Subscribe now for full access and no adverts
For many people, the image of an armoured knight riding into battle or thundering around the tournament ground on a warhorse captures the very essence of the Middle Ages. Valuable, highly trained, and prized for its physical characteristics and temperament, the warhorse was both a weapon that changed the face of battle and a symbol of elite status – the very word ‘chivalry’, meaning the social code by which medieval nobles lived, was derived from the French for ‘horseman’.
All very interesting, but what has any of that got to do with archaeology? Surely medieval knights and their horses are best left to historians interested in the study of battles, warfare, and the lives of the medieval great and good? To prove this wrong, Warhorse: A Medieval Revolution? – an Arts and Humanities Research Council-funded research project – has pioneered an archaeologically driven approach to the study of warhorses to create new knowledge and afford a more rounded understanding of an iconic animal with uniquely close links to human society.
The work has changed our understanding not only of the medieval warhorse and its military role but also its broader social, economic, and cultural significance. Medieval horses were bred and bought for more than battle: they were essential in transport, agriculture, and industry. A great range of horse-types existed, and an equine hierarchy mirrored the human social strata of the Middle Ages. At the pinnacle was the ‘destrier’, an elite, prized military mount that could cost 200 or even 300 times as much as a packhorse. Beyond military-grade steeds, though, a great variety of different types of horses were bred for different purposes – the swift ‘courser’ used for hunting, the smooth-gaited ‘palfrey’ for everyday riding, and the humble ‘sumpter’, the workhorse of the medieval world.
Spanning AD 800-1550 (the late Saxon period up to and including the reign of Henry VIII), our project aimed to create a new picture of warhorses specifically, but also of medieval horses more generally, including how and where they were bred and trained. The research was run as a collaboration between the University of Exeter and the University of East Anglia, working in partnership with other stakeholders including the Portable Antiquities Scheme and the Royal Armouries, and it embraced the contributions of specialists from many different branches of archaeology and history. In this way, we were able to make use of the widest variety of evidence possible, from biomolecular science to fine art. Most previous studies of medieval warhorses have been undertaken by historians, using documents as their key source of evidence, but we set out to gather and interrogate a much wider range of material, including zooarchaeological data (horse bones and teeth); the physical remains of horse apparel, equipment, and armour; visual depictions of horses in art and sculpture; standing structures such as stables; and the evidence of the historic landscape itself. What can these various strands of evidence tell us, and how, when combined, have they created a more rounded picture of the medieval warhorse than that provided by documents alone?
Illuminating adornments
Portable objects associated with medieval horses come in a great variety of forms. Organic items such as leather saddles and fabric ‘caparisons’ (decorative coverings) are exceptionally rare survivals and are found only sporadically in museum collections. Instead, archaeological evidence for warhorse equipment mainly takes the form of metal artefacts. Systematic archaeological surveys of medieval battlefields in England and Wales are unusual and, where they have been conducted, have yielded precious few equestrian items. In order to learn more about the material culture of the medieval warhorse, we must therefore turn to a much broader set of evidence, including artefacts in museums and private collections, but especially data produced by the Portable Antiquities Scheme (PAS).
The PAS website (https://finds.org.uk) includes records of almost 1.8 million archaeological finds reported by members of the public, and the Warhorse project partnered with the PAS to make use of the exceptional research value of this resource, compiling a database of more than 14,000 separate medieval horse-related artefacts, mainly metal-detected items. The most familiar and recognisable category of equine artefact is the horseshoe, but there are relatively few of these documented by the PAS. The fact that they are made of iron – a metal typically not targeted by metal-detectorists – means that they are instead dwarfed by a huge array of other smaller bits and pieces of horse equipment made of copper alloy. Many of these objects are only recognisable by finds experts, as they formed component parts of larger items of horse gear. A good example is the ‘stirrup-strap mount’ – a cast object that protected the stirrup strap where it looped through the apex of an iron stirrup. Their presence speaks of horses used for riding, and particularly showy decorated forms appear in the archaeological record in the early 11th century, testifying to a rise in horse ostentation in the period immediately before the Norman Conquest.
Also of great research value are ‘harness pendants’, which are similarly diagnostic of riding horses. They hung from the breast-bands of these steeds – often in large numbers, to judge from iconographic sources. Numerous from the 12th century but rare after the 14th century, these were purely decorative rather than utilitarian items, and they tell us about the social cachet of the medieval horse during this period. Although individual examples may seem small and insignificant, the cumulative data from the PAS, which extends to several thousand pendants, provides valuable insights into the distribution and dating of horse gear. Given that many pendants depict heraldic symbols, in some cases it was possible to map the footprints – or ‘hoofprints’ – of noble families too, and distributions of these objects might even tell us about the locations of medieval tournament sites.
Equine armour
Another very distinctive category of equestrian material culture is horse armour, which survives from the early 15th century. Components like the shaffron (head covering) survive in museum collections in reasonable numbers, and can provide information on horse size and conformation, particularly for animals bred for battle or the tournament ground. Work carried out by the Warhorse project included an analysis of the earliest piece of surviving medieval horse armour in Europe: the so-called ‘Warwick Shaffron’, held by the Royal Armouries in Leeds. Our study, which was published in the journal Arms & Armour (open access at http://www.doi.org/10.1080/17416124.2024.2308447) explored evidence of battle damage on the piece, and demonstrated that it would have fitted a horse of above average stature for the period. It was not the largest shaffron measured during the project, however: that title goes to ‘VI.1’, which is also held by the Royal Armouries and was made c.1515 for a horse of the young Henry VIII (see http://www.royalarmouries.org/collection/object/object-2627 for more on this artefact).
We can gain further clues from depictions of warhorses in medieval manuscript illustrations, seals, wall paintings, stained glass, sculpture, and even everyday objects such as livery badges and aquamanilia (vessels for holding water at the meal table), which were sometimes shaped like knights on horseback. Where these images are in colour, we can learn more about medieval attitudes to different shades of coat: for instance, iconographic sources show a very strong association between white horses and especially noble or high-status riders – a link popularised by images of St George on his white charger – whereas black horses are more commonly seen in agricultural settings. Such evidence needs to be treated critically, as medieval artists skewed representations to reflect their worldviews and doubtless to please their patrons, but these depictions nevertheless have particular value for telling us about aspects of horse appearance that are not discoverable from other sources, such as images of organic items of apparel that do not survive in the archaeological record.
The most famous visual source for medieval warhorses is, of course, the 11th-century Bayeux Tapestry, which depicts more than 170 horses, many of them Norman mounts in battle. In the period before the 11th century, and before the widespread production and survival of manuscript illustrations, stone sculptures are the prime body of evidence for understanding horse appearance and equipment. Particularly informative are Pictish sculptured stones, like the famous 8th-century Aberlemno stone (as seen on this issue’s cover), with its detailed depiction of mounted warfare, and Viking Age sculptures from the 9th and 10th centuries in northern England. Often showing male riders in profile, in military gear, they testify to the increasing horse-mindedness of aristocratic warriors. Some of the finest sculptural representations of the period show what might equate to an ‘ideal’ horse of the Viking Age: an elegant animal with a prominently arched neck and, significantly, an ambling gait – a trait which, genetic evidence shows, was introduced to Britain during the Viking period.
Surprising sizing
Moving from artefacts, armour, and art to the physical remains of the horses themselves, zooarchaeological analysis of bones and teeth provides information about the size, health, lifestyle, and activities of these animals. These remains are not always easy to come by: while the custom of burying entire horses is well-known from Anglo-Saxon pagan funerary practices, this custom declined with the rise of Christianity in the 7th century. The fact that horses were rarely eaten after this period, combined with their long lifespans relative to many other species, also means that horse bones are encountered much less commonly on medieval sites than the other main species, typically comprising less than 3% of animal-bone assemblages. Nonetheless, zooarchaeological analysis of the available evidence has proven illuminating and, importantly, included a broader range of horse-types than just those used for warfare.
For this aspect of the project, Warhorse primarily focused on analysing excavated bones from a wide range of sites, including castles and settlements, and collating data on metrics from published reports and grey literature. We began with traditional zooarchaeological methods, collecting standardised measurements of bones to assess horse size and shape over time, and this work included detecting pathologies hinting at military use, such as joint and spinal issues, and dental damage from bits. An interesting example of the latter was highlighted in a specimen from Barnard Castle, Co. Durham, which clearly showed the impact of a ‘curb’ bit – equipment mainly restricted to higher-status use.
The project applied cutting-edge GMM (Geometric Morphometric) and isotopic sampling methodologies to medieval horses for the first time, which has enabled aspects of the skeletal signatures of breeding and training regimes to be reconstructed. A headline finding about horse stature from a project database of over 8,500 bones, published in the International Journal of Osteoarchaeology (open access at http://www.doi.org/10.1002/oa.3038), is that most medieval horses were below the size of modern ponies, standing around 1.48m to the withers, or shoulder. This discovery was backed up by systematic measurement of more than 100 pieces of medieval horse armour, which were again mainly intended to fit pony-sized mounts. While popular perceptions of medieval warhorses picture large beasts, not a single bone from the entire project indicated a medieval horse that would have met the size criteria for a modern police horse (1.68m).
Small samples from bones and teeth underwent scientific analysis too, including a case study that focused on reanalysing the unusual later medieval and Tudor horse cemetery at Elverton Street in Westminster, London (see CA 162 and CA 411). There, our programme of isotopic analysis, published in the journal Science Advances (open access at http://www.doi.org/10.1126/sciadv.adj5782), allowed us to track how individual horses had moved between locations in the early years of their life, revealing that the cemetery contained a large number of imported animals. This, combined with the cemetery’s location, the unusual pattern of deposition, and the fact that the horses were large for the period and predominantly male, suggests we might have been dealing with the burials of warhorses or tournament horses, feasibly from the nearby jousting site at Westminster Palace.
Elite animals
What about structures associated with these animals? Warhorses were rarefied creatures whose breeding, training, and care required bespoke facilities and infrastructure, although surviving archaeological evidence for these is largely ephemeral. Such is the low survival rate of medieval stables that these buildings have been described as an ‘archaeological black hole’. The earliest surviving examples are chiefly from the Tudor period, found at high-status sites like Hampton Court Palace and Kenilworth Castle, although some medieval stables are hiding in plain sight, as with an excavated structure at Launceston Castle, Cornwall, originally thought to be a hall but whose true purpose was identified by our project.
As for horse-breeding sites, the Warhorse project has had considerable success in mapping the network of royal studs across England and Wales that produced many of the finer mounts used in military campaigns through the Middle Ages. Equine officials’ accounts held at the National Archives (the equitium regis series) were combed for information about the locations, character, and physical appearance of these barely known but hugely important installations, and the landscapes in which they were situated. This research made it clear that medieval horse studs mainly consisted of compounds within enclosed deer parks, which were in turn characteristically located at the junction of landscapes of different character, such as wetlands, forests, or upland zones. Where all the various lines of evidence come together, we can build a comprehensive picture of these establishments, such as at Odiham, Hampshire, which was a busy, specialist horse centre dedicated to training colts, lying close to a royal castle. But while we can pin down the general location of medieval horse studs, physical evidence on the ground, such as earthworks, remains elusive.
As we have discussed above, the appearance of the classic medieval warhorse of the later Middle Ages, including its armour and equipment, can be established with a fair degree of certainty, and an archaeological approach to the subject has also allowed us to trace the ancestry and evolution of this animal. But what are the main ways in which the use of horses changed through the Middle Ages? Horse-breeding was well established in the Roman Empire, which had an extensive network of studs, a developed understanding of equine veterinary science, and utilised horses for various purposes, including in military roles. By contrast, early medieval Europe initially experienced a decline in organised horse-breeding, but the later Anglo-Saxon period marked a crucial shift in the use and perception of horses, with a revival of interest in Roman equestrian practices in areas like art, military tactics, and literature. In the pagan era (5th to 8th century), horses played a prominent role in rituals, including cremations and burials. However, between the 9th and mid-11th century, these animals became symbols of aristocracy and featured prominently in warfare, especially during the Viking wars. Organised studs were certainly established by the 10th century, and equestrian gear from this period reflects the elite’s appreciation of horses as status symbols in addition to their martial utility.
Military mounts
Historians often credit the Normans with battlefield dominance due to their effective use of shock cavalry, supported by a sophisticated culture of horse-breeding. Yet, in terms of equestrian material culture, the Norman Conquest has left a minimal signature. The introduction of the curb bit is the key indicator of change, testifying to differences in horsemanship rather than horse appearance, and there are tantalising signs that the period immediately after the Conquest saw a small but definite dip in horse stature, seemingly part of a pan-European trend rather than a particular effect of the Norman takeover.
We see long-term continuity as well in the landscapes that horses were bred and trained within either side of the Norman Conquest. If we are looking for a more profound ‘watershed moment’ in the appearance of warhorses, this came in the middle years of the 12th century, especially in the troubled reign of King Stephen. In these years, a boom in tournament culture and the rise of an aristocracy keen on power projection seems to have driven developments in equine apparel, and it is around this time that the largest Norman horses are found in the archaeological record. By the 13th and 14th centuries, the various categories of military and non-military horses were well established and clearly differentiated in material culture, and average horse stature crept up gradually, although changes in their size pale into insignificance relative to the Agricultural Revolution of the 17th century.
Throughout the medieval period, it is clear that exceptional care was taken to breed and train appropriate horses, including importing or obtaining by gift suitable animals from overseas to enrich the stock. While selective breeding in studs formed a key part of warhorse management, the famed destrier cannot be considered to be a ‘breed’. Rather, it represented a selection for appropriate temperament and body conformation, accompanied by intensive training that resulted in highly prestigious steeds that might be compared to modern-day supercars. While the ‘great horse’ was likely larger than the average medieval equid, metrical evidence from bones, horse armour, and other material culture confirms that raw size was considered less of a key trait than other physical and mental characteristics. Overall, therefore, archaeology shows how the romanticised image of the warhorse is not entirely reflective of historical reality. The finest medieval warhorses were more akin to polo ponies – strong, agile, and able to change direction rapidly, rather than physically massive.
Above all, the Warhorse project’s integrated approach has made a powerful statement about how barriers between traditionally discrete research fields within archaeology can be broken down to produce new understandings. By adopting this multidisciplinary model, our work has challenged traditional narratives about the medieval warhorse and its role in society: it is clear that archaeological evidence has the potential not just to fill gaps left by historical documents, but also to create new knowledge about these revolutionary animals.
Further reading:
• Oliver H Creighton, Robert Liddiard, Alan K Outram, Katherine Kanne, and Carly Ameen (eds) Medieval Warhorse: equestrian landscapes, material culture and zooarchaeology in Britain, AD 800-1550 (Liverpool University Press, £50, ISBN 978-1836243359).
• The project’s digital research materials are freely available through the Archaeology Data Service: https://archaeologydataservice.ac.uk/archives/view/1005150.
