In October 1997, an Anglo-Saxon burial ground was found at the Eriswell Cemetery in Suffolk.2 This site has been dated to approximately 550 ad and is notable for the horse buried next to what is assumed, because of the presence of the horse and other artefacts, to be a high-ranking nobleman.3 This assumption is undoubtedly correct.4 However, whether the horse communicated something more specific than prestige is less easy to determine. In fact, interpreting Anglo-Saxon horses, whether in burial sites or poetic texts, requires data from a vast range of disciplines if it is not to depend upon leaps of logic and faith; the horse buried whole and Hrothgar’s gift of mearas…æppelfealuwe ‘dappled dun steeds’ (Beowulf 2163b and 2165a) are clearly meaningful, but the code is difficult to untangle.5 This is not to say that the ‘Dark Ages’ have shrouded the Anglo-Saxon horse in impenetrable mystery. A great deal of useful work, drawing upon a wide range of sources including poetic texts like Beowulf, has been done to untangle the development of the horse through the Middle Ages.6 For the most part such work has used the horses in Beowulf to contribute to an understanding of horses in the ‘real’ world. In this investigation, however, I use the information available from other sources, including archaeology, law codes, wills, chronicles, glossaries, colour semantics and art history, to contribute to an understanding of Beowulf. Thus I first address the use of horses in Anglo-Saxon England, the types available and how they were acquired. This context provides the basis for a re-examination of the horses in Beowulf, which pays particular attention to the difficult colour-terms that describe the horses used as both transport and treasure within the poem. The details, and, especially, the omissions in the poet’s depiction of horses contrast strongly with the context provided in the first part of this discussion. The contrast is particularly striking with respect to the horses described in tenth and eleventh century wills. As a result, I suggest that Beowulf derives from a date not later than the tenth century and that the depiction of horses within the poem would have struck the eleventh century audience of the manuscript as extremely archaic.
THE HORSE IN RITUAL
Horses appear to have had a number of different functions in Anglo-Saxon England. Archaeological discoveries, for example, suggest that horses may have played some role in the Anglo-Saxons’ pre-Christian religious beliefs; there may even have been an Anglo-Saxon cult of the horse,7 perhaps containing practices similar to those of the continental Germanic tribes described by Tacitus. Tacitus suggests that the prominent role played by horses in augury was characteristic of these tribes:
…proprium gentis equorum quoque praesagia ac monitus experiri. publice aluntur isdem nemoribus ac lucis, candidi et nullo mortali opere contacti; quos pressos sacro curru sacerdos ac rex vel princeps civitatis comitantur hinnitusque ac fremitus observant. nec ulli auspicio maior fides, non solum apud plebem, sed apud proceres, apud sacerdotes; se enim ministros deorum, illos conscios putant. (De origine et situ Germanorum 10)8
There is, unfortunately, no evidence that the Anglo-Saxons actually practised augury in this way. In fact, the burial at Eriswell contradicts many of the elements mentioned by Tacitus: not only is there no sign of a ‘sacred chariot’ in the burial, but the equipment buried with the horse indicates that the animal had participated in ‘worldly labour’ — that is, it had carried a human burden. Thus, although the Anglo-Saxons might have maintained special horses untouched by human service for augury, modern archaeology has not found any evidence of them. At the same time, the horse burial that archaeologists can investigate cannot be considered ‘characteristic’ of the Anglo-Saxons: burials like the one at Eriswell do not appear to have been common, although a similar one, dated to the sixth or seventh century ad, was found under mound 17 at Sutton Hoo in 1991,9 and there are numerous instances, particularly at Spong Hill, of the cremation of horses, as well as other animals.10 It may thus be possible to assume some kind of ritual importance for horses in early Anglo-Saxon England, even if the extant evidence does not confirm (or, indeed, disprove) the practices described by Tacitus.
‘Some kind of ritual importance’ does not provide much insight into the Anglo-Saxon attitude toward horses, although it seems reasonable to infer first, that horses were valuable property intended to reflect the status of those with whom they were buried and second, that the buckets, food, equipment and horses buried at Eriswell and Sutton Hoo reflect a belief in the need for such things in an afterlife. We can accept both these assumptions as reasonably accurate albeit creative reconstructions,11 but we should remember that the Anglo-Saxons themselves have left no documentary evidence to confirm the details of such beliefs:12 burying food, weapons and horses may indeed indicate a belief in a warriors’ paradise like Valhalla,13 but they may also indicate something else. No explanation of the meaning of a buried horse, whether whole, cremated, in pieces or decapitated, has survived.14Beowulf, for example, provides no hint of a ritual importance for horses: none of the high-status burials described in the poem contains horses, and the Danes’ desperate resort to pagan practices does not include the augury of horses, despite the apparent appropriateness of consulting well-respected auspices as the leaders of the people
hwæt swiðferhðum selest wære
wið færgryrum to gefremmanne. (Beowulf 172b–4).15
In contrast with their potential but nebulous religious significance in the Anglo-Saxons’ pre-Christian history, horses maintained well-documented roles as treasures and means of transport throughout the Anglo-Saxon period. Yet they were, perhaps surprisingly, not crucial to Anglo-Saxon life. To a great extent, they were a luxury, not a necessity. Cows and oxen were more economical and useful than horses: horses were not used for ploughing,16 were not used for milking,17 were not a usual food source,18 cost more to buy19 and cost more to feed over the winter.20 The horse’s relative uselessness thus rendered it an important indicator of wealth and prestige, as the poor could not afford to maintain an animal which provided few essential benefits for such a great cost. For the elite, on the other hand, the horse became increasingly important, both as a marker of status and as a tactical component in war.
Horses could mark more than a single division between rich and poor, however; they could indicate multiple, very particular gradations of wealth and prestige. At some point in Anglo-Saxon England, a range of different kinds of horses arose, with different functions and different levels of prestige attached to them. Thus there are terms to describe horses fit only to pull carts (stot, crætehors),21 horses fit to carry luggage (ealfara, seamhors),22 horses fit to ride (hors, radhors, hengest), horses fit for breeding (stodmyre, gestedhors, stodhors), horses fit for kings and the nobility (friþhengest, steda, blanca, mearh, wicg) and perhaps even horses fit for war (eoh).23 Some of these terms distinguish an even more exalted type of horse: the noble accoutrement of heroic poetry, the blanca, mearh or wicg. Such terms do not mean that the mounts of heroes were always white (blanca) nor that they were always mares: the Old English word for ‘mare’ is miere, not mearh. Blanca, mearh and wicg are part of the special vocabulary restricted to (and thus constitutive of) the elevated art of poetry.24 Old Norse literature, too, recognises a hierarchy of horses, with some — blakkr (OE blanca) and jór (OE eoh), for example — appearing almost exclusively in poetry and indicating horses fit for heroes and gods.25 That there were living examples of distinct classes of horses underlying these vocabulary items is supported by an entry in The Laws of the Dunsæte, which notes different levels of compensation payable for a lost horse:
Hors man sceal gyldan mid XXX scillingum oððe be ðam ladian; myran mid XX scillingum oððe be ðam, & wintersteal ealswa; wilde weorf mid XII scillingum.... (Laws of the Dunsæte 7)26
The meanings of both wintersteal and weorf are uncertain. Wintersteal might suggest a horse kept in a stall over the winter, perhaps one of the equi in halla listed in Domesday Book,27 but ‘a year old stallion’ and ‘a yearling foal (?)’ have also been suggested.28Weorf may indicate an ass or a generic term for beasts of burden but here probably refers to untamed horses of low quality destined to be pack animals.29 Regardless of their specific meanings, these terms indicate distinctions and rankings which were in place by the tenth century.30 A century later, Domesday Book records a similar number of distinctions: horses, rounceys, mares, unbroken mares and forest mares.31 The fuller information provided by later texts makes the distinctions between the different classes of horses even clearer. These classes included ‘affers’ (cheap horses for pulling carts and harrowing), ‘sumpters’ (slightly more valuable animals used as pack-horses), ‘rounceys’ (ordinary riding horses), ‘palfreys’ (horses for long-distance travel) and ‘destriers’ (warhorses, too expensive even for most knights).32 HORSE BREEDING
The differences between these categories of horses were both obvious and difficult to achieve. Left to their own devices, horses will take care of themselves, nourish themselves as best they can and breed enough to sustain their population.33 There is evidence that the Anglo-Saxons allowed some of their horses to do precisely that: although some horses may have been kept in the vicinity of — if not housed in — the halla, Domesday Book also mentions equae silvestrae ‘forest-mares’ and equae indomitae ‘unbroken mares’, horses apparently allowed to roam and breed relatively freely.34 This method of raising horses was inexpensive, but the resulting population belonged to that category of horses fit only to pull carts — an ‘unimproved’ animal not much different from the basic wild horse. Such a horse stands about eleven hands (110 cm) high at the shoulder (approximately the size of a Shetland pony)35 and can be characterised by a basic colour: dun, which ranges from dull yellow to grey, often with a black mane and tail, a dark stripe down the back (the ‘eel-back’36) and zebra striping on the legs,37 as exemplified by the Tarpan and Przewalski’s Horse,38 types of horse thought to be closest to the original wild horse from which all domesticated breeds descend.39 No true wild horses remain outside captivity today.40 Although there may have been some genuine wild populations in the Anglo-Saxon period,41 most of the horses used by the Anglo-Saxons were probably not entirely ‘unimproved’ by human contact and interference in their breeding; the equae ‘mares’ may have roamed relatively freely, but stallions apparently did not. The populations of feral horses still surviving, for example, in the Camargue marshlands in France and in the New Forest and Exmoor in England provide some indication of the appearance of the Anglo-Saxons’ wildu weorf and equae silvestres:42 forest mares in thirteenth century Wales, for example, possessed the same dun colouring displayed today by the Exmoor ponies.43 It has been argued that the equae indomitae and equae silvestrae were, in fact, Exmoor ponies.44
Modern investigations into Przewalski’s Horse and current populations of feral horses can thus offer insights into some Anglo-Saxon horses. Horses of this type were useful and valuable. For example, they would be perfectly adequate to pull a cart, even a war-chariot; the Britons may have employed such horses to pull their chariots as they resisted the Roman invasion, although they apparently exploited cavalry as well.45 For a horse good enough to ride, however, more care would be required. The most obvious issue is size; a comfortable riding horse should be large enough that its rider’s feet do not drag on the ground and strong enough that it does not tire too quickly from carrying the rider’s weight. Luckily, greater size is not a difficult objective to achieve. Merely feeding pregnant mares and young foals, rather than letting them forage on their own, can make a difference. Prolonged care of this kind can result in significant increases in height after a few generations.46 More drastic differences — changes in conformation,47 for example — require selective breeding and may depend upon the importation of superior individuals from elsewhere.
Improving the size and quality of horses is thus relatively simple. The problem is that such improvements are extremely expensive and are as easily lost as gained. To keep control of prized traits like size, selected individuals must be fed, even though feeding a horse over the winter could cost more than buying one.48 It is also important that these selected animals do not breed indiscriminately. Such control requires the gelding of non-selected males, selective slaughter of unwanted individuals and, most importantly, fences, which require even greater investments of wealth.49 The history of horse breeding in Europe is a history of gradual improvements over long periods of time, which were all lost during times of economic crisis — times when people could not afford to keep the fences fixed and the horses fed.50 In such conditions, horses are notorious for reverting back to their basic, wild form and losing the distinctions previously bred into them.51 This process is well-documented in modern times by the development (or degradation) of the mustang of the American plains and the Brumby of Australia, two breeds which, after reverting to the wild, developed great hardiness but also deformities and a tendency toward the ‘primitive’ dun colour of the wild horse.52 A similar process can be observed in the Middle Ages in the fate of the Welsh Powys. In 1166, after infusions of stock from Spain, the Powys was a prestigious, much sought-after breed, purchased by kings like Henry II. In the late thirteenth century, however, probably as a result of frequent, indiscriminate breeding with feral mares, the breed was mostly remarkable for its predominantly dun colouring and the fact that its price was only thirty-seven percent as high as that of other breeds.53 That is, the traits acquired by careful breeding were lost, and the breed began to revert back to the basic characteristics of the wild horse.
The level of commitment to horse breeding thus largely determines the kinds of horses that are available at any time in a particular culture. This discussion of the issues that contributed to the creation and maintenance of the Anglo-Saxon horse should allow us now to examine in more detail the appearance and function of this animal before going on to consider what this information can contribute to an understanding of Beowulf.
THE STATE OF THE ANGLO-SAXON HORSE
Was the Anglo-Saxon horse a dun pony or a fine-bred steed? The answer, of course, is both. Throughout the Anglo-Saxon period there were horses like the wilde weorf in the forests and moorlands, horses left to roam freely, largely ‘unimproved’ by careful feeding or selective breeding, and so worthy of only twelve shillings in compensation, unlike the ‘horse’, ‘mare’ and wintersteal, which, as stated above, were worthy of compensation at the level of thirty, twenty and twenty shillings respectively. Such distinctions testify that some care was taken with more valuable animals by the tenth century, at least. In addition, at some point, horses like the one ridden by Harold in the Bayeux Tapestry appeared. The horse on which Harold rides to Bosham (Sussex) is depicted as being not only of a distinct colour in comparison with the two other English horses nearby; it also displays a shape different from the many Norman horses depicted in the tapestry. Sarah Larratt Keefer suggests that its arched neck, fine muzzle and short back indicates a horse derived from Arab blood.54 That is, this animal appears to be an expensive foreign import, or perhaps one bred from expensive imported stock.
It may be assumed that there were always horses of poor quality available in Anglo-Saxon England; in addition to the evidence of Domesday Book, the practice of leaving some animals to run wild and fend for themselves is referred to by Chaucer.55 The question is when the Anglo-Saxons began to control breeding and feeding strictly enough to create and preserve distinctions in quality in some horses and when they began to import quality horses from abroad.56 Archaeology might be able to provide some answers to these questions, but no large-scale study of Anglo-Saxon horse-remains has, to my knowledge, been undertaken thus far, and the two Anglo-Saxon horses that have received detailed scrutiny, those buried at Sutton Hoo and Eriswell, provide more warnings than certainty.57 Given the valuable artefacts accompanying the two burials, these animals could be expected to represent the highest quality horses available at their time. Unfortunately, their quality has proven difficult to evaluate. For example, estimates of the sizes of the horses buried at both Sutton Hoo and the Eriswell cemetery have varied significantly in the course of their investigation. The Eriswell horse, although originally described as measuring sixteen hands (160 cm) high, is now said to measure about fourteen hands (140 cm) high.58 The Sutton Hoo horse was apparently measured as being thirty centimetres smaller but has since been reassessed and ascribed the same height as the Eriswell horse — fourteen hands high.59 The difference is significant: a horse standing sixteen or even fourteen hands high could have been an individual produced by a careful feeding and breeding program; a horse standing eleven hands high fairly certainly was not. It would be helpful to know other characteristics as well — details such as the profile of the skull, which, as in the Bayeux Tapestry, can indicate oriental bloodlines — but, with the basic issue of size apparently uncertain, it seems hazardous to speculate more closely about the animals’ genotype.60 The currently accepted height of these two animals suggests the existence of prestige animals produced at some expense even in the seventh century, despite assertions by earlier critics that Anglo-Saxon horses before the Norman conquest remained consistently small (and thus presumably undeveloped).61 In fact, there are indications from historical sources that horses received considerable attention throughout the Anglo-Saxon period. For example, Keefer suggests that the story of Cynewulf and Cyneheard demonstrates the existence of a mounted royal guard by the mid-eighth century.62 The story of regicide occurs in The Anglo-Saxon Chronicle:
Ða on morgenne gehierdun þæt þæs cyninges þegnas þe him beæftan wærun þæt se cyning ofslægen wæs, þa ridon hie þider, & his aldormon Osric, & Wiferþ his þegn, & þa men þe he beæftan him læfde ær. (Anglo-Saxon Chronicle 755 ( = 757))63
Although it does not explicitly state that there was an official mounted guard, the Chronicle also does not suggest that the king’s men suddenly had to acquire horses for this unexpected crisis; the men þe he beæftan him læfde ær ‘men whom [the king] had left behind earlier’ appear already to possess horses suitable for hard riding. Such a mounted guard would require a certain amount of organised breeding to ensure that sufficient numbers of quality horses were available to transport a mobile escort for the king. This organised breeding may have been a royal endeavour or one undertaken individually by the nobles accompanying the king; later evidence (discussed below) would support either possibility. By the ninth century, such organised breeding was even more necessary, for the whole army was mounted — for transportation, if not for combat — as Alfred chased marauding, mounted Viking invaders across the country. For example, the Anglo-Saxon Chronicle notes that se cyning Ælfred æfter þam gehorsudan here mid fierde rad oþ Exanceaster ‘King Alfred rode after the mounted army with his army to the fortification at Exeter’ (ASC 877 ( = 876)). By the tenth century, the Anglo-Saxons possessed horses whose value was on a par with precious stones and priceless religious relics — according to William of Malmesbury, at least. William states that in 926 Hugh the Great, Dux Francorum, sought the hand of Athelstan’s sister with extravagant gifts, including foreign race-horses:
Is, cum in conuentu procerum apud Abbandunam proci postulata exposuisset, protulit munera sane amplissima, et quae cuiuslibet auarissimi cupiditatem incunctanter explerent: odores aromatum qualia numquam antea in Anglia uisa fuerant; honores gemmarum, presertim smaragdorum, in quorum uiriditate sol repercussus oculos astantium gratiosa luce animaret; equos cursores plurimos cum faleris, fuluum (ut Maro ait) mandentes sub dentibus aurum.... (Gesta Regum Anglorum II.135)64
Although not described first, these horses ornamented with classical learning as well as their own trappings precede a long catalogue of expensive gifts, including a piece of the cross of the crucifixion. They are neither an afterthought nor a trivial appetiser; they are gifts fit for a king. Such animals would not only adorn a fashionable king; they would prove invaluable to the improvement of native breeds.65 William of Malmesbury may not be an accurate witness to the state of the horses in Anglo-Saxon England, but these sources, taken together, suggest that there were horse-breeding programs throughout much of the Anglo-Saxon period. The state of such programs probably fluctuated. The existence of well-bred horses buried at Sutton Hoo and Eriswell, even if proven, could guarantee little about the state of horses in different times and places: the quality of Cynewulf’s thegns’ horses might not be comparable to the quality of Alfred’s army’s mounts, and Athelstan might not have possessed horses as good as Alfred’s, much less any comparable with those brought by Adulf. Like literacy, horse breeding undoubtedly waxed and waned in response to economic and political conditions, as it did on the continent and in later times.