Memory and Performance: Visual and Rhetorical Strategies of Il Fior di Battaglia

(First presented at the Renaissance Society of America’s Venice conference in 2010. Presented also as part of an academic session followed by an armoured combat demonstration, organized by Dr. Regina Pskai, at the American Association for Italian Studies conference at University of Oregon, 2013)

This paper is part of a larger study on medieval and Renaissance martial arts manuscripts, their art historical context, their relationship to medieval arts of memory, and the practical interpretation of the arts they represent. I will address the work of Mary Carruthers and Kathryn Starkey on medieval techniques of reading to show how a medieval martial arts manuscript makes use of visual rhetorical devices to address the problems inherent in notating fencing actions. MS Ludwig XV 13, dated to 1410 and commonly known by its title Il Fior di Battaglia or Flower of Battle, is a Northern Italian manuscript by a military captain named Fiore dei Liberi. The manuscript, currently held by the J. Paul Getty Museum, is a complex performance document which employs specific notational techniques to record for later use the elements of a physical performance.

The difficulties of understanding and interpreting historical martial arts texts lie partly with their semiotic remoteness from the present day. It is not simply that the teachers of those traditions are now long dead, or that the manuscripts themselves invariably seem to assume some prior knowledge of the arts they record, but also that they employ a literary, academic, and artistic vocabulary that is different from our own. To arrive at reasonable interpretations of the physical performance and use of these arts requires study of the complete cultural context in which these arts were performed. Only with this type of study can we begin to assign degrees of confidence to our interpretations of these arts.

The study of European medieval martial arts bears many similarities to the study of medieval music. As Cynthia J. Cyrus observes in her essay “Introduction to Medieval Music:”

“The medieval musical experience is impossible to recapture, for most of the music of daily life is lost to us. […] Even the music that ‘survives’ does so in a fashion that leaves unanswered fundamental questions about how it originally sounded. The medieval musician, professional or amateur, expected to improvise, adding and changing musical materials as he or she performed a piece. […] Medieval notation can be frustratingly vague for the modern scholar attempting to reconstruct a plausible and historically-informed medieval sound” (emphasis in original). ((Cyrus, Cynthia J. 2003. Introduction to Medieval Music. The ORB: On-line Reference Book for Medieval Studies. ; accessed August 22, 2007).))

In both music and martial arts, surviving evidence of the practice of the arts leaves us with an incomplete picture, with much knowledge lost either through decay and destruction of primary source materials, or because that knowledge formed part of an oral tradition never recorded. Physical arts are learned primarily through performance, and books are auxiliary materials to the physical performance.

The Fior di Battaglia contains critical characteristics of composition that make it useful to compare to Mary Carruthers’ work on the medieval ars memoria, and Kathryn Starkey’s study, Reading the Medieval Book. Starkey’s monograph on the Munich-Nuremberg manuscript (c. 1270) of Wolfram von Eschenbach’s Willehalm (c.1215) argues that the manuscript was composed as a self-conscious hybrid product that mediates between oral performance and written manuscript. By examining the manner in which the Fior di Battaglia applies innovative strategies of mediation between presentation and performance, strategies similar to those outlined by Starkey, and by observing the ways in which the manuscript conforms to and employs the memorial arts that informed medieval scholarship, we can arrive at a deeper understanding of the nature of the text itself. This will illuminate the connection between text and practice—that is, between the representation of a complex physical practice on a flat page and its realization by agents in three-dimensional space.

Critical to this connection is the concept of orthopraxy – literally “correct practice,” which is related to orthodoxy, or “correct belief.” (( ; accessed 22 August 2007))  The term “orthopraxy” is usually employed in religious contexts as describing correct ways to live (as opposed to correct beliefs to hold)  ((“Orthopraxy” The Concise Oxford Dictionary of World Religions. Ed. John Bowker. Oxford University Press, 2000. Oxford Reference Online. Oxford University Press. University of Oregon. 22 August 2007
Orthopraxy and orthodoxy are related concepts, but there can exist a divide between belief and practice. We are concerned only with practice.)), but Carruthers employs the concept of orthopraxy in her treatment of physical practices, which she defines as follows:

“Any craft develops an orthopraxis, a craft knowledge which […] can only be learned, by the painstaking practical imitation and complete familiarization with exemplary master’s techniques and experiences. Most of this knowledge cannot even be set down in words; it must be learned by practicing, over and over again.” ((Carruthers, Mary. The Craft of Thought. Cambridge: The Cambridge University Press (1998): 1-2. Carruthers peaks specifically of monstic meditation as “the craft of thought.”))

This is also a fully functional definition for the means of learning a martial art.

The concept of orthopraxy, which formed the basis of practice in craft guilds and monasteries in medieval Europe, has profound implications for our study of medieval martial arts manuscripts and the way in which we study from them, for like scholars of medieval music, we seek to arrive at a credible, informed practice of the arts themselves. The surviving books, however useful to those who employed them when they were written, are today mere physical artifacts of a lost practice. Our study of these physical artifacts is problematic at the outset, as the books themselves were peripheral to the actual practice. We are attempting to recreate an art from signs left by its passage.

Visual Program and Mnemonic Structures of MS Ludwig XV 13

The Fior di Battaglia is part of a group of five related manuscripts. In addition to the Ludwig manuscript in the Getty Museum, another copy, MS M.0383 in the Morgan Library in New York, has very similar artwork and script, but is incomplete. Whether it was unfinished or portions were lost is unknown. ((The instructional sequences in MS M.0383 that do correspond to Ludwig XV 13 have fewer individual techniques, and the introduction, though almost identical, makes no mention of a patron. This may suggest that MS M.0383 is an earlier, possibly abandoned, draft.)) There exists a facsimile executed in 1902 of a third copy, the original of which is in the private collection of the Pisani-Dossi family in Italy.


A fourth, MS.Latin 11269, is held by the French Bibliothèque Nationale. Finally, a manuscript by the Pisan fencing master Fillipo Vadi dated to 1482 appears to be a partial copy of the Pisani-Dossi manuscript. There are important differences as well as substantial similarities between the five known manuscripts.


The Getty manuscript begins with an introduction that covers folio 3, recto and verso, and folio 4, recto only. The 315 pen and ink illustrations, executed in a Northern Italian, possibly Venetian style, begin on 8 recto and continue to 49 recto ((Kren, Thomas, and Kurt Barstow. Italian Illuminated Manuscripts in the J. Paul Getty Museum. Los Angeles: Getty Publications, 2005. The Pierpont Morgan Library’s catalog notes to their similar copy of the manuscript state, “A manuscript of the same text, which also was executed in the Venetian region in the early 15th century, can be found in the J.Paul Getty Museum (Ms. 83.MR.183); see the Ludwig catalogue, v. 4, p. 282-288.”)) . Most pages have a grid of four images on them, with occasional groupings of two and three images, three instances of a single image and a single grouping of five images. The manuscript is written in a Venetian dialect of Italian, and the script is a “small humanistic script” similar to Italian legal or notarial cursive ((,1&Search%5FArg=MS%20M%2E0383&Search%5FCode=GKEY%5E&CNT=50&PID=TMgFvkNMLFrXpsH4hCavG1PAElmk&SEQ=20100324155956&SID=2 Last accessed March 24, 2010.)).


This is a sophisticated didactic manuscript. Notable is the variety of instruction: in all, there are 10 combinations of weapons taught for combat on foot, and five on horseback. The text of the Fior di Battaglia is organized into logical units of related actions, beginning with abrazare (wrestling and grappling arts) and moving to dagger combat. A bridging section of dagger against sword brings us to techniques for the use of the sword in one hand, which is followed by the use of the sword in two hands ((The medieval longsword was capable of being wielded in one or two hands.)) . A short section follows showing various combinations of sword, spear, and stick. At this point, at folio 32 recto, there is a thematic diagram of the key principles of the art, called by Fiore the segno. This mnemonic diagram is worth a study in itself, but the key element for present purposes is the Lynx, at the top, which signifies prudentia – an important memorial virtue.


The next three sections show the use of techniques for fighting in and against a harness composed of mail and plate, using sword, poleaxe, and spear. After this we are shown equestrian combat principles, with the armored figures now on horse. These cover use of sword, lance, and even wrestling on horseback. The manuscript concludes with an entreaty from Fiore to recall his humility, virtue and nobility. The final folio shows a single image of two horses tied to a tree.


The text contains detailed technical instruction, and Fiore gives us a logical exposition of the system and its inter-related elements, expounding on critical martial principles and their applications, but the manuscript is not composed as a structured lesson plan. Fiore speaks to an experienced practical martial artist who requires theoretical discussion and specific example to advance his established understanding of the arts.

The audience for this manuscript is clear: all four surviving copies appear to have some relationship to a now-lost presentation copy in the Estense family library, made for Niccolo III d’Este. The four surviving copies do not appear to have belonged to the Estense family, but since in his prologue Fiore mentions five of his famous students, all of whom are condottieri with biographies known to history, it seems reasonable to assume that these five, or men very much like them, were the intended audience. We thus have a manuscript made for, and derived from, the needs of the aristocratic fighting man.

Fiore’s introduction explains the visual program of his manuscript. He discusses key elements of the first section, the abrazare or wrestling, and then explains the system of visual notation he will use throughout. Briefly, he employs a system of masters, scholars and players to demonstrate key principles and techniques of his system. Each section of the manuscript begins with one or more crowned “First Masters” who show principles and poste (or guard positions); these figures are unopposed.


They are followed by one or more crowned “Remedy Masters” who show defenses against attacks, with the attacks being made by a “Player.” The Remedy Masters are followed by their Scholars, who wear a “device” or garter on one leg. The Scholars show the plays that stem from the defensive technique of the Remedy Master, and they execute these against the Players.  Finally, there is a “Counter Master,” who wears both a crown and a garter, who shows the technique that defeats the original Remedy Master, and thus all of his Scholars. These visual elements function as mnemonic markers, necessary for learning the art.


Orthopraxy, Performance, and the Ars Memoria

A notable element of the scholastic intellectual tradition was the use of the arts of memory. To medieval scholars, books were precious and rare things, and the means to copy information were either expensive and permanent (as in the case of parchment or vellum) or transitory (such as wax tablets drawn on with a stylus). Professional scholars, who were usually churchmen, needed a system of memory training that would render impermanent information permanent and secure.

Very briefly, the mechanics of the ars memoria was the process of lectio, or studying a text through an orderly method of exposition and analysis, followed by a meditatio or meditative phase in which the text is committed to memory using systems that involve memory locations, or loci. This process was a deliberately learned skill and common to scholastic education. It was not an instant process: the scholastic systems for memory training detailed methods whereby texts were broken into manageable segments and committed to memory over the course of time. It seems clear that a good scholar could thus commit large amounts of text into his memorial “library,” and a brilliant scholar (such as Thomas Aquinas) could accomplish astonishing feats of memorization.

The real utility for the memorial library, however, was not in simple tricks of memorization, but in composition of new texts. Scholars could mentally “read” their memorial libraries as an aid to composing original works. The process of composition, or invention, began with a phase of reminiscence, a gathering of information from one’s well-ordered memory, and resulted in the res, or the well-formed idea of the composition. This well-formed idea was akin to a good rough draft, requiring only polishing. This entirely mental draft is then set down in written form, known as a dictamen. ((Carruthers, Mary. The Book of Memory. Cambridge: The Cambridge University Press (1998): 194-195.))

The images in memory, and the loci used to hold them, were a key element of memorial effectiveness. These systems were generally derived from classical sources. One such system was laid out by Thomas Bradwardine (1290-1349), an English cleric who eventually became chaplain to King Edward III.

In his manuscript De Memoria Artificiali Bradwardine explains the creation of a memory locus in terms that sound almost identical to a medieval picture. First he states that “Locations are like tablets on which we write, the images like letters fixed on them.” He says that they must be moderate in size, and no greater than can be taken in with a single look; they must be rectangular, neither too dark nor bright, and in a continuous and straightforward order. The images should be put into regions deserted by men, and empty, and

“the whole image should have some other quality such as movement, that thus it may be commended to memory more effectively through tranquility or repose…if you can, make the image you have fashioned a right hand and a left hand; and place the second image to the right of the first, so that with its right the image of the first hold, drags or strikes the second image…or…the second behaves in such a way to the first, so that their activity will be, as it were, a fastening together of their order in the series.” ((Thomas Bradwardine, On Acquiring a Trained Memory, trans. Mary J. Carruthers, reprinted in Carruthers, Mary. The Medieval Craft of Memory. Philadelphia: University of Pennsylvania Press (2002): 207-214))

This is an apt visual metaphor for a martial arts manuscript, the contents of which must necessarily reside in the memory of the martial artist in order to be useful. The actions of connecting and striking are perfectly suited to texts of this sort, and the fastening of images into a series is clearly marked in the text by the system of crowned masters and gartered scholars. This a very deliberate and well-conceived choice. Carruthers notes that:

“the symbiotic relationship between memorial effectiveness and the layout of books throughout the Middle Ages is apparent at the level of principle and general rules; the more difficult problem is to know to what extent the selection of images and decorative elements reflected particular mnemonic techniques and themes.” ((Carruthers, Mary. The Book of Memory. Cambridge: The Cambridge University Press (1998): 243.))

When Fiore states in his introduction that “there is so much to this art that even the man with the keenest memory in the world will be unable to learn more than a fourth of it without books.” ((Fior di battaglia, 2 verso, translation Tommaso Leoni.)) it seems clear that he is making acknowledgement of the arts of memory.

It is important to realize that the images in the manuscript were not mere support for the more important words. In the medieval period, the reading of text and the reading of images were both rhetorical acts, performed in the same manner and given the same weight. Retention and retrieval of information were considered best performed by use of visualization, and the individual looked at the contexts of memory: the mediator, as Carruthers notes, is visual ((Carruthers, Mary. The Book of Memory. Cambridge: The Cambridge University Press (1998): 17, 21.)) . We have also seen in Bradwardine’s description of memory loci the intensely visual nature of the process. What is thus given to us on the page of the manuscript is an integrated, compact rhetorical unit comprised of both text and image. The visual nature of this procedure is well-suited to the skills necessary to learn and employ martial arts.

The artwork of the Fior di Battaglia appears to offer a realistic portrayal of human figures. A casual glance tells us that it appears to employ depth of perspective as well as light and shading in a way that is pleasing to our eyes and easy to follow. It shows a wide variety of faces on the figures, with few resembling each other, as well as different forms of dress, different ages, and other decorating and distinguishing marks. It is tempting, when viewing these images, to see them as exemplary of the differences between medieval and Renaissance images.

However, there are some additional considerations to take into account. The first is that although the artist or artists are skilled, there are some limitations to the depictions. The most obvious comes with the distortion of individual figures, so that limbs are not always proportionate. In an image found at 8 recto, upper right, we have a Counter Master with disproportionately short legs demonstrating a counter to a grappling play. In the same page in the lower right, we can see that the Counter Master’s left arm is also strangely elongated. In an image at 12 recto, upper right, the scholar’s left arm is elongated and disproportionate. Disproportionately long arms are found throughout the manuscript.


Additional examples show that we may not be viewing “realistic” perspective with fully developed depth-of-field, but a partially flattened one, reminiscent of relief sculpture.

Aspects of Medieval Art

One issue we have to consider when viewing and attempting to understand medieval art, or the concepts portrayed by it, is the simple fact that meanings in medieval artwork are not transparent to the modern viewer. William Diebold has argued that “Because the early medieval conception of art was fundamentally different from our own, we cannot recover the meaning of that art simply with our eyes.” ((Diebold, William J. Word and Image: An Introduction to Early Medieval Art, Boulder: Westview Press, (2000): 6)) This holds true for later medieval art as well (and I will argue that the Fior di Battaglia partakes of the characteristics of medieval art, despite appearing to be a product of the Renaissance.) For modern-day users of the Fior di Battaglia, this presents a difficult problem: if the meaning of what appears to be a primarily visual text cannot be understood with the eyes, how can it be understood at all? The answer lies in the uses to which medieval artwork was typically put, and the ways in which the techniques of artistic representation were adapted to the needs of these martial arts manuscripts.

The majority of art made in during the medieval era was Christian art, and the purposes of it were laid out in the late 6th and early 7th centuries by Pope Gregory the 1st. In the year 600, he wrote, “What writing offers to those who read, a picture offers to those who look; in it they read who do not know letters.” ((Diebold: 1))

He had written a year earlier to reprimand Bishop Serenus of Marseilles for the destruction of images, saying:

“I note that some time ago it reached us that you, seeing certain people adoring images, broke the images and threw them from the churches. And certainly we praise you for your zeal lest something manufactured be adored, but we judge that you should not have destroyed those images. For a picture is displayed in churches on this account, in order that those who do not know letters may at least read by seeing on the walls what they are unable to read in books.” ((Quoted in Diebold: page reference.))

This was a watershed moment for Christian art and art in general: at that time, it wasn’t certain if the church would have a place for art. A strict reading of the Second Commandment – “Thou shalt not make unto thee any graven image – “ could easily forbid any form of representational art, but Gregory chose a looser interpretation, in that so long as the image was not adored and was used for religious instructional purposes, it could be condoned. It meant that Christian art would develop primarily along a narrow path specific to devotional purposes, with instruction being largely confined to religious concerns. Since the church was the major producer and consumer of art in the centuries following Gregory, it would prove to be a significant force in the shaping of art.

Modern Photo-Realism

When considering the uses of art in the 13th, 14th and 15th centuries, we also need to understand that there has been a fundamental shift in modes of artistic representation between then and now. These shifts have affected our perceptions of the world around us, and our understanding of how to represent and interpret it artistically. As 21st century people we are steeped in photographic realism and have a strong tendency to view and to judge images in terms of our relationship to the brute mechanical illusion of realism given by photography. A brief look at this major shift in artistic representation can illustrate some of the issues we face when reconstructing medieval martial arts from primarily visual sources.

A revolution in world view came in 1878 with the work of Eadward Muybridge, who used photographic technology to answer what was at the time a seemingly unanswerable question: when a horse gallops, do all four hooves leave the ground at one time?

Muybridge created a new photographic emulsion process, developed electric-controlled shutters that operated at a greater speed than any previous camera shutter, rigged 12 cameras to tripwires, and sent a champion horse and driver down the track. The result was to show that the horse was actually airborne at several points during his run: all four hooves were off the ground. These are the images as they might have been seen in a zoopraxiscope, a device invented by Muybridge to show the sequence of photographs in motion.


The photographs were re-printed in newspapers across the world, Muybridge became a celebrity, and the French painter Jean Louis Ernest Meissonier, who was a famous painter of equestrian scenes, realized that he had been mis-representing horse’s strides for his whole career and vowed to never again touch a brush. (He relented, by the way, and continued to paint.)

What Muybridge did was to stand the world’s view of artisitic representation on its head. According to his biographer, Rebecca Solnit, Muybridge initiated a “fundamental change in the nature of photography” that took “humanity across the threshold into a new world of knowledge and representation.” ((Smithsonian article – create citation)) Time and motion had become linked to images in a way they never had been before.

It is not that the art of the Fior di Battaglia is deficient or poorly executed, but rather that the use of perspective in the drawing does not employ a fully-developed depth-of-field using the techniques of perspective employed later in the Renaissance – it is not nor is it intended to be what we would consider “photo-realistic.”. Art produced in the context of late medieval scholasticism was not the apparently “naturalistic” art of the later Italian Renaissance, and still less photo-realistic, but was conceptual in nature: images were typically used to express concepts rather than to represent material objects or states of being, and the techniques of medieval art reflected this use. But we should not, to paraphrase Carruthers, judge the image good or bad as a representation of reality; ((Carruthers, Mary. The Book of Memory. Cambridge: The Cambridge University Press (1998): 24.)) it is the usefulness of the image as a mnemonic marker that is important. Thus we will never be “exactly like the picture” but we must perform what the picture tells us – what we read in it. We must meditate on the compact rhetorical unit formed by text and image to derive the essential lesson found in it.

Orthopraxy: Following the Practices of an Exemplary Master

This returns us to orthopraxy. Again, Carruthers states that:

“Any craft develops an orthopraxis, a craft knowledge which […] can only be learned, by the painstaking practical imitation and complete familiarization with exemplary master’s techniques and experiences. Most of this knowledge cannot even be set down in words; it must be learned by practicing, over and over again.” ((Carruthers, Mary. The Craft of Thought. Cambridge: The Cambridge University Press (1998): 1-2.))

Put simply, the craft must be performed to be learned.

It is the necessity of constant and continual performance of these arts which makes them problematic when set down in a text. The text does not “do” anything in the active physical sense, and the act of reading the text also does not “do” anything in a physical sense. But Starkey’s analysis of the Munich-Nuremberg redaction of Willehalm suggests a different possibility for conceiving of a text, one that relates well to the qualities of the ars memoria. The Munich-Nuremberg manuscript is a product, to quote Starkey, of a period when “the introduction of writing as a means for transmitting courtly texts gave rise to a dynamic and discursive period during which authors, artists, and audiences reflected on medium, and literature, and on the storytelling process. ((Starkey, Kathryn. Reading The Medieval Book. University of Notre Dame (2004). 3))” She further notes that the illustrations in Willehalm “are part of a hybrid semiotic system that draws on word, image, and performance. ((Starkey. 5.))” A major element of the presentation of Willehelm is the figure of a narrator, incorporated into the illustrations that comprise the visual program. The narrator is a self-conscious figure who addresses the audience directly, and interprets the text of the story as it is told. This figure, Starkey argues, is “constructed as an oral performer of the story,” using in part strategies of mediation between text and audience as might have been done by an actual performer. ((Starkey. 84, 17.))

What is critical here is that the text itself becomes a performance, or at least a hybrid product that consciously incorporates performance. The Fior di Battaglia is a depiction of a martial system that must be performed. It shares an interesting characteristic with Willehalm: the presence of self-consciousfigures who form a narrative frame that describes to the audience the actions that are performed. The narrator mediates between the performed action and the audience of the text.

In the Fior di Battaglia, each “rhetorical unit” of text and image has a distinct first-person narrator. Thus the First Masters explain their principles and poste (or guard positions), the Remedy Masters and Counter Masters explain their various actions, and the Scholars explain the way in which they perform their individual plays. There is a distinct hierarchy: Scholars frequently state that they have followed the teachings of their masters, the Remedy Masters guide us to an understanding of the actions of their scholars, and the First Masters set the tone by speaking about key principles with a specific gravitas appropriate to their authority and experience. These figures create a linked narrative frame that explains, guides, and controls the audience’s experience of the text – text considered here as the combination of word and image, rather than only words and sentences.

If we consider orthopraxis, to quote Carruthers, as “a set of experiences and techniques, conceived as a “way” to be followed, leading one to relive the founder’s path to enlightenment,” ((Carruthers, Mary. The Craft of Thought. Cambridge: The Cambridge University Press (1998): 1-2.)) we can relate this to the practice of the physical arts of combat, where the experience of practicing verbatim the master’s set-plays and specific techniques leads to an understanding of guiding principles that can then be applied to new situations. The acts of teaching and learning martial arts are a type of performance in which specific roles are adopted and acted out. These roles are a ritualized preparation for still another performance, the composition of new martial knowledge, and ultimately for the spontaneous expression of that knowledge in the context of a fight. This accords well with the practice of memoria, where an exemplary text or texts are learned and placed in memory as a new experience now belonging to the reader, and are then available for the act of inventio, or composition of a new text based on principles established by authority. These new “texts” can then be performed as responses to new situations.

The ability to perform new knowledge based on authority is a critical step in learning a martial art, for no system can be an exhaustive catalog of specific techniques for application to any situation. The Fior di Battaglia is not such a catalog, but it is complete when considered as an expression of key, linked principles by an exemplary master. In the Fior di Battaglia, many of the figures make reference to techniques or principles that have been established earlier, and at the end of the first section of the two-handed sword, the reader is informed that

“Here ends the wide play of the two-handed sword. These plays are all linked, and have remedies and counters both from the right and left side, counter-thrusts and counter-cuts to each action, with breaks, parries, strikes and binds—all things that can be understood very, very easily. ((dei Liberi, Fiore. Fior di battaglia. Translation Tommaso Leoni.)) ”

These are signs that the student of the art, and of the text, is to consider what he has learned and apply it to new situations.

The text itself is an integral part of the process. By examining the manuscript in the light of the memorial process that informed medieval scholarship, and by the innovative strategies of mediation between presentation and performance outlined by Starkey, we can arrive at a deeper understanding of the nature of the texts themselves. To return to the initial comparison of medieval music studies and musical notation, we now have a more complete understanding of the notation systems employed in the text and in the practice of medieval martial arts, and can use that understanding in our continuing efforts to reconstruct, from books and from the educated memories they represent, the practice of those arts.