Sacred Texts  Classics  Index  Previous  Next 



A STUDY of the Roman prayers which have come down to us, whether they be found in their original integrity as in the Carmen Saliare, the Carmen Arvale, the prayer of the Umbrian Attiedii, those of the farmer recorded by Cato in his De Agricultura; or whether, like those embedded in the Fasti of Ovid, they have been modified by the poets to meet the requirements of meter--all will, we believe, show that they are, in their essential nature, magic incantations. Our problem is not whether prayer has developed from incantation or whether a spell is a degenerated prayer. 1 It would seem, after a detailed study of a large number of Roman prayers, that these always retained some of the characteristics of an incantation; and if there is any difference between the two it lies chiefly not in the prayer or incantation as such, but in the mental attitude of the person toward the object of his incantation or prayer, and in the consequent change in the tone and, in a limited degree, in the form of the incantation.

It will be necessary at the outset to define what we mean by "incantation." An incantation is a command (or, rarely, a wish), usually chanted, addressed to the subject of the magic rite. 2 The shepherdess in the song of Alphesiboeus in Vergil's eighth Eclogue repeats nine times the words "Lead Daphnis home from the city, my charms, lead Daphnis home." In certain rites of riddance described by Ovid  3 a child is to be protected by an incantation and a magic act. The protection is commanded in these words: "Birds of the night, spare the entrails of the boy." Such incantations containing a command are found commonly in private magic, and traces of their influence may be seen in the Roman State religion. If the spell was intended to harm a person, the State could interfere to protect him. 4 For instance, it was not unusual for a farmer whose crops had failed to accuse another of having, by a spell (carmen), lured the crops away. Tibullus, in a poem  5 in which he complains that an old beldame has bewitched Marathus, takes the opportunity to recount various feats of witches, such as transferring crops from one field to another. Similarly, Pliny the Elder records  6 that a certain freedman, Furius, by using better implements and better methods than his neighbor, was able to obtain richer crops from a smaller strip of land. A neighbor haled Furius before the tribes and accused him of having bewitched his field. But when they saw his sturdy slaves and his implements of witchcraft--hoes, rakes, and ploughs--they acquitted him.

The Romans of later days restricted the use of the term carmina to slander and libel, giving the names dirae and defixiones to such spells as we have just mentioned.

The tendency of the growing mind of early man was to personify the object addressed; and along with this tendency came another--to assign a spirit to the object (animism), causing the spell to change its character somewhat. But, as we have suggested, the difference between a spell and a prayer lies not so much in any inherent change in the nature of either as in a shift in the attitude of mind toward the object to be influenced and in the consequent alteration in the tone of the prayer. In magic the process is purely mechanical. The person performing the rite wills a certain effect which is bound to ensue if the magic act and the incantation have been flawless. The volition lies with the person. In the case of prayer, however, the worshiper addresses a divinity, all-powerful in his sphere, whose will he must win just as he would win the will of a person. Here, then, lies the fundamental difference between incantation and prayer, so far, at least, as the mental attitude of the person is concerned: in the incantation the will of the person, in the prayer the will of the divinity, determines the effect. Furthermore, when with the passing of the years ancient religions degenerated into mere forms without meaning, and the notion of the divinity involved in the rite was lost, the prayer once more took on the nature of a spell.

We have a clear illustration of this degenerative tendency in the case of the curious goddess Carna who in ancient times had control over the vital organs of the human being. Carna was a recognized divinity of the Roman State religion; but in the popular religio she was considered a witch. The goddess has survived in modern Tuscany where she is known as Carradora, a kindly spirit with substantially the same functions as the ancient Carna. 7

We shall attempt in this chapter to show that a Roman prayer and a Roman incantation had six elements in common and that in only one essential point were they different. In order to do this we must first notice the characteristics of a Roman incantation and then see whether these characteristics are to be found in the Roman prayer.

As characteristics of the Roman incantation we may mention the following: (1) It was in the form of a command; (2) it was chanted; (3) it was uttered in an undervoice; (4) in order to be effective it must be repeated; (5) the wording of the incantation must be exact; (6) the usual purpose of the incantation was secretly to secure evil ends; (7) no god was involved in the incantation.

Prayer a Command

In magic rites, the incantation is almost always in the form of a command. Thus a charm for driving away gout reads: 8 "Away, away from my feet and from all my limbs, gout and every muscular pain." We have seen in the rites of Crane, the witch who possessed the power to keep off bloodsucking vampires, that these baleful creatures are addressed in the form of a command. 9 This direct address in the form of a command is found in rites of the State religion. At the Festival of the Spirit of the Mildew (Robigalia), the Priest of Mars commands Robigus to spare the young blades of grain. 10 Again, the shady merchant who desired to rid himself and his wares of evils attending a questionable deal commanded the waters of the Spring of Mercury to wash away his past perjuries. 11 In the rites of treaty-making preserved by Livy  12 the formula begins with a command: "Hear, Jupiter, hear, spokesman of the Alban People, hear, Alban People." In declaring war, the fetial began: 13 "Hear, Jupiter, and thou, Janus Quirinus, and all ye gods celestial and terrestrial and ye gods infernal, hear." Romulus in vowing a temple to Jupiter prayed: "But do thou, father of gods and men, at least keep the enemy from here, take away terror from the Romans, and stay their foul flight."  14

Prayers Chanted

That incantations were chanted is a matter of common observance and scarcely needs illustration. That the word carmen means "song" is evidence of this. For example, in magic rites, the purpose of which was to induce a dislocated or broken bone to come together, the incantation was sung (cantare). 15 Again, a witch composed a charm for Tibullus, to be chanted three times, after which he had to spit; then Delia's husband would believe gossip about other lovers of Delia, but not about her and Tibullus. 16

The two oldest prayers of the Romans which we possess--the Carmen Arvale and the Carmen Saliare--were both chanted. Livy writes  17 that "the leaping priests went through the city chanting their hymns." There is reason to believe that the old prayers which Cato has preserved for us in his treatise on agriculture were originally in metrical form; but in the directions given to the worshiper the verb dicito, and not cantato, precedes the prayer, showing that, in Cato's time at least, such prayers were "spoken" rather than "sung." However, these prayers, even in the form in which they are found in Cato, are predominantly spondaic, in keeping with the slow movement of the chant and with the solemn religious character of the rites. In ceremonies intended to bring thunderbolts down from the sky, incantations were used. 18

Prayers Uttered in an Undervoice

Normally prayers were uttered aloud by the Romans. But we have reason to believe that at times they were chanted in an undervoice. 19 We know that magic rites were commonly accompanied by an incantation which was either sung almost inaudibly or muttered. 20 The prayer of the Fratres Attiedii at Iguvium was so repeated, as were the prayers recorded in Cato. Cicero definitely states that men pray silently to the gods, but he does not tell us the sort of prayers thus offered. 21 Horace, however, sheds some light on the question where he satirizes the man whom the people believe to be good, who sacrifices a pig or an ox and cries to Janus and to Apollo that all may hear, but in an undervoice prays to Laverna, the goddess of theft, that he may be able to deceive his fellows. 22 In the country festival described by Tibullus, 23 the worshiper is bidden to call upon the god audibly for the flocks but in undertones for himself. Juvenal represents  24 a mother praying in an undervoice for beauty for her sons but in louder tones for the same gift for her daughters. Persius complains  25 that his fellows petitioned aloud in the temples for a sound mind and an honorable name, but inaudibly for the death of a relative or the finding of treasure. Seneca laments  26 the fact that men in their madness whisper the vilest prayers to the gods. Finally, the Chief Priest, while lashing an errant Vestal, prayed inaudibly. 27 It seems, then, that prayers were normally uttered aloud; but when a worshiper wanted to pray for evil (as would be the case also in a magic rite) he uttered his prayers so that no one but the god could hear.

The fact that people prayed for things of which they were ashamed may explain why they whispered their prayers; but it will not explain why the priests mumbled their prayers. The reason probably lies in the fact that they were of magical significance, and if they reached the ears of the unholy or the stranger

they might prove dangerous. Thus Ovid informs us  28 that it was considered impious for one to know the charms and arts by which Jupiter was drawn down from the sky. That the priests repeated their prayers inaudibly is shown by a passage in Ovid  29 where the poet bids the reader stand at the side of the priest officiating at the Carmentalia; by keeping thus close, he will hear the priest mutter two names not before known to him, referring to Porrima and Postverta, two goddesses who had control over the manner of birth of the child, whether head or foot foremost.

Prayers Repeated

Again, repetition characterized the magic incantation. For instance, the incantation of the lover in Vergil's eighth Eclogue, already referred to, was repeated nine times; the incantation which the witch formulated for Tibullus had to be uttered three times. 30 At the conclusion of the prayer to Pales we read the words:  31 "With these words the goddess must be appeased. So do you, facing the east, utter them four times. . . ." The verses of the Carmen Saliare were each chanted three times, as the Leaping Priests of Mars danced in threefold measure. Fowler, who on the whole is not inclined to identify spell and prayer, writes  32 that the verses "seem certainly to belong rather to the region of magic than of religion proper." Repetition was also characteristic of the Carmen Arvale and the prayer of the Fratres Attiedii.

Exactness in Naming the God and in the Wording of the Prayer

To the savage mind, merely mentioning the name of the person whose will was to be influenced was sufficient to work a spell. It was necessary for the name to be correct; and from this habit of the age of magic arose the scrupulous exactness with which the Romans addressed their divinities. If the divinity had several names, the worshiper would address him by them all. Horace, for example, invokes Ilithyia (Diana) to protect mothers and adds, "whether you prefer to be addressed as Lucina or as Genitalis."  33 Again, Horace bids Father Matutinus begin his song; but he adds cautiously, "or, if you prefer the name, Janus."  34 Lucius in Apuleius' Metamorphoses 35 addresses the Queen of Heaven as Ceres, Venus, the sister of Phoebus, Proserpina, or "by whatever name, with whatever rite, in whatever appearance it is right to invoke thee." In the formula of devotio prescribed in Macrobius we read:  36 "Dis Pater, or by whatever name it is right to address thee . . . " Again, in Servius, we read  37 that the pontiffs prayed: "Jupiter Optimus Maximus, or by whatever name you wish to be addressed." Finally, if the sex of the divinity was unknown, they would add "whether god or goddess, male or female."  38 Thus the sex of Pales and Pomonus (Pomona) was indeterminate.

Not only was it necessary for the divinity to be properly addressed, but the prayer had to be uttered exactly as prescribed for the sacrifice involved; if there was a mistake in the wording, the whole rite had to be repeated or an atoning sacrifice had to be made. 39 We learn from the accounts of the Fratres Attiedii that if the brethren made an error in the prayer they had to repeat it. Juvenal represents  40 a woman standing before an altar with veiled head praying that her paramour may win a prize for his lyre-playing. The words she utters are dictata--repeated after the priest.

Prayers for Ill

Magic acts were more often calculated to do harm than good. Such magic practices were rampant as early as 450 B.C., as is evidenced by the provisions against them in the Laws of the Twelve Tables. Rome thus officially looked with disfavor upon magic intended to harm, although magic principles of similarity and contact were to be found in almost every rite of the Romans. Servius states the attitude of a much later generation of Romans toward the art when he writes: "While the Romans adopted many rites, they always condemned magic, for it was considered a base art."  41

We shall give a few examples of prayers for evil, gathered from various Latin authors. Catullus prays to the gods to inflict evil upon one who has harmed his friends. 42 Vergil offers us three illustrations of prayers for evil--at least from our point of view. As the body of his son lies at his feet, slain by the sword of Pyrrhus, Priam prays  43 the gods to bring a like fate upon the murderer. Evander invokes  44 the gods that the murders and other high-handed acts of the tyrant Mezentius be visited in kind upon the Etruscans themselves. Ascanius, before entering into combat with Remulus, calls 45 upon Jupiter to aid him. In return for the god's assistance Ascanius will offer sacrifice at his altar. We have already seen how the Roman of Horace's day would pray for success in theft and that his sins and cheadngs might be cloaked. Propertius, like many of his fellow Romans, was quite willing to pray for evil. When, for example, a Roman praetor has displaced him in Cynthia's favor, Propertius prays: "But do you now, Venus, aid me in my grief, that he (the praetor) may destroy himself through his persistent lechery."  46

The story is told in Velleius  47 that Merula, who, before Cinna's arrival in Rome, had abdicated his office as consul, having opened his veins with suicidal intent, implored the gods to vent their wrath on Cinna and his party, a petition which might well have been made in the form of an incantation. In Juvenal's satires we have many instances of prayers offered to the gods for questionable ends: riches, the largest money chest in the whole city, beauty for one's sons, a prize for lyre-playing at the Capitoline Games. 48 Petronius laments  49 the fact that because of the degeneracy of the times the people pray, not for eloquence or the blessings of philosophy, but for the death of a rich neighbor or the unearthing of buried treasure. Persius, like Juvenal, was wholly out of sympathy with the things for which his fellows prayed in their temples: while they petitioned aloud for a sound mind, a fair name, and trustworthiness, their real prayers were for the death of a kinsman or the finding of a treasure. Such people, in order to make their prayers acceptable, would plunge their heads two or three times in the holy waters of the Tiber.  50 Sejanus, the commander of the praetorian cohort under Tiberius, prays  51 for honors and for wealth; but in these he finds, as it were, just so many stories of a lofty tower from which he is doomed to plunge all the farther to his destruction. The Lares to whom Tibullus prays with tenderness and affection, Juvenal represents the Romans of his day invoking as follows:  52 ". . . Little Lares of mine, whom I usually entreat with flakes of incense or spelt or slender garland, when shall I bag some game to give me security for my declining years, to protect me against the mat and the staff of the mendicant?" The objects of this man's prayer include twenty thousand sesterces in interest, small dishes of plain silver, two sturdy slaves, a "stooping engraver" (curvus caelator), and a painter. These he considers a wretched return for his piety.

No God Involved

We come at this juncture to the one element in a prayer which distinguishes it from an incantation. In a prayer the worshiper addresses a divinity, all-powerful in his sphere, whose will must be won by sacrifice and prayer. In magic, on the other hand, no god is involved. 53

Of course we must recognize that a prayer, in which a god is addressed, may possess one or all of the characteristics of an incantation; and the more of these elements it contains the closer it will be to pure incantation. There will also be corresponding changes in the psychology of the worshiper; for in the stage between the pure incantation, where no god is involved, and the prayer, in which a god is invoked, there must be times when the worshiper is uncertain whether he himself controls the result or whether there is a controlling force superior to him.

It will be necessary to show how the idea of god, all-powerful in his particular field, developed. In the age of magic, man believes that there is a mysterious quality (mana) residing in things and that this quality, if good (positive mana), can be made useful to him by the performance of a magic act of compulsion, accompanied by an incantation or charm; or, if it be harmful (negative mana), it must be avoided; that, when this is impossible, rites of purification must be performed to rid one of the contagion of the tabooed thing. Gradually, with the failure of magic to do what was expected of it, another stage developed in which spirits, not unlike man himself, with human emotions much like his own--spirits not yet developed into gods in the usual sense of that word--were thought to invest the things about him; and they had to be induced to serve him in much the same way as he would induce his friends to assist him, by entreaty and by offering gifts. There was a term for these undeveloped gods, numina; and the Romans never got far beyond this stage in their conception of gods.

We have tried to show in this chapter that the Roman prayers which have come down to us have certain elements in common with the incantation: they were chanted; they were usually in the form of a command, often uttered in an undervoice; they were, either in whole or in part, repeated; the wording had to be exact; the purpose of many of them was evil. We have, further, tried to show that there was one definite advance in prayer over the incantation, which was the result of a change in the mental attitude of the worshiper; that whereas no divinity was invoked in the incantation, in the prayer a numen or a fully developed god was invoked. Finally, in magic, the volitional element appears in the person; in religion, it rests with the divinity.


1 See W. Warde Fowler, The Religioms Experience of the Roman People, pp. 185-187.

2 See F. B. Jevons, The Idea of God, p. 115ff.

3 Fasti VI. 155-162.

4 Apuleius, Apologia XLVII.

5 I. 8, 17-23.

6 Naturalis Historia XVIII. 41-42.

7 Macrobius, Saturnalia I. 12, 31-33; see C. G. Leland, Etruscan Roman Remains, P. 108.

8 Marcellus, De medicina XXXVI.

9 Macrobius, Saturnalia 1. 12, 31-33.

10 Ovid, Fasti IV. 911.

11 Ovid, Fasti V. 681-682.

12 I. 24, 7.

13 Livy I. 32, 10.

14 Livy I. 12, 5.

15 Cato, De Agricultura CLX.

16 I. 2. 41-62.

17 I. 20, 4.

18 Ovid, Fasti 111. 323-325.

19 See Pease's note on Cicero, De Divinatione I. 57, 129.

20 See Lucan, Bellum Civile VI. 685-686.

21 Cicero, De Divinatione I. 57, 129.

22 Epistulae I. 16. 57-62.

23 II. 1. 84.

24 X, 289-292.

25 II. 6.

26 Epistulae Morales X. 5.

27 Plutarch, Nyma X.

28 Fasti III. 323-325.

29 Fasti I. 631-632.

30 I. 2. 54.

31 Ovid, Fasti IV. 777-778.

32 The Religious Experience of the Roman People, p. 187.

33 Carmen Saeculare 13-16.

34 Sermones II. 6. 20-23.

35 XI. 2.

36 Saturnalia III. 9, 10.

37 On Vergil's Aeneid II. 351.

38 Cato, De Agricultura, CXXXIX.

39 See Cato. De Agricultura CXXXIX.

40 VI. 390-392.

41 On Vergil's Aeneid IV. 493.

42 XXVIII. 14-15.

43 Aeneid II. 535-539.

44 Aeneid VIII. 484.

45 Aeneid IX. 625-629.

46 III. 8, 13-14.

47 II. 22, 2.

48 See VI. 385-388; X. 23-25; X. 289-292.

49 Satyricon LXXXVIII.

50 Persius II. 3. 16.

51 Juvenal X. 243-245.

52 Juvenal IX. 137-140.

53 See, for example, Lucan, Bellum Civile VI. 523-526.

Next: Chapter VII: Naturalism and Animism