CHAPTER I.
ON SYMBOLS COMMON TO DIFFERENT RACES.
The variety of symbols seems at first to be as boundless as the combinations of the human imagination. It is not uncommon, however, to discover the same symbolical figures amongst races the furthest apart. These coincidences can hardly be explained by chance, like the combinations of the kaleidoscope. Except in the case of symbols found amongst peoples who belong to the same race, and who, consequently, may have carried away from their common cradle certain elements of their respective symbolism, there are only two possible solutions: either these an*logous images have been conceived independently, in virtue of a law of the human mind, or else they have pa**ed from one country to another by a process of borrowing.
There exists a symbolism so natural that, after the manner of certain implements peculiar to the stone ages, it does not belong to any definite region or race, but constitutes a characteristic feature of humanity in a certain phase of development.
To this category belong, for example, the representations of the sun by a disc or radiating face, of the moon by a crescent, of the air by birds, of the water by fishes, also by a wavy line, and so forth.
Perhaps certain more complicated an*logies should be added to these, such as the symbolizing of the different phases of human existence by the life of the tree, the generative forces of nature by phallic emblems, the divine triads, and generally every triple combination whose members are equal, by the equilateral triangle, and, lastly, the four main directions of space by a cross.
What theories have not been built upon the existence of the equilateral cross as an object of veneration amongst nearly all the races of the Old and the New World! Of late years orthodox writers have protested with good reason against the claim of attributing a pagan origin to the Cross of the Christians because earlier creeds had included cruciform signs in their symbolism. And the same objection might be urged against those who seek for Christian infiltrations in certain other religions under the pretext that they possess the sign of the Redemption.
When the Spaniards took possession of Central America, they found in the native temples real Crosses, which were regarded as the symbol, sometimes of a divinity at once terrible and beneficent —Tlaloc, sometimes of a civilizing hero, white and bearded—Quetzacoalt, stated by tradition to have come from the East. They concluded from this that the Cross had reached the Toltecs through
Christian missions of which all trace was lost; and, as legend must always fix upon a name, they gave the honour to St. Thomas, the legendary apostle of all the Indies. Although this proposition has again found defenders in recent congresses of Americanists, it may be regarded as irrevocably condemned. It has been ascertained beyond all possibility of future doubt that the Cross of pre-Columbian America is a kind of compa** card, that it represents the four quarters whence comes the rain, or rather the four main winds which bring rain, and that it thus became the symbol of the god Tlaloc, the dispenser of the celestial waters, and, lastly, of the mythical personage known by the name of Quetzacoalt. 1
By a similar process of reasoning the Assyrians
FIG. 1. IDEOGRAM OF ANU. Fig. 1. Ideogram of Anu.
(Rawlinson. Western Asia Inscriptions, vol. ii. pl. 48, fig. 30 obv.)
were led to represent by an equilateral cross their god of the sky, Anu. The ideogram of this god is formed by four cruciform characters which radiate at right angles from the circle or lozenge denoting the sun in the cuneiform inscriptions. Is
not the sky indeed the space in which light radiates?
It is proper to remark that amongst the Assyrians themselves the equilateral cross, as denoting the main directions in which the sun shines, became also the symbol of that luminary, and consequently, here again, of the god who governs it. It was the same with the Chaldæans, the Hindus, the Greeks, the Persians, and perhaps with the Gauls, and the ancient civilizers of Northern America (fig. 2).
FIG. 2. SOLAR CROSSES. Fig. 2. Solar Crosses 1.
In China, the equilateral cross inscribed in a square
represents the earth, and according to Mr. Samuel Beal, a saying is met with there to the effect that "God made the earth in the form of a cross." 2
Egyptian writing utilizes among its signs the Greek and even the Latin Cross. In connection
with this we find in a recent article by the Abbe Ansault a characteristic example of the readiness with which one may go astray in the identification of symbols, if satisfied with a merely superficial resemblance. On the famous Damietta stone, the Greek words Πτολεμαῖος Σωτήρ, "Ptolemy the Saviour," are rendered by the demotic characters forming the equivalent of Πτολεμαῖος, followed by the sign
; from which the author concludes that the term Saviour being rendered by a cross, this sign was with the Egyptians, an allusion to the future coming of the Redeemer. 1 Unhappily for this ingenious interpretation, M. de Harlez, who has taken the trouble to refute M. Ansault's article, points out to him that in
FIG. 3. HIEROGLYPH OF THE HAMMER. Fig. 3. Hieroglyph of the Hammer
(E. Coemans, Manuel de langue égyptienne, p. 47, § xviii.)
demotic the sign
is the simplest form of a hieroglyph representing a hammer, or a boring tool, and is usually employed to express the idea of grinding, avenging, and by amplification, "the Grinder," "the Avenger," a not uncommon epithet of Horus, and some other gods. 2
Even in the presence of an an*logy of signification combined with a resemblance of forms, it is well to look twice before identifying symbols. The St. Anthony's Cross (croix potencée, literally "gibbet-cross")
is found, with almost the same symbolical signification, in Palestine, in Gaul, and in ancient Germany, in the Christian Catacombs, and amongst the ancient inhabitants of Central America. Among the Phœnicians and their kindred races, it was the character known by the name of tau, and from an oft quoted pa**age in Ezekiel 1 we learn that it was accounted a sign of life and health. Among the Celts and the ancient Germans it was the representation of the celestial Two-headed Mallet which was accounted an instrument of life and of fecundity. Amongst the early Christians it was a form sometimes given to the Cross of Christ, itself called the Tree of Life. 2 In Central America where, according to M. Albert Réville, the Cross was surnamed the Tree of Plenty, 3 it a**umed also the form of the tau.
Are we to conclude from this that all these gibbet-crosses have the same origin and the same aim? That would be a rather hasty conclusion. The symbolical signification of the tau is explained by its resemblance to the Key of Life or crux ansata of Egypt, so widely diffused throughout all Western Asia. The Double Hammer of Thor and of Tarann is a symbol of the lightning, and, for this reason, could not fail to represent the vivifying forces of the storm, according to the tradition common among the Indo-European nations. 4 Similarly, if in pre-Columbian America, the Cross became an emblem of fertility, it is, as we have seen, because it represents the rain-god. As for the early Christians, if they made of the Cross a symbol of life, it is especially in the spiritual sense; and, if they sometimes gave it the form of the palibulum, it was because such was the instrument employed among the Romans in the punishment by crucifixion.
In the mythology of primitive nations the contest between the sky, or sun, and the clouds is frequently represented by a fight between an eagle and a serpent. This subject has been treated more than once in ancient art. 1 Already in the Homeric ages it had become a symbol of victory, for we are told in the Iliad that the Trojans were on the point of abandoning the a**ault on the Greek entrenchments through having seen an eagle which held a serpent in its claws take flight, being wounded by its prey. 2 Now according to the tradition of the Aztecs, the founding of Mexico is said to have been resolved on owing to the apparition of an eagle which, perched upon an agave, and with wings outstretched towards the rising sun, held a serpent in its talons. 3 The first conquerors of Mexico saw therein an emblem of future greatness, and to the present day this emblem figures in the arms of the capital. Yet it is unlikely that the Aztecs had read Homer.
On the other hand, the ease with which symbols are borrowed is indisputable. Represented on the ordinary productions of industry, favourite subjects with artists, they pa** unceasingly from one country to another, with commodities of commerce and articles of ornament; as witness the specimens of Hindu, Chinese, and Japanese symbolism and pictorial art which have penetrated
among us with the vases, the fabrics, and all the curiosities of the far East. The centres of artistic culture have always been foci of symbolic exportation. Have there not been found even among the "treasures" of our mediæval churches on the one hand, and among those of Chinese and Japanese temples on the other, masterpieces of Sa**anian art, which themselves reproduce the symbols of cla**ic paganism? 1
In olden times soldiers and sailors and travellers of every profession never left home without taking with them, under some form or another, their symbols and gods, a knowledge of which they thus spread in remote parts—bringing back from abroad others in return. Slavery, so largely known in the ancient world, must likewise have favoured the importation of symbols through the medium of those innumerable captives whom the fortune of war, or the chances of piracy, brought everywhere from the most distant regions, without depriving them of the memory of their gods and the forms of their worship. Lastly, coins have never failed to propagate through immense distances the symbols of the nations who put them into circulation. Gallic coins are nothing but counterfeits of the Greek coinage of Phillip or Alexander, and even in the tumuli of Scandinavia native coins have been found which roughly imitated the coinage of Bactria. 2
Now nothing is so contagious as a symbol unless perhaps a superstition; they are all the more so when combined together, as they generally were among the nations of antiquity, who scarcely
adopted a symbol without attaching a talismanic value to it. Even to-day there are tourists who return from Naples with a coral horn, suspended, according to the s** of the traveller, from the bracelet, or the watchchain. Do they really believe that they have found a preservative against the evil eye in this Italian survival of an old Chaldæan symbol? To many of them it is a**uredly only a local curiosity, a trinket, a souvenir of travel. But there are some amongst the number who let themselves be influenced, even unconsciously, by the Neapolitan superstition. "It can do no harm and may perhaps do good" they would be tempted to reply, in imitation of certain gamblers when bantered about their fetiches.
We have here an argument which is almost universal among polytheistic populations, where everyone thinks it safe to render homage not only to his own gods, but also to those of others, and even to unknown divinities, for do we ever know of whom we may have need in this world, or in the next? Egyptian scarabæi have been found by the thousand from Mesopotamia to Sardinia, wherever either the armies of the Pharaohs or the ships of the Phœnicians penetrated. Everywhere too in these latitudes there have been found native scarabæi made in imitation of those of Egypt, and reproducing with more or less exactness the symbols which the engravers of the valley of the Nile displayed so lavishly on the flat side of these amulets. It is thus again that, long before the diffusion of coins, the pottery, the j**els, the statuettes of Greece and of Etruria, furnished Central and Western Europe with divine types and symbolic figures.
Are there any indications which enable us to determine whether an*logous symbols have been
produced independently, or were derived from the same source?
Intricacy and singularity of forms when they exceed certain limits may justify the second of these propositions.
We may well suppose that in the desire to symbolize the strength or activity of superhuman beings, the Egyptians, the Aztecs, the Hindus, and the Chinese, have been separately led to enrich by several pairs of arms and legs, or even by several heads, certain figured representations of their superhuman beings. 1 But does this hypothesis of an independent origin hold good when, for example, we see both on the ancient coins of Lycia and in the feudal coat of arms of the Isle of Man, a figure at once so precise as the triscèle or triquetra, those three legs, joined together at the thigh, which radiate from a central point?
FIG. 4. TRISCÈLE ON THE SHIELD OF ENCELADUS. (DE WITTE and LENORMANT. Monuments Céramographiques, vol. i., pl. viii.) Fig. 4. Triscèle on the Shield of Enceladus.
(De Witte and Lenormant. Monuments Céramographiques, vol. i., pl. viii.)
There is nothing for it but to ask one's self how this ancient solar symbol can have pa**ed from one country to the other. The intermediate stage may perhaps be found in Sicily, where the triscèle was used in the coinage, from the time of Agathocles, to symbolize the configuration of the island with three promontories. As the Isle of Man also presents this geographical peculiarity, it is very possible that, at the commencement of the
middle ages, a Norman baron, or even a Crusader, or simple adventurer, returning to his home after a sojourn in Sicily, applied to his native country a symbol still living in the cla**ic traditions of the ancient Trinacria, save that in order to suit the age he added spurs to the heels.
We are familiar with the Double-headed Eagle of the old German empire, still blazoned on the armorial bearings of Austria and of Russia. What was the surprise of the English travellers Barth and Hamilton, when, in exploring Asia
FIG. 5. BAS-RELIEF OF EYUK. (PERROT and CHIPIEZ. L'Art dans l'antiquité, vol. iv., fig. 343.) Fig. 5. Bas-Relief of Eyuk.
(Perrot and Chipiez. L'Art dans l'antiquité, vol. iv., fig. 343.
[paragraph continues] Minor about fifty years ago, they discovered a Double-headed Eagle of the same type sculptured in the midst of religious scenes on Pterian bas-reliefs dating back to the civilisation of the Hittites?
It is difficult to admit that, on both sides, there was conceived spontaneously, on identical lines, a representation of the eagle so contrary to the laws of nature. M. de Longpérier has solved the enigma by reminding us that it was only about 1345 that the Eagle with Two Heads replaced the monocephalous one on the armorial bearings of the Western Empire. It would seem to have been Flemish princes who, during the Crusades, appropriated the device from the coins and standards of the Turkomans, then the masters of Asia Minor. The latter had adopted it as the symbol of omnipotence, perhaps the hamka, the fabulous bird of Moslem tradition, which carries off the buffalo and the elephant, as the kite carries off the mouse. "Thus," observes M. Perrot, "there would seem to have been transplanted into our modern Europe a symbol belonging originally to an Asiatic creed of the highest antiquity; and by a singular turn of fortune the Turks saw, at Belgrade and at Lepanto, their advance towards the West barred by that same eagle which had conducted them triumphantly along the banks of the Euphrates and the shores of the Bosphorus." 1
Perchance the Turkomans themselves had borrowed this symbol from the sculptures carved by their mysterious predecessors on the rocks of Eyuk and of Iasili-Kaïa. But it is equally possible that they acquired it through the medium of the Persians. We find in the collection of M. de Gobineau an intaglio, attributed by him to the time of the Arsacidæ, on which is engraved the traditional type of the Double-headed Eagle holding, as at Eyuk, a hare in each claw.
M. de Longpérier observes that if the stem of certain ferns (Pteris aquilina) be cut transversely a fairly accurate image of the Double-headed Eagle
FIG. 6. ARSACIAN INTAGLIO. (Revue Archéologique of 1894, vol. xxvii., pl. v., No. 391.) Fig. 6. Arsacian Intaglio.
(Revue Archéologique of 1894, vol. xxvii., pl. v., No. 391.
is obtained. Now the fern is named in Greek πτέρις or πτερία, as is the province where the bas-reliefs of Eyuk are found. The learned archæologist wondered if it was not this similarity which caused the Double-headed Eagle to be chosen as the symbol of Pteria: 1 but we know now that the bas-reliefs in question date from a period far earlier than that of the appearance of the Greeks in this part of the world, and, besides, it is probable that the Greeks had given a name to the fern before knowing Pteria. The most that can be admitted is that the resemblance of the Hittite symbol to the bicephalous figure obtained from the fern led the Greeks to name the country after the plant.
The Greeks, whom we have seen adopting as a symbol of victory the figure of an eagle holding a serpent between its talons, sometimes replaced the serpent by a hare which corresponded with the Hittite scheme. Only they rejected anything monstrous that the latter might offer, and contented themselves with faithfully copying nature. 2 India on the contrary seems to have accepted without hesitation the bicephalous type which Persia probably transmitted to it. We there find the Double-headed Eagle on ancient coins, where
it holds an elephant instead of a hare, not only in each talon, but also in each beak. Moor saw in this a representation of Garuda the solar Eagle which Vishnu rode. In any case we here draw singularly near the hamka of the Turks; and it may even be that the latter derived their legend of the fabulous bird from some representation of this kind, where the part of the hare was taken by an elephant, or buffalo.
FIG. 7. ANCIENT INDIAN COIN. (MOOR. Hindoo Pantheon, pl. 103, fig. 3.) Fig. 7. Ancient Indian Coin.
(Moor. Hindoo Pantheon, pl. 103, fig. 3.)
It cannot, however, be said that Greece had nothing to do with the production of this emblem.
M. Clermont-Ganneau has shown how, in the popular iconography, complex monsters were frequently produced by a false interpretation of groups formed of separate individuals. There is, for example, an image of Phœnician origin which shows us Orthros in the form of two dogs distinctly apart. "Hellenic image makers," he goes on to say, "unite the two animals, while fable goes still further and endows the imaginary creature with a third head which it does not always possess in ancient art." 1 Thus again the Chimæra originates in the group, so widespread in Lycian art, of a lion devouring a goat; the two animals having been taken for one by the Greek copyist. In the same manner the triple Geryon slain by Herakles owes its existence to a wrong interpretation of the scene, taken originally from Egypt, in which a king is seen raising his club as if to strike three barbarians, who are grouped in such a manner as to give the
illusion of a single body with three heads, six arms, and six legs.
I had long since surmised a similar origin in the Double-headed Eagle, when, turning over once more the pages of Schliemann's Mykenæ, I discovered the solution of the problem in some golden fibulæ dug up by the famous archæologist among the tombs of the ancient Mycenæ (fig. 8 and 9). We there find two eagles, as Schliemann says, "leaning against each other with their whole bodies and even with their claws while turning their heads in opposite directions."
At Eyuk the two eagles are fused in one. In this instance it is not Greek mythology which has clumsily interpreted a Phœnician image, but the Asiatic sculptor who has misunderstood the real meaning of the Mycenian image.
When, therefore, the Double-headed Eagle changed sides in the thirteenth century of our era, during the struggle which has waged for more than thirty centuries between Europe and Asia, it did nothing else than return, after many wanderings, to its original home.
FIG. 8. JEWELS FROM MYKENÆ FIG. 9. JEWELS FROM MYKENÆ Fig. 8, 9. Jewels From Mykenæ.
(Schliemann, Mykenæ.)
I will cite yet another symbol, come from afar, the Semitic origin of which is not to be gainsaid; even when we cannot identify all the stages of the route it followed in order to reach us.
Christian symbolism has often represented God the Father, or Providence, "the Hand of God," by a hand emerging from a cloud. In some of these figures the finger-tips emit rays of light, "as if it were a living sun," observes M. Didron in his Histoire de Dieu; and a miniature of the ninth century in the Bibliothèque Nationale of Paris, shows the Divine Hand in the middle of a nimbus encircling a Cross.
FIG. 10. THE HAND OF GOD. (DIDRON. Histoire de Dieu, p. 32.) Fig. 10. The Hand of God.
(Didron. Histoire de Dieu, p. 32.)
M. Gaidoz has compared this figure with certain Gallic amulets, the solar Wheels formed of four rays through which a hand pa**es. 1
Might not the two symbols have their prototype upon an Assyrian obelisk, where two hands are shown to issue from a solar Disk, the right open and exhibiting the palm, the left closed and holding a bow?
FIG. 11. ASSYRIAN SYMBOL. (G. RAWLINSON. The Five Great Monarchies, vol. ii. p. 233.) Fig. 11. Assyrian Symbol.
(G. Rawlinson. The Five Great Monarchies, vol. ii. p. 233.)
The representation of the open, or uplifted, hand, intended to typify the divine might, is, moreover,
common to all branches of the Semitic race; it appears already amongst the Chaldæans, for a cylinder, of Babylonian origin, exhibits an uplifted hand, which emerges from a pyramidal base, between persons in an attitude of adoration; this is precisely the type of our "Hand of Justice."
FIG. 12. CHALDÆAN CYLINDER. (Lajard. Mithra, pl. xxvii., fig. 5.) Fig. 12. Chaldæan Cylinder.
(Lajard. Mithra, pl. xxvii., fig. 5.)
According to M. François Lenormant, the celebrated pyramid of Borsippa was called "the Temple of the Right Hand," and one of the names of Babylon was that of "the city of the Hand of Anu," or, what amounts to the same thing, of "the Celestial Hand." 1
The hand uplifted towards the sky is an oft repeated image on the ex voto of Carthage, and even at the present time it is figured on native houses in Palestine and Marocco to ward off evil spirits from the dwellers therein. 2 Moreover this symbol pa**ed also to India, where it decorates the
pedestal of the Holy Tree in a bas-relief at Bharhut (see pl. iv., fig. h).
In default of adequate proofs drawn from singularity of form, identity of meaning and of use may afford strong presumptions in favour of the relationship of symbols.
There would be nothing surprising in Hindus and Egyptians having independently adopted, as a symbol of the sun, the lotus-flower, which every morning opens under the first rays of that luminary to close again at eventide, and which seems to spring up of itself on the surface of the placid waters. However, the hypothesis of a transmission becomes very plausible when, in the iconography of the two nations, we see this flower serving both as a support to the solar gods, Horus and Vishnu, and figuring in the hand of the goddesses a**ociated with these gods, Hathor and Lukshmi, the Venuses of Egypt and India. Lastly, this plausibility becomes a sort of certainty when, on both sides, we find the lotus employed to interpret the same shade of thought in some indirect and subtle enough applications of solar symbolism.
It must indeed be remarked that on either side the Lotus-flower symbolizes less the sun itself than the solar Matrix, the mysterious sanctuary into which the sun retires every evening, there to acquire fresh life.
This miracle, which was believed to be renewed every day, was regarded as the origin of whatever exists. The Egyptians, who believed that the world had sprung from the liquid element, made the sun to proceed from a Lotus which one day had emerged from the primordial waters; 1 this they rendered in their iconography by representing
p. 29
Horus as springing forth from a lotus-shaped calyx held by Hathor. 1 In the same way the Indian sacred books constantly speak of gods as sprung from the lotus; it is on a golden Lotus that Brahma appears in the beginning of time, and it is with the different parts of this plant that he created the world. 2 A Hindu legend, related by Father Vincenzo de Santa Catarina, states that Brahma keeps watch six months of the year and sleeps the other six in a Lotus-flower of extraordinary size and beauty. 3
Hence a fresh enlargement given to the figurative meaning of the lotus. The symbol of solar renascence, it became, with the Egyptians, the symbol of human renascence and, generally, of life in its eternal and unceasingly renewed essence. On a sarcophagus in the Louvre there is depicted a scarabæus emerging from a Lotus between Isis and Nephthys in their characteristic attitude of guardians and protectresses of the dead. 4 Thus were represented both the sun and the deceased pa**ing through the tomb to renew their existence in the luminous fields of space. The lotus was even adopted with this signification in the funeral symbolism of Europe. It is met with again, not in the Greek traditions relating to the Lotophagi, those fabulous people who partook of the lotus in order to forget life and its troubles, but even in the inscriptions on tombstones which are met with, dating from the latter centuries of Paganism, from Libya to Belgium. 5
Renascence has but few attractions for the}
p. 30
[paragraph continues] Brahmans, and still fewer for the Bhuddists. The latter adopted the ancient Flower of Life but to symbolize, according to their different schools, nature in the sum of her manifestations—the eternally active matter—the innumerable worlds which fill space—the Buddha dwelling in each of them—lastly, the teaching of the Master, that is to say, the means of escaping from that chain of causes and effects which engenders personal existence. It is thus that they carried to the confines of Asia the Lotus of the Good Law; and even to-day in the Himalayas there is no valley so remote that the traveller does not hear everywhere on his approach, as an utterance of sanctification and of welcome, the mystic formula: om mani padmi om,—"Oh! the Jewel in the Lotus."
However, popular traditions and engraved monuments would suffice to remind us of the ties which unite the Lotus of Buddhism to that of Egypt. A legend relates that, when the Buddhisattva appears, a miraculous Lotus springs out of the earth, and he seats himself thereon, and takes in all the worlds at a glance. 1 Buddha, besides, is everywhere represented seated on the Lotus-flower like Vishnu and Horus. It is perhaps not impossible to fix the intermediate stages of this symbolism. The Lotus pa**ed from Egypt to the monuments of Phœnicia and, towards the eighth century before our era, to those of Assyria, 2 which in their turn transmitted it to Persia. Thus, in the sculptures of Phœnicia goddesses are found who hold in the hand a Lotus-flower, and, in the Sa**anian bas-reliefs at Tagh-i-Bostan, the solar god Mithras stands upon a Lotus-flower. 1 Lastly, among the Mesopotamians and the Persians it is not uncommon to find the Lotus blossoming on shrubs in which may be recognized either the Sacred Tree of the Semitic religions, or the Iranian tree which secretes the Elixir of Immortality. 2
Nowadays the beautiful rose-blossomed lotus, Nymphæa Nelumbo, observed on the monuments of Egypt, no longer grows in that country in its wild state, but by a curious coincidence, it has remained in the flora, as in the symbolism, of India. 3
FIG. 13. CAVES OF KANERKI. (FERGUSSON and BURGESS. Cave Temples of India, pl. x., fig. 35.) Fig. 13. Caves of Kanerki.
(Fergusson and Burgess. Cave Temples of India, pl. x., fig. 35.)
[paragraph continues] We may add that it has been imported from India to China and Japan, so that it is still one of the principal symbolical figures by which at the present day we recognize the sacred vases and other objects employed in religious services by the Buddhists of these countries