This is a passage on which I shall have to comment again: at present I will content myself with noting how accurately the poet, who of all others best understood the instincts of the less civilised Italians of his own day, has used his knowledge to express the antique feeling that there were places which man must shrink from entering—a feeling far older than the invention of legal consecratio by the authorities of a City-state.

Lastly, the principle of taboo, or religio, if we use the Latin word, affected certain times as well as places. Just as under the ius divinum of the fully-developed State certain spots were made over to the deities for their habitation and rendered inviolable by consecratio, so certain days were also appointed as theirs which the human inhabitants might not violate by the transaction of profane business. But I have just pointed out that the consecration of holy places in this legal fashion was a late development of a primitive feeling or religio; exactly the same, if I am not mistaken, was the case with regard to the holy days. These were called nefasti, and belong to the life of the State; but there were others, called religiosi, which I believe to have been tabooed days long before the State arose.

When we come to examine the ancient religious calendar, it will be found that I shall not then be called upon to deal with dies religiosi, for the very good reason that they are not indicated in that calendar—there is no mark for them as religiosi, and some of them are not even dies nefasti, as we might naturally have expected.67 What, then, is the history of them? We may be able to make a fair guess at this by noting exactly what these days were; Dr. Wissowa has put them together for us in a very succinct passage.68 He begins the list with the 18th of Quinctilis (July), on which two great disasters had happened to Roman armies, the defeats on the Cremera and the Allia; and also the 16th, the day after the Ides, because, according to the legend, the Roman commander had sacrificed on that day with a view to gaining the favour of the gods in the battle. We may regard the story about the 18th as historical; but then we are told that all days following on Kalends, Nones, and Ides were likewise made religiosi (or atri, vitiosi, which have the same meaning) as being henceforward deemed unlucky by pronouncement of senate and pontifices;69 thus all dies postriduani, as they were called, were put out of use, or at any rate declared unlucky, for many purposes, both public and private, e.g. marriages, levies, battles, and sacred rites,70 simply because on one occasion disaster had followed the offering of a sacrifice on the 16th of Quinctilis. It is difficult to believe that thirty-six days in the year were thus tabooed, by a Roman senate and Roman magistrates, in a period when the practical wisdom of the government was beginning to be a marked characteristic of the State. Some people, we are told, went so far as to treat the fourth day before Kalends, Nones, and Ides in the same way; but Gellius declares that he could find no tradition about this except a single passage of Claudius Quadrigarius, in which he said that the fourth day before the Nones of Sextilis was that on which the battle of Cannae was fought.71

I am strongly inclined to suggest that the traditional explanation of the tabooing of these thirty-six, or possibly seventy-two days was neither more nor less than an aetiological myth, like hundreds of others which were invented to account for Roman practices, religious and other; and this supposition seems to be confirmed as we go on with the list of dies religiosi as given by Wissowa. The three days—Sextilis 24, October 5, November 8—on which the Manes were believed to come up from the underworld through the mundus (to which I shall return later on) were religiosi;72 so were those when the temple of Vesta remained open (June 7 to 15),73 those on which the Salii performed their dances in March and October,74 two days following the feriae Latinae (a movable festival),75 and the days of the Parentalia in February and the Lemuria in May, which were concerned with the cult and the memory of the dead.76 Now the religio or taboo on these days obviously springs either from a feeling of anxiety suggested by very primitive notions of the dead and of departed spirits; or in the case of the temple of Vesta, by some mystical purification or disinfection preparatory to the ingathering of the crops, which I noticed in my Roman Festivals (p. 152 foll.); or again in the case of the Salii, by some danger to the crops from evil spirits, etc., which might be averted by their peculiar performances. In fact, all these dies religiosi date as such, we may be pretty sure, from a very primitive period before the genesis of the City-state, and were not recognised—for what reason we will not at present attempt to guess—as religiosi by the authorities who drew up the Calendar. Some of them appear in that calendar as dies nefasti, but not all; and I am entirely at one with Wissowa, whose knowledge of the Roman religious law is unparalleled for exactness, in believing that a religio affecting a day had nothing whatever to do with its character as fastus or nefastus.77

If all these last-mentioned dies religiosi are such because ancient popular feeling attached the religio to them, we may infer, I think, that the same was really the case also with the dies postriduani. The fact that the authorities of the State had made one or two days religiosi as anniversaries of disasters, supplied a handy explanation for a number of other dies religiosi of which the true explanation had been entirely lost; but that there was such a true explanation, resting on very primitive beliefs, I have very little doubt. Lucky and unlucky days are found in the unwritten calendars of primitive peoples in many parts of the world. An old pupil, now a civil servant in the province of Madras, has sent me an elaborate account of the notions of this kind existing in the minds of the Tamil-speaking people of his district of southern India. The Celtic calendar recently discovered at Coligny in France contains a number of mysterious marks, some of which may have had a meaning of this kind.78 Dr. Jevons has collected some other examples from various parts of the world, e.g. Mexico.79 The old Roman superstition about the luckiness of odd days and the unluckiness of even ones, which appears, as we shall see, in the arrangement of the calendar, was probably at one time a popular Italian notion, not derived, as used to be thought, from Pythagoras and his school.

I therefore conclude that we may add times and seasons to the list of those objects, animate and inanimate, which were affected by the practice of taboo in primitive Rome; and I hold that the word religiosus, as applied both to times and places, exactly expresses the feeling on which that practice is based. The word religiosus came to have another meaning (though it retained the old one as well) in historical times, and the Romans could be called religiosissimi mortalium in the sense of paying close attention to worship and all its details. But the original meaning of religio and religiosus may after all have been that nervous anxiety which is a special characteristic of an age of taboo.80 To discover the best methods of soothing that anxiety, or, in other words, the methods of disinfection, was the work of the organised religious life of family and State which we are going to study. But I must first devote a lecture to another class of primitive survivals.

NOTES TO LECTURE II

LECTURE III

ON THE THRESHOLD OF RELIGION: MAGIC

Taboo, the traces of which at Rome we examined in the last lecture, is, as we saw, closely allied to magic, even if it be not, as Dr. Frazer thinks, magic in a negative form. We have now to see what traces are to be found of magic in the proper or usual sense of the word—active or positive magic, as we may call it. By this we are to understand the exercise of a mysterious mechanical power by an individual on man, spirit, or deity, to enforce a certain result. In magic there is no propitiation, no prayer. "He who performs a purely magical act," says Dr. Westermarck,81 "utilises such mechanical power without making any appeal at all to the will of a supernatural being." Religion, on the other hand, is an attitude of regard and dependence; in a religious stage man feels himself in the hands of a supernatural power with whom he desires to be in right relation.

If we accept this distinction, as I think we may (though one school of anthropologists is hardly disposed to do so), it is plain that magical practices are of a totally different kind from religious practices, as being the result of a different mental attitude towards the supernatural; they belong to a ruder and more rudimentary idea of the relation of Man to the Power manifesting itself in the universe. True, they have their origin in the same kind of human experience, in the difficulties man meets with in his struggle for existence, and his desire to overcome these; but unlike religion, magic is a wholly inadequate attempt to overcome them. This inadequacy was long ago well explained by Dr. Jevons.82 He showed that man in that early stage of his experience did not understand the true relation of cause and effect; that, "turned loose as it were among innumerable possible causes (of a given effect), with nothing to guide his choice, the chances against his making the right choice were considerable." As a matter of fact he usually made the wrong one, and is still apt to do so. There is probably more magic going on behind the scenes even in civilised countries, and more especially both in Greece and Italy, than either men of science or men of religion have any idea of. In its various forms as they are now classified,83 e.g. contagious magic, and homoeopathic magic, the exercise of the mysterious will-power, real or imaginary, is to be found all the world over, accompanied usually with a spell or incantation which is believed to enforce and increase that power—a kind of telepathy, which seems to be the psychological basis, so far as there is one, of the whole system. In these rites the virtue resides in some action, which, together with the spell or incantation, enforces the desired result by calling out the will-power, or mana, if we adopt the convenient Melanesian word lately brought into use. Whatever percentage of psychological truth may lie at the root of such performances, it is obvious that they must in the main be wholly inadequate, and must constantly tend to pass into mere quackery and become discredited; and it was the special function of the religious organisation of early society to eliminate and discredit them.

But it was a long stage in the evolution of society before man arrived at a better knowledge of his relation to the Power manifesting itself in the universe; before he reached the idea of a god or spirit realisable and nameable, and thus capable of being addressed, placated, worshipped. When this stage is reached, there supervenes almost always a strong tendency to regulate and systematise the methods of address, placation, and worship; and among some peoples, e.g. the Romans, for reasons which it is by no means easy to explain, this tendency is much stronger than among others. Wherever it has been strong, wherever these methods of putting oneself in right relation with the Power have been systematised by a central authority or priesthood, and thus made into religious law, there, as we might naturally expect, the performances and performers of magic have been most vigorously discountenanced and outlawed. The interests of religion and its officials are wholly antagonistic to those of magic and magicians. In civilised communities and in historical times magic is in the main individualistic, not social; magical ceremonies for the good of the community seem to be confined to races in a very early stage of development. The examples on which Dr. Frazer relies for his theory of the development of the public magician into a king84 are of this primitive kind, or are mere survivals of magic in a higher stage of civilisation—such survivals as there will always be among forms and ceremonies, of which it is man's nature to be tenacious. But religion, once firmly established, invariably seeks to exclude magic; and the priest does his best to discredit the magician, as claiming to exercise mysterious powers outside the pale of the legally recognised methods of propitiation and worship. As Dr. Tylor observed long ago, the more civilised the race, the more apt it is to associate magic with men of inferior civilisation.85 In the Jewish law, though magic was well known to the Jews and privately practised, there is no recognition of it; the magical books attributed to Solomon were suppressed, according to tradition, by the pious king Hezekiah.86 So too at Rome, where the outward forms of religion were also very highly systematised, magic, as it seems to me, was rigorously excluded from the State ritual, though it continued in use in private life under certain precautions taken by the State; in the few genuine examples of it in the rites belonging to the ius divinum (i.e. those used and sanctioned for the purposes of the community), it is nothing more than a survival of which the magical meaning was unknown to the writers from whom we hear of it.

A good example of such survivals is the curious ceremony of the aquaelicium, without doubt a genuine case of magical "rain-making"—one of the many inadequate and blundering attempts on the part of primitive man to obtain what he needs. Probably it may be classed under the head of "sympathetic magic," but the evidence as to what was done in the ceremony is not quite explicit enough to allow us to do this confidently.87 It was, of course, not included in the religious calendar, as it would be only occasionally called for, and could not be fixed to a day; but there is clear evidence that it was sanctioned by the State, for the pontifices took part in it, and the magistrates without the toga praetexta, and the lictors carrying the fasces reversed.88 A stone, which lay outside the walls near the Porta Capena, was brought into the city by the pontifices, so far as we can make out the details, and it has been conjectured that it was taken to an altar of Jupiter Elicius on the Aventine hard by, this cult-title of the god of the sky having possibly some relation to the technical name of the ceremony. What was done with the stone we unluckily do not know; but it has been reasonably conjectured that it was a hollow one, and that it was filled with water which was allowed to run over the edge, as a means of inducing the rain-god to suffer the heavens to overflow.89 It was called lapis manalis; and the epithet here can have nothing to do with the Manes, as in the case of another lapis manalis, of which I shall have a word to say later on, but must mean "pouring" or "overflowing." One or two other fragments of evidence point in the same direction, and I think we may fairly conclude that the rite was originally one of sympathetic magic—that as the stone overflowed, so the sky would pour down rain. In my Roman Festivals I have pointed out a remarkable parallel to this in the collections of the Golden Bough; in a Samoan village a stone represented the god of rain, and in a drought his priests carried it in procession and dipped it in a stream.

This parallel I owe to Dr. Frazer's wide knowledge of all such practices among savage peoples. But this ever helpful and friendly guide, in treating of the Jupiter Elicius concerned in this ceremony, has gone beyond the evidence, and attributed to the Romans another kind of magic of which I believe they were quite innocent. He has been led to this by his theory that kings were developed out of successful magicians. In his lectures on the early history of the Kingship90 he maintains that the Roman kings practised the magical art of bringing down lightning from heaven. "The priestly king Numa passed for an adept in the art of drawing down lightning from the sky.... Tullus Hostilius is reported to have met with the same end (as Salmoneus, king of Elis) in an attempt to draw down Jupiter in the form of lightning from the clouds." To support these statements Dr. Frazer quotes Pliny, Livy, Ovid, Plutarch, Arnobius, Aurelius Victor, and Zonaras—truly a formidable list of authorities; but without any attempt to discover where any of these late writers found the stories. Yet he had but to read Aust's admirable article "Jupiter" in the Mythological Lexicon91 to assure himself that legends which cannot be traced farther back than the middle of the second century B.C. cannot seriously be assumed to be genuinely Roman. Pliny happens to mention Calpurnius Piso as his authority; this was the man who is well known in Roman history as the author of the first lex de repetundis of the year 149 B.C., a good statesman, but as an annalist much given to indulging a mythological fancy.92 We happen to know that he wrote with happy confidence about the life and habits of Romulus, and a story about wine-drinking which he attributes to that king is obviously transferred to him from some more historical personage. Romulus would not drink wine one day because he was going to be very busy on the next. Then they said to him, "If we all did so, Romulus, wine would be cheap." "Nay, dear," he replied, "if every one drank as much as he wished; and that is exactly what I am doing."93 I quote the story simply as a good example of the way in which Roman historians could deal with their kings, and of the absolute necessity of acquainting oneself with their methods before building hypotheses upon their statements. I hardly need to add that another of Dr. Frazer's authorities, Arnobius, informs us that he took the story from the second book of Valerius Antias, a later writer than Piso, whose name is a byword even with the uncritical Livy for shameless exaggeration and mis-statement.94

But how did these writers come by such legends, which, as Dr. Frazer shows, are to be found also in Greece and in other parts of the world? Why should they have wished to make Roman kings into magicians? Rain-making we can understand at Rome,—it had a practical end in view, the procuring of rain for the crops,—but why lightning and thunder, which were so much dreaded that every bit of damage done by a thunderstorm had to be carefully expiated by a religious process? Rome is not in the tropics, where rain and thunder so often come together, and where an attempt to produce rain by magic might naturally include thunder, as in some of Dr. Frazer's examples from tropical lands. I entirely agree with the latest and most sober investigators of Roman ritual that this kind of magic is quite foreign to Roman ideas and practice;95 there is no vestige of it in the Roman cult; these stories must have come from outside. And there is every probability that they came from Etruria, where the lore of lightning had become a pseudo-science, a waste of human ingenuity, for the origin of which we must look, as we are now beginning to understand, to Babylonia and the Eastern magic.96 The Jupiter Elicius of the Aventine had nothing to do with lightning; he took his cult-title from the rite of aquaelicium; but as soon as the Romans began to interest themselves in the Etruscan lightning-lore, of which this electrical magic was only a part,97 they perverted the meaning of the epithet to suit their new studies, and began to attribute to their legendary kings powers which properly belonged to Etruscan or Oriental magicians. The second century B.C., when Piso wrote his Annals, is exactly the period when we should naturally expect such studies to come into fashion, and with such perversions of "history" as their consequence.98

I go on to note one or two more examples of real magic in the State religion; but they are hard to find. Pliny tells that even in his day people believed that a runaway slave who had not escaped out of the city might be arrested by a spell uttered by the Vestal virgins.99 I take this to mean that any one who had lost his slave might get the Vestals to use the spell as a means of keeping the runaway within the city. The word for spell is here precatio, i.e. a prayer, not carmen, which is the usual word for a spell; and Pliny evidently thinks of it as addressed to some god. But no doubt it was originally at least a genuine spell, of the same kind as others used in private life, which we shall notice directly; and it implies a belief in some magical power inherent in the Vestals, of whom we are told that if they accidentally met a criminal being led to punishment they might secure his release.100 As the spell in this case seems to be telepathic, i.e. an exercise of will-power projected from a distance, it may perhaps be paralleled with certain mystical powers exercised by women, especially when their husbands are at war, among some savage peoples;101 but we have no information about it beyond the passage in Pliny, and further guessing would be useless.

This last is a case of genuine magic, but it is outside the ritual of the State, though exercised by a State priesthood. Within that ritual there is one other very curious case of what must be classed as a magical process, and one that has accidentally become famous. At the Lupercalia on February 15, the two young men called Luperci, or, more strictly, belonging respectively as leaders to the two collegia of Luperci, girt themselves with the skins of the slaughtered victims, which were goats, and then ran round the base of the Palatine hill, striking at all the women who came near them or offered themselves to their blows, with strips of skin cut from the hides of these same victims. The object was to produce fertility; on this point our authorities are explicit.102 Thus this particular feature of the whole extraordinary ritual of the Lupercalia is unmistakably within the region of magic rather than of religion. Some potency was believed to work in the act of striking, though apparently without a spoken spell or carmen, such as usually accompanies acts of this kind; and this part of the rite, grotesque though it was, was allowed to survive by the grave religious authorities who drew up the calendar of religious festivals. It was probably a superstition too deeply rooted in the minds of the people to admit of being excluded; and, strange to say, it survived, in outward form at least, until Rome had become cosmopolitan and even Christian. The Lupercalia has always been a puzzle to students of early religion, and as each new theory is advanced, this strange festival is seized on for fresh interpretation;103 but for our present purposes it must suffice to point out that we clearly find embedded in it a piece of genuine magic, dating beyond doubt from a very primitive stage of thought.

There is one other very curious performance, occurring each year on the ides of May, which in my view is rather magical than religious, though the ancients themselves looked upon it as a kind of purification: I mean the casting into the Tiber from the pons sublicius of twenty-four or twenty-seven straw puppets by the Vestal virgins, in the presence of the magistrates and pontifices. Recently an attempt has been made by Wissowa to prove that this strange ceremony was not primitive, but simply a case of the substitution of puppets for real human victims as late as the age of the Punic wars.104 These puppets were called Argei, which word naturally suggests Greeks; and Wissowa has contrived to persuade himself not only that a number of Greeks were actually put to death by drowning in an age when everything Greek was beginning to be reverenced at Rome, but (still more extraordinary to an anthropologist) that the primitive device of substitution was had in requisition at that late date in order to carry on the memory of the ghastly deed. And the world of German learning has silently followed their leader, without taking the trouble to test his conclusions by a careful and independent examination of the evidence. It happens that this fascinating puzzle of the Argei was the first curiosity that enticed me into the study of the Roman religion, and for some thirty years I have been familiar with every scrap of evidence bearing on it; and after going over that evidence once more I can emphatically state my conviction that Wissowa's theory will not hold water for a moment. I shall return to the subject in a later lecture dealing with the religious history of the second Punic war; at present I merely express a belief that, whatever be the history of the accessories of the rite,—and they are various and puzzling,—the actual immersion of the puppets is the survival of a primitive piece of sympathetic magic, the object being possibly to procure rain. It is, in my opinion, quite impossible to resist the anthropological evidence for this conclusion, though we cannot really be certain about the object; for this evidence I must refer you to my Roman Festivals, and to the references there given.105

This rite of the Argei, then, was a case of genuine magic, and exercised by a State priesthood, virgins to whom certain magical powers were supposed to be attached; it was, I think, a popular performance, like one or two others which are also outside the limit of the Fasti,106 and was embodied in a more complicated ceremonial long after that calendar had been drawn up. In the ritual authorised by the State, with public objects in view, i.e. for the benefit of society as a whole, there is hardly a trace of anything that we can call genuine magic apart from the examples I have just been explaining. There were, I need not say, many survivals of magical processes of which the true magical intent had long been lost—ancient magical deposits in a social stratum of religion, which I shall notice in their proper place. This is not peculiar to the religion of the Romans; it is a phenomenon to be found in all religions, even in those of the most highly developed type, and it is one apt to cause some confusion as to the true distinction between magic and religion.107 It is easy to find magical processes even in Christian worship, if we have the will to do so; but if we steadily bear in mind that the true test of magic is not the nature of an act, but the intent or volition which accompanies it, the search will not be an easy one.

The modern French school of sociologists, which now has to be reckoned with in investigating the early history of religion, claims that magic was not originally, as we now see it, a matter of individual skill, but a sociological fact, i.e. it was used for the benefit of the community, as religion came to be in a later age. If this be true, as it very possibly is, we see at once how the dead bones of magical processes might survive, with their original meaning entirely lost, into an age in which higher and more reasonable ideas had been developed about the relation of Man to the Power manifesting itself in the universe. To take a single example from Rome, divination by the examination of a victim's entrails was originally a magical process, according to the opinion of most modern authorities;108 but it ceases to be magic when it is used simply to determine in the State ritual whether in a religious process the victim is perfect and agreeable to the deity. In fact magical formulae, magical instruments, unless they are used in the true spirit of magic, to compel, not to propitiate a deity, are no longer magic, and may be passed over here. When we come to discuss the ritual of sacrifice and prayer, of lustratio, of vows, of divination, we may find it necessary to recall what has here been said. On the whole, we may conclude that organised religious cult, from its very nature and object, everywhere excluded magic in the true sense of the word; it implies prayer and propitiation, both of which are absolutely inconsistent with the object and methods of magic. Religion is the product of a higher stage of social development; it is the expression of a real advance of human thought; and in telling the story of the religious experience of the Roman people we are but indirectly concerned with those more rude and rudimentary ideas which it displaced.

But in private life, outside of the organised cult of the State and the family, magic was all through Roman history abundant, even over-abundant, and in this form I cannot pass it over entirely. Though the State authorities seem to have taken pains to exclude it rigidly from the public rites, and though there is little trace of it in the religious life of family and gens, yet there is evidence that it was deeply rooted in the nature of the people, and that they must have passed through an age in which it was an important factor in their social life. This fact, taken together with its almost complete elimination from the public religion, throws into relief the persistent efforts of the State authorities, from the framing of the old religious calendar to the time of the Augustan revival, to keep their relations with the Power clear of all that they believed to be unworthy or injurious. No better example can be found of the inherent antagonism between religion and magic.

Private magic may be divided into two kinds, according as it was used to damage another, or only to benefit oneself. In the former case the State interfered to protect the person threatened with damage, and treated this kind of magic as a crime. The commonest form of it was that of the spell, or carmen, no doubt often sung, and accompanied by some action which would bring it under the head of sympathetic magic; but the spell alone is taken cognisance of by the State. Pliny has preserved three words from the XII. Tables which tell their own tale: "qui fruges excantassit."109 Servius, commenting on the line of Virgil's 8th Eclogue, "atque satas alio vidi traducere messes," writes, "magicis quibusdam artibus hoc fiebat, unde est in XII. Tabb. 'Neve alienam segetem pellexeris.'" These last words, with the verb in the second person, are probably not quoted exactly from the ancient text,110 but they help to show us the nature of this hostile spell. There must have been a belief that the spirit, or life, or fructifying power of your neighbour's crops could be enticed away and transferred to your own. This is confirmed by a remark of St. Augustine in the de Civitate Dei;111 after quoting the same line from Virgil, he adds, "eo quod hac pestifera scelerataque doctrina fructus alieni in alias terras transferri perhibentur, nonne in XII. Tabulis, id est Romanorum antiquissimis legibus, Cicero commemorat esse conscriptum et ei qui hoc fecerit supplicium constitutum?" Given the belief, the temptation can be well understood if we reflect that the arable land of the old Romans was divided in sections of a square, and that each man's allotment would have that of a neighbour on two sides at least.112 If one man's corn were found to be more flourishing than that of his neighbours, what more likely than that he should have enticed away the spirit of their crops? The process reminds us, as it reminded Pliny, of the evocatio of the gods of foreign communities, a rite which belongs to religion and not to magic, though it doubtless had its origin in the same class of ideas as the excantatio.

In more general terms the old Roman law (i.e. originally the ius divinum) forbade the use of evil spells, as we see in another fragment of the Tables, "qui malum carmen incantassit." In later times this was usually taken as referring to libel and slander, but there can be no doubt that the carmina here alluded to were originally magical, and became carmina famosa in the course of legal interpretation. Cicero seems to combine the two meanings in the de Rep. (iv. 10. 2) when he says that the Tables made it a capital offence "si quis occentavisset, sive carmen condidisset quod infamiam faceret flagitiumve alteri" (to bring shame or criminal reproach on another). In the later sense these carmina have a curious history, into which I cannot enter now.113 In the earlier sense they existed and flourished without doubt, in spite of the law; or it may be that, as the words of the Tables were interpreted in the new sense, the old form of offence was tolerated in private. "We are all afraid," says Pliny, "of being 'nailed' (defigi) by spells and curses" (diris precationibus).114 These dirae, and all the various forms of love-charms, defixiones, accompanied by the symbolic actions which are found all the world over, lie outside my present subject, and are so familiar to us all in Roman literature that I do not need to dwell on them.115

Nor of the common harmless kind of magic need I say much now. It survived, of course, alongside of the religion of the family and State, from the earliest times to the latest, as it survives at the present day in all countries civilised and uncivilised; and being harmless the State took no heed of it. Some assortment of charms and spells for the cure of diseases will be found in Cato's book on agriculture, and one or two incidentally occur in that of Varro.116 They performed the work of insurance against both fire and accident, and even such a man as Julius Caesar was not independent of such arts. Pliny tells us that after experiencing a carriage accident he used to repeat a certain spell three times as soon as he had taken his seat in a vehicle, and adds significantly, "id quod plerosque nunc facere scimus."117 Such carmina were written on the walls of houses to insure them against fire.118 Pliny has a large collection of small magical delusions and superstitions, many of which have an interest for anthropologists, in the 28th book of his Natural History.

Another kind of harmless magic, to which the Romans, like all Italians ancient and modern, were peculiarly addicted, is the use of amulets. Here there is no spell, or obvious and expressed exercise of will-power on the part of the individual, but the potent influence, mana, or whatever we choose to call it, resides in a material object which brings good luck, like the cast horse-shoe of our own times, or protects against hostile will-power, and especially against the evil eye. This curious and widely-spread superstition was probably the raison d'être of most of the amulets worn or carried by Romans. A modern Italian, even if he be a complete sceptic and materialist, will probably be found to have some amulet about him against the evil eye, "just to be on the safe side."119 A list of amulets, both Greek and Roman, will be found in the Dictionary of Antiquities, and in Pauly-Wissowa, Real-Encyclopädie, s.v. "amulet," and it is not necessary here to explain the various kinds in use in Italy; but I must dwell for a moment on one type, which had been taken up into the life of the family, and in one sense into that of the State, viz. the bulla worn by children, both boys and girls.

The bulla was a small object, enclosed in historical times in a capsule, and suspended round the child's neck. It was popularly believed to have been originally an Etruscan custom,120 and borrowed by the Romans, like so many other ornaments. It is, however, much more probable that the custom was old Italian (as indeed the "medicine-bag" is world-wide), and that the Etruscan contribution to it was merely the case or capsule, which was of gold where the family could afford it—gold itself being supposed to have some potency as a charm.121 The object within the case was, as Pliny tells us, a res turpicula as a rule,122 and this may remind us that a fascinum was carried in the car of the triumphator as medicus invidiae, to use Pliny's pregnant expression. The triumphing general needed special protection; he appeared in the guise of Jupiter himself, and was for the moment lifted above the ordinary rank of humanity. Some feeling of the same kind must have originally suggested similar means for the protection of children under the age of puberty. They also wore the toga praetexta, which, though associated by us with secular magistrates, had undoubtedly a religious origin. There are distinct signs that children were in some sense sacred, and at the same time that they needed special protection against the all-abounding evil influences to be met with in daily life.123 Thus this particular form of amulet became a recognised institution of family life, and in due time little more than a mark of childhood.

Yet another kind of charm must be mentioned here which was used at certain festivals, though apparently not at any of those belonging to the authorised calendar. At the Compitalia, Paganalia, and feriae Latinae we are told that small images of the human figure, or masks, or simply round balls (pilae), were hung up on trees or doorways, and left to swing in the wind.124 At the Compitalia the images had a special name, maniae, of which the meaning is lost; but inasmuch as the charms were hung up at cross-roads on that occasion, where the Lares compitales of the various properties had their shrine, it was not difficult to manufacture out of them a goddess, Mania, mother of the Lares.125 The common word for these figures was oscilla, and the fact of their swinging in the wind suggested a verb oscillare, which survives in our own tongue with the same meaning. Until lately it used to be believed that they were substitutes for original human sacrifices: a view for which there is not a particle of evidence, though it was originated by Roman scholars.126 Modern anthropology has found another explanation, which is by no means improbable. Dr. Frazer, in an appendix to the 2nd volume of the Golden Bough, has collected a number of examples of the practice of swinging by human beings as a magical rite; they come from many parts of the world, including ancient Athens, and even modern Calabria. He also points out that at the feriae Latinae the swingers seem to have been human beings, if we accept the evidence of Festus, s.v. "oscillantes"; thus we are left with the possibility that the oscilla were really imitations of men and women, though not of human sacrificial victims.

Dr. Frazer is obviously hard put to it to explain the original meaning and object of this curious custom. In the Paganalia, as described by Virgil in the second Georgic,127 the object would seem to be the prosperity of the vine-crop.