Digitized by Microsoft® ?A ?12 CORNELL UNIVERSITY LIBRARY Digitized by Microsoft® 9.^t^ r' Cornell University Library PA 2027.F78 Roman essays and interpretations 3 1924 021 608 975 Due ^ fiP Palo-aseHt fi^^^mdm LOA IV 200i- Digitized by Microsoft® This book was digitized by Microsoft Corporation in cooperation witli Cornell University Libraries, 2007. You may use and print this copy in limited quantity for your personal purposes, but may not distribute or provide access to it (or modified or partial versions of it) for revenue-generating or other commercial purposes. Digitized by Microsoft® ROMAN ESSAYS AND INTERPRETATIONS By W. WARDE FOWLER M.A. HON. Lh.D. EDINBURGH, ^c. Author of The Roman Festivals of the Republic, Social Life at Rome in the Age of Cicero, The Religious Experi'mce of the Roman People, Rom^n Ideas of Deity in the Last Century of the Repiiblic, &c. OXFORD AT THE CLARENDON PRESS 1 920 Digitized by Microsoft® OXFOKD UNIVEKSITY PRESS LONDON EDINBURGH GLASGOW NEW YOEK TORONTO MELBOURNE CAPE TOWN BOMBAY HUMPHREY MILPORD PUELISUER TO THE UHTVERSITY /i7f/7GI Digitized by Microsoft® PREFATORY NOTE I HAVE included in this selection of papers none that were simply critical of the work of others : only those in which I seemed to myself, rightly or wrongly, to be moving towards some fairly definite conclusion on points of permanent interest. Should I be criticized for including some short and apparently trifling papers which I have called ' parallela quaedam ', I should reply that I like to show that the apparently marvellous may be sometimes wholly or in part authenticated by modern parallels. At the end I have placed character sketches of two great Roman historians, Niebuhr and Mommsen, and an essay on the Julius Caesar of Shakespeare. Whether I am right in reprinting and revising papers, many of which were written long ago, the critics will decide. I can only say that they are fragments of work into which I have put my best abilities, and in the writing of which I have found much pleasure, whether in the hurry of a busy tutorial life at Oxford, or in the leisure of old age in the country. I have to acknowledge gratefully the permission of the Council of the Society for the Promotion of Roman Studies, and of the Council of the Classical Association, to reprint papers originally published in their Journals, A considerable part of the material of the volume has not been published before. I am greatly indebted to my old friend Mr. P. E. Matheson for kind help in reading the proofs. W. W. F. KiNGHAM, June 19, 1919. A 2 Digitized by Microsoft® Digitized by Microsoft® CONTENTS PART I PAGE The Latin History of the woed Religio. (TransacUons of the Congress for the History of Religions, 1908) . 7 The Original Meaning of the word Sacer. {J. R. 8., 1911) 15 MxjNDUs Patet. (/. R. S., 1912) . • . - „ 24 The Oak and the Thunder-god .... 37 The Religious Meaning of the Toga Peaetexta of Roman Children 42 Was the Plaminica Dialis priestess of Juno ? (CI. Rev., 1895) 52 The Origin of the Lar familiaris .... 56 FoRTUNA Primigenia 64 Passing under the Yoke. {CI. Rev., 1913) . . 70 Note on privately dedicated Roman Altars . . 75 The Pontifices and the Periae : the Law of Rest- days ......... 79 PART II On the Date of the Rhetorica ad Herennium . 91 The Lex frumentaria of Gaius Gracchus. {E. H. R., 1905) . . 99 The Carmen saeculare of Horace and its First Per- formance. (CI. Qu., 1910) Ill On the Laudatio Turiae and its Additional Frag- ments. (CI Rev., 1905) 126 An Unnoticed Trait in the Character of Julius Caesar. (CI. Rev.j 1916) 138 Digitized by Microsoft® 6 CONTENTS PART III PAGE Ancient Italy and Modern Boeneo. (J. B. S., 1916) 146 Parallbla quaedam : . . • • • .165 The plague of locusts in 125, and a modern parallel 165 Plagues of field-voles in ancient and modern times 167 ' Armati terram exercent ' : and a modern parallel . 169 The disappearance of the earliest Latin poetry : and a modern parallel . . . ■ • .171 Roman Leges datae and English Enclosure awards 173 The Geeat Seepent oe the River Bagradas . . 178 PART IV Vergiliana : The Swans in Aen. i. 390 ff. The Harbour in iii. 633-6. Note on Dido and Aeneas, Aen. v. 5-6. On the word nefas in v. 197. Notes on Aen. ix, x, and xi 181 Notes on Horace, Odes, iii. 1-6 .... 210 Beethold Georg Niebtjhr : a sketch . . . 229 Theodor Mommsen : his life and work . . . 250 The Tragic Element in Shakespeare's Julius Caesar 268 INDEX OF LATIN WORDS AND PHRASES . 289 Digitized by Microsoft® THE LATIN HISTORY OF THE WORD RELIGIO This word, which in its modern form is in use all over Europe, had a remarkable history in its own Latin speech and literature. That history seems to me to have more than a mere linguistic interest, and I propose in this paper to indi- cate in outline where that interest lies. Of the much disputed etymology of the word I will only say this : that the question stands now very much as it did in the time of Cicero and Lucretius, who took conflicting views of it. Professor Conway, whose authority is great, tells me that apart from the evidence of usage and the feeling of the Romans themselves, there is nothing to decide whether it is to be connected with ligare, to bind, as Lucretius thought, or with legere, to string together, arrange, as Cicero believed. His feeling is in favour of Cicero's view, as less prejudiced than that of Lucretius ; so is mine. But our feelings are not of much account in such questions, and I may pass on at once to the history of the word. • In Latin literature down to Christian times, religio is used in a great variety of senses, and often in most curious and unexpected ones ; but all these uses can, I think, be reduced to two main types of meaning, one of which is probably the older, the other derivative. The one reflects the natural feeling of the Latin when face to face with the supernormal or supernatural, before the State with its priesthoods and religious law had intervened to quiet that feeling. The other expresses the attitude of the citizen of a State towards the supernatural, now realizable without fear or doubt in the shape of the recognized deities of his State. I must explain these two uses to begin with. I. Religio is the feeling of awe, anxiety, doubt, or fear, which is aroused in the mind by something that cannot be explained by a man's experience or by the natural course of Digitized by Microsoft® 8 THE LATIN HISTORY OF cause and efEact, and which is therefore referred to the super- natural. This I take to be the original meaning of the word, for the following reasons : 1. Religio is not a word which has grown out of any State usage, or been rendered technical by priestly law or ritual. It has no part in the ius divinum, like the word sacrum : we search for it in vain in the indices to the Corpus Inscriptionum, where it would inevitably be found if it were used in a technical or legal sense. In its adjectival form, as applied to times and - places, we may also see the results of this non-technical m Baning. Dies religiosi, loca religiosa, are not days and places which are proclaimed as such by the official administrators of the ius divinum : they are rather such days and places as man's own feeling, indep3ndently of the State and its officials, has made the object of religio. ' Religiosum stands in contrast with sacrum as indicating something about which there is awe, fear, scruple, and which has not been definitely brought within the province of State law, nor handed over to a deity by ritualistic formulae.' ^ If this be so, then we may safely refer the origin of the word to a pjriod when powerful State priesthoods had not as yet, by ritual and routine, soothed down the natural awe which in less perfect social forms man feels when obstructed, astonished, embarrassed, by that which he cannot explain or overcome. 2. That this is the true and the oldest meaning of the word seems also proved by the fact that it survived in this sense throughout Latin literature, and was indeed so used by the ordinary Roman layman. It is familiar to us in a thousand passages. Religio may stand for a doubt or scruple of any kind, or for anything uncanny which creates such doubt or scruple. To illustrate this I may select a single passage from Caesar, as a writer who would be sure to use a word in a sense obvious to every one. In describing the alarm of the soldiers of Q. Cicero when besieged at Aduatuca, he says : Alius castra iam capta pronuntiat, alius deleto exereitu atque imperatore victores barbaros venisse contendit ; pleri- ' See a paper by the writer in the Hibbert Journal for 1907, p. 847. Digitized by Microsoft® THE WORD RELIQIO 9 que novas sibi ex loco religiones fingunt, Cottaeque et Titurii calamitatem, qui in eodem occiderint castello, ante oculos ponunt.^ Here Caesar might almost as well have simply written metus instead of religiones ; but he wishes to express not only natural fear and alarm as to what may happen, but that fear accentuated by the sense of something wrong or uncanny, for which the soldiers or their leaders may be responsible — in this case the pitching of a camp in a place which they believed to have been the scene of a former disaster. Let us note that these soldiers were out of reach of the protecting arm of their own ius divinum : they were on foreign soil, ignorant of what supernatural powers might be present there. Their commander-in-chief, it is true, was the chief adminis- trator of that ius. Caesar was pontifex maximus : but Caesar was not there, and if he had been, his presence would in those days and in such a place have made little difference. They are in the same position towards the supernatural as their ancestors had been before the State arose, and in describ- ing their alarm Caesar uses the word religio in the same sense in which it had come into use in those primitive ages. Livy, writing of a p3stilence and its moral effects, says that ' nee corpora modo affecta tabo, sed animos quoque multiplex religio et pleraque externa invasit ' : ^ where by religio he means the feeling of anxiety which took practical shape in the performance of various rites, foreign for the most part Such examples could be multiplied a hundredfold : and the word came at last to be used for anything that produces a feeling of wonder or even of curiosity, seeing that we do not understand it. Thus Pliny says that there is a religio in men's knees, because we kneel on them to supplicate, and clasp the knees of those from whom we ask mercy ; ^ there is something uncanny about that part of the body — something we cannot explain. In the same way he says that no animal is ' religionis capacius ' than the mole, because its heart and its teeth are supposed to have some mysterious medicinal powers.* ' B. 0. vi. 37. ' iv. 30. ' H. N. xi. 250. ' Ibid. xxx. 19. Digitized by Microsoft® 10 THE LATIN HISTORY OF In this way the adjective religiosus came to be applied to human beings in a sense not far removed from that of super- stitiosus, which is, so far as I know, always used of persons addicted to rites or fancies outside the pale of the Roman State-religion. This sense seems to be an early one : it occurs in the fragment of an ' antiquum carmen ' quoted by Aulus Gellius : ^ ' Religentem (attentive) esse oportet, religiosus (over-anxious) ne seis.' Lucretius' use of the substantive may also be mentioned in this context : for him all that we call religion was superstitious and degrading, and could therefore be properly called by that word which the Romans invariably used to express their doubts, fears, and scruples. Lastly, before I go on to the second chief meaning of the word, I may mention the significant fact that religio is never personified as a deity, as were Pietas, Sanctitas and almost all the virtues at one time or another. It is not a virtue : it does not necessarily lead to a definite course of action, and embodies no sense of duty or moral value : it is primarily and essentially a feeling to which human nature is liable under certain circumstances. II. I now come to the seeond chief sense in which the word is used, and which brings it a step nearer to our own use of it. This sense was mainly due, I think, in Roman literatiu'e to Cicero, though it may be far older in common use : and is perhaps the result of the Greek originals, e. g. Posidonius, whom he was following when writing the de Legibus and the de Natura Deorum, &c. ; but this is a point which I must here pass over. From Cicero in any case I can best illustrate this new turn of meaning which the word acquires. When Cicero was a young man, not yet too learned or philosophical, he defined the word clearly according to its common usage, with an addition of some importance. ' Religio est quae superioris cuiusdam naturae, quam divinam vocant, curam caerimoniamque affert ; ' ^ i. e. a feeling of awe that inevitably suggests the discovery of the proper rites by which the object of that feeling may be propitiated. But later on ' iv. 9. 1 ; of. Baehrens, Fr. Poet. Rom., p. 36. ' De Invent, ii. 161. Digitized by Microsoft® THE WORD RELIGIO 11 in his life, in the second book of the de Legibus, which deals with the State religion, he uses the word with much freedom of the particular cults, or all of them together, which are the result of the feeling. Thus in x. 25 * sues deos aut novos aut alienigenas coli confusionem hahet religionum ', i. e. private persons may not introduce new cults ; for there would in that case be a confusion both of religious feeling and duty. In X. 23 he calls his own imaginary ius divinum a constitutio religionum, a system of religious duties. Thus the word is passing into the sense of the forms of cult, as ordered and organized by the State, the feeling, the religio proper, being only aroused when scruple is felt as to the accurate performance of these rites. In vii. 15 we read ' qua mente, qua pietate colat religiones ', where it answers almost exactly to religious duties. In xvi. 40 he tells how the Athenians consulted the Delphic oracle ' quas potissimum religiones tenerent ', and the answer was, ' eas quae essent in more maiorum '. Again in xi. 27 we find ' religio Larium ', the cult of the Lares. But the feeling which prompts the cult, and which is aroused afresh if it be neglected, is seldom entirely absent. The phrase religio sepulcrorum (xxii. 55) suggests quite as much the feeling as the ritual : and a little further down we are told that the pontifical law of burials ' magnam religionem eaerimoniamque declarat ' — the word caerimonia being necessary to express the ritual following on the feeling. And lastly this word may- be used to gather up and express in totality a number of acts of cult, because the same feeling is at the root of them all. Thus in xix. 47 the question is raised whether a pontif ex should know the civil law. The answer is, ' quod cum religione coniunctum est : de sacris, de votis, de feriis, de sepulcris ', the pontifex has to do with these matters, which can all be expressed together by the word religio. These examples seem to show how the word might pass into the sense in which we still use it ; the feeling which prompts us to worship, and also the forms under which we perform that worship*. The feeling is common to human nature, civilized or not : that is the original meaning of the Digitized by Microsoft® 12 THE LATIN HISTORY OF word : the worship, organized by a priesthood, is the work of the State— that is the second, or as we may call it, the Ciceronic meaning. And in the same age it is also so used by Lucretius, who includes under it all that was for him the world's evil and folly, i. e. both the feeling and the cult— delusion, myth, superstition, as well as the organized but futile worship of the family and the State. ' Tantum religio potuit suadere malorum.' ^ In an age of cosmopolitanism, when the old local character of the cults was disappearing, and in an age of philosophic-religious syncretism when men like Posidonius, Cicero, Varro, and others were thinking and writing about the, nature of the gods and kindred questions, a word was wanted to gather up and express all this religious side of human life and experience : it must be a word without a definite technical meaning, and such a word was religio. To take a single example, besides those already quoted from Cicero, there is the famous aphorism which St. Augustine ^ ascribes to Varro : ' expedit falli in religione civitates.' Thus while religio continues to express the feeling only, or the cult only, if called on to do so by Latin writers, it gains in the Ciceronian age a more cbmprehensive connotation, as the result of the contemplation of religion by philosophy as a thing apart from itself ; and this, as we shall see directly, enabled the early Christian writers, who knew their Cicero well, and modelled their prose on his, to use it in much the same sense as that in which we use it to-day. Time fails to trace the word in the pre-Christian literature- of the early Empire, and to see how it is affected by the finer quasi-religious Stoicism, or again by the Caesar -worship of the day, — the nearest approach in antiquity, as it has been called, to a cosmopolitan religion. So far as I can see, it did not take from either of these sources any new turn or type of meaning. Seneca, for example, has but little use for it ; though he was, as Professor Dill has said of him, one of the few heathen moralists who warm moral feeling with the emotion of modern religion, he had little real interest either ' Luor. i. 101 2 0i^ j)gi^ jy_ 27. Digitized by Microsoft® THE WORD RELIGIO 13 in the feeling or the cult. If he made himself a religion out of his Stoic principles, it was not one that he could have described by the word religio. For him, though tinged by emotion, it was still sapientia : he could hardly have assented to the later teaching of Lactantius ^ that sapientia and religio are inseparably connected. Nor did the worship of the Caesars bring any new turn of meaning : here it could express the cult (' caelestes religiones ' ^), but the feeling at the root of a genuine religious cult was not there to be expressed. This is perhaps significant both of the true meaning of the word, and also of the weak point in Caesar-worship : but I must not now dwell upon it. I will only mention one passage in which Pliny the Younger uses it of the cult of Trajan, because the kind of feeling which it there represents— loyalty and devotion to an individual — ^is in some sense a new one, and may be a foreshadowing of the Christian use. Pliny writes to Trajan from Bithynia reporting celebrations on the Empef or's birthday : ' Diem . . . debita religione celebravimus, com- mendantes dis imperii tui auctoribus et vota publica et gaudia.' * Here it means the feeling of devotion prompting the ' vota et gaudia ', as well as those acts themselves. There is nothing in it of the old fear, scruple, anxiety : it is the devotion and gratitude which expresses itself in religious festivities. But there was to be a real change in the meaning of the word, the last but one in its history. The second century A. D. was that in which the competition was keenest between various religious creeds and forms, each with its own vitality, and each clearly marked off from the others. It is no longer a question of religion as a whole contemplated by a critical or a sym- pathetic philosophy : the question is, which creed and which form is to be the true and the victorious religion. Our wonder- ful word again adapts itself to the situation. Each separate religious system can now be called a religio.* The old poly- theistic system can now be called religio Deorum by the ' De Vera Sap. iv. 3. ^ Tac. Ann. L 10. ' Ep. X. 102. « ' e cohorte rcligionis unus,' Apul. xi. 14, of Isis. Digitized by Microsoft® 14 THE LATIN HISTORY OF Christian, while his own creed is religio Dei. In the Octavius of .Minucius Felix, written probably in the first half of the second century A. D., the word is already used in this sense. His nostra religio, vera religio, distinguished from all other religiones, is the whole Christian faith and Christian practice as it stood then ; the depth of feeling and the acts which give it outward form. The one true religion can be expressed by this word, though it is quite different from anything the word has as yet been called on to mean. In Lactantius, Arnobius, TertuUian, this new sense of the word is to be found on almost every page : but a single noble passage of Lactantius must suffice to illustrate it. ' The heathen sacrifice,' he says, ' and leave all their religio in the temple ' : thus it is that such religiones cannot make men good, or firm in their faith. ' No- stra religio eo firma est et solida et immutabilis, quia mentem ipsam pro sacrificio habet, quia tota in animo colentis est.' ^ Religio here is not awe only or cult only, or scruple about details of cult, but a mental devotion capable of building up character. ' The kingdom of God is within you.' It is worth noting that it can now be explained by the word pietas, which was not possible in the old days, because pietas was a virtue and religio was not a virtue but a feeling. Lactantius says that philosophy, ' quae veram religionem, id est summam pietatem, non habet, non est vera sapientia '.^ Thus the word has meant successively (1) the natural fear and awe which semi-civilized man feels in the presence of what he cannot explain ; (2) the cult by which he strives to propitiate the unseen Powers, together with the scruple he feels if the propitiation is in the least degree imperfect ; (3) the whole sphere of worship, together with all belief in the supernatural, as viewed from the standpoint of the philosopher ; (4) the competing divisions of that sphere of worship and belief, each being now a religio, and the Christian faith being for the Christian the vera religio. There is one later stage in the history of the word,*" which I can only mention here. It suffered a degradation when it was made to ■ De histitiu, V. 19. " De Vera Sap. iv. 3. Digitized by Microsoft® THE WORD RELIGIO 15 mean the monastic life : the life of men who withdrew them- selves from a world in which true religion was not. But even in this degraded form it reveals once more its wonderful capacity to express the varying attitude of humanity towards the supernatural. Outside the monasteries — the homes of the religiosi — were a thousand fears, fancies, superstitions, which the old Roman might have summed up by his word religio, the anxious fear of the supernatural : inside them, for many ages at least, was still something of the vera religio of the early Fathers, the devotion and the rituiil combined, the pure life and training, religio Dei} THE ORIGINAL MEANING OF THE WORD SACER In Roman religious law the word sacer indicated that the object to which it was applied was the property of a deity, taken out of the region of the profanum by the action of the State, and passed on into that of the sacrum. We have an exact account of it which can be traced through Verrius Flaccus to a scholar apparently of the age of Cicero, Aelius Gallus. ' Gallus Aelius ait sacrum esse quodcunque more ^ atqiie instituto civitatis consecratum sit, sive aedis sive ara sive signum sive locus sive pecunia sive quid aliud quod dis dedicatum atque consecratum sit : quod autem privati suae religionis causa aliquid earum rerum deo dedicent, id ponti- fiees Romanos non existimare sacrum.' * This very explicit passage makes it plain that the state, through its religious authorities, had appropriated the word, and fixed it to a definite meaning, at some period when there were already temples in which deities could dwell and enjoy the possession ' For further comments on the word, see index to my Bdigious Experience, &c. ; and for a different view, W. Otto in Archiv, xii. 533 if. ^ This is Laohmann's correction for MS. ' quocunque niodo '. See Mar- quardt, StaatsverwaUung, iii, 145. ' Festus, p. 424 (-Lindsay). Cf. Gains, ii. 5. Digitized by Microsoft® 16 THE ORIGINAL MEANING OF ofvtheir own property, made over to them by the State to do them honour and propitiate them. But this highly developed idea of deities dwelling in fixed spots in the city, and holding property, is at Rome a compara- tively late one. The earliest document of the ius divinum, the so-called calendar of Numa, can be placed with confidence in the regal period, between the inclusion of the Quirinal in the city of the four regions and the building of the temple of Diana on the Aventine ; ^ and this temple of Diana, and that of the Capitoline trias which belongs to the same age, are the first two temples in the proper sense of the word, and the earliest in which any kind of statue is known to have been placed. Before that the fanum was a small open enclosure with a rude ara, probably of turf, and nothing more.^ The word sacer must have developed its later technical meaning in and after this period. Is it possible to discover with any approach to certainty what meaning it had in still earlier times ? We might naturally look for a meaning of the same general type, but less accurately defined, and so to speak, less theological. For until deities or spirits come to be localized in particular spots and to have special priests attached to them, the vocabulary of worship must be necessarily less clearly cut than in an age when that worship was becoming the most important part of the State's ' cura '. Perhaps this earlier meaning of sacer is indicated in a curious passage of Macrobius, who wrote it Mdth a book before him -De religionibus, by Trebatius Testa, the friend of Cicero. ' Hoc loco non alienum videtur de condicione eorum hominum referre quos leges sacros esse certis dis iubent, quia non ignore quibusdam mirum videri quod, cum cetera sacra violari nefas sit, hominem sacrum ius fuerit occidi.' ' The explanation that follows is of no value to us ; but the fact > See Rel. Exp., p. 94. " Marquardt, op. cit., p. 161 ff ; Wissowa, Eel und KuU. der Romer (2) 468. ' Macrob. Sat. iii. 7, 5. The explanation is a curious example of the semi-mystipal teiidency of Trebatius' time. The souls of homines sacrati were dis debitae, and might therefore be sent ad caelum as soon as possible, i. e. by any one who had the chance. Digitized by Microsoft® THE WORD SAGEn 17 that some Romans were puzzled by the impunity of the slayer of the sacer homo is one of the utmost interest. They were puzzled, because they had always understood the word sacer in the sense in which it was defined by Aelius Gallus. A thing that was sacrum was known by all to be the property of a deity, and to violate it was nefas, a deadly crime. Yet here was an object called by this solemn adjective, homo sacer, which might be violated without any nefas : a man whom any one might slay with impunity. Evidently sacer was used here in an exceptional sense, and surely in a very ancient sense ; for no one will deny that the h)m,o sacer is a survival from a primitive age into one of highly developed civil and religious law. Sacer esto is in fact a curse ; and the hamo sacer on whom this curse falls is an outcast, a banned man, tabooed, dangerous. We may compare him with the primitive Semitic outcast described by Robertson Smith in an appendix to his Religion of the Semites} He has been showing that the ' holy ' thing is not originally something made the property of a god, but something simply tabooed for whatever reason, without reference to gods or spirits. Then he goes on : ' Closely allied to this curse is the ban by which impious sinners or enemies . . . were devoted. The ban is a form of devotion to the deity, and to ban is in the 0. T. sometimes rendered " consecrate ".' So too the homo sacer, we may suppose, was cursed and consecrated at ' p. 434. Dr. Marett, Threshold of Religion, p. 126, compares sacer and taboo, but is thinking of sacer in aiyts senses. Of course that which is the property of a god can be called taboo as much as the accursed man ; but in that sense it is a survival from an older age into the religious law of a theological one. Dr. Marett tells me, what is very interesting in this connexion, that taboo tends in the Pacific to connote ' prohibited by religious law '. He also sends me an illuminating note on the magico- religious aspect of the feeling about the man who has shed blood, &c. — the sinner, in fact (rather than the criminal). ' If you have anything to do with him, something awful wiU happen. What follows then from a social point of view ? Trom the first, he 's an outlaw : which means that no one will stand up for his rights. But he is also a leper, a plague-spot in society. The practical moral is, "Get rid of him " That was the simple duty of savage society. The devotio to a deity is a further elaboration along the magico-religioua line of thought.' 2252 B Digitized by Microsoft® 18 THE ORIGINAL MEANING OF the same moment. He is therefore sacer, not in the sense appropriated by the framers of the ius divinum, of things made over to a deity in order to please and glorify him, but in the more" primitive sense of ' accursed and left to a deity to avenge himself on if he be so pleased '. And as he was not in any true sense the property of the god, or valued by him as such, like objects called sacra under the religious law, any one putting him to death would not be committing what was nefas. In no sense whatever could he be thought of as a sacri- ficial victim ; if he had been such, it would certainly have been nefas for any one but a magistrate or priest, or the authorized assistant of such officials, to lay hands on him. Let us pursue this point a little further. In the ritual of sacrifice at the altar under the ius divinum, the victim must be wholly acceptable to the deity ; it must be pure and perfect, and its passage out of the region of the profanum into that of the sacrum is only consummated when it has been slain, and its entrails examined to see whether they show any flaw that might make it an undesirable gift to the god.^ The sacer homo, on the other hand, was made or declared sacer by the community or its authorities,^ and his slaughter, in whatever way it might ensue, would not seem to have anything to do with its passage from the profanum to the sacrunt. Again, all sacrifice at the altar was accompanied with prayer, as Pliny expressly tells us (N. H. xxviii. 10), and the language of the oldest prayers makes it clear that the deity was believed to be glorified or strengthened by the process (e. g. macte his suovet&urilibus esto) ; ^ but in the case of the homo sacer such an idea is unthinkable. Whoever in short will go carefully through the altar ritual will see that ' This is, I think, the right way to look on the procesa of sacrificium. The preliminary steps, e. g. the pouring on the victim of mola salsa and libations of wine, are only consummated by the actual slaughter, and that again might fail to put the victim into the region of the sacrum, if its exto were not found perfect. ° See below at the end of this paper. ' Of. the prayers in Cato, De Agric. 132, 134, 139, Ul. Cf. Bel. Exp. pp. 183 ff.. Digitized by Microsoft® THE WORD 8ACEB 19 it is in every point wholly inapplicable to the homo sacer. This will explain a passage of Festus which seems to have puzzled the lawyers. ' Homo sacer is est quem populus iudicavit ob maleficium : neque fas est eum immolari, sed qui occidit parricidii non damnatur.' Here Festus, or rather Verrius Flaccus, seems to me simply to mean, ' in this case there is no question of altar sacrifice, though the word sacer might lead one to fancy so : any one may kill the sacer homo '. So Lactantius, with Varro before him, writing of the Argei, who were supposed to have been at one time human victims, says, ' non quidem ut homo ad aram immolaretur, sed uti in Tiberim de ponte Milvio mitteretur '. ^ And indeed there is no record of a homo sacer being slain at the altar, or slain with the axe at all. The shedding of his blood, for whatever reason, seems to be carefully avoided. The harvest thief is hung ; the man who had suffered sacratio capitis et bonorum in historical times might be thrown from the Tarpeian rock ; the parricide, who must have been sacer, though we are not expressly told that he was, suffered the horrible penalty of the sack and was thrown into the sea.^ So too the guilty vestal was buried alive. The only case of a human victim being slaughtered at an altar is that of the two mutinous soldiers', if such they were, who were beheaded at the Ara Martis in the Campus Martins by order of Julius Caesar, and their heads fixed up on the Regia : * a strange ritual which is so closely analogous to that of the yearly sacrifice of the October horse that we must suppose it to have been a somewhat wanton imitation of that rite. Lastly, in • Festua, s.v. sacer homo. Lactantius, Inst. i. 21. Wisaowa (op. cit., p. 388, note 11) haa aeen that the homo sacer cannot be the subject of a sacrificium : ' der mit Strafaohuld beladene Verbreoher konnte ebensowenig als eine Ehrung den Gottem dargebracht werden, wie die Miisgeburt, die man stillachweigend beseitigt.' I am glad to find A. Rosenberg of the same opinion as that expressed'in the text (Hermes, 1913, p. 363). " Oio. Pro Bosc. Amer. 26. 72. See an interesting parallel, both in feeling and practice, in The Pagan Tribes of Borneo, ii. 196. ' Die Cassius, xliii. 24. Mommsen, Strafrecht, p. 913 ; Wissowa, op. cit., p. 421, note 2, considers it an undoubted case of imitation of an ancient rite. See also my Roman Festivals, p. 249, note 2. B 2 Digitized by Microsoft® 20 THE ORIGINAL MEANING OF the case of the ver sacrum, though the animals were believed to have been sacrificed at the altar, the human beings were kept till they were grown up and then driven beyond the frontier. So far then the distinction between the homo sacer and a sacrificial victim seems clear. But here we meet with a difficulty in our argument. When we examine the records of the ancient rules of law relating to the hxmio sacer, we find that in most instances he is placed in connexion with a deity or deities to whom he might seem to be ' sacrificed '. Not indeed in every case : Festus, s. v. Terminus, tells us that Numa Pompilius ' statuit eum, qui terminum exarasset, et ipsum et boves sacros esse ' without any clear reference to a deity Terminus. So too in the XII tables : ' Patronus si clienti fraudem fecerit sacer esto ; ' where it is only from a Greek writer that we learn that the man was to be sacer ' to ' Jupiter, i. e. apparently Zevs opios} But of the harvest thief it is said that ' suspensum Cereri necari iubebant : ' ^ though it is to be noticed that the word sacer is not here used. The husband who sold his wife was to be sacrificed (if we may so translate Plutarch's QveirOai) to the infernal deities : * and of the son who struck his father it was written, ' divis parentum sacer estod '. * Here let us notice that with the exception of Ceres, it is the di inferi who are mentioned ; and even Ceres may reasonably be supposed to have been in this context originally Tellus Mater, whose place she frequently usurped in historical times.* Now these are the deities of the devotio : Decius for example, after having been made sacer under the directions of the pontifex ^ (so the process may be explained) and having - invoked all the gods of Rome to help the State, finished with the words 'ita pro re publica Quiritium, exercitu legionibus auxiliis populi Romani Quiritium legiones auxiliaque hostium ■ Serv. Aen. vi. 609 : cf. Dion. Hal. ii. 74. ' Plin. xviii. 8, 12. « Plut. Bom. 22. ' Festus (p. 260, Lindsay). ^ Wissowa, op. cit., p. 192 ff. " Livy viii. 9. 6-8 : and cf. the explanations of the ritual by Professor Deuhnev in Archiv fur Beligionsvnssenschaft, 1905, p. 69 ff. Digitized by Microsoft® THE WORD 8AGER 21 mecum di's Manibus Tellurique devoveo '} Evidently there is some analogy or close connexion between the devotio and the consecratio of the sacer homo : and as the self -immolating victim of the devotio was a kind of vicarious sacrifice for the whole host, so we may perhaps infer that the sacer homo was in some sense made over to the infernal deities in expiation for the mischief he had brought on the community. If this be so, the word sacer must here be translated not ' sacred to ' but ' accursed and devoted to ' ; and this is why he is an outcast and ' holy ' or dangerous. These infernal deities had no regular ordered altar sacrifices : ^ if one wished to appease them with a victim one must curse him and make him sacer in the old sense of ' taboo ', and then leave him to his fate, as Decius rushed into the middle of the enemy. If this is the right meaning of the word sacer in sacer esto, we may, I think, trace it back to the older stage in which it meant simply ' taboo ' without reference to a deity ; and we have ■seen that it seems to be so used in one or two of the ancient laws. But with the growth of the State and its religious law the attachment to a deity expressed in the dative case became more usual, though this deity could not as yet be one of those to ,whom altar sacrifice was regularly made, nor could the word sacer be used here of the criminal in the same sense in which it was used of the true sacrificial victim. Later on again, however, we come upon cases in which a man who violated a lex sacrata, particularly that which made the tribunus plebis sacrosanct, was declared sacer lovi, and his familia was to be sold at the (plebeian) temple of Ceres ' In a later formula quoted by Maorobius (iii. 9, 10) as used at the siege of Carthage, the deities are Dis Pater, Vediovis, Manes. Dis Pater is the Greek name for Orous. " This was so until the Graeco-Koman period and the introduction of Greek deities of the underworld (e. g. at the Indi saeeidares, when black victims were sacrificed at night). For animal sacrifice at the Parentalia the only evidence is that of Virg. Aen. iii. 67, which need by no means be taken as proving a Roman practice. Vediovis and the Manes had no temples or altars, so far as we know, till towards the end of the republican period. By the devotio a man's life was put into their power : and so with the saeralio. (See article Inferi in the Mythological Lexicon, p. 256.) Digitized by Microsoft® 22 THE ORIGINAL MEANING OF Liber Libera.^ I take this to be still the old sense of the word : the man is, so to speak, taboo and any one may kill him, e. g. by throwing from the Tarpeian rock ; ^ and the same was probably the case of the man ' qui regni occupandi eonsiha iniisset ' (Liv. ii. 8. 2) . But the mention of the heavenly deity in the dative is a novelty, and strictly speaking, an anomaly. The Jupiter in this case was probably the great deity of the Capitol, whose cult was specially connected with the idea of good faith and covenant. But even here I think it would be safer not to speak of the victim as being ' sacrificed to Jupiter '. If we now ask how the homo sacer came to be declared sacer, since he did not become so^by any sacrificial act, we may perhaps see three stages of the process, answering to the three main periods of the development of Roman society. (1) In the age of taboo proper, before the appearance of the State and its ius divinum, we have of course no Roman evidence to help us ; and perhaps we cannot well go further than to call it a collective or sociological declaration. Dr. Marett, to whom an Oxford man naturally goes for help in such matters, seems to make it the result of public opinion among savages, and the penalty by no means always a measurable quantity.^ Sacer esto could be the verdict of the group most immediately affected by the crime, either by what the Australians call ' growling ', or by the voice of an authority : but what the group or the authority was in Latium we cannot be sure. (2) In the age of the early city- state and its ius divinum, we may assume that the declaring authority was the rex, aided no doubt by the pontifices : for in historical times it was the pontifical college that declared an act nefas, or a man impius,* and the inference is a safe ' Livy, iii. 55. 7. " Mommsen, StrafrecM, p. 933. 3 The Threshold of Religion, pp. 90-2. " Thia is the view of Marquardt, Staatsverwaltung, iii. 277 ff. I cannot find a definite proof of the statement that the pontifex maximus declared a man impius, but the assumption seems a safe one. In the case of Clodius and the Bona Dea mysteries the pontifical college declared the act nefas before further steps could be taken. Cic. Att. i. 13. 3. Digitized by Microsoft® THE WORD SACER 23 one that in this matter of religious law they also were the possessors of the final formula of sacratio. From the pontifex of the republican age we naturally argue back to the Rex of the earlier period. (3) Under the republic, though the ponti- fices may have possessed the necessary formula, there is strong evidence that this was preceded by a judicial trial. The passage of Festus referred to at the beginning of this paper runs ' homo sacer is est quem populus iudicavit ob maleficium ', and under the sacratae leges some form of trial might certainly be expected. Mommsen therefore assumes that a trial and magisterial pronouncement were always necessary.^ To sum up what has been said : the relation between a deity or numen and any object brought into connexion with him, can always be indicated by the word sacer, but that relation is not always of the same kind. Originally the word may have meant simply taboo, i. e. removed out of the region of the profanum, without any special reference to a deity, but ' holy ' or accursed, according to circumstances. Natur- ally this word was seized upon by the framers of a ius divinum, to express that which is consecrated or sacrificed to a deity, as the idea of benevolent numina, with dwellings within the city at particular spots, gradually developed itself : hence the prevailing idea of the word throughout Roman literature is not a sinister one, but rather one suggesting a happy relation (pax) between the Roman and his gods. Nevertheless the older meaning of taboo in the sense of. accursed could not be forgotten or extinguished ; and it was retained in another department of the ius divinum for the criminal who was declared to be left to the infernal deities, or their agents, to be disposed of, and later again for the man whose caput and bona were ' consecrated ' in historical times.^ But by ' I am indebted for this statement of Mommsen's view, which he says is maintained throughout the StrafrecM, to the late Master of Balliol, who kindly allowed me to read the proof of the first chapter of his book on Roman Criminal Law. This reading suggested to me the subject of this paper, and also supplied me conveniently with a few of the passages I have noticed. " The compound consecratio retained this meaning throughout : e. g. Digitized by Microsoft® 24 ORIGINAL MEANING OF THE WORD 8ACER the end of the republican period such cases as these were rare, and the other and cleaner meaning of sacer had so entirely come to prevail, that, as Aelius Gallus said, it was puzzling to many Romans that an object called sacer could be violated with impunity.^ MUNDUS PATET 24th august, 5th OCTOBER, 8th NOVEMBER The mundus of Rome was believed to be a hole or under- ground pit or vault on the Palatine.^ It was said to be closed by a stone called the lapis manalis, which same name, oddly enough, is also given to an entirely different kind of stone, with which the pontifices occasionally worked some sort of magic in a drought.* Plutarch, in the chapter in which he in Pliny's Panegyric on Trajan, 64, we read : ' ille iuravit expressit ex- planavitque verba quibus caput suum, domum suam, si sciena fefellisset, deorum irae consecraret.' ' I may add in a footnote the curious use of the word (apparently in an antique form) mentioned by Varro in two passages in book ii of his De re rustica i. 20 ; iv. l6). Sacrificial animals were not reckoned fit for sacrifice until a certain number of days after birth (e. g. ten in the case of pigs) ; after this they were called in Varro's time ipuri, but formerly sacres. In each case he quotes a line of Plautus in the Menaechmi, which, as we have it, stands thus (ed. Lindsay, Oxford, ii. 289) : ' quibus hie pretieis porci veneunt sacres sinceri ? ' Cf. Festus, 318 (420 Lindsay). It may be that Plautus was here translating a Greek word {oaios ?), and so misled Varro into fancying that sacres was here really used in a primitive sense. The word sinceri seems to be added to make the meaning clear to a Roman audience. The ignis sacer of medical writers (Plin. N. H. xxvi. 121 ; Virgil, Oeorg. iii. 566), i. e. an eruption on the skin, may also be mentioned : the word here may be supposed to mean ' uncanny ' or ' dangerous '. Lastly, the use of the word in the term sacra via suggests that that path was originally reserved for religious purposes under the ius divinum of the city of the four regions ; but its early history, except so far as excavation has thrown light on it as a material object, is entirely lost to us. ' I leave its position to experts. See Year's Work in Classical Studies, 1915, pp. 12-13. 0. L. Richmond in J. S. S., 1914, p. 225 ff. " See my ijpmaii Festivals, p. 232. Digitized by Microsoft® MUNDUS PATET 25 describes the foundation of. Rome,^ says that the mundus, like the process of marking out a city, was of Etruscan origin ; that firstfruits of all kinds were thrown into the pit, and that each new settler brought a bit of earth from his own country and cast it into the pit ; he places the pit in the Comitium instead of the Palatine, but notes the word mundus as applied to it there, and the identity of this word with that for the heaven or universe. Plutarch says nothing of another notion, namely that on three days in the year, those noted above, the lapis was removed to give egress to the denizens of the underworld. This ,we learn from Varro, quoted by Macrobius : ' mundus cum patet, deorum tristium atque inferum quasi ianua patet.' ^ So too Ateiu's Capito quoted by Festus * : ' Mundus ter in anno patere solet, diebus his : postridie Volcanalia (et a.d. Ill non. Oct.) et ante diem VI id. Nov. Qui quid ita dicatur sic refert Cato in commentariis iuris civilis : Mundo nomen impositum est ab eo mundo qui supra nos est. . , . Bius inferiorem partem veluti consecratam dis manibus clausam omni tempore nisi his diebus qui supra scripti sunt maiores (censuerunt habendam), quos dies etiam religiosos iudicauerunt.' Here it is necessary to note that the only words of Cato are those in italics : * there are other words of his following these, to which I shall refer directly, but Cato had nothing to say of the lapis manalis and the ghosts, so far at least as we know : for these ideas Varro is our oldest authority, followed by Ateius Capito in the age of Augustus. Since I wrote my book on the Roman Festivals I have often wondered why these three days, August 24, October 5, November 8, were selected as holidays, so to speak, for the ' Bomulus, 11. ^ Macrobius, i. 16. 18. He adds evidence that the days were rdigiosi : an army might not give battle, nor any military operation of importance be performed ; nor might a marriage take place. ^ Festus, 144. Paulua, 145, gives the dates, which are mutilated in Festus. ■■ See the fragment in H. Jordan's Oatonis Libri Deperdili, p. 84, with his note. Digitized by Microsoft® 26 MUNDUS PATET ghosts. If the old Eomans really believed in their return to the upper world on those days, the days must have had some special importance in connexion with ghost-life ; but no one, so far as I know, has ever yet discovered what this importance is. The days fixed in the old Calendar of Numa as those on which ghosts would be roaming about in apparent freedom, and on which they might be expelled from the house by the paterfamilias, were 9th, 11th, and 13th May (Lemuria), and the more civilized festival of the dead was in February (Parentalia). Why should three other days be allowed them for freeddm in late summer and autumn ? In the book just referred to,^ taking a hint from 0. Miiller's Etrusker, I suggested that the ghostly function of the mundus was an accretion, perhaps or probably of Graeco-Etruscan origin, on a very simple original fact. The pit might be the penus of the new city, i.e. the underground storing-place for the grain ; and thus we can understand why it should be open on a day (August 24) which follows the Cbnsualia, a festival which almost beyond doubt has reference to har- vesting, and immediately precedes the Opiconsivia, which almost as certainly represents the storage of the grain as completed.^ ' Nor is it difficult to understand why, when the original use and meaning had vanished, the Graeco- Etruscan doctrine of the underworld should be engrafted on this simple Roman stem. Dis and Proserpina (Greek deities) claim the mundus : it is ianua Orci, faux Plutonis, fancies familiar to Romans who had come under the spell of Greek and Etruscan religious beliefs.' Quite lately I have been able to develop this suggestion a little further. I think, unless I am under a delusion, that I can explain not only August 24, but with some little prob- ability, also October 5 and November 8 as days on which we might expect the mundus to be open, not for the egress of ghosts, but for a very practical purpose of the farmer. I conjecture that it was the place in which was stored, not, or ' Soman Festivals, p. 211. Cf. Miiller-Deecke, Etrusker, ii. 100. 2 See Wissowa, Selig. und Kvltus der Earner, p. 168 (ed. 2, p. 203). Digitized by Microsoft® MUNDUS PATET 27 not only, the grain of the last harvest which would be needed for food; and for which the storehouse (penus) would need to be frequently opened in the old farmhouse, but the place of safety in which the seed-corn was stored. This was a sacred treasure almost more precious than the grain destined imme- diately for food : and it must be housed securely and hidden most carefully from enemies of all kinds. The mundus as Cato describes it, though on the Palatine in his day it would be only a symbolic survival from the original storing-place, seems to me strongly to suggest a use for human beings as well as ghosts. ' Mundo no men impositum est ab eo mundo qui supra nos est : forma enim eius est, ut ex his qui intravere cognoscere potui.' ^ The mundus then was a place into which a man might descend : we may imagine it as a kind of cellar with an opening in the centre of its roof, which was closed, except on the three days, by a stone, after the fashion of a trap-door. On the top of this there was no doubt a covering of earth, for the sake of concealment, an obvious safeguard which seems to be reflected in the descrip- tions- both of Plutarch and Ovid.^ The poet wrote : Fossa fit ad solidum. Fruges iaciuntur in ima et de vicino terra petita solo, fossa repletur humo. . . . But I must now go on to explain my justification for this very matter-of-fact conjecture. In August the opening of the mundus took place the day before the Opiconsivia, i.e. the 24th ; and in the latter festival it is pretty well'agreed that we should see a representation of the completed storage of the corn of the recent harvest. My conjecture is that on the previous day the seed-corn for the autumn sowing was separated from the rest of the grain, and deposited in an underground storing-place, for the se- curity that was absolutely essential for the existence of the ' Potui is Scaliger'a emendation for potuit of the codex. The pit dis- covered by Professor Boni in 1914 in the peristyle of the Flavian palace answers fairly well to Cato's description. See Year's Work, 1915, p. 12. ' Fasti iv. 821 : of. Plutarch, Bom. 11. Digitized by Microsoft® 28 MUNDUS PATET community. Varro tells us that in his time the finest ears were separated on the threshing-floor from the rest of the corn, in order that the semen (seed-corn) might be as good as possible.^ As a rule the corn seems to have been threshed as soon as it was brought home from the field : Varro and Columella imply this,^ though they do not state it in so many words ; and in primitive times, when your enemy might at any moment make a raid on you, this would be desirable in order to secure the precious treasure as quickly as possible. We know nothing from literary sources of the place of storage, but I venture to think that not only the curious underground altar of Consus, opened at the Consualia on the 21st, but also the opening of the mundus on the 24th, suggest the method that would obviously be the safest, that of concealing the treasure underground.* It is of course possible that both the grain for food and the seed-com were deposited in the same place. But apart from the extra security which two storing-places would give to the farmer, I think that the dates of the other two openings of the mundus may suggest that it was the receptacle of the seed-corn only. The oldest kind of grain used for food in Italy was that rough kind of wheat called far, which in historical times was used in the city only for religious purposes. But in some districts it was still grown, and Pliny tells us that the sowing '■ Varro, B. E. i. 62, init. ' Quae seges grandisaima atque optima fuerit, aeorsum in aream seoemi oportet spicaa, ut semen optimum habeat ' (i. e. the farmer). Cf. Pliny, xviii. 195 ; Columella, ii. 9, 11 ; and also Virgil, Qeorg, i. 197, who says that the farmer must pick out the largest by hand, or they will degenerate in the keeping. ' Varro, B. B. i. 50, 51 and the beginning of 52 already quoted. When the com has been reaped, it must be brought to the area (threshing-floor), which Varro then describes in c. 51 ; then, returning to the crop, he Urges the separation of the seed-oom from the rest. The same is clearly implied in Columella, ii. 21. ' See Mommsen's note in C. /. L. i, ed. 2, p. 326, followed by Wisaowa, Bel. und Kidt, p. 167 (ed. 2, p. 201). As from 5 to 10 modii of various kinds of seed were needed for each ivgerum, a fairly roomy receptacle would be necessary. There is a good account of modem ' siloa ' in Tristram's Land of Israel, p. 108. Digitized by Microsoft® MUNDUS PATET 29 went on through the month of October ."^ A date as early as the fifth, in the practice of the later people of the city, would suit well enough for the opening of the storing-place for the purpose of taking out the necessary amount of grain for sowing. But the third of the days of opening, November 8, bears more remarkable testimony to their original meaning. All readers of Virgil will remember that in his first Georgic (219, fE.) he urges the postponement of the sowing of wheat (triticum) till after the setting of the Pleiades : and in this he is borne out by Columella.^ No doubt Virgil represents the traditional practice of the Italian farmer. Now the apparent or cosmical setting of the Pleiades, i. e. that which alone can have been known to the husbandman throughout early Roman history, seems to have taken -place on or about November 9 ; different ancient authorities give different days, but all at the beginning of November, and the writer of the article Astronomia in the Diet, of Antiquities fixes the actual day of apparent setting as the 9th.* That the opening of the mundus should have taken place on the 8th is thus ' Pliny, H. N. xviii. 205 : cf. Varro, i. 34. Cf. Cornford in Essays and Studies presented to W. Bidgeway, p. 155, for early autumn sowing in ancient times, quoting Fhit.»Frag. 23. ' ' Ante tibi Eoae Atlantides absoondantur, Gnoaiaciue ardentis decedat Stella Coronae, Debita quam sulcis oommittas semina quamque Invitae properes anni spem credere terrae.'' Varro, i. 34, rather vaguely describes sowing as extending from the equinox to the bruma ; but Columella, ii. 8, quotes and supports Virgil : only in this passage he seems to be thinking of the true morning setting of the Pleiades, i. e. October 24, though in other places he obviously alludes to the apparent setting. See Diet, of Antiquities, a. V. Astronomia, p. 227. ' Diet, of Antiquities, loc. cit. ' The true morning setting was at Rome at that epoch on 29th October, the apparent morning setting on 9th Novem- ber.' This date has been confirmed for the time of the Roman kings by Dr. Fotheringham, who has most kindly made elaborate calculations for me. He sums them up thus in a letter : ' Anyhow, you will see that the date given in the Diet, of Antiquities (9th November) appears to apply excellently to the time of the kings. It does not seem to apply so well to the time of Julius Caesar, to which it was intended (in the dictionary) to refer.' In a later letter he wrote : ' As the Roman 8th November did not occupy a fixed place in the natural year before the time of Julius Caesar, I presume that a general and not an exact coincidence with the cosmical setting of the Pleiades is all that is required.' Digitized by Microsoft® 30 MUNDUS PATET a striking fact, and strongly suggests that the seed-corn of the better wheat crops, as distinct from the more ancient far, was at this time being taken out of the mundus for the November sowing. Now supposing that our hypothesis is a reasonable one, and that the mundus was originally a receptacle for seed-corn, how are we to account for the accretion on this simple and useful practice of the doctrine of the mundus as faux Plutonis, ostium Orci and so on, and of the liberation of the ghosts when the stone trapdoor was removed ? In the first place, there is no difficulty in attributing a religious character with taboos such as Varro mentions ^ to such receptacles of the means of man's subsistence : that is sufficiently well shown by the sacred character of the store- chamber of the house, which produced in time its own spirits or deities, the Penates : and the underground altar of Census points in the same direction. Professor Deubner has lately shown ^ that there are two main periods in early Roman religious thought, and that undoubtedly the purely rituahstic one is the oldest, when deities and a theology are only in the making, if as far advanced as that. To this older stratum belongs the original use of the mundus as I explain it. No deity is here concerned, unless it be the Ops Consiva of the day following that of the opening of the mundus in August, and that deity is plainly no more than the store itself with its religious character beginning to take tangible shape in a worship. Upon this older stratum of religious ideas there lies what we can only suppose to be a later stratum deposited by another race, in which the idea of existence after death in an underworld was more important than the practical ideas of the pure agriculturist. Such a race was the Etruscan. In a valuable summary of our present knowledge of Mediter- ranean burial, kindly sent me by the author, Professor von Duhn, he attributes a somewhat grossly material idea of the dead alike to the oldest population of Italy, and to the ' Maorobius, i. 16, 18. " In Netie Jahrbiicher filr das klassische Alterlum, 1911, p. 323. Digitized by Microsoft® MUNDUS PATET 31 Etruscans, both of which races buried their dead and supplied them with such objects as they were supposed to need, in contrast to the true Italic peoples (Sabines excepted) who used cremation, and show signs of being the ancestors of those who developed the orderly, sensible ritual of the ParentaUa. The conjecture in my recent volume ^ that the notion of an underworld and its horrors was Etruscan, but resting on a substructure of much more primitive belief, is not so wild as I feared at one time that it might be. Exactly how the new way of looking at the simple old practice came about it is impossible to say ; but each of us can make some kind of a guess for himself if he pleases. My own guess is that the primitive storing pit was transferred from the farm or the pagus to the newly-founded city, and the three days of opening were retained and fixed ; that in due time its original meaning was lost, owing to the city ceasing to be a practical centre of agricultural operations : and that as this cessation happened about the same time as the Etrus- can dominion in Rome, the mundus took on a new meaning connected with the Etruscan ideas of a nether world.^ The stone, of which we are told on a single authority that it was called lapis manalis, the same name as that of the stone of Jupiter Elicius, took on the name of that other stone through a misinterpretation of the word manalis, which was wrongly supposed to mean ' belonging to the Manes ' .^ ' The Religious experience of the Roman people, pp. 391 ff. ' The best account of the word mundus known to me is in Nettleship's Contributions to Latin Lexicography, p. 528. I have abstained, from invoking the aid of etymology ; but if Nettleship is right, the word may be developed from a root mu, meaning to enclose, or fence round. In regard to an Etruscan origin of a similar word, see MuUer-Deeoke, Die Etrv^her, ii. 100, n. 65a. ' Cf. Paulus, 128. In case the contrast between the original Latin meaning of the mundus and that here assumed to have been superimposed, should astonish any one, let me refer him to the remarks of Dr. J. B. Carter in Hastings' Diet, of Religion and Ethics, i. 464. He points out that the Romans do not seem to have been much interested in the lower world, and that every bit of description of it comes from writers under Greek influence, and all the details are identical with those of the Greeks. Hence it is probable that the Roman lower world was not mythologicaUy adorned till Greeks (and Etruscans) did it for them. As we have seen, the idea of a stone Digitized by Microsoft® 32 MUNDUS PATET Now I may reasonably be asked why, if I make so much of the seed-corn and its place of deposit, we do not find more distinct traces of the importance of these among other peoples, Mediterranean or other. My answer is that I do so find them, though they seem to me to have lain unnoticed since Mann- hardt developed his theory of the corn-spirit. For the ani- mistic period that theory undoubtedly holds good, and has been confirmed by the immense mass of additional evidence brought together by Sir James Frazer in his Golden Bough, but I have for some time felt that there is a yet more primitive way of looking at the mystery of the renewal of vegetation ; and in many of the examples of the familiar forms of the corn-spirit I am' inclined to see traces of the sacred character of the seed-corn itself, and of the place in which it was stored. My friend Sir James Frazer has most kindly pointed me out a number of such unindexed examples in the second volume of the Golden Bough (ed. 2), though without expressing de- finite approval of my views on this subject ; but in order to weigh the matter thoroughly, it is advisable to read the whole of chapter iii in that volume, as well as to let the mind dwell on isolated instances. I think I see signs that the last sheaf of the harvest, which in innumerable instances is treated with reverence or made into human form, may represent the precious seed-corn set aside at the time of threshing. A good example is taken by Sir James covering the abode of the dead, the removal of which gave egress to the ghosts, is found only in Festus, 144, and nowhere alluded to in Roman literature. It has been compared to the Dillestein of German mythology (Preller-Jordan, ii. 67), but a perusal of the description of that mysterious stone in Grimm's Deutsche Mythologie, iii. 806 (Engl, trans.) makes it clear to me that there is nothing in common between the two. The Dillestein was a ceiling or grating of the underworld, lying at the bottom of our earth. I may add to this note a few words of Sir James Frazer' s, contained in a letter to me : ' The ancient explanation of the n^undus is perhaps not wholly irre- concileable with your theory. For observe that the spirits of the dead are often supposed to watch over or further the growth of the crops : that is why the firstfruits are often presented to them. For examples see the Golden Bongh (ed. 2), ii. 459, seq.' On the connexion at Rome between Tellus Mater, the dead, and the crops, see my Bdigioiis experience of the Roman peopl^ pp. 121, 138 ; cf. Dieterich, Mutter Erde, cap. iv. Digitized by Microsoft® MUNDUS PATET 33 Prazer from Mannhardt.^ At Westerhiisen in Saxony the last corn cut is made into the shape of a woman, brought to the threshing-floor, and kept there till the threshing is done.^ Just below on the same page we have an example from Tarnow, Galicia, in which the last corn cut is made into a wreath and called the wheat-mother, &cl, and kept till spring, when some of the grain is mixed with the seed-corn. The last sheaf is often larger and heavier than the rest, and this Sir James Frazer explains (p. 176) as a charm, working by sympathetic magic, to ensure a large and heavy crop in the following harvest. Is it not rather a survival of the selection of the finest ears to use as seed-corn ? For it seems that this last sheaf is often taken from that part of the field where the corn is finest : examples of this practice will be found on p. 184 (from Kent), on p. 189 (Scotland), p. 193 (ancient Peru), a passage to which I will return directly, and p. 195 (ancient Mexico) ; p. 200 (Malay peninsula) and in Sumatra (p. 198), the best grains of rice are picked out to form the rice- mother, and are sown in the middle of the bed, with the common seed planted round them. When the time comes to transplant the rice from the nursery to the field, the rice- mother receives a special place either in the middle or in a corner of the field, and is planted with a prayer or charm. Further, this last sheaf of fine grain is sometimes deposited in a special place, and even in an underground cavity or cellar, like the firstfruits which Plutarch tells us were deposited in the Roman mundus, a practice which I take to be the fore- runner of those numberless instances in which the last sheaf or some puppet representing it, is kept stuck up on the farm- house during the winter. The great care taken of the njaiden, as this puppet, garland, or sheaf, is so often called, would be a survival of the care originally taken of the precious seed- corn.* , A good example of storage in a special granary occurs ' Golden Bowgh (ed. 2), ii. 172. 2 i. e. it is kept separate, as intended for seed-corn. Cf. Mannhardt, Mythologische Forsehungen, p. 334, translated in G. B., p. 181. " According to Festus (p. 125, Lindsay) the mundus was known as 'Oereris mundus '. g252 g Digitized by Microsoft® 34 MUNDUS PATET on p. 193 {0. B. vol. ii), from ancient Peru, described by the historian Acosta : a portion of the most fruitful of the maize is thus deposited with religious ceremony. So in G. B. ii. 459, a little hollow filled with grain is left on the threshing- floor, according to Prazer (or his informant Casalis), as a thankoffering to the gods. Is this explanation the right one ? Again (p. 194), in Mexico the priests, with the nobles and people, went in procession to the maize fields, where they picked out the largest and finest sheaf, brought it home, and laid it upon an altar. ' After sacrificing to the harvest- god, the priests carefully wrapped it in fine linen and kept it till seed-time. Then it was carried once more to the field from -vrtiich it had been taken, and deposited in a subterranean chamber, which was closed and covered over with earth. Then followed the sowing, after sacrifice had been made for an abundant harvest ; and finally, when the time of harvest drew near, the buried sheaf was solemnly disinterred by the priests, who distributed the grain to all who asked for it.' This I take to be an animistic and magical development of the simple practice of storing the seed-corn. One more example : in Java (pp. 201-2), two garlands are made of ears of rice, and called the rice-bride and rice-bridegroom, whose wedding is celebrated just before harvest. ' Later on, when the rice is being got in, a bridal chamber is partitioned off in the barn, furnished with a new mat, a lamp, and all kinds of toilet articles. Sheaves of rice, to represent the wedding guests, are placed beside the bride and bridegroom. Not till this has been done may the whole harvest be housed in the barn. And for the first forty days after the rice has been housed, no one may enter the' barn, for fear of disturbing the newly-married pair. ' ^ I read this to mean that the sheaves here called wedding guests were really those reserved for ' Does this mean that the first use of the grain for sowing ooourred forty days after it was thus deposited ! It is curious that the time between the Opioonsivia on August 25, and the opening of the mundus on October 5, is exactly forty days, a coincidence which I do not in the least wish to emphasize ; but the number forty has often a religious significance .♦ Digitized by Microsoft® MUNDU8 PATET 35 seed-corn, only after which reservation the housing of the general harvest could begin. Lastly, I will just allude to a feature analogous to some of those just noticed, in the ritual of Demeter and Persephone at the Thesmophoria. Miss Harrison has described this and commented on it in, her Prolegomena to the study of Greek Religion, chapter iv, translating a valuable passage from the scholiast on Lucian, Dial. Meretr. ii. 1. ' At some time not specified ' (so she sums up our information), ' but during the Thesmophoria, women carefully purified for the purpose let down pigs into clefts or chasms called megara or chambers. At some other time not precisely specified they descended into the megara, brought up the rotten flesh and placed it on certain altars, whence it was taken and mixed with seed to serve as a fertility charm. As the first day of the festival was called Kathodos and Anodos it seems likely that the women went down and came up on the same day.' This account is curiously confirmed by a discovery of Sir Charles Newton at Cnidus, quoted by Mss Harrison on p. 125. There, in the sanctuaryof Demeter, he found a crypt which had originally been circular, though later compressed by an earthquake, in which were bones of pigs and other animals, and the marble pigs which now stand near Demeter of Cnidos in the British Museum. This crypt seems to remind us of the mundus, and so perhaps do the megara described by the scholiast.^ We do not know what the mundus contained, though the description of it given by Cato ^ strongly suggests that it contained something, or was originally meant to do so. But the crypt at Cnidus, and the megara of Demeter, contained pigs, which in Greece were the special victims of the deities of earth and fertility, and these were used as a charm, mixed with the seed-corn, to obtain good crops. All this belongs, however, to an age of religion and fully • In his Modern Oreek Folklore and Ancient Greek Religion, Mr. J. C. Lawson has some interesting remarks about the beehive structures at Mycenae, suggesting that they may possibly have been megara, ' temples of ohthonian deities such as Demeter ' : see p. 94 ff. ' Apud Fest. (p. 144, Lindsay). C 2 Digitized by Microsoft® 36 MUNDUS PATET developed deities ; and I would here again suggest that behind it there lies the simple custom of storing the seed- corn for safety in a subterranean crypt. The seed and the crypt are both holy, as we might expect, and as we gather from the fact that women alone, and fully purified, were allowed to descend into the crypt and bring up the necessary supply of seed. It is not without interest to note that the Thesmophoria, when this took place, is in autumn (11th Pyanepsion), and presumably about the time of the autumn sowing. Dr. Farnell's more elaborate account ^ of the Thesmophoria and kindred festivals of Demeter and Persephone has also many points of interest in connexion with my subject, and I think it may be worth suggesting that experts in Greek religious usages should see whether my theory has any bearing on doubtful points. I note with interest his reference to a fragment of Anacreon^ in support of the possibility that one early (and lost) meaning of dfcr)x6s was dria-avpos. Is it remotely possible that the objects carried at that festival, as indicated by its name, were baskets of seed for sowing ? Dr. Parnell tells us that Triptolemus'was believed to have distributed the seed for this purpose.^ As so many strange explanations of this mysterious word have been suggested,* I need hardly fear to suggest yet another. Sir James Frazer ^ has hazarded the conjecture that the sacra were called Bea-fMOi because they were the things laid down, or as I would add, put into a thesauros. I only go a step further and suggest that these sacra were originally portions of seed-corn : for the Thesmophoria was a late autumn festival and clearly connected with sowing. In conclusion, all I have been doing in this paper is to turn over a stone to see if there is by any chance anything there. I am not at all sure that there is anything there really worth picking up ; the explanation of the three days may lie some- 1 Cults, iii. 105 ff. = Bergk, Poet. Lyr. iii. 271. ' Gvlta, iii. 184. * See e. g. Miss Harrison's Prolegomena, pp. 137 and 143. ' See Miss dlarrison, op. cit., p. 137. Digitized by Microsoft® MUNDUS PATET 37 where else, and I do not forget that the beginning of November is a great time for ghosts in many parts of the world, a fact which is reflected in the CJhristian calendar. Or there may be some mysterious connexion between firstfruits and seed- corn, and between both of them and the dead, which has not yet been entirely fathomed. I hope I may be allowed to hazard a hypothesis without doing any one any serious harm.^ THE OAK AND THE THUNBER-GOD In the third edition of Ths Golden Bough (1911), the dis- tinguished author inserted a chapter on the worship of the oak, and the intimate connexion of that tree with the Aryan god of the sky and thunder.^ The second part of the chapter sums up the evidence for that connexion ; evidence which has been collected, not only by Sir James Frazer himself, but by Dr. A. B. Cook, with great diligence and ingenuity in the Classical Review (vols, xvii and xviii), in Folklore (vols, xv and xvi), and in his great work on the Sky-god, of which the first volume has recently appeared. This evidence has made it certain that of all trees the oak ^ is the one most generally held in reverence by the peoples dwelling in the temperate zone which is its habitat ; this at least is proved for Greeks, Italians, Celts, Teutons, Slavs, and Lithuanians. It is equally certain that among all these races, except perhaps the Celts, the deity of (or in) the oak was also regarded as the deity of (or in) the sky, who descends to earth in the form of lightning, or in more anthropomorphic conception, hurls his bolts upon the earth. So far all is clear. But then comes a serious difficulty. How are we to explain ' For further developments see Miss Harrison's paper, ' Ichneutae ' , in Essays and Stiidies presented to William Bidgeway (1913), and Dr. Comford's ' 'Awapxcu and the Eleusinian Mysteries ', in the same volume. '■ ii. 349-75. " I use the word in a generic sense, so as to include Quercus ilex as well as Q. rdbur and its varieties. Digitized by Microsoft® "38 THE OAK AND THE THUNDER-GOD this intimate association between the religious character of the oak and that of the sky, between this particular species of tree and the mystery of thunder and lightning, in the mind of the primitive Aryan ? How are we to answer the question whether the tree or the lightning came first in his religious thought ? At the end of his chapter Sir James Frazer attempted an explanation of these diificulties. When writing it he had come to the conclusion, but not without hesitation, that the oak is the primary object of religious awe in this connexion, and the thunder and lightning a secondary one. It is unlikely, he argued, that a god of thunder should come to be regarded as a god of the oak merely because thunder and rain come from the sky, and because the oak reaches skyward, and is often struck by lightning. He preferred to think that the oak was the primary object of worship, and that the worshippers may have connected it with lightning when they kindled fire with oaken sticks ; the appearance of the spark in the wood suggesting perhaps the idea that lightning was the result of a similar process worked by some great Being up in the sky. In the summer of 1912 a different solution of the difficulty accidentally occurred to me, based on evidence which was not known to my friend Frazer when he published his chapter in the previous year. I straightway communicated it to him, and in his reply he told me that it looked to him as if it might turn out convincing. The following year, 1913, on p. ix of the preface to vol. vii of the new edition of The Golden Bough {Balder the Beautiful) he fully accepted my solution in two or three charming sentences, such as he alone knows how to write, and now believes that the oak was secondary, the lightning primary, in the order of Aryan worship. I published two very brief papers on the point, one in the Archiv fur Religionswissenschaft (1912, p. 317), and another in Folklore for that year. It may be worth while to recapitulate once more the substance of these papers. In June 1910, while I was in Edinburgh, there occurred in the soul^Ji of England a period of constant electric storms. Digitiz&d by Microsoft® THE OAK AND THE THUNDER-GOD 39 culminating at my Oxfordshire home in a storm of the utmost ^violence, of which I have given an account in Kingham Old and New, ch. viii. So many trees were struck that on my return I could make them the objects of my walks, and became much interested in the different ways in which trees of different species were treated by the lightning. And why among the damaged trees was there not a single beech ? It is a common belief, I found, that beeches are never struck ; but I could find no scientific book to tell me the truth or the reason of this. At last my friend Mr. Gadney, the Oxford bookseller, sent me the fourth and last volume of Schlich's great work on Forestry} In this volume I found, in a remote corner (so to speak) near the end, a few precious pages about the effects of lightning on trees, which instantly struck me as likely to throw light on the questions raised in The Golden Bough. Here, in a tabulated form, I found the results of sixteen years of patient observation carried on and recorded by the foresters of Lippe-Detmold. T'ortunately the stock of trees in the forest was known, so that a perfectly accurate conclusion could be drawn as to the comparative danger from lightning to each species. The forest was stocked as follows ; oak 11 per cent., beech 70, spruce 13, Scotch fir 6. Thus the beech is far the most abundant tree in the forest, and the oak comparatively rare. Yet in the sixteen years only 33 beeches were struck, while the stricken oaks numbered 310, spruces 34, Scotch firs 108. The danger to a beech being taken as 1, that to a spruce was 6, to a Scotch fir 37, and to an oak no less than 60. These results were borne out by other observations made in France and Bavaria, and I believe that in Germany the subject had been further investigated before the recent war broke out. There is practically no doubt at all that the oak is more frequently struck than any other species of tree.^ ' This volume was the work of Dr. Hess, Professor of Forestry in the University of Giessen. ' Why this is so need not concern us here, if we are confident as to the fact. Various explanations have been given : see pp. 662 ff. of vol. iv. Digitized by Microsoft® 40 THE OAK AND THE THUNDER-GOD To this I may add from my own observation, that the oak is more conspicuous when it has been struck than other species : the bark is stripped more completely, the tree begins to die at once, and in a year's time is little more than a naked skeleton, discernible by the eye at a great distance. An un- trained eye would notice a stricken oak at once, but not a stricken elm or ash. On reading this chapter of Dr. Hess's volume I became assured not only that there is a good deal of truth in the belief of our folks about the immunity of the beech, but also that we have here a natural explanation of the religious connexion between the oak and the thunder. For the oak must have been a very abundant tree both in Europe and Asia in the days of the migrations ; Sir James Frazer has convincingly shown that its timber was very largely used by man in remote ages, e.g. by the people of the European lake-dwellings,^ who also seem to have used acorns largely to feed their pigs. Certainly the proportion of oaks to other trees must have been far larger than in the forests of lippe-Detmold, and the damage done to them so much the more conspicuous. And if we may argue from the well-known fact that in Greece and Italy at least, and probably also in Northern Europe, any spot or object struck by lightning was made sacred, and became matter of religious awe, each stricken tree would be liable to become an object of worship,^ and the number of oaks thus consecrated would greatly exceed that of all other trees. But there remains the question whether this liability of the oak suggested the idea of a god in the sky who descended into a tree in the lightning-stroke, or whether on the other hand the Sky-god was a primary conception, and the worship ' Goldm Bough, ed. 3, ii. 352 ff. ^ See XJsener, Kleine Schriften, iv. 477 ff. (Keraunoa). For Greece, Frazer, op. cit., p. 361, note 4 : for Italy, Wisaowa, R. K., ed. 2, p. 122. Festua (p. 377, Lindsay) : ' Serufertarios dioebant, qui quaedam aaorificia ad arbores fulguritaa faciebant, a ferto scilicet quodam sacrificii genere.' Cf. the gloaa on Fulguritum (p. 82, Lindsay). The Zulu idea is that anything struck by lightning has in it the power of the lightning : Crawley, Mystic Jtose, p. 232 (from Calloway). Digitized by Microsoft® THE OAK AND THE THUNDER-GOD 41 of the oak a consequence of his activity and power. Here we are in the region, not of fact, but of conjecture ; and I do not suppose that the truth can ever be ascertained for certain. For myself, however, I am disposed to think (as Sir James IVazer now thinks) that the facts put together above point to an extension of the conception of a Sky-god and his powers under the influence of new experience, material and religious, in a land of forests. The benevolent heaven- god of the steppes, worshipped by a pastoral people, might become associated with the oak in the.mind of that same people when he was found to strike that tree especially in the forest which they ■^ere labouring to clear for purposes of agriculture. It may be that recent theories and discoveries about tree- spirits and vegetation deities have kept our attention too exclusively fixed on Man as a worshipper in the agricultural stage, or in the struggles with nature which eventually landed him in that stage. But in the earlier nomadic life which I suppose we may postulate for the settlers in the Greek . and Italian peninsulas,^ the Sky-god was, I believe, a real Sky-god, ' pure as the naked heaven, majestic, free '. If on the very threshold of Roman religious history we iind him associated with the oak as Jupiter Feretrius, we have now an explanation which so far seems to cover the facts.^ ' ' So far as I can see, all the existing political societies, in the ancient wo^ld round the Mediterranean, and the modem world of Europe, seem to have arisen ultimately out of a state of things in which peoples who began their existence on the great grasslands which lie to the east, in south Brussia and beyond, and to the south in the deserts of Arabia beyond Jordan, have been forced or tempted to leave them and migrate into moister and more forest-clad regions, nearer the Mediterranean and Atlantic' Professor J. L. Myres, The Dawn of History, p.. 14. " Liv.i.lO.i; Dion. Hal. ii. 34 ; Aust, inRoscher'siea:»co»,B. v. 'luppiter', p. 671 ; Rd. Exp., pp. 129 fi. Digitized by Microsoft® 42 THE TOGA PRAETEXTA OF ROMAN CHILDREN Roman children of free birth, both boys and girls, from the time they could walk freely to the age of puberty, wore the toga praetexta, a robe with a purple border running along its straight edge,^ which border could be made conspicuous when the toga was properly adjusted. When his parents thought a boy sufficiently developed they had him publicly invested with the manly toga (toga virilis or pura) which was entirely white and without a stripe ; and girls went through a somewhat similar ceremony. What was the meaning or object of this purple-edged toga when thus worn by children ? They wore of course another ornament, the bulla, an amulet hung round the neck ; but we can hardly call the praetexta an amulet in historical times, whatever it may have been in its origin.^ Nor was it only a mark of tender years, like the jacket of a junior Eton boy, for it was worn also by grown men under certain special circumstances. Let us glance at these other uses of the toga praetexta. In civil life it was worn only by curule magistrates, i. e. those whose authority descended constitutionally from that of the Rex. In the first volume of his Staatsrecht (p. 402) Mommsen was inclined to correlate this part of the magisterial insignia with the right of being accompanied by lictors and fasces, and thus to explain its eventual extension to magistrates of municipia ; but the censors are an awkward difficulty in this reasoning, seeing that they had the toga praetexta but not the lictors and fasces. I should be disposed to think that the privilege belonged to those only who had the right of performing sacrifice on behalf of the community, which belonged originally to curule magistrates only. In course ' For the exact nature of the toga and its stripe see Companion to Latin Studies, ed. 2, p. 191. " The account given by Macrobius, Sat. i. 6. 7 ff., is obviously an impossible one. » Digitized by Microsoft® THE TOGA PRAETEXTA OF ROMAN CHILDREN 43 of time there were naturally extensions of the privilege ; and from a passage in Cicero's speech against Piso, taken together with the comment of Asconius which is luckily preserved, we learn that even magistri collegiorum, as well as magistri vicorum, used to wear it at the Compitalia, a movable feast of great antiquity. Cicero expresses disgust that Sextus Clodius, homo impurus, had worn it in this capacity.^ No doubt sacrifice of some kind was offered by these magistrates when they wore the praetexta. Next we note that this toga was worn by all the priests of the most ancient priesthoods during the performance of sacrificial duties, and that the Flamen Dialis, the most im- portant and ancient of them, wore it always and everywhere.^ The Fratres Arvales used it on the first two days of their great festival, and laid it aside on the third at the end of their sacrificial work.^ In the ceremony of devotio the person offering himself, who is at the same time priest and victim, puts it on for the act of self-sacrifice.* The Vestals did not wear it ; but here again the connexion of the purple stripe with sacrifice is apparent, for we are told that the suffibulum or head-dress of the Vestals, worn when they sacrificed, was praetextum.^ It is clear, then, that the praetexta was a holy garment, worn by priests during the time of sacrifice, by the priest of Jupiter at all times, and by magistrates who had the right to sacrifice on behalf of the State. As worn by children too it must originally have been a holy garment, for the children of ingenui, both boys and girls, were regularly employed in the houjsehold as ministrants attending on daily sacrifice ; that a special name attached to them in this capacity {camilli ' Cio. in Pisonem, iv. 8 ; Asconius, p. 7 of Clark's edition. For the connexion of the Compitalia with the vici and the collegia, see Boman Festivals, p. 280. The praetexta was no doubt worn by these magistri vicorum or collegiorum for the sacrificial part of their duties : see below. '' Serv. Aen. viii. 552 ; xii. 169. ' Henzen, Act. Fr. Arv., pp. 11, 14, 21, 28. ' Liv. viii. 9 ; x. 28 ; Sel. Exp., pp. 207 ff. ' Festus (Lindsay, pp. 474, 475). Digitized by Microsoft® 44 THE TOGA PRAETEXTA OF ROMAN CHILDREN and camillae), shows at once the antiquity and the regularity of the practice. '^ In the r^eligious service of the State the children of the priests themselves were thus employed originally, a privilege which later extended to other children horn of parents married under confarreatio and still living.^ These children must be investes, i.e. still unripe for the toga virilis ; they must not have gone out into the world, where they might meet with contarninating influences both material and spiritual, and "so be made unfit for those home duties that called for perfect purity. They were ' holy ', and wore the holy garment which their fathers used only when per- forming religious duties ; but like the Plamen Dialis, they wore it at all times and places, for like him they were ex- tremely precious both to family and State .^ There must have been a time when all children of ingenui assisted at the family sacra, attending on their father as priest. As the religion of the State outgrew that of the household, the idea of holiness and the corresponding dress survived in the State only for priests and magistrates having the right to sacrifice, i. e. capable of performing those priestly functions which were of supreme importance to the State. But the dress was retained for the children, not only because of the constant demand for them as ministrants, but because they were in reality unspotted from the world — ^an ethical idea here superimposing itself on the ceremonial one. And as a dis- tinction began to assert itself between ingenui and non- ingenui, the toga praetexta carae also to have the significance which used to be regarded as the chief one, i. e. it was looked on as a mark of free birth.* But though these two ideas of ethical purity and free birth ' Wissowa, E. K., ed. 2, p. 496 and note ; Marquardt, Staatsterwaltung, iii, p. 227 ff. ; Serv. Aen. i. 730. 2 See J.R.S., 1916, p. 187. " There are some suggestive remarks on holy garments in R. Smith's Bdigion o/ the Semites, p. 433. For the preciousuess of children in the patriarchal family see Myres, The Dawn of History, p. 20. ' Maorob. i. 6. 12, says that it was never originally lawful for children of libertini or peregrini to wear this toga, though libertini eventually secured the privilege* Digitized by Microsoft® THE TOGA PRAETEXTA OF ROMAN CHILDREN 45 got the better of the older idea of the children's toga, there are passages in Roman literature which seem, to me to indicate that the true meaning' was never entirely lost to the conserva- tive mind of the Roman of good family. Horace, Epode v. 7. The boy victim of Canidia's sorcery cries ; Per liberos te, si vocata partubus Lucina veris adfuit, per hoc inane purpurae decus precor, per improbaturum haec lovem, quid ut noVerca me intueris aut uti petita ferro belua ? There is here surely the faint echo of a religious appeal in the third line, for it is an appeal to Jupiter that immediately follows it. The toga praetexta makes the boy in some sense sacred. Persius, Sat. v. 30, alludes thus to his first manhood ; Cum primum pavido custos mihi purpura cessit bullaque succinctis laribus donata pependit . . . The word custos is significant ; one may think of it as expressing something more than . an ethical guardianship based on a respect for tender years. We could not apply the word to any part of the modern boy's dress, however characteristic of boyhood. Persius evidently reckons the toga with the bulla, as having the power to keep off evil influences. Pestus, s.v. Praetextum sermonem (pp. 282-3, Lindsay) : Praetextatis nefas erat obsceno verbo uti, ideoque prae- textum appellabant sermonem, qui nihil obscenitatis haberet.^^ By praetextati here boys are evidently meant, for just above he has been writing of the three praetextati patrimi et matrimi (boys) who conduct a bride home after a wedding. By obsceno is meant no doubt ill-omened. The praetexta ' I have given in the text the gloss as it stands in Paulus : in Festus (p. 282) it runs : ' Praetextum sermonem quidam putant diei quod prae- textatis nefas sit obsceno verbo uti : alii quod nubentibus, depositis prae- textis, a multitudine puerorum obsoena clamentur.' For the obscena verba, OT fescennini versus, see Marquardt, PrivaUeben, i. 52, note 4, Digitized by Microsoft® 46 THE TOGA PRAETEXTA OF ROMAN CHILDREN was a holy garment still, and when the bride's three hoys wanted obscena clamare, as was the practice on such occasions, they took off the praetexta beforehand. " Quintilian, Declam. 340 (the most striking passage I have found) : Ulud sacrum praetextarum quo infirmiiatem pueritiae sacram facimus et venerabilem (cf. Culex, pp. 25-6). This makes an excellent comment on the passages given above from Horace and Persius. Pliny, Nat. Hist. ix. 127 (writing of the various uses of purple dye) ; Fasces huic securesque Rornanae viam faciunt, idemque pro maiestate pueritiae est. Distinguit ab equite curiam, dis advo- catur placandis, omnemque vestem inluminat, in triumphali miscetur auro. Quapropter excusata et purpurae sit insania. In the last words purpura is the purple-fish, which Pliny in his quaint manner thinks of as having a mystical or uncanny power (cf. the religio in men's knees, above, p. 9). In the case of the boy the purple symbolizes his maiestas ^ — a word which entirely bears out the idea of boyhood which our other extracts have suggested, though expressing it still more strongly. Perhaps what was in Pliny's mind was that the serious awfulness of the Roman priest and magistrate, repre- senting the State, is also present in the free-born boy when he wears the praetexta. To these passages I will just add one from Columella ; it does not mention the toga, but gives a good idea of the germ from which the idea of its sacredness originally sprang. He quotes a number of authors, Greek, Carthaginian, and Roman, as agreeing that the duties of the household should be under- taken only by those who are chaste and continent ; and that for certain things the service of boys and girls is necessary, e. g. for going into the penus, the storehouse and seat of the Penates, in order to bring out provisions.^ One of the Roman authors mentioned here by Columella is C. Matius, the friend 1 Maiestas, a strong word to use of boyhood, may be explained as combin- ing the ideas of potency, dignity, and inviolability. ' Columella, xii. 4, Digitized by Microsoft® THE TOGA PRAETEXTA OF ROMAN CHILDREN 47 of Cicero and Caesar, who wrote on domestic economy. Thus we may be sure that the idea of the ceremonial purity of the boy, apart from liis ethical innocence, was not wholly extinct at the end of the republican period. I have another word to say about the Roman children, and especially the boys, before I leave them. What I have already said will have shown how valuable they were in religious ceremonial both in the family and the State ; but there are other facts which prove that they were far indeed removed from that savage condition of childhood which has been so much brought forward of late to throw some dim light on features of primitive Greek usage. It would seem that where there are elaborate initiatory rites at the age of puberty, ■ lasting perhaps for months, and symbolizing some notion of a death and re-birth, there also the boy is up to that time regarded as a nonentity, and of no value to the tribe. As Miss Harrison puts it, ' till he has utterly put away childish things, he cannot be a full member of the tribe, he may not know the tribal secrets or dances, he cannot perform any of the functions of the full-grown man '} Now all this might be said of the Roman boy, but in a very different sense. Erom his infancy until he takes his toga virilis, which is the sole sign of anjiihing in the nature of an initiatory ceremony, he is growing in knowledge as well as stature, and not only in knowledge of useful arts, but of the practice of public life. In the good old days of the republic, though incapable of ' performing the functions of a grown man ', he is learning all about them from his father. ' The older lads were con- stantly with their father, and by seeing what he did and sharing in it, they learned their future duties. They assisted in the service of the gods ; they listened as their father gave advice to his clients ; and from him they learnt the traditions of the family and the national heroes.' ^ The fact is that among the Latins, and probably among their kindred in Italy, the rites of initiation at puberty, if ' Themis, p. 19. ^ Mr. Murison, in Companion to Roman Stndies, ed. 2, p. 227. Cf. my ' Bome ', in the University Library series, pp. 61 if. Digitized by Microsoft® 48 THE TOOA PRAETEXTA OF ROMAN CHILDREN there had ever been any such as those of which Miss Harrison finds traces in Greece, had disappeared under the influence of that wonderful family life for which it seems impossible to find an adequate explanation ; and under the same benign influence the importance and value of children, especially the boys, had steadily increased, to the lasting advantage of the race. Among the Pagan tribes of Borneo, where I have found such an instructive commentary on primitive Roman usage, I note the same phenomenon less fully developed. Messrs. Hose and McDougall tell us that they have no initia- tory rites at puberty ; ^ and as we might expect after what has been said above, the boys help their parents in the house at an early age and onwards, and are in this way educated in social duties. The value, nay the maiestas, of Roman boyhood, is most beautifully pictured in the Aeneid ; for just as Aeneas is surely meant to represent the ideal Roman in manhood, so is Ascanius meant to represent him in boyhood. And the study of their characters is all the more interesting, because they are not what has been happily called ' static ' characters, but develop and grow in grace as the action of the poem proceeds. I have elsewhere written of the gradual strengthen- ing of the father's character ; ^ let me conclude these remarks with a brief sketch of the development of the son. It will confirm what I have said already, and help us to understand better what Juvenal meant by his four famous words, ' Maxima debetur puero reverentia '. When we first meet with Ascanius at the destruction of Troy, he is a mere child ; Eece autem complexa pedes in limine coniunx haerebat, parvumque patri tendebat lulum ; ^ big enough indeed to walk, for he trotted beside his father as they left Troy : Dextrae se parvus lulus implicuit sequiturque patrem non passibus acquis.* » Pagan Tribes of Borneo, i. 164 ; cf. ii. 24 and 185. It is now known that rites at initiation are entirely absent in many primitive races. « Bel. Exp., pp. 422 ff, ' Aen, ii, 674 if, • Aen, ii. 723, Digitized by Microsoft® THE TOGA PRAETEXTA OF ROMAN CHILDREN 49 And it is most important to notice that he is chosen as the object of one of those fire-omens of which the Romans were fond, and of which we have another example in vii. 71 ff. ; an omen which prompted Aeneas to call aloud upon Jupiter, who ' subito fragore Intonuit laevum '} In the first stages of the wanderings he is still a small boy, and made much of by women, first by Andromache on the coast of Epirus, and then by Dido. But he is growing ; Andromache, seeing him, is reminded that her boy would be growing too : Sic oculos, sic ille manus, sic ora ferebat ; et nunc aequali tecum pubesceret aevo.^ Dido might take him on her knee, but he was big enough to learn to ride, and delighted in his pony. Virgil loved this boy fondly, as this picture proves : Gaudet equo iamque hos cursu, iam praeterit illos, spumantemque dari pecora inter inertia votis optat aprum, aut fulvum descendere monte leonem.^ On the arrival in Sicily we note a great advancement in his career ; he takes part with his father, doubtless as camillus, in the rites of the Parentalia at the grave of his grandfather.* In the games that follow his part is to lead the boy-riders in the ludus Troiae ; and here we discover that he has a tutor or guardian, Epytides, custos et comes, and also a boy-friend Atys, Genus unde Atii duxere Latini, parvus Atys, pueroque puer dilectus lulo.^ Almost directly after this he suddenly takes on himself a man's part ; at the first news of the burning of the ships by the women he gallops off to the spot, and calls on them to stay the fatal crime. ^ Heu miserae cives ! nori hostem inimicaque castra Argivum, vestras spes uritis. en ego vester Ascanius ! 1 Aen. u. 681 S. ' Am. iii. 491 £f. ' Aen. iv. 157 ff. ' Aen. V. 74 ff. ' Aen. v. 568 ff. ' Aen. v. 667. 2262 D Digitized by Microsoft® 50 THE TOGA PRAETEXTA OF ROMAN CHILDREN His frightened magistri'^ could not stop him (669). Clearly Virgil meant to show that he was old enough to act for himself, and not only to take a rapid resolution, but to speak with force and point. But he' is not yet more than a boy, and he may not accompany his father, it need hardly be said, on his descent into Hades ; in the sixth book Ascanius is not even mentioned. When we come to the last six books, the books of war and bloodshed, we find Virgil in a difficulty about the boy. He could not make him grow up of a sudden, and probably did not wish to do so ; yet if he remained a boy, he could not take part in the fighting — that was what no Roman boy did as praetextatus.^ Ascanius was in fact in danger at this point of falling out of the story altogether. Virgil ingeniously saves him from this fate by introducing him just where there is no fighting at the moment ; he is on the very edge of it, but (except for an instant) is kept out of it. In the seventh and eighth books we see little of him ; he is once more the medium of an omen, and soon after this he hunts on horseback while his father is on his mission to Etruria, unwittingly killing a favourite stag and helping to kindle the war.* In Book VIII he is only mentioned as the subject of a prophecy of the foundation of Alba Longa.* But in the ninth book we see more of him than in any other. Critics have often puzzled their heads to explain why the immortal episode of Nisus and Euryalus is inserted in this book, without any very obvious connexion with the story.* I am rather inclined to think that one reason — certainly not the only one,- — -was to bring Ascanius once more to the reader's mind as a boy; just verging on his first manhood. He cannot join the two young heroes in their attempt to reach Aeneas at Pallanteum, though he would gladly do so. They, so_to speak, have taken the manly toga and are ripe ■ Servius explains magistri by custodes, of whom Epytidea was perhaps chief. Another was Butes, ix. 649. ' See Marquardt, Privatleben, p. 131, note 7, for a possible exception. 3 Aen. vii. 116 and 478 S. ' Am. viii. 48 ff. ; of. i. 267 fi, ' See e, g. Heinze, Virgil's epische Technik, Ip, 438. Digitized by Microsoft® THE TOGA PRAETEXTA OF ROMAN CHILDREN 51 for service, though Euryalus is really still a boy and addressed as such by Ascanius. But Aeneas's son can take his father's place without going into the battle, and does so. He is at the council to which Nisus and Euryalus ask admittance in their excitement (trepidos), and is the first to bid them enter and speak. In the scene that follows, one of the most beautiful in all poetry, the old counsellor Aletes gives the hint to Ascanius, ' with all his life before him '} to join with the gods and his father in promising never to forget the heroic pair ; and the boy catches at it, and in a speech that is almost inspiredi rises above his own boyhood, addressing Nisus with an appeal to the Penates and Vesta such as his father himself might have made, and then turning to the younger with the language of fond boyish affection, he promises to care for his widowed mother as if she were his own, should death overtake Euryalus.^ All accompany the youths to the gates of the camp ". Nee non et pulcher lulus, ante annos animumque gerens curamque virilem, multa patri mandata dabat portanda ; sed aurae omnia diseerpimt et nubibus inrita donant.^ This of itself would be enough to show what Virgil thought a noble Roman boy might do when suddenly called on to act in the absence of his father. But there is still more ; in the attack on the camp that followed the deaths of the two youths, Ascanius is for the moment drawn into the fight by the taunts of Remulus, and kills him with an arrow. But this must not go further ; Apollo intervenes for the destiny of Rome in the world : Macte nova virtute, puer, sic itur ad astra, dis genite et geniture deos.* These famous words are not supposed to be heard by the boy himself, but mark Apollo's secret delight ; then he descends in the form of the guardian Butes, and bids him stay his hand. Ascanius is withdrawn almost entirely from our eight during the fighting of the next two books. » This is Mr. Mackairs happy translation of ' integer aevi ' (ix. 255). e J fin. ix. 295 ff. ' Aen. ix. 310 ff. ' Aen. ix. 641 3 3 Digitized by Microsoft® 32 THE TOGA PBAETEXTA OF ROMAN CHILDREN He reappears in the twelfth book, but not to shed blood ; he is still a boy. He assists his father at the sacrifice that should seal the treaty with Latinus ; ^ he assists him too when wounded by the spear of Turnus.^ And he is still a boy, but on the very verge of his first manhood, when his father bids him farewell before going to his last fight : ^ Ascanium fusis circum complectitur armis summaque per galeam delibans oscula fatur ; ' disce, puer, virtutem ex me verumque laborem, fortunam ex aliis. Nunc te mea dextera bello defensum dabit et magna inter praemia ducet. Tu facito, mox cum matura adoleverit aetas, sis memor et te animo repetentem exempla tuorum et pater Aeneas et avunculus excitet Hector.' In such clear strong touches' Virgil has left an enduring picture of the growth, physical and mental, of a noble Roman boy, whose toga praetexta suggested not only the weakness of boyhood and its need of protection by a holy garment, but kept daily before the eyes and mind of its wearer that duty to family and State which was the foundation of all that was best in the Roman character.* WAS THE FLAMINICA DIALIS PRIESTESS or JUNO ? We always used to believe that as the Flamen Dialis was the priest of Jupiter, so his wife was the priestess of Juno. I do not know who was originally responsible for the inference, which was natural enough in the days when we used to believe that the Italian Jupiter and Juno were man and wife. I do not find it in Ambrosch, who is perhaps the surest-footed of the earlier writers on the Roman religious system. But soon afterwards the assertion was made by Preller, though not quite so definitely as by later writers ; the Flaminica, he says, was ' Aen. xii. 168. ' Aen. xii. 385. . ' Aen. xii. 433 ff. * The substance of this account of Ascanius was inserted in The Death of Turnus, pp. 87 ff. Digitized by Microsoft® THE FLAMINICA DIALIS 53 ' eine priesterliche Dienerin der Juno '. On this statement Jordan in his edition added no comment (i. 122). Marquardt in his book on the Koman religious constitution says plainly that she was the priestess of Juno, and here again Wissowa, who edited the second edition, made no remark. Roscher in his tract on Juno and Hera went a little further ; the Flaminica Dialis ' hatte den Opferdienst der Juno zu versehen '. Hence the same statement has found its way into the article on Juno in the Mythological Lexicon, and is repeated still more emphatically in Aust's excellent article on Jupiter in the same work. I do not wish to say a word that might seem to be scoffing at this succession of distinguished scholars, to whom I owe so much ; but I confess that this is not the only occasion on which I have found German professors , following each other like sheep through a hole in a hedge, without inquiring whether they were going the right way. But I must own that I might myself never have been led to test the value of these assertions, if they had not been used to support a much more important inference, and one of the utmost interest for the student of early Italian religious ideas. Preller allowed himself to assert that the Flamen and his wife appeared before the people as in some sense the living images ■of the deities of light whom they served. Roscher took the hint, and after his manner, carried it out to its logical conse- quences. He saw in the Flamen and his wife and the rules of life which governed them, a means of getting at the ideas which lay at the root of the cult of Jupiter and Juno. In his view these are husband and wife, as well as gods presiding over marriage {Myth. Lex., s. v. ' Juno ', p. 590 ; cf . ' Juno and Hera ', p. 63). Or, as the author of the article on Jupiter puts it, ' die alterthiimliche Institution des Flamen und der Flaminica beweist auch, dass die paarweise Gotterverehrung in Italien eine urspriingliche war '. This inference is to the explorer at first sight as water in a thirsty land. He knows that the cult is the only absolutely safe guide in the study of old Italian religion ; he knows that the question — a vital one — whether the oldest Romans Digitized by Microsoft® 54 WAS THE FLAMINICA DIALIS thought of any of their deities as married couples, cannot be decided by literary evidence alone.^ But if it can be proved that the priest of Jupiter was the husband of the priestess of Juno; he feels at once that he has hold of something definite and trustworthy. The peculiar sanctity of the marriage tie in this case, together vsdth the strange restrictions under which the pair was placed, and the undoubted antiquity of the priesthoods, taken in comparison with evidence from other races as to the relation of gods and priests, prepare him to accept the inference as one of great interest. If the water so much desired by the explorer should not turn out to be a mirage, we may fairly believe that Jupiter and Juno were really a married couple, and that the oldest Italians had got at least as far as this on their way towards polytheism. But alas, the water is no water, and the mirage has deceived many. We know that the Hamen Dialis was attached to the cult of Jupiter ; but what is the evidence that the Haminica was priestess of Juno ? All the writers I have quoted, and some others of less importance, cite but a single passage, and that from an author whose authority on such' a subject is far from weighty, and who in this particular instance expresses himself doubtfully. Plutarch in the 86th Roman Question writes of the Maminica as Upav rrjs "Hpas elvai bonovcrav. In this 86th quaestio Plutarch may have been drawing, directly or indirectly, on a gloss of Verrius Placcus (cf . Pestus, s. v. Mains mensis, Lindsay, p. 120) ; but there is nothing in Pestus to bear out his remark about the Flaminica, and the word boKovcrav shows pretty clearly that what he says of her is simply his own suggestion, which of itself must be quite worthless. Apart from this passage I can find no ancient authority for the idea that the Plaminica had anything to do with the cult of Juno.^ The truth is rather to be found in such glimpses as we find in good Latin authorities, who were interested in the actual facts of the cult. ' See my Religious Experience oj the Roman People, pp. 114 ff. ^ Nor haje I ever seen one quoted since 1895, when this paper was first published. Digitized by Microsoft® PRIESTESS OE JUNO ? 55 In a Verrian gloss on flammeum I find the following (Paulus, p. 82, Lindsay) : ' flammeo vestimento flaminica utebatur, id est Dialis uxor et lovis sacerdos, cui telum fulminis eodem erat colore.' This definite statement, coming from a good authority, that the Flaminica was lovis sacerdos, is borne out by a passage of Macrobius. On the nundinae, he says, quoting Granius Licinianus, she offered a ram to Jupiter in the regia ; ^ on the other hand, on the Kalends of every month from March to December, which were specially sacred to Juno, it is not the Elaminica who sacrifices to Juno, but the regina sacrorum (Macrob. i. 15. 19). And so far as I know, none of the rites in which the Flaminica was concerned have any reference to Juno . Thus it is impossible to hold any longer to the old fancy that the Flaminica was Juno's priestess, or to build on this assumption important conclusions about the early Italian conceptions of the relations of Jupiter and Juno .^ It is as well also to remember that by the old Roman methods of marriage, which brought the bride under the manus of her husband, she thereby took upon her the duty of attend- ing to the sacra of her husband's family, and abandoned for good and all those of her own father's. Especially was this so when a pair were married by confarreatio, from which there was practically no possibility of release (Festus, p. 79, Lindsay), and which was necessary to the marriage of the Flamen and Flaminica Dialis. Thus the Flaminica after marriage must have been devoted to the cult of Jupiter, unless it can be shown that the Flamen had any part in the worship of Juno ; and for this there seems to be no evidence at all.^ ' SaU'i. 16. 30. " See e. g. Sir James Frazer's Adonis, Attis, Osiris, p. 410. ' In the second edition of his Melig. und Kultus der Homer, p. 191, Wissowa has at last declared his adhesion to the view that before the introduction of the Capitoline worship from Etruria there was no relation between Jupiter and Juno such as to suggest that they were a married pair. Digitized by Microsoft® 56 THE ORIGIN OF THE LAR FAMILIARIS There is a pretty story, connecting the great family of the Valerii with the origin of the ludi saeculares, which has come down to us through one of the name, Valerius Maximus. It has an interest as being one of the many family legends which contributed to the semi-fictitious history of Eome in which Romans themselves delighted ; and incidentally it serves my purpose at the present moment by giving us a glimpse of what I believe to be the wrong idea of the lar familiaris. Valerius, or Valesius as the writer calls hirtt, was a rich Roman who had a villa near Eretum in the Sabine country. He had three children, two boys and a girl, and was a happy man. But a terrible pestilence fell upon the land, and all three children fell desperately ill ; the doctors could do nothing for them. The father went to the hearth to get warm water for them, and there it occurred to him to fall on his knees and pray the Lares familiares to transfer the plague from the children to himself. Then a voice was heard telling him to take boat down the Tiber ' to Tarentum ' and fetch warm water for them from the altar of Dis Pater and Persephone. Sore puzzled, he obeyed the voice, and eventually discovered that underground altar at a place called Tarentum in the Roman Campus Martins, the scene of the first festival of the name (ludi Tarentini), and of the nightly celebrations of the ludi saeculares in the year 17 B. c.^ The story cannot well be much earlier than the second century B.C.; the origin of the Greek (or Graecized) form of the ludi dates from the year 249 B. c, and it is this form which we see in the Valerian legend.^ And in the very first sentence of the story, in the phrase lares familiares, we have additional proof of its lateness. Valerius in fact is simply using the language of his own time, when the plural Lares ' Valerius Maximus, ii. i, 5. " See Sel. Exp., p, 440. Dis Pater and Persephone are Greek deities. Digitized by Microsoft® THE ORIGIN OF THE LAR FAMILIARIS 57 had come to be used vaguely for the household deities in general, and when the specially defined range of function of each of them was fast being forgotten. We all know of course that Vesta was the spirit of the hearth-fire, from which position she was never deposed ; so too with the Penates, the spirits of the penus or storeroom. But towards the end of the republican period the new arrangements of the house in the city, and more or less also in the villa rustica, seem to have rubbed off the individualities of these spirits. The house came to be divided for practical purposes into two parts, the Roman part and the Greek part, and of these the Roman part was retained for reception rooms, while the household spirits, Vesta, Penates, and Lar or Lares, retired into the Greek part, the peristylium and its belongings, which was reserved for the family and its intimate friends.^ It is easy to see how such a change as this would affect the domestic spirits. They were all reckoned together henceforward as Lares or Penates, or Lares and Penates.^ When you prayed to them, it was convenient to address them all by one collective name, and this is exactly what Valesius did when he knelt before the hearth. As his immediate object was to get hot water, and as the hearth was at all times ideally the religious centre of the house, he naturally addressed the Lares there ; but what he meant by Lares was the whole group of household deities, which, by the way, he might just as well have called di penates. And it does not in the least follow from this that the genuine original Lar had anything whatever to do with the hearth-fire. One of the many valuable contributions of Professor Wissowa to our knowledge of the real Roman religion is his convincing proof (so at least I regard it) that the Lar of the primitive Roman family was not an inhabitant of the '■ See Social Life at Borne in the Age of Cicero, p. 242. " Wissowa, S. K., ed. 2, p. 168 and note 9. In this way Lares came to be used for the home — a use for which I need not quote proofs. How this came about may be seen, among other passages, by Cic. de Legibus, ii 42 ' vexati nostri Lares familiares ', &c. (of what happened to his house during his exile). Digitized by Microsoft® 58 THE ORIGIN OF THE LAR FAMILIARIS house, much less the spirit of an ancestor, as has so often been maintained since the early days of our knowledge of animism. Preller, who wrote before the appearance of Tylor's Primitive Culture, and also before that of Fustel de Coulanges' famous book La Cite antique, came very near the truth, for he was not using the comparative method, which needs very careful handling in Roman matters. These two books seem to have overwhelmed Preller and the early Roman evidence, but the authorities they go upon are nearly all of them writers of the Empire. I may quote here a few lines of de Coulanges : ^ Ces ames humaines divinisees par la mort etaient ce que les Grees appelaient des demons ou des heros. Les Latins leur donnaient le nom de Lares, Manes, Genies. ' Nos ancetres ont cru, dit Apulee, que les Manes, lorsqu'ils etaient mal- faisants, devaient etre appeles larves, et ils les appelaient Lares lorsqu'ils etaient bienveillants et propices.' On lit ailleurs ; ' Genie et Lare, o'est le meme etre ; ainsi Font cru Jios ancetres.' Et dans Ciceron ; ' Ceux que les Grecs nomment demons, nous les appelons Lares.' (This last is from Cicero's Timaeus, ll ; and the other passage is from Censorinus, de die natali, 3.) ^ ITorty years after the appearance of these two remarkable books, when we had been thoroughly inured to the belief that the Lares were dead ancestors, two others of considerable note reiterated it with a much larger knowledge of things Roman — the excellent Familienfeste of E. Samter, and the no less valuable work of the Italian De Marchi on la Religione della vita domestical I am not writing an article for one of those vast encyclopaedic collections of facts and theories, or facts manipulated to suit theories, which are now in fashion in the learned world ; and I may leave these two writers to the student, merely noting that after studying them again and again I remain unconvinced either that the Lares were ' La Cite antique, p. 19. " In spite of what Cicero here says in his translation of Plato, we shall soon find him helping us in the right direction. Apuleius and Censorinus are simply using the language of their own time. " Samte^ pp. 105 ff. ; De Marchi, i. 28 ff. Digitized by Microsoft® THE ORIGIN OF THE LAR FAMILIARIS 59 originally inhabitants of the house, or that they were the dead ancestors of families. When I wrote my Roman Festivals I was still to some extent in doubt, and allowed myself to write of the Lares that ' they may have been the spirits of dead ancestors duly buried '} But since then the appearance of Wissowa's Religion und Kultus, and his controversy with Dr. Samter on this point,^ has left no ftirther doubt in my mind. I am only here intending to fill up what was a gap in Wissowa's original argument ; but I may note that he has made use of my fencing in the second edition of his great book. Wissowa's main arguments are the following ; 1. There is no place for the Lares in the primitive house. Its centre point, the hearth, was occupied by Vesta ; the penus by the di senates ; and these two represent the essentials of human life, together with the spirit of the spring (whatever that was, it was certainly never known as a Lar). 2. There is no association of the Lares with the cult of the dead ; the spirits of the ancestors are di parentes (a sufficient appellation), and their cult, if it can be so called, went on at the grave of the family. Ingenious attempts have been made to prove that the dead were originally buried in the house ; but the archaeological evidence is entirely against this.^ ' If the Lares worshipped in historical times at the hearth (as in the story of Valesius) were really the spirits of dead ancestors, then those ancestors must have been buried under the hearth ' ; and this neither Samter nor De Marchi is bold enough to maintain. 3. OTtr oldest sources of information connect the Lares not with the house but with the land. These are (a) the hyimi of the Arval Brethren,* whose special duty it was to look after » p. 337. ' ArcMv for 1904, pp. 42 S. ' The terremare of North Italy, from the inhabitants of which the Romans were almost certainly descended, had their cemeteries outside the settle- ment : Peet, Stone and Bronze Ages in Italy, pp. 364 ff. This principle held good with their descendants. * Henzen, Acta, p. 26 ' Enos Lases iuvate neue luetve Marmor sins inourrere in pleores, satur fu fere Mars ', &,c. The semunes are the only other Digitized by Microsoft® 60 THE ORIGIN OP THE LAR FAMILIARIS the religious protection of the ager Romanus : here they are associated with Mars in the invocation during the sacred dance — a deity whom no one has ever yet tried to connect with the cult of the dead ; (6) the Lares are the chief object of worship at the Compitalia, a very ancient festival of the farm ; and their seats were not in the house but at the compita, i.e. the spots of ground where in the parcelling out of agricultural allotments, divided by semitae or viae, one of these crossed another at right angles. At the point of crossing we learn that there was a kind of chapel with as many niches as there were allotments here touching each other, each niche containing the image of a Lar in historical times, and,- as we may presume, the Lar himself in the spirit, before images came in.^ What exactly was the relation of each Lar to the allotment he represented is not so easy to say ; but we shall not be far wrong if we guess that he was a protecting spirit, upon whom the family owning that bit of land could call for help, exactly as the Arval Brethren of the developed State called on the Laresas a whole, ' Enos Lases iuvate '. Thus we get a gap filled up in the category of spirits in whom th^ family was interested ; for if the Lares were spirits of the house and of deceased ancestors, where are we to look for individual guardians of the land belonging to that house, and once cultivated by those ancestors ? The interpretation of fanuly deities cannot be rightly undertaken without a clear idea of the economy of the family (in which we include all members of the economic unit, free and unfree, all who worked on its land) ; and if we insist on bringing the Lares into the house in that early age of which I am speaking, we have no spirits left to perform the necessary duties of guardians of the family's land. Yet the Lar, originally one for each family, did undoiibtedly spirits mentioned, but of these unluckily we know nothing certain {Roman Festivals, p. 136). ' See Roman Festivals, p. 279, where the references are given. Cf. Cic. de Legibus, ii. 19 ' lucos in agris habento et Larum sedes ' ; ib. 27 ' Nee ea quae a maioribus prodita est cum dominis turn famulis, posita infundo villaeque Ainspectu religio Larium '- Digitized by Microsoft® THE ORIGIN OF THE LAR PAMILIARIS 61 find his way into the house : that is made certain by innumer- able allusions in Latin authors of the literary age, among the earliest of which is a well-known prologue of Plautus, spoken by the Lar himself {Aulularia, 2 fE.) : Ego Lar sum familiaris ex hac familia unde exeuntem me aspexistis. hanc domum iam multos annos est quam possideo et colo patri auoque iam huiius qui nunc hie habet. And this Lar seems to inhabit the focus ; for he goes on to say that the present owner's grandfather entrusted him with a treasure secretly, and buried it ' in medio f oco '. A hundred other passages will confirm this evidence of the Lar in the house ; and the question for us is, how did he come off the land and get a footing in the house ? My own belief is that he came with the slaves of the familia. The familia, I need hardly say, was an economic unit settled on the land, including unfree members as well as free ; ^ and it is important to understand that the unfree, in early Roman days, would not be slaves drawn from distant regions, as in the later ages of the republic, but Italians, captives from neighbouring cities, or debtors condemned to slavery. Had these slaves any share in the religious life of the economic unit ? Assuredly they had no part in the worship of Vesta and the Penates, or in the cult of the di parentes of the family. The hearth was the peculiar care of the daugh- ters of the family, as the penus was of the materfamilias and the children ; and in what we know of the cult of Vesta and the Penates as embodied in the State worship of a later age, and in that of the ancestors at the Parentalia, there is no trace whatever of a share taken by any strange or unfree person. But it is not so with the cult of the Lar ; we know that slaves took part in the Compitalia, where the Lares of the ccynvpita were the object of worship, as well as in other festivals which were clearly descended from the religion of > Cf. Mel. Exp., p. 70. It is worth noting here that the adjective familiaris came to be used for a slave : Seneca, Ep. 47. 14 ' Dominvun patrem familiae adpellaverunt, servos, quod etiam in mimis adhuc durat, familiares.' See other references in Marquardt, Privatleben, p. 172. Digitized by Microsoft® 62 THE ORIGIN OP THE LAR PAMILIARIS the farm, the Paganalia and probably the Saturnalia. For the Paganalia we have the evidence of Ovid : ■*■ Vilice, da requiem terrae, semente peracta ; da requiem terram qui coluere viris. TibuUus is generally thought to be referring to this in a passage on which I have commented elsewhere ; ^ in any case he is speaking of one of the country festivals in which groups of familiae joined : Tunc nitidus plenis confisus rusticus agris ingeret ardenti grandia ligna foco, turbaque vernarum, saturi bona signa coloni, ludet et ex virgis extruet ante casas. And at the Saturnalia in December there can be little doubt that the well-known custom by which slaves were treated as being in a position of entire equality with their masters, was descended from the religion of the farm.* As to the Compitalia, which was a festival originally of the rural folk owning allotments of land, under the old Latin system of rectangular divisions separated by semitae, which (as was said above) crossed each other at compitM, there is practically no doubt that unfree labourers took par^; in it. Dionysius tells us that this was a privilege conceded to them by King Servius TuUius in explicit terms, since the Lares were pleased with the service rendered by these workers.* He is giving a popular explanation of the Compitalia of the city in his own time ; but that explanation itself has its root in the old country practice, as may be seen if we compare with it what Cato says in his treatise on agriculture about the relations between the slaves and the Lar of the compitum. In the fifth chapter Cato tells us that the steward (vilicus) must not sacrifice ' in compito aut in foco ' except at the Compitalia ; ^ words which seem to me to mean that at the ' Fasti, i. 667. As the vilicus was usually himself a slave, there is little doubt that slaves are here meant : of, Wallon, Hist, de VEsclavage, iii. 211, ? Tib. ii. 1. 21. = Roman Festivals, p. 272. < Diony^ iv. 14, ad fin. ' ' Rem divinam nisi Compitalibus in compito aut in foco ne faciat,' Digitized by Microsoft® THE ORIGIN OF THE LAR PAMILIARLS 63 Oompitalia, and only then, he might sacrifice for the unfree workers to the Lar at the compitum (in convpito), or to the Lar in the homestead {in foco), the ' ant ' suggesting that where the Lar had already found his way to the house, there the vilicus might follow him for the purpose of sacrificing. So too his wife, the vilica, might and should adorn the fo^us with a garland on calends nones and ides, and on the days of the lustratio agri she might pray to the Lar familiaris ' pro copia ' ; representing the female slaves of the house, she thus had a modest share in the worship of the Lar of that economic unit (familiaris), but not in that of any other spirit or deity .^ In all other i:ites of this unit the paterfamilias alone had the right of sacrifice, except when he deputed this duty to his steward. As in the Anglo-Saxon manor the duties of an absentee landlord were performed by the Gerefa, usually a villanus or serf, so in the frequent absence of the Roman rural paterfamilias, becoming more constant as Rome grew more powerful, his duties passed to the unfree vilicus. It seems fairly clear, then, that the worship of the Lar at the compitum or in the house came more and more distinctly to be the right of the vilicus and his wife, and through them of the slaves of the familia, perhaps even without the necessity of obtaining leave or receiving orders to that effect from the master (dominus) ; and thus the Lar came to be called familiaris, which plainly indicated that in his cult the slaves were included. Now we know that it was the old custom for the slaves to sit at the meals of the family on benches (subsellia) ' below the salt ' ; ^ it was therefore quite natural that they should see and worship there the only deity of .the farm to which they were closely attached. What more natural than that they should bring the Lar with them from the compitum to the house, especially in the frequent absence of the master ? In other words, as the slaves came to be more and more ' Cato, 143 ' kalendis, idibus, nonis, festus dies cum erit, coronam in fooum indat, per eosdemque dies lari familiari pro copia supplicet.' ' Marquardt, PrivaUeben der Bomer, p. 171, Digitized by Microsoft® 64 THE ORIGIN OP THE LAR PAMILIARIS distinctly recognized as members of the economic community of which the house was the centre, the one deity whom they had always worshipped on the land followed them into the house . In this, as in a hundred other instances , a more accurate knowledge of the details of Roman religion, once so much decried and despised, will be found to have a most important bearing on the social and economic history of the wonderful Roman people. FORTUNA PRIMIGENIA It is quite in keeping with the tendencies of the time that I am unlearning in my old age some of the settled conclusions of my yoUnger days. Lately I convinced myself that the Romans did not see Jupiter in the triumphator, as I used (more or less doubtfully) to think. Now I am going to retract another opinion, expressed by me in print at least three times over, viz. that Fortuna primigenia of Praeneste was the firstborn daughter of Jupiter. I owe this change of mind mainly to my friend Mr. Mackail, who interested himself in the question after reading what I said of Portuna in Roman Ideas of Deity, Lect. III. I sent him Jordan's treatment of the subject,^ which has been meekly followed, German fashion, by all who have touched on it since, including myself ; and being quite free from traditional bias, he at once pointed out that though Jordan rightly interprets the passages on which alone we depend for our knowledge of the word primigenius, he perversely arrives at a wrong conclusion as to the meaning of the word itself. Jordan held that it means ' quicquid primum genitum sit primaeque servet geniturae signum ', and proceeds to argue that as a title of Portuna, primigenia must mean ' firstborn '. What brought him to this I will presently explain ; but let us first look carefully into the passages which he quotes from Varro to show us the meaning '■ SymiiglaR ad Historiam religionum Italicarum alterae : Konigsberg, 1885. See also my Soman Festivals, pp. 223 ff. Digitized by Microsoft® FORTUNA PRIMIGENIA 65 of this rare word. Varro seems to like it and to wish to save it from becfoming obsolete. Varro, B. B. i. 40 ' Primigenia semina dedit natura, reliqua invenit experientia coloni. Nam prima, quae sine colono priusquam sata nata ; secunda, quae ex iis collecta neque priusquam sata nata.' Nature has given the original seeds ; the rest were discovered through experiments made by the farmer. Virgil, in Georg. ii. 9 ff ., seems to be following this passage, in distinguishing wild plants from cultivated ones by supposing that the former ' came of themselves ' ; Varro's book was probably just out when Virgil was at work on the Oeorgics. The same idea is found however in Lucretius, ii. 1156 fE., and this famous pass.age, and indeed the idea generally, seem to fall in with the favourite notion of a golden age long past, about which Greek poets, Hesiod and Aratus, had sung before him. Varro, of course, should have distinguished more clearly between natural and artificial planting ; but it is in any case plain that by 'primigenius he means ' natural ', existing before man began to meddle with agriculture ; the original seed of all, and therefore the most natural. Note that the termination genius may here be rather active than passive in meaning ; the original seed has a dynamic principle within it. Certainly firstborn, in the ordinary sense of that word, is not the meaning here. Nor is it the meaning in Varro, B. B. ii. 2 iuit., where he says that this book is to be devoted to the subject of stock, and that he will begin with primigenia pecvuria {res), i.e. the most ancient kind of stock, for, as he goes on to explain, the sheep was the first wild animal to be domesticated by man.^ Varro seems beyond doubt to have been fond of the worS.for in de Ling. Lat. vi. 36 and 37 he uses it four times in the course of a few lines. He seems to have found the phrase primigenia verba in a grammatical treatise by one Cosconius, who said that of such ' original ' words there were in Latin about ' The word primigenius ia here put into the mouth of Attious, who takes part in this dialogue ; and I incline to think that he is supposed to be alluding lightly to the fondness of Varro for the word, exemplified in the previous day's conversation (i. 40). 2252 E Digitized by Microsoft® 66 FORTUNA PRIMIGENIA a thousand. From this Varro concludes that of words derived in various ways from these there were about fifty thousand. He then goes on to explain the sense he and his predecessor give to primigenia verba. ' Primigenia dicuntur verba ut lego, scribo, sto, sedeo, et cetera quae non sunt ab alio quo verbo, sed suas habent radices', i. e. they are not grafted on or propagated from other words, but grow from their own root. ' Contra verba declinata sunt quae ab alio quo oriuntur, ut ab lego legis legit legam, et sic indidem hinc permulta.' He does not mean th&t primigenia verba are what we call roots, of which of course he knew nothing ; but what he believed to be the original forms, later modified by the ingenuity of man. As with the primigenia semina, the notion is unscientific, but we can understand clearly enough what he means. I may add here that the pedantic Emperor Claudius, when he doubled two of his legions (XV and XXII), gave to the original ones of these numbers the title primigenia. Festus, representing Verrius Flaccus, had a gloss on primi- genius sulcus, now hopelessly mutilated (Lindsay, p. 270), but Paulus preserved it in this form : ' Primigeidus sulcus dicitur, qui in condenda nova urbe tauro et vacca designa- tionis causa imprimitur.' ^ This must mean the first furrow made by the plough round the urbs that was to be, which furrow was to be enlarged afterwards into a foss. This, though not so useful as the other usages, tends to confirm what was said above of the active or dynamical meaning in the word ; seeds, words, foss, are all developed out of an original to which the epithet can be applied. Lastly, Arnobius in the fourth century A. D. writes of prirriigenios ortus (ii. 61 and 70) ; but his text is uncertain and his meaning not very clear. I do not think that he throws any light on the meaning of. the word ; and I know of no other occurrence of it in the least likely to help us. The meaning as Varro understood it is perfectly clear. If we apply it to Fortuna primigenia, the meaning of the title would seem to be, the Fortuna who (or whose cult) was the ' I do ntit know why Jordan, p. 6, says that this ' e Varronis libris fluxit ' . Digitized by Microsoft® FORTUNA PRIMIGENIA 67 original of all the Fortunae afterwards suggested or devised by man's experience, the one that gave rise to the whole series. And it is not impossible that this may really have been the meaning.. There were many Fortunae in Latium, among which she of Praeneste always retained her supremacy, or , at any rate her claim to it. At Antium, where as at Prae- neste she presided over something in the nature of an oracle, she had, so far as we know, no cult-titles. But since Jordan wrote his dissertation every one has followed him in translating (Fortuna) primigenia ' firstborn '. I must now explain why this was. I feel fairly sure that Jordan would not have thus translated the word, if it had not been for the discovery, three years before he wrote, of an inscription from Praeneste which runs thus (Dessau, no. 3684 ; C. I. L. xiv. 2863) : Orcevia Numeri nationu gratia Fortuna Diovo fileia Primo- genia donom dedi. Here Fortuna appears to our astonishment as the daughter of Jupiter, a position she claims nowhere else in Italy. There are indeed two later inscriptions at Praeneste in which she appears as lovis puero (i. e. the daughter of Jupiter, by an old usage of the word puer, as it is generally understood) ; in one of these the word primigeniae follows, in the other it is absent. The Praenestines must have been most incon- sistent people. The older inscription has the familiar filia, the later ones have the antique use of puer, meaning daughter. One man calls Fortuna primigenia, another does not. Still more astonishing is it that Cicero, who seems to have known something about the cults of Praeneste, declares that there actually was among them one of Jupiter puer, who was seated with Juno in the lap of his mother Fortuna, and was fondly worshipped by mothers {de Div. ii. 85 ; cf . Eoman Festivals, p. 224). But this seems to be a case of confusion arising from a misinterpretation by Praenestines of statues and inscriptions, none of which were really primitive or of pure Italian origin. Even the oldest, the first quoted above, cannot well be earlier than the fourth century B.C., unless I am greatly mistaken, e2 Digitized by Microsoft® 68 FORTUNA PRIMIGENIA and is probably still younger. It is hardly safe, where there is such confusion as this, to conclude that in Orcevia's dedica- tion primigenia is to be taken with Diovo fileia, and under- stood in the sense of firstborn. Nor even if it must be so taken, ,as some will argue, should it be considered authoritative evidence of a common belief. Except this particular dedica- tion, there is not only no ground for such a belief, but two special points which make against it, 1 . It is certain that Fortuna is never mentioned or suggested by any Latin author as the daughter of Jupiter, and that no Latin deity was thought of, except after becoming Graecized, as the son or daughter of another deity. Praeneste "was saturated with Greek influence, and Greek works of art were early introduced there ; after which, we know not how, some notion must have got abroad such as is expressed in these inscriptions. But it never spread beyond Praeneste ; when the great goddess of that city was taken to dwell in a Roman temple in 194 b. c, she went as Fortuna primigenia simply,^ and no Roman ever suspected her of being Jupiter's child. 2. The word primigenius never means firstborn in Roman literature ; that I hope I have already proved. If it had ever had this meaning, why was it not retained ? why be content with the awkward ' natu maximus ' ? The make of the word does not suggest this meaning to us, nor did it suggest it to Romans. When Cicero in his de Legibus, ii. 28, is discussing the various forms of Fortuna, and their particular objects, he simply says of Fortuna primigenia that she is a gignenda comes — a new aspect of the deity indeed, and perhaps a fancy one, but with no relation whatever to firstborn children, or to the deity herself as any one's firstborn. Nor even at Prae- neste (Div. ii. 85) does he know of primigenia in such a sense. Without doubt Fortuna primigenia is the real title of the Praenestine deity, and until we have more explicit evidence we may let the Diovo fileia drop out. Without doubt too she " The hppelessly mutilated gloss in Festus (p. 272, Lindsay), might have solved some difficulties, as Verrius knew Praeneste. Digitized by Microsoft® PORTUNA PRIMIGENIA 69 was not called primigenia because she was a firstborn daughter. What the title really meant was, so far as I can see, unknown to the Romans, like many other such cult-titles, e. g. Gradivus as used of Mars. We ourselves can only guess at it, and I cotafess that I have no definite conviction. I have already said that if we hold to the meaning of the word as Varro understood it, we should think of a Fortuna who claimed to be the original deity of the name, before man began to inter- fere with her, adding artificial titles such as Fortuna huiusce diei, Fortuna muliebris, Fortuna virilis, and so on ; as the first of the series, which gave rise to the rest, and therefore not standing in the relation of an eldest child to the younger ones. On the other hand, there may be something in the ' a gig- nendo comes '^ of Cicero. She n^y originally at Praeneste have been thought, of as the equivalent of genius, the spiritual power which the Romans called Juno, attending on women as genius on men. This is what Mr. Mackail seems disposed to see in her. Or to use his own words in a letter to me, ' Fortuna primigenia would be the power which determines the whole movement of any life from its outset.' He refers to Iliad vi. 488-9, and Od. vii. 197-8 ; and there can be no objection in the case of a Praenestine cult in thus referring to Homer. This would fall in with the fact that at Praeneste and elsewhere Fortuna was specially worshipped by women, without doubt in special connexion with childbirth. Possibly she held at Praeneste the same place, or something like it, which Juno Lucina held at Rome and elsewhere ; for Lucina, I believe, is not known to have inhabited Praeneste, where Juno was of secondary importance. It is worth noting that there was a lunonarium, i. e. a cella or shrine, in the great temple of Fortuna there. I have not yet mentioned that the cult-title primigenius is found; in a few inscriptions applied to Hercules, both at Rome and in the province of Baetica (C. I. L. ii. 1436, 1545, 2463 ; and Dessau, no. 3433), and that Jordan claims (op. cit., p. 8) ' If this reading be right, there is a lacuna immediately after comes. Digitized by Microsoft® 70 FORTUNA PRIMIGENIA that here too we are to understand it as firstborn, (of Jupiter). I think that after what has been said above we may be sure that whatever it does mean in this instance, it certainly does not mean that. PASSING UNDEE THE YOKE In ancient Italy, when an army surrendered in the field, there were three alternatives before the victors for dealing with the vanquished. First, they might put them to death ; there was nothing to prevent this but the feeling and tradition among Italian peoples in historical times against unnecessary bloodshed.^ Secondly, they might keep them as prisoners of war, and sell them as sl*ves ; but in early times this was practically out of the question, partly owing to the difficulty of feeding and guarding them before they were sold, and partly because the machinery of sale, the slave-agents at hand and the slave-markets in the cities, had not then been invented. Thirdly, they might let their captives go free, with or without imposing conditions on them to be ratified by their State. This was really the simplest and easiest plan, and was adopted in the few cases recorded by Livy in which whole armies were captured ? But before the vanquished were dismissed, they were made to go through the ceremony of ' passing under the yoke ' (sm6 iugum missi), which Livy, when he first mentions it, explains as a kind of dramatized form of degradation. Two spears were fixed upright in the ground, and a third was fastened horizontally by each end to the tops of them ; under this extemporized arch the conquered army had to pass, disarmed, and apparently wearing nothing but an under-garment, probably the subliga- culum, in later times the dress rather of slaves than of soldiers or citizens. Livy's language is explicit : ' ut exprimatur ' See Phillipson, International Law and Custom of Ancient Greece and Some, ii. 253 ff. ,LiT. ix. 3 shows the feeling against bloodshed. ' Liv. iii. 28 ; ix. 6 ; a. 36. Digitized by Microsoft® PASSING UNDER THE YOKE 71 confessio subactam domitamque esse gentem.' In historical times this was certainly the commonly received view, and it is still possible that it may be the true explanation. But there is real reason to believe that the passage under the yoke had originally another object and meaning. In the second edition of his Oolden Bough, iii. 406, Sir James Frazer, after collecting a number of practices in which human beings or '' animals are passed through a space limited in some way on either side, suggested in a footnote that the passage under the yoke might have been a rite of purification, ' destined to strip the foe of his malignant aaid hostile powers before dismissing him to his home '. In the third edition of his work, vii. 194, he took up this suggestion and worked it out more fully in his text, bringing to bear on the problem of the yoke, (1) the tigillum sororium, and (2) the porta triumphalis. It happened that I was doing the very same thing at the same time in the Classical Review for March 1913. What I wrote then I reproduce now, with full acknowledgement to my friend Frazer of his original suggestion in his second edition.^ On the steep slope of the Carinae, just above the hollow where the Coliseum now stands, there was a small street called the vicus Cuprius ; and where it was crossed by another {ad compitum Acili), a beam {tigillum) stretched from one side of the street to the other, called the tigillum sororium. Festus, p. 380 (Lindsay), tells us that this had originally rested upon two other beams, like the spear under which the captured army passed ; and that it was set up in this manner by the father of Horatius, who in the legend had murdered his sister, after his acquittal by the people, and after certain sacrifices to Janus Curiatius and Juno Sororia, whose altars remained there to the latest age of Roman history.^ But what was the ■ In March 1914 1 received from Professor Zachariae, of Halle a. d. Saale, an article written by him in the Berlin Zeitschrift des Vereinsfiir Volhakunde, entitled ' Das Kaudinische Joch '. He too had reached the same kind of conclusion, unknown to me or to Sir James Frazer, taking a hint, as I had done, from the second edition of the Golden Bough. ^ The best account of the position is in Hiilsen-Jordan, S6m. Topographie, iii. 322 ff., where also the texts relating to the tigillum are quoted in fuU Digitized by Microsoft® 72 PASSING UNDER THE YOKE object of this beam ? Livy, Festus, and Dionysius all agree that Horatius was made to pass under it ' velut sub iugum missus ', as an expiatio for the crime he had committed. His acquittal by the people was not enough ; he was not, by that method of procedure, ' liberatus omni noxia sceleris ' (Festus). Something had yet to be done in order to make him fit to mingle with his fellow citizens, and what was done was of a religious character. He is represented as having his head covered, which is a sure sign of this, even if we cannot trust the tradition about the two altars below the beam. He passed under the beam, as under the yoke, and was then clear of all scelus with the approval of the augurs. I have used the Latin word scelus, not wishing to commit myself to the notion that the thing thus got rid of was moral guilt ; had it been that, our modern minds would naturally suppose that it was wiped away by the acquittal. Far from it ; the sister's blood had been shed, and no lay court could possibly get rid of that stain. Horatius was undoubtedly sacer, i. e. taboo, in an infectious condition, dangerous to society. But in this peculiar form of passage under an arch or iugum, the religious authorities were able to apply a method of disinfection,^ which in the language of a later age is naturally spoken of as expiatio or purification. What the primitive Romans may have believed that it effected can only be guessed by one who studies carefully the evidence collected in the Oolden Bough? The second object in Rome which may be compared to the iugum was the porta triumphalis. A victorious army, before entering the city, had to pass under an arch or gateway (Liv. i. 26 ; Festus (p. 380, Lindsay) ; Dionys. iii. 22. 7). For the two altars below the beam see Auct. de viris Ulustribus, iil. 4. It is singular that the beam itself seems to have been an object of worship ; the Fasti Arvales have a note on October 1 : ' tigillo sororio ad compitum Aoili ', Henzen, Acta Fratr. Arv., p. ocxxxviii. It is easy to get out of one's depth in specula- ting on these worships and their connexion with the gens Horatia, as Pais does, Storia di Soma, i. 298 ff. My own opinion on these puzzling points is expressed in my Soman Festivals, pp. 237 fE. ' See Sel. Exp., p. 28. ' It will be found in Balder the Beautiful, ii. 169. Digitized by Microsoft® PASSING UNDER THE YOKE 73 which was called by this name. Its exact position we do not know for certain, but it is quite clear from a passage in Josephus {Bell. lud. vii. 5. 4) that it was in the Campus Martins, i.e. that it was an archway standing by itself outside \ the walls, and it is probable that it was close to the famous ara Martis, but separated from it by the stream called Petronia amnis, which had a religious importance of its own.^ We may suppose that in ancient times the army was lustrated in mmpo, near the ara Martis, and that it then crossed the sacred stream and passed under the porta triumphalis. What this porta was like we do not know ; neither description nor representation of it survives. I may perhaps conjecture that the oldest surviving triumphal arch, that of Augustus at Rimini, was not wholly unlike it. When triumphal arches were first introduced as memorials of some military achieve- ment, it is reasonable to suppose that they took the form of the old porta triumphalis familiar to Roman soldiers, and that a more elaborate style of ornamentation was only developed in course of time.^ If we look at the arch at Rimini, of which there is a cut handy in the Dictionary of Antiquities, s. v. arcus triumphalis, we shall notice that the most striking part of-it consists of two upright Corinthian pillars with an archi- -, trave laid across them ; within this is an arch proper built on the usual plan. I hope I am not too fanciful in suggesting that we may have here a reminiscence of the horizontal beam or spear resting on two upright ones. This was in fact the very oldest form of gateway, as one sees in the Lion gate ' See Hiilsen-Jordan, op. cit., iii. 494 and references ; also Domaszewski, in Abhandlungen zur rom. Religion, pp. 222 ff. For purification water running from a spring was necessary ; Festus (p. 296, Lindsay) ; Liv. i. 45-6. __ ' Nearly all forms of Roman art are now thought to be traceable to Ctreek originals, and without doubt the ornamentations of the triumphal arches which have survived are Hellenistic ; but such a thing as a triumphal arch is not known in Hellas, and I believe that the idea is far more likely to be Boman. According to Servius on Aen. xi. 6, the original object of the arch was to supply an elevated position for trophies, which certainly formed a part of the ornamentations on the arches on later date. The question is discussed in a sensible way, though withoup anthropological knowledge, in Courbaud's Le Bas-relief romain, p. 370 ff. 1 Digitized by Microsoft® 74 PASSING UNDER THE YOKE at Mycenae, and thus a course of development may be traced from very primitive beginnings, to the most elaborately decorated archways of the Roman Emperors. Now is there any common idea in the use of these three arches (or whatever we choose to call them), the iugum, the tigillum, and the porta triumphalis ? Let us compare the objects in the three cases, so far as they can be made out. The object is clearest in the case of Horatius. He had to be cleansed from something dangerous and infectious, and the consummation of this cleansing is signified by his passage under an object which formed a limit between the region of the ' sacrum and that of the profanum : he could not mingle with his fellow citizens if he were sacer, but the sacrifices and the passage under the tigillum effectually rid him of this burden. It was what has been well called a ' rite de passage ', and may safely be compared with those in which a human being passes through some divided object such as a cleft tree, or between two^osts, in order to get rid of some disease or other trouble.^ The passage through the porta triumphalis may well have had the same meaning. The army was guilty of bloodshed, like Horatius; and had been moving about in places where there were strange beings. Human and spiritual, with whom it would be unsafe to come in contact ; unsafe not only for the soldiers themselves, but for the citizens with whom they might mingle on their return. Even if we cannot exactly say that the returning host was sacer, we know that it needed lustration, and that it underwent this process immediately before passing through the porta ; which was exactly what happened to Horatius, according to the story .^ Lastly, let us return to the iugum of the three spears. I think we are now justified in assunaing that there was a religious ' See Van Gennep, Bites de passage, oh. ii, pp. 19 ff. ; Frazer, Oolden Bo'ugTi, ed. 3 (Balder the Beautiful), ii. 169 ff. It is absolutely necessary to read these pages carefully in order to be convinced of the meaning of rites of this kind in different parts of the world. ^ The same fear of the evil influence of an enemy is seen in the belief that spoils taken from him should be destroyed by fire, which has been elucidated by M. S» Reinach in his Cvltes, Mythes et Bdigions, iii. 233 ff. I have discussed this in The Death of Turnus, p. 155 (Aen. xii. 938 ff.). Digitized by Microsoft® PASSING UNDER THE YOKE 75 or magical element in this degrading ceremony. The many analogous examples collected in the Golden Bough strongly suggest it, and so also does the tigiUwn sororium, which plainly reminded the Romans, as we saw, of the iugum. And the passage through the 'porta triumphalis, though it is the passage of a victorious and not a surrendered army, had beyond doubt a religious or magical meaning, which should suggest an analogous one in the case of the iugum. Still there is a difficulty. What can have been the object of sub- jecting the captives to the same kind of ceremony as the murderer, or the victorious host ? Sir James Frazer thinks that it was to deprive them of their malignant and hostile powers before sending them home. I do not see that we can find a better explanation, though I might put it somewhat differently. They had to be brought out of one status into another ; they must not be any longer the same beings they were before the surrender ; just as in historical times the dediticius passed out of his former status into a new one, and became absorbed in the body politic of the conqueror, to be henceforward harmless. NOTE ON PRIVATELY DEDICATED ROMAN ALTARS I HAVE recently been asked a question which seems simple enough, yet is not after all very easy to answer. If a Roman in the provinces of the Empire wished to erect an altar, in fulfilment as a rule of a vow, how did he set about it ? How could he make it sacred, secure it against sacrilege ? Man does not erect altars to his gods unless he can protect them from harm ; how is this protection procured for the innumer- able ex votos of the Roman provinces ? It may be worth while to put together such evidence as I can collect on the subject of the dedication of altars whether at Rome or in the provinces ; in this way we may Digitized by Microsoft® 76 NOTE ON PRIVATELY be able to answer, more or less directly, the question raised above. Within the limits of the ager Romanus the transfer of an object to the gods, e. g. of a temple, altar, or sacellum of some kind, might be either public or private according as it was made over by the State under public law, or by an individual or a corporation privati iuris. If it were the act of the State, the process of dedication was completed by the consecratio of the pontifices, and the object became sacrum, i.e. it was taken out of the region of ordinary human usage, and became the property of the god to whom it was dedicated.^ As usual in such, solemn acts, the priest dictated the formula while the magistrate repeated it after him,^ and the result, if the object were a temple, was a lex, or written covenant destined to guarantee the sanctity of the site and building for all time. How far the Roman pontifical authority and law spread beyond the Roman territory into Italy we do not know for certain ; in any case this could only happen if the Italian communities adopted it. But municipia and coloniae, whether in Italy or the provinces, had their own religious authorities, pontifices and augurs,* and when the community wished to erect a temple or altar, these no doubt provided the necessary formulae. Did they also assist the private individual or corporation in a dedica- tion, and thereby give it sanctity and security ? I find no ' Aelius Gallus ap. Festum (p. 424, Lindsay) ' Gallus Aelius ait sacrum esse, quocunque modo atque instituto civitatis consecratum sit, sive aedis, sive ara, sive signum, sive locus, sive pecunia, sive quid aliud quod dis dedicatum atque consecratum sit.' Gaius, ii. 4 ' sacrae sunt (res) quae dis superis consecratae sunt : religiosae quae Dis Manibus relictae sunt ' : sec. 5, though mutilated, seems to state that in order to be sacrum an object must be dedicated ' auctoritate populi Bomani '. Cf. Liv. ix. 46. 7. ^ Wiasowa, S. K., pp. 394 ff. ; Marquardt, pp. 145 ff., 270 ff. ^ The plainest evidence is that of the Lex Coloniae Genetivae (C. /. L. ii. 5439, sec. 62; Dessau, Inscr. Lat. Sdectae, ii. 1, p. 502). Toutain, Cultes paiens dans I'Emp. rem. i. 277 ff., tells us that pontifices and augurs are not found in inscriptions of the three Gauls, Britain, Germany, and Rhaetia ; but in those provinces Roman towns were comparatively few. Where municipia and coloniae are in plenty, there is also plenty of evidence of the existence of these priesthoods. Digitized by Microsoft® DEDICATED ROMAN ALTARS 77 proof of this in the provinces, but we may here admit the evidence of certain inscriptions at Ostia, which was an ancient colony and had its own priests.^ Among these priests the chief was the pontifex Volcani et aedium sacrarum, who was in charge of all the temples of the colony, and whose permission seems to have been necessary before statues could be erected in sacred precincts or gifts of importance dedicated in sanc- tuaries.^ It is not unlikely that this priest superintended the process of other pr'ivate dedications. But the legal result of these dedications, even when sanc- tioned by the local priesthood, was not as complete as it was within Roman territbry. The Roman lawyers assure us that the objects dedicated could not become sacra in the proper sense of that word, but were held to be religiosa only, like burial-places and spots struck by lightning ; ' that is to say, they were not made over to a deity as his property, but were objects of reverence and awe without the legal protection gained by consecratio. The word sacrum is of course found in votive inscriptions all over the Empire, but when so used it is not a term of Roman religious law. ' See Ruggiero, Diz. Epigr., i, s.v. aedes and ara, who says that private dedications did not need the help 6t a priest, but gives evidence of an occasional instance of such help : e. g. C. I. L. vi. 746 ' ara posita asstante sacerdote Secreusina Secundo ut voverant ' ; p. 412 ' aram posuit per C. Fab. Germanum sacerd.' Cf. De Marchi, La Rdigione neUa vita domestica, i. 290. In the Boman ius divinum the pontifioes and augurs seem to have been called in to assist private individuals or families in religious details : Cicero in his imaginary code {Legg. ii. 8. 20) saj^ ' quoque haeo privatim et publice modo rituque fiant, discunto ignari » publicis sacerdotibus '. Commenting on this in sec. 47, he reckons vota among the matters of private religion over which the pontifices may have superintendence, together with feriae, sepidcra, et si quid eiusmodi est. Cf. what Cic. says of the great P.M. Coruncanius and of P. Crassus in de Oral. iii. 134 ' ut ad eos de omnibus divinis et humanis rebus referretur.' Eel. Exp., p. 281. From this we may infer that the same held good of the priests in rrmnicipia and coloniae in Italy and the provinces. ^ For Ostia see Miss Lily Taylor's monograph on the Oidts of Ostia: Bryn Mawr monographs, xi. 15 ff. ' Festus, /. c. ; Marcianus, Dig. i. 8. 6, sec. 3 ' si quis privatim sibi sacrum constituerit, sacrum non est sed profanum.' Gains, ii. 7, seems to imply that in this case the object was not even technically religiosum, but was considered to be pro rdigioso. Digitized by Microsoft® 78 NOTE ON PRIVATELY Outside Roman towns, or within them on ground owned by an individual {in privato solo ^), it was common enough for the private person to set up altars, temporary or permanent, or to dedicate any other object,^ on land to which he had some kind of a legal title. When Cicero wished to erect a shrine (fanum) to his daughter TuUia in a public place, he had to purchase land to suit his purpose.' There is a curious case from older Roman history, of the erection of an altar within a sacellum in a private house, in order to be the rival of an older cult of a public character.* In the fifth Eclogue of Virgil we find shepherds erecting altars, apparently meant to be permanent, to Apollo and Daphnis.^ There is another example in a poem of Martial ; the vilicus of an estate has erected altars to Jupiter and Silvanus, which the owner entrusts to a friend during his absence, with a request that he will perform all the necessary rites.* Such cases as this last illustrate the usual method of instituting private cults in Italy and the provinces. Dedications by corporate associations, of a private char- acter, of which there are very large numbers, stood on the same legal footing as those made by private individuals ; the objects dedicated did not become technically sacra.'' ' It will be remembered that Augustus built his great temple of Apollo on the Palatine ' in solo privato ' ; the political meaning of this is well stated by J. B. Carter, Religion of Numa, p. 166. ■= Great numbers of these were merely votive tablets or inscriptions, but many were aedes or aedicidae, as well as arae. ' See my Bdigious Experience of the Romans, p. 385, and the letters quoted in notes 23 and 24. ' Liv. X. 23. The older cult was that of Pudioitia patricia in the Forum boariwm ; the new one was to Pudicitia plebeia : ' in vico Longo, ubl habita- bat, ex parte aedium quod satis esset loci modioo sacello exclusit aramque ibi posuit.' ° Eel. V. 65 ff. Note the expression solennia vota. " Mart. X. 92 : Commendo pinus iliceaque Faunorum et semidocta vilici manu structas Tonantis aras horridique Silvani. ' Waltzing, Les Corporations professionndles chez les Romains, ii. 434 ff. ; iv. ' 457 f^ ; De Marohi, La Religione gentilizia (II Cvlto privato, pt. ii), pp. 134 ff. Digitized by Microsoft® DEDICATED ROMAN ALTARS 79 They were protected from abuse simply by the fact that they were Roman, if they were in the provinces ; if in Italy, by their religious character. We may perhaps assume that though they were not the property of the deity iure' divino, they were so regarded by the vulgar, and were truly religiosa, i. e. objects of reverence and awe. This was certainly the case with all those that were dedicated by soldiers under the licence, and when possible the supervision, of the Caesar or his representative.^ Thus we may roughly say that there was little or no question of the mutilation or destruction of these objects throughout the Empire, since they were always Roman, and very often Roman in a military sense. THE PONTIFICES AND THE FERIAE: THE LAW OF REST-DAYS The subject of Roman rest-days is an interesting one, for it helps us to understand not only the religious feeling of the Romans in Regard to times and seasons, and the policy of the pontifices in dealing with that feeling, but incidentally throws sidelights on the life and work of the Roman farmer. It has been touched upon by Bouche-Leclercq in his book Les Pontifes, pp. 119ff., and recently by Dr. Hutton Webster in his Best-days, pp. 94 fE. and 121 ff. ; but there is room for further discussion of a subject still obscure in some points. We all know the lines in the first Georgic (268 fE.) in which Virgil touches on the rule that feriae were rest-days, and on some of the exceptions to that rule which were sanctioned by the -pontifices. Quippe etiam festis quaedam exercere diebus fas et iura sinunt ; rivos deducere nulla religio vetuit, segeti praetendere saepem, insidias avibus moliri, incendere vepres, balantumque gregem fluvio mersare salubri. saepe oleo tardi costas agitator aselli vilibus aut onerat pomis, lapidemque revertens incusum aut atrae massam picis urbe reportat. ' See Domaszewski, Bdigion des romischen Heeres, p. 110. Digitized by Microsoft® 80 THE PONTIFICES AND THE FERIAE : I must make one note on this passage before going on. By the words ' fas et iura ' Virgil undoubtedly means the action of the pontifices in making or relaxing rules. It has been usual, from Servius downwards, to explain the two words as meaning religious and civil law respectively ; ' divina humanaque iura ', as Servius puts it. But it is certain that the pontifices alone had cognisance of these rules, and their ius pontificium was a part of the ius divinum which governed all religious matters. Virgil, who is not using technical language, adopts the term fas, then just coming to be used in this sense, to mean this pontifical law, and adds iura to make it more intelligible to his readers. ^ ' Dies festi ', as Virgil calls them, or more strictly Feriae,^ were days made over to the gods, just as templa were the definite spots of ground in which those gods had consented to take up their abode. For this reason both days and places were under the control of the pontifices, not of the civil magistrates ; and the outcome of this priestly control was known as ius divinum ('fas et iura ', as Virgil calls it), or as ius pontificium in reference to its regulators. I am not here specially concerned with the primitive history and origins of rest-days, for which I can now refer the reader to Professor Webster's book just mentioned, and to his own and other articles on Sabbath (primitive) in Hastings's Encyclopaedia of Religion and Ethics. When Roman history begins, the feriae and the pontifices are already there ; and any attempt to trace them farther back must, in the present state of our knowledge, be pure speculation. We may, however, be fairly sure that whatever be the primitive origin of the institution of fixed rest-days, the idea in the minds of the early Romans was that these days had become the property of deities, and that man honoured these deities by abstaining from work on ' On this meaning of fas, see Appendix IV to my Rdigiaus Experience of the Romans. " Dies festi is the wider and less technical expression, and might include nundinae (markets) and Ivdi. Feriae in historical times was limited to days dedicate^ to a deity (Serv. Oeorg. i. 268 ; Wissowa, Rd. und Kult., ed. 2, pp. 432 n. and 441 n.). Digitized by Microsoft® THE LAW OF REST-DAYS 81 those days, just as he honoured them with sacrifices in their temples. We may indeed find an indication here and there that there were much older ideas at the root of the rules as the later Romans knew them, of which a word will be said later on ; at present I am dealing with the subject as it appeared to the Roman farmer in the two centuries from Cato to Columella. Our earliest information about what might be done on feriae comes from Cato {de Agric. ii. 4, Keil). The owner of the estate is supposed to be telling his bailiff what he might or ought to have done, during his master's absence, better and more carefully than he has done. There were many things, for example, which he might have done in bad weather ; and then he goes on ' per ferias potuisse fossas veteres tergeri, viam publicam muniri, vepres recidi, hortum fodiri, pratum purgari, virgas vinciri, spinas runcari, expinsi far, munditias fieri '. This is characteristic of Cato, and doubtless of the Roman farmer generally in that age of money-making. No time was to be lost on the farm ; the slaves were not to be idle even on feriae, for there was plenty that they might do without transgressing the divine law. The examples of such work are interesting ; most of them are concerned with cleaning out and cleaning up, as distinct ifrom starting new work. One or two are interesting for other reasons. The repairing of a public road seems to imply hard work, but if it was work made necessary by a proximity of the estate to a via publica, repairable by the neighbouring owners,^ the bailiff should have been glad to get it done on days when important work on the farm was not possible. Expinsi far refers, I think, to the preparation of salt-cake for the daily domestic sacrifice ; if so, it was natural enough that the pounding should be done on ' This method of repairing roads is alluded to by Sioulus Flaccus, de conditionibus agrorum (Gromatici, p. 146) : ' Vicinales viae de publicis quae devertuntur in agros, et saepe ipsae ad alteras publioas perveniunt, aliter muniuntur, per pages, id est per magistros pagorum, qui operas (slaves) a possessoribus ad eas tuendas exigere soliti sunt. Aut, ut comperi- mus, uni euique possessori per singulos agros certa spatia adsignantur, quae suis impensis tueantur.' 22S2 ^ Digitized by Microsoft® 82 THE PONTIFICES AND THE PERIAE : days devoted to the gods. Far, in Cato's time, would hardly be used for ordinary bread ; in religious rites it survived, and was pounded in a mortar instead of being ground in a mill (Roman Festivals, p. 149). -One other rule we learn from Cato which may be mentioned here ; it occurs in oh. 138. Oxen might be yoked on jeriae for draught purposes, but mules, horses, and asses have no feriae ' nisi si in familia sunt '. What these last words mean I am not sure ; Professor Webster translates (p. 95), ' For mules, horses and asses there are no other holidays than those of the family '. The words would more naturally mean that the animals had no holidays unless they belonged to the familia in some special sense, e.g. lived in the homestead. Cato is probably using familia in its true sense, of the economic unit settled on the land. Of this unit the animals in question were certainly members ; but what animals were those which were not in the familia ? Perhaps an explanation may be found in the distinction drawn by Varro {B. R. ii. 6. 4) between the villatica pastio and the pastio agrestis ; the latter being far away, while the former was at or near the homestead. We hear nothing more of the law of feriae until we come to the great Pont. Max. Mucius Scaevola, of whom Cicero has much to tell us in the Brutus ; a man of fine character and liberal ideas of law. Macrobius tells us (i. 16. 10-12) that this man, when asked ' quid feriis agi liceret ', answered succinctly ' quod praetermissum noceret '. Feriae were made for man, not man for feriae ; if an ox fell into a drain or reservoir (specus), no one need scruple to pull it out on a dies festus. But at the same time he insisted that the old rule of sacred law still held good, that if a man knowingly broke the peace of the feriae, no piaculum could absolve him ; he became impius. Another Pont. Max., called simply Umbro by Macrobius, laid it down that no man was polluted who did any work ' ad deos pertinens or sacrorum causa ' (such perhaps as the preparation of the salt-cake), or anything 'ad urgentem vitae utilitatem respiciens ' — an interpretation even wider than that of Scaevola. It is pretty clear that by Digitized by Microsoft® THE LAW OF REST-DAYS 83 this time the sting had been taken out of the old rigid rules ; the ancient scruples and superstitions were losing strength. Half a century later it is interesting to find that Cicero, when drawing up his imaginary code of religious law (which, by the way, he does not call fas), gives the law of jeriae for the first time a quasi-moral meaning. ' Let contentions of every kind cease among the free members of a community, and let slaves enjoy these days as holidays ; for this purpose they were ordained at special times.' ^ The policy of Augustus in matters of religion was to restore as far as possible the old feeling of duty to the gods ; and it may be that there was a revival of interest in the rules of pontifical law, which is perhaps shown in Virgil's lines. It is remarkable that far the larger part of our knowledge both of rules and exceptions comes from the agricultural writer Columella, who, though a Spaniard by birth, was a practical farmer in Italy in the generation next after Augustus, and wrote his book in old age under Nero.^ I will now examine what he and Virgil have to tell us about these rules. ' Pontifices negant segetem feriis saepiri debere ', says Columella ; but here Virgil difiers. On this Keightley remarked, and probably with justice, that the poet must have been thinking of the mending of a fence, while the pontifices meant that you must not begin a new one. For Servius (Interpol) commenting on G. i. 272, says that any work was lawful which was not a beginning, ' quicquid fieri sine institu- tione no vi operis potest ' ? This is an interesting point, and may help to explain other rules. A moment's thought will show how important in a religious sense the Romans believed the beginning of any serious undertaking to be : the beginning of a literary work, an oration, a war, the foundation of a city, are among the examples that instantly occur to us ; and the > Cio. de Legibus, ii. 8. 18, and 12. 29 ; de Div. i. 45. 102. Cf. Webster, Rest-days, p. 97. ^ Sohanz, Oesch. der Edm. Lit. (ed. 3), pt. ii. 2, p. 501. The passage of Columella is bk. ii. 22. ' Verrius Flaccus (ap. Macr. i. 15. 21) cynically observed that on this principle feriae were better suited for marriages of widows than of, maids ! F 2 Digitized by Microsoft® 84 THE PONTIFICES AND THE FEEIAE : intense desire for good omens at the- beginning of a journey or a battle, points in the same direction. The association of Janus with beginnings, though it may not have been primi- tive, will also help us here. We all know the emotion we experience in taking a decisive step — in crossing a Rubicon, as Dr. Marett has lately put it ; when once the work is begun, the resumption of it never again suggests quite the same feeling. In the second chapter of Dr. Webster's Rest-days, on ' tabooed days at critical periods ', plenty of evidence will be found of the ' religio ' attaching to newness or beginnings in many parts of the world at the present day.^ Especially, as we might expect, it is the important matters of food, war, and so on, most deeply affecting the life of the community, which have taboos attached to their beginnings ; and these have descended to civilized peoples, in many cases, in the form of religious rules of law. I cannot pursue the subject here ; but it will not fail to interest a student of human nature, seeing that it has its root in truth and fact. Virgil says that sheep may be washed on a dies festus : ' balantumque gregem fluvio mersare salubri.' Here again he has to be explained or corrected, but both Columella and Macro bius are most anxious to save his credit. ' Maro omnium disciplinarum peritus ', says Macro bius, ' sciens lavari ovem aut lanae purgandae aut scabiae curandae gratia, pronuntiavit tunc ovem per ferias licere mersari, si hoc remedii causa fieret.' Oolumella lays special stress on the word salubri, and is followed by Servius : they take it as indicating that the poet thought of the washing as done for medical reasons only, a remedy mentioned in G. iii. 441. However this may be, the rule emerges that the important operation of sheep-washing, occurring regularly at a certain time of year, may not be done on a dies festus ; but a casual remedy for scab may be applied at any time. Perhaps the association of all medical remedies with religion in early Rome and elsewhere had something to do with this exception ; and we must also remember that no time was to be lost in ' See especially pp. 14-17, 32, and 34 fE. Digitized by Microsoft® THE LAW OF REST-DAYS 85 applying the remedy, which would come under Scaevola's rule of ' quod praetermistom noceiret '. Virgil says ' Rivos dedueere nulla Religio vetuit '. Rivos deducere is explained by Servius as siccare ; not to be taken as irrigare, which is expressed in this first Georgic by inducere, as in 106. What is meant by deducere is the drawing the water off the fields which have already been irrigated. Servius goes on to quote Varro as saying that it was ' contra religionem ' either to irrigate land or wash animals on feriae, because water-spirits (Nymphae ^) ' sine piaculo non possunt moveri '. Varro is here reflecting a real religio about the use of fresh water, of which it unluckily happens that we have but few traces in Latium .^ The distinction is between the pure fresh water, full of life and healing, which was let in on the fields from brook or river, and the water which had become stagnant after doing its work there. We may remember that only fresh running water was of avail for religious purposes, and that the Vestals fetched their supply daily from Juturna's spring. I take it that the used-up water had been abandoned by the spirits before it was drawn off ; they did not inhabit a farmyard pond, but, like trout, rejoiced in the living water of a running stream. Columella tells us that you might on feriae cultivate a hired vineyard or olive yard, one, that is, which you undertake for its owner at a certain price. This exception has been explained as being necessary in order to enable the lessee to fulfil engagements within a specified time.^ Or it may be that as Roman law originally recognized no other form of possession but absolute ownership,* other forms which came in later were more easily made subjfect to exemptions of this kind. There might be a difference of feeling about land not your own, which would therefore not be a part of the economic ' Varro used the word Nymphae because the Latin water-spirits were nameless (Wissowa, if. K., p. 223). ' Aen. viii. 68, and a note in my Death of Turnus, p. 55 £f. C£. Varro, E. B. i. 11. ' Bouche-Lecleroq, Pontiles, p. 119. * Gains, ii. 40. Digitized by Microsoft® 88 THE PONTIFICES AND THE EERIAE ; (see my Roman Festivals, pp. 90 ff.).^ Perhaps the solution lies in the number of puppies in a litter that were not kept for training ; you selected the best, Varro tells us,^ when they were quite young, and destroyed the rest. But it might be (Convenient to keep some for purposes of insurance ! Let us now consider these pontifical rulings as a whole. Professor Webster, in a footnote on p. 98, writes : 'the ponti- fical law in such matters was as minute, tyrannical, and absurd as the rabbinical ordinances relating to the proper observance of the Sabbath '. Here I cannot at all agree with him. The rabbinical ordinances were a much later develop- ment of law than the pontifical, relatively to the history of society in Judaea and Rome respectively ; they were the result of a durious passion for minutiae of tiresome and mean- ingless casuistry. ' The brief prohibition of work on the Sabbath which is found in the Pentateuch, and which hardly at all enters into detail, was in the course of time developed in so many-sided a manner as to form of itself an important branch of knowledge. For of course the Rabbis could not rest satisfied with simple prohibition. They must also accurately define what work was forbidden . And consequently they at last, with much ingenuity, found that on the whole thirty-nine kinds of work were prohibited, of which very few are anywhere alluded to in the older law.' ^ The thirty- nine prohibited works include the chief operations of algricul- ture, but descend to such things as making or untying a knot, sewing two stitches, catching a deer, and killing, skinning, and salting it, writing two letters, putting out a fire, lighting a fire, and so on. And each of those rules needed further discussion of their range and meaning ; e. g. to what kind of knots the rule applied ; and it was decided that the knot of camel-drivers and that of sailors was forbidden, both for tying and ilntying. What depths of absurdity can be reached by those who once launch on this kind of inquiry, may be ' Pliny, xxix. 58, notices the use of puppies ' plaeandis numlnibus hostiarum vice ', as if it were a well-known practice. " R. R. ii. 9. Schiirer, II. ii. 96. Digitized by Microsoft® THE LAW OF REST-DAYS 89 seen in Schiirer's chapter on ' Life under the Law ' in his Jewish People in the time of Christ, II. ii. 97 ff. I can see little of this kind of absurdity in the rules we have been examin- ing. Something of the same love of minutiae is seen in the lists of the Indigitamenta ; ^ but in the practical matters of farm work we expect and we find more good sense. We may assume, I think, that in the earliest times there were no exceptions to the ancient taboos on days made over to deities ; no work at all could be done on those days, of which in rural life there were probably not too many. This seems to me to be suggested by the survival of primitive ideas about the spirits of earth and water, and the strict maintenance of the associated taboos even in the time of Columella. I should imagine that the same rigidity attached to the rules even after the city calendar was drawn up in the age of the later kings. Blit when we come down to the fourth and third centuries b. c, which is in other ways the age of special pontifical activity, we find that though these rules are fully kept up in theory, the growing difficulties of the farmer are being gradually resolved by pontifical decisions of which Cato a century later gives us some specimens. What- ever the share of the pontifices may originally have been in fixing the rules, their work was now to discover the means of relaxing them — a process in which I seem to see, from the examples examined above, a considerable amount of common sense. This relief was called for, we may conjecture, mainly by the exigencies of more extensive and more scientific farming. The rules which could do no harm while the peasa^nt had only his two iugera of heredium and his rights of common of pasture, "were likely enough to hamper his activity when he farmed 200 acres or more at a distance from Rome, or ran large herds of cattle on the saltus of the hilly country. The process of enlarging farms was exactly contemporaneous with the period when pontifical activity was at its height, when plebeians had a share in land newly acquired from » Sel. Exp., pp. 164, 286. Digitized by Microsoft® 88 THE PONTIFICES AND THE PERIAE ; (see my Roman Festivals, pp. 90 ff.).-"^ Perhaps the solution lies in the number of puppies in a litter that were not kept for training ; you selected the best, Varro tells us,^ when they were quite young, and destroyed the rest. But it might be Convenient to keep some for purposes of insurance ! Let us now consider these pontifical rulings as a whole. Professor Webster, in a footnote on p. 98, writes : 'the ponti- fical law in such matters was as minute, tjrrannical, and absurd as the rabbinical ordinances relating to the proper observance of the Sabbath '. Here I cannot at all agree with him. The rabbinical ordinances were a much later develop- ment of law than the pontifical, relatively to the history of society in Judaea and Rome respectively ; they were the result of a durious passion for minutiae of tiresome and mean- ingless casuistry. ' The brief prohibition of work on the Sabbath which is found in the Pentateuch, and which hardly at all enters into detail, was in the course of time developed in so many-sided a manner as to form of itself an important branch of knowledge. For of course the Rabbis could not rest satisfied with simple prohibition. They must also accurately define what work was forbidden . And consequently they at last, with much ingenuity, found that on the whole thirty-nine kinds of work were prohibited, of which very few are anywhere alluded to in the older law.' ^ The thirty- nine prohibited works include the chief operations of algricul- ture, but descend to such things as making or untying a knot, sewing two stitches, catching a deer, and killing, skinning, and salting it, writing two letters, putting out a fire, lighting a fire, and so on. And each of those rules needed further discussion of their range and meaning ; e. g. to what kind of knots the rule applied ; and it was decided that the knot of camel-drivers and that of sailors was forbidden, both for tying and Untying. What depths of absurdity can be reached by those who once launch on this kind of inquiry, may be ' Pliny, xxix. 58, notices the use of puppies ' placandis numinibus hostiarum vice ', as if it were a well-known practice. ^ n. ie.*ii. a Schiirer, II. ii. 96. Digitized by Microsoft® THE LAW OF REST-DAYS 89 seen in Schiirer's chapter on ' Life under the Law ' in his Jewish People in the time of Christ, II. ii. 97 ff. I can see little of this kind of absurdity in the rules we have been examin- ing. Something of the same love of minutiae is seen in the lists of the Indigitamenta ; ^ but in the practical matters of farm work we expect and we find more good sense. We may assume, I think, that in the earliest times there were no exceptions to the ancient taboos on days made over to deities ; no work at all could be done on those days, of which in rural life there were probably not too many. This seems to me to be suggested by the survival of primitive ideas about the spirits of earth and water, and the strict maintenance of the associated taboos even in the time of Columella. I should imagine that the same rigidity attached to the rules even after the city calendar was drawn up in the age of the later kings. But when we come down to the fourth and third centuries B. c, which is in other ways the age of special pontifical activity, we find that though these rules are fully kept up in theory, the growing difficulties of the farmer are being gradually resolved by pontifical decisions of which Cato a century later gives us some specimens. What- ever the share of the pontifices may originally have been in fixing the rules, their work was now to discover the means of relaxing them — a process in which I seem to see, from the examples examined above, a considerable amount of common sense. This relief was called for, we may conjecture, mainly by the exigencies of more extensive and more scientific farming. The rules which could do no harm while the peasa^nt had only his two iugera of heredium and his rights of common of pasture, Vere likely enough to hamper his activity when he farmed 200 acres or more at a distance from Rome, or ran large herds of cattle on the saltus of the hilly country. The process of enlarging farms was exactly contemporaneous with the period when pontifical activity was at its height, when plebeians had a share in land newly acquired from » Bd. Exp., pp. 164, 286. Digitized by Microsoft® 90 THE PONTIFICES AND THE FERIAE conquered enemies, and when even the Pontifex Maximus might be a plebeian. The chief result of these decisions was that many minor operations came to be ruled lawful on feriae, especially those which could be carried on at the homestead, in order to gain more time and labour for those of greater importance, such as sowing, ploughing, and harvest. Rules about beginning a new work, or such as might disturb the ^ ^ spirits of land and water, were retained, but subsequent work of the same kind, which involved no disturbance to the spirits nor scruple to the farmer, was allowed. Whether an increase in the number of the feriae was also a contributory cause is a difficult question. The number in any case is hard to fix for a farm, for we do not know the exact relatian between city calendars and rural ones. For early Latium, however, we may reckon sixty-one feriae of city life, with two or three more that were conceptivae (movable) and do not appear in the oldest city calendar, and further an unknown number of feriae privatae, i.e. those of the gens or family, such as dies natales and feriae denicales, all of which were under pontifical rule.^ The whole number is not likely to have been less than 70 or 80 out of a total of 355 days, not counting nundinae, or the ludi of the city which became so numerous in later times. For real honest work on a big farm in the fourth and third century e.g. this number of holidays may well have been a serious inconvenience. Hence it was that not only light work and secondary work came to be allowed on feriae, but even more serious operations if piacula had been offered beforehand as an insurance against divine wrath. All this seems to me to show good sense. By the time of Scaevola and Cicero a more liberal and generous interpretation of the old rules was general and acknowledged ; and though there may have been a revival of strictness under Augustus, as was suggested above, the ancient taboos gradually gave way everywhere. ' Necessitas feriis caret ', said Palladius in the fourth century of the Empire (i. 6). " WisBowa, Sel. und Kult., p. 433 ; Macrob. I. xvi. 7. Digitized by Microsoft® 91 ON THE DATE OF THE RHETORICA AD HERENNIUM {Rhet. ad Herenn. iv. 54. 68) In 1880, at the suggestion of the late Professor Henry Nettleship, I very carefully examined a now well-known passage near the end of the fourth book of the Ehetorica ad Herennium, with the object of determining whether it could be referred to Marius, instead of to Sulla, to whom scholars had till then been in the habit of referring it. On November 19 of that year I read a paper on the subject to the Oxford Philological Society, giving reasons for my belief that the person referred to is Marius, and that the date of the work may be 83 or 82 b. c. The paper appeared in the Journal of Philology for 1882, pp. 197 ff. ; it was the first of its kind I ever published, and I remember that I took all possible pains with it. It has occasionally been referred to in German works, e. g. in the second edition of Teuffel's Roman Literature, and also in the second edition of Schanz's great work on the same subject. But there is no mention of it in the very elaborate edition of the Rhetorica by F. Marx (1894). I should not be disposed to go back on it now, but that Marx, though he too, like most scholars, is now of opinion that the passage refers to Marius, has found himself compelled, by the rules of textual criticism which have been suggested to him by his investigation of the MSS., to 'restore' the passage to. very much the same lamentable condition in which it formerly stood. It occurs in iv. 54. 68, and is an example of brevitas in oratory — a point to be noted carefully. It is preceded by another example, to which I will return later on. In Kayser's separate edition of the Rhetorica (1854), which was not improved on by Baiter and Kayser in their joint edition (1860), it runs thus : Digitized by Microsoft® 92 ON THE DATE OP THE Item : modo consul quodam is deinde primus erat civitatis. Turn proficiscitur in Asiam, deinde hostis est dictus, post imperator et populi Eomani consul factus est. Here the two most doubtful points are : (1) the obviously corrupt quodam is, and" (2) the words populi Bomani, which in an example of brevitas are as obviously absurd, and are the emendation of the early editors for populorum, the reading of the best group of MSS.^ I proposed to read the passage as follows : Item : modo consul quotannis deinde primus erat civitatis. Tum proficiscitur in Asiam, deinde hostis est dictus, post imperator populorum et consul factus est. The correction quotannis (or quodannis) appeared on the authority of H. Jordan in Hermes, May 1881, midway between the reading of my paper and its publication. The reference to the six successive consulships of Marius seemed thus fairly well established, and since then there has been a general consensus, so far as I know, that Marius and not Sulla is the person referred to. And the brevitas of the example was undeniable.^ Marx however, though he agrees about Marius, spoils the brevitas, as it seems to me, as well as the neaJtness of the Latin. In his text the words appear thus ; Item'- Modo consul quondam, is deinde primus erat ciuitatis ; tum proficiscitur in Asiam, deinde hostis est dictus, post imperator et postremo factus est consul. Will any one maintain that ' Modo consul quondam ' is a natural way of alluding to six successive consulships of Marius only some twenty years before the Rhetorica was written ? or that ' is ' has any meaning in an example of ' See the edition of Marx, p. 373, and Prolegomena, p. 154. ^ In a letter written to me at the time by the late Professor Robinson Ellis, which I have fortunately preserved, he says of my version : ' As an example of breuitas the sentence now becomes excellent : antithetical, and with just the amount of deviation from ordinary expressions which would strike an attentive reader ... If quodannis is right, then I think, as I suggested at the meeting, there can be no doubt that modo, i. e. " just now ", " but recently '" is in antithesis to deinde.' Digitized by Microsoft® RHETORIGA AD HERENNIUM 93 breuitas ? or that ' postremo ' is not superfluous for the same reason, especially as Marius was elected to his seventh consul- ship only just after Cinna had illegally made him imperator ? But Marx's canon of textual criticism compels him, it seems, to spoil the passage thus. Let me explain. The text of the Rhetorics depends on two groups of MSS., each of which may be taken to represent a lost archetype : (1) an older group (ninth and tenth centuries) in which there are lacunae ; and (2) a younger group (c. twelfth century) constituting a complete text. If pp. 41-5 of Marx's Prolego- mena are studied carefully, it will appear that the reading of the older group M must under certain circumstances give way to those of the younger group E, especially where the former is corrupt. Now in our example M has ' Modo consul quodam is ', and this is retained'by Marx, except that he writes quondam for quodam,^ inserting the additional letter on the evidence of his experience of M. The younger group E has ' quondam tribunus ', and this is also the reading of C, one of *he MSS. of group M. This last fact makes a difficulty, and causes Marx to retain the reading of M as against that of E (see Proleg., p. 43), which is almost certainly wrong, for to refer to the trib'uneship of Marius in such a passage as this would be surprising as well as superfluous. But in the last words of the example E is preferred for reasons which I confess I do not quite understand ; and ' postremo ' is read from E instead of ' populorum ' from M. ' Postremo ' may seem natural until we remember that we are dealing with an example of breuitas, ' res tantummodo verbis necessariis expedita,' as it is defined in this same chapter ; but when we bear that in mind it becomes superfluous, and is indeed almost absurd, as I will show directly. For populorum, on the other hand, I da'n find a very good sense, appropriate to the time, and to the facts of the life of Marius. As it is now generally accepted that Marius and not Sulla ' This is in spite of the fact that on his own showing the word qiumdam is nowhere found in the whole four books of the work. See Marx, index, and Prolegomena, p. 154. Digitized by Microsoft® 94 ON THE DATE OP THE is the person whose public life is here briefly expressed, I need not reproduce in full the arguments I used in the Journal of Philology to prove this point. But as the last words of the example do not seem to me to have been correctly understood, I must advert to them for a moment ; I mean the words which in M run thus : ' pos't imperator et populorum consul factus est ', or as Marx would read them, ' post imperator et postremo factus est consul '. First, the word ' imperator ', if Sulla be mearit, is meaning- less, or at least needs a justification which it has not yet found. It is true that Sulla was technically imperator from the day on which he left Rome in 87, to the day on which he resigned his Dictatorship in 79. But the word, if here used of Sulla, must be used in some such special and extended sense as was afterwards given'it by Augustus ; and we have no evidence whatever that in this point Sulla anticipated the Empire. In the inscription on the equestrian statue, which (according to Appian, B. C. i. 97) ran KopvrjXio'v 'SvXXas rjye- fMovoi evTvxovs, we must understand ^ye/^oVos as meaning Dictatoris and not Imperatoris, since the latter word would have been rendered avTOKparoip by a Greek writer.^ Marius, on the other hand, was furnished by the consul Cinna with the proconsulare imperium and the fasces, doubtless in order to give him, technically hostis as he still was, a definite position in the eyes of his soldiers ; illegally no doubt, and doubly so, as Cinna himself was not legally consul. But in the eyes of the Italians, who were to form the bulk at least of his army, and in the eyes of the Auctor of this work, whose leaning to the side of the populates is universally recognized, the title might well be accepted de facto. That it was conferred is stated by Plutarch in a passage too explicit to be the result of misapprehension, and probably derived from Posidonius, who was in Italy a few months later, and had an interview with Marius on his death-bed.^ Secondly, the mysterious word ' populorum ' can have no • See Mommsen in 0. I. L. i. 168. ' Plut. Marius, 41 ; of. 45. H. Peter, Qudlen Plutarch's, p. 103. Digitized by Microsoft® RHETORIGA AD HERENNIUM 95 meaning if Sulla be the person spoken of ; but it fits in well enough with the position of Marius in the middle of 87, after he had been made imperator by Cinna. The army he got together, like those of Cinna, Carbo, and Sertorius, at this time, was composed of the Italian populi still in arms,^ together with numbers of the new Italian cives who were discontented with the inferior position assigned them by the Senatorial government in a limited number of tribes. This campaign in fact, though commonly called the first Civil War, was only a new phase of the Social or Marsic War ; the new feature being that one party at Rome was now heading the Italians against the other. If we had the eighth and ninth decades of Livy, we should no doubt find, as we may guess from his Epitomist, that throughout the war the Italians were called ' populi Italici ', or ' populi ' only ; ^ and in an example of breuitas, framed by a person writing soon after the war, it would be perfectly natural to term a general at the head of an Italian army ' imperator populorum '. So I propose to read here ' Post imperator populorum et consul f actus est '.^ Marius became consul for the seventh time on January 1, 86,- having been elected (or appointed by Cinna) but a short time after he had been made imperator ; thus the word ' postremo ', which Marx insists on reading, in spite of the unanimity of the M group of MSS., for populorum, is certainly superfluous, even to absurdity, in an example of breuitas. Let me now turn to the example of breuitas which immedi- ately preceded the one we have been so far discussing. The ' Liv. Spit. 80 (Samnium) ; Appian, B. C. i. 67 (Etruria). Cf. Kiene, Bundesgenossenkrieg, p. 298 ; Mommsen, £. H. (Eng. trans, iii. 317 ff.). ^ Liv. Epit. 72 ' Italici populi defeoerunt '- 73 ' Complurea populi ad hostes defeoerunt '. 75 ' L. Sulla . . . aliquos popidos recepit '■ 80 ' Italicis populia a aenatu civitaa data eat '. 86 ' SuUa cum ItaUcia populia foedua percusaifc '. A writer of the age of the Auctor might well write populi simply and be underatood at once. ' The transposition of the et from before popvlorum to follow it, may- be justified by the muddle that had arisen at a very early date in the copying of these words. Digitized by Microsoft® 96 ON THE DATE OP THE latter, I may note, is now accepted as giving us the year 86, and the last consulship and death of Marius, as a terminus ex quo for the date of the Bhetorica ad Herennium?- But I am strongly inclined to think that we can make a more exact conjecture as to that date, if we carefully examine the first example in relation to the events of the time. I believe that it refers to the movements of Lucullus and his fleet, co- operating with Sulla, and providing for the safe progress of his army to the Hellespont in the spring of 84 B.C. In Marx's edition the text of this passage stands thus : Lemnum praeteriens cepit, inde Thasi praesidium reliquit, post urbem Viminacium sustulit, inde pulsus in Hellespontum statim potitur Abydi. These few words, in spite of corruption, indicate clearly that some one with a fleet is watching the Thracian coast, trying to guard it from a hostile fleet by securing the two islands, Lemnos and Thasos, which (as Strabo says) lie immediately off it, and having the command of the Hellespont as his ultimate object. This at once suggests that some one else is approaching along that coast by the via Egnatia, for whom the naval commander is making his march safe. Now early in 84 Sulla was advancing from Greece by the via Egnatia, and had commissioned Lucullus, who after various adventures had established himself with a fleet off the Troad,^ to clear the Aegean and the Hellespont of the forces of Mith- ridates. The situation is thus exactly that which is reflected in the words just quoted. Lucullus took Lemnos, occupied Thasos, then did two other things which the corruption of the text lesives obscure, and finally, exactly as Appian relates,^ returned to the Hellespont, and seized Abydos on the Asiatic side, so as to prevent Mithridates from establishing himself there. With Abydos in Roman hands, the Hellespont was • Schanz, Rom. Litt. i. 2 (ed. 3), p. 467. ^ pj^t Lucullus, 3. ' App. Mithridalica, 56, implies that Lucullus seized Abydos in advance of Sulla, in order to secure the safe passage of the Hellespont for his chief. It was tljjs passage of Appian that originally suggested to me that the Auctor is referring to these movements. Digitized by Microsoft® BHETORICA AD HERENNIUM 97 also theirs, and Sulla's march was secured. But there are two textual difficulties here about which I must say a word. The first of these lies in the name Viminacium (M uimina- chium^). Under the Empire there was a well-known city of that name some 300 miles inland from the Hellespont, on the Danube, which probably suggested itself to a copyist who could not make out the name in his copy. But it was unknown to the Romans of Marius's time, if indeed it existed then at all. Lysimachia was adopted by Kayser ; a city on the isthmus of the Thracian Chersonese, and therefore in one sense a key to the Hellespont ; and the only city in these regions, sa far as I can discover, whose name ends in machia or acMa, the reading of M.^ But I fear that Lysimachia must be rejected, partly as difficult to reconcile with the MSS., and partly as being probably too strong to be destroyed by a force landing from a fleet. ' Sustulit ', about which there is no doubt, must mean total destruction ; and Lysimachia did not meet with this fate, for it is mentioned by Strabo as existing in his time (vi. 54). It is more likely, I think, that this town may have been some small one, on the Thracian coast, a spot commanding the via Egnatia, of which the name has vanished since its destruction. Anyhow, I have not as yet been able to discover a name which can help us, on that coast. ^ ' Three MSS. of group E have Bithynia, which is as impossible as Vimi- naoia. Surely the name must have become hopelessly corrupted before the date of the archetypes of either M or E. Lysimachia was the conjecture of Spengel. Of. Liv. xxxvii. 36. My friend Mr. J. A. R. Munro ingeniously BUggests that the words post urbem may cAaceal the name Pistyrum^ a small town opposite to Thasos which is mentioned in Herodotus vii. 109. In that case Viminacium may conceal the name of some tribe in the genitive plural ; or the latter part of the word might be, as he suggests ad lacum. An old pupil who has been much on that coast during the war, suggests that urbem might be the town of Thasos, which threatened the garrison that had been left there {vi minacem is his conjecture) and was promptly destroyed. ' It was not in Thasos, as the words praesidium rdiquit prove. It is to be looked for somewhere on the Thracian coast between the Strymon and the Hebrus. 3252 „ Digitized by Microsoft® 98 ON THE DATE OF THE The second difficulty lies in the word ' pulsus ', which might suggest that Lucullus had experiended a reverse immediately after the destruction of the town, whatever it was, and had been compelled to fly to his head-quarters. But pulsus is not in the MSS. ; those of group M have sulsus, all except C, which seems to stand by itself, and to give us occasionally the true reading. C has rursus, and this suggests to me reversus} i. e. Lucullus then retraced his steps to the parts whence he had started to prepare the way for Sulla, and immediately seized Abydos, as we have seen. He had started from the Troad, where he had been watching the entrance to the straits, and having got into touch with Sulla, probably after the destruction of the little town, returned to. the Helles- pont to await his arrival. We know that he then todk Sulla across the strait, and that the interview with Mithridates followed which brought about a peace. These events happened in the spring of 84. If the Auctor aUudes to them, he had probably heard of them from some one serving in the fleet, and put them in at the very end of his work not long after they happened, perhaps the very next year. In any case we now have some ground for bringing our terminus down to the year 84.^ I may add that it seems to me not impossible that the author of this work, who had made no secret in it of his sympathy with the cause of the populares and Italians,' may have perished in the Sullan reign of terror which followed close on its publication. This would account in some degree for the mystery which has shrouded its authorship, and for the fact • Eeversus, I find, was adopted by Baiter and Kayser in the edition of 1860. ' Kayser (notes, p. 310) saw in iv. 52, 65 (example of sermocinatio) an allusion to an outrage at Larinum adverted to by Cicero in pro Cluentio viii. 25, which must have occurred in 83, after Sulla's return to Italy. But this cannot be proved from a comparison of the two passages : for in the former no names are given, and in the latter no details. ' Among other passages which indioatei^ this conclusion I may note iv. 9. 13, 22. 31, 34 46, and 55. 68. The name of Herennius, to whom the work is dedicated, also suggests Italian and Marian associations (Plut. Marina 5). See Schanz, op. cit., I. ii. (ed. 3), 471. Digitized by Microsoft® RHETORIC A AD HERENNIUM 99 that we know of no subsequent work by the same hand. It is at any rate quite fruitless to attempt to identify the author with any individual known to have been living at a later date, whether bearing the name Cornificiu^ (as used to be thought probable) or any other. THE LEX FRUMENTARIA OF GAIUS GRACCHUS The details of this bill are clear as far as our knowledge goes, though that is but a little way. Corn was to be sold in . the capital to any one applying for it at the rate of 6i asses per modius ;^ no one was to be excluded from purchasing at this price on account of position or means.^ What amount any one could buy at one time is uncertain ; the amount needed by an individual was about five modii per month, and it is usually assumed^ that this was the amount named in the bill. The average price of a modius is believed to have been about 16 asses or a denarius ; * but it was liable to jgreat fluctuation, according to the fortune of the harvest. The price fixed by Gracchus may have been calculated as that to which the value of the modius might eventually be reduced, by careful development -of the corn-growing industry, and of the means of regular transport to Rome. Nothing is easier for a modern than to condemn this law as an unpardona le blunder, if not as a crime. When once the laws of political economy have been clearly formulated, it needs no appreciable mental effort to see that you cannot put an artificial price on com, even within a limited space, • Liv. Epit. 60, with Mommsen'a certain emendation of ' senis cum triente '- • This appears from the language of Cicero in Tusc. Disp. iii. 20. 48. Piso Frugi, a oonsularis, had applied for purchase at the cheap rate, though he had spoken against the bill. " On the strength of a passage in Sallust {Hist. Fragmenta, iii. 61. 19, Dietsch). • Marquardt, Staatsverwaltung, ii. Ill, and note. For a good instance of fluctuation under Augustus, Dio. 56. 22 ff. Q 2 Digitized by Microsoft® loo THE LEX FRUMENTARIA OF without producing serious moral and economic damage ; * but in ancient times it was the urgent difficulty of the moment on which the legislator acted, rather than on reasoned con- sideration of causes and their effects. Even as late as the year 1800, beheving that the country was threatened with starva- tion, Pitt, who had been brought up on Adam Smith, wished to put a guaranteed price on importations of grain, and was only prevented from doing so by the urgent remonstrances of Grenville.^ To judge of Gracchus's policy fairly we must do more than simply state the fact that he brought the State power to act artificially on the price of corn in Rome, reducing it by more than one-half, thereby creating a heavy charge on the treasury, putting an unwholesome premium on life in the great city, and bribing the mob to support his other measures. We must endeavour to realize the difficulties he had to meet, and the amount of experience and knowledge which he could bring to bear on them. His legislation comes at the end of a period which had been marked by the rise of new and populous cities under the rule of the Diadochi, and by a great increase in the population of many old cities. Beloch has compared this period with the nineteenth century in m6dern Europe, which has seen exactly the same phenomenon.^ What the causes were in each case need not now detain us ; the fact is sufficient that among the innumerable cities of the Hellenic world some developed in wealth and population to an extraordinary degree, while others sank into comparative insignificance. Hence arose new problems of administration, analogous to those which faced the commanders of large armies serving for a long period, as compared with the far simpler administrsftion of the burgher forces of the city-states. The task of feeding the populations of Alexandria, Antioch, or Carthage, may not have been difficult, since the districts in which such cities were placed were rich in the fruits of the earth, and lent themselves easily ■ Written before the War. ' Stanhope's Life of Pitt, iii. 244. ' BeiiSlkerung, p. 479. Cf. Pohlmann, t/berv6lHrung der antiken Qroas aladtt, ch. ii. Digitized by Microsoft® GAIUS GRACCHUS 101 to the work of transport ; but the new mistress of the Mediter- ranean world was not so happily situated. Italy, though capable of raising sufficient corn for its many cities of moderate size, is in reality a peninsula of hill and moimtain, and must, even in the time of the Gracchi, have still been largely occupied by forest and undrained marsh. At an early period the inhabitants discovered that it was worth their while to spend immense labour in draining the lakes among their hills, in order to provide themselves with fertile corn-land.^ Even now the dwellers in some districts of the Apennines live chiefly on meal made not of grain but of chestnuts. The Italy of Gracchus's day could hardly do more than feed its own country population ; for it must be remem- bered that the people lived almost entirely on grain, con- suming but little meat, and that they used their sheep and cattle chiefly for the production of wool and leather. Even if the country could have been stimulated to a larger produc- tion of grain, the transport of it to Rome would have been both difficult and costly ; for example, the great fertile plain of Cisalpine Gaul, the most valuable corn-land in the country, would have had to send its products to the Roman market in vessels coasting round the heeFand toe of the peninsula, for a land-transport of three hundred miles over the Apennines was practically out of- the question. Yet the population of Rome was steadily increasing, and at the end of the second century B. c. was probably not less than half a million souls.^ And the causes operating to produce this result were so intimately connected with the astonishing rise of the Roman dominion, with slavery, with professional soldiery, with commercial development, with the increase of capital and the conveniences and luxuries of city-life, that to counteract ■ See a very interesting paper by Prof. Tenney Frank, in the American Economic Review, vol. ix, no. 2 (June 1919), on' Agriculture in Early Latium. 2 Beloch {Bevolkerung, oh. ix, sec. 2) puts the population of the city in the early empire at about 800,000, basing this conclusion on three several methods of calculation. The same facts are not available for calculating the population before Sulla's time. Beloch handled the same subject more recently in Klio (iii. 490) and arrived at much the same result. Digitized by Microsoft® 102 THE LEX FRUMENTARIA OF them successfully was almost as impossible as to renounce that dominion itself. So at least it must have seemed to an intelligent statesman of that day. It may be indeed that a highly intelligent governing class, with leisure to attend to home problems as well as to foreign wars, might have done much to check the increase when once the long struggle with Hannibal was over ; but that fatal war inevitably led to others, distracted attention from Italian problems, and warped and narrowed the policy of the Roman aristocracy towards the Italian peoples who had helped to prolong the struggle. The feeding of the great city had to be done at haphazard, without reference to the economic condition of the peninsula of which it was the pohtical centre. Long before the Gracchan age, and even before the Hannibalic war, the aediles, whose duty it was to supervise the supply of grain, had been used to procure the necessary quantity from Sicily and Sardinia ; Sicily in fact paid its tribute to the Roman state in this form, and was forbidden to export corn to other countries.-"- With corn-growing possessions so close at hand, whence the transport was short and safe, the State might well be tempted to keep its growing city popula- tion comfortable and quiet at the expense of these pro-vinces. Cato had described Sicily as the store-cupboard of the Roman people,^ and its possession enabled the government to regulate both the supply and the price of its chief product. The hungry plebs usually expected a commander returning from a success- ful campaign to bring corn -with him for their benefit ; we have records of corn thus obtained being sold at nominal prices even before the Hannibalic war, and again as soon as ' Marquardt, Staatsverwaltung, ii. 109, notea 3 and 4. Polybius (xxviii. 2) says that the Rhodians, -wishing to buy com in Sicily, -were unable to do so without leave from Rome. On the importation of com by Greek cities long before this, see Dr. Grundy's book on Thuoydides, ch. iii. I do not think it has been noticed that Caesar, in describing his o-wn difficulties of supply during the campaign of Dyxrhachium, tells us that the inhabitants of those parts, i. e. Dyrrhachium and Apollonia, consumed for the most part imported grain {Bell. Civ. iii. 42). ' ' Cella penaria reipublicae nostrae, nutrix populi Romani ' : quoted by Cic. Verr. ii. 2. 5. Digitized by Microsoft® GAIUS GBACCHUS 103 it was over.i Such a case is mentioned by Plutarch as occur, ring during the first tribunate of C. Gracchus. A large quantity of corn was sent by a propraetor from Spain for distribution ; Gracchus himself persuaded the Senate to have this corn sold and the value of it returned to the provin- cials from whom it had been stolen.^ But was it not better, he may have thought, to avoid all such dealings in the future — dealings which were unjust to the provincials, and doubtfully beneficial to the Roman plebs ? Let us consider this question from his own point of view, as we may reasonably imagine it. Was it not possible to bring the price of corn permanently low enough to make it worth no man's while to gain a temporary popularity by treats of this kind, for which the treaters themselves paid little or nothing ? Was it not possible so to organize and manipiilate the existing supervision of the corn-supply by the State, as to solve, for the time at least, the problem of keeping the huge population of the city alive and in good humour ? Was it really necessary to leave such dangerous work to the mercies of individual capitalists, or even to the inexperience of yearly- changing aediles, guided by their own intuitions rather than by permanent legal regulations ? It should not be forgotten that these aediles might have from time to time to encoimter special difiiculties, the -result of bad harvests or the neglect of their predecessors. Just before the first tribunate of Gracchus an extraordinary plague of locusts had done irreparable damage to the crops in Africa ; and it has been suggested that this was one of the immediate causes of Gracchus's action.' We know how in later times ' The instances are collected in Marquardt, op. cit., p. 110. A reference to the bibliography of the subject given in the same work, p. 106, note 4 (which might now be considerably enlarged), will serve' to show how care- fully this matter has been investigated. ' Plutarch, Gaiv^ Gracchus, ch. 6. This was at the moment of Gracchus's greatest personal influence. The stroke was a fine one : he would get the grain, gain credit with the provincials for righteous dealing, and at the same time baulk the. personal ambition of the propraetor Pabius. ' Nitzsch, Die Gracchen, p. 393. The plague of locusts is mentioned in Liv. Spit. 60, and by Orosius, v. 11, where it is placed in the year 125 B. c. Digitized by Microsoft® 104 THE LEX FRUMENTARIA OF the security of the corn-supply became a matter of the utmost moment, not only as affecting the lives and temper of the Roman plebs but the political situation of the moment and the ambitions of public men. I have long been convinced that Gracchus's object was not merely, as is so often assumed by historians, to ' bribe ' the hungry plebs into acquiescence with his legislative projects, but (1) to prevent sudden and violent fluctuations in the market-price of corn, which were dangerous both politically and economically ; (2) to stimulate the production of corn in Italy, in harmony with the spirit of his brother's agrarian law, and in immediate connexion with his own ; •*■ to keep the agricultural population on the land, and to facilitate the transport of their produce and its safe warehousing at Rome. Let us consider these two points separately, though in my view they are two parts of one and the same scheme. 1 . We have in the history of our own country a good example of the acceptance by a State of the policy of maintaining a fair and steady price for the staple food of a nation, at a period when economic circumstances were apt to produce sudden and unwholesome fluctuations. Mr. R. E. Prothero (Lord Ernie), in his History of British Farming, has shown how the English Government in the seventeenth and eighteenth centuries, though they had no such dangerous circumstances to contend with as the Roman Government in Gracchus's time, yet found it desirable to do what they could to maintain ' a just price '? ' When prices were below a certain level, foreign imports were practically prohibited, exports of home- grown corn permitted, and the quantity of production stimulated by bounties. When home prices rose above a certain level, the bounties ceased, exports were prohibited. See also Dr. Greenidge'a History of Borne, p. 205. For a modern parallel see this book, p. 165. ' In my view the corn-law and the land-law of C. Gracchus were both passed in his first tribunate and about the same time. Dr. Heitland is of the same opinion (Roman Republic, ii. 301-2) ; but he differs from me in thinking that the two laws were irreconcileable with each other, from the point of view ©f the agrarian interest, and that Gracchus did not perceive this. ' Prothero, English Farming Past and Present, pp. 255 and 257. Digitized by Microsoft® GAIUS GRACCHUS 105 and imports of foreign corn admitted duty free or at reduced rates.' This was the method, which differed of course from that of the Roman legislator, but the object was essentially the same in both cases. Prices of corn both in England and ancient Italy, owing to difficulties of transport and warehousing, were much more sensitive then than now to fluctuations in the yield of harvests. Again, in Italy as in England, the amount of corn produced (even if we include Sicily in the former case) was as a rule not much more than was needed for the popula- tion, and when a bad harvest was anticipated .or realized, corn might run up to panic prices. Thus the general aim of legislators, as Mr. Prothero puts it for England, was to maintain an abundant supply of food at fair and steady prices ; and to save the country from violent oscillations between cheapness and dearness. It is interesting to find that among the practical steps taken for this end, one at least is found in ancient Rome as well as in modern England. ' The erection of public granaries, in which farmers might store the surplus of one year against the shortage of the next was borrowed from Holland, and urged on the country by Royal proclamation. In 1620 the King's Council wrote letters into every shire ... to provide a granary or storehouse, with stock to buy corn, and keep it for a dear year.' We are expressly told by Plutarch, whose authority for this biography was unquestionably a contemporary one,^ that Gracchus paid special and personal attention to the erection of granaries, and this must have been with the same object of keeping the supply as constant as possible. 2. It may be argued that Gracchus could not have meant to stimulate production by this bill, since it lowered the price of corn by more than one-half in the city, and would thus rather discoui'age agrarian effort. But I doubt if this argu- ment will hold good. I take it that the ordinary small holder under the Gracchan legislation would not usually look on the ' Plutarch, G. Gracchus, oh. 6. For an excellent and judicious investiga- tion of the value of Plutarch's Uves of the Gracchi, see the introduction to Mr. G. E. Underhill's edition, 1892. Cf. English Hist. Rev., 1905, pp. 212 ft. Digitized by Microsoft® 106 THE LEX FRUMENTARIA OF Roman market as his main object ; but after a good harvest he would be only too glad to sell to the government what grain he had beyond the needs of his family or of the nearest town, even if the price he could obtain for it were not a high one. And that government might be willing to pay him more for the corn than they could get for it in Rome under the lex frumentaria. In any case the grower would benefit, if he could only get his corn conveyed to Rome for deposit in the public granaries ; and this seems to have been the object of Gracchus's road-making, on which Plutarch lays so much stress. Of his own lex agraria we know nothing, though we may be fairly sure that it practically re-enacted his brother's and set the agrarian commission on its feet again ; but we do know the important fact that it was accompanied by a lex viaria (road-law), and Plutarch describes the way in which this was carried out under Gracchus's personal super- vision.^ These roads must have been meant to assist the small farmers who were to benefit under the agrarian laws ; the assistance they would most naturally need would be that of increased facility for transport from their farms to the great roads.^ I think, therefore, that I am justified in concluding that roads and granaries taken together with the corn-law and the agrarian law are all parts of a plan for the encourage- ment of the Italian corn-grower by enabling him to find eventually a market in the capital, after supplying the needs of his own household or locality. Now supposing that these were the more spacious aims of Gracchus, apart from the narrower and immediate one which is usually ascribed to him, it is none the less inevitable that we should condemn his method. He no doubt honestly thought that he could fix the price of corn in the city, just ' Plutarch, 0. Gh-acchits, ch. 7. " Such a law was attached to (or wag a part of) the lex Servilia Rvtti in 63, and again to another lex agraria which failed to pass, in 51 b. c, Cio. ad Fam. viii. 6. 5 : ' (Curio) transfugit ad populum et pro Caesare loqui ooepit : legemque viariam, non dissimilem agrariae Bulli, et alimentariam, quae iubefc aedilea metiri, iactavit.' The concurrence here of a lex viaria and a lex alimentaria is significant. Digitized by Microsoft® GAIUS GRACCHUS , 107 as long afterwards Diocletian fancied that he could fix prices for articles of all kinds. If we compare such a policy with that of the English Government to which I referred just now, we see the blunder at once. The EngUsh policy was an honest attempt to keep prices at a reasonable average by adjusting the imposition of duties and bounties to the economic cir- cumstances of the moment. It inay be compared to the action of a coxswain who keeps a straight course by moving the helm first to one side, then to another, as tide, current, and wind act upon his boat. But Gracchus, living in the infancy of State-navigation, would seem to have known no better than to tie the tiller-ropes fast, in the expectation that the boat would go straight ; he fixed by law a permanent price, hoping that it might be possible to abide by it. But there is no doubt that he miscalculated the cost to the State and the drain on its resources, though there is some ground, as we shall see directly, for thinking that he did not neglect to make some estimate of it. And there is also no doubt that he failed to see that less scrupulous leaders of the ' people ', exaggerating his policy, would sooner or later be ready to feed the city-mob entirely at the expense of the State — as actually in due time happened.^ All this seems to us so strange, especially to beginners in the study of Roman politics who have some chance acquain- tance with political economy, that I have often been at pains to try "to realize for myself and my pupils not only the condi- tions of the problem before Gracchus, but the mental equip- ment he could bring to bear on it. We are so apt to think of these crises of ancient history simply in terms of modern life, and to judge of the actors in them by modern standards of experience, that it may be worth while to ask how a man of such remarkable ability and integrity as Gracchus can have been so curiously misled. First, I would point out that there was then no political philosophy in existence which took into account this particular ' The facts will be found put together in Marquardt, StaatsvervxUtung, ii, loo. cit- Digitized by Microsoft® 108 THE LEX FRUMENTARIA OF difficulty of life in great cities. The old Greek thinkers dealt with the ttoXls, the city which ideally at least needed no support from without ; ^ and those of the Hellenistic period did not greatly trouble themselves about the practical problems of government. Thus the Roman statesman, even if educated, like the Gracchi, by Greek philosophic teachers, had no intellectual inheritance to draw upon in such matters ; nor indeed had he any sound tradition of Roman experience to work on. Experience had indeed shown that the senate and magistrates had so far been able to feed the plebs urbana and keep it quiet by means of an unlimited power of organization ; but when the reforming statesman arose, and was brought into conflict with that oligarchy, compelled to figure as the leader of the sovereign people against it, he himsdf became inevitably responsible both for the order and for the food-supply. Yet Gracchus, in facing these difficulties, at one of the most critical moments of Roman history, had no experience to guide him but that of his political opponents, and no organiza- tion but such as he could create by the force of his own genius. Later legislators, Sulla, Caesar, Augustus, being backed by military force, had both time and organization at command ; but Gracchus was' in the peculiar position of being in opposition yet forced to govern, and unable to reckon for certain on having sufficient time to think out and work out his problems. Here, as in almost every political question of the age, we see the utter inadequacy of the machinery of the city-state to cope with the difficulties of an imperial system. It was empire that had produced the vast increase of the city population, and it was by means of empire that that population was destined to be fed ; ^ yet the constitution ' The Greek remedy for over-population was colonization, and thia was recognized by Plato {Laws, 708 b). But the problem of feeding a city-popula- tion does not seem to have attracted the attention of philosophers, though if Dr. Grundy is right (Thucydides, oh. iii) they must have been aware of it. In the Hellenistic age the mainland of Greece was sufltering from depopulation ; the o'f er-populated cities were far away, and the philosophers of (jjie post-Aristoteliftn schools do not seem to have been interested in their difficulties. " The treasury was filled, mainly from the provinces, and the com supply Digitized by Microsoft® GAIUS GRACCHUS 109 remained that of the city-state with its yearly changing magistrates, and the only hope of efficient administration seemed to lie in the permanent council, the senate. At one moment it looked' as if Gracchus were about to overpower this oligarchic council by sheer weight of ability and by the courteous tact which distinguished him ; ^ but in fact he had no chance against its traditions and prejudices. Yet it is not unlikely that if he had been able from the first to dominate or reform it, no permanent lex frumentaria might have been needed. Again, if Gracchus could have had time and the means of organization, his true policy would have been to police the city adequately, as Augustus afterwards did ; to check the growth of slavery, and consequently of manumission ; to encourage industrial undertakings in the city, as well as agricultiural activity in Italy ; and to organize the foreign corn-supply efEectively, so as to keep the natural price per- manently low, while holding fast to the principle that the State should not expend its wealth on the maintenance of its ' unemployed '. But for the moment, as it must have seemed to him, the one necessary condition of getting any of these reforms started was to keep the sovereign mob com- fortable at any price. The treasury was full, yet the citizens were almost untaxed, and the temptation was great. Counter- acting measures might follow, agrarian bills, colony bills, road-making bills, the eventual result of which might be to relieve the treasury of its new burden. It may be that he saw the necessity of such measures as antidotes to the corn-law ; at any rate he devoted to the execution of them such an amount of personal energy as suggests this strongly. But for the moment he was compelled to find cheap corn for the people, because for the moment they were masters of the situation. If he seriously believed that he was thus creating a per- manent charge on the treasury, he was certainly to be blamed, and if he hoped to avoid it he was as certainly over-sanguine ; itself tended more and more to depend upon Sicilian, African, and Egyptian harvests. ' See the story about the corn from Spain in Plutarch, C. Oracclms, ch. 6, and the description of his relation to the senate there given. Digitized by Microsoft® no THE LEX FRVMENTARIA OF GAIUS GRACCHUS but we must remember that we have none of those speeches in which he developed his policy and defended it. ' Read his speeches ', wrote Cicero in a memorable passage, ' and you would call him a veritable patrdnus aerarii '.^ This makes it more than probable that he had gone carefully into the financial question ; and to me it also suggests that he was devising schemes for neutralizing the drain on the treasury by so developing the production and transport of corn, as to bring down the price at Rome by natural instead of artificial means. In conclusion, I would raise the question whether these alimentations, as we may call them, were really so vicious in their consequences as is generally assumed. That they constituted a serious evil no one can deny ; but I am inclined to think it possible that they saved Rome from still worse evils. When the government of an empire is concentrated in a single city, and that city is in the power of an ignorant, hungry, and idle mob, the statesman can only be described as sitting on a volcano. Until military organization and disci- pline could be brought to bear on it, till the centre of political gravity could be shitted away from it, the danger was chronic and extreme. Bad government and the lawless population of Alexandria brought the Empire of the Ptolemies to ruin ; bad government and an unfed populace might have done the same for the Roman Empire. No Roman statesman for a century after Gracchus was able to counteract the tendencies which kept this great mass of population crowded in the city ; for all of them circumstances were too strong, and all but Sulla acquiesced in the Gracchan remedy. It was an ugly running sore in the Roman system, and no physician could be found to attack the cause at its root. All that could be done was to prevent the disease becoming mortal ; and though he probably had higher aims, this seems the result of Gracchus's policy. ' Tusc. Disp. iii. 20, 48. Cicero is here contrasting the words and acts of Gracchus. But Gracchus was no mere rhetorician, as we know from Cicero's own evidence {Brutus, 126 ' grandis est verbis, sapiens sententiis, genere toto*gravis '), and this passage must be taken as indicating a serious attempt to deal with the financial aspect of his bill, Digitized by Microsoft® Ill THE CARMEN 8AECVLARE OF HORACE AND ITS PERFORMANCE, JUNE 3rd, 17 B.C. The great object of Augustus in celebrating Ludi saeculares in 17 B. c. was to encourage the belief in himself and the consequent active loyalty to himself, as the restorer of the jpax deorum — the good relation between the divine and human inhabitants of Rome. So far he had tried to attain this end by the ancient usual and proper means, i. e. by carrying out the various regulations of the ius diuinum, so many of which had long been neglected. But in that year he determined to undertake a special celebration, with the design of more efEectually stamping the impression already made on the minds of the people ; and it so happens that we have more detailed knowledge of this celebration than of any other Roman rite of any period. This is fortunate, for it stands on the margin between an old and a new regime, like the Aeneid of Virgil, who had died two years earlier : that great religious poem was just becoming known, and there is an allusion to it in the hymn of which I am going to speak.^ The Ludi were the outward or ritualistic expression of the idea immortalized by the poet, that a regeneration is at hand of Rome and Italy, in religion, morals, agriculture, government : 'old things are now to be put away,^ a new and glorious era is to open. Henceforward the Roman was to look ahead in hope and confidence, trusting in Augustus, the Aeneas of the actual State. Thus the study of the ritual of this festival is in every way most instructive, and every one can study it for himself in the several sources from which our information is derived : in the account given by Zosimus,* in the Sibylline oracle • Line 40 fE. * For the meaning of saecvlum and saecvlum condere, see Mommsen, B6m. Chronologie, ed. 2, p. 172, and Wissowa, Abhandlungen zur romischen Rdigions- wnd Stadtgeaehichte, pp. 200-2. ' Zosimus, ii. 6 : the oracle is in ii. 6. Both are printed in Wickham's Horace in the introduction to the Carm. Saec. Digitized by Microsoft® 112 THE CARMEN SAECVLARE OF HORACE AND which he has fortunately preserved, in the hymn sung on the last day of the Ludi, and in the inscribed Acta of which a great part was discovered in 1890 hy the Tiber bank near the Ponte St. Angelo.^ Soon after the pubhcation of this latest source, it was discussed from three several points of view at one of the most interesting meetings of the Oxford Philological Society which I have ever attended. I do not remember that any of us who took part in that discussion laid stress on the new light which it threw on the performance of the Carmen saeculare : but it soon became apparent that it had a direct bearing not only on the performance but on the matter and composition of the hymn, and would give rise to controversy on these points. For whereas up till then we only knew that it was sung on the Palatine,^ before the temple of Apollo which Augustus had lately built there ' in priuato Bolo ', we learnt from the Acta (line 147) that it was sung also on the Capitol : Sacrificioqtie perfecto pueri XXVII quibus denuntiatum erat patrimi et matrimi et puellae totidem carmen cecinerunt (i. e. on the Palatine) eodemque modo in Capitolio. Carmen composuit Q. Horatius Flaccus. Mommsen in commenting on this in Ephemeris EpigrapMca VIII, where he published the inscription, insisted that the argumentum of the hymn forbids us to suppose that the whole of it was sung both on the Palatine and on the Capitoline, for only the middle part of it, where Jupiter and Juno the Capitoline deities are rather obscurely hinted at, is suited to the Capitol, while Apollo and Diana (who was associated with Apollo in the Palatine temple) ^ are prominent both at the beginning and end. The first part, he contended, was sung on the Palatine, the middle part on the Capitoline, and the last part again after the return of the choirs to their original station. As to singing en route, he did not express a definite opinion. » Ephemeris EpigrapMca VIII, 255 ff., contains the text and Mommsen's commentary. Dessau, Inscript. Selectae, ii. 1. 282, does not give the whole document. C. I. L. vi. 32323. = So Zosimus, who adds that the hymn was sung both in Latin and Greek : but of this we have no confirmation. ' Propertius, ii, 31. 15. It also seems to be implied in Plin. N. H, xxxvi. 13. Digitized by Microsoft® ITS PERFORMANCE, JUNE 3rd, 17 b. c. 113 This view gave. rise to a good deal of controversy, as will be seen by reference to the last edition of Wickham's Horace ; but it has recently been reasserted by Wissowa, who in reprinting a paper written in 1894 about the Ludi added a note in which he declared that the threefold division of the Carmen ' springt in die Augen ', that lines 37-52 belong to the Capitoline worship, all the rest to that of the Palatine, and assumes that what belongs to each temple area was necessarily sung there. He too declines to speak with con- fidence about singing during the procession from one area to the other. ^ It may clear the ground if I state my reasons for believing that the hymn was not sung in procession at all. True, such singing was not unknown at Rome. In 207 b. c. a choir of twenty-seven virgins sang a carmen in Saturnian verse of Livius the poet, as they went in procession from the shrine of Apollo in the Campus Martius to the Forum, where they stopped and danced in a peculiar manner with a rope,^ and thence proceeded, whether singing or not Livy does not say, to the temple of Juno Regina on the Aventine, which had been struck by lightning. This performance, as Diels has observed in his Sibyllinische Blatter,^ stands half-way between the old semi-magical singing and dancing of the Salii and the Fratres Arvales, and the singing of our Carmen, which was really only a carmen (so to speak) by courtesy, having no magical intent whatever,* and, as being in Greek lyrical metre, does not ' Wissowa, Gesammelte AbhancUungen zur romischen Religions- und Stadtgeschichte, p. 206 and note. Mommsen in Ephemeris JBlpigraphica, viii. 256. 2 Liv. xxvii. 37 ' Septem et uiginti uirgines, longam indutae uestem, carmen in (i. b. addressed to) lunonem Keginam oanentes ibant : ilia tempestate forsitan laudabile rudibus ingeniis, nunc abhorrens et incon- ditum, si referatur. ... A porta (Carmentali) lugario nioo in forum uenere : in foro pompa constitit : et per manus reate data, uirgines sonum uocis pulsu pedum modulantes incesserunt.' Diels, Sib. Blatter 91, puts this rope-dancing down as Greek, not Eoman, and connects it with the ropes which occur in lists of articles paid for by the Upoirowi in Delian inscrip- tions. ^ p. 91, note 1. ' For the original magical meaning of the word, see Jevons in Anthropology and the Classics , pp. 94 ff. 2352 H Digitized by Microsoft® 114 THE CARMEN 8AE0VLARE OF HORACE AND suggest dancing in the sense of any ol(i Roman religious practice. Rhythmic movements of some kind there certainly were, as I hope to show directly, but in the two sacred areas, not in procession from one to the other. . And further, the words of the Acta seem to me explicit : the hymn was sung on Palatine and Capitoline, and nothing is there said of any point between the two. Practically too there would have been serious .difficulty in marshalling- fifty-four boys and girls, if they sang as they Went down the steep hill from the Apollo temple to the Sacra via, along that irregular way and through the narrow fornix Eabianus into the Eorum, and finally up the steep ascent to the temple of Jupiter. "Gardt- hausen, in his work on Augustus,^ suggested that they might have stopped at particular points to sing, e. g. in the Forum : but there is nothing in the hymn or the Acta to support this — no deity of the Forum is mentioned, nor did the Forum play any part in the religious rites of the Ludi. Once more, if the children had to sing in procession, accompanied as I presume they were by instruments,^ careful rehearsal would be needed more than once : and if this were done in public as it niust have been, it would destroy the novelty of the performance on June 3. We know from Odes iv. 5 ad fin. that Horace took pains with his rehearsing : but he says nothing there that can suggest processional singing. It is far better, I think, to accept the words of the Acta as giving us the simple fact. They are in other matters curiously explicit, and it is unlikely .that in this one particular they should have been unnecessarily concise. I prefer even to accept the literal statement that the hymn was sung right through once on the Palatine and once on the Capitoline, and that the performance came to an end there. At any rate we will for the moment assume that the children were not compelled, after singing nineteen stanzas in one place and thesame number at another ' Augustus und seine Zeit, vol. i, pt. ii, p. 630. Ferrero, Greatness and Decline of Borne, v. 94, note, is right in obj acting to this kind of interpretation. " Aenatores are mentioned in line 88 of the Acta': but these belong to another f)art of the Ludi. I imagine that the boys and girls were accom- panied by tibieines, Digitized by Microsoft® ITS PERFORMANCE, JUNE 3bd, 17 b. c. 115 nearly half a mile away, to plod back again and go through them all once more at the original starting-point. But how are we to reconcile this limitation of the places of performance to two, with the contents of the poem itself ? In order to explain how I think this may be done, I must digress for a moment, and consider what the instructions must have been which Augustus gave to his poet-laureate. We must not of course imagine that on such an occasion Horace was left to himself. I suppose there is no bit of Latin poetry which has more constantly been in my mind than this hymn : and the impression it always gives me is that Augustus wrote out in prose what he wanted put into it, and that his laureate did this with consummate skill and concinnitas ; but the result, for me at least, is that it is as flat as such composi- tions usually have been. Nay, it is occasionally prosy, as e. g. in the fifth and sixth stanzas. Recently Signer Ferrero has glorified it as a most magnificent poem, full of inspiration, in language which suggests as great a want of judgement in literary, as he so often exhibits in historical, criticism.^ At any rate the inspiration came from Augustus and not from the poet's native genius. First, I should like to point out (though it does not directly concern our problem) that Augustus clearly wished Horace to combine in the hymn the three ideas of religion, morality, and the fertility of man, beast, and crop. The Princeps, I have long been confident, had grasped the fundamental idea of the old Roman worship, still alive in the hearts of most Italians, that this general fertility, without which the State could not go on and prosper, depended on the dutiful attention {pietas) paid to the divine beings who had taken up their abode in farm or city ; ideas which covered the ordered life and. religion both of family and State — ^both morality and religious duty. All these three ideas will be found duly expressed in the hymn.^ Secondly, Horace must have had instructions not to mention Augustus personally — that would be unnecessary, > op. cit., V. 90 ff. ' e. g. in lines 13 ff., 29 ff., 45 ff., and 57 ff. H 2 Digitized by Microsoft® 116 THE CARMEN SAEGVLARE OF HORACE AND owing to the prominent part taken by him in the whole ritual of the Ludi ; but to give his poerii a strong ApoUine colouring, Tt^hich was much the same thing as giving it a strong Augustan colouring, so completely had Augustus by this time come to be associated with the god whom he had settled close to his own house on the Palatine and on his own land. This of course suited Horace exactly as a poet, and he expresses his satisfaction in Odes iv. 6. 29 ff. I think it is quite possible that he went a little beyond his instructions ; for Apollo and Diana are far more prominent than Jupiter and Juno of the Capitol, who are only alluded to, not named. ^ No doubt it was part of Augustus's policy to put the great Jupiter of the republic somewhat in the background as compared with his own Apollo : this can be proved in many ways : ^ yet I half suspect that Horace here went a little beyond what was required of him. The Capitoline deities had as a matter of fact been so far, i. e. up to the third day, on which the hymn was sung, more prominent in the ritual than Apollo. Thirdly, Horace must have been told that the hymn must .contain allusions to all the deities invoked in the ritual diu-ing the three previous nights and two days, as well as those of the third day. Now offerings had been made on the first night, at the underground altar of the Tarentum, near the Tiber bank, to the Moirae, on the second to Ileithyia (or the plural), and on the third to Tellus or Ceres : ^ and these all duly appear in stanzas 4 to 8 inclusive. Then by day the sacrifices had been ofEered to the Capitoline deities — and they ' i. e. in 45-52. ' See e. g. J. B. Carter, Religion of Numa, pp. 166 ff., who has many interesting remarks on the ApoUinism of Augustus. ' In line 29 it is tempting to write Tellus with a capital T : but here Ceres seems to be performing her part as deity. The two run very closely together throughout the early history of the Roman religion : see my Roman Festivals, pp. 73 ff., Wissowa, Rel. und Kult. der Romer, pp. 192 ff. Mr. Stuart Jones has drawn my attention to Petersen's very interesting suggestion of a connexion between this stanza and the slab from the Ara Paoia in tiie TJffizi at Plorence ; see Petersen, Ara Pads Augu-itae, pp. 48 ff. : Mrs. Strong's Roman Sculpture, p. 42. Digitized by Microsoft® ITS PERFORMANCE, JUNE 3ed, 17 b. c. 117 too are here, though somewhat obscured, in lines 45 to 52. These Capitoline deities are followed by Fides, Bax, Honos, &c., of whom we do not hear anything in the ritual : but this point I must for the moment postpone. Then Apollo and Diana come to the front again, and the hymn ends with a kind of summing-up of all the deities, Jupiter, Apollo, and Diana being mentioned by name. The result of all this is that |to a casual reader the hymii is a jumble of divine names, Roman and Greek, with Apollo appearing oftener than the rest, and almost abruptly in lines 33 and 61. Even when we have learnt all about the ritual and the policy of Augustus, it is very hard to divide the poem intelligibly : and I confess that no threefold division of it has ever " sprung into my eyes ', as into Dr. Wissowa's.^ Those five stanzas concerned with the deities of the night- ritual absolutely forbid it. So far as I can see, it runs thus : (1) an Apolline introduction or proodos of two stanzas, with an invocation of Apollo as Sol, which I will explain directly ; (2) five stanzas concerned with the deities of the nightly worship at the Tarentum ; (3) a return to Apollo for three stanzas ; (4) an appeal to the Capitoline deities, whom we identify by the white victims in line 49, followed by two apparently rather irrelevant stanzas about the prestige of Rome and her virtues ; (5) another return to Apollo and Diana ; and a concluding stanza, summing up the whole. How are we to reconcile all this apparent confusion with the singing of the hymn on Palatine and Capitoline only — ^i. e. on two sites with which Jupiter and Juno, Apollo and Diana were respectively alone concerned ? To sing of Jupiter or the Parcae or Tellus at the temple of Apollo on the Palatine ' Ahhanilungen, p. 206, note 1, quoting Vahlen, whose paper I have not been able to see. Wissowa seems to take the whole down to line 36 as standing together and Apolline (p. 207, note). But to me lines 13 to 32 are plainly in honour of the deities of the Tarentum, though Ileithyia is introduced first instead of second, perhaps in order to run her into a dim kind of identification with Diana Luoina, or Juno Lucina, or both. This would suit the last and Apolline day of the festival : and we must note that the Tarentine deities are not now Dis and Proserpina, i. e. sinister deities of the underworld, but helpful ones (Wiss. 208). Digitized by Microsoft® 118 THE CABMEN 8 AEG VL ARE OF HORACE AND would seem inappropriate, if we assume that the whole hymn was gone through there, as the Acta plainly imply ; and still more inappropriate would it seem that Apollo and Diana and the deities of the undergroimd altar should be celebrated in the precincts of the great Jupiterof the Capitol. The solution of these difficulties which I now propose for criticism has been suggested by a consideration of the nature of the two sites on'which we know for certain that the hymn was sung : combined with the further consideration, in which no doubt every one will agree with me, that this hymn was not sung by two choruses of boys and girls standing stock still all the time, but making certain movements like the simple evolutions of the Greek chorus. This is now made clear by line 21 of the Acta, which probably belongs to a letter of Augustus to the quindepemviri datable some three months before the festival. This left plenty of time for choir training, and the inference is that there was plenty to learn. The words of Augustus's letter show that there was more than learning the hymn by heart ; the necessary steps are to be taken ' ad carmen canendum chorosque habendos '. Here some kind of evolutions must be meant, if not exactly dancing.^ We cannot certainly know what those movements were ; but we may be sure that they would add to the interest and pleasure both of performers and spectators : and perhaps what I am going to say about the nature of the two sites will help us in guessing at some of them. The only possible site of the temple of Apollo, says Hiilsen,^ iu at the north-eastern corner of the Palatine. But there are serious objections to this view, and of late the conviction has been growing that the real site was where the temple of Jupiter Victor is generally supposed to be, overlooking the Forum Boarium and the Circus Maximus, whence there is an uninterrupted view over the Campus Martins, with the Capitol in the foreground a little to the right- — a point of importance for my interpretation of the Carmen, as will be ' For Hos&ce'a uae of the word chorus, see Odes iv. 7. 6 ; i. 4. 5. C£. Propertius, ii. 2. 28. ^ Jordan-Htilsen, Eom. Topogr. iii. 72. Digitized by Microsoft® ITS PERFORMANCE, JUNE 3bd, 17 b. c. 119 seen directly. Ovid suggests this site strongly in the lines in his Tristia (iii. 1. 59 fE.) in which he imagines his book arriving from. Tomi and climbing up to this temple : Inde te/nore pari gradibus sublimia celsis Ducor ad intonsi Candida templa dei. The words teriore pari must mean ' going straight on ', equiva- lent to uno tenore, as Professor A. C. Glark suggests to me. As Ovid's book is supposed to enter the Palatine by the temple of Jupiter Stator from the sacra via, and to arrive first at the domus of Augustus, it would have to turn sharp to the left if the Apollo temple were on Hiilsen's site, but would go straight on if it were beyond the domus at the western edge of the hill. Here was space enough for a grand area, enclosing the temple to north, south, and west, and in this open space a few very simple movements would enable the chorus to command every other site of religious or historical interest in the city, now adorned in all directions with new or restored buildings. There was no building as yet on the Palatine that could interfere with this view. A little to the right was the Capitol, with its own splendid temple, rising above the Forum ; and beyond that again, plainly visible in the distance just to the left of the Capitoline hill, was the site of the Tarentum, where the midnight ceremonies had been held. Let us now apply our knowledge of this splendid prospect to the subject-matter of the Carmen. At once we see that the first three stanzas hang together, and contain a happy allusion to the view from the area : ' possis nihil Roma uisere mains.' But they contain also another allusion, which (so far as I know) has not been noticed in this connexion. On the fastigium of the temple there was, as Propertius tells us,^ a figure of Sol with a quadriga : Turn medium claro surgebat mar more templum Et patria Phoebo carius Ortygia. In quo Solis erat supra fastigia currus . . . If we suppose that the first two stanzas were sung by the united choirs in position, or as they wheeled into position on ' Propertius, ii. 31-10 (ed. H. E. Butler) foil. Digitized by Microsoft® 120 THE CARMEN SAEGVLARE OF HORACE AND the area, facing the view, we may safely conjectiire that when they reached the third stanza they would wheel again to face the temple and Sol looking down from his fastigium. If the real sun were shining at the time the effect of this fine stanza would be very impressive. It is in my humble judgement the best in the poem : Alme Sol, curru nitido diem qui Promis et celas, aliusque et idem Nasceris, possis nihil urbe Roma Visere maius. As they sang these last words the choirs may have wheeled again to face the prospect of the city. If so, they would then be in the right position for celebrating the next group of deities, those of the Tarentum and the nightly rites, who must of course be taken before those of day and light, as in the order of the festival. With what movements, if any, the following five stanzas were sung, which plainly refer to the nightly rites, invoking the female deities there worshipped, it is impossible to say. They may have all been sung by the girls, and softly : they are certainly not so well suited to the boys. Nor would it be easy to explain why there are so many of them — ^five in all — if Augustus had not given his poet strict orders to bring in the lex de maritandis ordinibus ^ — ^a task accomplished deftly in that prosaic stanza to which I referred just now. Doubtless Horace was very glad to get back to Apollo in the next group of stanzas, to which I now turn. The. more closely I examine the hymn, the more convinced I become that it is purposely written so as to keep its ApoUine character persistently in the minds of the audience ; three of its five parts are Apolline — ^the first, third, and fifth : in be- tween these we have the deities of the Tarentum and those of the Capitol. This reversion to the Apolline character would be emphasized, at the point we have now reached, by a movement of the choirs which would bring them once more into position facing the temple. 1 This les Julia had come into effect the year before that of the Ludi, viz. 18 B. c. Digitized by Microsoft® ITS PERFORMANCE, JUNE 3bd, 17 b. c. 121 There is a difficulty here as to how far we are to consider this reversion to Apollo and Diana as continuing. Wissowa ^ would limit it to one stanza, putting a full stop (I suppose) after puellas, and connecting the four following stanzas together as Capitoline, i. e. as addressed to Jupiter and Juno. By general consent the last of these, beginning ' Quaeque uos bobus ueTieratur albis ' is so addressed : for the white heifer was the special victim of Jupiter, and was never offered to Apollo : the Acta inform us that Jupiter and Juno had on the two previous days been propitiated with boues pulchri and pulchrae, while to Apollo had been offered only calces of various kinds. There is also a strong opinion (Mommsen, Vahlen, Wissowa) that the stanza immediately before this one, ' Di probos mores, &c.', belongs to the Capitoline deities, and with this I agree, seeing that probi mores, and the general well-being of the gens Romula would be much more natxirally connected with Jupiter and Juno than with ApoUo and Diana, But I cannot, with Wissowa, begin the Capitoline part with ' Roma si uestrum est opus ' ; Apollo was the protecting god of Troy, and the way in which he is treated in Odes iv. 6, written at this same time, seems to me to make it clear that Augustus wished to encourage the idea that Rome was in a legendary sense at least the work of Apollo. I should therefore put a full stop after relictis, and there end the ApoUine diversion : and imagine the choirs turning towards the Capitoline temple in front of them to begin — perhaps after a pause — the address to Jupiter and Juno with ' Di, probos mores docili iuuentae '. With this must be connected, not only the next stanza, -about which there is no doubt, but that which follows it ; in both these we have a distinct expression of the imperial idea, and the mission of Rome in the world, and this idea could be associated with no other deity than Jupiter Capitolinus, and with no other temple than his. Every outward sign of the Roman imperium was thus associated in the minds of the people, and Augustus must have known well enough 1 Abhandlungen, p. 207, note. Digitized by Microsoft® 122 THE CARMEN SAECVLARE OF HORACE AND that any change in this could' onlylbe very gradually accom- plished. I will also hazard a conjecture that the fifteenth stanza, with the names of Fides, Pax, and other deified abstractions, belongs to this Capitoline section of the hymn. The abode of Fides, an ancient goddess, was on the Capitol, and there is hardly a doubt that she was 'closely connected with Jupiter. Wissowa thinks that she was an offshoot from the ancient Dius Fidius, who must be identified with Jupiter.^ The notion in Augustus's mind, which Horace had here to reproduce, was, I think, to lay stress on the Pax Romana, which must rest on the basis of treaties and good faith ; and I look upon Fides as here playing the part of a callida iunctura, connecting Pax, Honos, Virtus, and Pudor, with the great deities of the Capitol. Pax was in Augustus's head at this time, just hover- ing, so to speak, on the verge of deification, as may be seen in Tibullus i. 10 : and the Ara Pacis was begvm only four years later. ^ Honos et Virtus, though -separated in the hymn for metrical reasons, must go together as they always did at Rome ; their character is military, and they suggest Mars and the warlike virtues, which would otherwise be unmen- tioned in the hymn.^ The Pax Romana, let us remember, depended on these virtues as well as the domestic ones. Lastly, Pudor would probably have been Pudicitia if the metre had admitted of it ; the latter was the female family virtue at Rome, and here I think we may see a compliment to Livia, and through her to the Roman matrons. Valerius Maximus not long afterwards, in the preface to his sixth book, de Pttdicitia, thus addressed her : ' Tu prisca rehgione consecrates Vestae focos incolis, tu Capitolinae lunonis » Rd. und Kvlt, p. 129 ff. " Wisaowa, E. K., p. 334. ' Wissowa, iJ. iT., pp. 149 £E. It used to be supposed that there was a temple' to these deities on the Capitol (see e. g. Burn, Borne, and the Campagna, p. 193), the work of Marius. The site of Marius's temple is, however, uncertain, though this passage of the Carmen might be used to support the old hypo- thesis. The best-known temple was near the Porta Capena : and it is probably of this temple that Dio Cassius writes (liv. 18) that Augustus in this yeap 17 B. c. fixed the date of its festival on May 29 ; which is almost the same thing as saying that he rebuilt it. Digitized by Microsoft® ITS PERFORMANCE, JUNE 3kd, 17 b. c. 123 puluinaribus incubas, tu Palatii columen augustos Penates sanctissimumque luliae genialem torum adsidua statione celebras.' This brings us to the end of the Capitoline section of the hymn as I understand it, all of which would be sung with the great Capitoline temple as the most conspicuous object in view. Then we return once more to Apollo and Diana, the choir wheeling round so as to face the temple behind them, and possibly turning towards the Aventine at line 69. Here with great skill the poet introduces Apollo as aug%r, probably in allusion to the augurium salutis, which Augustus had revived in 29 B. c.,^ and with the physical welfare of the people still more distinctly in lines 63 and 64. The verbs become indicative,^ anxious prayer changes to confident assertion : and the prosperous future of Rome is thus happily associated with the Augustan Apollo at the end of the perform- ance on the Palatine. The nineteenth and last stanza, which sums up the whole ceremony as Horace and his choirs are about to return to their homes, could not, I think, have been sung here ; it was kept to the real end of the performance, was added as an odd or lucky number, and would be more appropriate to the temple which was still the spiritual home of the idea of Roman greatness. Towards that temple the procession would now make its way, down the steep ascent to the Sacra via, and so through the Forum up to the area Capitolina. Let us now in the last place shortly consider how the hymn was adapted to this site as perfectly as to the other. The area Capitolina was even larger than that of the Apollo temple, and the view from it was equally magnificent ; these two religious sites were in fact the only two in which the choirs would have had ample space for evolutions, and from which at the same time they would be able to see ■ Dio Cass. li. 20 ; Suet. Aug. 31. We know hardly anything about this antique ceremony : but the language of Dio in xxxvii. 241 shows that the word salutis (Wissowa, S. K. 525) is not the deity Salus, but the health of ' the people : cf. Cic. de Legibus, ii. 21 ' augures . . . salutem populi auguranto. ' The medical character of Apollo is apparent in lines 63-4. ^ See Wickham's commentary. Digitized by Microsoft® 124 THE CARMEN SAECVLARE OE HORACE AND almost every other important religious site in Rome. A good idea maybe formed of the size of the Capitoline area from Lanciani's map to scale of the Sacra via/ which includes hoth the Coliseum and the Capitoline ; there it will be seen that the area is at least as large as the whole space occupied by the Coliseum. The temple stood in the middle of it, which accounts for the somewhat astonishing fact that (at one time) chariot-races — a mild form, I presume — used to be held here at the time of the Latin festival.^ As time went on other temples were built here, but there was plenty of room for meetings of Comitia up to the end of the republican period. If we apply these facts to the performance of the Carmen here, we see at once that it could be gone through with motions as perfectly appropriate as on the Palatine. The new temple of Apollo, which they had just left, was in full view across the Forum Boarium and the Velabrum, with the quadriga of Sol on its fastigium — probably in its newness the most brilliant object in sight. Doubtless the choirs would be facing it as they sang the first three stanzas. The site of the Tarentum across the Campus Martius was of course visible from the southern end of the area, and here the choirs would be during the next five stanzas, while they would wheel again to the west when they reached the second Apolline passage. They would be drawn up in front of the great temple during the CapitoUne stanzas that follow, and would wheel about once more for the three Apolline ones with which the singing had concluded on the Palatine. There would then remain the nineteenth stanza, summing up the whole performance : Haec louem sentire deosque cunctos * spem bonam certamque domum reporto, doctus et Phoebi chorus et Dianae dicere laudes. This is an extremely clever stanza ; Horace contrives to bring in Jupiter as after all the presiding genius of Rome, ' In his Ruins and Excavations of ancient Rome, fig. 72. ' Plin. iV. H. xxvii. 45. Digitized by Microsoft® ITS PERFORMANCE, JUNE 3rd, 17 b. o. 125 upon whose good will the future of the State depends, and as also the presiding deity among all the rest — dei cuncti, of all of whom, including Apollo, there were statues in the area Capitolina.^ To me it seems impossible that this concluding stanza should have been sung anywhere but in that area. But at the same time Horace has most dexterously managed to make the final touch an ApoUine one, as would in fact be fitting on a day especially dedicated to the Augustan Apollo, Phoebus and Diana are not here alluded to as the controllers of the destinies of Rome, but as the deities in whose honour the choirs, now about to disperse, have learnt and sung this hymn. Beyond doubt this last stanza was sung by both boys and girls. How the rest of the Carmen was distributed between them I think it is impossible to determine, though many attempts have been made. I have made attempts myself, but never reached a satisfactory conclusion : we simply have not the necessary data. I have found it much more interesting and instructive to myself to correlate the divisions of the hymn with the two sites in which we know it was sung, and the views from them. It is difiicult to realize to the full, even for one who has been constantly occupied with the religious side of Roman life, how intensely local all Roman worship was — how intimate the association between place and cult.^ It was in fact a perfectly right instinct that prompted Mommsen and others to assume that the ApoUine part of the hymn must have been simg on the Palatine, and the Jovian part on the Capito- line, and thus even to strain the plain words of the inscription, ' Eodemque modo in Capitolio . . .' But it was clearly im- possible to carry out such a principle logically on this June 3 ; for to do it the choirs would have had to make a pilgrimage ' Servius, ad Aen. ii. 319. The statue of Apollo here was a remarkable one, thirty cubits high, brought from ApoUonia by M. Luoullus, as Pliny tells us N. H. iv. 92 and xxxiv. 30. ^ The same holds good in the case of Ludi, which were in origin only a form of cult. The .ludi Romani e. g. were in the cult of Jupiter, and originally took place on the dedication day of the Capitoline temple, the ides of September. The Megalesia were celebrated before the temple of Magna Mater, ' in ipso Magnae Matris oonspeetu ' (Cic. Harusp. Resp. 24). Digitized by Microsoft® 126 ON THE LAVDATIO TVBIAE AND of about a mile and back right across the Campus Martius to the Tarentum, and there to sing, in broad dayhght instead of at night, the stanzas appropriated to the Tarentine deities. Yet to leave these out would have been to violate the plan of Augustus for including in the last performance all the deities invoked in the festival. Instead of this Augustus^ chose the tw.o finest religious sites in Rome, from each of which every- thing could be seen that was to bs alluded to in the hymn, for the complete performance ; so far yielding to popular feeling and conviction as to fix the second and last performance for the Capitoline, the real religious centre of the whole empire : but astutely taking care that the interest of this third day's entertainment should be closely connected with himself and the new regime, and that the religious colouring of the ritual and the hymn should be emphatically Apolline. Nothing could please his poet better : Spiritum Phoebus mihi, Phoebus artem carminis nomenque dedit poetae. ON THE LAVDATIO TVBIAE AND ITS ADDITIONAL FRAGMENT {C.I L. vi. 1527) All students of Roman law know the inscription which goes by this name ; and that part of it which raises a compli- cated question of legal inheritance is to be found in the later editions of Brims 's Pontes luris Romani. The whole series of surviving fragments, partly preserved in the Villa Albani in the original marble, pa.rtly in the form of copies made long ago of fragments now lost, contain a record of domestic life of exceptional human interest ; the heartfelt utterance of a husband on the death of a wife absolutely devoted to him for f ortj'-'One years, and addressed, unlike all other laudationes, to herself and not to an audience. The_ portrait which he draws of ier is no rhetorical exaggeration, but mainly a record of facts, and she lives in it for ever as a woman of'extraordinary Digitized by Microsoft® ITS ADDITIONAL FRAGMENT 127 energy, ability, and good sense, yet a real tender-hearted unselfish woman, devoted to her household duties and to the interests of her husband and her relations, unfortunate only in having never borne him a son. The most touching passage in it is perhaps that in which, apparently after the death of an only daughter, he records how she implored him to divorce her and raise up seed by another wife ; he breaks out into a passionate protest against the very thought of such treachery to one who had rescued him by her prudence and self-devotion from imminent dangers, and had lived with him in unbroken harmony for so many years. For a study of this famous inscription, which may almost count as a fragment of Roman literature, something had been done before 1863 when Mommsen took it in hand, but he for the first time made it intelligible^ as a whole. He read a paper on it to the Berlin Academy, which was pubHshed in a separate form, and is now reprinted, with the additional fragment which is chiefly the subject of this paper, in the fijst volume of his Oesammelte Schriften. This fragment was foimd at Rome in 1898 near the Via Portuense, and was first published in the Notizie del Scavi of that year by Vaglieri ; it has since been printed with a short commentary by 0. Hirschfeld in the Wiener Stiidien for 1902, who also inserted it in its proper place in the whole inscription as editor of the volume of Mommsen's works just mentioned. There can be no doubt that it belongs to the Laudatio Turiae. Though it consists of only ten lines, none of which seem to contain more than about three-quarters of the original ones, i. e. the latter part of each line, it seems to fit very naturally into a large gap in the middle of the whole inscription ; but it does not entirely fill this gap, for it does not join on at either end to the text as we have it. It gives us the only letters we possess of the original heading, which can be completed (u)xoeis ; but unluckily the wife's name is not preserved with it. We must therefore wait for further discoveries in order to make absolutely sure of the identity of this wonderful woman. Up to 1898 Mommsen and most scholars have accepted the Digitized by Microsoft® 128 ON THE LAVDATIO TV MAE AND view that the lady was Turia, wife (as we know from Val. Max. 6. 7. 2) of a Q. Lucretius, whose romantic adventures in the proscriptions of the year 43 are recorded by Appian, Bell. Civ. iv. 44. But both Vaglieri and Hirschfeld insist that this new fragment puts that view out of court, on grounds, as I think, by no means convincing. I propose to show in this paper why I think that the new fragment adds to the probability that Mommsen's view was the correct one, and that the lady was actually Tmria, wife of Q. Lucretius Vespillo. In order to explain the place which the new fragment should take in the inscription as we have it, it is necessary to understand that the LauAatio obviously consisted of two parts, roughly answering to the two parts of the surviving fragments, which are divided, as has been said above, by a gap which may have been a considerable one.^ In the first part, which is mutilated at the beginning, the chief topics are the prudence, energy, and unselfishness of the wife in rescuing her father's will from an attack made on its validity by her relations, and the way in which she and the speaker dealt with the patrimonium they thus inherited ; these matters are only interrupted by two paragraphs ^ in which he speaks of the long period of their happy married life, and of his wife's many domestic and other virtues. This digression looks to me as if the speaker thought that he was getting too legally technical, and that the laudatio proper was not suffi- ciently obvious. However this may be, it is, I think, quite clear that in this first part of the document he never really travels beyond the beginning of their married life ; according to an almost certain completion of the text ^ (line 3), the marriage had not taken place when the parents of the wife were suddenly murdered together (perhaps by their own slaves, as Mommsen suggested), and the affair of the will ' The most convenient edition of the whole inscription is now that in Dessau's Inscriptiones Sdectae, vol. ii, pars, ii, pp. 924 fE. " Lines 27-36 in Dessau, p. 925. ' ' Orbata es re{pente ante nuptiar)um diem utroque pa( rente in penatium soli)tudine,una oc{oi3is).' It is hard to see how the first five words can be otherwise completed. Digitized by Microsoft® ITS ADDITIONAL FRAGMENT 129 must have happened soon afterwards, whether when the speaker and his wife were still only betrothed or actually married is uncertain. But before we reach the end of part i, the details of the management of the patrimonium clearly show that the marriage has been completed. Then comes the gap which is imperfectly filled by the new fragment. The second part, before the discovery of the new fragment, began with a mutilated passage which seems to refer to a return from absence or exile, which the husband owed quite as much to the energy and pietas of his wife, as to the clemency of some one in power ; and as the well-preserved succeeding paragraphs tell the story of a wonderful escape, of the vain efforts of the wife to persuade Lepidus to carry out the restitution accorded to her husband by Octavian (Caesar Augustus, as he is called by anticipation), of the brutal conduct of Lepidilp, and the final dementia of Caesar, it has been assumed, and perhaps rightly, that this powerful person was Octavian himself. The lavdatio then proceeds to the happy time of peace after Actium (' pacato or be terrarum, restituta republica '), the want of children, the proposed divorce, and the speaker's horror at the bare idea of it, and the death of the wife ; ending with words which in a religious sense have not obtained the attention they deserve : ' te di Manes tui ut quietam patiantur atque ita tueantur opto.' ^ 1 now give the correct text of the new fragment, from Hirschfeld.2 V XORIS svhsi DIA • rVGAE • MEAB • PEABSTITISTI • 6eNAMBNTIS • CVM • OMNB ■ AVBVM • MABGABITAQTJB • COBPOBI trad IDISTI ■ MIHI • BT • SVBINDB • PAMILIA • NVMMIs. • FK'VCTIBVS ■ 5 a DVEKSABIOEVM • CVSTODIBVS • APSENTIAM • MBAM • LOCVPLBTASTI ITIS • QVOD • VT ■ OONAEEEE • VIBlf S • TVA • TB • HORTABATVB VNIBAT • CLEMBNTIA • EORVM • CONTBA • QVOS • BA • PABABA3 V 6x ■ TVA • EST ■ BIEMiTATE • ANIMI ■ ilMISSA. ETIS • HOMnilBVS • i. ■ MILONB • QVOIVS • DOM^S • EMPTIONE 10 BXV[l] ■ BELLI • CIvIlIS • OCCASIONIBVS • LNEVPTVEVM defe ndistI ■ domvm • nostbam ' See my Rdigious Experience of the Roman People, p. 389. 2 As printed in Dessau, op. oit., the fragment is further oonjecturally completed, but without any gain of certainty. 2252 X Digitized by Microsoft® 130 ON THE LAVDATIO TVRIAE AND From the position in this fragment of the word (U)xoris, the only one we as yet possess of the original heading of the inscription, and obviously the last one, it is clear that Hirsch- feld was right ^ in placing this fragment in the big lacuna between the two main portions of the lavdatio as we have it. But what was the size of the gap between the end of this and the beginning of the next fragment we cannot be sure. The one ends with a fairly clear indication of an attack on a house belonging to the pair, warded off by the wife in her husband's absence ; the other begins with an allusion to a return of the husband from exile or enforced absence. It has been assumed both by Vaglieri and Hirschfeld that they follow close on one another and refer to the same circumstance, viz. the escape of the husband from the proscriptions of 43 b. c. This seems to me to be quite impossible. It has arisen, I think, simply from unconscious prepossession in favour of the story as it was formerly known to us. They refer, I feel sure, to quite different times and events, and after a careful revision of this paper (in 1918) I am still more confident. That Dessau had an inkling of the truth I am also inclined to think ; for at the end of the new fragment he indicates a considerable gap.^ Let us consider this fragment a little more closely : in spite of the loss of a considerable part of each line, its general bearing is pretty clear. Krst, we have a fv^a of the husband ; secondly, at his departure his wife gave him as subsidia all the gold and pearl ornaments she had about her. It is futile to connect this, as Hirschfeld does, with the story of a certain Acilius told by Appian (4. 39), who persuaded the soldiers to whom he was betrayed to take a communication to his wife, on promise of a rich reward: she gave them all her jewels, and they procured his escape to Sicily. In the husband's own account she gave the jewels to himself- — ■' tradidisti mihi ' ■ — which is a very different thing. Next we find her sending him slaves (familia), m.oney, and fructus, in his absence. • ' Mommsen's Gesammdte Schrifien, i. 403. ' ' Perierunt non pauoa.' Digitized by Microsoft® ITS ADDITIONAL FRAGMENT 131 This is quite out of keeping with the hairbreadth escapes of 43, and would have been apt rather to attract attention to the man than to effect his security. As we read through the long list of escapes in Appian, it is clear that it was with the utmost difficulty that the proscribed eluded notice, hiding themselves, often ineffectually, in all sorts of holes and corners ; and of those who reached Sicily safely we are told that they were glad to receive food and clothing at the hands of Sextus Pompeius. I may add that the words ' apsentiam meam locupletasti ' also seem to me ill suited to a time of such imminent peril for the fugitives, when hardly any part of the empire was without its spies and assassins. Again, the words that follow in lines 6, 7, 8, though they are by no means clear in detail, evidently refer to some effort on the part of the wife undertaken on behalf of her husband ; and if this is to be explained of the part she played after he was proscribed, the speaker has told the same story twice over in a most unnatural way, for he immediately proce.eds to tell it again in lines 14 to 20 of the next fragment. If on the other hand we could explain it of some earlier danger and escape, the order of events in the Lavdatio which is in the main presented through all that remains of itj would be sufficiently preserved. But the most effectual proof, as I think, that he is here speaking, not of 43 b. C. but of an earlier time, lies in the mention of Milo in line 9 of the new fragment as if he were alive at the time spoken of. ' Mirum ', says Dessau in a note, ' orationem reverti ad annum ' 706 (48 b. c.) ; but it is not mirum at all, unless you insist on referring the new fragment to 43, with Hirschfeld and Vaglieri. Milo was killed in the spring of 48 B. 0. after being recalled by Caelius Rufus from his exile at Massilia in order .to join him in a mad sedition against Caesar's government and legislation (Caes. Bell. Civ. iil. 20-2). Caesar's own account of this miserable business is unluckily very corrupt, but the story can be made out in outline with the help of Dio Cassius (42. 24). It would seem I 2 Digitized by Microsoft® 132 ON THE LAVDATIO TVRIAE AND that when Caelius was ejected from Rome, he went to Cam- pania and was there joined by Milo, who still had in his pay the remains of gladiatorial bands which he had formerly collected there ; that they made a combined but futile attempt on Capua, and that Milo was then sent south to the region of Thurii ' ad soUicitandos pastores ', while Caelius attacked Casilinum. Or it may be that Milo had left Campania before the attempt to surprise Capua. But in any case it is clear that Milo, as Dio Cassius says,^ gathered a band of desperadoes together, and roamed southwards seeking whom he could devour ; in Bruttium he began to open the ergastula, and met his death in an attempt on Cosa.. It seems to me hardly possible to refer the imperfect lines 9, 10, 11 of the new fragment to any other event than this. The wife ^ is in a country house, as we are entitled to guess, from the fact that she supplied her husband with fmctus as well as with slaves and money. .Milo may have had a grudge against the pair for having bought cheap either this identical villa, or some other house which had formerly belonged to him and was sold cheap after the forfeiture of his property by exile. We know something about such sales from Asconius (in Milonianam, p. 54, ed. Clark), and from Cic. ad AU. v. 5 ; Fam. viii. 8. 3. Milo apparently attacked the house, which was successfully defended by the wife. On this interpretation the whole of the new fragment would refer to the events of 49 and 48 B. C. But if so, it will be asked what was the fuga of line 2, for which the husband received from his wife so much provision in the form of jewels and gold, and during which she supplied him with slaves, money, and fructus ? Let us notice (1) that in line 5 she is evidently represented as having eluded or corrupted ' adversariorum custodes ', and that adversarius is exactly the word which would be used of one side at the opening of * "JEs Tc T^v 'lTa\iav dxpiKfTO^ Kal TroWoiis av6p6jirovs, tovs ju^f Piov Seofituovs, Tovs Si Ko} TipLojpiav Tiva SeSioras, (TuWefas tt/v tc X'^P^^ fKaKovpytt kt\. 42. 24. ' I use the word ' wife ' for convenience : as willJje seen directly, it is not clear whettier the marriage had as yet taken place. The completion, (defe)ndistu is almost certain. Digitized by Microsoft® ITS ADDITIONAL FRAGMENT 133 a civil war, but not of assassins going about to catch and slay the victims of a proscription ; ^ (2) that in lines 4 and 5 of Part I, the husband is spoken of as being in Macedonia soon after the sudden murder of the parents of the wife, while her sister's husband Cluvius had gone to Africa ; (3) that if this fragment refers to the events of 49 and 48, the dementia spoken of in line 7 can hardly be other than that of Julius himself, of whom the word is so often used from the very outset of the civil war. Putting these things together, we may divine, not with certainty, but with great probability, as I think, that the fiiga was nothing more than a flight of the husband from the country house at which they were staying when the war broke out ; if it was the one attacked by Milo in the following year, it would probably be between Campania and Bruttium, and open to Caesar's troops marching in pursuit of Pompey to Brundisium. We may guess that the husband reached Brundisium safely and crossed with Pompey to Macedonia ; the wife remained, and was treated with courtesy by Caesar's orders, after a display of the spirit and courage that was natural to her (' quod ut conarere virtus tua te hortabatur : vox tua est firmitate animi emissa '). This is indeed guesswork ; but it is entirely in keeping with the part of the lines left to us, and inconsistent with nothing that is recorded in the rest of the laudatio. In any case, if it be true that this fragment refers to events having nothing whatever to do with the proscriptions of 43, and can be itself referred with confidence to 49 and 48, we are now in a position to recast our ideas both as to the date of the marriage and the- identity of the pair. On both these points we may now, in my view, safely return to the conclusions of Mommsen in his paper of 1863. As regards the first, Mommsen put the marriage between 48 and 42 B. c. The pair were certainly married at the time of ' It is interesting to find that this word is used no less than four times by Pompeius himself in his dispatches to Domitius preserved in Cic. Att. viii. 12 ; i. e. it is used of the opposite party and its leader in January 49, the very time to which, as I believe, the first lines of the fragment refer. Digitized by Microsoft® 134 0^ THE LAVDATIO TVRIAE AND the proscriptions, which took place in the autumn of 43. At the time of the murder of the parents they were probably not married but only betrothed ; but the condition of the first few lines of Part I, on which this conclusion is chiefly based, is not such as to make it quite certain. If however it is correct, the marriage remained uncelebrated while the future husband was in Macedonia, and a legal defence of the will, as well as the defence of the house alluded to in the new fragment (a house perhaps left them by her father) took place also during the period of betrothal. As Mommsen assumes, on the return of Caesar from the east in the autumn of 47, the affianced husband received a free pardon, like Cicero and so many others ; or possibly this was after the battle of Pharsalia. The marriage would naturally follow, and we should not be far wrong in putting it at the end of 48 or some time in 47. As they were married for 41 years (as he tells us in line 27), this would bring the date of the death of the wife, and of the laudatio itself, to 7 or 6 b. C. Secondly, as regards the identity of the husband, we may return to the hypothesis, recently discarded by Vaglieri and Hirschfeld, that he was that Q. Lucretius Vespillo whose adventures in the proscription of 43 are recorded by Appian and Valerius Maximus ; for what the laudatio tells us of these adventures is not changed, according to my view, by anything in the new fragment. It may be as well to recapitulate the evidence for this identification, especially as the story of Lucretius's escape is incidentally of singular interest. Caesar in Bell. Civ. iii. 7 mentions that on arriving off Oricum from Brundisium he found Lucretius Vespillo and another man in command of eighteen ships from Asia, iJ e. a part of Pompey's fleet ; and this exactly suits the statement of the laudatio that the speaker had gone to Macedonia while his wife's sister's husband, C. Cluvius, had passed to Africa—" the two provinces where operations were being carried on by the Pompeian party in 49-48 b. c. This however is rather a confirmatory point than a matter of substantial evidence. The real argument lies in a comparison of the accounts of Digitized by Microsoft® ITS ADDITIONAL FEAGMENT 135 Appian and Valerius Maximus of the escape in 43, with the hints afforded by the laudatio. What the laudatio tells us is this : ' Why (he says) should I pluck from my inmost thoughts once more the story of my rescue ? how you sent me a sudden message of warning, how you repressed my audacia, and when I yielded to advice, you prepared fida rece{ptacula), with the knowledge only of Cluvius and your sister.' Appian's story is as follows : Lucretius was wandering in the country with two faithful slaves, and being in difficulty for provisions was trying to return to his wife in Rome, and had actually arrived at the gate, when he saw a troop of soldiers coming out. It suddenly occurred to him that this was the very place where his father had been arrested ^ in the SuUan proscription, and he slipped into one of the tombs that there lined the road. One of his slaves had hurt his leg, and he was leaning on the arm of the other when this happened. While they were hiding here they were surprised by some tomb-wreckers (what a picture here of the insecurity of the times !), and to these the slave gave himself up to be stripped while Lucretius fled to the gate — the soldiers having now presumably disappeared. At. the gate, one reads with aston- ishment, he waited for the slave, shared his clothes with him, and reached the house in safety. There his wife hid him between the ceiling and roof of a chamber until the storm had passed over. Valerius Maximus, who gives the name of the wife as Turia, merely tells how he was hidden ' intra cameram et tectum cubiculi ' at the great peril of his wife, who shared the secret with one handmaid only. Now the only contradiction between these combined accounts and the story of the laudatio is in the statement of Valerius Maximus that no one knew of the hiding-place but the maid, while the laiidatio speaks of Cluvius and his wife being in the secret. This however is not a serious difficulty ; • Appian doBS not say that he was killed. If that had been so, the son would have been forty when he himself was proscribed, and could not have been married till he was about thirty-five, which is perhaps unlikely. Digitized by Microsoft® 136 ON THE LAVDATIO TVRIAE AND we may assume that the maid was the only person in the ^OMsewho knew, but that Cluvius and his wife were acquainted with the fact also, as being either in Rome or not far away. In any case Valerius Maximus was careless in regard to detail. Appian's account agrees strikingly with that of the laudatio, if we may assume that Lucretius was making for Rome on the advice of his wife, instead of exposing himself to his enemies in the country districts. She sent him a sudden warning and repressed his rashness, preparing meanwhile a safe hiding-place in their house in Rome. The return to the city was obviously made by night and in disguise ; this is suggested by the mention of the tomb-wreckers, and the changing of the clothes with the slave at the gate ; thus though the peril was undoubtedly great, it was less exactly to be described by the word audacia than the attempt to escape from Italy, which brought sor many to their end. This identification is of course by no means certain, but it may hold the field until another fragment is discovered. No other of Appian's many stories of wonderful escapes tallies in any degree with the laudatio ; and the whole tenor of the document shows that the speaker was a sufficiently important person to have been included in such a collection of stories. If he was Lucretius Vespillo, he held the consulship in 19 B. c. ; and here Hirschfeld has raised the objection that there is no mention of the consulship in the laudatio. But with singular and touching delicacy, the speaker throughout keeps himself in the background, attributing his wealth, his safety, his happiness, entirely to the wonderful woman he celebrates. Once, indeed, when he is speaking of their joint management of their property, he breaks off with the words, ' of this I will say no more, lest I should seem to be claiming a share in your praises ' (Part I, line 40). Could such a man have dreamt of referring to his consulship while recalling the happiness of his domestic life ? Supposing that my reasoning holds good, I would reconstruct the whole astonishing story as follows : Turia's parents were murdered at the very outbreak of the civil war in January 49, Digitized by Microsoft® ITS ADDITIONAL FRAGMENT 137 at a time when we might naturally expect such things to happen. Shortly afterwards Lucretius, then affianced to her, had to leave Italy and act under Pompey in Epirus. Turia, left behind in Italy, with only her sister to help her, whose husband Cluvius had gone to Africa, also to fight on the Pompeian side, had now to face a series of dangers and difficulties, all of which she overcame by her wonderful courage and address ; she traced out the murderers of her parents and secured their punishment ; she obtained the protection of Caesar during his march through Italy to Brundisium ; she contrived to smuggle supplies to the absent Lucretius ; and resisted and finally defeated an attempt to upset her father's will, under which she and Lucretius were the chief if not the only inheritors. The next year, 48 B. 0., during the attempted revolution of Caelius and Milo, she was attacked by the ruffian following of the latter in a villa in the country and contrived to- beat them. off. At the end of that year, or some time in 47, Lucretius returned like Cicero to Italy, and obtained a pardon from Caesar. The marriage was now celebrated, and until Caesar's assassination they presumably lived in tranquillity. When a second triumvirate was formed and the proscrip- tions began, Lucretius's name appeared on the lists, whether at the instance of Octavian or Lepidus is not clear ; the restitution came from Octavian, and the conduct of Lepidus suggests that he had a personal spite against the pair. Then followed the extraordinary escape I have already described, which must have happened at the end of 43 or beginning of 42. For some months Lucretius must have been kept in conceal- ment of some kind, for when at last an edict was obtained for his restitution, Octavian the author of it was absent ; he had gone to the campaign of Philippi, and his departure seems not to have taken place till the summer of 42. Turia took this document to Lepidus, who was consul and in charge of Rome and Italy, and was received, according to her husband's account, with insults and even with blows. The return of Octavian at the end of the year set this matter right ; Digitized by Microsoft® 138 AN UNNOTICED TRAIT IN THE and Lucretius hints that Lepidus's brutality was not forgotten by him. The rest of the story, which is of unique interest as a picture of Roman domestic life, does not properly belong to the subject of this papBr.^ It is to be hopsd that other fragments may be discovered which may help to complete it, and may afford us a more certain identification of the husband and wife ; and this is not impossible if, as Vaglieri thinks, the original site of the inscription was in the locality where this new fragment was found, viz. the Via Portuense on the right bank of the Tiber. AN UNNOTICED TRAIT IN THE CHAR- ACTER OF JULIUS CAESAR Caesar did fewer foolish things than most men with his opportunities have done ; so far as we can judge from his own writings and the accounts of those who knew him, a want of practical wisdom was not one of his weak points. But on one occasion, early in his political life, he did what seems to us a foolish thing, and one which no one has ever attempted to explain as a wise one. I am thinking of the revival of a quaint antique and semi-religious procedure for the con- demnation ^ of Rabirius in 63 b. c. The circumstances are familiar, and have been discussed recently in this country by the late Master of Balliol in his Problems of Roman Criminal Law, and by Dr. Hardy in the Journal of Philology? The leaders of the popular or Marian party, Caesar and Crassus, wished to make it highly unsafe to put Roman citizens to death without trial under the ' last decree ' of the Senate, or in any other way, in times of political excitement. They did not so much want to impugn the legality of that decree ' See Social Life at Borne in the Age of Cicero, pp. 158 £f. ^ The condemnation, because it is quite clear that the duumviri did not judge the case, but only pronounced sentence (Liv. i. 26). ' Strachaa-Davidson, Problems of Criminal Law, i. 188 ff. Hardy in Journal of Philology, xxxiv. 12 S. Digitized by Microsoft® CHARACTER OF JULIUS CAESAR 139 {senatus consultum ultimum), for that would have been almost impossible ; ^ but to make it dangerous for the consul to take violent action under it. They wanted, no doubt, to impress this deeply on the minds of the city population, and Caesar hit upon the plan of reviving a curious and obsolete procedure, which would bring the possible results of such poUtical violence and murder vividly before their eyes. The only example of the use of this procedure known to us dates from the age of the kings, and is embodied in the legend of the victorious Horatius, who slew his sister on his return from battle. For this murder he was not tried, but straightway condemned ; the king appointed two duumviri to perform this duty, being unwUling, Livy says, to undertake such an ill-omened job himself.^ A lex horrendi carminis governed the procedure. The duumviri were to pronounce sentence ; against this sentence the condemned man might appeal to the people ; if their verdict went against him, ' caput obnu- bito, infeUci arbori reste suspendito, verberato vel intra pomerium vel extra pomerium '.^ This procedure belonged to an age when civU law had not yet been fuUy disentangled from religious law. The words last quoted make it probable that Horatius was a homo sacer in some sense, and the sequel to the story shows how difficult it was to restore him to the condition of an ordinary citizen ; for this point, which does not bear directly on our present subject, I may refer to p. 72 of this volume.* What could have induced Caesar to imitate this strange, semi-religious ritual ? Was it simply that it gave him an opportunity to exhibit the infelix arbor, or, as Cicero calls it, the crux, in ' This follows from the acquittal of Opimius, in 121 B. c, for killing C. Gracchus, under the sermtus consiUtum ultimum. See e. g. Heitland's Eoman Republic, ii. 318. Hardy, op. cit., pp. 16 ff. ' Liv. i. 26 ' Rex, ne ipse tarn tristis ingratique ad vulgus iudicii ac secundum indicium supplicii auctor esset, concilio populi advocate, Duum- viros, inquit, qui Horatio perdueUionem iudioent, secundum legem faoio.' I suspect that the Rex appointed duumviri in order that the sacred kingly office might not be polluted. ' Liv. i. 26. 6 : Strachan-Davidson, op. cit., i. 135 ff. » Cf. Journal o/ Roman Studies, i. 58 ft. Digitized by Microsoft® 140 AN UNNOTICED TRAIT IN THE the Campus Martius, with the executioner (carnifex) who was to do the ugly work, unless the victim were acquitted on appeal ? ^ Certain it is that, having found the old Rabirius, who was said to have killed Saturninus in the disturbances of the year 100, Caesar and Crassus, with Labienus as their agent, contrived to pass a law which revived this old pro- cedure ; that Caesar and a relative of the same name were appointed duumviri under it, that they condemned Rabirius, and that on his appeal the Senate interfered and declared the whole foolish proceedings to be invalid.^ (The speech of Cicero, of which we have a considerable part, was deHvered in an ordinary trial before the tribune and his comitia tributa, and with this we are not concerned.) At the moment of condemnation, and before the Senate had quashed the pro- ceedings, it seems possible that crux and carnifex were actually on view in the Campus. Yet the desire to imponieren seems hardly sufficient to explain why a sane man like Caesar should have chosen to go back to such primitive practice. Nothing else that we know of him in that year 63 shows any parallel to such injudicious statesmanship. I think that there are traces in Caesar of a tendency, common at the time, to take an interest in ancient procedure, especially that of religion ; and it is possible that for once he may have been tempted to give this intellectual interest a practical application. It would be interesting if we could discover whether he was already pontifex maximus when he condemned Rabirius ; but it does not seem possible to determine this. It is, however, in any case likely that his thoughts were running on the probable vacancy, and the duties of the office, for which he and his mother seem to have been equally desirous.^ It is worth remembering that Varro dedicated his great work on the religious antiquities of Rome to this pontifex ' Cicero seems to imply this in pro Rahirio, sees. 10, 11, 16. So, too, Strachan-Davidson, i. 197. ' So Hardy, op. oit., p. 28. Strachan-Davidson thinks that Cicero interfered, either as consul or through the agency of a tribune. ' ' Cum mane ad comitia descenderet, praedixisse matri osoulanti fertur, domum se nisi pontificem non reversurum ' (Suet. lul. 13). Digitized by Microsoft® CHARACTER OF JULIUS CAESAR 141 ma^imus, which he would hardly have done if Caesar had shown no interest in such things. It is also worth remembering that, as a boy, Caesar had been, presumably by his family, intended to fill the most ancient of Roman priesthoods, and that a Flamen Dialis was daily and hourly engaged in caeri- monia. I have elsewhere casually suggested that, as in the case of C. Valerius Flaccus, recorded by Livy as happening about a century earlier, the object of the family may have been to keep the lad out of mischief.^ On the other hand, it is possible that Caesar's mother Aurelia, who seems to have shared with him his ambition to be the head of the Roman religious system, may have been one of those good ladies who venerate all forms of priesthood, and are ready to dedicate their sons at an early age to the lifelong service of the religion of the State. Undoubtedly these things were arranged within the family in collusion with the pontifex maximus, as in the case of the Vestals ; and it is noticeable that, according to Suetonius, the Flamen Dialis destinatus was immediately provided with a wife, young as he was, doubtless because the office could only be held by one who had a Flaminica ready to assist him in his duties.^ Incidentally, I may remark that it was this wife Cornelia who saved Caesar for the world. As she was Cinna's daughter, Sulla ordered the boy to give her up, which he promptly refused to do, and was at once deprived of his priesthood (or, rather, the prospect of it), and of his wife's dowry and other property. Suetonius evidently thinks of the priesthood as an honour which Caesar would have been glad to retain ; but even if Aurelia looked at it in this light, it does not follow that at the early age of fifteen the boy was not glad to be safe from the shackles of such an office. A few years after this (69 B. c), when delivering an oration at the funeral of his aunt Julia, he dwelt on the fact that she was descended, on the mother's side, from a rex, Ancus > Rdigious Experience oj the Roman People, p. 343. According to Velleius ii. 43, Caesar had been made a pontifex during his absence in Asia as a young man, and hurried home to Italy to take up the of&ce, which suggests that he was in earnest about these priesthoods. ' Suet. IiU. i, ad init. Frazer, Adonis, &c., p. 409. Digitized by Microsoft® 142 AN UNNOTICED TRAIT IN THE Marcius, and on the father's side from Venus, the reputed ancestor of the gens Julia. His comment on this, as quoted by Suetonius from the original, is remarkable : ' Est ergo in genere et sanctitas regum, qui plurimum inter homines poUent, et caerimonia deorum quorum ipsi in potestate sunt reges.' ^ The use of the word caerimonia here is peculiar : it seems to mean that the Julii had an inherited instinct for looking after the cult of the gods — an instinct which, perhaps, the devoted Aurelia discerned in her son.^ The whole sentence, a good specimen of the Attic style, breathes the young man's feeling that the Roman State cannot dispense with its gods, and that caerimonia is necessary in order to keep them in full vigour of benevolence. On the whole, then, I think it quite possible that in the imprudent revival of the obsolete procedure of the duumviri Caesar may have been prompted by this instinct for caeri- monia ; or, if I may be allowed a little latitude of conjecture, I should guess that Aurelia suggested to him a course which he was not unwilling to take. She was at this time living in his house, and her anxiety about caerimonia is well illustrated in the afiair of the Bona Dea in the following year, when (as Plutarch tells us) she took all possible pains to prevent any disturbance of the rites. When Clodius was discovered she put an end to them at once : evidently she had the lead in the house at the time.* I have little doubt that the divorce which followed was also the work of this strong-minded mother. But apart from these indirect inferences we have a story, which seems quite worthy of credence, that many years afterwards Caesar again had in his mind a piece of antique ritual, when he punished two mutinous soldiers in 46 B. o. Though the story is told only by Dio Cassius,* it cannot have • Suet. lul. 6. ' It is worth remembering that the Julii were charged with the care of the cult of Veiovia at Bovillae. 0. 1. L. i. 807, and Wissowa, Bel. und Kult. der Homer, ed. 2, p. 237. ' Plutarch, Caesar, ch. 9, is very explicit about this. Whence did he get his information about Caesar's private life ? " D. 0. xliii. 24. Digitized by Microsoft® CHARACTER OF JULIUS CAESAR 143 been invented by him. These two men were put to death, the historian says, ' in a sort of priestly fashion '. 'I cannot explain it ', he goes on, ' for no Sibylline verse or other oracle is quoted for it ; but the fact is that they were sacrificed {(Tvdricrav) in the Campus Martius in the presence of the pontifices and the Flamen Martialis, and their heads were afterwards fixed up on the Regia.' It has long been recognized that this strange and barbarous procedure closely resembles that of the sacrifice of a horse to Mars on the Ides of October, which I fully discussed in my Roman Festivals, p. 241. After a chariot-race in the Campus Martius, the near horse of the winning pair was sacrificed to Mars ; the tail was cut off and carried to the Regia (the official residence of the pontifex maximus), and the warm blood allowed to drip on the hearth there. The head was also cut off and decorated with cakes ; and formerly there was a fight for it between the men of the Via Sacra and those of the Subura. If the former carried off the prize, they fixed it on the wall of the Regia ; if the latter, on the Turris Mamilia. Caesar himself took part, according to Dio Cassius, in the quelling of this mutiny, which was especially dangerous as taking place at Rome, where his position was not too secure. Dio puts it in the year 46 ; and in that year Caesar returned to Rome from the African war on July 26, and stayed there till after November 26, when we hear of him in a letter of Cicero.^ He was thus beyond doubt in the city, as Dio states, at the time of the mutiny, and I am much inclined to suspect that the sacrifice of the mutineers took place on the Ides of October, and concurrently with that of the horse, or as a substitute for it. The motive was perhaps much the same as in the case of Rabirius, to make an impression on the city mob, who might easily be infected with the spirit of mutiny. We may doubt whether in either case the desired effect was produced. I do not think that Caesar was ever at home in • Cf. de Bell. Afr. 98, and Cic. ad Fam. vi. 14 ; which letter is dated A. D. 5 Kal. intercalares priores (two intercalary months were that year inserted between November and December). Digitized by Microsoft® 144 AN UNNOTICED TRAIT IN THE the city, or understood its motley population ; as I have said elsewhere,^ they knew little of him, and had received no great benefits from him. Augustus understood them far better, and made no such strange attempts to frighten them into acquiescence. There is yet another curious story of Caesar, which may be set by the side of these two examples of perverted caerimonia. It was said that when he crossed the Rubicon he ' conse- crated ' a number of horses, and set them free to wander where they would. Suetonius tells us that in the days pre- ceding his assassination these horses persistently refused to eat, and even shed abundant tears ; he seems in this chapter to be depending on the authority of Cornelius Balbus, Caesar's intimate friend and secretary.^ The legend of their refusal to eat seems to be alluded to in the fifth Eclogue of Virgil, and was adduced by H . Nettleship as evidence for the identifica- tion of Daphnis in that poem with Caesar : ^ non ulli pastes illis egere diebus frigida, Daphni, boves ad flumina ; nulla neque amnem libavit quadrupes nee graminis attigit herbam. But why did Caesar release these horses, and to what god did he consecrate them ? Is this, after all, only one of the legends which gathered round a famous event ? It may be so ; but, on the other hand, it is not a common form of marvel, but looks rather as if it had a bottom of truth, and we may note that Asinius Pollio was with him at the time, who after- wards wrote a history of the Civil Wars.* If the horses were consecrated at all they were consecrated to Mars, for whom he had a special regard, and to whom war-hojrses were sacred.^ But it is difficult to fathom his motive, or even to be sure of the fact. In the last place, let us note that in his somewhat elaborate ' Soman Ideas of Deity, ■p. 118. ^ Suet. lul. 81. ' Ancient Lives of Virgil, p. 40. " Plutarch, Caesar, 32. * Roman Festivals, p. 330. Caesar seems to have been fond of horses, and rode one of which Suetonius tells strange things (Jvl. 61), and which would allo\g no one to mount him but Caesar. He afterwards placed a statue of this horse in front of his temple of Venus Genetrix. Digitized by Microsoft® CHARACTER OP JULIUS CAESAR 145 account of the civilization of the Gauls great prominence is given to religion {natio admodum dedita religionibua), and especially to the Druids and the details of their human sacrifices, as well as to their gods.^ Even now this account forms a considerable part of what we know about early Celtic religion. There is then, I think, some reason to believe that Caesar, among his many various interests, included the caerimonia of deities at Rome and also elsewhere ; and that once or twice in his life he translated his interest and knowledge somewh^ strangely into practical procedure. This does not mean, of course, that he was in any sense ' superstitious ' : what interested him was the ritual of State or tribe. He may have had his trifling superstitions. Pliny tells us^ that after a certain carriage accident he always used to repeat a sort of spell three times when he took his seat — but this he may have done just as I take off my hat to a magpie. As we might expect, we have it on good authority that he never allowed a religio to alarm or delay him in any undertaking about which he had made up his mind : when in his African campaign the victim fled from the sacrificing priest, he went none the less determinedly to meet his enemy.* But of serious ritual he thought without contempt, and the careful pains which Augustus bestowed on this department of State activity may, after all, though we have not been used to think so, be due in some measure to his uncle's precepts. As Dr. Hardy reminds me, the uncle was as careful in the constitution of his colonies of the maintenance of caerimonia as Augustus himself could have been ; of this his lex Ursonensis gives abundant proof. • De Bdl. Gall. vi. 13-19. " Nat. Hist, xxviii. 21. ^ Suet. Ivl. 59 ' Ne religione qmdem idla a quoquam incepto absterritus unquam vel retardatus est.' 2252 Digitized by Microsoft® 146 ANCIENT ITALY AND MODERN BORNEO : ANCIENT ITALY AND MODERN BORNEO: A STUDY IN COMPARATIVE CULTURE ^ The valuable work of Messrs. Hose and McDougall on the Pagan Tribes of Borneo was published in 1912, and contains an account of the methods of divination practised by some of these peoples. In compiling it the authors had been led to consult Smith's Dictionary of Classical Antiquities, so striking did the parallelism appear between the augural practices of Borneo and those of ancient Italy. And they were not mistaken ; the parallelism is even stronger than they suspected, and induced me to write a notice of the book in the Journal of Roman Studies, in order to call the attention of students to the subject.^ Since then I have again gone carefully through the work, and noted a nurnber of other points in which the habits of the one people remind me of the other. It seems worth while to bring these together. But before doing this, I' must profess myself both unable and unwilling to theorize on comparative evidence of this kind. These singular analogies may arise naturally from the fact that those tribes of Borneo are now in much the same state of culture as were the earliest Latin settlers and their ancestors of the pile-dwellings and terremare of northern Italy .^ It is primarily on these earlier stages of Italian lite, ' Eead to the Oxford Anthropological Society, April 1915. 2 J. 11.8. ii (1912), 269 f. ^ For example, the Kayana, &c., live chiefly by agriculture, but they also keep domestic animals, particularly the pig (the favourite sacrifice), and they hunt mid animals, pigs, deer, &c. See Hose and McDougall, chs. vi and ix. That the peoples of the pile-dweUings and terremare were in much the same condition as regards their food is proved by the remains of it which have been found. Seeds of cultivated plants have been discovered among the earliest of these settlements, and the people seem to have become more agricultural as the settlements became more permanent. They had domesticated the pig, but continued to hunt it in its wild form. In the terremare we find evidence of distinct advance on the same lines : the people practised agriculture more elaborately, but had not ceased to be hunters. Peet, Stone and Bronze Age in Italy, chs. xiii and xiv ; Modestow, Introduction a I'histoire romaine, ch. iv. Digitized by Microsoft® A STUDY IN COMPARATIVE CULTURE 147 incidentally on the more advanced civilization of early Latium, that the Kayans and Kenyahs of Borneo may possibly throw some light. Into the question of race I am quite incompetent to enter ; but two points may be here noticed as bearing on the problem : (1) there is now practically no doubt that the pile-dwellers and terremare folk of northern Italy came into that country from the north, and mainly from the vaUey of the Danube/ to which they may probably have found their way from central Asia ; (2) the Kayans are probably in part of Caucasian blood, and are believed to have found their way to Borneo from central Asia either by way of Burmah or Bengal.^ Thus it is not impossible that, as is suggested by the authors (ii. 265), some of the analogous customs, e. g. that of divination, may have had a common root. The Kayans and Kenyahs live in what may roughly be called village communities, on the banks of rivers, and usually, like the Etruscans, at points where the river forms an angle or where a tributary joins it. These village communities consist of one or more house communities, each containing some forty or fifty families. The house is practically a pile- dwelling, built not in the water, like most of those of northern Italy before the age of terremare, but not more than twenty or thirty yards above it. The shape of the house is rectangular, but its length is much greater than its breadth ; it is placed upon piles driven into, the ground, and at a height sufficient for security against enemies. The whole length of the house is about 200 yards, and the breadth from 40 to 60 feet : in front a gallery for common use runs along the whole length, with doors opening into rooms behind appropriated to ihdi- vidual families.^ After reading the description of these curious ' long houses ' in the third and fourth chapters of Messrs. Hose and McDougall's ' Peet, op. cit., pp. 505 f. 2 H. and MoD. ii. 233. On the next page the authors state their reasons for believing that the Kayans have passed on their culture to other tribes. = ibid. i. 39 and 51. K 2 Digitized by Microsoft® 148 ANCIENT ITALY AND MODERN BORNEO : work, it is interesting to turn to Mr. Peet's account of the pile-dwellings on the Italian lakes.^ There we find evidence, slight indeed but complete so far as it goes, that here too the house was a long rectangular one, which must have contained many families (if families were there, as I suppose we may assume) : but the breadth was greater in proportion than in Borneo. It would seem too that the house was not invariably built over the water, but sometimes in the peat which formed its shore ; ^ and the facts seem to point to an advance from the lake-dwelling properly so called to the terramara or fortified position at some distance from the lake, though- rarely far removed from water. Such a change also indicates an advance in civilization, and security, and increased confidence in the good faith of neighbours ; and it is perhaps here that we should place the first glimmerings of that famous ' ius fetiale ', common to the descendants of the terremare folk,^ a most remarkable result of combined human interest and religious experience. I 'cannot, of course, prove this ; but there are two facts, of very different kinds, which make the guess less hazardous. One is that in much the same stage of civilization some Bornean chiefs had discovered, even before the arrival of direct European influence, that war and rapine, indulged in without warning given, in other words, without something in the nature of a ' ius fetiale ', are not desirable.* The other fact is a curious archaeological one. The spear thrown into hostile country by the ' pater patratus ' under the fetial law was a survival from an age when iron was not in common use ; Livy describes it (i. 32) as having a point either of iron or of wood hardened and' sharpened in the fire and smeared with blood ; i. e. iron was a late substitute for the old practice. Now in the period of the terremare iron had not yet made its appearance, and • Peet, op. cit., ch. xiii. For the shape of the house (80 yards by 30) see pp. 291 f. ' ibid. p. 291. " Wiasowa, Rd. und Kult. ier Romer, ed. 2, p. 550 ; Marquardt, Staats- verwaltung, iii. 419. * H. and JIcD. ii. 205-6. The poUoy was started among the Kenyahs before they were incorporated in the Kaj of Sarawak. Digitized by Microsoft® A STUDY IN COMPARATIVE CULTURE 149 wood was sometimes used instead of bone or horn. Wood was occasionally "hardened by burning in the age of the pile- dwellings : and the same practice is found in Borneo ; ' for the defence of a house short sharp stakes of split bamboo are thrust slantingly into the ground,' so as to present the fire-hardened tip towards the feet of the coming foe '} Of the government of the old Italian long houses we of course know nothing ; but it is none the less interesting to examine that of the more advanced tribes in Borneo. Each house has a chief and, if there is more than one house in a village, the village has a chief ; but the most instructive fact is that the chiefs and their families form a social class like the patricians at the head of the community.^ The members of this class rarely marry outside it, and special attention is paid to their methods of marriage, a fact which reminds us of the patrician confarreatio. Cohabitation of men of the upper class and women of that below is not unknown, but the men wUl eventually marry into their own class. The members of this upper class can be distinguished by the superiority of their personal appearance as well as by their dress ; but the origin of the class and its chieftainships does not seem to be known. They do not appear to be of different blood from the rest,* though some of them claim divine descent, nor do they seem to have attained their position by skUl in ' ftiagic ' ; there is nothing here to support Frazer's theory of the origin of kingship. The government of the chiefs has the three characteristic features of the old Roman king- ship, i. e. they are leaders in war, in religion, and in the settle- ment of disputes and the punishment of the guilty. The ' See Peet, op. oit. 355, 297 ; H. and MoD. i. 161. For the practice o£ hardening wood in the fire see my Roman Festivals, p. 203, and Skutsch in Classical Quarterly, 1910, p. 270. I find that it is familiar to anthropologists, and I have been shown spears thus hardened in the Pitt Rivers Museum at Oxford. For the smearing with blood. Professor Reid sends me an interesting parallel in Ammianus, xix. 2, 61. ^ There are three classes in all : the lowest consists of slaves, mainly belonging to the upper class, and the middle class of all other members. See H. and McD. i. 68 f. ^ But see ii. 10. Digitized by Microsoft® 150 ANCIENT ITALY AND MODERN BORNEO : moderation and good sense with which all this work is done, and the comparative absence of savagery and magic, suggest that the tendency we sometimes meet with to interpret Greek and Roman practices by reference to those of such primitive populations as the Australian aborigines, is apt to be misleading, and that there is need of more careful com- parative study of peoples in a more advanced social condition. I may note in passing one or two other characteristics of these chiefs, which will further help us in fqrming an idea of the intellectual condition of the ancestors of the Latins. They have a natural gift for oratory, and frequently use it in their deliberations : and they preserve the traditions of the history of their tribe, for at least several generations back.^ How far they preserve them accurately we cannot tell ; but the fact is worth noting just now, when there are signs that we are beginning to criticize the destructive criticism of the last century or so, which brushed away aU old Italian tradition as worthless.^ It is as well to reflect that the members of the upper class, from which the chiefs were drawn, were both in Borneo and in Italy educated by their own. experience and that of their ancestors ; they had religious, military and judicial duties to perform, and in performing them must have learnt much. There is no sign in either country that they were intellectually petrified, merely the creatures of collective habit ; there was room for individuality to ripen and fructify. As Professor Conway has lately said, the ' unknown Roman statesman who shaped the Roman con- stitution ' is surely the great Valerius Publicola, if we are to treat tradition with any sort of respect. The reader of The Pagan Tribes of Borneo, and especially its twentieth chapter, will find such a chief as Publicola in Tama Bulan of the " H. and MoD. i. 63 and 68. Other points relating to the chiefs and their class will be mentioned below. A photograph of a chief haranguing his followers will be found opposite i. 70. ^ See e. g. Professor Conway in Classical Review, 1914, p. 275. Of course in ancient Italy the great difficulty is to distinguish what is Greek in origin from genuine Italian tradition, and especially in matters of religion. In civil history and law we are rather better able to see through the mist. Digitized by Microsoft® A STUDY IN COMPARATIVE CULTURE 151 Kenyahs, not to speak of others. "^ In writing of peoples of this stage of civilization it is preposterous to talk of ' repre- sentations collectives ' or ' mentalite prflogique ' ? At every turn we have the clearest evidence of individual character and its value. The functions of the chiefs in three departments, war, jurisdiction (if I may use the word), and religion, may suggest a convenient division of the rest of my subject-matter. Whatever is done of importance in these departments is done by the chiefs and members of their families, i. e. the upper class, as by the patricians in earliest Rome, who were doubt- less the descendants of an upper class of chiefs and their families in the age of the pile-dwellings and the ferrewjare. About the customs of war there is not much to say, but it is interesting to find that enlightened chiefs are gradually overcoming the savagery of former years, and initiating a pacific policy destined to put an end to sudden raids, and so to set an example to modern Europe.* Nothing definite, however, in the nature of a ' ius fetiale ' seems as yet to have been noticed. Except in the matter of head-hunting, which seems not to have been an ancient practice of this people, and is now dying out, war is not accompanied by special cruelties among the higher tribes. Prisoners are eagerly taken by the Kayans, and become slaves, but are kindly treated, and as in ancient Latium, often form a part of the family Hfe.* It is curious to find the classical ' testudo ' in use among the Kayans. ' If a strong party determined to attack a house in face of an alert defence, they may attempt to storm it in broad daylight by forming several compact bodies of about twenty-five men. Each body protects itself with a roof of ' Conway, loo. oit. ; H. and MoD. index, s. v. Tama Bulan. ^ Individuality is everywhere apparent in Hose and McDougaU's book. Even dancing, which is so often represented nowadays as the result of ' collective mental action ' , is often performed by a single person (e. g. ii. 157). For the ' mentalite prelogique ', &c., to which allusion is made above, see e. g. Levy-Bruhl, Lzs Fonctions mentcUes dans le's socUtes primitives, eh. iv. 3 H. and McD. ii. 205. ' ibid. i. 184. For head-hunting, see end of oh. x. Digitized by Microsoft® 152 ANCIENT ITALY AND MODERN BORNEO : shields held closely together, and the several parties move quickly in upon the house simultaneously from different points, and attempt to carry it by assault.' ^ Compare the ' testudines ' as sculptured on the column of Marcus Aurelius, in Schreiber's Atlas of Classical Antiquities, plate 44, fig. 8. There is evidently a certain kind of sacredhess (for want of a better word) about warfare among the Bornean tribes, especially at its commencement : the warriors are subject to strict taboos throughout the raid or campaign.^ Two of these show that curious instinct for avoiding new things and adhering to old ones, which is familiar to aU students of old Italian religion ; they must use only their home-made earthen pots, and- fire must only be made by friction.^ They were clearly thought to be in a ' holy ' or dangerous state, a belief which might easily be illustrated from other parts of the world, and which has been fully dealt with by Sir James Frazer in The Golden Bought But the most cmious ceremony is that which the men undergo who are to take part in the actual attack. Some distance from the vUlage to be attacked .' two solid platforms are built about twenty feet apart, and a large beam is laid from one to the other. The chiefs and principal men take their seats on the platforms, and then • every man of the party in turn approaches this beam, the fighting leader coming first. If he is willing to go through with the business, he slashes a chip from the beam with his parang and. passes under it. On the far side of the beam stands a chief holding a large frond of fern, and as each man passes under, he gives him a bit of the leaf.'^ The magic properties of fern are well known,^ and here probably the idea is that some special protection would be gained by. carrying it about the person. But the passage under the beam also, in all probabihty, contributed to this result ; here the last 1 H. and McD. i. 181. = ibid. i. 170 f. " See e. g. Henzen, Acta Frairum Arvalium, pp. 26 and 27 ; Wissowa, Bd. und Kult. der Homer (ed. 2), p. 160. For the making of fire by friction 'at Rome of. Festus (p. 94, ed. Lindsay). • See a. B.^pt. ii (Taboo, &o.), pp. 157 f. ' H. and McD. i. 171. « G. B. ii. 66 f. Digitized by Microsoft® A STUDY IN COMPARATIVE CULTURE 163 edition of The Golden Bough is again helpful in convincing us that the passage through an aperture or under an arch or beam enables a man or a host to get rid of evU influences.^ In another paper I have discussed this question in connexion with the passing of a conquered army under the yoke in ancient Italy, and of the return of an army into the city through* the Porta triumphalis.^ We do not know for certain that a Roman army setting out for a war had to pass under any kind of arch ; but we do know that it had to be ' lustrated ' on its return in the Campus Martius ; and it is possible that the so-called temple of lanus, which was really a double archway always standing open during war-time, may have been a survival from a time when the- host actually passed through an arch on its way to a campaign.* Turning to the administration of justice, we find that custom, sanctioned by religion, and collective responsibility, are the chief agents among the Kayans in enabling the chiefs to perform this part of their duties without much difficulty.* There is ample opportunity for the tact and ability of an individual chief like Tama Bulan, but the real foundation of justice is in the community itself. The following sentences » G. B., pt. vii, vol. ii, pp. 189 f. 2 See above, p. 70 fE. ; of. G. B., pt. vii, vol. ii, pp. 193 f. ^ See Domaszewski, Ahhandlungen, p. 222 : he seems to guess that the army originally passed out under this arch, and later on, after the building of the ' Servian ' wall, by the Porta Carmentalis (of. the story of the Fabii, Ov. Fasti, ii. 201 ; Liv. ii. 49 ' infelioi uia, dextro iano portae Carmentalis, profeoti ad Cremeram flumen perveniunt '). Dom. as usual gets out of his depth here, but his main suggestion is worth attending to. This porta had two iani or passages (Wissowa, E. K., ed. 2, p. 104), and so had the lanus Quirinus or lanus bifrons. Is it possible that this feature had some relation to the exit and return of ah army ? Note that aooording to Wissowa this is the real meaning of the term lanus geminus, the double head being now proved to be later. I find an interesting fact from India bearing on this matter : ' In northern India it is a common charm to drive the cattle imder a rope fixed over the village cattle-path, and among the Dravidians of Mirzapur two poles and a crossbar are fixed at the entrance of the village with the same object, i. e. to protect them against disease ', &o. (Crooke, Folklore of Northern India, ii. 299). « H. and McD. at the beginning of oh. xx. The whole of this chapter is of very peculiar interest. Digitized by Microsoft® 154 ANCIENT ITALY AND MODERN BORNEO : may help us to realize the conditions of social life in a Latin community like early Rome, where the basic principle of justice lay beyond all doubt in the minds of the people, that is, of the upper class, known to us in Rome as patricians. ' The principle of collective or communal responsibihty of the household ' (i. e. of the ' long house '), ' which is thus recognized in face of the spiritual powers, as well as in face of other communities, gives every man an interest in the good behaviour of his fellows, and at the same time develops in him the sense of obligation towards his community. The small size of each community, its separation and clear demarca- tion by its residence under a single roof, its subordination to a single chief, and its perpetual rivalry and conflict with other neighbouring communities of similar constitution, all these circumstances also make strongly for the develop- ment in its members of a strong collective consciousness, that is to say, of a clear consciousness of the community and of his place within it, and a strong sentiment of attach- ment to it.' ^ The authors add that a member can hardly leave the community even if he would : he would be an outcast, without helm or sail, and he would find it almost impossible to gain admittance to any other community.^ All these facts explain ' how smoothly the internal life of the community generally runs, how few serious offences are committed, how few are the quarrels, or the instances of insubordination towards the chief, and how tact and good sense can rule the house without inflicting any other punish- ment than fines and compensatory payments '. And the same class of facts, if we think of them as existing in early'Latium, will help to explain the nature of the authority of the pater- familias, and of that higher form of the same broadly based authority, conferred by the patrician body, which we know as that of the Rex. There are, however, occasionally serious crimes, breaches ' H. andMcD., pp. 194 f. '^ In early Italy this was not so, as Professor Reid reminds me ; the provisions for a ius exilii were already elaborate. Digitized by Microsoft® A STUDY IN COMPARATIVE CULTURE 155 of taboo, the most dreaded of which is incest, and .these are held to bring grave peril to the house, and danger to the crop of padi} The same feeling can be recognized in early Rome in the belief that crime disturbed or violated the ' pax deorum ', and that the deity concerned would punish the community unless some expiatory step were taken to re- establish the right relation between the human and divine inhabitants qf the city.^ At Rome this was done by making over the criminal, the man who had broken taboo, as ' saeer ' (cursed and consecrated) to the deity whose special rights he had violated, with the result that any one might put him to death who was so disposed.* At this point we find a remark- able similarity of feeUng among the two peoples with whom we are concerned, in the dislike of shedding blood even in these cases of punishment for sinful crime. I have noticed this in my paper on the original meaning of the word ' sacer ' : the harvest thief is hung ; the man who had suffered ' con- secratio capitis et bonorum ' might be thrown from the Tarpeian rock ; the guilty vestal was buried alive ; and the ' parricida ' was sewn up in a sack and east into the water. For incest in Borneo the usual punishment is, strangely enough, also to shut up the offenders in a wicker cage and to throw them into the river.* This is a substitute for an older punishment which involved the shedding of blood, for tbis was felt to be the blood of the community. It may be that the strength of communal feeUng, which was extremely strong in ancient Italy, may there also help to explain the intense dislike of shedding human blood, either in punishment or sacrifice or in the two combined. The idea that the blood of the individual is the blood of the community has been long familiar to us.® In Borneo even the process of cupping, which involves some loss of blood, has to be atoned for by an offering made to the patient by the surgeon. Kayans have no scruple in shedding the blood of their enemies, but 1 H. and McD. ii. 196. 2 Bdigious Experience of the Romans, pp. 272 f. 3 See J. B. S. i (1911), 59. - H. and McD. ii. 196. ^ See e. g. Robertson Smith, Religion of the Sermtes, pp. 254 £. Digitized by Microsoft® 156 ANCIENT ITALY AND MODERN BORNEO : they rarely go to war with other Kayans : and the shedding of Kayan blood by Kayans is of rare occurrence. To shed human blood, even that of an enemy, in the house, is against custom.^ I now turn to the subject of religion proper, where our comparison is in some points of remarkable interest. In writing of the religion of the early Romans I have insisted that their beliefs must be considered as essentially animistic ; ^ andthat, though showing traces of the survival of ideas which Dr. Marett has described as animatistio, they were on their way to develop something in the nature of deity out of their world of spiritual agents. The Romans in fact were fairly well advanced in their conception of the supernatural. The same is the case with the Kayans and Kenyahs of Borneo. ' They may be said to attribute a soul or spirit to almost every natural agent and to all liviag things, and they pay special regard to those that seem most capable of affecting their welfare for good or evU. They feel themselves to be surrounded on all sides by spiritual powers, which appear to them to be concentrated in those objects to which their attention is directed by practical needs ... we may say that they have differentiated from a continuum of spiritual powers a number of spiritual agents with very different degrees of definiteness. Of these the less important are very vaguely conceived, but are regarded as being able to bring harm to men, who must therefore avoid giving offence to them, if they should by ill-chance have been offended. The more important, assuming individualized and anthropomorphic forms and definite functions, receive proper names, are in some cases represented by rude images, and become the recipients of prayer and sacrifice.' ' This language might be used with little alteration of the earliest Roman religion as I have described it in my sixth Gifiord lecture ; and that description may gain some force from the comparison with another people not so far behind the ' H. and McD. ii. 199. « ' See Edigious Experience, of the Romans, ch. vi. ' H. and MoD. ii. 2, note. Digitized by Microsoft® A STUDY IN COMPARATIVE CULTURE 157 early Latins in general culture. Life, soul, power, animation, or whatever we choose to call it, is the essential characteristic of the spirits of the Kayans and Kenyahs, as it was of the old Roman ' numina ' : ^ and inanimate objects were not supposed to be endowed with such powers by either people. There is no true worship of stocks and stones. Before every Kayan house there stand wooden posts, very roughly carved to indicate the head and limbs of a human form. ' But the post cannot be called an idol : it is more of the nature of an altar.' Perhaps the faint traces that we find at Rome of a tendency to make images of a deity, e. g. Pales,^ should be explained in this way : as I wrote of them some years ago, ' it is hardly safe to take them as genuine examples of iconic worship '. Or the stone of Terminus may be compared with these curious objects (survivals from a preanimistio age, or an age of animatism, as Dr. Marett calls it) ; whether any spirit or ' numen ' was thought actually to be resident in the stone, may be doubtful, yet that stone was also in some sense an altar, as is shown by the fact that when first set up as a boundary-mark it was sprinkled with blood and adorned with garlands.^ In one point, which is of importance in estimating the process of the development of deity out of a vague world of spirits, the religion of the Kayans and that of the Romans are strongly contrasted. The former have no priesthood ; not even in their chiefs do we see the characteristics of the true priest. Neither have they permanent temples, which are usually found in conjunction with a priesthood. The fact seems to be that priesthood arises where there is a fixed and permanent settlement on the land, and where a ' numen ' or deity is caught and isolated on a particular spot, which becomes his own property, and so needs an agent or official.* ' H. and McD. ii. 3; Bd. Exp.-, p. 119 ; Roman Ideas of Deity, p. 27, &c. " Bel. Exp., pp. 147 and 165, note 7, and Tibullus, ii. 5, 27. ' See my Roman Festivals, pp. 325-6. * H. and MoD. ii. 74; Rel. Exp., p. 123. See also A. Lang, GifEord Lectures, The Making of Religion, p. 284 ; Robertson Smith, Bel. of the Semites, pp. 104-5. Digitized by Microsoft® 158 ANCIENT ITALY AND MODERN BORNEO : The Romans and their Latin cousins were beginning to be definitely settled when we first catch a vague glimpse of them : Latium had at least some communities that we may call cities, of which the sites were never afterwards changed, nor the habitations of their deities. But the Borneo tribes are more in the condition of the pile-dwellers of the Italian lakes ; their houses and villages, though often unchanged for long periods, are not as yet bound to the soil by any passionate regard for it. This is probably the reason why priest and temple are not found among them in any true sense of the words ; and this makes it seem strange that they should have got as far as they have in the direction of anthropo- morphic deity. But this may be the result of contact with other peoples. Though the process of god-development is of some interest tt) us, it is hardly so with the gods thus developed. There is one, Laki Tenangan, who seems to be in some sort a supreme and intertribal deity ; but little is known of him, and -that little does not show any real analogy with Zeus or lupiter.^ He is not a god of the sky, nor of the thunder ; but his title Laki seems to answer to the Graeco-Roman use of the word ' father ', as applied to deities, i. e. his power can be used benevolently, and he is not unwilling so to use it for mankind.^ So with the other gods of the Kayans and Kenyahs, who can be approached through the birds of omen, as I wUl explain directly, or by way of messages or prayers sent through the channel of sacrificed animals. On the other hand, the great mass of spiritual beings, the Toh, are thought of as hostile to mankind. They are wUd and have not been reclaimed : but ' those of the locality in which a man dwells are regaled by him as less dangerous than those of other parts : for experience has shown him that in the neighbourhood of his own village he may behave in certain ways with impiinity, whereas in distant regions all ' H. and McD. ii. 6 and 10 f. ^ ibid., ii.»6, wtere this is proved by reference to the sacrifice and prayer ofiered to this god. Digitized by Microsoft® A STUDY IN COMPARATIVE CULTURE 159 is uncertain '.^ This is simply the feeling of ' religio ', gradually soothed away in settled habitations, but recurring as they are left farther and farther behind, e. g. by an army, as I have described it more than once in my Gifford lectures.^ Nothing can show more conclusively the importance of permanent settlement in the mental development of mankind. This is illustrated too in one detail which reminds us of the early Latin settler. In clearing a patch of jungle for sowing padi, a few trees are usually left standing on some high point of the ground, in order not to offend the Toh of the locality, who are believed to use them as resting-places.^ We may aptly compare this with the instructions given by Cato to the Latin farmer for making a new clearing ; he is to offer sacrifice and to pray the unknown spirits of the wood or trees to excuse the liberty he is taking : and when he proceeds to dig the ground, he is to take- further steps to set himself right with the spirits.* Midway between the gods and the Toh are spirits of a higher order than the latter, i. e. those attached to men or animals, alive or dead ; but they do not seem to throw any new light on ancestor worship or the idea of Genius, as we might have expected. True, the description given of the ' secret helper ' of an individual, which occurs mostly among the Sea-Dyaks, rarely among the tribes we are specially concerned with, is of great psychological interest ; but it does not seem as yet to be sufficiently investigated to enable us to compare it with familiar Greek and Roman beliefs.^ The most instruc- tive point about it is perhaps the strong proof contained in it of the development of the individual as distinct from the group. I proceed in the last place to consider the system of divina- tion among the Kayans and Kenyahs, which affords such striking points of comparison with that of the Romans and 1 H. and McD. ii. 25. " Rd. Exp., pp. 41, 316, and many other passages (see index). See above, p. 8 ff. •■ H. and JIcD. ii. 23. ' Cato, de Agric, ch. 139, and see above p. 87. ' H. and McD. ii. 90 f. Digitized by Microsoft® 160 ANCIENT ITALY AND MODERN BORNEO : Etruscans that it caused the authors to look up the details of the process in ancient Italy. They suggest, though with all due caution, that the two systems had a common root : that ' while the Aryans carried the system westward into Europe, the Indonesians, or some Caucasic people that has been merged in the Indonesian stock, carried it eastward '.^ The bird whose flight is most important in augury is a hawk (Haliastur intermedins). In the work we are examining (ii. 52) is a most interesting account of the way in which this bird is consulted. A party of Kenyahs under Tama Bulan are about to undertake a journey, with the object of making peace with ajiother tribe. The record is in the form of a diary, and the writer very naturally remarks at the end that he felt as if he had just lived through a book of the Aeneid. After the chief had been purified with water and the blood of pigs, three men of the upper class — patricians in fact, who alone can take the auspices^ — sat under an extempore shelter and searched the sky for hawks. After a while they caught sight of a hawk high up and far away, and began to try and persuade it to fly towards the right. This they luckily accomplished ; then they settled down to watch for another that should fly towards the left, and a third which should circle round and round. ' In about half an hour two hawks had obligingly put in an appearance, and behaved just as it was hoped they would behave ; and so this part of the business was finished.' What followed, i. e. sacrifice and prayer, also has its interest for us. Tama Bulan, though a trifle unwilUng to perform these ancient rites, nevertheless had to act in a priestly capacity, like the Roman rex : he sprinkled on the congregation the blood of the victims (a practice obsolete in historical Rome), and recited a prayer in a rapid murmur, like the priests of Iguvium (tacitus pre- cator).^ ' Meanwhile four boys were pounding at two big drums to keep away from the worshippers all sounds but the ' H. and McD. u. 255. ' Bel. Exp., p. 304. ' JeTong in Anthropology and the Classics, pp. 94 f. ; Rel. Exp., p. 187. Tama Bulan seems to have disliked the magical character of such praying. Digitized by Microsoft® A STUDY IN COMPARATIVE CULTURE 161 words of their own prayers.' So at Rome the priest covered his head with his toga, and tibicines played aU the time in order that no unlucky sound or word might be heard which would make it necessary to start afresh with a new victim. Here the Borneo practice seems useful in dissuading us from abandoning this old interpretation of the veiled head in favour of one of which the Romans themselves knew nothing at all, viz. that it was a sign of separation and consecration.^ A still more remarkable parallel with Roman custom is the consultation of BaU Flaki, the divine hawk, before sowing and harvesting the rice crop.^ On some quiet spot by the river bank, what is called a ' tegulun ' is set up, i. e. a horizontal pole supported about a yard above the ground by a pair of vertical poles. The augur lights a small fire beside the ' tegulun ', and sits on the ground behind it, so as to see through it a square patch of sky, and waits until a hawk becomes visible on this patch. When the hawk becomes visible he waves it towards the left : for he knows that if it fUes towards the left he wiU prevail over his enemy, but that if it goes to the right his enemy is too strong for him. Here we plainly have the old ItaUan ' templum ' of augury, the rectangular space marked out by the ' lituus ', within which birds might be observed ; ' and here too we find the rule which both in Borneo and in Italy applied to some birds though not to all, that the left was the auspicious side. Then again we have ' oscines ' as well as ' ahtes ', e. g. the wood- pecker has two notes, one of which is lucky, the other unlucky.* ' See Rd. Exp., p. 195, note 35, and references there given. Add Aen. iii. 405 f., which leaves no doubt as to the Roman interpretation. So too Serv. ad loo. = H. and MoD. ii. 56. ' The templum was a rectangular space, af which the frame, so to speak, was an imaginary one ; but I suspect that it was originally the space visible from the door or opening in the operator's tabernaculum, which would naturally be a rectangular one ; cf. von Jhering, Evolution of the. Aryan (Eng. trans.), p. 364. » H. and MoD. ii. 59. At Rome there were many minutiae in the doctrine of omens from oscines : here too the Black Woodpecker (Picus Marlins) was an important bird in augury. Festus (Lindsay, p. 214). On the subject generally, see Bouche-Leclercq, Hist, de Divination, iv. 199 f, 325? j^ Digitized by Microsoft® 162 ANCIENT ITALY AND MODERN BORNEO ; If the augur has secured good omens from other birds, his men wUl avoid heariag the woodpecker by singing and rattling their paddles on the side of the boat, which seems to savour of Roman ways, if it cannot be exactly paralleled at Rome. We may remember that the famous MarceUus of the second Punic war used to ride in his litter with the blinds drawn down, so that he should not see anything of evU omen.^ The method of divination by examination of a victim's liver is also of great interest for students of old Italian religion. The sacrifice of a pig, the commonest victim, is always accompanied by such an examination, and Messrs. Hose and McDougall have given us a photograph of two Kayans in the act of inspection.^ It does not appear that they are trained experts, like the haruspices ; but these were the result of a long period of development, and we have to look for a closer parallel in the earUest form of the reUgion of Rome, before the Etruscans had passed on their elaborate code of doctrine to the Roman augurs.* As in early Rome, the liver was marked out in certain divisions. ' The rule generally followed is to identify the imder surface of the right lobe with the territory of the party that kUls the pig and makes the inquiry ; the adjacent part of the left lobe with the territory of any party involved in the question, which adjoins that of the first party ; and the under surface of the caudal extremity witE that of any remoter third party.' It is interesting to note that the omens thus obtained are held to be the answer of the god to the prayers carried to him by the spirit of the pig ; and also that the killing of a pig is always the occasion for, ' Cic. Div. ii. 36. 77. ^ ;;, go. " Wissowa maintains {R. K., ed. 2, pp. 418 f.) that the Romans did not examine the entrails with a view to divination, except to see whether the god would accept the victim, till they learnt the art from the Etruscans ; but I do not feel quite convinced on this point. (Wissowa states.the same view in his article on Koman divination in Hastings's Bncycl. iv. 824. ) The appearance of the liver seems everywhere to have been looked on as a message from the deity ; what that message was, whether simply approval of the victim or communication of a more detailed character, is difficult to deter- mine in eaeh case. Some good remarks on this subject, based chiefly on ThuUn, op. oit., will be found in Prof. Halliday's Divination in Cfreece, pp. 192 f . Digitized by Microsoft® A STUDY IN COMPARATIVE CULTURE 163 or occasioned by, some religious rite, and even if the occasion is not primarily a religious one, the victim is invariably charged with a message to a god.^ We do hot find here any system so highly elaborated as that of the Etruscans, illustrated as it now is by the famous bronze Uver of Piacenza, as well as by the work of Martianus CapeUa, and indeed now also by our knowledge of the Baby- lonian science of hepatoscopy.^ But the method of dividing the surface of the liver is the same in principle in both systems, though the Etruscans divided their livers into seats of deities instead of into territories belonging to the parties in a quarrel : and the Borneans seem to be in a fair way of developing their methods with something like Etruscan extravagance. On the whole we find an extraordinary growth of this pseudo- science in distant Borneo, in Babylonia, and in ancient Italy, such as is not found, so far as I know, in any other part of the world. What conclusions, if any, are to be drawn from this I will not attempt to guess, unless it be to suggest that we are to see in the mental build of these three peoples, or the • H. and MoD. ii. 64. The authors seem to have no doubt about this ; the message is carried, they say, by the spirit of the pig. But this purpose of sacrifice is rare, so far as I know. It is found in connexion with human sacrifice in Greece, but not in Italy. There is the famous case of the Getae in Thrace, who sent a messenger by sacrifice every five years to their god Zalmoxis, telling him what they were in need of (Herodotus, iv. 94) : and Herodotus teUs us that they expected to go to this god themselves after death. Again in Euripides {Hec. 422-3), Polyxena, about to be sacrificed, asks what message she is to bear from Hecuba to Priam and Hector : which according to Mr. Lawson (Modern Greek Folklore, &o., pp. 340 and 346), who found the same idea in the island of Santorin surviving at the present day, was no mere poetic conceit, but a feature of the popular religion.- Lastly, in Aen. ii. 547 (also quoted by Mr. Lawson), Pyrrhus makes Priam a messenger at the moment of his slaughter : ' referes ergo haec et nuntius ibis Pelidae genitori.' It seems not lumatural that the idea shoiild spring up in connexion with human sacrifice, but I do not find it associated with ordinary animal sacrifice : is it possible that in Borneo it may be a late interpretation of animal sacrifice combined with prayer, and with the answer of the god in the liver of the sacrificed victim ? Messrs. Hubert et Mauss in their essay on Sacrifice {Mtianges d'histoire des rdigiona) do not reach this view, though once or twice they approach it, e. g. on p. 124. 2 Thulin, Martianus Ca^ella imd die Leber van Piacenza, p. 7 ; Hastiags's Encyd. of Rdigion and Ethics, iv. 784. L 2 Digitized by Microsoft® 164 ANCIENT ITALY AND MODERN BORNEO ; upper classes among them, considerable resemblance to each other. In conclusion I will add two or three points of some interest which have not been noticed in this paper so far : 1. Marriage. The ceremony is described as a blending of symbolical capture with actual purchase (ii. 174). In the Roman ceremony these two elements are also to be traced as survivals. The subject is too elaborate for me to dwell on it here : but see my article on Roman Marriage in Hastings's Encycl. of Beligion and Ethics} 2. Initiation at puberty. There is no elaboration of cere- mony for this purpose in Borneo, as among the Australians and other primitive peoples, and (according to Miss Harrison) among the Greeks.^ At Rome too we know of no such practice, unless the change of toga can be regarded as a survival of it. 3. Mythology. Though the Kayans, &c., are fond of stories, and possess a mythological instinct, their myths, like those of ancient Italy, have little or no relation to religion and the gods.^ Roman divine myths are of Greek origin : the Borneans have none, properly so-called. 4. Magic. ' The Kayans dislike and discourage all magical practices, with the exception of those which are publicly practised for beneficent purposes, and have the sanction of custom. In the old days they used to kill those suspected of working evil by magic. There are no recognized magicians among them.' * The same dislike is a characteristic of Roman life, where magic was excluded from state ritual except in ' In the Roman ceremony we have the simulated rape of the bride at the deduotio, the parting of her Tiair with a spear, and the lifting her over the threshold of the bridegroom's house, together with the legend of the rape of the Sabines. For the similar rites of the upper class among the Kayans, see H. and MoD. ii. 172 f. But there is no actual trace of exogamy among either people, apart from these forms, which may be explained, as Mr. Crawley first showed us, in quite a different way. See his Mystic Rose, pp. 350 f. There was, of course, no actual purchase at Rome, but the form of coemptio may suggest that it was in use at some remote time, though personally I doubt this ; see my article on Roman Marriage in Hastings, Encycl. of Bel. and Mh., vol. viii. ^ See her Themis, eh. i and elsewhere. » H. and McD. ii. 136 ff. « ibid. ii. 115 f. Digitized by Microsoft® A STUDY IN COMPARATIVE CULTURE 165 the form of meaniBgless survivals, and practised privately under precautions taken by the State. See my Roman Religious Experience, pp. 49 and 107 (cf. index s. v. Magic). 5. The intimate relation (a quasi-religious one) of Bomean women to the seed-corn and its growth, and in fact to all agricultural operations, finds illustration rather in Greece than in Italy ,^ but we may aptly remember that the deities of Italian agriculture were mostly female, Tellus, Ceres, Pales, Ops, Vesta, Feronia (probably), Venus, Flora, and Pomona : a fact not without bearing on the question. In historical Rome the women did not work in the fields, as in Borneo, but the idea of a close connexion between that sex and the fruitfulness of grain seems to have survived in these female deities. PARALLELA QUAEDAM THE PLAGUE OF LOCUSTS IN 125 B. C. : AND A MODERN PARALLEL Two years before the first tribunate of Gaius Gracchus, the Roman province of Africa, then the chief source of the corn supply of Rome, was ravaged by a terrible invasion of locusts. The fact is well attested ; Livy probably described it at some length, for his epitomist (Ep. 60) devoted a sentence to it : ' PestUentia in Africa ab ingenti locustarum multitudine et deinde necatarum strage fuisse traditur '. We can gather further details from Julius Obsequens's book of prodigia (ch. 30, ed. Jahn), from St. Augustine {Civ. Dei. iii. 31), and from Orosius (v. 11). All these accounts probably owe their details chiefly to Livy ; but as Augustine and Orosius were both intimately connected with the African province, the one ' H. and MoD. i. 111. Here we are in touch with the idea of the "sital principle in the seed, which is carried over from one harvest to another through the seed-corn, and is, so to speak, immortal. It may be at the root of many practices still here and there surviving, as I have suggested in ' Mundus Patet ', above p. 33 if. Digitized by Microsoft® 166 PARALLELA QUAEDAM as a native, the other as an immigrant in early life, it is just possible that they may have picked up some traditions of the disaster still iioating about at Utica or Carthage. The most curious feature in their account is the fact — for such, as I shall show, we may call it — that the swarms of locusts were blown out to sea by a sudden storm and drowned. Then, as the story runs, they were washed up on the shore by the waves, and as their numbers were boundless along a great reach of coast, the result was a poisonous miasma which produced a terrible pestilence. Orosius shall speak for himself : At vero quanta fuerit hominum lues, ego ipse dum refero toto corpore perhorresco : siquidem in Numidia, in qua tunc Micipsa rex erat, octingenta miha hominum, circa oram vero maritimam, quae maxime Carthaginiensi atque Uticensi litori adiacet, plusquam ducenta miUa perisse traditum est. He adds that 30,000 soldiers, the garrison of the province, were wiped out ! That he did not get these numbers from Livy is sufficiently clear from the story of the 30,000 legion- aries ; but on the whole we may assume that there was a pestilence at the time, to which the stench of the dead insects in some way contributed : so much seems to be proved by the words of the epitomist of Livy.^ Whether the corn supply of Rome was affected by the visitation we do not know for certain ; but Nitzsch in his monograph on the Gracchi (1847) went so far as to suggest that C. Gracchus found in it a good opportunity for his lex frwmentaria ; ^ and it is quite possible that in the following year, 124, when Gracchus was canvassing for the tribunate, the question of supply and of prices was brought home to him in this way. My modern parallel is from South Africa, and was luckily recorded by a man of an accurate and scientific habit of mind. Sir John Barrow, afterwards F.R.S. and Secretary to the Admiralty for some forty years, happened to be at the Cape in 1796, on his way honie from Lord Macartney's mission to China, and was the first Englishman to travel as far as the ' Cf. Joel, ii. 6 ft. ; Tristram, Nat. Hist, of Bible, p. 314. » p. 393 ; cf. Greenidge, Hist, of Borne, p. 203 Digitized by Microsoft® PARALLELA QUAEDAM 167 Orange river, at a time when the country was in our hands for a few years before being returned to the Dutch at the peace of Amiens. In his autobiography he wrote as follows of a locust plague of that year : The present year was the third of their continuance in this part of the colony. Their last departure, with its result, is described as rather singular, and it was confirmed by the inhabitants of the lower part of the colony. All the full- fledged insects were driven by a tempestuous north-west wind into the sea, and afterwards thrown back upon the beach, where they formed a bank three or four feet high, between the mouths of the Bosjesmans river and the Beeka, a distance of nearly fifty miles ; and our present comparty assured me that when this mass became putrid, the stench was sensibly felt in several parts of Sneuwberg (p. 181). The Sneuwberg range is about 150 miles as the crow flies from the mouth of the Bushman river. The stench must have been quite enough, one would imagine, at the coast, to be the contributory cause of a pestilence, but of this Barrow gives no hint. We must, however, remember that the coast in question was at that time very thinly populated as compared with that of Roman North Africa in 125 b. c, where for the last twenty years a Roman population had been adding itself to that of Carthaginians and Libyans. But whether or no the pestilence was due to the locusts, the heaping of the dead locusts on the shore, and the stench thereby occasioned, are fully and independently confirmed by Barrow's narrative. Those who are apt to find fables everywhere in history, may learn here how exaggeration and marvel may gather in course of time round a perfectly authentic fact. PLAGUES OE FIELD- VOLES {Arvicola arvalis) IN ANCIENT AND MODERN TIMES In the Zoologist for September 1892 there was an interesting account of a plague of these destructive little animals, which ravaged Thessaly in the spring of that year. The account was Digitized by Microsoft® 168 PARALLELA QUAEDAM translated from the Centralblatt fur Bacteriologie, and the writer was Dr. LoefBer of Greifswald, who had been invited by the Greek Government to attempt the extirpation of the pests by introducing among them a bacillus fatal to mice. Not long before this there had been a very serious plague of the same voles in the south of Scotland, when birds of prey arrived in large numbers and did much execution on them. In both cases the harvest, which promised to be very good, was not only threatened but seriously damaged. These sudden plagues of mice were well known in the ancient world, especially in Greece. Aristotle {Hist. Anim. 580 b) says that ' such an inexpressible number of field-mice has sometimes appeared that very little food remained. Their destructive power is also so great that some small farmers, having on one day observed their corn ready for harvest, when they went the next day to cut it, found it all eaten. The manner of their disappearance also is unaccountable, though up to that time they could not be exterminated either by smoking or digging them out.' He adds that nothing can get the better of them but rain. Whether he was right about the rain I do not know ; but the report of the Greek Government, quoted in the Zoologist, amply confirms the most striking feature of Aristotle's description. ' One evening a field was visited which was to be mown the next day : but when the labourers came to the field next morning they found nothing left to cut. The voles had destroyed the entire crop in a single night.' A little later I drew attention to these facts in the Classical Review, and suggested that the cult of Apollo Smintheus, the Mouse Apollo, might be better explained as having some relation to such plagues, than as a relic of totemism, as Mr. Lang had shown himself disposed to look on it in his Myth, Ritual, and Religion. Mr. Lang wrote to me from Edinburgh that he would not ' go to the stake ' for the totemio explanation, and that the other might perhaps bei the right one. A few days later I found in the Daily News a short buf delicious article, in which he urged the farmers Digitized by Microsoft® PARALLELA QUAEDAM 169 of the Lowlands to set up the worship of Apollo Smintheus without delay, in case they should again be attacked by the voles. Armati terram exercent, semperque recentes Convectare iuvat praedas et vivere rapto. I have always thought of Virgil's lines (Aen. vii. 748-9) as giving us our best available picture of old Italian life in the age of the first settlements made by invaders who after- wards became Latins, Umbrians, and Samnites. The picture is reproduced in bk. ix. 607 ff. : At patiens operum parvoque adsueta inventus , aut rastris terram domat aut quatit oppida beUo. omne aevum ferro teritur, versaque iuvencum terga fatigamus hasta. In the former passage Virgil is speaking of the Aequi, the very people who in Livy's earlier books descend so often from their hills to raid the Latin territory in the plain below : ' Horrida praecipue cui gens adsuetaque multo Venatu nemorum, duris Aequicola glebis.' In the ninth book he is drawing a picture of the Rutuli, who stand in the poem for the wilder tribes then inhabiting Latium itself. Where did the poet get his knowledge of this uncomfortable kind of life, which seems characteristic of tribes that still Uve largely by hunting, but keep sheep and cattle for purposes of clothing, and have begun to cultivate patches of ground cleared in the forest, and perhaps to hide the produce in pits underground, the prototypes, as I believe, of the Roman mundus of later times ? Perhaps he knew of such tribes in the foothills of the Alps, near his own Mantuan home ; or he and Livy may both have known of traditions in Latium of the old days when neither your cattle nor your com was safe from the raids of your neighbours. However that may be, Virgil's terse and intensely graphic picture seems to appeal to the reader instantly as being true to the life. I have recently stumbled on an equally graphic picture of this very condition of life, Digitized by Microsoft® 170 PARALLELA QUAEDAM dating no further back than 1864. The author, Dr. Tristram, knew his Bible almost by heart, but had forgotten his Virgil, or he would have quoted the passages I have printed above. After a first attempt to penetrate into the hill-country beyond Jordan, in which he and his friends were themselves raided and robbed, the party hired a strong Arab escort and tried again successfully. The Bedouin tribes in this land of GUead were, and stUl seem to be, in exactly the state described by Virgil, hunting, cattle-keeping, and cultivating small patches, which may at any moment be raided by a hostile tribe. On approaching a certain hUl-village, the party created a panic among the inhabitants, who took them for enemies out for plunder. ' Cows were driven away in all directions, men snatched up guns and stood on the defensive. When the mistake had been discovered and the travellers passed on, the men, with their reaping-hooks and plough- shares, but of course with their guns slung on their shoulders, were hurrying back to their peaceful occupations in the fields. What a country to live in, with a plough in one hand and a firelock in the other ! ' In other parts of his book [The Land of Israel, 1866) Dr. Tristram has also described the underground pits in which these tribes bury the produce of their fields for safety (e. g. p. 108). I have long been of opinion that it is exactly this condition of hfe which was represented in central Italy by the cult of Mars, who has so often puzzled investigators as being at once a numen connected with agriculture, and a deity of war, as we know him in historical times in the Roman State- worship. ' A spirit who dwells on the outskirts of civiUzation, and can with profit be propitiated both for help against the enemies beyond, and for the protection of the crops and cattle within, the boundaries of human activity.' ^ ' Religious Experience of the Roman People pp. 133 £E. Digitized by Microsoft® PARALLELA QUAEDAM 171 THE DISAPPEARANCE OF THE EARLIEST LATIN POETRY ; A MODERN PARALLEL The foUowirig note is part of the result of a conversation with a Rhodes scholar about the famous Ballad theory of Niebuhr, of which we seldom hear now except in connexion with Macaulay's Lays. Niebuhr, of course, let his imagination carry him too far, and beheved in the existence of a Roman national epos as the foundation of the ' history ' of the regal period. He thus laid himself open to trenchant criticisni, e. g. that of Schwegler in the first volume of his Roman History, which wiped out his theory, and rather unluckily diverted attention from the texts on which it was based. Recent criticism seems rather to look on these as proving not indeed a national epos, but the practice in the early republican period of singing at meals and funerals the praises of famous men to the accompaniment of the tibia. H. Nettle- ship accepted this conclusion,^ E. Meyer raises no difficulty about it,^ and Pais has adduced parallels to the practice among other peoples.' And, indeed, whoever reads these once famous passages will find it hard to deny that they must represent a sound tradition of an actual Roman practice. But what these songs were really like, we know no more than we do what the music was hke which accompanied them ; for words and music have alike utterly vanished. It may be as well to quote these in full, as otherwise my suggestion of a modern parallel would be less inteUigible. Cic. Tusc. iv. 2. 3 ' Gravissimus auctor in Originibus dixit Cato morem apud maiores hunc epularum fuisse ut deinceps qui accubarent canerent ad tibiam clarorum virorum laudes atque virtutes. Ex quo perspicuum est,, et cantus tum fuisse rescriptos vocum sonis et carmina.' Cic. Brutus, 75 ' Atque utinam exstarent Ula carmina, quae multis saecuhs ante suam aetatem in epuHs esse cantitata a singulis convivis de clarorum virorum laudibus, in Originibus scriptum reliquit Cato ! ' ' Essays in Latin Literature, p. 58. " Gesch. des Altertwms, ii. 2. 397. " Storia di Boma, i. 9, note. Digitized by Microsoft® 172 PARALLELA QUAEDAM Cic. de Oral. iii. 51. 197 ' Nihil est tarn cognatum mentibus nostria quam numeri atque voces . . . quorum ilia summa vis carminibus est aptior et cantibus, non neglecta, ut mihi videtur, a Numa, rege doctissimo, maioribusque nostris, ut epularum sollemnium fides ac tibiae, Saliorumque versus indicant.' Valerius Maximus, ii. 1. 10 ' Maiores natu in conviviis ad tibias egregia superiorum opera carmine comprehensa pange- bant, quo ad ea imitanda iuventutem alacriorem redderent.' Varro, ap. Nonius, p. 77 (Lindsay, i. 107) ' In conviviis pueri modesti ut cantarent carmina antiqua, in quibus laudes erant maiorum, et assa voce (i. e. without accompaniment) et cum tibicine ' {de Vita Pop. Bom. lib. ii). Now the complete disappearance of this poetry was made use of as one of the most damaging criticisms of Niebuhr's theory ; Schwegler, for instance, asks why no commentator of the age of Roman learning cared to unearth them for his purposes. Niebuhr himself ascribed this disappearance mainly to the influence of Ennius, i. e. of a foreigner whose genius simply overwhelmed the old Italian poetry and gave the Romans something better to think about. And this is indeed the only explanation possible. Ennius came in on the crest of the wave of Greek tendency, at a time of momentous change in the modes of thought and habits of the Romans. Greece, in fact, invaded Italy with a great general, Ennius, at the head of her victorious host. It has occurred to me that we may find an instructive parallel to this sudden disappear- ance of a native art in the history of our own country. In the first half of the eighteenth century, Germany and Italy invaded musical England in the person of Handel, a German by birth and an Italian by training. The result was that our own ancient national music almost entirely vanished from the minds of Enghshmen. It could not, indeed, wholly vanish from the world, for it was in manuscript or in print, and some of it survived in our cathedral services. The obliteration was less complete than with the old Latin songs. But none the less Handel was the Ennius of our musical history. So effectually did he and his great German successors wipe out the memory of the English music of the Digitized by Microsoft® PARALLELA QUAEDAM 173 fifteenth and sixteenth centuries, that not even the all- pervading German research of to-day has ever reached it. The Germans know as little of it as the Greeks knew of the old Latin poetry ; like the learned scholars of the Roman Empire they remain unconscious of a fine field of inquiry. On this point I may refer to Dr. Ernest Walker's excellent History of Miosic in England (Oxford, 1907). The parallel may be pushed even further. For at least a century after Ennius Rome produced no great poet. When at last Roman poetry suddenly became great, it was because men of real genius had so thoroughly absorbed aU that the Greeks had to teach them, that they were now free to assert their own poetical individuality, to express their own national traditions in their own way. They were not hampered by their Greek training, but rather aided by it, in expressing themselves. So in English music ; for a century after Handel's death there was no really great English composer, and it is only in the last half-centiu'y that two or three have been able to express their own individuality in spite of the overwhelming influence of German music. Doubtless this process of gaining freedom will increase in strength, for the Germans can now no longer persuade themselves and the world that the secrets of the musical art are their own pecuhar possession. ROMAN LEGES DATAE AND ENGLISH ENCLOSURE AWARDS To the Roman political mind it seemed natural and neces- sary that every municipium, every town, that is, included in the Roman body politic in any sense or degree,^ should have a document, a charter, in the Roman speech a lex, as the legal basis of its municipal existence, fixing its constitution and other details. The Roman legal instinct required such documents, and did not look on them as mere ' scraps of paper ' ; for every province, for example, it demanded a lex ' I use the word in the widest sense, in which throughout Roman history a municipium could mean any town having the whqle or a part of the Roman civitas. Digitized by Microsoft® 174 PARALLELA QUAEDAM provinciae, for every temple a lex templi, regulating the rights, customs, revenues, &c., of the institution. How were these charters obtained ? The lex templi was the legal work of the college of pontifices ; but the charter of a province or a muni- cipium had its legal basis in an Act of the Legislature, which in each case appointed a commissioner or commissioners, to frame the necessary document. This document, being imposed or bestowed on the town, was called a lex data, in contrast to the lex rogata of the Comitia from which it derived its legal authority. Of such leges datae, as every student knows, we have several specimens preserved more or less completely, but among them only one relating to an Italian town — a fragment of the lex data of Tarentum, discovered in 1895. While studying the history of enclosures in England, and in particular the Enclosure award of my own village history, I have often been struck by the close analogy between the lex data of a Roman municipium and the Enclosure award of an English parish. Like the lex data, the Enclosure award had its legal basis in an Act of the Legislature ; and it was imposed (or bestowed) on the parish by commissioners (one or more) appointed under that Act. Thus in England as at Rome, we find examples of the regulation of local institutions by commissions deriving their legal authority from Acts of the Legislature. Let me explain this a little more fuUy. An Enclosure is a rearrangement of the land of a parish in respect of ownership, whether of individuals or corporations, and in respect of rights, such as those of common pasture or fuelling. Before the eighteenth century it was often effected by the owner or owners of land without recourse to Parliament ; but in the eighteenth and nineteenth centuries it was deemed necessary to give it legal sanction by the legislative process. An Act, private indeed, but fully open to criticism and defeat, empowered commissioners to go to the parish in question, and after due inquiry, and under oath to act impartially, to effect the necessary changes.^ The result, when completed • See Gonner, Common Land and Enclosure, pp. 56 ff., and chap, iii ; Digitized by Microsoft® PARALLELA QUAEDAM * 175 by the commissioners, was an award binding on the parish, a charter from which there was no escape except by further legislation. The resemblance to a lex data will be obvious to every student of Roman history. The work of the commissioners was naturally found to increase in complexity with the increase and variety of the parishes to be enclosed ; but ' so far as the general -structure of these private Acts is concerned, and by consequence of the awards resulting from them, a fairly constant form seems to have been attained by the end of the reign of George II ' (1760).'- It is probable that Roman commissioners also developed a ' forma communis ', as Mommsen called it,^ without the immediate aid of any general law regulating the aims and procedure that should govern the action of commissioners. Such a tendency to uniformity was in England no doubt the result, as Professor Gonner suggests,* of the practice of sending each bill to a committee, which again led to certain standing orders of the House of Commons on the subject. Yet beyond doubt there was much variety of practice, arising from the soil, history, usages, &c., of the parishes to be enclosed. Towards the end of the eighteenth century, after thousands of enclosures had been efEected by these awards, we find a growing desire to have a general Act which should secure such uniformity as had been already attained, and facilitate and improve the work in each case. After some previous attempts, such a general Act was passed in 1801. Before we return to Italy, let me give a brief indica- tion of the nature of this general Act. It is described as consohdating in one Act certain provisions usually inserted in private Enclosure Acts.* For example, it took some forty clauses commonly found in these, and Hammond, The Village Labourer, chap, iii ; Prothero, English Farming Past and Present, pp. 249 ff. ' Goraier, op. cit., p. 65. ° See Legras, La Table latine d'Hiradie, pp. 268 ff. " op. oit., p. 65. « The best account is given by Gonner, op. cit., pp. 62 ff. ; cf. Hammond, op. oit., p. 77. Digitized by Microsoft® 176 PARALLELA QUAEDAM provided for their incorporation in future bills. It also introduced a few new features ; e. g. it allowed afftdavits to be taken as evidence, and relieved witnesses from coming to London to make oath before parliamentary committees ; and it instructed commissioners to have special regard to the convenience of small holders of land. No change in the methods of enclosure was introduced ; the commissioners and their ' leges datae ' went on as before. But their work was simplified, and their tendency to uniformity increased, by the insertion in this general Act of principles of action which experience had already approved. Let us now return to Italy, and see whether this last phase in the history of enclosing in England has any bearing on that of municipal legislation at Rome. It is curious that there is a lively dispute at the present time as to whether there ever was or could have been a general law' regulating the action of commissioners appointed to draw up leges datae for municipia} I must very briefly explain how this dispute has arisen. It is of course in the period immediately following tlie enfranchisement of Italy, 89 and following years, that we should most naturally expect to hear of municipal leges datae. Almost all Itahan communities which had formerly been ' allies ' of Rome {civitates foederatae) now became municipia, lost their foedera, which had so far supplied the legal basis for their relation to Rome, and needed fresh charters to justify and regulate their civic existence. Whether in these years of confusion and civil war, or even for many years after the death of Sulla, this need was sufficiently supplied, is a disputed question into which I need not enter here. I will only express my own opinion, after careful study of the controversy, that it was only supplied gradually and sporadically. I see no reason why a town should not have continued to use its old constitution, together with its new rights of civitas, while ' See Reid, Municipalities of the Roman Empire, p. 147 ; Mommsen in his Commentary on the lex data of Tarentum, Ephemeris Epigraphica, ix. 1 ; and Harflky in J. B. S., 1914, pp. 96 if,, criticizing H. Legras, op. cit., pp. 233 ff, Digitized by Microsoft® PARALLELA QUAEDAM 177 in expectation of a charter that might be long delayed. It is as well to remember that such work as the ' constituting ' ^ of municipia would hardly be sought for by distinguished Romans, who were busy making war or making money in this period ; and that the wars and proscriptions had greatly thinned the ranks of persons qualified for such delicate legal operations. The difficulty of answering this question as to the con- stituting of municipia in this period has led to the 'dispute among the learned to which I alluded just now. The question is this : Was there, apart from single kges datae for individual municipia, any general law intended to affect municipia in general, to lay down principles on which commissioners should act in each case, so as to secure at least a certain amount of uniformity ? And is the document known to us as the lex lulia municipalis, or at least the latter part of it, which deals with certain municipal institutions in general terms, such a general law ? Formerly, and in fact until 1903, it was assumed by most inquirers that Mommsen was right in claiming the document as a general municipal law ; but in that year, in commenting on the newly discovered fragment of the lex data of Tarentum,^ he entirely recanted this opinion ; and ever since then there has been a growing tendency to deny that such a general law was necessary or even possible. The whole question wUl be found reasoned out with exactness and discrimination by Dr. Hardy in his recent criticism of the revolutionary work of a French scholar on the so-called lex lulia municipalis. Dr. Hardy has come to the conclusion that there was such a general law, and that it was intended ' to fix the forma communis to bind the commissioners in respect to principles, and to afford them guidance though naturally with greater laxity, in matters of detail ' (p. 100). He also conjectures that the lex lulia municipalis ' was not the first formulation of a general municipal law, but a revision ' The word was constituere, as in Caesar, Bdl, Civ. i, 15, of Cingulum, quod opiMdum Labienus constituerat '. = See above, p. 176 note 1. 2252 M Digitized by Microsoft® 178 PARALLELA QUAEDAM and re-enactment of earlier laws, together with certain new provisions, of which two are preserved on the Table of Heraclea ' ; and that the date of that law was probably 45 B. c. (p. 108). It seems to me that the evidence of our English method of deaUng with certain local problems, which beyond doubt closely resembles that of the Roman in curiously similar circumstances, strongly confirms Dr. Hardy's conclusions. If there was nothing impossible or even unlikely in the enactment in England of a general law of regulation, even after the process which it was intended to regulate had been going on for some time, I do not see why such a law should have been unlikely or impossible for Roman lawyers and statesmen.^ THE GREAT SERPENT OF THE RIVER BAGRADAS WhUe turning over Niebuhr's Lectures on the Punic Wars, I came on a tale which had never yet attracted my attention. The modern historian does not trouble himself with portents, monsters, miracles, and such fairy tales ; he has more im- portant things to attend to. Yet I have always found it interesting and sometimes useful, to turn over the pages of Julius Obsequens's' Record of Portents, for it teUs us something of the mind of a people, and occasionally something of the natural history of the Mediterranean. For example, the great plague of locusts in 125 b. c. I have shown by analogy to be essentially true, and not without historical significance. Niebuhr, who let himself talk in his lectures about anything that interested him, did not scruple to tell his hearers at Bonn the tale of the great serpent of the River Bagradas (Mejerda), though he fancied that it was an invention of the poet Naevius, who wrote in verse the history of the first Punic War. It has reached us through the epitomist of Livy's lost books, who could not forbear to mention it as an integral part of the Jor ft Prussian general municipal law, to which the constitutions of individual cities were required to conform, see Niebuhr's Life and Letters, III. xxxix. Digitized by Microsoft® PARALLELA QUAEDAM 179 eighteenth book, just as in book sixty he selected the locusts as likely to interest people who had not time to read the full text of Livy. His way of introducing it is amusing, and might suggest that the serpent was an even more formidable enemy than the Carthaginians. ' AtiUus Regulus in Africa serpentem portentosae magnitudinis cum magna olade miUtum occidit, et cum aUquod proeliis bene aduersus Carthaginienses pu- gnasset . . .' &c. To get a fair idea of what Livy really wrote about this marvel we must go on some foxu? centuries to Orosius,^ as we did with the locusts ; for these things happened in Africa, to which province Orosius belonged, and that excellent Christian was not only fond of marvels but fond of connecting them with his own country. He shall speak for himself. Regulus, bellum Carthaginiense sortitus, iter cum exercitu faciens haud procul a flumine Bagrada castra constituit : ubi cum plurimos militum aquandi necessitate ad flumen descendentes serpens mirae magnitudinis devoraret, Regulus ad expugnandum bestiam cum exercitu prof ectus est. Sed nihil in tergo eius proficientibus iaculis atque omni telorum ictu inrito, quae per horrendam sqamarum cratem quasi per obUquam scutorum testudinem labebantur mirumque in modum, ne corpus laederent, ipso corpore peUebantur, cum insuper magnam multitudinem morsu comminui, impetu proteri, haUtu etiam pestifero examinari uideret, baUistas deferri imperavit, per quas murale saxum spinae eius incussum conpagem totius corporis soluit. (At this point Orosius digresses to inform his readers of the way in which snakes move along the ground, which account, strange to say, is correct in the main. He goes on to teU us that the creature was mortally wounded by a stone from a ballista, and then kiUed with javelins.) Corium autem eius Romam devectum, quod fuisse centum viginti pedurn, spatio ferunt, aliquamdiu cunctis miraculo fuit. {Hist. iv. 8.) That the monster had to be killed by having siege machinery brought to bear on him, does not seem to me incredible ; but we may reduce the number of Roman soldiers whom he devoured to one or two. The length of the beast, 120 ft., may be reduced by one-half or more, to come within the 1 Valerius Maximus (i. 8. 19) also tells the story from Livy. Silius ItaJious has an amusingly absurd account of it : vi. 146 ft. M 2 Digitized by Microsoft® 180 PARALLELA QUAEDAM * range of reality, as we shall see directly. Niebuhr remarks that the length is exactly that number of feet which so often occurs in Roman institutions, namely a multiple of 10 and 12. He is thinking, I suppose, of the threefold division of the legion of the second century B. c, into bodies of 1,200 each. But Pliny, who had been attracted by the story, quite accepts the 120 ft., and adds that the skin and jaws were preserved in a temple at Rome ' usque ad bellum Numantinum ' (Nat. Hist. viii. 37). This seems to me to help us in believing the main outline of the tale ; for skin and jaws would be exactly the parts which it would be natural to preserve, if the creature were really a serpent, and not e. g. a gigantic crocodile. I could wish that Pliny had told us what temple was honoured with these remains. There were no museums in those days, and if such a gift were given to a temple it must have been as a religious offering ; the only one to which such an offering would seem appropriate was that of Aescula- pius in the Tiber Island. It was handy to a ship arriving from Africa, and the offering may have been thought useful for completing the conquest of the Carthaginian territory. Regulus, when the incident occurred, was in the full swing of his first success, and had not yet achieved the fame that arose from his defeat and capture. Why the skin remained in the temple till the Numantine war and no longer I cannot divine ; but it may be that snakes became unpopular at that time, for in Julius Obsequens I find that the unlucky Mancinus was warned against going to Spain by a snake coming on board his ship at Genoa (ch. 24, ed. Jahn). Supposing that there is a bottom of fact in the story, what could this monster have been 1 A crocodile it assuredly was not, for the crocodile was well enough known in the Mediter- ranean world. I leave it to experts to decide this question : only remarking that water serpents in great rivers do even now reach a very large size. Bates in his Naturalist on the Amazons (p. 215, ed. 5) heard of specimens of the anaconda which measured forty-two feet ; and adds that the natives of the Amazons country universally believe in the existence of a monster water-serpent, said to be many score fathoms Digitized by Microsoft® PARALLELA QUAEDAM 181 in length, which appears successively in different parts of the river. ' They call it the Mai d'agoa — the mother, or spirit, of the water. This fable was doubtless suggested by the occasional appearance of serpents of unusual size.' Aristotle knew of great serpents in Africa {Hist. An. viii. 27. 6), which attacked ships as they sailed along the coast. On the whole there is sufficient evidence, I think, to make it possible that this was the survivor of a species even then aU but extinct. VERGILIANA THE SWANS IN AENEID I. 390 S. Namque tibi reduces socios classemque relatam nuntio et in tutum versis Aquilonibus actam, ni frustra augurium vani docuere parentes. aspice bis senos laetantis agmine cycnos, aetheria quos lapsa plaga lovis ales aperto turbabat caelo ; nunc terras ordine longo aut capere aut captas iam despectare videntur : ut reduces illi ludunt stridentibus alls et coetu cinxere polum cantusque dedere, haud aliter puppesque tuae pubesque tuorum aut portum tenet aut pleno subit ostia velo. This is not a simile but an omen ; we need not look for too exact a parallelism between swans and ships. This is one thing to remember in reading the passage ; another is that neither swans nor ships had at the moment reached their destination, though they are on the point of doing so. This is perfectly clear in the case of the swans, who at the moment of vision are ' laetantes agmine ', and also by the rather puzzling hne 398 (' et coetu cinxere polum cantusque dedere '), which if interpreted by the preceding line (' ut reduces illi ludunt stridentibus alis ') can only mean that they have recovered their formation and after circling round the sky, they have decided where to alight and are about to do so. To me it is quite clear that all the birds are still flying ; and if so, there is only one meaning for ' capere terras ', the one in which Servius saw no difficulty, viz. ' take in with their minds ', as in Georg. ii. 230 ' ante locum capies ocuhs '. Our word ' spot ', which will soon become classical, exactly Digitized by Microsoft® 382 VERGILIANA gives the meaning. In vain Mr. Page dogmatically insists that capere cannot mean this, and must mean that the leading swans have alighted on the earth ; he at once gets into hope- less confusion. Swans have a leader, as Virgil knew (xii. 250), and when the leader has fixed the point to make for, they can already look down on it as they draw nearer. ' Terras ordine longo aut capere aut captas iam despectare videntur.' No need to alter despectare ; the MSS. GMR, of which G goes back to the fourth century, are sufficient evidence. Now for the ships. Like the swans, they have not yet reached land ; they are ' making ' the port. Portum tenet need not mean that they have entered the harbour, but as in V. 159 ' iamque propinquabant scopulo metamque tene- bant ', where, as the meta was on the scopulum, it is plain that the boat was only about to reach, it (cf. xii. 754). And this is made certain by the words that follow, ' aut pleno subit ostia velo '. The sails are stiU unfurled, i. e. of the ships that are not in the van ; they cannot therefore have even almost reached the harbour ; but they are approaching it quickly, as the swans stiU flying look down on the land. portum tenet. Virgil uses tenere much as a Somerset friend of mine uses the word ' hold '. He loves to plan a long country walk to end in catching a train, with no time to spare ; and he always says we ' hold the train ', even a mile or two away from it, if we have still enough time left to get to the station. — R. S. C. THE HARBOUR IN AENEID III. 533-6 Portus ab euroo fluctu curvatus in arcum, obiectae salsa spumant aspergine cautes, ipse latet : gemino dimittunt brachia muro turriti scopuli refugitque ab litore templum. Here Mr. Page makes difficulties where there are none. Has he ever been to Lulworth Cove in Dorset ? If so, he will have seen the very scene Virgil describes. There the sea has broken through a rampart of Portland rock at one point only, and has washed away in a wide curve the softer deposits of sand and chalk within that rampart. On each side of the Digitized by Microsoft® VERGILIANA 183 narrow entrance the rocks of harder material stand firm — ' obiectae salsa spumant aspergine cautes ', and ' gemino dimittunt brachia muro turriti scopuM ', whUe the cove within is quiet (' ipse latet '). Even the temple is there ; for withdrawn a little way up the low cliff to the west of the cove is the remnant of the ancient abbey of Bindon, which stood here till it was moved to Wool. I do not know whether Virgil is describing a scene which he knew personally ; but at any rate he is describing one which is perfectly natural, and needs no elaborate explanation, no forcing of simple language. NOTE ON DIDO AND AENEAS The tendency in modem times has been to throw the whole blame upon Aeneas ; very strong language has been used, e.g. by Fox, Matthew Arnold, and many others. To the chorus of execration Dr. Glover is an exception ; and this is simply because he is less affected by the ideas of the nine- teenth century, and understands Roman feeHng better than most. Moreover, he understands the character of Aeneas as a whole, and is not so whoUy absorbed in the passion of the fourth book as to lose his balance in judging of the question. Aeneas is throughout the poem, and to its very end, the impersonation of family life and affection, and of the other ties which bind civilized men together, friendship, hospitality, good faith, and justice. This is why at the very end of the poem all that was best in the pure and wholesome Italian tradition of family life and social relationship is placed in striking contrast with the wantonness of an individual warrior without scruple and without mercy. But this is no doubt what the modern reader does not readily see, because such impersonations are not personally attractive. I suppose that David is a much more popular Bible hero than Moses, who is the impersonation of a Jewish ideal, just because David is a human being of warm feelings and frequent failings. I am not of course comparing Moses and Aeneas ; I only mean that it takes both knowledge and insight to see the human traits in either of them ; they are too serene and quiet Digitized by Microsoft® 184 VERGILIANA for the casual reader to catch at a glance. These traits in Aeneas are mostly to be found in the last six books, which are not nearly so weU known as the earlier ones. But I doubt if we can do full justice to Aeneas without a real knowledge of them, and I hope that in my recently published little volumes I may have contributed something to that end. Aeneas was a married man and a father when Troy was captured, though he was not the head of a family ; it would be as well if every one were to go back to the second book (559fE.) before beginning the fourth. There he tells Dido how at the death of Priam the probable ruin of his own happy family life suddenly flashed on his mind : At me tum primum saevus circumstetit horror, obstipui ; subiit cari genitoris imago, ut regem aequaevum crudeli vulnere vidi vitam exhalantem ; subiit deserta Creusa et direpta domus et parvi casus luli. This terrible prospect so maddens him that a wild desire for once gets the better of him to slay the woman who is the cause of aU the trouble. ^ When he leaves the burning city mth his father on his back, his little son grasping his hand, his wife following, the disappearance of Creusa drives him back in search of her at the risk of his Hfe. This he teUs to Dido. I need not quote what follows, for Virgil has put all his heart into it, and I would advise all who wish to form an opinion about the Dido story to begin by laying it to heart. I will only draw attention to the hues (781 ff.) in which the promise is given him of a new and royal family life in Italy, on the banks of the Tiber.^ His father Anchises, who up to his death continued to ' I have not the smallest doubt that this passage (567 ff.) is really Vii-gil's, and that the first editors cut it out simply because of the apparent violation of Pietas which the slaying of a woman would have implied. " It is sometimes asked, why should the new city be at Rome rather than in Africa ? The answer is that all foundations of cities were matters of religion ; and that in this case the oracles had so directed. If it be answered, Why trouble about the oracles ? all I can say is that the foundation of colonies without the oracle was in Greek history unthinkable ; and Cicero had reminded the Romans of this (de Div. i. 1. 3). We cannot be just to Virgil if we forget such things as this. Digitized by Microsoft® VERGILIANA 18S govern the family group, was the next to leave him. Let us remember that the visit to Carthage happens just when this loss has come upon him. He is certainly meant to appear at the first somewhat helpless ; he loses heart at the storm (i. 92 fif.), and then, landing in Africa, falls unwittingly a victim to the superb charms of Dido — charms used with consummate skill. I might go on to show his recovery after that fall, and the alteration of his character in the later books ; but this I have already done in the eighteenth chapter of my Religious Experience of the Romans, to which I may refer the reader. Suffice it to say here that courage combined with gentleness, tenderness combined with justice, are shown again and again in his relations both with friends and enemies ; with his own boy, with Pallas entrusted to his care by his father Evander, and even by the death of Laiisus, the son of his most ferocious enemy, which brings all the feelings of a father welling up in his mind. Let us turn to Dido. We know that Virgil altered the story of Dido to suit his own purpose, and the change should teU us what that purpose was. In the traditional form, the queen, ' who had resolved to remain a widow after the mm-der of her husband Sychaeus, was pressed, not without threats, by a neighbouring prince, larbas, to become his wife. Her own subjects urged her to comply, but she, professing that she was going to perform some rite which should absolve her from her vows, erected a great pile of wood near her palace, which she kindled, and then threw herself into the flames '?■ Now Virgil makes Dido fall violently in love with Aeneas ; when she has induced him to fulfil her passionate desire, he is warned by Jupiter to forsake her and make for Italy. This leads to a frenzied outburst on the part of the queen, which ends in her suicide while Aeneas sails away. To me it is quite clear that Virgil altered the story in order to contrast the fury of ungovernable love, love of the animal type, with the settled order, affection, and obedience, of the Roman family life. Let me explain this. Up to the departure of Aeneas the true nature of Dido's ' NettlesMp, Vergil, p. 58. Digitized by Microsoft® 186 VERGILIANA passion is not revealed, yet it is clear that there is something wrong about it. She had vowed never to marry again ; but already at the end of book i the arts of Venus and Cupid have begun to destroy her resolution (i. 715-22), and she cannot sate her passion for her guest. When the hero's tales are finished, at the beginning of book iv, her sister Anna leads her on to beheve that he may be kept at Carthage (iv. 51), that the ghost of Sychaeus would not mind (34), and so on ; and that sacrifices to the gods may legitimatize a union with Aeneas (' pacem per aras exquirunt '). Now mark the poet's comment : Heu vatum ignarae mentes ! quid vota furentem quid delubra iuvant ? est moUis flamma medullas interea et tacitum vivit sub pectore vulnus. (iv, 65 ff.) The point is in the word ' furentem '. Dido was possessed by a spirit of madness (' heu furiis incensa feror ', 376), which was quite incompatible with serious marriage. This is ' human passion bent on its own fulfilment in contempt of the gods, and ending, as it can only end, in infatuation and ruin '.^ This madness is of course increased tenfold when she discovers that the object of it is deserting her. The great scenes in which she gives vent to her indignation^ — scenes which have won the pity of the world — are in reality meant to exhibit a spirit and temper entirely out of place in Roman family Ufe. And she knew perfectly well that she was sinning. The terrible lines (450-73) in which the poet has described her dreams, prove that she knew she was breaking a vow, sinning against pudor. The modem reader asks. Why should she not have settled down with Aeneas — why should they not have reigned at Carthage together ? The usual answer given is that the fates had decreed otherwise, and that the great deity of the future Rome was determined to save Aeneas for Italian civilization. But this would not, if I am not mistaken, have been the whole of Virgil's answer. He would have said that Dido's character, '■ Nettleship, Essays in Latin Literature, p. 125. Digitized by Microsoft® VERGILIANA 187 as he conceived it, was utterly incompatible with Italian ideals. She does not understand the combination of virtues which made up the ideal Roman matron. She has lost all power of self-restraint ; not coniugium, but Amor, is her aim ; ' improbe Amor, quid non mortaUa pectora cogis ? ' Yet in the background of her mind there is always the thought of her Sychaeus (460 ff.), and when she meets Aeneas in the shades, it is to the company of Sychaeus that she turns for comfort. Tandem corripuit sese atque inimica refugit- in nemus umbriferum, coniunx ubi pristinus iUi respondet curis aequatque Sychaeus amor em. (vi. 472 ff.) Virgil had good reason to draw this terrible picture of an infatuated woman. In his own day Cleopatra had poisoned the mind of one, if not two, great Romans, and the escape of Augustus from her charms was a matter of enormous impor- tance in the history of Rome, as all his contemporaries knew. Nettleship believed that he was following the idea of certain Greek tragedies where the conflict of individual incHnation with the divine will is represented (pp. 124 ff.). But the passion of ungovernable love, even without the treatment of it in tragedy, must have been before the poet's mind ever since he studied Lucretius, or wrote his own third Georgic. It was common to man and the animals ; not so the ordered life of the family, which to the Roman of the older and nobler type was as an institution from heaven. This explains why Venus is represented as bringing the madness about through the agency of Cupid (bk. i. 657 ff.) : she knows that such love is incompatible with the Ufe to which Aeneas looks forward, and that Jupiter will take care that he is not led by it to give up the hope of Italy. There are one or two subsidiary questions on which it may be as well to say a word. For example : Was Aeneas himself in love with Dido 1 Of this there is not a trace tUl after the fatal meeting in the cave. It is then expressed in the passage beginning ' At plus Aeneas ', which has been so obnoxious to some critics. 'Multa gemens magnoque animum labefactus Digitized by Microsoft® 188 VERGILIANA amore, lussa tamen divum exsequitur classemque revisit.' His emotion is also visible in lines 448-9 : ' magno persentit pectore curas : Mens immota manet, laorimae volvuntur inanes.' But he was only half won. ' As Caesar was half won by Cleopatra, Aeneas is half won by Dido ', writes Nettleship. I was quite wrong when I wrote of his ' passionate love for Dido ' in Religious Experience, p. 416. Then again, what are we to say of the part of the gods in all this, more especially that of Juno and Venus ? I confess that it is to me one of the least attractive features in the poem. There is no sort of pleasure in reading of plot and counterplot ; and we cannot bring our modern minds to think of them as standing for right and wrong designs of Powers over whom man's free will has no control. But so much as this may be said : Juno stands all through the Aeneid for the Power working against Rome, and on behalf of Rome's most deadly enemy. Her plan is a treacherous one ; she does ' not mean real marriage of a strict Roman type. ' Conubio iungam stabili propriamque dicabo,' (126) like 99 ff., is not said of Roman marriage, and so Dido takes it : ' coniugium vocat, hoc praetexit nomine culpam ' (172). The counterplot of Venus is the strangest thing of aU. She sent Cupid to make Dido fall in love with Aeneas — not the respectful love of Roman marriage, but the animal love of a ' furens '. If she can succeed, Aeneas is saved ; because this wild love, as I have said, is incompatible with true marriage ; and Aeneas is not the man to enslave himself to it, or to renounce the joys of sober family comfort that he has known so well. It will drive him away ; Dido knew this, and Anna expressed it clearly for her in iv. 51. Anna sees the condition of her sister's mind, and the necessity of detaining Aeneas by every art in her power. They succeeded in driving him away, as Venus intended, and this explains his precipitation. Not a moment was to be lost. But the escape was a narrow one. He was already beginning his functions as co-founder of , Carthage when the messenger of Jupiter found and roused him wit?i that stirring speech that sounds once again the great fugue subject of the Aeneid. Digitized by Microsoft® VERGILIANA 189 NOTE ON AENEID V. 5-6. AMORE POLLUTO Interea medium Aeneas iam classe tenebat certus iter fluctusque atros Aquilone secabat moenia respiciens, quae iam infelicis Elissae eonlucent flammis. quae tantum accenderit ignem causa latet ; duri magno sed amore dolores pollute, notumque furens quid femina possit, triste per augurium Teucrorum pectora duount. Polluto is usually taken as meaning spoilt or ruined by Aeneas's desertion ; amor being ' sacred ', as Mr. Page says. But was amor sacred in any Roman sense ? Far from it. Mackail translates ' trampled ', which seems to me far from the mark. Servius has ' laeso '. But these interpreters do not seem to notice that Virgil is speaking of the effect of the fire on the mind of the Trojans, who would be far from attributing the ' pol- lutio ' to their loved leader. To me it seems more Hkely that Virgil is here reflecting his own view of the mischief, viz. that Dido's love was, or had become, poisoned by her own madness ; ^ hence the words that follow, ' furens quid femina possit ', which cannot be separated in meaning from those which precede them. The mischief was in Dido's own heart ; with true oriental ungovemableness she had given full way to her passion, and deserted, in spite of pvdor, her vows of loyalty to Sychaeus's 'memory. Even at the end of bk. i we find that ' vulnus aUt venis et caeco carpitur igni '. She becomes madly careless of her name and fame. ' Heu fiuiis incensa feror ' (376), and when Virgil uses the word furiae he always means some sort of violent disorder of the mind (e. g. xii. 101, 946). She sees horrible omens, dreams dreadful dreams ; she pretends magic and is half in earnest about it. The whole course of her love had gone wrong ; and chiefly of her own doing. This is in the poet's mind ; he has used all his resources to draw a woman whose real nature was ' Weise, Langtiage and Character of the Roman People, Eng. trans., p. 24 ; ' amor was to the Roman a malady, a consuming fire, a fatal wound.' In the laudatio Turiae (see ante, p. 126 ff.) the word is never used ; the love of husband and wife is pietas, Concordia, constantia, &c. Digitized by Microsoft® 190 VERGILIANA that of Medea, or Clodia, or Cleopatra ; women whose nature was utterly incompatible with all Roman ideals of family and social hfe. True, Aeneas was himself to blame, but in my opinion more sinned against than sinning. AENEID V. 197 (Cf. IL. XXIII. 408) hoc vincite, cives, et prohibete nefas. The commentators from Servius downwards have under- stood the word nefas in this line to mean disgrace or shame, with the exception of Henry, who very properly observes that they all seem determined to emasculate Virgil, that is, to make him mean something commonplace when he really means more. Certainly nefas does not mean shame or disgrace simply, but must always have a reUgious allusion implied in it. Here the religious allusion can be easily detected. These games were as much funeral games as if Anchises had just died ; they therefore had an unmistakably religious character ; ^ this is perfectly clear in the account given of them in lines 72 ff . Secondly, the prizes are of a religious char- acter, e. g. the sacred tripods. Thirdly, the prayer and vow of Cloanthus, and its reception by the deities of the sea, add to ' the rehgious atmosphere. Lastly, the one competitor who failed was rewarded by Aeneas with a kind of consolation prize, of which the object was, I think, to avoid, not shame or disgrace, so much as a failure in reUgious duty. In the competition of shooting Aeneas is careful to give every competitor a prize. Notice how one is reminded of the connexion with Anchises consistently throughout. Aeneas is ' satus Anchisa ' in 244 and 424 ; ' Anchisiades ' in 407 ; in 536 the prize is a bowl which belonged to Anchises ; 550, ' ducat avo turmas '. I may also note the curious appeal to God's will in 465 S. ' This religious character, if not entirely absent, is hardly perceptible in Homer, II. xxiii ; which to a modem reader may seem the more natural of the two accounts. In Virgil it is plainly shown by the coronae (lines 72 ff. and Serv. ad loc.) ; and also by the tuba (113 and Serv.),.for the use of which in funeii/ls see Marquardt, Privatleben, p. 341. Digitized by Microsoft® VERGILIANA 191 AENEID IX. 15 Dixit, et in caelum paribus se sustulit aUs ingentemque fuga secuit sub nubibus arcum. The second line, as so often, explains the first ; or rather, helps it out. Iris opened her wings and soared heavenwards, taking her way along the great bow under the clouds. Mr. Mackail as usual sees into the poet's meaning ; ' flashed under the clouds in a long flying bow.' Iris was the rainbow, and as she progressed the bow shone out ; this must be why Servius explains secuit by duxit. By getting rid of the notion of cleaving or cutting we get a far more beautiful picture. So in X. 107 ' quam quisque secat spem ', where the idea is that a man goes along with his hope, not simply follows it. H. Nettleship, in his Essays in Latin Literature, p. 370, in a note of great interest, concludes that there is a fair amount of evidence for a Latin word seco, which meant originally ' to go ', ' to go after ', or ' to follow ' ; or as I should put it on Virgilian evidence, ' to go along with '. Servius quotes vi. 899 ' ille viam secat ad naves ', which seems to me to be an exact parallel to our passage ; meaning not that he followed a road, but that he went on his way. I am not sure that in vui. 96 ' viridesque secant placido aequore silvas ', the real meaning is not of the same kind. Oxa poet is faithful to nature, as usual, and manages to combine nature with myth very pleasantly. Turnus, gazing up from the earth, sees the cloud quickly vanish, and Iris and her bow with it ; then the bright sky appeared, brighter in contrast with the cloud. In the clear sky he sees strange things — the stars by daylight, or perhaps by the earliest hght of dawn, as Etruscan wise men had seen them : ' Vulsiniis prima luce flamma caelo emicare visa : cum in unum coisset, OS flammae ferrugineum ostendit, caelum visum discedere, e cuius hiatu vertices flammae apparuerunt.' ^ Without doubt Virgil knew Cicero, de Div. i. 42. 97 ' caelum discessisse visum est atque in eo animadversi globi.' What these apparitions ' Jvliua Ohsequens, Prodigiorum liber, ed. Jahn, oh. 52 ; a book not without its occasional usefulness. Digitized by Microsoft® 192 VERGILIANA were, globi vertices flammae, palantes stellae ', or whether ' they were anything but the dreams of an Etruscan, we cannot tell. None the less the picture is a beautiful one, and its effect on the solitary Turnus is as striking : sic effatus ad undam processit summoque hausit de gurgite lymphas multa deos orans, oneravitque aethera votis. , Lines 30 ff. : Ceu septem surgens sedatis amnibus altus Per taciturn Ganges, aut pingui flumine Nilus Cum refluit campis et iam se condidit alveo. This fine simile is Virgil's own ; but he is applying his knowledge of the Padus to rivers he had not seen. Mantua was just the place where the main tributaries of the Po (excluding the Adige)r combine to form a great silent stream ; but why he did not use his own river I cannot guess. Fullness and silence are the two points of the simile ; the troops collecting on the open field swell in numbers and slacken in pace, silently advancing, as the Ganges collects its great tributaries, and now ' too full for sound or foam ' flows with the gathered force of silent waters. Can any other language find such words to express a majestic idea 1 surgens — of volume, not overflow — sedatis, altus, per taciturn ! This is another proof of Virgil's way of looking at a river ; he thinks of it as a whole, tributaries and all. The seven tributaries of the Ganges may safely be left to the commen- tators. AENEID IX. 79 Prisca fides facto, sedfama perennis. This is almost the only real fairy story told by Virgil himself ; those in the third book are told by Aeneas, and the Cacus story in bk. viii by Evander. By a fairy story I mean one of those Mediterranean marvels, usually about the sea, storms, birds and so on, such as abound in the early books of the Odyssey. Heinze fancied that Virgil invented this one of the ships turned into nymplis ;^ but I cannot agree with him. This is a transforma- ' Virgil's epische Teclinik, p. 298. Digitized by Microsoft® VERGILIANA 193 tion story like that of the birds of Diomede, and stands quite apart from the ordinary run of divine miracles ; and from most of the metamorphoses in Ovid. The nearest approach to it in Homer is the change of a ship into a rock by Poseidon at the bidding of Zeus, Od. xiii. 140 ff., no doubt suggested by the shape of a particular rock. The tale took Virgil's fancy ; what its ancient original was we cannot guess, nor did it matter for him. The difficulty was to introduce it ; for he was not writing a fairy poem, but one of great seriousness. He does it very happily ; he appeals to the spirit of poetry, the Musae ; let them tell it. And then, before he begins, come the half apologetic words, ' prisca fides facto sed fama perennis '. He carefully avoids saying that his own age believes in such wonders ; but they love a good story still, whether they believe it or no. The words are a beautiful example of the Latin genius for condensed expression, and also of the fullness of meaning in Latin words. But prisca has been rather troublesome to the learned, from Servius downwards. The ' sed ' gives the clue to the shade of meaning wanted in prisca. Henry's ' old-fashioned ' will do very well, for it is not the story which is prisca, but the belief in it. ' Old-fashioned folks may still believe it, but we all like to listen to it.' ^ AENEID IX. 214 Solita aut si qua id fortuna vetabit. Henry proposed saltern for solita, but we must accept the manuscript reading, which after all makes quite a good sense. Servius's note shows that a difficulty was felt about it, no doubt because in his day (and long before) most of the slain in battle were recovered and buried on the field, and the bodies of all chiefs would be identified. But Virgil is no doubt right in assuming that in ancient times the usual fate of a slain man was to be lost, and to remain unburied. ' In the prisca fides of vi. 878 (Marcellus) the adjective has something of the same meaning ; righteousness hke that of his ancestors. 8252 j^ Digitized by Microsoft® 194 VERGILIANA , The idea of the cenotaph, which Virgil here introduces, is found also in iii. 304 and vi. 505. Cenotaphs were not common at Rome, so far as I know ; probably for the reason given above. The most famous instance is that of the young Caesars, grandsons of Augustus, who were worshipped with a regular ritual at their cenotaphs at Pisa, though their bodies were in the Mausoleum at Rome.^ But, as Marquardt * says, it was quite in keeping with the Roman idea of the monu- mentum, as distinguished from the Etruscan practice of burial in a house sealed up for ever, or only opened for new inmates. The desire not to be forgotten, which was urgent under the Empire in the gilds, is common to both Greeks and Romans, but is not expressed in Etruscan burial. THE EPISODE OF NISUS AND EURYALUS : WHY INTRODUCED HERE ? I suggest four motives : (i) The desire to put off typical fighting of the Homeric kind ; Virgil knew that it could not be treated again with the same felicity, and it was not after his own heart. ^ (ii) His love of boys just growing up, * an amorpius, as with this pair (296), love of young manly beauty, and the happy relation between elder and younger, (iii) A wish to imitate and surpass the Dolpneia of Homer, in which he was beyond doubt successful, (iv) There was here a chance to bring Ascanius forward ; see on xii. 87 fE. Being too young to fight, he might have fallen out of these last books. His part here is admirably conceived ; it is dignified, affec- tionate, and yet quite boyish ; e. g. his astonishing promises of rewards (263 if.), which so puzzled Gossrau that he thought the poet must have been slumbering, are simply the result of bojdsh enthusiasm. See below on 252 ff. ' C. I. L. xi. 1421 ; Dessau, Inscr. Select, i. 140. ^ Privatleben, p. 355, note. " See Aeneas o' the Site of Borne, p. 1. * Death of Turniis, p. 90, Digitized by Microsoft® VERGILIANA 195 AENEID IX. 252 ff. Aletes, annis gravis atque animi maturus, says to the youths about to start on their perilous journey : Quae vobis, quae digna, viri, pro laudibus istis praemia posse rear solvi ? pulcherrima primum di moresque dabunt vestri : turn cetera reddet actutum pius Aeneas atque integer aevi Ascanius meriti tanti non immemor unquam. A passage of great importance ior the mind of Virgil and his idea of the Roman mind at its best. Conington refers to Servius, Servius to Cic. Phil. ii. 44 ^ ('satis in ipsa conscientia pulcherrimi facti fructus '), but neither of them seems to feel the ethical force of the words, or of the appeal to ' mens conscia recti ' (i. 604). No earthly reward is called for in the first place — such things are purely secondary. ' Pulcherrima primum Di moresque dabunt vestri.' This is the true Roman spirit of the good old times, and it is the spirit too of men like Scipio Aemilianus, Cato the elder, the Gracchi, and many more, before the days of unnatural ambition for triumphs and supplicationes. Di moresque were the two pivots of old Roman social life, and a faithful pietas toward god and good custom needed no reward. Rare indeed, compared with ours of to-day, were rewards for Roman soldiers, who were fighting for the divine inhabitants of their city, and for the good traditions of the human ones. Read Gellius, v. 6 (de coronis militaribus), and especially the quotation from Masurius Sabinus in sec. 13 ff., and from Cato in sec. 25. Note that these words (252 ff.) are spoken by one who is ' annis gravis atque animi maturus ', and that the readiness of Ascanius to promise praemia, which so puzzled the rather wooden-headed Gossrau, is by way of contrast, as the enthusiasm of a boy, who seizes on the chance given him by Aletes (' immo ego vos, excipit Ascanius '), for the moment forgetting the di moresque. ^ Add Phil. i. 9 ; de Rep. vi. 8 ' sapientibus conscientia ipsa factorum egregiorum amplissimum virtutis est praemium '. This ia, however, the Stoic idea of virtus, which is like that of the old Romans, but not lelated to it. N 2 Digitized by Microsoft® 196 VERGILIANA AENEID IX. 275 ff. Ascaniua has just been promising Nisus an astonishing number of good things (praemia) if he returns safely. Then he turns to Euryalus : Te vero, mea quem spatiis propioribus aetas insequitur, venerande puer, iam pectore toto accipio et comitem casus complector in omnes. nulla meis sine te 4uaeretur gloria rebus : seu pacem seu bella geram, tibi maxima rerum verborumque fides. What does Virgil mean by venerande puer ? If he wrote the Culex, he had used the same expression of a boy who can, in my opinion, hardly be other than Octavius, afterwards Augustus.^ There the boy is also sanctus (26 and 37) ; and sanctus might be used of any praetextatus, for well-born children were all in some sense ' holy ' till they took the toga virilis. But venerandus is a still stronger word, and in Virgil always (except here) implies worship. As applied to Octavius in the Culex it can hardly have a reference to divine descent, for Octavius was not yet Octavianus ; but it may mean reverence for a superior in birth and quaUty, though not in age. Here it is used by a princely boy of one who is not much older than himself and below him in rank, but one whom his enthusiastic affection has made his model and leader. Euryalus is to be his adviser and counsellor in the days to come when he will reign over the new city. For Nisus he feels as a prince to one of his best chieftains ; but towards Euryalus he feels as a younger to an elder boy, and the com- bined affection and reverence produce the word venerande. The whole passage to 313 shows a depth of feeling rarely equalled in poetry. AENEID IX. 435 These three exquisite lines, says Mr. Page, ' are borrowed partly from II. 8. 306, partly from CatuUus, 11 . 22 (cf . 62. 39 ff .) '. If the reader will look out these passages, instead of taking them for granted, he will see that the essential beauty of the ' Ovlex, 25; Classical Review, 1914, p. 119. Digitized by Microsoft® VERGILIANA 197 Virgilian simile is not borrowed from either of them. In Homer the victim, Gorgythion, is a chance one, and our pity is not specially roused for him among a hundred others. In Catullus the thought is a lovely one, but comes at the end of an unpleasant poem. Virgil's hnes are inspired by the fact that it is a beautiful boy, integer aevi, ' with all his Hfe before him ', who is the subject of the comparison. Euryalus has been before us all through the story ; we can almost love him as Virgil loved him, as Ascanius loved him ; and the simile touches us keenly, we care not whence it comes. What was simply a charming thought in the older poets has become a perfect bloom, full of beauty, colour, and feeling. A word about the poppy. Whether Homer's plant was really a poppy I carmot say, but Virgil's undoubtedly was. It is an interesting fact that our own common scarlet poppy hangs its head only before the bloom unfolds itself, in order to protect its most valuable organs from the rain. When old enough to resist the showers it holds up the head and the bloom begins to appear. If our poppy was that of Virgil, or it other species have the same habit, perhaps he had noted the fact without fuDy understanding it. AENEID IX. 641 ff. Apollo sees Ascanius from a cloud, and addresses him in words which the boy cannot hear : ' Macte nova virtute, puer, sic itur ad astra ' ; then descends to earth in the form of Butes, and speaks to him in person. Servius asks, why Apollo ? and the answer he gives shows that such an unusual event as an appearance of Apollo not only needed an explana- tion, but did not get a very satisfactory one. It is, says Servius, because Ascanius had killed Remulus with an arrow ! What is the real reason, one naturally asks ? This is almost the only occasion in the Aeneid when Apollo is seen personally ; he is alluded to as a prophet half a dozen times. In xi. 785 Aruns prays to him as the deity of Soracte, and he answers the prayer. Nowhere else is he prominent, unless it be for an instant on the shield of Aeneas (viii. 704), and this is very good proof that Augustus had not asked Digitized by Microsoft® 198 VERGILIANA the poet to make much of his favourite deity in the Aeneid. Here he is introduced as a prophetic deity, beyond doubt. There are three distinct prophetic allusions : (1) ' dis genite et geniture deos ' ; (2) the closing of the temple of Janus, as in i. 294 ; (3) the assurance that Troy is not to be the centre of the future empire. All these are references to Augustus in the years when the Aeneid was being written. ' Macte nova virtute ' is exactly the Irish ' more power to you '. Bel. Exp., pp. 182 if. For ' sic ituc ad astra ' see The Death of Turnus, p. 70. AENEID IX. 731 ** Continuo nova lux ocuUs effulsit et arma horrendum sonuere, tremunt in vertice cristae sanguineae, clipeoque micantia fulmina mittunt. mittunt PR : mittit M. CUpei P On the whole the line as written above has the best manuscript support. But Conington and Mr. Page read mittit. Let us remember what is happening. Turnus has penetrated into the new city, and has escaped notice, till suddenly ' nova lux effulsit, &c.', and the essential ' frightfulness ' of Turnus becomes conspicuous. In the Gathering of the Clans we had this frightfulness fully described ; it consisted of two objects, the crest on the helmet breathing flame, and the smooth shield with its artistic rehefs. These are the two points in his equipment on which Virgil insists. So in x. 270 ff. with the armour of Aeneas ' stans celsa in puppi ' as he approaches the land ; ' ardet apex capiti cristique a vertice flamma Funditur et vastos umbo vomit aureus ignis.' It seems to me impossible to take the shield in the passage I am discussing as an enemy's (so my friend Professor Conway in litt.) ; it must be that of Turnus, as every one will allow who compares the two other passages. If it be that of Turnus, what do the four words mean, ' cUpeoque micantia fulmina mittunt ' ? I take cristae as subject to mittunt ; the blood-red crest flickers with motion of the warrior's head, and lights up the shield — passes on its lightning to the shield. Professor Digitized by Microsoft® VERGILIANA 199 Conway says that the shield, held on the left arm, could not catch the flashing crest. If that is a fatal objection — I do not myself think it is — ^we are reduced to taking mittunt in. a neuter sense, which does not recommend itself to me at all. The armour is not that of ordinary human make ; as in X. 270, its illumination is superhuman ; and I see no reason why crest and shield should not combine to flash their light. AENEID X. 1-117 This curious scene of the celestial senate, and the bitter speeches of Venus and Juno, followed by the strange decision of Jupiter, seem unnatural to the modern reader, though Virgil is but following Homer in the fourth and eighth books of the Iliad. Servius indeed says (comment on line 104) that LucUius in his first book did the same thing, and that ' totus hie locus ', by which I think he means the whole scene, was taken over from that poet. But Servius is apt to speak rashly of Virgil's ' imitations ', and it is almost inconceivable that our poet should have been thinking of what must have been a travesty, with intent to scoff, of Homer's sittings of the divine council. But why does he introduce the scene at all 1 Merely to find opportunity for that skiU in rhetoric in which'Macrobius thought him quite as much distinguished as in poetical expression ? No doubt this motive was present ; it is difficult for us to reaUze how strong was the demand for oratory, whether in prose or verse, among the Roman educated classes of that day. There was also, I think, the desire to postpone further fighting yet awhile, as uncongenial to the poet himself, and monotonous if indulged in. Perhaps yet another motive is t^t he wished to pause and bring the rights of the whole struggle before the minds of his readers at a critical moment ; for of the two speakers Juno is obviously meant to seem on the wrong side in her bitterness against the Trojans. However this may be, it is perhaps worth while to examine the whole scene as evidence of one characteristic of Virgil's poetic mind. Johnson complained of Paradise Lost that it comprises neither human actions nor human manners. ' The Digitized by Microsoft® 200 VERGILIANA man and woman who act and suffer are in a state which no other man and woman can ever know. The reader finds no transaction in which he can be engaged ; beholds no condition in which he can by any effort of imagination place himself. . . . Milton knew human nature only in the gross, and had never studied the shades of character nor the combinations of concurring or the perplexity of contending passions ' (Preface to Shakespeare). This is certainly not true of Virgil ; so far from it, that even in the council of the gods he finds room for Roman and Italian actions and manners, passions and rhetoric. So it is throughout the poem ; it is character in which Virgil rejoices, and in which he rejoices more and more to the very end of his poem. His characters are drawn on a grand scale, more especially that of Aeneas, but from time to time we find those delightful details which make them living and human for those who are not blind. These characters are governed by a supervising destiny, yet there is always to be seen in them that free action of the will which was a Roman's inheri- 'tance. And as I have noticed elsewhere, these characters, or some of them, are not static but grow and change ; Aeneas of the last books, a consummate warrior, is not the hesitating Aeneas of the earlier books, and his son grows in wisdom and stature all through the poem. It was then the natural instinct of Virgil to think of human beings, their moods and passions, not only in dealing with men but with gods. Whether this has any advantage for the reader I will not attempt to determine. Jupiter's attitude at the end of this episode has often been described as weak and vacillating, e. g. by Professor Mclnnes.^ To my mind this is not only a great mistake, but quite im- possible. The whole passage seems to me meant to show the god's tremendous power as compared with that of all other gods ; a truly Roman idea. Venus enlarges on it at the beginning of her speech, and in line 100 he is ' pater omni- potens, rerum cui prima potestas ' ; as he begins to speak there is universal silence, and he ends with that awful oath by the Styx on which I have commented in The Death of ' Claasiad Review, 1910, p. 169, where other alleged instances are given. Digitized by Microsoft® VERGILIANA 201 Turnus, pp. 141 £f. A few words on the relation of Jupiter and the other gods to the Fates may be useful here. After reading everything I can find on the subject I have come to a fairly definite conclusion ; but it should be clearly under- stood that Virgil is not wholly consistent throughout the poem, as was inevitable for a poet who had to adapt a decajdng system of polytheism to the philosophical needs of his story. I find that Virgil shared the view of almost all the thinking men of his day that there was a great driving force at work in the world, which was responsible for the rise and growth of Rome ; and that his word for this is usually fata, which however expresses it rather as it was interpreted to men by oracles and signs than as a thing in itself. It is in fact the Stoic idea of Destiny, and might be thought of as the will or plan of the Soul of the universe. Man was subject to it, but none the less his will was free ; a view which was natural and traditional with all educated Romans. But what of the gods, who are after all only a superior sort of men ? Were they subject to it ? Undoubtedly the deities in the Aeneid could not effectually oppose it, though they might hamper and delay. Juno and Venus for example had different views of what it ought to be, the one taking the side of Carthage, the other of Rome, and quarrelling between themselves they ask a decision of Jupiter. Jupiter clearly says, ' Enough of this fooKng '. But now comes the crucial question ; was Jupiter himself capable of making a decision outside the driving force of destiny ? Beyond aU doubt he stands far above the other gods, partly owing to his kinship with the Zeus of Homer, partly perhaps to his Roman inheritance as a heaven god ; but whatever the cause, no other deity has anything like the power over human life which he has. Can we, however, say either that he can overrule destiny, or that his will is identical with destiny ? On the whole I think the answer must be No. If we read carefully his great speech in bk. i. 257 ff., though it is by no means perfectly clear, I think that the general result is that Jupiter is rather the high priest (so to speak) of destiny than identical with it; as also in i, 375-6 -he rather consults it Digitized by Microsoft® 202 VERGILIANA than controls it. True, he sometimes uses language which seems to put him in the place of destiny, as in i. 278 ; but that should not be a serious difficulty. A prophet may not un- naturally speak as if the events he foretells were of his own making. In the passage before us he seems to claim the power of interfering with destiny, but will not exercise it. His last words mean ' destiny will find its own way ' (not, will find out a way) ' without any help or interference from me '. The two goddesses have power enough to hamper the action of destiny, and Jupiter refuses to check them by deciding for one or the other. The evolution of things will work out most satisfactorily if the squabbles of goddesses be left to take their own course. This is really, as Servius saw, equivalent to a decision in favour of Aeneas and Rome ; and the reason of it is that Juno must be allowed to have her turn with Carthage in a struggle for world-power. Had Jupiter decided to sjippress Juno altogether, there could have been no Punic Wars ; and that he could have suppressed her I can have no doubt. In the difficult lines 107-12, it is best, I think, to foUow Henry, who puts a full stop at the end of line 108, and encloses the first three words of 111 in a parenthesis. AENEID X. 287 ff. Interea Aeneas socios de puppibus altis pontibus exponit. multi servare recursus languentis pelagi et brevibus se credere saltu, per remos alii, speculatus litora Tarchon qua vada non spirant nee fracta remurmurat unda sed mare inoffensum crescenti adlabitvu" aestu. . . . Landing by pontes was the orthodox way, but it took time ; and many, not caring to wait their turn, jumped into the shallow water as the wave retired, or sUd down the oars. This last mode of landing amuses Mr. Page ; but he has forgotten the length and size of the oars of a ship of Virgil's day — ten to thirteen feet — rather too long either for use as a walking-stick or (as Heyne fancied) a leaping-pole. In the next line spirant is such an expressive and beautiful Digitized by Microsoft® VERGILIANA 203 word, and is so fully borne out by words that follow, ' nee fracta remurmurat unda ', that it is a pity to find sperat in our Oxford text. True, there is good manuscript authority for sperat ; but then the copyists would have found spirant rather beyond them. Henry's note is here excellent (Mr. Page follows him without acknowledgement) ; quoting Lucr. vi. 890, of a spring bubbUng up in the sea and crinkling the surface, and Seneca, Thyestes, 755, of the heaving of the arteries in a man. But Virgil himself decides the reading : Georg. i. 327 ' fervetque fretis spirantibus aequor '. Here Keightley notes : ' By spirantibus is meant the foaming and boiling up of water when driven against land or rocks.' Here I think it rather means the slight disturbance of the surface caused by a rock or stone below it ; such a disturbance as is well known to fishermen, suggesting that a trout is lurking in the lee of a stone below the surface. The next line, the last I have quoted above, is, in contrast with the foregoing ones, one of Virgil's most expressive and beautiful ones. AENEID X. 356 ff. Magno discordes aethere venti praelia ceu tollunt animis et viribus acquis ; non ipsi inter se, non nubila, non mare cedit ; anceps pugna diu, stant obnixa omnia contra. These curious lines are almost unnoticed by the comihen- tators. I suppose they are not troubled as I am to discover what aerial phenomenon the poet is reaUy thinking of. He does not seem to be imitating II. xvi. 765, where Eurus and Notos are at strife in a mountain ravine, breaking the trees, nor can I find the original among the storms of Lucretius. The contending armies make no way, either of them ; ' expel- lere tendunt Nunc hi nunc iUi ; haeret pede pes densusque viro vir ■'. When does suc^ a balance of forces occur in nature ? A passage in Sir A. Geikie's Love of Nature among the Bomans, p. 214, has suggested to me that the poet may have been thinking of the solemn quietude that sometimes precedes a thunderstorm, a phenomenon which he must often have noticed in Italy. It sometimes occurs in this coyintry ; see Digitized by Microsoft® 204 VERGILIANA e. g. an account of the extraordinary storm of August 9, 1843, described by the Rev. J. Jordan in his History of Enstone, p. 401 : ' Sitting in the house I endeavoured, by opening windows and doors, to obtain a draft, but the air was stagnant. . . . The storm rose very slowly and majestically, and directly in the teeth of the wind, which, though imperceptible in the slightest degree, was indicated by the weathercock to be in the east, while the storm came from WNW.' ' Anceps pugna diu, stant obnixa omnia contra.' The introduction of the sea in 358 is puzzling ; it is an imaginative touch added to the familiar phenomena. AENEID X. 362 ff. The geography here is not very clear, and has unfortunately been confused by Boissier, who imagined that the Arcadian and Etruscan cavalry here mentioned had come by land from Agylla, crossing the Tiber. Of that I see no sign ; they had come with their horses on board ship (see 181, where Astur, ' equo fidens ', was on board), and landing rather farther down the coast than the rest, were pushing up one of the rocky ravines or defiles that lead from the sea into the hilly region. Pallas was with them, or watching them, and he saw that they had dismounted on account of the rough track, and were suddenly attacked in front. In giving way they must make for the sea ; hence his words ' ecce maris magni claudit nos obice pontus, Deest iam terra fugae : pelagus Troiamne petamus ? ' The sea was behind them ; the new city Troia but a little to their left rear. ' Don't go back ; you have only the sea behind you ; you must cut your way to the camp through the enemy in front of you.' There remains the difficulty that they were near the Tiber, since in 420 ff. PaUas addressed the river-god ; but as Pallas did actually bring them victoriously through the foe, they would be close to the river, and in touch with the new city. AENEID X. 537 £f. This is to me one of the strangest passages in the Aeneid. ■ A priest of Apollo and Diana (' Phoebi Triviaeque sacerdos ') Digitized by Microsoft® VERGILIANA 205 is fighting on the Italian side against A^eas and Fate, for the two deities whom, but two years after Virgil's death, Horace had to treat as the chief deities of the Carmen saeculare, at Augustus's bidding. Aeneas sacrifices this priest, ' lapsum- que superstans Immolat ingentique umbra tegit ' ; and a trophy of his spoils is dedicated to Mars under his ancient cult-title of Gradivus, of which the meaning is lost. If we read insignihus albis with Probus, Virgil may have been thinking of the white dress of Greek priests (art. Sacerdos in Diet, of Antiquities, 572 b) ; no Roman priest wore white. But I think we should keep the armis of the manuscripts. If so, and as the arma were made into a trophy, the fact that he was fighting may have been fatal to him, priest though he was. Or it may be that an enemy priest had no claim to be spared on the field of battle. In xi. 768 ff. Chloreus is ' sacer Cybelo oUmque sacerdos ', i. e. at Troy ; but he is not killed by Camilla. Cf. vi. 484 : ' Cereri sacrum Polyboeten.' Ingenti umbra tegit : Henry is clear that it was Aeneas's body that threw the shadow. I think so too ; but the idea of death may be also impUed. It is obvious that Aeneas's wrath is terrible when the battle-fever is on him ; his speech 557 ff. is almost too much for us. ' Onerabit membra sepulcro ' reminds us of Byron's ' On thee shall weigh no ponderous tomb '. AENEID XI. 5 Ingentem quercum decisis undique ramis Constituit tumulo fulgentiaque induit arma, Mezenti ducis exuvias, tibi, magne, tropaeum Bellipotens. The description of the trophy follows. Servius says ; ' Constituit tumulo, in coUe : quia tropaea non figebantur nisi in"eminentioribus locis. Sallustius de Pompeio : Devictis Hispanis tropaea in PjTeneis iugis constituit. Ex quo more in urbibus tropaea figebantur in arcubus exaedificatis ' (i. e. when the arch was completed they put the trophy on the top). Does this statement of Servius answer to the fact so far as we know it ? Digitized by Microsoft® 206 VERGILIANA The evidence is slender, but the point is interesting as bearing on the origin of the triumphal arch. The list of such arches begins with Augustus, but we need not suppose that there was nothing of the kind before his time ; the principle that trophies must be erected in conspicuous positions might lead to something in the nature of an arch before it was intro- duced into the city by Augustus. We know of three arches erected in Augustus's time, one at his triple triumph in 29 B. c, one in 20, when he had recovered the standards of Crassus, and one after the recovery by Tiberius of the standards lost by Varus. Only the second of these is known in any detail ; a denarius of 19 B. c. (Cohen, Monnaies imperiales, nos. 82-4) has Augustus on a quadriga on the top of an arch, with Parthians presenting him with a military ensign and an eagle ; no bas-reliefs adorned this arch so far as we can see, and the quadriga on the top with Augustus may have been an innovation in place of a trophy. In the provinces triumphal arches sometimes had among their ornaments sculptured trophies, as at Orange. ' Arcus cum tropaeis ' is the expression of Suetonius {Claudius 1) for one erected to Drusus the elder on the via Appia ; and 'A\j/(s TpoTTaLO(l)6pos occurs in Dio Cassius, xlix. 15 ; li. 19, &c. If we had more representations of these earlier triumphal arches, we might be able to confirm Servius's statement ; as it is we must be content with guessing. My information as to these arches is derived from Courbaud, Le Bas-relief romain, p. 116, who discusses the date of the arch of Orange on p. 331 ff. He beheves it to be of the time of Tiberius, and to reflect the style of the Pergamean sculp- tures. See also the same work, p. 373. He thinks that the architrave was an addition to the arch ; but I am not sure that the process was not exactly the reverse, and that the Porta triumphalis, originally a iugum perhaps, was not the origin of the triumphal arch, combining with the notion that a trophy must be elevated. Courbaud allows that triumphal arches properly speaking did not exist in Greece ; would it have occurred to the Romans to imitate the Greeks in this particular form ? I doubt it. Digitized by Microsoft® VERGILIANA 207 NOTE ON VIRGIL, AENEID XI. 160 ' Vivendo vici mea fata ' In a paper in the Classical Qvurterly for January 1917, by Miss L. E. Matthaei, on the Fates in Virgil, there is a passage which so completely puzzled me that I fear I wasted some time over it ; for I find after six months that it baffles me as much as ever. I am greatly indebted to the article in many ways, and have constantly had recourse to it ; but I now see clearly, what I suspected at the first reading, that Miss Matthaei has put more undigested philosophy into Virgil's mind than ever was there when he was writing the Aeneid, and has made too little allowance for the feehng and language of a poet. On p. 12, commenting on Aen. xi. 160, in the lament of Evander for his slain son (' Vivendo vici mea fata '), Miss Matthaei says that there is in ' this very odd expression ' a contrast ' very dimly expressed between what Hfe is and what it should be ', and fate is made to stand for what it should be, its best possibilities. With this she contrasts x. 154, where the Etruscan people is said to be libera fati, because they have accepted the leadership of Aeneas, as they were directed to do by a haruspex (vii. 499) ; they were free, that is, in the sense in which a man was free of a vow when he had performed it. Then comparing the two passages, she finds that in the one ' fate is what life ought to be contrasted with what life is ' ; whUe in the second ' fate is what hfe is con- trasted with what it ought to be '. Well may she ask what stable notion can be rescued from such a contradiction ! It convinced her that she now had her finger on the whole gist of the Virgilian problem of the fates. ' I believe that Virgil inherited a gloomy and pessimistic definition of fate, as of a malignant pursuer of men, a power to be feared and placated, but without moral inspiration or elevation of any kind. . . . But the Romans had always struggled against this miserable doctrine, and foremost among them is Virgil. The consequence is that he is always swaying between the thought of a moral world and the thought of a sad and unjust world.' Digitized by Microsoft® 208 VERGILIANA This attitude of mind is of course common to all who give any sort of serious thought to the changes and chances of this mortal life ; but Miss Matthaei goes on to develop her view that VirgU was a pessimist in spite of his better instincts, and the Aeneid a pessimistic poem, in which fate makes mockery of humble human effort. This is a view which I cannot possibly share, after an intimate acquaintance with our poet for over half a century ; and I doubt whether Miss Matthaei would have reached such a conclusion if she had not begun by mistaking the meaning of the word fata in the expression ' Vivendo vici mea fata '. What the meaning really is is made quite clear by Dr. Henry in one of his best notes. Fata is not here used in any metaphysical sense, but simply as we use the word ' lot ', both the Latin and the English words beiQg used for the span of life, though originally denoting the idea of a method by which the future might be foretold. All that Evander means is that he has reached beyond the ordinary span of human life, his wife and son having gone before him ; just as in xii. 395, we find proferre fata used of seeking to prolong a man's life. And all the other passage means is that the Etruscans have obeyed the utterance of the soothsayer who bade them look for a foreign leader. In both cases there is a reminiscence of Virgil's favourite Euripides.^ Henry quotes Alcestis, 939 : eyoi 8*, bv ov xp'j fv", 7rape\s to iji6pin}),ov, \v7rp6v Sidfia ^iotoV apn fiavBdva. and 694 : ail yoiiv dvai8S>s Sifiid)(ov to fifj daveiv, KOL ^^s irapeKSiov t7]v TTeTrpa^evrjv rvx^^t TavTtjv KaroKTas. Conington quotes for ' libera fati ' Eur. Phoen. 999 : 01 /icV 6ecT(f)aT