APPENDIX A
NOTE ON THE SOURCES
In a field already so well explored as that of Normandy and England in the eleventh and twelfth centuries, there is little need to enter into a detailed discussion of primary materials. A brief review, however, of the sources upon which the present volume is based may be a convenience and serve a useful purpose.
Among the narrative sources for the life of Robert Curthose, the Historia Ecclesiastica[1] of Ordericus Vitalis is, of course, by far the most important. One of the greatest historical writers of the twelfth century, the monk of Saint-Évroul has treated of Robert’s character and career at great length and with much vivacity and insight. And while one may admit with Gaston Le Hardy[2] that he was no friend of the duke, indeed, that as a churchman and as a lover of peace and of strong and orderly government he was strongly prejudiced against him and sometimes treated him unfairly, still it must be confessed that in the main his strictures are confirmed by other evidence and are presumably justified. Unfortunately, Ordericus Vitalis stands almost alone among early Norman writers in paying attention to the career of Robert Curthose. Some assistance, however, has been gained from William of Poitiers[3] and from the Gesta Normannorum Ducum, a composite work once solely attributed to William of JumiÈges, but now at last made available in a critical edition which distinguishes the parts actually written by William of JumiÈges, Ordericus Vitalis, Robert of Torigny, and others.[4] The Roman de Rou of Wace[5] has also been drawn upon, sometimes rather freely, but it is hoped always with due caution and discretion, for much picturesque detail concerning events in western Normandy, about which the author clearly possessed special information. For Robert’s relations with Maine, the contemporary Actus Pontificum Cenomannis in Urbe degentium[6] have been an almost constant guide, often confirming and even supplementing the more extensive but less precise narrative of Ordericus Vitalis. Matter of much importance has also from time to time been gleaned from the works of French and Flemish writers, such as the famous Vie de Louis le Gros by Abbot Suger of Saint-Denis,[7] the anonymous Chronique de Morigny,[8] and the Histoire du Meurtre de Charles le Bon by Galbert of Bruges.[9]
The English writers of the period have naturally proved invaluable. Of these, William of Malmesbury,[10] as we should expect, possesses the keenest insight into Robert’s character; but the Anglo-Saxon Chronicle treats[11] of the events of Robert’s life with greater fulness and in more coherent and trustworthy chronological order. Florence of Worcester[12] is in general dependent upon the Chronicle, but occasionally he presents a different view or supplementary matter of independent value; and the same may be said of the Historia Regum, which is commonly attributed to Simeon of Durham,[13] in its relation to Florence of Worcester. Henry of Huntingdon,[14] who is also largely dependent upon the Anglo-Saxon Chronicle, professes himself a first-hand authority from the accession of Robert Curthose and William Rufus to the ducal and royal thrones in 1087;[15] and his narrative becomes increasingly valuable as it advances, though he cannot be considered a really independent writer before 1126, i.e., a score of years after the close of Duke Robert’s active career at the battle of Tinchebray. For all the facts bearing upon Robert’s life with which it deals, the Historia Novorum in Anglia of Eadmer,[16] the companion and confidential adviser of Archbishop Anselm, is a strictly contemporary narrative of the highest value, though its specialized character considerably restricts its usefulness for the purposes of the present study. The brief chronicle of Hyde abbey,[17] which was compiled during the reign of Henry I, has often proved helpful, as have also other minor monastic narratives such as the chronicle of Abingdon[18] and the annals of Winchester,[19] of Waverley,[20] etc.
The documentary sources for the life of Robert Curthose are very meagre; but, such as they are, they are now all conveniently accessible. As a result of prolonged researches in the archives and libraries of Normandy and in the BibliothÈque Nationale, and after a careful sifting of all the printed materials, Professor Charles H. Haskins has been able to give us, in another volume of the Harvard Historical Studies, a definitive edition of seven hitherto unpublished ducal charters, together with a complete and annotated list of all the charters of the reign.[21] The best guides to the remainder of the documentary material bearing upon Robert’s life are the Regesta Regum Anglo-Normannorum by H. W. C. Davis[22] and the Calendar of Documents preserved in France illustrative of the History of Great Britain and Ireland by J. H. Round.[23] While both these works leave something to be desired, they have proved invaluable in the preparation of the present study; and it is earnestly to be hoped that the publication of the second volume of Davis’s work, containing the charters of Henry I, will not be long delayed.[24] For the full texts of documents, and for other scattered materials not calendared by either Round or Davis, it has been necessary to consult many special collections, e.g., the Livre noir of Bayeux cathedral,[25] the Chartes de Saint-Julien de Tours,[26] the Cartulaire de l’abbaye de Saint-Vincent du Mans,[27] the letters of Pascal II,[28] of Ivo of Chartres,[29] and of St. Anselm,[30] which are too numerous to be listed here in detail, and which have been fully cited in their proper places in footnotes.
The Crusade forms a special chapter in the record of Robert’s life for which it is necessary to draw upon a different group of sources. Of works by contemporary or early writers on the Crusade, the anonymous Gesta Francorum[31] is, of course, invaluable for all the facts with which it deals; but the Historia Hierosolymitana of Fulcher of Chartres[32] has proved of even greater service in the present study, because of the author’s close association with Robert Curthose on the Crusade from the time when the expedition left Normandy until it reached Marash in Armenia; concerning later events also Fulcher was by no means ill informed. The Historia Francorum qui ceperunt Iherusalem of Raymond of Aguilers[33] is also a first-hand narrative by an eyewitness; and, while the author is at times rather hostile to Duke Robert and the Normans, he is nevertheless invaluable as representing the point of view of the ProvenÇaux. Inferior to any of the foregoing, but still by a writer who was in the East and who was well informed, the Gesta Tancredi of Ralph of Caen[34] has proved of great assistance, as has also the voluminous, but less trustworthy, work of Albert of Aix,[35] which, when it has been possible to check it with other evidence, has contributed valuable information. Of western writers on the Crusade who did not actually make the pilgrimage to the Holy Land, apart from Ordericus Vitalis,[36] who has already been mentioned, Guibert of Nogent[37] and Baldric, archbishop of Dol,[38] have been most helpful. The English writers, except William of Malmesbury,[39]—whose account is based almost wholly upon Fulcher of Chartres, and, apart from an occasional detail, is of little value—have not treated the Crusade with any fulness, and are of little service except for the beginnings of the movement. Of the Greek sources only the Alexiad of Anna Comnena[40] has been of much assistance. The Oriental writers are in general too late to be of great importance for the First Crusade, and they had, of course, no particular interest in Robert Curthose; but their writings have not been overlooked, and Matthew of Edessa,[41] Ibn el-Athir,[42] Kemal ed-Din,[43] and Usama ibn Munkidh[44] have been of service. The contemporary letters bearing upon the Crusade have been admirably edited, with exhaustive critical notes, by Heinrich Hagenmeyer.[45] Of charters, or documents in the strict sense of the word, there are almost none relating to the Crusade; but such as there are, they have been rendered easily accessible by the painstaking calendar of documents dealing with the history of the Latin Kingdom of Jerusalem by Reinhold RÖhricht.[46] It would be going too far afield to describe at this point the scattered materials from which the attempt has been made to draw up a list of the known associates and followers of Robert on the Crusade. They are fully cited in Appendix D.
For the chapter on Robert Curthose in legend, with which the narrative part of the present volume ends, it has been necessary to depart from the narrow chronological limits within which the rest of our researches have been conducted, and to explore a wide range of literature extending to the close of the Middle Ages. Most of the Robert legends make their appearance early, and can be traced to a certain extent in William of Malmesbury and Henry of Huntingdon and in Robert the Monk and Ralph of Caen. But their elaboration was in the main the work of chroniclers and romancers of a later period. Among Norman and English sources, the works of Geoffrey Gaimar, Wace, William of Newburgh, Ralph de Diceto, and Ralph Niger have proved most helpful for the twelfth century; of Roger of Wendover, Matthew Paris, and Robert of Gloucester, together with the anonymous Flores Historiarum and Livere de reis de Engletere, for the thirteenth; of Peter Langtoft, Ranulf Higden, and Henry Knighton, together with the anonymous Eulogium Historiarum, for the fourteenth; while Thomas Walsingham in the fifteenth century has occasionally been of service. Much material of a legendary character relating to Robert’s exploits in the Holy War has also been gleaned from the various versions of the poetic cycle of the Crusade, the most notable of which are the Chanson d’Antioche of the late twelfth century, the Chanson de JÉrusalem, which probably dates from the thirteenth century, and the Chevalier au Cygne et Godefroid de Bouillon, edited by the Baron de Reiffenberg, which belongs to the fourteenth or fifteenth century. Such detailed criticism as it has seemed necessary to make of these widely scattered materials bearing upon Robert Curthose in legend has been placed in the footnotes of Chapter VIII, where the editions used have also been fully cited.
APPENDIX B
DE INIUSTA VEXATIONE WILLELMI EPISCOPI PRIMI[1]
The anonymous tract De Iniusta Vexatione Willelmi Episcopi Primi[2] is worthy of more attention and of a more critical study than it has yet received.[3] Since it gives the only detailed account which we possess of the dispute between William Rufus and William of Saint-Calais, bishop of Durham, and of the trial of the latter before the curia regis at Salisbury upon a charge of treason in connection with the rebellion of 1088, final judgment as to the bishop’s guilt or innocence must in large measure depend upon a just estimate of its value. Freeman was very reluctant to recognize its high authority as compared with his favorite ‘southern writers,’ the Anglo-Saxon chronicler, Florence of Worcester, and William of Malmesbury;[4] but his distrust appears to be unwarranted.
The tract is manifestly made up of two distinct parts: (1) the main body of an original libellus, concerned exclusively with the bishop’s ‘vexation,’ and beginning (p. 171), “Rex Willelmus iunior dissaisivit Dunelmensem episcopum,” and ending (p. 194), “rex permisit episcopo transitum”; and (2) introductory and concluding chapters, which contain a brief sketch of the bishop’s career before and after his unfortunate quarrel with the king and his expulsion from the realm. The joints at which the separate narratives are pieced together are apparent upon the most cursory examination. Not only is there a striking contrast between the detailed and documentary treatment found in the body of the libellus and the bare summaries which make up the introductory and concluding paragraphs, but the reader is actually warned of the transition in the last sentence of the introduction by the phrase (p. 171), “quam rem sequens libellus manifestat ex ordine.” The two parts of the tract are evidently derived from different sources and written at different times by different authors.
The libellus properly so called, i.e., the central portion of the tract, is a narrative well supplied with documents; it has all the appearance of being contemporary and by an eyewitness, and is manifestly a source of the greatest value for the facts with which it deals. Liebermann, with his unrivalled knowledge of mediaeval English legal materials, has declared that there is no ground for doubting its authenticity;[5] and Professor G. B. Adams, who also finds abundant internal evidence of its genuineness, points out, as an indication that it was written by an eyewitness in the company of Bishop William, the fact that no attempt is made to tell what went on within the curia while the bishop and his supporters were outside; and further, he considers it more “objective and impartial” than Eadmer’s better known account of the trial of Anselm before the council of Rockingham.[6] The author, it may be conjectured, was a monk of Durham who stood in somewhat the same favored position among the intimates of Bishop William as that occupied by Eadmer with regard to Anselm; and while we know nothing of his personality, it is perhaps worth remarking in passing that he may very well be the ‘certain monk’ (quendam suum monachum) who acts on at least two occasions as the bishop’s messenger (pp. 172, 175). The account in the earlier instance is so intimate and personal as strongly to support this hypothesis: “Ipsum quoque monachum episcopi, qui de rege redibat, accepit et equum suum ei occidit; postea peditem abire permisit.”
The introduction and the conclusion of the tract, on the other hand, are not a first-hand narrative; and fortunately we possess the source from which they are derived. The introduction (pp. 170 f.), dealing with the bishop’s career prior to 1088, contains nothing which is not told with much greater fulness in the opening chapters of the fourth book of the Historia Dunelmensis Ecclesiae of Simeon of Durham.[7] It is in fact a mere summary of those chapters; and while the author is no servile copyist, he evidently had no other source of information. It seems safe to conclude, therefore, that he was not identical with the author of the original libellus. Judged by style and method, the conclusion of the tract (pp. 194 f.) appears to be by the same author as the introduction. It, too, is clearly an abridgment of certain chapters of the Historia Dunelmensis Ecclesiae,[8] though with this notable difference from the introduction, that it contains some matter not to be found in the Historia, e.g., the statement that the exiled bishop was intrusted by the duke with the administration of all Normandy, and the notices of the expedition of William Rufus against King Malcolm in 1091, and of the presence of the Scottish king at the laying of the first stones in the foundation of the new cathedral at Durham in 1093. Apparently, for these more recent events, the writer was drawing upon his own first-hand knowledge. The date at which the introductory and concluding chapters were appended to the original Durham libellus cannot be fixed with exactness. The reference to Anselm as “sanctae memoriae” (p. 195) shows that they were written after his death in 1109;[9] and since, as will appear below, they in turn were used in the Historia Regum, which is commonly attributed to Simeon of Durham, the terminus ad quem cannot be placed much later than 1129.[10]
The relationship between the above mentioned additions to the Durham libellus and the Historia Regum may be displayed by the following quotations.
The introduction to the Durham tract closes with the following sentence (p. 171):
… sed orta inter regem et primates Angliae magna dissensione, episcopus [i.e., William of Durham] ab invidis circumventus usque ad expulsionem iram regis pertulit, quam rem sequens libellus manifestat ex ordine;
and the conclusion opens as follows (pp. 194 f.):
Anno sui episcopatus octavo expulsus est ab Anglia, sed a Roberto fratre regis, comite Normannorum, honorifice susceptus, totius Normanniae curam suscepit. Tertio autem anno, repacificatus regi, recepit episcopatum suum, ipso rege cum fratre suo totoque Angliae exercitu, cum Scotiam contra Malcolmum tenderent, eum in sedem suam restituentibus, ipsa videlicet die qua inde pulsus fuerat. Tertio Idus Septembris, secundo anno suae reversionis, ecclesiam veterem, quam Aldunus quondam episcopus construxerat, a fundamentis destruxit.
The account of the rebellion of 1088 in the Historia Regum—at this point almost wholly independent of Florence of Worcester—ends with the expulsion, not of Bishop William of Durham, but of Bishop Odo of Bayeux:
… et ita episcopus [i.e., Odo] qui fere fuit secundus rex Angliae, honorem amisit irrecuperabiliter. Sed episcopus veniens Normanniam statim a Rodberto comite totius provinciae curam suscepit; cuius ordinem causae libellus in hoc descriptus aperte ostendit. Etiam Dunholmensis episcopus Willelmus, viii. anno episcopatus, et multi alii, de Anglia exierunt.[11]
And in a later passage the king’s restoration of Bishop William to his see is thus recorded:
Veniens Dunelmum, episcopum Willelmum restituit in sedem suam, ipso post annos tres die quo eam reliquit, scilicet iii. idus Septembris.[12]
Thomas Arnold, the editor of Simeon’s Opera, remarks upon the clause “cuius ordinem causae libellus in hoc descriptus aperte ostendit” of the Historia Regum, “This ‘libellus,’ describing Odo’s administration in Normandy, appears to be lost.”[13] Taken by itself the passage is obscure, and it is perhaps not surprising that the editor wholly mistook its meaning. But a comparison of it with the clause “quam rem sequens libellus manifestat ex ordine” of the Durham tract at once reveals dependence and resolves the difficulty. The verbal similarities are striking, and the author of course uses the puzzling “causae” because the source from which he drew was in fact the account of a causa, viz., the trial of William of Saint-Calais before the curia regis. It is clear, therefore, that the libellus to which the author of the Historia Regum refers his readers is not a lost treatise on the administration of Bishop Odo in Normandy—as Arnold supposed—but in fact the Durham tract on the ‘unjust vexation’ of Bishop William, which Arnold had himself already published in the first volume of Simeon’s works. A further comparison of all the passages which have been indicated by italics in the foregoing excerpts fully confirms this conclusion and reveals the extent of the debt of the Historia Regum to the Durham tract. Not only the verbal agreements but the close similarities in thought are so marked as to preclude every possibility of independence.
We are now in a position to see how the author of the Historia Regum worked. Having before him the chronicle of Florence of Worcester—which he regularly followed—with its dark picture of Bishop William’s treason, and the elaborate Durham tract in his defence, he chose to suppress all reference to the bishop of Durham in connection with the rebellion, and substituted for him Odo of Bayeux as a scapegoat. Then at the end of his chapter he added, apparently as an afterthought, and borrowing directly from the Durham tract, that Bishop William ‘departed’ from England in the eighth year of his episcopate. The statement of the Historia Regum, therefore, that Odo of Bayeux upon his expulsion from England after the fall of Rochester went to Normandy and had the ‘care’ of the whole duchy committed to his charge, is valueless. If that honor belongs to any one, it is to William of Saint-Calais, bishop of Durham, as set forth in the conclusion of the tract De Iniusta Vexatione.[14]
But the author of the Historia Regum was a clumsy borrower, and we have not yet reached the end of the confusion which has arisen as the result of his easy way of juggling with his sources. In a later passage in which he deals with the return of Bishop William to his see at the time of the expedition of William Rufus against King Malcolm in 1091, he explains that the restoration of the bishop took place on the third anniversary of his retirement, “that is, on the 3d before the Ides of September.” Freeman, relying upon this text, but apparently mistaking Ides for Nones, states that the arrival of the king in Durham and the reinstatement of the bishop took place on 3 September.[15] Comparison with the parallel text of the Durham tract, however, makes it clear that the author of the Historia Regum has here again made an unintelligent and altogether misleading use of his source, copying almost verbatim, but detaching the phrase “iii. idus Septembris” from the next sentence, where it properly refers to an event of the year 1093. It is necessary, therefore, to get back to the evidence of the De Iniusta Vexatione, which not only says that Bishop William was reinstated on the third anniversary of his expulsion, but fixes that earlier date with exactness: “Acceperunt ergo Ivo Taillesboci et Ernesius de Burone castellum Dunelmense in manus regis, et dissaisiverunt episcopum de ecclesia et de castello, et de omni terra sua xviii. kal. Decembr.” (p. 192). The bishop’s restoration, accordingly, should be dated 14 November 1091. If it cause surprise that William Rufus should have undertaken a campaign in the northern country so late in the season, it may be noted that he previously had his hands full with an expedition against the Welsh,[16] and that Florence of Worcester in describing the campaign makes the significant statement, “multique de equestri exercitus eius fame et frigore perierunt.”[17]
It remains to raise a question as to the authorship of the Historia Regum. As is well known, the evidence on which both it and the Historia Dunelmensis Ecclesiae are attributed to Simeon of Durham is not contemporary and not conclusive,[18] though a better case can be made out for the latter than for the former. Without discussing this evidence anew, and without entering at this time upon the more extended inquiry as to whether it is credible that two works of such different character and of such unequal merit can be by a single author, it is still pertinent here to remark their striking difference in point of view with regard to the controversy between William Rufus and the bishop of Durham. The Historia Dunelmensis Ecclesiae speaks of the quarrel and of the bishop’s expulsion and exile without any reserve; and, moreover, it contains remarkably full information concerning his fortunes while in exile.[19] In all this it is freely reproduced in the additions to the Durham libellus (pp. 171, 194 f.). And they in turn are used by the author of the Historia Regum.[20] Yet with these additions and the original libellus and Florence of Worcester all before him, he suppresses every reference to the alleged treason of Bishop William, persistently declines to use such words as expulsion and exile in connection with him, and steadily ignores the quarrel. For him the bishop ‘went out’ of England, although he unconsciously slips into an inconsistency in a later passage when he notes that the bishop was ‘restored’ to the see which he had ‘left.’[21] If the Historia Dunelmensis Ecclesiae and the Historia Regum are by one and the same author, then assuredly he had a bad memory for what he had himself previously written, and his point of view had curiously shifted during the intervening years.
APPENDIX C
ARNULF OF CHOCQUES, CHAPLAIN OF ROBERT CURTHOSE
Arnulf of Chocques, who went on the First Crusade with Robert Curthose and ended his dramatic career in 1118 as patriarch of Jerusalem, is a character of more than ordinary interest, and his provenance and early career are worthy of more careful investigation than they have yet received.[1] The foundation for such a study was laid in 1904, when, by the publication in a new and scholarly edition of a little-known text of the early twelfth century, entitled Versus de Viris Illustribus Diocesis Tarvanensis, the Belgian scholar Charles Moeller identified Arnulf’s birthplace as the village of Chocques in the diocese of ThÉrouanne on the river Clarence, an affluent of the Lys.[2] Thus Moeller returns to the view of the Flemish annalists Meyer and Malbrancq,[3] who manifestly knew and used this text; though modern writers upon the Crusades, overlooking it and relying mainly upon Albert of Aix,[4] have said that Arnulf was ‘of Rohes, a castle of Flanders,’ which no one has ever been able to identify.[5] If further evidence were needed to establish the correctness of Moeller’s conclusion, it is found in a charter of 15 August 1095 by Robert Curthose in favor of Rouen cathedral, among the witnesses to which appears “Ernulfo de Cioches capellano meo.”[6] This document is also important as confirming and supplementing the meagre notices of the chroniclers, on which one is compelled to rely almost entirely for all that is known about Arnulf of Chocques before he went on the Crusade and came into prominence and controversy.
As to Arnulf’s family, practically nothing is known; though one may safely infer that he was of lowly origin from the speech which his friend and former pupil, Ralph of Caen, puts into his mouth when he makes him say to the princely leaders of the Crusade, “You have promoted me from a humble station, and from one unknown you have made me famous, and, as it were, one of yourselves.”[7] His enemies openly charged that he was the son of a priest;[8] and that their accusations were not without foundation is evidenced by a letter of Pope Pascal II, replying in 1116 to complaints which had been made against Arnulf, and reinstating him in the patriarchal office from which he had been suspended by the papal legate. While clearing him entirely from two of the charges which had been brought against him, the pope announced that the third complaint, viz., the general belief as to a stain upon his birth, was to be overlooked, ‘by apostolic dispensation,’ in view of Arnulf’s great services and of the needs of the church.[9] The statement sometimes made that Arnulf had a niece named Emma, or Emelota,[10] who figures in the charters of the Latin Kingdom,[11] and who was the wife, first of Eustace Gamier, lord of Caesarea, and then of Hugh II, count of Jaffa, appears to rest upon the sole authority of William of Tyre.[12]
Considering the age in which he lived, Arnulf doubtless received an excellent education,[13] though where it is impossible to say; and while still a young man he appeared in Normandy as a teacher, presumably at Caen. Ralph of Caen, who later became the distinguished historian of the First Crusade, was among his pupils; and upon the completion of his great work, the Gesta Tancredi, dedicated it in grateful remembrance to his old master.[14]
Far more important for Arnulf’s future, however, was the connection which he early established with the Anglo-Norman ruling family when he was made tutor in grammar and dialectic to the oldest daughter of William the Conqueror, Cecilia, the pious nun of La TrinitÉ at Caen, who later became the second abbess of her mother’s great foundation.[15] It was probably through the friendship thus established with the royal princess that the Flemish schoolmaster succeeded in rising to higher things; for Cecilia is said to have obtained from her indulgent brother, Duke Robert, the promise of episcopal honors for Arnulf, in case any of the Norman bishoprics should fall vacant;[16] and while he never gained that preferment, it can hardly be doubted that it was through her influence that he entered the service of the duke as chaplain. The charter to which attention has been called above furnishes proof that Arnulf already held that position in August 1096 (supra, n. 6). But his official connection with the ducal court undoubtedly began at least a year earlier, for the contemporary biographer of Abbot William of Bec states very specifically that on, or shortly after, 10 August 1094 he went on an important official errand for the duke in the capacity of ‘chancellor.’[17]
One other fact remains to be noticed as indicating Arnulf’s intimate relationship with another member of the Conqueror’s family. Although he was chaplain of the duke before and during the Crusade, he is said to have set out for the Holy War in the company of Robert’s uncle, Bishop Odo of Bayeux, who upon his death at Palermo, early in 1097, left him the greater part of his splendid outfit.[18]
APPENDIX D
ROBERT’S COMPANIONS ON THE CRUSADE
It cannot be said with certainty that every one who appears in the ensuing list actually went on the First Crusade with Robert Curthose. Since it was desired to make the list as complete as possible, doubtful names have been included and marked with an asterisk (*). The evidence is fully set forth in each case, so that no confusion can arise.
1. Alan, “dapifer sacrae ecclesiae Dolensis archiepiscopi.” Baldric of Dol, in H. C. Oc., iv, p. 33; Ordericus, iii, p. 507.
2. Alan Fergant, duke of Brittany. His presence is recorded at the siege of Nicaea (Albert of Aix, in H. C. Oc., iv, p. 316) and at the siege of Antioch (Baldric of Dol, ibid., p. 50, n. 9, being the variant from MS. G). His absence from Brittany during the Crusade is indicated by his disappearance from the charters of the period. The latest document which I have noted in which he appears before his departure is dated 27 July 1096. Cartulaire de l’ abbaye de Sainte-Croix de QuimperlÉ, ed. LÉon MaÎtre and Paul de Berthou, 2d ed. (Paris, 1904), no. 82, pp. 234-235. He was back again in Brittany 9 October 1101, when he made grants in favor of the abbey of Marmoutier. P. H. Morice, MÉmoires pour servir de preuves À l’histoire ecclÉsiastique et civile de Bretagne (Paris, 1742-46), i, cols. 505, 507; cf. col. 504.
3. Alan, son of Ralph de Gael. He was present with Robert at Nicaea, and advanced with him from there. Baldric of Dol, in H. C. Oc., iv, p. 33; Ordericus, iii, p. 507.
4. Alberic of Grandmesnil. Ordericus, iii, p. 484; cf. supra, p. 107, n. 88.
5. Anonymous, engineer of Robert of BellÊme: “ingeniosissimum artificem, … cuius ingeniosa sagacitas ad capiendam Ierusalem Christianis profecit.” Ordericus, iii, p. 415.
6. *Anonymous, wife of Thurstin, prÉvÔt of Luc. See no. 44 infra.
7. *Anonymous, son of Thurstin, prÉvÔt of Luc. See no. 44 infra.
8. Arnulf of Chocques, chaplain of Robert Curthose. Raymond of Aguilers, in H. C. Oc., iii, pp. 281, 302. Cf. Appendix C.
9. Arnulf of Hesdin: “Ernulfus de Hednith,” who was accused of complicity in Robert Mowbray’s conspiracy, and cleared himself by a judicial duel; but “tanto dolore et ira est commotus, ut abdicatis omnibus quae regis erant in Anglia, ipso rege invito et contradicente, discederet; associatus autem Christianorum exercitui, Antiochiam usque devenit, ibique extremum diem clausit. Cumque ei infirmanti principes medicorum curam adhibere vellent, respondisse fertur, ‘Vincit Dominus quare medicus me non continget, nisi ille pro cuius amore hanc peregrinationem suscepi.’” Chronicon, in Liber de Hyda, pp. 301-302. Arnulf ceases to appear in charters from about the period of the First Crusade. Cf. Davis, Regesta, nos. 315, 319; Round, C. D. F., no. 1326.
10. *AubrÉe la Grosse. See no. 20 infra.
11. Bernard of Saint-Valery. Baldric of Dol, in H. C. Oc., iv, p. 33; Ordericus, iii, p. 507. Ralph of Caen credits him with having been the first to scale the wall of Jerusalem. H. C. Oc., iii, p. 693.
12. Conan de Lamballe, second son of Geoffrey I, called Boterel, count of Lamballe. He was present with Robert at Nicaea and advanced with him from there. Baldric of Dol, in H. C. Oc., iv, pp. 28, 33; Albert of Aix, ibid., p. 316; Ordericus, iii, pp. 503, 507. He was killed by the Turks at Antioch 9 February 1098. Ralph of Caen saw his tomb there years afterwards. H. C. Oc., iii, p. 648.
13. Edith, wife of Gerard of Gournay and sister of William of Warenne. Her husband died on the Crusade, and she returned and became the wife of Dreux de Monchy. Interpolations de Robert de Torigny, in William of JumiÈges, pp. 277-278.
14. Emma, wife of Ralph de Gael and daughter of William Fitz Osbern. She accompanied her husband on the Crusade. Ordericus, ii, p. 264; Interpolations de Robert de Torigny, in William of JumiÈges, p. 287.
15. Enguerrand, son of Count Hugh of Saint-Pol. He died at Marra in Syria. Albert of Aix, in H. C. Oc., iv, pp. 372, 451; Raymond of Aguilers, ibid., iii, p. 276.
16. *Eustace III, count of Boulogne. It seems impossible to determine the route taken by Eustace of Boulogne on the First Crusade. According to the Anglo-Saxon Chronicle (a. 1096), Henry of Huntingdon (p. 219), and Albert of Aix (H. C. Oc., iv, p. 314), he went with Robert Curthose; Baldric of Dol (ibid., p. 20), Ordericus Vitalis (iii, pp. 484-485), and Robert the Monk (H. C. Oc., iii, p. 732), on the other hand, all say that he went with his brother Godfrey of Bouillon. Cf. G. F., p. 465, n. 17.
17. Fulcher of Chartres, historian of the Crusade. See the introduction to Hagenmeyer’s edition of the Historia Hierosolymitana.
18. Geoffrey Chotard, one of the barons (proceres) of Ancenis: “anno dedicationis Maioris Monast. ab Urbano papa facte statim post Pascha, cum dominus abbas noster tunc temporis Bernardus rediret a Nanneto civitate per Ligerim, anno scilicet ordinationis sue .xiii. venit ad portum Ancenisi,” and Geoffrey Chotard, “post parum temporis iturus in Ierusalem cum exercitu Christianorum super paganos euntium,” came to him and granted to Saint-Martin freedom from customs on the Loire. P. H. Morice, Preuves, i, col. 488.
19. Gerard of Gournay. Ordericus, iii, pp. 484, 507; Albert of Aix, in H. C. Oc., iv, p. 316; Baldric of Dol, ibid., p. 33. He was accompanied by his wife Edith, and died on the Crusade. Interpolations de Robert de Torigny, in William of JumiÈges, pp. 277-278. Cf. no. 13 supra.
20. *Gilbert, an architect (?). “Tunc Gislebertus, quidam laicus, de Ierusalem Rotomagum venit, et a praefato patre [i.e., Abbot Hilgot of Saint-Ouen, 1092-1112] ad monachatum susceptus, ecclesiae suae digniter profecit. Opus enim basilicae, quod iamdudum admiranda magnitudine intermissum fuerat, assumpsit; ibique pecuniam Alberadae Grossae, dominae suae, quae, in via Dei moriens, thesaurum ei suum commendaverat, largiter distraxit, et inde, aliorum quoque fidelium subsidiis adiutus, insigne opus perficere sategit.” Ordericus, iii, pp. 432-433.
21. Gilbert, bishop of Évreux. He was present at the council of Clermont as legatus of his fellow bishops. Ordericus, iii, p. 470. He was with Bishop Odo of Bayeux at the time of the latter’s death at Palermo early in 1097. Ibid., iv, pp. 17-18; iii, p. 266. Cf. no. 29 infra. If Gilbert completed the Crusade, he must have returned from Jerusalem far more quickly than most of his comrades, for he was back in Normandy by the middle of November 1099. Ordericus, iv, p. 65; cf. v, pp. 159, 195-196.
22. *Guy, eldest son of Gerard le Duc. He received five solidi from Saint-Vincent of Le Mans “cum pergeret ad Ierusalem cum Pagano de Monte Dublelli.” Cartulaire de S.-Vincent, no. 666. The editors, without good reason, date the document “circa 1096.” Cf. no. 30 infra.
23. *Guy de SarcÉ, a knight of Saint-Vincent of Le Mans. He surrendered his fief to the abbot and monks of Saint-Vincent, and received from them 20 livres manceaux and 300 solidi. This was done in the chapter on 22 June 1096, “eo videlicet anno quo Urbanus papa adventu suo occiduas illustravit partes, quoque etiam innumerabiles turbas populorum admonitione sua, immo vero Dei suffragante auxilio, Ierosolimitanum iter super paganos adire monuit.” It is not improbable that Guy’s brothers, Nicholas and Pain, accompanied him on the Crusade. Cartulaire de S.-Vincent, no. 317. This charter was witnessed, among others, by William de Braitel, who is no. 47 of our list infra.
24. *Hamo de Huna. He made a grant to Saint-Vincent of Le Mans on 29 July 1096; and “post non multum vero temporis … antequam Ierusalem iret quo tendere volebat,” he added another gift, and received from the monks 20 solidi. “Hoc actum fuit in domo monachorum apud Bazogers, in adventu Domini iv die ante natale Domini.” Cartulaire de S.-Vincent, no. 460. This was 22 December, presumably of the year 1096. Hamo, therefore, did not accompany the other crusaders in the autumn, but he may very well have overtaken them in Italy the following spring.
25. HervÉ, son of Dodeman. He is named among those who advanced with Robert after the capture of Nicaea. Baldric of Dol, in H. C. Oc., iv, p. 33; Ordericus, iii, p. 507; cf. n. 6, ibid., where Le PrÉvost remarks that ‘Breton chronicles’ name HervÉ, son of Guyomark, count of LÉon, in place of HervÉ, son of Dodeman.
26. Hugh II, count of Saint-Pol. He set out from Normandy with Robert in 1096. Ordericus, iii, p. 484. He was present at the siege of Nicaea, and advanced with Robert from there. Ibid., pp. 502-503, 507; Baldric of Dol, in H. C. Oc., iv, pp. 28, 33. He was present at the siege of Antioch. Albert of Aix, ibid., p. 372.
27. *Ingelbaudus: “Ego Ingelbaudus illud Sepulchrum volo petere.” In view of the proposed journey he made various grants to Saint-Vincent of Le Mans. Cartulaire de S.-Vincent, no. 101. The editors date the document “circa 1096,” but there are no chronological data. Most of the documents among which this appears are of the late eleventh century.
28. Ivo of Grandmesnil. Ordericus, iii, p. 484. Cf. supra, p. 107, n. 88.
29. Odo, bishop of Bayeux. He was present at the council of Clermont as legatus of his fellow bishops. Ordericus, iii, p. 470. He was in touch with Abbot Gerento of Saint-BÉnigne of Dijon, the Pope’s special agent, who was promoting the Crusade in Normandy during the summer of 1096. Haskins, pp. 75-76. But it seems probable that he undertook the Crusade rather to escape the wrath of William Rufus than from any religious zeal. Ordericus, iv, pp. 16-17. He died at Palermo, in February 1097 according to Ordericus Vitalis (ibid.), though his obit was celebrated in Bayeux cathedral on Epiphany (6 January). Ulysse Chevalier, Ordinaire et coutumier de l’Église cathÉdrale de Bayeux (Paris, 1902), p. 410. He was buried by his fellow bishop, Gilbert of Évreux, in the cathedral church of St. Mary at Palermo, and Count Roger reared a splendid monument over his grave. Ordericus, iv, pp. 17-18; iii, p. 266; cf. Guibert of Nogent, in H. C. Oc., iv, p. 233. Odo’s epitaph is published, from a late seventeenth century MS., by V. Bourrienne, in Revue catholique de Normandie, x, p. 276.
30. *Pain de Mondoubleau. See the quotation from Cartulaire de S.-Vincent, no. 666, in no. 22 supra. The editors accept this as convincing evidence that Pain de Mondoubleau went on the First Crusade, but in the absence of any definite date there is no proof. And indeed it seems hardly likely that we have to do here with the First Crusade, since in 1098, according to Ordericus Vitalis—who, however, is a very untrustworthy guide in matters of chronology—Pain was in Maine and handed over the castle of Ballon to William Rufus. Ordericus, iv, p. 47; cf. Latouche, Maine, p. 47; Auguste de TrÉmault, “Recherches sur les premiers seigneurs de Mondoubleau,” in Bulletin de la SociÉtÉ archÉologique du VendÔmois, xxv (1886), pp. 301-302. The latter mentions no evidence of Pain’s having gone on any crusade.
31. Pain Peverel. The distinguished Norman knight who acted as Robert’s standard-bearer on the Crusade, and who upon his return was granted a barony in England by Henry I, and became the patron of Barnwell priory. He is described as “egregio militi, armis insigni, milicia pollenti, viribus potenti, et super omnes regni proceres bellico usu laudabili.” He endowed the church of Barnwell with notable relics which he brought back from the Holy Land: “reliquias verissimas super aurum et topazion preciosas, quas in expedicione Antiochena adquisierat cum Roberto Curthose, dum signiferi vicem gereret, necnon quas a patriarcha et rege et magnatibus illius terre impetraverat.” Liber Memorandorum Ecclesie de Barnewelle, ed. J. W. Clark (Cambridge, 1907), pp. 54, 55, 41, 46. According to the editor this anonymous work was written in its present form in 1295-96; the author had access to documents, and probably based his narrative on the work of an earlier writer (introduction, pp. ix-x, xiv). The part dealing with our period contains notable chronological inaccuracies, but for the fundamental facts of the life of Pain Peverel it may probably be relied upon.
32. Philip of BellÊme, called the Clerk, fifth son of Roger of Montgomery. He set out with Robert from Normandy in 1096, and died at Antioch. Ordericus, iii, pp. 483, 426.
33. *Rainerius de Pomera. “Ista quae narravimus [i.e., the details of a miracle wrought by St. Nicholas of Bari] a quodam bono et fideli homine, nomine Rainerio, de villa quae dicitur Pomera, didicimus, qui haec vidit et audivit et iis omnibus praesens affuit, dum rediret de itinere Ierusalem.” Miracula S. Nicolai conscripta a Monacho Beccensi, in Catalogus Codicum Hagiographicorum Latinorum in Bibliotheca Nationali Parisiensi, ed. the Bollandists (Brussels, 1889-93), ii, p. 427.
34. Ralph de Gael. Baldric of Dol, in H. C. Oc., iv, pp. 28, 38; Ordericus, iii, pp. 484, 503, 507; Interpolations de Robert de Torigny, in William of JumiÈges, p. 287. Emma, his wife, and Alan, his son, went with him. Cf. nos. 14 and 3 supra.
35. Richard, son of Fulk, of Aunou-le-Faucon: “quidam miles, genere Normannicus, vocabulo Ricardus, filius Fulconis senioris de Alnou.” After the capture of Jerusalem he was saved from shipwreck off the Syrian coast through the miraculous interposition of St. Nicholas of Bari; and upon his return to Normandy he became a monk of Bec. Miracula S. Nicolai conscripta a Monacho Beccensi, in Catalogus Codicum Hagiographicorum Latinorum in Bibliotheca Nationali Parisiensi, ed. the Bollandists, ii, p. 429. On Fulk of Aunou, see Ordericus, ii, p. 75.
36. Riou de LohÉac. He died while on the Crusade, but sent back to LohÉac a casket of precious relics, among them a portion of the true Cross and a fragment of the Sepulchre: “Notum sit … quod Waulterius, Iudicaelis filius de Lohoac, quidam miles nobilissimus et illius castri princeps et dominus… Sancto Salvatori suisque monachis quoddam venerandum et honorabile sanctuarium, quod frater suus, videlicet Riocus, dum iret Hierosolyman, adquisierat, et post mortem suam, nam in itinere ipso obiit, per manum Simonis de Ludron sibi transmiserat, scilicet quandam particulam Dominice; Crucis et de Sepulchro Domini et de ceteris Domini sanctuariis, cum maximis donariis que subter scribentur, honorifice dedit et in perpetuum habere concessit.” These relics were placed in the church of Saint-Sauveur at LohÉac in the presence of a great concourse of clergy and people, among them being the famous Robert of Arbrissel, “quidam sanctissimus homo.” The document was attested, among others, by Walter and William, Riou’s brothers, and by Geoffrey his son, Gonnor his wife, and Simon de Ludron. “Hoc factum est in castello de Lohoac, iuxta ipsam aecclesiam monachorum, .iii. kal. Iul., in natali apostolorum Petri et Pauli, anno ab incarnatione Domini millesimo centesimo .i., luna .xxix., epacte .xviii., Alano comite existente, Iudicahele episcopatum Sancti Maclovii obtinente, et hoc donum cum suo archidiacono Rivallono annuente, data .vi. non. Iulii.” Cartulaire de l’abbaye de Redon, ed. AurÉlien de Courson (Paris, 1863: Documents inÉdits), nos. 366, 367. Baldric of Dol names him among those who advanced with Robert from Nicaea. H. C. Oc., iv, p. 33.
37. Robert of Jerusalem, count of Flanders. One of the well known leaders, who was closely associated with Robert Curthose during most of the Crusade and who returned with him at least as far as southern Italy. See Chapter IV, passim.
38. *Robert the Vicar (vicarius). Before he went to Jerusalem (priusquam Ierusalem pergeret) he made donations to Saint-Vincent of Le Mans—his wife, son, and brothers consenting—and received from Abbot Ranulf and the monks four livres manceaux. Cartulaire de S.-Vincent, no. 522. The document is undated, but the mention of Abbot Ranulf places it between 1080 and 1106. The editors date it “circa 1096.”
39. Roger of Barneville. G. F., p. 185; Ordericus, iii, p. 503. He was captured and beheaded by the Turks at Antioch early in June 1098; and was buried amid great sorrow by his fellow crusaders in the church of St. Peter. Kreuzzugsbriefe, p. 159; Raymond of Aguilers, in H. C. Oc., iii, p. 252; Ordericus, iii, pp. 549, 538; Robert the Monk, in H. C. Oc., iii, pp. 808-809; Albert of Aix, ibid., iv, pp. 407-408.
40. Rotrou of Mortagne II, son of Geoffrey II, count of Perche. His father died during his absence, having made provision for Rotrou to succeed him in the countship upon his return from the Crusade. Ordericus, iii, p. 483; v, p. 1.
41. Simon de Ludron. It was he who brought back the relics which had been obtained by Riou de LohÉac while on the Crusade. See the extract from the Redon cartulary quoted in no. 36 supra.
42. Stephen, count of Aumale. He was one of the Norman rebels who had previously sided with William Rufus against Robert Curthose. Ordericus, iii, p. 475. But he was on friendly terms with the duke by 14 July 1096—doubtless as a result of the pacification which had been brought about by the Pope—since Robert attested a charter by Stephen on that date. Gallia Christiana, xi, instr., col. 20; cf. Haskins, p. 67, no. 5. Stephen also attested a charter by the duke in 1096. Archives de la Seine-InfÉrieure, G 4069 (Inventaire sommaire, iii, p. 255). Albert of Aix records his presence at Nicaea; and Ralph of Caen names him among those who at Antioch were obligated to Robert Curthose by gifts or homage. H. C. Oc., iv, p. 316; iii, p. 642.
43. Stephen, count of Blois and Chartres. One of the well known leaders of the Crusade. He was closely associated with Robert Curthose at least as far as Nicaea. He became faint-hearted and turned back home after the expedition had reached Antioch. See Chapter IV, passim.
44. *Thurstin, son of Turgis, prÉvot of Luc-sur-Mer. In 1096 he pledged his allod (alodium) of forty acres at Luc for four marks and a mount (equitatura): “si ipse Turstinus aut uxor eius vel filius post vi annos rediret, redderet Sancto Stephano ad finem vi annorum iiii?? argenti marcas.” Probably the Crusade was in contemplation, though it is not specifically mentioned. R. GÉnestal, RÔle des monastÈres comme Établissements de crÉdit (Paris, 1901), p. 215; cf. pp. 29-30.
45. Walter of Saint-Valery. Ordericus, iii, pp. 483, 507; Baldric of Dol, in H. C. Oc., iv, p. 33.
46. Wigo de Marra, a crusader from Perche. “Rediens a Ierosolimitano itinere, tempore profectionis communis Aquilonensium et Occidentalium,” he passed through Tours; and while he rested there with the monks of Saint-Julien, he gave them his church at Bellou-sur-HuÎne, a gift which he afterwards confirmed upon reaching home. Chartes de S.-Julien de Tours, no. 51. The document is dated 1099, “regnante Willelmo rege Anglorum et duce Normannorum,” and is of special interest as indicating the early date at which some of the crusaders got back to western Europe.
47. *William de Braitel (en Lombron), son of Geoffrey the vicomte. With the consent of his brothers he made a donation to Saint-Vincent of Le Mans in 1096, “eo videlicet anno quo papa Urbanus occidentales partes presentia sua illustravit.” Cartulaire de S.-Vincent, no. 738. The similarity of dating between this charter and no. 317 of the same cartulary (cf. no. 23 supra), as well as the fact that many of the witnesses are identical in both, makes it seem not improbable that they were drawn up on the same occasion. If William actually went on the First Crusade, his return appears to have been delayed until 1116. In that year a precious relic which he brought back from Jerusalem for Adam, a Manceau who had become a canon of the church of the Holy Sepulchre, was presented to the cathedral church of Le Mans. Actus Pontificum, p. 407. Cf. Samuel Menjot d’Elbenne, Les sires de Braitel au Maine du XI? au XIII? siÈcle (Mamers, 1876), p. 38.
48. William, son of Ranulf de Briquessart, vicomte of Bayeux. He is named among those who advanced with Robert from Nicaea. Baldric of Dol, in H. C. Oc., iv, p. 33; Ordericus, iii, p. 507.
49. *William de ColombiÈres. On 7 June 1103 Henry de ColombiÈres granted to Saint-Martin of Troarn “all that his father William had given and granted before he went on crusade (Ierosolimam pergeret).” Round, C. D. F., no. 471.
50. William de FerriÈres. He is named among those who advanced with Robert from Nicaea. Baldric of Dol, in H. C. Oc., iv, p. 33; Ordericus, iii, p. 507.
51. William de Percy, benefactor of Whitby abbey. He died while on the Crusade. “Denique nobilissimus Willielmus de Perci Ierosolimam petens, apud locum qui vocatur Mons Gaudii, qui est in provincia Ierosolimitana, migravit ad Dominum, ibique honorifice sepultus est.” Cartularium Abbathiae de Whiteby, ed. J. C. Atkinson (Durham, 1879-81), i, p. 2. The quotation is from the “Memorial of Benefactions,” which, according to the editor, was written in the second half of the twelfth century, certainly before 1180. It is probably only a legend that William’s heart was brought back and buried at Whitby abbey. His son had evidently succeeded him by 6 January 1100. Davis, Regesta, no. 427.
52. William du Vast. On 9 September 1096, “vadens in Ierusalem,” he pledged his land to the abbey of FÉcamp for a loan of three marks until his return. LÉopold Delisle, LittÉrature latine et histoire du moyen Âge (Paris, 1890), pp. 28-29.
APPENDIX E
LAODICEA AND THE FIRST CRUSADE
Laodicea, as a commodious port on the Syrian coast directly opposite the fertile island of Cyprus, was a maritime base of the utmost importance to the crusaders, and it has a special interest for the life of Robert Curthose. Its history during the period of the First Crusade is obscure, and it may be admitted at the outset that it will not be possible to elucidate it entirely from such meagre and contradictory materials as have survived. Nevertheless, the problems are by no means hopeless; and the sources, such as they are, are worthy of a more careful and critical examination than they have yet received.[1]
From the oriental sources it seems reasonably certain that during the period immediately preceding the arrival of the crusaders in Syria Laodicea was in the hands of the Turks. Previous to 1086 it had belonged to the Munkidhites of Shaizar;[2] but it passed from their hands into the possession of Malik-Shah when in that year he established himself at Aleppo.[3] Malik-Shah granted it to Kasim ed-daula Aksonkor, who held it until his death in 1094.[4] There is no evidence that it passed out of Turkish control between this date and the arrival of the crusaders and their associates from the West in 1097; and, in view of the precarious situation of the Eastern Empire and the preoccupation of the Greek Emperor with other problems during this period, there seems to be no ground for such a supposition. According to Kemal ed-Din—who wrote towards the middle of the thirteenth century, and whose statement would perhaps deserve little consideration were it not so specific—a fleet of twenty-two ships came from Cyprus on the 8th of the month of Ramadan in the year 490 of the Hegira (19 August 1097), entered the port of Laodicea, pillaged the town, and carried off all the merchandise.[5]
The western sources dealing with Laodicea in 1097-98 are numerous; but at some points they are contradictory, and at best they yield but scanty information. It will be well to analyze them separately with some care:—
(1) The letter of Anselm de Ribemont to Archbishop Manasses of Rheims, written from Antioch near the end of November 1097, states definitely that Laodicea had been taken—evidently by some one acting in the interest of the crusaders, and pretty clearly before the arrival of the land forces at Antioch on 21 October 1097.[6]
This statement is confirmed by the anonymous Florinensis Brevis Narratio Belli Sacri,[7] as it is also by the account of Raymond of Aguilers.
(2) Raymond of Aguilers, who, because of his actual presence in Syria and his close association with the count of Toulouse, is by all odds the best and most reliable chronicler dealing with the events now under consideration, seems to have received but little attention from modern scholars in this connection. According to his account, which is quite full, English mariners, who were fired with enthusiasm for the Crusade, sailed via Gibraltar to the eastern Mediterranean, and with much labor obtained possession of the port of Antioch (evidently Port St. Simeon is meant) and of Laodicea before the arrival of the land forces. And during the siege of Antioch, together with the Genoese, they rendered important services to the crusaders by means of their fleet, keeping open commercial intercourse with Cyprus and other islands, and in particular protecting the ships of the Greeks from attack by the Saracens. Finally, when the crusaders were about to advance from Syria upon Jerusalem, the English, finding that their ships had been reduced by wear and tear from thirty to nine or ten, abandoned them or burned them, and joined the land forces on the southward march.[8]
Now, of the actual presence of English mariners on the Syrian coast acting in coÖperation with the crusaders, there can be no doubt. Apart from the foregoing narrative, the fact is proved beyond question (a) by the well known letter of the clergy and people of Lucca in which they state that their citizen Bruno had journeyed from Italy to Antioch “with English ships,” had taken part in the siege, and had stayed on for three weeks after the victory;[9] and (b) by the letter of Patriarch Dagobert, written from Jerusalem in the spring of 1100, which mentions the presence of English ships, apparently at Jaffa.[10] While the English ships referred to in these letters are not necessarily, or even probably, identical with those mentioned by Raymond of Aguilers, the letters are still of great importance as demonstrating the general fact of the presence and activity of English mariners at this period in these distant waters.
As will appear below, Raymond’s account receives some further confirmation from Ordericus Vitalis and from Ralph of Caen.
(3) The narrative of Ordericus differs widely from that of Raymond of Aguilers. According to him, at the time when the Christians were themselves being besieged at Antioch (6-28 June 1098), a great number of pilgrims from England and other islands of the ocean landed at Laodicea and were joyfully welcomed by the inhabitants, who accepted their protection against the Turks. The chief among these pilgrims was Edgar Atheling.[11] Taking Laodicea under his protection, Edgar afterwards handed it over to Robert Curthose, whom he loved as a brother. Thus Robert gained possession of Laodicea, and came and dwelt there for some time with Normans, English, and Bretons. Then, leaving his own garrison in the fortresses, Robert pursued his way to Jerusalem. But meanwhile Ravendinos, protospatharius of Emperor Alexius, and other Greek officers came with an expedition by sea, and laid siege to Laodicea; and the citizens, sympathizing with the Greeks, their compatriots, expelled the men from beyond the Alps and admitted imperial governors.[12]
William of Malmesbury is the only other writer who mentions a journey of Edgar Atheling to the Holy Land, and his account is very different from that of Ordericus Vitalis. He makes no mention of English mariners, and he places Edgar’s arrival in the East, in company with a certain Robert, son of Godwin, at the time of the siege of Ramleh by the Saracens (May 1102).[13]
(4) Raymond of Aguilers is authority for the statement that Robert was absent from Antioch in the third month of the siege, apparently about Christmas 1097.[14]
A fuller explanation of this absence seems to be supplied by Ralph of Caen, who says that Robert, disgusted with the tedium of the siege, withdrew to Laodicea in the hope of ruling there; for the English at that time were holding it for the Emperor, and being menaced by a wandering band, had called in Robert as their protector. Robert accordingly went to Laodicea and gave himself up to idleness and sleep. Yet he was not altogether useless, for, having come upon opulence, he shared it generously with his needy comrades at the siege. Laodicea was then the only city on the Syrian coast which was Christian and which obeyed the Emperor; and Cyprus had filled it with an abundance of wine, grain, and cattle. Robert was very loath to turn his back upon such ease and plenty; and it was only after he had been thrice summoned, and even threatened with excommunication, that he reluctantly yielded to the entreaties of his comrades and returned to the hardships of the siege.[15]
From the place which this incident occupies in Ralph’s general narrative one would judge that it belongs to the spring of 1098; but he does not date it exactly, and his chronology at best is confused and by no means trustworthy. It may be conjectured that this account is to be connected with the above mentioned briefer but more trustworthy statement of Raymond of Aguilers, thus placing the episode in the winter of 1097-98. Ralph’s chronology is not to be regarded as impossible, however, since there is no record of Robert’s presence at Antioch between 9 February and the end of May, or even the first of June, and he may very well have enjoyed more than one sojourn in Laodicea.
Further evidence of the duke’s connection with Laodicea is found in a curious statement of Guibert of Nogent that Robert had once held it, but that when the citizens were unable to bear his excessive exactions, they drove his garrison from the fortresses and threw off his domination, and out of hatred abjured the use of the money of Rouen.[16]
Finally, the twelfth-century poet Gilo remarks that English victors gave Laodicea to the Norman count.[17]
(5) The problem of Laodicea in its relation to the First Crusade is still further complicated by a statement of Anna Comnena that the Emperor wrote—she gives no date—to Raymond of Toulouse, directing him to hand over the city to Andronicus Tzintzilucas, and that Raymond obeyed.[18] Both Riant[19] and Chalandon[20] accept this statement and assign the Emperor’s letter to the first half of 1099. Their reason for so doing appears to be found in the strange narrative of Albert of Aix, which is unique among the sources.
(6) According to Albert of Aix, while Baldwin and Tancred were at Tarsus on the way to Antioch (circa September 1097) a strange fleet approached the Cilician coast. It proved to be made up of ‘Christian pirates’ from “Flanders, Antwerp, Frisia, and other parts of Gaul [sic],” who under their commander, a certain Guinemer of Boulogne, had been pursuing their calling for the past eight years. But when they learned of the Crusade, they concluded a treaty with Baldwin, and, landing, joined forces with him and advanced as far as Mamistra. But here they turned back, and, reËmbarking, sailed away to Laodicea, which they besieged and took. Then resting there in the enjoyment of ease and plenty, they sent no aid to their Christian brothers at Antioch. But presently they were attacked and cut to pieces by ‘Turcopoles’[21] and men of the Emperor, who recovered the citadel and threw Guinemer into prison, Godfrey and the other chiefs at Antioch being ignorant of the whole affair. Later Guinemer was released at the request of Godfrey.[22]
Elsewhere Albert sets forth another version of these curious events. Guinemer and his pirates, he tells us, had assembled their fleet in conjunction with the ProvenÇaux of the land of Saint-Gilles under the dominion of Count Raymond.[23] Then, sailing to Laodicea, they had taken it and driven out the Turks and Saracens whom they found there. Then, after the siege of Antioch, they had handed their prize over to Count Raymond. Still later, Guinemer, the master of the pirates, had been captured by the Greeks, and after long imprisonment had been released through the intervention of Duke Godfrey. Then, when the advance to Jerusalem had been decided upon, Raymond had restored Laodicea to the Emperor, and so kept his faith inviolably.[24]
Thus, if we could rely upon Albert of Aix, Laodicea came into the hands of the count of Toulouse after the siege of Antioch, and Alexius might naturally be expected to write him demanding its restoration to the Empire, as Riant and Chalandon suppose in accepting the above mentioned statement of Anna Comnena regarding the Emperor’s letter. It should be noted, however, that from Albert’s statement that Raymond handed over Laodicea to Alexius when the advance to Jerusalem had been decided upon,[25] it follows that the transfer could not have taken place later than 16 January 1099, the date on which Raymond moved southward from Kafartab;[26] whereas Chalandon has shown that the letter of which Anna speaks cannot be earlier than March 1099.[27] Albert of Aix and Anna Comnena, therefore, are not mutually confirmatory.
(7) Finally, note should be taken of the statement of Cafaro of Genoa—who passed the winter of 1100-01 at Laodicea, but who wrote as an old man years afterwards—that, at the time of the capture of Antioch by the crusaders, Laodicea with its fortresses was held by the Emperor, and was under the immediate command of Eumathios Philocales, duke of Cyprus.[28]
So much for an analysis of the sources. It remains to consider what conclusions may reasonably be drawn from them. And since the efforts which have been made to accept them all as of equal validity and to bring them into reconciliation have plainly not been successful, it will be well to begin with a consideration of some things which must probably be eliminated.
And first, it seems clear that the account of Ordericus Vitalis, which represents Edgar Atheling as landing at Laodicea between 6 and 28 June 1098 at the head of a great body of English pilgrims, cannot be accepted without serious modification; for we know from reliable English sources that towards the end of 1097 Edgar was engaged in Scotland, assisting his kinsman, another Edgar,[29] to obtain the Scottish throne;[30] and it would, it seems, have been impossible for him to have made the necessary preparations for a crusade and to have journeyed from Scotland to Laodicea within the limitations of time which our sources impose. It is perhaps conceivable that he should have made a hurried trip to Italy in the winter of 1097-98 with a small band of attendants, and sailing from there, have reached the Syrian coast by June. But according to Ordericus he arrived at the head of “almost 20,000 pilgrims … from England and other islands of the ocean.” Further, if the account of Ordericus were to be brought into chronological accord with the other sources which deal with Robert’s sojourn at Laodicea, the arrival of Edgar Atheling would probably have to be placed several months earlier, indeed, in the early winter of 1097-98, almost at the very time he is known to have been in Scotland. The chronology of Ordericus, therefore—which in general is notoriously unreliable—seems at this point unacceptable; and William of Malmesbury, who places Edgar’s arrival in the East in May 1102, appears to give the necessary correction. In view of the testimony of both Ordericus Vitalis and William of Malmesbury, it can hardly be doubted that Edgar Atheling actually went to the Holy Land; but that he reached Laodicea in time to have anything to do with the calling in of Robert Curthose seems highly improbable, if not impossible.
The tale of Guinemer of Boulogne and his fleet of Christian pirates, as told by Albert of Aix, must also meet with rougher handling than it has yet received, and for the following reasons: (1) The description of this fleet with its “masts of wondrous height, covered with purest gold, and refulgent in the sunlight”[31] is not such as to inspire confidence, particularly in such a writer as Albert of Aix, where one expects at any time to meet with the use of untrustworthy poetical materials. (2) As the narrative proceeds it becomes self-contradictory. At one point we are told that Guinemer was captured by the Greeks during the siege of Antioch, whereas at another he seems to have held Laodicea throughout the siege—since he turned it over to Count Raymond after the siege—; and his capture and imprisonment by the Greeks are placed still later. (3) Albert of Aix is in direct contradiction with Raymond of Aguilers, the best of all our authorities, who tells us that the English held Laodicea during the whole of the siege of Antioch and rendered important services to the crusaders; whereas, according to Albert’s account, Guinemer and his pirates held it and refused to aid the crusaders. (4) Not a scrap of evidence concerning Guinemer and his pirates has come to light in any source except Albert of Aix—unless perchance their fleet is to be identified with the ships which, according to Kemal ed-Din, came from Cyprus 19 August 1097, pillaged Laodicea, and sailed away;[32] and this seems unlikely. (5) In any case, outside the pages of Albert of Aix, evidence is lacking that such a piratical fleet held Laodicea for any considerable period; and apparently the only reason why Riant and Chalandon have accepted this fantastical tale of Guinemer and the Christian pirates is the fancied possibility of connecting it with the letter which, according to Anna Comnena, the Emperor wrote at an undetermined date to Raymond of Toulouse, directing him to hand over Laodicea to Andronicus Tzintzilucas. But Riant and Chalandon have somewhat arbitrarily assigned this letter to the first half of 1099. If Raymond was directed to hand Laodicea over, he must have possessed it. Therefore, so the argument seems to run, the Guinemer episode should be accepted as explaining how Raymond came into possession of Laodicea. But, as has already been pointed out, this explanation involves a serious chronological inconsistency. Further, the evidence is not conclusive that the letter ever existed—it rests upon the sole statement of Anna Comnena—and, if it did exist, it may with more reason, and with less violence to Anna’s chronology, be assigned to the period between September 1099 and June 1100, when Raymond is known to have been in possession of Laodicea and on terms of close understanding with the Emperor.[33]
The foregoing considerations are not, it may be conceded, sufficient to prove that there is no shadow of truth in the tale of Guinemer and the pirates; but they do constitute a strong case against the narrative as it stands, and suggest the probability that it is one of the strange pieces of fiction occasionally to be met with in the pages of Albert of Aix.
Having now somewhat cleared the ground, it is possible to set forth the probable course of events at Laodicea on the basis of the more reliable sources.
There can be little doubt that Laodicea had already been taken from the Turks when the crusaders arrived at Antioch, 21 October 1097;[34] and we may accept without question the statement of Raymond of Aguilers—which Riant and Chalandon appear to ignore without reason—that it was taken by the English, who had come by sea, and who held it during the siege of Antioch and assisted the land forces by protecting commerce and keeping communications open with Cyprus and the other islands. These English mariners were unquestionably acting in coÖperation with the Emperor,[35] who at this time, as Chalandon has shown, was supporting the crusaders in accordance with his treaty obligations.[36]
At some time during the siege of Antioch by the Christians Robert Curthose was called to Laodicea by the English—probably because of dangers on the landward side which made their situation there precarious—and he remained there for a time, in the enjoyment of ease and plenty, until he was obliged by repeated summonses and by a threat of ecclesiastical censure to return to Antioch.[37] The date of Robert’s sojourn at Laodicea cannot be determined with certainty, but it may probably be assigned to December-January 1097-98,[38] 8 February being the extreme limit for his return to the siege.[39] Yet there is no record of his presence at Antioch between 9 February and the beginning of June, or between the end of June and 11 September; and the possibility of his having paid more than one visit to Laodicea must be recognized. The accounts of Ralph of Caen and of Ordericus Vitalis, interpreted strictly, point to sojourns in the spring and in the summer of 1098; but the chronology of these authors is not trustworthy, and it is not unlikely that they have fallen into inaccuracies here, and that they really refer to Robert’s earlier sojourn at Laodicea, for which we have the indirect but more reliable evidence of Raymond of Aguilers.
The arrangements which were made at Laodicea upon Robert’s final departure before his advance to Jerusalem must remain a matter of doubt. According to Ordericus Vitalis and Guibert of Nogent he left a garrison, which was later driven out by the citizens. Guibert is curiously circumstantial. He says that the citizens, unable to bear the duke’s excessive exactions, drove his men from the citadel, threw off his domination, and abjured the use of the money of Rouen. But this incident is confirmed by none of the early writers who were in the East; and in the absence of any other evidence of Robert’s having attempted to secure for himself a private possession in Syria, we may well wonder whether Guibert and Ordericus have not blundered through a misunderstanding of the actual situation in the East and of the spirit in which Robert undertook the Crusade.
Finally, what is to be said of the statement of Cafaro of Genoa that, at the time of the capture of Antioch by the crusaders, Laodicea was under the rule of Eumathios Philocales, duke of Cyprus? It would not be surprising if Cafaro, writing long after the event, should be mistaken on a point of this kind; yet he is by no means to be ignored, and on the whole his account does not seem inconsistent with established facts. The sojourn of Robert Curthose at Laodicea was apparently a passing episode rather than a lasting occupation. But throughout the period under consideration the Syrian port was clearly in the hands of crusaders, mainly English mariners, who were acting in coÖperation with the Greeks. Under existing treaty obligations the place might fairly be regarded as a Greek possession from the moment the Turks were expelled[40]—unless there were a Bohemond or some other like-minded chief to seize and hold it in defiance of imperial rights. And the Emperor would most naturally delegate authority over Laodicea to the head of his administration in Cyprus. From the Greek standpoint, therefore, it might well be regarded as subject to Eumathios Philocales, though actually held by the Emperor’s allies, the crusaders.
Between the departure of the crusaders from northern Syria early in 1099 and their return in September after the capture of Jerusalem, Laodicea seems to have become definitely a Greek possession; but whether there was any violent expulsion of the garrison of a crusading chief, as Ordericus and Guibert suppose, or any formal transfer,[41] must remain uncertain. When the crusaders moved southward from northern Syria to Jerusalem, their influence at Laodicea must, it seems, inevitably have declined, while that of the Greeks increased; and without any formal transfer it is conceivable that the place might gradually and almost imperceptibly have passed under full Greek control.
But for this later period there are some further scattered notices in the chronicles of Albert of Aix and of Raymond of Aguilers and in the anonymous Gesta Francorum, which must now be considered, and which make it clear that at this time Laodicea was still in Christian hands and served as a most important base for the further prosecution of the Crusade.
Albert of Aix, who is the fullest and most specific, explains that the crusaders still remaining in Syria gathered in council at Antioch on 2 February 1099, and, determining upon an advance to Jerusalem, fixed 1 March as the date for a general rendezvous of all the forces at Laodicea, a city which was then under Christian dominion.[42] Pursuant to this decision, Godfrey, Robert of Flanders, and Bohemond assembled their forces at Laodicea on the appointed day. And from Laodicea Godfrey and Robert moved on southward to the siege of Jebeleh; but Bohemond, ever suspicious and anxious lest through some fraud he should lose a city which was ‘impregnable by human strength,’ returned to Antioch.[43] This very specific account of Albert of Aix is confirmed by the much briefer statements of the Gesta Francorum, which record the meeting of the leaders at Laodicea, the advance of Godfrey and the count of Flanders to the siege of Jebeleh, and the return of Bohemond to Antioch.[44] It is also clear from Raymond of Aguilers that in the spring and summer of 1099—at least until June—the port of Laodicea was open to the ships of the Greeks, Venetians, and Genoese who were engaged in provisioning the crusaders at Arka and at Jerusalem.[45]
There can be little doubt, therefore, that until June 1099, Laodicea was held in the interest of the crusaders, and that its harbor was open to the ships of Greeks and Italians without distinction. Albert of Aix nowhere explains what he means when he says that Laodicea was “under Christian dominion”; but, in the absence of valid evidence of its retention by any of the crusading chiefs, or by the fleet of any Italian city, the most reasonable hypothesis appears to be that it was held by the Greeks in the interest of the common enterprise.
We get our next information concerning Laodicea when, in September 1099, Robert Curthose, Robert of Flanders, and Raymond of Toulouse, upon their return from Jerusalem, found the place undergoing a prolonged siege at the hands of Bohemond, who was assisted in his nefarious enterprise by a fleet of Pisans and Genoese.[46] Since the early summer, when ships of Genoese, Venetians, and Greeks had all enjoyed free entry to the port, a complete change had come over the situation at Laodicea.[47] What had happened to produce this? As is well known, it was the fixed policy of the Emperor to turn the Crusade to his own advantage, and to utilize the efforts of the Franks for the recovery of the lost provinces which had formerly belonged to the Greek Empire in Asia. To this end, he had been on the whole successful in coÖperating with the crusaders. But in Bohemond of Taranto he had encountered opposition from the beginning; and, since the capture of Antioch by the crusaders, it had been the little disguised policy of this crafty and ambitious leader to hold it for himself, and to make it the capital and centre around which he hoped to build up a Norman state in Syria. It was, of course, inevitable that the Emperor should set himself to thwart such plans by every means at his disposal; and when the departure of the main body of the crusaders for Jerusalem left Bohemond with a free hand in the north, open hostilities became imminent. Undoubtedly foreseeing what was to come, Bohemond had separated from Godfrey and Robert of Flanders at Laodicea in March, and had returned to Antioch to mature his plans.[48] A few weeks later, ambassadors from the Emperor arrived in the crusaders’ camp at Arka and lodged a complaint against Bohemond.[49] But the Emperor was in no position to take vigorous measures at that time. Such a course might even have endangered his friendly relations with the other leaders. But neither was Bohemond in a position to resort to an overt act against Laodicea so long as he was powerless to meet the imperial fleet at sea. In the late summer of 1099, however, all this was changed by the arrival of a Pisan fleet under the command of Dagobert, archbishop of Pisa; for Bohemond, with true Norman adaptability and shrewdness, came to an understanding with the Pisans and secured their aid for an attack upon Laodicea.[50] And with this, the slight naval supremacy which the Greek Emperor had been vainly striving to maintain in the eastern Mediterranean came to an end.[51]
Such was the situation at Laodicea when in September 1099 Robert Curthose and the counts of Flanders and Toulouse arrived at Jebeleh on their way home from the Crusade. The siege had already been going on for some time and was making progress. The place seemed to be on the point of falling.[52] But never were the plans of Bohemond to end in more egregious failure. His unprovoked attack upon a friendly city which had rendered important services to the crusaders roused the indignation and jealousy of the returning leaders. The archbishop of Pisa suddenly discovered that he had been led into a false position by the crafty Norman, and, deserting Bohemond, he threw his powerful influence on the side of Raymond, Robert Curthose, and Robert of Flanders. The Greeks too, who, though hard pressed, were still holding out, well understood that Bohemond was their real enemy and that it behooved them to make terms quickly with the leaders who had kept faith with the Emperor. Accordingly, an agreement was promptly reached among the Pisans, the Laodiceans, and the returning leaders. An ultimatum was despatched to Bohemond demanding that he withdraw forthwith; and thus suddenly confronted with superior force, he had no choice but to yield. Wrathfully he retired under the cover of darkness; and next morning Robert Curthose and the counts of Flanders and Toulouse entered Laodicea with their forces, and were enthusiastically welcomed by the inhabitants.[53]
Count Raymond placed a strong garrison in the citadel, and raising his banner over the highest tower, took possession of the city[54]—in the Emperor’s name, it may be supposed, since by this time he clearly had an understanding with Alexius.[55] A few days later he met Bohemond outside the city and concluded peace.[56]
After a fortnight’s sojourn at Laodicea the two Roberts and a large number of humbler crusaders took ship and proceeded on their homeward way. But Raymond, still suspicious of the prince of Antioch, remained to keep a close guard upon Laodicea and Tortosa until the following summer, when he went to Constantinople and entered the Emperor’s service.[57]
APPENDIX F
THE BATTLE OF TINCHEBRAY[1]
The tactics of the battle of Tinchebray have been the subject of much discussion among recent writers, including the specialists in military history. There is general agreement as to the strategical stroke by which the victory was won, viz., a surprise attack upon the flank of the ducal forces by a band of mounted knights from Maine and Brittany. But as to the disposition of the troops in the two main armies, widely different views are held upon two points.
(1) Oman thinks that the battle formation on each side was an extended line made up of a right, centre, and left.[2] Ramsay, on the other hand, holds that the opposing forces were “marshalled in column, in successive divisions”;[3] and this view is accepted by Drummond,[4] by DelbrÜck,[5] and by Davis,[6] the two latter conjecturing a formation in Échelon. Ramsay’s view is pretty clearly supported by the sources. Ordericus Vitalis (iv, p. 229) designates a first, second, and third acies, or division, on the side of the king, and a first and last (extrema) acies on the side of the duke; and, according to his account, only the first acies, i.e., the leading elements, of the two opposing forces engaged in the fighting. The contemporary letter of a priest of FÉcamp, which is discussed below, is also specific with regard to the royal forces, describing a first and a second acies.[7]
(2) The larger question in debate between the specialists, however, turns upon the relative importance of cavalry and infantry in the battle of Tinchebray. Oman, relying upon a very specific passage in Henry of Huntingdon (p. 235), and placing a strained interpretation upon Ordericus Vitalis (iv, p. 229), holds that the battle was almost wholly an affair of infantry, and therefore almost without precedent in the tactics of the period.[8] For Ramsay, on the other hand, it was mainly an engagement of cavalry, the foot soldiers playing but a minor part.[9] Drummond has gone even further and taken great pains to demonstrate that it was a “ganze normale Schlacht des XII. Jahrhunderts,” i.e., a battle between mounted knights, the foot soldiery that happened to be present being held entirely in reserve;[10] and Drummond’s conclusions have been accepted without question by DelbrÜck.[11]
It is surprising that in none of the discussion above noted has any account been taken of the most important extant source for the tactics of Tinchebray, viz., a letter from a priest of FÉcamp to a priest of SÉez written a very few days after the engagement, and describing with exactness certain tactical features of the battle. If not actually by an eyewitness, the letter is still by one who was in touch with the king and who was well informed as to the disposition of the royal forces. It is, therefore, entitled to rank as an authority above any of the accounts in the chronicles. It was first discovered by Paul Meyer in an Oxford manuscript,[12] and published in 1872 by LÉopold Delisle as a note in his great edition of the chronicle of Robert of Torigny (i, p. 129). But, strangely overlooked by all the military historians, it remained unused, and was rediscovered by H. W. C. Davis and published with extensive comment in 1909 in the English Historical Review (xxiv, pp. 728-732) as a “new source.” As afterwards turned out, Davis’s transcription of the letter had been exceedingly faulty—rendering, indeed, a part of the text which was fundamental for tactics quite unintelligible—and in a later number of the Review (xxv, p. 296) it was again published in a corrected text. By a comparison with the original edition of Delisle[13] it appears that, by an almost unbelievable coincidence, the same omission of an entire line of the manuscript was made there as in the edition of Davis. Yet all transcripts have been made from a single manuscript, viz., Jesus College, Oxford, no. 51, fol. 104. We have, then, at last, a correct edition of this important source in the English Historical Review, xxv, p. 296.[14]
Davis, in commenting on the tactics of the battle in the light of this letter, but from his own faulty transcript, maintains that neither of the extreme views is correct, and suggests “a third interpretation of the evidence, midway between the two existing theories.”[15] He holds that infantry played an important part in the action, but still assigns much prominence to the cavalry. Apropos of the corrected text of the priest’s letter, however, he remarks: “Taking the omitted words into consideration, it is clear that the foot soldiers played a larger part in the battle than I allowed in my article. The second of Henry’s divisions, like the first, was composite, containing both infantry and cavalry.”[16] This, indeed, is the correct view. Our conception of the battle of Tinchebray must be based upon the sources, and not upon a preconceived theory of the all-importance of the mounted knight in twelfth-century warfare. Drummond and DelbrÜck have quite unjustifiably ignored Henry of Huntingdon in favor of Ordericus Vitalis. Whatever the theorists may hold, foot soldiers did play an unusually large part in the battle of Tinchebray. In view of the explicit statement of Henry of Huntingdon (p. 235) and of the priest of FÉcamp[17] it cannot be denied that, on the king’s side at least, some knights were dismounted and fought on foot, in order that they might stand more firmly (ut constantius pugnarent). On the other hand, Oman, while perfectly justified in pointing out the unusual prominence given to foot soldiers, certainly exaggerates in representing the battle as almost wholly an affair of infantry. The large part played by cavalry is clear both from the explicit statement of the priest of FÉcamp and from the account of Ordericus Vitalis. The battle of Tinchebray may, therefore, still claim to stand as an important precedent in the development of mediaeval tactics because of the unusual combination of infantry and cavalry in the fighting line.
APPENDIX G
THE ROBERT MEDALLION IN SUGER’S STAINED GLASS WINDOW AT SAINT-DENIS
A recent writer has described Suger’s reconstruction of the abbey church of Saint-Denis as “le fait capital de l’histoire artistique du XII? siÈcle”;[1] and certainly among the most remarkable features of that great achievement were the stained glass windows, which were the abbot’s pride, and which he caused to be wrought “by the skilful hands of many masters from divers nations.”[2] The oldest painted windows of known date which survived from the Middle Ages,[3] most of them were destroyed during the French Revolution; and there would be no occasion to mention them in connection with the life of Robert Curthose, were it not that a series of ten medallions from one window, representing scenes from the First Crusade, has been preserved for us by the venerable Benedictine, Bernard de Montfaucon, in copperplate engravings of the early eighteenth century.[4] The eighth scene in the series has given rise to much discussion. It portrays a Christian knight in the act of unhorsing a pagan warrior with a mighty thrust of his lance, and bears the inscription: R DVX NORMANNORVM PARTVM PROSTERNIT.[5] Clearly we have here some spectacular victory of Robert Curthose over a Saracen; and it is the oldest graphic representation of the duke now extant. The only problem is to identify it either with a historic or with a legendary exploit of Robert on the Crusade. Ferdinand de MÉly, assuming that it had nothing to do with veritable history, has supposed that it represented Robert’s legendary combat with the emir ‘Red Lion’ during the great battle of the Franks against Kerboga, as related in the Chanson d’Antioche;[6] and at Riant’s suggestion he has gone further and proposed that it may offer a terminus ad quem for determining the date of composition of that poem.[7] Gaston Paris has very properly rejected both these hypotheses. But he still holds that the Robert medallion can only be explained by reference to the Chanson d’Antioche, and he identifies the scene portrayed with Robert’s legendary victory over Kerboga himself rather than with that over Red Lion.[8] On the other hand, Hagenmeyer, who is better qualified to speak upon such matters, sees not legend at all but sober history in the scene in question. Indeed, upon comparison of the whole series of Montfaucon’s engravings with the original narratives of the First Crusade, he finds all the scenes portrayed to be in remarkably close agreement with historic facts. “L’artiste qui a fait ces peintures,” he says, “a ÉtÉ, sans aucun doute, trÈs au courant des ÉvÉnements marquants de la premiÈre croisade… A proprement parler, aucune de ces peintures ne contient d’Épisode lÉgendaire.” And the scene in the Robert medallion he considers to be no more than a pictorial rendering of a text from the Gesta Francorum describing the battle of Ascalon: “Comes autem de Nortmannia cernens ammiravisi stantarum … ruit vehementer super illum, eumque vulneravit usque ad mortem.”[9]
Although MÉly in quoting Hagenmeyer’s opinion does not accept it,[10] there can be little doubt of its correctness. The scenes from the Crusade in Suger’s window do not, it is true, agree in every minute detail with the primary literary sources, but the deviations are certainly not greater than should be expected from a mediaeval painter striving to produce an artistic result within the limitations of his craft. The arrangement and numbering of Montfaucon’s engravings leave some doubt as to the original sequence of the medallions, but so far as it is possible to determine, the outstanding events of the Crusade from the siege of Nicaea to the battle of Ascalon appear to have been portrayed in chronological order. About the first six scenes, as arranged by Montfaucon, there can be practically no doubt. And the great battle against Kerboga is set in its proper place between the capture of Antioch and the storming of Jerusalem; and there is no indication that Robert played a special part in it, any more than there is in the strictly historical literary sources.
The last four medallions as given by Montfaucon present peculiar difficulties; and it will be well to describe them briefly, preserving his numbering.
No. 7. The flight of defeated horsemen through a gate into a walled city. Inscription: ARABES VICTI IN ASCALON FVGIVNT.
No. 8. The Robert medallion which has been described above.
No. 9. A single combat between a Christian and a pagan horseman, each supported by a band of warriors who fill the background. Inscription: DVELLVM PARTI EX ROTBERTI FLANDRENSIS COMITIS.
No. 10. A general combat between Christian and pagan warriors fighting on horseback. Inscription: BELLVM AMITE ASCALONIA IV; and an unfilled space at the end seems to indicate that it is incomplete. Evidently this inscription has become corrupt in transmission, and as it stands it is not wholly intelligible. It seems clear enough, however, that we have here a representation of the great battle of the Franks against the Egyptian emir Malik el-Afdhal near Ascalon.
Now if the four medallions in question be taken in the order in which they have just been described, it is difficult, if not impossible, to reconcile them with the literary sources as a representation of actual events in chronological order. But it is very doubtful whether Montfaucon has placed them in their proper sequence. We have no way of checking him as to the arrangement of nos. 8 and 9; but a glance at his engravings reveals the fact that nos. 7 and 10 are not perfectly circular like the rest, but are considerably cut away, the former in the upper right hand sector and the latter in the upper left hand sector.[11] Clearly they were placed side by side at the top of the window in the restricted space beneath the pointed arch, no. 10 being on the left and no. 7 on the right. Now the general sequence of the medallions in the window appears to have been from the bottom to the top; and in that case nos. 10 and 7 must have been the last two of the series. If this arrangement be accepted the interpretation of the last four medallions does not seem to offer greater difficulties than that of the first six. All four have to do with events centring around Ascalon and the great contest of the Franks with the Egyptian emir. Nos. 8 and 9 portray the individual feats of arms of Robert Curthose and Robert of Flanders as set forth in the literary sources.[12] No. 10 (with the corrupt inscription) probably represents the general engagement in which the exploits of the two Roberts were such notable features. And no. 7, properly belonging at the end, represents the flight of the vanquished pagans through the gate within the protecting walls of Ascalon. It is true that our best literary sources in describing the pursuit which followed the battle make no mention of this particular feature. But we know that the inhabitants of Ascalon closed their gates and successfully bid defiance to the crusaders;[13] and it certainly does not seem improbable that some of the fugitive Saracens should have escaped thither. At any rate, the artist might very well have assumed that they so escaped.
[1] Émile MÂle, in AndrÉ Michel, Histoire de l’art (Paris, 1905-), i, p. 786. On the rebuilding of the church see Otto Cartellieri, Abt Suger von Saint-Denis, 1081-1151 (Berlin, 1898), p. 105, and the references there given; Michel FÉlibien, Histoire de l’abbaye royale de Saint-Denis en France (Paris, 1706), pp. 170-176; Paul Vitry and Gaston BriÈre, L’Église abbatiale de Saint-Denis et ses tombeaux (Paris, 1908), pp. 9-10; and above all Anthyme Saint-Paul, “Suger, l’Église de Saint-Denis, et Saint Bernard,” in Bulletin archÉologique du ComitÉ des Travaux historiques et scientifiques, 1890, pp. 258-275.
