The Furs

Since the origin of armorial bearings in the middle of the twelfth century, the furs ermine and vair have been featured in heraldry. Heraldically, the furs are patterns of two or (occasionally) more colours, but are treated more or less like solid colours. Because they are treated like regular colours, they are classed among the heraldic tinctures. The furs are originally based on the real-life patterns of medieval fur clothing.

Ermine is the popular name of a weasel species (mustela erminea). The wintertime pelt of the ermine is white right down to the tip of the tail, which is black. Historical ermine fur clothing would generally be white, with the black tails scattered around the surface. The complexity of ermine lay in the tails, because the shapes of ermine tails are by no means standardized in artistic representation. There are many variations on ermine spots recorded in art, some resembling the thin black point of a tail, others more fanciful and stylized. Ermine clothing is common in medieval and Renaissance art, especially in the lining of cloaks, hoods and other garments. It remains in active use today, most notably in the official robes of UK peers whose scarlet robes include an ermine mantle denoting rank.

Ermine as it appears in heraldic art is very similar to ermine in medieval art. Many of the same variations on the ermine spot were used in the past, but heraldic art has gradually settled on one basic stylized design, with only minor variations. Because all the variations are at least loosely based on the use of ermine in clothing, the design has remained recognizable in all its iterations.

The colours of heraldic ermine have not remained so traditional. There are three named variations to ermine: Ermines features white spots on a black field; Erminois has black spots on a gold field, and Pean consists of gold spots on a black field. The design strayed from its origins in fur clothing by introducing unnatural colours, but retained its essential patterning.

The other heraldic fur, vair, is not necessarily associated with a specific species, but is considered to be formed from the pelts of the Eurasian red squirrel, Sciurus vulgaris, whose winter coat turns blue-grey on the back and white on the belly in the colder parts of northern and central Europe. Because a squirrel pelt is tapered at the head and wider at the feet, the most efficient organization for squirrel pelts is side by side and head to foot. By alternating the back and belly pelts, the furrier makes a wavy or chequered pattern of blue and white.

In medieval art, and presumably in usage, vair was usually employed in the linings of mantles and cloaks. By the Renaissance, however, vair had virtually disappeared from non-heraldic art, and, unlike ermine, vair does not enjoy the modern patronage of the UK peerage.

Like ermine, heraldic depictions of vair have variations of pattern and colour. When of any colours other than blue and white, vair is termed vairé of the appropriate colours. In extremely rare cases, vair is fancifully made up of three or even four colours. Different arrangements of the pelts are called vair-en-point, vair in pale and counter-vair. When the pelts are stripped of their natural shape, and become essentially T-shaped, the patterns are called potent and counter-potent.

There is still more variation in vair, however, because of subtle changes in the depiction of pelts over time. In early depictions, the pelts were rounded, so creating a wavy appearance. Over the centuries a stylistic modification arose whereby the curves became angular and began to resemble shields with triangular ‘ears’. These two designs were used concurrently for centuries, until the older, curved version gradually fell out of use. Even today, both designs are recognized and considered correct depictions of vair. They are sometimes called vair ancient and vair modern by armorists.

The importance of fur in medieval art is not limited to its identification and recognition. The furs bear a powerful and readily apparent symbolism. Sumptuary laws restricted the use of fur, which therefore became an indicator of aristocratic rank. Because the finest furs were imported at great expense from Scandinavia and Russia, furs were also a symbol of wealth, and because of their distinctive patterns, furs were an effective means of signalling a subject’s wealth and power in a work of art. In iconographic terms, this is probably the essential symbolism of fur in medieval art.

Proto-Vair In French Manuscripts

The artistic use of vair seems to have arisen in France around the beginning of the twelfth century, in manuscript art. The earliest depictions of vair differ from later medieval conventions. Rather than the two symmetrical lines of blue and white waves of vair ancient, the pattern consists of a blue field with regularly arranged white figures: regular horizontal lines and deeply scalloped curves that begin at the horizontal line above and reach their nadir at a point of tangency with the horizontal line below. Between the descending panels are visible triangles of material with shallowly scalloped upper angles. The larger descending panels are white, and the smaller, triangular underpanels are in blue. To distinguish this early vair pattern from the roughly standardized vair ancient and vair modern, I have termed this pattern ‘proto-vair’.

A pair of early illustrations of fur-lined mantles or cloaks may be found in a French copy of Gregory I’s Moralia in Iob dating from the 1120s.Footnote 1 One historiated initial, coming at the beginning of Book xxv, shows a long-haired rider astride a horse, with a falcon perched on his gloved right hand and his left hand holding up his long cloak to display a lining of vair (fig 1).Footnote 2 The other initial, placed at the top of Book xix, shows a knight slaying a dragon (fig 2). The cloak is clearly displayed and lined in the same pattern.Footnote 3 Vair was usually depicted as a cloak lining in medieval art, so the Moralia in Iob illustrations represent vair in its usual context.

Fig 1 Knight with a falcon and vair-lined cloak. Photograph: © Bibliothèque municipale de Dijon, ms 173, fol 174; coll E Juvin

Fig 2 Knight with a vair-lined cloak slaying a dragon. Photograph: © Bibliothèque municipale de Dijon, ms 173, fol 20; coll E Juvin

The earliest clearly identifiable use of a vair pattern in medieval art is perhaps a dozen years older, and came from the same scriptorium. The Bible of Cîteaux (also known as the Bible of Stephen Harding, or the Bible d’Étienne Harding) dates from about 1109–11.Footnote 4 Its illustrations have some stylistic similarities to the Bayeux Tapestry, especially in its representations of armoured soldiers and grotesque figures. The manuscript was made at the abbey of Cîteaux in Burgundy, the founding abbey of the Cistercian Order. Both the Bible of Cîteaux and the Moralia in Iob manuscript mentioned above were made under the direction of Stephen Harding, the monastery’s third abbot. Both manuscripts are believed to have been illustrated by the same workshop, and both manuscripts show proto-vair illustrated in the same way.Footnote 5 Proto-vair appears five times in the Bible of Cîteaux, in two different contexts. In the first two instances, it is subtly, but unmistakably, used in the linings of mantles. In the other three, an entire tunic is formed from proto-vair.

The first example is a full-page illustration featuring King David with his harp, seated on a throne (fig 3).Footnote 6 There is a framework around King David in the form of an embattled wall, manned by a garrison of knights very reminiscent of the Norman knights in the Bayeux Tapestry. The soldiers wear the helmets of Norman knights, and wield lances, axes, gonafons, shields and bows all identical to those seen in the Bayeux Tapestry. Within the framework, King David wears a long woollen tunic. On his shoulders he wears a mantle draped over his left arm. A close look at his mantle where it hangs behind his right arm, and again where it is turned up over his left knee, reveals that his mantle is lined with small fur pelts.

Fig 3 David enthroned, wearing a vair-lined cloak. Photograph: © Bibliothèque municipale de Dijon, ms 14, fol 13v; coll E Juvin

The second illustration is at the beginning of the Book of Daniel (fig 4).Footnote 7 Shadrach, Meshach and Abed-Nego stand in the furnace protected by an angel. A servant feeds the fire, and three more servants lie engulfed in flames at the base of the furnace. Nebuchadnezzar and two associates look on. Nebuchadnezzar is wearing an open crown and the usual loose tunic, with a mantle over the left shoulder, pinned on the right shoulder. The outside of the mantle is decorated with what appears to be a floral pattern, and is lined inside with proto-vair like the mantle of King David. In her catalogue of Dijon manuscripts, Manuscrits Enluminés de Dijon, Yolanta Załuska describes Nebuchadnezzar as ‘debout, couronné, portant chlamyde en étoffe brodée, doublée d’hermine’.Footnote 8 Rather than ermine (hermine), however, the pattern is vair.

Fig 4 Nebuchadnezzar wearing a vair-lined cloak. Photograph: © Bibliothèque municipale de Dijon, ms 14, fol 64; coll E Juvin

The third Cîteaux illustration is a decorated initial showing King Antiochus IV Epiphanes of Syria, who incited the Maccabean Revolt, seated on his throne overseeing a massacre of the Jews (fig 5).Footnote 9 Antiochus, who is seated facing forward, wears a long-sleeved, ankle-length tunic, curled and pointed shoes and a cap. There is a slit centred in the front of the tunic, from the knees down, and another V-slit at the front of the neck. The neckline and cuffs are coloured bright red. The outer surface of the tunic is made of contrasting-coloured pelts, in a pattern identical to the mantle linings of King David and Nebuchadnezzar.

Fig 5 Antiochus IV in a vair tunic. Photograph: © Bibliothèque municipale de Dijon, ms 14, fol 191; coll E Juvin

The design of Antiochus’ tunic is closely echoed in a historiated initial showing Solomon at his anointing as King of Israel (fig 6).Footnote 10 Solomon is seated, and is being anointed from a ram’s horn by Zadok and Nathan. Solomon wears a long, loose tunic of proto-vair. With the exception of the location of the slit at the bottom of the tunic, and white cuffs rather than red, it is identical in pattern to that of Antiochus.

Fig 6 Solomon being anointed, in a vair tunic. Photograph: © Bibliothèque municipale de Dijon, ms 14, fol 44v; coll E Juvin

The final example of proto-vair in the Bible of Cîteaux is on a page illustrating the life of King David in extended cartoon form (fig 7).Footnote 11 At the top of the page Goliath is dressed as a Norman soldier, with a chain-mail hauberk, a conical helmet, a spear and a pointed Norman shield. He is so large that he extends above and below the margins of the narrative. Near the bottom of the same page, David is anointed King of Israel at Hebron. He is seated, sporting a crown on his head and a sword in his right hand. He wears a mantle of light green, draped over his left arm. The lining of the mantle cannot be seen, but under it he wears a long tunic of the same pattern as those of Antiochus IV and Solomon, in white and blue proto-vair. The cuffs of the tunic are again trimmed in red, and like most of the figures on fol 13r, David wears red hose under his tunic. Załuska describes David at Hebron, in the scene from fol 13r, as ‘vêtu d’une tunique d’hermine et d’une chlamyde [mantle]’.Footnote 12 Załuska was right to recognize the pattern as a fur, but again mistakenly identifies vair as ermine.

Fig 7 David in a vair tunic, enthroned at Hebron. Photograph: © Bibliothèque municipale de Dijon, ms 14, fol 13r; coll E Juvin

The presence of the proto-vair pattern – depicted as both a traditional cloak lining and as a tunic material in the same manuscript – leaves no doubt that the Bible of Cîteaux represents tunics whose outer surface is composed entirely of fur. However, the absence of this fashion in later medieval art raises the question of whether fur tunics in this style ever existed in common use, or whether they were merely an invention of the Cîteaux scriptorium.

A manuscript in Tours suggests the former. An illustration in an early twelfth-century copy of the Tractatus in Evangelium Johannis shows a clerical scribe at work on a manuscript with knife and pen.Footnote 13 Behind him stands another tonsured cleric, observing the scribe’s work. The standing cleric wears a tunic composed of vair, but the tunic differs dramatically from the Cîteaux illustrations. It is looser, and the vair pattern follows an artistic convention more familiar from later manuscript art. The almost shield-shaped white pelts are spaced further apart than in the Cîteaux manuscripts, and there is blue visible all around them. If the vair tunic had been a scribe’s invention, then the idea must have been transmitted a great distance, without an attendant transmission of the stylized vair pattern. More likely, vair tunics were a real, if understandably rare, article of clothing in early twelfth-century France.

The Bayeux Tapestry: Nascent Vair

Three times in the Bayeux Tapestry, Norman notables are depicted wearing an odd garment of contrasting light and dark colours in a roughly chequered pattern. This garment has been a mystery of the Tapestry ever since it was subjected to scholarly study, and has been described as possibly a gambeson, or scaled leather armour, or as some unknown form of noble dress.Footnote 14

The unexplained Norman garment is seen on three occasions in the Bayeux Tapestry. In the first, Guy, Count of Ponthieu (c 1048–1101), is receiving messengers from William, Duke of Normandy. In the second, William, Duke of Normandy (1027–87), is wearing a very similar garment as he rides past the abbey of Mont-Saint-Michel on a military expedition. In the third example, in the midst of the Battle of Hastings, William’s half-brother Odo, Bishop of Bayeux (c 1032–97), wears a garment that is again similar to the previous two. None of the three garments are depicted identically, but they are clearly all of a type.

In 1064, Harold Godwinson, Earl of Wessex, was taken captive by Guy, Count of Ponthieu, in Beaurain. The Duke of Normandy took a particular interest in Harold’s situation and sent messengers to secure his release. This meeting between Guy and the messengers of William is depicted in the Bayeux Tapestry (fig 8). Guy stands with his right hand on his hip, his left hand holding a single-bladed battle axe just slightly shorter than himself. He wears a past-knee-length garment that seems to be made of scales in alternating horizontal rows of terracotta and pale green, edged in green thread. The scales are rounded at the bottom, and each row obscures the top of the row directly beneath. The bottom hem of the garment is a band of terracotta. On his legs are horizontal bands of colour, probably garters. On his shoulders he wears an unlined woollen mantle, draped over his left arm. Rather than continuing the chequered pattern, his right arm is embroidered in the same green that outlines the scales of his garment.

Fig 8 Guy, Count of Ponthieu, receives messengers from Duke William: detail from the Bayeux Tapestry. Photograph: courtesy of the City of Bayeux

After the Duke of Normandy secured his release, Harold Godwinson joined him in a campaign against Conan, Duke of Brittany. In the Tapestry, Harold and William are depicted passing the abbey of Mont-Saint-Michel and approaching the treacherous shores of the River Couesnon (fig 9). Just to the right of the abbey we can identify William astride a horse and flanked by retainers, some in chain mail and some unarmoured.Footnote 15 William carries what appears to be a baton of office or mace, and wears a garment similar to that previously worn by Count Guy. It is green at the collar, has long sleeves and comes to his knee. It is different from Guy’s in that the colours are arranged in an irregular checkerboard pattern of terracotta and dull yellow, rather than in horizontally arranged scales. William does not wear a mantle, nor do any of his soldiers.

Fig 9 Duke William (or Odo) rides by Mont-Saint-Michel: detail from the Bayeux Tapestry. Photograph: courtesy of the City of Bayeux

Near the Tapestry’s end are two great panoramic views of the Battle of Hastings. Norman cavalry charge at the English shield wall and this time Harold and William command the opposing armies. In the second battle panorama, on the right side of the shield wall, is one of the most celebrated scenes in the Tapestry (fig 10). The mounted Norman knights have been repulsed by the English shield wall and the Norman commanders struggle to rally their troops. William raises his helmet to show his demoralized soldiers that he still lives, and Odo, Bishop of Bayeux, waves what may be a baton of office or a mace and encourages the fleeing men.Footnote 16 Odo wears a Norman helmet tipped with a small golden ball and a scaled garment, with no mantle. The scales are worked in black or very dark blue and a terracotta colour that matches nearby horses. The scales are nearly triangular in this garment, though some are slightly curved, and alternate in a chequered pattern like William’s scales at Mont-Saint-Michel, only smaller and more regularly arranged. The tunic descends just below the knee and the sleeves come to mid-forearm where they end in pale yellow cuffs. Any collar is concealed behind Odo’s upraised arm. The warrior-bishop is clearly wearing chain mail under his chequered garment. The arms of a mail shirt are visible at his wrists, and a mail coif covers his head and extends below the collar of the tunic. Citing an example from the biography of William Marshall, Earl of Pembroke, Legge claims that Odo is not wearing a full-length hauberk and that the chequered garment marks him out as a non-combatant.Footnote 17 It is possible that the chequered garment is a sign of a non-combatant, but it would be impossible for Legge to know the length of Odo’s hauberk under a garment that clearly covers mail at the neck and wrists and which goes all the way to his boots.

Fig 10 Odo at Hastings: detail from the Bayeux Tapestry. Photograph: courtesy of the City of Bayeux

Of all the scholars who have written on the Bayeux Tapestry, few have attempted to explain the chequered Norman garments. Most recently, Michael Lewis classed the chequered tunics as styles of armour, and notes especially that the odd garments differentiate the important characters from those surrounding them.Footnote 18 Lewis also recognizes the structural difference between Guy’s scaled tunic and William and Odo’s more triangular patterns. Accepting that they represent armour, Lewis conjectures that the scaled/plated armour was more expensive to create, and is therefore a marker of wealth and status in the Tapestry.Footnote 19

Planché and Legge both conclude that the ‘garments so chequered’ are gambesons.Footnote 20 A gambeson was a quilted, padded garment worn under the hauberk, with the dual purpose of softening blows and making the mail more comfortable to wear. But while a diagonally quilted gambeson would not be out of the question, the multiple colours of the Bayeux garments would be needlessly complex. A gambeson was composed of two layers of tough fabric, with softer layers in between as padding, all sewn together (quilted) to prevent the layers from shifting. To make a multi-coloured gambeson, one would have to sew on extra patches to add colour, a needless affectation and expense for an item of clothing normally hidden under armour, where it would inevitably attract sweat and blood stains. Furthermore, if a gambeson is meant to be worn under the hauberk, why would Odo wear one outside his mail coat? William, Odo and Guy are the three most prominent Frenchmen identified in the Tapestry. Since they are the only ones depicted in the chequered garment, it seems likely, as Lewis has argued, that the garment is a marker of status. A gambeson, even a colourful one, could not be a marker of status. It was a basic part of the armour, and any professional soldier would have had one.

It has also been suggested that the chequered garments represent armour formed of leather scales, or even lamellar armour. Such armour might have offered greater comfort and economy than heavy chain mail, although at the cost of decreased protection. But if the chequered garments represent leather armour, why would they be worn over chain mail? It is conceivable that scaled leather armour would have offered increased protection over mail alone, but if it were an effective supplement then we should expect the practice to have been more widespread. Other knights would also have made the relatively minor investment in supplemental scaled-leather armour. Also, while leather could certainly have been obtained in brown and grey (although not pale green), there would have been no reason beyond the purely decorative to use scales of alternating colours. In fact, more complicated patterns would have been possible.

Recognizing fundamental similarities in pattern, Wolfgang Grape cited the proto-vair garments in the Bible of Cîteaux as evidence that the Bayeux garments were not an invention of the Tapestry’s designer.Footnote 21 However, Grape did not recognize the Cîteaux pattern for what it was: an early version of vair. Because of this, Grape does not venture an explanation of what the garments represent, but does suggest, based on their twin appearances in the Bayeux Tapestry and the Bible of Cîteaux, that they were a fashion among European nobles in the second half of the eleventh century, but were unknown in Britain. Grape describes the garment as ‘horizontal and diagonal lines [creating] a triangular pattern in two colours’.Footnote 22

What Grape, Planché, Legge and Lewis all miss is that the diagonal lines on the Bayeux garments are not straight but rather are subtly curved. We can expect a hauberk to be quilted with straight stitches, while lamellar armour is formed of rectangular scales. The Bayeux Tapestry garments are scalloped, much like the pelts of the Cîteaux illustrations.

A Comparison Of The Bayeux And Cîteaux Patterns

A close examination of Odo’s garment at the Battle of Hastings shows that the shape of its scales closely matches the proto-vair of the Bible of Cîteaux. Duke William’s chequered garment is more irregularly embroidered; the pelts are not precisely arranged, but the garment is noticeably similar to Odo’s, and Lewis classifies them together.Footnote 23 Guy’s garment is different from the other two. It shows only descending scales, with no scalloped ascending panels. The colours are arranged into horizontal rows instead of chequered. Horizontal rows are not unheard of in later heraldic depictions of fur, but a fur tunic without ascending panels would be a less efficient use of pelts. Certainly the three garments are more similar to each other than to anything else in the Tapestry, and each garment has a structure consistent with squirrel fur.

It is possible that Guy’s garment is intended to represent another form of dress entirely, but it is more likely that the difference may be ascribed to different embroiderers or embroidery workshops producing and adding their subtle graphic reinterpretations to the various scenes.Footnote 24 Chain mail in the Tapestry shows similar variety in design, sometimes formed of tiny circles and other times embroidered in a straight crosshatch pattern, so the differences between the three Bayeux garments may represent a similar phenomenon.Footnote 25 If the Tapestry is the product of an English workshop, and the fur tunics had been unknown in England prior to the Conquest, then it is also possible that the designer and embroiderers were working from a limited knowledge of fur tunics in practice.

Putting pattern aside, the cut of all three Bayeux garments corresponds to the long proto-vair tunics worn in the French manuscripts. The Cîteaux tunics come to the ankle, the Tractatus tunic is knee length and the Bayeux tunics come just below the knee. William’s garment has a prominent dark-coloured collar. Guy’s collar is obscured by his mantle and Odo’s is obscured by his arm. King David’s collar in Hebron is obscured by his mantle, but the collars of Antiochus and Solomon are visible and prominent. The only difference between the visible collars in the Tapestry and the Cîteaux Bible is that the visible Cîteaux garments, like most woollen garments in the Bible, have V-necks.

The most important difference between the Bayeux garments and the Cîteaux garments – and indeed all later depictions of vair – is the colour. The Bayeux Tapestry garments do not conform to the usual blue and white colouring for vair, but the colour difference should not be taken as a disproof. The Bayeux Tapestry may pre-date the earliest clear depictions of a vair pattern by years or decades. It is no surprise that the earliest depictions of vair do not conform to the later, settled conventions regarding its use. Even if the designer had been aiming for blue-and-white vair, the Bayeux Tapestry was embroidered with a limited colour palette, which could only approximate white by leaving the surface bare and unembroidered.Footnote 26 The Tapestry does use some variations on blue, but not in any shades that would conform to later artistic practice for depicting vair. The colours in the Bayeux Tapestry are often fanciful, with blue and sea-green horses.Footnote 27 It is therefore no surprise to find the fur garments in the Tapestry picked out in black or dark blue, terracotta, buff and pale sea-green. Since horses are shown in the same colours, the colours may be interpreted as representing natural hues, suitable for genuine squirrel pelts.

Given the fanciful use of colour in the Bayeux Tapestry, not to mention the stylized colours conventionally used in later medieval depictions of vair, the chequered pattern is more telling than the colours involved, because it is an inherent result of the efficient use of fur from small mammals. When lighter front pelts and darker back pelts are sewn together, the result is necessarily chequered. If the Bayeux garments had been gambesons, or scaled leather armour, the chequering would be purely decorative. If they represent fur, then the chequering is inherent to the material.

Designer Intent

Two of the three vair tunics in the Bible of Cîteaux are worn by enthroned kings and the third is a coronation scene. Certainly this confirms Grape’s argument that the tunics were an expensive status symbol among the French nobility. The later widespread use of vair in cloak linings suggests that the material was not strictly restricted to enthroned kings, and the illustration from the Tours Tractatus in Evangelium Johannis shows that at least one scriptorium thought vair tunics were appropriate for senior clerics.

None of the chequered tunics in the Bayeux Tapestry are worn during enthronement scenes, or by kings. Count Guy wears his in what appears to be an informal meeting with William’s messengers, rather than a formal audience. In that scene, Guy is holding an axe but his attendant and William’s messengers, while armed with lance and sword, are shown unarmoured. In the Mont-Saint-Michel scene, William wears his tunic with no visible mail or helmet during an expedition against Brittany. Most of his accompanying knights are unarmoured, but all are armed; most carry shields and two men directly behind him ride in mail. The strangest setting for a fur tunic is Odo’s appearance at Hastings. The enthronement scenes of the Cîteaux manuscripts have little in common with the Tapestry’s battle scenes. The mail at his cuffs reveals that, at the least, Odo was not unarmoured, but given the cost of such a fur tunic, why would Odo expose it to the danger of a battle?

Conceivably, a fur tunic could have served to identify Odo on the battlefield, or even announce his non-combatant status.Footnote 28 However, the Bayeux Tapestry is not a photographic representation, and there is no need to conclude that Guy, William or Odo wore any such garment at the moments depicted in the Tapestry, or indeed ever. The garments are used primarily to distinguish their wearers in the pictorial narrative and, if they truly represent vair tunics, to identify their wearers as men of great wealth and status. It was not necessary that such garments were worn – only that they were meaningful to the designer and the Tapestry’s intended audience. Lewis argued that the Tapestry’s designer was familiar with lay clothing fashions and that the depiction of clothing was largely drawn from real life rather than artistic convention.Footnote 29 It seems most likely that the designers of both the Cîteaux and Tours manuscripts and the Bayeux Tapestry were artistically rendering real tunics they had seen, rather than following previous conventions.