STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world by JSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. THE ORIGIN OF THE CULT OF THE LARES By Gobdon Laing In a recent number of the American Journal of Archaeology 1 Miss Margaret Waites revives ^he controversy in regard to the origin of the cult of the Lares. Adopting in substance the view held by Samter, De-Marchi, Rohde, von Domaszewski and others, she traces it to the worship of the Lar familiaris, in whom she recognizes "a good spirit, closely attached in each case to a particular family, to its dwelling and to the territory immediately silrrounding the house." The worship of this spirit, she argues, "contains many features which seem to point to a chthonic cult and which imply that the Lar was originally worshiped as the spirit of the ancestor who had founded the family and still watched with devotion over the fortunes of his descendants." One's immediate reaction on reading this article is that among all the theories of cult origins that die hard, this ancestor theory of the worship of the Lares has one of the first places. It is probably as old as Plautus, certainly goes as far back as Varro, and has the support of other Roman writers; it has established itself in the commentaries on many Latin authors, is found in more than one dictionary of antiquities, and is defended even by those specialists in Roman religion and folklore whom I have mentioned above. Within recent years, to be sure, its prestige has suffered in no inconsiderable degree through the opposition of Wissowa in his article in Roscher's Lexicon, in his volume Religion und Kultus der Romer, his discussion in the Archivfur Religionswissenschaft, 2 and in other places. More- over, Warde Fowler, after a grudging admission of its possibility in his Roman Festivals, 3 has definitely rejected it in his Religious Experience of the Roman People (pp. 77-79). It is disregarded by Clifford Moore in his incidental reference to the cult in his book on 1 XXIV (1920), 241-61, "The Nature of the Lares and Their Representation in Roman Art. " 2 VII, 42-57. 3 P. 337: "The Lares, who may have been the spirits of dead ancestors." [Classical, Philology XVI, April, 1921] 124 The Origin of the Cult of the Lares 125 The Religious Thought of the Greeks, 1 and by Fox in his volume on Greek and Roman Mythology. Yet in spite of this trend of contem- porary opinion it raises its head once more in this article in the American Journal, confident, unabashed, unafraid, and set forth with all that plausibility which Miss Waites's essays invariably show. I must confess that I was surprised to see it. I have never won- dered at finding it in editions of Latin authors. One gets trained to seeing anything there. Nor need its appearance in general works of reference cause anyone dismay. Many of these are scissors-and-paste compilations and tend to become vast repositories of error. But its advocacy by so brilliant a scholar as Miss Waites and its pub- lication in so sober a periodical as the American Journal of Archae- ology almost drive me to the conclusion that it is one of those academic vagaries which we shall always have with us. In my opinion the theory has nothing to stand upon; and while I do not agree with all the points which Wissowa — for he, as Miss Waites rightly says, is its most vigorous opponent — has attempted to make against it, not one of the arguments urged in its favor is convincing. That the theory, in one form or another, appears in Roman literature is a familiar fact. We find it in Varro in fragments pre- served by Arnobius 2 and Augustine 3 respectively, in Verrius Flaccus, 4 and later in Apuleius 8 and Martianus Capella.* It is true also that 1 P. 228: "In this first period the religion of the family also was already fixed in the form which it retained to the end of antiquity. Vesta of the hearth-fire, the Penates of the larder, the Lar of the farm, the Genius of the pater familias, were the divine powers which were worshipped in the house. Rites were paid also to the Manes, the shades of the dead." 2 Hi. 41: Varro .... nunc esse illos manes et ideo Maniam matrem esse cog- nominatam Larum, nunc aerios rursus deos et heroas pronuntiat appellari, nunc antiquorum sententias sequens Larvas esse dicit Lares, quasi quosdam Genios et functorum animos. 3 De civ. dei vii. 6 : inter lunae vero gyrum et nimborum ac ventorum cacumina aerias esse animas, sed eas animo non oculis videri et vocari heroas et Lares et Genios. * Paul, p. 121 : Lares .... animae esse putabantur hominum redactae in numerum deorum; ibid., p. 239: deorum inferorum, quos vocant Lares. 6 De deo Socrat. 15, p. 15, 15 ff. Lutjoh.: ex hisce ergo Lemuribus qui posterorum suorum curam sortitus placato et quieto numine domum possidet, Lar dicitur f amiliaris. 6 ii. 155: hie [cf. Varro above, inter lunae gyrum et nimborum ac ventorum cacumina] igitur Lares, hie post membrorum nexum degunt animae puriores, quae plerumque si meritorum excellentia sublimantur etiam circulum solis ac flammantia saepta transiliunt. 126 Gordon Laing the word Lares in a number of cases is translated by the Greek fjpues. 1 None of these passages takes us farther back than the last century of the Republic. But Miss Waites 2 sees evidence of an earlier date in Plautus, Merc. 834, familiai Liar pater, and in the Lar familiaris of the prologue of the Aulularia. With Leo she thinks that in these two places Lar is a translation of ?}pcos. This is prob- ably correct, and the theory may easily be as old as the translators and adapters who played so important a part in Roman literature in the second and third centuries B.C. These writers were continually confronted with the problem of equating Greek with Roman gods. At any rate the belief is older than Varro, for there is no reason to doubt Arnobius' statement (iii. 41) that Varro in expressing the opinion that the Lares were the souls of the deceased was following the views of the ancients (antiquorum sententias sequens). But it makes very little difference when the theory is first referred to. The only question with which we are concerned is whether it contains the right explanation of the origin of the cult. Let us look at this ancient testimony for a moment. It is not, on analysis, very formidable, consisting, as we have seen, of half a dozen passages in authors and a slightly larger number of references to the equation of Lares and ijpwes. But it is even less significant than it appears at first sight, for an examination of the content of the passages shows that they are very largely influenced by a single writer, namely Varro. For example, the two statements in Verrius Flaccus probably go back to the view expressed by Varro in the fragment in Arnobius, while in Apuleius and Martianus Capella we have an elaboration of the Varronian theory found in the fragment in Augustine. Moreover, it is especially noticeable that Varro's own views, as stated by Arnobius, are far from giving the impression of any profound conviction: Varro .... nunc esse illos manes, nunc aerios rursus deos. According to this he is uncertain whether 1 Cf. Mon. Ancyr. Gk. x. 11 and xviii. 23 =Lat. iv. 7 and vi. 33; Dionys. Hal. iv. 70 and iv. 2, where 6 (tar' oidav ijpws =Lar familiaris; Corp. Gloss., II, 121, 14: Lares familiares =^po>ts kotoikZSuh ; also the poetical inscription of Acerrae (CIL, X, 3757), heroes qui Augusti nomen gerunt, correctly interpreted by Mommsen as Lares August! ; and Prudentius Adv. Symm. i. 190: tot templa deum Romae, quot in urbe sepulchra heroum numerare licet, quos fabula manes nobilitat, nosier populus veneratus adorat, where the reference is to the shrines of the Lares compitales. 2 P. 243. The Origin of the Cult of the Lares 127 the Lares are to be classed among the di inferi, or whether they should be assigned to the regions of the air. In fact, what Varro and the others offer us is not the definite statement of a widely- accepted traditional belief, but a complex of vague speculations which attest little else than the uncertainty in which the whole question was involved. That Roman writers should err in such a matter is not surprising. Their theorizing in the field of religion is notoriously untrustworthy. In a large number of cases it is as wild as their etymologizing. They seem to have lacked that spirit of self-castigation without which progress in this subject is impossible. The haphazard methods which they employed in the identification of Latin and Greek gods show how reckless and unscientific they were. Furthermore, even if the first identification of Lares and ijpaes is as early as Plautus, there is no evidence of its having estab- lished itself till a much later date. None of the examples cited (see footnote 1, p. 126) antedates the age of Augustus. Cicero equates Lares with daiiwves, and apparently is in some doubt about the matter. 1 The testimony of the ancients, then, affords but feeble support for the theory that the cult of the Lares was a worship of souls. What can be said about the rites, ceremonies, and practices con- nected with it? Can evidence in favor of the theory be found here ? De-Marchi, 2 Samter, 3 and others answer in the affirmative. They cite the fact that at the festival of the Lares compitales it was customary to hang up at night on the crossroad shrines little woolen images of human beings and balls. This is recorded by Festus. 4 The images, according to another passage in Festus, 5 represented and corresponded numerically to the free members of the household, while the number of balls was determined by the number of slaves. 1 Tim. 38: quos Graeci Saiitovas appellant, nostri opinor Lares, si modo hoc recte conversum videri potest. See Wissowa in Roscher's Lexicon, II, 1870. 2 11 culto private, I, 34 S. 3 Familienfeste, pp. 1 1 1-14. 4 121. 17: Laneae effigies Compitalibus noctu dabantur in compita. 6 239. 1: Pilae et effigies viriles et muliebres ex lana Compitalibus suspendebantur in compitis, quod hunc diem festum esse deorum inferorum, quos vocant Lares, putarent, quibus tot pilae quot capita servorum, tot effigies quot essent liberi pone- bantur, ut vivis parcerent et essent his pilis et simulacris contenti. 128 Gordon Laing Macrobius 1 tells us that the Compitalia had been reorganized by Tarquinius Superbus; that this reorganization had included among the divinities worshiped not only the Lares but also Mania, the mother of the Lares; that, in accordance with an oracle of Apollo, slaves had been sacrificed to them, ut pro capitibus supplicaretur. Macrobius goes on to say 2 that this practice continued for some time, but after the expulsion of Tarquin, Junius Brutus substituted for the slaves heads of garlic and poppy. In the same passage he states that images were hung up to Mania before the doors of the houses to ward off any danger that might threaten the family. In commenting on these passages the advocates of the theory point to Festus' statement (footnote 5, p. 127) quod hunc diem (i.e. the festival of the Compitalia) festum esse deorum inferorum, quos vocant Lares, putarent. Samter, moreover, lays special emphasis on the time at which this ceremony of suspension took place, namely the night. This, he argues, shows that we have to do with a rite of a propitiatory character celebrated in honor of chthonic deities. The woolen images of men and women he explains as substitutes for an original human sacrifice. 3 It is his opinion that such a rite can belong only to a cult of souls or of some divinity of the earth and concludes with the far from convincing argument that insomuch as he has already proved that it is the Lar of the household (Lar familiaris) and not the Lar of the field that is the starting-point of the cult, all possibility of connection with earth divinities is excluded and it must be to the Lares as souls of the deceased that the images are suspended. But he has not proved that the Lar familiaris is the starting-point of the cult. Nor is there any evidence that the images were substitutes for human sacrifices. They may not have been offerings at all. 4 The regular offering at the festival seems to have been cakes and, if i Sat. i. 7. 34. 2 Loc. cit.: Idque aliquamdiu observatum ut pro familiarium sospitate pueri mactarentur Maniae deae, matri Larum, quod sacrificii genus Iunius Brutus consul pulso Tarquinio aliter constituit celebrandum. Nam capitibus alii et papaveris supplicari iussit, ut responso Apollinis satis fieret de nomine capitum, remoto scilicet scelere infaustae sacrificationis ; factumque est ut effigies Maniae suspensae pro singulorum foribus periculum, si quid immineret familiis, expiarent. 3 He compares them with the puppets thrown into the Tiber by the Vestal Virgins on the occasion of the rite of the Argei on May 15. * See Wissowa, Arch. f. Rel., VII, 55. The Origin of the Cult of the Lares 129 Propertius iv. 1. 23 1 refers to the Compitalia, a pig. Further, nocturnal rites were not confined to earth divinities or spirits of the dead, 2 and, what is of more importance for our inquiry, the Compi- talia was probably celebrated in the daytime. This accords best with the description given by Dionysius of Halicarnassus. Finally I may draw attention to the lack of positiveness in Festus' statement (footnote 5, p. 127), who says that they hung up the images because they thought that this was a festival of the dei inferi called Lares. There is in this and in the passage in Macrobius (footnote 2, p. 128) the same atmosphere of speculation as in the citations from ancient authors referred to at the beginning of this article. Samter 3 sees further evidence of the Lares being the souls of the defunct in a custom referred to by Pliny the Elder xxviii. 27. 4 That writer states that if any pieces of food fell from the hand they were gathered up and burned as an offering to the Lar. That this is a practice closely connected with the worship of souls, Samter attempts to show by the citation of parallels from the superstitions of other peoples. He mentions first of all the Greek belief that any bits of food that fell to the ground from the hand or the table were the portion of the spirits that dwelt in the house. 5 He finds evidence of the custom among other races also. For example, at funeral meals in Prussia in primitive times it was usual for the guests to throw under the table something from each dish. This was for the refreshment of the souls of the deceased. What fell accidentally from the table was not picked up but was left for the "poor souls," i.e. the souls of those who had no relatives or friends among the 1 Parva saginati lustrabant compita porci. 2 Wissowa (Joe. cit., p. 55) cites as an example the practice of the consul rising nocte silentio to take the auspices. The consul did this, not because the souls of the dead were wandering about at night, but because the results of his inquiry must be available at daybreak. He thinks that the nocturnal suspension of the images was a preliminary preparation for the festival which began next day. s Op. cit, p. 108. 4 Cibus etiam e manu prolapsus reddebatur utique per mensas vetabantque mun- ditiarum causa deflare, et sunt condita auguria, quid loquenti cogitantive id accident, inter execratissima, si pontifici accidat dicis causa epulanti. In mensa utique id reponi adolerique ad Larem piatio est. 6 He cites Diogenes Laertius viii. 34, who quotes Aristophanes; Athen. x. 427e; Suid. ii. 2, p. 553, 4; and Iamblichus Vti. Pyth. 126. See also Rohde, Psyche, I, 245. 130 Gordon Laing living to provide them with nourishment and were therefore depend- ent on this chance sustenance. In the Palatinate crumbs that have fallen are collected and flung into the fire as food for the "poor souls." Among the Lithuanians bread crumbs are thrown upon the floor for the benefit of the souls of ancestors. In Bohemia crumbs are dropped into the fire for the souls; if they fall upon the ground and anyone steps on them, the souls, according to the native folklore, weep bitterly. In certain parts of the Tyrol also crumbs thrown into the fire are supposed to reach the souls of the poor. These parallels are interesting but not convincing. Wissowa 1 meets them by contending that in the Pliny passage Lar is used for the whole group of domestic gods, and that Pliny might just as well have used the word Penates. However that may be, Samter's inferences are in the highest degree doubtful, Summed up his reasoning is as follows: Among many peoples it has been the custom to throw into the fire, as an offering to the souls of the deceased, pieces of food that have fallen on the floor. Among the Romans these were burned as an offering to the Lar. Therefore the Lar was the spirit of someone deceased. But this conclusion does not necessarily follow. Such offerings might have been intended for spirits other than the souls of the deceased. Probably the custom was of purely secular origin. For cleanliness' sake crumbs that fell on the floor were swept into the hearth fire. Later there grew up the idea that it was an offering to some divinity of the hearth. In countries where the souls of the departed were believed to haunt the hearth, they were thought of as the recipients. But any spirit connected with the hearth (as the Lar familiaris in its later develop- ment certainly was) might with equal plausibility be regarded as the deity to whom the offering was made. Another point urged by the defenders of the ancestor theory is that the Lares must be connected with the goddess Larentia (also called Larentina and Larunda), whose festival, the Larentalia, was celebrated on December 23. But there is a difficulty in the fact that the a is short in Lares but long in Larentia, Larunda, and Larentalia. Samter 2 endeavors to meet this difficulty by quoting Zielinski's i Arch. f. Rel., VII, 45. 2 Op. cit., p. 115. The Origin of the Cult of the Lares 131 contention that Ovid, who has Larentalia with long a in Fasti iii. 57, was forced to use a false quantity there because with short a Laren- talia would not fit into a dactylic hexameter. He adds that the long a of Larentia used by Ovid in line 55 of the same book was due to the poet's desire to avoid inconsistency with the quantity which he was giving Larentalia two lines lower down. But all this is merely linguistic squirming; and the same may be said of Samter's attempt to discredit the long a of Larunda by pointing out that our sole authority for it is the fourth-century poet Ausonius, from whose works he cites a list of false quantities. De-Marchi 1 accepts the difference in quantity, but insists that it does not imply a difference in root any more than the different quantity of statio with short o and stare with long. But even if there were not this difference in quantity and if a connection between the Lares and Larentia could be established, this would not throw much light on the question. To be sure, the ceremony of the Larentalia did consist of some sort of funeral rites, apparently in honor of Larentia, but there is the greatest uncertainty as to who she was. According to some she was the foster-mother of Romulus and Remus; others said that she was the mistress of Hercules. 2 She has been identified also with the mother of the Lares, but this is idle speculation and has no bearing on the question of the origin of the cult. The Lares certainly go far back in the history of Roman religion and belong to a period when fathers and mothers of gods were unknown. Divine genealogies belong to the period of Greek influence. Before concluding my criticism of this theory I must refer to two other points. The first turns on the passage in Plautus, Merc. 836, where Charinus says: ego mihi alios deos penates persequar, alium harem, aliam urbem, aliam civitatem. If the Lar was an ancestor, how could this young man hope to get another one? One of the things that cannot be changed in this world is an ancestor. The Romans had a proverb that a man must run a long way to escape from his relatives, but I take it that not even they thought a person 1 Op. tit., I, 36. 2 Fast. Praenest., Dec. 23: Accae Larentiae. Hanc alii Remi et Romuli nutricem, alii meretricem, Herculis scortum (fuisse dicunt) : parentari ei publice, etc. 132 Gordon Laing could change his blood kindred. I do not think that either of Miss Waites's 1 suggestions that Plautus was consciously aiming at a comic effect or that in Roman law ancestors could be changed by adoption really meets the difficulty. My other point is one that by itself is sufficient to disprove the ancestor theory. It is the relative unimportance of the Lares on the occasion of the Parentalia, the regularly organized cult of the dead, the celebration of which occupied the days from the thirteenth to the twenty-first of February. If the Lares had been from the beginning regarded as the spirits of ancestors, surely their worship would have formed a much more prominent part of the festival than it did. It was only on the day after the festival, the twenty- second, when the Caristia was held — a sort of reunion of the surviving members of the family after they had fulfilled their duties to the dead — that any attention was paid to the Lares; and this seems to have been nothing more than what was due to them as domestic divinities on an occasion so important to the family as a whole. Another view of the origin of the cult has been set forth by Walter Otto. 2 While dissenting from the ancestor theory and expressing the opinion that the Lares were spirits of fruitfulness, 3 he claims that a connection between them and the lower world is beyond doubt. His argument is based largely upon the emendation of a fragment of Varro, through which we are given to under- stand that maniac were. hung up on the shrines of the Lares, pre- sumably the Lares compitales. Otto interprets maniae as meaning not dolls, as others say, but grotesque masks* He compares these maniae with larvae, emphasizes the fact that they were hung up at night, and on this flimsy substructure seeks to establish a connection with the lower world. But he does not explain the bearing of the other objects which even with his reading of the fragment were suspended at the same time as the maniae, namely reticula (hair nets) and strophia, which were the closest Roman approach to corsets. We must admit, however, that he does not make the mistake of which Wissowa is guilty when he says that the fragment describes 1 hoc. cit., p. 243. 2 "Mania und Lares," Arch. f. Ut. Lex., XV, 113-20. 3 He refers especially to their being appealed to in the Carmen Arvale. The Oeigin of the Cult of the Lares 133 the offering made to the Lares by Latin girls on attaining puberty. For while all would concede that it would be highly appropriate for girls on the threshold of womanhood to give up their dolls (as Wissowa interprets maniae), most of us would contend that it would be a uniquely inappropriate time to discard corsets. 1 Let us now examine the view held by Wissowa. As I have already indicated, most of his arguments against Samter seem to me cogent and conclusive. But his own explanation of the origin of the cult is not, as it seems to me, tenable. His theory is that the Lares were originally divinities attached to places. 2 They had definite localities within which they manifested their power, e.g., on the farms, at the crossroads, in the farmhouse, and along the highways. For just as the crossroads (compita) had their Lares compitales, so the roads (viae and semitae) had their Lares viales and Lares semi- tales. To these Lares viales travelers-by-land prayed, just as those who journeyed by sea made vows to the Lares permarini. It was, according to Wissowa, in the fields that the cult began. From them it made its way into the house. In building up his theory that the cult is connected with places and not with persons, he contends that while we have many references to the Genius of this or that person or of a college or a curia or a legion and so forth, we do not find allusions to the Lar of a person or of a group of persons. My first criticism of this theory is that his statement that there are no Lares of persons is not supported by the facts. 3 The familia, to which the Lar familiaris belonged, was a group of persons. If the reference had been to the house instead of the household, the spirit would have been regularly called Lar domesticus, but this term, as compared with Lar familiaris, is rare. Moreover, we have not only the innumerable references to the Lares Augusti, that is the Lares of the emperor, 4 but also many other examples. Perhaps the best-known of these is the case of the Lares Hostilii, i.e. the Lares of the Hostilian family. 6 We find also in an inscription of 1 Otto, loc. cit., pp. 114 f. 2 Religion und Kultus, p. 169. 3 See Waites, loc. cit., p. 244. 4 Cf. Lares Caesaris nostri, Dessau, Inscrip. Lai., 3542. 6 Waites, loc. cit, p. 245. 134 Gokdon Laing Spain (CIL, II, 804) diis Laribus Gapeticorum gentilitatis; and in one belonging to the city of Rome (ibid., VI, 10266) Lares Volusiani. In the tenth volume of the Corpus (8061, 1) we have the words Larum Galillensium on a seal, and from another inscription (ibid., 7852) we know that Galillenses was the name of a people. In these instances the Lares are obviously connected with persons, and attempts to explain such references as local should convince no one. There is even an example of Lares curiales given in the Bulletino Archeo- logico Communale, 1909, page 19, and commented on by Tomassetti there. It is not a local idea that we have in Lares praestites, or in the later Lares publici, 1 or Lares paterni. 2 Nor is the idea of locality to be found in the Lares militares: Dessau, Inscrip. Lat., 451 fin., 3637, and 3638. The first of these inscriptions occurs in the acts of the Arval Brothers; the other two stones were set up by military tribunes in Pannonia. Again, it would be interesting to know what local idea could be found in the inscription discussed by Zangemeister in Rhein. Mus., XIX, 49 ff . : Martis et Pads Lari. Noticeable also is the other inscription cited by Zangemeister, in which we find the phrase nostri publice Lar populi. Compare the metrical inscrip- tion found at Nicopolis in Lower Moesia (CIL, III, 754) : Lar mihi haec quondam, haec spes haec unica vitae, where a husband, lamenting his dead wife, says that in her lifetime she had been his Lar. It is quite clear that it had never occurred to this man at least that only places, not persons, could have a Lar. And he little dreamed to what extent he was running counter to the ancestor theory when he spoke as if her service as a Lar had terminated with her life. The passages I have quoted are, I think, enough to disprove Wissowa's limitation of the Lares to places. Another mistake that he makes is in his treatment of the Compitalia. One gathers from his discussion that it is here that the origin of the cult must be looked for. Possibly this is due to the fact that we happen to have more data about this phase of the worship of the Lares than about others. It was a popular festival and has won a place in the literary tradition. Then we have Tibullus' 3 remark that the Lares were agri custodes, 1 Dessau, Inscrip. Lot., 99, 3625, 3629. 2 Orelli, Inscrip. Lai., 1667. »i. 1. 20. The Origin of the Cult of the Lakes 135 and Cicero 1 speaks of the shrines of the Lares as being in agris and of the religio Larum as in fundi villaeque conspectu. Wissowa's reconstruction of all this is that each farm had its Lar and that at the crossroads (compita) where adjoining properties converged a shrine was set up, with as many altars as there were individual properties contiguous to it. This he believes was the shrine of the Lares compitales, and in support of his view cites the description of boundary-line shrines in Grom. Lot., p. 302, 20. But the word compitum is not used in that description, 2 Compitum means a place where roads cross, and this does not necessarily coincide with the boundaries between the holdings of different individuals, to which the gromaticus obviously refers in the passage cited. This insistence upon the field origin of the Lares has been the cause of endless error. It has diverted the inquiry. It was not, however, Wissowa who first drew this herring across the trail; it was Jordan 3 in the note in which he described the extension of the cult from the fields to the house. But Wissowa has gone farther than Jordan, He has emphasized unduly the rural aspect of the cult- This after all was only one phase of it. It is significant that out of the list of twenty-one classes of Lares given in Roscher's Lexicon (1885-87) only one refers definitely to the country. That is the Lares rurales mentioned on the Capitoline base, and even in its case the reading is uncertain. The reference in the Arval hymn 4 does not necessarily involve a close connection with the fields. Doubtless in the primi- tive period also the Arval Brothers appealed to divinities whose scope included more than the arva. It is not even certain that Mars, the appeal to whom in the hymn is side by side with that to the Lares, is addressed in any other capacity than as a protecting divinity of the people. In regard to the inscription 5 said to have been on the 1 De leg. ii. 8. 19; ibid. 11. 27. 2 See Samter's discussion of this in Arch. f. Bel., X, 369 ff. 3 Preller-Jordan, Bom. Myth., II, 102, n. 2: "Lases .... sind ihrem Wesen nach ursprunglich Flurgotter (als solche werden sie geradezu definirt bei Cicero De leg. II. 8. 19, angerufen im Arvallied, verehrt in Campanien und im Arvalhain), haften daher an domus familiaque des grundbesitzenden romisehen Burgers .... und behiiten wie auf der Flur so in der Stadt Weg und Kreuzweg." 4 Henzen, Act. Frat. Arv., p. 26: enos Lases iuvate. 6 Quoted by Tertullian Sped. 5: Consus consilio, Mars duello, Lares t ooillo potentes. 136 Gordon Laing altar of Consus, the uncertainty of its text and the quality of the information implied in the derivation of Consus from consilium combine to make it negligible. In the passages cited above (pp. 134, 135), Cicero and Tibullus were obviously referring to the Lares compitales, many of whose shrines were of necessity in the rural districts. Furthermore, Wissowa's account of the Lares compitales would lead one to suppose that they were not spirits of the cross- roads at all, but merely spirits of the fields worshiped at the cross- roads. He does not recognize the fact that they should be linked with the Lares viales to which we have so many references. But it would be presumptuous in me to criticize so old a belief as the ancestor theory and so distinguished a scholar as Wissowa without offering an explanation of my own. I can give my opinion in a few words. To begin with, I think that we should recognize that we have in the appeal to the Lares in the Arval hymn and in the reference to them in the formula of devotion of Decius 1 a stratum of data that, meager as it is, must be considered by itself. It confuses the issue to consider these references side by side with those belonging to a much later period and to a more developed stage of the cult. In both we see the Lares — without limiting or descriptive epithet — as they were when Roman religion was just beginning to emerge from that pandemonism which was the most notable characteristic of its primitive state. They were Saifioves in the broadest sense in which the Greeks used that word. Their name was a generic term for a class of divinities the range of whose powers was not precisely defined in the minds of their worshipers. To the early Latins the world swarmed with spirits, nor can we for a moment assume that such specialization as we find, for example, in the prayer to the twelve gods of agriculture at the beginning of the sowing season existed in the earliest period, or even in later times extended to all gods or groups of gods. And among the groups that in those early days were without precise delimitation of function I would place the Lares. That they were thought of en masse, as it were, need surprise no one. Think how long the Manes, though belonging to a wholly different sphere, were conceived of in the same. way. But 1 Liv. viii. 9. 6: Jane, Juppiter, Mars pater, Quirine, Bellona, Lares, divi noven- eiles, di indigetes, divi quorum est potestas nostrorum hostiumque, diique manes. The Origin of the Cult of the Lares 137 these early Lares were not merely without specific field of activity; they were also without limitation in number. To the primitive Latin mind their name 1 carried no more definite numerical connota- tion than the words brownies, spirits, angels, do to us. When we meet them in the Arval hymn and in the formula of devotion, it is true that other Roman gods had emerged from the pandemonic swarm and attained individuality. In the hymn, it will be remem- bered, Mars is appealed to at the same time as the Lares, and in the devotio of Decius they are mentioned together with Janus, Jupiter, Mars, Quirinus, and Bellona. In the devotio, however, they go more closely with the latter part of the list, namely the groups divi novensiles, di indigetes, divi quorum est potestas nostrorum hostiumque, and Manes. They were mentioned with these because they were thought of as belonging to the category of gods whose powers to help or harm were not confined to any special field, but were of a comprehensive and general character. But there is other evidence of the vague and general character of the Lares in the early period. This lies in the later practice of attaching to the word Lares an epithet indicating the special field in which they were supposed to operate. For example, to the word Lares, which by itself would indicate merely spirits that could either protect or injure, there was added, apparently in a fairly early period, the epithet praestites, and the obvious interpretation of Lares praestites is "the spirits that protect the community." And in the same way Lar familiaris is the spirit that protects the familia (i.e. the household group: the family proper and especially their slaves and freedmen); Lares compitales, the spirits that haunt the crossroads; Lares permarini, those that follow the sea; Lares militares, those that help in war. We hear even of a Lar agrestis, spirit of the woodland. There is no need to give here the names of all the classes of Lares that have come down to us. Doubt- less these are only a part of those that the Latins knew. It was in the nature of such vaguely conceived beings that subdivision into classes could continue indefinitely. It is idle to pick out one of these classes and say that it is the original, and that from it the 1 Fowler speaks of the multiplex groups in early Roman religion, "like the Fauni» Silvani, Lares, Penates, Semones, Carmentes," RF, p. 337. 138 Gobdon Laing others are derived: to say, for example, that those worshiped at the compita were the earliest and that they crowded from the farm into the farmhouse, and from the farmhouse passed to the city-house, and that, once the idea of them as protecting deities had arisen, their guardianship was extended to other spheres, and that in this way all the classes came into existence. Much simpler and, as it seems to me, more probable is the assumption of such Lares as I have posited for the early period: spirits of so general a type that they could be connected with persons or places or activities widely diver- gent. Certainly the existence of all these different classes — each with its determining epithet — is significant, and the high potentiality of diversity of field and function which it indicates is of the very essence of the original character of the cult. Probably such a type as the Lares praestites arose at an early date, but it is neither possible nor especially important to determine in what order the other sub- divisions were detached from the parent swarm. In the fact that some of the classes were of relatively late development we have simply an indication that the original idea of the Lares survived throughout the history of the cult. Of the pertinacity of the original concept we have another striking piece of evidence in this that Latin writers sometimes use the word Lares as the practical equivalent of dei. 1 There was of course some differentiation accorded to them, or they never would have attained to a name of their own. We have some information on this point, and we get it from Cicero, who speaks of them as dei plebeii: that is, they were an inferior order of divinities, as compared with those whose individuality had emerged with more or less definiteness from the throng of spirits that pervaded the world. The passage in Ovid Ibis 81 f. bears directly upon this point: plebs superum, Fauni Satyrique Laresque Fluminaque et Nymphae semideumque genus. Just what the word Lares means no one can say with any degree of positiveness. Its etymology has been much discussed, but with meager results. 2 Among the more 1 Wissowa, Religion und Kultus, p. 170, speaks of this as a late development: "die in spaterer Zeit auftretende Verallgemeinerung des Larenbegriffes, vermoge deren das Wort so ziemlieh mit deus oder numen gleichbedeutend wird. " But this usage was not a late development. It was a survival of the original meaning. 2 See Walde, Lat. Etymol. Wort, p. 413. The Origin of the Cult of the Lakes 139 plausible of the derivations is that which connects it with lascivus. 1 Certainly the waywardness and liveliness which that word connotes may be said to accord well with the familiar representation in sculp- ture and painting of the dancing Lar, as well as with Naevius' 2 phrase Lares ludentis and all that has come down to us in regard to the merriment of the Compitalia. Indeed one of the mysteries of the ancestor theory is how anyone can reconcile such merry sprites with the Romans' idea of their grandfathers. I am aware that some of the readers of this article may say that the conception of the original Lares which I have suggested is too vague to be tenable. It will be argued that a more .definite sphere and more specific functions should be assigned to them. In reply I would point out that the cult goes back to that dawn of religion of which vagueness was a dominant characteristic. 3 The neat pigeon- holing of divine beings, which is the obsession of later ages, made but scant appeal to the primitive mind. They believed in the existence of these spirits; they believed that they could protect or injure persons or places, and that they operated in many spheres and in many ways; and they believed that it was highly important to establish friendly relations with them. But they did not push their analysis farther. Yet while the original Lares were spirits of the general type that I have attempted to describe, it goes without saying that the different groups, as they became detached from the mass, tended to develop characteristics and rites that were their own. The Lares compitales are a case in point. And while we may well distrust the explanations of the ancients in regard to the woolen images and balls hung up on the shrines, there is no reason for doubting the fact of suspension. There may have been some design of conciliation in the practice. For the Lares were not necessarily and inevitably beneficent. We may reasonably assume a certain degree of capriciousness in them. 'See Ehrlich, Zeitschr. f. vergleich. Sprachforsch., XLI (1907), 299: "Das Indische kennt ein lasati, 'er ist Iebhaft' = *leseti, lasati, 'er verlangt'; damit verbindet man passend \i\aloii