task was sometimes performed by ḥajjārīn or with their help. However, naqqābīn also had the role of preparing the substructure underneath the walls during sapping work. For example, Abū Shāma, quoting al-Iṣfahānī, wrote that the naqqābūn undermined the walls of Bayt al-Aḥzān (Vadum Jacob), shored them up and then burnt them.38 Ibn al-Athīr wrote that in the occupation of Āmid in 579/1183 and Darb Sāk in 584/1188 by Saladin, the naqqābūn undermined a wall and one of the towers, respectively, and shored them up.39 Both times, the Arabic term used to describe the situation was “‫( ”علقوه‬ʿallaqūhu, “shored it up”), which means here that the object was shored up by something supporting it from below. A detailed description of the undermining and the wood construction was given by Usāma Ibn Munqidh (1095–1188), who observed this act during the siege of Kafar Ṭāb in 1115.40 This sheds light on the nature of the naqqābūn’s job, which necessitated a wide knowledge of construction in wood41 – usually the work of builders (bannāʾīn, ‫)بنائين‬.42 We cannot confirm that this specific task was always done by naqqābūn, but if it was, this would mean that on many occasions they probably oversaw everything that was happening in the breaches. It is also possible that some of these miners were actually ḥajjārīn (although these would not be the men who built the wood constructions; see discussion below).43 The events mentioned by Bahāʾ al-Dīn Ibn Shaddād were also recorded by ʿImād al-Dīn al-Iṣfahānī, who, like Ibn Shaddād, witnessed and participated in them all. However, al-Iṣfahānī’s use of assonance style occasionally meant 38

39

40 41 42 43

Abū Shāma, Rawḍatain, 3:25; idem, RHC Or. 4:204. Naqqābūn was translated as mineurs. Ibn Wāṣil abridged the same text of al-Iṣfahānī, but did not mention that the naqqābūn shored up the walls (Ibn Wāṣil, Mufarrij, 2:82). Ibn al-Athīr, al-Kāmil f ī al-taʾrīkh, vols. 9 and 10, ed. Muḥammad al-Daqqāq (Beirut, 1987), 10:119, 173; Ibn Wāṣil, Mufarrij, 2:135, 267. For Āmid see ʿIzz al-Dīn Ibn Shaddād, al-Aʿlāq al-khaṭīra f ī dhikr umarāʾ al-shām wa’l-jazīra, vol. 3, part 2, ed. Yaḥyā ʿAbbāra, Ta’rīkh al-jazīra (Damascus, 1978), p. 515, and for Darb Sāk see al-Nuwayrī, Nihāya, 28:274. See also Abū Shāma, Rawḍatayn, 3:99; 4:23. Here Abū Shāma quoted al-Qāḍī al-Fāḍil, who ignored the information about the shoring up of the walls. In the translated version of Abū Shāma only Darb Sāk was mentioned and in this instance naqqābūn was translated as mineurs (RHC Or., 4:376). In this instance, the miners were from Khurasān. See Usāma Ibn Munqidh, Kitāb al-iʿtibār, ed. Hartwig Derenbourg (Paris, 1886), pp. 54–55. Ibid. See also Prouteau, “‘Beneath the Battle’?” pp. 107–9, 112. On this point, unlike on several other points, I agree with Prouteau. The builders usually needed to have a wide knowledge of wood construction in order to set up scaffolds and frames. Naqqābīn have been also described as filling the breaches with wood and setting fire to them in Jaffa (Ibn Shaddād, Nawādir, pp. 331–32) and Tiberias (Ibn al-Athīr, Kāmil, 10:96) (perhaps a description of the shoring up). For all the editions and translations of Ibn al-Athīr’s manuscript see D. S. Richards, The Chronicle of Ibn al-Athir for the Crusading Period from al-Kamil fi’lTa’rikh, Part 1, Crusader Texts in Translation 13 (Aldershot, 2006), p. 6.

from DR 4 Khamisy.indd 78

16/07/2015 17:09



Ayyūbid and Mamluk Military Terms

79

that he chose and added many words solely to sustain the rhyming, sometimes at the expense of the accuracy of details. Assonance is the repetition of vowel sounds to create internal rhyming within phrases or sentences, and together with alliteration and consonance is one of the building blocks of verse. Thus, assonance style is somewhat confusing and could lead the reader to erroneous conclusions. In addition to Ibn Shaddād, al-Iṣfahānī mentioned both naqqābīn and ḥajjārīn working on the destruction of the walls of Tiberias during its siege by Saladin in 1187. In that same paragraph he stated that the naqqābūn began to undermine the walls.44 This may indicate that the naqqābūn were indeed the sappers (Massé, for example, translated the word as sapeurs),45 but it is also possible that he chose to designate both groups here as naqqābūn because this word fit in with his use of assonance. Al-Iṣfahānī did not specify the tasks performed by the ḥajjārīn in this event. However, the absence of mangonels from this siege implies that the ḥajjārīn did not prepare or launch ballista stones, as suggested by Massé, who described ḥajjārīn as “ceux qui lancent des pierres au moyen de machines” (those who throw stones using machines).46 The mangonel operator was later called manjanīqī by al-Iṣfahānī; this term appeared alongside naqqāb and ḥajjār during the siege of Tyre,47 where perhaps the ḥajjār was responsible, inter alia, for preparing ballista stones. But here Massé again translated ḥajjār as stone throwers (lanceurs de pierres),48 naqqāb as miner (mineur) and manjanīqī as

44

45

46

47 48

ʿImād al-Dīn al-Iṣfahānī, al-Fatḥ al-qussī f ī al-fatḥ al-qudsī, in al-Mawsūʿa al-shāmiyya f ī ta’rīkh al-ḥurūb al-ṣalībiyya, ed. Suhayl Zakkār, 40 vols. (Damascus, 1995), 13:5832. The same events were quoted by Abū Shāma, but the slight difference suggests that he was using al-Barq al-shāmī of al-Iṣfahānī (Rawḍatayn, 3:179; idem, RHC Or. 4:264–65). In the French edition of Rawḍatayn, the word naqqābīn was translated as mineurs and the ḥajjāriyya as les hommes des balistes. Later Massé made a nearly identical suggestion (see below and n. 5). In the letter sent to Baghdad by ʿAbd Allāh Ibn Aḥmad al-Maqdisī, informing of the victory at Ḥiṭṭīn, he mentioned the naqqābīn in the attack on Tiberias. This letter was quoted by Abū Shāma from the book of Muḥammad Ibn al-Qādisī. According to this text, al-Maqdisī was an eyewitness (Abū Shāma, Rawḍatayn, 3:190–91; in RHC Or. 4:286). Here, naqqābīn was translated into French as mineurs. ʿImād al-Dīn al-Iṣfahānī, Conquête de la Syrie et de la Palestine par Saladin, Documents relatifs à l’histoire des croisades 10, trans. Henri Massé (Paris, 1972), p. 24 (hereafter al-Iṣfahānī, Conquête). Ibid., p. 24. Prouteau understood this translation of ḥajjārīn to be a description of the Khurasāniyya, who were mentioned between the naqqābīn and ḥajjārīn. See Prouteau, “‘Beneath the Battle’?,” p. 112. It should be noted that the only two historians who mentioned the use of mangonels in the siege of Tiberias were al-Ṣafadī (1297–1366), who wrote his account some 150 years after the events, and al-ʿAynī (1360–1451), writing at least 250 years after the events. See Ṣalāḥ al-Dīn Khalīl al-Ṣafadī, Kitāb al-wāf ī bi’l-wafayāt, ed. Aḥmad al-Arnāʾūṭ and Turkī Muṣṭafā, 29 vols. (Beirut, 2000), 6:117; al-ʿAynī, ʿIqd, 2:60. However, these sources cannot be compared with the contemporary accounts mentioned above. al-Iṣfahānī, Fatḥ, 13:5882–83. The term manjanīqī is derived from manjanīq, which is mangonel in Arabic. Thus, the manjanīqī was the man who operated the mangonel. Meaning the men who throw stones using machines, as already mentioned before. Probably Massé did not prefer to repeat the sentence ceux qui lancent des pierres au moyen de machines

from DR 4 Khamisy.indd 79

16/07/2015 17:09

80

Rabei G. Khamisy

ballista crew (servants de balistes).49 If so, Massé claims that the ḥajjār and the manjanīqī worked together, with the ḥajjār responsible for launching the stones. We suggest that if they did work together, the ḥajjār was responsible for preparing the stones and the manjanīqī for launching them.50 Another task possibly handled by ḥajjārīn was to prepare the platforms on which the mangonels stood. Al-Iṣfahānī mentioned that as part of the preparations for attacking Tyre, the platforms for the mangonels were leveled (‫)وتسوية مناصب المجانيق‬.51 This job could have involved ḥajjārīn if the platforms were carved in the bedrock or partly built of stone. Al-Iṣfahānī again mentions a ḥajjār later in his description of the siege of Tyre, but without indicating the nature of his job; Massé, however, translated it as lanceurs de pierres.52 Al-Iṣfahānī also mentioned a naqqāb in his description of the siege of Ascalon,53 and in a later paragraph describing the siege of Tiberias.54 Again, the assonance dictated the choice of words. In his previous mention of the event (see above), al-Iṣfahānī mentioned both naqqābīn and ḥajjārīn. The two terms appear together twice during his description of the siege of Latakia, where no mangonels are mentioned;55 Massé, however, translated the term naqqāb as sapper on both occasions, and the ḥajjār as on jeta pierres, and balistier.56 The first translation refers to a man who threw stone, and the second clearly indicates the belief of Massé that the ḥajjār threw ballista stones by mangonels. Although al-Iṣfahānī, in his Fatḥ, did not mention ḥajjārīn as participating in the occupation of Ṣahyūn (Saone) in 1188,57 he seems to have done so in his description of the same event in al-Barq, which presumably is what Abū Shāma had quoted.58 Al-Iṣfahānī’s final mention of ḥajjārīn is also his most important description of the nature of their work in a context other than warfare. He wrote that in 1192 fifty ḥajjārīn with their salaries were sent by the prince of Mosul in order

49 50 51 52 53

54

55 56 57 58

every time he wanted to translate the word ḥajjār, and thus, in the later mentions of the term, he abstracted its translation to lanceurs de pierres. al-Iṣfahānī, Conquête, p. 65. See the suggestions by Gibb and Amitai in n. 5 above. al-Iṣfahānī, Fatḥ, 13:5884. Ibid.; al-Iṣfahānī, Conquête, p. 67. See n. 48 above. al-Iṣfahānī, Fatḥ, 13:5855. Abū Shāma, Rawḍatayn, 3:210. Here also, naqqābīn was translated as mineurs (Abū Shāma, RHC Or. 4:313). Ibn Wāṣil also mentioned the naqqābīn as controlling part of the outer fortifications (Ibn Wāṣil, Mufarrij, 2:209). al-Iṣfahānī, Fatḥ, 13:5910. Massé translated it as sapeurs (al-Iṣfahānī, Conquête, p. 95). This is actually a letter from al-Iṣfahānī to Sayf al-Islām in Yaman, informing him of the occupation of Jerusalem (al-Iṣfahānī, Fatḥ, 13:5907–16). al-Iṣfahānī, Fatḥ, 13:5939. al-Iṣfahānī, Conquête, pp. 129, 130. al-Iṣfahānī, Fatḥ, 13:5943. Abū Shāma, Rawḍatayn, 4:16–17; idem, RHC Or. 4:366. In this case the term was translated as pierre. Perhaps al-Rijāl al-ḥalabiyya (the men from Aleppo) who were mentioned here are the naqqābīn, but it is not certain. For identifying the Aleppan men as miners see Prouteau, “‘Beneath the Battle’?”

from DR 4 Khamisy.indd 80

16/07/2015 17:09



Ayyūbid and Mamluk Military Terms

81

to help rebuild the walls of Jerusalem and excavate its moat.59 They worked for half a year, during which time they certainly quarried stone from the bedrock in order to create a moat, but also dressed the stones to prepare them for use in construction work. We are specifically told that the stones taken from the moat were used for building.60 This time, Massé translated the term ḥajjārīn as stonemasons.61 Al-Iṣfahānī later wrote a chapter about thanking the prince of Mosul for sending jaṣṣāṣīn (‫ )جصاصين‬to excavate the moat;62 Gibb called these men “moat-diggers,” and Massé translated the term as terrassiers (diggers).63 In this specific instance, it appears that the ḥajjārīn and jaṣṣāṣīn were one and the same. Abū Shāma quoted a letter by al-Qāḍī al-Fāḍil,64 one of the foremost figures during the reign of Saladin, as mentioning naqqābīn taking part in the occupation of Jerusalem in 1188.65 Al-Qāḍī al-Fāḍil wrote that the naqqābīn worked to undermine Jerusalem’s walls and described their job in detail; according to him, these men pulled stones out of the walls, and used pickaxes to crush stones into dust. The work described here seems to pertain to both naqqābīn and ḥajjārīn (see below); therefore, the reference to naqqābīn in this passage might actually relate to men of both professions (further on we will suggest that a ḥajjār could be described as a naqqāb, but not vice versa). I assume that the letter’s assonance again played an important role in al-Qāḍī al-Fāḍil’s choice of words. Al Fatḥ al-Bindārī (thirteenth-century historian, ca. 1190–1245), who abstracted material from the al-Barq al-shāmī by al-Iṣfahānī, mentioned that naqqābīn were employed at Manbaj in 1175, Vadum Jacob in 1179, Sinjār in 1182 and Ascalon in 1187.66 It should be noted that both Abū Shāma’s and al-Bindārī’s descriptions mentioned above are generally taken from texts written in assonance, apart from those passages taken from Bahāʾ al-Dīn Ibn Shaddād. Accounting for the styling of words, it is possible that in most or all of these events, both naqqābīn and ḥajjārīn took part. We are also told by Ibn al-ʿAdīm (1192–1262) that in 1197 both Jabla and Latakia were ordered to be demolished by ḥajjārīn and zarrāqīn (‫زراقين‬, those who throw Greek fire), due to the Muslim fear of Frankish invasion.67 This historian wrote that Latakia was undermined 59 60 61 62 63 64 65

66 67

al-Iṣfahānī, Fatḥ, 13:6164. Ibid. al-Iṣfahānī, Conquête, p. 356. al-Iṣfahānī, Fatḥ, 13:6175. Gibb, “Armies of Saladin,” p. 90 n. 77; al-Iṣfahānī, Conquête, p. 367. C. Brockelmann and Claude Cahen, “Al-Ḳāḍī al-Fāḍil,” in EI², 4:376–77. Abū Shāma, Rawḍatayn, 3:231. Ibn Wāṣil recorded the same information, but in a different manner (Ibn Wāṣil, Mufarrij, 2:213), but he also cited the letter in a later paragraph (Mufarrij, 2:239), as was done by ʿIzz al-Dīn Ibn Shaddād (al-Aʿlāq al-khaṭīra f ī dhikr umarāʾ al-shām wa’l-jazīra, vol. 2, part 2, ed. Sāmī Dahhān, Ta’rīkh lubnān, al-urdunn wa-filasṭīn (Damascus, 1962), p. 206 (hereafter ʿIzz al-Dīn Ibn Shaddād, Aʿlāq, 2/2). Al-Fatḥ ibn ʿAlī al-Bindārī, Sanā al-barq al-shāmī, ed. Fatḥiyya al-Nabrāwī (Cairo, 1979), pp. 99, 169–70, 208, 308. Ibn al-ʿAdīm, Zubdat al-ḥalab min ta’rīkh ḥalab, in Zakkār, Mawsūʿa, 16:7303. The term zarrāqīn will be discussed in a future study.

from DR 4 Khamisy.indd 81

16/07/2015 17:09

82

Rabei G. Khamisy

and shored up; this is the first and only indication we have that the shoring up could be done by either ḥajjārīn or/and zarrāqīn. However, it is likely that many other men participated in this effort, some of whom were perhaps familiar with this work – probably naqqābīn or bannāʾīn. In Ibn Wāṣil’s version of the event, he claimed that ḥajjārīn and naqqābīn undertook destruction work in Beirut, without mentioning Latakia.68 Ibn al-ʿAdīm mentioned naqqābīn several times when describing events in the years 1126, 1147 and 1176, all of which were connected to warfare;69 the first ended without any conflict, while the other two involved walls being undermined, leading to the surrender of the garrisons inside the castles. Ibn Wāṣil referred to ḥajjārīn and naqqābīn several times in passages dealing with events that had occurred before his birth,70 but later on mentioned ḥajjārīn three times and naqqābīn four times. Both ḥajjārīn and naqqābīn were working together to destroy Jerusalem’s walls in 1219 because of fears of Frankish action, and naqqābīn were mentioned as sapping the walls of Āmid in 1232.71 However, this historian did not witness either event; in 1219 he was a mere twelve years old and not present in Jerusalem, while his description of the second event is based on quotes from a letter detailing what had happened, and his phrasing makes it quite clear that he was not there himself.72 Here, again, the letter was written in assonance and might not be indicative. His other mentions of both terms are related to events in the Mamluk period and will be discussed below. When comparing these works to Ibn al-Athīr’s, we find a stark difference: Ibn al-Athīr did not use the term ḥajjār at all, although he did use naqqāb on several occasions in describing events that took place between 498/1104 and 617/1220.73 Few of these events were mentioned by Bahāʾ al-Dīn Ibn Shaddād 68 69 70 71

72 73

Ibn Wāṣil, Mufarrij, 3:71. Ibn al-ʿAdīm, Zubda, 16:7172, 7203, 7242 See nn. 14, 35, 37, 38, 39, 53, 65, 68, 73, 117. See also his mention of the naqqābūn during the sieges of Ḥimṣ and Kawkab in 1130 and 1188, respectively (Ibn Wāṣil, Mufarrij, 1:42; 2:274). Ibn Wāṣil, Mufarrij al-kurūb f ī akhbār banī ayyūb, vols. 4–5, ed. Ḥasanayn Muḥammad Rabīʿ (Cairo, 1972–77), 4:32; 5:21. Ellenblum, while dealing with the text regarding Jerusalem, translated ḥajjārīn as stonecutters and naqqābīn as cave diggers (Ronnie Ellenblum, “Frankish Castles, Muslim Castles, and the Medieval Citadel of Jerusalem,” in In Laudem Hierosolymitani: Studies in Crusades and Medieval Culture in Honour of Benjamin Z. Kedar, ed. Iris Shagrir, Ronnie Ellenblum and Jonathan Riley-Smith (Aldershot, 2007), p. 104). Ibn Wāṣil, Mufarrij, 5:18. Ibn al-Athīr, Kāmil, 9:79, 128, 331, 354; 10:76, 96, 119, 153, 155, 173, 177, 404. I will not deal with these events. In each case, Richards translated the term as sappers (Richards, Chronicle, part 1, pp. 86, 140, 373; idem, The Chronicle of Ibn al-Athir for the Crusading Period from al-Kamil fi’l-Taʾrikh, Part 2, in Crusader Texts in Translation, vol. 15 (Aldershot, 2007), pp. 23, 243, 266, 291, 329, 331, 352, 356; idem, The Chronicle of Ibn al-Athir for the Crusading Period from al-Kamil fi’l-Taʾrikh, Part 3, in Crusader Texts in Translation, vol. 17 (Aldershot, 2008), p. 208). In the RHC edition the term was translated as sape, sapeurs and mineurs (see for example RHC Or. 1:225, 444, 471, 638, 696, 700, 730, 737). Some of these events were also recorded by Ibn Wāṣil, who like Ibn al-Athīr used the term naqqāb. See for example his description of the occupations of al-Rahā (Edessa) and al-ʿAzīma in 1144 and 1148 respectively (Ibn Wāṣil, Mufarrij, 1:94, 114).

from DR 4 Khamisy.indd 82

16/07/2015 17:09



Ayyūbid and Mamluk Military Terms

83

and al-Iṣfahānī, who actually witnessed them and provide us with the most detailed descriptions of Saladin’s wars (especially Ibn Shaddād, who did not use assonance and thus seems to have been at liberty to use the accurate terms). This, again, might strengthen our assumption that the ḥajjār had been sometimes designated as naqqāb, but not vice versa. Al-Nuwayrī, another later historian, wrote that in 1227 the city of Tinnīs was demolished by naqqābīn and ḥajjārīn sent by al-Malik al-Kāmil.74 This is further proof that these two groups of men worked together in destroying walls. They were also mentioned simultaneously in a letter written by al-Malik al-Nāṣir, informing the Caliph al-Mustanṣir of the occupation of Jerusalem in 1240. He mentioned both the naqqābīn and ḥajjārīn as undermining the walls, but also noted that the ḥajjārīn pulled stones out of David’s Tower.75 Ḥajjār and Naqqāb in Texts Relating to the Mamluk Period Information about the two terms under discussion can be also found in texts relating to the Mamluk period, and especially in those describing the reigns of Baybars (1260–77) and Qalāwūn (1279–90), who captured most of the Frankish strongholds and cities. One should note, however, that Ibn Wāṣil stated that ḥajjārīn worked on the demolition of Damieta in November 1250, at the very beginning of the Mamluk period.76 It is worth noting that Ibn Wāṣil was probably an eye-witness of the event, as he was in Cairo and the region at the time.77 Ibn ʿAbd al-Ẓāhir wrote that in 1263 Baybars’s men undermined a tower near Acre, and, later in the text, added that Baybars rode again to the tower which was shored up by the naqqābūn.78 Here again, the naqqābūn seem likely to have 74 75

76

77 78

al-Nuwayrī, Nihāya, 29:91; al-ʿAynī, ʿIqd, 4:186. ʿIzz al-Dīn Ibn Shaddād, Aʿlāq, 2/2:226–27. Part of this description was translated by Max van Berchem, where he translated the naqqābūn as sapeurs and the ḥajjārūn as mineurs (Max van Berchem, Matériaux pour un corpus inscriptionum arabicarum: deuxième partie Syrie du sud, 3 vols. (Genève, 2001), 1:137–38, n. 5). The text of Ibn Shaddād was quoted in its entirety by Ibn al-Furāt with only minor changes (Ibn al-Furāt, Ta’rīkh al-Duwal wa’l-Mulūk, trans. Ursula Lyons and Malcolm Cameron Lyons, ed. Jonathan Riley-Smith, Ayyubids, Mamlukes and Crusaders: Selections from the Ta’rīkh al-Duwal wa’l-Mulūk of Ibn al-Furāt, 2 vols. (Cambridge, UK, 1971), 1:76, 2:62). The event was mentioned also by the contemporary Ibn Wāṣil, who wrote his own account without referring to naqqābīn and ḥajjārīn (Ibn Wāṣil, Mufarrij, 5:247). Ibn Wāṣil, Mufarrij al-kurūb f ī akhbār banī ayyūb, ed. ʿUmar ʿAbd al-Salām Tadmurī, 6 vols. (Sidon-Beirut, 2004), p. 145. Although not mentioned by him, Ibn al-Furāt’s account is clearly based on Ibn Wāṣil’s (Ibn al-Furāt, Ta’rīkh, 1:47, 2:38). Lyons, throughout their translation of Ibn al-Furāt’s text, used the words sappers for naqqābīn and masons for ḥajjārīn. For the presence of Ibn Wāṣil in Egypt at the time see Peter Jackson, The Seventh Crusade, 1244–1254: Sources and Documents, Crusader Texts in Translation 16 (Aldershot, 2007), p. 5. Ibn ʿAbd al-Ẓāhir, al-Rawḍ al-zāhir f ī sīrat al-malik al-ẓāhir, ed. Abdul-Aziz al- Khowaiter (al-Riyāḍ, 1976), pp. 158, 160; al-Nuwayrī, Nihāya, 30:163; Ibn al-Furāt, Ta’rīkh, 1:73, 2:59. Sadeque, who was the first to publish part of Ibn ʿAbd al-Ẓāhir’s manuscript, translated it as “shored it up with props” (Syedah Fatima Sadeque, The Slave King Baybars I of Egypt, 2 vols. (Dacca, 1956), 1:313; 2:168).

from DR 4 Khamisy.indd 83

16/07/2015 17:09

84

Rabei G. Khamisy

constructed the supports in the breaches. That same year, Baybars brought with him Egyptian ḥajjārīn, bannāʾīn, najjārīn (‫نجارين‬, carpenters) and ṣunnāʿ (‫صناع‬, craftsmen)79 to the attack on al-Karak. Ibn ʿAbd al-Ẓāhir previously mentioned that Baybars supposedly gathered them for construction work on Mount Tabor,80 but actually these ḥajjārīn were to destroy walls. The same event had been recorded by Ibn Wāṣil, who added the naqqābīn to the previously mentioned list.81 Later, during his first major invasion of the Frankish Kingdom in 1265 – just before his attack on Caesarea – Baybars sent messages to various castles, asking them to send him some men, including ḥajjārīn; however, Ibn ʿAbd al-Ẓāhir claimed that the walls of Caesarea could not be undermined because of the construction method used there. Yet later in the text we are told that the walls were indeed undermined.82 Presumably this job was done by the ḥajjārīn, who possibly also helped in cutting or preparing ballista stones; however, they did not launch those stones, as al-manjanīqiyya are mentioned alongside them in the same text.83 The work of the ḥajjārūn during the occupation of Safed is clearer; Ibn ʿAbd al-Ẓāhir wrote: “‫ – ” فتقدم الحجارون وأخذوا في النقوب‬meaning, “The stone workers (ḥajjārūn) went ahead and started digging.”84 Later on, we are told that Baybars promised the ḥajjārīn that he would give each of them one hundred dinars for every stone pulled out of the walls, up to the tenth stone. In the same passage, the biographer wrote that many breaches were made and that the naqqābūn entered those breaches, leading the sultan to grant them three hundred dinars.85 This paragraph suggests that the ḥajjārūn undertook the first and most difficult stage of undermining: extracting stones from the outer face of the walls. This task was far from easy; firstly, despite always being protected by many other soldiers and war machines, the ḥajjārūn were exposed to the defenders’ retali­ ations from within the walls. Secondly, pulling out stones – especially large ones – was much more difficult than digging in the mortar. Perhaps that is why each ḥajjār was promised a hundred dinars per stone, while the naqqābūn were collectively paid three hundred dinars. Even if, as translated and suggested by

79

80

81 82 83 84 85

We will discuss the term ṣunnāʿ in more detail later in this article, in reference to Qalāwūn’s occupation of al-Marqab, as well as in a future article. The najjārīn are carpenters who carry out work such as cutting, preparing, decorating and constructing in wood. This term also will be discussed in a future work. Ibn ʿAbd al-Ẓāhir, Rawḍ, pp. 162–63; al-Nuwayrī, Nihāya, 30:50. Sadeque translated the ḥajjārīn as stone-cutters, the bannāʾīn as masons, the najjārīn as carpenters and ṣunnāʿ as artisans. Moreover, she mistakenly identified al-Ṭūr with Tyre (see Sadeque, Slave, 1:315; 2:170). Ibn Wāṣil, Mufarrij, 6:371–72. Ibn ʿAbd al-Ẓāhir, Rawḍ, pp. 230–31; al-Nuwayrī, Nihāya, 30:168; Ibn al-Furāt, Ta’rīkh, 1:85, 2:69. Ibn ʿAbd al-Ẓāhir, Rawḍ, p. 231. Ibid., p. 257; al-Nuwayrī, Nihāya, 30:182; Ibn al-Furāt, Ta’rīkh, 1:117, 2:92. Ibn ʿAbd al-Ẓāhir, Rawḍ, p. 258; Ibn al-Furāt, Ta’rīkh, 1:117, 2:92; al-Nuwayrī did not mention the hundred dinars (see al-Nuwayrī, Nihāya, 30:183).

from DR 4 Khamisy.indd 84

16/07/2015 17:09



Ayyūbid and Mamluk Military Terms

85

Lyons, each naqqāb was individually paid three hundred dinars,86 this sum pales in comparison to the thousand dinars that the individual ḥajjār could potentially earn. We do not actually know how many stones were pulled out or how much the ḥajjārīn were ultimately paid, but the nature of Baybars’s promise suggests that it was very difficult to remove even ten stones, shedding light on the arduousness of this task. This work could be done only by someone skilled in breaking and removing large stones. Later in this passage, Ibn ʿAbd al-Ẓāhir stated that Baybars set up a tent with doctors to tend to the injured men, and that the ḥajjārīn were brought there.87 The fact that the ḥajjārīn are the only group to be mentioned by name as receiving this medical attention seems to indicate that they were particularly exposed to counter-attack. This would appear to strengthen our hypothesis about their work and its importance. Al-Yūnīnī quotes Kmāl al-Dīn Aḥmad’s letter to Ibn Khālikān, in which he wrote that the ḥajjārūn undermined the walls of Safed and set them on fire.88 This could be understood to mean that the ḥajjārīn did the burning, but we believe that this letter, like many others reporting particular conquests, is un-indicative in some respects because it is written in assonance. It is more likely that the ḥajjārīn pulled out the stones and then joined in the undermining work alongside the naqqābīn, who perhaps shored up the walls. During a later attack on the Acre region in 1267, we are told that Baybars divided “the stone workers and the people” (al-ḥajjārīn wa’l-nās)89 and sent them to destroy the buildings and wells or cisterns, and to cut down the orchards.90 Again, the demolition work seems to have been done by ḥajjārīn and with their help. In 1268, according to Ibn ʿAbd al-Ẓāhir, ḥajjārīn excavated for twenty days to access groundwater found in Jerusalem.91 This endeavor is reminiscent of the moat excavation at Jerusalem during Saladin’s reign (see above), which was another task that the ḥajjār could do when not engaged in warfare. Before attacking ʿAkkār (Giblacar), Baybars called for ḥajjārīn, and it would appear, according to Ibn ʿAbd al-Ẓāhir, that in this instance the ḥajjārīn removed stones and excavated the bedrock in order to clear a path that would allow the mangonels and other army forces to reach the castle. They may have also prepared

86

87 88 89 90 91

Ibn al-Furāt, Ta’rīkh, 2:92. It is not clear from the text whether this sum was paid to each sapper individually or to all of them collectively, but we assume that it was paid collectively. Every text stated that the naqqābūn were paid three hundred dinars, but none of them suggested that this sum was paid out to each naqqāb, as was the case with the ḥajjārīn in the same paragraph. Ibn ʿAbd al-Ẓāhir, Rawḍ, p. 158; Ibn al-Furāt, Ta’rīkh, 1:118, 2:92. Quṭb al-Dīn Al-Yūnīnī, Dhayl mirʾāt al-zamān, 4 vols. (Hyderabad, 1954–61), 2:341. For the term nās and awlād al-nās see Ayalon, “Studies, II,” pp. 76–78. In our case, the word should be translated as people. Ibn ʿAbd al-Ẓāhir, Rawḍ, p. 281; al-Nuwayrī, Nihāya, 30:188; Ibn al-Furāt, Ta’rīkh, 1:130, 2:103. Ibn ʿAbd al-Ẓāhir, Rawḍ, p. 289; al-Nuwayrī, Nihāya, 30:96.

from DR 4 Khamisy.indd 85

16/07/2015 17:09

86

Rabei G. Khamisy

platforms for the mangonels, which according to the text were set up in very problematic locations.92 No naqqābīn were mentioned in Ibn ʿAbd al-Ẓāhir’s description of the occupation of Montfort, whereas he wrote that the ḥajjārīn were promised one thousand dirhams for every stone pulled out.93 As with the attack on Safed, the sums that Baybars was ready to spend bear witness to the importance of this work. The remains of walls of the upper ward which still exist at Montfort prove that their foundations, where the mining work was done,94 are built of large stones. One breach, in the outer southern wall, is completely preserved: It is almost eight meters long, one meter high and three meters deep. The outer face of this section of the wall is built with large, partly dressed field-stones, measuring approximately one meter square. Our assumption is that the naqqābīn could do nothing with such walls unless the ḥajjārīn first removed the large stones. It is worth noting that the stone-workers were paid less here than at Safed,95 probably because the occupation of Safed was much more difficult and more important.96 The situation at Crac des Chevaliers seems to have been different. After its fall, Ibn ʿAbd al-Ẓāhir sent an assonance-styled letter to Ibn Khālikān in which he mentioned the naqqābūn as doing the mining.97 ʿIzz al-Dīn Ibn Shaddād also claimed that the job was done by the naqqābūn of al-Malik al Saʿīd.98 However, it is probable that this work was done with the help of ḥajjārīn. Before the occupation of al-Marqab (Margat) in 1285, Qalāwūn brought many ṣunnāʿ (craftsmen) to participate in the siege.99 Ṣunnāʿ could refer to skilled laborers such as metalworkers, carpenters and so on.100 Although ḥajjārīn Ibn ʿAbd al-Ẓāhir, Rawḍ, p. 380; Ibn al-Furāt, Ta’rīkh, 1:187–88, 2:148. Ibn ʿAbd al-Ẓāhir, Rawḍ, p. 386; al-Nuwayrī, Nihāya, 30:213; Ibn al-Furāt, Ta’rīkh, 1:192, 2:151. 94 The author has discussed this point elsewhere. See Rabei G. Khamisy, “Montfort Castle (Qalʿat al-Qurayn) in Mamluk Sources,” in Montfort: History, Early Research and Recent Studies, ed. Adrian J. Boas and Rabei G. Khamisy (Brill, forthcoming). 95 One dinar at the time was equal to 40 low-quality (Egyptian) dirhams or to nine Syrian dirhams of the higher quality. (Adrian Boas, Domestic Settings: Sources on Domestic Architecture and Day-to-Day Activities in the Crusader States (Leiden, 2010), pp. 222, 225 tab. 2; Robert Irwin, “The Supply of Money and the Direction of Trade in Thirteenth-Century Syria,” in Coinage in the Latin East: the Fourth Oxford Symposium on Coinage and Monetary History, ed. Peter Edbury and David Michael Metcalf, British Archaeological Reports 77 (Oxford, 1980), pp. 86–87). Thus, depending on their type and quality, a thousand dirhams would equal either 111 dinars or 25 dinars. However, it seems that the dirhams in this case were those minted by Baybars, called ẓāhirī dirhams. This type of dirham was rated at 20–28.5 dirhams per dinar (Irwin, “Supply of Money,” p. 87), meaning that the thousand dirhams promised to the ḥajjārīn in Montfort were equal to 35–50 dinars. 96 See Khamisy “Montfort Castle” (forthcoming). 97 Al-Yūnīnī, Dhayl, 2:447. 98 ʿIzz al-Dīn Ibn Shaddād, Aʿlāq, 2/2:117. 99 Ibn ʿAbd al-Ẓāhir, Tashrīf al-ayyām wa’l-ʿuṣūr f ī sīrat al-malik al-manṣūr, ed. Murād al-Kāmil (Cairo, 1961), p. 78. 100 Ibid. p. 177. 92 93

from DR 4 Khamisy.indd 86

16/07/2015 17:09



Ayyūbid and Mamluk Military Terms

87

and naqqābīn were not specifically mentioned, they were presumably among the ṣunnāʿ, because extensive mining work was done at the site and seems to have been the main reason for the castle’s surrender.101 After the occupation of Margat, the sultan took Marqiyya unopposed and ordered it to be demolished. This was done, inter alia, by the emir Badr al-Dīn Biktāsh al-Nijmī, accompanied by a hundred ḥajjārīn.102 Finally, when Qalāwūn decided to attack Acre in 1290, just before his death, he prepared large numbers of troops, including ṣunnāʿ and many ḥajjārīn.103 Important information can be found in Ibn al-Dawādārī’s account of Qalāwūn’s occupation of Tripoli in 1289. Ibn al-Dawādārī wrote that his father, who participated in this event, told him that the city was bombarded by nineteen mangonels, and that the number of ḥajjārīn and zarrāqīn present totaled 1,500.104 It is possible that in this event the ḥajjārīn were involved with mangonel operations – for example, preparing ballista stones. However, another possible reason that ḥajjārīn were mentioned alongside the mangonels and zarrāqīn is that these three were regarded as being of the utmost importance to sieges and occupations.105 It should be noted that Baybars al-Manṣūrī, a contemporary and bio­­ grapher of Qalāwūn who did not witness the event, wrote that the undermining was done by naqqābīn.106 In this case Ibn al-Dawādārī is presumably more reliable, due to his detailed account. Additionally, in many cases ḥajjārīn could be described as naqqābīn, as suggested above, and perhaps this is one such instance (see also the conclusion). 101 Ibid,

pp. 78–79. In Gibb’s discussion of Saladin’s army, he suggested that the attacker troops of the foot-soldiers were called ṣunnāʿ, which he preferred to translate as technicians. He claimed that the ḥajjārīn, naqqābīn and khurasāniyya (the men from Khurasān, which manned the “cats,” according to Gibb) are the three groups which were frequently mentioned as being ṣunnāʿ (Gibb, “Armies of Saladin,” pp. 83–84, 90 nn. 78–79). However, the term ṣunnāʿ appeared several times alongside ḥajjārīn (for example, before the attack on al-Karak in 1263, and in preparation for the attack on Acre in 1290). It seems that the ḥajjārīn are either a separate group or one of the most important elements of the ṣunnāʿ, as suggested by Gibb. The current study prefers Amitai’s translation of ṣunnāʿ, “craftsmen” (Amitai, “Foot Soldiers,” p. 246). The term ṣunnāʿ will be explored in a forthcoming study. 102 Ibn ʿAbd al-Ẓāhir, Tashrīf, p. 89. 103 Ibid., p. 177. Al-Nuwayrī explicitly mentioned only the ḥajjārīn while referring to the other professionals as “‫”وغيرهم‬, meaning “and others”. al-Nuwayrī, Nihāya 31:124. 104 Ibn al-Dawādārī, Kanz al-durar wa jāmiʿ al-ghurar: al-durra al-zakiyya f ī akhbār al-dawla al-turkiyya, ed. vol. 8, Ulrich Harmann (Cairo, 1971), p. 283. Ibn al-Dawādārī was born in 1289 and his work Kanz was written between 1309/10 and 1335/6 (Northrup, From Slave to Sultan, pp. 47–48). Amitai, based on Holt, mentioned several times that Ibn al-Dawādārī had flourished around 1338–40 (see for example Reuven Amitai-Preiss, “Whither the Ilkhanid Army? Ghazan’s First Campaign into Syria (1299–1300),” in Warfare in Inner Asian History (500–1800), ed. Nicola Di Cosmo (Leiden, 2002), p. 229). The same numbers – nineteen mangonels and 1500 ḥajjārīn and zarrāqīn – were given by al-Nuwairī, although his account is slightly different (al-Nuwairī, Nihāya, 31: 33). 105 The descriptions by contemporary historians usually mentioned the damage caused by the mangonels and the undermining, which seems to have been the most effective aspect of sieges. 106 Baybars al-Manṣūrī, Zubdat al-fikra f ī ta’rīkh al-hijra, ed. D. S. Richards (Beirut, 1998), p. 266.

from DR 4 Khamisy.indd 87

16/07/2015 17:09

88

Rabei G. Khamisy

Al-Nuwayrī mentioned that in 1282 two markets in Damascus caught fire, and ḥajjārīn took part in demolishing buildings and other structures in order to block and contain the fire.107 But his most important addition is his description of the raid on Sīs by Sultan Ḥusām al-Dīn Lājīn al-Manṣūrī (1297–99) in 1298. He reported that an emir managed to reach one wall of Nujayma castle and undermine it, pulling out three of its stones. During this action, eleven of his men and two ḥajjārīn were killed.108 This is conclusive evidence regarding the role of ḥajjārīn in removing stones to enable undermining. The fact that they pulled out only three stones (certainly large enough), and that two of them were killed alongside eleven other men, is also proof of the highly risky and difficult nature of their task. It is possible that these eleven men had been trying to protect the ḥajjārīn whilst they were doing their job. The later historian al-Maqrīzī mentioned various activities of ḥajjārīn on different occasions that were unrelated to war;109 he also noted several instances when they did participate in military action – sometimes clearly as part of mining work, and twice alongside naqqābīn.110 The last hint to be discussed in this study regarding the nature of the ḥajjārīn’s work is provided by Ibn Taghrī Birdī. While copying the description of Cairo’s boundaries, he used the word ḥajjārīn twice: the second instance is part of a phrase, dakākīn al-ḥajjārīn (the shops of the ḥajjārīn), and the text’s editor has suggested that the first instance refers to the same place, although the word dakākīn is missing.111 This might mean that the ḥajjārīn had shops (where they perhaps sold their artwork and crafts) or workshops. This information could expand our understanding of their line of work. It is probable that ʿIzz al-Dīn Ibn Shaddād, Baybars’s second biographer, had mentioned the terms ḥajjārīn and naqqābīn when describing the sultan’s invasions and occupations. Unfortunately, the biography’s first volume – relating all events prior to the year 1272, including Baybars’s acts against the Franks – is lost. The second volume includes almost no information about siege activities.112 In his other major work, al-Aʿlāq, ʿIzz al-Dīn Ibn Shaddād mentioned these terms several times, but usually did so when quoting other sources. For example, in addition to the events mentioned above,113 he wrote that Atabak Nihāya, 31:59. p. 215. 109 Aḥmad ibn-ʿAlī al-Maqrīzī, Kitāb al-sulūk li-maʿrifat duwal al-mulūk, ed. Muḥammad ʿAṭā, 8 vols. (Beirut, 1997), 2:380; 3:160, 296, 299; 5:224; 6:406. These activities included building, probably as masons, and excavating (probably digging in bedrock and removing stones). Just once are they mentioned as participating in another activity – playing an unspecified game (ibid., 3:299). 110 See ibid., 3:399, 404; 5:245; 6:121. Some of the events mentioned by Ibn ʿAbd al-Ẓāhir and al-Nuwayrī were also recorded by al-Maqrīzī; See ibid., 1:556; 2:18, 34, 44, 212. 111 Ibn Taghrī Birdī, al-Nujūm al-zāhira f ī mulūk miṣr wa’l-qāhira, ed. Muḥammad Shams al-Dīn, 16 vols. (Beirut, 1992), 4:38, 39, 38 n.5. Ibn Taghrī’s description will not be discussed here. 112 See ʿIzz al-Dīn Ibn Shaddād, Ta’rīkh al-malik al-ẓāhir, ed. Ahmad Hutait (Wiesbaden, 1983). 113 See above nn. 39, 65, 75, 98. 107 Al-Nuwayrī, 108 Ibid,

from DR 4 Khamisy.indd 88

16/07/2015 17:09



Ayyūbid and Mamluk Military Terms

89

Shihāb al-Dīn Ṭughrul asked some ḥajjārīn to quarry stones from one of the moats of Aleppo in order to widen it. One can infer from the description that this happened sometime after the year 620/1223.114 Later, he described a section of Aleppo’s walls which had been raised by the same Atabak but was not strongly built, and claimed that the Mongols succeeded in sacking Aleppo by gaining access to the city via this section, because only there did their naqqābīn succeed in their undermining.115 Conclusion Before addressing the differences and similarities between ḥajjār and naqqāb, an important observation should be made regarding sources. The sources used here can be divided into two main categories: those written in assonance and those written in freestyle form. As mentioned above, the assonance style is somewhat confusing and could lead the reader to erroneous conclusions. For example, it seems likely that in many instances where naqqābīn are mentioned alone, they actually worked alongside ḥajjārīn. One could say that, technically, these historians were not mistaken; the ḥajjārīn who had pulled out stones and sapped walls could also be described as naqqābīn. However, by not addressing minute details, these historians fail to portray the scene precisely, thus reducing the accuracy of their descriptions. It should be noted, however, that sometimes these two terms do appear simultaneously in assonance-styled texts, which may point to a distinction between them. These issues do not apply to texts written in freestyle form, as their writers are free to choose their words without any linguistic obstacles. Therefore, their descriptions are the most important and helpful for clarifying the meanings of both terms. One of the current study’s most important observations is the identification of the relationship which exists between the two terms under discussion and the events in which they appear. The term naqqāb appears only in the context of warfare, which means that this role was directly related to wartime action and the undermining of walls during sieges. The naqqāb was also skilled in wood construction. On the other hand, the term ḥajjār repeatedly appears not only al-Dīn Ibn Shaddād, al-Aʿlāq al-khaṭīra f ī dhikr umarāʾ al-shām wa’l-jazīra, vol. 1, part 1, ed. Dominique Sourdel (Damascus, 1953), p. 18 (hereafter ʿIzz al-Dīn Ibn Shaddād, Aʿlāq, 1/1). Ibn Shaddād stated that this happened sometime after 1223, and we may add that it occurred before the death of Ṭughrul on 16 October 1233 (for a short biography of Ṭughrul see al-Nuwayrī, Nihāya, 29:132). 115 ʿIzz al-Dīn Ibn Shaddād, Aʿlāq, 1/1:27. The information about the building work in 620/1223 was certainly quoted from another historian or recorded from an eye-witness, because at the time Ibn Shaddād was only six years old. However, Ibn Shaddād witnessed the Mongol occupation of Aleppo, and thus the information about the naqqābīn seems to be authentic. As these men were part of a foreign army with which he was unfamiliar, Ibn Shaddād described them collectively as naqqābīn based solely on their action of undermining the wall, although this group might have also included ḥajjārīn and other men. The same might be true regarding Ibn Wāṣil’s description of the event (Ibn Wāṣil, Mufarrij, 6:269). 114 ʿIzz

from DR 4 Khamisy.indd 89

16/07/2015 17:09

90

Rabei G. Khamisy

in descriptions of sieges and warfare, but also in many instances unrelated to military action; we are told of ḥajjārīn excavating moats, cutting stones, digging underground, demolishing buildings to try to contain fire and so on. They are also mentioned as having shops or workshops. We would therefore suggest that the term ḥajjār should generally be translated as “stone worker,” someone who is skilled in working with stone. This could be a mason, an artist, a man who carved cisterns or dug ditches – or perhaps someone experienced in all of these occupations or some combination of them. One difference between these two professions is that the ḥajjār could do the work of digging into mortar, whereas the naqqāb usually could not deal with stones. On the other hand, building wooden constructions was not one of the ḥajjār’s professional skills. Therefore, it is to be assumed that walls with found­ ations built of large stones could not be undermined without ḥajjārīn (or, at least, their undermining became much more difficult). In other words, ḥajjārīn could handle the undermining work if they were supported by men skilled in wood construction, whereas naqqābīn could not go through with the undermining if the walls contained large stones. Thus, we conclude that the term naqqābīn, when mentioned on its own, might also refer to ḥajjārīn, but not vice versa. Finally, it can be suggested that these men’s work during sieges was regarded as being essential and of the utmost importance to the sieges’ success, with the ḥajjārīn’s job usually being the most difficult, dangerous and vital. The two references to high sums that Baybars was ready to pay to the ḥajjārīn attest to this. It might be assumed, with a high degree of certainty, that in many cases mining was impossible without the presence of ḥajjārīn, especially when the foundation stones were large, as at Safed and Montfort. Relying on several Arabic and Latin sources, Prouteau assumed that the Aleppans were professional naqqābīn (miners),116 which seems to be correct. Furthermore, I should like to present several other descriptions that strengthen and broaden his argument: Ibn Wāṣil, in his description of the siege of Damascus by al-Malik al-Afḍal and al-Malik al-Ẓāhir in 1199, wrote that the Aleppan naqqābūn had reached the wall and undermined it.117 Two years later, we are told that during the attack on Manbaj Castle, al-Malik al-Ẓāhir shouted to the Aleppans (probably to encourage them), and they jumped into the moat and made some breaches in the castle walls.118 In 1203, al-Malik al-Ẓāhir sent some forces to al-Malik al-ʿĀdil in Egypt to help him defend it against a supposed Frankish invasion. These forces included, inter alia, one hundred Aleppan foot-soldiers who were categorized as ḥajjārīn, naqqābīn and zardkhāna (arsenal, or maybe men of the arsenal) (‫)مائة راجل من الحلبيين حجارين نقابين وزردخانة‬.119 This last ­reference is also the only one that our search for mentions of Aleppan ḥajjārīn has yielded. 116 Prouteau,

“‘Beneath the Battle’?”. Wāṣil, Mufarrij, 3:100. 118 Ibid., p. 120. 119 Ibid., p. 135. Both Gibb and Nicolle claim that zardkhāna is the arsenal of weapons that were distributed before campaigns. See Gibb, “Armies of Saladin,” p. 84; David Nicolle, “The 117 Ibn

from DR 4 Khamisy.indd 90

16/07/2015 17:09



Ayyūbid and Mamluk Military Terms

91

To reinforce our hypothesis regarding the ḥajjārīn’s nature of work in wartime, we would like to compare these historical findings with an example of stone-workers in the twentieth century. Twenty-two kilometers to the northeast of Acre, in the centre of the western Upper Galilee, lies Miʿilyā, a village whose population traditionally subsisted on agriculture. As the village’s lands are situated in mountainous, rocky terrain, every farmer had to break and remove large stones using traditional methods in order to cultivate the land. However, a few locals were known as specialists in this field. In Miʿilyā’s oral tradition, stories are still told of the late Riḍā Ibrāhīm Ghabrīs (1908–87), the most esteemed of these stone-workers. A man who could break any rock of any size, he was known as Ḍarrīb Shāqūf.120 In the nearby village of al-Buqayʿa, the same title was given to Farīd Ḥabīb Makhkhūl (1912–94), whose skill is recalled in a story still recited in the village today. It is said that one day, as several men were making great efforts to break a rock lying in a field, Farīd walked by. They called out to him for help. He looked at the rock for a while and then took the shāqūf, telling them that no strength was needed to break the rock. Pointing to a particular spot, he said, “You have to strike it here,” and with one blow he broke the rock, then continued to smash it into small pieces.121 In one of the author’s conversations with Farīd (in the early 1990s), he said that a stone is like a map; one has to know how to read it. He added that every kind of stone has its own weak points and therefore each stone is broken a different way. He also said that sometimes strength is needed, sometimes speed and sometimes both; but what one must always know is exactly where to strike the rock and at what angle to land the blow.122 We believe that the same principles applied to the ḥajjārīn in this study. They were familiar with the different types of stone, could “read” them and locate their weak points, and thus knew exactly how to break them – striking the right blow in terms of strength, speed and direction – and when to extract them from the wall. I believe that during the twelfth and thirteenth centuries, as well as in later periods, these professionals existed everywhere: in every village where people used to deal with stones when cultivating their lands, and in every city where Manufacture and Importation of Military Equipment in the Islamic Eastern Mediterranean (10th–14th Centuries),” in Egypt and Syria in the Fatimid, Ayyubid and Mamluk Eras, ed. U. Vermeulen and J. Van Steenbergen, 3 vols. (Leuven, 2001), 3:150. It is also possible that al-Malik al-Ẓāhir sent one hundred Aleppan foot-soldiers together with ḥajjārīn and naqqābīn supported by zardkhāna. Due to a lack of punctuation marks, the exact meaning of the text remains unclear. 120 Many thanks to the late Ḥanna Khamīsy/Ghabrīs, Dr Shukrī ʿArrāf, Mr Salīm Tūnī, Mr George Khamīsy and others for providing this information. Ḍarrīb Shāqūf means one who can hit (very well) with a large hammer. The word ḍarrīb, “hitter” or “striker,” is derived from the root Ḍ.R.B (‫“ )ض ر ب‬hitting”. Shāqūf is one name for a 10 kg hammer (also called Mahadda, which could be translated as destroyer). Both words refer to objects that can cause destruction. 121 Personal communication with Mr Mubaddā Khūrī. 122 Personal communication with the late Farīd Makhkhūl. Similar comparisons can be made to people who still live in traditional settings and work in breaking, cutting and dressing stone with tools similar to those that existed in the twelfth to fourteenth centuries AD.

from DR 4 Khamisy.indd 91

16/07/2015 17:09

92

Rabei G. Khamisy

they carried out masonry work. These men were ready to participate in wars when summoned by the lords and the sultans. The case with the naqqabīn seems to be different; because they were mentioned only as participating in wars, one can conclude that they were trained specifically to carry out the undermining and shoring up jobs. This could strengthen the suggestion by Prouteau about the tradition of creating professional sappers in Aleppo and Khurasān. However, I would like to suggest that this job was not monopolized by a specific city or area but, rather, that many emirs and lords had naqqabīn among their troops. The jobs performed by these men outside of wartime are not known, but due to their knowledge in wood construction, it is possible that they worked in building, perhaps constructing scaffolding or framing. It is also possible that they spent part of their time training for their wartime job, for which they were probably paid. Finally, although the majority of the ḥajjārīn were certainly not kept in castles, it is probable that a few of them were attached to garrisons trained to break stones. This latter group might have been a section of the naqqabīn.

from DR 4 Khamisy.indd 92

16/07/2015 17:09

5 Helgastaðir, 1220: A Battle of No Significance?1 Oren Falk

On a cool autumn Saturday in 1220, the combined forces of three warlords closed in on the northern Icelandic farmstead of Helgastaðir, trapping Bishop Guðmundr Arason and a number of his followers inside (Figure 1). This was hardly the first time in his long and tempestuous career that the nearly sixtyyear-old bishop, known to posterity as Guðmundr the Good, came under attack, nor would it be the last. Terse annals, as well as several more prolix sagas – which dignify the encounter with the title bardagi á Helgastôðum, “the Battle of Helgastaðir” – record Guðmundr’s involvement in at least four other battles, and one commando-style raid besides, in the course of his thirty-four-year episcopate (1203–37).2 This time, the hostilities were to last the better part of twenty-four hours, involving heavy barrages of missiles, hand-to-hand combat, and even the operation of some sort of siege equipment. In addition to the human toll of 1

2

The origins of this paper, and its fondest debt, may be traced back to Preben Meulengracht Sørensen’s Saga and Society: An Introduction to Old Norse Literature, trans. John Tucker (Studia Borealia, 1, ed. Gerd Wolfgang Weber; [Odense, 1993; orig. 1977]), pp. 60–61. I am deeply indebted to Arngeir Friðriksson, the current farmer at Helgastaðir, for patiently assisting my research, and to my parents-in-law, Dov and Hana Levitte, for photos and notes they took at Helgastaðir. A preliminary version of this paper was presented at a De Re Militari session at the 42nd International Congress on Medieval Studies, Kalamazoo MI, in May 2007. I am grateful to Kelly DeVries for organizing that session and for inviting me to submit a revised version for publication; he and Clifford Rogers also offered valuable critical comments. References to the Family Sagas (see n. 11 below) are to the Íslenzk fornrit editions, abbreviated as ÍF, by volume and page numbers; the four redactions of Guðmundar saga (see further n. 17 below) are abbreviated as GSA, GSB, GSC, and GSD. All translations are mine; hearty thanks to Ellert Þór Jóhansson for his assistance with deciphering a few thorny turns of Old Norse phrasing, and to Joel Anderson for reading a draft and offering perceptive comments. I retain the sagas’ seemingly free alternation between past and present tense. Except in direct quotations, Icelandic names and terms are normalized to thirteenth-century orthography. In accordance with Icelandic convention, patronyms are not treated as surnames; Icelanders are identified by their first names, primarily. All remaining errors and infelicities are mine. Guðmundr’s supporters also saw action at Víðines (1208), the See of Hólar (1209), Hólar again (1221), and Grímsey (1222), in addition to staging a daring raid to break the bishop out of jail (1219). These battles are only a fraction (albeit a sizeable one) of the fighting that took place in Iceland during these turbulent years. For the relevant annalistic entries, see Annales Reseniani, Henrik Høyers Annaler, Lögmanns-annáll, and Gottskalks Annaler, all s.a. 1220 (in Islandske Annaler indtil 1578, ed. Gustav Storm [Christiania, 1888], pp. 23, 63, 255, 326; the different annals’ idiosyncratic titles are Storm’s).

94

Oren Falk

Figure 1.  Map of Iceland, marking the two bishoprics (Skálholt and Hólar), the northern bays of Skagafjôrðr (home district of Arnórr Tumason), Eyjafjôrðr (home district of Sighvatr Sturluson), and Skjálfandi, and the approximate location of Reykjadalr, in which Helgastaðir is situated (see Figure 2).

dead and injured – among them, some of the highest-ranking combatants – the chapel at Helgastaðir would also sustain heavy damages, still visible a century later. Yet, for all its martial bluster, Helgastaðir’s claim to a place in a military hall of fame appears tenuous. The performance of the combatants was noted for neither tactical nor strategic acuity; very few casualties were sustained on either side; and the attackers’ primary quarry, the bishop himself, got away safely (as did the chief men in his entourage), even after having been effectively trounced. It seems difficult to reconcile the stridently militant language used to describe this clash with its insipid outcomes. Helgastaðir appears to lend credence to the slur often repeated against the Icelandic sagas: that, despite their pretensions to recount sober, solemn history, they are mere tales of “farmers at fisticuffs.”3 3

The phrase “bændur flugust á” is attributed to the eighteenth-century Icelandic scholar, Jón Ólafsson frá Grunnavík, and is emblematic of the disdain felt by educated Enlightenment-era Icelanders towards the deeds of their forebears, as depicted in the medieval sagas; cited by Jesse L. Byock, Medieval Iceland: Society, Sagas, and Power (Berkeley et al., 1988 [repr. 1990]), p. 45 n. 28.



Helgastaðir, 1220

95

In this paper, I consider the Battle of Helgastaðir from several points of view: military, political, social, cultural, and religious. In spite of the lackluster initial impression it may give, I argue, Helgastaðir deserves notice on more than one count. First, in the perspective of thirteenth-century Icelanders themselves, this battle gave the measure of an ongoing, seismic shift in Iceland’s political landscape, a shift that was opening up new opportunities for some, while threatening the stability of others. Even as its needle registered these tremors running through the social stratigraphy, the battle also charted the secular magnates’ unwillingness to travel new routes Bishop Guðmundr was striving to map onto the island’s politics – a statement to all contemporaries of where the country’s chief men drew their lines in the sand. But from the perspective of the presentday historian of medieval Iceland, this unimpressive engagement arguably has still greater significance. It offers keyhole views onto several remarkable vistas – social, cultural, organizational, and technological – exposing to our gaze a more rugged and contoured terrain than the idealized landscape that the standard accounts of Icelandic history – rooted in legal texts and in the better streamlined, heroic narratives of the classic sagas – tend to depict. In the first part of the paper, I survey briefly the sources available for accessing medieval Iceland, some of the historiographical debates these sources have sparked, and the reports of the fighting at Helgastaðir, in particular. I then lay out the background to the incidents of 1220 and provide an account of the succession of events at Helgastaðir on that fateful autumn day. Finally, I analyse the battle narrative and draw from it some general conclusions, both for our understanding of medieval Iceland’s history and, more broadly, for the kinds of constraints that enabled the participants in conflicts like the one at Helgastaðir. Such analysis requires gradual shifting of the historian’s lens, from the circumstantially technical and tactical to the more conceptually strategic and social. A proper military history of Helgastaðir must take into account the impact of religious reform on the structures of Icelandic society – long Gregorian shudders rippling across the medieval centuries to be amplified, muffled, or mutated by the medium of local culture. The Icelandic polity, in the form it took until its incorporation into the Norwegian crown in the 1260s, is widely regarded as a medieval anomaly: a country without a king, enjoying a roughly egalitarian ethos to about the same extent that it groaned under the bloody violence attendant on statelessness.4 This 4

See the indignant words put into the mouth of Cardinal William of Sabina, a papal legate who attended the coronation of Norway’s Hákon Hákonarson in 1247: “kardínal[i] … kallaði vsannlikt at land þat þionaði ecki vndir eínn-huern konnung sem ơll ơnnur i verôlldínní” [the Cardinal … called it improper that that country [sc. Iceland] was not subject to any king like every other in the world] (Sturla Þórðarson, Hákonar saga Hákonarsonar cap. 257, ed. Marina Mundt (Oslo, 1977), p. 144). Cf. Einar Ól. Sveinsson’s caustic remark: “the city republics of Italy must for a moment have slipped His Eminence’s mind,” The Age of the Sturlungs: Icelandic Civilization in the Thirteenth Century, trans. Jóhann S. Hannesson (Ithaca, NY, 1953; orig. 1940), p. 14. Hákonar saga was written c. 1265, shortly after Norway had taken over the

96

Oren Falk

contrast with the rest of Europe is likely overdrawn. On the one hand, in Iceland, egalitarian tendencies obtained (if at all) only within quite narrowly circumscribed, elite circles.5 On the other hand, large swathes of medieval Europe were hardly any more securely governed by their nominal central authorities than was Iceland by its decentralized political cadres.6 But the important issue, for present purposes, is not so much the realities behind perceived Icelandic exceptionalism as the evidentiary reasons for it. Our perception depends to a large extent on Iceland’s unique configuration of primary sources. Like many medieval societies, Iceland has left us a number of continuous annals, mostly laconic and extremely pithy, heavily interdependent, and none extant in a form that predates the fourteenth century.7 Like most medieval societies, it has also bequeathed to us its law codes, the earliest of which – Grágás, allegedly in effect until the

5

6

7

government of Iceland. Its author, an Icelander (see n. 12 below), was then in exile in Norway: his biography of King Hákon, whom he had spent most of his life opposing, was probably part of his effort to be reconciled with the reigning monarch, Hákon’s son Magnús; see David Ashurst, “The Ironies in Cardinal William of Sabina’s Supposed Pronouncement on Icelandic Independence,” Saga-Book 31 (2007), 39–45. On Icelanders’ inequalities, see, e.g., William Ian Miller, Bloodtaking and Peacemaking: Feud, Law, and Society in Saga Iceland (Chicago and London, 1990), pp. 26–34 (esp. 33–34); and Orri Vésteinsson, The Christianization of Iceland: Priests, Power, and Social Change, 1000– 1300 (Oxford, 2000), p. 124. Paul R. Hyams, Rancor and Reconciliation in Medieval England (Ithaca, NY, 2003), argues persuasively for the endurance of feuding norms in Angevin England, for example; see also Bruce Lenman and Geoffrey Parker, “The State, the Community and the Criminal Law in Early Modern Europe,” in Crime and the Law: The Social History of Crime in Western Europe since 1500, ed. V. A. C. Gatrell, Bruce Lenman, and Geoffrey Parker (London, 1980), pp. 11–48, and Warren C. Brown, Violence in Medieval Europe (Harlow, 2011). Cf. the incredulity Bartholomaeus Anglicus, Sturla Þórðarson’s contemporary (fl. 1240s), voices concerning Frisian independence: “Gens quidem est libera extra gentem suam, alterius dominio non subjecta.… [S­]­ubsunt tamen iudicibus, quos annuatim de seipsis eligunt, qui rempublicam inter ipsos ordinant & disponunt” [This nation is indeed free (of any) outside its own nation, not subject to another’s rule.… For they are subordinate to judges whom they elect annually from among themselves, who between them rule and direct the commonwealth] (Bartholomæi Anglici De genvinis rervm coelestivm, terrestrivm et inferarvm Proprietatibus, Libri xviii, ed. Georg Barthold Pontanus of Breitenberg (Frankfurt, 1601; repr. 1964), lib. 15, cap. 61, pp. 655–56; for the dating, see M. C. Seymour et al., Bartholomaeus Anglicus and his Encyclopedia (Aldershot and Brookfield, 1992), pp. 29–35; cf. p. 167 on Bartholomaeus’s sources). I am indebted to Han Nijdam for calling my attention to Bartholomaeus’s comment. See Storm, Islandske Annaler, pp. xi–xii, xxi–xxii, lxxxiii, as well as Björn Þorsteinsson, “Síðasta íslenska sagnaritið á miðöldum,” in Afmælisrit Björns Sigfússonar, ed. Björn Teitsson, Björn Þorsteinsson, and Sverrir Tómasson (Reykjavík, 1975), pp. 47–72; Inge Skovgaard-Petersen, Jakob Benediktsson, and Gösta Holm, “Annals,” in Medieval Scandinavia: An Encyclopedia, ed. Phillip Pulsiano et al. (New York and London, 1993), pp. 14–16; Eldbjørg Haug, “The Icelandic Annals as Historical Sources,” Scandinavian Journal of History 22 (1997), 263–74; and Elizabeth Ashman Rowe, “The Flateyjarbók Annals as a Historical Source: A Response to Eldbjørg Haug,” Scandinavian Journal of History 27 (2002), 233–42. An English translation by Elizabeth Ashman Rowe is reportedly in production. Storm grouped these so-called older annals (distinguished from some younger ones, first compiled as late as the seventeenth century) into three genealogies, all ultimately going back to an archetype of c. 1280.



Helgastaðir, 1220

97

1270s – has held special allure for scholars seeking to lay hands on remnants of independent Icelanders’ social sensibilities.8 But like (arguably) no other medieval society, Iceland has also left us a wealth of seemingly dispassionate, detailed, and compellingly narrative histories, the justly famous Icelandic sagas. Whereas most medieval histories celebrate the deeds of royalty, church prelates, or the saints, only a fraction of Icelandic sagas do so, and these are usually relegated nowadays to secondary status, overshadowed by the substantial corpus of sagas that tell the tales of much more ordinary folk, Iceland’s bœndr (“farmers,” “householders,” singular bóndi) and goðar (“chieftains,” “seigneurs,” singular goði).9 Moreover, few European chronicles can compete with the sagas’ eye for quotidian detail, their nuanced and realistic characterization, or their understated talent for recounting the gripping, moving, familiar dramas of the human condition. Certainly, historians of other regions have done wonders in approximating the Middle Ages through the lenses provided by Froissart, Salimbene, Richard of Devizes, Thietmar, Einhard, Procopius, Hydatius, and countless other chroniclers; but, in aggregate, the sagas appear to allow a uniquely intimate proximity with the society they purport to record. It is this saturation of the historical senses, I would argue, that is primarily responsible for Iceland’s anomalous image: for all of the inevitable gaps in our knowledge, historians feel that we can observe Iceland, if not at first hand, then at least in much higher resolution than any other society prior to the documentary explosion of the early modern

8

9

Vilhjálmur Finsen edited Grágás in 3 volumes; see Grágás: Islændernes Lovbog i Fristatens Tid, udgivet efter det kongelige Bibliotheks Haandskrift [Konungsbók], Parts 1–2 (Copenhagen, 1852), abbreviated as Grágás Ia and Ib; Grágás efter det Arnamagnæanske Haandskrift Nr. 334 Folio, Staðarhólsbók, ed. Vilhjálmur Finsen (Copenhagen, 1879), abbreviated as Grágás II; and Grágás. Stykker, som findes i det Arnamagnæanske Haandskrift Nr. 351 fol. Skálholtsbók og en Række andre Haandskrifter (Copenhagen, 1883), abbreviated as Grágás III [all repr. Odense, 1974]. For an excellent English translation, see Laws of Early Iceland: Grágás, ed. and trans. Andrew Dennis, Peter Foote, and Richard Perkins, 2 vols. (Winnipeg, 1980–2000). Charters and other documents that might attest to how the law was implemented in practice (collected in Diplomatarium Islandicum: Íslenzkt Fornbréfasafn, sem hefir inni að halda Bref og Gjörninga, Dóma og Máldaga, og aðrar Skrár, er snerta Ísland eða íslenzka Menn, var. eds. 16 vols. [Copenhagen and Reykjavík, 1857–1952] are hardly preserved prior to the fourteenth century. Bœndr were the basic participants in the Icelandic political community; all Icelandic bœndr were (in principle) each other’s peers. Goði was a position of local leadership, elevated above other bœndr but not firmly separated from them; by securing the loyalty of his followers, a goði could promote his own interests and act as arbiter among fellow bœndr, though not (in principle) enforce his will against opposition. Most historians have followed Grágás in supposing the number of goðar to have been fixed at forty-eight. Whether the law reflected practice seems doubtful, however, especially as the two main codices in which it is preserved date from about the time when Grágás was superseded by a different law code, or shortly thereafter; see Thomas McSweeney’s study in progress, “Imagining a World through Law: Freedom and Equality in Grágás” (working title). It seems likely that the precise number of goðar fluctuated as spheres of authority and power grew or contracted. See Helgi Thorláksson, “The Icelandic Commonwealth Period: Building a New Society,” in Vikings: The North Atlantic Saga, ed. William W. Fitzhugh and Elisabeth I. Ward (Washington, DC and London, 2000), pp. 175–85.

98

Oren Falk

period. The sagas allow us to see in medieval Iceland things that, in the rest of Europe, quite simply do not heave into view until many centuries later.10 It is conventional to distinguish several sub-genres of sagas. Nowadays the so-called Family Sagas (or Sagas of Icelanders) tend to dominate our perception of the corpus and, consequently, of medieval Icelandic realities. The Family Sagas were mostly penned in the thirteenth century (and later), but are set during the long tenth century, roughly between the discovery and first settlement of Iceland in the 870s and the generation or so after the country’s Christianization, c. 1000.11 They thus purport to recount events two, three or even more centuries in the past, forcing on scholars a stark choice between belief in long, unprovable spans of oral transmission and a bone-deep, potentially paralyzing skepticism regarding the sagas’ historical authenticity. For the history of Guðmundr’s time, we must turn to sagas of other sub-genres. Chief among these is Íslendinga saga, a sprawling chronicle of the fiery Age of the Sturlungs (c. 1200–64), so named after the Sturlung family (descendants of Sturla Þórðarson, 1116–83, known from his residence at Hvammr as Hvamm-Sturla). Íslendinga saga itself was written by a namesake grandson (1214–84) of the eponymous patriarch.12 During this era, the fabric of statelessness began to fray: social inequalities deepened, powerful individuals could monopolize control of several chieftaincies, and a handful of families – foremost among them, on this account at any rate, the Sturlungs – dominated the island’s politics, carving out for themselves 10

11

12

One illustration of many is provided by considering the rare evidence for a popular religious custom in Spain: putting vermin (such as locusts or rats) on trial and excommunicating them, when other means for combating their plague have proved ineffective. This custom is first recorded in the late sixteenth century; see William A. Christian, Jr., Local Religion in SixteenthCentury Spain (Princeton, 1981), pp. 29–31. I am indebted to Duane J. Corpis for this reference. Cf. Eyrbyggja saga’s account, probably from the second half of the thirteenth century, of ghosts banished through an analogous process (cap. 55, ÍF 4:150–52). There are, nowadays, many excellent surveys of this literature available in English (and, of course, in other languages, too). Meulengracht Sørensen’s concise Saga and Society remains the single best introduction to medieval Iceland’s history and literature. See also Old Norse– Icelandic Literature: A Critical Guide, ed. Carol J. Clover and John Lindow (Ithaca, NY, 1985); Jónas Kristjánsson, Eddas and Sagas: Iceland’s Medieval Literature, trans. Peter Foote (Reykjavík, 1988; orig. 1974); Margaret Clunies Ross, Old Icelandic Literature and Society (Cambridge, 2000); eadem, A History of Old Norse Poetry and Poetics (Cambridge, 2005); eadem, The Cambridge Introduction to the Old Norse-Icelandic Saga (Cambridge, 2010); Heather O’Donoghue, Old Norse-Icelandic Literature: A Short Introduction (Malden, MA, 2004); and A Companion to Old Norse-Icelandic Literature and Culture, ed. Rory McTurk (Malden, MA, 2005). Most Family Sagas are available in English translation; for the most comprehensive set, see The Complete Sagas of Icelanders, including 49 Tales, ed. Viðar Hreinsson et al. 5 vols. (Reykjavík, 1997). On the younger Sturla Þórðarson, see W. P. Ker, “Sturla the Historian” (orig. 1906), in his Collected Essays, ed. Charles Whibley, 2 vols. (Freeport, NY, 1967; orig. 1925), 2:173–95; Guðrún Ása Grímsdóttir, “Sturla Þórðarson,” in Sturlustefna: Ráðstefna haldin á sjö alda ártíð Sturlu Þórðarsonar sagnaritara 1984, ed. Guðrún Ása Grímsdóttir and Jónas Kristjánsson (Reykjavík, 1988), pp. 9–36 (esp. p. 31 on the connections between both Sturla and his father Þórðr and Bishop Guðmundr); and Marlene Ciklamini, “Sturla Þórðarson,” in Medieval Scandinavia: An Encyclopedia, pp. 613–14.



Helgastaðir, 1220

99

nascent principalities. At over 100,000 words, Íslendinga saga forms the backbone of a collection, compiled around 1300, which commemorates the long unraveling of the Icelandic commonwealth (c. 1117–1264); the compilation is known by the collective title Sturlunga saga, and its components are somewhat misleadingly labeled Contemporary Sagas, to distinguish the briefer gap between event and narration (mere decades) from the centuries spanned in the case of the Family Sagas.13 In light of its pedigree and relative proximity to the period it covers, Sturlunga saga has tended to avoid the kind of fretting over historical reliability that has dogged the Family Sagas (although it has also failed to win the literary acclaim accorded to them).14 Like the classic Family Sagas, however, the texts that make up Sturlunga saga are primarily secular in orientation, unsentimental and, if anything, even more grittily realistic. Though they tend to focus on the ever more exclusive upper echelons of society, the so-called stórgoðar and stórbœndr (“mega-chieftains” and “mega-farmers”),15 the polity they portray is still recognizably the one familiar from the Family Sagas. And if its ruling goðar are further removed from their bóndi brethren, still they are locked with them in an agonistic political embrace. Probably composed about 1280, Íslendinga saga is a mammoth history, picking up the story at Hvamm-Sturla’s death and carrying it down to the capitulation to Norway, eight decades later. It thus also spans Guðmundr’s career, and may be the earliest extant source for it; an older biographical sketch up to Guðmundr’s elevation to the episcopate, the so-called Prestssaga, was probably drawn up not long after his death, possibly by his disciple, Lambkárr Þorgilsson (d. 1249), but it survives only in later, interpolated revisions.16 The bishop was 13

14

15

16

See Sturlunga saga, ed. Jón Jóhannesson, Magnús Finnbogason, and Kristján Eldjárn, 2 vols. (Reykjavík, 1946). My word-count is based on a digitized text scanned from a later edition: Sturlunga saga, ed. Örnólfur Thorsson, 3 vols. (Reykjavík, 1988); I am greatly indebted to Eiríkur Rögnvaldsson of the University of Iceland for giving me access to this digitization. An English translation by Alison Finlay is eagerly anticipated; the existing translation (Sturlunga Saga, trans. Julia McGrew and R. George Thomas, 2 vols. (New York, 1970–74) is defective and unreliable. Úlfar Bragason has long been the chief advocate for reading Sturlunga saga more under the sign of literature and less under that of history; see, e.g., his Ætt og saga: Um frásagnarfræði Sturlungu eða Íslendinga sögu hinnar miklu (Reykjavík, 2010). His conclusions are generally accepted, but his efforts have had a limited impact on the tenor of the scholarship; see, e.g., Jón Viðar Sigurðsson’s review of Ætt og saga, in Speculum 88:3 (2013), 863–64. For these non-medieval terms, see, e.g., Byock, Medieval Iceland, pp. 71–75; idem, Viking Age Iceland (London, 2001), pp. 341–50; Jón Viðar Sigurðsson, Chieftains and Power in the Icelandic Commonwealth, trans. Jean Lundskær-Nielsen (Odense, 1999), pp. 62–83; Orri Vésteinsson, Christianization, pp. 238–46. For a reconstruction of the Prestssaga, see Sturlunga saga, 1:116–59. For the (tentative) attribution to Lambkárr, going back to the nineteenth century, see Stefán Karlsson, “Guðmundar sögur biskups: Authorial Viewpoints and Methods” (orig. 1985), in Stafkrókar: Ritgerðir eftir Stefán Karlsson gefnar út í tilefni af sjötugsafmæli hans 2. Desember 1998, ed. Guðvarður Már Gunnlaugsson (Reykjavík, 2000), pp. 153–71, here p. 155. Perhaps in conjunction with the writing of the Prestssaga, records were assembled in the thirteenth century with an eye towards composing a full hagiography, but these were lost in a fire and the effort was abandoned until

100

Oren Falk

to get his own proper saga, four times over, in the fourteenth century. The youngest redaction, Guðmundar saga D, is attributable with fair confidence to a known monastic author and must have been written c. 1344; the three older versions (only one of whose authors may be tentatively identified) all date from c. 1315 or shortly thereafter. The chronology of their composition cannot be definitively established, but they do fall into a progressively elaborated logical sequence. Guðmundar saga A, the most ‘primitive’ redaction, is largely a pastiche of passages relevant to the bishop’s life, lifted from various antecedent sources and roughly stitched together into a continuous narrative; B and C follow a similar pattern, but embellish and hone it with greater editorializing flair.17 Further enhancing the historical significance of Guðmundr’s sagas (especially of the A redaction), the version of Íslendinga saga on which they drew may have been older than the one currently preserved in the c. 1300 form incorporated into Sturlunga saga.18 Present-day scholarship on Guðmundr is hampered, however, by a haphazard state of publication: the A version is served by a state-of-the-art, semi-diplomatic edition, but the D version is available only in a critical edition from the nineteenth century, and of B and C mere

17

18

the following century. See GSB cap. 82: “eru nú allir dauðir þeir er [þessa frásôgn] ætluðu langa at gera.… Bar ok svá til um þat er menn hôfðu fjôlða brèfa ritat ok í einn stað komit í Laufásskirkju, ok brunnu þau þar inni ôll í kirkjubruna einum, ok munu þau aldri síðan ritin verða” [they are all dead now, those who were long planning to prepare (this account).… It happened, moreover, that people wrote many documents and deposited them in one place, in Laufás church, and they all burned in a church fire and were never thereafter (re)written] (= Biskupa sögur, 1:565–66; Annales regii and Skálholts-Annaler, both s.a. 1258, report the burning of the church at Laufás, in Storm, Islandske Annaler, pp. 133, 192). Of the four later redactions, A, B, and C were likely all written shortly after Guðmundr’s remains were elevated in 1315. (GSC has been tentatively attributed to Bergr Sokkason, fl. 1312–45.) The youngest version, D, written by Arngrímr Brandsson (fl. 1340s–60s; see Margaret Cormack, The Saints in Iceland: Their Veneration from the Conversion to 1400, Preface by Peter Foote [Brussels, 1994], p. 102, for the dating), is largely dependent on C; it survives in six vellum manuscripts, suggesting its greater popularity relative to earlier redactions. D is often thought to have been translated from a Latin original, intended primarily for a foreign (presumably papal) readership; see, e.g., Marlene Ciklamini, “Hidden and Revealed: The Manifest Presence of the Virgin Mary in Bishop Guðmundr’s Life,” Frühmittelalterliche Studien 40 (2006), 223–61, here 225 n. 11, 227–28. This view has recently been challenged; see Gunnvör Sigríður Karlsdóttir, “Himnesk hönd í jarðneskri veröld: Um jarteinir í Guðmundar sögu Arasonar eftir bróður Arngrím” (unpublished MA thesis, Reykjavík, 2009), pp. 125–33; I am grateful to Gunnvör for sending me a copy of her thesis. On the different redactions, see further Margaret Cushing Hunt, “A Study of Authorial Perspective in Guðmundar saga A and Guðmundar saga D: Hagiography and the Icelandic Bishop’s Saga” (unpublished Ph.D. dissertation [Bloomington, 1985], pp. 5–36; Guðrún P. Helgadóttir’s “Introduction” to her edition of Hrafns saga Sveinbjarnarsonar (Oxford, 1987), pp. xi–cxvi, here pp. cxii–cxiii; Stefán Karlsson, “Guðmundar sögur biskups”; and Marlene Ciklamini, “Sainthood in the Making: The Arduous Path of Guðmundr the Good, Iceland’s Uncanonized Saint,” Alvíssmál 11 (2004), 55–74. Anton Zimmerling, “Bishop Guðmundr in Sturla Þórðarson’s Íslendinga Saga: The Cult of Saints or the Cult of Personalities?” in Scandinavia and Christian Europe in the Middle Ages: Papers of the 12th International Saga Conference, Bonn, Germany, 28th July–2nd August 2003, ed. Rudolf Simek and Judith Meurer (Bonn, 2003), pp. 557–67, here 558–59.



Helgastaðir, 1220

101

scraps have so far been published.19 In the discussion that follows, I shall use Guðmundar saga A’s account of the Battle of Helgastaðir (reproduced, with some of the more important variants, in an Appendix) as the point of departure for my analysis, referring to divergent details from other texts as needed. Two of the men who led the attack on Helgastaðir, Sighvatr Sturluson of Eyjafjôrðr and his brother-in-law, Arnórr Tumason of Skagafjôrðr, must be reckoned among the top potentates in Iceland at the time. Sighvatr (c. 1170–1238), one of Hvamm-Sturla’s three legitimate sons, outstripped his siblings in both ambition and ruthlessness. Arnórr (c. 1184–1221), meanwhile, was brother not only to Sighvatr’s wife but, just as importantly, also to Kolbeinn Tumason (c. 1173–1208), who until his death had counted as the undisputed leader in the north of Iceland. Arnórr, though hardly equal to his brother, inherited Kolbeinn’s position of authority. Even among goðar, Arnórr and Sighvatr thus stood head and shoulders above most contenders. Their ally, the third leader of the antiepiscopal coalition, was one Ívarr of Múli, a nearly anonymous neighborhood farmer: he is known only by his domicile (just down the valley from Helgastaðir), not by his patronymic, and he merits mention in the sources only in connection with Guðmundr’s travels through the region. For all that he is referred to as a mere bóndi,20 however, Ívarr seems to have been quite prominent in the local community, and, as we shall see below, his real power (as distinct from his de jure authority) may not have fallen far short of Sighvatr’s or Arnórr’s. The engagement at Helgastaðir was the latest in a series of hostile encounters between Guðmundr and the secular magnates. Following twelve years of arm-wrestling – and, most recently, imprisonment and mistreatment at Arnórr’s 19

20

For GSA, see Guðmundar saga A, in Guðmundar sögur I, ed. Stefán Karlsson (Copenhagen, 1983), pp. 15–255 (superseding Saga Guðmundar Arasonar Hóla-biskups, hin elzta, in Biskupa sögur, gefnar út af Hinu íslenzka bókmentafèlagi, ed. Jón Sigurðsson, Guðbrandur Vigfússon, Þorvaldur Björnsson, and Eiríkur Jónsson, 2 vols. [Copenhagen, 1856–78], 1:405–558). For GSD, see Saga Guðmundar Arasonar, Hóla-biskups, eptir Arngrím ábóta, in Biskupa sögur, 2:1–187. GSB and GSC are cited from transcriptions made by the late Stefán Karlsson († 2006), now held by Hið íslenzka fornritafélag; I am indebted to Þórður Ingi Guðjónsson, general editor of the Íslenzk fornrit series, for allowing me to consult these transcriptions, and to Ásdís Egilsdóttir and Gunnvör Sigríður Karlsdóttir for making them available to me. Both semi-diplomatic (Editiones Arnamagnæanæ, ser. B vols. 7–9) and critical (ÍF 18–19) editions of Guðmundr’s sagas are anticipated, but have been delayed by Stefán’s untimely death. Some variant readings from GSB and GSC are cited in the apparatus criticus of GSA in Biskupa sögur, 1:405–558 (under the sigla B and 395, respectively); additional excerpts from GSB are printed as Brot úr miðsögu Guðmundar, in Biskupa sögur, 1:559–618; and an additional fragment of GSC has been edited by Peter G. Foote in “Bishop Jörundr Þorsteinsson and the Relics of Guðmundr inn góði Arason,” in Studia centenalia in honorem memoriae Benedikt S. Þórarinsson, ed. B. S. Benedikz (Reykjavík, 1961), pp. 98–114. An English translation of Guðmundr’s vita is available, based mostly on the A version (with some interpolations from B): The Life of Gudmund the Good, Bishop of Holar, trans. G. Turville-Petre and E. S. Olszewska (Coventry, 1942). And possibly a priest; GSD attributes the actions that other versions ascribe to Ívarr of Múli to an (unnamed and undomiciled) “prestr” (cap. 55, in Biskupa sögur, 2:112–13).

102

Oren Falk

hands – Bishop Guðmundr had escaped to spend most of the summer of 1220 in Reykjadalr, a shallow valley of rolling grassland running south–north, which stretches some twenty-five miles inland from the bay of Skjálfandi (Figure 2). Reykjadalr was probably typical of many micro-environments in Guðmundr’s diocese, though perhaps better off than most. As elsewhere in medieval Iceland, farmsteads, not hamlets or villages, were the rule, and pasture (of both cattle and horses) trumped tillage. The central segment of the valley supported four major farms in the Middle Ages – Grenjaðarstaðir, Múli, Helgastaðir, and Einarsstaðir – strung out at no more than about three miles apart along the valley floor.21 All four were ecclesiastical estates, and they were supplemented by several smaller, probably dependent farms.22 Guðmundr also had a personal connection to this region, having spent part of his youth at Grenjaðarstaðir.23 Here, then, was a theater capable of sustaining an itinerant bishop and his entourage, an entourage that grew to be about one hundred strong as “many people then flocked to him,” the saga reports.24 Perhaps even more important in the bishop’s calculus, Reykjadalr was without a resident goði. Guðmundr was itinerant, after all, because he had been forced to flee his see for fear of Arnórr and his allies. The depredations of the magnates had already driven him into a four-year exile in Norway (1214–18), 21

22

23 24

In the Middle Ages, ‘Reykjadalr’ referred to the whole length of the valley between the inland lake Másvatn and Skjálfandaflói; in modern times, the name has been reserved for the middle section of a valley system, between Seljadalur to the south and the coastal Aðaldalur to the north. See Jóhann Skaptason, “Suður-Þingeyrarsýsla, austan Skjálfandafljóts,” Ferðafélags Íslands árbók (Reykjavík, 1978), pp. 9–111, esp. 11–12, 28–31, 45–50; Deanna Swaney, Iceland, Greenland and The Faroe Islands: A Travel Survival Kit, 2nd edn (Hawthorn, Vic., 1994), pp. 259 (map), 262; Iceland Road Guide, ed. Örlygur Hálfdánarson, 3rd ed., trans. Einar Guðjohnsen and Pétur Kidson Karlsson, texts by Steindór Steindórsson frá Hlöðum, maps by Narfi Þorsteinsson (Reykjavík, 1981; orig. 1973–74), pp. 305–6, 332 (map), 333; and Þorsteinn Jósepsson, Landið þitt: Saga og sérkenni nær 2000 einstakra bæja og staða, 2nd ed. (Reykjavík, 1967), pp. 8, 70, 134–35, 253–54. For livestock bred in Reykjadalr, see, e.g., GSA capp. 169–70 (pp. 184–85). Nowadays, Arngeir Friðriksson’s farm at Helgastaðir subsists mostly on cattle, supplemented by some horse breeding and farming of barley and potatoes (personal communication). Both Helgastaðir and Grenjaðarstaðir were venerable churches, established before 1200 (and both rich and important enough to possess, by the late 1300s, glass windows, as well as a bell tower at Grenjaðarstaðir; see Diplomatarium Islandicum, 3:574, 581, 709 [§§487, 490, 586], 4:21 [§14]), and both Múli and Grenjaðarstaðir were “major churches with many clerics” (Orri Vésteinsson, Christianization, p. 16; see also pp. 98, 207 n. 35). As early as 1318, Múli had a copy of a Guðmundar saga, the first of only three medieval churches (including the cathedral at Hólar) known to have owned a saga about Guðmundr (Diplomatarium Islandicum, 2:435 [§252]; see Joanna A. Skórzewska, Constructing a Cult: The Life and Veneration of Guðmundr Arason (1161–1237) in the Icelandic Written Sources (Leiden, 2011), pp. 256–57. Einarsstaðir was, according to GSA, farmed in partnership by two men, one of them a priest, in 1220 (cap. 169, p. 184); Stefán Karlsson has, moreover, traced the lineage of two of Guðmundr’s ecclesiastical confidants to Einarsstaðir: see “Icelandic Lives of Thomas,” p. 215; cf. Sturlu saga cap. 1, in Sturlunga saga, 1:63. See GSA cap. 13 (pp. 37–38). GSA cap. 168: “Dreif þa til hans mart folk.… ok hafðe hann nẻr tiutigum manna” (p. 184).



Helgastaðir, 1220

103

Figure 2.  Map of Reykjadalr, drafted in 1931; elevations marked in 20 meter intervals.

104

Oren Falk

and would do so again in the next decade (1222–26). Shortly after the bishop’s return from exile, Arnórr had caught up with him, treating him bruisingly and throwing him unceremoniously into a makeshift prison until he might be carted off to Norway again. Guðmundr was spared further indignity only when his secular right-hand man, Eyjólfr Kársson, managed to spring him from jail, in a brilliant sting operation that included the preparation of getaway steeds, a change of costume, and – the master stroke – the planting of a decoy prisoner to delay pursuit (“the men who kept watch were saying that [Grey]whiskers was sleeping in late this morning”).25 But Guðmundr knew from bitter experience that Arnórr would continue to nurse his long-held grievance and would come after him again as soon as opportunity presented itself. Reykjadalr, somewhat removed from the fjords directly under Arnórr’s and Sighvatr’s dominion, appeared relatively safe, and for some months it was so. Meanwhile, however, Guðmundr was making new enemies. In the fragile, tenuously balanced ecology of a northern Icelandic valley, his large retinue were leaving a heavy footprint; the residents of Reykjadalr could not help fearing, and indeed feeling, that they were being eaten out of house and home. For a while, much as the local bœndr chafed at the size and appetite of Guðmundr’s following, they endured it; even Ívarr received the bishop courteously, we are told, on his first visit to Múli. When Guðmundr came a second time, however, he found Ívarr “most unwilling to receive him, and he had 40 men, equipped as for battle, and he arrayed his men in ranks.” The bishop backed off, but still the farmers felt sufficiently beleaguered to send word to Sighvatr and Arnórr, asking for their help in eliminating this episcopal nuisance.26 The magnates did not need much goading; they speedily mobilized their forces. Sighvatr brought with him two of his sons, Tumi and Sturla.27 The joy of the local population 25 26

27

GSA cap. 164: “sogðo menn er uorð helldo. at kanpe suẻfe lenge um morgininn” (p. 182). For the full narrative of Guðmundr’s captivity and rescue, see GSA capp. 161–64 (pp. 179–82). GSA cap. 169: “Nv ferr G[uðmundr] byskup iMula. ok toc Juar uið honum liðlega. Var þar sẻmilega við tekit. þess at þat matte sea at eigi var af astsemd ueitt af Iuare. Skilea þeir þo uel. ok for G[uðmundr] byskup a brot ok setz a … Greniaðar støðum. Var þat orð a at hann munde þaðan iMula iannat sinn. en Iuar uillde uist eigi við honum taka. ok hafðe hann .xl. manna ok sua bunir sem til bardaga. ok skipaðe hann mønnum istoþur” [Now Bishop Guðmundr goes to Múli and Ívarr received him obligingly; (the hospitality) which, it could be seen, was not extended by Ívarr out of love was accepted graciously. Still, they parted on good terms, and Bishop Guðmundr went away and stayed at … Grenjaðarstaðir. Word was that he would go thence to Múli once more, but Ívarr was …] (p. 184). For the Reykjadalr farmers’ appeal to the goðar, see GSA cap. 170: “bẻndr gera orð Sigh[uate] Stur[lo] s[yne] ok Ar[nore] T[uma] s[yne] at þeir rydde af þeim ufriðe þessom” [The farmers send word to Sighvatr Sturluson and Arnórr Tumason that they should rid them of these disturbances] (pp. 184–85). Most accounts of the battle do not mention their presence up front, but reveal it only in passing as events progress. The author of GSD evidently felt the need to clarify matters: “dregr Arnórr lið saman fyrir vestan fjöll, en Sighvatr um Eyjafjörð, mætast síðan ok halda þat fólk til Reykjadals; í þeirri ferð eru tveir synir Sighvats, Sturli [sic] ok Tumi. Tumi var þá ungr, en þó vill faðir hans at hann venist meingjörðum móti herra Guðmundi, skyldist honum ok þat verk, sem brátt mun lýsast” [Arnórr recruits a troop west of the mountains, but Sighvatr in Eyjafjǫrðr; afterwards they meet up and lead that troop to Reykjadalr. Sighvatr’s two sons, Sturli [sic] and



Helgastaðir, 1220

105

at seeing their liberators bearing down on the bishop’s party is captured in the words attributed to the bóndi’s brother at Helgastaðir: “Look there now, lads, where the troop of the chieftains rides! The locks on the pantries of the men of Reykjadalr click shut at last!”28 Having thus set the stage – sources, personnel, and geography – I now turn to a blow-by-blow account of the battle itself. Saturday, 29 August 1220, was the feast day of Saint John the Baptist’s beheading, and Guðmundr, ignorant of the brewing storm, traveled from Einarsstaðir, at the foot of the hills, to Helgastaðir, close to a bend in the river, to celebrate the anniversary service of the dedicatio ecclesiae.29 Eyjólfr Kársson rode ahead to Múli, in the foothills farther north, to arrange for a similar ceremony there later in the day, but was turned back just as rudely as the bishop himself had been some days before. Meanwhile, the jaws of the goðar’s trap were snapping shut as their men moved into position up and down the valley. We have no clear idea of the numbers involved on either side. The earlier statement putting Guðmundr’s following at one hundred may be taken as a firm maximum, from which an unknown fraction of non-combatants (as well as any who just happened to be away that day) should be subtracted; a year or two later we hear of “70 bold men” and “30 women and beggars” who followed the bishop when he sought refuge on some offshore islands.30 Ívarr’s forty perhaps also give an indication of numerical relations. All in all, it is

28

29

30

Tumi, take part in this expedition. Tumi was young at the time, but still his father wants him to become proficient at acting feloniously against Lord Guðmundr, and he takes to that task, too, as will soon be shown] (cap. 55, in Biskupa sögur, 2:112–13). On the economic burden Guðmundr’s following placed on the residents of areas they passed through, see, e.g., GSA capp. 130, 135, 238–39 (pp. 155, 159, 240–41). On one occasion, when Guðmundr arrives for an unexpected visit with a large entourage, he considerately arranges for a miraculous whale to beach itself nearby, securing enough food for everyone (GSA cap. 83, pp. 111–12, redoubled in GSD cap. 65, in Biskupa sögur, 2:133; cf. further provision miracles in GSA cap. 25, p. 62, and GSB capp. 121, 123–24, 128–29, 140–41 = Biskupa sögur, 1:591, 593–95, 597, 611). Helgastaðir was an old and established church, so I assume that Guðmundr’s purpose there that day, at vígja þar kirkju, was to commemorate the anniversary of the consecration. The redactors of GSB, GSC, and GSD seem to have taken the verb at face value, and thought that the bishop came to Helgastaðir to consecrate a new church (see their retrospective comments on the damage done in the battle to a “newly consecrated church”). By the late fourteenth century, Helgastaðir was dedicated to Saint Nicholas and the Virgin, not John the Baptist; the fact that Guðmundr meant to perform the same service that day at Múli, a church likewise known to have had Saint Nicholas and the Virgin as its patron saints in the 1300s (see Diplomatarium Islandicum, 3:573, 576 [§§487, 489]; GSA cap. 171, pp. 185–86), may suggest that his schedule planner did not stand too much on calendrical niceties. On the ceremonial of dedicatio ecclesiae, see Dominique Iogna-Prat, La Maison Dieu: Une histoire monumentale de l’Église au Moyen Âge (v. 800–v. 1200) (Paris, 2006), pp. 265–77. I am indebted to Jennifer A. Harris for this reference. GSA capp. 177, 195: “.lxx. rauskra manna” (towards the end of 1221) and “.xxx. kuenna og stafkarl[a]” (in early 1222; pp. 192, 205); seventy arms-bearing men are again mentioned in the later instance.

106

Oren Falk

hard to imagine that more than two or three hundred participants, all told, were involved, and perhaps far fewer than that.31 Only towards noon did the bishop’s men realize that they were about to come under attack. Late August in these parts tends to be cool and clear, so – barring inclement weather, which the saga does not mention – visibility over the gently undulating terrain of Reykjadalr should have presented few difficulties, and the attackers would easily have been spotted from three miles off or more as some of them spilled downhill into the valley’s southern basin, while others galloped along its eastern ridge: “Afterwards they see that the troops rode down to Einarsstaðir.… Later they saw another company ride up along Vatnshlíð.”32 We know from other sources that Helgastaðir had been fortified in the late twelfth century, likely by circumvallating earthworks, and developments on the second day of the battle confirm that the ramparts, while they may have fallen into disrepair, still provided some protection, slowing the attackers down by bottlenecking them into a finite number of entry points and so freeing the defenders to economize their efforts.33 From the wording of the battle account in Guðmundar

31

32

33

GSD notes in passing that Ívarr came “til fulltíngs viðr þá með mikinn flokk bygðarmanna” [to assist them with a great company of local residents] (cap. 55, in Biskupa sögur, 2:113). Cf. the largest muster we hear of in GSA, in 1234, which numbers a thousand men, four hundred on one side and six hundred on the other, though battle is on this occasion averted (cap. 256, p. 254); see also Guðmundr’s entourage of 600 in GSB cap. 123 (= Biskupa sögur, 1:593). Elsewhere in Sturlunga Saga there are musters of as many as 1700 on a side; see Andreas Heusler, Zum isländischen Fehdewesen in der Sturlungenzeit (Berlin, 1912), p. 34. Elsewhere, we do hear about foul weather when it has narrative significance; see, e.g., GSA capp. 41–46, 80, 114–15, 179–80 (pp. 77–83, 108–9, 141–44, 193–94). Northern Iceland sees over fifteen hours of daylight in late August (over seventeen hours from first light to darkness, though the sun is visible on average for only 4.5 hours a day during the month). Average temperatures in August are 11.1°C, ranging from an average low of 8.3°C to an average high of 15.3°C, with winds averaging 4.2 meters/second and precipitation 1.2 millimeters/day. I derive astronomical and meteorological data (using Akureyri as a proximate reference point) from Almanak Hins íslenska þjóðvinafélags 2005 / Árbók Íslands 2003, ed. Þorsteinn Sæmundsson / Heimir Þorleifsson, vol. 131 (Reykjavík, 2004), pp. 55, 82. A. E. J. Ogilvie, “The Past Climate and Sea-Ice Record from Iceland, Part I: Data to A.D. 1780,” Climatic Change 6 (1984), 131–54, and “Climatic Changes in Iceland A.D. c. 865 to 1598,” Acta Archaeologica (Copenhagen) 61 (1990), 233–51, surveys our (scanty) knowledge of medieval Iceland’s climate. See Guðmundar saga dýra cap. 3, describing events in 1187: “hvárir tveggja [á Grenjaðarstöðum ok Helgastöðum] höfðu virki um bæ sinn” [they each had ramparts around their farms (at Grenjaðarstaðir and at Helgastaðir)] (in Sturlunga saga, 1:165). On the fortification of Icelandic farms, see Adolf Friðriksson, “Sturlunga minjar,” in Samtíðarsögur: Níunda alþjóðlega fornsagnaþingið, forprent / The Contemporary Sagas: Ninth International Saga Conference, Preprints (Akureyri, 1994), 1:1–15, here 3–10, and cf. Birgir Loftsson, Hernaðarsaga Íslands 1170–1581 (Reykjavík, 2006), pp. 63–80. Adolf seems skeptical of the quality of construction at Helgastaðir: “Ekki er auðvelt að ráða af textanum hvort hér hafi verið vegleg mannvirki eða ekki. Athygli vekur að á Helgastöðum, rúmum 30 árum síðar (1220) … er virkis ekki getið.” It is not easy to decide from the text (of Guðmundar saga dýra) whether there was a high-quality fortification here or not. It merits attention that, about 30 years later (1220), no fortification is mentioned at Helgastaðir; nevertheless, he points out the significance of the vígflaki (discussed



Helgastaðir, 1220

107

saga A and its analogues, it seems that the walls enclosed the churchyard, but not the farmhouse proper; as Olaf Olsen has pointed out, the primary function of churchyard enclosures was to demarcate sacred space, so a secular farmhouse would hardly have belonged within the walled churchyard. Arngeir Friðriksson, whose family have owned Helgastaðir since 1907, points to an oval area next to the present-day farmhouse as the site of the old church, torn down in 1872. This area, which he estimates at a little under twenty meters (abuot sixty feet) across on its longest axis, was planted in recent decades with birch and other trees, and serves nowadays as a graveyard; in spite of the transformations it has undergone, it is still visibly bounded by an embankment rising several feet above the surrounding landscape, which may well be a remnant of the medieval rampart (Figure 3). Such crowded quarters, though they would have ranked Helgastaðir’s churchyard alongside some of Iceland’s largest, further limit the number of men that we can safely imagine to have been holed up together with Guðmundr. The old farmhouse itself, Arngeir reports, had stood west of this perimeter. The house’s turf walls have left no readily identifiable traces, but there is little reason to doubt that this local tradition accurately preserves the memory of the site: this could be where the disgruntled bóndi of Helgastaðir and his brother had stood in 1220, applauding the assault from a critical distance.34 Perhaps taking courage from the advantage of their fortified position, the bishop’s men kept calm and carried on; they hastened to gather stones and prepare for defense in the churchyard. The attackers then closed in: “The companies arrived at nearly the same time; it was noon.… Arnórr Tumason advanced from the south with his detachment, but Sighvatr advanced from the home meadow towards the gate, while Ívarr stormed from the north.” It is not known quite where the home meadow lay – possibly southwest of the farm35 – but it is clear

34

35

below) in overcoming the defenses, noting: “Þar sem voru góð vígi, þurfti tilfæringar við atlöguna” [Where there were good ramparts, equipment was needed for attacking] (p. 7). Olaf Olsen, Hørg, hov og kirke: Historiske og arkæologiske vikingetidsstudier (Copenhagen, 1966), p. 201, and Arngeir Friðriksson, personal communication; Arngeir was kind enough to pace the old churchyard for me, measuring its north–south axis at thirty paces (c. 60 feet) and its east–west axis at twenty (c. 40 feet). Cf. the churchyards at Saurbœr in Eyjafjôrðr (16 × 18 paces, i.e. c. 32 × 36 feet); Hofstaðir in Skagafjôrðr (16 × 19 meters = c. 50 × 60 feet); Gunnsteinsstaðir in Langadalr (16 × 18 meters = c. 50 × 60 feet); Gásir in Eyjafjôrðr (24 × 19 meters = c. 80 × 60 feet); and Vatnsfjôrðr in Ísafjôrðr (c. 47× 43 meters = c. 155 × 145 feet); these measurements are taken from Guðbrandur Jónsson, Dómkirkjan á Hólum í Hjaltadal: Lýsing íslenzkra miðaldakirkna (Reykjavík, 1919–29), plates xii–xiv and p. 79, and Olsen, Hørg, hov og kirke, pp. 199–200. Arngeir’s local knowledge may, of course, reflect the location of the farmhouse only in the nineteenth century rather than in the thirteenth; the wording of Guðmundar saga dýra (quoted above, n. 33) suggests that the medieval farmhouse may have been inside the ramparts, and the size of the enclosed space would probably have allowed for this. Olsen tentatively accepts that particularly large enclosures, such as at Vatnsfjôrðr and possibly at Gásir, may also have served a defensive function; see Hørg, hov og kirke, p. 201, and cf. Guðbrandur Jónsson, Dómkirkjan á Hólum í Hjaltadal, pp. 77–84. On the date when the church of Helgastaðir was demolished, see Þorsteinn Jósepsson, Landið þitt, p. 134. This is where the core of Helgastaðir’s home meadow lies nowadays. According to Arngeir Friðriksson, the home field covers some 35 hectares, mostly located to the south and southwest

108

Oren Falk

Figure 3.  View from the south of Helgastaðir’s cemetery, formerly the site of its church, which was torn down in the late nineteenth century. The current farmhouse is just visible behind the trees, planted in the mid-twentieth century. An oval embankment (visible at the foot of the trees) surrounds the cemetery; it may be the remnant of the medieval rampart.

enough that the attacking columns locked Guðmundr’s men in a pincer, with Arnórr (and probably Sighvatr, too) heading north, following in Guðmundr’s tracks from Einarsstaðir (Figure 4), and Ívarr riding in the opposite direction, down from Múli along the Vatnshlíð ridge that borders Reykjadalr to the east. As soon as the attackers came within hailing distance of the bishop’s men, they wasted no time on pleasantries: “The war-cry was then raised and the attack was launched.” Their clamor, no doubt a means of synchronizing the tripartite assault, also served to demoralize those who were bracing for impact within the churchyard; on a different occasion, we hear of a besieging force whooping for calculated psychological effect.36 Although the attackers had arrived on

36

of the farmhouse, between about 5 and 9 o’clock (with another small field in the northwest, at 9–11 o’clock); personal communication. Furthermore, reference to Sighvatr’s assault on hliðin, “the gate,” may suggest that Helgastaðir had only one main gate. In a later context, GSB refers to a “gateway in the fence … which faced towards Glaumbœr” (cf. p. 114 below). Glaumbœr is a tiny farm about a third of the way between Einarsstaðir and Helgastaðir, just west of the present-day road connecting the two. If this is the gate Sighvatr attacked, his men may have followed a path similar to that of the modern roadway, a little to the west of a direct southerly approach from Einarsstaðir. See GSA cap. 141 (recounting the goðar’s assault on Hólar in 1209): “Þorðr moður broðir Sturlo s[ona] lagðe þat rað til at þeir skylldo ẻpa um quelldit herop. enn ganga fyr at enn um



Helgastaðir, 1220

109

Figure 4.  View to the south from Helgastaðir. The farms Akrar, built in 1939 on the southern half of historic Halldórsstaðir (A), Öndólfsstaðir (B), and historic Glaumbær, split up in 1953 into two adjacent farms, Glaumbær and Grundargil (E), are visible. Farther in the distance is Laugar, a small village (cf. the three farms Stóru-Laugar, Laugar, and Litlu-Laugar in Figure 2), some four miles south of Helgastaðir (C). Einarsstaðir, though visible from other spots in Helgastaðir, is here hidden behind a small rise (D). The mown field in the foreground is part of Helgastaðir’s home meadow.

horseback, the assault itself was carried out on foot. This is typical of combat in Iceland, where a native knighthood never emerged. The action appears to have consisted of a headlong charge from three directions, met with volleys of missiles, followed by disorganized thronging of spears at close quarters.37 “Intense fighting now broke out with stones and shot. Sturla [or, in another version, Sturla’s slightly older brother, Tumi] Sighvatsson got hit by a stone. Then Sighvatr proclaimed: ‘The bishop’s men spare nothing! Now they strike the boy, Sturla [Tumi], the same as other men!’” He took his followers to task,

37

morginin. ok ẻtlaðe hann at þa mundi þun skipaðra a husunum en um quelldit” [Þórðr, the maternal uncle of the sons of Sturla, offered this advice, that they should yell the war-cry in the evening but not advance until the morning; and he thought that the ranks would be thinner then than in the evening] (pp. 163–64). See also n. 41 below. Cf. Régis Boyer, “La guerre en Islande à l’âge des Sturlungar (XII–XIIIe siècles): Armement, tactique, esprit,” Inter-Nord: Revue internationale d’études arctiques et nordiques 11 (1970), 184–202, speaking of Icelandic combat in this period in general terms: “Toute bataille … commence par un violent échange de cailloux et de pierres.… Ceci passé, on en venait au corps à corps qui, sauf exceptions, se faisait confusément” (p. 190).

110

Oren Falk

and “at this egging on, Sighvatr’s men pressed the assault hard.” In Guðmundar saga B and C, Sighvatr too is struck by a stone, losing his footing and almost precipitating a general rout. A ditty is remembered in this context, highlighting the significance ascribed to Sighvatr’s stumble. Curiously, on this island beset by volcanic eruptions and earthquakes, the little quatrain about the Battle of Helgastaðir, associated with an otherwise unknown Guðmundr klasi, provides one of only a handful of medieval references to seismicity – and that only in metaphor.38 As the combat intensified, we hear of first casualties, who, notably, are also the last to die fighting (though not the last to die): “One of the bishop’s men, named Gísli, fell before the spears. A little later a man from Arnórr’s troop received a spear thrust to the eye.” But the bloodshed led neither to decision nor to abatement: “Then Bishop Guðmundr walked out of the church and to his men. Then they turn very fierce and fight most valiantly.” The mêlée continued in this way for an indeterminate (but apparently quite long) while, perhaps until dusk, before the attackers pulled back to blockade the farmstead, offering a facesaving gloss for their failure to overrun the defenders: that, wishing to preserve the lives of men (their own? or also Guðmundr’s?), they preferred to triumph through stratagems rather than brute force, and that they were mindful of the bishop’s personal safety in particular. Inconclusive negotiations followed, and during the night Eyjólfr hríðarefni, one of Guðmundr’s men, tried to slip the encircling ring but was caught, “jostled and beaten and dragged back half-dead, and they killed him later.” By the time the assault was renewed the following morning, the attackers had equipped themselves with a vígflaki, literally a “battle-barrier.” This could conceivably have been a shield-wall, a formation of men holding their shields edge-over-edge to present a continuous front. Here, however, it appears (at least on the evidence of two words in the Íslendinga saga version, which notes that it was made “of beams”) to be used as the vernacular term for what elsewhere in Europe was known as a testudo, “tortoise” (among other names): a mobile structure, typically of wickerwork or wood, intended to provide shelter from projectiles while the attackers dismantle any obstacles in their path. Snorri Sturluson describes such an apparatus in his Heimskringla: “King Óláfr had large testudines made of the gnarly and soft parts of trees.… Then he had staves propped under them, so thick and so tall that it was easy to fight beneath [them], and robust enough to withstand stones, if they should be hurled from above.” (Whether constructions such as Snorri describes were a native Norse invention or a distant echo of Roman military engineering seems ultimately undecidable.)39 Nothing is said about the provenance of this contraption at 38 39

Cf. Oren Falk, “The Vanishing Volcanoes: Fragments of Fourteenth-Century Icelandic Folklore,” Folklore 118 (2007), 1–22. Óláfs saga helga cap. 13: “Óláfr konungr lét gera flaka stóra af viðartaugum ok af blautum viði.… Þar lét hann undir setja stafi svá þykkt ok svá hátt, at bæði var hœgt at vega undan ok ýrit stinnt fyrir grjóti, ef ofan væri á borit” (in Heimskringla, ÍF 27:15–16). Cf. Publius Renatus



Helgastaðir, 1220

111

Helgastaðir; Íslendinga saga does mention in this context a convenient copse – a rare feature, in Iceland’s mostly treeless landscape – just south of the farmstead. Perhaps the twentieth-century plantings at Helgastaðir commemorate an older, native clump of birches that would have stood only a short distance off. The vígflaki was borne against the south wall of the churchyard, enabling the attackers to breach it. The defenders promptly withdrew into the church itself. (The redactor of the D version, markedly more interested in hagiography than in military history, simplifies the narrative here by omitting any mention of the vígflaki, crediting instead the attackers’ numerical superiority, though he does note that the defenders were holed up in the church as inside “a fortress.”) Perhaps they hoped to find sanctuary in the building, though hallowed space seldom delayed Icelandic victors for long; and indeed, according to Guðmundar saga B and C’s accounts of the battle, Arnórr’s men injured one of the bishops’ hangers-on, a certain Þorsteinn hryggr, inside the church, and later ransacked it.40 Alternatively, the defenders may have been preparing to make a last stand in the doorways, or, perhaps most likely, they hoped to use the protective walls as cover while they bargained for terms. Before Arnórr, Sighvatr, and Ívarr could make up their minds whether to pursue their enemies into the church or enter into negotiations, however, a new player appeared on the stage: Ísleifr Hallsson from the neighboring valley to the east, about six miles away by the shortest route. (In keeping with the trend

40

Vegetius Flavius, Epitoma rei militaris, ed. Alf Önnerfors (Stuttgart and Leipzig, 1995), lib. 4 capp. 13–16, pp. 212–16, distinguishing a variety of similar constructions (testudines, vineae, plutei, and musculi). Bernard S. Bachrach argues vigorously for continuity of Roman military science into medieval Francia, Anglo-Saxon England, and ultimately perhaps even Scandinavia; see “‘A Lying Legacy’ Revisited: The Abels-Morillo Defense of Discontinuity,” Journal of Medieval Military History 5 (2007), 153–93, esp. 170–71 n. 62, 187–88, and contrast Richard Abels and Stephen Morillo, “A Lying Legacy? A Preliminary Discussion of Images of Antiquity and Altered Reality in Medieval Military History,” Journal of Medieval Military History 3 (2005), 1–13. Though Bachrach, to my mind, persuasively debunks specific errors of fact and argument in Abels and Morillo’s discussion, he does not invalidate their fundamental points: that medieval authors routinely stuffed the realities of their own time (or recent past) into a Procrustean vocabulary they had inherited from Antiquity (see, e.g., “A Lying Legacy?” pp. 3 n. 8, 6, 10–12), and that many of the precepts medievals might have learnt from the Ancients were “such basic, common sense measures that any intelligent [military leader] should have been able to work them out independently” (Stephen Morillo, Warfare under the Anglo-Norman Kings, 1066–1135 [Woodbridge, 1994], p. 118 n. 89). Þorsteinn’s nickname, hryggr, literally means “spine”; he is described as a vesalingr, which means “miserable, poor wretch” in either a pejorative or a compassionate sense. Taken together, the terms might be taken to suggest some sort of deformity (perhaps “Þorsteinn the Hunchback”). Orri Vésteinsson, Christianization, p. 54 n. 58, notes the abundant evidence for both a norm, which seems to have been well established by the thirteenth century, of using church buildings as sanctuaries, and for breaches of this norm. Cf. Michael Gaddis, There Is No Crime for Those Who Have Christ: Religious Violence in the Christian Roman Empire (Berkeley and Los Angeles, 2005), pp. 54–55, 79–87, discussing the formation of “armed soldiers violently invading a church” as a polemical and hagiographic topos in Late Antiquity (p. 79).

112

Oren Falk

we have observed before, the D version omits his name, referring to him with hagiographic generality as “a certain good farmer.”) We do not know how Ísleifr had been alerted to the fighting at Helgastaðir. There were plenty of vectors for spreading rumors in medieval Iceland – Njáls saga, for instance, famously dramatizes the impact of gossip-bearing women traveling between households; perhaps the unfortunate Eyjólfr hríðarefni was not the only one among Guðmundr’s men who preferred to take leave of the holy father’s company overnight, and others may have been more successful at making a clean getaway.41 Be that as it may, when Ísleifr arrived on the scene, he invited Guðmundr to come with him, the bishop accepted, and the magnates did nothing to hinder their departure. Such last-minute redemption was not unusual. Over a decade earlier, when a huge army made up of the followings of seven goðar had successfully besieged Guðmundr at Hólar cathedral, Snorri Sturluson – though a member of the attacking coalition – had similarly intervened to provide the bishop with an out, once his forces had been reduced to submission.42 Little is known about Ísleifr, but his family connections suggest that his sympathies in all likelihood lay with Guðmundr; the implication of the statement that “his men had not joined in the fighting” could well be that they stood ready to throw their weight behind the bishop, had Ísleifr’s peaceable intervention been thwarted (an interpretation further endorsed by the wording of the C redaction, which emphasizes Ísleifr’s lack of hostility towards one side only).43 The goðar need not have consulted their Vegetius to determine that 41

42

43

See Brennu-Njáls saga cap. 44, 92 (ÍF 12:112–14, 230–31), and cf. Svínfellinga saga cap. 6 [215] for a siege broken by the fear of reinforcements summoned by a fugitive (in Sturlunga saga, 2:91–93). During the siege of Hólar in 1209, the attackers (as noted in n. 36 above) used a deliberate ruse to encourage Guðmundr’s supporters to bleed away overnight: “Þat for sem Þorðr gat til at margir menn leynduz um nottina fra G[uðmunde] byskupi. en sumir til vuina hans. en sumir annan ueg a brot. sua at fatt var manna eptir a husunum” [It happened so as Þórðr had figured out, that many men slipped away from Bishop Guðmundr during the night, some to his enemies’ (camp) but some away in other ways, so that few men remained behind at the cathedral] (GSA cap. 141, p. 164). See GSA capp. 141–42: “Nu gerðo þeir byskupi .ij. koste anan þann at hann skyllde taka þa or banne. en þeir mundo þa gefa þeim sumum mønnum griþ er ikirkio voro. En byskup skyllde fara af staðinum. ok koma þar alldri. siþan. ella mundo þeir drepa þa alla. er i kirkiunne voro. ok eira øngo uẻtta. en reka þo byskup abrot af staðinum. suivirðelega. Han kaus huarngan kost.… Eptir þetta byðr Snorre Sturlo s[on] G[uðmunde] byskupi til sin” [Now they gave the bishop two options: the one that he should remove the ban (see n. 59 below) from them, but they would then give quarter to some of the men who were in the church, but the bishop should leave the cathedral and never come back there; otherwise, they would kill all those who were inside the church and spare not a soul, but still drive the bishop away shamefully from the cathedral. He picked neither option.… After that Snorri Sturluson invites Bishop Guðmundr to his home] (pp. 164–65); cf. Íslendinga saga cap. 24 [29] (in Sturlunga saga, 1:252). Ísleifr’s brother Eyjólfr – at one time priest in nearby Grenjaðarstaðir, later abbot of the important monastery of Saurbœr – had been a staunch supporter of Guðmundr’s, counseling him, for instance, about assuming the episcopal mantle (GSA cap. 97, p. 126). Sturlunga saga’s editors believe Eyjólfr may have been one of three unnamed abbots in Guðmundr’s company at Víðines in 1208 (see the index entry for Eyjólfr, 2:381–82, here 382). Another brother,



Helgastaðir, 1220

113

their successful siege might well turn into a bloody fiasco if a relief force were to hem them in.44 Ísleifr thus reversed what had been a sure-fire victory for the attackers, serving to extricate Guðmundr from a sticky, and potentially very dangerous, situation. Once Guðmundr was out of the picture, his men surrendered. Most of them, including the most prominent fighters, were given quarter – the saga, mindful of the prestige of all concerned, is careful to note who gave safe passage to whom – but some insignificant nobodies were led out and executed. The various saga redactions that have come down to us cannot agree on a body count at this juncture. In general, the death toll, not surprisingly, appears to climb as the story is elaborated: whereas Íslendinga saga and Guðmundar saga A both speak of two men put to death, recording their (otherwise unknown) names,45 the B redaction counts six, but is much more interested in the manner of their death than in

44

45

Ásbjörn, is known to have held Helgastaðir some thirty years earlier (Guðmundar saga dýra cap. 3, in Sturlunga saga, 1:166). A scene in Guðmundar saga dýra cap. 3 may be analogous to Ísleifr’s arrival at Helgastaðir, spelling out the intervening party’s leverage more clearly: “Þá gengr Guðmundr í millum ok sagði, at hann myndi þeim í mót, er eigi vildi hlýða því, er hann mælti” [Then Guðmundr (dýri) went between them and said that he would go against anyone who didn’t want to listen to what he pronounced] (in Sturlunga saga, 1:164; cf. Þorgils saga ok Hafliða cap. 19, in Sturlunga saga, 1:36; Laxdœla saga cap. 87, ÍF 5:246; Harðar saga cap. 10, ÍF 13:27). John Gillingham, especially, has been instrumental in stressing the degree to which generals in the High Middle Ages were informed by sophisticated military theoretical thinking, founded ultimately in Vegetius’s late-Roman manual; see Gillingham’s “Rejoinder: ‘Up with Orthodoxy!’ In Defense of Vegetian Warfare,” Journal of Medieval Military History 2 (2004), 149–58, responding to the critiques of Clifford J. Rogers, “The Vegetian ‘Science of Warfare’ in the Middle Ages,” and Stephen Morillo, “Battle Seeking: The Contexts and Limits of Vegetian Strategy,” both in Journal of Medieval Military History 1 (2002), 1–19 and 21–41; among them, these three articles also contain references to essential earlier scholarship on the subject, and see also Bachrach, “‘A Lying Legacy’ Revisited,” esp. pp. 188–91, focusing on the Carolingian milieu. Vegetius’s manual was among the most frequently copied and translated secular texts in the Middle Ages, but there is some dispute over whether medieval commanders had any practical use for its provisions; see, e.g., Morillo, “Battle Seeking,” p. 21 n. 3. We have no evidence that Vegetius was translated into Old Norse, nor that Icelanders had any awareness of Vegetian precepts. Traces of Vegetian thinking are arguably discernable in the King’s Mirror, a Norwegian conduct guide of the mid-1200s, but the similarities are faint and superficial; cf., e.g., Vegetius, Epitoma rei militaris, lib. 1 capp. 14 (spear throwing), 15 (archery), 16 (shooting stones by hand or with slings and staff-slings), and 18 (vaulting onto horseback), pp. 28–31, 32–33, with Konungs skuggsía, ed. Ludvig Holm-Olsen, 2nd ed. (Oslo, 1983), pp. 58–59 (spear throwing, archery, and shooting with slings and staff-slings, and horsemanship). I cannot agree with Rudolf Simek’s opinion that “[d]er Autor des Königsspiegels muß … entweder das Werk des Vegetius selbst oder das [fast wörtlich übernommen] Speculum doctrinale des Vinzenz von Beauvais gekannt haben”; see his “Zum waffenkundlichen Abschnitt des Königsspiegels,” in Speculum Regale: Der altnorwegische Königsspiegel (Konungs Skuggsjá) in der europäischen Tradition, ed. Jens Eike Schnall and Rudolf Simek (Vienna, 2000), pp. 103–25, here p. 124 (and cf. pp. 112–13). I am grateful to Jens Eike Schnall for calling this article to my attention. Íslendinga saga provides one detail missing in GSA: the nickname háleygr, “of Hålogaland,” associated with one of the executed men. This may suggest that at least one of the dead was a Norwegian émigré, presumably more expendable than men with Icelandic connections.

114

Oren Falk

Figure 5.  View to the west from Helgastaðir, showing another small portion of the home field (where the horses are grazing), as well as the road passing through Reykjadalr. About a kilometer (a little over half a mile) to the northwest of the farm, Leiðarhús, site of some ruins suspected as the location of a local assembly (which have never been excavated, however), is marked by an arrow; the hills where some of Guðmundr’s men were put to death after the Battle of Helgastaðir probably overhang this site.

their identities: all six remain anonymous but are said to have been hanged “like thieves” (sem þjófa), two on either side of the Glaumbœr gate and four at a place called Þjófahólar [Thieves’ Hills] (Figure 5).46 The C version, which otherwise tends to agree with B in its account of the battle, says nothing of executions, instead summarizing the fight with an aggregated body count that tends to agree 46

The reference to Þjófahólar suggests that, in the fourteenth century at least, the saga author knew (or thought it reasonable to imagine) a permanent gallows site associated with Helgastaðir. Cf. evidence for a handful of such sites in the Frankish Levant, unearthed by Rabei G. Khamisy, “The History and Architectural Design of Castellum Regis and Some Other Finds in the Village of Mi’ilya,” Crusades 12 (2013), 13–51, here 42–43 (esp. 42 n. 57); I am indebted to Benjamin Z. Kedar for bringing this article to my notice, and to Raphael Falk and Edna Hauser Kenner for making a copy of it available to me. Cf. also the Thiefdounes attested as an execution site in fifteenth-century Sandwich; see Borough Customs, ed. Mary Bateson, 2 vols. (London, 1904–6), 1:74. Clifford Rogers has suggested to me the translation “as thieves” for sem þjófa; this would imply that the local bœndr had identified specific individuals they held accountable for excesses carried out in the bishop’s name, and had singled them out for execution. Such a reading seems syntactically possible, but it also extends the speculative chain by several links. In any case, regardless of what any actual Reykjadalr farmers might have thought they were doing, it seems certain that the saga author meant to castigate their actions as manifest injustice; for reasons of both parsimony and fidelity to authorial tenor, therefore, I translate “like thieves.”



Helgastaðir, 1220

115

Figure 6.  Hypothetical reconstruction map (not to scale) of Helgastaðir and its immediate environs in 1220, as it may have been envisioned between c. 1280 and c. 1344 by the saga authors whose accounts of the battle of Helgastaðir survive. I have drawn the churchyard as an oval, 1.5 times as long north–south as it is wide east–west, and located just east of the farmhouse itself. I have marked two gates in the churchyard ramparts, one facing southwest towards Glaumbœr and the home meadow (marked by a circle, south of the farmhouse), the other facing north towards Múli and the coastal district; it is, of course, possible that there were more openings or fewer, and that they were otherwise situated. Also indicated are a small stand of trees, just south of the churchyard, and a gallows in the hills overlooking Leiðarhús, northwest of the farm.

with A’s tally: “Four men fell of the bishop’s men, but of Arnórr’s, one.”47 The D version gives the highest overall numbers, insisting, moreover, that the casualties were all on the bishop’s side: four are allowed to go down fighting, seven more swing. Thus ended the Battle of Helgastaðir (Figure 6). The corpses dangling at the ends of their ropes, as well as the unlucky overnight runaway’s mangled remains, serve to remind us that, on that Sunday in 1220, quasi-judicial and extra-judicial killings claimed between 150 percent and 350 percent more lives than had the combat proper.

47

In Íslendinga saga and GSA, the bishop loses one man (Gísli) in fighting, another (Eyjólfr) to a lynching, and two more (Þorgeirr and Þórðr) to summary justice, while of Arnórr’s men, one (Hámundr) falls in combat.

116

Oren Falk

The first tactical point to make about the succession of actions I have narrated is that the level of military proficiency they reveal seems, at first glance at any rate, quite abysmal. Thirteenth-century Icelandic combat as practiced at Helgastaðir appears to be a low-cost, low-tech, low-intensity affair, far removed from the chivalric lethality of the contemporary Continent, as evident even in the choice of weapons: spears and stones, not the aristocratic swords given pride of place in a majority of sagas.48 Such weapons have the decided advantage of being cheap and easy to procure, even on the spot (as the bishop’s men demonstrate). It is not entirely clear whether the stones the defenders gathered were shot or merely hurled; Hjalmar Falk, the foremost authority on Icelandic weaponry, dismisses the possibility that Norsemen availed themselves of slings, but his opinion reflects only the negative evidence of silence in the sources, which could just as easily point to the ubiquity and low regard in which such arms were held.49 At least one rare mention hints as much; the mother of one Family Saga’s sling-toting hero offers her son the following disparaging advice: “I’d like you … to bear weapons and not go about unarmed, like a woman.”50 It may also be significant that the mid-thirteenth-century King’s Mirror, a Norwegian conduct manual for the aspiring yeoman, includes training with slings among the pursuits which it is “both good and appropriate” to undertake.51

48

49

50 51

On spears, see Hjalmar Falk, Altnordische Waffenkunde (Christiania, 1914), pp. 66–90, and Birgir Loftsson, Hernaðarsaga Íslands, p. 28: “Spjót voru algeng [í Sturlungaöld] og í eigu bænda sem annarra, enda einföld í gerð og ódýr” [Spears were common (in the Sturlung Age) and owned by farmers, too, since (they were) simple to make and cheap]. On swords, see Falk, Altnordische Waffenkunde, pp. 9–65; Aslak Liestøl et al., “Sverd” (1972), in Kulturhistorisk leksikon for nordisk middelalder, fra vikingetid til reformationstid, var. eds. 22 vols. (Copenhagen, 1956–78), 17, coll. 511–42, at coll. 511–28, 537–42, and Ian G. Peirce, Ewart Oakeshott, and Lee A. Jones, Swords of the Viking Age (Woodbridge, 2002). On chivalric combat, see Richard W. Kaeuper, Chivalry and Violence in Medieval Europe (Oxford, 1999). See Falk, Altnordische Waffenkunde: “die kleine Hand- oder Stabschleuder [fehlt] in der einheimischen a[lt]nord[ischen] Literatur völlig.… Anstatt solcher Schleuder [bewarfen] unsere Vorfahren … die Steine aus freier Hand” (p. 192). See also Boyer, “La guerre en Islande,” p. 190, and Birgir Loftsson, Hernaðarsaga Íslands, pp. 29–30. Some accounts indicate the effectiveness of stones against unprotected body parts, or, conversely, the effectiveness of armor in protecting the wearer from such projectiles (see, e.g., Íslendinga saga capp. 124, 138 [129, 143], Þorgils saga skarða cap. 32 [260], all in Sturlunga saga, 1:406, 432, 2:160, respectively); but see the report of an undoubtedly armored “mikill maðr ok sterkr” [big man and strong], killed by a stone blow to his midriff, which may suggest more force than could be generated by hand alone (Þórðar saga kakala cap. 42 [205], in Sturlunga saga, 2:76; cf. p. 79). Kjalnesinga saga cap. 3: “vilda ek, at þú … bærir vápn, en færir eigi slyppr sem konur” (ÍF 14:10). Konungs skuggsía, p. 59: “Su skemtan er oc god oc halldkem æf maðr ƿæn sec at kasta af slongo hvartƿæggia langt oc þo beint” [This entertainment is also good and fitting, if a man accustoms himself to cast (stones) from a sling, (shooting) both far and also precise]. Many historians rightly doubt whether the King’s Mirror reflects actual martial practices in thirteenthcentury Norway; see, e.g., Simek, “Zum waffenkundlichen Abschnitt,” pp. 103–4, 108 (with references to earlier scholarship).



Helgastaðir, 1220

117

With possibly one exception, spears appear to have been wielded at Helgastaðir as thrusting weapons only. Þorsteinn hryggr is the sole casualty mentioned en passant to have been wounded at a distance with an edged weapon, and that only in the B redaction (C, too, mentions his having been shot, but leaves open the possibility that his leg was crushed by a rock).52 Various reasons might account for a reluctance to hurl javelins at one’s adversaries – perhaps doing so effectively would have required a higher level of martial training than any of the participants could lay claim to; or perhaps, lacking the kind of complementary armament that allowed Roman legionnaires to launch their pila before closing to fight with gladii, the defenders especially were loath to throw away the only arms available to them for close combat – but all such explanations are, of necessity, speculative.53 We can do little more than note as fact the division of labor between stones, sent flying, and spears, reserved for skewering at close quarters. As with weaponry, so also the deployment of forces at Helgastaðir does not seem to commend the leaders’ aptitude in the science of war. We hear nothing about forces held in reserve (as occasionally happens in other battle accounts) or of any attempt to organize the attackers into a formation suitable to punch through the defenses: a free-for-all is the impression the saga gives of the fighting, with individual warriors delivering or dodging blows as best they may.54 The low casualty figures – really only two men slain in combat, and one

52

53

54

GSA describes the initial phase of the battle as involving “hørð hrið með griote ok með skotum” [intense fighting … with stones and shot] (Íslendinga saga has “með grjóti ok lögum” [with stones and stabs], and GSD has “grjót ok fólkvápn” [stones and weapons of war], instead). This wording may suggest a distinction between stones and some other projectiles, or the two juxtaposed terms might be used synonymously, in contradistinction from thrusting and hacking weapons; cf. other frequent synonym pairings, often (but not always) alliterative, such as “ár ok friðr” [prosperity and peace], “sultr ok seyra” [hunger and starvation], “frið ok grið” [peace and truce]. Cf. also the language of Þórðar saga kakala cap. 42 [205]: “Var fyrst grjóthríð, en þá gengu spjótalög” [first there was stone shooting, and then began spear thrusting] (in Sturlunga saga, 2:76). On Roman arms and battle tactics, see, e.g., Adrian Goldsworthy, Roman Warfare, gen. ed. John Keegan (London, 2000), pp. 43–45, 125; I am indebted to Dave Blome for this reference. The author of the King’s Mirror advises: “ƿarazc at alldrigi later þu spiot þitt ifylkinghu laust næma þu hafir tƿau. þƿi at bætra er æitt spiot ifylkingu aƿælli en tƿau sƿærð til barðaga” [take care that you never let loose your spear (when fighting) in formation, unless you have two; because one spear is better for battle in formation on the field than two swords] (Konungs skuggsía, p. 60). The theme of a defender short on supplies who is able to recycle the missiles fired at him is famously dramatized in Brennu-Njáls saga cap. 77 (ÍF 12:187–88); cf. Íslendinga saga cap. 39 [44]: “Loftr bað sína menn eigi kasta aftr ok bíða þess, at grjóti þyrri þeim” [Loptr ordered his men not to fire back and to wait until (their adversaries) ran short of stones], and cap. 138 [143]: “Kastið þér eigi grjóti í lið þeira, því at þér takið stór högg af því sama grjóti, þá er þeir senda þat aftr” [Don’t throw stones at their troop, because you’ll get great blows from the same stones when they send them back] (both in Sturlunga saga, 1:281, 432). Cf. Íslendinga saga cap. 85 [90] for Sturla Sighvatsson’s positioning of his men for an attack, while he himself “ok nökkurir menn með honum gengu hjá, ok skipaði hann þar til, er honum þótti þurfa” [and several men with him drew near, and he detailed (them) where it seemed to

118

Oren Falk

at most who sustains lasting injury, this despite the absence of any helmets, body armor, or shields, as far as we can tell55 – likewise suggest that, even at its most ferocious, this is pretty desultory and ineffective mêlée. But, despite its apparent crudeness, the fighting at Helgastaðir bristles with clues giving away the sophistication of the proceedings. The attackers did have enough tactical sense and, more important, their men had enough discipline to form into three distinct columns, timed (albeit without any means of longdistance communication) to converge on Helgastaðir simultaneously. Later, at night, the magnates’ troops were also able to maintain an effective perimeter around the farm and seize at least one man trying to break out. Finally, their ability to come up with a vígflaki by morning speaks to their discipline (even if we only imagine them advancing under cover of overlapping shields), training, and technical skill. If, as seems likely, they were able to first fabricate a protective screen (overnight!) and then march forward under the shade of its hastily constructed frame, they must have been even more capable of decisive and coordinated action. Even Guðmundr klasi’s corrosive verse, expressly designed to demoralize, could not put these bold men off their mettle. So far, the advantages I have listed all seem to be on the side of the attackers. Let us not forget that the bishop’s men – taken by surprise, almost certainly outmanned and outmaneuvered, if not also outgunned – did manage to hold off the attackers all day Saturday. They, too, seem to have had more wits about them than first meets the eye. The stores of stones Guðmundr’s men laid up in preparation for battle performed a crucial tactical function in the fighting. The strongest testimony to their importance is, in fact, the decisive role the vígflaki played in breaking down resistance on the second day of fighting: only once the short-range tactical arsenal had been neutralized could the goðar’s men gain the upper hand. At the same time, it bears recalling that both men slain in combat were stabbed, not bashed, and that Sturla’s projectile injury appears to have inflicted no irreparable harm (nor are any other serious stoning injuries reported). Hurtling stones, while sufficient to force the attackers’ heads down, could not kill reliably at a distance. As long as they could lay down a curtain of suppres-

55

him there was need] (cf. Brennu-Njáls saga cap. 139, ÍF 12:371–73, where the Machiavellian Snorri goði eggs two other chieftains on to attack a third by promising to position his own men as a tactical reserve); cap. 99 [104], which speaks of a formation “á norrænu” [in the Norwegian style]; cap. 156 [161], for a (suspected) feigned retreat; and the detailed description of troop positioning in Þórðar saga kakala cap. 42 [205] (all in Sturlunga saga, 1:352–53, 375, 462, and 2:75–76, respectively). The best-known Norse battle formation is the wedge-shaped “svínfylking” [boar formation]; see, e.g., Konungs skuggsía, pp. 59–60, and Færeyinga saga cap. 19 (ÍF 25:42; see p. 42 n. 1 for a parallel in Vegetius, Epitoma rei militaris, lib. 3 cap. 19, pp. 163–64), and cf. Birgir Loftsson, Hernaðarsaga Íslands, pp. 61–63. Elsewhere in GSA, we do hear of various sorts of protective gear; see, for example, capp. 196, 199, 217 (pp. 206, 209, 227). The failure to mention such items at Helgastaðir is hardly dispositive evidence of their absence, of course; it seems unlikely that the attackers would not have prepared for combat, and the defenders, too, had been involved in clashes and could well expect more in the future.



Helgastaðir, 1220

119

sive fire, it seems, the bishop’s militia were able to slow down the attackers enough to win a fighting chance in hand-to-hand skirmishes. The magnates’ first priority when the fighting resumed on Sunday morning was to find a way to motivate their men forward through this lapidary rain, enabling them to bring their superior force to bear. At close quarters, Guðmundr’s followers did not really stand a chance against the concerted assault of determined, disciplined opponents. Putting things this way implies, however, that the attackers’ objective was Clausewitzian decisive victory. This may not have been the case. Despite Guðmundr’s aspirations to martyrdom (see below), the chieftains had, in the past, shown a marked disinclination to harm him, and would do so again in the future. When Arnórr had captured Guðmundr in 1218, and again when Sighvatr laid hands on him in 1222, they may have roughed him up a little (“They seized Guðmundr in his bed and dragged him down through the cathedral. He jams his hands and feet against the doorposts and planks, but they only pulled him the harder, so that he suffered great injuries”),56 but the only substantial sanction they directed at him were attempts to send him to Norway, to answer for himself before the archbishop of Trondheim. The attackers’ willingness to grant quarter to the bishop’s most valiant warriors in the aftermath of Helgastaðir likewise forces us to take their self-professed intention “rather to win by counsels than by endangering men, and also, [not] to fight so that the bishop’s life would be in danger,” a little more seriously. Perhaps this was more than just lip-service. (The defenders, too, may not have been seeking to score as many kills as possible. One more reason why we do not hear of hurled spears, besides the cost-benefit analysis of throwing away one’s side-arm for a risky shot, may precisely be due to spears’ greater lethality: raining stones down on the attackers’ heads might disincentivize them from proceeding, without drawing the kind of blood that would make a fight to the death seem necessary.) The key to understanding the attackers’ intentions, it seems to me, lies in grasping the kind of threat Guðmundr posed to them. The Gregorian Reform was late in coming to Iceland, its progress halting and idiosyncratic – in some respects, it never took, and so in the sixteenth century the island’s last Catholic bishop was still merrily begetting sons, who fought and died at his side in a doomed effort to repel Lutheranism – but under Guðmundr, it made a stormy appearance on the political stage.57 From the start of his episcopal career, 56

57

GSA cap. 161 (speaking of 1218): “Þeir toko G[uðmund] ihuilo sinni. ok drogo hann ofan eptir husunum. Hann lẻtr hendr e[ða] fẻtr idyre stafe e[ða] þile. en þeir drogu hann þui harðara. sua at uið stormeizlum var buit” (p. 179); cf. cap. 207 (speaking of 1222): “Þetta sumar it sama leto þeir G[uðmund] byskup utan fara. ok helldr harðliga leikin. af ouinum s[inum]” [That same summer they had Bishop Guðmundr go abroad, and he was rather roughly handled by his enemies] (p. 217). Gottskalks Annaler s.a. 1551 note the execution of Jón Arason, the last Catholic bishop of Hólar (in Islandske Annaler, p. 375); for Jón’s biography, see Guðbrandur Jónsson, Herra Jón Arason (Reykjavík, 1950). Already in the twelfth century, Bishop (later, Saint) Þorlákr of Skálholt had sought to introduce new, stringent standards of conduct, informed by a Reformed agenda; see

120

Oren Falk

Guðmundr seemed to mimic Thomas Becket with almost intentional precision,58 taking up the cause of clerical autonomy against his erstwhile patron, Kolbeinn Tumason. An illegitimate son of a prominent northern family, well connected but without much of a personal power base, Guðmundr was in the 1180s and 1190s a priest whose star was in the ascendant. His piety and energy attracted the attention of Kolbeinn, the strongest secular leader in the country, who appointed him his house chaplain. When the see of Hólar fell vacant, Kolbeinn saw to it that Guðmundr would be elected to fill it, doubtless in the belief that he would find in him a pliant puppet. He was soon disabused of any such hope. Guðmundr at once began to insist on the economic and juridical independence of his Church. Matters quickly went from bad to worse, with Kolbeinn getting the bishop’s men outlawed in the secular courts and Guðmundr excommunicating the goði’s followers left and right.59 Still, perhaps some accommodation could have been reached, had a skirmish

58

59

Orri Vésteinsson, Christianization, pp. 167–72. His success at enforcing new mores, however, is open to question. Orri, who notes that Þorlákr made no real effort to promote the central Gregorian project of “differentiating clearly between the secular and ecclesiastical spheres” (Christianization, p. 200), nonetheless regards Þorlákr as an effective reformer of sexual conduct (p. 171). Even in this respect, however, Orri’s estimation of Þorlákr’s achievements may be too sanguine, as suggested by the well-known tale of Jón Loptsson, the most powerful secular chieftain in late twelfth-century Iceland (and an ordained deacon), who took the bishop’s own sister as his concubine and insouciantly refused Þorlákr’s demands that he should distance himself from her, “til þess er Guð andar því í brjóst mér at skiljask viljandi við [henni]” [until God should inspire in my breast the will to part from (her)] (Þorláks saga B cap. 27, ÍF 16:178); Orri discounts this anecdote as an interpolation from the 1270s or later, Christianization, pp. 115–16, 123, but see his comment on its basis in “stories [that] are in essence true” (p. 169). Only in the 1270s were Gregorian guidelines on sexual conduct definitively written into Icelandic law; see Orri, p. 293; Bjørn Bandlien, Strategies of Passion: Love and Marriage in Old Norse Society, trans. Betsy van der Hoek (Turnhout, 2005), p. 243. GSA cap. 13 (p. 36) gives notice of Becket’s martyrdom in Guðmundr’s childhood. On the reception of Saint Thomas in Iceland, see Eiríkr Magnússon, “Preface” to his edition of Thómas saga erkibyskups, 2 vols. (London, 1875–83), 2:xxvii–xxxiv; P.G. Foote, “On the Fragmentary Text Concerning St Thomas Becket in Stock. Perg. Fol. Nr. 2,” Saga-Book 15 (1957–61), 403–50; Stefán Karlsson, “Icelandic Lives of Thomas à Becket: Questions of Authorship,” in Proceedings of the First International Saga Conference, University of Edinburgh 1971, ed. Peter Foote, Hermann Pálsson, and Desmond Slay (London, 1973), pp. 212–43; Stefanie Würth, “Thomas Becket: Ein literarisches und politisches Modell für die isländische Kirche im 13. Jahrhundert,” in Samtíðarsögur: The Contemporary Sagas, 2:878–91; and Cormack, Saints in Iceland, pp. 156–57. One Norse translation of Becket’s vita (c. 1200) was by the priest Bergr Gunnsteinsson, a personal companion of Guðmundr’s (see Foote, “Fragmentary Text,” pp. 443–44). Würth rightly comments: “Wie weit Thomas Beckets tatsächlich Vorbildcharakter für auf das Leben der isländischen Bischöfe hatte, wage ich nicht zu entscheiden – für die Sagas der betreffenden Bischöfe und damit auch für die durch diese Werke vermittelte kirchenpolitische Botschaft gab seine Biographie aber sicher das Modell ab” (“Thomas Becket,” p. 890). Guðmundr liberally sprinkled his secular rivals with bann, ‘excommunications’ or ‘interdicts’; see, e.g., GSA capp. 124–29, 135 (pp. 149–54, 159–60), matched with secular sentences passed against men associated with Guðmundr in capp. 124–25, 127–29, 142–43 (pp. 149–54, 165–66). See Jón Jóhannesson, A History of the Old Icelandic Commonwealth: Íslendinga saga, trans. Haraldur Bessason (Winnipeg, 1974; orig. 1956), p. 164, and Peter D. Clarke, The Interdict in



Helgastaðir, 1220

121

in 1208 not turned unexpectedly ugly. When Kolbeinn cornered the bishop’s entourage at Víðines, on the road from Hólar, and forced a confrontation, his motive seems to have been simply a desire to contain Guðmundr within traditional power grids: to arrest the clerics he had outlawed and to explode the bishop’s fantasy of unimpeded ecclesiastical jurisdiction. To everyone’s surprise and indeed dismay, Kolbeinn was himself killed in the encounter by a stray rock; that the bishop’s men were not much less distraught than their adversaries is suggested by the fact that, contrary to custom, no one stepped forward to acknowledge having loosed the fatal shot. Stones may not have been reliably deadly at a distance (as Helgastaðir demonstrates), but by the same token, neither could they be relied on to spare high-value targets (as Víðines illustrates). Once flung in the heat of battle, there was no telling where a tumbling projectile might touch down, and no controlling the damage it might do.60 Hence Sighvatr’s (only slightly trumped up)61 outrage when his son was clocked: plebeian warfare was again threatening to claim patrician lives. Guðmundr’s supporters might even be suspected of recidivism – did they display a pattern of intentionally trying to pick out the other side’s foremost men? – even as Sighvatr and his allies took pains to avoid consummating Guðmundr’s Becket fantasy. Kolbeinn’s misfortune had changed the stakes but not the rules of the game: the coalition of goðar who descended on Hólar in 1209 meant, first, to avenge his death on his killers, and second, to complete Kolbeinn’s life mission by ousting Guðmundr from his power base. A third goal, to punish the bishop personally for his responsibility (as leader of the opposition) for Kolbeinn’s death, only coalesced gradually in the wake of a summons from the archbishop for both Guðmundr and his lay rivals to appear before him. It seems clear that the secular magnates did not feel cowed by the prospect of contending with Guðmundr in ecclesiastical court: we know of enough non-Gregorian clergy in Iceland at the time, and of a lukewarm-enough response to Guðmundr’s unheard-of innovations across the ecclesiastical board, that we may safely assume similar attitudes prevailed among the Norwegian chapter and prelates.62 Indeed, Guðmundar

60

61

62

the Thirteenth Century: A Question of Collective Guilt (Oxford and New York, 2007), pp. 1, 75–76, for distinctions among ecclesiastical sanctions. The best literary and historical analysis to date of the various accounts of Kolbeinn’s killing is in Elizabeth M. Walgenbach’s undergraduate Honors Thesis, “A Problem of Corpses: Wondrous Killings and their Consequences in Thirteenth-Century Iceland” (Ithaca, 2007), esp. pp. 28–38. The main reason to suspect Sighvatr of exaggerating his outrage is his reference to Sturla (or, in GSC, Tumi) as a “boy.” Born in 1199, Sturla was young, but certainly not so young as to make him an illegitimate target by anyone’s standards (and Tumi was a year his senior). For expressions of pious doubt about Guðmundr (including by his future patron, Bishop Brandr), see GSA cap. 25 (pp. 61–63). One of Guðmundr’s contemporaries, Bishop Páll Jónsson of Skalhólt (r. 1195–1211), is a poster boy for anti-Gregorian attitudes; his saga’s account of Guðmundr’s conflicts with Kolbeinn is partial to the latter (Páls saga cap. 15, ÍF 16:321–23). Guðmundr is said to have been au courant with new religious fashions already in the 1180s: “Marga lute toc hann þa upp til trv ser er enge maðr uisse aðr at ne ein maðr hafe gert aðr her a landi” [he then adopted many things into his worship which no one had known anyone to have practiced here in (this) country] (GSA cap. 25, p. 62). Arons saga cap. 14 asserts that praying

122

Oren Falk

saga B imagines a lengthy interview in Trondheim, at which an unsympathetic archbishop grilled Guðmundr on his aberrant conduct (though, as the saga tells it, his good subordinate ultimately won him over handily), and independent records confirm that Guðmundr’s relations with his superiors were hardly idyllic.63 Nor did lower clergy necessarily warm to Guðmundr’s reforming initiatives: throughout his career, he found many of the shepherds under his pastoral care just as truculent as the lay sheep they attended to. On Guðmundr’s arrival in Reykjadalr in 1220, for example, the priest who farmed a moiety of the estate at Einarsstaðir promptly flocked to Ívarr of Múli’s following, taking his livestock with him; the priest’s son later set on one of Guðmundr’s followers, narrowly missing the man but killing his horse under him.64 It may well be that Iceland’s unusual political culture received Gregorian ideals in a way that made them unrecognizable from a Continental perspective, opening as deep a gulf between Guðmundr and his metropolitan brethren (who themselves were not all of one mind, of course) as separated him from the secular goðar. Throughout Europe, reformist prelates were routinely thrown into alliances of necessity with restive factions of the aristocracy against their kings and emperors – as happened to Pope Gregory VII in the struggle against Henry IV, or to a series of bishops and archbishops maneuvering among a welter of pretenders in twelfth- and thirteenth-century Norway.65 In less neatly strati-

63

64

65

prone with arms outspread, chanting Benedicte and Ave Maria, was an innovation taught by Bishop Guðmundr (in Sturlunga saga, 2:268). I am indebted to Joel Anderson for stimulating my thinking about these issues. See GSB capp. 99–101 (= Biskupa sögur, 1:573–83); Ciklamini, “Hidden and Revealed,” p. 248, speculates on the charges against which this imagined debate polemicizes. Two letters of Gregory IX’s, dated 11 May 1237 (almost two months after Guðmundr had, in fact, died), instruct the archbishop to suspend clergy Guðmundr had ordained irregularly, and even to relieve the bishop from office: “Qvia … privatus penitus lumine oculorum clericos ad ordines sacros promovit, ipso manum imponente, ac diacono legente verba, que idem in oratione facienda dicere tenebatur” [because … he, being practically deprived of the light of his eyes, has promoted clerics to sacred orders, laying on his own hand but having a deacon read out the words which he ought to have spoken in prayer]; “mandamus quatinus … eumdem moneas attentius et inducas, ut anime sue saluti providere consulens cedat episcopatui” [we order that … you sternly urge and persuade him that, taking counsel to look to the salvation of his own soul, he step down from the episcopacy]; Diplomatarium Norvegicum: Oldbreve til Kundskab om Norges indre og ydre Forholde, Sprog, Slægter, Sæder, Lovgivning og Rettergand i Middelalderen, ed. C.A. Lange and Carl R. Unger et al., 22 vols. (Christiania/Oslo, 1847– 1995), 1:14–15 [§§17–18]. Other sources suggest the archbishop had not waited for papal instruction, deposing Guðmundr already in 1232 (Annales Reseniani, Henrik Høyers Annaler, Annales regii, Skálholts-Annaler, Lögmanns-annáll, Gottskalks Annaler, and Oddaveria Annall, all in Islandske Annaler, pp. 25, 64, 129, 187, 256, 327, 480; Íslendinga saga capp. 76, 99 [81, 104], in Sturlunga saga, 1:337, 374). See GSA capp. 169–70 (pp. 184–85) for the hostile priest at Einarsstaðir and his quarrelsome son, Oddr. The priest is later reported to have been killed, alongside his son Gunnsteinn (scribal discrepancy? or perhaps a different son?), by his former partner in farming, the target of Oddr’s assault (cap. 189, p. 201). See, e.g., I. S. Robinson, Henry IV of Germany, 1056–1106 (Cambridge, 1999), pp. 122–23; John Eldevik, Episcopal Power and Ecclesiastical Reform in the German Empire: Tithes, Lordship,



Helgastaðir, 1220

123

fied Iceland, however, there was no crown against which bishops and second-tier secular magnates might unify. Instead, members of a barely differentiated political class made up of the land-owning goðar and bœndr, who together constituted perhaps 5–10 percent of the total population, were uniformly motivated by their structural position to resist any reforming efforts, denying maverick churchmen footholds for making common cause with disgruntled aristocrats against an overlord.66 Reformers like Guðmundr were thus left with little choice but to seek allies among the dregs of sub-political society, the tenants and hired hands who were officially designated dependents of bœndr, less than full legal persons. Although a handful of Guðmundr’s supporters (for the most part, his relatives) were drawn from the ruling elites, the sort of men we might expect to find in other sagas wielding well-crafted swords and performing deeds of heroism,67 the vast majority of those who fell in behind his call for reform were decidedly common men, members of the classes who had little access to gear finer than spears and stones. If Guðmundr’s agenda appears no less Mendicant than Gregorian, then (and bearing in mind that Mendicant ideals may already have been in the air during his lifetime, and had surely reached Iceland by the time his sagas were written), this may be due not just to his biographers’ anachronism, but also to the real exigencies of the political landscape within which he roamed. No Archbishop of Trondheim, ensconced in his lofty romanesque cathedral (Figure 7), was likely to have enormous empathy with such uncivilized rusticity. Respectful of the archbishop’s summons, Arnórr took ship to Norway already in 1212, but was unable to confront Guðmundr there before he returned to

66

67

and Community, 950–1150 (Cambridge, 2012), pp. 168–71, 207–8; Sverre Bagge, From Gang Leader to the Lord’s Annointed: Kingship in Sverris saga and Hákonar saga Hákonarsonar (Odense, 1996), pp. 74–80, 83–84, 119–21. The situation in Norway is made opaque by the fact that the sources, in partisan support of the centralizing royalist ideology promoted by Sverrir and Hákon (ultimate victors in the struggles for the crown), tend to focus exclusively on claimants to the throne, and to regard those aristocrats who had no royal aspirations as unproblematically subordinate to them. Orri Vésteinsson, Christianization, does not much concern himself with the question of reform (though see pp. 167–78). He focuses on Guðmundr’s “inability to select his targets tactfully,” a personal failing, as having caused him to “forfei[t] any sympathy he might have had among the local householders” and become “the worst agent [reformers] could hope for,” rather than on structural reasons for the wedge driven between bishop and bœndr (pp. 177, 175). As his discussion makes clear, however, the social and economic interests (and often the personal identities) of the landholding classes as such and those of conservative churchmen as such were almost entirely coterminous (see esp. pp. 182–209). See, e.g., Guðmundr’s cousin, the goði Ǫgmundr sneis, stripped of his chieftaincy as a consequence of his attempt to aid the bishop in 1209 (GSA cap. 143, p. 166; cf. cap. 141, p. 163), and Aron Hjǫrleifsson, one of the bishop’s champions in the early 1220s, who wears “auruggri bryniu. ok godum hialme. og sterkri hlif” [an impregnable byrnie and a good helmet and a strong shield], wielding “mikit sax. sem stor suerd. og uar þat hit agiætasta uopn” [a large dagger, big as a sword, and it was the most excellent of weapons] (cap. 196, p. 206). Aron cut a figure heroic enough to receive his own saga, which has been compared to the classic Gísla saga.

124

Oren Falk

Figure 7.  Western façade of Trondheim cathedral.

Iceland in 1214, just as his nemesis was traveling in the opposite direction. The bishop’s four-year stay in Norway may, perhaps, be read in part as a reflex of the traditional Icelandic penalty of outlawry and exile. Outlawry came in two forms, lifelong and temporary, the latter (according to the idealizing law code) to last at least three years.68 By staying abroad for a period that matched or exceeded the sentence allotted to outlaws, Guðmundr may have hoped to expiate his offence in Arnórr’s eyes. But in 1218, as soon as Guðmundr was back on Icelandic soil, Arnórr again tried to force him into a confrontation on the archbishop’s turf. Even now, Arnórr evidently did not feel that the case he 68

See Grágás Ia §53: “Ef hann komr fyr ut hingat en hann hafi .iii. vetr abrott verit … verðr hann scogar maðr.… Ef hann cømr út hingat þa er hann hefir iii. vetr abrott verit. þá er hann sva sycn sem aldrigi quæmi su secþ ahendr honom” [If he comes out hither before he has spent 3 winters abroad, he becomes a lifelong outlaw.… If he comes out hither when he has spent 3 winters abroad, then he is as reprieved as if he had never suffered conviction] (p. 91). See Jesse L. Byock, “Outlawry (in Free State Iceland),” in Medieval Scandinavia: An Encyclopedia, pp. 460–61; Oren Falk, “Outlawry,” in The Wiley-Blackwell Encyclopedia of The Viking World, ed. Judith Jesch and Christina Lee (Chichester, forthcoming).



Helgastaðir, 1220

125

had would prove a liability in an ecclesiastical court. Thanks to Eyjólfr Kársson’s audacity, however, Guðmundr again slipped his grasp before the two sides’ skills at litigation could be put to the test. In all of these instances, as well as in the second four-year exile forced on Guðmundr in 1222 after his men had killed Sighvatr’s son Tumi, we may see the goðar acting against the person of the bishop in accordance with the Icelandic feuding idiom of personal responsibility, albeit limiting their sanction to attempts to neutralize Guðmundr politically and send him into exile, rather than kill him outright.69 His foremost captains, meanwhile, the men directly responsible for Tumi’s slaying in 1222, were ruthlessly destroyed: Eyjóflr Kársson was killed, another man’s grave – he had died of natural causes in the interim – was defiled, and a third was forced to spend the next four years on the run throughout Iceland, a hunted fugitive, before he managed to find passage abroad and beyond the grasp of the Sturlungs.70 Similarly, in 1209, Arnórr had made every effort to pinpoint the identity of Kolbeinn’s slayer: “They compelled four good men, 2 priests and 2 laymen, to bear iron [swearing] that they had not struck down Kolbeinn, and they were all exonerated.”71 Kolbeinn’s blood cried out to be avenged, but Arnórr did his utmost to find an unmitred head on which he might visit capital vengeance. The case in 1220 is slightly different. Here, the motivation for action came not from the goðar but from local farmers alarmed by the locust-like menace of the bishop’s swarm. Perhaps because of his personal experiences of hardship in his youth, perhaps under the influence of a climate of ideals not unlike that which drove his Italian contemporary, Saint Francis, Guðmundr is depicted as taking notions of charity much more seriously than did many of his predecessors and contemporaries.72 Guðmundr’s biographers are forced to admit (even 69

70

71 72

Cf. Þorláks saga B cap. 26: “Jón ætlaði enn sem fyrr byskup með kúgan kløkkan at gera, heldr en vápn á hann at bera” [Jón intended once more to cow the bishop through oppression as before, rather than bear weapons against him] (ÍF 16:175). Cf. also Íslendinga saga cap. 156 [161], where a highly partisan Bishop Sigvarðr of Skálholt steps, in full regalia, into the thick of a battle to pronounce excommunications on his adversaries: “fló grjótit á hvára tveggja hlið honum ok yfir höfuðit, sem í drífu sæi. En er menn kenndu, vildu engir honum mein gera, ok stöðvaðist þá bardaginn” [the stones flew on either side of him and over his head, so that it looked like a blizzard. But when men recognized (him), no one wanted to do him harm, and then the battle died down] (in Sturlunga saga, 1:464). See GSA capp. 191–207 (pp. 202–17) for the hard-fought 1222 expedition, and capp. 209–23 (pp. 218–32) for the fate of Aron Hjôrleifsson, the sole survivor among Guðmundr’s lieutenants, in the subsequent years. GSA cap. 143: “Fiora goða menn. presta .ij. ok leicmenn .ij. nauðgøðo þeir til iarnburðar vm þat. at þeir hefðe eigi unnit a Kolb[eine]. ok urðo þeir allir uel skirir” (p. 166). For some of the privations Guðmundr suffered in his youth, see GSA capp. 11 and 14 (pp. 34, 43–47). Around 1207, Guðmundr is said to have tried to force the bœndr to pay not just “tiundar … e[ða] kirkna fiar [gjôld enn] við tøko við fatẻkum frẻndum sinum. Bẻndr toko þui þungliga” [tithes … or (payments) of church dues (but also) care of their poor relatives; the farmers took this badly] (GSA cap. 128, p. 153, emphasis mine; misconstrued by Skórzewska, Constructing a Cult, p. 182). Care for one’s destitute kinsman was a legal duty, according to Grágás Ib (§128, pp. 3–7). Svanhildur Óskarsdóttir considers the possibility of Franciscan, and

126

Oren Falk

as they go to great lengths to identify any reputable men among his supporters) that the huge following he attracted was largely composed of men and women of decidedly low status.73 Such people were sternly frowned on by respectable Icelanders, especially if they were able-bodied (as many in the bishop’s ravenous retinue doubtless were): a society clinging to precarious life on the margins of the Arctic Circle could scarcely afford to feed idle mouths. At the same time, the increasingly competitive political environment of the thirteenth century led to the frequent recruitment of ad hoc levies, men snatched from the plow (or, in Iceland, more likely from the shepherd’s crook or the fisherman’s net) to march on campaign with their chieftains; even at the height of thirteenth-century political turbulence, no Icelandic goði ever commanded a standing army.74 Like the routiers and later Free Companies on the Continent, such conscripts often found themselves, in the lulls between skirmishes among the goðar, possessed of the skills for harrying and looting, but no employer to invest their violence with decorum.75 Joining the bishop’s following, helping him to assert the freedom of Mother Church from secular domination as well as to collect the “tithes” and “alms” he and his followers felt were their due, was an alluring opportunity for such men, untethered from the normal social grids. This synergy of Guðmundr’s humble style of religious reform, on the one hand, and real social agitation, on the other, can further explain the anxiety of the Reykjadalr farmers in 1220, and it is what seems to have determined the immediate objective of the assault on Helgastaðir: to disperse Guðmundr’s host. It was Ívarr of Múli, then, not the more exalted goðar, who took charge of setting this campaign’s strategic parameters, and for him, shattering the mass of commoners posed a higher priority than humiliating Bishop Guðmundr or

73

74

75

esp. Cistercian, influence on Guðmundr; see “Að kenna og rita tíða á millum: Um trúarviðhorf Guðmundar Arasonar,” Skáldskaparmál: Tímarit um íslenskar bókmenntir fyrri alda 2 (1992), 229–38 (and references to earlier scholarship in a similar vein at p. 230 n. 2). See, e.g., GSA cap. 130, which names three prominent men “ok margir aðrir menn þeir er røsquir voro” [and many other people, those who were valiant], but also concedes the presence of “nockurir o spectar menn” [some riotous men] among Guðmundr’s followers (p. 155). On the poor in Iceland, see Miller, Bloodtaking and Peacemaking, pp. 147–54, Orri Vésteinsson, Christianization, pp. 81–84, as well as Wolfgang Gerhold, Armut und Armenfürsorge im mittelalterischen Island (Heidelberg, 2002); and, specifically on the poor among Guðmundr’s adherents, see Jón Jóhannesson, Commonwealth, pp. 200, 212–14, and Meulengracht Sørensen, Saga and Society, pp. 60–61. We have seen, above, references to Arnórr’s, Sighvatr’s, and perhaps also Ívarr’s raising of levies from their respective districts. The most striking examples of such practice occur in Þórðar saga kakala, set in the 1250s, which tells of one of Sighvatr’s sons’ failed bid to gain hegemony; Þórðr spends much of his time trying to raise troops, going so far as to employ an unscrupulous recruiting sergeant whose methods include kidnapping a man’s wife in order to force him to enlist, and murdering a reluctant bóndi, along with his brother (see Þórðar saga kakala capp. 6, 14 [169, 177], in Sturlunga saga, 2:12–13, 28). Cf. Philippe Contamine, La Guerre au Moyen Age (Paris, 1980), pp. 396–405 (trans. Michael Jones, War in the Middle Ages [Oxford, 1984], pp. 243–49), and Matthew Strickland, War and Chivalry: The Conduct and Perception of War in England and Normandy, 1066–1217 (Cambridge, 1996), pp. 291–321.



Helgastaðir, 1220

127

even neutralizing his warlike captains. We can see these priorities in action in the lynching of Eyjólfr hríðarefni, the less hurried but no less arbitrary hangings of Þórðr Arason and Þorgeirr (who may not even have been an Icelander), the maiming of the wretched Þorsteinn hryggr, and perhaps even in the battlefield slayings of Gísli and Hámundr; what all these men have in common is their all-but-anonymity, and the documented marginality of two or three among them suggests that they all may have been allotted their fates precisely because they were ciphers, synecdochic of that rabble which the attackers were determined to strike fear into and scatter to the winds. Capturing the bishop himself would surely have been a nice bonus for Arnórr and Sighvatr – so also, incidentally, would the death of Ívarr in the course of fighting, which could have created a local political vacuum into which they might project their own authority – but it was not a sine qua non on which to insist. Ísleifr’s timely arrival to break up the fight may thus have seemed as fortunate to the attackers as to the defenders. After nearly twenty-four hours of bloodletting, adrenaline, and sheer physical exertion, everyone on both sides would have been fatigued and edgy. Even the attackers’ morale may have grown brittle. Recalling Kolbeinn’s demise, and Sturla’s (and perhaps Tumi’s and Sighvatr’s) more recent close call, the chieftains might have reflected that, even if those in their crosshairs were disposable hoi polloi, they themselves were also exposed to some danger as long as shots and blows flew unchecked. Thanks to Ísleifr, the besiegers were now spared having to make the decision: are they sufficiently committed to stay the course, ordering their men into the church and practically ensuring the sacrilege of a bloodbath (and continuing to run the risk, however statistically small it may have been, that they themselves would suffer Kolbeinn’s fate), or do they wish to uphold their stated intention “to prevail by counsels,” and reap the fruits of an incomplete (but by no means unsatisfactory) victory? Sighvatr, godless ruffian that he was, might not have minded a massacre too much, and would hardly have cared whether it took place on secular ground or sacred (though he too would doubtless have appreciated the folly of pressing on with Ísleifr’s flying column at his back).76 But Arnórr, to all 76

Sighvatr’s (and his sons’) thuggish style of leadership is captured in an incident from 1217, when Sturla assaulted a prominent Eyjafjôrðr bóndi, injuring him gravely. Sighvatr learnt of Sturla’s exploits, and “spurði hann Sturlu, hvárt þat væri satt, at hann hefði vegit eða særðan inn bezta bónda.… Sturla lézt ætla, at því myndi verr, at hann myndi eigi dauðr. Síðan tók Sighvatr á inum mestum hrakningum við Sturlu.… Síðan átti Tumi hlut at ok segir, at þeir skyldi svá fleiri fara, at þeir væri barðir, sagði reynt, at bændr mátti eigi með góðu tryggja.… Sighvatr bað [Sturlu] ganga í lokrekkjuna til sín. Ok er hann kom þar, tók Sighvatr til orða: ‘Ekki þykkir mér þetta svá illa sem ek læt, ok mun ek um klappa eftir. En þú lát sem þú vitir eigi.’ Síðan … fekk [Sighvatr] sætta þá, ok váru gervir þrír tigir hundraða, ok kom þat fé seint fram” [he asked Sturla whether it were true that he had killed or injured the best of bœndr.… Sturla said it was all the worse, to his mind, that he hadn’t died. Then Sighvatr began to reprimand Sturla sternly.… Afterwards Tumi chimed in and says that so might it happen with many others, that they should be beaten; he said (this) proved that treating bœndr well would not make them reliable.… Sighvatr asked (Sturla) to come in to him in his bed-closet. And when he had come there, Sighvatr began to speak: “This seems to me nowhere as bad as I let on, and I will patch it

128

Oren Falk

appearances, was a genuinely pious man; he had no quarrel with God, only with His servant Guðmundr. We have already seen some evidence to suggest that Arnórr thought his cause just, including his willingness to resort to divine judgment to ferret out who had killed Kolbeinn – himself, incidentally, a devotee of the Virgin, who wrote hymns in her honor.77 The contrast between Sighvatr and Arnórr’s religious sentiments is teased out gently (and not without ambiguity) in the conversation the saga author puts into their mouths the previous evening, when the former sarcastically proposes that God had intervened to grant Arnórr respite from his ailments in order to allow him to fight against the miracle-working bishop, while the latter resorts to pious agnosticism: “I call it a circumstance … but no miracle.”78 For Arnórr, at least, the prospect of attacking Guðmundr’s men inside the church would likely have appeared markedly less attractive than that of beating them in the open air. Ísleifr may thus have been as much deus ex machina for the besiegers as he was for Guðmundr, allowing them to accept a face-saving compromise without having to put either their moderation or their violent self-confidence to the ultimate test.79 They welcomed his intervention. True, it rescued Bishop Guðmundr from their grasp; but it also left

77

78

79

all up later. But you let on as if you know nothing.” Afterwards … (Sighvatr) got a settlement, and thirty hundreds were awarded, but that money was slow to be paid over] (Íslendinga saga cap. 32 [37], in Sturlunga saga, 1:261–62). Kolbeinn’s Marian poetry, none of which survives, is mentioned in GSB cap. 40 (= Biskupa sögur, 1:569–70) and GSC capp. 46 and 74. For the extant fragments of his verse, see Den norsk-islandske skjaldedigtning, ed. Finnur Jónsson; A: Tekst efter håndskrifterne, 2 vols.; B: Rettet Tekst, 2 vols. (Copenhagen and Christiania, 1910–15), B2:45–49. This edition is to be superseded by Skaldic Poetry of the Scandinavian Middle Ages, ed. Margaret Clunies Ross et al., 9 vols. expected (Turnhout, 2007–); so far, only part of Kolbeinn’s surviving oeuvre has appeared: see his Jónsvísur, ed. Beatrice La Farge, in 7, part 1:223–27. This exchange has received considerable attention, and although scholars disagree on how it should be interpreted, there is consensus on its significance. Space prohibits me from treating the dialogue thoroughly here, but I elaborate on it further in This Spattered Isle: Violence and Risk in Medieval Iceland, Chapter Two (unpublished manuscript). See also W. P. Ker, “Gudmund Arason” (orig. 1907), in his Collected Essays, 2:152–72, here 152–53, 164; Einar Ól. Sveinsson, The Age of the Sturlungs, pp. 123–24; Peter Foote, “Secular Attitudes in Early Iceland” (orig. 1974), repr. in his Aurvandilstá: Norse Studies, ed. Michael Barnes, Hans BekkerNielsen, and Gerd Wolfgang Weber (Odense, 1984), pp. 31–46, here pp. 45–46; Guðrún Nordal, Ethics and Action in Thirteenth-Century Iceland (Odense, 1998), p. 185; and Joel Anderson, “The Miraculous Water of Guðmundr Arason and the Limits of Holiness in Medieval Iceland” (unpublished MA thesis, Reykjavík, 2008), pp. 20–24. All of these authors agree in reading the exchange as an expression of “secular attitudes,” as Foote puts it (adding that “engaged as they are in a hot battle against a bishop who himself had a reputation for wonder-making, the remark attributed to Sighvatr could hardly be more coolly cynical or wickedly ironic,” p. 46). Contrast Lars Lönnroth, “Saga and Jartegn: The Appeal of Mystery in Saga Texts,” in Die Aktualität der Saga: Festschrift für Hans Schottmann, ed. Stig Toftgaard Andersen (Berlin and New York, 1999), pp. 111–23, here pp. 115–17; Arnved Nedkvitne, Lay Belief in Norse Society, 1000–1350 (Copenhagen, 2009), p. 227; and cf. Skórzewska, Constructing a Cult, pp. 120–22. On moderation as an Icelandic chieftainly ideal, see Theodore M. Andersson, “The Displacement of the Heroic Ideal in the Family Sagas,” Speculum 45 (1970), 575–93; Byock, Medieval Iceland, pp. 128–29.



Helgastaðir, 1220

129

them in a position to make leisurely examples of two (or six or seven) small fry, who could be led out and unceremoniously strung up, driving home to the remainder of Guðmundr’s ragtag following a message: the wages of service in the bishop’s brigade are death. The insignificant Battle of Helgastaðir thus turns out to have significance for the historian of medieval Iceland as a prism through which the agendas of different social players at various points in time are refracted. Nothing is quite as it initially seems: the military history of this unremarkable skirmish is neither so brutish nor so inept as it first appears, the attackers’ intentions are hardly transparent, and Ísleifr’s tie-breaker role is more polychromatic than that of a (partisan) white knight in shining armor. Even the ground of social realities on which Helgastaðir rests appears unstable, its strata “all a-tremble beneath the warriors”: those who die in the bishop’s defense all seem to be marginal men, as well we might expect them to be, but among the attackers, leadership roles are not apportioned quite in accordance with what ascribed status would have led us to predict. Finally, it is notable that despite Icelanders’ reputation for unbridled violence – “their literature,” after all, is “all about killing each other,” as one modern commentator famously declares – the combat at Helgastaðir fells remarkably few victims, whereas summary justice plows a deeper, in some versions considerably deeper, furrow.80 Indeed, it seems possible that design, rather than blind luck, lay behind the low casualty figures during those interminable hours of thrusting and parrying. No similar restraint stayed the headsman’s hand after the defenders’ surrender. For the military historian of medieval Europe, this curious engagement offers various lessons. Helgastaðir definitively punctures any illusions we may harbor about the utility of strict evolutionary models of military science or technological determinism.81 The combatants in northern Iceland in 1220 resort, on the one 80

81

Tom Shippey, “Je sui uns hom” (reviewing Byock, Medieval Iceland), London Review of Books 11:11 (1 June 1989), 16–17, here 17. Shippey’s phrase is often quoted, as, e.g., on the back cover of the paperback edition of Byock’s Medieval Iceland. Cf. the Battle of Hólar in 1209, where six or seven of the bishop’s men die fighting (while the chieftains lose five or six), and three others are executed in the aftermath (GSA capp. 141–42, pp. 164–65; Íslendinga saga cap. 24 [29], in Sturlunga saga, 1:252–53). One of these executions is every bit as spectacular as that of the eighteenth-century regicide, Damiens (on which see Michel Foucault, Surveiller et punir: Naissance de la prison [Paris, 1975], pp. 9–11, trans. Alan Sheridan, Discipline and Punish: The Birth of the Prison [London and New York, 1977], pp. 3–5), though doubtless for different reasons; see GSA cap. 142: “Þa toc Sueinn Jons s[on] til orða. Gera man ec kost a ut at ganga. en þeir spurðu hverr sa vẻre. Hann svar[aðe]. Ef þer limit mic at høndum ok fotum aðr þer halshøggit mic.… Var Suein þa limaðr ok song Mariu vers meðan. En siþan rette hann halsin undir høgit” [Then up spoke Sveinn Jónsson: “I shall make you a condition in order to come out.” But they asked him what it would be. He replied: “That you hack off my arms and legs before you behead me.” … Sveinn then had his limbs hacked off while he sang Ave Maria. But afterwards he stretched out his neck for the blow] (p. 165). Cf. Birgir Loftsson, Hernaðarsaga Íslands, p. 30: “Sumir íslenskir sagnfræðingar hafa tiltekið grjótkast Íslendinga sem vott um frumstæða bardagaaðferð þeirra. En Íslendingar voru ekki

130

Oren Falk

hand, to mobbing tactics, feuding mentality, and (iron-tipped) sticks and stones. On the other hand, they are savvy about applying just the right amount of force to both adversaries’ and allies’ pressure points for achieving quite specific goals, they remain disciplined and cool-headed (not to mention technically competent) under fire, and they prove agile in exploiting emergent opportunities and fluid circumstances. No single index we might focus on, such as weapons technology, military organization, or logistics, is a reliable predictor of these Icelanders’ overall martial aptitude, nor does considering the complete picture allow us to conclude with confidence whether the combatants at Helgastaðir were simply lucky to have somehow muddled through or whether, indeed, they deserve recognition as martial geniuses. Perhaps, rather than fetishize the consummate generalship of a Robert Guiscard or a Richard Coeur-de-lion (or, for that matter, an Arnórr, a Sighvatr, or an Eyjólfr Kársson), and by implication disparage the mediocrity of most other medieval masters of war, we may be well advised to admit our inability to pierce through the shroud of ignorance imposed on us by sources that are almost everywhere – Iceland excepted – too thin to sustain the kind of close reading I have attempted here. We may also suspect, based on the view from the Thieves’ Hills overlooking Helgastaðir, that the noose may often have been mightier than the sword. In the Middle Ages, at any rate – and can parallels with the early twenty-first century really be dismissed as coincidental? – “legal” measures adopted to obviate the need for warfare may routinely have exacted far heavier dues than the storm of battle itself. For medieval Icelanders, meanwhile, the chief significance of the Battle of Helgastaðir may have been in introducing a new idiom into their political discourse. Over the course of the preceding century, as the sizes of war posses mushroomed and as potentates made increasingly bold bids to attain absolute lordship, politics had grown steadily more rarified, a sport for the elites in which common farmers counted only as peons. At Helgastaðir, however, no longer were magnates of the first order – goðar and bishops – merely duking it out among themselves in accordance with the structures of family feud (if ever they had in fact acted out this saga commonplace).82 Instead, mid-level landed men, bœndr like Ívarr and his neighbors so fearful for the integrity of their pantries, found it possible to leverage magnate support without paying an exorbitant cost.83 Moreover, and just as importantly, these rank-and-file farmers

82

83

einir um það og þrátt fyrir að Norðmenn notuðu mikið handboga í bardögum á 12. og 13. öld gripu þeir einnig til grjótsins eins og Íslendingar” [Some historians have taken Icelanders’ (resort to) stone-throwing as testimony to their primitive military method. But the Icelanders were not alone in this, and although Norwegians made much use of bows in battles of the twelfth and thirteenth centuries, they, too, like the Icelanders, resorted to throwing stones]. The idiom of family feud is evident at Helgastaðir in Sighvatr’s passing allusion to the Battle of Melr; see n. 90 below. Neither Bishop Guðmundr nor Sighvatr had had anything to do with Melr, so the reference is a bald attempt to motivate one of Sighvatr’s supporters by seeking to align his familial grievance against Eyjólfr Kársson with the impersonal cause of fighting on Sighvatr’s behalf against the bishop. A clear example of the tariff bœndr might be expected to pay for a goði’s support is provided in



Helgastaðir, 1220

131

were looking both up and down, not only manipulating their betters but also, at the same time, keeping a watchful eye on their inferiors. The cause to which they enlisted magnate support at Helgastaðir was a gambit aimed at quelling a menacing Gregorian seepage from below, the swelling tide of low-status laborers and vagrants determined to exercise agency in the political community. Ívarr and his fellow bœndr had no intention of upsetting or displacing the power structures that propped the goðar up. Their attention was consumed by the risky rise of militarized commoners, a rabble of beggars and troublemakers; only incidentally did they discover (or perhaps only make patent the truth they had known all along) that appealing for patronage was also an inspired method of bending the goðar to their will. If there was reform at work here, it was a thoroughly reactionary reform: mid- and upper-echelon Icelanders locking arms to march together against a newfangled bishop and his plebeian adherents. Yet we must not succumb to the temptation of lumping participants into overly neat categories: the men who came together at Helgastaðir were historical subjects, not sociological types. Individual bœndr made choices that steered some of them in this direction, others in another. For every Ívarr, we meet an Ísleifr; for every bishop-bashing priest of Einarsstaðir we have his partner in farming, a staunch supporter of Guðmundr’s. Introducing Iceland’s riffraff into a game they had never before been permitted to play was the least of Bishop Guðmundr’s achievements: by pulling paupers and vagrants into the political arena, he also created, e nihilo, novel options for the more seasoned players, those smallish landholders who had formerly occupied the bottom rungs and could now choose between resisting their displacement from below and embracing it as an opportunity for upward mobility. Had fortune smiled on the bishop’s party a little more consistently, men like Eyjólfr Kársson might have risen to become the movers and shakers of the mid-thirteenth century, borne up on the shoulders of inconsequential and nameless men like those whom the victors at Helgastaðir were at such pains to put to death; it bears remembering that Sturla Sighvatsson, arguably the most ambitious and dangerous magnate in Iceland until he fell at the Battle of Ørlygsstaðir in 1238, came within an inch of being brained in 1220. What would have happened, had a few more stones found their mark that day? “Warfare,” says Clausewitz, “is the realm of danger … of uncertainty … of coincidence.”84 What might the history of the island have looked like if the experimental plot Guðmundr watered had chanced to bear sturdier fruit, if the men of the old guard had perished while innovators had survived and thrived?

84

Eyrbyggja saga cap. 31, where a certain Úlfarr “handsalaði … Arnkatli fé sitt allt” [handed over all of his property to Arnkell] in order to gain his protection (“ok gerðisk hann þá vernaðarmaðr Úlfars” [and he then became Úlfarr’s guardian], ÍF 4:84). For solid analysis of this episode, see Byock, Viking Age Iceland, pp. 99–117. Carl von Clausewitz, Vom Kriege, lib. 1 cap. 3: “Der Krieg ist das Gebiet der Gefahr.… Der Krieg ist das Gebiet der Ungewißheit.… Der Krieg ist das Gebiet des Zufalls,” in Hinterlassene Werke, 10 vols. (Berlin, 1832–37), 1:56–58, trans. Michael Howard and Peter Paret, On War (Princeton, 1976), p. 119.

132

Oren Falk

It also bears pondering whether, or to what extent, the fate of the Bishop of Hólar and his followers may mirror the fates of other prelates and their retinues across Europe, in an age of contested ideological innovations and escalating social tensions. For those of us who do not work on Church history proper, Gregorian Reform may often seem like a bit of a distant abstraction, a question of clerical celibacy, investiture, and spiritual red tape that, most of the time, would have had little impact on how most people outside the cloister actually went about their business (so long as their kings did not involve them in headlong conflict with the Princes of the Church over its implementation). In truth, reform must have looked very different from different vantage points across medieval Europe, and for some, it may have amounted to little more than that. But Helgastaðir illuminates some of the ways in which the revolutionary thinking of a clerical intelligentsia might set in motion tangible landslides in the social and political realms inhabited by worldly, calloused laymen, even in a place as remote and uncouth as Iceland. The spiritual and the secular spheres were inextricably joined in an armillary that would not allow one to revolve without affecting the other’s rotation; exalted ideas assumed substance in a social medium, and aspirational ideologies like poor relief and ecclesiastical independence came to rest on flesh-and-blood vagrants and clerics – not all of them necessarily the most laudable, wholesome, deserving characters.85 Much as papal monarchs on the model of Innocent III (and his predecessors and successors) would have liked to think that it was they who were calling the shots for all of Christendom, their dependence on the acquiescence of ecclesiastical subordinates and laymen opened up ample opportunities for humbler lay and ordained Christians in far-flung provinces to hitch their idiosyncratic carts to the Gregorian engine: universal Church reform, in other words, proved an effective rallying bugle, on occasion perhaps even a force multiplier, for a variety of far more local causes.86 In Iceland, Church reform tunnelled its way into fissures in the political landscape, driving some opportunistic bœndr into the arms of a 85

86

The point is starkly illustrated by the incident that sparked the final conflict between Guðmundr and Kolbeinn Tumason in 1208, a dispute over the right to prosecute an acolyte who (among other misdeeds) had recently impregnated a woman. GSA cap. 129 makes no bones about this cleric’s turpitude: “[hann var] osiðuandr at uapna burð. ok klẻðnaðe. hann var ein hendr. þenna man høfðo Austmenn handhøgit at Gasum. þa er Guðmundr en dyre mẻlti eptir hann” [(he was) ramshackle in bearing weapons and in habit. He was one-armed; Norwegians had hacked off his arm at Gásir, that time when Guðmundr dýri sued on his behalf] (p. 153; cf. Guðmundar saga dýra cap. 26, in Sturlunga saga, 1:212). See also GSA cap. 142 for the acolyte’s beheading after the Battle of Hólar in 1209 (p. 165); Miller reviews his tatty career (Bloodtaking and Peacemaking, pp. 1–10; see also idem, Eye for an Eye [Cambridge, 2006], pp. 51–53). See, e.g., Benjamin Arnold, Medieval Germany, 500–1300: A Political Interpretation (Toronto and Buffalo, 1997), pp. 62–66, 99; Eldevik, Episcopal Power and Ecclesiastical Reform, pp. 197–99, 243–46, 254. Cf. also Marc Bloch’s classic “From the Royal Court to the Court of Rome: The Suit of the Serfs of Rosny-Sous-Bois” (orig. 1939), trans. Sylvia L. Thrupp, in Change in Medieval Society: Europe North of the Alps, 1050–1500, ed. Sylvia L. Thrupp (New York, 1964), pp. 3–13. For introducing me to this enduringly entertaining and illuminating little article, I am indebted to Amnon Linder.



Helgastaðir, 1220

133

Gregorian bishop and his vagabond levies, while others, no less opportunistic, made common cause with the power establishment. In the ensuing collision, for all the valor of men like Eyjólfr Kársson and for all the sacrifices of those like Eyjólfr hríðarefni, the newly forged conservative alloy of landholding elites proved hardier than the bid of an apolitical underclass for social emancipation. At Helgastaðir, the uprising of the proletariat was successfully put off. Some foot-soldiers of the revolution fell, others fled as best they might, while the ideologues and propagandists lived to preach and fight another day. Hardly anyone, then or now – certainly beyond Iceland’s shores – took notice. Perhaps we ought to be paying closer attention. The story of that bloody weekend, 29–30 August 1220 at Helgastaðir, may be yet another instance in which the Icelandic sagas precociously reveal processes that, elsehwere, would have to wait for 1381, 1789, 1848, or perhaps even 1989 (and beyond) to acquire visibility.

Appendix Afterward, Bishop Guðmundr goes to Einarsstaðir, and planned to continue to Helgastaðir to consecrate the church there on Saint John’s autumn feast, which he did. Eyjólfr Kársson rode to Múli and suggested to Ívarr that the bishop should consecrate the church there the next day. But Ívarr did not want this and said there would be a fight if the bishop wanted to come over. News now comes to the bishop that troops are converging from all sides. [GSD: That same morning, when he was prepared to perform the service, news of hostilities are brought to him; then some of his men say that he should skip the church consecration, if things unfold more quickly than it (can) be concluded. The bishop then reveals once again what is in his breast, and says: “I shall perform my service, because the time will suffice for this.” It happened as he had said.]87 Guðmundr finishes the church consecration. After that they prepare for defense there in the churchyard, and collected many stones. Afterward they see that the troops rode down to Einarsstaðir. The bishop’s following is then in the churchyard, but the farmer Þorljótr and his brother Sigurðr and their serv87

GSD cap. 55: “Sama morgin, sem hann er búinn til þjónustugerðar, berst honum njósn af úfriðnum, tala þá nökkurir af hans mönnum, at kirkjuvígsla muni um líða, ef þá berr at fljótara en lykt verði. Biskupinn birtir þá enn hvern hann hefir innan brjósts, ok talar: gera mun ek þjónustu mina, þvíat nægjast mun tíminn til þess. Svá var sem hann sagði” (in Biskupa sögur, 2:113). Similarly in GSC cap. 90: “Þann sama morgun er byskup býz til kirkjuvígslu á Helgastöðum kom honum njósn, at flokkar þeira Sighvats sé mjök nálægir. Byskup sagðiz lúka mundu kirkjuvígslu sem hann hafði ætlat ok kveðr verða munu stund til þess, ok svá var” [That same morning, when the bishop prepared for the church consecration at Helgastaðir, news came to him that the troops of Sighvatr and the others were very close by. The bishop said he would finish the church consecration as he had planned and that there would be enough time for it, and so it was].

134

Oren Falk

ants stood apart, and Sigurðr declared: “Look there now, lads, where the troop of the chieftains rides! The locks on the pantries of the men of Reykjadalr click shut at last!” Later they saw another company ride up along Vatnshlíð [Lake-Slope]. The companies arrived at nearly the same time; it was noon. The war-cry was then raised and the attack was launched. Arnórr Tumason advanced from the south with his detachment, but Sighvatr advanced from the home meadow towards the gate, while Ívarr stormed from the north. Bishop Guðmundr was in the church. Intense fighting now broke out with stones and shot. Sturla [GSC: Tumi]88 Sighvatsson got hit by a stone. Then Sighvatr proclaimed: “The bishop’s men spare nothing! Now they strike the boy, Sturla [GSC: Tumi],89 the same as other men! Where are you, Guðmundr G[ísl]sson? Don’t you see that Eyjólfr Kársson is in the churchyard? Don’t you remember the Battle of Melr?”90 At this egging on, Sighvatr’s men pressed the assault hard. One of the bishop’s men, named Gísli, fell before the spears. A little later a man got stabbed with a spear in the eye, [one] of Arnórr’s troop; he was named Hámundr and was the son of Þorvarðr. He lived through the night, but expired the following morning. Then Bishop Guðmundr walked out of the church and to his men. Then they turn very fierce and fight most valiantly, but not one of them better than Jón Ófeigsson [Eyjólfr Kársson’s half-brother]. [GSB: Sighvatr then got a great blow from a stone, and he went down, and Guðmundr klasi who stood near him cried: “Do you fall now, dear Sighvatr?” “I bent down, you filthy scum,” said he, “but I didn’t fall.” Then Klasi began to call out and urged people to flee before the terror that overcame (him), “because the earth is quaking beneath us.” This was then recited: Klasi began to call three times: “Go on, men, and run away – the earth is all a-tremble beneath the warriors – from the encounter with the bishop.”]91 88 89 90

91

GSC cap. 90. GSC cap. 90. Sighvatr’s reference is to an encounter c. 1216 between the family of Guðmundr Gíslsson on one side and Eyjólfr Kársson and his kinsmen on the other, allegedly sparked by various humiliations of Gísl’s family. Guðmundr Gíslsson’s uncle was killed in the fighting. See Íslendinga saga cap. 33 [38] (in Sturlunga saga, 1:262–65). GSB cap. 106: “Sighvatr fekk þá steinshôgg mikit, ok fell hann við, en Guðmundr klasi kallaði, er stóð hjá honum: ‘Felltu nú, Sighvatr sæll?’ ‘Lágr varð ek, fýla veil,’ sagði hann, ‘en eigi fell ek.’ Þá tók Klasi at kalla ok bað menn flýja undan ógn þeiri er yfir var komin, ‘því at jôrðin skelfr undir oss.’ Þá var þetta kveðit: Klasi nam kalla þrysvar: / ‘kosti menn ok renni / – jôrð bifask ôll und fyrðum – / undan byskups fundi’” (= Biskupa sögur, 1:513 n.1, transcribing komi instead of kosti). Similarly in GSC cap. 90: “Í þessari svipan fekk Sighvatr steinshögg mikit svá hann fell við. Þá tók til orða Guðmundr klasi er stóð hjá honum: ‘Felltu nú, Sighvatr sæll?’ kvað hann. ‘Ekki fell ek, fýla veil, en lágt fór ek,’ sagði hann. Þá tók Klasi at kalla ok bað menn flýja undan ógn þeiri er yfir var komin – ‘því at jörðin skelfr öll.’ Þetta var kveðit um: Klasi nam kalla þrisvar: / ‘Kosti menn ok renni / – jörð bifaz öll und fyrðum – / undan



Helgastaðir, 1220

135

And at that, Sighvatr and his men held off for now, and they announced that they wanted rather to win by counsels than by endangering men, and also, they didn’t want to fight so that the bishop’s life would be in danger. Now they position men so that none of the bishop’s men might get away against their will. Afterwards they, the magnates, went south into the field. Arnórr asked Sighvatr: “Doesn’t it seem to you like a hard battle has been fought, kinsman?” – “Hard indeed,” said Sighvatr. Arnórr spoke: “I was sickly over summer, but when word came to me from the men of Reykjadalr that they needed help, all ailments lifted from me, so that I feel in no way ill.” – “That must seem to you a miracle,” said Sighvatr. Arnórr answered: “I call it a circumstance,” he said, “but no miracle.” Afterward they encircled the churchyard and accords were sought, but nothing came of it, and the night wears on. Bishop Guðmundr and his men were in the church, but the others surrounded it. One of the bishop’s men made it to the river and out of the church. He was called Eyjólfr hríðarefni [the Tempestuous], a troublemaker. He was grabbed and beaten and they dragged him back half-dead and killed him later. Early on the Lord’s day in the morning [30 August 1220] they made a battlebarrier and under cover of it they dug up the churchyard [wall]. [Íslendinga saga: Early on the Lord’s day in the morning they made a battle-barrier of beams and carry it to that woodland which stood to the south of the (church) yard.]92 Just as soon as an opening had been made in the churchyard [wall], the bishop’s men gave way and went into the church. They had positioned the battle-barrier at the southern edge of the churchyard. [GSD: Afterwards they make a plan to overpower the Lord Bishop’s men more by trickery than by endangering men. Afterwards they besiege the church as (if it were) a fortress, so that each keeps watch over the other through the night. But early Sunday morning they use superior numbers: some attack then, but some break a great opening in the defenses, so that then everything was given up.]93 But Ísleifr Hallsson had arrived there and had not fought [GSC: against the bishop],94 nor any of his men. He lived then at Þverá in Laxárdalr. Ísleifr invited

92

93

94

byskups fundi’ [At that moment Sighvatr got a great blow from a stone so that he went down. Then Guðmundr klasi who stood near him began to speak: “Do you fall now, dear Sighvatr?” spoke he. “I didn’t fall, you filthy scum, but I bent down,” said he. Then Klasi began to call out and urged people to flee before the terror that overcame (him) – “because the earth is quaking beneath us.” This was recited about (it): Klasi began to call three times: / “Go on, men, and run away / – the earth is all a-tremble beneath the warriors – / from the encounter with the bishop”]. Íslendinga saga cap. 37 [42]: “Dróttinsmorgininn snemma gerðu þeir vígflaka af röftum ok bera hann at lundi þeim, er stóð sunnan at garðinum” (in Sturlunga saga 1:276). It seems curious that the vígflaki should be borne to the woods; emending at to af would yield better sense: “carry it from that woodland.…” GSD cap. 55: “Síðan gera þeir ráð, at vinna herra biskups menn meirr meðr klókskap en mannhættu; setjast síðan um kirkjunna sem einn herkastala, vaka svá hvárir yfit öðrum um nóttina. En árla sunnudags myrgininn njóta þeir liðsmunar, sækja þá sumir, en sumir brjóta stór hlið á verndinni, svá at þá er allt upp gefit” (in Biskupa sögur, 2:113). GSC cap. 90: “móti byskupi.”

136

Oren Falk

the bishop home with him [and] he went away with him. [GSD: A certain good farmer invites Lord Bishop Guðmundr to his (home); he rides away with him the same day.]95 But those who were left behind then looked for quarter. Arnórr gave quarter to Eyjólfr Kársson, and Tumi Sighvatsson to Jón Ófeigsson. Two men were then put to death (drepnir)96 in the hills north of the farmstead. One was called Þorgeirr [Íslendinga saga: Þorgeirr háleygr (the Halogalander)]97 and the other Þórðr Arason.98

95 96

97 98

GSD cap. 55: “Herra Guðmundi biskupi býðr til sín einn góðr bóndi, ríðr hann brött með honum sama dag” (in Biskupa sögur, 2:113–14). For the narrower, juridical sense of the verb drepa (normally simply “to strike, kill”) in the fourteenth century, see Karl von Amira, Die germanischen Todesstrafen: Untersuchungen zur Rechts- und Religionsgeschichte (Munich, 1922), p. 118. Íslendinga saga cap. 37 [42] (in Sturlunga saga, 1:277). GSA capp. 171–72: “Siðan ferr G[uðmundr] byskup aEinars staðe. ok ẻtlaðe þaðan a Helga staðe. at vigia þar kirkiv. Jons messo um haustit. sem hann gørðe. En Eyiolfr Kars s[on] reið iMula. ok bauð Juare at byskup vigðe þar kirk[io] eptir vm d[aginn]. en Iuar uillde þat eigi. ok quaz uige mundo veria ef byskup uillde til koma. Nu kemr byskupi niosn. at flockar dragaz at ollum megin. G[uðmundr] lykr þa kirkiu vigslo. Eptir þat buaz þeir til varnar þar ikirkiu garðinum. ok baro at ser griot mart. Siþan sea þeir at flockarnir riðo ofan at Einars støðum. Er þa sueit byskups ikirkiu garðe. en Þorliotr bonde. ok Sigur[ðr] broðir hans. ok heima menn voro einir ser. Þa mẻlti Sigurðr. Se þer nu sueinar flocc þeira høfðingianna. huar riðr. enda skellr þar nu lass firi buren þeira Reykdẻlanna. Siþan sa þeir annan flocc riða neðan eptir Uaz hlið. Flockarnir quomo nẻr iafn snemma. Þa var non d[ags]. Var þa slegit up herope. ok skipat til at gønngu. Gecc Arnorr Tuma s[on] sunnan at með sina sueit. en Sigh[uatr] gecc at hliðino. ok or tunino. en Jvar sotte at norðan. G[uðmundr] byskup var i kirkiv. ok gengr nu hørð hrið með griote ok með skotum. Sturla Sigh[uatz] s[on] fecc steins høg. þa mẻlti Sigh[uatr]. Engu eira byskups menn. nu berea þeir sueinin Sturlo sem aðra menn. e[ða] hvar ertu Guðmundr Gils s[on]. Ser þu eigi at Eyiolfr Kars s[on] er ikirkiu garðinum. e[ða] man tu eige bardag[ann a] Mel. Við a egian þessa knuðuz Sigh[uatz] menn fast til at gø[n]gu. Fell þa maðr af byskups mønnum. firi spiotum er Gisle het. Litlu siþar fecc maðr lag af spiote i augat af liðe Arn[ors]. Sa het Hamundr. ok var Þorv[arðz] s[on]. Hann lifðe um nottina. en andaðiz um morg[ininn] eptir. Þa gecc G[uðmundr] byskup or kirk[io] ok til manna sinna. Ero þeir þa allakafir ok beriaz alldrengil[iga]. ok enge einn hverr betr enn Jon Ofeigs s[on]. Ok við þetta leggia þeir Sigh[uatr] fra at sinne. ok letuz helldr uilea uinna með raðum. en mannhaska. Villdo þeir ok eige beriazt sua at byskupi vẻre at þui lifs haske. Setea þeir nu men til at enge af byskups mønnum. kẻmiz i brøt an uilea þeira. Siþan ganga þeir suðr a uøllin firi menninir. ok spurðe A[rnorr] Sigh[uat]. Þicke þer eigi hørð hrið gengit hafa magr. Hørð vist. s[agðe] Sigh[uatr]. [Arnorr mẻlte.] J sumar var mer kuellinga samt. en er mer komo orþ Reycdẻla at þeir þyrfte liðs. hof af mer allar vamur, sua at ek kenne mer hverge illt. Þat man þer þickja iartein. s[agðe] Sigh[uatr]. A[rnorr] svar[aðe]. Þat kalla ek atburð s[agðe] hann enn eige iartein. Siþan seteaz þeir um kirk[io] garþin. ok var leitat um sẻttir. ok var þo sem ecke vẻre. ok liðr af nottin. Var G[uðmundr] byskup ok hans menn ikirkiu. en hinir satu um hverfis. Einn byskups maðr komz yfir ána. oc or kirk[io]. Sa het Eyiolfr hriþar efne ospakr maðr. Var hann dreginn ok barðr ok drogu þeir hann heim halfdauðan ok drapu hann siðan. Drottins morginin snemma gerðo þeir vigflaka. og grofu þar undir kirk[iu] gardinn. Uar þat miok iafnskiott ad hlid uard ȁ kirk[iu] gardin[um]. og byskups menn gafust up og foro j kirk[iu]. Þeir settu uigflakann sunnann undir kirk[iu] gard[inn]. Enn Jsleifur Halls s[on] uar þar komin. og hafdi eigi barizt. ok eingin hans manna. Hann bio þa at Þuerr ȁ. j Laxar d[al]. Baud Jsleifur byskupi heim með sier. Med honum for hann j burt. enn þeir er eptir uoro geingu þa til grida. Gaf Ar[nor] grid Eyolfi Kars s[syni].



Helgastaðir, 1220

137

[GSB: but when the bishop had got away, Arnórr and Tumi (Sighvatsson)’s men took six men out of the church and hanged them like thieves; two of Bishop Guðmundr’s men were strung up, one on either side of the gate which faced towards Glaumbœr, but four were driven up into Þjófahólar (Thieves’ Hills). There was a man named Þorsteinn and he was called hryggr (Backbone). He was shot with a spear in the thigh inside the church, and he was a certain pauper (accompanying) the bishop, but lived long after. Many atrocities were committed there against a holy, newly consecrated church, and one may still see it on her nowadays, how poorly she – O immaculate and undeserving! – fared in that battle. And people later said this: that it looked as if the church’s wounds were knitting wherever she had received injury from weapons or stones.]99 [GSC: Four men fell of the bishop’s men, but of Arnórr’s one. Þorsteinn who was called hryggr, a certain pauper among the bishop’s (following), was shot in the thigh inside the church, and he lived long after. They committed many martial deeds in a holy, newly consecrated church, and one may still see it nowadays, how poorly she – O immaculate and undeserving! – fared in that encounter. And people later said this: that it seemed as if the church’s wounds were knitting, those which she had received from weapons and stones during the battle.]100 [GSD: This was a shameful action, a newly consecrated church so wounded and beaten with stones, so that the entire age saw it on her, more than a hundred years (later), when this account was compiled.… Four men, of the Lord ­Bishop’s (following), fell to weapons; but at that time when he had gone away, Arnórr and Tumi’s men take 7 men who belong to him and string them up on the gallows.]101

99

100

101

enn Tume Sigh[uats] s[on] Jone Ofeigs s[yni]. Tueir menn uoru þar drepnir nordr a hơlunum fra bænum. Het annar Þorgeir. enn annar Þordur Ara s[on]” (pp. 185–88). GSB cap. 106: “En er byskup var brotu tóku þeir Arnórr ok Tumi sex menn ór kirkju ok hengðu sem þjófa. Menn Guðmundar byskups tveir váru festir upp sínum megin garðshliðs hvárr, því er veit til Glaumbjár, en fjórir váru fœrðir upp í Þjófahóla. Þorsteinn hét maðr er kallaðr var hryggr. Hann var skotinn með spjóti í lærit í kirkju ok var vesalingr einn með byskupi ok lifði lengi síðan. Margt illvirki var þar gôrt á heilagri kirkju nývígðri, ok enn má þat sjá á henni í dag hversu hôrðu hon mœtti í þeim bardaga, saklaus ok ómakleg, ok svá hafa menn sagt síðan at svá sýnisk sem sár hafi gróit á kirkjunni þar sem hon fekk áverka af vápnum ok grjóti” (= Biskupa sögur, 1:514 n. 1). GSC cap. 90: “Af byskupsmönnum fellu fjórir menn en einn af Arnóri. Þorsteinn er kallaðr var hryggr, vesalingr byskups, var skotinn í lærit inn í kirkjunni, ok lifði hann lengi síðan. Margt hervirki gjörðu þeir í heilagri kirkju nývígðri, ok enn má sjá þat í dag hversu hörðu hon mætti á þessum fundi, saklaus ok ómaklig, ok svá hafa menn sagt síðan at því sýniz líkt sem sárin hafi gróit á kirkjunni, þau sem hon fekk af vápnum ok grjóti í bardaganum.” GSD cap. 55: “Hörmulig var sú framferð, kirkja nývígð er svá særð ok barin grjóti, at alla æfi sá þat á henni, meirr en hundrat ára, þá er þessi frásögn var samsett.… Fjórir menn fèllu vápnsóttir af herra biskupi; en þann tíma, sem hann var á bröttu, tóku þeir Arnórr ok Tumi vij menn, er honum til heyra, ok festa upp á galga” (in Biskupa sögur, 2:113–14).

4 Por la guarda de la mar: Castile and the Struggle for the Sea in the Thirteenth and Fourteenth Centuries Nicolás Agrait

The naval prominence of Castile in the late medieval and early modern period has been amply researched. However, the rise of the Castilian navies and their fundamental role in the thirteenth and fourteenth-century Castilian expansion have not been addressed properly. One of the principal reasons has been the relatively small amount of coverage of naval affairs in the chronicles, and the dearth of documentation in the archives in comparison to ground warfare and forces.1 After all, the Castilian elite, which takes up most of the attention in the sources, did not routinely train in maritime or naval matters as part of its upbringing. Maritime and naval historians, perhaps, have somewhat exacerbated the situation by appearing to treat naval issues separately from those of fighting on land. Such a demarcation is untenable in Castile. The growth of its navies coincided with, and was important, if not fundamental, in its expansion and consolidation, which was accompanied by an increasing awareness and embrace of the importance of seafaring and the conquest and defense of the coasts.2 Castile struggled against not only the Muslims in the Iberian Peninsula and North Africa, but also the kingdoms of Aragon and Portugal and the Italian city-states, especially Genoa, which sought to establish and protect commercial interests in the Mediterranean and the Maghreb. Since all of these conflicts invariably included a naval component, more analysis is necessary to fully understand how, when, and why the Reconquest unfolded as it did. Just as the Muslim territorial domination gave way to Christian southern expansion, so too the coastlines and sea lanes, in the Atlantic, the Strait of Gibraltar, and the Mediterranean, were gradually wrested away from total Islamic domination during the thirteenth and fourteenth centuries. This was accomplished by following a familiar strategy of 1

2

Eduardo Aznar Vallejo, “La guerra naval en Castilla durante la Baja Edad Media,” En la España Medieval 32 (2009), 168. The problem is not limited to Castile as Chrisopher Allmand pointed out in The Hundred Years War: England and France at War c. 1300–c. 1450 (Cambridge, 1988), p. 87. Michael Anthony Furtado, “Islands of Castile: Artistic, Literary, and Legal Perceptions of the Sea in Castile-León, 1248–1450” (unpublished Ph.D. dissertation, Ann Arbor, MI, 2011), pp. 293, 295–97 and 300.

140

Nicolás Agrait

the capture and defense of important strongpoints, and only through difficult and slow progress.3 As Castile and Aragon emerged as the two top naval powers, their uneasy balance and later union in the fifteenth century signaled the end of any serious Muslim naval aspirations or capability, whether in Granada or North Africa.4 Navies played only a small role in the final defeat and annexation of Granada by the Catholic kings in 1492. Even so, it must be determined whether medieval Castilians themselves thought that navies and warfare at sea were important. First, what reasons compelled them to create and sustain navies and their accompanying infrastructure in the thirteenth and fourteenth centuries? Connected to this, second, how much material and financial resources were they able and willing to spend on their naval enterprises? Third, how did they utilize their ships in warfare? Addressing these issues, in turn, will show that the Castilian policy makers indeed assigned great importance to naval warfare and considered it an integral part of the military apparatus. Therefore, they consciously pursued policies to achieve dominance over the surrounding coasts and seas. The first Castilian naval centers arose in the northern sections of the Iberian Peninsula. Some of the early loci of Castilian maritime activity were in the Asturian and Galician coastal regions, which had a history of enduring constant raids from the sea, especially by Muslim pirates coming from the south in the eleventh and twelfth centuries. To combat the threat, many of these towns put in place ad hoc measures such as the construction of defensive towers and fortified bridges in the estuaries. Gelmirez, archbishop of Compostela, implemented one of the first significant naval programs, in 1120 hiring a Genoese shipbuilder to construct two galleys for defense against Islamic privateering.5 The effort did not long survive him and, as the Muslim threat waned, this initiative did not lead to the creation of any long-term naval institutions. Trade and other maritime commercial activities were left in private hands, while defense of the sea was still a local affair. The most successful of the ports focused on the activities developing to the northeast, in the English Channel, the Bay of Biscay, and beyond.6 The region of Vizcaya, favored by its geographical location and its notable forests and deep deposits of iron ore,7 withstood the incursions of the Norsemen 3 4

5 6

7

Manuel Flores Díaz, “Las ciudades portuarias, objetivo en la expansión política y económica de los Estados medievales ibéricos,” Revista de Historia Naval 91 (2005), 10–14. Hugo O’Donnell y Duque de Estrada and José María Blanco Núñez, “Las marinas medievales y la guerra en el mar: medios, técnicas, acciones,” in Historia Militar de España. II. Edad Media. ed. Miguel Ángel Ladero Quesada (Madrid, 2010), p. 428. Lawrence Mott, “Iberian Naval Power, 1000–1650,” in War at Sea in the Middle Ages and Renaissance, ed. John B. Hattendorf and Richard W. Unger (Woodbridge, 2002), p. 105. Elisa María Ferreira Priegue, “Castilla: la génesis de una potencia marítima en Occidente,” in Itinerarios medievales e identidad hispánica: XXVII Semana de Estudios Medievales, Estalla, 17 a 21 de julio de 2000 (Pamplona, 2001), pp. 25–27. Ana María Rivera Medina, “Paisaje naval, construcción y agentes sociales desde el medioevo a la modernidad,” Istas Memorias: Revista de Estudios Marítimos del País Vasco 2 (1998), 49.



Castile and the Struggle for the Sea

141

and then integrated into the northern sea-trade networks, developing significant fishing, whaling, and maritime exchange industries.8 The first of these towns to receive royal privileges was Castrourdiales,9 followed by San Sebastián,10 Laredo,11 Guetaria,12 and San Vicente de la Barquera.13 This allowed them to further cement their presence in Aquitaine, Flanders, and England.14 In 1296, these cities formed the Hermandad de las villas de la marina de Castilla (Confraternity of the Towns of the Navy of Castile), or Hermandad de las Marismas.15 This effective union provided for the defense of traditional privileges granted since the twelfth century, including exemptions from military service and taxation, and the members’ right to choose their own lords. It also exempted them from paying tolls to the bishop of Burgos, freed them from the jurisdiction of the Castilian almirante mayor (high admiral) in most respects, and enabled them to monopolize the export of iron, and to defend common interests against Gascon, English, and Portuguese shipping.16 Even though Alfonso XI (1312–50) tried to suppress it, the Hermandad de las Marismas continued, and sustained these towns’ commercial and strategic importance to the point that, during his fratricidal civil war with his half-brother Enrique of Trastámara, Pedro I (1350–69) promised the Black Prince overlordship over the Viscayan region in exchange for the Englishman’s aid against Enrique.17 8

9

10 11

12 13

14

15

16 17

Xabier Alberdi Lonbide and Álvaro Argón Ruano, “La construcción naval en el País Vasco durante la Edad Media,” Itsas Memoria: Revista de Estudios Marítimos del País Vasco 2 (1998), 17–18. “Fuero [royal charter] granted by Alfonso VIII to Castro Urdiales (10 March 1163)” in Rogelio Pérez Bustamante, Historia de Castro Urdiales desde sus orígenes hasta la época moderna (Santander, 1988), doc. 1, p. 175. “Fuero granted by Sancho VI to San Sebastián (c. 1180)” in Cesáreo Fernández Duro, La marina de Castilla (Madrid, 1894), doc. 35, p. 459. “Alfonso VIII grants the fuero given to Castro Urdiales to Laredo (Belorado, 25 January, 1200)” in Documentación medieval de la Villa de Laredo, 1200–1500, ed. Virginia M. Cuñat Ciscar (Santander, 1998), doc. 1, p. 55. “Royal Privilege from Alfonso VIII to Guetaria (1 September 1209)” in Real Academia de la Historia (Madrid), Vargas Ponce Collection, vol. 42, fols. 124–25. “Fernando III confirms a privilege to San Vicente de la Barquera exempting them from paying portage fees across the kingdom (10 July 1241)” in Colección documental de Pedro I de Castilla, 1350–1369, 4 vols., ed. Luis Vicente Días Martín (Salamanca,1997–99), doc. 406, 2:138–39. “Maritime treaty between Castile and Bayonne (1 December 1293),” “Treaty between Bayonne and Laredo, Castro Urdiales and Santander (19 July 1311),” in Fernández Duro, La marina de Castilla, docs. 6 and 9, pp. 388–89 and 398–404. “Alliance between the town councils of Santander, Laredo, Castrourdiales, Vitoria, Bermeo, Guetria, San Sebastián and Fuenterrabia (Castrourdiales, 4 May 1296)” in Antonio Benavides, Memorias de Don Fernando IV, 2 vols. (Madrid, 1860), doc. 57, 2:81–85. Luis Suárez Fernández, “Evolución histórica de las Hermandades castellanas,” Cuadernos de Historia de España 16 (1951), 22–23. Pedro López de Ayala (d.1407), Crónica del rey Don Pedro in Crónicas de los reyes de Castilla, ed. Cayetano Rosell, 3 vols. (Madrid, 1875–78), year 17, chap. 24, 1:549; “Letter from Pedro I to the Prince of Wales with the grant of Bermeo, Bilbao, Lequitio, Castrourdiales and other locales (23 September 1366)” in Colección documental de Pedro I, doc. 1343, 4:243.

142

Nicolás Agrait

It is important to note that the growth of these naval centers was accomplished without much direct input from the Castilian crown. As their private and local maritime and naval resources grew, the monarchy was happy to confirm these towns’ liberties and benefit from the further increase in trade, which provided the royal coffers with much welcome income from indirect taxation. These cities’ naval prowess freed the crown from spending resources to protect the northern coast, allowing it to concentrate on its struggle with the Muslims in the south. Yet the Cantabrian and Vizcayan ports understood that their freedom was not limitless, since the Castilian monarchs often called upon them to assist in royal campaigns. In fact, it was these cities’ involvement in the Reconquest that provided the initial stimulus for the formation of the Castilian navy in the south. In 1246, the crown prince of Castile, the future Alfonso X (1252–84), found that, unlike the rest of Murcia, he could not conquer the port of Cartagena from land. He entrusted his vassal Roy García de Santander with the task of assembling and arming a fleet, sailing it through the Strait of Gibraltar, and then blockading the stubborn Murcian holdout. With its sea supply route cut off, Cartagena surrendered and became the first and only Castilian port directly on the Mediterranean.18 Two years later the Vizcayan and Galician towns were again called upon and provided a navy of thirteen galleys and round ships to assist Fernando III (1217–52) in his siege and capture of Seville. The Christian ships managed to break through the Muslim galleys guarding the port, capturing three, impeding any supplies from reaching the city through the Guadalquivir river, and opening up an access route into the city itself.19 With such important ports in its hands, Castile could now challenge the Muslim dominance on the Atlantic, Strait of Gibraltar and Mediterranean shores. Yet, understanding the need to develop the naval capacity necessary to continue its expansion and actually building it were two different things. Even though Castile had considerable catching up to do in establishing a naval presence in the southern Iberian Peninsula, it did not have to start from scratch, as it could use the long naval traditions and institutions of Muslims who had come under Castilian rule as a foundation. Prior to its collapse, the Cordoban emirate had built an impressive naval and maritime web in the Mediterranean and the Atlantic ocean. From west to east, the principal Atlantic Islamic ports included Lisbon, Alcaçer do Sal, Silves, Santa María del Algarve, Saltes, Seville and Cádiz; in the Strait of Gibraltar, Tarifa, Algeciras and Gibraltar; on the north of Africa, Tangier and Ceuta; and in the Mediterranean, Málaga, Almuñécar, Almería, Denia, Valencia, Tortosa, and the ports in the Balearics. The two principal naval bases were Alcaçer do Sal in the Atlantic and Almería in the Mediterranean. The Almoravid and Almohad regimes sustained a great deal of this network, respectively using Tarifa and Gibraltar as their principal contact point 18 19

O’Donnell and Blanco Núñez, “Las marinas medievales,” p. 432. Primera Crónica General de España, 2 vols., ed. Ramón Menéndez Pidal (Madrid, 1977), chaps. 1075, 1078–80, 1089, 1093–97, 1108, 1119–20, 2:748–50, 754–57, 760–61, 765–66.



Castile and the Struggle for the Sea

143

between al-Andalus and Africa, but could never equal the Umayyads.20 The collapse of the Almohad regime after 1212 and the disintegration of the Muslim polity into taifas or petty kingdoms, however, made all these points vulnerable and ultimately led to the dismantling of Muslim naval power.21 Yet the true birth of the Castilian navy as a royal institution came after the conquest of Seville and the acquisition through treaty of all the territories of the lower Guadalquivir valley, including Lebrija, Arcos de la Frontera, Jerez, Sanlúcar, and even Cádiz, which was not fully reconstructed and repopulated until 1260.22 Fernando III also commenced rebuilding the Almohad shipyards, which lay between the city’s southwestern walls and the Guadalquivir river. This project was continued during the reign of Alfonso X, eventually producing the massive edifice known as El Arenal. It consisted of a rectangular structure 182 meters long, facing the river, and divided into sixteen, eventually seventeen, naves of various sizes. These individual naves, separated by the robust pillars supporting the arches above them, were roughly 12.5 meters in height and easily accommodated most ships during construction or repair. El Arenal also served as a supply depot.23 This facility remained the only large dockyard under the auspices of the Castilian crown until 1372, when Enrique of Trastámara (1369– 79) established a new northern royal shipyard at Santander.24 Since Alfonso X intended the Sevillian shipyards to provide a constant supply of armed galleys, he assembled a group of shipwrights, patrones and other specialists to be stationed there. In return, he granted them lands in Huévar. He also permitted logging in the nearby royal forests, the cut logs to be floated down the Guadalquivir river toward the city. The ships were then built at the shipyard. Each galley’s individual captain or comitre,25 in return for a salary and a share of any booty captured during naval expeditions, promised

20 21

22

23 24 25

Flores Díaz, “Las ciudades portuarias,” pp. 9–12. Portugal conducted an impressive series of conquests including Alcaçer do Sal (1217), along with Faro (1249) and Silves (1250), placing it in a strategically favorable position on the Iberian Atlantic coast and in direct conflict with Castile. Aragon, for its part, looked to cement its position in the Mediterranean, already significant after its acquisition of Tortosa (1148), by conquering Mallorca (1229), Menorca (1231), Peñíscola (1233), Castellón (1233), Valencia de Ibiza (1235), Mértola (1238), and Denia (1244). This, in turn, exacerbated the border conflicts with Castile and put Aragon in direct competition with Italian city-states, especially Genoa, which had engaged in a progressive economic penetration of the area. (Flores Díaz, “Las ciudades portuarias,” pp. 13–14). Concepción Cereijo Martínez, “La política marina de Alfonso X: la toma de Salé en la crónica de Alfonso X y en las fuentes musulmanas,” Revista de historia naval 96 (2007), 37–38 and 44. Florencio Pérez-Embid, “La marina real castellana en el siglo XIII,” Anuario de Estudios Medievales 6 (1969), 159; Ferreira Priegue, “Castilla,” 34. José Luis Casado Soto, “El Cantábrico y las galeras hispanas de la Edad Media a la Moderna,” Itsas Memoria: Revista de Estudios Marítimos del País Vasco 4 (2003), 541. The cómitres were the individual captains of each galley, and were in charge of all the aspects of the ship they commanded. In Alfonso X, Las Siete Partidas, ed. Real Academia de la Historia (Madrid, 1807), Partida 2, Title 24, Law 4, 2:260. (Hereafter, Partidas, Number, Title, Law).

144

Nicolás Agrait

to outfit and maintain the vessel and to completely refit it every nine years. The crown was responsible for funding the construction of a new galley, should one be lost during the course of a military operation.26 The largest group of these galley captains was composed of Genoese merchants and shipmasters, who had already established an influential community in Seville by the middle of the thirteenth century. With the infrastructure in place to provide for a steady stream of ships, Alfonso expanded Castilian naval influence by refurbishing and repopulating new ports from Seville to the Atlantic ocean. The route from Seville to the open sea headed west on the Guadalquivir river with Sanlúcar de Barrameda, Puerto de Santa María, and Cádiz as stops along the way.27 As the Muslim territory in the Iberian Peninsula shrank, the struggle moved from the Atlantic to control over the Strait of Gibraltar and the northwestern African coast.28 The most important Muslimcontrolled ports on the north of the Strait were Tarifa, Algeciras, and Gibraltar; their counterparts in the south were Tangier and Ceuta. The three Iberian ports were alternatively under the overlordship of Granada or the Marinid emirate of Morocco. Alfonso understood the Strait’s strategic importance as he moved ahead with two operations, a raid on the Marinid port of Sale in 1260 and later a failed expedition to capture the fortified port of Algeciras in 1279.29 His heirs continued the struggle in spite of stiff Granadan and Moroccan resistance. Sancho IV (1284–95) annexed the crucially important city of Tarifa, which was positioned with direct access to both the Atlantic and the Strait of Gibraltar, in 1292.30 After a pause due to his royal minority, Fernando IV (1295–1312) continued the onslaught, attempting to capture Algeciras in 1309,31 but ending up instead with the strategic, though isolated, prize of Gibraltar, which remained in Christian hands until 1333. His son Alfonso XI, after having overcome his own lengthy royal minority, border wars with Navarre and Portugal, and the open revolt of Castile’s aristocracy, also renewed the Castilian offensive in the Strait, relieving Tarifa from a Marinid assault and later conquering Algeciras in 1344 after a lengthy siege.32 His untimely death prevented him from recapturing

26 27

28

29 30 31 32

Cereijo Martínez, “La política marina de Alfonso X,” p. 39. Victor Muñoz Gómez, “Para el conocimiento de la costa de la Andalucía atlántica (siglos XIV– XVI): descripciones, relaciones y documentación náutica,” Historia. Instituciones. Documentos 40 (2013), 184. This has been termed by historians as the Batalla del Estrecho. The best general treatment of this work is Miguel Ángel Manzano Rodríguez, La intervención de los Benimerines en la Península Ibérica (Madrid, 1992). Crónica de Alfonso X in Crónicas de los reyes de Castilla, ed. Cayetano Rosell 3 vols. (Madrid, 1875–78), chaps. 19, 70, 72–73, 1:13–15, 54–57. Crónica del rey don Sancho el Bravo in Crónicas de los reyes de Castilla, 3 vols., ed. Cayetano Rosell (Madrid, 1875–78), chap. 9, 1:86–87. Crónica del rey don Fernando IV in Crónicas de los reyes de Castilla, ed. Cayetano Rosell 3 vols. (Madrid, 1875–78), chap. 17, 1:163–64. Crónica del rey don Alfonso, el Onceno in Crónicas de los reyes de Castilla, ed. Cayetano Rosell 3 vols. (Madrid, 1875–78), chap. 336, 1:388–90.



Castile and the Struggle for the Sea

145

Gibraltar in 1350, but even so Castile was then in a dominant position in the Strait of Gibraltar. However, it could not claim mastery over the region, since its naval power was directly challenged by Aragon with its significant fleets, Granada and Morocco (certainly in decline but not to be discounted), and Portugal, which also moved to assert some claims over the Atlantic north African coast. In addition, the commercial rivalries involving Genoese, Catalan, Vizcayan, and even English merchants ensured that the area remained in a state of frequent conflict.33 The conflict between Castile and Aragon, called the Guerra de los dos Pedros (War of the Two Peters), involved many festering disputes between the two kingdoms, including the nasty sticking point of Pedro IV’s grant of asylum to Enrique de Trastámara, Alfonso XI’s illegitimate son and a pretender to the Castilian throne. Yet it was an incident of piracy in 1356 in Seville that sparked open warfare. A Catalan fleet of ten ships captured two Genoese fishing vessels close to Sanlúcar de Barrameda. Pedro I threatened to retaliate against the Catalan merchant community in Seville if the ships were not returned. The Catalan captain ignored the threat, sold off the ships, and left the area. The Castilian monarch made good on his threat, confiscating the goods of the Catalans in Seville and selling them off for profit. After this response, there was hardly any way for a peaceful solution and both sides prepared for war. The original Castilian fleet was composed of eighteen galleys – twelve Castilian and six Genoese. These ships, however, were lost in a storm off Guardamar. In response to the naval disaster, Pedro had seven galleys and six naos built in 1358 and outfitted in all haste, and looked for assistance outside Castile. By the next year, the combined Castilian-Granadan-Portuguese armada boasted forty-one galleys, three galliots, eighty naos, and four small support vessels. This fleet, with the capacity to match Aragonese sea power, inflicted significant damage, especially in the Balearic Islands, but was unable to take the ports of Ibiza or Barcelona. As the war stalemated and became much more costly for both sides, Pedro retired his fleets in 1359, although he continued to organize naval operations against Aragon, having to stop only when the war against this half-brother turned decisively against him.34 Even after Enrique of Trastámara gained the Castilian throne in 1369, relations between his kingdom and Aragon remained hostile. Peace was fully restored only with the signing of the Treaty of Almazán in 1375.35 Thereafter, the alliance of the two principal Christian Iberian powers in the Strait of Gibraltar spelled the end for any Granadan naval 33

34

35

Manuel González Jiménez, “Genoveses en Sevilla (siglos XIII–XV),” in Presencia italiana en Andalucía, siglos XIV–XVII: actas del I Coloquio Hispano-Italiano, ed. Bibiano Torres Ramírez and José Hernández Palomo (Sevilla, 1985), pp. 123–27; Pérez-Embid, “Navigation et commerce dans le port de Seville au Bas Moyen Âge,” Le Moyen Age 2 (1969), 284–85. Ayala, Crónica del rey Don Pedro, year 7, chap. 7; year 9, chap. 9; year 10, chap. 10, 12–17, 19–20; year 11, chap. 13; year 15, chap. 1–4; year 16, chap. 1, 1:473–74, 486, 494–98, 505–6, 531–32 and 534–35. Ángel Luis Molina Molina, “Proyección mediterránea del reino de Murcia en la Edad Media,” Miscelánea Medieval Murciana 17(1992), 70.

146

Nicolás Agrait

aspirations. Castilian economic and military pressure reduced Granada’s navy to a mere shadow of its former self, constituting little more than a coast guard by the fifteenth century.36 With the new dynasty on the throne, Castilian foreign policy also took a sharp turn toward France, away from Pedro I’s alignment with England. At stake here was which side would enjoy the favor of the Basque and Cantabrian fleets. In 1372, a Vizcayan fleet defeated an English fleet blockading La Rochelle, thoroughly destroying it and capturing its commander.37 On the heels of this victory, Enrique continued to encourage naval activity, setting up the royal shipyards in Santander to provide a constant supply of galleys to patrol the English Channel.38 The Castilian fleets continued their actions with naval operations in 1376 in the coast of Brittany and a series of raids in 1375–77 on the English coast.39 While the Vizcayan commercial and naval fleets remained a force throughout the rest of the medieval period, the cessation of hostilities between England and France, due to the Truce of Leuningham of 1389, also brought a tense peace between the Castilian and English fleets. This period also was marked by a series of conflicts, alternating with some interludes of cooperation, between Castile and Portugal that were the result of border and dynastic disputes, in addition to the continuing the struggle for control of the Algarve region with its easy access to the Atlantic ocean and Strait of Gibraltar. By the end of the fourteenth century, Castile had emerged victorious, at least in naval and maritime terms. In 1337, the Castilian navy achieved victory in the battle of Cape St. Vincent.40 The rivalry between the two kingdoms again heated up after 1369, when Castile turned toward France and Portugal allied itself with England. That year King Fernando (1367–83) sent a fleet that blockaded Seville for more than a year. This, however, took such a devastating toll on the Portuguese fleet that when it attempted to exit the Guadalquivir and into the Atlantic, a Castilian contingent intercepted it. Because of its weakened condition, all the Portuguese fleet could do was leave the area – but it did not escape without losing several ships.41 Nine years later, the Portuguese monarch again initiated hostilities and specifically gathered a large fleet to attack the Castilian ports. This disorganized fleet, however, was thoroughly

36 37 38 39 40

41

Manuel Flores Díaz, “Fases del poder naval en la Edad Media Hispana,” Revista de Historia Naval 77 (2002), 18. Pedro López de Ayala, Crónica del rey Don Enrique Segundo de Castilla in Crónicas de los reyes de Castilla, ed. Cayetano Rosell 3 vols. (Madrid, 1875–78), year 6, chap. 10, 2:91. Jesús Ángel Solórzano Telechea, “Santander, puerto atlántico medieval,” in Santander: Puerto, Historia, Territorio (Bilbao, 2011), p.113; Casado Soto, “El Cantábrico,” p. 541. Mott, “Iberian Naval Power,” p. 109. Crónica do rei Afonso IV in Crónicas des sete primeiros reis de Portugal, ed. C. de Silva Torouca, 3 vols. (Lisboa, 1952–53), chap. 39, 2:264–65; Crónica del rey don Alfonso, chap. 181, 1:289–90. Ayala, Crónica del rey Don Enrique, year 5, chaps. 4 and 5, 2:82–83.



Castile and the Struggle for the Sea

147

routed by a Castilian flotilla near Saltes.42 Three years later, it was Castile which took the initiative, sailing its vessels up the Tajo river and besieging Lisbon. Even though the Portuguese were able to drive the Castilians away,43 Portugal’s naval capacity had been so weakened that it would not be able to seriously challenge Castile again until the next century. Even then, it had to do so by sailing in the Atlantic ocean and toward the western African coast, since, for all intents and purposes, it had been shut out of the Strait of Gibraltar and the Mediterranean. From the mid-thirteenth century onward, Castile’s navies were characterized by their inclusion of two different seafaring traditions: northern and southern. The colder and rougher waters and weather of the Bay of Biscay, Atlantic ocean and Baltic sea assured that Castille’s western ports adhered to a northern maritime tradition. Their shipwrights used a clinker construction method, in which nails securing the planks were driven in from the outside and hammered flat on the inside. The actual frame was placed later in the building process.44 This hull-first shipbuilding technique was more expensive, due to its greater use of iron, but it turned out the sturdier ships that were necessary to cope with the difficult seafaring conditions of the northern seas. Their southern counterparts, influenced by the Mediterranean seafaring tradition, utilized a frame-first or skeleton construction in which the keel and the framework of a vessel were built first and the planks were fitted later.45 The end result was the relatively faster production of lighter ships that were ideal for the smoother and much more predictable seagoing conditions in the Mediterranean sea.46 The Castilian fleet’s vessels, regardless of their origin, were divided into those propelled by oars and those moving by sail. From the twelfth through the fourteenth centuries, the principal oared vessel was the galley.47 This ship evolved from earlier ones such as the Roman liburna or Byzantine dromon and averaged around forty to fifty meters in length and five to six meters at the widest point, with a very low freeboard.48 There were three types of galleys: the light galley, which had twenty-four banks of oars with space for three oarsmen at each bank; the bastard galley, which had twenty-six banks; and the great galley, which had twenty-eight banks. All could also be outfitted with masts 42 43 44

45 46 47 48

Idem., Crónica del Rey Don Juan Primero de Castilla y León in Crónicas de los reyes de Castilla, ed. Cayetano Rosell, 3 vols. (Madrid, 1875–78), year 3, chap. 4, 2:76. Ibid., year 6, chap. 7–8, 2: 90–91. Ian Friel, “Winds of Change? Ships and the Hundred Years War,” in Arms, Armies and Fortifications in the Hundred Years War, ed. Anne Curry and Michael Hughes (Woodbridge, 1994), p. 183. Arcadi García I Sanz and Núria Coll I Julia, Galeres mercants catalanes dels segles XIV i XV (Barcelona, 1994), pp. 58–64. Felipe Fernández Armesto, “Naval Warfare after the Viking Age,” in Medieval Warfare. A History, ed. Maurice Keen (Oxford, 1999), pp. 231 and 234. Partidas 2, 24, 7, 2: 263–64. Fernando Bordejé Morencos, “La Edad Media. Los años obscuros del poder naval. Primera Parte,” Revista de Historia Naval 40 (1993a), 90–91.

148

Nicolás Agrait

and rigging.49 Assuming two or three men per bank would result in a corps of oarsmen of either 96 or144 for the light galleys, 104 or 156 for the bastards, and 112 or 168 for the largest class. All were galleys alla sensile, each rower pulling on a single oar, with the oars varying in size, the shortest being the one closest to the gunnel.50 When prepared for war, a light galley was fitted with an aftercastle in its stern to accommodate soldiers.51 The great galleys were outfitted with larger fore- and aftercastles and even topcastles, which were small platforms and mastheads used for observation, and also to throw diverse projectiles onto the decks of enemy ships. These ad hoc structures gave the galleys an advantage over lower opposing vessels, and lessened the advantage of enemy ships with larger freeboards.52 Other types of Castilian oared vessels included the galeota or galliot, which at the time was about half the length of a galley, with no mast and less space for oars,53 the tarida, which was generally smaller than the usual galleys and specifically used for the transport of horses, so that it was built with doors at the stern for disembarking them,54 and saetias and zabrias, smaller and lighter vessels with less space for oars and single masts.55 All these served as support ships for the galleys.56 Oared vessels were, of course, not limited to southern Castile, as they were also present in the northern ports. In the twelfth century, the ships utilized in the northern ports were descended from the Scandinavian drakkar.57 Even so, the galley design made its way north, and by the thirteenth century the northern shipyards were building such ships, adapting them for the rougher conditions of the Atlantic.58 While some Mediterranean fleets were composed exclusively of galleys, most would have some sailing ships as well.59 The two principal vessels of this type present in the Castilian navies, as attested by the sources, were the cocas

49

50

51

52 53 54

55 56 57 58 59

José Cervera Pery, El poder naval en los reinos hispánicos (Madrid, 1992), p. 248; John E. Dotson, “Merchant and Naval Influences on Galley Design at Venice and Genoa in the Fourteenth Century” in New Aspects of Naval History, ed. C. L. Symonds (Annapolis, 1981), p. 23. Federico Foerster Laures, “La táctica de combate de las flotas catalano-aragonesas del siglo XIII, según la describe Ramón Muntaner (1265–1315),” Revista de Historia Naval 16 (1987), 24. Antonio Torremocha Silva, Algeciras entre la Cristiandad y el Islam: estudio sobre el cerco y la conquista de Algeciras por el rey Alfonso XI de Castilla, así como de la ciudad y sus términos hasta el final de la Edad Media (Algeciras, 1994), p. 173. Friel, “Winds of Change?” p. 185. Bordejé Morencos, “La Edad Media,” p. 91. Foerster Laures, “The Warships of the Kings of Aragón and Their Fighting Tactics during the 13th and 14th Centuries AD,” The International Journal of Nautical Archeology and Underwater Exploration 16:1 (1987), 25. Bordejé Morencos, “La Edad Media,” pp. 91–92. Partidas 2, 24, 7, 2:264. Bordejé Morencos, “La Edad Media,” p. 89. Casado Soto, “El Cantábrico,” p. 537. The distinction drawn here is for ships strictly built to be propelled by wind power. All galleys were built with masts in order to take advantage of propitious wind conditions.



Castile and the Struggle for the Sea

149

(cogs) and naos (roughly the same as French or English “nefs”).60 Used interchangeably, these terms described relatively short vessels with high freeboards that had one or more masts. As they evolved, they gave way to larger ships such as the carraca or carrack, which boasted three masts with square or lateen sails.61 These round ships were slower and less maneuverable than the galleys, especially when faced with windward conditions. As such, they were generally used for transporting supplies and troops rather than in battle. However, when they were in combat their principal asset was that their higher freeboards, especially when augmented with castles in the bow and stern, made it difficult to board them from ships closer to the water like galleys.62 Though small by modern standards, these ships could become virtual fortresses at sea. To deal with the difficult task of keeping track of all these diverse activities, the crown created the office of Almirante mayor or high admiral. Under Fernando III there were royal officers who assumed naval leadership roles on an ad hoc basis, but it was not until 1254, during Alfonso X’s reign, that the high admiral became a legally sanctioned post and started to appear in royal documentation. After 1260, the term Adelantado mayor de la mar (Greater Military Governor of the Sea) gave way to Almirante,63 and the duties and privileges of the office were laid out in the Siete Partidas.64 The Almirante was the undisputed naval commander, with the duty of preparing a fleet whenever necessary, keeping written records on the state of the realm’s naval resources, and in most cases personally leading the naval contingents. The royal shipyards both in Seville and in Castrourdiales, the shipwrights who worked them, and the town quarters where the shipbuilders settled, also came under his jurisdiction. In peacetime, the high admiral was entrusted with the efficient operation and judicial administration of the realm’s ports, holding jurisdiction over all maritime conflicts.65 He was also in charge of patrolling the Strait of Gibraltar, making sure privateers and other hostile forces were kept at bay. Among the privileges the admirals enjoyed were a share of any booty accumulated during a naval battle or operation; receipt of payment for the transport of any goods in ships under their specific authority; and the right to levy 60 61 62 63 64 65

Partidas 2, 24, 7, 2:263. Marcel Pujol Hamelink, “La tipología naval medieval en Cataluña (siglos VIII–XV): Las fuentes de información,” Revista de Historia Naval 88 (2005), 53. Joé M. López Piñero, El arte de navegar en la España del Renacimiento (Barcelona, 1986), p. 222. José Manuel Calderón Ortega, El Almirantazgo de Castilla: Historia de una institución conflictiva (1250–1560) (Alcalá de Henares, 2003), pp. 168–70. Partidas 2, 24, 3, 2:259–60. The reach of his authority was challenged by the traditionally independent-minded northern Castilian ports. The issue was not settled until 1399, when Enrique III officially and definitively extended the High Admiral’s jurisdiction to include all of the kingdom’s ports. (“Letter from Enrique III [Oropesa, 22 February 1399]” in Calderón Ortega, El Almirantazgo de Castilla, pp. 332–33).

150

Nicolás Agrait

the almirantazgo, a tax over the loading and unloading of merchandise in any Castilian port, and ancoraje, the payment for each ship to drop anchor in the ports. Such substantial benefits made the office particularly attractive and much sought after by the high nobility, yet it was never reserved for members of the aristocracy, and different monarchs filled it according to their needs and the qualifications of the individual candidates. Thus, the office was held at times by Genoese like Benedetto Zaccaria (1292–94),66 Egidiol Boccanegra (1341–66),67 and his son Ambrosio Boccanegra (1368–76).68 The process of gathering a naval contingent, the most important duty of the high admiral, was a complex and time-consuming task not to be taken lightly. It entailed impressing, repairing, or constructing the vessels; recruiting the maritime and military crews; and equipping the ships with equipment and weaponry. The crown’s decision to assemble a fleet invariably constituted a difficult undertaking requiring large quantities of material, human, and financial resources. If anything, the fact that Castile constantly engaged in naval operations serves as proof of the great importance assigned by its kings and commanders to such activities. One established mechanism for the Castilian crown to acquire ships, especially from its northern ports, was through requisition.69 This must have been particularly galling for merchants and other ship owners, who lost the use of their own property and received little or no guarantee of the return of their ships or compensation in the event of damage or destruction. Even though most documents regarding this custom reflect either royal grants of exemption from it or conflict regarding the provision of vessels, the practice was clearly widespread during the thirteenth and fourteenth centuries, in all likelihood resembling that of the English crown.70 This was the certainly the way Ramón Bonifaz gathered a flotilla on behalf of Fernando III to be used in the conquest of Seville in 1248.71 In 1311, San Sebastián managed to declare itself free from having its ships impressed by the high admiral, claiming that the privileges contained in its fuero exempted it.72 Yet the fact that the town’s council continued to lobby to have this privilege reaffirmed during the next three decades suggests that royal officials kept requisitioning ships when necessary.73 Despite its unpopularity, the 66 67 68 69 70 71 72 73

Crónica del rey don Sancho, chap. 1, 1:69–70. Georgius et Iohannes Stellae (d. 1418), Annales genuenses, ed. Giovanna Petti Balbi (Bologna: Zanichelli, 1975), p. 134. Ayala, Crónica de Don Pedro I, year 7, chap. 5, 1:462. Cervera Pery, El poder naval, p. 219. Michael Prestwich, Armies and Warfare in the Middle Ages. The English Experience (New Haven, 1996), pp. 269–70. Primera Crónica General, chap. 1075, 2:748–49. “Royal Privilege granted by Fernando IV to San Sebastián (Toro, 26 August 1311)” in Benavides, Memorias de Don Fernando IV, doc. 556, 2:818–19. “Alfonso XI confirms a royal letter from Fernando IV to San Sebastián ordering that no ships be impressed for the royal fleets (Valladolid, 1318),” “Privilege from Alfonso XI to San Sebastián’s council that no more ships be impressed for the royal fleet (Burgos, 23 May 1345)”



Castile and the Struggle for the Sea

151

practice of requisitioning ships survived and was utilized by Pedro I (1350–69) when he reorganized the navies in his conflict with Aragon.74 Another means to acquire ships was through the specific obligations of certain towns to provide outfitted vessels to the Castilian monarch whenever a fleet was assembled. Known as galera (galley), this obligation was established no later than the thirteenth century and continued into the fourteenth.75 Santander, for example, was required to arm a round ship and a galley any time that the king went on campaign and needed naval support.76 Laredo, in place of paying servicios (levies), provided an armed galley outfitted with sixty oars, sixty men carrying shields and helmets, six crossbows and bolts, and a few dozen lances for a period of service of three months. After that, it was stipulated that the men were to be given the choice of returning home or remaining with the royal fleet.77 In similar fashion, the shipyards of Castrourdiales were obligated to provide a sailing ship or galley for three months’ service, after which the king could retain the ship, but the crew would have the same choice as that from Laredo. While the ship was being built and in use, the city would enjoy freedom from all levies.78 How widespread and effective the galera was is difficult to determine with certainty because almost all the documentary references come from northern Castile. Even though Algeciras was required to outfit two galleys whenever a royal fleet was assembled after 1344,79 we do not know if such obligations also fell upon the rest of Castile’s Andalusian and Mediterranean ports. Regardless, this levy remained a point of contention between the crown and the individual towns. In the cortes of 1329, Alfonso, upon being asked to uphold galera exemptions for some towns, responded that he would respect those granted the privilege during the previous three reigns, but only if he confirmed them. He also affirmed that when a city did fulfill its galera duties it was free from all other royal levies.80 His heir, Pedro I, was less committed toward these liberties, asserting that galera did not carry with it a blanket exemption from all taxation. He was, however, more open to the notion of galera service securing an exemption from fonsadera (scutage).81 In fact, he availed himself of this mechanism in

74 75 76 77 78 79 80

81

in Colección de documentos medievales de las villas guipuzcoanas (1200–1369), ed. Gonzalo Martínez Díez et al. (San Sebastián, 1991), docs. 134 and 232, pp. 135–36 and 246–47. Ayala, Crónica de Don Pedro I, year 9, chap. 10, 1:486. Miguel Ángel Ladero Quesada, Fiscalidad y poder real en Castilla (1252–1369) (Madrid, 1993), p. 45. Libro becerro de las behetrías, 3 vols., ed. Gonzalo Martínez Díez (León, 1981), entry 108, 2:180–81. Ibid., entry 328, 2:567–68. Ibid., entry 330, 2:569–70. Ayala, Crónica del Rey Don Enrique, year 1, chap. 7, 2:442. Ordenamiento de las Cortes de Madrid (Ordinances from de Courts held at Madrid) (1329), in Real Academia de la Historia, Cortes de los antiguos reinos de León y de Castilla, 5 vols. (Madrid, 1861–1903), items 50–51, 1:421. Cuaderno segundo dado a petición de los procuradores de las ciudades y villas del reino en las Cortes de Valladolid (Second written record of the Courts held at Valladolid granted at the

152

Nicolás Agrait

1358 when he assembled a large fleet, impressing ships in all the northern ports except for those specifically exempt.82 The evidence thus indicates that neither side could take this duty, or freedom from it, for granted. In addition to requisition and galera, ships were secured by building or repairing them at one of the royal shipyards. One of these was located at the Vizcayan town of Castrourdiales and was administered by the high admiral or his officials from the nearby city of Burgos.83 The second was the El Arenal shipyard in Seville. Because of all the military operations witnessed by the southern regions of Castile, El Arenal must have undergone a period of intense activity during the late thirteenth and most of the fourteenth centuries. Needless to say, outfitting vessels for war was a time-consuming process; construction supplies had to be gathered and prepared, and then the ship itself had to be refurbished or constructed from scratch. Once it was deemed seaworthy, then it had to be fitted with the proper seafaring equipment such as oars, sails, tools, and rope; extra structures such as castles; weaponry for hand-to-hand combat like knives, hand axes, swords, armor, helmets, and shields; missile weapons, which in the case of Castilian vessels consisted almost exclusively of crossbows, javelins and spears, with the occasional siege engine placed on board; and specialized weaponry such as grappling hooks, soap which was used to make the decks of enemy ships slippery, quicklime which was used to blind opponents, and sulfur, resin, or tar, which all served as incendiary substances.84 After that, a large enough crew had to be recruited, including oarsmen, sailors, naval officers,85 and soldiers. The final step would be to gather the supplies necessary to feed the crew, which included dried vegetables, dried and salted meats and cheese, bizcocho (biscuit), oil, garlic and onions, and vinegar. Even though it was discouraged, cider and wine in all likelihood also made it into the hold.86 Finally, each ship had to take on as much fresh water as it could carry, which for galleys of this period ranged between 800 and 1500 gallons, depending on their size and purpose – less water for supply trips, more for combat missions.87 Even though a newly constructed or refitted galleys or naos must have been a welcome sight, the time it took to get them built – months on end, even in the best of circumstances – would have stretched the patience of any commander.

82 83 84 85

86 87

request of the procurators of the cities and villages of the kingdom) (1351), in Cortes de los antiguos reinos de León y de Castilla, item 48, 2:73. Ayala, Crónica de Pedro I, year 9, chap. 10, 1:486. Fernández Duro, La marina de Castilla, p. 334. Partidas 2, 24, 9, 2:265 ; Crónica del rey don Alfonso, chaps. 263 and 316, 1:338 and 375. Within each individual nao or galley, the highest authority was the comitre, the equivalent of the modern captain, and under him would have been officers named naucheres – masters or pilots – who were navigational experts. Below them were the individual sailors in charge of all the other tasks of running the ship. (Partidas 2, 24, 4–5, 2:260–61). Account drawn up by Juan Matheo de Luna for Sancho IV (1293–94), Archivo Histórico Nacional (Madrid), Codex 985b, fols. 157anv.173r.–74v. John H. Pryor, Geography, Technology and War: Studies in the Maritime History of the Mediterranean 649–1571 (Cambridge, 1988), p. 77.



Castile and the Struggle for the Sea

153

In particular, assembling a fleet in this manner would have made it impossible to respond quickly to unforeseen events or crises. To supplement domestic resources, Castilian monarchs often resorted to contracting mercenary vessels from foreign kingdoms such as Genoa, Aragon, and Portugal. In 1308, Fernando IV negotiated with James II of Aragon (1291–1327) to have the latter send a fleet of ten galleys and three support ships, and assumed all the costs associated with it so that he could continue his offensives in the Strait of Gibraltar.88 Alfonso XI secured the services of fifteen armed and manned galleys to serve in the Strait in 1340 after the Castilian fleet was destroyed.89 He also hired galleys from the kingdom of Aragon, with which he was generally on friendly terms. From 1339 to 1344 there was always an Aragonese component in the Castilian fleets.90 Finally, the kingdom of Portugal was a source of mercenary vessels, even though their service in 1340 and 1342 was very brief.91 Regardless of how the monarchy assembled a fleet, the financial outlays that accompanied these undertakings were so great that they show the crucial importance assigned by the Castilian monarchs to naval operations. This is unsurprising when one considers that naval expenditure began from the moment that an operation started, and continued to mount as long as it lasted. In addition to the aforementioned construction, outfitting and supply costs, one must also add the remuneration given to the crews; and the ongoing repair costs needed to keep the vessels seaworthy; unforeseen emergencies; and the replenishment of victuals, lost or damaged weaponry, oars, tools, and other equipment. Determining the exact monetary sums spent by Castile on naval affairs during this period is a difficult task, since the chancery records for most of the thirteenth and fourteenth centuries have been lost and no local accounts have been uncovered so far. The existing sources, including some from outside Castile, do permit of some estimates, but they also come with their own limitations.92 One especially galling problem is that the documents concentrate exclusively on the galleys, leaving out the naos. While the former composed the bulk of any fleet at the

88 89 90

91 92

“Letter from Fernando IV to James II of Aragon (Alcalá de Henares, 20 December 1308),” Benavides, Memorias de Don Fernando IV, doc. 419, 2:625–26. Crónica del rey don Alfonso, chap. 212, 1:245–46. “Letter from Pedro IV of Aragon to Alfonso XI of Castile ratifying their alliance and sending some galleys to Tarifa under admiral Jofre Gelabert de Cruylles (Barcelona, 15 June 1339),” “Credentials and Instructions given to Juan Escrivá, Aragonese ambassador to Alfonso XI (Valencia, 8 January 1341),” “Letter from Pedro IV to Alfonso XI (Barcelona, May 15, 1342),” “Letter from Pedro IV to Alfonso XI sending ten galleys to his service (Barcelona, 1 July 1343)” in Colección de Documentos Inéditos del Archivo General de la Corona de Aragón, ed. Próspero Bofarull y Mascaró , 41 vols. (Barcelona, 1851–1909), docs. 9, 11, 41, 46, 7:99–100, 102, 157–59, 168–69. Hereafter, CDIACA. Crónica do rei Afonso IV, chap. 55, 2:313–16; Crónica del rey don Alfonso, chap. 264, 1:338– 40. The situation is not the same in the case of Aragon-Catalonia as the records have survived in much better condition, especially in the crown archives of Aragon in Barcelona.

154

Nicolás Agrait

time, especially in the Mediterranean, the upkeep of the sailing ships would also have required significant expenditures. Unlike those of his son and grandson, some of Sancho IV’s royal accounts, specifically for 1292–94, are extant. For the six-month siege of Tarifa in 1294, he ordered a fleet of ten galleys to blockade the port city. The overall cost of the Castilian portion of the fleet, including construction and maintenance, totaled 304,817 maravedíes (mrs.) or about 15,240 florins (fls.).93 This would imply that a ten-galley flotilla cost approximately 50,800 mrs. (2,900 fls.) per month, with each galley costing 5,800 mrs. (290 fls.).94 These vessels were supplemented with three Genoese galleys under the command of Benedetto Zaccaria, each costing 9,414–12,860 mrs. (471–643 fls.) per month. There was also a contingent of twenty-five Catalan ships, for which, unfortunately, there is precious little information. The sole reference is contained in the accounts of the fonsadera collected in the archbishopric of Toledo, stating that 25,000 mrs. (1,250 fls.) be given to Francés Espin, a notary of James II of Aragon, to be put toward the cost of preparing the Aragonese vessels.95 The overall expenditures for this naval operation, including the outlays required to maintain the four or five ships that remained to guard Tarifa until February of 1295, totaled over 930,000 mrs. (46,500 fls.) or roughly 77,500 mrs. a month.96 The actual total expenditure of the crown was likely higher, since the surviving accounts do not make mention of any of the smaller, supporting vessels that would have been present. By Alfonso XI’s reign, the costs had increased. The Genoese fleet he procured in 1340 cost 16,000 mrs. (800 fls.) per galley per month, with the exception of the flagship, which came in at 30,000 mrs. (1,500 fls.) a month. In addition, the Genoese admiral was to be paid 1,500 fls. a month and the flotilla was to be provided with sufficient bizcocho to supply its ships every month. The total cost of this fleet per month would have been 284,000 mrs. (14,200 fls.), without counting each month’s supply of victuals and water for each ship.97 Projecting these costs over one year yields sums of 2,800,000–3,000,000 mrs. (140,000–150,000 fls.). However, other evidence shows that galleys perhaps need not have been quite that expensive. In a 1346 ledger drawn up in the Aragonese court to tabulate Alfonso’s debts to Pedro IV (1336–87) for the period 1339–44, the monthly cost of a galley was set at 9,000 mrs. (450 fls.) – 93

94 95 96

97

Exchange rates are difficult to calculate for the medieval period. I am using an exchange rate of 20 mrs. per florin as calculated by Peter Spufford et al., Handbook of Medieval Exchange (London, 1986), p. 158. Account drawn up by Juan Matheo de Luna for Sancho IV (1293–94), Archivo Histórico Nacional (Madrid), Codex 985b, fols. 156v–157r. Accounts of the scutage collected in the Archbishopric of Toledo (18 September 1294) in ibid., fols. 109r–v. Francisco García Fitz, “La defensa de la frontera del Bajo Guadalquivir ante las invasiones benimerines del siglo XIII” in Las relaciones de la Península Ibérica con el Magreb(siglos XIII–XVI), ed. M. García Arenal y M. J. Viguera (Madrid, 1988), pp. 295–99. Crónica del rey don Alfonso, chap. 212, 1:309.



Castile and the Struggle for the Sea

155

there were supposed to be twenty from 1339 to 1341, and then ten during the Algeciran siege of 1342–44 and afterwards. This would imply that the price of a fleet of ten Aragonese galleys would rise to at least 90,000 mrs. (4,500 fls.) a month, and over 1,000,000 mrs. (50,000 fls.) for one year’s service.98 The figure of 9,000 mrs. is problematic because there were many circumstances that could drive the cost of employing a fleet either up (emergency repairs, supplies, new instructions) or down (reduction in crew, use of naos rather than oared vessels, or reliance on domestic rather than foreign ships). Yet even if one uses a theoretical average price range of 7,000–13,000 mrs. per ship per month, attempting to account for both the higher-priced Genoese ships and the lower expenses of a domestic impressed cog, the overall costs are still astonishingly high. In March 1343, Alfonso supposedly amassed a combined armada of fifty Castilian and Genoese galleys, ten Catalan galleys and forty naos to surround Algeciras from the sea.99 Since he was not paying for the Catalan galleys, the overall expense of just the ninety galleys and the cogs would have been an enormous 630,000–1,170,000 mrs. (31,500–58,500 fls.) per month, and an astronomical 7,560,000–14,040,000 mrs. (378,000–702,000 fls.) over one year. To put these figures in some context, this amount of money could have equaled or signficantly surpassed Alfonso’s annual royal budget, judging by the claim in 1316–17 that the crown’s expenditures amounted to 9.6 million mrs.100 These operations must have strained the royal treasury to breaking point. Considering the burden to the royal treasury, the kings were forced to use several mechanisms to generate the necessary resources to maintain the navies at sea.101 First, there was the revenue from royal rents, indirect taxation, and extraordinary levies.102 Second, they could earmark the revenue from certain local taxes for determined periods of time to support the fleets, as Alfonso often did.103 Third, they borrowed heavily from the Castilian church, foreign rulers, 98

99 100 101

102

103

Account drawn up by Juan Ferrández Muñoz, official in charge of the finances of Pedro IV’s fleet (1346), Archivo del Museo Naval de Madrid, Colección Sanz Barutell, Ser. Barcelona, MS. 355, fols. 213–19. Crónica del rey don Alfonso, chap. 287, 1:293. Ibid., chap. 10, 1:180–81. This is not the place for a detailed discussion of the fiscal mechanisms of the Castilian crown. I have covered these issues much more in depth in “El asta de la lanza: los mecanismos de financiación de la guerra durante el reinado de Alfonso XI (1312–1350) (The Shaft of the Lance: Financing War During the Reign of Alfonso XI [1312–1350]),” Gladius 32 (2012), 103–20. “Fernando IV requests more extraordinary collections from the archbishopric of Toledo to finance the fleets and the protection of Tarifa and Gibraltar (Toledo, 15 July 1312)” in Benavides, Memorias de Don Fernando IV, doc. 583, 2:862; Hilda Grassotti, “Los apremios fiscales de Alfonso XI,” in Historia económica y de las instituciones financieras en Europa, ed. Manuel J. Peláez (Málaga, 1990), pp. 3442–43. “Letter from Alfonso XI to Astudillo, El Barco, Piedrafita, and other places in Alcorneja that he would appropriate all the income from the public notaries to support the fleets (Valladolid, 12 February 1335)” in Colección documental del Archivo Municipal de León (1219–1400), ed. José Antonio Martín Fuertes (León, 1998), docs. 96–97, pp. 138–42; “Letter from Alfonso

156

Nicolás Agrait

and the papacy.104 Fourth, they granted privileges to individuals and groups so as to generate revenue or defer payments to a later date.105 While the Herculean efforts of the Castilian crown to create and maintain its fleets are proof that it considered such naval assets important, the historical record also shows that the kings were keenly aware of the different ways in which navies could be used to advance strategic and military goals. Fleets served to guard the coasts against any hostile intruders, to scout and gather intelligence, to provide support and supplies for ground forces or for strongpoints near the coasts, to conduct amphibious and joint operations with ground troops, to raid enemy ports, to blockade specific areas or ports, and, of course, to search for and engage enemy ships in battle. One of the most important roles that navies assumed was to patrol the waterways around the port areas to ensure the safety of fishing and commercial fleets, especially from pirates.106 The Castilian sources use the expression guarda de la mar (safeguard of the sea) or guarda del Estrecho (safeguard of the Strait) when referring specifically to the Strait of Gibraltar. As the thirteenth century

104

105

106

XI in which he accedes to the petition by the abbot of Sahagún to return to the monastery the income from the public notaries he had taken to supply his fleet (Segovia, 24 September 1344)” in Colección documental de Alfonso XI. Sección del Clero, ed. Esther González Crespo (Madrid, 1985), doc. 301, pp. 502–3; “Privilege granted by Alfonso XI to the council of Briviesca that it would again receive the income of its public notary after five years of it being used by the king to sustain his navy (Burgos, 2 May 1345)” in Colección documental de Pedro I, doc. 488, 2:229–31. “Letter from Alfonso XI to the collectors of the royal thirds, tithes and crusading preaching campaigns granted by John XXIII in 1328 (Madrid, 28 April 1329)” in Manuel García Fernández, “Regesto documental andaluz de Alfonso XI (1312–1350),” Historia, Instituciones, Documentos 15 (1988), doc. 146, 34; Crónica del rey don Alfonso, chaps. 273 and 303, 1:291 and 368; “Loan granted by Clement VI to Alfonso XI (Avignon, 14 Jun 1343)” in Luciano Serrano, “Alfonso XI y el Papa Clemente VI durante el cerco de Algeciras,” Cuadernos de los Trabajos de la Escuela Española de Arqueología e Historia de Roma 3 (1912), doc. 4, 27–31. “Royal privilege granted by Alfonso XI to Guetaria (Royal Camp at Algeciras, 25 June 1343)” in Colección documental de Pedro I, doc. 602, 2:323; “Royal grants to Alfonso Jufre Tenorio, high admiral of Castile (Madrid, 5 July 1329),” “Donation of the village of Moguer to Alfonso Jufre Tenorio, high admiral of Castile (Seville, 3 October 1333)” in “Regesto documental andaluz de Alfonso XI,” docs. 153 and 229, pp. 36, 51; “Royal privilege from Alfonso XI granting Palma del Río to high admiral Edigiol Boccanegra in return for his service to the crown (Royal camp at Algeciras, 2 September 1342),” “Royal privilege from Alfonso XI granting high admiral Egidiol Boccanegra the alcázar at Manifle, and several properties in Algeciras (Seville, 25 May 1344),” “Alfonso XI grants several houses in the Santa María parish of Seville to high admiral Egidiol Boccanegra (Seville, 2 June 1344),” Real Academia de la Historia, Colección Salazar y Castro, MS. M-114, nº 57.240, Sign. 9/920, fols. 1–22. The topic of piracy is too large to be dealt with properly here. As the literature is vast, suffice it so say that some good starting points are Robert I. Burns, “Piracy as an Islamic-Christian interface in the thirteenth century,” Viator 11 (1980), 165–78; María Dolores López-Pérez, “Las interferencias Corsarias y piráticas” in La corona de Aragón y el Magreb en el siglo XIV (1331–1410) (Barcelona, 1995), pp. 577–841; and Cristóbal Torres Delgado, “El Mediterráneo Nazarí: diplomacia y piratería. Siglos XIII–XIV,” Anuario de Estudios Medievales 10 (1980), 227–36.



Castile and the Struggle for the Sea

157

came to a close and the fourteenth progressed, the term came to refer almost strictly to a series of missions to impede the access of supplies and forces by sea from the Moroccan ports of Tangier and Ceuta to the fortified Muslim cities on the Iberian coast like Algeciras and, after 1333, Gibraltar. The possibility of a Muslim invasion added urgency to these activities. In this light one can understand why the Castilian kings were so keen on maintaining a fleet in constant vigilance along the coasts and took the necessary steps to make it so. The fleets were also particularly useful for scouting and gathering intelligence. Certainly the ability of all port cities in the kingdom to observe conditions and relay that information back was augmented by their constant patrols. By the reign of Alfonso XI, the Castilian fleet in the Strait of Gibraltar formed the first line of his intelligence-gathering apparatus, without which his strategic interests would have been seriously compromised. Any information gathered would have reached one of the ports and then been relayed either by land or by sea until it reached the king’s commanders or his court. The waterborne route consisted of ships docking at Tarifa and then stopping at Cádiz, Santa María, and Sanlúcar de Barrameda and then heading up the Guadalquivir river to Seville.107 For example, the Castilian military build-up that led to the battle of Salado (1340) and later operations commenced because of reports of increased military and naval activity in the North African and Iberian Marinid ports.108 In May 1342, almirante Egidiol Boccanegra conveyed the news that the combined Granadan and Marinid fleet in the Strait, if the numbers can be trusted, numbered more than eighty galleys and naos.109 Throughout Alfonso XI’s apparently interminable but ultimately successful siege of Algeciras, part of the Christian armada was always assigned the task of surveilling the Moroccan ports of Ceuta and Tangier. Such activities could be combined with spying, such as when a diplomatic mission was sent to the court of Abū l-ḥasan (r. 1331–51). While awaiting the emir’s response, the Castilian officials headed to the Ceutan docks to survey and take note of the size and condition of the Moroccan ships and their capacity to ferry troops across the Strait.110 Another common method of acquiring information regarding the movements and plans of enemy fleets was through the capture of individual vessels. In some very lucky instances, ships carrying military messages were intercepted. In addition to letters, the captured crews provided, voluntarily or otherwise, whatever information they may have known.111 While the Crónica del rey don Alfonso, el Onceno places a great deal of importance on this last intelligence-gathering mechanism, it is difficult to determine how much meaningful information could be extracted in this manner,

107 108 109 110 111

Muñoz Gómez, “Para el conocimiento de la costa,” p. 184. Crónica del rey don Alfonso, chap. 207, 1:306. Ibid., chap. 263, 1:338. Ibid., chap. 273, 1:347. Ibid. chaps. 282, 300 and 317, 1:354, 365 and 376.

158

Nicolás Agrait

since it is never specified who exactly – shipmaster, fighter, sailor, oarsman, other – was interrogated or what techniques were used. Another fundamental service provided by navies was the transport of troops and, more importantly, supplies.112 Considering Castile’s instability during the late thirteenth and fourteenth centuries, when roads remained susceptible to attack from highway robbers and from Muslim raiders in the southern sections,113 sea and river routes could provide a reliable alternative, in spite of the inherent dangers of waterborne transit. For a Castilian army on campaign in the vicinity of the Strait of Gibraltar, supplies could be brought from two different directions. The first route started in Seville, where victuals and other goods were gathered from the surrounding towns, loaded onto vessels and transported down the Guadalquivir to the port of Salúcar in the Atlantic. From there, a supply fleet continued to Puerto de Santa María, which was also a depot for materials and other goods coming from the northern ports as well as from foreign sources. It also had a shipyard for any required maintenance or repairs. The last link in this supply chain was Tarifa, which was on the Strait itself. Its proximity to Morocco and other Muslim cities in the area endowed it with great strategic importance, but also made it difficult and costly to defend. The second supply route involved ships traveling in a westerly direction from Murcian towns such as Cartagena or other Catalan or Genoese ports through the Strait to Tarifa or any of the other Castilian Atlantic ports.114 This supply system served Fernando IV well during his attempt to capture Algeciras in 1309. He arranged for grain supplies to be brought from Seville through the Strait while he moved his forces over land and established the siege works.115 Alfonso continued this practice, moving not only foodstuffs, materials, and troops but also the components of siege engines both for his failed campaign to retake Gibraltar in 1333 and his more than twenty trebuchets and catapults used to pound Algeciras’ walls in 1342–44.116 Finally, the importance of this supply chain for the Castilian navy at Algeciras should not be underestimated, for it provided food and fresh water to those ships blockading the port, relieved them every few months, and ensured access to shipyards for the repairs that ships engaged in war constantly required. Castilian ships were not limited to serving as a merchant navy. This is unsurprising, since the military personnel serving in individual vessels generally were the same as those on land. To take Alfonso XI’s unsuccessful siege of Gibraltar in 1333 as an example, the capacity of his navy for amphibious operations made

112

113 114 115 116

Jean Gautier Dalché, “A propos d’une mission en France de Gil de Albornoz: operations navales et dificultés financières lors du siège d’Algeciras” in El Cardenal Albornoz y el Colegio de España, ed. Evelio Verdera y Tuells, 3 vols. (Zaragoza, 1972), 1258–59. García Fernández, “La Hermandad General de Andalucía durante la minoría de Alfonso XI de Castilla: 1312–1325,” Historia, Instituciones, Documentos 12 (1985), 354. Torremocha Silva, pp. 195–96. Crónica del rey don Fernando IV, chap. 17, 1:163. Crónica del rey don Alfonso, el Onceno, chaps. 118–19 and 277, 1:251–52, 350–51.



Castile and the Struggle for the Sea

159

it ideal to transport effectives to and from different points around the Gibraltar peninsula.117 When the king arrived with his armies he found a Muslim host from Algeciras lying in wait. He commanded that his vanguard establish a camp near Gibraltar, while some of his nobles and their retinues formed a detachment to protect his rear-guard, the true target of the Islamic soldiers, as it moved from its position in the Sierra Morena down to the safer coastline. This new unit was under orders to confront the Muslims, but to limit pursuit of retreating enemies to short distances. Once the skirmish began, however, the Castilians turned back the enemy and chased them all the way to the Palmones river, close to Algeciras. Suddenly, they found themselves facing the larger Algeciran army. Even with reinforcements, the Christians were outnumbered, overextended, and caught between the Palmones and the suddenly swollen Guadarranque river. With fatigue setting in, disaster appeared imminent until help arrived from the Castilian fleet. Admiral Jufre Tenorio anchored his galleys close by and gathered an infantry unit of about one hundred men. Using a small ship, they traveled up the Palmones river and disembarked. These reinforcements allowed the Christians to match the Muslim force, which chose not to attack. Then Alfonso’s troops retreated safely back to camp.118 Such joint operations continued to be important in the coming years. During the Castilian-Portuguese border conflict of 1337, Alfonso had a navy shadow his host’s march into Portuguese territory. When the Guadiana river blocked his entrance into Portugal, he ordered some of the galleys to travel up the river to meet him. They were then anchored next to each other and, by placing wooden planks above them, formed a bridge that his armies could use to cross.119 During Alfonso’s successful attempt at relieving Tarifa and his subsequent victory at the battle of Salado (1340), the combined Castilian, Genoese, and Aragonese fleet dispersed the Muslim blockade of the city and also prevented supplies from reaching the Marinid armies.120 Coordinated efforts like this one continued later during the siege of Algeciras (1342–44). The Christian navies anchored as close as they could to the ground forces so that each could more easily come to the other’s aid. In August 1343, the Castilians managed to entice a significant Algeciran force into an attempt to break the siege lines. Genoese crossbowmen were drawn from the naval force and helped to drive back the Muslims. The following December the Granadan ruler, Yūsuf I (r. 1333–54), organized his last serious attempt at breaking the Algeciran siege, assembling a large army and a fleet composed of thirty galleys. Alfonso responded by leading a large contingent to meet him at the Palmones river while his high admiral, Egidiol Boccanegra, flanked him by sea with a flotilla of Genoese galleys. As the Castilian ground forces fought their counterparts to a standstill and then forced the Muslims to

117 118 119 120

Ibid., Ibid., Ibid., Ibid.,

chap. chap. chap. chap.

118, 1:251–52. 116, 1:249–50. 183, 1:291. 242, 1:318.

160

Nicolás Agrait

withdraw, the Christian ships scattered the Granadan fleet, effectively neutralizing its ability to assist its companions on land. Thus, his fleet helped Alfonso to garner a significant victory.121 Navies were also used to raid enemy ports and coastal areas. In this case, there was no need to meet an army on land, since the plan was to identify and attack a specific target, and then disembark the forces already present on the ships. Raiding by sea served the same purpose as on land: to acquire more knowledge of a specific area, to cause damage to hostile lands, to secure booty, to extend the range of a conflict, to demoralize the opponent, and at times to establish a forward post in enemy territory.122 One particularly important instance illustrating the Castilian monarchs’ strategic interest in the African Atlantic coast occurred in 1260 when a fleet of thirty-seven ships raided and pillaged the Moroccan port of Salé. After sacking the port mercilessly, the Castilian ships took to the sea and left before relief columns could arrive.123 Although some have speculated that this campaign amounted to a failed invasion on the part of Alfonso X, its timing – September, right before the stormy season in the Strait of Gibraltar area – and the relatively low number of ships involved make it much more likely that this operation was conceived as a raid from its start and therefore achieved its purpose.124 Raiding remained a common activity throughout the thirteenth and into the fourteenth centuries, as can be seen in the Castilian navy’s attacks on the Portuguese coast during the conflict between Portugal and Castile in the 1330s,125 the assaults upon the Catalan and Balearic coasts during Pedro I’s war against Aragon in the 1350s,126 and the Cantabrian and Vizcayan raids upon the English coasts in the 1370s.127 Another important function of navies was to establish blockades. While it was not possible for any medieval navy to completely block any large area, it could attempt, successfully at times, to impede access to a maritime route or, more significantly, a port.128 Nowhere was this more significant than in the struggle for control of the Strait of Gibraltar, where Castile was just one of the kingdoms vying for dominance. Alfonso X blockaded the port of Algeciras as

121 122

123 124

125 126

127 128

Ibid., chaps. 270, 310 and 325–31, 1:344, 371–72, 381–85. Michael Metzeltin, “La marina mediterránea en la descripción de Ramón Muntaner,” in Medieval Ships and the Birth of Technological Societies, ed. Christiane Villain-Gandosi, Savino Busuttil and Paul Adam, 2 vols. (Malta, 1991), 2:317. Cereijo Martínez, “La política marina de Alfonso X,” p. 46. José Manuel Rodríguez García, “La marina alfonsí al asalto de África. (1240–1280) Consideraciones estratégicas e historia,” Revista de Historia Naval 85 (2004), 47–49; Cereijo Martínez, “La política marina de Alfonso X,” p. 53. Crónica del rey don Alfonso, chap. 183, 1:291. Ayala, Crónica del rey Don Pedro, year 7, chap. 7; year 9, chap. 9; year 10, chap. 10, 12–17, 19–20; year 11, chap. 13; year 15, chap. 1–4; year 16, chap. 1, 1:473–74, 486, 494–98, 505–6, 531–32, 534–35. Mott, “Iberian Naval Power,” p. 109. Flores Díaz, pp. 10–11.



Castile and the Struggle for the Sea

161

part of his attempt to take it in 1279, until the lack of supplies and the length of the siege so depleted his fleet that a much smaller Marinid flotilla was able to destroy it in one attack and kill or capture most of the sailors.129 Fernando IV used his fleets to surround both Algeciras and Gibraltar in 1309, managing to take the latter. Alfonso XI used this tactic often as he attempted to limit contact between the Muslim southern Iberian coasts and the Marinid Moroccan ports. In his failed attempt to retake Gibraltar in 1333, his fleet managed to obstruct the Muslims’ access from the sea, putting the city in dire straits. Ultimately, the siege failed and the port was reopened.130 For his siege of Algeciras (1342–44), Alfonso made sure to have enough vessels to cut it off from the sea, though it proved difficult at times, especially at night when small Muslim ships managed to run the blockade and sneak into the port under cover of darkness. Even so, the city’s supply of goods and grain dwindled to a trickle. In March 1343, the king ordered that pine trunks be floated to the nearby Isla Verde from points in the Iberian mainland. The trunks were tied together with chains to serve as a barrier against any ships trying to pass. Unfortunately, this plan backfired when a storm hit the area, breaking the chains and sending the logs toward the shore, providing the Algecirans with much-needed firewood.131 The following January, the Castilian commanders devised another mechanism to prevent access to the Algeciran port. A series of weighted barrels were arranged to surround the city, the first line extending from the shore south of the city and the second from the north, with the two meeting at the Isla Verde. Masts fitted with millstones were lowered into the water so that the barrels could be tied to the masts sticking up above the surface. Additionally, ships were anchored at regular intervals along the new barrier to guard against any attempts to pass over it. This particular barrier, which must have exacerbated the desperation inside Algeciras, survived until the end of the siege a few months later. Only one small ship managed to break through it – not enough to change the outcome of the operation.132 For a city facing a blockade, the most obvious solution was for a fleet to come and break the encirclement. Certainly, the port of Tarifa welcomed the arrival of the Castilian fleet in 1340 prior the battle of Salado, as did Seville when it was relieved after a year-long siege by the Portuguese fleet in 1369–70.133 Yet, when that was not possible another defensive tactic was for ships to form an extrafortified line around the beleaguered port, separating the attacking fleet from the city. In particular, the besieged looked to deprive access to river mouths and other parts of the coast where the besiegers could replenish their stocks of fresh water and forage for supplies. This tactic served Barcelona well in 1359 when Pedro I surrounded it with his fleet. Even though they were outnumbered,

129 130 131 132 133

Crónica de Alfonso X, chaps. 72–73, 1:54–57. Crónica del rey don Alfonso, chap. 126, 1:257–58. Ibid., chaps. 285 and 356, 1:287 and 358. Ibid., chaps. 334–35, 1: 387–88. Ayala, Crónica del rey Don Enrique, year 5, chaps. 4–5, 2:82–83.

162

Nicolás Agrait

the Barcelonan garrison and the sailors prevented the capture of their town and forced the Castilians to retreat.134 The Portuguese used the same tactic against Castile in 1378 when they defended Lisbon from a lengthy naval blockade. In spite of inflicting great damage to the area and severely weakening Portugal’s naval and maritime capacity, the Castilians were ultimately forced to leave.135 Arguably the most important purpose of the diverse Castilian fleets was to search for enemy vessels and engage them in battle. Yet even when fleets prepared for battle, most encounters between hostile vessels at the time tended to be on a small scale. For instance, in 1338, Admiral Jufre Tenorio’s fleet came upon a very large pirate carrack. His ships surrounded it and demanded its submission. When it refused, Tenorio’s ships attacked, boarded, and captured the carrack. After killing its captain and a great many of the crew, they towed the ship to the port of Santa María and refitted it to form part of the Castilian navy.136 Four years later, a Castilian fleet of ten galleys defeated a Marinid force near the port of Bullones. A month later, the same fleet also dispatched another Muslim flotilla launched from Algeciras.137 However, this does not mean that the Castilian fleets during this period did not witness large sea battles. The preparations for such engagements actually began prior to a fleet raising anchor, with each ship taking on additional military personnel, like infantrymen and crossbowmen, and more equipment than usual. The ships would also be outfitted at this time with any extra military structures. In general, these soldiers would have been equipped for hand-to-hand combat with helmets, light cuirasses and shields, and armed with spears, short swords, axes, or daggers. It was up to the officials in each ship to provide the necessary weaponry, including the crossbows or bows along with a supply of bolts or arrows, as well as a substantial reserve to compensate for those lost during the course of the operation.138 While gunpowder weapons were of little consequence during the thirteenth and fourteenth centuries, there is evidence that in addition to the crossbow and other light projectiles ships could boast smaller versions of large stone-throwing engines used on land, such as mangonels.139 As battle became imminent, activity within the galleys and naos must have reached a fever pitch, with officials making last-minute preparations and rallying their crews. The fighters donned their protective gear, prepared their weapons,

134 135 136 137 138

139

Ayala, Crónica de Don Pedro I, year 10, chap. 12, 1:495. Idem., Crónica del Rey Don Juan, year 3, chap. 4, 2:76. Crónica del rey don Alfonso, chap. 242, 1:242–43. Ibid chaps. 263–65, 1:338–42. “Chapters by the king Pedro IV regarding the maritime feats and acts (Barcelona 10 December 1340)” in The Consulate of the Sea and Related Documents, ed. and trans. Stanley S. Jados (University, AL, 1975), pp. 267–68; “Account drawn up by Juan Matheo de Luna for Sancho IV (1293–94),” Archivo Histórico Nacional (Madrid), Codex 985b, fol. 174v. Frederick W. Brooks, “Naval Armament in the Thirteenth Century,” Mariner’s Mirror 14 (1928), 120.



Castile and the Struggle for the Sea

163

and arranged themselves in the fore- and aftercastles. During the first stage of the encounter, and assuming these were mixed fleets, the admirals of the two fleets typically arranged their ships into a tripartite formation similar to that of ground forces: the vanguard would have been composed of galleys, which were the fastest and most likely best-armed vessels; the second line would have been the heavier, slower ships, perhaps the naos or cogs; and the rear line would have been ships dedicated to supply and support roles, perhaps even including a hospital ship. Another common formation, especially when outnumbered, was to have all ships coalesce into one unit, even at times going as far as to tie the ships together, so as to create a floating fortification, maximizing its defensive capabilities. The hope was to make the enemy ships break formation and attack one by one, giving the defending ships the advantage.140 Once organized, each flotilla attempted to outmaneuver the other so as to place itself between its opponents and the direction of the wind. Ships with the wind at their backs would have been able to move at higher speeds and with more flexibility. The other obvious advantage went to the fleet that clearly outnumbered the other.141 As with any other military encounter, those navies that maintained disciplined formations tended to win out. As the ships advanced toward each other, the ships’ crews communicated by voice, if they could; by signals using trumpets, standards, or fire;142 or by having smaller boats go from ship to ship conveying orders. As soon as they were in range, each side unleashed a volley of artillery fire targeted at oarsmen, soldiers, and ships’ sails. As they got closer, other missiles such as javelins would have supplemented crossbow fire.143 Each side then flung rocks, liquid soap to make enemy decks slippery, pots of quicklime, and burning sulfur, resin, tar, or other incendiary substances onto the enemy’s decks.144 While nets or barriers placed over defenders could provide some protection, they were obviously ineffective against flaming liquids. Once the ships were close enough, attacking vessels tried to fasten themselves to opposing ships with grappling irons and hooks.145 With missiles still flying, the next phase of the battle commenced. The infantry attempted to board the opposing vessels and overpower their defenders. If the enemy was put to flight, then the rest of the attackers’ fleet moved in to complete the task. The victor would be the side that managed to capture the most opposing ships or destroy the capacity of the enemy to fight. While the object of the whole conflict was not to destroy ships per se, but rather to incapacitate or capture them, plenty

140 141 142 143

144 145

Foerster Laures, “La táctica de combate,” p. 27. Cervera Pery, El poder naval, pp. 255–57. Crónica del rey don Alfonso, chap. 320, 1:377. “Letter from Pedro IV to Alfonso XI communicating that the Aragonese admiral had left the Strait of Gibraltar to attack Muslim ships along the Barbary Coast (Barcelona, 15 May 1342)” in CDIACA, doc. 40, 7:158. Partidas 2, 24, 9, 2:265; Crónica del rey don Alfonso, chaps. 263 and 316, 1:338 and 375. Crónica del rey don Alfonso, chap. 181 and 209, 1:289 and 307–8.

164

Nicolás Agrait

were lost because of fire or damage to their hulls. There were also factors that were beyond the control of the combatants, such as changes in the winds or sudden storms. One such encounter was the battle of Lisbon or Cape St. Vincent (September 1337) between the Castilian and Portuguese navies. The Portuguese fleet was initially composed of five galleys, seven naos and eight other ships, while Castile’s boasted forty velas (sails). However, before Castilian numbers could win the day, both sides were forced to deal with a sudden storm. By the time both sides recovered, reinforcements had made the fleets more evenly matched, and in fact throughout much of the engagement it was unclear which side would emerge victorious. The tide turned after the Portuguese flagship, captained by the Genoese Manuel Pezano, was captured. At this point, perhaps pushed by a change in the winds, the remaining Portuguese vessels fled. This was a considerable victory for Admiral Jufre Tenorio, who headed for Seville with the captured vessels and a large group of prisoners to present to the king.146 The Castilian admiral’s good fortune ran out three years later when the Castilian fleet attempted to disrupt the transfer of Marinid effectives and ­­materiel from Ceuta and Tangier across the Strait. In spite of leading a fleet that was outnumbered and in serious need of relief, he tried several times to draw the Marinid flotilla, anchored at Gibraltar, into battle, as well as trying to disrupt further enemy convoys. However, on 8 April 1340 he found himself under a surprise attack and forced to fight in very unfavorable conditions. The Castilian fleet, which boasted thirty-three galleys and nineteen naos, faced fortyfour Muslim galleys and thirty-five armed leños, smaller but very mobile oared vessels, that had picked up reinforcements in Algeciras and Gibraltar. Even though the Christian chronicles depict the battle as a heroic episode for the Castilian admiral, in which he ordered his galleys into battle against the odds,147 it appears more likely that the Muslims caught the Christian ships in a state of relative unpreparedness while anchored off the beach at Getares, near Algeciras.148 In the ensuing battle, the Castilian admiral’s galley was overwhelmed and captured, with Jufre Tenorio ending up decapitated, his head flung overboard, and his body kept as a trophy. With the flagship taken, panic quickly spread through the Castilian fleet, enabling the Muslim ships to inflict further damage. By the end of the day, Castile had lost thirty-seven vessels to capture, wreck, or sinking. Only five galleys managed to make it back to Tarifa, while about ten naos carried survivors eastward toward Cartagena.149 146 147 148 149

Crónica do rei Afonso IV, chaps. 37–38, 2:258–65; Crónica del rey don Alfonso, chap. 181, 1:289–90. Crónica del rey don Alfonso, chap. 209, 1:307–8; Crónica do rei Afonso IV, chap. 54, 2:311– 12. Manuel López Fernández, “La actuación de las flotas de Castilla y Aragón durante el cerco meriní a Tarifa en el año 1340,” Aljaranda: revista de estudios tarifeños 64 (2007), 4. “Letter from Pedro IV of Aragon to Don Juan Manuel (Barcelona, 10 April 1340)” in Andrés Giménez Soler, Don Juan Manuel. Biografía y Estudio Crítico (Zaragoza, 1932), doc. 61, p. 636; Crónica del rey don Alfonso, chaps. 210–12, 1:307–8.



Castile and the Struggle for the Sea

165

The last naval battle of significance in the fourteenth century occurred in 1381 between Castile and Portugal. In this case, the result was determined by the indiscipline of the Portuguese fleet and its incompetent leadership. The Portuguese fleet had twenty-one galleys, along with a galliot, four naos and a crew of 6,000 men. The Castilian fleet had twenty-one galleys, with only seventeen in seaworthy condition, which explains why it switched direction and headed toward Huelva when it first came within sight of its opponent. The Portuguese admiral ordered his fleet to give chase, but he also allowed it to become disorganized in its pursuit. The Castilian fleet assembled into a disciplined formation and waited for the next move by the Portuguese. In spite of the lack of coordination and the exhaustion of the oarsmen, the Portuguese admiral gave the order to attack. His galleys of the first line were captured and sunk, and then when the rest of fleet moved in it found itself outnumbered. The Castilian navy achieved a crushing victory, capturing all of the galleys and putting the sailing ships to flight. The Castilians then pillaged along the coasts near the Tajo river, virtually unopposed.150 By the fifteenth century, the Castilian navy had achieved a dominant position in both the northern European seas and the Strait of Gibraltar, well set to benefit from any new developments in maritime and naval technology. Castile’s alliances with Genoa and Aragon (and even more so its eventual union with the latter) assured that its only rival in the Atlantic ocean was Portugal. The growth of Castilian and later Spanish naval power in the Mediterranean and the Atlantic ocean, which is so well documented,151 had its foundations in the preceding two hundred years. The naval institutions developed in the thirteenth and fourteenth centuries were fundamental to the expansion and development of Castile. Furthermore, the crown consciously fostered the development of its fleets and accompanying infrastructure because it recognized their military potential and importance. Medieval Castilian naval strategists thought in similar ways to their counterparts on land, recognizing the need to control strong points and understanding that the acquisition of territory was a slow process, and that securing of communication lines was fundamental. As the Reconquest progressed, the sea and land frontiers came together. If medieval Castilian policy makers did not understand this crucial connection, then how can one explain their readiness to invest such great material, financial, and human resources, time and time again? If anything, naval warfare cannot be dissociated from general military affairs, since, after all, real advances against the Muslims within the Iberian Peninsula

150

151

Saturnino Monteiro, Batalhas e combates de marinha portuguesa, 1139–1521. Vol. I Batalhas e combates de marinha portuguesa, ed. Armando da Silva and Idem (Lisbon, 1989), pp. 33–36; Ayala, Crónica del Rey Don Juan, year 3, chap. 4, 2:76. Fernández Armesto, Before Columbus; Exploration and Colonization from the Mediterranean to the Atlantic, 1229–1492 (Philadelphia, 1987), pp. 207–12; Aznar Vallejo, “La colonización de las Islas Canarias en el siglo XV,” En la España Medieval 5:8 (1986), 195–218; John Lynch, Spain 1516–1598; From Nation State to World Empire (Cambridge, 1991), pp. 116–22 and 329–41.

166

Nicolás Agrait

were accompanied by the capture of the points along the Atlantic, Strait of Gibraltar, and Mediterranean coasts. It is time for a new, more comprehensive way of looking at the Castilian Reconquest that fully recognizes and integrates naval affairs.

7 The Battle of Hyddgen, 1401: Owain Glyndŵr’s Victory Reconsidered Michael Livingston

High up in the central mountains of Wales, in a parking lot beside the dam that forms the reservoir of Nant-y-Moch, is a memorial, dedicated in 1977, intended “to commemorate Owain Glyndŵr’s victory at Hyddgen in 1401.” It is a simple monument, fitting for the stark wilds of the countryside surrounding it, and we can say one other thing for certain about this construct of stone: it is not in the right place. The battle is named for the mountain of Hyddgen, next to or upon which it was reportedly fought, and that mount is roughly three-and-a-half miles northeast from this point as the crow flies, up dirt paths and the trackless rock-andbog slopes beyond. The memorial was placed where it is as a simple matter of convenience. What the setting of the monument lacks in terms of geographical accuracy, however, it makes up for in terms of its geographical precision: the wild feel of the place, subjective and immeasurable though that might be, is appropriate, as is the looming closeness of the mountains that surround Nanty-Moch: for the story of the battle of Hyddgen starts with the fastness of these remote mountains, an environment that is now, as it was then, high and hard and foreboding. Cold in summer, frozen in winter, ever deep and dire and deadly, these mountains of Wales have long stood as sentinel reminders of the smallness of man; and that power is evident whether you stand before the monument or at the foot of the mount of Hyddgen itself. These peaks have driven back kings and armies. A refuge for the hunted, a bane to the hunter, they have stood and withstood for ages. And in the summer of 1401 they made Owain Glyndŵr a prince. Or they didn’t. Because one of the first and most important questions about the battle of Hyddgen is one that cannot be answered definitely: Did it happen at all? Certainly there is little doubt displayed in the scholarship. J. E. Lloyd thought it was real.1 So did R. R. Davies.2 Ian Fleming thought its reality so certain that he wrote a book about the event, reconstructing this “crucial 1 2

J. E. Lloyd, Owen Glendower / Owen Glyn Dŵr (Oxford, 1931), p. 39. R. R. Davies, The Revolt of Owain Glyn Dŵr (Oxford, 1995), p. 266. A full accounting of critics accepting the historicity of the battle is not possible, as it occurs even as a passing notice in disparate contexts; see, for instance, Timothy Scott Jones, Outlawry in Medieval Literature (New York, 2010), p. 58.

168

Michael Livingston

battle” that is the “pivotal victory” from which “all the subsequent events of the Glyndŵr Revolt flow.”3 If indeed the battle happened at all, then Fleming is quite correct in attributing to Hyddgen a central role in establishing the legitimacy of what would become known as the Glyndŵr Revolt. Though the story of the rebellion that would carry Owain’s name as it is told to Welsh schoolchildren today begins with the crowning of the man as Prince of Wales on 16 September 1400, it remains unclear whether it was, in its beginnings, a fully coordinated affair under his authority.4 It is true that it was through this symbolic act that Owain openly declared himself a rival to the English prince who would one day take the throne as Henry V, but there is no evidence that the seizure of Anglesey by the brothers Rhys ap Tudur and Gwilym ap Tudur in that month – or even their seizure of Conwy Castle on 1 April 1401 – was done in Owain’s name. Yet even if the origins of the rebellion in Wales remain somewhat murky – perhaps because it is in the nature of the act of rebellion that questions of origin are difficult to resolve – we can say without doubt when it truly did become his rebellion, when the war in Wales became the war of Owain Glyndŵr: it was in the summer of 1401. Immediately after his 16 September 1400 declaration of rebellion, Owain and his men made what Kelly DeVries has described as “a series of chevauchée-style attacks” against relatively soft targets in North Wales: Ruthin, Denbigh, Rhuddlan, Flint, Hawarden, Holt, Oswestry, Felton, and Welshpool.5 These assaults were troublesome, to be sure, enough to cause King Henry IV to confiscate Owain’s lands on 8 November 1400, and for numerous Welshmen to be put to an 11 November inquisition for treason.6 By the end of the month, however, the king was granting peace to those rebels who would submit to the crown.7 And while the damages that had been done to English holdings were serious enough (in practice or in theory) to merit legal crackdowns on Welsh rights in the English Parliament in February 1401,8 it is nevertheless the case that even as late as 10 March 1401 the king was issuing a general pardon for the Welsh rebels – albeit with the exception of Owain and Rhys and Gwilym ap Tudur.9 The crown of England does not yet appear to have considered the rebellion, and Owain Glyndŵr, a truly significant threat. The next we hear of Owain, however, is in a royal commission of 26 May 1401 that makes clear that the situation has changed. Gone from the king is the

3 4 5 6 7 8 9

Ian Fleming, Glyndŵr’s First Victory: The Battle of Hyddgen 1401 (Talybont, 2001), p. 7. For a brief look at Owain’s career before this declaration, see Anthony Goodman, “Owain Glyndŵr before 1400,” Welsh History Review 5 (1970), 67–70. Kelly DeVries, “Owain Glyndŵr’s Way of War,” in Owain Glyndŵr: A Casebook, ed. Michael Livingston and John K. Bollard (Liverpool, 2013), pp. 435–50, here 440. These sources appear in both original language editions and translations as Items 10 and 11 in Owain Glyndŵr: A Casebook, ed. Livingston and Bollard. Owain Glyndŵr: A Casebook, ed. Livingston and Bollard, Item 12. Owain Glyndŵr: A Casebook, ed. Livingston and Bollard, Item 13. Owain Glyndŵr: A Casebook, ed. Livingston and Bollard, Item 14.



The Battle of Hyddgen, 1401

169

authoritative position of magnanimous pardon. To the contrary, Henry IV seems more or less in a panic: on this Thursday the 26th day of May we received certain news at our castle of Wallingford that Owain Glyndŵr and other rebels from our country of Wales are raising arms and have made a new assembly on the marches of Carmarthen. They intend to enter our realm with a strong force in order to destroy our English tongue and all our loyal lieges and subjects, which God defend! To resist the malice of these rebels we will depart tomorrow from this castle and make our way toward Worcester. We mandate that with knights, squires, men at arms, and sufficient archers of the county of which you are our lord you come to us without fail to that same place in all possible haste through the loyalty and allegiance that you owe us and as you desire the salvation of us and of our realm.10

In the two-and-a-half months since the king had offered general pardon to the Welsh rebels, they had coalesced into a force that the crown now considered a palpable threat against South Wales and Carmarthen in particular. Despite this commission, Henry IV did not enter into Wales at this time, presumably because our next notices of Owain and his men reveal them to be on the run on the southern edges of Snowdonia in North Wales: John Charlton reports defeating Owain near a place that has been tentatively identified as Mawddwy on 1 June,11 and Hotspur reports defeating some of Owain’s men in a separate engagement not far away at Cader Idris on 4 June.12 Charlton states that the rebels had been scattered but were rumored to be headed south again, probably toward Carmarthen. Although it might seem from such English victories that Owain was in retreat, destined to become little more than a guerilla nuisance, on 18 September Henry IV was once again sending out calls for the gathering of an army at Worcester to combat what he describes as a very clear and present Welsh threat led by Owain.13 Within weeks, this English army had at last entered Wales with the clear intent on finding and killing Owain in what amounted to a punitive expedition from Worcester to Carmarthen, ravaging the abbey (and perhaps

10

11 12 13

Nous vous salvons en vous signifiant que yce Jeudy le xxvj. jour de May a nous estoit apportee certeine nouvelle a nostre chastel de Walyngforde que Oweyn Glendourdy et autres noz rebelx de nostre pays de Gales se sont levez et de nouvelle assemblez en les marches de Kermerdyn aiant en purpos dentrer en nostre roiaume ove fort main pour destruir nostre lange Angloys et tous noz foialx lieges et soubgiez, qui Dieux defende! Et pour resister a la malice de noz ditz rebelx nous suymes ordennez a departir demain de nostre dit chastel et de tener nostre chemyn vers les parties de Wircestre. Par quoi vous mandons que ovec les chivalers, escuiers, gentz d’armes, et le pluis suffisantz archiers de nostre countee dont vous estes nostre vis counte vous soiez devers nous [par la ou nous soions] en tout hast possible sauns defaute sur la foye et ligeance que vous nous devez et come vous desirez la salvacioun de nous et de nostre roialme.” Text and translation from Owain Glyndŵr: A Casebook, ed. Livingston and Bollard, Item 16.1–13. Owain Glyndŵr: A Casebook, ed. Livingston and Bollard, Item 17. Owain Glyndŵr: A Casebook, ed. Livingston and Bollard, Item 18. Owain Glyndŵr: A Casebook, ed. Livingston and Bollard, Item 19.

170

Michael Livingston

even beheading a monk) of Strata Florida for supporting Owain, and somewhat famously executing Llywelyn ap Gruffudd Fychan of Caeo and his eldest son at Llandovery for failing to give up Owain’s whereabouts.14 The summer of 1401 thus saw the Welsh Revolt become Owain’s War. And it was sometime in those months that the battle of Hyddgen is said to have occurred, in the mid-Wales mountains between Snowdonia and Carmarthen. To this we must turn. Our most complete medieval record, and indeed our only direct reference to the battle of Hyddgen, comes from the Annales Oweni Glyndwr, which was probably written between 1422 and 1450, but which survives in only three manuscripts, the oldest of which was copied between 1556 and 1564 in the hand of Gruffudd Hiraethog (d. 1564). The report of the event in the Annales is brief but tantalizing: The next summer after that, Owain rose up with six score wicked men and thieves, and he brought them as to war into the uplands of Ceredigion. And fifteen hundred men from the lowland of Ceredigion and Rhos and Pembroke assembled there and they came to the mountain to try to capture Owain. And on Hyddgant Mountain was the encounter between them, and as soon as the English host turned their backs to flee, two hundred of them were killed. And then great praise came to Owain, and there rose up with him a great part of the youth and the wicked men from every region of Wales until there was a great host with him.15

The only other possible corroboration we have for this event comes from the contemporary chronicle of Adam Usk, this part of which was written on or around 14 February 1402: All this summer Owain Glyndŵr and several of the Welsh chieftains, whom the king regarded as traitors and outlaws from his kingdom, severely devastated West and North Wales, taking refuge in the mountains and woodlands before emerging either to pillage or to slaughter those who tried to attack or ambush them.16

Such evidence is not a lot to build upon, and it is troubling that we have no other corroboration for a military action on the scale reported by the Annales: such an assemblage of men, and such a loss of them, might be thought to have left some sign in troop musters or payment accounts – or at the very least in multiple reports of contemporary recriminations or lauds from either side. Yet none of 14 15

16

Owain Glyndŵr: A Casebook, ed. Livingston and Bollard, Item 22.17–38. “Yr haf nessaf wedi hynny y kyvodes Ywain ar y chwigeinved o ddireidwyr a lladron ac y ddaeth ef ac wynt ar ryvel i flaenav Keredigion, ac yno ydd ymgynvllawdd mil a haner o wyr bro Geredigion a Rros a Phenfro ac y ddaethant i’r mynydd i geissiaw dala Owain. Ac ar Vynydd Hyddgant i bv yr ymgyfarvod ryddynt, ac yn gytnaid ac y troes y llv Seissnic eu kefnav y ffo y llas CC o honyw. Ac yna y ddaeth gair mawr i Owain ac i kyvodes attaw ran vawr o’r yfiengtid a’r direidwyr o bob gwlad o Gymry oni oedd gidac ef llu mawr.” Text and translation from Owain Glyndŵr: A Casebook, ed. Livingston and Bollard, Item 70.6–13. “Tota illa estate Oweyn Klyndor cum pluribus Wallie proceribus, regni exules et regis proditores habiti, in montanis et siluestribus delitentes, aliquando depredando, aliquando insidias et insultus eis inferentes interficiendo, partes West Walie et North Walie non modice infestarunt.” Text and translation from Owain Glyndŵr: A Casebook, ed. Livingston and Bollard, Item 22.12–15.



The Battle of Hyddgen, 1401

171

this exists. Lloyd, aware of these problems, nevertheless accepted the general historicity of the Hyddgen story, observing that “some such success is required to account for the situation as it appeared to the king at Wallingford.”17 In other words, it is reasonable to assume that there had to be events in Wales prompting the king’s panic in the summer of 1401, and none of the other known or rumored events of Owain’s career so neatly fits the royal reaction as Hyddgen. By this logic, even if we cannot be certain that the battle of Hyddgen occurred as the Annales reports, we can be relatively confident in saying that something very much like it must have occurred – for nothing else adequately explains the actions of the English. For his part, Lloyd therefore surmises that the Annales must have a core of historical truth, and that Hyddgen must have occurred in mid to late May: it was this battle that gave rise to the king’s call to arms on 26 May, a call to arms that fizzled out when news arrived of the Welsh defeats less than a week later. While I agree with Lloyd’s rationale for accepting a Hyddgen event, and I also concede the essential historicity of the account in the Annales – at least until evidence appears of either a rival contender event or a convincing reason to disbar the Annales report – I think Lloyd and those who have followed him are mistaken on the timing of the event. If the battle of Hyddgen occurred – and, again, I agree with Lloyd that it is more likely to have happened than not – the engagement probably happened later in the summer, between Owain’s defeats in early June and the king’s second call to arms in September – the call that really did lead to an invasion of Wales. Such evidence as we have suggests that the sequence of events began when Owain and his rebels made a chevauchée of some kind into South Wales, no doubt raiding for supplies and attempting to disrupt the English systems of trade while working to agitate the Welsh marches around Carmarthen into rising up against English rule. The first months of Owain’s rebellion had been focused in North Wales, and the opening of this new southern theater of operations revealed some of the grand measure of his designs – and the corresponding scale of the threat that the English faced.18 It was this military action, not a defeat in the high mountains of Hyddgen, that likely prompted the king’s letter from Wallingford at the end of May. The king did not invade Wales at that time, however, because as preparations were being made he learned that John Charlton and Hotspur had handed the rebels significant defeats in West Wales. These English victories struck Owain’s men as they were completing a somewhat triumphant circuit of the country, returning to their bases in North Wales flush with goods from their southern chevauchée.

17 18

Lloyd, Owen Glendower, p. 39 n. 3. As Richard Suggett has suggested, the possible scale of the destruction from these and other raids may well be found in the “apparent absence of houses built before 1430” that survive in Wales today; see Suggett, “The Chronology of Late-medieval Timber Houses in Wales,” Vernacular Architecture 27 (1996), 32.

172

Michael Livingston

John Charlton’s report indicates that Owain was now very much on the run, and it further suggests that he was moving southward again. Though we cannot know for certain, it is reasonable to assume that as the rebels did so they looked to the mountains for safety – this had been and would continue to be their modus operandi across the revolt, much to the annoyance of Henry IV. Keeping to the safety of the mountains as they rode south from their defeats, the Welsh rebels would thus have passed close upon the steep slopes and high valleys around Mynydd Hyddgen, a fitting place for Owain to lick his wounds and recover. Meanwhile, if we are to give any credence to the report in the Annales – and, as we have seen, the circumstances of what we know about that summer seem to support its basic accuracy – English-aligned forces from the areas previously pillaged by Owain in the south, particularly the Anglo-Flemings from the region often called Little England Beyond Wales, had gathered their own forces to repel Owain’s incursions.19 What is more, these troops – even if not the 1,500 men of the Annales’ report, certainly they would have been several hundred strong – made the decision not just to maintain a defensive position against the rebels, but to move directly against them in the mountains. They were taking the fight to Owain where he would least expect it: to Owain’s own base of operations. According to Cledwyn Fychan’s study of the oral traditions passed down through the generations of shepherds in the area, that location was just north of Hyddgen mountain, a natural rock cleft called the Siambr Trawsfynydd, of which it was said by a Mr. Pugh in February 1966 that his father had taught him that it was here “where Owain Glyndŵr’s soldiers slept when the English came at their worst upon them and that many of them were buried on the mountain nearby. Mr Pugh said that the old people used to tell this story as if they truly believed it and not like some of the other traditions.”20 In a landscape generally devoid of shelter, this relatively protected site of the Siambr Trawsfynydd would indeed have made for a logical base of operations for Owain. In addition to its qualities as a shelter in an unforgiving land, the site is both well hidden and well suited for keeping watch over threats from the north and west in particular: the ridge line just above it affords a view across the low spurs of the Dyfi Valley and its tributaries, all the way to the ridge lines where Charlton and Hotspur had handed the Welsh their recent defeats. Fleming, who has provided the only lengthy account of the Battle of Hyddgen, follows the suppositions of Mr. T. O. Morgan, who, in an 1851 issue of the Archaeologia Cambrensis, opined that the battlefield was located upon the very

19

20

The variability of allegiances in this chaotic period can be observed in the fact that, as Ralph A. Griffiths suggests, many men who took part in this assault on Owain likely supported his cause in the following years; see Griffiths, Conquerors and Conquered in Medieval Wales (New York, 1994), p. 52. Quoted in Elissa Henken, National Redeemer: Owain Glyndwr in Welsh Tradition (Cardiff, 1996), p. 77.



The Battle of Hyddgen, 1401

173

summit of Mynydd Hyddgen itself.21 Fleming additionally supposes that the Anglo-Flemish force, coming from the south, split into two arms in order to approach Owain’s camp at the Siambr Trawsfynydd. One of these arms swung west in order to cut off Owain from a ride north, while the other marched up the Hyddgen valley directly from the south (47–48). The rebels, surprised between them, retreated for the high ground of the summit and were subsequently encircled. Fleming supposes that they were there besieged for “a number of days” before the Welsh made a final desperate charge downhill, sending the AngloFlemings into confusion and killing roughly 200 of them (53). To Fleming’s credit, his reconstruction does place the battle on the mountain of Hyddgen, as reported in the Annales, and his downhill charge also fits with the account of the battle provided in the Memoirs of Owen Glendowr, written by Robert Vaughan or Thomas Ellis (or perhaps co-authored by the two of them) in the second half of the seventeenth century, which supposedly utilized additional medieval sources that have now been lost: The Flemmings of Roos, Pembroke and Cardigan, whom Owen distressed most of all raised 1500 men and went agst him being full of confidence that they would either kill him or take him. they hemmed in on all sides at a place called mynydd hyddgant, that he could not possibly get off wth.out fighting at a great disadvantage. He and his men fought manfully a great while in their owne defence agst them. Finding themselves surrounded & hard put to it they resolved at length to make their way through or perish in the attempt. So falling on furiously wth. courage whetted by despaire they put the enemy after a sharpe dispute into disorder, and they pursued so eagerly their advantage that they made them give ground and in ye end to fly outright leaving 200 of their men dead in the place.22

Yet for all its grand theater, Fleming’s reconstruction, I would suggest, is unlikely. For one thing, it is highly improbable that an Anglo-Flemish force of such size could have marched upon the Welsh rebels, besieged them upon a mountaintop, been summarily routed, and then left no record beyond the paltry account in the Annales. In order to account for the lack of evidence for this force – again, assuming that there is some measure of historicity in the battle of Hyddgen at all – the numbers involved must have been far fewer than the 1500 reported in the Annales. Only a much smaller force – probably in the range of only a few hundreds – might plausibly have left no mark in the records. The smaller the numbers involved, however, the more implausible the split-force approach suggested by Fleming becomes, not to mention that the English forces in and around the Dyfi Valley that had just recently defeated Owain would 21 22

T. O. Morgan, “Historical and Traditional Notices of Owain Glyndŵr,” Archaeologia Cambrensis, new series 2 (1851), 30–32. This text follows that of the earliest manuscript record of the Memoirs, National Library of Wales MS 14587B, 59–60, which dates to the last quarter of the seventeenth century; I am grateful to Gruffydd Aled Williams for providing this transcription. The first printing would not appear until roughly a century later, in Thomas Ellis, Memoirs of Owen Glendowr (London, 1775), p. 58.

174

Michael Livingston

have already served as a significant deterrent to retreat in a northern direction. A smaller-sized Anglo-Flemish force would also make it far more difficult to encircle Owain’s force upon the high and wide summit of Hyddgen – if indeed the Welsh took position upon the summit, which I think they did not: the idea of encircling this mount – even with 1500 men – is rather striking in its implausibility. Lines that somehow perched on the mountain slopes with an aim to enclose the flat summit alone would need to be nearly a mile long, while lines better positioned for besieging the mountain itself might need to be seven miles long or more. Besides which fact, it would be foolish indeed to take position upon an exposed mountaintop, cut off from supplies and lines of retreat, and history would prove Owain no fool. Odder still, how are we to imagine that Owain and his rebels – who knew the mountains better than any English commander could possibly hope to achieve – were surprised by so many men marching from the Anglo-Flemish reaches near Carmarthen? The distance from Carmarthen to Hyddgen is fifty-six miles as the crow flies, and far more by the twisting paths that they would have needed to follow – when they indeed had paths to follow at all. It would be a most extraordinary lack of intelligence and, even, simple guard placement for the Welsh rebels to have been so surprised that they had no choice but to scramble in desperation up to the desolate, windswept summit of Hyddgen. In 1931, Lloyd gave an alternative location to this summit site, stating in his famed history of Owain that the battle occurred “on the banks of the river Hyddgen”;23 Lloyd’s deserved authority has consequently caused a valley setting to become the dominant presentation of the engagement. Ian Skidmore, for instance, imagines that the Anglo-Flemings trapping the Welsh “in a mountain glen in the Hyddgen valley,” where they “ringed the rim of the glen and their downward charge gave them a formidable advantage over the Welsh, forced to fight uphill” – thus achieving a field situation that is both in complete opposition to the one reported in our sources and, as near as I can tell, impossible, given the actual terrain of the area.24 None of these locations is tied to argument, nor are they presented as rebuttals to the Hyddgen summit. Instead, they seem to be drawn to a valley location simply due to the presence there of the so-called Cerrig Cyfamod Glyndwr [Owain’s Covenant Stones].25 These “Covenant Stones” were described by a Royal Commission site visit on 6 May 1910 as Two unhewn blocks of white quartz, almost certainly placed with intention, with a small natural outcropping boulder midway between them, to which the local tradition points as marking the site where Owen Glyndŵr “held parley, and made his covenant.” The boulders are on the north-western slope of Plynlymon, 100 yards above the right 23 24 25

Lloyd, Owen Glendower, p. 39. Ian Skidmore, Owain Glyndŵr: Prince of Wales (Swansea, 1978), pp. 59–60. That the Stones were perceived by Lloyd to have a direct bearing on the location seems plausible, given that he does include mention of them in a footnote discussing the Annales as the “sole authority for the battle” (Owen Glendower, p. 39 n. 3).



The Battle of Hyddgen, 1401

175

bank of the river Hyddgen, and in a shallow depression between Carn Gwilym and the carneddau on Banc Llechwedd Mawr. They are not marked on the Ordnance Survey sheet. Each stone is two feet high, and shows no trace of tooling. They are 60 feet apart and are aligned exactly north and south. Though not placed in an elevated position, they can be seen from afar, and show up conspicuously in the sunlight.26

More modern surveys have noted that the stones do not follow an exact north– south alignment, but they do perhaps align with a third stone roughly 3,600 feet south, upon the slopes of Plynlimon Fawr. So strong has been the pull of Lloyd’s authority and the draw of these quartz blocks that writers such as Chris Barber have flatly stated that they “mark the site of the battle.”27 Whether or not these blocks are a deliberate placement; whether, if so, the tradition connecting them to Owain is accurate; and whether, if so, the rebel’s “covenant” has any connection at all to the battle of Hyddgen are questions that cannot now be answered. But they do have a kind of romantic appeal, which certainly lies behind any valley placement for the battle. After all, there is no evidence to otherwise support such a supposition, and it works completely against our earliest reports of a battle site on the mountain itself. If the battle is unlikely to have occurred on the summit of Hyddgen as Fleming conjectures, and the Covenant Stones likewise have little to recommend them as indicators of the battle site’s being located in the valley itself – if indeed they are related to Owain at all – then what alternative location might make the best sense of the source evidence, the physical topography, the conditions of the two forces, and what is known of Owain’s generalship? In the remainder of this essay I will present an alternative site, albeit with the significant caveat that any attempt at locating a battle that cannot itself be proven to have occurred must, of course, be exceedingly tentative. Our earliest accounts make clear that the battle of Hyddgen occurred upon the mount of Hyddgen, as Fleming noted, but this need not mean that the battle occurred upon the mountain’s summit. To the contrary, it need only mean that it occurred somewhere upon the mountain, and indeed I think it did: just much farther down its slopes, tantalizingly close to the Covenant Stones themselves. Rising directly east from the Covenant Stones is the west side of Banc Lluestnewydd, the hard, southern knee of the mount of Hyddgen. There is a natural draw up the slope in this location, fed by a freshwater spring. An easy climb up this draw allows one to look back down upon the Stones from this lower part of the mountain, and it also reveals upon the slopes of the Banc here the existence of a surprisingly flat area, which is approximately the size of a football pitch. The field is large enough to allow for the presence of a significant force, and while it is protected on two sides by steep cliffs, the avenues of exit to either side are surprisingly smooth, albeit steep (Figure 1).

26 27

Http://www.coflein.gov.uk/en/site/303679/ [accessed 31 May 2014]. Chris Barber, In Search of Owain Glyndŵr, 2nd ed. (Abergavenny, 2004), p. 69.

176

Michael Livingston

Figure 1.  The site of the battle of Hyddgen

Because of the topography of the Hyddgen area, this unnamed field is in a prime location: any force located on the Banc overlooks the confluence of – and thus has control over – the three main river valleys in the watershed: the Afon Hyddgen, Afon Hengwm, and Nant y Llyn. The location also provides an excellent vantage point from which to observe the southern approach toward Owain’s camp further up the valley. If Owain was in this area, as tradition, circumstance, and tactical logic dictate, then a position on the Banc was surely one of his primary watches. And what a position it would be! The outer edge of this flat ground beside a running spring is lined with a ridge line of rock more than a man’s height: a natural battlement protecting anyone on this field from view, while affording significant advantages as a defensive position, should an attack come. If there was a Welsh force located at Hyddgen in the summer of 1401, if indeed an Anglo-Flemish force marched north against them, then the line of their approach from the south would have been under clear view for miles from this protected position. It is simply unbelievable that the Welsh were surprised, as previous critics have imagined. To the contrary, they had to know the attack was coming, and they thus had time to take up the best position possible: which was surely this hidden field amid the ridges of the Banc Lluestnewydd, a toponym which indeed means the “Bank of the New Encampment.”28 The field has access to water, relative shelter compared to the barren desolation of the summit, two easy 28

How long this toponym has been associated with this location has not been ascertained. While it



The Battle of Hyddgen, 1401

177

lines of retreat, and significant natural defensive advantages: quite remarkably, while hundreds of men could have a position here – well supplied and well ready to shoot down the slope at any attackers – the location is hidden from view from the valley below. It is unlikely that anyone would be any the wiser about the force in place until the killing began. If Owain and his men took their positions here in the summer of 1401, they could easily have waited for the enemy to pass below this place on their route toward the rebel camp further up the valley, perhaps at Siambr Trawsfynydd itself. It is likely that the AngloFlemings would have followed some kind of track along the valley’s eastern edge: this would put them on the correct side of Afon Hyddgen – at the foot of Mynydd Hyddgen – and the land here is slightly drier and more firm than that on the lee side of the valley. If their path took them close to the location of the modern road – a suitable assumption, given how little the terrain in this wilderness has changed – then the Anglo-Flemings would have been passing less than two hundred yards below the rebels, who no doubt shot down into them, goading them into an attack uphill. From our limited sources, it seems that the Anglo-Flemings, who had come so far for this moment, complied. They charged uphill, and perhaps for a few minutes felt they had trapped Owain and his little band among the rocks; as the Memoirs reports, the Welsh were “hemmed in on all sides.” Only at the last moment did Owain spring his trap: he had far more men hidden upon Banc Lluestnewydd than could ever be seen from below, and they came down upon the Anglo-Flemings like a terror. What connection, if any, these events might have to the nearby Covenant Stones cannot be said. Perhaps there is none. Or Owain staged a “flight” from a location at or near the Stones in order to provoke pursuit up the northern draw, one that would funnel the enemy into charging this single approach to his hidden field. Owain might then have moved a force out from the back side of his position, not only flanking the enemy but effectively cutting off their immediate line of retreat. Slaughter would have ensued as the Anglo-Flemish force would have been caught in an action with a remarkable parallel in Owain’s greatest victory, the battle of Bryn Glas, which happened the following year, in the summer of 1402. There, too, his outnumbered force took the high ground, prompted the excited enemy into a difficult uphill assault, and then executed a flanking maneuver from a hidden position that amounted to a devastating ambush.29 Few would have survived as they were run down in flight upon the valley floor. A Banc Lluestnewydd location for the battle of Hyddgen would thereby account for all our written sources, the traditional association with the Covenant Stones, the oral reports of the local shepherds, the logistics necessary for all parties, and the known tactics of an able military commander who could turn an efficient defense into a devastating offense. It was thus most likely upon these

29

may owe something to memories of Owain’s actions in 1401, it may also be simply attributable to the more modern lodgings of shepherds in the area. For the most recent study of this battle, see Michael Livingston, “The Battle of Bryn Glas, 1402,” in Owain Glyndŵr: A Casebook, ed. Livingston and Bollard, pp. 451–74.

178

Michael Livingston

lower slopes of Mynydd Hyddgen – not high upon its summit, and at least not initially in its wet valley below – that the battle happened in 1401. It was here that they fought and they died. It was here that the revolt in Wales became the revolt of Owain Glyndŵr. I began this article by observing how well the mountains of Wales – the peaks and valleys that, as we have seen, may well be said to have given Owain’s rebellion birth – have withstood the predations of man. Alas, it seems they may at last have met their equal. Plans are underway to place a wind farm upon the slopes of Mynydd Hyddgen. The aim of this renewable energy machinery is without doubt noble, but the wind farm will also change the landscape irrevocably. Footings will be driven into rocks upon which Owain and his men may once have stood, and the turning blades of the modern world will cast long shadows upon the valley in which he fought. Such is often the price of progress. Still, one might well hope that this incursion of the modern world into the mists of Mynydd Hyddgen may serve as a chance to bring to light the medieval world to which we owe so much: the inevitably new access routes to the mountain will provide an excellent opportunity to bring new access, new eyes, new interest, and new observations upon the battlefield where Owain won such a lasting victory.

8 The Provision of Artillery for the 1428 Expedition to France1 Dan Spencer

Introduction On 24 October 1428 the commander of the English army besieging Orléans, the earl of Salisbury, whilst observing the city from the newly captured fortification of the Tourelles, received a mortal injury as a result of a cannon-ball striking a wooden ledge which forced a fragment of wood into his face, and died eight days later.2 The loss of England’s leading commander was a critical event for the eventual outcome of the siege. By 8 May, under the inspired leadership of Joan of Arc, the French were able to lift the siege of Orléans. This was undoubtedly a turning point in the war. As the earl of Salisbury experienced to his cost, artillery was increasingly important by 1428.3 However, relatively little evidence survives for gunpowder weapons in English armies in the early fifteenth century.4 This is largely due to the lack of surviving records for the Tower of London, the chief arsenal for the royal ordnance.5 Fortunately, extensive archival evidence survives in Exchequer records for the expedition of 1428, which culminated in the siege of Orléans. This includes the earliest known set of accounts for artillery for a fifteenth-century English expeditionary army: The National Archives, Exchequer Accounts Various, E101/51/27 and E101/51/30.6 This material permits major insights into the provision and transportation of artillery. It shows that substantial resources were invested in the 1 2 3

4

5 6

The author would like to thank Professor Anne Curry for reading and commenting on an earlier draft of this paper. Kelly DeVries, Joan of Arc: A Military Leader (Stroud, 1999), p. 61. For the growing importance of gunpowder weapons in the early fifteenth century, see Bert S. Hall, Weapons and Warfare in Renaissance Europe: Gunpowder, Technology and Tactics (Baltimore, 1997), pp. 63–64. By comparison, more evidence survives for Burgundian artillery of the period, for which see Robert D. Smith, and Kelly DeVries, The Artillery of the Dukes of Burgundy 1363–1477 (Woodbridge, 2005). David Grummitt, “The Defence of Calais and the Development of Gunpowder Weaponry in England in the Late Fifteenth Century,” War in History 7 (2000), p. 255. These accounts have also been discussed by Mark William Warner, “The Montagu Earls of Salisbury, 1300–1428,” (unpublished Ph.D. thesis, University College London, 1991); Clifford J. Rogers, “The Military Revolutions of the Hundred Years’ War,” The Journal of Military

180

Dan Spencer

creation of gunpowder weapons and that their potential importance in warfare was recognized. It is therefore of crucial importance in understanding how English gunpowder weapons developed during the reign of Henry VI. Background The year 1428 can be seen as the high-water mark of English success in France. Henry V’s military successes in France, including victory at the battle of Agincourt and his conquest of Normandy, culminated in the Treaty of Troyes in 1420. By this treaty Henry V became heir to Charles VI, king of France. This was contested by the Dauphin Charles and his Armagnac supporters. The treaty committed Henry and his supporters to continue the war until the Armagnacs were defeated. This necessitated extensive military expenditure and campaigning, using troops within France and Burgundy, as well as expeditionary armies from England. Henry’s death in 1422, and the succession of the child king Henry VI, meant that the war was directed principally by the duke of Bedford, as Regent of France. Bedford continued the conquest of Normandy, which was completed by the summer of 1424, and achieved a significant victory at the battle of Verneuil in August of the same year.7 English objectives were then expanded to include the neighbouring provinces of Maine and Anjou.8 On 10 August 1425, Le Mans, the capital of Maine, was captured, and by 1427 the complete conquest of Maine and Anjou appeared to be imminent.9 Despite these successes the English suffered a setback at the siege of Montargis in the same year.10 In 1428, however, a shift in focus occurred as a result of the decision to send a large army from England under the command of the earl of Salisbury to campaign in the Loire valley. Thomas Montagu, fourth earl of Salisbury, was, as Anne Curry remarks, England’s “most talented and experienced commander,” with extensive military experience in France.11 Under Henry V he had played an important role in the sieges of Caen, Falaise and Rouen, which led to his appointment, on 26 April 1419, as the king’s lieutenant-general in Normandy. Following Henry’s death, he fought in the battles of Cravant and Verneuil, and campaigned in Champagne. In 1425 he took command of the invasion of Anjou and Maine, with his use of artillery key to the capture of towns, including Le Mans.12 Two years later, in 1427, he returned to England to gather reinforce-

7 8 9 10 11

12

History 57 (1993), 270 n. 127; Kelly DeVries, “Gunpowder Weaponry and the Rise of the Early Modern State,” War in History 5 (1998), p. 141 n. 76. Ralph A. Griffiths, The Reign of King Henry VI: the Exercise of Royal Authority, 1422–1461 (London, 1981), p. 185. Betram Wolffe, Henry VI (London, 1981), p. 53. Ibid., p. 54. A. J. Pollard, John Talbot and the War in France, 1427–1453 (London, 1983), p. 12. Anne Curry, “Montagu, Thomas, Fourth Earl of Salisbury (1388–1428),” Oxford Dictionary of National Biography, Oxford University Press, 2004; online edition, January 2008, http://www. oxforddnb.com/view/article/18999 [accessed 9 September 2013]. Warner, “The Montagu Earls,” pp. 172–73.



The 1428 Expedition to France

181

ments for a major expedition to France. This was the army which set sail from England in July 1428 and for which the artillery detailed in E101/51/27 was provided. In 1433, reflecting back on the matter, Bedford stated “of the siege of Orléans [it was] taken in hand God knows by what advice.”13 This reveals the difference of opinion which there had been in 1427–28 as to how the war should be pursued. Bedford was fully committed to the conquest of Anjou and Maine, areas which had been granted to him in 1424 in anticipation of their conquest. In May 1428 a decision was taken at the Grand Conseil in Paris to attack Angers, the capital of Anjou.14 But in England a decision was taken to send an expeditionary army to target the Dauphin’s power base. This divergence stemmed from political divisions between the two uncles of Henry VI, Humphrey, duke of Gloucester, as Protector in England, and John, duke of Bedford, as Regent of France. The 1428 campaign was an ambitious one, intended as part of an aggressive strategy to directly target the lands of the Dauphin by capturing Orléans, and to secure English control of the Loire.15 The Loire valley was an area of enemy territory – as opposed to Anjou, which was already partly conquered – so the English army would be isolated and at greater risk of being cut off from support from Lancastrian France and Burgundy. The 1428 expedition under Salisbury was therefore an ambitious and risky venture, with 2,400 troops from England intended to participate; it also saw the largest force sent to France since the reign of Henry V.16 This included a large artillery train sent with the expedition, which was assembled in England and financed by the English Exchequer. The course of the expedition Salisbury attended Privy Council meetings regularly from July 1427 until 1 June 1428. On 24 March 1428 he entered into an indenture for an expeditionary army of 600 men-at-arms and 1,800 archers to serve for six months. Salisbury subsequently made use of a clause in the indenture which allowed him to substitute 450 archers for 156 men-at-arms, so that the eventual size of the army sent

13 14

15 16

Proceedings and Ordinances of the Privy Council of England, ed. N. Harris Nicholas, 7 vols. (London, 1835), 4:223. As late as 14 September 1428, two months after Salisbury had started his campaign in France, money was voted by the Estates of Normandy for the siege of Angers, but this aide had to be diverted in February 1429 to fund the siege of Orléans. Wolffe, Henry VI, pp. 53–54; Warner, “The Montagu Earls,” p. 191; B. J. H. Rowe, “The Estates of Normandy under the Duke of Bedford, 1422–1435,” English Historical Review 46 (1931), 564; Letters and Papers Illustrative of the Wars of the English in France during the Reign of Henry the Sixth, King of England, ed. Joseph Stevenson, 3 vols. in 2 parts (London, 1864), 2.2:79–84. DeVries, Joan of Arc, p. 57. Anne Curry, “English Armies in the Fifteenth Century,” in Arms, Armies and Fortifications in the Hundred Years War, eds. Anne Curry and Michael Hughes (Woodbridge, 1994), p. 47.

182

Dan Spencer

from England was 2,694 men.17 Both the indenture and the warrant for issue given the following day mentioned provision for artillery and for an ordnance company.18 Both specified the granting of 1,000 marks (£666 13s 4d) to John Parker of Cheshunt, who was the king’s master of ordnance “under” Salisbury, for the purchase of artillery and other necessities.19 The particulars of account in E101/51/27 state that such equipment was for the king’s forts, towns, castles and fortifications in France, but in the context it appears that they were intended for a campaign of conquest, possibly to make up for losses of artillery in the failed siege of Montargis, which had occurred in the summer of 1427.20 On 28 April a commission was given to John Parker to take and provide cannons, stones, carts and other necessities for the earl of Salisbury to take overseas.21 On 16 May two cannons were purchased for £4 4s, and two days later Parker took receipt of the 1,000 marks granted to him by the Exchequer.22 Over three days at the beginning of June (1, 2 and 4 June) payments totalling £177 19s 4d were made by Parker out of this sum for the construction of three large cannons and seventeen large carts.23 On 16 June a mandate was granted to John Hunt and William Rider, bailiffs of the city of London, to arrest the George of London and other ships to conduct artillery to Normandy. Eight days later charcoal was purchased at the price of £9 11s 4d for the forging of cannon.24 On 6 July Parker was paid £66 7s 9d for shipping expenses and an indenture was made with the master and owner of the George of London.25 On the same day Parker made payments totalling £205 6s 2d for a large cart, gunpowder and four large cannons.26 Further payments followed on 8 July for forty-six cannons, sixteen hand-cannons (canonis manualibus), cart wheels, axles, one large sled, brass pulleys, iron and 1,214 cannon-stones, totalling £214 11s 8½d. Costs included the carriage of the latter to the Tower of London for subsequent loading onto ships.27 On the same day a commission was granted to William Rider and John Exham to muster “those about to cross the sea with the king’s ordnance, and to certify the king and council of the sufficiency of their array.”28 The George of London was indented to transport artillery from 10 August to 8 September to Harfleur in Normandy.29 Four further ships, the George of Baldesey, the Trinity 17

18 19 20 21 22 23 24 25 26 27 28 29

The National Archives, Kew [hereafter TNA], E101/71/2/28, printed and translated in Letters and Papers Illustrative of the Wars of the English in France during the Reign of Henry the Sixth, 1:402–14; Curry, “English Armies in the Fifteenth Century,” pp. 46–47. TNA, E404/44/336. TNA, E101/51/30. TNA, E101/51/27 m.2. CPR 1422–29, p. 493. TNA, E101/51/27 m.2. TNA, E101/51/27 m.2. TNA, E404/44 f.321; E101/51/27 m.3. TNA, E101/51/27 m.2, 3. TNA, E101/51/27 m.2. TNA, E101/51/27 m.2, 3. CPR 1422–29, p. 499. TNA, E101/51/27 m.4.



The 1428 Expedition to France

183

of Gravesend, the Janys Kynte and the Trinity of Norwich, were indented on 11 August to transport artillery between 10 July and 20 September to the city of Rouen.30 Thus, within six months of Salisbury’s indenture, the artillery and all other equipment had been constructed, purchased and shipped to Normandy. This demonstrates that artillery could be assembled and transported relatively quickly to France. One large bombard (grosse bombarde) with stones and carts was received by William Appleby, master of the ordnance in Normandy; this was likely to have been transported in the George of London and sent to the arsenal at Harfleur.31 The vast majority of the ordnance, however, was sent to Philebert de Moulins, master of the duke of Bedford’s artillery, in the city of Rouen.32 From here it is likely that the ordnance was shipped at de Moulins’ order up the river Seine to Paris, and subsequently carted overland with Salisbury’s troops, which had crossed from Sandwich to Calais and then moved to the Loire via Paris. Salisbury began his active campaign in mid-July. By September he claimed, in a letter, to have captured forty towns and to have begun the investment of the city of Orléans itself.33 The artillery played a prominent role in the bombardment of Orléans, which began on 17 October and by 24 October had resulted in the capture of the tower of the Tourelles. The army’s ordnance was subsequently deployed in boulevards which were constructed around the perimeter of the town, and was fired at the city throughout the course of the siege. It was also used defensively to protect these fortifications from French attacks. Much of this artillery appears to have been captured by the French following the lifting of the siege of Orléans in 1429.34 The ordnance In all, the particulars of account list the construction or purchase of seventyone cannons, consisting of seven large cannons (magnis canonis), forty-eight fowlers (ffoulers) and sixteen hand-cannons (canonis manualibus).35 Direct payments for artillery comprised £471 17s 8½d. Of this, £346 5s 8d was spent on the construction of the seven bombards. The remaining £125 12s½d, was used to purchase sixty-four ready-made cannons. The seven bombards collec30 31

32 33 34 35

TNA, E101/51/27 m. 4; E101/51/30 f.2, 7, 8, 10, 12. TNA, E101/52/3; C. T. Allmand, “L’artillerie de l’armée Anglaise et son organisation à l’époque de Jeanne d’Arc,” in Jeanne d’Arc, une Époque, un Rayonnement , ed. R. Pernoud (Paris, 1982), p. 76. TNA, E101/52/3. Warner, “The Montagu Earls,”, pp. 191–93. DeVries, Joan of Arc, pp. 59–62, 91; Kelly DeVries, “The Use of Gunpowder Weaponry by and against Joan of Arc during the Hundred Years War,” War and Society 14 (1996), 9–11. TNA, E101/51/27 m.2, 3. The “large cannons” are elsewhere described as “bombardes” in TNA, E101/52/3 and in this paper are hereafter described as bombards. For definitions of bombards and fowlers (the latter were also known as veuglaires), see Smith and DeVries, The Artillery of the Dukes of Burgundy 1363–1477, pp. 204–11, 230–36.

184

Dan Spencer

tively weighed 35,650 lbs, whereas the sixty-four fowlers and hand-cannons weighed in total 6,465 lbs. This illustrates the difference in the weight of the bombards and the other cannons, which was substantial. The average weight of each bombard is 5,093 lbs, whereas the average weight of the other sixty-four cannons is 101 lbs. The great disparity in weight can also be seen with the sixteen fowlers of 1428, which are described as being 1½ feet in length and which fired cannon-shots which weighed 2 lbs, whereas no measurements are given for the seven bombards. The bombards of 1428 also represent an increase over the weight of the cannons which had been available to Henry V in 1415, such as the “Messenger,” at 4,480 lbs, and the “George,” at 3,678 lbs.36 Whilst the bombards were intended primarily for breaking down defences, the fowlers, with their light, 2 lb cannonstones, were not designed for this purpose. The fowler was a small type of cannon which is first recorded in English sources in 1419.37 Forty of the fowlers were provided with three chambers, and three with two chambers. This would have allowed them to maintain a high rate of fire and therefore to be effective in an anti-personnel role.38 The sixteen hand-cannons are likely to have been used in a similar fashion. The weights of the hand-cannons are not provided, but they fired lead shot and each had two chambers. The documents for the 1428 expedition also provide evidence for the purchase and construction of the guns. The sixty-four smaller bronze guns (i.e. both the fowlers and the hand-cannons) were purchased from two men described as citizens and (metal) founders of London, Thomas Coston and Robert Warner. Three of the bombards were constructed by a smith, John Matthew of London; of these, one fired shot eighteen inches in diameter, and two fired shot sixteen inches in diameter. These weapons were tested with 80 lbs of gunpowder, by the same John Matthew, to ensure that they did not burst under the strain of firing. Construction work, by the gunner Walter Thomasson at his house in Southwark, was started on three other bombards, one firing stones of eighteen inches in diameter, another of sixteen inches in diameter and a third of fourteen inches in diameter. These new bombards were shipped over to Rouen, where the construction was finished by another gunner, Godfrey Goykyn. Goykyn, in addition, constructed another bombard firing stones sixteen inches in diameter. These bombards were forged using iron from the stores of the king, which were supplemented by additional purchases from Richard Warbelton and Thomas

36 37 38

TNA, E364/49 m.3d. TNA, E403/642 m.3, where they are referred to as “ffowlers”; also see E101/188/1 for a slightly later example. For more information on gun chambers, see Kelly DeVries and Robert Douglas Smith, “Breech-loading Guns with Removable Powder Chambers: A Long-lived Military Technology,” in Gunpowder, Explosives and the State: A Technological History, ed. Brenda J. Buchanan (Aldershot, 2006), pp. 251–65.



The 1428 Expedition to France

185

Halyday. The latter appears to have been based in Rouen and was entrusted with the custody of the iron left over from the forging of the bombards.39 Efforts were also made to weigh these cannons, as their price was determined by weight. Four of the bombards were weighed at the crane at Bolbroke Wharf by a John Vincent. It is unclear when the other three bombards were weighed, but these possibly correspond to the three that Goykyn finished in Rouen, so they may have been weighed there. Weights were cited for only two of the other cannons, two bronze fowlers, which implies that such information was less important to be recorded for the smaller cannons. Accessories and personnel for the construction of the ordnance The accounts also describe the payments made for accessories for the artillery train.40 This included materials such as charcoal, wood and iron, which were used for the construction of cannons, carts and tools; as well as tools such as twelve felling axes and lead mallets purchased from London smiths. Additional items included thirty-six brass pulleys, cables and bolts of iron, provided by William Pikworth, citizen and (metal) founder of London for the price of £7 8s 4d. It is likely that the pulleys and cables were used to move the cannons, particularly the heavier bombards, either onto or from ships or carts. One large cart for the movement of large cannons was purchased from two wheelwrights, Robert Redelwot and Thomas Whitby, for 20s 6d. Eighteen wheels for this cart, and others, were shipped to Rouen, where they were attached in the forge of William Rantost. This was the same forge where Goykyn finished the three bombards. Axles and ropes were also purchased for carts, the latter for fastening and hauling cannons and other equipment. A large sled of wood was also acquired for temporarily moving cannons from Towerhill to Towerwharf, for 3s 7d. In addition, purchases were made for 1,214 cannon-stones, of which seventy-four were twenty-four inches in diameter, 245 were eighteen inches in diameter, 260 were sixteen inches in diameter, 435 were fourteen inches in diameter and 200 weighed 2 lbs. These were purchased from five masons for £78 11s 4d, who appear to have used stone from Maidstone. These quantities would suggest that a significant proportion of this ammunition was intended for other cannons in France, as, for instance, none of the seventy-one cannons in the accounts fired cannon-stones of twenty-four inches in diameter. This may imply that Lancastrian France was dependent, at least in part, on supplies of ammunition from England for its ordnance. The fowlers, firing cannon-stones of 2 lbs in weight, also seem to have been underprovided with ammunition, most likely because the smaller cannon-stones could be made fairly easily by masons at the site of sieges. The sixteen hand-cannons were also provided with 1,200 lead shot at the price of £10. The accounts specify only the purchase of 180 lbs

39 40

TNA, E101/52/3. TNA, E101/51/27 m.2, 3.

186

Dan Spencer

of gunpowder for £6 from Gerard “Ducheman,” 80 lbs of which was expended on testing three of the bombards before departure. This suggests that the actual gunpowder for use in the field was acquired in France itself, as England was reliant on imports of saltpetre from saltpetre farms in Germany.41 It is clear that a large number of individuals were involved in the construction and transportation of the ordnance. The particulars of account list 143 individuals by name, consisting of: one knight, one ladder-maker, one joiner, one timbermonger, two gunners, two officials, three (metal) founders, three merchants, three wheelers, four unspecified, five masons, six smiths, six labourers, seven carpenters and ninety-eight sailors in four ships (although the receipts for the account also include a further thirteen sailors in one ship). This demonstrates that the assembling of an artillery train was a significant undertaking which necessitated the employment of a diverse range of tradesmen. Ordnance Company In addition, there is evidence of a support or ordnance company for the expedition. The earl of Salisbury’s indenture, in addition to 2,400 men-at-arms and archers, specifies the employment of a group of eighty tradesmen consisting of carpenters, masons, fletchers and others.42 The particulars of account include payments made as a “reward” for a master of the works and carpenters, two carpenters, and another man who appears to have been master of the labourers.43 This suggests that at least some of these men were hired in England, but, as was often the case with ordnance companies for the English in France, a significant proportion of the final company is likely to have been of French ethnicity.44 Salisbury’s indenture also includes terms for four master cannoneers (the equivalent English term is master gunner); these were each to receive 40d per day, 12d for their own wages and the remaining 27d to pay for the wages of their three archers each receiving 9d per day, making twelve archers in all. It is possible that these archers were intended as escorts for the cannoneers, but their high rate of pay of 9d per day (archers normally received 6d per day), suggests that they were assistants to the cannoneers. This would suggest that the ordnance company for Salisbury’s artillery train of seventyone cannons consisted of ninety-six men, comprising four master cannoneers, twelve assistants, and eighty carpenters, masons, fletchers, labourers and others. The ordnance company of 1428 was mustered at Gravesend in early July and appears to have accompanied the artillery to Normandy between August and September; whereas Salisbury’s main army embarked from Sandwich in July.45

41 42 43 44 45

David Cressy, Saltpeter: The Mother of Gunpowder (Oxford, 2013), p. 45. TNA, E101/71/2. TNA, E101/51/27 m.4. Allmand, “L’artillerie de l’armée Anglaise,” p. 77. CPR 1422–29, p. 499; TNA, E101/51/27 m.4; E101/71/2.



The 1428 Expedition to France

187

The most significant individual in the ordnance company appears to have been Godfrey Goykyn, a gunner from the Low Countries. Goykyn had been responsible for the construction of one bombard and for finishing three further bombards in Rouen. He also appears to have been one of the four master cannoneers specified in Salisbury’s indenture. The particulars of account include a payment of 10s given to him, which at the rate of 40d a day works out at three days’ wages for himself in addition to three archers or assistants. This payment is likely to have been issued during the muster at Gravesend. Goykyn was not merely a skilled craftsman capable of constructing large artillery but also served with the expedition itself. He also had extensive military experience in the service of the English. He had served with Henry V’s 1415 expedition, and so had participated in the siege of Harfleur.46 Thereafter he appears to have been part of the garrison of Calais, as the treasurer’s accounts for the town for 1413–17 include the receipt of 1,862 lbs of gunpowder from Godfrey Goykyn and John Bedford “gunners of the king in the town of Calais.”47 The important role he occupied at Calais can be seen by the treasurer’s accounts for 1421–24, which make mention of a great cannon of iron called “Goykyn.”48 Goykyn’s growing stature can be seen by his inclusion in the enrolled accounts for William Philipp, treasurer of war to Henry V, for the years 1421– 22.49 These accounts specify that Sir William Wolf was made responsible for organizing a team of workers in the construction of a palace for the king in Rouen. Goykyn was the only gunner included, at the rate of 12d per day for 380 days, and seems to have been responsible for overseeing the work of sixteen smiths, each at the rate of 6d per day for 200 days, in the construction of artillery for the king. In a warrant for issue of 1423 Goykyn was described as one of four German “gunnemestres” who received part payment of arrears for royal service.50 Following the siege of Orléans he remained in English service, as can be seen by his sale for £56 10s of one large cannon of iron called “The Henry,” weighing 6,780 lbs, in Calais on 27 May 1430.51 This example suggests that the four master cannoneers in the expedition were experienced professionals with highly valuable skills. The names of the other three master cannoneers are unknown, but they are likely to have been French or German.52 Research carried out by Andy King on the ordnance companies of Normandy for the 1430s and 1440s indicates that master gunners rarely stayed in English service for long periods of time.53 This suggests that it was intended that the ordnance company for Salisbury’s army should be made up of the most highly skilled individuals 46 47 48 49 50 51 52 53

TNA, E101/69/8 f.511. TNA, E364/69 m.12d. TNA, E101/188/1. TNA, E364/69 m.7. TNA, E404/40 f.128. TNA, E101/52/22. Allmand, “L’artillerie de l’armée Anglaise,” p. 77. Andy King, “Gunners, Aides and Archers: The Personnel of the English Ordnance Companies in Normandy in the Fifteenth Century,” Journal of Medieval Military History 9 (2011), p. 74.

188

Dan Spencer

available, which is why John Parker recruited perhaps the most experienced gunner in English service for the expedition. Transportation and Shipping The accounts also provide information for the transportation and shipping of the ordnance. It appears that four of the large cannons were moved on a wooden sled from Tower Hill to Tower Wharf. Tower Hill is likely to have been linked to the arsenal at the Tower of London and this is possibly where some of the cannons were tested by the maker of three of them, John Matthew. From Tower Wharf they were conducted by labourers to Bolbroke Wharf, where they were weighed by John Vincent. They were then shipped to Rouen, in the custody of Godfrey Goykyn and John Merch, master of the works, and carpenters. The three unfinished bombards were conducted from the house of Walter Thomasson in Southwark to Bridge Wharf, where they were shipped to France. The accounts also imply that additional cannons, which were not purchased for the expedition, were shipped to France at this time. There is mention of a large cannon conducted from the house of John Cornwall, knight, to Billyngesgate, and of a large cannon of Raby being transferred from one ship to another at “Brianestrane.” The cannons and other equipment were transported from London to Normandy in five ships, over the course of July to September. These were the George of London (forty-two sailors), the George of Baldesey (twenty sailors), the Janys Kynte (twenty sailors), the Trinity of Gravesend (sixteen sailors) and the Trinity of Norwich (thirteen sailors). Three of these ships are described as being crayers and one as a cog (the description of the George of London is illegible). This suggests that they were relatively small craft.54 As mentioned before, carts were purchased for moving the cannons. This included one great cart and seventeen other carts. The terms of Parker’s commission to “take and provide” carts and cannons, given on 28 April, also includes the possibility that additional carts were requisitioned for the expedition which were not included in the particulars of accounts.55 The accounts provide no information on the carters who would have moved the cannons, nor on the animals they used; possibly these were provided for the artillery train in France itself. Evidence from the Burgundian archives suggests that considerable resources are likely to have been required. Of three bombards transported in 1436 for the siege of Calais, one required eighteen horses, another thirty and the third needed eighty-four horses.56 The cannons for the 1428 expedition lacked

54

55 56

TNA, E101/51/30; Maryanne Kowaleski, “Warfare, Shipping, and Crown Patronage: The Impact of the Hundred Years War on the Port Towns of Medieval England,” in Money, Markets and Trade in Late Medieval Europe: Essays in Honour of John H. A. Munro, eds. Lawrin Armstrong, Ivana Elbl and Martin M. Elbl (Leiden and Boston, 2007), p. 249. CPR 1422–29, p. 493. Smith and DeVries, The Artillery of the Dukes of Burgundy 1363–1477, pp. 208–9.



The 1428 Expedition to France

189

wheeled carriages, as they were moved in carts; this meant that they would have needed be transferred from their carts to a wooden frame, a process often referred to as “stocking.” The purchase of the thirty-six brass pulleys is likely to have been for this purpose, or for the loading of cannon-stones. Conclusion The preceding discussion has shown how the artillery for the 1428 expedition to France was assembled and transported to Normandy within six months. This highlights the importance of London in the manufacture of gunpowder weapons and accessories. The needs of the expedition involved the services of a variety of different tradesmen in London, with very little purchased from the regions, with the exception of some lead from a merchant of York. This contrasts, for instance, with the expedition of Henry V in 1415, which involved the construction of a large bombard at Bristol.57 The accounts also provide an insight into the relationship between England and Normandy in the provision of gunpowder weapons. It supports the argument that Lancastrian Normandy was dependent on England for its artillery; however, it also shows that facilities did exist at Rouen for the manufacture, or at least the finishing, of bombards.58 The documents also indicate that important supplies for the expedition were obtained in Normandy, such as gunpowder, trestles and some of the cannon-stones. This was also the case with many of the personnel for the ordnance. By 1428 gunnery was an established profession, with skilled and experienced individuals such as Godfrey Goykyn, many of them of Continental origin. However, the campaign represented a new departure for the conduct of the war in France. It was the first expedition since Henry V’s conquest of Normandy in 1417–19 to rely principally on English as opposed to Lancastrian French resources for the provision of artillery. This can be seen by the relatively few entries in the Issue Rolls of the Exchequer for expenditure on guns in England during the years 1422 to 1427.59 Prior to 1428 the resources of Normandy were used to produce large numbers of cannons for the war in France. This can be seen from the work carried out at a palace in Rouen by Godfrey Goykyn, as specified in the accounts of Henry V’s treasurer for war in 1421–22.60 Goykyn was paid for 380 days, at least in part to oversee the work of sixteen smiths paid 200 days for the construction of artillery. The large numbers of workers employed and the length of their employment suggests that a substantial number of cannons were forged. Artillery continued to be constructed in Normandy after 1428, as can be seen from the estimates for military expenditure from October

57 58 59 60

TNA, E403/614 m.15. Anne Curry, “Guns and Goddams. Was there a Military Revolution in Lancastrian Normandy, 1415–50?” Journal of Medieval Military History 8 (2010), 183–4. For expenditure on guns in 1422 and 1423, see TNA, E403/655 m.7, m.18; E403/660 m.10. TNA, E364/69 m.7.

190

Dan Spencer

1433 to September 1434.61 Few inventories survive, but some idea of the artillery available in Normandy can be seen from the inventory compiled in Rouen castle following the death of the duke of Bedford in 1435, which included one bombard and nine fowlers.62 However, future expeditions from England to France, such as in 1430–32 and 1443, followed the trend of 1428 by obtaining a significant proportion of their artillery from England, as opposed to Lancastrian France.63 This is likely to have had the long-term effect of encouraging the development of the armaments industry in England, particularly in London. As this is one of only two sets of artillery accounts to survive for the first half of the fifteenth century, the other being for the 1430–32 expedition, it is relatively difficult to compare the numbers of cannons used in 1428 with those of other English armies. For the coronation expedition of 1430–32, John Hampton, master of the ordnance, was allocated the large sum of £2,200 for making purchases, more than three times the £666 13s 4d allocated to John Parker in 1428.64 However, fewer cannons were constructed in England specifically for this campaign, consisting of three bombards called “Henry,” “Hampton” and “Crown” in addition to twenty fowlers. Instead, significant quantities of supplies and weaponry were supplied from Lancastrian France, including three bombards called “Fflourdelice,” “Newgate” and “Towerwharf of Cumbria,” nineteen fowlers and fourteen organ guns.65 The reason for the much higher expenditure in 1430 was due to the purchase in England of large quantities of gunpowder, saltpetre and sulphur for the expedition.66 By way of comparison, the duke of Somerset, for his expedition to France in 1443, was allocated a total of £1,100 for the construction of ordnance; this implies that the numbers of cannons sent from England in 1428 were relatively low by the standards of some of the later expeditions.67 This was undoubtedly due, at least in part, to the differences in the sizes of the expeditions: Somerset’s army consisted of a total of 4,549 men, whereas Salisbury’s army in 1428 consisted of 2,694 men.68 It also appears to be a reflection of the intensification of the war in France. The resurgent French threat in the 1430s and 1440s required greater resources from England to support Lancastrian France.69 The number of cannons to men in the expedition seems high, with seventyone cannons for 2,694 men-at-arms and archers, together with ninety-six men 61 62 63 64 65 66

67 68 69

Letters and Papers Illustrative of the Wars of the English in France during the Reign of Henry the Sixth, 2.1:559–60. Ibid., 2.1:565–68. Curry, “Guns and Goddams,” p. 184; TNA, E404/59 fols.162, 247. Curry, “Guns and Goddams,” p. 184. TNA, E364/69 m.17, 17.d, 18. Consisting of 1,204 lbs of gunpowder, 5,164 lbs saltpetre and 5,200 lbs of sulphur purchased in England, in addition to 8,642 lbs of gunpowder, 8,035 lbs of saltpetre and 6,320 lbs of sulphur obtained from Lancastrian France, TNA, E364/69 m.17, m.17.d, 18. TNA, E404/59 f.162, 247. Curry, “English Armies,” p. 47. Ibid., p. 45; Pollard, John Talbot and the War in France, pp. 32–33.



The 1428 Expedition to France

191

in the ordnance company; a proportion of one cannon to thirty-nine men.70 The accounts also indicate that the cannons were provided for sieges, as opposed to use in the field. The lengthy process of transferring the cannons from their carts to frames and trestles meant that they could not have been deployed quickly. These were also essentially the same type of cannons used by Henry V earlier in the fifteenth century.71 This means that the significant changes in gunpowder weapons that occurred during the course of the fifteenth century, as revealed by the accounts for Edward IV’s 1475 expedition to France, had not yet occurred by 1428.72 However, the documents for the 1428 expedition show that artillery was fundamental to the conduct of the war in France by this date.

Appendix Selected passage from E101/51/27 (translated from Latin) Expenditure on cannons and the ordnance: In the same account, 14 perins canonis of bronze called ffoulers, firing stones weighing 2 lbs, each of which are 1½ feet in length and each with three chambers. 3 other canonis of bronze of various sorts, each with one chamber. 29 other canonis of various sorts, 26 of which with three chambers and three of which with two chambers. 16 canonis manualibus firing lead shot, each with 2 chambers. (All cannons) weighing in total 6,297 lbs, price per lb 4½d. 1,200 lead shot for the same canonis manualibus hand-cannons, price per piece 2d. Purchased of Thomas Coston, citizen and founder of London, on the 8th day of July in the 6th year of King Henry VI the sum of £121 8s ½d And of 2 canonis of bronze called ffoulers, firing stones weighing 2lbs, each of which are 1½ feet in length, which in total weigh 168 lbs, price per lb 6d. Purchased of Robert Warner, citizen and founder of London, on the 16th day of May the same year the sum of £4 4s. And of 3 large canonis of iron, of which one fires stones 18 inches in diameter, and 2 which fire stones 16 inches in diameter, weighing in total 15,500 lbs, price 70 71 72

Curry, “English Armies,” p. 47. TNA, E364/49 m.3. TNA, E101/55/7.

192

Dan Spencer

per 100 weight of iron, 18s 4d. Purchased of John Matthew of London, smith, on the 6th day of July of the same year the sum of £142 20d. And of 1 large canona, which fires stones 16 inches in diameter, weighing 5,350 lbs, price per 100 weight of iron, 20s, 12 bars of iron purchased for the same, price the piece 4s 6d. Purchased of Godfrey Goykyn, gunner, on the 6th day of July the same year the sum of £56 4s. And of 3 other canonis of iron, of which one fires stones 18 inches in diameter, one of which fires stones 16 inches in diameter, and one of which fires stones 14 inches in diameter, with one bar weighing in total 14,800 lbs, price the hundred weight 20s. Purchased of Walter Thomasson, gunner, and afterwards by Godfrey Goykyn, gunner, on the 4th day of June the same year the sum of £148.

9 1471: The Year of Three Battles and English Gunpowder Artillery1 Devin Fields

In the second half of the fifteenth century gunpowder began to change the face of warfare. While the impact of gunpowder has often been cited on the Continent, historians have failed until recently to give proper consideration to the role of gunpowder weapons in England’s dynastic struggle known as the Wars of the Roses. John Gillingham, for example, has stated that “for all the growing importance of artillery, it remains true that this arm played only a minor role in the campaigns of the Wars of the Roses.”2 He maintained that the reason for the failure of the English to use gunpowder effectively was the lack of sieges in the extended conflict. Anthony Goodman believed that English strategy of the period was too focused on bringing the enemy to battle quickly and the armies were too hastily organized to allow the creation of a large siege train or the use of artillery.3 Kelly DeVries delivered a paper in 2001 which has since been published, showing the wide use of gunpowder artillery in local conflict by nobles, towns and even religious houses. Yet he argued that in only one instance did gunpowder weapons affect the outcome of the fighting in the Wars of the Roses.4 Conclusions such as those of Gillingham, Goodman and DeVries are beginning to be eroded because the evidence suggests that, in reality, English gunpowder weapons were as numerous and advanced as those on the Continent. David Grummitt examined the history of the Calais garrison, the largest professional military establishment in England during the Wars of the Roses. 1

2 3 4

There are a few people whom I must thank, especially since this is a first article. The first are the board of editors of the Journal of Medieval Military History and the reviewers, whose comments greatly improved the finished product. I must also thank James King, who served as my advisor and the chair of my MA thesis. That project provided much of the foundation for this article. Finally, John Buesterian, who allowed me to take this project and prepare it for publication as an article in a research methods course at Texas Tech University. John Gillingham, The Wars of the Roses: Peace and Conflict in Fifteenth Century England (Baton Rouge, 1981), p. 27. Anthony Goodman, The Wars of the Roses: Military Activity and English Society, 1452–1497 (London, 1981), p. 171. Kelly DeVries, “Gunpowder Weapons in the Wars of the Roses,” in Traditions and Transformations in Late Medieval England, ed. Douglas Biggs, Sharon D. Michalove and A. Compton Reeves (Boston, 2002), p. 31.

194

Devin Fields

He ultimately determined that the Calais garrison served as the conduit for technologies and tactics into the English military establishment.5 Striking evidence has also been provided by the survey carried out on the battlefield at Bosworth. This survey found lead shot in such abundance that the survey team shifted the location of the battle several kilometers away from where many historians believed the majority of the fighting had taken place. Foard speculated in 2009 that perhaps the largest gathering of artillery in the British Isles occurred at Bosworth.6 However, in the final published volume on Bosworth, Foard was more circumspect, arguing that the fifteenth century, and in particular the Wars of the Roses, “may span a key transition in the battlefield use of gunpowder weapons – at least so far as artillery is concerned.”7 There has been a considerable re-evaluation of English gunpowder weapons in the late fifteenth century. While these weapons are now recognized as having been more numerous and sophisticated, no one has proposed an occasion in the Wars of the Roses where events were affected by gunpowder weapons used to actually harm the enemy. This article will take the new information on gunpowder weapons and re-examine the fighting that occurred in 1471, the year when Edward IV returned from the Continent and defeated the Lancastrian dynasty. This year included three battles and is documented remarkably well for the Wars of the Roses.8 The battles of Barnet, Tewkesbury and London represent a sophisticated use of gunpowder weapons which directly influenced the outcome of the Wars of the Roses. In 1470 Edward was driven from England after losing the support of Richard Neville, the earl of Warwick. Warwick offered his support to the Lancastrians and Edward was forced to flee to Burgundy. In the spring of 1471 Edward returned to England with the support of Charles the Bold, the duke of Burgundy. The battle of Barnet, fought on 14 April 1471, was the first battle that he fought to regain the throne of England. At Barnet, Edward faced down the earl of Warwick and several of the leading Lancastrian nobles who had recently joined him. It was one of the most decisive battles of the Wars of the Roses, with both Warwick and his brother dying on the field. It is also one the most welldocumented battles. It was recorded in several contemporary chronicles, and

5 6

7 8

David Grummitt, The Calais Garrison: War and Military Service in England 1436–1558 (Woodbridge, 2008), pp. 139–40. “Update on Battle of Bosworth Project,” statement by Glenn Foard (2009), http://www. battlefieldstrust.com/resource-centre/warsoftheroses/battlepageview.asp?pageid=824 [accessed 24 February 2012]. Glen Foard and Anne Curry, Bosworth 1485: A Battlefield Rediscovered (Oxford, 2013), pp. 196–97. Historie of the Arrivall of Edward IV, in England and the Finall Recouerye of His Kingdomes from Henry VI. A.D. M.CCCC.LXI., ed. John Bruce (London, 1838), p. 18. John Warkworth, A Chronicle of the First Thirteen Years of the Reign of King Edward the Fourth, ed. James Orchard Halliwell (London, 1834), p. 16.



English Gunpowder Artillery

195

there are also two letters which provide details about the battle.9 Gunpowder weapons figure in these sources. The comparative strength of the sources and the fact this was first of Edward’s battles fought in 1471 make it the first battle to receive attention here. In 1470, Edward IV was driven into exile by an alliance of the Nevilles, Lancastrians and Edward’s own brother. As already noted, while in exile, Edward gained the support of his brother-in-law, Charles the Bold, duke of Burgundy. With the financial and military of support of Burgundy and the English merchant community in the Low Countries, Edward returned to England.10 Warwick and the Lancastrians were not completely unprepared and Edward was forced to sail further north than he had planned, in an effort to find a safe place to land with his small army and to rally his remaining followers. He was finally able to make landfall in April 1471, somewhat poetically at Ravenspur, where Henry Bolingbroke had landed in 1399 when he returned from exile and deposed Richard II.11 As Edward marched south under the pretense that he wanted only the lands he should have inherited from his father. This claim held up long enough that the Lancastrian resistance in the north allowed him to march to Nottingham without bloodshed.12 By a clever night march, Edward was able to disperse two nearby columns of Lancastrian troops, giving him a further advantage against Warwick, who was trying to gather an army in the Midlands.13 By the time he reached Leicester, Edward’s army numbered around five thousand men and he felt comfortable enough to openly claim that he had returned to regain the throne.14 The earl of Warwick had gathered his forces at Coventry; sources differ on the number of men he had with him, but he probably had a force which was 9

10 11

12 13 14

The chronicles of prime importance for the battle of Barnet are: Warkworth, Chronicle, and Historie of the Arrivall of Edward IV, known simply known as the Arrivall. Phillipe de Commynes, Memoirs: The Reign of Louis XI 1461–83 (Baltimore, 1972). The letters are one from the Hanseatic merchant Gerhard van Wesel, which is available in John Adair, “The Newsletter of Gerhard von Wesel, 17 April 1471,” Journal of the Society for Army Historical Research 46 (1968), 65–69; and another from Margaret, Duchess of Burgundy, Edward’s sister, which can be found preserved in Jean De Waurin, Anchiennes Chroniques D’Engleterre 3 (Paris, 1863), 3:210–13. Commynes, Memoirs, pp. 193–94. Peter W. Hammond, The Battles of Barnet and Tewkesbury (New York, 1990), pp. 54–55. Henry Bolingbroke was later crowned Henry IV and was the grandfather of Henry VI, the Lancastrian claimant who had been placed back on the throne. Arrivall, p. 2; Hammond, Battles of Barnet and Tewkesbury, p. 56. Arrivall, p. 7. Warkworth, Chronicle, p. 14. Hammond, Battles of Barnet and Tewkesbury, p. 63. Estimate of armies is always somewhat difficult. The Arrivall tracks pretty well the large parties which join Edward on his march. The size of the army that Edward left to Burgundy with seems reasonable. In total the numbers come to 5,600 men; to account for the weakness of estimates, I have lowered my estimate here to 5,000. It is possible that Edward had fewer than five thousand men, but it is also possible he had more. I prefer to give the author of the Arrivall the benefit of the doubt with his estimate here, since he neither makes Edward appear too strong nor too weak, which suggests that while he may be writing for Edward he is conveying the truth. Arrivall, pp. 1, 7–8.

196

Devin Fields

equivalent in size to Edward’s army.15 Edward then attempted to bring Warwick to battle. He advanced to Coventry and challenged Warwick. However, Warwick was a cautious commander and withdrew his men inside the walls of Coventry. Warwick expected more men to arrive, and that when they did so he and his reinforcements would outnumber Edward’s army and be able to defeat him.16 The nearest army from which Warwick could expect support was the forces of the Earl of Clarence.17 However, Warwick had made a terrible mistake in trusting Clarence. Edward had worked to heal the breach with his brother, using his mother and sisters as intermediaries., Clarence met Edward outside Coventry and worked out an agreement to rejoin his brother. Warwick was dealt a double blow, since Clarence had been one of seven men trusted to recruit the army that would fight Edward. The four thousand men that he had recruited, instead of fighting for Warwick and Henry VI, joined Edward.18 With the situation around Coventry now in his favor and with Warwick refusing to give battle, Edward was faced with a choice. Besieging Coventry to get at Warwick would be time consuming and would have turned into a disaster, since Coventry was not an enemy fortress but a city which had demonstrated reluctance to support either side. The Wars of the Roses were noted for very little violence against civilian populations because neither side wanted to alienate populations on whom they would rely for support after they had won. When violence did occur, it could turn the population against one of the parties, as it did in 1460 when the Lancastrian garrison in the Tower of London turned its guns on the city.19 Edward decided instead to bypass Coventry and marched on London. Warwick left quickly to try to catch Edward before he could get to London; but his artillery, which is mentioned in the Arrivall, probably slowed him down, ensuring that Edward would arrive in London with time to spare.20 When Edward entered London he captured Henry VI, the Lancastrian claimant to the throne, and released the prisoners whom Warwick had taken to keep them from supporting Edward. This added perhaps as many as another thousand fighting men to his army. While Edward’s gamble had paid off, he was far from secure: Warwick’s army was still in the field. In addition to the threat of Warwick’s forces, Henry VI’s wife, Margaret, was expected to arrive from France with an army of her own and with Henry VI’s heir, Prince Edward, who was old enough to fight. With the two armies threatening Edward, he needed to act quickly. He took his army, leaving a token force behind to hold London, and left to catch Warwick on the march. As Edward marched north, his outriders ran into Warwick’s army in the town of Barnet, ten miles from London. After a 15 16 17 18 19 20

Warkworth reports that Warwick had four thousand men. The Arrivall states that Warwick had six or seven thousand. Warkworth, Chronicle, p. 15; Arrivall, p. 9. Warkworth, Chronicle, p. 14. The Earl of Clarence was one of Edward’s younger brothers. Warkworth, Chronicle, p. 14. Arrivall, pp. 9–11. DeVries, “Gunpowder Weapons,” p. 36. Arrivall, p. 13.



English Gunpowder Artillery

197

brief skirmish by the scouts, Warwick’s men fell back to his main force, which was just north of Barnet. Edward’s main force began to arrive in Barnet at the end of the day, but he was eager to force Warwick to battle; so under the cover of night, he marched his army out of Barnet to within a short distance of Warwick’s army.21 The author of the Arrivall estimated Warwick’s army at thirty thousand. We can be pretty sure that this was an exaggeration. Edward’s brother Clarence, had been working with the Lancastrians and Warwick before he joined his brother in the Midlands, and the Arrivall reports that he brought only a few thousand men with him. Warwick trusted few of the nobility with the recruitment of an army to fight Edward. Clarence was one of them, and so he undoubtedly had sizable proportion of the men recruited by Warwick.22 P. W. Hammond, in his widely regarded book on the events of Warwick’s rebellion against Edward, estimated the size of the armies at about twelve thousand for Edward and fifteen thousand for Warwick.23 The Arrivall also acknowledged that Warwick had managed to gather more guns than Edward had, which he would try to use to his advantage in the early stages of the battle by firing on the Yorkist men while they rested the night before the battle. Fortunately for Edward’s men, the artillery fire passed over their heads and they were spared.24 Warwick had placed his men along the St. Alban’s High Road. The position roughly corresponded to the area of Hadley Green. Under the cover of darkness, Edward had deployed his army opposite to Warwick’s, close enough that Warwick had to give battle. In the dark Edward could not be sure of the of the exact location of the enemy. This led to a misalignment of opposing forces, with the right flanks of either army extending further than the enemy opposite to them. Edward was eager to bring Warwick to battle the next day, especially after Warwick had earlier declined battle at Coventry. He ordered his men to remain quiet the whole night and not to light any fires. This would have allowed the Yorkists to get as close as possible to the Lancastrian forces without revealing themselves. Warwick seems to have known the Edward was deploying somewhere in front of him, because he ordered his artillery to fire at where he thought Edward was deployed. Warwick’s gunners fired into the night, but because of Edward’s precautions all of their fire passed over the heads of the Yorkist army.25 Early the next morning, Edward ordered his men to attack under the cover of fog that had settled overnight, and the battle quickly became a confused affair. Edward’s right flank, under the command of his brother, Richard, seemed to have become lost in the fog. His left flank, meanwhile, was destroyed by an attack by the earl of Oxford, who commanded Warwick’s right flank. Oxford’s 21 22 23 24 25

Ibid., p. 18. Ibid., p. 20. Hammond, Battles of Barnet and Tewkesbury, p. 74. Arrivall, p. 18. Ibid.

198

Devin Fields

Figure 1.  Diagram of the battle of Barnet based on the work of Burne and Hammond.

men pursued Edward’s left flank from the field as far as Barnet. In Barnet, after some looting, Oxford was able to get some of his men under control and returned to the battle. As they marched back they were mistaken for more of Edward’s men and were attacked by their fellow Lancastrians. Some of the chronicles claim that Oxford’s men cried treason and that this panicked the Lancastrian forces. P. W. Hammond argued this was possible because Warwick’s army was cobbled together from Lancastrian and Neville retainers who did not necessarily trust each other.26 It was also about this time that Richard and his men found their way back to the battle and attacked the Lancastrian army from its left flank. These two events led to a break down of the morale in the Lancastrian army. In the panicked flight that followed, the earl of Warwick and his brother, the marquis of Montague, were both killed.27 The battle of Barnet has been the subject of some historical debate, mostly over the location of the battlefield and the deployment of the armies. What is beyond doubt is that Warwick’s and Edward’s forces were misaligned by the night deployment. In 1892 James Ramsay published his account of the battle, arguing that Warwick had deployed along the road running north from Barnet 26 27

Hammond, Battles of Barnet and Tewkesbury, p. 77. Arrivall, p. 20; Warkworth, Chronicle, p. 16.



English Gunpowder Artillery

199

so that he could then ambush Edward as he marched out of Barnet the next day.28 A. H. Burne considered this a preposterous idea, since it would open up Warwick’s flank to attack from Edward, who was in Barnet as the sun went down. He felt that a more probable placement was for Warwick’s men to be faced south, arrayed to either side of the road which ran north from Barnet. Burne was able to identify a hedge running along the east–west line which offered a place where he believed Warwick’s men could have spent the night.29 An important part of the overall determination of Burne’s account is that he took seriously the claim that Oxford was shot at by Lancastrian troops under the earl of Somerset.30 Burne’s work replaced the interpretation of Ramsay, and today historians agree as to a general east–west alignment of the armies. Some will argue more specifically for the location of the field of battle, but what matters is that Warwick was in fact facing south.31 While Burne’s east–west alignment has been accepted, what has managed to survive of Ramsay’s interpretation is the reason for the north–south alignment. Oxford returned to the field and was mistaken for Edward; the north–south alignment allowed Ramsay to argue for a turn of the battle lines based on the resulting flank attacks. A turn allows for Oxford to return in the rear of his own army and for him to be attacked by the Lancastrian troops without their having to contend with the Yorkist forces, which would otherwise be in between them. Some have tried to defend the turn and incorporate it into Burne’s battlefield layout. P. W. Hammond argued that the lines turned, and that when Oxford returned to the battle he and his men were believed to be Edward’s forces conducting a flank attack, which led to some fighting and the reported cry of treason.32 Hammond believed that Oxford’s men fought hand to hand with their Lancastrian allies. His interpretation requires the turn to have placed Warwick’s forces in a position to charge Oxford. A charge by Warwick’s forces against his ally asks us to disregard The Warkworth Chronicle, which mentions that in the fog on that day the badge of the earl of Oxford was mistaken for the badge most often used by Edward.33 Historians have debated the merits of this claim, some believing that Oxford’s emblem of a silver star could never be confused for Edward’s preferred sigil, the golden sun. What is important to mention here is that men who can see well enough to hit each other with melée weapons are not likely to mistake a silver star for a golden sun, no matter how thick the fog. The supposed turn of the armies has survived into accounts such as Hammond’s because it helps to explain how Warwick’s forces could have attacked Oxford. Evidence for the armies’ turning, though, is contained in only one primary 28 29 30 31 32 33

Sir James H. Ramsay, Lancaster and York (London, 1893), p. 370. Alfred H. Burne, Battle Fields of England (London, 2002), pp. 263–65. Ibid., pp. 260–61. Battle Field Trust, “The Battle of Barnet,” http://www.battlefieldstrust.com/resource-centre/ warsoftheroses/battleview.asp?BattleFieldId=5 [accessed 16 July 2014]. Hammond, Battles of Barnet and Tewkesbury, p. 77. Warkworth, Chronicle, p. 16.

200

Devin Fields

source: a letter from Edward’s sister, who was in Burgundy. The letter which she wrote to her mother-in-law claimed that Edward had been “en face de village” prior to the battle.34 Ramsay took a more literal interpretation and supported claims of a turn to explain how Edward came to be facing Barnet during the battle. Considering the nature of second- and third-hand accounts, some confusion is hardly surprising. We could explain his sister’s claim of Edward facing Barnet if the skirmish of the day before is included in the battle.35 If the skirmish is included the battle, instead of being separated from the main action the next day, the idea of any turn at the battle of Barnet becomes harder to defend, since for Edward’s sister the battle began when scouts met in Barnet. The Warkworth Chronicle offers another option, however, as it specifically mentions that Oxford was shot at by Warwick’s men.36 This could only have been done with ranged weapons, either archers or artillery, possibly Warwick’s archers. This would explain how Oxford’s men were attacked before they were sufficiently close to be no longer be mistaken.37 Some records of the Hundred Years’ War indicate archery was used to weaken the enemy before the general crush of hand-to-hand combat began. When this occurred, the archers would sometimes shoot into the flanks of the French army facing them, or they would join in the melée as well, fighting opportunistically against the more heavily armed French.38 Either way, they seem to have focused more on the immediate melée and less on parties that might be arriving in the rear of the enemy. When English armies fought each other, both would have a ready supply of archers equipped with longbows, so they would have to contend with enemy archers shooting at them in return. In at least one case, the return volley from archers seems to have influenced the events of a battle.39 In such circumstances, the speed with which the archers shot their arrows might increase as they tried to shoot the enemy before the enemy took aim at them. It is possible that the archers would run out of arrows at some point before the battle was over.40 Finally, in order to shoot at Oxford’s returning men, the arrows would have to pass over the heads of Edward’s army. It is not impossible that this occurrence describes exactly what had happened when Warwick’s artillery fired on Edward overnight. So, I believe we should consider the role of artillery. Gunpowder weapons had been growing in efficacy throughout the preceding century. The first instance where gunpowder weapons break a charge is in 1382, in the Low Countries.41 Also, the English had been expelled from France by 34 35 36 37 38 39 40 41

De Waurin, Anchiennes Chroniques D’Engleterre, 3:212–13. Ramsay, Lancaster and York, p. 370. Warkworth, Chronicle, 16. Sadler, Red Rose and the White Rose, p. 196. Desmond Seward, The Hundred Years War: The English in France 1337–1453 (New York, 1978), pp. 90–91; and Anne Curry, A New History of Agincourt (Gloucestershire, 2005), p. 212. Santiuste, Edward IV, pp. 55–56. Peter Reid, A Brief History of Medieval War (Philadelphia, 2008), p. 54. Robert Douglas Smith and Kelly DeVries, The Artillery of the Dukes of Burgundy (Woodbridge, 2005), pp. 15–16.



English Gunpowder Artillery

201

the strength of the French artillery train, not just in siege, but at the battle of Castillon.42 Grummitt’s argument that Calais functioned as conduit for tactics and technology enables us to accept that it was theoretically possible for English artillery to be capable of these same types of action.43 By 1471, it was possible for artillery fire to break the will of a surprised enemy. While it is all well and good to acknowledge the theoretical ability, the quality of the guns would be a major consideration in terms of their potential as weapons on the battlefield. A series of Patent Rolls from 1469 to 1475 provides a list of the types of guns possessed by the royal artillery train. While it gives no information on methods of construction, it provides perhaps the most in-depth information of the quality of English arms, indicating the relative modernity of the types used in England. Warwick would have had access to these weapons when he left for Coventry. The royal artillery train, headquartered at the Tower of London, included “bombards, cannons, culverins, folwers, serpentynes and other cannon.”44 Additional research on the artillery of Burgundy has shown that significant differences existed in the types of gunpowder weapons, suggesting a difference between heavy guns specifically for sieges, with a large caliber and short barrels, and lighter ‘field’ artillery with a smaller bore but longer barrels.45 Bombards were large cannons used primarily for smashing down walls. The serpentine was one of the types of guns with a longer barrel-to-caliber ratio, suggesting a role for it on the battlefield. Cannon was a more generic term, with no distinguishing characteristics identified. Culverins, as listed in the Patent Rolls, were actually the colouvrine of the Continent. In Burgundian records, they are quite small and are often called colouvrine à main. The à main refers to the hand, so these were handguns.46 The serpentine was an especially important type of artillery to consider because, while a very limited selection of records have survived, some of these sources provide some fascinating information on serpentines. After 1453, when the English lost their possessions in France, they also lost much of their artillery train. The guns could not be removed from France before the English were overwhelmed, and many can still be found in France today.47 Even Henry VI’s feeble government recognized this problem and worked to correct the deficiency in England’s arsenal of artillery. Soon afterwards, Jon Judde, a local London merchant, contracted with the crown for the production and delivery of sixty serpentines and a large quantity of saltpeter.48 Judde was able to deliver on this contract, so it is a given that during the Wars of the Roses the persons 42 43 44 45 46 47 48

Clifford J. Rogers, “The Military Revolutions of the Hundred Years War,” Journal of Military History 57 (1993), pp. 266–67. Grummitt, The Calais Garrison, pp. 139–40. Calendar of the Patent Rolls Preserved in the Public Record Office, 1467–1477 (1900, repr. Nendeln, Liechtenstein, 1971), pp. 163, 338, 365, 398, 474, 494. Smith and DeVries, Artillery of Burgundy, pp. 32. Ibid., pp. 204–19, 227–30. DeVries, “Gunpowder Weapons,” p. 28. Oliver F. G. Hogg, English Artillery, 1312–1716 (London, 1963), pp. 97–99.

202

Devin Fields

who controlled the royal government, like Warwick, had access to a significant number of serpentines for their use. For perspective, sixty serpentines was actually more serpentines than were in the Burgundian artillery train, widely regarded as one of the best in Europe.49 Warwick’s artillery train had deployed the night before the battle and fired on Edward’s army to little effect. This was because the men were firing where they thought Edward’s army was; their fire actually overshot Edward’s men and landed behind them. Oxford’s men would have crossed this area as they returned to the battle. There can be no doubt that Warwick’s artillery was able to hit Oxford’s men. Edward’s muster documents for his later invasion of France make clear that the English artillerymen went to war not as soldiers, but as craftsman, and list individually the various professions that would be needed to keep the artillery going. Every common soldier had a profession outside of war, but when it was time to call out men to fight, the soldiers are called armed men, with no acknowledgement of their premobilization lives.50 If it is a commission of array, no mention of soldiers is given at all.51 It seems likely that the artillerymen recruited as masons, smiths or carpenters would not get involved in the melée. They were hired for their skills, and unable to fire on the hand-to-hand combat, for fear of hitting their own men. By this point in the battle they probably would have returned to camp. The question is, where was the camp? Warwick seems to have picked the location of the battle as the sun was going down after a day of marching. The artillerymen were then also ordered to spend the night firing on Edward. There is also the opinion of Brune, which places Warwick’s camp pretty contiguous with the hedge that marks the battle line. The guns were probably placed in the same rough space as the camp. Then Oxford began his return with his easily mistaken banner. Warwick’s artillerymen fired on Oxford, believing that they were firing on reinforcements for Edward that would arrive and turn the battle decisively in his favor. When Oxford’s men were fired on, they panicked because the fire came from their own side. Oxford and his men fled from the field of battle with a cry of treason, and spared Edward an attack from behind.52 No historian in an institutional setting has yet argued for an event shaped by artillery, despite a growing body of evidence. It is the case that the sources for Barnet argue for the importance of gunpowder. It is possible that some historians may ultimately question what was decisive and effective. At Barnet it is probable that artillery fire, admittedly aimed mistakenly, broke the morale of a group of men who were in a position to attack Edward IV from behind. The battle of Barnet provides an example of the potential of English gunpowder weapons. In that battle, however, that role was shaped by chance. The Yorkists

49 50 51 52

Smith and DeVries, Artillery of Burgundy, p. 343. Calendar of Patent Rolls, 1467–1477, p. 363. Ibid., p. 252. Arrivall, p. 19; and Warkworth, Chronicle, p.16.



English Gunpowder Artillery

203

and Lancastrians would fight two more battles that year, each with a clearer application of artillery. After he had defeated Warwick, Edward was faced with conducting another campaign in the West Country. While gunpowder artillery had changed the events at Barnet, this was by accident. At the subsequent battle of Tewkesbury the cannons performed as expected. Furthermore, both Edward and Queen Margaret judged that their presence on the campaign was important. While Edward and Warwick were playing cat and mouse in the Midlands, other events were moving ahead in favor of Warwick and his new-found allies, the Lancastrians. According to the Arrivall, on 13 April Margaret and her son, Prince Edward, both left for England along with the remainder of the exiled Lancastrians.53 Queen Margaret and Prince Edward received news of the defeat and death of Warwick in the West Country, together with the duke of Somerset and John Courtenay, the Lancastrian earl of Devonshire. The news of Warwick’s death immediately threw the Lancastrian plan into doubt. Warwick had been expected to defeat Edward, or at least to have a force in being to challenge whatever force Edward could gather. The loss of Warwick was a devastating blow to the Lancastrians. Somerset and Devon managed to persuade Margaret, however, that the death of Warwick actually strengthened their position in England because Edward had already recruited one army and he likely would not be able to recruit a new one. They argued that Warwick’s death also was an advantage for the Lancastrians because of his dubious loyalty, and that by killing Warwick, Edward had alienated any Neville supporters who might have joined him. So Margaret and her supporters began recruiting from the West Country and gathered their men at Exeter.54 Meanwhile, in the wake of Barnet, Edward dismissed his army and was busy providing for the wounded when news arrived in London of Margaret’s landing. Edward hurriedly called his men back to London and began laying in supplies for his army. The Arrivall makes it clear that this included artillery which would be brought with the army. Most likely the artillery was the more mobile serpentines, although it included the hand-held coulovrines as well. While the preparations were being made in London, Edward went to Windsor and gave thanks to St. George for his victory against the Earl of Warwick.55 P. W. Hammond views the time that Edward spent at Windsor as dawdling.56 However, it does take time to organize an army, especially when that army is going to include an artillery train which would require the gathering of ammunition, gunpowder and the wagons and animals necessary to pull the guns and 53

54 55 56

Arrivall, pp. 22–23. Prince Edward was the son of Margaret of Anjou and Henry VI. As such, he was the next in line to the throne through the Lancastrian succession. To avoid the possibility that referring to both Prince Edward and Edward IV might lead to some confusion, Prince Edward will be referred to as ‘the Prince’. Arrivall, p. 23; Hammond, Battles of Barnet and Tewkesbury, p. 81. Arrivall, p. 24. Hammond, Battles of Barnet and Tewkesbury, p. 82.

204

Devin Fields

their supplies. It would also undoubtedly take several days to gather enough men, since many of those returning to London did so after just being released from service. As a result, a rest period would be desirable, if not necessary, for the morale of the army. Furthermore, Edward was preparing to fight his enemy in the West, which had traditionally been a region of Lancastrian strength, so the last thing he needed was a shortage of supplies. When Edward did advance, he marched west to Malmesbury on 1 May, preventing Margaret from gathering further reinforcements.57 On 30 April, Margaret was informed that Edward had arrived in the West, seeking to bring her to battle. According to the Arrivall, she was more than willing to oblige. The same day that Edward arrived at Malmesbury, Margaret marched her army to Bristol. In strategic terms, this took her west instead of north and ensured that Edward would catch her army. In Bristol, she acquired men, money and artillery. With her new recruits and guns, she attempted to face Edward near Sudbury in Gloucestershire. Her plan began to fall apart, however, when Edward’s men began arriving before her preparations were complete.58 Margaret fell back on her original plan to march north and rendezvous with Jasper Tudor, who was gathering men in Wales. She departed for Gloucester, which was the first place where the river Severn could be crossed. When it became clear to Edward that Margaret had decided against battle, he sent messengers to Gloucester to warn the mayor to expect an imminent attack. With that warning, his supporters at Gloucester prepared to resist any attack by Margaret. While she may have intended to assault Gloucester, she was deterred by her awareness that Edward’s army was somewhere nearby. It was obviously unwise to storm the town and risk a surprise attack from Edward on her otherwise distracted army.59 As Margaret was leaving the vicinity of Gloucester, it was claimed by the Tudor historian Edward Hall, the mayor sallied from the town and attacked the rear of her army. Hall asserts that the mayor’s forces seized at least some of Margaret’s artillery. This raises some concerns, since Hall was not born until after the battle of Bosworth, and thus his chronicle is based on the accounts of others. From the sources for the battle of Tewkesbury it seems as if only the Yorkists had artillery. Hall’s chronicle then explains the disappearing artillery and should be given some credence.60 As Margaret moved on with her army, her next chance to cross the Severn was at Tewkesbury, which had a reliable ford. By the time it arrived in Tewkesbury, the Lancastrian army was exhausted. Hammond estimates that, after leaving Bristol, Margaret’s forces had marched nearly fifty miles in thirty-six hours. Since Bristol, Edward’s army had been shadowing Margaret’s army and had spent most of the day before the battle within five miles of it. From the higher 57 58 59 60

Arrivall, p. 24. Ibid., p. 26. Ibid., pp. 24–26. Edward Hall, Chronicle, ed. H. Ellis (London, 1809), p. 300.



English Gunpowder Artillery

205

road which Edward’s army occupied, he could probably see Margaret’s army even if his own outriders were not providing regular reports of the Lancastrians’ progress.61 Both armies by this time would have completed grueling crosscountry marches. However, it was not in Edward’s interest to wait and avoid battle, since Margaret was marching in the direction of further reinforcements. A battle at Tewkesbury would probably have occurred even if Margaret’s forces had been more rested. Tewkesbury is sited near the junction of three rivers, the Swilgate, Avon and Severn. These three rivers formed a boundary to the east, the west and the north of the battlefield, placing Margaret’s army in a box with Edward’s army, coming from the south, making the fourth side.62 It is hard to imagine, as Edward was so close behind, that any attempt to cross the rivers in any sort of order could have been completed before he attacked. Because of the terrain, a battle would have been forced on Margaret unless she was willing to face the risk of having her army divided by a river when Edward attacked. The Arrivall makes clear that the battle began with Edward using artillery to attempt to draw Margaret’s army into an unwise attack. He allowed his archers and artillery to fire on the enemy. Thanks to the actions of the mayor of Gloucester, Edward had the only artillery on the field and the Lancastrian army could fire back only with bows and arrows. Attention must be paid to the archers in this battle, since the actual records of the armies are sparse; it is known that Edward paid the wages for over three thousand archers after the battle.63 This forces us to consider the question of longbow or cannon as the weapon of bombardment, which has already been examined for the battle of Barnet. Margaret deployed her army with the standard three divisions, each under the command of a prominent noble. At the center of the army, the Prince held command, accompanied by several more experienced nobles. The right flank was placed under the command of the duke of Somerset, while the left was under the command of the earl of Devon. With his ranged weapons Edward deliberately targeted the duke of Somerset, who had the strongest reputation as a soldier. The attempt worked and Somerset was forced to lead his men onto the offensive.64 While Somerset was forced to launch an attack because of Edward’s fire, the rest of the Lancastrian army remained on the defensive. This spelled disaster for Somerset’s attack, since Edward could defeat his smaller force with little fear of the rest of the Lancastrians.65 Fortunately for Somerset, while he was forced onto the offensive by missile weapons, the hedges which crisscrossed the battlefield obscured the view of Edward’s artillery and archers, giving his men a respite as they crossed the broken terrain. For a medieval army the maintenance of unit cohesion was espe61 62 63 64 65

Arrivall, p. 27; Hammond, Battles of Barnet and Tewkesbury, p. 87. English Heritage Society, “Battle Field Report: Tewkesbury 1471” (1995), p. 2. Arrivall, p. 25; Hammond, Battles of Barnet and Tewkesbury, p. 95. Arrivall, p. 35; Hammond, Battles of Barnet and Tewkesbury, pp. 93–94. Hammond, Battles of Barnet and Tewkesbury, pp. 95–97.

206

Devin Fields

cially important, since part-time soldiers provided the vast majority of troops. Somerset’s attack was able to avoid the damage to unit cohesion which missile weapons could produce. Somerset’s men reached Edward’s lines with very little damage and could have at least held their own until the rest of the Lancastrian host swung into action. Unfortunately for the Lancastrian cause, as Somerset’s men made contact with Edward’s men, they were also surprised by the unexpected attack of about two hundred mounted spearmen. When Edward had entered the vicinity of Tewkesbury, he had identified a forested hill further to his left which might be used to conceal an ambush. To avoid a nasty surprise by Margaret, Edward sent men to investigate. He also gave orders that if they found nothing, they were to use their own initiative to enter the battle at an opportune moment. The mounted party who attacked Somerset’s forces from an unexpected direction broke Lancastrian morale. Faced with a counterattack from their front and a surprise attack from their flank, Somerset’s men broke and fled back toward the rest of the Lancastrian army, which in the confusion joined the flight. The subsequent flight was an unmitigated disaster for the Lancastrian cause. In the immediate aftermath, the earl of Devon was killed. A greater loss for the Lancastrians was the death of Prince Edward. In the wake of the battle, Somerset fled to the sanctuary of Tewkesbury abbey, which was forcefully entered by Edward’s men. The nobles who were there were later tried and executed. Thus, the Lancastrians lost all three of their field commanders and the Lancastrian heir, their one hope for holding together a coalition to oppose the Yorkists.66 The role played by gunpowder artillery at the battle of Tewkesbury and the events which preceded it are unquestionable. The Arrivall, the best account of the battle, testifies to the role missile weapons played in forcing the Lancastrians into an uncoordinated attack. Edward had an unquestioned supremacy in gunpowder artillery. He had regained London and thus acquired the royal gunpowder arsenal, together with the Neville arsenal, following the defeat of Warwick. The apparent slowness of his response to Margaret reflects the time that Edward spent in gathering a new army which included field artillery. After Margaret lost a large part of her artillery outside Gloucester, Edward had the only significant artillery at Tewkesbury. Field artillery had the added benefit of heavier ammunition, which maintained its killing ability at a great range. With possible weakness in the bows and small arms of the armies at extreme distances, the field artillery provided a heavy weapon which could and did force battlefield commanders to attack instead of stand on the defensive against the cannon of their enemies.67 At Tewkesbury, gunpowder weapons gave Edward IV the ability to engage his enemy at a great range and force that enemy to come to him and attack under conditions which he dictated. With the exception of Bosworth in 1485 66 67

Arrivall, p. 30; Hammond, Battles of Barnet and Tewkesbury, pp. 97–101. Clifford J. Rogers, “Tactics and the Face of Battle,” in European Warfare, 1350–1750, ed. Frank Tallett and David J. B. Trim (Cambridge, 2010), pp. 235–44.



English Gunpowder Artillery

207

and Towton in 1461, no other battle was as decisive in these wars. Edward’s artillery functioned efficiently in providing the needed support. His gun crews were able to keep up with an army in the field and strike the Lancastrian army when it came time to do so, and to force the Lancastrian offensive after an exhausting three day march. It therefore cannot be concluded that the relatively new gunpowder artillery was not an effective and decisive element in the battles of the Wars of the Roses. If anything, Tewkesbury makes it clear that even in poorly fortified England, gunpowder artillery was a weapon of growing importance, ignored by leaders at their own peril. Before the fighting ended in 1471, an often-overlooked battle occurred which clearly demonstrated the sophistication attained by English artillery. Unlike the battles of Barnet and Tewkesbury, occurring in the countryside of England as armies were on the march and seeking to bring one another to battle, this last fight took place in London. The leader of the attack was Thomas Neville, known as the Bastard of Fauconberg. He was the illegitimate son of William Neville, the earl of Kent, Warwick’s uncle. Following Warwick’s defeat in 1469, Thomas Neville was forced to go into exile, where he was joined by his cousin. Thomas Neville brought with him several ships from Kent which were outfitted for war. The arrival of these forces increased the resources that Warwick had at his disposal and, moreover, Thomas Neville was more than eager to fulfill the role of raider assigned to him by his cousin, attacking English and Burgundian shipping.68 When Warwick returned from France in 1470 to expel Edward IV and place Henry VI back on the throne, he found a use for Thomas. In return for the support he received from France, Warwick was obliged to make war on Burgundy.69 While Warwick was gathering the men required by the French treaty, his cousin took command of an English fleet with the two-part mission of keeping Edward from landing in England and raiding Burgundian shipping. The sea is a big place, however, and Thomas failed to stop Edward from making the return crossing. When Edward landed in northern England, Warwick recalled his cousin to Kent to raise troops there.70 Thomas was unable to respond, however, because he was at sea and could receive no new orders.71 On 2 May, while Edward was beginning his pursuit of Margaret in the West Country, Thomas returned from the sea and landed in Kent with three hundred men from the Calais garrison. The men of Kent proved loyal to the Neville family and joined Thomas when he called for troops. With the men he gathered around him, he marched from Canterbury to London. The ships under Thomas’s command were ordered to sail around Kent and up the Thames to rendezvous 68 69 70 71

P. M. Kendall, Warwick the Kingmaker (New York, 1957), p. 302. Commynes, Memoirs, pp. 193–94. Kendall, Warwick, p. 355. It is unknown where he was at sea when Warwick sent his order. He had wandered quite a bit however, raiding shipping off the coast of Brittany. C. F. Richmond, “Fauconberg’s Kentish Rising of May 1471,” The English Historical Review 85 (1970), 676.

208

Devin Fields

with him outside of London. Before he arrived there, he sent a letter ahead asking permission to cross the river so that he could travel west and reinforce Margaret. Although Thomas claimed that he intended to come to London peacefully, his actions did not match his words. After his arrival outside of London, he put the borough of Southwark to the torch and destroyed the southern gate of London Bridge. If his men were allowed to enter the city, it was uncertain what he might do. In addition to Neville’s men, there were also some Lancastrian sympathizers within the city who might join forces with Thomas. The Yorkist leadership therefore suspected that if it came to violence, there would be attacks against the Yorkists by opportunists looking to take advantage of the chance to enrich themselves. In light of this information, the Yorkist garrison and city leaders decided to close the city gates to Thomas.72 In retrospect it seems that they were correct, since Thomas clearly hoped to capture London by subterfuge if he could. He further hoped that by taking the city, he would also take possession of Henry VI and provide a rallying point for Lancastrian discontent. At the same time that Thomas appeared outside London, the news arrived that Margaret’s army had just been defeated. A rallying point was now needed more than ever: the Tudors were stirring up discontent in Wales and another rebellion was forming in the north. A victory in London might well have strengthened these rebellions.73 Unlike Warwick earlier, who had been unable to leave a strong enough force to hold London, Edward left London with the support of the local leaders when he moved against Margaret. Chief among those left behind by Edward was Lord Rivers, the brother of the queen. This group of nobles and their men were able to provide strong support for London officials who denied Thomas Neville’s request to enter the city.74 The London government was unsure what would happen if it opened it gates to Thomas. It feared he was lying about his intentions and that once he was in the city he and the Lancastrian supporters would turn on the government, capture the city and release Henry VI from captivity in the Tower of London.75 Frustrated at the refusal of entrance after the destruction of Southwark, Thomas sent a party up the Thames to cross the river and come back to attack Westminster. The Arrivall believed that Thomas hoped the destruction of Southwark and other “suburbs” of the city would convince the Londoners to open their gates. The ships under Thomas’s command were left at London, along with the men who unloaded the cannons from his ships and placed them on the south bank of the Thames to bombard the city.76

72 73 74 75 76

Arrivall, p. 21. Hammond, Battles of Barnet and Tewkesbury, p. 107. Hammond, Battles of Barnet and Tewkesbury, p. 105. Arrivall, p. 21. Arrivall, p. 35; Hammond, Battles of Barnet and Tewkesbury, p. 107.



English Gunpowder Artillery

209

In the 1470s, the use of artillery on ships like Thomas’s was not an innovation. From 1379, Venetians hand been mounting cannons on their galleys.77 At least one Burgundian ship from the fifteenth century mounted five larger cannons which had a caliber of just under four inches. They were perhaps not large enough to demolish city walls, but the guns from Thomas’s ships, placed on the south bank of the Thames, faced London from a direction which did not have a defensive wall, because of the river. Thomas’s guns could thus fire directly into the city. Against the wooden structures of a fifteenth-century city, a four-inch cannon would have been more than sufficient. His choice of location clearly indicates that Thomas hoped to use his guns to weaken the resolve of his enemy, rather than to attempt to punch a hole directly through the defenses of London.78 The leaders of London were not unprepared. They recognized the flaws in the defenses of the city. Some large casks which were used to transport wine were filled with gravel and placed along the shore of the Thames to form a crude barricade and prevent a crossing into the city proper. Beyond these preparations, some of the buildings of St. Katherine’s Hospital, just outside the city walls to the east, were destroyed to clear lines of fire from the Tower.79 The defenders of London certainly had guns of their own. Although Edward probably took many of the guns from the royal artillery train with him, it is clear from both the Arrivall and the Warkworth Chronicle that the defenders used guns against Thomas Neville’s forces. It is likely that Edward, who needed speed above all, left many of the heavier weapons, such as bombards, in London. An important aspect of smaller weapons like handguns was that the training needed for their use was less than for the longbow, and hence they were more suitable for an urban militia and the defense of towns. In some German towns, the crossbow had already been completely replaced.80 The actual number of weapons in London at this moment is unknown, but they were there and they played a major role in the defense of the city. It is possible from the position of the artillery on the south side of the river that Thomas Neville might have been planning to bombard the city into submission, and not to assault it. The news of Edward’s victory at Tewkesbury meant that Neville now had only a limited time, and the last thing he would have wanted was to be caught between a city and an army with a force as small as the one he had. It is unsurprising, then, that Thomas decided to take the city by storm. He divided his force into three groups and planned to attack three different gates simultaneously. He had three thousand men on the north side of the Thames and he ordered them to attack both Aldgate, near the Tower of 77

78 79 80

John F. Guilmartin, “The Earliest Shipboard Gunpowder Ordinance: An Analysis of Its Technical Parameters and Tactical Capabilities,” The Journal of Military History 71 (2007), 658. Smith and DeVries, Artillery of Burgundy, pp. 343–44; Arrivall, p. 36. Arrivall, p. 36; Hammond, Battles of Barnet and Tewkesbury, p.107. Hall, Weapons and Warfare, p. 147.

210

Devin Fields

London, and Bishopsgate, which was north of that. On the south side of the Thames, a third party was ordered to force its way across London Bridge.81 On 14 May, the attack on London was launched by the different parties. The party which attacked Aldgate had the most success. They succeeded in forcing open the first gate and pushing the defenders back behind the portcullis. The defenders were so closely pressed that when the portcullis fell, some of Neville’s men were killed and seven were actually caught on the inside of it. They were quickly cut down by the defenders. Things could have continued to go poorly for the defenders, but as the attackers and defenders faced each other, separated by the portcullis, the Arrivall reports that there was a “continuynge of muche shote of gonnes and arrows a greate while, upon bothe parties.”82 After the attack had been blunted, the city was able to reorganize. The mayor and many of the leading citizens ordered the portcullis to be raised so they could attack the enemy. This initial assault had some success, as Neville’s men were forced back to a church outside the gate. At that moment, a second attack was launched by Earl Rivers from a postern gate near the Tower. The Neville attack on Aldgate collapsed as the London defenders counterattacked. The assault on Bishopsgate was stopped by the earl of Essex, and his men who arrived at London during the assault. The Neville men were pursued from London and chased five miles to Blackwell, where they reboarded their ships and returned to the south bank of the river.83 The attack of the third party against the gates of London Bridge met with significantly less success than the other two. The gate at the southern end of London Bridge had already been destroyed. Before Thomas’s men launched their assault, they set several fires to burn down houses on the bridge to help clear the way. Despite this “success” the assault came to an abrupt stop as they encountered “suche ordinaunce in theyr ways.” The Arrivall makes clear that it was the artillery that stopped them, turning an open bridge into a death trap as guns were sighted down the length of the bridge. Furthermore, guns’ effectiveness likely contributed to the success of the counterattack from the Tower, as the men who had been protecting the bridge could now join Rivers in his part of the counterattack on Aldgate.84 Thomas Neville, the Bastard of Facounburg, could have played an important role in English history, had he been able to take London. He was not able to do so, however, because the defenses of London proved too strong. It is very significant that artillery which had not been taken into the field with Edward’s army proved decisive in the defense of London Bridge. During Thomas’s attack on London both attackers and defenders used gunpowder weapons, which played a significant role in the defenders’ success in protecting the city. The three battles discussed here demonstrate the role which gunpowder 81 82 83 84

Arrivall, p. 36. Ibid., p. 36. Arrivall, p. 37; Hammond, Battles of Barnet and Tewkesbury, p. 108. Arrivall, p. 36.



English Gunpowder Artillery

211

artillery played in late fifteenth-century English warfare. Commanders such as Edward and Margaret delayed their marches to ensure that they would have access to gunpowder weapons. Barnet and Tewkesbury demonstrate the importance of English artillery in the open field. The assault on London offers further evidence of the importance of the weapons in England, enabling a successful defense of the kingdom’s largest city. The limited time of the campaigning in 1471 shows the importance of gunpowder weapons and their use in the Wars of the Roses.

10 “Cardinal Sins” and “Cardinal Virtues” of “El Tercer Rey,” Pedro González de Mendoza: The Many Faces of a Warrior Churchman in Late Medieval Europe1 L. J. Andrew Villalon

A king was here, his name was Corsablis… Archbishop Turpin has listened to his speech, And hates him worse than any man that breathes. His golden spurs he strikes into his steed, And rides against him right valiant for the deed. He breaks the buckler, he’s split the hauberk’s steel, Into his breast driven the lance-head deep, He spits him through…. [stanza 95] Archbishop Turpin goes riding through the field; Ne’er was mass sung by any tonsured priest That of his body could do such valiant deeds! He hails the Paynim: “God send the worst to thee! Thou hast slain one for whom my whole heart grieves.” Into a gallop he urges his good steed, He strikes him hard on his Toledo shield, And lays him dead upon the grassy green. [stanza 121]2 1

2

This article is based on research presented over the years at various academic conferences – the Central Renaissance Conference (1987), the sixth Annual Conference of the Medieval Association of the Midwest (1990), and the Texas Medieval Association Conference (2006). There are a number of individuals and institutions that have aided in my study of the house of Mendoza over a period of decades, going back to my first dissertation director at Yale, Dr. Ursula Lamb, and a former head of the Sección de Osuna in the Archivo Histórico Nacional, Consuelo Gutiérrez del Arroyo. The list includes my friend and co-editor, Donald Kagay, my mentor, Jack Hexter, my dissertation director, John Boswell, and a number of others, prominent among them Judith Daniels, Daniel Gottlieb, William Maltby, Sally Moffitt, Helen Nader, Ann Twinam, and Thomas White. I also wish to thank the following academic institutions: the Archivo Histórico Nacional and the Real Academia de la Historia in Madrid, the Archivo General de Simancas, the University of Cincinnati library services, in particular, interlibrary loan and photoduplication, and the University of Texas library system. As always, special thanks go to the Department of Romance Languages and Literatures at the University of Cincinnati, which took me in from the cold. Unless otherwise stated, all translations in the article are my own. These passages have been taken from a translation of the Chanson de Roland by an early twentieth-century medieval scholar turned mystery-story writer, Dorothy L. Sayers, The Song of Roland (Harmondsworth, 1937), pp. 99–100, 113.

214

L. J. Andrew Villalon

Introduction The two stanzas reproduced above come from one of the most famous works of medieval literature, the Song of Roland, first of the chansons de geste or “song of deeds,” depicting in a highly fictionalized manner what would ironically become Charlemagne’s most remembered battle, the defeat of his rearguard at Roncevaux. Prominent among those named in the chanson as having died fighting at Roland’s side was Archbishop Turpin of Rheims, to whom the passages refer. Although the appearance of the poem around 1100 converted Turpin into the prototype “warrior churchman” of the age, there is no certainty that such an individual actually existed outside of legend. To the extent that the fictional character accords with reality, he was probably a conflation of several clerics who held the see in the late eighth century, the best-known of whom was no warrior at all. In contrast to such a shadowy figure, one encounters throughout the medieval period real flesh-and-blood warrior churchmen about whom a good deal can be known, men who make it possible to arrive at a better understanding of just what it meant to combine the two roles.3 One such individual who lived in the late Middle Ages was the cardinal-archbishop of Toledo, Pedro González de Mendoza (1428–95), who occupied the primate see in the Castilian church near the end of the fifteenth century4 and was dubbed by one of his early biographers el gran cardinal.5 Like a number of contemporary prelates, including his chief rival and predecessor as archbishop of Toledo, Alonso Carrillo, Mendoza took up arms as he rose through the Church hierarchy. What is more, he continued to exercise his military function until the last three years of his life (1492–95), well after he became both an archbishop and a cardinal. In 1476, three years after having been raised to the cardinalate, he covered himself in military glory at the battle of Toro, the battle that assured the succession of the Catholic Monarchs, Ferdinand and Isabel.

3

4 5

In chapter 1 of his history of the medieval inquisition, Henry Charles Lea explores the many sins of the contemporary Church that alienated Christians and helped promote heresy. Lea spends several pages supplying prominent examples of clerical violence as manifested by warrior churchmen: Henry Charles Lea, A History of the Inquisition of the Middle Ages, 3 vols. (New York, 1888), 1:9–12. See also: James R. King, “The Friar Tuck Syndrome: Clerical Violence and the Baron’s War,” in The Final Argument: The Imprint of Violence on Society in Medieval and Early Modern Europe, ed. Donald J. Kagay and L. J. Andrew Villalon (Woodbridge, 1998), pp. 27–53. The archdiocese of Toledo was widely recognized as the primate see of Spain, comparable to Canterbury in England or Mainz in Germany. See: Pedro Salazar de Mendoza, Crónica de el Gran Cardenal de España Don Pedro González de Mendoza [hereafter Salazar de Mendoza, Crónica] (Toledo, 1625), esp. pp. 66–67. This same designation is also applied on occasion to Mendoza’s younger contemporary, protégé, and successor as cardinal-archbishop of Toledo, Francisco Jiménez de Cisneros, who in his own right became another major adviser of the Catholic Monarchs, Ferdinand of Aragon and Isabel of Castile.



Pedro González de Mendoza

215

A letter written in May 1488, seven years before the cardinal’s death, by an Italian humanist named Pietro Martyr, provides some insight into just how important Mendoza was in the history of late medieval Spain. Having recently left his native Italy to escape political turbulence that “daily went from bad to worse,”6 Martyr journeyed to Iberia under the protection of the cardinal’s nephew, Iñigo López de Mendoza, count of Tendilla,7 who was returning in triumph after having helped to prevent a war between Naples and the papacy.8 Now, in a letter to his protector’s uncle, the newcomer conferred upon the aging prelate a nickname that highlights his significance; he thanked Mendoza for his support in “a world … where you are justly called the third king [el tercer rey],” and where “the entire government of the realm rests upon your shoulders.”9 Pietro Martyr’s letter to Mendoza is one of those set pieces of flattery that flowed so readily (and unblushingly) from humanist pens.10 Nevertheless, his praise for the cardinal contains considerably more than a grain of truth. For nearly thirty years, Pedro González de Mendoza had stood at center stage in a realm torn by both civil and foreign wars, wars in which he was a frequent participant, donning his armor under his clerical garb and appearing in the forefront of battle. Starting as early as 1460, he also served as a leading royal adviser, first to Enrique IV (r. 1454–74), who bore the unfortunate sobriquet of “el impotente,” later to Enrique’s illustrious successors, the Catholic Monarchs (los reyes católicos), Fernando of Aragon (r. 1474/1479–1516) and Isabel of

6

7

8

9

10

Originally written in Latin, the letters have been collected and translated into Spanish. See: Pedro Martir de Anglera, “Epistolario, estudio y traducción por José López de Toro,” in Documentos Inéditos para la Historia de España vols. 9–12 (Madrid, 1953). This letter complaining of the contemporary situation in Italy appears in Martir, “Epistolario”, 1:5–6 (letter 2: 29 August 1487). See also: Martir, “Epistolario”, 1:3–5 (letter 1) and 1:6–7 (letter 3). A partial genealogical chart of the house of Mendoza and the royal house of Trastámara can be found in L.J. Andrew Villalon “Deudo and the Roots of Feudal Violence in Late Medieval Castile,” in The Final Argument: The Imprint of Violence on Society in Medieval and Early Modern Europe, ed. Donald J. Kagay and L.J. Andrew Villalon (Woodbridge, 1998), p. 56. For the count of Tendilla’s peace mission to Italy, see the leading chronicle of the reign of Ferdinand and Isabel: Fernando del Pulgar, Crónica de los Reyes Catolicos, 2 vols. [hereafter Pulgar, Crónica] edición y estudio por Juan de Mata Carriazo (Madrid, 1943), 2:206–8. Various documents concerning this Spanish diplomatic initiative are printed in two collections: Documentos sobre Relaciones Internacionales de los Reyes Católicos, ed. Antonio de la Torre, 6 vols. (Barcelona, 1949–66), 2:237, 243–46, 249–51, 257, 278–85, 288–89, 296–99, 301, 313–14, 386–87; Politica Internacional de Isabel la Católica, Estudio y Documentos, ed. Luis Suárez Fernández, 2 vols. (Valladolid, 1965–66), 2:339–63, 366–68, 370–77, 386–90, 392–98. For a summary of the mission, readers should consult the best work in English dealing with the house of Mendoza, Helen Nader’s The Mendoza Family in the Spanish Renaissance, 1350– 1550 (Rutgers, 1979), pp. 152–54. The letter identifying Mendoza as “el tercer rey” appears in Martir, “Epistolario”, 1:31–2 (letter 24). In another letter from this period, Martyr conveys his understanding that “in the realm, he occupies the third place.” Martir, “Epistolario”, 1:37–38 (letter 29: 31 May 1488). The sort of obsequious flattery to which even the “Prince of Humanists,” Desiderius Erasmus, might stoop comes through clearly in the pages of Roland Bainton’s biography. See: Roland Bainton, Erasmus of Christendom (New York, 1982).

216

L. J. Andrew Villalon

Castile (r. 1474–1504).11 Throughout both reigns, it was Mendoza’s significant political and military role rather than any religious activity on his part that primarily accounted for his rise to the apex of power in a kingdom that was on the verge of becoming the foremost European state of the early modern period. In this respect, he differed somewhat from his protegé and successor as archbishop of Toledo, Francisco Jiménez de Cisneros. As advisor to the Catholic Monarchs later in their reign, Cisneros played at most a minor role in military affairs, but a far greater role in religious matters than did Mendoza. Under the circumstances, any flattery that Martyr may have lavished upon his new patron would have been hard put to overestimate that individual’s true significance.12 This article will examine from various perspectives the career of Pedro González de Mendoza. At any one time, this younger son, who achieved prominence not only on the battlefield but in many other endeavors, had many irons in the fire. After briefly mentioning the available sources, then summarizing Mendoza’s career, we shall focus upon those military activities that justify placing him among the great warrior churchmen of his age. On the other hand, while closely examining his participation in the warfare of the period, we shall also consider the many other roles played by this versatile prelate, what are referred to in the title as “the many faces of a warrior churchman.” It is, after all, these other “faces” that help to distinguish a real figure like Mendoza from fictional counterparts such as archbishop Turpin. What I hope to demonstrate in this article is the fact that many if not most real warrior churchmen of the Middle Ages, as distinct from their counterparts in song and story, had too many “faces” to ever be summed up by mere allusion to their warrior status.

11

12





The dates given for the Catholic Monarchs require some explanation. Ferdinand and Isabel, who had married in 1469, succeeded in her name to the Castilian throne in 1474. In 1479, Ferdinand came into his own inheritance, the crown of Aragon. Isabel died in 1504, designating Ferdinand as regent for their unbalanced eldest surviving daughter, Juana la Loca. However, the Castilian nobles backed Juana’s Flemish husband, Philip “the Handsome,” who became Philip I. After Philip’s premature death in 1506, Ferdinand returned as regent for Castile, a position he retained alongside his own crown of Aragon until 1516, when he was succeeded in both realms by his grandson, Charles. Writing about a century later than Pietro Martyr, the cardinal’s first biographer summed up his famous subject along much the same lines: [The cardinal] held a high position with King Enrique and an even higher one with the Catholic Monarchs. During the twenty-one years he lived under their rule, he never failed to accompany the king in war and the queen in peace. [The royal pair] never determined any matter of importance without consulting him, nor would they deny him anything he asked of them.… See: Francisco de Medina y Mendoza, Vida del Cardenal D. Pedro González de Mendoza [hereafter Medina y Mendoza, Vida] in Memorial Historico Espanol: Colección de Documentos, Opúsculos, y Antiguedades que Publica la Real Academia de la Historia, vol. 6 (Madrid, 1853), pp. 298–99.



Pedro González de Mendoza

217

Sources Given the pivotal nature of the cardinal’s career, there is no lack of source material for studying him. He lived during a period when many chroniclers were at work in Castile; as a result, he appears prominently in all principal chronicles dealing with the reign of Enrique IV – Diego Enriques del Castillo, Alonso de Palencia, Diego de Valera – as well as those that cover the Catholic Monarchs – Palencia, Fernando del Pulgar, Andrés Bernaldez, and Alonso de Santa Cruz.13 Together, these works are of particular importance when it comes to revealing the churchman’s public persona, including his participation in the warfare of the period, and therefore they have been used extensively in writing this article.14 In turn, the contemporary chronicles provide much of the basis for two early, highly laudatory biographies, both of which appeared some decades after the prelate’s death. The first, entitled Vida del Cardenal D. Pedro González de Mendoza, dates to the second half of the sixteenth century and was written by Francisco de Medina y Mendoza, a vassal of Iñigo López de Mendoza, fifth duke of the Infantado, who, through his mother, was directly descended from the cardinal.15 A second, longer, and considerably better-known Crónica de el Gran Cardenal de España Don Pedro González de Mendoza was the work of the churchman’s own great-great grandson, Pedro de Salazar y Mendoza, a canon of the cathedral of Toledo and historian of some note, who published it in 1625.16 In addition to the narrative sources, an extensive collection of the cardinal’s papers (wills, mayorazgos,17 property deeds, charitable donations, etc.) entered the Mendoza family archives, preserved over the centuries by the dukes of the Infantado and housed since 1915 in the Archivo Histórico Nacional.18 In the late seventeenth century, copies of many of these documents also found their 13 14

15 16 17 18

I supply the full bibliographical reference to each chronicler when I first cite his work. Chronicle accounts often supply historians with much if not most of their surviving information about medieval warfare; their importance in this respect is stressed in an article by Kelly DeVries, “The Use of Chronicles in Recreating Medieval Military History”, Journal of Medieval Military History 2 (2004), pp. 1–15. Medina y Mendoza, Vida. See n. 5. Mayorazgo is the Iberian version of what the English call an “estate in tail” or simply entail. Both first appeared around the thirteenth century. Legislation creating the Archivo Histórico Nacional [AHN] in 1866 designated it the public archive for the realm (archivo público general del Reino). Today, it numbers among Spain’s five most important archives for research into medieval and early modern Spanish history, alongside the Archivo General de Simancas [AGS], the Archivo de la Corona de Aragon [ACA], the Real Academia de la Historia [RAH], and the Archivo General de Indias [AGI]. Of the fifteen secciones contained in the AHN, the eleventh, entitled the Sección de Osuna [hereafter AHN, Osuna], contains the family papers of numerous noble houses, including the Mendozas. Although formerly preserved in the central archival complex in Madrid, the Osuna papers have now been moved to new quarters in Toledo. For extensive treatment of the AHN and its holdings, see: Luis Sánchez Belda, Guia de Archivo Histórico Nacional (n.p., 1958); and Carmen Crespo Nogueira, Archivo Histórico Nacional: Guia (Madrid, 1989).

218

L. J. Andrew Villalon

way into the Real Academia de la Historia in a section called the Coleccíon Salazar, named for one of the leading European genealogists of the early modern period, Luis de Salazar y Castro. A number of these documents were subsequently printed by Salazar y Castro as appendices to his massive genealogical histories.19 Taken together, they supply a better understanding of the cardinal’s personal life. Perhaps the only real hindrance to exploring the churchman’s career grows out of a near absence of writings penned by the man himself,20 an absence that may help to account for the relative paucity of modern biographies dealing with such a key figure of the period.21 “El Gran Cardenal” Pedro González de Mendoza was born on 3 May 1428, and died on 11 January 1495, at the age of sixty-six.22 The line to which he belonged, what would later become known as the Infantado branch of the house of Mendoza, was one of the greatest in late medieval Castile, where it played a significant role since migrating there from the Basque province of Álava shortly before the middle

19

20

21

22

Other than the AHN, Spain’s most extensive documentation devoted specifically to the history of Castilian noble families resides in the RAH, established in 1738; and in particular, in the Colección Salazar. Luis de Salazar y Castro later printed a number of relevant documents which he had compiled in several of his genealogical histories; for example, in the Pruebas para la Historia de la Casa de Lara (Madrid, 1697), which is the fourth and final volume of his Historia Genealógica de la Casa de Lara. Nevertheless, the actual collection dwarfs that part of it which made it into print. Some idea of its immensity can be formed by consulting the forty-nine volume catalogue, drafted between 1949 and 1979. See: Baltasar Cuartero y Huerta and Antonio de Vargas-Zuñiga y Montero de Espinosa, Marquess de Siete Iglesias, Indice de la Colección de Don Luis de Salazar y Castro, 49 vols. (Madrid, 1949–19). Some letters of the cardinal do survive that deal with his foundation of the Colegio de Santa Cruz in the city of Valladolid. See: Cartas del Cardenal Pedro González de Mendoza al Colegio de Santa Cruz de Valladolid (Valladolid, 1989). Unfortunately, no personal correspondence of the sort that sheds maximum light on an individual’s character has ever surfaced. The closest we have may be those speeches scattered throughout Pulgar’s chronicle, said to have been delivered in meetings of the royal council. See, for example, see: Pulgar, Crónica, 1:201–5, 219–20; 2:195–96. Twentieth-century works include Ramón La Cadena y Brualla, marqués de La Cadena, El gran Cardenal de España (Madrid, 1942); Abelardo Merino Alvarez, El cardinal Mendoza (Barcelona, 1942); F. J. Villalba Ruiz de Toledo, El Cardenal Mendoza (1428–1495) (Madrid, S.A., 1988) Antonio Herrera Casado, La huella viva del Cardenal Mendoza (Guadalajara, 1995). There is no English-language biography. The best accounts in English are to be found in Nader’s The Mendoza Family and Lynette M. F. Bosch, Art, Liturgy, and Legend in Renaissance Toledo: The Mendoza and the Iglesia (Park, PA, 2000), esp. chap. 2: “The Archbishops of Toledo and their Monarchs: Politics in Fifteenth Century Castilla,” pp. 106–14. Several other books of recent decades mention primarily the cardinal’s progeny: Ian MacPherson and Angus MacKay, Love, Religion, and Politics in Fifteenth Century Spain (Leiden, 1998), chap. 5; James Boyden, The Courtier and the King (Los Angeles, 1995), pp. 24–29. Medina y Mendoza, Vida, pp. 154, 296. Salazar de Mendoza, Crónica, p. 61.



Pedro González de Mendoza

219

of the fourteenth century.23 The cardinal’s great grandfather, the first Pedro González de Mendoza, for whom he was named,24 fought on the winning side in the civil wars that brought the Trastámaran dynasty to power.25 By the time the older Pedro González met a heroic death fighting the Portuguese at the battle of Aljubarrota (1385),26 he had managed to firmly establish his family among the upper nobility of the kingdom. The cardinal’s grandfather, Admiral Diego Hurtado de Mendoza, despite a relatively early death, became a royal favorite of 23

24

25

26

Two standard family genealogies, which in several instances contradict each other, especially when treating the family’s remote origins, are: Luis Salazar y Castro’s seventeenth-century Historia Genealógica de la Casa de Haro (Madrid, 1959); and Diego Gutiérrez Coronel’s eighteenth-century Historia Genealógica de la Casa de Mendoza (Cuenca, 1946). The most important modern treatment of the family is Francisco Layna Serrano’s four-volume Historia de Guadalajara y sus Mendozas en los Siglos XV y XVI (Madrid, 1942) [hereafter Layna Serrano]. In addition to its highly detailed text, this work reproduces a number of key documents dealing with the family. As indicated earlier, English readers should consult Nader’s The Mendoza Family. Virtually all scholars who have dealt with the house of Mendoza stress the importance of the cardinal’s namesake in the aggrandizement of his line. For more information about the first Pedro González, see my dissertation: “The Law’s Delay: The Anatomy of an Aristocratic Property Dispute (1350–1577),” (Ann Arbor, MI, 1984), especially chapter 2. For detailed treatment of the Trastámaran civil wars which propelled the Mendozas into the upper aristocracy, see my earlier articles, “Pedro the Cruel: Portrait of a Royal Failure,” in Medieval Iberia: Essays on the History and Literature of Medieval Spain, ed. Donald J. Kagay and Joseph T. Snow, Iberica Series, vol. 25 (New York, 1997), pp. 201–16; “Seeking Castles in Spain: Sir Hugh Calveley and the Free Companies’ Intervention in Iberian Warfare (1366– 1369),” in Crusaders, Condotierri, and Cannon: Medieval Warfare in Societies around the Mediterranean, ed. L. J. Andrew Villalon and Donald J. Kagay (Leiden, 2003), pp. 305–328; “The Battle of Najera and the Hundred Years War in Spain,” in The Hundred Years War: A Wider Focus, ed. L. J. Andrew Villalon and Donald J. Kagay (Leiden, 2005), pp. 3–74; “‘Cut off their Heads, or I’ll Cut Off Yours’: Castilian Strategy and Tactics in the War of the Two Pedros and the Supporting Evidence from Murcia,” in The Hundred Years War: Different Vistas, ed. L. J. Andrew Villalon and Donald J. Kagay (Leiden, 2008), pp. 153–82. On 14 August 1385, on the battlefield of Aljubarrota 110 kilometers north of Lisbon, a numerically superior Castilian army encountered a better-disciplined Portuguese force, including a contingent of English longbowmen, that occupied a strongly fortified defensive position. As in earlier encounters between the two, the overconfident Castilians launched a series of costly and unsuccessful frontal assaults. By the end of the day, they had suffered one of the most crushing defeats of the century, leaving large numbers of dead on the field, numerous prisoners in enemy hands, and what remained of the army, including King Juan I (1379–90) in headlong flight toward the border. In this sauve qui peut atmosphere, Mendoza supplied a shining moment that became the stuff of legend. After surrendering his horse to the fleeing monarch, he reentered the fray, where he met his death. This example of knightly prowess became the subject of a poem in the Romancero General reprinted in volume 79 of the Boletín de la Real Academic de la Historia (p. 185) that contains a short biographical sketch of this first Pedro González as well as a number of key documents concerning his life. For his part, the cardinal alluded to this episode in a 1476 speech to the war council indicating that the house of Mendoza remained well aware of its forbear’s heroism. See: Pulgar, Crónica, 1:220 For an excellent recent treatment of the battle, based on both the historical literature and new archaeological evidence, see: João Gonveia Monteiro, “The Battle of Aljubarrota (1385): A Reassessment,” Journal of Medieval Military History 8 (2009), 75–103.

220

L. J. Andrew Villalon

Enrique III (r. 1390–1407) and a leading political figure around the turn of the fifteenth century. His father, Iñigo López de Mendoza, first marquess of Santillana, enjoyed considerable renown as both a warrior and a poet.27 As a younger son in an aristocratic family, Mendoza found himself destined, like many another in his position, to a career in the Church.28 After receiving a sound education, family connections played a key role in his early preferment. In 1442, at the age of fourteen, he assumed his first Church benefice, when his father had him appointed curate in the family-held town of Hita. Three years later, he moved to the city of Guadalajara, also dominated by his family, where, despite his youth, he assumed a lucrative archdeaconship. In 1452, after concluding university studies, Pedro González entered the court of the old monarch, Juan II (r. 1407–54), as a royal chaplain and two years later became bishop of Calahorra at the still relatively young age of twenty-six. For the rest of the decade, he appears to have fulfilled the role of a resident cleric, the longest of only two occasions in a long career when he did so. Although these higher clerical benefices – the royal chaplaincy and the bishopric – were conferred upon Mendoza by the crown, the old monarch hardly knew his young churchman. Instead, these, like the earlier appointments, clearly reflected a royal desire to please a powerful noble house and can therefore also be attributed to family influence rather than any personal actions.29 It was only after his elevation to the episcopate that Mendoza embarked upon what might be called his secular career, serving the crown in the politics and warfare of the period. In 1461, Enrique IV reached an agreement with the Mendozas, ending several years of estrangement between the crown and family.30 As part of their rapprochement, Enrique demanded that the bishop take 27

28 29 30

Iñigo López de Mendoza, first marquess of Santillana and count of Real de Manzanares (1398–1458), was one of the major poets of late medieval Castile. For over four decades during the reigns of Juan II and of his son, Enrique IV, he also participated actively in war and politics. A nineteenth-century biography of the warrior-poet, which for the first time made extensive use of the Mendoza family archives, was written by José Amador de los Rios, Vida del Marquess de Santillana (Buenos Aires, 1947). Layna Serrano devotes most of his first volume to the marquesses’ career. The writings of Santillana are printed in: Obras del Marquess de Santillana [hereafter Santillana, Obras], ed. Augusto Cortina (Madrid, 1964). For a brief account in English, see: E. Michael Gerlí, “López de Mendoza, Iñigo,” in Medieval Iberia: An Encyclopedia, ed. E. Michael Gerlí (New York, 2003), pp. 518–19. Don Pedro was the fifth-born son of the marquess of Santillana. Medina y Mendoza, Vida, p. 157. Salazar de Mendoza, Crónica, pp. 66–67. By the late 1450s, the king found himself at odds with a number of the great nobles, among them the Mendozas. In 1461, a royal decision to intervene in the internal affairs of Aragon put a premium on securing peace at home. While most of the nobles rejected his overtures, the Mendozas proved willing to settle their differences with the crown. According to several chroniclers, after this settlement “fueron siempre firmes constantes muy leales servidores del rey.” For the history of this reign, see: Alonso de Palencia, Crónica de Enrique IV [hereafter Palencia], introduction by A. Paz y Melia, 3 vols. (Madrid, 1973–75). Diego Enriques de Castillo, Crónica del Rey Don Enrique el Cuarto de este Nombre por su capellan y cronista Diego Enriquez de Castillo [hereafter Castillo] in Crónicas de Los Reyes de Castilla 3 (Madrid, 1953), pp. 98–222. Mosen Diego de Valera, Memorial de Diversas Hazañas [hereafter Valera]



Pedro González de Mendoza

221

up permanent residence at court, in effect serving as a hostage for the family’s good behavior.31 The recruitment of the young churchman ranks as one of the best such decisions ever made by a king with a penchant for selecting mediocre advisers and disastrous favorites. Very quickly, Mendoza began to demonstrate some of those qualities which, over the course of more than thirty years, would make him indispensable to the monarchy. During those decades, he served as royal adviser, diplomat, liaison with Rome, troubleshooter, spokesman, propagandist, and, of course, despite his clerical garb, warrior. As a result of this multifarious service, he moved progressively up through the Church hierarchy. Under Enrique, he was translated from Calahorra to the more prominent see of Siguenza in 1468, then became both cardinal and archbishop of Seville in 1473. Throughout the five years following Enrique’s death in 1474, when a war of succession raged within Castile, the cardinal staunchly supported the Catholic Monarchs in their fight for the crown.32 He travelled with the queen during her efforts to quell domestic disturbances and joined the king on the battlefield. At the decisive battle of Toro in 1476, as on several past occasions, he again donned his armor and by sound military advice as well as heroic actions cemented his reputation as a warrior churchman. From the position of archbishop, first of Seville and later of Toledo, he backed Ferdinand and Isabel in their efforts to establish a Spanish national church, similar to the Gallican Church of France. Despite his advancing age, he once again rendered extensive military service (albeit of a somewhat less strenuous nature) in the decade-long war against Granada (1481–92), the final act in the Spanish reconquista that had started nearly eight centuries earlier.33 Mendoza retained his preeminent position in both the political and military councils of

31

32

33

edición y estudio por Juan de Mata Carriazo (Madrid, 1941); Lorenzo Galíndez de Carvajal, Crónica de Enrique IV [hereafter Carvajal] in Juan Torres Fontes, Estudio sobre la “Crónica de Enrique IV” del Dr. Galindez de Carvajal (Murcia, 1946), pp. 160–61. While Palencia, the official chronicler, is extremely critical of his subject, Castillo, the royal chaplain, has a far friendlier take on this monarch. Carvajal’s later chronicle was based on the earlier ones and is a compromise between their varying positions. The bishop’s status as a hostage is suggested by the fact that at the same time the king also required the bishop’s brother, the second marquess of Santillana, to send his son Juan to court “en rehenes, condicionalmente que no saliese della sin licencia y mandado del rey.” Carvajal, p. 161. In addition to Hernando del Pulgar’s chronicle that deals with all but the last years of the reign, the major sources for the Catholic monarchs in which Mendoza appears are: Alonso de Palencia Crónica, esp. vol. 3; Andrés Bernaldez (Curate of Los Palacios), Memorias del Reinado de los Reyes Católicos [hereafter Bernaldez] (Madrid, 1962); and a Crónica incompleta de los Reyes Católicos (1469–1476), prólogo y notas de Julio Puyol (Madrid, 1934). The other major source for the reign, the chronicle of Alonso de Santa Cruz, takes up the story after Mendoza’s death. The Reconquista was the movement to drive Islam from its westernmost foothold in Europe, the Iberian Peninsula, known to the Moslems as Al-Andaluz. It is considered to have begun in the year 718 at the battle of Covadonga. Here, a small group of Visigothic survivors, having fled into the northern mountains, banded together under the leadership of Don Pelayo and turned back a half-hearted Moorish attempt to penetrate the region, an encounter that Christian chronicles depict as miraculous. The movement continued for nearly eight centuries, ending

222

L. J. Andrew Villalon

the Catholic Monarchs until the surrender of Granada in 1492, at which time he retired to manage the affairs of his archepiscopal see. Just before departing from the captured city, Mendoza delivered his last and perhaps most significant piece of advice to the crown when he recommended that Ferdinand and Isabel give serious consideration to the proposal of a little-known Italian mariner named Christopher Columbus.34 Mendoza lived for much of his last three years in Guadalajara, where he died in 1495. A splendid funeral procession conveyed his earthly remains to Toledo, where they were laid to rest within the cathedral.35 The Warrior Churchman Over the course of his life, the cardinal’s greatest virtue, frequently remarked on by the chroniclers, proved to be his nearly unswerving loyalty to the crown. It was a trait he demonstrated during two fundamentally different reigns; not only under the highly competent and charismatic monarchs Ferdinand and Isabel but even more tellingly, during the troubled reign of their predecessor, Enrique IV, when aristocratic perfidy reached its peak and loyalty of the sort displayed by Mendoza was indeed rare. The years when Mendoza served the hapless Enrique witnessed a kaleidoscopically shifting pattern of aristocratic alliances with and against the king, as great noble families maneuvered to acquire power, influence, and property, frequently to the considerable detriment of the crown. However, even when the king’s blatant failure to consult his closest allies over the issue of the succession occasioned a temporary rift with the house of Mendoza, the bishop and his family simply absented themselves from court as a means of

34

35

only in January 1492, when the last Moorish stronghold on the peninsula, the city of Granada, fell to the Catholic monarch. The third major chronicle of the Catholic monarch’s reign was a continuation of the work of Hernando del Pulgar, written by a prominent geographer, Alonso de Santa Cruz. Since it begins only in 1491, on the eve of the cardinal’s retirement from royal service, it therefore contains the fewest references to this long-dominant figure. On the other hand, it is this source that supplies the best account of his connection with Columbus. See: Alonso de Santa Cruz, Crónica de Los Reyes Catolicos, edición y Estudio por Juan de Mata Carriazo, 2 vols. (Seville, 1951), 1:64–65. To escape an outbreak of plague in the city of Nuremberg, a German physician named Hieronymus Munzer, Dr. Munzer (sometimes called Monetarius), set out on a lengthy journey through Switzerland, France, Spain, and Portugal, a Latin account of which he subsequently published under the title Itinerarium sive peregrinatio excellentissimi viri, atrium ac utriusque medicinae doctoris Hieronymi Monetarii de Feltkirchen, civis Nurembergensis. The sections dealing with Spain and Portugal have been translated into Spanish. Arriving in central Spain in the summer of 1495, Munzer witnessed the nearly regal funeral cortege afforded this great churchman, one that he said covered the twenty-two leagues from the family seat in Guadalajara to the great archiepiscopal city of Toledo, and that proceeded “with so much splendor, pomp, and solemnity, that no more would be possible.” Jerónimo Munzer, Viaje por España y Portugal, 1494–1495, trans. José López Toro (Madrid, 1951). See also: “Voyage de Philippe le Beau en Espagne en 1501 par Antoine de Lalaing, Sr. de Montigny,” [hereafter Lalaing] in Collection des Voyages des Souverains des Pays-Bas, ed. L. P. Gachard, 2 vols. (Brussels, 1874–82), 1:167; Salazar de Mendoza, Crónica, pp. 262–65.



Pedro González de Mendoza

223

protest, rather than enter into the self-interested plotting so characteristic of nobles in this period.36 Despite his clerical status and unsurpassed value as a political adviser, Cardinal Mendoza’s loyalty, courage, and exceptional sang froid, coupled with the troubled era in which he lived, rendered it likely that his service to the crown would sometimes take a military form. In fact, as things turned out, the image of the warrior cleric, ignoring Church strictures against blood-letting, well suited Don Pedro, who, wearing a steel helm and armor under his clerical garb, participated actively not only in the two major battles of the period – Olmedo (1467) and Toro (1476) – but also in lesser engagements. What is more, during many of the sieges through which the Catholic Monarchs finally reduced Granada, the aging cleric was a military fixture. There was, however, one significant difference in the nature of the military service he rendered during the two successive reigns. Enrique IV’s engagement in a foreign war came early on, during the first half dozen years of his reign, when he repeatedly attacked (ultimately with limited success) the mountainous kingdom of Granada, the last Muslim stronghold on the Iberian Peninsula, which lay to the south of Castile.37 When this latest round in the reconquista took place, the young bishop of Calahorra appears to have been safely ensconced in his new diocese far to the north, and although members of his immediate family, including his aging father and his older brothers, took part in the king’s abortive campaign, he himself played little or no role.38 Instead, his military service to Enrique would come only later, during years of turbulence within Castile that frequently verged on civil war, in a period when the near absence of major engagements served to limit the military reputation he could acquire. Nevertheless, there were incidents during those years that amply demonstrate not only the churchman’s fidelity, but also his courage. Throughout Enrique IV’s reign, the king’s boyhood friend and favorite, Juan Pacheco, marquess of Villena (d. 1474), played a significant role in fostering the turbulence that engulfed Castile and did much to render the reign a shambles.39 On one occasion in 1464, as opposition to the crown magnified and Castile descended toward 36 37 38

39

Castillo, p. 178; Carvajal, p. 333. Castillo, pp. 104–5, 133; Carvajal, pp. 82–83, 93–94, 96–98, 103–7, 114–17, 119–120, 126–29. According to the chroniclers, when Enrique called for a renewal of the Granadan War during his first cortes at Cuellar, the estates elected Mendoza’s father, the marquess of Santillana, as their spokesman to announce their support for the king’s proposal. In December 1455, the marquess sat in the royal council of war, and the following spring he marched with the monarch’s first expedition against the Moors. According to the chronicles, he was accompanied by four of his sons: Diego Hurtado, who would soon succeed him, Pedro Laso, Iñigo López, and Lorenzo Suárez. The chroniclers do not list Pedro González de Mendoza as part of this family contingent. Castillo, p. 105; Carvajal, pp. 83, 89. For a recent biography of this enigmatic figure, see Nancy Marino, Don Juan Pacheco: Wealth and Power in Late Medieval Spain (Turnhout, 2006). For my rather more negative view of Pacheco, see my article “Don Alvaro de Luna and the Indictment against Royal Favoritism in late Medieval Castile,” chapter 11 in The Emergence of León-Castile c.1065–1500: A Festschrift for Professor Joseph P. O’Callaghan, ed. James Todesca (Farnham, England, 2015).

224

L. J. Andrew Villalon

civil war, Pacheco, now leader of the disloyal opposition, requested a parlay with the king. Although Enrique granted the request, while riding to the designated meeting place, he began to receive warning that his old friend planned to use the conference as a means of seizing his person. At this point, Bishop Mendoza, who had accompanied the royal entourage, volunteered to ride on ahead and see if the reports were true. Even before arriving at the rendezvous, he and his companion, the chaplain and chronicler, Diego Enriquez del Castillo, encountered new messengers bringing confirmation of Pacheco’s treachery. He now sent Castillo and some of their escort back to alert the royal party, which escaped to the safety of Segovia while he and a small group of horsemen calmly rode into the enemy camp, where he lectured an angry group of nobles concerning their conduct.40 It is certain, my lord counts, that what you have tried here today bears an ugly name. When your king, with faith in you as subjects and good vassals, comes to meet with you in hopes of bringing peace and ending discord, you come with thoughts of seizing his person.41

The bishop then remounted and with his small group calmly rode away from the frustrated nobles. Despite such small but telling exhibitions of personal courage, it would not be until three years later, at the one pitched battle of Enrique IV’s reign, that the bishop of Calahorra made his debut on the battlefield. His eldest brother, Diego Hurtado, who had succeeded their father as marquess of Santillana, was one of three leading nobles who raised the royal army that triumphed at Olmedo in 1467.42 Although the chronicles single out the marquess for his performance on that field, several of his brothers, including the bishop, figured prominently in the battle, fighting at his side.43 A decade later, during the civil war (1474–79) which brought the Catholic Monarchs to power, Pedro González de Mendoza, now cardinal-archbishop of Seville, cemented once and for all his military reputation. On almost all occasions when the principal chronicler of the reign, Hernando del Pulgar, names those who supported the royal couple in their greatest crisis, he places the cardinal’s name first on the list. Shortly after seeing to the burial of the old king,44 Mendoza hurried to join them, becoming one of the first to rally to their banner. He also rallied the members of his powerful house to support them.45 In 1475, 40 41 42 43

44 45

Castillo, pp. 136–37. Carvajal, pp. 217–19. Carvajal, p. 218. Castillo, pp. 162–63; Carvajal, pp. 295–300; Valera, pp. 124–28. Castillo and Carvajal credit Diego Hurtado with having entered the fray with particular gusto where, at one point, he rescued his hard-pressed son-in-law, Beltran de la Cueva, who was a special target of the enemy. On the other hand, Valera makes no particular mention of the marquess and attributes Cueva’s salvation to the possession of a good warhorse. Castillo, pp. 162–63, Carvajal, pp. 295, 300; Valera, p. 128. Pulgar, Crónica, 1:64. Ibid., 1:90.



Pedro González de Mendoza

225

when King Alfonso of Portugal laid claim to the Castilian throne in the name of Enrique’s daughter, Juana “la Beltraneja,” Bishop Mendoza vigorously aided Ferdinand in preparing northern Castile to resist a Portuguese invasion. The following year, when word reached court that the Portuguese army had crossed the frontier, the cardinal gathered his vassals and, along with his relatives, including his elder brother, the marquess of Santillana, rushed to the king’s side.46 As Ferdinand ran short of funds with which to pay his army, the bishop was one of those who melted down their plate and lent it to the royal treasury. In the events leading up to the battle of Toro on 19 March 1476 and during the battle itself, Mendoza displayed a Turpin-like prowess, enjoying his greatest success on the battlefield. Some weeks earlier, the Portuguese king had laid siege to the city of Zamora in northwestern Castile, where King Ferdinand had established his headquarters. The cardinal, who had been with the queen, immediately hurried forward with all the forces he could gather to join the beleaguered monarch. By the end of February, with supplies running low and the siege turning out to be much more trouble than it was worth, the Portuguese king decided to retreat to the neighboring city of Toro, which he already ­occupied.47 In order to facilitate his withdrawal, he sent messengers to Ferdinand, asking for a fifteen-day truce, offering in return to leave Zamora in peace and pull back his forces. In turn, Ferdinand took the matter before his council of war, which included a number of the realm’s leading nobles, and sought their opinion of the offer. Not a few favored granting the truce, arguing that the Portuguese, in having to request it, had already clearly demonstrated their weakness and that the retreat would simply increase the impression that they were losing the struggle. Those Castilian nobles who had adhered to his cause would then begin to drop away, eventually forcing him to leave the realm. Some of the nobles who took this position argued that not only should the truce be granted, but it should be open ended.48 At this point, Ferdinand turned to the cardinal and asked his opinion. His reply, as recorded by Hernando del Pulgar, stands very much in the tradition of Archbishop Turpin. With the army gathered and fully ready to fight, it would be not only dishonorable but unwise to permit an enemy already on the ropes to escape with impunity. I say it is not to be suffered, my lord, by any knight, much less a king as powerful as you that a foreign monarch come within his realm and lay siege where he wishes, then withdraw when constrained by necessity without incurring any damage. To allow this would be a demonstration of royal weakness and a violation of the laws of chivalry that forbid enduring such an injury when one has adequate force to avenge it.49

46 47 48 49

Ibid., Ibid., Ibid., Ibid.,

1:193. 1:201. 1:202. 1:203.

226

L. J. Andrew Villalon

In rendering this advice, the cardinal made clear to all that he spoke not as a man of the cloth, but as a warrior descended from a line of such men. “I, my lord, speak in this matter not as a son of religion or the bearer of this habit that I took on, but as the son of my father, the marquess of Santillana, who by his own exercise of arms and that of his forbears, achieved proficiency in the military discipline.”50 Faced with the risk of death or capture in a battle, Mendoza quipped, “I am far more disposed to see how God will treat my soul in the next life than [to see] what these Portuguese might do with my person in this one.”51 The cardinal’s arguments, as well as his humorous bravado, carried the day. The Portuguese messengers were summoned and informed by Mendoza, speaking for the crown, that the king refused to accede to any truce of even minutes, not to mention fifteen days, if it left matters as they were. Only if the Portuguese king agreed to end hostilities and evacuate Castile would any truce be possible.52 With this message in hand, the messengers returned to their monarch, who then lifted his siege of Zamora and began the retreat toward Toro, with the Castilian army in pursuit. As the two armies approached Toro, the road wound through a narrow defile between the hills and the Duero river “through which many men could not advance together, but only a few at a time (salvo pocos a pocos).”53 With the Portuguese having passed through, the Castilian force arrived and Ferdinand called another hurried council of war to decide whether or not he should send forward his advance units or return to Zamora now that he had clearly driven the enemy away from that city. Once again, opinion was divided. When some recommended withdrawal, Cardinal Mendoza offered to replicate his feat of a dozen years earlier, by riding forward with only a small escort to scout the enemy position. As a result, he and one other officer passed through the defile, eventually returning with word that the passage was clear and the enemy was awaiting, drawn up for battle outside the walls of Toro.54 Acting on the cardinal’s report, Ferdinand advanced through the passage, then drew up his forces and launched what would prove to be the decisive encounter of the civil war. During the fighting, Mendoza could be seen in the thick of the action “wearing his clerical habit over a coat of mail.”55 Within several hours, most of the Portuguese army broke and fled into the city, with the Castilians giving chase. There remained on the field only the battalion commanded by the Portuguese prince, which took up a position on a slight rise where the prince hoped to provide a rallying point for his father’s fleeing army. Here he stubbornly remained even after it became clear that the day was lost. Seeing this sole demonstration of defiance marring a complete 50 51 52 53 54 55

Ibid. Ibid., 1:205. Ibid., 1:206–7. Ibid., 1:209. Ibid., 1:210. Bernaldez, pp. 59–60. See also: Pulgar, Crónica, 1:215.



Pedro González de Mendoza

227

Castilian victory, the cardinal and the duke of Alba tried to rally the victorious Castilians for one last attack. Their effort, however, was in vain. With the approach of nightfall, the tired men, many of whom had not eaten all day, simply took possession of the field, where they settled down for the night. The following day, the cardinal and the duke gathered the army and set off after the king, who had already returned to Zamora as soon as the victory was secured, to make certain that no remnants of the Portuguese force would manage to seize the city in the army’s absence. In the days after the battle, defeated Portuguese fleeing from Castile encountered the implacable hatred of their enemy whose lands had been destroyed and plundered and women raped. Many of the fugitives were killed; not a few were castrated. At this time, the royal council debated whether or not to grant the survivors a safe conduct from the realm. Once again, the cardinal took a strong stand, this time against those who, hungry either for vengeance or profit, recommended that the defeated Portuguese be either massacred or enslaved. In the end, Mendoza convinced Ferdinand that the wisest course as well as the most chivalrous one was to be merciful and grant them the safe conduct they sought.56 Again, Pulgar quotes his words in the council of war: To kill those who have surrendered is better called an act of ugly vengeance than glorious victory. If you, knights, had killed these Portuguese in battle, it would be a knightly act, but if they surrender to you and you kill them, it will be reputed cruelty, offensive to the ways of Castilian nobility…. It is certainly a thing far removed from all virtue to kill the unarmed who now cannot defend themselves when we could not kill them in armed conflict. These Portuguese who now wish to return to Portugal, entered at the command of their king, and as they have acted within this realm, we would have done in theirs if our king had led us there.57

Cardinal Mendoza conceded that killing the enemy in the immediate aftermath of conflict, when tempers ran high, was to an extent understandable “for it is difficult to sheath the sword in the hour of anger.” “But it would be an inhuman thing for this fury to endure so long that after ten days those who came begging for mercy would be killed…. Let us strive to win and not think of vengeance, for victory is a thing for strong men, while vengeance is for weak women.”58 Although the civil war would drag on for three more years, until the two sides signed the treaty of Alcocovas in 1479, for all intents and purposes, it was decided that day by a military victory outside the walls of Toro, a victory in which our warrior churchman, Pedro González de Mendoza, played a key role. In contrast to Archbishop Turpin, Mendoza survived his Roncevaux. And although he would never again enjoy such a singularly heroic moment, he would live to fight again – and again – for over fifteen years. For it was only two years 56

57 58

Pulgar, Crónica, 1:219–20. It was in this speech to the war council that the cardinal alluded to the heroism displayed by his namesake, the first Pedro González de Mendoza, at Aljubarrota, who had been killed “fair and square” in the actual fighting. Ibid., 1:219–20. Ibid., 1:220.

228

L. J. Andrew Villalon

after the civil war ended that the war to conquer Granada (1481–92) began. Once again, the house of Mendoza played a major role in the conflict that would finally end the Castilian chapter of the Reconquista. And with the death of his elder brothers, the duke of the Infantado and the first count of Tendilla, the cardinal now became the acknowledged head of the family as well as being principal adviser to the crown. In the new war, the aging churchman’s role began to shift somewhat away from combat and toward logistics. Not that he abandoned the field entirely. Mendoza participated in several campaigns, including reprovisioning the captured city of Alhama in 1482,59 raids on Granadan territory in 1490,60 and the final siege of Granada in 1491.61 What is more, in 1484, while Ferdinand was in Aragon trying desperately, and in the end unsuccessfully, to raise money for a war with France over the lost provinces of Cerdagne and Rousellon, the cardinal assumed command of the Castilian field army operating against Granada and would have led its annual campaign, had not the king, frustrated in his other project, resumed command.62 Nevertheless, for the most part, Mendoza remained at the queen’s side, advising her and helping to organize the war effort. In 1483, when Isabel deemed it necessary to remain in the north and devote herself to problems other than the conflict, he remained with her.63 He also accompanied the queen during her visits to the front.64 Consequently, when Malaga surrendered in 1487, he was on hand to consecrate a cathedral church and help establish a law code for the newly conquered city.65 And at Baza two years later, he quickly placed his own vicar in the town, thus bolstering his claim to be reuniting it with the see of Toledo, to which it had formerly belonged.66 At one point, Mendoza offered to undertake the annual provisioning of Alhama at the head of three thousand horsemen that he and his relatives would supply, and to contribute the money necessary to pay the troops. In fact, he proposed to gather his men and proceed directly to Ferdinand’s camp; however, in the words of the chronicler, “his company was a great consolation to the queen and his counsel a great relief and remedy for the things which befell, so she did not give him permission to leave her.”67 Later that year, however, when word reached court of a Moorish relief force marching against the king, the 59 60 61 62 63 64

65 66 67

Ibid., 2:23. Bernaldez, p. 217. Ibid., p. 232. Pulgar, Crónica, 2:113–15, 118. Ibid., 2:73. Mendoza is specifically mentioned as having travelled with the queen to the siege camps at Malaga and Baza; however, it is a safe assumption, given his importance to her as an adviser, that he accompanied Isabel on other occasions as well. See: Pulgar, Crónica, 2:334–36, 418; Bernaldez, p. 199. Pulgar, Crónica, 2:334–36. Ibid., 2:429. Ibid., 2:196.



Pedro González de Mendoza

229

cardinal did hurry south to aid his royal master.68 He was prepared to do so again several years later when the queen put out a desperate call for men, at which time he offered to pay out of his own purse any cavalry that would join him in hurrying to aid the king.69 There is even some indication that Mendoza may have helped to care for and conduct to the frontlines foreign volunteers who took this opportunity to participate in the final crusade against Islam in Western Europe.70 On those occasions when the war effort took a turn for the worse, the cardinal made an added contribution by shoring up the queen’s confidence in the justice and efficacy of her cause. The best example occurred in 1485, when the count of Cabra’s advance guard was smashed while attempting to establish preliminary siege lines around Moclin. When news of this reverse reached court, Mendoza addressed the worried monarch in his calming manner: You must realize, my lady, that no conquest of lands or territories was ever undertaken where the victors were not upon occasion vanquished, for if there is no resistance during a conquest, it would better be said that possession was taken rather than acquired by war. Remember that the Moors are war-like men and possess harsh and mountainous lands, which none of your predecessors were able to conquer in the past, given the disposition of the ground which is its major defense. You, my lady, should give thanks to God because thus far you have been able to persevere more than any of them in this endeavor and that by his grace, you have acquired more cities and towns and lands in three years than your predecessors were able to win in 200.71

Not until the final surrender of Granada at the beginning of January 1492 did Pedro González de Mendoza finally relinquish his role of warrior cleric and take up his duties as a full-time churchman for the first time since the opening years of his career. The Warrior Churchman’s Non-military Service to the Crown As valuable as Mendoza’s military activities proved to be, his non-military services as adviser, diplomat, royal spokesman, and all-around troubleshooter at least equaled what he accomplished on the battlefield. Such services began during the reign of Enrique IV, and while there are many examples, including the fact that the cardinal was consulted on most matters of importance, we shall 68 69 70

71

Ibid., 2:199. Ibid., 2:276. The Granadan War took on something of an international flavor as adventurers flocked to Spain from many parts of Europe. For example, as early as 1483 a party of Swedish fighters joined the Castilian army. Three years later, a company of Frenchmen as well as English archers and men-at-arms led by the count of Escalas arrived in Castile. The count, long a supporter of Henry Tudor, was in search of new adventure following the latter’s accession to the English throne; he found it at the siege of Loja, where he was twice wounded. In treasury records for 1486, there is an entry specifying payment to one of the cardinal’s servants for bringing the Frenchman Marco de Pria to the frontlines. Cuentas de Gonzalo de Baeza, Tesorero de Isabel la Católica, ed. Antonio de la Torre and E. A. de la Torre, 2 vols. (Madrid, 1955), 1:126. Ibid., 2:195–96.

230

L. J. Andrew Villalon

mention only a few. In 1465, when opposition nobles gathering in the city of Ávila deposed Enrique IV in absentia and crowned in his place his younger halfbrother, Alonso, it was the bishop of Calahorra whom the monarch delegated to frame an official response. As royal spokesman, he skillfully laid out his arguments against deposition before a gathering of undecided nobles and helped to convince a number of these men to remain loyal.72 In 1467, the bishop was sent to meet with the count of Alba, a notorious waverer, when the latter failed to answer several summons to join the royal army.73 In 1469, when the royal court left the city of Cordoba, Mendoza remained behind to settle a quarrel between two of the region’s leading magnates.74 In 1470, the bishop played a major role in welcoming and winning over the papal legate, Rodrigo de Borgia, who would several decades later become Pope Alexander VI. (It would be Borgia who helped Mendoza obtain his appointment as the Castilian cardinal.) In 1474, with heavily armed aristocratic factions facing each other across a field and threatening a full-scale battle over the town of Carrion de los Condes, Mendoza, by now a cardinal, approached both sides and helped to negotiate a last-minute solution that headed off conflict.75 Later that same year, when a struggle broke out over the grand mastership of the military order of Santiago, he helped to arrange for one of the claimants to release a rival whom he had seized and imprisoned.76 During the civil war that followed Enrique’s death, it was the cardinal who tried, albeit unsuccessfully, to prevent his fellow prelate and long-time rival, the archbishop of Toledo, Alonso Carrillo de Acuña, from going over to the Portuguese side.77 Month after month, he accompanied the queen as she rode around Castile, drumming up support and putting down rebellion.78 When the Portuguese king, Alfonso, claimed the Castilian throne, Mendoza participated in drafting the Catholic Monarchs’ diplomatic reply that refuted Portuguese claims and called upon Alfonso to set aside his plans for an invasion.79 Later, when that first effort failed, he wrote letters to the Portuguese monarch, trying to arrange a last-minute truce and reopen negotiations.80 At the same time, he directed parallel negotiations with the French, aimed at preventing them from coming to the aid of Portugal.81 Finally, with victory imminent, Mendoza helped 72 73 74 75 76 77 78 79 80

81

The text of Mendoza’s argument is printed in: Memorias de Don Enrique IV de Castilla 1 (Madrid, 1835–1913), pp. 489–90 (document cxviii). Castillo, p. 166; Carvajal, p. 304. Castillo, p. 185; Carvajal, 352. Castillo, p. 218–20; Carvajal, pp. 445–47. Castillo, pp. 220–221; Carvajal, pp. 456–57. Pulgar, Crónica, 1:62–63. Ibid., 1:151–52, 158, 164, 266, 268–70, 417. Ibid., 1:96. The attempt to arrange negotiations foundered on Alfonso’s demand, as the price of abandoning his claims was not only a large indemnity, but also the Castilian cities of Toro and Zamora, and the region of Galicia from which Portugal itself had emerged. Ibid., 1:113–14, 149. Ibid., 1:252–53, 364–66.



Pedro González de Mendoza

231

to convince Ferdinand and Isabel that moderate treatment was the best way of reconciling those Castilian nobles who had fought on the losing side.82 In 1480, as peace returned to Castile, the Catholic Monarchs called the cortes of Toledo, the most important such meeting to be held during their reign. Here, the cardinal strongly supported royal measures aimed at restoring much-needed stability throughout the kingdom, to which end he helped push through the Redemption Act that reclaimed from the nobility, including members of his own family, a substantial amount of crown property alienated by Enrique IV.83 During the trying years of the Granadan War, Cardinal Mendoza helped receive and entertain ambassadors. In 1483 he and his nephew, the bishop of Palencia, joined the monarchs in welcoming a papal nuncio authorized to preach the bulla de cruzada and collect a clerical subsidy for the war.84 Again, in 1485 and 1486, when the pope authorized renewed subsidies, Mendoza took charge of their assessment and collection.85 When the couple’s fifth child, Katherine, was born in Alcalá de Henares in 1485, the cardinal, in whose archiepiscopal see the city was located, staged a magnificent tournament to celebrate the occasion.86 Five years later, he helped to arrange a match for the monarch’s eldest daughter, Isabel, with Prince Alonso of Portugal, and led the extensive escort that accompanied her to the frontier near Badajoz.87 It was these many nonmilitary services, even more than his accomplishments on the battlefield, that prompted contemporaries to dub Don Pedro “el tercer rey.” The Warrior Churchman as Scholar, Bibliophile, and Patron of Learning In addition to his royal service, the warrior churchman wore many other faces. As mentioned earlier, Don Pedro defies the stereotypical image of the uneducated aristocrat occupying a place in the upper clergy due to his family’s wealth and social position, an image not uncommonly encountered in the literature of the period, especially when it manifests an anticlerical bent. Many years ago, in a classic study, J. H. Hexter explored the manner in which aristocrats of the early modern period bought into a literate culture.88 While Hexter centered his attention on English aristocrats of a slightly later date, his observations fit many of their peers throughout Western Europe, starting in the later Middle Ages. From at least as early as the fourteenth century in Italy and radiating out from there, an ever-growing sector of the second estate came to see a need for

82 83 84 85 86 87 88

Ibid., 1:221. Ibid., 1:415–21. Ibid., 1:50. Ibid., 1:189, 243–44. Despite her birth in Castile, she is known in English history as Katherine of Aragon, later the wife of Henry VIII, whose divorce would touch off the English Reformation. Pulgar, Crónica, 2:440; Bernaldez, pp. 221–22. J. H. Hexter, “The Education of the Aristocracy in the Renaissance,” in Reappraisals in History (London, 1961), pp. 44–70.

232

L. J. Andrew Villalon

literacy, not only for practical reasons, but as a point of pride. This was especially true of families that laid claim to a heritage of intellectual achievement. The Mendozas were just such a family.89 Closely aligned as they were to the house of Ayala that had produced Castile’s foremost chronicler of the later Middle Ages,90 a Mendoza tradition for learning began to take root even before the first marquess of Santillana. The cardinal’s eldest brother, Diego Hurtado de Mendoza, first duke of the Infantado, voiced the family’s intellectual ethos when he entailed his library in order to preserve it for future generations. I bequeath to my son, the count, in mayorazgo … the books which will be found in my library and my room, and it is my will that neither he nor his successors be able to alienate them. Instead, they shall remain, like other entailed goods, within the mayorazgo … since I very much desire that he and his descendants will devote themselves to the study of letters … as did my father, the marquess [of Santillana], and myself, and many of our ancestors, believing sincerely that in so doing we would better ourselves and our line.91

Under the circumstances, it is not surprising that care would be taken in educating a family member such as Pedro González who was destined for a life in the Church. By the age of fourteen, the future cardinal could read and write Spanish and Latin, skills afterwards refined by three years of living and studying in the household of his cousin, Gutierre Alvarez de Toledo, then the archbishop of Castile’s primate see.92 In 1446, Don Pedro entered Castile’s foremost university at Salamanca, where he read canon law, while continuing his study of 89

90

91

92

Professor Nader’s work lavishes praise upon the cultural activities of the house of Mendoza, crediting the family with almost single-handedly inspiring what she has called “the caballero Renaissance of the fifteenth century.” Pedro López de Ayala (c. 1332–1407) was a warrior, diplomat, and royal official who became, in his later years, not only a poet but also the leading chronicler of late medieval Spain, producing accounts of the four monarchs under whom he served: Pedro I “the Cruel,” Enrique II “the Bastard,” Juan I and Enrique III (the latter work remaining unfinished at the time of the chronicler’s death.) The first biographical sketch of Ayala dates to the early fourteenth century. See: Fernán Pérez de Guzmán, Generaciones y Semblanzas (Madrid, 1965), pp. 37–39; Guzmán, Generaciones y Semblanzas (Buenos Aires, 1947), pp. 30–31. For a modern, full-length treatment, see Michel Garcia’s Obra y personalidad del Canciller Ayala (Madrid, 1983.) This work appends several of the scarce contemporary sources that supply much of our information concerning Ayala himself. English readers should consult Nader’s third chapter in The Mendoza Family, entitled “Pedro López de Ayala and the Formation of Mendoza Attitudes,” which provides a fine capsule biography of the great chronicler, a penetrating analysis of his chronicle of Pedro I and a valuable bibliography. See also: Clara Estow, Pedro the Cruel, 1350–1369 (Leiden, 1995), esp. xiv–xx; Estow, “Royal Madness in the Crónica del Rey Don Pedro,” Mediterranean Studies 6 (1996), 13–28; and Constance L. Wilkens, “López de Ayala, Pero,” in Medieval Iberia: An Encyclopedia, ed. E. Michael Gerlí (New York, 2003), pp. 515–16. The will of the first duke, entailing the library, can be found along with other Mendoza wills in AHN, Osuna 1762. The will is substantially reprinted in: Layna Serrano 2, appendix, pp. 465–74. Chroniclers refer to Don Gutierre as Mendoza’s uncle (tio); he seems, however, to have been an elderly cousin.



Pedro González de Mendoza

233

history and the classics. Here, Mendoza’s proficiency in Latin reached a point where he could translate into the vernacular several ancient authors, including 93 the poets Virgil and Ovid and the historian Sallust. From these student days comes a rare personal letter written to him by his father, the marquess of Santillana. Despite the latter’s many literary achievements, he appears never to have mastered Latin, his knowledge of ancient literature having come to him by way of translations. The aging nobleman remarked philosophically “when we cannot have what we want, we must want what we can have.” He could have translations, and so he besought his better-educated son to translate into Spanish a Latin version of the Iliad recently sent to them from Italy.94 When Mendoza left Salamanca to take up his duties in the royal chapel, the affairs of church and state increasingly absorbed his time; as a result, his life as an active scholar came to an end. Despite this, he never lost that early interest in classical learning. Decades later, the cardinal became one of a handful of Spaniards mentioned in the memoirs of the greatest Renaissance bookseller, Vespasiano di Bisticci,95 who would praise his learning and extoll his librarybuilding efforts. 93

94

95

In his Vida del Cardenal, Mendoza’s first biographer, Medina y Mendoza states “I have had in my possession various handwritten books translated by him, dedicated to his father, the marquess, in order that he could read them not in Latin, but in Castilian (castellano), including a volume of Ovid and the Aeneid of Virgil. His Castilian prose was very clear and elegant, demonstrating both his understanding and his eloquence.” Medina y Mendoza, Vida, pp. 156–57. Although Medina y Mendoza writes of the cardinal’s university career, it is Salazar de Mendoza who supplies the date of his matriculation. See: Salazar de Mendoza, Crónica, p. 66. Several sixteenth-century writers, including Lucio Marineo Siculo, have asserted that the marquess of Santillana did know Latin, a claim which more recent writers have echoed. See, for example, the prologue to Augusto Cortina’s Obras del Marquess de Santillana, 12: “Despite his native language, he knew Latin (more the ecclesiastical than the literary), Italian, French, Provencal, Galician-Portuguese, and Catalan.” However, the poet’s own testimony appears to contradict these statements. In the letter to his son, which Cortina places among Santillana’s Obras, the marquess expressed regret at having failed to master the ancient language, but viewed his failure in this regard philosophically. Medina y Mendoza bears out the letter when he states that the marquess asked his son, the future cardinal, to undertake his translations “because he read Castilian, but not Latin.” Medina y Mendoza, Vida, p. 156; Santillana, Obras, pp. 43–44. Born near Florence, Vespasiano di Bisticci (1421–98) rode the wave of interest in things Greek, making his name selling copies and transcriptions of Greek manuscripts currently entering the Florentine market from Constantinople. Eventually, his establishment in Florence became the meeting place for Renaissance bibliophiles, Italian or otherwise, a fact which accounts for his acquaintanceship with most of the great book collectors of his day. After retiring, he wrote short biographies of 105 of them, called collectively the Vita di uomini illustri del Secolo XV. The work has been translated into English as Renaissance Princes, Popes, and Prelates: The Vespasiano Memoirs, Lives of Illustrious Men of the XVth Century, translated by William George and Emily Waters (New York, 1963): for the sketch of Cardinal Mendoza, see pp. 155–56. Although the bookseller mentions the cardinal as having spent time in Rome, there is no confirming evidence that I can find. The error probably results from Vespasiano’s having conflated the leader of the family with his nephew and protogée, Diego Hurtado de Mendoza, who did visit the city.

234

L. J. Andrew Villalon

Messer Piero de Mendoza, of noble Spanish birth, was made cardinal by Pope Sixtus by reason of his merit. He had a thorough knowledge of the Canon Law; also of the humanities and of theology, and … caused many books to be transcribed, both in sacred and secular learning, in view of forming a library.96

His life-long fondness for learning manifested itself in the cardinal’s most important charitable foundation – the College of Santa Cruz, devoted to the study of theology and canon law.97 In May 1479, having received a papal license from Sixtus IV, Mendoza selected as the site of this college the northern city of Valladolid,98 where he began construction the following year. In November 1483, well before the building’s completion, the college officially opened its doors in several nearby houses purchased from the Bishop of Segovia to serve as temporary quarters. Construction went forward under the direction of the cardinal’s secretary, Diego de Muros, reaching completion in 1492, the year he retired from royal service. Ultimately, the College of Santa Cruz would house a faculty of twenty, including six theologians, nine canon lawyers, two legists, and three medical doctors, as well as a staff of eight to serve its needs. Students, each of whom enjoyed a small private room, were allowed to move about the city, but only in pairs and wearing academic robes and hoods. According to the cardinal’s chronicler, Salazar de Mendoza, the great churchman’s initial reaction to the building was one of disappointment, so much so that he even contemplated tearing down his creation and starting afresh, had not the Catholic Monarchs dissuaded him. By contrast, in describing the college, the chronicler leaves no doubt that he himself believed it to be an elegant edifice: On all sides, it is of plain ashlar masonry, very white, very well laid out, and very strongly built. Around the top of the structure, there is an entablature composed of many large, separately carved stone tablets and above that stands a parapet with balusters of hewn stone.… The facade is highly adorned and composed of many square pillars … and the principal portal is highly ornate. There are three embellished columns on either side and carved figures and coats of arms. The windows are well placed and well proportioned. Most of this well laid-out edifice is in the Gothic style, though combined with much that is Roman. An elegant plaza, surrounded with stone pillars, where bulls can be run and jousts fought is located in front of the college. A door from the large patio leads to a spacious, ornamented chapel. Equally majestic are the library, refectory, lecture halls, living quarters, and offices, both large and small … as well as the vegetable and flower gardens.99

In 1501, six years after the cardinal’s death, the Flemish noble Antoine de Lalaing, seigneur of Montigny, wrote a highly detailed account of his journey 96

97 98 99

Although Vespasiano knew of the family’s elevated social position in Spain, he was apparently unaware of the marquess of Santillana’s prominence as a poet, characterizing the latter as “not given to letters.” On the other hand, this may also have reflected an Italian prejudice with regard to those who did not function in Latin. Salazar de Mendoza, Crónica, pp. 262–66. The license authorized establishing a college at either Valladolid or Salamanca. Ibid. Salazar de Mendoza, Crónica, pp. 265–66.



Pedro González de Mendoza

235

to Spain in which he echoed the chronicler’s praise for Mendoza’s work as a builder.100 Cardinal Mendoza … founded at Valladolid [a] college which is all new and one of the most beautiful to be seen. There are two or three gilded chambers, like those of a bishop. Nothing is lacking. [The] scholars study medicine, “physick,” canon law (decrets), and other branches of knowledge. Their library exceeds others in beauty and richness.101

Despite the importance of his college, in the long run, Mendoza’s indirect influence on Spanish education would prove even greater. During the 1460s he became the patron of a young cleric named Francisco Jiménez de Cisneros who was at odds over the possession of a benefice with Mendoza’s major ecclesiastical rival, Alonso Carrillo, archbishop of Toledo and primate of the Castilian church.102 In what was otherwise a highly unequal contest, Mendoza extended his protection to Cisneros. The latter would eventually enter the Franciscan order and, at Mendoza’s recommendation, become the queen’s confessor. Ultimately, Cisneros succeeded both his nemesis and his protector as archbishop of Toledo and was appointed cardinal of Spain. And it would be Cardinal Cisneros who late in his career built the University of Alcalá into one of Europe’s most famous educational institutions. Mendoza’s lifelong predilection for learning and for learned men like Cisneros proved to be one of his most fruitful virtues. The Warrior Churchman as Ecclesiastical Pluralist Unfortunately for the Church, Mendoza’s virtue as a servant of the state was not equaled in the ecclesiastical sphere. This was the heyday of pluralism and the cardinal would ultimately rank among its leading practitioners, standing alongside such figures as Thomas Wolsey in England and Albrecht von Hohenzollern in Germany. By 1468, when Mendoza exchanged Calahorra for the more important see of Siguenza,103 he had begun to amass what would become a breath-taking array of benefices. In short order, he became abbot of two monas-

100

101 102

103

In 1501, the Catholic Monarchs’ eldest surviving daughter, Juana (widely known by the unfortunate sobriquet “la loca” or “the mad”), and her Hapsburg husband, Philip, journeyed to Spain. Among the noblemen who traveled in their entourage was Antoine de Lalaing, seigneur of Montigny. His detailed account of their visit provides much useful information near the end of the joint reign of the Catholic Monarchs. See: Lalaing. As a result of this marriage, the Hapsburg prince would briefly occupy the Castilian throne as King Philip I, between the death of Queen Isabel in 1504 and his own premature demise two years later. Lalaing, 1:167. Juan Meseguer Fernández, OFM, “El arzobispo Carrillo y el Cardenal Cisneros,” Archivo Ibero-Americano 45 (1985), 167–88; Angus MacKay, “Jiménez de Cisneros, Francisco,” Medieval Iberia: An Encyclopedia, ed. E. Michael Gerli (New York, 2003), pp. 440–41. Castillo, pp. 170–1. Carvajal, p. 319. Medina y Mendoza, Vida, 179–81. Salazar de Mendoza, Crónica, pp. 121–24.

236

L. J. Andrew Villalon

teries, one in Valladolid, the other in Carrion.104 In 1473, he attained both the archbishopric of Seville and the red hat of a cardinal, without surrendering any of his other ecclesiastical positions, including the see of Siguenza.105 Later in the 1470s, while he was conducting diplomatic negotiations with France, the French king conferred upon him a prestigious abbey at Fiescamp in Normandy.106 And in 1482, Pope Sixtus IV made special provision for him to hold yet a fourth such office, the Cistercian abbey of Santa María de Morevuela.107 Finally, in 1483, Don Pedro capped the acquisitorial side of his church career by securing the primate see of Toledo after the death of its previous holder, Alonso Carrillo.108 To achieve this final advance, he had only to give up the archdiocese of Seville, having been permitted to retain the rest of his benefices. The other famous pluralists of the age had nothing on the Spanish cardinal. These vast ecclesiastical holdings returned an income second only to that of the monarchy. The Flemish traveler, Lalaing, has left some figures for 1501, figures that would not have changed all that much since the cardinal’s death.109 At the time, the see of Toledo yielded approximately 52,000 gold florins and that of Siguenza 14,000, for a grand total of 66,000 gold florins.110 Lalaing’s figure does not include the income Mendoza must have derived from his possession of four abbacies;111 nevertheless, even without taking this tidy sum into account, the cardinal’s annual revenues would have surpassed those garnered by any of Spain’s lay nobles, the two most richly endowed of whom were the dukes of Frias and Medina Sidonia, whose incomes the Flemish traveler placed at 62,000 and 61,000 gold florins, respectively.112 Undeniably, pluralism could pay big dividends! The Warrior Churchman as Nepotist If pluralism and blood-letting numbered among Mendoza’s “cardinal sins,” so too did nepotism. Over the decades, Cardinal Mendoza helped to launch and 104

105 106

107 108 109 110 111 112

Salazar de Mendoza points out that a number of bishops had held the abbey in the recent past, alleging that “in that century, it was permitted to hold incompatible benefices.” Salazar de Mendoza, Crónica, pp. 125–7. Castillo, p. 216. Carvajal, pp. 432–3. Medina y Mendoza, Vida, pp.193–94. Salazar de Mendoza, Crónica, p. 135. The abbey of Fiescamp was located on the Somme River near the port city of Dieppe. According to Salazar de Mendoza, the Cardinal did not actually derive any income from the abbey, but instead transferred the revenues to the old abbot in Rome, who had been dispossessed by Louis XI. Salazar de Mendoza, Crónica, pp. 63–64. Ibid., pp. 180–82. Medina y Mendoza, Vida, pp. 248–49. Salazar de Mendoza, Crónica, pp. 180–82. Lalaing, 1:231–37. Ibid., 1:231–32. Lalaing states that “concerning [the income of] abbeys, I could find nothing.” Ibid., 1:232. The other Castilian nobles who appeared highest on Lalaing’s income list were the duke of Alba, with 44,000 gold florins; the count of Benavente, 44,000; the head of the Mendoza family, the duke of the Infantado, 40,000; and the duke of Medinaceli, 40,000. Ibid., 1:233.



Pedro González de Mendoza

237

then advance the religious careers of many relatives, chief among them his nephew, Diego Hurtado de Mendoza, a younger son of his brother Iñigo López de Mendoza, first count of Tendilla. While still bishop of Calahorra, Mendoza took Diego into his household. Here, the boy received a sound education befitting an individual who was being groomed to become the family’s next great churchman. In 1458 the younger Mendoza accompanied his father, the count, on a diplomatic mission to Rome;113 then, after a period of further training at the Holy See, he returned to Spain and reentered his uncle’s household. Ultimately, Cardinal Mendoza would use his growing power within the Church and the monarchy to assure his nephew’s success. In 1481, while still archbishop of Seville, he obtained for Diego the lucrative diocese of Palencia.114 In 1483, on the death of Alonso Carrillo, he himself acquired the archbishopic of Toledo and two years later, when his successor in the see of Seville died suddenly, he lobbied the crown to obtain the position for his nephew. There were powerful competitors on the scene. In Rome, Pope Innocent VIII had made provisions for Mendoza’s old friend, the papal vice-chancellor, Rodrigo Borgia, to hold the archbishopric in commendam. For his part, King Ferdinand wanted the archbishopric for his illegitimate son, Alonso of Aragon. In the end, however, the cardinal prevailed when both of the Catholic Monarchs threw their support to his nephew, arguing with the papacy that such an important see should not be placed in the hands of an absentee foreigner.115 Rome relented and Diego Hurtado de Mendoza became the new archbishop of Seville, with annual revenues of 23,000 gold florins.116 Finally, during the closing years of his life, Mendoza positioned his nephew to succeed him as cardinal.117 While Diego provides the most spectacular instance of nepotism, others among the cardinal’s relatives benefitted from his ecclesiastical clout. To another nephew, Garcia, son of the duke of the Infantado, he transferred the abbacy of the monastery in Valladolid, while a cousin, Luis Hurtado, received that of Carrion.118 Since the eleventh-century reform movement, the Church had had in place strict anti-nepotism statutes; despite which it habitually turned a blind eye to this besetting clerical sin. Whatever the attitude of the Church, a medieval churchman’s family considered nepotism both desirable and necessary. We can witness an example of this attitude in the will drawn up in June 1475, by the cardinal’s eldest brother, Diego Hurtado de Mendoza, first duke of the Infantado. Noting the close connection between himself and the cardinal, the 113 114 115

116 117 118

Castillo, p. 112. Medina y Mendoza Vida, pp. 242–43. Salazar de Mendoza, Crónica, pp. 134–35. Medina y Mendoza, Vida, pp. 255–57. Apparently, Ferdinand realized that the case for his son was badly weakened by a combination of the boy’s illegitimacy and the facts that he was underage and, as an Aragonese, could also be classified as a “foreigner” in respect to the Castilian church. Lalaing, 1:231. Medina y Mendoza, Vida, p. 300. Luis Hurtado de Mendoza was a brother of the count of Castro, Alvaro de Mendoza. Medina y Mendoza Vida, p. 306. Salazar de Mendoza, Crónica, p. 417.

238

L. J. Andrew Villalon

aging duke, who was soon to fight in his last campaign, called upon the cardinal to look after his family once he was gone.119 In particular, he commended to his brother’s care an illegitimate son named Alfonso: “since I desire my son to become a cleric, I beg my brother the cardinal to receive the boy into his household, and bestow upon him some benefice.”120 The duke could rest easy, knowing that his brother would undoubtedly honor the request. The Warrior Churchman as Estate Builder As extensive as his ecclesiastical revenues ultimately became, they constituted only part of the cardinal’s total wealth. Over the years, he amassed a vast personal fortune, the great majority of which came not from inheritance or even from the Church, but from the crown, which he had served so long and faithfully. In 1455, when Mendoza’s father, the marquess of Santillana drew up his last will and testament, the recently appointed bishop had already embarked upon his career in the Church; consequently, the old man saw no need to make extensive provision for this younger son.121 Aside from a share of the divisible goods, the bishop received only two small parcels of real estate, identified as the places (lugares) of Monesterio and Canpello, both of which had once belonged to his mother. He was also excluded from several divisions of family money. In the will, the marquess went on to limit the young man’s control over even the small portion of the inheritance that he had been given. Since Don Pedro, as a cleric, could not leave any legitimate sons or daughters, he was to hold the properties only during his lifetime, after which they would revert to whomever had inherited the family’s principal entailed estate.122 When the marquess died in 1458, his sons drew up an agreement to make minor alterations in the property division that, among other things, somewhat increased the bishop’s share.123

119

120 121

122 123

A dignified though poignant passage in the will contains this request which clearly shows the nobleman’s trust in his younger sibling: “Although I could not ask the reverend lord cardinal, my brother, due to his status, to be an executor of my will, as I asked him to be an executor in all other such documents drafted before he was made a cardinal, I beg him out of the great love there has always been between us, that he will take cognizance of the terms of my will and see that they are fulfilled … and I strongly recommend to his care my wife, the marquise, and my children, especially those who are still young and therefore cannot guide themselves. AHN, Osuna 1762: Will of the first Duke of the Infantado (1475), in Layna Serrano, 2:473–74. AHN, Osuna 1762. Layna Serrano, 2:47l. In the warrior-poet’s words: “I … command my son … to be content with that part of my property which he will ... inherit even though it will not be as large a part as would normally go to him.” AHN, Osuna 1762: Will of the marquess of Santillana (1455). The document is reprinted in Layna Serrano, appendix, 1:316–24. Ibid., 1:319–20, 322. By the terms of this supplementary document (dated 9 May 1458), the eldest brother, Diego Hurtado de Mendoza, agreed to turn over to each of his surviving siblings, including the bishop of Calahorra, lands containing three hundred vassals, in satisfaction of their share of the inheritance. The bishop received the places (lugares) of Pioz, Pozo, Retuerta, Yelamos de



Pedro González de Mendoza

239

Nevertheless, Don Pedro’s inheritance remained that of a younger son who had entered the clergy and was therefore expected to live primarily on his clerical income. Most of the cardinal’s personal fortune would come in later years from the monarchy that he had served so well. He received several significant grants from Enrique IV, including one issued in 1469 permitting him to collect certain royal rents (the tercias reales) in the city of Guadalajara.124 However, his major concessions from the crown dated to the reign of the Catholic Monarchs and involved the spoils from the conquest of Granada. In March 1490, the crown separated from the newly conquered city of Guadix eight towns and villages, known collectively as the Zenete,125 and in a pair of grants transferred these properties to the cardinal.126 In both of these, the monarchs carefully specified that the lands were to go “to you personally, rather than to any of the churches or ecclesiastical offices which you occupy” – an obvious precaution against the Church laying claim to the Zenete following the cardinal’s death. Influence with the crown could also bring wealth by means other than a direct royal grant, as one can see in the case of the town of Maqueda. Beginning early in the 1460s, Enrique IV’s secretary, Alvar Gómez de Ciudad Real, began to prosper from the favor bestowed upon him by the king to such an extent that he was able to amass various properties, most important among them the town of Maqueda. Despite this, when rebellion broke out against Enrique, the royal secretary joined the rebels, on one occasion convincing several high nobles on their way to join the king that he intended to seize them, as a result of which they too went over to the other side. Having learned of this betrayal by his trusted and well-rewarded servant, Enrique determined to make an example of him.127 At a critical moment, Mendoza intervened. Immediately following the royal council meeting in which the decision was made to arrest the secretary, the bishop hurried to his lodgings, telling him what had happened and suggesting that he leave at once for the nearby town of Buitrago, long a possession of the house of Mendoza, where he would be safe. No second warning was needed; consequently, when royal marshals arrived, they found that their quarry had flown the coop.128

124 125 126

127 128

Suso, all located in the territory of Guadalajara, and the places of San Agostin and Pedrezuela in the nearby bishopric of Segovia. A substantial excerpt from this “Iguala y avenencia entre los hijos del primer marquess de Santillana” is also printed in the appendix to Layna Serrano, 1:334. AHN, Osuna 1862 n. 3–6. In March 1490, the crown separated the towns of Alquise, Andeyr, la Calahorra, Ferreyra and Dolar from the jurisdiction of the recently conquered city of Guadix. AHN, Osuna 1893 n. 1. AHN, 1887 n. 2; 1893 n. 1. Symbolic of their transfer was the fact that henceforward these towns would no longer have to take their civil and criminal cases before the city magistrates of Guadix, nor would they pay taxes or perform services that the city might require of those living within its jurisdiction. Valera, p. 94. Carvajal, pp. 229–30. According to the cardinal’s chroniclers, Gómez eventually moved to Guadalajara, where he

240

L. J. Andrew Villalon

Mendoza’s involvement in the affair gave rise to a story which, whether true or apocryphal, aptly illustrates the arrogant self-assurance of a man in his elevated position. At the next council meeting another prelate made a pointed remark about the disloyalty of anyone who would help a criminal to flee from royal justice. The bishop replied in a quiet voice: I understand. You are speaking of me because I warned Alvar Gómez to look to his safety. It’s true, I warned him, because he is my friend.… If you, my lord, had done it, they would cut off your head. But not mine; and that is the difference between you and me.129

Mendoza realized full well that his support and that of his family were too important to risk by overzealously pursuing the fugitive. And indeed, having learned the source of the warning, the beleaguered monarch quietly let the matter drop. Although his actions had saved the disgraced royal secretary at least from imprisonment and possibly worse at the hands of an angry monarch, such protection had its price tag. Several years later, the former secretary found himself compelled to make a disadvantageous “trade,” in which he gave his new protector the town of Maqueda and various royal rents bestowed on him in better days in return for the less valuable properties Mendoza had inherited. In November 1469, the bishop turned around and exchanged these new acquisitions for one of his most important secular holdings – the town of Jadraque with its fortress of the Cid.130 The Expenditures of a Warrior Churchman The cardinal-archbishop used his princely income in various ways. A fair percentage of his revenues were consumed in supporting the functioning of a large clientele, both religious and secular. Not surprisingly, in the case of a warrior churchman, raising, equipping, and placing in the field a military force that might on occasion number several thousand entailed a major expenditure, one that could test the resources of even the great cardinal. In addition, Pedro González de Mendoza lived as befitted a great medieval noble – magnificently. The cardinal’s splendid palace in Guadalajara, a city dominated by the house of Mendoza, serves as an example of his lifestyle. Jerome Munzer, a German who visited Spain in the 1490s, left the following account of this building:

129 130

established himself in the shadow of the bishop’s palace. Salazar de Mendoza, Crónica, p. 165. Medina y Mendoza, Vida, p. 186. Salazar de Mendoza, Crónica, pp. 166–67. The contract setting forth the details of this trade is dated 2 November 1469. This, as well as several other relevant documents, is contained in: AHN, Osuna 1703 n. 3. In return for both Maqueda and the alcaldia mayor of Toledo, Mendoza received Jadraque, the fortresses of the Cid and Corlo, and other land necessary to bring to 1,100 the total number of vassals turned over to him. See also: Medina y Mendoza, Vida, p. 186. Salazar de Mendoza, Crónica, p. 422.



Pedro González de Mendoza

241

It has two most beautiful galleries, one above the other, off of which are rooms and chambers, all with gilded and paneled ceilings ... with each panel different from the others. Two summer rooms open onto the garden with marble columns and there is so much gold about that it is hard to believe. Oh, what a majestic chapel, large although not very broad, on the altar of which there are exquisite paintings of Saint Peter and Saint Paul and of the blessed Virgin; and on the sides of which are pictured Saint Gregory and Saint Helen with the cross, from which he took his title. [Mendoza was Cardinal of Santa Cruz.] There is a most beautiful garden in the center of which stands a fountain watering it. There is also an immense aviary, enclosed with copper wire, which contains so many kinds of birds that it is impossible to describe them all. There were turtle doves and pigeons of various kinds from both Spain and North Africa; there were numerous African hens which were black with white spots, like dice, and which had hard, grey crests, very long legs, and a short tail. There were partridges of various species, many ducks of a dark, purplish color, with black tails and beaks, beautiful cranes with white crests behind their heads, and a large number of others.131

The German visitor went so far as to say, “I have seen the many great palaces of the cardinals in Rome, yet never in my whole life have I seen one so spacious and with the rooms so well distributed!”132 On the other hand, in addition to living en noble, the cardinal devoted much of his wealth to those charitable activities so typical of great nobles and princes of the Church. In addition to the college in Valladolid, he established a hospital in Toledo to treat the sick and care for foundlings.133 Although he died before this project got far beyond the planning stages, the queen whom he had served, with some help from the pope who had been his friend,134 saw to its completion, for the most part using monies he had bequeathed for the purpose.135 Mendoza left a fund to supply dowries for orphan girls and a large sum to ransom Christian captives. He also provided for vast amounts of grain to be distributed to the “deserving poor.”136 In each diocese where he had held a church office, he contributed generously to building new churches or repairing old ones.137 During his lifetime, he made extensive contributions to the convent of the Blessed Virgin of Sopetran near the town of Hita, and, by the terms of his will, left the convent money to rebuild its church. At Guadalajara, he rebuilt the Church of Our Lady outside the walls and contributed to the repair of the Church of Santa María de la Fuente. In Calahorra, the cardinal generously endowed the town fund for the upkeep of religious buildings, as well as constructing outside of the walls the new Church 131 132 133 134 135 136 137

Munzer, Viaje por España y Portugal, 1494–1495, 116. Lalaing left a comparable, if somewhat less elaborate, description of this palace. See Lalaing, p. 228. Ibid. Salazar de Mendoza, Crónica, pp. 381–95. Rodrigo Borgia, who had visited Spain in the 1470s as papal legate and helped secure for Mendoza his capelo, ascended the papal throne as Alexander VI in 1492. Salazar de Mendoza, Crónica, pp. 382, 395. Ibid. Ibid., pp. 402–6.

242

L. J. Andrew Villalon

of Our Lady de los Huertos. To Santo Domingo de la Calzada, he bequeathed a generous building fund. His legacy to Seville consisted in reconstructing of the parochial church of Santa Cruz and repairing the monastery of San Francisco. Finally, in Toledo, he built the parochial Church of San Juan de la Leche and in the cathedral redecorated the chapel of St. Peter and the cloisters. To many places, Mendoza supplied clerical vestments and religious ornaments.138 Nor did his ecclesiastical bequests stop at the boundaries of Castile: he donated to the rebuilding and adornment of his cardinalate church in Rome, and to the hospital attached to it. And through the Venetians, he contributed to work on the Holy Sepulcher in Jerusalem.139 While much of what the cardinal spent on these acts of charity came out of his ecclesiastical purse, at least some of the funds were taken from his personal fortune, including a certain amount originally ear-marked for heirs. For example, in 1494, through a codicil to his will, the cardinal removed from the inheritance of his eldest son, Rodrigo, a number of houses in the city of Guadalajara, the subsequent sale of which helped to provide funds for the hospital he was building in Toledo.140 In that same year, he removed from the inheritance of his second son, Diego, the substantial sum of four cuentos which had originally been earmarked for the purchase of income-producing property and redirected them to his charitable undertakings.141 “Los Pecados del Cardinal” The main reason that Mendoza, despite his a vast clerical income, felt compelled to amass a personal fortune resulted from his most ostentatious failing as a cleric. Like many another churchmen of the period, Don Pedro sired several children, whom his contemporaries dubbed affectionately “los pecados del Cardinal” (the sins of the cardinal). He acknowledged and provided for three sons – Rodrigo, Diego, and Juan. In the mid-1460s, a burgeoning scandal engulfed Enrique IV’s Portugueseborn queen,142 and negatively affected the fortunes of the young women who had accompanied her to Castile. Although some found the Castilian husbands they had been promised, others were left to fend for themselves. One of them was

138

139 140 141 142

For example, to the Church of the Cross in the Toledan parrish of Saint Nicholas, Mendoza contributed religious ornaments chalices, and “all that was necessary” for conducting religious services. Ibid., p. 403. Ibid., p. 406. AHN, Osuna 1878 n. 1. AHN, Osuna 2021. Starting not long after her arrival in the Castilian court, Juana of Portugal began to earn the reputation of a “wanton woman,” due at least in part to the propaganda put forth by Enrique’s opponents. Consequently, during the 1460s, the rumor arose that her daughter had been sired not by the king, but by the royal favorite, Beltran de la Cueva, hence her designation by Spanish historians as “Juana la Beltraneja.”



Pedro González de Mendoza

243

Mencía de Lemos.143 Enter the virile young bishop of Calahorra, by now in his early thirties, a major figure at court, and by no means immune to temptations of the flesh. Following what his earliest chronicler refers to as “the customary practice of the period,” Mendoza began to court Doña Mencía, who became his mistress. Later, when the realm was plunged into civil war, and the queen was sent back to Portugal, he removed the lady to the safety of a family stronghold, and there sired his two eldest sons, Rodrigo144 and Diego de Mendoza. Some years later, he had a second romantic liaison with one Inés de Tovar, resulting in the birth of a third son, Juan de Mendoza.145 During the final decades of his life, Mendoza devoted much time and effort to providing for these offspring; especially for the two eldest. Having amassed the necessary property, he secured from both the Church and the crown legitimation of the young men and permission to leave them his ever-increasing store of worldly goods. Mendoza’s first papal license appears to have been granted as early as 1478 by Sixtus IV. In October 1486, Innocent VIII (himself a confessed father146), issued a bull legitimating the cardinal’s brood. There followed two other bulls, dated December 1486, and January 1488, as well as an apostolic letter in October 1487, which together permitted Mendoza to bequeath his property as he saw fit.147 During these years, Cardinal Mendoza also approached the crown, which extended similar permission. As early as 1484, the Catholic Monarchs affirmed the churchman’s right to bequeath the royal rents in Guadalajara bestowed on him by their predecessor. Soon afterwards, Queen Isabel added her legitimation of the three sons to that already acquired from the papacy. Finally, in May 1487,

143

144

145 146

147

In the words of Medina y Mendoza “after the affairs of King Enrique began to unravel and Queen Juana was detained, all of her ladies found themselves on their own.” According to the chronicler, “I have met people who knew Doña Mencía when she lived in Manzanares who said she was a most beautiful and gentle lady, gracious and endowed with great spirit.” Bishop Mendoza, who was then thirty-two years of age, was taken with her and “according to the usages of that time, courted and served her when she lived in the royal palace and for a long time, she became his mistress.” Medina y Mendoza, Vida, p. 168. According to Salazar de Mendoza, the young man’s full name was Rodrigo Diaz de Vivar y Mendoza, a name he had been given due to his father’s belief that the family could trace its ancestry back to the famous medieval hero of the same name, better known as “the Cid.” Nevertheless, in all of the documentation I have encountered, he is referred to as simply “Rodrigo de Mendoza.” Salazar de Mendoza, Crónica, p. 422. Ines de Tovar is identified in the biographies simply as a “woman from Valladolid.” Medina y Mendoza, Vida, p. 304. Salazar de Mendoza, Crónica, pp. 465–66. As one scholar writing about the period points out, “the Sovereign Pontiff was a family man, who openly acknowledged the paternity of seven bastards, and whose chief concern appears to have been their settlement in life.” Frederick Baron Corvo [Frederick William Rolfe], Chronicles of the House of Borgia (New York, 1962), 72. All papal licenses are cited in every subsequent transfer of property to one or another of the cardinal’s sons. See: AHN, Osuna 1703, 1760, 2021, several of which are partially reprinted in the appendix to Layna Serrano, 2:478–80.

244

L. J. Andrew Villalon

the queen issued another royal license permitting the cardinal to entail upon his offspring any or all of his property.148 Armed with these papal and royal documents, Mendoza proceeded to divide his fortune. He chose to do this primarily through inter vivos donations rather than in a last will and testament, in all likelihood his way of making certain that the transfer went off without a hitch. In November 1489, some six years before his death, Mendoza transferred to his eldest son, Rodrigo, the lion’s share of the property he then held, including the town of Jadraque and the fortress of the Cid, as well as some urban property in Guadalajara, and various rents which he held of the crown.149 Less than two years later, he added to this bequest his newly acquired lands in Granada.150 And in the autumn of 1492, shortly before his retirement from public life, Mendoza put the finishing touches to Rodrigo’s settlement, seeing to it that all of the properties would be gathered into a single mayorazgo or entailed estate. At this time, he prevailed upon the crown to bestow on the young man and his heirs the dual title of marquess of the Zenete and count of the Cid. Finally, he arranged for and presided over Rodrigo’s marriage to the daughter of another great noble house.151 The estate which Cardinal Mendoza had created for his eldest son brought in an annual rent of some 16,000 gold florins, placing its owner thirteenth on Lalaing’s list of Castilian aristocrats in respect to income.152 During this same period, the cardinal established a smaller but still significant estate for his second son, Diego.153 In November, 1489, he drafted a mayorazgo for Diego which incorporated the town of Almenara in the diocese of Cuenca, the tercias reales of Guadalajara, and four cuentos which were to be deposited with the cardinal’s treasurer until such time as they could be used for the purchase of some revenue-producing property that could then be added to the entail. Mendoza retained the right to alter this mayorazgo during his lifetime, and in June 1494 he made use of the clause in order to remove the four cuentos, which were then redirected to charitable undertakings. However, he helped the young man to launch a successful military career that ultimately enabled him to win with his own sword the county of Melito within the Italian territories 148

149 150

151 152 153

There survives not only mention, but also a lengthy paraphrase of this final document in various of the deeds through which the cardinal transferred the property to his sons. See: AHN, Osuna 1862 n. 3; Osuna 1760 n. 6; Osuna 1858 n. 7. See also: Layna Serrano, 2:478–80. AHN, Osuna 1703. There exist two deeds, both drafted on the same day (2 March 1491), in the same place (Guadalajara), by the same person (the cardinal’s secretary, Diego Gonzales), and both purportedly transferring the Zenete to Don Rodrigo. While one transfers the property with the condition attached that “of all the abovesaid goods, you are to establish a perpetual and indivisible mayorazgo,” the other actually establishes the mayorazgo. The presence of two documents, containing quite different provisions, remains something a mystery. AHN, Osuna 1760 n. 6 (printed in Layno Serrano 2, appendix, pp. 478–80.) AHN, Osuna 1760 n. 10–11. The bride was a daughter of the count of Medinaceli. Medina y Mendoza, Vida, pp. 293–94. Salazar de Mendoza, Crónica, p. 423. Lalaing, pp. 231–7. AHN, Osuna 2021.



Pedro González de Mendoza

245

ruled by Spain. The cardinal’s second son became one of the most important members of the family to carry forward its medieval military heritage into the early modern period. By contrast, the youngest son, Juan de Mendoza, did not fare as well. Mendoza’s plan for a clerical career for him foundered on the boy’s rejection of the religious life. Since most property had by then been divided or otherwise committed, Juan received only a modest inheritance. Disappointment at his share of the property may have led to his participation in the Comuneros Revolt against the crown in 1520. Shortly afterwards, he left Spain, entered the service of Francis I, and married into the French aristocracy.154 Conclusion Archbishop Turpin, with whom we began this essay, is what novelist and literary critic E. M. Forster once referred to as “a flat character”; in other words, one who can be largely summed up in a single sentence. In Turpin’s case, that sentence might be, “he was a brave and loyal warrior churchman.”155 What this article has tried to demonstrate is the fact that real warrior churchmen of the Middle Ages, whom the surviving sources permit us to study in depth, can be distinguished from their fictional counterparts, as a result of having too many “faces” to ever be summed up by merely alluding to their warrior status. To capture the reality of a figure like Pedro González de Mendoza necessitates close examination of not only his martial qualities, but those other faces as well. In fact, most of what might be dubbed the “cardinal virtues” of Don Pedro – loyal service to the crown, courage especially in battle, his sumptuous lifestyle and charitable donations, his efforts to protect, promote, and provide for family and friends, his patronage of learning, even his personal scholarship – could have characterized either a great churchman or a great secular noble of the later Middle Ages. By contrast, it would be his “cardinal sins” – most notably nepotism, pluralism, absenteeism, clerical concubinage, and the production of progeny, as well as periodic blood-letting – that indelibly marked him as a member of the first estate. Not infrequently, secular virtues and clerical sins were just opposite sides of the same coin. For example, Mendoza’s constant attendance at court which helped make him such a fine servant of the state also made him an absentee bishop. His overriding concern for family, so typical of a medieval aristocrat, led him into the paths of nepotism, and accounted, at least in part, for his ventures into clerical concubinage and fatherhood. And his military activities, so valued by Enrique IV and the Catholic Monarchs, in particular his service on the field of battle, violated ancient and oft-reaffirmed church laws against blood-letting. 154 155

Medina y Mendoza, Vida, p. 304. Salazar de Mendoza, Crónica, pp. 466–67. E. M. Forster, Aspects of the Novel (New York, 2002) is a work of literary criticism first published in 1927. Forster introduces the concept of the flat character to help his readers analyze works of literature.

246

L. J. Andrew Villalon

Not to put too fine a point on it, in Pedro González de Mendoza the historian encounters a figure who, admirable as he may have been in many respects, led a life that flew in the face of the dictates of his clerical profession. His younger contemporary and ecclesiastical successor, Cisneros, would lead a very different lifestyle, unmarred by the clerical sins that characterized Mendoza; in particular, his warlike activities. Nevertheless, the commission of such clerical sins by the older man did not mark him as particularly exceptional among high churchmen of his age. It was, after all, a time when even popes acknowledged their illegitimate children and provided for them, often at the expense of the Church. Alexander VI’s active support for the state-building efforts of his son, Caesare Borgia, was only the most flagrant example. Even in what might have been Mendoza’s most distinctive sin, to wit, his active participation in warfare, he enjoyed a good deal of company, including his chief ecclesiastical rival, Alonso Carrillo, against whom he fought at both Olmedo and Toro. And it was their contemporary, Giuliano della Rovere, who ascended the throne of Peter as Julius II and became known to history as “the warrior pope.”156 The career of this fascinating and powerful figure exemplifies both the best and the worst qualities of a high churchman in late medieval Europe, one living on the eve of the Reformation. Several decades after Mendoza’s death, an obscure German monk named Martin Luther would add his voice to the swelling chorus of complaints directed against the sins of high churchmen like Don Pedro. The monk’s indictment would become a regular theme of the reform movement that he and like-minded contemporaries touched off and that changed Western Christianity forever.

156

This appellation has long been applied to the pope, who, when not feuding with Michelangelo over paintings in the Sistine Chapel, spent much of his papacy (1503–13) on the battlefield, reasserting control over the papal states or, alternatively, stirring up the latest round in the Italian Wars. See, for example: Christine Shaw, Julius II, The Warrior Pope (Oxford and Cambridge, 1993). Shaw entitles one of her chapters with another sobriquet commonly applied to Julius II by his contemporaries – “il papa terribile.”

11 Late Medieval Divergences: Comparative Perspectives on Early Gunpowder Warfare in Europe and China1 Tonio Andrade

When historians discuss the revolutionary changes associated with guns, they tend to focus on Europe’s early modern period (1500–1800), but the late medieval period (1300–1500) was arguably more significant. It was then that guns became widely adopted, grew in power and size, and, finally, toward the 1480s, took on their classic form, which would remain largely the same for the next three centuries. The work of scholars such as Kelly DeVries, Robert Smith, Bert Hall, and Clifford Rogers has examined these developments in Europe, establishing that the late medieval period was a key watershed in the history of guns, as important as – perhaps more important than – the early modern period.2 But what is intriguing is that this was not just true of Europe. Recent work by historians of China has demonstrated that during the thirteenth, fourteenth, and

1

2

The author wishes to thank the Harry Frank Guggenheim Foundation, Emory University’s College of Arts and Sciences, and the Emory University History Department for helping to make possible the research that underlies this piece. Emory Woodruff Library’s outstanding staff helped to obtain materials, particularly Marie Hansen, Guo-hua Wang, and Alain Saint Pierre. Thanks also to the editors and anonymous reviewers of Journal of Medieval Military History, whose excellent and stimulating advice greatly improved this article. Robert D. Smith and Kelly DeVries, The Artillery of the Dukes of Burgundy, 1363–1477 (Woodbridge, 2005); Kelly DeVries and Robert D. Smith, Medieval Military Technology, 2nd ed. (Toronto, 2012); Kelly DeVries, “The Technology of Gunpowder Weaponry in Western Europe during the Hundred Years’ War,” in XXII. Kongress der internationalen Kommission für Militärgeschichte (Vienna, 1997), pp. 285–99; Robert D. Smith, “Artillery and the Hundred Years War: Myth and Interpretation,” in Arms, Armies and Fortification in the Hundred Years War, ed. Anne Curry and Michael Hughes (Woodbridge, 1994), pp. 151–60; Robert D. Smith, “All Manner of Peeces: Artillery in the Late Medieval Period,” Royal Armouries Yearbook 7 (2002), 130–38; Robert D. Smith, Rewriting the History of Gunpowder (Sundby Lolland, Denmark, 2010); Clifford Rogers, “The Military Revolutions of the Hundred Years War,” The Journal of Military History 57:2 (1993), 241–78, “‘Military Revolutions’ and ‘Revolutions in Military Affairs’: A Historian’s Perspective,” in Toward a Revolution in Military Affairs? Defense and Security at the Dawn of the Twenty-First Century, ed. T. Gongora and H. von Riekhoff (Westport, CT, 2000), pp. 21–35 and “The Artillery and Artillery Fortress Revolutions Revisited,” in Artillerie et fortification: 1200–1600, ed. Nicolas Prouteau, Emmanuel de CrouyChanel, and Nicolas Facherre (Rennes, 2011), pp. 75–80; Bert S. Hall, Weapons and Warfare in Renaissance Europe: Gunpowder, Technology, and Tactics (Baltimore, 1997).

248

Tonio Andrade

fifteenth centuries Chinese guns were also evolving rapidly, transforming East Asian warfare.3 Reading the evidence from China in conjunction with that from Europe reveals puzzles that I believe illuminate both historiographies, because although gunpowder warfare evolved rapidly in both Far Eastern and Far Western Eurasia, it also evolved differently: Western Europeans focused on gunpowder artillery; Chinese focused on smaller guns, particularly firearms (handheld guns). Thus, whereas it seems that firearms did not play a major role on Western European battlefields until the mid-1400s at the earliest (they were present, but in relatively small proportions), there were around 150,000 dedicated firearms units in the Chinese armed forces by the 1370s, a proportion of 10 percent of infantry forces, a ratio that would rise to 30 percent by the mid-1400s.4 This latter figure would not be matched in Europe until the mid-1500s. In contrast, wall-smashing gunpowder artillery began to appear in Europe by the late 1300s and did not appear in Chinese warfare until introduced by Europeans, starting in the 1500s. What accounts for these divergent developments? And, equally importantly, why in Europe did all forms of guns – handheld and artillery – improve so rapidly in the second half of the 1400s, whereas in China development slowed or ceased, so that by the early 1500s Chinese recognized that European guns were superior and began adopting them? The answers I offer here are preliminary, and much more research is needed, but I believe that the explanations may be relatively straightforward. War makers in China were able to make effective use of handguns because they already had 3

4

And not just China. The pioneering work of Gábor Ágoston has made a similar case for the dynamism of the Ottoman Empire’s gunpowder weaponry. See especially Gábor Ágoston, Guns for the Sultan: Military Power and the Weapons Industry in the Ottoman Empire (Cambridge, 2005) and Gábor Ágoston, “Firearms and Military Adaptation: The Ottomans and the European Military Revolution, 1450–1800,” Journal of World History 25:1 (2014), 85–124. For the historiography on China, see especially Wang Zhaochun 王兆春, Zhong guo huo qi shi 中国火器史 (Beijing, 1991); Wang Zhaochun 王兆春, Zhong guo ke xue ji shu shi: jun shi ji shu juan 中国 科学技术史：军事技术卷 (Beijing, 1998); Li Huguang 李湖光, Da Ming di guo zhan zheng shi: da Ming long quan xia de huo qi zhan zheng 大明帝国战争史：大明龙权下的火器战 争 (Beijing, 2010); Liu Xu 刘旭, Zhong guo gu dai huo yao huo qi shi 中国古代火药火器史 (Zhengzhou, 2004); Stephen G. Haw, “The Mongol Empire – The First ‘Gunpowder Empire’?” Journal of the Royal Asiatic Society 23:3 (August 2013), 1–29; Peter Lorge, The Asian Military Revolution: From Gunpowder to the Bomb (Cambridge, UK, 2008) Peter Lorge, War, Politics, and Society in Early Modern China, 900–1795 (London, 2005); Sun Laichen, “Chinese Military Technology and Dai Viet: c. 1390–1497,” Asia Research Institute Working Paper Series No. 11 (September 2003); Sun Laichen, “Military Technology Transfers from Ming China and the Emergence of Northern Mainland Southeast Asia (c. 1390–1527), Journal of Southeast Asian Studies 34:3 (2003), 495–517; Yao Jiarong 姚家荣, “Xi pao de ying yong yu Ming dai de guo fang” 西砲的应用与明代的国防 (M.A. Thesis, Hong Kong Lingnan University, 2004); Harriet Zurndorfer “What is the Meaning of ‘War’ in an Age of Cultural Efflorescence? Another Look at the Role of War in Song Dynasty China (960–1279),” in War in Words: Transformations of War from Antiquity to Clausewitz, ed. Marco Formisano and Hartmut Böhme (Berlin, 2010), pp. 89–113. Wang Zhaochun, Zhong guo huo qi shi, pp. 103–6.



Early Gunpowder Warfare in Europe and China

249

large standing infantry armies with a long and vibrant tradition of drill. By using drill to instill techniques such as the famous volley fire method – which makes its definitive appearance in Western European records only in the 1500s, and, most scholars argue, toward the very end of that century5 – Chinese forces were able quite early (by the mid-1300s) to compensate for the great drawback of early firearms, their slow rate of fire. Western European polities, on the other hand, generally lacked standing infantry armies, and the tradition of infantry drill (including volley techniques) appears to have died out with the Western Roman empire. In most of Western Europe, it was not until the very end of the late medieval period, with the development of standing armies, that infantry drill began to be implemented on a significant scale, and at this point handguns could begin to be used in large numbers on the battlefield. And why did the Europeans develop powerful siege artillery, whereas the Chinese did not? Here I believe the answer lies in differing cultures of fortification. Chinese walls were an order of magnitude thicker than those of medieval Europe, and whereas fortifications in medieval Europe tended to be made of stone and were thus relatively brittle, Chinese walls had thick cores of tamped earth. Thus, existing European walls were far more vulnerable to artillery than were Chinese walls. It made little sense for anyone in China to invest the huge resources necessary to build large siege guns, because even the largest cannons would have had relatively little effect on existing walls. In Europe, in contrast, walls were susceptible to cannons, leading to an increased investment in and development of anti-wall artillery. As artillery became more effective, Europeans also began developing new fortification techniques, and the methods of construction they eventually adopted in some ways approximated the traditional methods of China: thicker, filled with earth, and battered (i.e., slightly sloped), although the Europeans, in contrast to the Chinese, also implemented principles of geometric defense, which during the early 1500s were codified into the famous trace italienne design of the artillery fortress. The final question of why European guns continued developing and became superior to those of China is most difficult, and my answer is most speculative. I believe it has to do with the frequency of warfare, a venerable but, I believe, compelling explanation. In the period between 1350 and 1450, China saw frequent warfare, with armies of hundreds of thousands raging through East Asia. In this environment, guns developed rapidly, with designs being passed back and forth, and not just within China itself. The polity of Dai Viet, for

5

See in particular Geoffrey Parker, “The Limits to Revolutions in Military Affairs: Maurice of Nassau, the Battle of Nieuwpoort (1600), and the Legacy,” The Journal of Military History 71:2 (2007), 331–72. Work on the Ottoman Empire suggests that the Ottomans, with their standing armies, may have deployed volley fire as early as 1526, and were certainly doing so by the early 1600s. See Gábor, “Firearms and Military Adaptation”; Günhan Börekçi, “A Contribution to the Military Revolution Debate: The Janissaries’ Use of Volley Fire during the Long Ottoman–Habsburg War of 1593–1606 and the Problem of Origins,” Acta Orientalia Academiae Scientiarum Hungaricae 59:4 (2006), 407–38.

250

Tonio Andrade

example, located in what is today the northern part of Vietnam, also developed significant innovations, which influenced China.6 After 1450, however, warfare in East Asia decreased. The Ming Dynasty, whose tumultuous beginning years had been full of armed conflict, established a general peace, and military innovation slowed. Meanwhile, the dynamic polities of Western Europe continued fighting and innovating. So when Portuguese ships brought guns to China, the Chinese recognized their effectiveness and rapidly copied them. Soon Western-style guns were bristling from the Great Wall itself.7 In the period of intense warfare that began in the 1550s and continued through the end of the seventeenth century, Chinese military innovation sped up again, as Chinese war makers adopted and adapted arquebuses, warships, advanced muzzle-loading cannons, and even principles of geometric defense. Early Guns It is all too easy, when looking at early guns, to fall into anachronism. To fully appreciate them and understand their use, we must take account of the fact that the gun evolved in China from a weapon known as the fire-lance, and that distinctions between the two weapons were probably less clear in the 1300s than they seem in retrospect. The fire-lance, as its name implies, was a staff at the end of which was affixed a tube filled with gunpowder. The gunpowder was lit and then, ideally, spewed forth. The fire-lance had perhaps emerged by the end of the 900s, but it probably did not become an effective and widespread weapon until the 1100s, when fierce warfare between the warring polities of Song China (960–1279) spurred rapid development in weaponry. Fire-lances were often used to eject things other than fire. One could add to the gunpowder mixture rocks, pellets, caltrops, etc. Historians have drawn a distinction between these objectspewing fire-lances and true guns based on the criterion of muzzle occlusion. Fire-lances’ projectiles did not occlude the barrel but were merely ejected along with the gunpowder mixture, leading the famous historian of Chinese science Joseph Needham to label them “co-viatives.”8 It is said that the first “true guns” appeared in the second half of the 1200s, but the distinction between these “true guns” and fire-lances should not be overdrawn.9 Aside from the fact that early guns in both China and Europe often had a 6 7

8 9

See especially Sun Laichen, “Chinese Military Technology and Dai Viet.” The literature on the Ming Dynasty’s adoption of Western cannons is extensive. For an overview of recent work, see Tonio Andrade, “Cannibals with Cannons: The Sino-Portuguese Clashes of 1521–1522 and the Early Chinese Adoption of Western Guns,” Journal of Early Modern History (in press, expected 2015). Joseph Needham, Science and Civilisation in China, V. 5 Part 7 Military Technology: The Gunpowder Epic (Cambridge, 1986), p. 9. Among the first firmly-dated extant exemplars is a piece from 1298, which is called the Xanadu Gun because it was discovered on the ruins of Xanadu (上都), the Mongol Yuan Dynasty summer palace in Inner Mongolia. Zhong Shaoyi 钟少异 et al., “Nei Meng-gu xin fa xian Yuan-dai tong huo chong ji qi yi yi” 内蒙古新发现元代铜火铳及其意义,Wen wu 文物 11



Early Gunpowder Warfare in Europe and China

251

similar form to the fire-lance – short barrels affixed to the ends of poles and held out in front – it seems that in both China and Europe early firearms were often used much in the way a fire-lance was used: to spew forth fire for conflagrative purposes, with other items added to the mix for added damage. Consider, for example, data from a battle of 1356, when the French attacked the English-held castle Breteuil. They were hurling stones at it from catapults and had also built a belfry, that is to say a siege tower on wheels. It must have been huge, because the chronicler Froissart says that each of its three stories could hold two hundred men. To counter the belfry, the English prepared “cannons throwing fire and large quarrels” (kanons jettans feu et grans gros quariaus), with which they planned to “destroy everything.”10 At first the English held these weapons in abeyance, fighting hand to hand from the walls. But when the French got the upper hand the English “began to shoot their cannons and throw fire onto and into the belfry, and with this fire they shot thick quarrels, and large ones, which wounded and killed large numbers, and made them [the French] so anxious [les ensonnyèrent] that they did not know what to do. The fire, which was Greek, took hold on the roof of the belfry, persuading those within to come out of it fast, or otherwise they would have been lost, turned to

10

(2004), 65–67. It is quite likely that certain other pieces predate the Xanadu Gun. One, for example, has an inscription that reads, roughly translated, “Made by bronzesmith Li Liujing in the year Zhiyuan 8 (直元), ningzi number 2565.” An expert suggests that Zhiyuan (直 元) should actually be zhiyuan (至元), which would put the date in the Western calendar at 1271 AD, a credible if not conclusive argument. Like the Xanadu Gun, it has a serial number (2565), which suggests that thousands of similar pieces were manufactured. There is reason for caution about the dating, because the reign name used is not a standard form, but the timing is not improbable. See “Ningxia fa xian shi jie zui zao you ming que ji nian de jin shu guan xing huo chong” 宁夏发现世界最早有明确纪年的金属管形火铳, Xinhua News Net, 2004– 06–09, http://www.nx.xinhuanet.com/newscenter/2004–06/09/content_2278435.htm [accessed 20 September 2012]. Other, circumstantially dated guns may be from the 1220s. For example, in 1980, a 200-pound bronze gun was discovered in a cellar in Gansu Province. There is no inscription data, but contextual evidence suggests that it may be from the late Xi Xia period (Gansu was part of Xi Xia territory), probably after 1214 but before the end of the Xi Xia in 1227. What is intriguing is that it was discovered with a pellet and a tenth of a kilogram of gunpowder in it. The pellet, about 9 cm in diameter, is a bit smaller than the muzzle diameter of the gun (12 cm), which suggests that it may have been a co-viative rather than a true bullettype projectile, although it is also highly corroded, suggesting that it used to be larger (thanks to Ben Sinvany for this idea). For more on this gun, which is known as the Xi Xia Bronze Cannon (西夏铜火炮) or the Wuwei Bronze Cannon (武威铜火炮), see Dang Shoushan 党寿 山, Wu wei wen wu kao shu 武威文物考述 (Wuwei: Guang ming yin shua wu zi you xian gong si, 2001), esp. pp. 103–13. See also Niu Dasheng 牛达生 and Niu Zhiwen 牛志文, “Xi Xia tong huo chong: wo guo zui zao de jin shu guan xing huo qi” 西夏铜火铳：我国最早的 金属管形火器, Dao gen 寻根 6 (2004), 51–57, esp. 51–52; and Liu Xiaolei 刘小雷, “Wuwei tong huo pao: shi jie shang zui gu lao de jin shu guan xing huo qi” 武威铜火炮：世界上最 古老的金属管形火器, Gansu Daily 每日甘肃, 2012–02–17, http://gansu.gansudaily.com.cn/ system/2012/02/17/012373433.shtml [accessed 21 September 2012]. Jean Froissart, Les chroniques de Jehan Froissart, publiées avec les variants des divers manuscrits part M. le baron Kervyn de Lettenhove, 25 vols. (Brussels, 1868), 5:376.

252

Tonio Andrade

ash.”11 Another source from the same year notes a similar mention of the use of cannons “to shoot quarrels and Greek fire,” in this case to set fire to the roofs of towers on a castle.12 This “Greek fire” was not the classic Greek fire of Byzantium, which was probably a petroleum-based liquid projected from siphons.13 The Europeans had a tendency to apply the label “Greek fire” to all kinds of incendiaries, and in this case what was used was almost certainly some kind of gunpowder mixture, perhaps similar to early Chinese gunpowder recipes, in which the active ingredients of saltpeter, sulfur, and charcoal were mixed with other incendiary materials such as resins, oil, or pitch.14 For example, a European recipe for “Greek fire” from circa 1450 reads as follows: “One calls ‘Greek fire’ a certain confection and brew [bouillement] of willow charcoal [charbon de saux], saltpeter, eau-de-vie, sulfur, pitch, and incense with a soft wool thread from Ethiopia.”15 This compound may have acted somewhat like early Chinese conflagrative gunpowder mixtures. Whether the gunners of the mid-1300s were using this recipe or another is impossible to determine, but it seems likely that some kind of gunpowder-like mixture was used.16 In any case, it seems quite possible that medieval European gunners sometimes used their guns as the Chinese used firelances, loading them with a gunpowder mixture and then, instead of ramming in a wooden plug to increase the projectile quality of the shot, leaving the plug out, stuffing the barrel full of gunpowder and conflagratives, and using the weapon to set fire to things. In China, fire-lances were primarily used, however, not as incendiaries but as anti-personnel weapons, discharging their pellets or shrapnel or rocks as co-viatives. Early European guns appear to have been used in the same way. For instance, the famous Loshult gun, perhaps the earliest-preserved European gun 11 12 13

14 15

16

Froissart, Les chroniques de Jehan Froissart, 5:377. Froissart, Les chroniques de Jehan Froissart, 5:389; Henry Brackenbury, Ancient Cannon in Europe, Part II: From AD 1351 to AD 1400 (Woolwich, UK, 1866), p. 39. John Haldon, “‘Greek Fire’ Revisited: Recent and Current Research,” in Byzantine Style, Religion, and Civilization: In Honour of Sir Steven Runciman ed. Elizabeth Jeffreys (Cambridge, 2006), pp. 290–325, here p. 291. Haldon, “‘Greek Fire’”, p. 290. Extracted in M. Reinaud and M. Favé, Histoire de l’artillerie, 1re partie: du feu grégeois des feux de guerre et des origins de la poudre a canon (Paris, 1845), p. 224. This type of recipe has a long pedigree in Europe. The recipe for Greek fire given by Marcus Graecus, probably written up in the early 1300s, goes as follows: “Take live sulfur, tartar, sarcocolla and pitch, boiled salt, petroleum oil, and common oil. Boil all these well together and then immerse tow in it…. When the fire is kindled it cannot be extinguished except by urine, vinegar, or sand.” J. R. Partington, A History of Greek Fire and Gunpowder (Baltimore, 1999 [first published 1960]), p. 50. To be sure, we cannot be certain that the “cannons throwing fire” were intentionally being used as incendiary weapons, because all black powder weapons from the late medieval and early modern periods shot out considerable amounts of fire along with their projectiles. Moreover, even if the cannons were being used as incendiary weapons, it is possible that the quarrels themselves were incendiary – perhaps tipped with burning gunpowder and other conflagratives, much as Chinese fire arrows were.



Early Gunpowder Warfare in Europe and China

253

(most experts date it to the first half of the fourteenth century), has a barrel with deep ruts and scratches.17 This has led researchers to conclude that the weapon was used to fire grapeshot or shrapnel.18 When those researchers built and tested a replica of the Loshult gun, they found that it could be used quite flexibly, shooting quarrels, grapeshot, and pieces of flint. They therefore concluded that the Loshult gun was likely used at close range, primarily as an anti-personnel weapon.19 This is precisely how Chinese fire-lances and early guns were used. But the most intriguing parallel between early European guns and Chinese guns from the same period is their size. One of the earliest firmly dated guns still extant is the Xanadu Gun from 1298, so named because it was discovered on the ruins of Xanadu (上都), the Mongol Yuan Dynasty summer palace in Inner Mongolia. It is 6.2 kg in weight and 35 cm long. Another early gun, the Heilongjiang Gun, which has been dated to 1288, is even smaller: 3.5 kg, and 34 cm long, with a 2.6 cm bore.20 Indeed, evidence suggests that nearly all other early Chinese guns – both extant and those whose existence we infer from historical records – were also small. For example, of the extant guns that we know for certain were early Ming pieces – i.e., from the 1350s through the early 1400s – nearly all are less than 80 kg in weight, and most weigh a couple kilograms or less. Guns considered “large” weighed only 75 kg.21 To be sure, there are three preserved guns from 1377 of relatively large size: a meter long, with 17

18

19 20

21

Although, as Kelly DeVries has argued, there are troublesome issues surrounding the dating of the gun. See Kelly DeVries, “Reassessment of the Gun Illustrated in the Walter de Milemete and Pseudo-Aristotle Manuscripts,” Journal of the Ordnance Society 15 (2003), 5–17. For more information on the gun, see the information page of the Swedish Historical Museum of Stockholm: http://mis.historiska.se/mis/sok/fid.asp?fid=114743 [accessed 3 January 2013]. See Smith, Rewriting, esp. pp. 115–18. Note that there is considerable controversy about this, with some researchers suggesting that its primary purpose was to shoot quarrels. On the controversy, see Wilfried Tittmann, “Die Eltzer Büchsenpfeile von 1331/3,” Waffen- und Kostümkunde, 36 (1994), 117–28; Wilfried Tittmann, “The Guns of Archbishop Baldwin of Trier and the Guns in the Milemete Manuscripts of 1326/7 Some Critical Comments,” Journal of the Ordnance Society, 17 (2005), 5–23; Klaus Leibnitz, “Fitting Round Pegs into Square Holes? Did Balduin of Luxemburg, Archbishop of Trier Use Gunpowder Artillery in the Siege of Eltz Castle 1331/33?” undated manuscript, available at http://www.vikingsword.com/library/ leibnitz_round_pegs.pdf [accessed 1 March 2013]. See Smith, Rewriting, pp. 115–18. Needham says 35 cm long. Lu Xu says 34 cm. Needham, Gunpowder Epic, pp. 289, 290. Cf. Liu Xu, Zhongguo gu dai huo yao huo qi shi, pp. 50–51. See also Wang Chong 王崇, A’cheng ban la cheng tong huo chong de lai li 阿城半拉城铜火铳的来历, Hei long jiang shi zhi 黑龙 江史志, 268 (2012, issue 3), 33–34. See Wang Zhaochun, Zhong guo huo qi shi, pp. 73–97; and Liu Xu, Zhongguo gu dai huo yao huo qi shi, pp. 106–7. Some historians – including the great Joseph Needham, have suggested that a collection of old cannons known as the Zhou cannons, and which weighed between 100 and 500 pounds, were forged by Zhang Shicheng in the 1350s or 1360s. This viewpoint has been firmly and persuasively refuted by Chinese scholars, who note that although the pieces in question do bear inscriptions for the dynastic term Zhou (周), which Shicheng did use, the guns were in fact most likely created during another Zhou period: the short-lived Zhou Dynasty of the warlord Wu Sangui in the 1670s. Needham, Gunpowder Epic, pp. 290–92. Discussion of controversy in Wang Zhaochun, Zhong guo huo qi shi, pp. 58–63.

254

Tonio Andrade

a muzzle diameter of 21 cm, and two handles on either side, shaped for easy grip for human transport.22 As Chinese scholar Wang Zhaochun notes, the existence of these guns shows – if any proof were needed – that the Chinese were entirely capable of making larger guns; but what is notable is that these are the only relatively large guns preserved from the early Ming period (pre-1500), and no other examples are known from either archeological or textual evidence.23 Chinese researchers have concluded that they were an anomaly, and that during the fourteenth and fifteenth centuries Chinese guns remained small and light.24 How large were the earliest European guns? The evidence is much less copious. Whereas we have dozens of extant Chinese pieces that are certainly from the 1300s, only one surviving European gun can be firmly dated to that century, and it has been dated to 1399.25 It is small, just 33 centimeters long and 1.24 kg in weight, but this means little, because we know that by then there existed huge bombards.26 Other candidates for early European guns, however, are also small. The Loshult Gun, which is the most famous early European gun, and which most experts date to the middle of the 1300s, weighs just 9 kg and is 30 cm long.27 It is thus remarkably close in size and weight to the earliest Chinese guns, such as the Xanadu Gun of 1298 (35cm long and 6.2 kg).28 Because archaeological evidence for early European guns is scant, we must turn to other sources. The earliest visual sources – two manuscript illustrations from the 1320s – depict guns that are relatively large, sitting on platforms, but these appear to be outliers.29 All other data indicate that early European guns were small. Some of this evidence is relatively direct. For example, sources suggest that the guns used by the English at Crécy in 1346 – one of the earliest 22

23 24

25

26 27 28 29

Cheng Dong 成东, “Ming dai qian qi you ming huo chong chu tan” 明代前期有铭火铳初探, Wen wu 文物 (1988, no. 5) 68–79, here 74. Cheng Dong argues compellingly that the outcroppings were handles and not trunnions. Wang Zhaochun, “Hong wu shi nian da tie pao,” in Zhong hua guo cui ci dian 中华国粹大辞 典, ed. Men Kui 门岿 and Zhang Xijin 张燕瑾 (Beijing, 1997), p. 183. Cheng Dong, “Ming dai qian qi,” p. 74; Wang Zhaochun 王兆春, Zhong guo gu dai bing qi 中 国古代兵器 (Beijing, 1994), pp. 170–71. See also Wang Fuzhun 王福谆, “Gu dai da tie pao” 古代大铁炮, Zhu zao she bei yan jiu 铸造设备研究 (2008, no. 3), 46–56, p. 48. On the lack of European guns, see Robert D. Smith, “Artillery and the Hundred Years War,” p. 154. For lists of extant fourteenth-century Chinese guns, see Liu Xu, Zhong guo gu dai huo yao huo qi shi, pp. 106–7 and 123–25. On possible candidates for fourteenth-century guns, including the 1399 Tannebergbüchse, see Gerd Strtickhausen, “Bemerkungen zu frühen Feuerwaffen im 14. Jahrhundert,” in Würfen hin in Steine / gröze und niht kleine: Belagerungen und Belagerungsanlagen im Mittelalter, ed. Olaf Wagener and Heiko Laß (Frankfurt, 2006), p. 52. Robert Coltman Clephan, An Outline of the History and Development of Hand Firearms (London, 1906), pp. 26–27. On issues with the dating of the Loshult Gun, see DeVries, “Reassessment.” Zhong Shaoyi et al., “Nei Meng gu,” pp. 65–68. On these and other early depictions of guns, see Valérie Serdon-Provost, “Les débuts de l’artillerie a poudre d’après l’iconographie médiévale,” in Artillerie et fortification: 1200–1600, ed. Nicolas Prouteau, Emmanuel de Crouy-Chanel, and Nicolas Facherre (Rennes, 2011), pp. 61–74.



Early Gunpowder Warfare in Europe and China

255

records of the use of guns in European history – were operated by one gunner or even, it seems, two guns to one person, which suggests that they were light.30 Other textual evidence is even more compelling. The nineteenth-century scholar Henry Brackenbury, reasoning that the cost of a gun must be directly related to the amount of metal used to make it, collected early sources recording the costs of forging and casting guns – there were many such sources – and then deduced gun sizes from price data. For example, one French receipt from 1342 records that 25 livres were paid for the manufacture of five bronze cannons and five iron ones. Using comparative price and wage data, Brackenbury determined that this sum would have sufficed to purchase five wrought iron cannons of 25 pounds each and five of bronze weighing 22 pounds each.31 After compiling large amounts of such pricing data, he concluded that until the middle of the 1300s guns averaged around 25 pounds in weight. None of the guns in his data would have exceeded 120 or so pounds. They increased slightly in size until the early 1370s, but not by much. This is quite similar to the sizes of guns of China. Guns of this size, he noted were “but feeble weapons in comparison with the great warlike engines of the period [i.e. catapults], which still were employed for the more serious operations.”32 Other scholars have corroborated his work and extended it, and Brackenbury has modern admirers.33 Robert Smith, one of today’s foremost historians of early artillery, writes that Brackenbury “is one of the few authors who uses the documentary evidence in a systematic and thorough manner and one which later writers have singularly failed to follow.”34 This is perhaps unfair to later writers, who have conducted outstanding systematic work – including Smith himself – but the point is clear: Brackenbury’s research still stands. Early European guns appear to have been small through 1370 or so. Of what use were such small guns? As Brackenbury notes, they could have “had little or no effect against the walls of cities or castles; they were quite incapable of making, or even assisting to make, a breach.”35 Historians have suggested that in fact guns were of marginal utility in European warfare until large guns emerged in the last quarter of the 1300s. Indeed, some have even wondered whether early firearms were used on the battlefield at all.36 To be sure, data from Crécy in 1346 show the use of guns, but historians have found 30 31 32 33

34 35 36

Henry Brackenbury, Ancient Cannon in Europe, Part I: From their First Employment to AD 1350 (Woolwich, 1865), p. 17. Brackenbury, Ancient Cannon, Part I, p. 8. Brackenbury, Ancient Cannon, Part I, p. 21. One prominent example of the price-to-weight method, applied to German sources, can be found in the detailed work of Bernhard Rathgen, whose findings corroborate those of Brackenbury. Bernhard Rathgen, Das Geschütz im Mittelalter (Düsseldorf, 1987 [originally published 1928]), pp. 6–7 and 151ff. For a modern and very positive take on Brackenbury, see DeVries, “Technology of Gunpowder Weaponry.” Smith, “Artillery and the Hundred Years War,” p. 153. Brackenbury, Ancient Cannon, Part I, p. 21. Brackenbury, Ancient Cannon, Part I, p. 21. He believes they were but feels there is room for doubt.

256

Tonio Andrade

no other reliable evidence of guns on European battlefields until the battle of Beverhoutsveld of 1382, and thereafter there are few examples of guns playing decisive roles until the mid-fifteenth century.37 Why? One possibility is technology: “[T]he technology was not yet there to make the gun an effective battlefield weapon. It was not until the late fourteenth and fifteenth centuries that guns began to appear regularly in battle.”38 Yet what is curious is that the Chinese were able to make their small guns work on the battlefield. From the 1350s through the early 1400s, a period when there are few sources showing guns being used effectively in European battles, Chinese records contain numerous descriptions of firearms used decisively: against Chinese rebels, Tai elephants, Vietnamese ships, and Mongol horsemen.39 Indeed, guns were considered so effective that, as I have already noted, 37

38 39

DeVries and Smith, Medieval Military Technology, p. 144; Kelly DeVries, “The Forgotten Battle of Beverhoutsveld, 3 May 1382: Technological Innovation and Military Significance,” in Armies, Chivalry, and Warfare in Medieval Britain and France, ed. M. Strickland (Stamford, 1998), pp. 289–303. Andrew Ayton and Philip Preston, The Battle of Crécy, 1346 (Woodbridge, 2007), pp. 154–55. I will discuss many such battles in a separate publication, but a quick mention of some examples is in order here. In the Siege of Shaoxing (1358), for example, guns (火筒) were used by a sallying force of defenders to drive back enemy troops. Xu Mianzhi 徐勉之, Bao yue lu 保越录 (1359); for a western-language account, see Herbert Franke, 2Die Belagerung von Shao-hsing im Jahre 1359,” in Krieg und Krieger im Chinesischen Mittelalter (12. bis 14. Jahrhundert), ed. Herbert Franke (Stuttgart, 2003), pp. 122–214. As Franke notes, at this siege, guns could not cause significant damage to walls or gates, and they were primarily used as anti-personnel weapons. Franke, “Belagerung,” p. 213. In the campaign of Poyang Lake in 1363, firearms were used in at least one battle on land and, more famously, in clashes on the water. See Wang Zhaochun, Zhong guo huo qi shi, pp. 57ff; in English, see Edward L. Dreyer, “The Poyang Campaign, 1363: Inland Naval Warfare in the Founding the Ming Dynasty,” in Chinese Ways in Warfare, ed. Frank A. Kierman, Jr. and John K. Fairbank (Cambridge, MA, 1974), pp. 202–42. In 1388, guns shooting in volleys defeated a force of Tai war elephants. See below, and see also Ming shi lu 明实录, Tai zu shi lu 太祖实录, Hongwu 洪武 11, Month 3, Juan 189. In a war of imperial succession in 1401, the usurper, the Prince of Yan (who would become the famous Yongle Emperor) found his huge forces routed by clever use of guns and poison crossbows, and he was nearly captured, an event that may have given him post traumatic stress disorder and likely stimulated him to pay close attention to guns in his future campaigns as a usurper and, even more spectacularly, in his subsequent bellicose reign. See Gu Yingtai 谷应泰, “Yan wang qi bing” 燕王起兵, Ming shi ji shi ben mo 明史纪事本末, Juan 16; a brief account in English is in David B. Chan, The Usurpation of the Prince of Yen, 1398–1402 (San Francisco, 1976), p. 72; for another battle in which guns were decisive in this succession war, see Chan, Usurpation, pp. 80–81. In the Yongle Emperor’s many campaigns, guns played decisive roles on various occasions. In the Vietnam campaigns, guns were used on both sides, as in the siege of Dobang City (1407), when guns proved decisive against Vietnamese war elephants. Ming shi lu 明实录, Tai zong shi lu 太宗文皇帝实录, Juan 62, Yongle 4, Month 12, Day 11. Cf Wang Zhaochun, Zhong guo huo qi shi, p. 110. For a brilliant discussion of the Ming Vietnamese wars, see Sun Laichen, “Chinese Gunpowder Technology and Dai Viet.” In 1410, far to the north, the Yongle Emperor’s firearm troops decisively defeated the Mongol leader Arudai near the Great Ghingan mountains, Ming shi 明史, Juan 154, Lie zhuan 42 列传第四十二, Liu Sheng 柳升; and Ming shi lu 明实录, Tai zong shi lu 太宗文皇帝实录, Yongle 8, Month 6, dingwei day (13 July 1410), Juan 105. In another battle, in 1414, Yongle himself led a vanguard



Early Gunpowder Warfare in Europe and China

257

by the 1370s the armies of the Ming Dynasty were fielding some 100,000 or so gunners, 10 percent of total infantry, a proportion that appears to have continued increasing until by the mid-fifteenth century they comprised some 33 percent of infantry units. Why might guns have been more effective on Chinese than on European battlefields? Was Chinese small-gun technology better? It is possible. But it is more likely that Chinese forces were simply better at deploying firearms than European forces because the effective use of handguns required coordinated drill. Guns were so slow to load that unless gunners worked closely with each other to maintain a constant rate of fire, they would find themselves overrun by the enemy before getting a chance at a second shot. Such coordination required painstaking and constant practice, i.e., collective drill. Although there is still much research to be done on military training in medieval Western Europe, it seems that there was not much training done collectively. As military historian Michael Prestwich writes in the Oxford Encyclopedia of Medieval Warfare, in Europe, “there is surprisingly little indication of this type of drill [i.e. ‘collective training’] taking place until the late fifteenth century.”40 This perspective is supported also in the work of Clifford Rogers, who notes that military ordinances from the medieval period say next to nothing about collective drill until the late fifteenth century, when the Burgundians began drilling; with the possible exception of Sicilian crossbowmen, who may have drilled in the eleventh century.41 There may of course be many examples of medieval European drill just waiting to be found, but it seems unlikely. Stephen Morillo has argued that the reason is simple: there were no centralized states to make them do it. “Drill,” he writes, “may only be instituted where there is a central authority strong enough to gather sufficient numbers of men together, and rich enough to maintain them while they are trained…. In effect, strong infantry depends on strong government.”42

40 41 42

of elite troops against a Mongol force, followed closely by firearm units. His gunners killed several hundred Mongols and threw their lines into disarray, at which Yongle’s elite cavalry – the iron horsemen – chased the remainder back up the hills, capturing many horses. In that same campaign, the Mongols later tried again to attack the emperor and his small elite force but were blasted by Ming guns firing from ambush positions, allowing his elite forces to rout the enemy: “the number of men and horses killed and hurt was uncountable, and they all screamed out in pain and left…. Henceforth that place was called ‘Barbarian Slaughtering Hold.’” Jin Youzi 金幼孜, “Bei zhen hou lu” 北征后录, one juan, http://www.guoxue123.com/other/gcdg/ gcdg/021.htm [accessed 29 October 2012]. Another source, the Ming Veritable Records, adds the intriguing detail that the guns “fired in continuous succession” [连发神机铳炮], which suggests volley fire. Ming shi lu 明实录, Tai zong shi lu 太宗文皇帝实录, juan 152. Michael Prestwich, “Training,” in The Oxford Encyclopedia of Medieval Warfare and Military Technology, vol. 1, ed. Clifford Rogers (Oxford, 2010), 370–72, p. 372. Clifford Rogers, Soldiers’ Lives through History: The Middle Ages (Westport, CT, 2007), pp. 68–69. Stephen Morillo, “The Age of Cavalry Revisited,” in The Circle of War in the Middle Ages, ed. Donald J. Kagay and L. J. Andrew Villalon (Woodbridge, 1999), p. 52. As a peer reviewer of

258

Tonio Andrade

The Chinese, in contrast, had a long and unbroken tradition of centralized states with standing armies. Even during times of division and fragmentation, China’s warring states maintained armies and trained them collectively. Drill was a core principle of Chinese warcraft. As the great military genius Zhuge Liang (AD 181–234) put it, “Disciplined troops under an incapable general cannot lose. Undisciplined troops under an able general cannot win.”43 Consider, for example, the case of volley fire, one of the most venerable methods by which forces armed with slow-loading missile weapons – such as crossbows, slings, or early guns – could maintain a constant hail of fire, thus avoiding the deadly pauses during which they might be rushed by an enemy armed with short-range weapons. The idea is straightforward. Soldiers stand in rows, usually one in front of the other, launch their missiles, and then move back to the rear to reload, while their comrades shoot. With several lines of soldiers, the shooting can be continuous. The principle may be simple, but it is very hard to train soldiers to do it properly. The men must be drilled, exercised, and trained exhaustively so that the sequence becomes second nature. Otherwise discipline evaporates once the men face an actual enemy in battle. In separate works, Geoffrey Parker and Olaf van Nimwegen have examined the difficulties that Dutch military leaders experienced as they sought to instill the musketry volley technique in their troops during the late 1500s and early 1600s.44 The challenges were significant. As Parker wrote, “Changing a pike square perhaps fifty deep into a musketry line only ten deep inevitably exposed far more men to the challenge of face-to-face combat, calling for superior courage, proficiency and discipline in each individual soldier. Second, it placed great emphasis on the ability of entire tactical units to perform the motions necessary for volley-firing both swiftly and in unison. The answer to both problems was, of course, practice.”45 The Europeans of course had classical precedents for the volley technique – the Greek military writer Aelian (second century AD) described countermarch

43

44

45

this article points out, however, Morillo does not take into account in his argument the urban militias, which were quite influential in the late Middle Ages. This is a topic that requires more research, although it may be that the decentralized structure of many urban militias – with artisan guilds responsible for supplying set numbers of militiamen, who sometimes fought under their own banners – was not as conducive to large-scale collective drill as the more centralized organization of a true standing army. Zhuge Liang 诸葛亮, Bing yao 兵要 (The Essence of War), in Zhuge Liang 诸葛亮, Zhuge Liang Ji 诸葛亮集, compiled by Zhang Shu 张澍 (1781–1847), ed. Duan Xizhong 段熙仲 and Wen Xuchu 闻旭初 (Beijing, 1960), juan 2. See especially Olaf van Nimwegen, The Dutch Army and the Military Revolutions, 1588–1688, trans. Andrew May (Woodbridge, 2010), pp. 100–12; and Parker, “Limits to Revolutions.” But see also J. P. Puype, “Victory at Niewupoort, 2 July 1600,” in Exercise of Arms: Warfare in the Netherlands, 1568–1648, ed. Marco van der Hoeven (Leiden, 1997), pp. 69–112. Geoffrey Parker, The Military Revolution: Military Innovation and the Rise of the West (Cambridge, 1996), p. 20.



Early Gunpowder Warfare in Europe and China

259

formations for javelin- and sling-units in his work Tactics.46 The Dutchmen who implemented the volley-fire technique in the early 1600s were inspired by the rediscovery and dissemination of Aelian and other ancient military thinkers carried out by the Flemish classicist Justus Lipsius (1547–1606).47 It seems, however, that the volley technique had died out with the Romans, at the very least in northwest Europe, and had to be revived and re-formulated nearly from scratch. In China, the volley technique also had deep classical roots – it was used as early as the Warring States period (475–221 BC) – but in contrast to Western Europe, it never died out.48 It remained vibrant because China’s standing armies usually fielded large numbers of crossbow units, which had slow rates of fire. A

46 47

48

Geoffrey Parker, The Cambridge Illustrated History of Warfare: The Triumph of the West (Cambridge, 2008), p. 4. Justus Lipsius, De militia Romana (Antwerp, 1614). For more on the influence of classical models on European warmaking in the early modern period, see Melissa Scott, “The Victory of the Ancients: Tactics, Technology, and the Use of Classical Precedent,” Ph.D. dissertation, Brandeis University, 1992. (For Scott’s examination of the musketry volley technique, see pp. 116–22.) Lipsius’s 1595 book De militia Romana influenced not just Willem Lodewijk but also his cousin Maurice of Nassau, who read it on campaign and used it in his reorganization of the Dutch army. Jeanine de Landtsheer, “Justus Lipsius’s De militia Romana: Polybius Revived, or How an Ancient Historian was Turned into a Manual of Early Modern Warfare,” in Recreating Ancient History: Episodes from the Greek and Roman Past in the Arts and Literature of the Early Modern Period, ed. K. A. E. Enenkel et al. (Leiden, 2002), pp. 101–22; Peter Dear, “The Mechanical Philosophy and Its Appeal,” in The Scientific Revolution, ed. Marcus Hellyer (Oxford, 2003), pp. 101–29; David Parrot, The Business of War: Military Enterprise and Military Revolution in Early Modern Europe (Cambridge, 2012), pp. 97–99; and, most important, Geoffrey Parker, “The Limits to Revolutions in Military Affairs.” A passage in Sun Bin’s Art of War, which was discovered in a tomb in 1972, may refer to the practice, saying, “Long range units should be in front, short-range behind, and the crossbows should flow to help their rapidity.” The passage appears in the section of the Sun Bin bing fa 孙 膑兵法 called “Questions of the King of Wei” 威王问. It is a short and enigmatic passage, and it certainly admits of other interpretations. The translation in the best English translation of the book is quite different. See D. C. Lau and Roger T. Ames, trans., Sun Bin: The Art of Warfare: A Translation of the Classic Chinese Work of Philosophy and Strategy (Albany, 2003), p. 100. Here is my translation of the entire passage: The Wei King said [to Sun Bin], “The enemy is numerous, and we are few; the enemy is strong and we are weak. What can be done?” Sun Zi said, “The tactic is called ‘Ceding to might.’ You must protect your rear so that you can withdraw. Long-range units should be in front, short-range behind them, and let the crossbows flow to help their rapidity.” My interpretation is informed by that of an enigmatic young Taiwanese blogger who writes under the name Shuo Xuehan 朔雪寒, and whose website on Chinese military thought, Ce lüe yan jiu zhong xin (策略研究中心) was an incredible resource, filled with original sources and ruminations on China’s military history. Alas, that website is now gone, and at present it seems not to have been rebuilt, which is a pity. I wish nonetheless to acknowledge that good parts of what I write about Chinese drill were inspired by it, particularly an article from the website on drill in history, in which Shuo Xuehan reflects on the film Zulu – the same film that Geoffrey Parker alludes to in his article, Parker, “Limits to Revolutions in Military Affairs.” http://www. cos.url.tw/tacticspattern/id10001.htm [accessed 26 March 2013].

260

Tonio Andrade

text from circa AD 801, the famous Tong dian (通典) of Tang Dynasty scholar Du You (杜佑, 735–812), notes the problem: “The crossbow is slow to load, and when battle is near it cannot shoot more than one or two times, and so battle is not a straightforward thing for the crossbow. At the same time, without the crossbow, it is not beneficial to do battle.”49 To compensate, Du You wrote, one should use the volley technique: [Crossbow units] should be divided into teams that can concentrate their arrow shooting.… Those in the center of the formations should load [their bows] while those on the outside of the formations should shoot. They take turns, revolving and returning, so that once they have loaded they exit [i.e., proceed to the outer ranks] and once they have shot they enter [i.e. go within the formations]. In this way, the sound of the crossbow will not cease and the enemy will not harm us.50

Similar passages can be found in other military manuals from the Tang Dynasty (618–907), the Song Dynasty (960–1279), and the Ming Dynasty (1368–1644), and there is also clear evidence of its use in battle, as when the official Song History describes how a famous general named Wu Jie (吴玠) and his younger brother Wu Lin (吴璘) deployed the volley technique to defend the Song against invasions by the Jin State during the 1130s: [Wu] Jie ordered his commanders to select their most vigorous bowmen and strongest crossbowmen and to divide them up for alternate shooting by turns [分番迭射]. They were called the “Standing-Firm Arrow Teams” [驻队矢], and they shot continuously without cessation, as thick as rain pouring down. The enemy fell back a bit, and then [Wu Jie] attacked with cavalry from the side to cut off the [enemy’s] supply routes. [The enemy] crossed the encirclement and retreated, but [Wu Jie] set up ambushes at Shenben and waited. When the Jin troops arrived, [Wu’s] ambushers shot, and the many [enemy] were in chaos. The troops were released to attack at night and greatly defeated them. [The Jin commander] Wushu was struck by a flowing arrow and barely escaped with his life.51

It is noteworthy that the Song History makes specific mention of this volley method, one of the few times that specific battlefield techniques are mentioned in the Song History, or indeed in any dynastic history, and it is likely that the Standing-Firm Arrow Teams fought many other engagements that are lost to history. As historians have noted, to come up with the idea of volley fire and to implement it are two very different things. It is hard to persuade troops to “stand firm” when the enemy is advancing. As Tang and Song manuals noted, “the accumulated arrows should be shot in a stream, so in front of them there must be

49

50 51

Du You 杜佑 (735–812), Tong dian 通典, juan 149, “Warfare Part 2” (兵二), in Du You, Tong Dian, 5 vols., ed. Wang Wenjin 王文锦 (Beijing, 1988), 4:3818. The Tong dian is also available from wikisource, http://zh.wikisource.org/wiki/通典/卷149 [accessed 27 March 2013]. Du You, Tong Dian, juan 149, p. 3818. Biography of Wu Jie, in Song shi 宋史, juan 366, “Lie zhuan 列传 125,” “Liu Qi, Wu Jie, Wu Lin” 刘锜 吴玠 吴璘.



Early Gunpowder Warfare in Europe and China

261

no standing troops, and across [from them] no horizontal formations.”52 Thus, crossbowmen had to face the enemy directly, with no troops protecting them in front. The only way to make certain that shooters stayed firm was to drill them regularly and intensively, and Chinese military manuals are full of techniques and regimens. The famous Wu jing zong yao (1044), for example, notes that “although large inspections of infantry and cavalry are not carried out frequently, one should nonetheless adopt a method of daily practice in each garrison [营] in order to teach sitting, rising, advancing, and retreating.… One should use drum sounds as signals.”53 The text describes specific large-scale exercises that combined different types of units – cavalry, spearmen, crossbowmen, archers, flagmen, drummers – whose movements were coordinated by drums, gongs, wooden clappers, and flag signals. It seems that drilling techniques – and the volley technique itself – were rapidly adopted for guns. The first clear evidence is from the late fourteenth century. In 1388, Ming forces were fighting in Yunnan Province to put down an insurrection by a Tai leader named Si Lunfa (思伦发, d. 1399). The Tai force was huge, on the order of 100,000. But the numbers of Tai fighters worried the Ming much less than did the Tai war elephants, which “wore armor and bore on their backs war towers with ramparts.”54 The Ming commander, Mu Ying (沐 英, 1344–92), gathered his officers together and admitted to them that the Ming attacks had so far failed against pachyderms, but, he said, “I do know what [the enemy] will be unable to withstand.”55 He explained his tactics carefully. His troops were to array themselves in three lines, setting up guns and some kind of gunpowder-fired arrow weapon (神机箭 – probably arrow-shooting guns but possibly rockets).56 “When the elephants advance,” he said, “the front line of guns and arrows will shoot all at once. If they do not retreat, the next line will continue this. If they still do not retreat, then the third line will continue this.”57 The following morning the troops were divided into three teams just as he had outlined. The enemy came out riding their armored elephants and advanced slowly and then, suddenly, rushed the Ming lines. “Our troops attacked them, shooting arrows and stones, the noise shaking the mountains and valleys. The

52

53 54 55

56 57

This is from Li Quan 李筌, Shen ji zhi di tai bai yin jing 神机制敌太白阴经 (c. 759AD), juan 6 卷六, “Jiao nu tu bian” 教弩图篇, online at http://www.cos.url.tw/book/3/O-1–023-d6. htm. Zeng Gongliang 曾公亮, Wu jing zong yao qian ji 武经总要前集 (Beijing, 1959 [1506–21, based on original from 1231, which in turn was based on the 1044 edition], fols. 23ff. Ming shi lu 明实录, Tai zu shi lu 太祖实录, Hongwu 洪武 11, Month 3, juan 189. Ming shi lu 明实录, Tai zu shi lu 太祖实录, Hongwu 洪武 11, Month 3, juan 189, my translation. Cf. Geoff Wade, Southeast Asia in the Ming Shi-lu, http://epress.nus.edu.sg/msl/ entry/2876 [accessed 26 October 2012]. Sun Laichen suggests that they were rockets, and he is one of the world’s experts on the topic. Sun Laichen, “Military Technology Transfers,” p. 500. Ming shi lu 明实录, Tai zu shi lu 太祖实录, Hongwu 洪武 11, Month 3, juan 189. Cf. Wade, Southeast Asia in the Ming Shi-lu.

262

Tonio Andrade

elephants shook with fear and ran.”58 The Ming pursued. According to the Veritable Records, half the elephants were killed, thirty-seven were captured, thirty thousand human heads were harvested, and ten thousand men were taken alive. Scholars have hailed this passage as the first evidence of continuous volley fire with firearms in world history.59 Of course, the evidence is not unambiguous – Mu Ying does not explicitly say that after the third row fires the first is to fire again and so on. On the other hand, there is no reason to doubt that a procedure that had been implemented for centuries with crossbows should not immediately have been adapted for guns. Chinese standing armies were able to draw on a deep and vibrant heritage of collective training. No “revolution in drill” was necessary.60 Other early Ming battles also show evidence of volley fire, such as the main battle of the bellicose Yongle Emperor’s 1414 expedition against the Mongols. According to the Veritable Records, “the commander Zhu Chong led Lü Guang and others directly to the fore, where they assaulted the enemy by firing continuously and in succession firearms and guns. Countless enemies were killed.”61 The interpretation hinges on two characters: lian fa (连发), to fire in succession, one after another. Yet the meaning seems clear. Lian means connected and continuous, one after another. There is no description of taking turns, but the enemy was on horseback, and if the Ming forces were not using volley fire, then there would have been long gaps between shots, or chaos as everyone fired and loaded at their own pace. Moreover, the practice of volley fire was so deeply entrenched that there was no need to describe it in detail. Thus, Sinophone historians have convincingly interpreted this battle as evincing volley fire.62 Other evidence similarly suggests volley fire with firearms, and it even seems likely that the Chinese were the first to implement volley fire with arquebuses. Whereas most scholars believe it was either the Dutch around 1600 or the Japanese in 1575 who first employed arquebus volley techniques, a Chinese military manual of 1560 contains clear descriptions of the practice.63

58 59 60

61 62 63

Ming shi lu 明实录, Tai zu shi lu 太祖实录, Hongwu 洪武 11, Month 3, juan 189. Cf. Wade, Southeast Asia in the Ming Shi-lu. Wang Zhaochun, Zhong guo huo qi shi, pp. 109–10; Sun Laichen, “Military Technology Transfers,” p. 500. The term “revolution in drill” is drawn, of course, from the famous article by Michael Roberts, “The Military Revolution, 1560–1660,” in The Military Revolution Debate: Readings on the Military Transformation of Early Modern Europe, ed. Clifford Rogers (Boulder, CO, 1995), pp. 15–16. Ming shi lu 明实录, Tai zong shi lu 太宗文皇帝实录, juan 152. Wang Zhaochun, Zhong guo huo qi shi, p. 110. There is also intriguing evidence that some kind of volley technique was used by the Ottomans in the battle of Mohács in 1526. See Gábor, “Firearms and Military Adaptation”; cf. Börekçi, “Contribution to the Military Revolution Debate.” For Chinese examples, see Qi Jiguang 戚继 光, Ji xiao xin shu 纪效新书 (18 juan version) (1560) (Beijing, 1999), p. 38 (in juan 2) and p. 94 (toward the end of juan 8). Other examples in Qi Jiguang 戚继光, Ji xiao xin shu: shi si juan ben 纪效新书:十四卷本 (Beijing, 2001), pp. 136, 152–3.



Early Gunpowder Warfare in Europe and China

263

Thus, the volley technique was part of an uninterrupted Chinese tradition of drill. In Europe, on the other hand, systematic drilling emerged only in the early modern period. This contrast may explain why during the late medieval period handheld guns were more prevalent on the battlefield in China than in Europe. Indeed, when we compare descriptions for early gun battles, we find that whereas Chinese sources rarely suggest that guns were fired all at once and are likely to say that they were “fired in succession,” the few early European sources that describe guns used on the battlefield (and there are far fewer than describe artillery battles), seem to indicate that the guns were fired off all at once. In the battle of Crécy, for example, there is no indication that guns were fired in turns, and one chronicle even suggests the opposite: “with many guns they vigorously attacked the French camp, firing all the guns at once.”64 Another significant gun battle – this one considerably later – also suggests that guns were fired at once. This was in 1382, at the battle of Beverhoutsveld. On one side was an army of the Flemish town of Ghent, who were attacking the town of Bruges. The chronicler Froissart writes: The Ghentenaars positioned themselves on a hill and gathered themselves together. Then they fired off more than three hundred cannons all at once and turned themselves about so that the Brugeois had the sun in their eyes…. Then they launched themselves into the [Brugeois’s] lines, crying “Ghent!” As soon as the Brugeois heard their voices and the cannons going off … they … threw down their weapons, turned tail, and ran.65

Kelly DeVries has rightly hailed the battle of Beverhoutsveld as “unique and special,” one of the only times in late medieval Europe that guns were decisive on the battlefield and “the first such decisive use in the history of western Europe.”66 The firing of guns all at once rather than in turns happened again at another famous early field battle with guns, the battle of Bulgnéville of 1431, when the victors “shot with fire from their cannons and couleuvres [smaller guns] all at the same time.”67 We certainly need more research into the comparative history of firearms, and particularly for the late medieval period, which was so seminal for their development, but it seems safe to suggest that one reason the Chinese were able to incorporate handheld guns into their infantry in such high proportions was that they were used to drilling units systematically, thanks to the fact that they had a long tradition of standing infantry armies. Intriguingly, it seems that in the Ottoman Empire, which also had standing armies, gunners were similarly 64

65

66 67

Anonymous, Storie pistoresi, in Raccolta degli storici Italiani dal ciquecento al millecinquecento, ed. L. A. Muratori (Città di Castello, 1897), p. 223. Cf. Brackenbury, Ancient Cannon Part I, p. 13. Note that Brackenbury mistranscribed the word scoccare for “soccare.” Froissart, Les chroniques de Jehan Froissart 10:31. With thanks to Kelly DeVries and Robert Smith, whose translation of this passage informed my own. DeVries and Smith, Artillery of the Dukes of Burgundy, p. 62; and DeVries, “Forgotten Battle,” p. 300. DeVries, “Forgotten Battle,” p. 303. Enguerrand de Monstrelet, Chronique, cited in DeVries and Smith, Artillery of the Dukes of Burgundy, p. 105.

264

Tonio Andrade

integrated into army units quite early on – by the 1390s at the latest – although the numbers and proportions seem to have been considerably lower than those obtaining in early Ming China.68 Much research remains to be done, but it is likely that scholars will find further Ottoman parallels with Ming China: standing armies, which were able to drill collectively, were likely better at deploying handheld guns than less permanent forces throughout Eurasia. But what is perhaps even more intriguing is that by the end of the 1300s, Western Europeans – and the Ottomans, too – were beginning to develop guns that were quite different from those of China. Previously, in both Europe and China, guns had been used primarily as anti-personnel devices and perhaps secondarily to attack and set fire to wooden structures. But in the last quarter of the 1300s, European guns became very large and were increasingly used to blast down fortifications, whereas guns stayed small in China. Why? The answer may have to do with the fact that the Europeans and Chinese built very different types of walls. Big Guns Although there are some disagreements about timing, most medieval military historians believe that the decisive period for the emergence of big guns was the 1370s, and the development of huge guns proceeded rapidly over the following decades.69 By the mid-1400s, guns ceased becoming larger in Western Europe, as gunsmiths focused more on mobility than size, but they continued to be aimed at walls, and successfully. In China, in contrast, no such developments took place. Guns remained small. Whereas Westerners were making guns that weighed many tons, “large” guns in China from the 1300s and 1400s weighed only 75 kg or so in weight, and the vast majority of Chinese guns weighed just a kilogram or two.70 Moreover, Chinese records make clear that although guns were ubiquitous at sieges, they were not used to blast down walls. Rather, they were aimed at people and, sometimes, wooden gates and towers. Why? Historians have suggested that Chinese gun makers did not need to destroy walls because China was a unified empire: “Since China was under a single sovereignty, gunpowder weapons were only needed on ships and for defence of fortified places against barbarian harassment. For both these purposes,

68 69

70

Ágoston, “Firearms and Military Adaptation,” 93–4; Ágoston, Guns for the Sultan, pp. 16–24. DeVries, “Technology of Gunpowder Weaponry”; Devries and Smith, Medieval Military Technology, pp. 140ff; Kelly Devries, “The Use of Gunpowder Weaponry by and against Joan of Arc During the Hundred Years War,” War and Society 15:1 (1996), 4–5; Philippe Contamine, War in the Middle Ages (New York, 1984), pp. 140ff; Smith, “Artillery and the Hundred Years War”; Rogers, “Military Revolutions of the Hundred Years War.” Wang Zhaochun, Zhong guo huo qi shi, pp. 58–63, 74. As I have noted above, there are extant guns that are larger, but they appear to be an anomaly, a road not travelled. Nor is there any evidence that they were used against walls.



Early Gunpowder Warfare in Europe and China

265

smaller and more mobile guns alone made sense.”71 But of course, walls stood in the way of many Chinese armies, and China was often not unified. I believe that a better explanation has to do with the culture of fortification. The Chinese built different types of walls than did the Europeans, walls that were much less vulnerable to bombardment. Toward the middle of the twentieth century, a European expert in fortification mused on how astoundingly large China’s walls were: “in China … the principal towns are surrounded to the present day by walls so substantial, lofty and formidable that the medieval fortifications of Europe are puny in comparison.”72 It is a significant observation, with profound implications. Even in prehistoric times, the ancestors of the Chinese surrounded their cities with massive fortifications. The capital of a Neolithic settlement from the Longshan period (3000 to 2000 BC), known as Chengziya (c. 2500 BC), was surrounded by a long wall eight to ten meters wide.73 In the Shang period (1600–1046 BC), Chinese city walls were even more massive. The walls of the Shang city of Zhengzhou, for example, which have been the object of considerable archeological research, stood ten meters high and had a width of more than twenty meters at the base and five meters at the top.74 For the next millennia, Chinese kept building huge walls, so that by the late imperial period, nearly all prefectural and provincial capitals were fortified. European walls were much thinner. The Romans were the great wall builders of European antiquity, and it seems that in the early Roman period (up to the end of the second century AD) Roman walls were often ten meters or so high, but they were not wide by Chinese standards, varying from 1.5 to 2.5 meters.75 They were “built on a standard of .25 meters of width for each 1 meter of height. Thus an 8 meter high wall would be 2 meters thick.”76 Among the most impressive Roman walls were those of Rome itself. The city’s Servian walls were up to 3.6 meters thick at their base, and during the period of the Emperor Aurelian they were rebuilt and eventually attained a thickness of around four meters and a height of six meters.77 Walls in the far reaches of the empire could also reach four meters thick, such as the Diocletian-era walls known as the Saxon Shore

71 72 73 74 75 76 77

William H. McNeill, “Men, Machines, and War,” in Men, Machines, and War, ed. Ronald Haycock and Keith Neilson (Waterloo, Canada, 1988), p. 14. Sidney Toy, A History of Fortification: From 3000 BC to AD 1700 (London, 1955), p. 181. Victor F. S. Sit, Chinese City and Urbanism: Evolution and Development (Singapore, 2010), p. 39. Ralph D. Sawyer and Mei-chün Sawyer, Ancient Chinese Warfare (New York, 2011), p. 125. DeVries and Smith, Medieval Military Technology, p. 189. J. E. Kaufmann, H. W. Kauffman, and Robert M. Jurga, The Medieval Fortress: Castles, Forts, and Walled Cities of the Middle Ages (Cambridge MA, 2004), p. 35. Kaufmann et al., Medieval Fortress, p. 35; DeVries and Smith, Medieval Military Technology, p. 191.

266

Tonio Andrade

Forts, which were 4.3 meters.78 This is still far thinner than Chinese walls of the same period, which were often twenty meters wide at the base.79 The most impressive walls of the Western world were those of Constantinople, which had a double fortification system: an outer wall two meters thick and an inner wall four meters thick, separated by a no-man’s land of about fifteen meters across.80 These defenses have been justly lauded. One author has written that Constantinople had “the most famous and complicated system of defence in the civilized world.”81 Another has called Constantinople’s “the most formidable development of fortification systems in the ancient world.”82 The intricacy of Constantinople’s defenses was certainly impressive, but if we stretch our definition of “civilized world” and “ancient world” to East Asia, we may perhaps find occasion to moderate such statements. Constantinople’s outer wall was a tenth the width of the wall of any reasonably sized Chinese city, and even the much stouter inner wall was merely a quarter or a third as thick. In fact, for much of the Middle Ages, most towns in Europe had no walls at all. Some scholars have argued that in the German lands around 1200, there were only twelve towns with proper walls, and nine were left over from Roman times.83 French and English towns were also usually wall-less, unless, again, they happened to have Roman walls. That is not to say that they were defenseless. Many European towns surrounded themselves with ditches, stockades, or low earthen ramparts. This was the case with the vast majority of German towns in the 1100s and 1200s.84 These earthen ramparts could sometimes be quite thick, but they tended to be low and rudimentary. An earthen fortification in twelfth-century Hereford, England, for example, was probably about fifteen meters wide at the base, but it was just three meters high and seems to have been protected from erosion only by a layer of gravel. No wonder that it was replaced by a stone wall in the 1200s.85

78 79

80

81 82

83

84 85

DeVries and Smith, Medieval Military Technology, p. 191. To be sure, some ancient northern European ringforts were much thicker than Roman-style walls, such as the ringwall of Otzenhausen, a Celtic fort in Saarland, Germany, parts of which were forty meters wide at the base. This was exceptional, however, and in any case Celtic fortbuilding practices died out during the early medieval period. See Efpraxia Paschalidou, “The Walls of Constantinople: An Obstacle to the New Power of Artillery,” in XXII. Kongress der internationalen Kommission für Militärgeschichte (Vienna, 1997), pp. 172–78. Paschalidou, “Walls of Constantinople,” p. 172. Richard Tomlinson, From Mycenae to Constantinople: The Evolution of the Ancient City (London, 1992), pp. 213–22, cited in James D. Tracy, “Introduction,” in City Walls: The Urban Enceinte in Global Perspective, ed. James D. Tracy (Cambridge, 2000), p. 10. DeVries and Smith, Medieval Military Technology, p. 269. This is perhaps an understatement, though, since it is not always easy to tell whether towns had walls or not. See Hans Planitz, Die deutsche Stadt im Mittelalter: von der Römerzeit bis zu den Zunftkämpfen (Graz-Köln, 1965), pp. 231–35. Planitz, Die deutsche Stadt, p. 260. John R. Kenyon, Medieval Fortifications (New York, 1990), p. 186.



Early Gunpowder Warfare in Europe and China

267

Hereford was not alone. In the 1200s and 1300s, new walls rose throughout Europe.86 They sometimes matched but rarely exceeded the thickness, height, and length of Roman walls. Late medieval French walls were usually two meters or less in thickness. England’s tended to be even thinner, with those of Southampton only 0.76 meters thick and those of Shrewsbury 1.37 meters thick.87 Other English towns boasted walls of French thickness: those of Bristol ranging between 1.5 and 2.5 meters, those of Bath 1.9 meters thick, those of Newcastle 2.1 meters thick.88 In fact, Western-based historians and archaeologists often use the phrase “very thick” to refer to walls that would be considered very thin in the Chinese context, as for example, when Kelly DeVries and Robert Smith write about the Southern French keep at Najac, begun in 1253: “Its walls were also very thick, measuring 2.2 meters in width.”89 Or when they write that most French walls in the late medieval period were “very wide, most measuring nearly two meters in thickness.”90 To be sure, DeVries and Smith are making comparisons within the European context, but it still bears noting that these “very wide” walls were less than a tenth the thickness of average Chinese city walls.91 Indeed, the marketplace of the Chinese city of Chang’an boasted walls thicker than the walls of European capitals, and that marketplace stood within the walls of Chang’an itself, which were far, far thicker.92 It is of course much easier to blast your way through a two-meter wall than a fifteen-meter wall, but it was not just the thinness of European walls that made them vulnerable to artillery. It was also the way they were built. European walls were made of stone, often with a filling of gravel or rubble, with limestone mortar often used as a bonding agent, a practice that went back to Roman times. Chinese walls, however, were much more resistant to artillery because they had an earthen core. Earthwork absorbs the energy of an artillery shot. An earthen-core wall might become riddled with holes during an attack, but those holes tended not to penetrate deeply, and the walls were resistant to shattering. One must not imagine that Chinese walls were filled with loose earth. The Chinese were able to create sturdy, hard walls by using an ancient earth-tamping method. One constructed a framework of wooden planks of the height and width one wanted. Then one poured in a layer of earth. This earth layer was tamped

86

87 88 89 90 91 92

For France, see Christopher Allmand, The Hundred Years War: England and France at War, c. 1300–c. 1450 (Cambridge,1988), p. 77. For England, see Kenyon, Medieval Fortifications, pp. 183–84. For the German lands, see Planitz, Die deutsche Stadt, pp. 229–42. Kenyon, Medieval Fortifications, p. 187. Kenyon, Medieval Fortifications, p. 187. DeVries and Smith, Medieval Military Technology, pp. 248–49. DeVries and Smith, Medieval Military Technology, p. 270. My italics. Kaufmann et al., Medieval Fortress, p. 36. Heng Chye Kiang, Cities of Aristocrats and Bureaucrats: The Development of Medieval Chinese Cityscapes (Honololu, 1999), p. 19.

268

Tonio Andrade

down until highly compact. Then another layer was added, and another.93 When the wall had reached its desired height, one removed the planks and used them to make the next section of the wall. The tamped-earth method produced walls that were surprisingly durable, and some ancient walls have survived four thousand years of rain and wind. Often these earthworks were encased in brick or stone to protect them from erosion, a practice that became more prevalent as time went on. The walls of Ming cities were built this way, and so was the Great Wall, rebuilt in the Ming period: a tamped earthen core (the earth sometimes interspersed with stone and rubble) encased in stone and brick. But the tamped-earth method was not the only thing that made Chinese walls resistant to artillery: Chinese walls were also sloped. Whereas a vertical wall that is struck by a projectile perpendicularly receives the full force of impact, a sloped wall deflects the projectile and absorbs less energy. What is intriguing is that as the Europeans began to adapt their fortifications to resist artillery, they ended up making them more like Chinese walls. In the course of the 1400s, the Europeans began building walls of sloped earthwork. As Kelly DeVries has shown, the practice seems to have begun in France and the southern Netherlands, where artillery warfare was particularly intense in the 1400s. Defenders added earthen outworks to their stone walls to protect them. These outworks were called boulevards in French and bolwercqen in Flemish, whence the English word “bulwark.” They were made with wooden planks on the outside and earth on the inside, and they were sloped to lessen the force of horizontal fire. As boulevards’ utility became recognized, they were made permanent and faced in stone.94 The boulevard was designed to protect vulnerable stone walls, but Europeans soon began building wholly new types of fortifications. These new walls were quite similar to traditional Chinese walls: filled with earth, encased in stone, and much thicker than the old stone walls. They were designed from the ground up to resist guns, and they worked. Artillery, which once breached walls rather regularly, was effectively countered, and sieges once again became long, drawnout affairs, with tight cordons that lasted months. Gone were the days in which garrisons surrendered after a cannonball or two pierced the wall. Traditional Chinese walls never needed to be rebuilt to resist artillery because they were never threatened by artillery. Consider, for example, the famous siege of Suzhou of 1366. This was one of the most important sieges of the violent early Ming Wars, when the forces of the Ming founder, Zhu Yuanzhang, were attacking the capital of one of his main rival states. Some experts have suggested that the siege of Suzhou proceeded as our standard image of a gun-era siege might lead us to expect. Ming specialist Edward Dreyer writes that “flaming 93 94

Yinong Xu, The Chinese City in Space and Time: The Development of Urban Form in Suzhou (Honololu, 2000), p. 113. Kelly DeVries, “Facing the New Military Technology: Non-Trace Italienne Anti-Gunpowder Weaponry Defenses, 1350–1550,” in Heirs of Archimedes: Science and the Art of War through the Age of Enlightenment, ed. Brett Steele and Tamara Dorland (Cambridge, MA, 2005), p. 48.



Early Gunpowder Warfare in Europe and China

269

arrows and rockets were used for incendiary purposes, while cannon of a more standard cast battered the walls.”95 Peter Lorge, whose excellent work has laid the groundwork for much current research on China’s military history, writes, “Zhu’s army completely enclosed the city [of Suzhou] in a circumvallation, and pounded it with artillery. Ten months of firing resulted in that rare occurrence in Chinese city fortifications, a wall breach.”96 But if we look carefully at the historical sources, we see that guns played a minor role in getting through Suzhou’s walls. That is not to say that gunpowder weapons were not important in the siege. Guns were certainly present, but they were aimed not at walls, against which they were useless, but at humans, against which they were lethal. The gates and towers of Suzhou were targeted not by guns but by weapons of a more traditional sort: trebuchets hurling huge stones and gunpowder bombs. The breach at Suzhou in 1366 was not even a proper wall breach. The breakthrough occurred not in the walls but at a gate. There is no record of cannons or catapults being used to effect this breach. It was probably done by traditional manual mining or battering, since the focus of the extant accounts describing it is on troops and men, not on machines of any kind.97 This should not surprise us. Not only were Ming guns relatively small, but Suzhou’s walls were also enormously thick, which is to say they were normal Chinese city walls. The city’s walls had been rebuilt in 1352 and records indicate that its seventeen kilometers of walls were 7.3 meters high, 11 meters thick at the base, and 5.1 meters thick at the top.98 Moreover, the walls were made of tamped earth and were battered, sloping markedly from bottom to top, thus providing further resistance to artillery.99 How might European cannons of the late medieval period have fared against such walls? According to a Florentine diplomat, writing in the early 1490s, “the French say that their artillery is capable of creating a breach in a wall of eight

95

96 97

98

99

Edward L. Dreyer, “Military Origins of Ming China,” in The Cambridge History of China, Volume 7, The Ming Dynasty, 1368–1644, Part I, ed. Frederick Mote and Denis Twitchett (Cambridge, UK, 1998), p. 93. Peter Lorge, The Asian Military Revolution: From Gunpowder to the Bomb (Cambridge, 2008), p. 74 and War, Politics, and Society in Early Modern China, 900–1795 (London, 2005), p. 105. Gu Yingtai 谷应泰, Ming shi ji shi ben mo 明史纪事本末, juan 4, “Tai zu ping wu” 太祖平 吴. The description in the Ming Veritable Records is almost identical. Ming shi lu, Hongwu shi lu, juan 25, Year of Wu 1, Month 9, Xinsi day. The Chinese measurements for 1352 are as follows: 23 chi high, 35 chi thick at base, and 16 chi wide at the top. I converted to the metric measures using a chi length of 31.6 cm, a reasonable estimate for the length of a chi during this period. In the late 1360s, records suggest a slightly greater thickness for the top – 18 chi, or about 5.6 meters (3 zhang 5 chi thick at base and 1 zhang 8 chi thick at the top). The figures are from Yinong Xu, Chinese City, p. 114. Frederick Mote says they were 25 feet thick at the base and 25 feet high, but his estimates are not quite as evidence-based as Xu’s. F. W. Mote, “A Millennium of Chinese Urban History: Form, Time, and Space Concepts in Soochow,” Rice University Studies 59:4 (1973), 53. Xu, Chinese City, p. 113.

270

Tonio Andrade

feet in thickness.”100 The diplomat suspected that his readers might not believe this, conceding that “the French are braggarts by nature,”101 but let us take the French at their word and suppose that European siege artillery circa 1490 were capable of creating breaches in walls of up to eight feet thick – very thick in the European context. Would that siege artillery have proven useful against the walls of Suzhou, which, at eleven meters (or thirty-six feet) were more than four times thicker than the Frenchmen’s hypothetical eight-feet walls? It is doubtful. To be sure, these French guns of the 1490s were not the massive bombards of the 1440s, but they were just as effective, making up in power and numbers what they lacked in size. More importantly, would the Europeans have bothered to develop wallsmashing artillery – either the huge bombards of the early 1400s or the lighter but more powerful guns of the late 1400s – if they had faced walls like those of China? Large guns were enormously expensive to make, to transport, and even to fire – the largest required a hundred pounds of powder or more for a single shot.102 Smaller siege artillery required less powder, but the expense was still significant. Scholars have estimated that a single shot from a sixteenth-century cannon cost the equivalent of a month’s wages for an infantry soldier.103 The kings and dukes of Europe were willing to pay these huge sums because the payoff was great. When your artillery train rolled up to a medieval town you could be fairly sure that you had a good chance of breaching its walls or, even better, intimidating its garrison into a quick surrender. In China, artillery would not have repaid the heavy investment. To be sure, the Chinese used catapults and guns to destroy wooden structures on walls, but the massive tamped-earth walls of China acted as a deterrent to the development of gunpowder artillery. Yet walls cannot explain everything about the military divergence. European guns did not just get bigger than Chinese guns. They also got more effective The Classic Gun in Europe In Western Europe, guns evolved rapidly in the second half of the 1400s. They came to have a long, tapered form, with a large length-to-bore ratio. This development had occurred by around 1490, and the new form was so successful that it would hold for the next three centuries. Historian Bert S. Hall has labeled

100 101 102

103

Cited in Philippe Contamine, “L’artillerie royale francaise a la veille des guerres d’Italie,” Annales de Bretagne, 71:2 (1964), 223. Cited in Contamine, “L’artillerie royale,” p. 223. Marios Philippides and Walter K. Hanak estimate that firing the huge bombards used by the Turks around the time of the siege of Constantinople in 1453 would have required up to 300 pounds of powder per shot. See Marios Philippides and Walter K. Hanak, The Siege and the Fall of Constantinople in 1453: Historiography, Topography, and Military Studies (Burlington, 2011), p. 423. Jack Kelly, Gunpowder: Alchemy, Bombards, and Pyrotechnics: The History of the Explosive That Changed the World (New York, 2004), p. 78.



Early Gunpowder Warfare in Europe and China

271

the development of the classic gun the “modern ordnance synthesis.”104 Robert Smith prefers to call the new gun design the “classic” gun.105 They disagree about the reasons for the development (Hall believes it had to do with powder corning; Smith suspects new casting techniques) but they agree on the importance and the timing of the developments. But why did the classic gun emerge in Europe and not in China? This is one of the key questions of global military history. No answer can be definitive – especially at this point, when we still know so little about global military history – but it seems that the forms of Chinese and European guns were developing along similar lines until around 1450. Until the mid-1400s, Chinese guns seem to have had length-to-bore ratios very similar to those of European guns, an average of 15 to 1 (Table 1).106 Table 1.  Ratio of length to bore, Chinese guns, 1288–1423

104 105 106

Year

Length (cm)

Bore (cm)

Ratio of length to bore

1288

  34

  2.6

13.1

1332

  35.3

10.5

  3.4

1334

  26.5

  2.3

11.5

1338

  47.5

10.5

  4.5

1338

 32

  2.2

14.5

1340

  21.5

  2.6

  8.3

1340

  31.5

  2.6

12.1

1351

  43.5

 3

14.5

1372

  36.5

11

  3.3

1372

  45.7

  2.54

18.0

1372

  43

 2

21.5

1372

  36.5

11

  3.3

1372

  44.6

  3.9

11.4

1375

  63

26

  2.4

1377

100

21

  4.8

1377

  42

  2.2

19.1

1377

  44.3

  1.9

23.3

1377

  44

  2.1

21.0

1377

  42

  2.1

20.0

Hall, Weapons and Warfare, p. 87. See especially Smith, “All Manner of Peeces,” p. 136; cf. DeVries and Smith, The Artillery of the Dukes of Burgundy, p. 42. The length-to-bore ratios of Chinese guns differed greatly according to the weight class, with smaller guns having much higher ratios, something that was just as true for Europe.

272

Tonio Andrade 1377

  36

  1.9

18.9

1377

  27

  2.3

11.7

1377

  38.5

  1.9

20.3

1377

  32.2

  2.1

15.3

1377

  43

 2

21.5

1377

  44

 2

22.0

1377

  44

 2

22.0

1377

101.6

  21.6

  4.7

1377

  44

 2

22.0

1377

  43.5

 2

21.8

1378

  36.4

14.9

  2.4

1378

  21.7

14.8

  1.5

1378

  36

  2.3

15.7

1378

 30

 2

15.0

1379

  26.5

 2

13.3

1379

  44.5

 2

22.3

1379

  44.2

  2.1

21.0

1385

 52

10.8

  4.8

1409

 55

10.4

  5.3

1409

 35

  1.5

23.3

1409

  35.5

  1.4

25.4

1414

  36

  1.4

25.7

1414

  36

  1.5

24.0

1421

  35.8

  1.5

23.9

1421

  35.7

  1.5

23.8

1423

  35.8

  1.4

25.6

Average Average with outliers excluded

15.2 18.9

Chinese guns through 1423 or so had a similar length-to-muzzle-bore ratio as those of Europe around the same period, or an average of 15.2, although the standard deviation, 7.9, is quite high because of the presence of a number of outliers (which is to say, guns whose length-to-muzzle-bore ratio is 6 or less). If we omit the outliers the average comes out as 18.9, with a standard deviation of 4.9.107

107

These data are gathered from various sources: Wang Rong 王荣, “Yuan dai huo chong de zhuang zhi fu yuan” 元代火铳的装置复原, Wen wu 文物 1962 vol., no. 3, 41–45; Yuan Xiao-



Early Gunpowder Warfare in Europe and China

273

Moreover, like European guns, Chinese guns were developing larger lengthto-bore ratios, as seems clear from Figure 1. Figure 1.  Trends in the Development of Chinese Guns, 1288–1423 30.0

25.0

20.0

15.0

10.0

5.0

0.0 1280

1300

1320

1340

1360

1380

1400

1420

1440

The vertical axis is the ratio of length-to-muzzle-bore, and the trend is quite clear. Through 1420 or so, Chinese guns were growing longer relative to muzzle bore, or toward the ratio that characterized the so-called “modern ordnance synthesis” and that was achieved in Europe in the late 1400s.

chun 袁晓春, “Shan dong Peng lai chu tu de Ming wan kou pao” 山东蓬莱出土的明碗口炮, Wen wu 文物, 1991 vol., no. 1, 91–92; Shi Baozhen 史宝珍, “Zhen jiang chu tu de Ming dai huo qi” 镇江出土的明代火器, Wen wu 文物, 1986 vol., no. 7, 91–94; Liu Hongcai 罗宏才, “Ding bian xian fa xian de yi jian Ming dai tie chong”定边县发现的一件明代铁铳, Wen bo 文博, 1988 vol., no. 4, 92–93; Liu Xu, Zhong guo gu dai huo yao huo qi shi, pp. 107, 117; Liu Shanyi 刘善沂, “Shan dong guan xian fa xian Ming chu tong chong” 山东冠县发现明初 铜铳, Kao gu 考古, 1985 vol., no. 10, 914; Yin Qichang 殷其昌, “He zhang chu tu de Ming dai tong chong” 赫章出土的明代铜铳, Gui zhou she hui ke xue 贵州社会科学, 1982 vol., no. 5, 71–90; Chen Lie 陈烈, “He bei sheng Kuang cheng xian chu tu de Ming tong chong” 河北省宽城县出土的明铜铳, Kao gu 考古, 1985 vol., no. 8, 759; Shi Wanlin 师万林, “Gan su Zhang ye fa xian Ming dai tong chong” 甘肃张掖发现明代铜铳, Kao gu yu wen wu 考古 与文物, 1986 vol., no. 4; Hu Zhenqi 胡振琪, “Ming dai tie pao” 明代铁炮, Shan xi wen wu 山西文物, 1982 vol, no. 1.

274

Tonio Andrade

To be sure, there is considerable variation in these data, caused by the fact that guns used for different purposes had quite different length-to-bore ratios, but in general the trend seems clear enough for provisional speculation. European guns continued farther along the path of a high length-to-bore ratio, and when the Portuguese brought their cannons to China in the 1510s, cannons that were of the “classical gun” type, the Chinese were highly impressed, recognizing the many advantages that the longer barrel length and thinner walls conferred. So what accounts for the fact that European guns kept evolving toward the “classic gun” form and the Chinese guns did not? It probably did not have to do with powder corning, because recent evidence strongly suggests that corned powder was used in China as early as 1370.108 Moreover, as Robert Smith and Kelly DeVries have shown, the evidence in Europe for the impact of powder corning is considerably less straightforward than the corning hypothesis would suggest: “corned powder,” they write, “was not the revolutionary development that some have asserted it to be.”109 Other explanations suggest that Europeans managed to develop guns with greater freedom because they had the advantage of being late adopters. As historian Peter Lorge has written, “Gunpowder weapons had become a fairly mature technology [in China]…. In Europe, by contrast, the technology was new and quickly demonstrated a narrow set of effective uses. Free from preconceived notions … [Europeans] set off with renewed creativity. Progress in China continued slowly.”110 This is an intriguing idea, but it does not account for the speed of European developments or the fact that Chinese guns had been trending toward the classic gun shape through the 1300s and early 1400s but that this development seemed to slow in China in the mid- and late 1400s, even as it continued in Europe. Another explanation suggests that the Chinese did not experiment as heartily with guns because they found them relatively useless against their most fearsome enemies, the horse-borne nomads of central Asia. To fight against nomads, logistics were the challenge, because the enemy could run away and draw out supply chains. They could sally and retreat, lure and engage at their will. Guns, no matter how good, simply were not so effective in such a context. The Europeans, in contrast, fought large infantry field armies and sieges, types of warfare that suited guns. As a result, historian Kenneth Chase argues, guns evolved more rapidly in Europe than in China.111 108

109 110 111

Shi Jungui 石俊贵 and Li Hong 李鸿, “Nei Meng gu chu tu de Ming chao chu nian tie ke di lei ji shi” 内蒙古出土的明朝初年铁壳地雷纪实, Qing bing qi 轻兵器 (Beijing), 2002 vol., no. 4, 36. DeVries and Smith, Artillery of the Dukes of Burgundy, p. 46. Lorge, Asian Military Revolution, p. 17. This argument is also put forth by the Mongol specialist Thomas Allsen, “The Circulation of Military Technology in the Mongolian Empire,” in Warfare in Inner Asian History (500–1800), ed. Nicola De Cosmo (Leiden, 2002), p. 286: “[I]t may be offered as a working hypothesis that those states and cultures with a lengthy history of interaction with the nomads, who for so long lived under the threat and the spell of an earlier ‘cavalry revolution,’ were the more



Early Gunpowder Warfare in Europe and China

275

The Chase hypothesis is a compelling one, but there are reasons to doubt it. For one, the Chinese themselves considered guns to be highly effective against nomads. The Ming used them to smash Mongol power in the late 1300s and early 1400s, as well as to defend against a huge Mongol invasion in 1449. Thereafter, guns were in great demand in China, especially on China’s northern frontiers, precisely where the inner Asian incursions arrived, and the fortifications of the North were studded with gun emplacements. And when Portuguese guns arrived in the 1500s, they were sought after not just to war against infantry armies but to use against nomads, their long barrels jutting out from the Great Wall. More importantly, the Chase hypothesis downplays the tremendous variety of warfare within China itself. Nomads may have been the primary enemies in the North, but southern China was often beset by warfare that was similar to that of Europe: huge infantry armies clashing with each other and attacking cities. The Chase hypothesis should certainly not be discarded, but it probably represents only part of the explanation. I believe that the most straightforward answer is most compelling: the Ming stopped improving their guns because they did not need to. Guns evolved quickly in the pre- and early Ming period, a time of constant and existential warfare. In the 1350s and 1360s, the founder of the Ming, Zhu Yuanzhang, fought incessantly with his rivals, and innovations in firearms provided an edge to those who made them. Even after he declared the establishment of the Ming Dynasty in 1368, he continued his conquests, west into Sichuan, north into Mongolia, southwest into Yunnan. These were massive wars against states whose militaries possessed, especially in the case of Sichuan, state-of-the-art weaponry. When Zhu Yuanzhang died, the Ming were rocked by an intense civil war, in which tremendous armies shot guns at each other in China’s heartland. The usurper won and immediately began carrying out other massive expeditions, most notably a massive war in what is today northern Vietnam and five great campaigns into Mongolia. His expeditions involved hundreds of thousands of men armed with tens of thousands of guns. They stimulated innovations in firearms manufacture, tactics, and administration, particularly, it seems, the Vietnam Wars, because the Dai Viet state was a sophisticated gunpowder empire in its own right. But after the usurper – the Yongle Emperor – died in 1424, the frequency and intensity of Chinese warfare decreased dramatically. From his death until the mid-1500s, there was only one dynasty-shaking military event: the Tumu Episode of 1449, when firearms played an important role in preserving the capital from a Mongol onslaught. Thereafter, as the Mongol threat lessened, warfare became less frequent, less intense, and, most importantly, less existential. Wars between 1449 and the 1540s tended to be police actions against minor

reluctant to accept the new technology diffused by the Mongols, the more hesitant to join the ‘gunpowder revolution,’ while those peoples on the extreme periphery of Eurasia, Europe and Japan, whose contact with the nomads was restricted and intermittent, were the more eager to interrogate and exploit its possibilities.”

276

Tonio Andrade

enemies. The Ming were overwhelmingly dominant. There was little challenge, little stimulus to further innovation. Europe, in contrast, saw no let-up in the nearly constant warfare that had marked the 1300s and early 1400s. That, more than anything else, probably explains why European guns kept improving. Europe was at the beginning of a long Warring States period, one that would last, with a few breaks, until 1945. China’s Ming Peace lasted a century, but in the 1550s, warfare began again, and it was intense and protracted, lasting through the late 1600s. During this period, China adopted, adapted, and innovated furiously, mastering muskets and advanced muzzle-loading cannons. By 1560, muskets were deployed with tactics superior to those of Europe; in the 1600s, the Chinese and their enemies were making wall-smashing cannons that in some ways were more sophisticated than those of Europe. This period of military innovation lasted until the next dynasty, the Qing (1644–1911), had concluded its own wars of conquest and consolidation in the mid-eighteenth century. We still have enormous amounts to learn about global military history, particularly for the key late medieval period, a watershed for the development of guns in both China and Europe. The speculations in this article are highly provisional, but I hope they may prove stimulating and help to provoke further discussion. Even more importantly, I hope that they help to bring together two very exciting fields – medieval European military history and Chinese military history – helping to point the way toward a truly global military history.

List of Contributors

Nicolás Agrait took his Ph.D. Fordham University 2003 and is currently Assistant Professor of History, Long-Island University-Brooklyn. His main interests are Iberian Medieval History, Medieval Military History, Castilian Military History 1200–1350. His recent publications include “The Battle of Salado (1340), Revisited,” Journal of Medieval Military History 10 (2013), 89–112; “El asta de la lanza: los mecanismos de financiación de la guerra durante el reinado de Alfonso XI (1312–50),” Gladius; Revista de estudios sobre armas antiguas, armamento, arte militar y vida cultural en oriente y occidente 32 (2012), 103–20 and “Castilian Military Reform under the Reign of Alfonso XI (1312–50),” Journal of Medieval Military History 3 (2005), 88–126. Tonio Andrade is Professor of History at Emory University. His latest book, The Gunpowder Age: China, Military Innovation, and the Rise of the West in World History, will appear next year. He is the author of two previous books, Lost Colony (2011) and How Taiwan became Chinese (2008), as well as articles in The Journal of Asian Studies, The Journal of World History, Late Imperial China, Itinerario, The Journal of Early Modern History, and other journals. David Bachrach is Professor of Medieval History at the University of New Hampshire. His main research foci are the military, administrative, economic, and political history of medieval Germany from the tenth to thirteenth centuries, and medieval England during the reigns of Henry III (1216–72) and Edward I (1272–1307). His recent publications include Warfare in Tenth-Century Germany (2012); Warfare and Politics in Medieval Germany c. 1000: On the Variety of our Times by Alpert of Metz (2012); The Histories of a Medieval City, Worms c. 1000–c.1300: Translation and Commentary (2014), and Widukind of Corvey, Deeds of the Saxons (2014). Oren Falk is Associate Professor, History and Medieval Studies, Department of History Cornell University. His interests focus on medieval Norse cultural history, especially clustering around issues of violence, gender, the Icelandic sagas in general and the Bishops’ Sagas in particular. Recent and forthcoming publications include articles on widowers, on bestiality and profanity, and a book entitled The Bare-Sarked Warrior: A Brief Cultural History of Battlefield Exposure. He is preparing work on Norse masculinity, on some modeling prob-

278

List of Contributors

lems having to do with the study of historical violence, and on translation of Icelandic sagas. Devin Fields is a Ph.D. student at Texas Tech University in Lubbock, Texas. He studies both Medieval and Military History. Michael Fulton is a Ph.D. candidate at Cardiff University studying the development of artillery and fortifications during the crusades under the supervision of Denys Pringle. He holds a B.A. and an M.A. in History from Queen’s University, Canada. His interests include various aspects of medieval and early modern siege warfare and the reciprocal development of masonry defenses across Europe and the Levant. He attempts to incorporate various elements of archaeology, geography and practical physics throughout his work. Forthcoming publications include a number of investigations of twelfth- and thirteenth-century sieges in and around the Latin East in which artillery played a prominent role, as well as an analysis of how artillery was treated by various sources in Western Europe. Thomas K. Heebøll-Holm was awarded a PhD from the Saxo-Institute, University of Copenhagen (2011). He is currently Assistant Professor of History, University of Southern Denmark. His principal publications include: “Apocalypse then? The First Crusade, Traumas of War and Thomas de Marle,” in Kerstin Hundahl, Lars Kjær and Niels Lund (eds.), Denmark and Europe in the Middle Ages, c: 1000–1525: Essays in Honour of Professor Michael H. Gelting (2014), pp. 237–54; Ports, Piracy, and Maritime War: Piracy in the English Channel and the Atlantic, c. 1280–c. 1330 (2013); “Between Pagan Pirates and Glorious Sea-Warriors: The Portrayal of the Viking Pirate in Danish TwelfthCentury Latin Historiography,” Viking and Medieval Scandinavia 8 (2013), 141–70; “Priscorum quippe curialum qui et nunc militari censentur nomine – riddere i Danmark i 1100-tallet,” Historisk Tidsskrift 109 (2009), 21–70. Dr Rabei G. Khamisy is an Honorary Associate, Cardiff University (SHARE). His main research is about the regions of Acre and Tyre in the Crusader period, due to be be published as a book titled : Fiefs, Fortresses, Villages and Farms in Western Galilee and Southern Lebanon in the Frankish period (1104–1291): Political, Social and Economic Activities. A particular topic is the treaties between the Mamluks and the Franks. Michael Livingston is an Associate Professor at The Citadel, the Military College of South Carolina. He is interested in the intersection of history and literature in the cultural remnants of the Middle Ages. Recent publications include casebooks on the Battle of Brunanburh (2011) and Owain Glyndwr (2013). His next book will be a casebook, co-edited with Kelly DeVries, on the battle of Crécy. Dan Spencer is a Ph.D. candidate in history, University of Southampton, supervised by Professor Anne Curry and Professor Mark Stoyle. His research investigates how and why English artillery developed over the course of the fifteenth century. His interests also include military history, castles studies, the battle



List of Contributors

279

of Agincourt, and naval and urban history. His publications include “Edward Dallingridge: Builder of Bodiam Castle,” Ex Historia 6 (2014), 81–98. L. J. Andrew Villalon did his degree work at Yale University. Having retired from the University of Cincinnati, where he is a professor emeritus, he moved to the Republic of Austin and is currently teaching as a senior lecturer at the University of Texas. A specialist in late medieval and early modern European history, Villalon has co-edited with Donald J. Kagay six collections of medieval essays, centering on violence and warfare, the most recent of which, dealing with the Hundred Years War, won the 2013 Verbruggen Prize for the best book on medieval military history. He has published over a score of journal articles and chapters in collections and has delivered numerous conference papers on a wide range of topics. In addition to research in his major field, he has also presented and published on automotive history and the history of World War I. He was vice president of the Texas Medieval Association (TEMA) in 2007–2008 and president in 2008–2009, at which time he organized the annual conference held in Austin. He is a founding member of De re militari: The Society for Medieval Military History, and is currently serving as its president. He was an associate editor of the Encyclopedia of Medieval Warfare and Military Technology (2010).

Journal of Medieval Military History 1477–545X

Volume I 1 The Vegetian “Science of Warfare” in the Middle Ages – Clifford J. Rogers 2 Battle Seeking: The Contexts and Limits of Vegetian Strategy – Stephen F. Morillo 3 Italia – Bavaria – Avaria: The Grand Strategy behind Charlemagne’s Renovatio Imperii in the West – Charles R. Bowlus 4 The Composition and Raising of the Armies of Charlemagne – John France 5 Some Observations on the Role of the Byzantine Navy in the Success of the First Crusade – Bernard S. Bachrach 6 Besieging Bedford: Military Logistics in 1224 – Emilie M. Amt 7 “To Aid the Custodian and Council”: Edmund of Langley and the Defense of the Realm, June–July 1399 – Douglas Biggs 8 Flemish Urban Militias against the French Cavalry Armies in the Fourteenth and Fifteenth Centuries – J. F. Verbruggen (translated by Kelly DeVries)

Volume II 1 The Use of Chronicles in Recreating Medieval Military History – Kelly DeVries 2 Military Service in the County of Flanders – J. F. Verbruggen (translated by Kelly DeVries) 3 Prince into Mercenary: Count Armengol VI of Urgel, 1102–1154 – Bernard F. Reilly 4 Henry II’s Military Campaigns in Wales, 1157–1165 – John Hosler 5 Origins of the Crossbow Industry in England – David S. Bachrach 6 The Bergerac Campaign (1345) and the Generalship of Henry of Lancaster – Clifford J. Rogers 7 A Shattered Circle: Eastern Spanish Fortifications and their Repair during the “Calamitous Fourteenth Century” – Donald Kagay 8 The Militia of Malta – Theresa M. Vann 9 “Up with Orthodoxy!” In Defense of Vegetian Warfare – John B. Gillingham 10 100,000 Crossbow Bolts for the Crusading King of Aragon – Robert Burns

Volume III 1 A Lying Legacy? A Preliminary Discussion of Images of Antiquity and Altered Reality in Medieval Military History – Richard Abels and Stephen F. Morillo

2 War and Sanctity: Saints’ Lives as Sources for Early Medieval Warfare – John France 3 The 791 Equine Epidemic and its Impact on Charlemagne’s Army – Carroll Gillmor 4 The Role of the Cavalry in Medieval Warfare – J. F. Verbruggen 5 Sichelgaita of Salerno: Amazon or Trophy Wife? – Valerie Eads 6 Castilian Military Reform under the Reign of Alfonso XI (1312–50) – Nicolás Agrait 7 Sir Thomas Dagworth in Brittany, 1346–7: Restellou and La Roche Derrien – Clifford J. Rogers 8 Ferrante d’Este’s Letters as a Source for Military History – Sergio Mantovani 9 Provisions for the Ostend Militia on the Defense, August 1436 – Kelly DeVries

Volume IV 1 The Sword of Justice: War and State Formation in Comparative Perspective – Stephen F. Morillo 2 Archery versus Mail: Experimental Archaeology and the Value of Historical Context – Russ Mitchell 3 “Cowardice” and Duty in Anglo-Saxon England – Richard Abels 4 Cowardice and Fear Management: The 1173–74 Conflict as a Case Study – Steven Isaac 5 Expecting Cowardice: Medieval Battle Tactics Reconsidered – Stephen F. Morillo 6 Naval Tactics at the Battle of Zierikzee (1304) in the Light of Mediterranean Praxis – William Sayers 7 The Military Role of the Magistrates in Holland during the Guelders War – James P. Ward 8 Women in Medieval Armies – J. F. Verbruggen 9 Verbruggen’s “Cavalry” and the Lyon-Thesis – Bernard S. Bachrach 10 Dogs of War in Thirteenth-Century Valencian Garrisons – Robert I. Burns

Volume V 1 Literature as Essential Evidence for Understanding Chivalry – Richard W. Kaeuper 2 The Battle of Hattin: A Chronicle of a Defeat Foretold? – Michael Ehrlich 3 Hybrid or Counterpoise? A Study of Transitional Trebuchets – Michael Basista 4 The Struggle between the Nicaean Empire and the Bulgarian State (1254–1256): Towards a Revival of Byzantine Military Tactics under Theodore II Laskaris – Nicholas S. Kanellopoulos and Joanne K. Lekea 5 A “Clock-and-Bow” Story: Late Medieval Technology from Monastic Evidence – Mark Dupuy 6 The Strength of Lancastrian Loyalism during the Readeption: Gentry Participation at the Battle of Tewkesbury – Malcolm Mercer 7 Soldiers and Gentlemen: The Rise of the Duel in Renaissance Italy – Stephen Hughes

8 “A Lying Legacy” Revisited: The Abels-Morillo Defense of Discontinuity – Bernard S. Bachrach

Volume VI

1 Cultural Representation and the Practice of War in the Middle Ages – Richard Abels 2 The Brevium Exempla as a Source for Carolingian Warhorses – Carroll Gillmor 3 Infantry and Cavalry in Lombardy (11th–12th Centuries) – Aldo Settia 4 Unintended Consumption: The Interruption of the Fourth Crusade at Venice and its Consequences – Greg Bell 5 Light Cavalry, Heavy Cavalry, Horse Archers, Oh My! What Abstract Definitions Don’t Tell Us about 1205 Adrianople – Russ Mitchell 6 War Financing in the Late-Medieval Crown of Aragon – Donald Kagay 7 National Reconciliation in France at the end of the Hundred Years War – Christopher Allmand

Volume VII 1 The Military Role of the Order of the Garter – Richard W. Barber 2 The Itineraries of the Black Prince’s Chevauchées of 1355 and 1356: Observations and Interpretations – Peter Hoskins 3 The Chevauchée of John Chandos and Robert Knolles: Early March to Early June, 1369 – Nicolas Savy 4 “A Voyage, or Rather an Expedition, to Portugal”: Edmund of Langley’s Journey to Iberia, June/July 1381 – Douglas Biggs 5 The Battle of Aljubarrota (1385): A Reassessment – João Gouveia Monteiro 6 “Military” Knighthood in the Lancastrian Era: The Case of Sir John Montgomery – Gilbert Bogner 7 Medieval Romances and Military History: Marching Orders in Jean de Bueil’s Le Jouvencel introduit aux armes – Matthieu Chan Tsin 8 Arms and the Art of War: The Ghentenaar and Brugeois Militia in 1477–79 – J. F. Verbruggen 9 Accounting for Service at War: The Case of Sir James Audley of Heighley – Nicholas Gribit 10 The Black Prince in Gascony and France (1355–6), according to MS 78 of Corpus Christi College, Oxford – Clifford J. Rogers

Volume VIII 1 People against Mercenaries: The Capuchins in Southern Gaul – John France 2 The Last Italian Expedition of Henry IV: Re–reading the Vita Mathildis of Donizone of Canossa – Valerie Eads 3 Jaime I of Aragon: Child and Master of the Spanish Reconquest – Donald Kagay 4 Numbers in Mongol Warfare – Carl Sverdrup 5 Battlefield Medicine in Wolfram’s Parzival – Jolyon T. Hughes

6 Battle-Seeking, Battle-Avoiding or Perhaps Just Battle-Willing? Applying the Gillingham Paradigm to Enrique II of Castile – L. J. Andrew Villalon 7 Outrance and Plaisance – Will McLean 8 Guns and Goddams: Was there a Military Revolution in Lancastrian Normandy 1415–50? – Anne Curry 9 The Name of the Siege Engine Trebuchet: Etymology and History in Medieval France and Britain – William Sayers

Volume IX 1 The French Offensives of 1404–1407 Against Anglo-Gascon Aquitaine – Guilhem Pépin 2 The King’s Welshmen: Welsh Involvement in the Expeditionary Army of 1415 – Adam Chapman 3 Gunners, Aides and Archers: The Personnel of the English Ordnance Companies in Normandy in the Fifteenth Century – Andy King 4 Defense, Honor and Community: the Military and Social Bonds of the Dukes of Burgundy and the Flemish Shooting Guilds – Laura Crombie 5 The Battle of Edgecote or Banbury (1469) Through the Eyes of Contemporary Welsh Poets – Barry Lewis 6 Descriptions of Battles in Fifteenth-Century Urban Chronicles: A Comparison of the Siege of London in May 1471 and the Battle of Grandson, 2 March 1476 – Andreas Remy 7 Urban Espionage and Counterespionage during the Burgundian Wars (1468–77) – Bastian Walter 8 Urban Militias, Nobles and Mercenaries: The Organization of the Antwerp Army in the Flemish–Brabantine Revolt of the 1480s – Frederik Buylaert, Jan Van Camp and Bert Verwerft 9 Military Equipment in the Town of Southampton During the Fourteenth and Fifteenth Centuries – Randall Moffett

Volume X 1 The Careers of Justinian’s Generals – David Parnell 2 Early Saxon Frontier Warfare: Henry I, Otto I, and Carolingian Military Institutions – David S. Bachrach and Bernard S. Bachrach 3 War in The Lay of the Cid – Francisco Garcia Fitz 4 The Battle of Salado (1340) Revisited – Nicolás Agrait 5 Chivalry and Military Biography in the Later Middle Ages: The Chronicle of the Good Duke Louis of Bourbon – Steven Muhlberger 6 The Ottoman-Hungarian Campaigns of 1442 – John Jefferson 7 Security and Insecurity, Spies and Informers in Holland during the Guelders War (1506–1515) –James P. Ward 8 Edward I’s Wars in the Chronicle of Hagnaby Priory – Michael Prestwich

Volume XI 1 Military Games and the Training of the Infantry – Aldo A. Settia 2 The Battle of Civitate: A Plausible Account – Charles D. Stanton 3 The Square “Fighting March” of the Crusaders at the Battle of Ascalon (1099) – Georgios Theotokis 4 How the Crusades Could Have Been Won: King Baldwin II of Jerusalem’s Campaigns against Aleppo (1124–5) and Damascus (1129) – T. S. Asbridge 5 Saint Catherine’s Day Miracle – the Battle of Montgisard – Michael Ehrlich 6 The Military Effectiveness of Alan Mercenaries in Byzantium, 1301–1306 – Scott Jesse and Anatoly Isaenko 7 Winning and Recalling Honor in Spain: Pro-English Poetry in Celebration of the Battle of Nájera (1367) – Donald J. Kagay and L. J. Andrew Villalon 8 The Wars and the Army of the Duke of Cephalonia Carlo I Tocco (c. 1375–1429) – Savvas Kyriakidis 9 Sir John Radcliffe, K. G. (d. 1441): Miles Famossissimus – A. Compton Reeves 10 Defense Schemes of Southampton in the Late Medieval Period, 1300–1500 – Randall Moffett 11 French and English Acceptance of Medieval Gunpowder Weaponry – Kelly DeVries

Volume XII 1 Some Observations Regarding Barbarian Military Demography: Geiseric’s Census of 429 and Its Implications – Bernard S. Bachrach 2 War Words and Battle Spears: The kesja and kesjulag in Old Norse Literature – K. James McMullen 3 The Political and Military Agency of Ecclesiastical Leaders in Anglo-Norman England: 1066–1154 – Craig M. Nakashian 4 Couched Lance and Mounted Shock Combat in the East: The Georgian Experience – Mamuka Tsurtsumia 5 The Battle of Arsur: A Short-Lived Victory – Michael Ehrlich 6 Prelude to Kephissos (1311): An Analysis of the Battle of Apros (1305) – Nikolaos S. Kanellopoulos and Ioanna K. Lekea 7 Horse Restoration (Restaurum Equorum) in the Army of Henry of Grosmont, 1345: A Benefit of Military Service in the Hundred Years’ War – Nicholas A. Gribit 8 The Indenture between Edward III and the Black Prince for the Prince’s Expedition to Gascony, 10 July 1355 – Mollie M. Madden 9 Investigating the Socio-Economic Origins of English Archers in the Second Half of the Fourteenth Century – Gary Baker 10 War and the Great Schism: Military Factors Determining Allegiances in Iberia – L. J. Andrew Villalon

De Re Militari and the Journal of Medieval Military History De re militari: Society for the Study of Medieval Military History (DRM) was founded in 1992 during the annual International Congress for Medieval Studies meeting at Kalamazoo, Michigan. Our organization takes its name from the military treatise by Flavius Renatus Vegetius, the famous Roman work. From its beginning, the Society has had its most significant presence at the Congress, where each year it holds its annual meetings and sponsors a number of sessions, as well as a special lecture delivered by a distinguished medieval military historian. In addition, the society has sponsored sessions at other conferences, including the International Medieval Congress in Leeds, UK, and the annual meeting of the Society for Military History (SMH). DRM maintains a website which has become a principal online source for scholarly information about warfare in the Middle Ages, easily available for use by scholars and students in the United States and abroad. On this website, the organization hosts many primary sources, articles, and dissertations, as well as an extensive book review section. The Society’s Journal of Medieval Military History (JMMH), founded in 2003, is published annually by Boydell & Brewer. It is very much in the interests of both the field of medieval military history, and scholars within that field, to support DRM and the JMMH. Consequently, we encourage scholars to take out membership. At the time of writing, the annual dues for DRM membership are $35 (if paid by check or money order); $38 (if paid by credit card). Payment should be made to the treasurer (or his/ her delegate), if at all possible by check or money order: please see the website for further details, www.deremilitari.org. A DRM membership includes: one hardcover copy of the annual JMMH the right to serve as an officer of DRM the right to vote in all elections of officers and on all issues relevant to the society preference in participating on panels sponsored by DRM the right to compete for the Gillingham Prize annually awarded to the best article in the JMMH the right to buy earlier copies of the JMMH at a reduced price. Those renewing their membership or joining for the first time can obtain an application form via the De re militari site at the following URL: www://deremilitari.org/society/drm-membership-form/ L. J. Andrew Villalon President, De Re Militari (2014–2017)

spine 24mm

The Journal’s hallmark is broad chronological, geographic and thematic coverage of the subject, bearing witness to ‘the range and richness of scholarship on medieval warfare, military institutions, and cultures of conflict that characterises the field’.  HISTORY CONTENTS DAVID BACHRACH THOMAS K. HEEBØLL-HOLM

Feudalism, Romanticism, and Source Criticism: Writing the Military History of Salian Germany When the Lamb Attacked the Lion: A Danish Attack on England in 1138?

MICHAEL S. FULTON

Development of Prefabricated Artillery during the Crusades

RABEI G. KHAMISY

Some Notes on Ayyu ¯bid and Mamluk Military Terms

OREN FALK

Helgastaðir, 1220: A Battle of No Significance?

NICOLÁS AGRAIT

Por la guarda de la mar: Castile and the Struggle for the Sea in the Thirteenth and Fourteenth Centuries

MICHAEL LIVINGSTON

The Battle of Hyddgen, 1401: Owain Glyndw ˆ r’s Victory Reconsidered

DAN SPENCER

The Provision of Artillery for the 1428 Expedition to France

DEVIN FIELDS

1471: The Year of Three Battles and English Gunpowder Artillery

L. J. ANDREW VILLALON

TONIO ANDRADE

“Cardinal Sins” and “Cardinal Virtues” of “El Tercer Rey,” Pedro González de Mendoza: The Many Faces of a Warrior Churchman in Late Medieval Europe

Cover: Coll. Bibliothèques municipales de Chambéry, MS13, fo. 105r (circa 1170–1190). The Italian artist who illuminated this copy of Gratian’s Decretum beautifully captured the deadly seriousness of medieval warfare – for townsmen as well as knights. Image used by generous permission of the Bibliothèques municipales de Chambéry. an imprint of Boydell & Brewer Ltd PO Box 9, Woodbridge IP12 3DF (GB) and 668 Mt Hope Ave, Rochester NY 14620–2731 (US) www.boydellandbrewer.com

FRANCE, DEVRIES, ROGERS (editors)

Late Medieval Divergences: Comparative Perspectives on Early Gunpowder Warfare in Europe and China

JOURNAL OF MEDIEVAL MILITARY HISTORY VIII JOURNAL OF MEDIEVAL MILITARY HISTORY

XIII

J ournal of

MEDIEVAL MILITARY HISTORY

·  X III · Edited by JOHN FRANCE, KELLY DEVRIES and CLIFFORD J. ROGERS