Download PDF | Marcus Bull - Eyewitness and Crusade Narrative_ Perception and Narration in Accounts of the Second, Third and Fourth Crusades-Boydell & Brewer Inc. (2018).

Download PDF | (Crusading in Context, Vol. 1) Marcus Bull - Eyewitness and Crusade Narrative_ Perception and Narration in Accounts of the Second, Third and Fourth Crusades-Boydell & Brewer Inc. (2018).

408 Pages 

Acknowledgements

Over the course of the research for this book I have incurred a great many debts of gratitude. I have benefitted from exchanges with numerous colleagues. I shall not list names for fear of leaving some people out, but I think it is right to mention William Purkis, with whom it is always a great pleasure to discuss ideas.

I am appreciative of the ways in which the Department of History and the Medieval and Early Modern Studies programme at the University of North Carolina stimulate a creative atmosphere in which the synergies between teaching and research are fostered by opportunities to teach to one’s evolving interests and to go in new research directions. I am therefore very grateful to my medievalist and other colleagues in Chapel Hill for contributing to a positive and energizing academic milieu. I would particularly like to thank the medieval postgraduate students in the Department of History. It is my very good fortune to have some small role in the academic formation of an exceptional group of talented and engaging young scholars. It is without question the best part of the job. They will recognize in this book many of the things that I have droned on about in recent years; but I hope that they will also get a sense of how much I learn from them.

I was fortunate to be awarded a fellowship at the National Humanities Center in the academic year 2014-15, during which time I did much of the reading that informs this book. I am very grateful to then Vice-President for Scholarly Programs Cassie Mansfield and her courteous staff for the wonderful scholarly ambience in which I found myself immersed at the Center. I am further grateful to Cassie and the Center for permitting me to retain my office there in the autumn of 2015, when I was in receipt of a Senior Faculty Research and Scholarly Leave from UNC.

I am extremely grateful to Caroline Palmer and Boydell and Brewer. It is always a great pleasure to work with such a professional and helpful team; and I am especially grateful for Caroline’s gracious patience and keen insight. To inaugurate a new series for the press is a particular honour.

To my wife Tania and my daughter Sasha, my hoofdklasse team, no expression of gratitude is sufficient. In generously creating the conditions in which it is easy to work, they are effectively the co-authors of this book.

Finally, an acknowledgement of two scholars who made a big difference for me. I had always hoped that one day I might have the chance to register my appreciation of Rosalind Hill (1908-97) in some way; and although she is best known for her edition (with Sir Roger Mynors) and translation of the Gesta Francorum, a First Crusade text, and this book begins with the Second, this seems an opportune moment. In her late seventies and well into her eighties, Rosalind would unfailingly attend the ‘Crusades and the Latin East’ seminars at the Institute of Historical Research; and what struck me at the time and remains an abiding memory is her great interest in and kindness towards the assembled postgraduate students, me included. British academic life is much the poorer for the passing of the generation whose values Professor Hill personified.

Jonathan Riley-Smith’s death in 2016 was a loss keenly felt by many. It is a characteristic of unusually kind and generous people that they leave those around them in their debt, even though they would not for one minute dream of calling in that debt to their own advantage. Jonathan was such a person. This book is dedicated to his memory.

Introduction: Medieval and Modern Approaches to Eyewitnessing and Narratology as an Analytical Tool

There is a category of historical evidence that historians are wont to characterize, and indeed to essentialize, as ‘eyewitness’. This is not just a technical term of art. The idea of being an eyewitness to something is deeply embedded in a wide range of cultural situations. We do not need to have a formal grounding in common law, for instance, to appreciate that the reliability of a witness who claims to have seen an event is normally greater than that of someone who is merely passing on hearsay. As is discussed in more detail below, sight and light are the basis of countless metaphors for understanding, realization and many other cognitive operations. 

So when we say that a historical source is ‘eyewitness’, we are making very large claims about it, even though the underlying assumptions about what we are saying have been surprisingly little studied relative to the importance of this category to the ways in which historians evaluate and deploy their evidence. All historical evidence that is the result or residue of human agency — as opposed to, say, tree-ring data and some types of archaeological deposit — has some experiential basis. But the particular appeal of eyewitness evidence is that, all other things being equal, it seems to close the gap between record and experience more than any other trace of the human past. It is through eyewitness evidence that we seem to get closest to validating the powerful instinct that people in the past must have led lives grounded in moment-by-moment sensory experience that was every bit as real to them as our lived experience is to us. 

In this way the category of eyewitness evidence seems to plug historical inquiry into basic human capacities that transcend cultural differences across time and space — or at least do so enough to grant us a meaningful point of entry into societies which in many respects can strike us as very dissimilar from our own. Eyewitnessing appears to be a powerful common denominator that permits us to understand and empathize with people in the past.!

We need, therefore, to unpack the assumptions that we build into the category of evidence that we label ‘eyewitness’. In order to do so, and to examine the mechanisms whereby the experiences of people who took part in a historical event are transposed into narratives about it, this study focuses on a selection of texts written in connection with the Second, Third and Fourth Crusades.” Individually and collectively, the chosen texts raise questions about the ways in which eyewitnessing informs substantial and detailed narratives that tell complex stories. The overall argument of this book is that, whereas we tend to appraise the eyewitness quality of a narrative source primarily in terms of the relationship of the source’s author to the events that the source narrates, there is more to be gained from looking inwards into the workings of the eyewitness narrative as text. 

This is not to argue that the history of events, histoire événementielle, is trivial or unimportant, or that the reality of the past simply collapses into its textual representations. But it is to suggest that event-centred historical reconstructions that are substantially grounded in narratives such as those that feature in this study would do well to ‘go the long way round’, methodologically speaking, when validating their truth claims with reference to a given source’s supposed eyewitness status. A fuller appreciation of the textual means by which this eyewitness, or ‘autoptic’, quality impresses itself upon narrative sources can deepen our understanding of both the past as lived experience and the means by which we are granted access to that experience.

A helpful point of entry into thinking about eyewitnessing is a well-known post-medieval narrative with a medieval setting. Kurosawa Akira’s Rashdmon (1950) is, alongside the same director’s Seven Samurai (Shichinin no Samurai, 1954), among the best-known Japanese period films in world cinema. Its setting and central plot device were inspired by two short stories by the popular writer Akutagawa Rytinosuke (1892-1927). It narrates from several viewpoints mutually exclusive versions of the story of the murder of a minor nobleman and the rape of his wife by a bandit in a remote forest clearing. Although the film’s pared-down mise-en-scéne does not mandate a precisely fixed period setting, the fact that some of the action takes place within the ruins of the Rashomon, or Rajomon, the gate that was the main southern entrance into Kyoto, Heian Japan’s capital city, points to a time shortly after the collapse of the Heian political system and of Kydto’s importance in the mid 1180s.4 

The film’s many ambiguities and subtleties are activated by placing what did, or did not, take place in the clearing within not one but two narrative frames. These are set off not only from the action in the forest but also from one another by means of starkly contrasting diegeses, or scenic settings: first, a courtyard-type space, characterized by clean geometric lines and bright sunlight, in which various of the characters, including the murder victim himself speaking through a medium, address an unseen and unspeaking judge; and second, the gloomy setting of the ruinous city gate that gives the film its name, under which a woodcutter who claims to have stumbled upon the scene of the crime and a priest who has also given evidence in the courtyard earlier that day are grilled by an inquisitive and aggressively cynical commoner — effectively the audience’s surrogate in the search for answers about what really happened — as they shelter from pounding rain.

Rashomon attracted a great deal of critical attention when it was released. It was awarded the Golden Lion, the top prize, at the 1951 Venice Film Festival, and it thus played a significant part in opening up Japanese cinema in particular and Japanese culture in general to global audiences after the isolation of the post-war years. Rashomon inaugurated a brief but highly creative period during which Japanese filmmakers, principally Kurosawa himself and Mizoguchi Kenji, produced what is probably the richest corpus of filmic explorations of the premodern world in the history of cinema.° Interpretations of the film are many and varied. Viewed as a product of its particular time and place, it can be read as a parable about Japan’s militaristic past, uncertain present and hoped-for future — this last element represented by the discordantly upbeat and sometimes criticized coda to the main action in which the woodcutter undertakes to care for an abandoned baby whom he, the priest and the commoner have chanced upon in the Rashomon’s ruins. 

The film can also be seen as a commentary on the Allied Occupation of Japan, or SCAP, still in place in 1950 and arguably represented by the unseen authority-figure in the courtyard scenes. It is noteworthy that the bandit, played by Mifune Toshiré, gestures in his wild and over-exuberant physical and vocal manner to Japanese stereotypes of Westerners, especially so when juxtaposed against the nobleman, who for the most part embodies the cold self-control of the Japanese warrior class. Rashomon can be read as a critique of contemporary constructions of masculinity, and it is also about some of the different forms that sexual violence can assume. Kurosawa himself was generally reluctant to volunteer a definitive interpretation of his film; when pressed, he tended to suggest that it made an ethical point about the human propensity for egotistical mendacity and self-deceit.’ 

The clichéd summary verdict on Rashdmon, however, has been that it concerns something called ‘the relativity of truth’; that is to say, it captured a certain post-war anxiety about the absence of fixed points of moral reference in human affairs as well as presciently anticipating the later postmodernist insistence that objective knowledge of the world and of the past is in fact impossible.

Whatever the interpretive loading that the film is made to bear, however, two aspects of its foundational meaning-making project and the manner in which it invites responses from its audience are pertinent to the present study. The first is the basic plot artifice that grabs the viewer’s interest; the presentation of multiple, irreconcilable versions of what happened violates our assumptions that eyewitnesses of an event are, or at least should be, reasonably accurate and reliable sources of information. True, we seldom insist on absolute uniformity among various witnesses to an event, and we typically tolerate minor discrepancies of detail and emphasis, but there is generally an expectation that different versions should converge on certain irreducible elements, what are sometimes termed plot cruxes or kernel events, which between them encapsulate and characterize what happened. This expectation is all the greater when an eyewitness is not simply someone placed in the role of observer but is her- or himself a protagonist in the action.

 The interactions of the three principals in the forest, the bandit, the nobleman and the woman, are in every version of events those of fully-engaged participants: this is their story, or rather stories. And while the woodcutter’s role seems on the surface to be that of a passive and disinterested observer — a point-of-view shot during his second go at recalling his experience would seem to situate him just behind the tree line gazing through the leaves into the clearing — there is some suggestion in his exchanges with the commoner in the outer framing narrative that he may have played a more active role, at the very least purloining the nobleman’s expensive dagger which was left at the scene and which may have been the actual murder weapon, and possibly even committing the murder himself.

We usually accept that a casual outside observer might miss or distort basic details, but how could people so intimately caught up in such memorable and personally consequential events seem to get it all so wrong? Indeed, the film itself plays with this very expectation in setting up a clear contrast between the principals, whose versions are full and circumstantial and focus on the all-important question of what happened in the clearing, and the minor contributions of other eyewitnesses, who can only flesh out brief and peripheral moments of little or no value as evidence. If these marginal scenes were excised from the film, their loss would scarcely affect the main plot. Thus, the constable who arrests the bandit is only in a position to recount the circumstances of the arrest. The priest, for his part, is an important figure in the outer framing narrative, complicating the antagonistic dynamic between the woodcutter and the commoner; but as a witness within the courtyard frame his contribution is self-highlighting in its triviality as he merely recalls fleetingly passing the nobleman and his wife on the forest path at some unspecified point before they encountered the bandit further down the road. If Rashomon had been made as a film in which the audience is invited to solve apuzzle as it pieces together fragmentary clues volunteered by those people, such as the constable and the priest, on the margins of the action, it would conform to more familiar genre expectations, those of the detective story or murder mystery. But it would almost certainly have been a lesser achievement. For it is in getting the eyewitness-participant principals themselves to disagree in fundamental ways about their recent, vivid, physical and life-changing experiences that Kurosawa profoundly destabilizes our normally unexamined expectations: expectations, that is to say, about how we lock on to the world around us by means of our perceptions of it, how we remember and narrate our experiences, and how in our routine social interactions we tend to repose trust in the self-narration of others whose perceptual and mnemonic capacities, whose own purchase on the world, we presume to be very similar to our own.

In addition, and following on from this, the film resolutely refuses to steer the viewer towards one preferred version of events. Each main account is framed and narrated in the same ways and presented as entire unto itself as an ethical space, obeying its own logic of cause and effect and of character motivation within the parameters of the particular storyworld that it constructs. True, the fact that the nobleman speaks through a medium might give us pause as far as his testimony is concerned, even if we choose to suspend disbelief and tell ourselves that recourse to mediums was standard late Heian judicial practice. But his version of events stands or falls by the same criteria of belief or disbelief that apply to the others as long as we accept that his stated reason for killing himself, as he does in his telling, namely his shame at being dishonoured by the bandit’s rape of his wife, is as plausible a motivation as those that inform the other accounts. There is no voiceover commentary to privilege one version over another, nor are significant differences present in the framing and sequencing of shots to nudge viewers in a particular direction. In addition, each of the three principal’s tellings seems to enhance its plausibility by owning up to responsibility for the death of the nobleman, rather than trying to evade blame or point the finger at someone else, as one might expect: the bandit gleefully and defiantly confesses in the courtyard; the wife claims she plunged the dagger into her husband, although the nature of what would in common law parlance be called her mens rea is left open; and the nobleman admits to suicide, as we have seen.

The experience of showing the film to groups of students suggests that many viewers instinctively gravitate towards the woodcutter’s second version of events in the clearing: it is placed last, thereby appearing to resolve the contradictions created by the competing versions that precede it; it is full of circumstantial detail, synthesizing in a seemingly plausible and coherent manner some of the motifs and diegetic bits and pieces, such as the dagger, that circulate within and between the three principals’ accounts; and, on the surface at least, the woodcutter would seem to lack the principals’ egotistical investment in spinning the story in a certain way. There are also built-in plausibilities, or reality effects, absent from the other renderings.* So, whereas in the bandit’s telling the swordfight that he has with the nobleman is a demonstration of ultra-masculine skill and bravado in obvious keeping with his own self-image (as well as with the conventions then governing the stylized depiction of duelling in Japanese cinema), when the two men come to blows in the woodcutter’s version, they are depicted as timorous, emasculated and rather pathetic figures, not only reduced to looking and acting like children but also brought down to the level of animals in their panicked, desperate scrambling around on the forest floor. This looks like what fighting someone to the death, shorn of its epic, masculine performativity, might actually be like.

But is the woodcutter such a privileged and reliable witness? After all, he is the only character who gets to tell his story twice, and he compromises his credentials in the process. His first version, as told to the unseen authority-figure in the courtyard, simply has him walking through the forest — a sequence shown in a famous montage — and literally stumbling over the nobleman’s body after the fact; whereas it is only in his second version, which he is eventually goaded into volunteering by the commoner’s cynicism, that he emerges as an eyewitness in the fuller sense of the word. Even then, as we have seen, doubts emerge as to the true nature and extent of his involvement in what took place. So, pace many students’ instinct to search for a resolution and to attach their faith to that version of events which seems best able to provide it, a preference for the woodcutter’s tale really comes down to the triumph of hope over reason — the desire or expectation that, somewhere in all this confusion, the truth will ultimately prove accessible after all because there was somebody there to witness it.

Rashomon is only one among a number of works of art that in various ways exploit the device of conflicting perceptions, memories and narrativizations: other examples include Robert Browning’s long poem “The Ring and the Book’ (1868-9), Fyodor Dostoyevsky’s The Brothers Karamazov (1879-80), G. K. Chesterton’s short story ‘The Man in the Passage’ (1914) and William Faulkner’s The Sound and the Fury (1929). But in its formal structure and its unwillingness to offer the viewer the easy closure of a resolution — at least on the narrative level of the events in the forest — Rashomon stages the problems of eyewitness perception and eyewitness narrative in particularly compelling, almost ‘textbook’, terms. This makes it an excellent point of entry into the questions that this book will address. There are probably only a handful of examples of what has been termed the ‘Rashomon effect’ in premodern history because the surviving sources seldom cluster in sufficient depth around events that were as compactly bounded in space and time and as readily observable as are the small-scale human confrontations played out in the clearing. And even when there are such source concentrations, they are more often than not derived from other written texts or are in conversation with oral traditions, not direct, independent and unmediated witnesses. Possible examples of the Rashomon effect at work might include certain episodes during the First Crusade, the murder of Thomas Becket, and some of the more notorious incidents during the Spanish conquest of the New World such as the capture and execution of Atahualpa.? But it is important to stress that Rashdmon’s lessons extend far beyond the small number of recorded moments in premodern history for which we have multiple more or less discrepant and more or less independent sources that record, or are otherwise informed by, one or more eyewitnesses’ perceptions.' Larger questions emerge. What do we think we mean when we describe someone as an eyewitness? What are the expectations and assumptions that we pack into the word eyewitness — that is to say, the very expectations and assumptions that are so profoundly destabilized in Rashémon? What is an eyewitness source, and why are we so often disposed to privilege it over other sorts of historical evidence? Should eyewitness sources be read in particular ways, and what challenges of interpretation do they pose?

The once fashionable primers of historiographical method typically held up eyewitness evidence as a — sometimes the — privileged route into reconstructing the past.!! And while more recent verdicts tend to be rather more guarded, eyewitness evidence, broadly conceived, retains its status as a central plank of historical research.!? Much of the potency of the word eyewitness derives from the fact that when historians apply the term to their sources, they are stepping outside the technical and methodological boundaries of their discipline and plugging into much larger circuits of linguistic and cultural practice. The effect is to make the mobilization of eyewitness evidence seem simply a matter of common sense. The portmanteau word ‘eyewitness’ is first attested in English in the sixteenth century.!? But the close combination of the two elements that the word captures goes back much further. Its equivalents in other European languages — for example, in the primary sense of the person who sees, témoin oculaire, testigo ocular, Augenzeuge, ooggetuige, ayenvitne — suggest that the tight and natural-seeming juxtaposition of the acts of seeing and of bearing witness transcends linguistic difference. Each of the two elements carries with it powerful associations that are magnified still further when they are combined.

One indication of the cultural importance of sight is that, in English as in many other languages, seeing is not confined to its literal semantic range. It extends figuratively into innumerable metaphors involving intangibles and abstractions, as well as mental actions of all kinds: ‘I see what you mean’, ‘She glimpsed the truth’, ‘What is your perspective on what happened?’, ‘This changes his worldview’, ‘Watch yourself’, and so on.!4 This is not intended as a ‘sightist’ observation at the expense of blind or visually-impaired people, simply a recognition of the fact that sight is much the most deeply sedimented and wide-ranging figurative resource among the five senses, especially so when we further factor in the many metaphorical applications of light and darkness. Sight is, alongside bodily orientation in space and physical motion, among the basic building blocks of what George Lakoff and Mark Johnson have termed the ‘metaphors we live by’.!> Indeed, it is closely bound up with questions of bodily situatedness and movement in that it typically seems the most responsive and versatile of the senses: to a large extent at least, the eye responds to the conscious will of the viewer, and attention can be purposefully directed towards a particular object. Sight is thus the sense that we tend to feel most effectively positions us as active subjects apprehending the world as opposed to passive recipients of the world’s acting upon us. When historians use the term eyewitness evidence, they are implicitly appealing to these powerful associations, for the expectation is that in the act of generating the source to hand the historical observer has more or less seamlessly made a transition from simple sensory perception to cognitive apperception, in other words from the workings of physiology to cultural articulation. She or he has introduced into the source some expression of the understanding and interpretation that the metaphorical acceptations of sight and light capture.

Witnessing is such a resonant idea because it is flexible, adapting to a wide range of human situations and needs. In the section (#10) of his An Enquiry Concerning Human Understanding (first published in 1748) concerning miracles, David Hume observed that ‘there is no species of reasoning more common, more useful, and even necessary to human life, than that which is derived from the testimony of men, and the reports of eye-witnesses and spectators’.!° That is to say, much and probably most of our knowledge of the world and the beliefs that this knowledge subtends reach us from what others reveal to us about their own experiences. This pooling of information locks us into the world as we believe it is, given that our general day-to-day experiences of human interaction tend to reassure us that there is a tolerably close correspondence between the testimony we receive from others, unless we happen to suspect mendacity, incapacity or error on their part, and how we believe the world is or plausibly might be. In the helpful formulation of C. A. J. Coady, whose 1992 study of testimony remains the best philosophical exploration of the subject: “The judgements of others constitute an important, indeed perhaps the most important, test of whether my own judgements reflect a reality independent of my subjectivity.

As Coady suggests, there may well be an egotistical or individualist slant within people’s self-awareness that predisposes them to overestimate the extent to which their knowledge of the world derives from their own experience, and correspondingly to underestimate their reliance on what they learn from others.!8 In other words, we are prone to exaggerate the extent to which we feel, epistemologically speaking, masters of the world around us. This has important implications for our understanding of historical eyewitness evidence, for it is deceptively easy to project this same sort of epistemological over-confidence onto the historical observer. Such a projection is made all the easier by the fact that ‘eyewitness evidence’ is not a precise and technical term of art but a very large and baggy category that subsumes a wide variety of human experiences and observer-observed relationships. As Marc Bloch noted, eyewitness evidence is usually nothing of the sort on a strict understanding of the term. His example is that of a general whose official account of the victory recently won by his forces necessarily draws on much more than the memories derived from his own sensory experiences, even if he enjoyed a good view of the battlefield. In order to craft a coherent account of what happened, he must also have recourse to the testimonies of informants such as his lieutenants.!? Bloch was here drawing upon the well-worn topos, familiar since Antiquity, to the effect that battles represent the limit case of individuals’ inability to grasp the scale and complexity of what is going on around them.”° But his larger point extends to all varieties of historical action and testimony. If anything, Bloch’s claim that a ‘good half of all we see is seen through the eyes of others’ would seem to be an understatement.7!

As we shall see, none of this study’s chosen texts — in this respect they are broadly representative of medieval eyewitness narratives in general — are autobiographical memoirs in the sense that they consistently foreground the author’s personal circumstances, perceptions and subjective reactions within the larger frame of the narrated action. In most of the sequences that our texts narrate, the eyewitness author is not ‘there’ at all in the sense of being overtly situated within the action in propria persona, still less precisely positioned relative to the action in such a way that he is able to bring an observant and tightly focused ‘camera eye’ to bear on what is happening. In some cases we can draw on external evidence to deduce that the author could not have been present at a given event or was at least unlikely to have been so. In other cases the physical proximity of the author to the narrated action may be suspected with greater or lesser degrees of confidence, but this is not expressly stated in the text. A minority of scenes or episodes include the sort of circumstantial detail that might seem to suggest that the author is recalling a direct eyewitness experience. But in some of these instances, perhaps the majority, we would probably do better to treat what we are reading as attempts to mimic the subjective texture of vivid recall rather than as direct evidence of the workings of eyewitness memory. When historians categorize a source as eyewitness, there is a tendency to allow this designation to blanket the material as a whole, whereas authorial autoptic perception may well inform only a small portion of the global content. We should always remember that when we label the authors of sources as eyewitnesses, we are for the most part simply saying that they found themselves placed in situations in which they were optimally exposed to the testimony of others, with intermittent opportunities for ‘topping up’ their knowledge with their own personal experience. Paradoxical as it might seem, being an effective historical eyewitness would normally seem to have had less to do with visual acuity or some happy knack of being in the right place at the right time, and more to do with being a good listener. But such is the resonance of the term eyewitness that when we apply it to a historical source, it can easily inflate our estimation of its unmediated experiential basis.

A good deal of medieval history-writers’ understanding of the value of eyewitness experience, their own and that of informants whom they considered trustworthy, was inherited from their ancient Greek and Roman predecessors, although there were some differences of emphasis, as we shall see. It was stock etymological wisdom among ancient historiographers that the word iotopia/ historia derived from the Greek verb iotopeiv, ‘to inquire’, ‘to observe’, with the result that it was felt to be incumbent upon the historian to establish his personal credentials as a researcher in order to pronounce authoritatively and credibly upon his chosen subject. Often this involved rhetorical appeals to good character and impartiality, as well as references to the time, effort and expense involved and the difficulties overcome — more practical and logistical than conceptual and epistemological — in the process of researching and writing.2? Where the historian’s subject matter was the recent past, moreover, mention could also be made of his own perceptions or those of others, with the experience of sight assuming a standard, though not automatic, pride of place over hearing and the other senses.” It is noteworthy that Herodotus chose very early in his Histories to illustrate the supposition that people believe the evidence of their eyes more readily than what they hear in his story of how Candaules, the king of Lydia, improvidently arranges for his favourite guard, Gyges, to spy on his naked wife in order to prove his boastful claims about her great beauty; Candaules’s explanation to Gyges assumes the force of an obvious and incontestable cliché when he observes that ‘it’s true that people trust their ears less than their eyes’.?4

Occasional notes of caution were sounded about the uncritical use of others’ testimonies: as Thucydides observed, it could be very hard work to appraise evidence rigorously ‘as eyewitnesses on each occasion would give different accounts of the same event, depending on their individual loyalties or memories’.*> Thucydides stages the mutability of the eyewitness’s gaze in a remarkable passage that forms part of his account of the Athenians’ ill-starred campaign against Syracuse in 413 BC. The Athenian land forces look on helplessly from the harbour as they watch the progress of the naval battle that will decide their fate. Various groups have different lines of sight on the action, and their responses to what they think they see play out in their different cries and bodily movements, until a collective understanding of the disastrous Athenian defeat gradually emerges:

For the Athenians everything depended on their ships, and their anxiety for the outcome was intense beyond words. Localized action varied throughout the theatre of battle, and so inevitably the men lining the shore had varying perspectives: the action was quite close in front of their eyes, and they were not all looking at the same arena. So if some saw their own side winning in their particular part of the battle, they would take instant encouragement and begin calling out to the gods not to deprive them of this hope of salvation; others who had witnessed an area of defeat turned to loud cries of lament, and from the mere sight of what was happening were in more abject terror than the actual combatants. Yet others, focused on a part of the battle which was evenly balanced, went through all the agonies of suspense: as the conflict lasted on and on without decisive result, their acute anxiety had them actually replicating with the movement of their bodies the rise and fall of their hopes — at any point throughout they were either on the point of escape or on the point of destruction. And as long as the battle at sea remained in the balance you could hear across the Athenian ranks a mixture of every sort of response — groans, cheers, ‘we’re winning’, “we’re losing’, and all the various involuntary cries let out by men in great danger.”°

In general, however, even though there was a lively tradition in ancient philosophy of questioning the reliability of the senses and the status of the knowledge derived from them, among historians reservations about eyewitness testimony simply attached to questions of the witness’s possible bias and partiality, not to more basic matters of human perception and cognition.?” The result was that when historians, as they often did, sought to establish their credentials as impartial and scrupulous authorities in contrast to the shortcomings of others, they were implicitly assuring the reader that the evidence gathered from the experience of their own eyes was impeccably reliable.

The trust reposed in the historian’s own sensory perceptions was magnified by an extension of ‘autopsy’ (adtdaty¢) in the strict sense of direct visual apprehension ofa given event (Gyzc¢) to include more wide-ranging personal experiences that aided a feeling of proximity to and understanding of the historical reality in question.?® Public affairs such as politics, diplomacy and war were considered the proper stuff of written history, and it was therefore routinely assumed that only those with personal experience of such matters were equipped to pronounce upon them.”? As Polybius, the ancient historian whose methodological remarks on this score are the most considered and developed, observed: The point is that, just as it is impossible for someone who lacks military experience to write about warfare, it is impossible for someone who has never acted in the political sphere or faced a political crisis to write good political history. Nothing written by authors who rely on mere book-learning [Polybius is here particularly attacking his bookish béte noire Timaeus] has the clarity that comes from personal experience, and nothing is gained by reading their work.??

Such experience extended to visiting historically resonant locations such as battlefields.3! One suspects that such a capacious understanding of autopsy, one that included the right sort of career path and opportunities for a kind of historical tourism as well as individual personal experiences and interviewing reliable third parties, was in part a self-serving way of justifying the fact that historiography was the preserve of a socially exclusive elite. There is some support for such a view: for example, the fourth-century BC writer Theopompus, whose historical works, including a history of Alexander the Great’s father Philip of Macedon, are substantially lost, was believed to have had the leisure to devote much of his life to his work, to have been able to spend very large amounts of money in conducting research, and to have had the social entrée to cultivate personal connections with important politicians, generals and intellectuals.** But more seems to have been at stake than indulging the opportunities of privilege. As Polybius shrewdly remarked, because the answers one elicits from respondents are only as good as the framing of the questions one poses, appropriate life experience was necessary to be able to extract the most useful information from witnesses to events.

It has been convincingly argued that in relation to all the ancient works of history that once existed but are now lost or known only from fragments (sadly, the great majority), the surviving corpus of substantially intact texts over-represents the sort of historical writing that focused on recent political and military events and thereby particularly lent itself to authorial appeals to eyewitness evidence.*4 This was, nonetheless, an important strand among the strategies of authorial self-fashioning and validation that were bequeathed to medieval historians. Because ancient Greek historians tended to be more expansive about their methodologies and sources than their Latin counterparts, western medieval historiographical culture drew much of its inspiration from the works of authors such as Josephus and Eusebius that both cast themselves as continuations of Greek traditions of historical writing and were available in late antique Latin translations. In his The Jewish War, for example, Josephus positioned himself as the heir of Polybius in his insistence that participation in the events that one narrates lends one’s account the important quality of vitality or vividness (évapyeia), which enhances its credibility. His leaning towards autopsy also informed the temporal and geographical scope of his narrative. As he insisted, ‘I shall relate the events of the war which I witnessed in great detail and with all the completeness of which I am capable, whereas events before my time will be run over in brief outline.’3>

The influence of Greco-Roman models on medieval historical writing plays out in innumerable ways, but is especially visible in the many prologues and other forms of front matter in which the author, in referring to the example set by ancient writers, positions himself in relation to his predecessors in a spirit of emulation or the continuation of tradition*° There was, however, a shift in historians’ methodologies and self-presentation, in that the ancient extension of autopsy into authorial life experience in the round tended to recede somewhat. This was a subtle change: it is important not to paint a caricatured contrast between mobile and cosmopolitan ancient historiographers searching out their material in a spirit of active inquiry, and their more sedentary and passive medieval counterparts trapped in a less interconnected world and obliged to wait for news to come to them in dribs and drabs. When due allowance is made for their very different social, cultural, religious and intellectual circumstances, writers as diverse as, for example, Gregory of Tours, Einhard, Liudprand of Cremona, Aethelweard and Otto of Freising were in their various ways the medieval equivalents of the ancient type of the autoptic historian: educated, mobile, wealthy, well connected and completely at home in the world of the powerful.

Nonetheless, the purchase of medieval historians’ own experiences and travels on the subject matter of their works is clearly more uneven in its application and relevance than it is within the (admittedly much smaller and less variegated) corpus of surviving ancient historiographical texts. When, for example, William of Malmesbury reached the point in his Gesta Regum Anglorum at which he began to narrate events during the reign of the king of his own day, Henry I (1100-35), and apologized to the reader that he was ‘a man far distant from the mysteries of the court’ (‘homo procul ab aulicis misteriis remotus’), he was seeking to explain the selectivity of his treatment and the fact that he was, so he claimed, ignorant of some of the king’s more important achievements, even as he also complained of the large amount of information that he still had to contend with. He was not making a point about how his limited experience meant he could not form an understanding of royal politics per se, nor that he was ill-equipped to picture what events played out in aulae, not just halls as such but all the privileged spaces of elite political action, looked like.

Similarly, William’s close contemporary Orderic Vitalis seems to be making a straightforward point about the spatial reach of his competence as observer and researcher, not admitting to the wrong sort of life experience, when he states in the general prologue to his monumental Ecclesiastical History:

For although I cannot explore Alexandrine or Greek or Roman affairs and many other matters worthy of the telling, because as a cloistered monk by my own free choice I am compelled to unremitting observance of my monastic duty, nevertheless I can strive with the help of God and for the consideration of posterity to explain truthfully and straightforwardly the things which I have seen in our own times, or know to have occurred in nearby provinces.*®

Orderic’s self-fashioning in this passage is quite subtle, for his construction ‘res alexandrinas seu grecas uel romanas’ cannot simply be read as a remark about the limitations of the geographical range of his work, even though it is set up in implied contrast to his secure grasp of events ‘in nearby provinces’ (‘in uicinis regionibus’). If meant merely as samples of the numerous places that Orderic had never visited, the series seems oddly precise and eccentric relative to those parts of the world that Orderic knew best, the English marches of his childhood memories and the Norman-French borderlands where his monastery of St-Evroult was situated. Rather, Orderic is here gesturing towards the broad subject matter of ancient history and positioning himself in relation to it even as he implicitly acknowledges that, although his epistemological range, the reach of his autopsy, is more restricted than that of the ancient histories he implicitly evokes, this does not itself diminish his competence as an historian functioning within his tighter spatial boundaries. (The irony is, of course, that the Ecclesiastical History spectacularly broke the geographical bounds of the sort of local history anticipated in this prologue, and that as a result Orderic found himself drawing upon a rich variety of sources including numerous narrative texts, charters, letters, inscriptions, oral reports and, to a limited degree, personal observation.)*?

The contraction of the ancient understanding of autopsy meant that, if only by default, greater emphasis than before was placed on eyewitness observation in the specific sense of the visual perception of action and events.*° Medieval history-writers were in particular nudged in this direction by the formulation of the most widely circulated and authoritative ‘dictionary’-like definition of history available to them, that to be found in Isidore of Seville’s Etymologies. According to Isidore:

A history is a narration of deeds accomplished by means of which those things that occurred in the past are discerned. History is so called from the Greek term ano tov iotopetv, that is ‘to see’ [videre] or ‘to know’ [cognoscere]. Indeed, among the ancients no one would write a history unless he had been present and had seen what was to be written down, for we grasp with our eyes things that occur better than what we gather with our hearing. Indeed, what is seen is revealed without falsehood.*!

In his sweeping, and to a large extent inaccurate, characterization of ancient historiographical practice, as well as in his significant narrowing of the semantic range of iotopeiv to include only acts of visual perception and apperception, Isidore lost much of the sense of active inquiry and wide-ranging experience that ancient notions of autopsy had captured. The emphasis was instead placed on history conceived as the reflex of the mechanisms by means of which information reached the historian. And even though Isidore’s ideal of historical writing based exclusively on the author’s personal participation in events and his own eyewitness observations was very difficult to effect in practice, at least over anything more than brief bursts of autobiographical reminiscence, it easily commended itself as a rhetorical posture to enhance the historian’s authority and the credibility of his narrative in the round.4” One consequently finds in medieval historical writings numerous prologues, dedicatory epistles and other forms of prefatory utterance that ring the changes on the epistemological primacy of sight and articulate variations of the ‘sooner by the eyes than by the ears’ topos. For example, Einhard, whose Life of Charlemagne (probably composed in the 820s) was very widely read and copied, claimed that no one could write a more truthful account of those matters of which he had first-hand experience and knew ‘with the faith of one’s eyes’ [oculata fide].** In the following century, in opening his Antapodosis Liudprand of Cremona apologized to his addressee, Bishop Recemund of Elvira, that for two years he had put off making good on Recemund’s request that he write a history of all the rulers of Europe, ‘not as one who, reliant on hearsay, can be doubted, but as one who is reliable, like one who sees’.

In a similar vein, Geoffrey Malaterra, who was a fairly recent arrival to southern Italy and Sicily when he wrote his history of Count Roger of Sicily and his brother Robert Guiscard in or soon after 1098, felt that he had to exculpate himself to his addressee Bishop Angerius of Catania by claiming that errors of chronology or omissions were to be ascribed not to the author himself but to his informants, in particular when it came to events that had taken place before he arrived in the area. His implication is that he would have been able to exercise much greater quality control had eyewitness participation in, or at least greater proximity to, the action been possible.*> In offering a summary of the reasons why Prince John’s intervention in Irish affairs in 1185 was a fiasco, Gerald of Wales, who had been in Ireland at that time, reasserts his eyewitness credentials by quoting John 3:11: ‘We speak of what we know. We bear witness to what we have seen.’4° And a similar epistemological leaning is evident in William of Malmesbury’s contrasting treatments of two (on the face of it similarly impressive and politically significant) ecclesiastical councils that took place in England only a few months apart. Of the earlier council, that held at Winchester in April 1141, William expresses the belief that because he had taken part in the proceedings and his memory of them was very good, he is able to narrate the full truth of what had transpired (‘integram rerum ueritatem’). But about the latter, at Westminster in December, he is much more guarded: ‘I cannot relate the proceedings of that council with as much confidence as those of the earlier one because I was not present.’47 As these and many similar remarks make plain, there was a clear tendency among medieval historians to believe that personal eyewitness was the single most secure and valuable resource at their disposal.

In practice, however, writers understood that the sort of expectations as to accuracy, amplitude and coherence that William of Malmesbury encapsulated in the term integra ueritas could not always be met on the basis of authorial autopsy alone. Already in the ancient period historians had lamented the physical constraints under which they worked and had fantasized about being able to be everywhere all at once;*? and their medieval successors likewise appreciated that, as was almost invariably the case, they needed to cast their net of sources as widely as possible if they were to write the history of public affairs in ways that suitably foregrounded the actions of third-party principals and aimed for a spatiotemporal reach greater than that of their own personal experience. The result was an often eclectic approach to the gathering of information that blended various types of sources of information in the interests of making the most of what were frequently acknowledged to be inadequate resources. An important exemplar was provided by Bede, himself following the lead of models such as Gregory the Great’s Dialogues and Eusebius’s Ecclesiastical History, which was available to him in Rufinus’s Latin translation.*° In the dedicatory epistle that begins his own Ecclesiastical History, Bede offers his addressee, Ceolwulf, king of the Northumbrians, a quite full and painstaking itemization of the sources (auctores) upon which he had drawn.>! For the longer-range portions of his work, Bede states that he had consulted writings gathered ‘here and there’ (hinc inde): he does not mention the authors by name, but we know that he drew upon Orosius, Constantius, Gildas and others, in very large part thanks to the unusually rich library resources available to him in his twin monastery of Wearmouth-Jarrow.

Bede also drew on archival materials in Canterbury and Rome, in written copies or in oral summary, through the good offices of Albinus, the abbot of the monastery of SS Peter and Paul in Canterbury whom Bede acknowledges as a major source of encouragement to write the Ecclesiastical History, and of his go-between Nothhelm, a priest from London (and future archbishop of Canterbury) who conducted research on Bede’s behalf during a visit to Rome. Once Bede reaches the all-important threshold moment of the arrival of Augustine’s mission to the Anglo-Saxons in 597 — that is to say, a span of a little more than 130 years before Bede was planning and writing the Ecclesiastical History, which he finished in 731 —his preface appeals to a kind of apostolic succession of elite ecclesiastical tradition as passed down from the time of the earliest evangelization of the English. Here named individuals and institutions are singled out, each typically dominating what Bede could discover of a part of the Anglo-Saxon world with which he was generally unfamiliar: thus Albinus himself for Kent and some other places; Bishop Daniel of Winchester for much of the south and south-west; the monastery of Lastingham, which although situated in Yorkshire preserved the memory of the evangelization of Mercia and Essex by its founders, Ched and Chad; an Abbot Esi, together with the writings and traditions of people in the past, for East Anglia; and Bishop Cyneberht, alongside other ‘trustworthy men’ (fideles uiri), for the kingdom of Lindsey (the area approximating to the later Lincolnshire).

This is not an exhaustive list, for there are many passages in the body of the text that must have been based on other sources of information. But its symmetries are meant to situate Bede’s research within a clear three-way matrix: written sources; information supplied by individuals who stand out by virtue of being named and whose trustworthiness is a compound of their personal relationship to Bede himself, their elite status within the Anglo-Saxon Church, and their careful cultivation of memories of their predecessors; and behind these foregrounded figures a hazier but important body of memories, sometimes fixed within a specific institutional setting such as the monastery of Lastingham, but more often a freer-floating ‘tradition of those in the past’ (traditio priorum) that at its outer limits dissolved into an even more imprecise category of “common report’ (fama uulgans). When Bede’s preface turns, however, to his home region of Northumbria, about which he knew much more and which duly enjoys a disproportionate amount of coverage in the Ecclesiastical History, he is aware that two shifts of emphasis come into play: he can draw on his own experience, and he has recourse to a much greater number, countless even, of informants.** As he notes: ‘But what happened in the church of the various parts of the kingdom of Northumbria, from the time when they received the faith of Christ up to the present, apart from the matters of which I had personal knowledge, I have learned not from any one source but from the faithful testimony of innumerable witnesses, who either knew or remembered these things.’>? Written texts continue to be important for the Northumbrian portions of the work, particularly what had been written about St Cuthbert at Lindisfarne. But the most important methodological lesson to be learned by those many later historians who looked to Bede for inspiration and whose projects were likewise set, in whole or substantial part, in their own localities, their own Northumbrias, was the essential epistemological binary between the sort of knowledge that one could obtain directly (in Bede’s phrase per me ipsum) and the indisputable testimony of reliable witnesses (certissima fidelium uirorum adtestatio).

Numerous examples of this binary appear in medieval historians’ programmatic utterances. William of Malmesbury, for example, who was particularly conscious of following in Bede’s footsteps, observes in the prologue to the first book of his Gesta Regum Anglorum, which serves as a general preface to the whole work, that whereas the reliability of his narrative, which begins with the end of Roman rule in Britain and the arrival of the Anglo-Saxons, substantially rests on that of his (written) authorities (auctores), his selection of material from more recent times is derived from what ‘I either saw myself or heard from men who can be trusted’.*4 Similarly, Henry of Huntingdon, another self-conscious heir to the tradition of Bede, in the opening of the seventh book of his Historia Anglorum, flags up the transition from reliance on old books and fama uulgans to the means by which he knows of recent events (Book VII begins with the reign of William Rufus, 1087—1100, and was first written in the early 1130s). He announces that ‘Now, however, the matters to be studied are those that I have either seen for myself or heard about from those who did see them.’>> Formulations of this sight-report binary were not confined to those writers who deliberately fashioned themselves on Bede as their principal model. They appear in a wide variety of texts, for example in the remarks of two historians of the twelfth-century Mezzogiorno, Falco of Benevento*® and the writer conventionally but almost certainly incorrectly known as Hugo Falcandus;>7 in Eadmer’s statement of purpose at the beginning of his Historia Novorum in Anglia;>® Rodulfus Glaber’s Five Books of the Histories;°? Wipo’s Deeds of Emperor Conrad II,®° and Helmold of Bosau’s Chronicle of the Slavs.°!

In a similar vein to William of Malmesbury and Henry of Huntingdon, William of Jumiéges, in the dedicatory letter of his Gesta Normannorum Ducum addressed to William I of England in about 1070, deployed the sight-informant binary in order to set up a contrast in relation to his written sources. He had relied on a guide text, the history of Dudo of St-Quentin, for the earlier portion of his narrative as far as the time of Duke Richard II of Normandy (996-1026); after this point, we are told, he had included material ‘partly related by many persons trustworthy on account equally of their age and their experience, and partly based on the most assured evidence of what I have witnessed myself, from my own store’.® It is true that one encounters several instances of authors praying in aid written sources in addition to personal observation and the reports of informants: for example, in Lampert of Hersfeld’s account of the foundation of the church of Hersfeld, written in the 1070s;°? and John of Salisbury’s Historia Pontificalis from the mid twelfth century. But more often than not the sight-reliable informant nexus was set up in contrast to the use of written authorities, not simply as a complement or amplification of it, and in terms that seem intended to suggest a qualitative shift of methodological orientation and epistemological ambition on the author’s part. This is particularly evident in those cases in which, either by virtue of their choice of subject matter or because of some external factors beyond their control, writers had to concede that in the absence of written evidence on which to base their histories they had no choice but to become flexible and creative. A good case in point is Lethald of Micy’s Miracles of St Maximinus, written in the early 980s, in which, confronted by the absence of adequate written records for the early history of the monastery of Micy, the author states that he directed his attention to the question of how best to deploy what he had himself seen and the truthful accounts of reliable informants.°

Does the frequent pairing of visus and auditus as complementary means to gain access to the past suggest that they were believed to be epistemologically equivalent, despite the recurrence of the trope about the superiority of the eyes that we have already noted? To some extent a projection of the perceptual and mnemonic capacities of the eyewitness onto third-party informants was implied by the semantic ranges of the Latin noun festis and verb testificari and their vernacular equivalents. In the same way that the English word ‘witness’ suggests both the experience of perception and the subsequent articulation of that experience, as in the bearing of witness, so the testes whose names, for example, appear in countless medieval documents were both witnesses to the transaction set out in the text and witnesses to the fact that the transaction had taken place; in this latter capacity their testimony could, potentially, be required to settle a legal dispute at some future date. The dual sense of being a witness would also have been very familiar from numerous reference to both witnessing-as-seeing and the bearing of witness in the Bible, especially the Gospels and Acts of the Apostles. It is reasonable to imagine that when medieval historians drew upon the festimonium of others, they were imaginatively projecting onto their interlocutors the same depth and acuity of eyewitness understanding that they would have expected to achieve had they been present themselves. In such cases, the informants were effectively autoptic surrogates. Thus, for example, Falco of Benevento invoked the testimonium of those who had been present at the anointing of Prince Robert II of Capua in 1128 in support of his belief that 5,000 people had been present at the occasion.

On the other hand, the kind of complete and explicit equivalence between seeing for oneself and having others seeing for you that one encounters in, for instance, Henry of Huntingdon’s formulation that we have already noted, to the effect that his treatment of recent affairs would be based on what he himself had seen or what he had heard from those who had themselves been eyewitnesses (‘uel ab his qui uiderant audiuimus’), was quite unusual.°7 More common was a studied imprecision about the exact relationship between one’s informants and the material that they had to offer: it was often sufficient just to have ‘been there’, as in, for example, Hugo Falcandus’s prefatory remarks that some of the events that he is going to recount he had seen himself, whereas others he had learned from the trustworthy reports of those who ‘had taken part’ (‘[qui in]terfuerunt’).°° The criteria by which a witness was judged to be trustworthy were typically age, education, social status, moral reputation and familiarity with the author, not visual or mnemonic acuity as such. Moreover, to be ‘present’ at an event was necessarily an imprecise notion, less a case of being granted opportunities for camera-eye visual perception, and more a cultural immersion in a given moment and a receptivity to the back-and-forth, the ‘buzz’, of other participants’ observations and reactions.

So, even as sight and hearing were often juxtaposed in historians’ methodological remarks, there was nonetheless a built-in imbalance between the two perceptual modes. Personal observation was believed to entail greater precision, a quality that lent itself to being emphasized by means of pronouns, as in ‘I myself saw’, and by intensifying constructions such as ‘with my own eyes’ or ‘with ocular trust’. This contrasted with the much baggier category of informants, some but not all of whom might themselves have been eyewitnesses to what they recounted. The use of phrases such as ‘with my own eyes’ suggests that the eyewitness historian was typically imagined close to but not caught up in the action: it is the role of observer that is highlighted, not participant. But exceptions to this sense of distance between observer and observed can be found, for example when the narrator presents himself as immersed in unusual or stressful collective situations in which his identification with other members of a beleaguered and threatened group is affirmed in acts of perception that assume arepresentative quality. That is to say, he casts himself as seeing on behalf of his co-sufferers. At the beginning of his account of the First Crusade, for example, Fulcher of Chartres, after drawing the reader’s attention to his own eyewitness credentials (‘oculis meis...perspexi’), emphasizes the sufferings and tribulations that the Franks had had to overcome during the expedition. Fulcher does so in order to invite the reader’s wonder at the manner in which ‘we, a few people’ (‘nos exiguus populus’) prevailed against greatly superior opponents.® In the prologue to his account of the murder of Count Charles the Good of Flanders and its consequences, Galbert of Bruges expresses a similar sense of participating in a momentous collective experience when he recalls that the genesis of his text was his being caught up in the aftermath of Charles’s death:

Nor was there a good place or time to write when I turned my spirit to this work, for our place [noster locus: Galbert principally means Bruges but also Flanders more generally] was so upset then by fear and anxiety that all the clergy and the people, without exception, were in immediate danger of losing both their goods and their lives. It was there, surrounded by impediments and so narrowly confined, that I began to compose my mind, which was tossing as if it had been thrown into Euripus [a narrow, turbulent channel of water], and constrain it to the mode of writing...I rest secure in the knowledge that I speak a truth known to all those who endured the same danger with me, and | entrust it to our posterity to be remembered. ”°

This degree of narratorial immersivity in and identification with collectivities in times of particular peril is, however, fairly unusual. The typical eyewitness gaze presupposed a degree of detachment from the thick of the action; the eyewitness can step back, so to speak, and take it all in. This relative distancing seems to have been grounded in the belief that one’s own autopsy, in theory at any rate, granted access to the sort of epistemological penetration and the confidence born of subjective personal experience that even the best informed and most reliable of third-party informants could never fully replicate.”! In this connection, it is significant that even first-hand individual informants are not as a rule expressly named as sources. The exceptions tend to relate to stories about miracles and wonders or to a broader category of the unusual or coincidental, typically the stuff of arresting but tangential obiter dicta, not the ‘routine’ substance of public events in the political, military or ecclesiastical spheres.’ A fortiori, longer chains of information are seldom traced out.’> An exception that helps to prove the rule, atypical both in its featuring very well-known figures and as an example of what we would nowadays term the uncanny, is a story told by William of Malmesbury in anecdotal mode. This, he insists, is not idle chit-chat (‘non friuolo auditu hausi’) but a true account that had been passed on to him by someone who swore that he had heard the story from none other than Abbot Hugh of Cluny (who had been dead more than ten years when William was writing). The narrative concerns Hugh’s first encounter with Hildebrand, the future Pope Gregory VII (1073-85), and turns on Hildebrand’s apparent ability to read Hugh’s mind, a knack so unnervingly acute that he is able to upbraid Hugh for the unfairly harsh first impressions that he had formed of him but had not articulated out loud.” Guido Schepens has observed of ancient historians’ attitudes to source criticism, in the basic sense of their categorization and evaluation of the material at their disposal, that they attached greater importance to the subjective process of discovery and less to the objective traces of the past in themselves. Research was principally a series of experiences, not an end product.’> Much the same can be said of medieval historians’ approaches: what might seem to be their methodological leaning towards eyewitness evidence on pragmatic grounds is actually rooted in the idea that historical writing was the end result of various types of experience, among which the ‘sightist’ assumptions that we have noted are embedded in moment-by-moment perception and cognition as well as in language tended to assume pride of place. That said, there are of course many instances in which the circumstances in which the historian was writing or his distance from his subject matter made it prudent to downplay or simply ignore the question of autopsy. Not all history was about recent events in one’s own part of the world. In the majority of such instances, the question of autopsy could simply be disregarded as irrelevant to the historian’s purposes.

There were, moreover, liminal cases in which the writer chose as a matter of particularly assertive authorial self-fashioning, or in anticipation of his readers’ suspicions about his methodology, to minimize, and even to rebut, the value of eyewitness evidence. A particularly clear example, remarks made by Guibert of Nogent, has been thoughtfully explored by Elizabeth Lapina in a study of the role of eyewitness authority in narratives of the First Crusade.’° Guibert’s Dei Gesta per Francos was a telling of the First Crusade that largely drew upon the anonymous first-hand account known as the Gesta Francorum et aliorum Hierosolimitanorum or a variant very close to the text as we now have it. As is well known, Guibert, alongside other learned historians of the crusade in northern France in the first decade or so after the fall of Jerusalem to the crusaders in 1099, was scornful of both the Gesta’s perceived lack of conceptual sophistication, in that it did not offer a sufficiently developed theological framework in which to situate the crusaders’ achievements within the scheme of providential history, and, related to this, its supposedly crude style, which was regarded as unfitting for such an elevated subject.’” Other writers who, like Guibert, did not participate in the crusade and also drew heavily upon the Gesta, Baldric of Bourgueil and Robert the Monk, made similar observations. But it is Guibert, the most methodologically self-conscious of the three, who works hardest to establish his trust in the truth value of the Gesta’s basic story matter while maintaining that his own retelling is superior. For Guibert, hearing (which subsumed the act of reading) was not necessarily inferior to seeing as a route to understanding, as attested by the many authoritative Lives of saints written by those who had not known their subjects.’* In Guibert’s view, eyewitnesses such as the author of the Gesta could easily make mistakes and had difficulty in probing the motivations of others; they over-emphasized one dimension of their experience — in the Gesta’s case the military aspects of the crusade — to the detriment of a rounded understanding; and in perceiving only surface realities they missed more important transcendent truths.’?

The relationship between Guibert and the recent events in the east amounted to a recasting in spatial terms of a sense of distance between author and subject matter that was more often expressed as temporal in nature. So, for example, Erchempert, whose history of the Lombards of Benevento was written around 889, admits that he is telling his story (which spans slightly more than a century) more on the basis of what he had heard than what he had seen. This he justifies with reference to the evangelists Mark and Luke, whose lack of eyewitness experience of what they narrated had not hindered them from basing their truthful and authoritative accounts (in the Middle Ages Luke was generally supposed to have authored the Acts of the Apostles in addition to his Gospel) on what they had heard from others.®° Similarly, Agnellus of Ravenna, writing in the second quarter of the ninth century, appeals to the impeccable examples of Mark, who never followed Christ’s footsteps in person or witnessed his miracles and thus drew on Peter’s memories in order to compose his Gospel, and of Luke, whose Gospel was the fruit of his relationship after the fact with Paul. These are the models for his narrative, which he says he has based not only on what he has seen but also (and by necessary implication to a greater extent, for the story stretches back many centuries to the origins of Christianity in Ravenna) on what ‘our elders’ had told him.®!

When a historian’s evolving project brought him from the distant past to relatively recent events, it did not automatically follow that this was greeted with relief now that the narrative could be reinforced by appeals to eyewitness evidence. For example, when the mid-seventh-century chronicler known as Fredegar reached the end of his reliance on the work of Gregory of Tours (d. 594), and realized that he must now strike out on his own to bridge the approximately fifty-year gap to his own time, he reassured the reader that his narrative of royal politics and warfare remained securely based on what he had read and heard and even, almost as an afterthought, on what he had seen as well.®* At work in these sorts of examples, as with Guibert, is the conviction that no, or minimal, autopsy was not a methodological impediment for the historian. On the other hand, it is noteworthy how these writers, again like Guibert, felt the need to position themselves in relation to eyewitnessing even as they sought to downplay its significance. This element of special pleading suggests that eyewitnessing did enjoy a privileged epistemic status, even as many historiographical projects did not, or could not, exploit it.

To a large extent, the important question is not what medieval historians thought about the value of eyewitness evidence in principle, but the ways in which their approaches fed through into their working methods. Most of the pronouncements that we have examined above are taken from prefatory statements that stage the author’s circumstances, aims and credentials: they are moves in what Luke Pitcher, in the very similar context of ancient historiographical practice, has nicely termed ‘author theatre’.*? But to what extent did theory and practice match up? The potentially most revealing way in which to approach this question is to consider instances of the author doubling up as an agent within his own storyworld: that is to say, as we saw above in relation to Rashomon, situations in which eyewitness perception is taken as read by virtue of its being subsumed within active participation in the events-as-narrated. For present purposes, we need to put to one side ego-texts in which the author-as-actor is the central protagonist for at least a substantial portion of the narrative as a whole, and which accordingly express much of the action in the first person singular or plural: for example, Liudprand of Cremona’s embittered and vitriolic account of his unsuccessful embassy to Constantinople on behalf of Otto I of Germany in 968-9; 84 and two well-known texts inspired by St Augustine’s Confessions, Guibert of Nogent’s Monodiae (written in 1115) and Peter Abelard’s Historia Calamitatum (c.1132), which have been central to the long-running debate about whether the twelfth century discovered, or rediscovered, the individual.®> These sorts of texts have attracted a good deal of scholarly interest, but they are few in number compared with the large majority of historical works in which the narrative centre of interest is an individual who is not the author, an institution such as an abbey or bishopric, a polity such as a kingdom or self-governing city, or a miscellany of actors within an area that happens to correspond to the reach of the author’s knowledge.

In such works, authors are for the most part sparing in their insertion of themselves into the action in propria persona. In many instances, there is no authorial intervention as a character at all within the storyworld; and in most of the cases in which there is some such presence, it tends to be occasional, brief and tangential to the main themes of the narrative, as when, for example, a personal reminiscence is cued by some name- or place-association. A good illustration is supplied by the twelfth-century chronicle of St-Maixent: when the author notes that a new abbot of the monastery of Cormery, near Tours, was elected in 1082, this is followed by the throw-away remark that it was at this same monastery that he, the author, had once encountered a monk named Litier who exercised such remarkable self-control that over a period of ten years he never touched a drop of wine or water except during Mass.*° Given that many authors grant themselves few or no ‘walk-on parts’ in their own storyworlds, it follows a fortiori that explicit references to autopsy or to conversations with informants that serve to validate a given assertion are also infrequent. For example, Henry of Huntingdon is typical of medieval historiographers in not routinely supplying the sources of his information; the narrative matter that constitutes his storyworld is for the most part delivered to the reader in the form of self-evident declarative statements. In a rare exception, however, he offers an intriguing glimpse of the temporal reach of oral tradition going back about a century when he reports the scheme hatched by King Aethelred II in 1002 to have all the Danes living in England seized and killed on St Brice’s Day (13 November). Henry remarks that in his childhood (he was probably born around 1088) he had heard it told by very old men (who must themselves have heard the story from their parents’ or grandparents’ generation) that the king had coordinated the plot by means of letters sent to every town in the kingdom.’” In many other cases, self-reference of this sort clusters around accounts of miracles, wonders and other unusual happenings. For example, although Rodulfus Glaber inserts himself into the ‘routine’ action of his Five Books of the Histories in a few passages of autobiographical reminiscence, the most sustained inclusion of individual agency on his own part comes in his account of various apparitions that he had experienced.®* Similarly, one of the fairly few self-references within the action, as distinct from editorial comment on it, to be found in William of Malmesbury’s Historia Novella concerns his learning about, though not apparently himself seeing, a solar eclipse in March 1140.8°

In a minority of cases, however, a greater degree of authorial self-insertion is visible. This could take one of two forms. Either the author is positioned close to the action in the manner of a roving reporter but does not function as an actor within the storyworld, at least not as an identified individual with discrete agency; Galbert of Bruges is a case in point, the /Jocus classicus of an observer placed very close to, but narratorially-speaking separate from, much of the action that he recounts. Or the author doubles up as an actor within the workings of the plot itself; a good example is provided by the chronicle of the polymath Odorannus of Sens, written in or shortly after 1032, which includes details of his personal involvement in events that took place in his monastery of St-Pierre-le-Vif in Sens, including a disagreement with some of his brethren that drove him into temporary exile in 1023, and his efforts to promote the cult of St Savinianus.”° A revealing limit case with regard to the inclusion of autobiographical material within a narrative that is substantially about something or someone else, here the career of a king, is Suger’s Life of Louis VI of France (1108-37).?!

As is well known, Suger, abbot of St-Denis between 1122 and 1151, was one of the most active and important figures in the Capetian regime, a close adviser to Louis VI and to his son Louis VII (1137-80). It would therefore not have seemed incongruent, an egregious distortion of political reality, for Suger to have included himself as a character in an account of the elder Louis’s reign. Sure enough, there are sequences in the Life that imply that Suger was a regular companion of the king and moved easily among the great and the good at the royal court. In a famous passage, for example, he states that he used to overhear Philip I (1060-1108) complain to his son, the future Louis VI (the imperfect ‘testabatur’ suggests regular and easy intimacy with the king and his heir), about the trouble that had been regularly caused him by the castle of Montlhéry, a few miles south of Paris.?> Suger also describes how he was one of the inner circle of close advisers (‘intimi...et familiares’) who in 1131 counselled the king to have his second son Louis crowned after his heir presumptive, Philip, had been killed in a riding accident; and he remarks that on one occasion Louis found him weeping in the royal chamber when he thought that the king was going to die.

In a few places Suger permits himself the luxury of passages in which he is the main protagonist acting in his capacity as a senior figure in the French Church, most notably in a fairly full account of the circumstances in which he discovered that he had been elected abbot of St-Denis in absentia while on a visit to the papal curia.?> But it is also noteworthy that his mentions of his meetings with Pope Gelasius II at Maguelone and Pope Calixtus II in Apulia, surely among the highpoints of his ecclesiastical career, make a point of noting that he had been sent there on royal business, as if to reassert his primary identity, for the purposes of the narrative at any rate, as a close royal confidant.?°

Given that Suger’s Life, which is by no means a comprehensive narrative of Louis’s reign and is highly selective in its coverage, chiefly concerns itself with the king’s military deeds — the frequent prosecution of just wars that cast him in the role of the defender of churches, the poor, widows and orphans — it is significant that the most extensive autobiographically-oriented sequence in the text concerns Suger’s contributions to the king’s campaigns against one of his most difficult opponents, Hugh III of Le Puiset, in 1111 and 1112. Hugh’s reputation is that of an archetypal robber-baron, thanks in large part to Suger’s lengthy excoriation of him in the Life, though in fact he was a member of an important kindred network with connections to the princely courts of northern France and to the Latin East. Hugh’s caput, the castle of Le Puiset, was situated in the Beauce, an agriculturally rich area south-west of Paris where the abbey of St-Denis owned a cluster of valuable estates centred on Toury, about four miles from Le Puiset. Having received complaints about Hugh from Count Theobald of Blois and his mother Agnes, as well in response to the petition of a coalition of bishops and monasteries, including St-Denis, whose lands Hugh threatened, Louis moved against him in 1111. As part of his forward planning, Louis sent Suger — at this stage in his career an up-and-coming political fixer in the royal circle — to Toury with instructions to improve its defences and to station a force of knights there in anticipation of its serving as a base of operations against Le Puiset. It is noteworthy that Suger, in what amounts to a mise-en-abyme of the ideological programme of the text as a whole, conflates his monastic and curial personae in insisting that he was sent to Toury, where he was already the abbey’s praepositus, or estate-manager, by the king; his abbot, Adam, gave his consent but did not initiate this move. Suger’s account of Louis’s assault on Le Puiset is told as though by someone close to the action. Although an eyewitness gaze directed at the fighting is not explicitly evoked, Suger’s substantial narrative, one of the longest military sequences in the text, includes the kinds of details, once it moves from generic evocations of the clash of arms to specifics such as a type of incendiary wagon pushed against the defenders’ gate and the brave actions of a bald priest in command of a local levy, that suggest that Suger was close to the action at least some of the time and in a position to be well briefed by others on what he had not seen for himself.’

Hugh of Le Puiset was taken prisoner when his castle fell, but in due course he was released and became an even greater nuisance than before thanks to a realignment of regional alliances around Count Theobald of Blois, who had fought against Hugh in the 1111 campaign but now joined forces with him in opposition to the king. Louis therefore resumed his offensive against Le Puiset, which had been largely demolished but was still serviceable as a dangerous base of military operations, in 1112.9° Because Toury was once again caught up in the unfolding action, Suger is able to work himself into the narrative of the second Le Puiset campaign as a prominent protagonist with a significant role to play in the plot, particularly in the early stages. Indeed, the action proper begins in Suger’s telling as he is tricked by Hugh, newly reinstalled in Le Puiset, into leaving the area in order to find the king and intercede with him on Hugh’s behalf, thereby freeing Hugh to launch an assault on Toury in his absence. Upbraided by Louis for falling for Hugh’s tricks, Suger hastens back ahead of the forces that the king assembles.

In the meantime his men have been resolutely defending Toury, benefitting from the strengthening of the site that had been undertaken the previous year. At this point in the narrative there is a remarkable, and in its explicitness quite unusual, moment of authorial autopsy placed within the diegesis. As Suger nears Toury, we are told how he and his party scrutinize the scene that confronts them: Persistently peering ahead, we beheld from afar one sure sign that the fortress had not yet been taken. Its three-storied tower could still be seen [apparebat] dominating the entire plain, whereas, if the fortress had fallen, the enemy would have immediately destroyed it by fire.°?

Suger-as-agent continues to serve as the narrative focus as he and his party boldly slip through the enemy positions and gain entry to Toury by means of a pre-arranged signal. The defenders’ morale is restored, though in his inability to refrain his men from taunting their opponents Suger implies that his personal control of the situation is less than complete.

Thereafter, as first a royal advance party and then the king himself arrive on the scene, Suger drops away as an active protagonist within the storyworld. The plot now works itself through in four phases that are all centred on Louis’s actions and reactions: a repulsed royal assault on Le Puiset in which Louis bravely covers his force’s retreat but which causes the royal host to scatter; renewed royal pressure on Le Puiset aided by the use of a nearby motte as a siege castle; Louis’s relocation of his forward base to Janville, closer to Le Puiset than Toury, and a royal victory won there; and the conclusion of the siege when Louis grants Theobald of Blois permission to slip away, leaving Hugh to be taken prisoner and Le Puiset to be razed once more. Although the narrative of these events is full and detailed, there is greater use of generic combat language and there are fewer circumstantial details than in the corresponding account of the fighting the year before. 

This would seem to suggest that Suger was not consistently positioned as an eyewitness close to the action, at least with respect to the combat sequences that account for much of the chapter, though he would have been well placed at Toury to keep abreast of developments. The remark that concludes the whole chapter, however, to the effect that when Le Puiset had been levelled, its walls shattered and it wells filled in, it resembled a scene of divine malediction, seems to evoke a moment of personal observation and reflection on the author’s part.

Suger’s Life of Louis VI usefully highlights many of the challenges that attach to autoptic evidence and our reading of it. At first sight, the passages concerning the campaigns against Hugh of Le Puiset in 1111 and 1112 would appear to be textbook eyewitness evidence: the intersection of the author’s personal circumstances and political events contrive to place Suger at the right place at the right times; and what seem to result are two full and detailed narrative sequences which, if for the most part not the direct result of visual perception on Suger’s part, at least capture his ability to keep up with sometimes fast-moving events and to correlate his experiences and impressions with those of informants who were likewise in or near the thick of the action. 

This quality seems to be confirmed by the fact that Suger’s treatment of Louis’s third and final campaign against Hugh, in 1118, is strikingly brief and undeveloped by comparison, no doubt in large part because Suger had ceased to be responsible for Toury in the intervening period. In the interests of reaching closure on Hugh, Suger is willing to break into his normal chronological structure in order to jump ahead to Hugh’s final defeat and eventual departure for the Holy Land, but the contrast with the detailed and circumstantial narratives that precede this chapter is stark.

The autoptic quality of the accounts of the first two Le Puiset campaigns further emerges in a comparison with Suger’s treatment of Louis’s actions against Thomas of Marle, who is often paired with Hugh as a representative of the aggressive castellan class that Louis was committed to subduing. In Suger’s presentation, the dynamics are in many ways the same: Louis is moved to act after learning of the complaints against Thomas raised by an ecclesiastical council at Beauvais in late 1114, and this enables Suger to position the king as a righteous protector of churches and the poor. The narrative of the royal campaign that results, and that of the second and final action against Thomas in 1130, likewise prompted by the ‘lamentations of churches’, are full and quite detailed, but they lack some of the circumstantial texture of the first two Le Puiset sequences, especially the first.!°?

It is possible that, as Lindy Grant has suggested, the attention that Suger devotes to Louis’s actions against Thomas of Marle reflects the fact that he was personally involved in these events.!™ As a senior counsellor of the king, at least by the time of the second campaign, Suger doubtless contributed to Louis’s strategizing and may have accompanied the king. But we need to be careful not to treat relative length and narrative density as unequivocal signs of authorial autopsy. Suger does not insert himself into the action in the Thomas sequences, in large part because, although St-Denis had many interests in the Laonnais, Thomas’s main theatre of operations, there would seem to have been no equivalent to Toury to place Suger at the centre of the storyworld at key plot junctures. 

Moreover, given the great deal of selectivity that Suger brings to the Life, the driving force behind his choice of incidents to include and the narrative detail devoted to them should be seen as ideological in inspiration far more than autobiographical. The text’s primary purpose is to construct an image of Louis’s kingship as that of a just ruler willing to put his military might at the service of the Church, his close relationship with which is exemplified in particular by his deferential posture towards the abbey of St-Denis.!°4 And, to this end, personal reminiscence is a useful resource, but one that surfaces only intermittently at those points in which it reinforces the ideological thrust of the text. 

The “perfect storm’ in the Le Puiset sections that aligns Suger’s autobiographical circumstances and his text’s principal thematic preoccupations is not typical of the narrative as a whole. Additionally, Suger’s Life is itself unusual among medieval historical works in the degree to which authorial autopsy plays through into the specifics of the storyworld, as opposed to simply hovering over it in shadowy and imprecise ways. If, therefore, we take the Life as an example of what an ‘eyewitness’ source can deliver, we need to be aware that in long and complex narratives the texture of the author’s autopsy will inevitably be uneven. A fortiori, this autoptic quality will vary considerably from one text to another.

The parameters within which authorial autopsy functions in Suger’s Life may usefully be compared with the mobilization of eyewitness agency and authority in a text that at first glance one would expect to be free of real-world constraints but which in fact ends up largely reinstating them: the Historia Turpini, often known as the Pseudo-Turpin. This curious text purports to narrate the history of Charlemagne’s campaigns in the Iberian peninsula, the last of which involves a version of the battle of Roncesvalles different from that familiar from the tradition preserved in the Oxford Chanson de Roland and later tellings. This is not real history, of course, but it would have been generally accepted as such by medieval readers. 

The earliest extant copy of the Historia is in the famous twelfth-century manuscript known as the Codex Calixtinus preserved in Santiago de Compostela. This contains a miscellany of materials themed around the cult of St James; the Historia, together with three documents that form a coda to it, comprises Book IV.!% The story told by the Historia is narrated by Archbishop Turpin of Reims, who claims eyewitness authority and is in a position to serve as narrator because, unlike his incarnation in the Oxford Roland tradition who is killed at Roncesvalles — he is in fact the last Frank to die before Roland! — he survives that battle by virtue of his being positioned with Charlemagne’s main army down in the valley and away from the thick of the fighting.!°’ Indeed, the Turpin of the Historia outlives Charlemagne — just.

There has been a great deal of debate about the origins of the Historia and its place within the Codex Calixtinus. Although, in its account of the privileges that Charlemagne grants to St-Denis after his return from the last of his Spanish wars, the text shows a good understanding of the more grandiose claims to political and ecclesiastical status within France that this abbey nurtured, the once common scholarly belief that it was written by a monk or monks from St-Denis is no longer generally supported.!°8 A case can be made for a Cluniac connection, but again this is open to debate. Perhaps the best that can be said is that the Historia, which is most probably to be dated to the 1140s, is a product of the many interactions between Spanish and French ecclesiastical culture that characterized the final decades of the eleventh century and the first half of the twelfth, in particular as they were energized by the growth of pilgrimage to Compostela. 

The Historia enjoyed enormous success; in its Latin original it survives, sometimes in abridged or expanded forms, in a very large number of manuscripts, about 170, the sort of total only achieved by the most conspicuous ‘bestsellers’ of medieval historiography; and from around the turn of the thirteenth century it was translated into several vernaculars and widely disseminated.! It is therefore highly likely that in its various guises the Historia was in fact the most widely read ‘eyewitness’ historical narrative in the later Middle Ages.

For our purposes, the problem of the text’s origins is less important than the form in which it presents itself in its earliest surviving version. The Historia is an eclectic mix of generic influences. The sweeping narration of warfare on an epic scale is its principal concern. Indeed, in its evocation of rulers who can quickly mobilize implausibly large armies and effortlessly traverse great distances, and in its breezy disregard of logistics, physical obstacles and all the difficulties of warfare, it closely resembles Geoffrey of Monmouth’s History of the Kings of Britain, which predates it by only about a decade.!!° But the text also contains many passages inspired by other genres, including vision and miracle literature, sermon exempla and moralizing commentary, religious polemic, royal biography, panegyric and chorography, or geographical description, which emerges in its listing of place names and its interest in local curiosities. 

There is even an excursus on the seven liberal arts cued by the claim that they were the subject of allegorical paintings that Charlemagne ordered to be made in his palace in Aachen.!!! In light of this pronounced composite quality, it is probably best to label the last person responsible for crafting the Historia in the earliest form in which it is preserved — if, as is likely, there was more than one creative agency at work in various places and at different times — as the ‘author-compiler’.

Link 

Press Here