As it was in the beginning
by T.R.Davies
Coat of Arms No.109, Spring 1979

We know from classical Greek painted pottery that the Corinthians and Athenians used shields with different designs, apparently painted on them, representing eagles, boars, swans, lions, chimerae, etc., in the sixth to fourth centuries BC. But there is no evidence that many of the other hundreds of city-states used any device on their shields. For example, in the case of Sparta, the only state with a permanent regular army, all artistic remains portray plain shields, as do those of Thebes. By Hellenistic times Athenian shields as well as Macedonian ones are invariably plain (figs. 1, 2, 3).

Roman shields are depicted on hundreds of monuments, often unadorned, but frequently all the soldiers are shown with the shields of one design, perhaps the thunderbolt of Jupiter, which suggests possibly a legion-sign. Such designs appear to be cut-out of thin metal and rivetted onto the field (fig. 4).


Fig. 1. Athenian and Corinthian shields, 5th. C. B.C. shown on pottery.
Fig. 2. Spartan shield - figurines and reliefs.
Fig. 3. Theban shield -statuettes and coins.
Fig. 4. Roman types - monuments.
Fig. 5. Viking oak-plank shield – mound burials.
Fig. 6. Norman shields - Bayeux Tapestry.
Fig. 7. Three of the Naumburg shields.
Fig. 8. Shield of William de Valence, Westminster.

We have tangible evidence that the Norsemen painted their shields. When the Viking longship was dug out of the burial mould at Gogstad on the shores of the Oslo Fjord in 1880, thirty-three round oak shields hung on each side, painted alternately all yellow and all black. Other Viking burial sites have yielded round shields made of five or six planks each plank painted alternately black and yellow, giving a barry effect, with a bronze boss in the centre, while the runic stone at Ronninge, Denmark, is inscribed as carved by 'Sote son of Asgot with the Red Shield' (fig. 5).

About 170 years later, their descendants, the Normans, are shown on the Bayeux Tapestry with kite-shaped shields each painted in two colours and bearing mainly geometric or symmetric curvilinear designs, though the troops of Count Gui bear shields charged with mythological creatures (fig. 6).

Perhaps the finest set of mediaeval statues extant adorns the Stifterchor of Naumburg Cathedral, representing those nobles and their wives who had been generous to that diocese during the eleventh century. Eleven out of the original twelve figures executed by the unknown 'Meister von Naumburg' remain and, though they date from the rebuilding of the choir in 1249, the dress and adornment of each conform in every detail to the eleventh century when the subjects were alive, having possibly been accurately copied from tomb effigies which have not survived. This series is usually admired for the lifelike faces, the expressions and the delineation of character, but our interest is in the equipment of the men. These belonged to the Wettiner and Askanier families and their relatives by marriage (two of the great families under the Hohenstaufen Emperors) and the seven male subjects died between 1032 and c.1110. All are portrayed in full length civilian clothes, four are bareheaded and three wear small close-fitting caps, but all carry sheathed swords and large kite-shaped shields. It will be noted that the period covered falls in the last century preceding the rise of heraldry and the shields are therefore of paramount interest. They are approximately three to four feet in height, half of that broad, and apparently bordered by a three-inch metal strip. Two of the shields - those of Hermann von Meiszen †1032 and of Dietrich von Brehna † c. 1110, have plain borders, while the other five are inscribed in Latin with the owner's name, title and his outstanding gift to the Church. Dietrich's field is now plain, but viewed from a certain angle an outline of a female figure can be faintly discerned - perhaps the Virgin. The others all show some object probably in thin metal, to strengthen the shield. Dietmar von Billing bears a narrow saltire couped with its extremities beaten into fleur-de-lys, the Wettiner Wilhelm von Kystritz has an upright branch with three side branches each having side twigs, the extremities of each being beaten out to form round 'fruit'. Ekkehard II von Meiszen and Sizzo von Kefernburg show designs partly obliterated, but the former may have shown a thin wrought-iron lion rampant upright in the centre with straight bars to the dexter for legs and a spiral to sinister for a tail, while both Sizzo and Hermann mentioned above appear to have had two different designs of curved scrollwork.

The significance of these shields lies in the fact that they were sculptured a century after the rise of heraldry and the noble families portrayed were still flourishing. It would have been easy for the gifted master to have given the Askanier males the argent with an eagle gules dimidiated with barry sable and Or for that house and the Wettiners fields or with a lion rampant sable. But the sculptor must have preferred to use the evidence of their tombs for the shields he showed, which accords with his accuracy of jewellery, swords, clasps, buttons and other details. Here we have a period when a knight chose whatever he wished to bear on his shield. Hermann and Ekkehard were brothers yet they used different designs: Dietrich and Wilhelm were brothers with contrasting shields (fig. 7).

The pre-heraldic mediaeval shield was of oak, hickory or other tough wood, the planks forming it being laid vertically usually and frequently made in a curved plane with the edge protected with iron. Experience showed that the top was more liable to hacking and damage than the lower part, while the dexter side - which was moved quickly across the body in combat -tended to take more blows than the sinister. We know from contemporary accounts that on occasions a great blow with an axe sheared a shield in half. It is evident that the experience of an individual knight would lead him to require strengthening of his shield by the addition of extra planks. Some would require the addition across the upper half, others would want the dexter side double the thickness of the sinister. Thus arose the chief, per fess and per pale, to be followed later by the fess, bend, pale, chevron, saltire and cross throughout. The latter is almost certainly due to shield-strengthening, while the cross couped, a later development, is the charge associated with the Faith.

Subsequently the need to reduce weight would give rise to the replacing of one wide plank with a number of narrower wooden rails or iron strips — producing barry, paly, bendy, etc. An example of this type of strengthening gave rise to the Lusignan arms shown on the tomb of William de Valence, †1294 at Westminster, depicted in two examples as either barry of twenty-eight or eighteen argent and azure with an orle of martlets gules (fig. 8).

Other forms of strengthening, such as the boss and eight starlike rays of iron, continued into the heraldic period as the escarbuncle, which by alternating the tinctures of the segments between produced gironny, while the trellis of hoop-iron, often employed in early Frankish shields, gave rise to fretty and lozengy, with their variants, in the twelfth century. It is important to appreciate that the so-called ordinaries existed on shields before armory arose and the innovation was to paint the divisions different colours.

In this article the word 'ordinaries' is used to its widest sense to embrace shields having only a) partitions, b) ordinaries and sub-ordinaries c) all their diminutives and d) escarbuncles and trellis which added other divisions to the field. If a shield has a design based alone on strengthening devices it is classed among the Ordinaries, but if one invented charge is added, say a mullet or a roundel, it is counted as a charged shield.

When and Where. The point and locality where heraldry came into being has never been established: the factors associated with its early years are mainly negative and concrete evidence only emerges towards the close of the second quarter of the twelfth century. When the scattered scraps are assembled on a date-basis the following pattern emerges :-

1. The Bayeux Tapestry of 1068 shows designs in colour but not of heraldic form.
2. The First Crusade of 1098 has so far failed to produce any evidence of the use of heraldry.
3. Between 1140 and 1159 heraldic shields are depicted on seals in eight West European countries: this suggests a somewhat earlier use on actual shields. The enamel plaque of Geoffroi d'Anjou at Le Mans, made in 1151, is the oldest heraldry in colour.
4. In spite of this however, no proof of the use of heraldry has been found associated with the Second Crusade of 1147.
5. No seals relating to the feudal states of the Latin Orient display heraldic shields until after 1160.

The alternatives, based on the above inadequate evidence, are either:
a) heraldry arose in Western Europe and spread rapidly from country to country, according to the supporters of this theory, by international tournaments: or b) it arose in the Holy Land during one of the Crusades, when knights from all countries gathered together and could exchange ideas, which accounted for the adoption of armory in a single generation throughout so many countries.
The first theory is supported by earlier evidence in the West: the second would more easily account for the international uniformity of the basic rules from the very inception of heraldry, which a short visit of a team of foreign knights at a tourney is unlikely to have produced.
Against a) is the extraordinary similarity of heraldic forms in every country from the beginning -the ordinaries, the attitudes of animal-charges: the common use of the same inanimate charges -mullets, billets, crescents, annulets, cinquefoils, etc., the tincture rule. Against b) stands the absence of any concrete evidence of heraldry in use in Outremer for twenty years after its appearance is known in the West.
It is generally accepted that the first essential of heraldry is colour -a different knight to one bearing a gold lion on a blue shield the first step had been achieved. The other necessary features were 1) the consistent use of the adopted design 2) the hereditary use of that shield by succeeding generations of the family and 3) the acceptance of recognised charges in heraldic form and constant attitude. All these features existed by the second half of the twelfth century.
The only surviving evidence consists of the earliest armorial seals, some monuments and the oldest rolls -heraldry was then simple, basic and essential, without the later complications of marshalling, etc.

Seals. There are no rolls relating to arms of the twelfth century and the main evidence for this period is in the form of seals. It must be emphasised that seals in themselves are not heraldry since they do not indicate tinctures; but they confirm the existence of heraldry and sometimes furnish proof that a shield found in rolls of the succeeding century had already been in use by forebears some generations earlier.

In considering the seals of this early period in Europe, there are several obstacles: a) comparatively few survive, b) those extant relate to great barons who granted manors, lands and charters to feudal tenants, c) between 1150 and 1200 not all baronial seals depict armorial designs, and d) as virtually no armorial seal survives of a knight bachelor or an esquire, it is not possible to obtain a proper cross-section of the characteristics of shields of that period. Of those that do survive from the twelfth century, unrepresentative though they may be, ordinaries alone appear on 48%, while 52% bear charges; of the latter over half (28%) are lions and 8% are single-headed eagles. Apart from these the other charges consist of garbs, fleur-de-lys, lucies and a single heraldic panther, so the rate of creative invention appears somewhat slow.

Rolls. It is proposed to examine the earliest representative roll of arms of England, France and the German states and attempt to analyse its contents in an endeavour to trace developments. Matthew Paris' shields have been ignored as not representative of English armory since they consist of the arms of foreign sovereign-rulers with those of the chief barons of England mentioned in the Monk's chronicles: instead Walford's and Charles' Rolls of the late thirteenth century are preferred. Of early French rolls, two have been redated more than a generation later than first attributed and another (Compiègne) is strongly suspected of being a later forgery, leaving the Bigot Roll, c. 1254, as the undoubted earliest. Similarly, for the German states, the mythological arms that illuminate Eneide des Heinrìchs von Veldeke have been passed over in favour of Grosse Heidelbergen Liederhandschrift, c. 1300, whose arms are drawn from the area of Thuringia to South Tyrol, and from between Alsace and Styria. The tabulation of these four rolls gives some interesting figures:-


Roll No. of Arms Ordinaries Charges
Bigot, c. 1254 Fr. 295 133 or 46% 155 or 54%
Walford's c. 1280 Eng. 101 (excl. 78 foreign arms) 49 or 49% 52 or 51%
Charles' c. 1295, Eng. 486 214 or 44% 269 or 56%
Gr. Heid c. 1300 Ger. 135 40 or 30% 94 or 70%

These rolls cover a period of fifty years stretching from a century to 150 years after the origin of heraldry: it is remarkable that when we compare the proportion of ordinaries to charges shown in the seals for the first fifty years of heraldry with the French and English rolls above,there is virtually no change. Only the German roll at the end of the period discloses a big increase in charges, suggesting that heraldry was developing faster and adopting new changes more freely, a trend which was to continue more rapidly in the German states than elsewhere in the following centuries.

As some of the rolls include one or two blanks, or damaged shields, or incomplete blazons, it will be noted that there are slight discrepancies between the number of entries and the proportion of ordinaries to charges.

The Bigot Roll, approximately a century after the rise of armory, shows little change in the percentage of charges but more variety. The charges consist of sixty-five lions, eight eagles, two double-headed eagles, sixteen martlets, ten mullets, eight fleurs, six roundels, five each of escallops and annulets, four each of escutcheons, crescents and fers de moline, three each of roses and étriers (stirrups), two each of croissettes and leaves and one example bearing a hind, duck, cock, crown, tower, maunch, hearts and a zuil (column).-

Excluding arms of foreign kings, sovereign-dukes and counts, only fifty-two blazons in Walford's Roll disclose charges, and these are disappointing in comparison with the French armorial roll since, although they are quarters of a century later in date, they show little progress in variety from the early seals of a century earlier. They consist of seventeen lions, six mullets, five fleurs, four roses, three each of martlets and roundels, two each of escallops, escutcheons, bougets and garbs, and single examples of lucies, crescents, eagles, hammers, popinjays and fermaulx (buckles).

Charles' Roll discloses a five per cent rise in the proportion of charges, but more important is the considerable advance in the number of different objects in use. The 269 charged shields include eighty four lions, twenty-one roundels, nineteen mullets, seventeen martlets, fifteen crosses couped, thirteen eagles, thirteen escallops, nine crescents, eight roses, four ravens, four escutcheons, three each of sixfoils, lucies, garbs, maunches, trumpets and buck's heads, two each of horseshoes, annulets, hammers, fermaux, lures and cocks and one each bearing horse-brays, talbots, greyhounds, hands, pinecones, bird-bolts, trees, leaves, snakes, gorges, bourdons, chaplets, gloves, axes, herons, popinjays, dolphins, hose, chessrooks, klarions and helmets.

Die Grosse Heidelbergen Liederhandschrìft only a few years later is not only remarkable for the substantial rise in the proportion of charged shields to 70% but also in the relative decrease in the preponderance of lions, eagles etc., with a corresponding rise in other varieties of charges. Lions and eagles now number less than ten per cent of the charges and a rich selection of new objects appear — the unicorn, the demi-stag, wolf, swan, ram, trout, bird in a cage, castle, heads of men, women, eagles and dogs, crab-claws, gemring, sickle, hachet, wooden-bats, harp, lute, leaftwig, cushion, key, horns, crested-pothelm and an eagle with three heads, among others.

In the first 150 years of heraldry it seems likely that there was no distinction between roses and cinquefoils, mullets and estoiles, chess-rooks and zuils. It is proof that the mullet was originally a star, and not a spur-rowel as maintained by some authorities, by its common appearance in the earliest rolls: the grave-monuments all show knights wearing prick-spurs up to 1325-27 in Western Europe and only after that date is the rowel spur depicted. The early occurrence of the mullet pierced does not alter the position, since cinquefoils, sixfoils and other charges are found pierced at the same period. The estoile may have originally have been a comet, and in Southern France and Italy appears to have been common, while the mullet was apparently unknown.

The Inference. Greek classical shields are interesting because, though not heraldic, they portray creatures which the Greeks saw everyday in their lives- eagles, swans, geese, bulls, snakes, lions (common in Mycenaean times but only a memory in stone by sixth century BC), Medusa-heads (over every early temple door), chimerae and sphinx (both favourite art motifs whose existence was not doubted). Similarly the Vikings adorned the sail of their longships with common objects they observed at home - the bear, the raven of Odin, the wolf of Fafnir, the eagle and the omnipresent dragon which figures in every form of Scandinavian art until far into the Middle Ages.

Why then at the rise of heraldry in West Europe did not knights follow the same trend? It is curious they neglected the animals seen or hunted in their own lands - why are there no boars, wolves, bucks, bulls, stallions, bears and hounds for the first 150 years of armory? Why are there no dragons, when Celtic, Anglo-Saxon and Carolingian Romanesque art made wide use of them? Everyone in mediaeval Europe believed there were dragons and even as late as the eighteenth century the worthy scientist Professor Scheuchzer of Zurich dedicated his Proof of the Existence of Dragons to the Royal Society, London, some of whose members paid the costs of his publication!

Could the answer to these questions be that Heraldry did not originate in the west but on the Crusades ? Are there other pieces of evidence which suggest or support the eastern origin rather than the western ?

It is submitted that there are several other factors that point to the same conclusion. First for consideration are the animal charges and, as mediaeval man -whether knight, artist or sculptor - had little idea of zoology, it is not surprising that he had curious ideas of the appearance of non-European animals, as a glance at any heraldic work showing panthers, tigers, antelopes, etc., will confirm. Western knights had heard of lions in the Scriptures and seen the curious 'cross between a mastiff and a poodle' which in Romanesque art did duty for a lion, but it is unlikely that anyone in the West had ever seen a lion in the twelfth century. The carved shield of de Montfort in the choir at Westminster shows how far away from the real animal it was possible to go! But Roman and Romanesque art had shown animals moving naturally, yet from the beginning of heraldry the lions (and other creatures) walk with both legs on the same side in the forward position and those on the other a pace behind, as in the walk of the camel. Similarly when 'rampant' the two legs of one side are raised while both on the other are lower - an impossible attitude!

Fig. 9. Double eagle, Hittite 14th. C. B.C. — gateway, Alaya Hüyüt.
Fig. 10. Double eagle, Hittite 13th. C. B.C. — rock gallery Yazilikaya.
Fig. 11. Byzantine textile, 11th. C. A.D. — St. Germ. Auxerre.
Fig. 12. Eagle displayed, Byz. 10th. C. A.D. — stone, B. Mus.
Fig. 13. Double eagle, Byz. 9th. C. A.D. based on the Alaya relief above, even to the crouching beasts beneath the talons
Fig. 14. Lion of curious gait, 6th. C. A.D. — mosaic, Istanbul.
Fig. 15. Lion of curious gait, 11th. C. A.D. — Byz. text, copied from Persian relief, Düsseldorf Mus.
Fig. 16. Lion rampant, Byz. text. 8th. C. A . D . — Lyons Mus.
Fig. 17. Griffin segreant, Byz. text. 7th. C. A.D. St. Urs. Köln, note that the Byz. griffin is among the first mythical beasts as a heraldic charge, long pre-dating the West European dragon.
Fig. 18. Head of a unicorn, Babyl. 6th. C. B.C., copied in Byz. art.
Fig. 19. The sun of Apollo and the moon of Artemis were the symbols of the protecting deities of pagan Byzantium and were retained by Christian Constantinople.

Roman, Scandinavian and Romanesque Art portrayed eagles in a natural position; yet from the inception of heraldry the eagle is always 'displayed' like a dead butterfly in a collection. Where could these curiosities of attitude for lions and other quadrupeds, and for eagles have originated? They were in fact already in existence in these forms - in Byzantine art and particularly on woven textiles!

Moreover to reinforce the case we have the presence in heraldry, at least from the thirteenth century, of two things which could only have come from Asia Minor or Syria, since they occur nowhere else - the double-headed eagle and the use of the field paillé in Northern French heraldry, particularly in Normandy.

The Double Eagle. For generations heraldists have occupied themselves in attempting to explain or account for the eagle with two heads. The favourite theories of last century were a) it originated in the dimidiation of two separate coats each bearing a normal eagle displayed, or b) Nisbet's theory, quoted by Woodward, that the Byzantines invented it to represent the dual Emperorship by two eagles standing one in front of the other so that only the head of the rear one showed.

But the double-eagle had been known in the ancient Near East from the third millennium BC. Its earliest appearance is on cylinder seals of the Sumerian civilisation found at Lagash on the Euphrates. The Sumerians travelled as far north as the Taurus Mountains in Asia Minor to obtain tin, and must have left their art-forms in North Syria and Anatolia, since this symbol was inherited by the Hittites who occupied those areas a thousand years later and frequently used this motif in their rock carvings and on their monuments. Two examples are shown in the accompanying illustrations.

The Byzantines therefore had examples of this ancient symbol extant on monuments in their own empire, and even as they made use of Persian and Babylonian motifs in their arts and particularly on their textiles, so they made use of the double-eagle of the Hittites in damask, embroidery etc. We know that isolated examples of these textiles showing double-eagles had reached Europe long before the period of the Crusades — one fragment was found in Charlemagne's tomb in Aachen and another had been used as a shroud in an eleventh century tomb at Auxerre -but neither apparently attracted the attention of western artists.

Paille. This field, peculiar to Normandy and the adjoining provinces of Northern France, is intended to represent a shield covered by eastern textiles from the looms of the Byzantine Empire. It is depicted as blue or green, semé of either lions and single eagles or double-eagles, each enclosed within an annulet, and each annulet linked to those around, the whole being in gold. It was treated as a fur and was used with ordinaries or charged like any other field. It appears to have relatively common in the early centuries of armory and Galbreath illustrates one example as late as 1640.

Paillé: this field represents Byzantine textiles, brocaded in gold and used to cover shields at tourneys and pageants (cf. French verb 'pailleter' — to spangle). Its use was peculiar to Northern France and was most common in Normandy. The design was always in gold on either blue or green backgrounds, a) and c) examples of paillé; b) part of the shield of the family Brisard-Tiville : barry of six azure paillé and ermine.

Heraldry's Debt to Byzantium. A study of the textiles of the Eastern Empire reveals where the Frankish knights found the charges which Western Europe had never seen before. Here is the lion of peculiar gait, the eagle displayed, the double-eagle, the unicorn, the griffin - both winged and wingless - and other fabulous creatures which may have produced the heraldic panther and antelope, and the curious Austrian panther.

But there was another art-motif that came from the same source: the city-symbol of Hellenistic Byzantium - a combination of the signs of its two patron deities - the sun of Apollo and the moon of Artemis. The moon was shown as a crescent, with the sun represented as a circle of rays without a disc above it: this symbol was retained by Christian Constantinople and was adopted by the Crusaders. It appears on seals, apparently as a sign of crusader-vows, often unfulfilled. So they passed into armory but did not become the emblem of Islam until the Turks captured Constantinople in 1453.

Conclusion. When all these links with the Byzantine arts are contrasted with the absence - until a century and a half after the first appearance of heraldry - of the objects which might have been expected from Western European culture, it appears more than probable that, while the ordinaries originated from the mechanics of strengthening shields in the West, the introduction of charges first occurred in the Near East.

As the earliest use of armory discloses a close uniformity in choice of charges, attitudes and basic rules in eight countries, the only opportunities that can be found when men from many western nations were in the Byzantine Empire was during the early Crusades. The second crusade began in 1147, but the evidence for heraldry on seals in the West goes back to 1140 and possible earlier, and as it is unlikely that knights adopted heraldry on shields and simultaneously had new seals made to reproduce the same design, heraldic shields may have originated five or ten years earlier.

Going back more than forty years before this period we find that the western knights all travelled across Europe by land to the First Crusade and most had already assembled at Constantinople by December 1096. But they remained there until May 1097 before plans were completed and supplies amassed to cross the Bosphorus and march across Anatolia. We know that the knights were amazed at the luxury, comforts and riches of the Byzantines, but they had a long and arduous campaign ahead of them before Jerusalem was captured in July 1099.

It is not suggested that heraldry arose during this campaign, but from 1100 onwards, the Crusaders were occupied in perfecting a network of feudal states on the western model -Edessa, Antioch, Tripoli and the kingdom of Jerusalem - as well as fighting the Arabs. The difference between the First Crusade and its successors lay in the knights as a whole staying permanently in Outremer and carving out a life and career there, while on subsequent crusades, usually lead by monarchs, knights only stayed one or two years and then returned to their homes in the West. In addition to the organised expeditions, smaller bands of knights arrived almost every year to reinforce the Franks, but after a sojourn returned to Europe.

The proposition for consideration is that heraldry arose among these crusader-states of Outremer during this period of consolidation, say after 1110, but more likely during the kingship of Fulk d'Anjou, who obtained the throne of Jerusalem in 1131. We know that the barons of the Holy Land by this date had adopted the luxuries of both Byzantine and Eastern cultures - silks, damasks, brocades, carpets, perfumes, spices, fruit cultivation, flower-gardens, etc., - which tastes by degrees spread to western Europe, encouraged by the importing merchants of Genoa, Amalfi, Pisa and Venice. It is suggested that returning 'short-term' crusaders brought back to their homelands the science of heraldry. The fact that no resident barons of the Kingdom apparently used heraldic seals until 1160 may be due either to the accident of non-survival of evidence or, more likely, to the local adoption of the armorial seal coming later, after heraldry was established in western Europe.