The most important temporal foci of divine concepts in Rome were their ‘holidays’.
By Dr. Michael Lipka
Professor of Religious Studies
University of Patras
Roman gods were invariably eternal. This explains why any Roman god, even the antiquated and forgotten ones, could be invoked at any times. Temporality was therefore an indispensable constituent concept of the concept of divinity. But ‘eternity’ was too unspecific a concept to be of any practical consequence in cultic terms. Therefore, it was narrowed down. The outcome was a series of occasions, i.e. temporal foci, on which the relevant deities were expected to be present and particularly benevolent. This cycle of temporal foci of Roman deities was recorded in the pontifical calendar.
Hardly any element of Roman culture has enjoyed such breathtaking success as the Julian calendar, of which Scaliger could justly say that it “marked a victory in the realm of culture more lasting than any Roman victory on land or sea”.87 Reaching back to the sixth century B.C., it was substantially revised by Caesar and, after a minor adjustment by Pope Gregory XIII at the end of the sixteenth century, commenced its triumphant march all over the globe. At an early stage, perhaps towards the end of the fourth century B.C., festivals, i.e. temporal foci of the worship of specific gods, were included. Later on, annual celebrations of public events, such as victories of the Roman armies and the rulers’ anniversaries, were also marked in it. From time immemorial, its redaction lay in the hands of the pontiffs. The first surviving copy belongs to the first half of the first century B.C. (Fasti Antiates Maiores), a wall painting from a Mediterranean seaside resort south of Rome. The principal importance of this copy lies in the fact that it represents a selection of festivals and ceremonies of the religious calendar as it was before the revisions initiated by Caesar in 46 B.C. Nevertheless, a selection it was, not the whole calendar.
The most important temporal foci of divine concepts in Rome were their ‘holidays’, Latin feriae. Ancient sources divide such ‘holidays’ into two main categories, feriae publicae and privatae.88 The former were relevant to public cults and are dealt with now, the latter were relevant to private cults and are dealt with below.
The feriae publicae of gods can be divided into those celebrated annually on the same day, and thus marked as such on the calendar (feriae stativae), and those whose specific dates were announced by the magistrates or priests (feriae conceptivae). Besides this, extraordinary feriae were ordered at the discretion of consuls and praetors (feriae imperativae), and later of the emperor.89
Initially, all feriae were proclaimed on the fifth or seventh day (Nonae) of each month by the rex sacrorum, i.e. all were initially feriae conceptivae.90 If so, the question arises why some maintained this status, while others turned into feriae stativae. The explanation is hardly to be found in specific seasonal events marked by the feriae: for not all feriae conceptivae depended on seasonal conditions, and some feriae that did depend on seasonal conditions were no feriae conceptivae.91 One might guess that those feriae conceptivae that had political connotations, such as the Feriae Latinae or the Saturnalia, were less likely to be changed so as to become feriae stativae. For such a change would have been tantamount to a de facto loss of political control on the part of those responsible for fixing the dates of the feriae conceptivae, most notably the pontiffs. After all, such political ‘holidays’ could always be politically exploited, either to promote one’s own cause or to obstruct the plans of one’s political opponent.
The core of feriae publicae may have been fixed in the sixth century, although the publication of a feriale may have been delayed until the end of the fourth century B.C.92 One cannot verify to what extent information that was self-evident to the target group of the published calendar (ordinary people as opposed to the aristocratic élite with its priestly monopoly?) was deliberately omitted. Interestingly, it appears that the earliest version of the published feriale was not modeled on the ‘earliest’ reconstructable Roman pantheon, as reflected by those gods that were represented by the flamines93 or, for that matter, by the names of months of the earliest Roman calendar.94 Furthermore, it seems that the feriale was not committed to written form immediately after its creation. This would explain irregularities in what seems to have been originally conceived as a consistent plan: for example, despite the overall scheme, at least on two occasions we find two feriae on the same day.95 Besides this, despite the purely religious character of the feriale, an additional ‘political’ holiday, i.e. a holiday commemorating a specific historical event, had found its way into the official calendar as early as the first half of the first century B.C., viz. the regifugium.96 The latter, however, appears to have been an extraneous element anyway, being exceptional in terms of timing: in violation of usual practice, it fell on an even day (March 24) and was immediately preceded by another ‘holiday’ (Terminalia, March 23) without a day intervening between the two feriae.
In practical terms, the feriale reflected a slowly but permanently shifting system of temporal foci. For the speed of this dynamic process one may compare the Calendar of Philocalus, composed in 354 A.D., some 400 years after the oldest preserved calendar, the Fasti Antiates Maiores. Of the forty-five festivals of the latter, only twelve are mentioned by name in the former, while other festivals were renamed, replaced by public Games, or simply forgotten.97
The feriae publicae of gods did not necessarily denote a homogeneous category. True, most feriae publicae mentioned by the feriale were specifically marked in the preserved epigraphic evidence by the mysterious sign NP, which has caused headaches to the most eminent epigraphists, starting with Mommsen. However, some feriae publicae were marked otherwise.98 Even if we cannot decipher the letters NP satisfactorily, the general nature of public holidays is made clear by the sources: Feriae publicae were days of promoting divine peace; business transactions and physical labour, especially by slaves, were restricted or completely avoided, while certain priests were not even allowed to see someone working on that day less they should be defiled.99 In theory, such regulations applied to all feriae publicae alike. In practice, though, there were manifold gradations. First, there were palpable differences in terms of popularity. On the one end of the scale one may mention the exceedingly popular Saturnalia (December 17), on the other the completely obscure Agonalia on May 21. Strangely, not even their divine patron is known with certainty (Vediovis?).100 One may also refer to the Fur(r)inaliaon July 25, whose deity by the time of Varro was almost completely forgotten.101 Second, there were practical needs. It was virtually impossible for the peasant to lay down his tools on any given dies feriatus, if weather conditions required otherwise. Hence, according to P. Mucius Scaevola (cos. 133 B.C.), an expert on pontifical law and pontifex maximus, on feriae one was allowed to do what could bring harm if left undone.102 Third, not all public festivals were relevant to both sexes alike: thus, it is a fair guess that men would have been less likely to observe the Matralia, the festival of matrons, on June 11, while women would not automatically participate in the Armilustrium, the ‘purification of arms’, on October 19. Fourth, many public festivals were specifically linked to a professional group. It is natural, therefore, that the Vinalia on August 19 and the Robigalia on April 25 were of special importance for the rural peasantry. Similarly, the Vestalia on June 9 were especially linked to bakers and millers,103 and the Quinquatrus on March 19 were sacred to Minerva and thus connected to all kinds of arts and crafts.104 These were mainly observed by urban craftsmen, artisans and skilled labours, but not self-evidently, say, by members of the senatorial order. Fifth, there may have been territorial differences concerning observances. Varro reports that the Septimontium was celebrated, not by the Roman people as a whole, but by the ‘people of the hills’, while the Paganalia were held by the members of a pagus.105
In short, there were a variety of ways in which feriae could be celebrated. Further diversity is suggested by the lack of a consistent terminology: in the imperial period, people could no longer distinguish between various forms of religiously relevant days such as dies nefasti, dies religiosi and dies atri,106 and even legal texts assimilated dies nefasti and feriae.107 The result was that the various, originally distinct concepts of time were assimilated to each other. Even more confusing is the officially sanctioned modification of the character of a number of holidays: Caesar transformed the legal marking of three feriae into NP-days, thus clearly reacting to changing religious attitudes.108
The religious life of the individual was determined not only by the feriale, but also—and predominantly—by private holidays (feriae privatae).109 Private holidays were either passed on within major clans (gentes), or derived from the personal biography of the celebrant, such as birthdays, anniversaries etc.110 The feriae, celebrated by the leading members of the family (familia), especially the pater familias, certainly affected other members of the family as well.111 Despite the fact that the feriae privatae were of paramount importance for the religious life of the individual citizens and frequently might have overshadowed public holidays, they have left almost no trace in ancient sources.112 It is important to note the ambivalent position of the private holidays of the imperial family, which became feriae publicae to the extent determined by the emperor.113
The way in which divine concepts were formed through the Roman calendar can now be demonstrated by a number of examples. The temporal foci of the most supreme Roman god, Iuppiter, are numerous. To begin with, the days of the full moon (Idus) were sacred to him.114 That explains why the Ides were marked in the calendar as NP-days, i.e. why they belonged to the same category as most public holidays.115 But, apart from the monthly rhythm, worship of Iuppiter focused on various dates of the annual cycle too. As a matter of fact, no Republican god equalled him in the number of ‘fixed holidays’: the Poplifugia ( July 5), the Vinalia (April 23 and August 19), the Meditrinalia (October 11), and possibly also the Regifugium (February 24).116 As to public Games, the ludi Romani in September117 and the ludi plebei in November were sacred to Iuppiter Optimus Maximus,118 while the Capitoline Games on October 15 were dedicated to Iuppiter Feretrius.119 The last two, at least, were not only among the oldest Games, but also the most extended religious events of the Republican year. However, the temporal focalization of Iuppiter went further. The anniversaries of Jovian temples fell exclusively on the ‘marked’ days of the month, i.e. the first (Kalendae) and fifth (or in March, May, July and October the seventh) day (Nonae) of the month, as well as—and unsurprisingly (since dedicated to Iuppiter anyway)—the 13th (or in March, May, July and October the 15th) day of the month (Ides).120 The only case that would contradict this rule has been convincingly explained as having originally fallen on the Ides too.121
To further illustrate the importance of temporal focalization, some short additional notes should be made. For instance, all Republican temples of Iuno were dedicated on the Kalendae, with one exception that was dedicated on the Nonae.122 Independent evidence suggests that the Kalendae were indeed sacred to Iuno.123 In the case of Mars, all but two festivals of the war god are found in his month, i.e. March. Furthermore, even the two exceptions to this rule are directly related to festivals celebrated in March.124
Naturally, unofficial or even banned cults likewise show temporal focalization. For example, we hear that the ‘calendar’ of the cult of Bacchus, at the beginning of the second century B.C., included regular initiations, which initially were carried out just three times a year. However, after reforms of ca. 210 B.C., initiations were performed on five days of every month.125 It is at least a plausible guess that the Liberalia on March 17, initially connected with scenic Games (ludi), also served as a temporal focus for the cult of Bacchus, given the general identification of Bacchus with Liber during the Republican period. The merger of the Games with the ludi Ceriales has been tentatively, though plausibly, connected with the Bacchanalian affair.126
In the same vein, the worship of foreign gods was temporally focused. For example, in the cult of Isis, the Menologia Rustica (first century A.D.) mention a festival called Heuresis (i.e. the recovery of Osiris—Sarapis, who had been killed and dismembered by Typhon) on November 15 (while the Calendar of Philocalus refers to the same festival on November 1 as the climax of a festival named Isia. This festival lasted from October 28 to November 3).127 Furthermore, the Menologia record an Isidis navigium on 5 March128 and two festivals in April, namely the sacrum Phariae (Pharia being an epithet of Isis), held somewhere between 22–24 April, and the Sarapia on 25 April.129 In addition, the Calendar of Philocalus mentions the lychnapsia on August 12.130 In the Egyptian calendar, clearly underlying most Roman dates, there may have been patterns of focalization that were lost when the dates were adopted into the Roman calendar. Nevertheless, temporal focalization characterized the cult of Isis even in Rome.
Turning to imperial worship, we have to bear in mind that the divinity of the emperor was modelled on that of traditional gods. This included the temporal focalization of his cult. Indeed, the terminology is telling: imperial ‘holidays’ were called feriae, just like the ‘holidays’ of traditional gods.131 More importantly even, ‘ordinary’ days were declared imperial holidays, resulting in their ‘day’ character being changed. They were marked as NP-days, as were the vast majority of traditionally ‘fixed’ holidays.132 Certain days served eo ipso as temporal foci, for instance the imperial birthday or the anniversary of the emperor’s enthronement (dies imperii). Thus, Augustus’ birthday (September 23) was declared a public holiday in 30 B.C. with Games being added later in 8 B.C.133 Consider the festival established on the occasion of the victory at Actium in 30 B.C. (September 2), apparently intended to form the beginning of a new era.134 Circus Games became the rule on either of these occasions and were continued under later emperors too.135 Many more such imperial ‘holidays’ were established by Augustus and his successors, most of which were of a temporary nature, though all served in varying degrees as temporal foci of the imperial cult.136
The distribution of ‘fixed holidays’ in the Republican calendar shows a remarkable consistency. All ‘fixed holidays’ fell after the Nonae of a month (for it was then that they were ‘announced’ by the rex sacrorum).137 In addition, all fell on uneven days and, as a consequence, no ‘fixed holiday’ ever follows immediately on another.138 Where a festival lasted for more than a day, days of non-festive character intervened. Consequently we find the Carmentalia on January 11 and 15, the Lemuria on May 9, 11 and 13 and the Lucaria on July 19 and 21. Exceptions to these rules are few. The Regifugium, dedicated to Iuppiter, fell on February 24, and the Equirria, dedicated to Mars, on March 14 (following the earlier Equirria on February 27). The only festival before the Nonae of a month is the Kalendae of March, while the Poplifugia fall exactly on the Nonae (July 5). These exceptions must be briefly commented on.
The Kalendae of March constitute the beginning of the Roman calendar in its first historical form. The names of the month following June (namely Quinctilis [later July], Sextilis [later August] etc.) are calculated from March onwards. A number of rites underpin the importance of the Kalendae of March as the beginning of the Roman year.139 The special festive character of the Kalendae of March is therefore not surprising. Concerning the dates of the Regifugium and the Poplifugia, their similar word-formation, their seeming reference to a specific historical event (otherwise unique among the ‘fixed holidays’), and their peculiar position within the calendar may indicate a close relationship with each other, and a secondary addition to the calendar. Lastly, the celebration of the March Equirria on an even day remains in fact unexplained.140
The final and perhaps most important principle of the Republican ‘fixed holiday’ is the fact that most of them formed a temporal focus for one, and only one god at a time. The parallel to the spatial foci of official cults is obvious.141 Still, a few exceptions to this rule must be considered.142
All Ides were sacred to Iuppiter and consequently, no other Republican ‘fixed holidays’ fell on the Ides, with the exception of the Ides of March. These were also sacred to Anna Perenna, who was worshipped then, on the first full moon of the first month of the old calendar (beginning in March), as the goddess of the ‘new year’ (as also suggested by the etymology of her name). Ovid may have preserved old beliefs that linked her to the moon, although as his own uncertainty shows, this tradition had been almost forgotten in his day.143 Apart from her function, there is no apparent reason why the festival dedicated to Anna Perenna could not have fallen on another day in March. It is clear that March 11, 13, 25 and 27 would have been available, as none of these was a ‘fixed holiday’. In short, Anna Perenna had received her place in the calendar on the basis of function at a time when it still mattered. Even later, when the beginning of the year had been moved to January 1, she successfully defended her place in the calendar. Two conclusions can be drawn from this test case: firstly, function was more important than the avoidance of any overlap of temporal foci of different gods, and secondly, temporal foci in the traditional calendar were as conservative as spatial foci and, once established, were virtually irremovable.
A similar coincidence of temporal foci of different cults is found on March 19. The day called Quinquatrus (= ‘the fifth day after the Ides’) was sacred to both Mars and Minerva, though again independently of each other. As a ‘fixed holiday’ of Mars, the Quinquatrus were connected with the purification of the ancilia, the mythical shields on which the prosperity of Rome allegedly depended. These were kept in the temple of Mars by the Salii. This festival can be seen to parallel the Armilustrium held on October 19, which was also linked to the purification of the ancilia by the Salii and fell on the same day of the month as the Quinquatrus. In other words, the date of the Quinquatrus as a ‘fixed holiday’ of Mars was firmly anchored in the calendar by its parallel ‘holiday’ in October.144
The attribution of March 19 to Minerva has been explained by equating Minerva here with Nerio, an otherwise obscure female consort of Mars.145 One may argue that groups with similar spatial or functional foci such as Ceres, Liber and Libera, could be worshipped jointly at the Cerialia (April 19). If a similar spatial or functional focus existed in the case of Mars and Nerio, one would consider this solution more seriously. However, no such focus is on record, while rituals performed on the Quinquatrus unequivocally mark craftsmen and artists as their target group. These stood under the protection of Minerva, and predominantly the Aventine Minerva, at least from the time of the second Punic War.146 In addition, the anniversaries of two ancient temples of Minerva, on the Aventine and the Caelian Hills, fell on the Quinquatrus (another tradition places the anniversary of the Aventine temple on the Quinquatrus Minores, i.e. June 19), and followed the traditional pattern of temple anniversaries celebrated on ‘fixed holidays’ of the relevant gods. In other words, we cannot explain away the fact that the Quinquatrus were dedicated to two independent divine notions, Mars and Minerva. Nor is this the only case of such ‘double’ attribution of a ‘holiday’: into a similar category falls the October horse, sacred to Mars, but sacrificed on the Ides of October, which—like all Ides—were traditionally sacred to Iuppiter; or the Liberalia celebrated on March 17, falling on the same day as the Agonalia of Mars. We may explain such double attribution as mere chance. But it is worth noting that in all three cases, double attribution occurs in connection with Mars. Without proposing an elaborate theory, which would necessarily remain hypothetical, let me remind the reader that the city of Rome was the result of a synoecism of the neighbouring peoples. One should at least grant the possibility that Mars may have played a special role in one of the synoecizing communities (e.g. that on the Palatine, whose priestly college of Salii Palatini was under explicit protection of the war god), and that the double attributions of ‘holidays’ as well as other inconsistencies in an otherwise consistent calendary system are residues of a unification of different calendars, which were employed by the communities in question.147
The coincidence of temporal foci may, on occasion, be due to complementary functional foci, in the same way that spatial foci were, at times, connected to complementary functional foci too.148 I have mentioned the example of Ceres, Liber and Libera, who are honoured jointly on the Cerialia. A further example is the joint worship of Iuppiter and Venus during the two wine-festivals, on April 23 (Vinalia Priora) and August 19 (Vinalia Rustica). Iuppiter was closely linked with viticulture due to his functional focus as a god of the ‘heavens’ and therefore of ‘weather conditions’. Venus for her part was a goddess of fertility and, more specifically, of gardens and gardening. In this sense, her functional foci amplified those of Iuppiter as a god of the weather. Two ancient sanctuaries devoted to her (in the grove of [Venus] Libitina outside the Esquiline gate,149 the other near the Circus Maximus) had their anniversaries on the Vinalia Rustica, the day celebrated by the kitchen-gardeners as their ‘holiday’.150 The impact of the cult of Venus on the Vinalia Rustica was so marked that even well-informed sources attributed the ‘holiday’ exclusively to Venus.151 In fact, the day may have originally belonged simply to Venus, for it was a female victim (a lamb, agna) that was offered on this occasion. Such an interpretation would, of course, mean disregarding Varro’s explicit statement: “this is a day sacred to Iuppiter, not to Venus”.152
Complementary functional foci may also be the reason for the coincidence on December 23 of the Larentalia, sacred to Larent(i)a or Larentina, to whom a sacrifice for the dead (parentatio) was offered on this occasion, and Iuppiter, in the form of Vediovis, that is as a chthonic deity.153 The specific sacrifice (parentatio) is well attested, inter alia by Cato and by the most eminent scholars of the Augustan age (Verrius Flaccus, Varro).154 In contrast, Latte does not succeed in proving his theory that a sacrifice to Larent(i)a on an altar (ara) by the pontiffs, as attested by Cicero, our earliest witness, excludes worship of the dead: at the very least, worship of chthonic gods (and that would include the worship of the dead, I assume) was conceivable under similar circumstances: Consus’ chthonic character is manifested by the fact that his altar (univocally called ara by the sources) was subterranean, while it was the sacerdotes, i.e. the pontiffs, who offered a sacrifice there on July 7.155
Complementary functional foci of various cults may merge to such a degree that a temporal focus, originally characteristic of one specific cult, is eventually attributed to other cults too. For instance, during the festival of the Lemures (Lemuria, 9th, 11th and 13th of May) beans were offered—according to one source—not to the Lemures, but to the Larvae. According to another source, however, the recipients of the sacrifice were the manes paterni.156 The reason for this confusion was largely the fact that the different notions of Lemures, Larvae and Manes were confusingly similar. In exact usage, the ordinary word for “ghost” in the sense of terrifying spooks was larvae, which was considered to be a synonym of the antiquated lemures.157 If, however, one referred to the ghosts as the venerable souls of the past, manes was the correct word to use.158 Besides this, the difference between Larvae (Lemures) and Manes was local. While Manes were the ghosts of the underworld, Larvae (Lemures) belonged to the upper world; this explains why Larvae figure conspicuously in Plautan daily life and Manes are absent there.159
In the case of Ceres, Liber and Libera, spatial foci interacted with temporal foci. Liber and his female counterpart, Libera, had their own ‘fixed holiday’ (Liberalia, March 17), including their own scenic Games.160 Interestingly, we find the two deities worshipped also during the Cerialia on April 19 (as already laid down by the Fasti Antiates Maiores). Meanwhile, the anniversary of the Aventine temple also fell on the Cerialia. Given these facts, the following scenario seems plausible: originally, Liber and Libera had their own ‘fixed holiday’ on March 17, which may have included scenic Games, while Ceres was honoured on April 19. When the temple of the Aventine triad was dedicated at the beginning of the fifth century B.C., the ‘fixed holiday’ of the most prominent member of the triad, Ceres, was chosen as the consecration day of the temple. Subsequently, Liber and Libera were ‘added’ to the ‘fixed holiday’ of Ceres thanks to their joint worship in the temple. Subsequently, perhaps in the wake of the Bacchanalian affair, the scenic Games of Liber were also merged into the fixed holiday of Ceres.161 The Aventine triad is, therefore, the only case in the Republican calendar in which a joint cult is evidently reflected as such, by joint spatial and temporal foci.
If we consider the interaction of fixed holidays and public Games in the Republic, it is apparent that until the time of Caesar, care was taken that no fixed holidays of gods intervened other than the ones honoured by the Games. For instance, the months in which the two most important public Games took place, September (ludi Romani) and November (ludi plebei), were free of all fixed holidays apart from the Ides, which were, as usual, sacred to Iuppiter. Since, however, both Games were devoted to Iuppiter anyway, there was no inconsistency of temporal foci here. Nor do we find an overlap in the case of other Republican Games, viz. the ludi Megalenses (April 4–10), Florales (April 28–May 3), Apollinares ( July 6–13) and the victory Games of Sulla, established in 82 B.C. and first held a year later (October 26–November 1). The exception is the ludi Cereris (April 12–19). They included the Ides of April, sacred to Iuppiter (April 13), and the Fordicidia, sacred to Tellus (April 15). I can offer no explanation, unless we assume that the ludi Cereris were very different in nature from the other Games. One may be tempted to consider the strong plebeian link of the cult of Ceres and her Aventine temple as a possible reason. The Games may have been conceived as merely political in the first place and, when they were established (in the fifth century?),162 perhaps as ‘opposed’ to the age-old ludi Romani. An indication in this direction may be the fact that they were held by the plebeian aediles (whose existence dates back to the beginning of the fifth century B.C.);163 and that plebeian families (under explicit exclusion of the patricians) invited each other to dinner during the Games (or possibly on the last day, the Cerialia in the strict sense).164 One may then speculate that after, or on the occasion of, the eventual compromise achieved by the Orders in the fourth century, the Games changed both date and addressee, and were now celebrated in November by the entire Roman people as ludi plebei, sacred to Iuppiter (while the old Games of Ceres were not abolished). One should bear in mind that the ludi plebei appear in historical records for the first time as late as 216 B.C., while the Games of Ceres are attested almost simultaneously, i.e. in 202 B.C., for the first time.165
Most major Republican Games show a remarkable connection with temple anniversaries of the relevant gods. Normally, the last day of the Games coincided with the temple anniversary of the god to whom the Games were dedicated. This is the case with the ludi Megalenses (April 10), which was also the anniversary of the temple of Magna Mater, and this may similarly have been the case with the last day of the ludi Apollinares ( July 13), possibly the original anniversary of the temple of Apollo Medicus.166 The last day of the ludi Ceriales, a ‘fixed day’ sacred to the goddess (Cerialia, April 19) was simultaneously the anniversary of her Aventine temple. By contrast, the anniversary of the temple of Flora fell on the first day of the ludi Florales (April 28). Games of Hercules Magnus, perhaps officially established by Sulla, are likely to have been connected with the anniversary of the temple in the Circus Maximus ( June 4).167 Last but not least, the anniversary of the Capitoline temple fell within the ludi Romani (September 13). This meant that all public Games until the victory Games of Sulla, established in 82 B.C., were directly linked to a specific temple via its anniversary. This tendency may well have continued in the Empire, for there is a reasonable chance that the temple of Mars Ultor was dedicated by Augustus on May 12 in 2 B.C., a day on which ludi Martiales are attested.168 Temporal foci such as temple anniversaries and Games were thus combined in these cases in order to reinforce each other.
It is not exactly clear to what extent the ‘announced holidays’ (feriae conceptivae) were adjusted to the pattern of the ‘fixed holidays’ (feriae stativae). We may tentatively turn to the sacrifice to Dea Dia, whose shifting dates are known from 21 B.C. onward thanks to the survival of the acts of the arvals. We find that during the imperial period the sacrifice to Dea Dia was performed either on May 17, 19, 20 or on May 27, 29, 30, i.e. on days that were not occupied by another god according to the Republican feriale. Even in the very few cases where the acts mention other dates for the festival, these dates do not as a rule coincide with the ‘fixed holidays’ of the Republican calendar. However, there may be one exception: the sacrifice to Dea Dia in 66 A.D. was performed, for whatever reason, on July 17–19–20, even though July 19 according to the Fasti Antiates Maiores was a fixed holiday (Lucaria).169 The obscurity of the deity involved may be the reason for pontifical indifference. However, one should be careful not to draw far-reaching conclusions from this seeming exception. The sacrifice to Dea Dia was a specific ceremony rather than a ‘holiday’ (feriae),170 and the information afforded by the acts started in or around 21 B.C.; there is therefore no direct link to the pre-Caesarian calendar, as represented by the Fasti Antiates Maiores. On the other hand, if the imperial dates of the sacrifice to Dea Dia did indeed take account of the Republican calendar, the manner of calculating them may actually be much older.171
‘Private holidays’ did not follow the pattern of ‘public holidays’. For instance, they could fall on an even day, such as the Caristia (or Cara Cognatio) on February 22. Or they could coincide with other ‘public holidays’. One may refer to the Parentalia, the ‘holidays’ of the di parentes, which began on February 13 with a sacrifice by a vestal virgin and ended with the Feralia on February 21. The Parentalia, therefore, included the Ides (February 13), sacred to Iuppiter, and the Lupercalia (February 15), which were sacred to Faunus. One may wish to argue that the participation of the vestals here indicates a public cult. But the very name and nature of the parentalia (referring to one’s ancestors) suggest otherwise. Only the last day of the Parentalia, the Feralia, were feriae publicae.172
The number of days in a calendar year was limited. Since the day was the basic unit for temporal foci of Roman gods, an overlap of such foci became inevitable over time. As in other areas, the Caesarian era forms a watershed here. When Caesar’s victory Games (dedicated to Venus Victrix) were established from July 20–30 in 46 B.C., they included no less than three ‘fixed holidays’, the second day of the Lucaria on July 21, the Neptunalia on July 23 and the Furrinalia on July 25. The Games were not connected to any temple anniversary, not even to that of Venus Genetrix, although her temple was dedicated just two months after the establishment of the Games (September 26).173
Technically speaking, it was Caesar’s authority as pontifex maximus that entitled him to interfere with hoary traditions of temporal focalization. He put his powers to good use during his famous calendary reforms.174 Similarly, Caesar’s imperial successors were all pontifices maximi, and all made similar use of their powers to tamper with the inherited Republican calendar. It is a fact that between 38 B.C. and 17 A.D. at least fourteen temples, which had been restored, had their dies natalis changed, some with the clear objective of ‘synchronizing’ their anniversary with an imperial holiday or other important imperial events.175 In contrast, temporal foci of the imperial cult, especially of the cult of Augustus, could influence the choice of imperial ‘holidays’. For instance, Caligula accepted the title pater patriae, bestowed on him by the senate, on September 21, perhaps having in mind the temporal closeness to Augustus’ official deification on September 17 as well as to Augustus’ birthday on September 23. In the same vein, Caligula dedicated the temple of Divus Augustus on August 30, a day before his own birthday. It was hardly by chance that the posthumous consecration of Livia, Augustus’ spouse, and her wedding anniversary with the princeps, as well as the dedication of the altar of the numen Augustum by Tiberius all fall on the same date, January 17.176
However, the emperor had the power not only to add, but also to remove ‘fixed’ holidays. Caligula abolished two Augustan ‘holidays’,177 Claudius rescinded even more imperial feriae, “for the greater part of the year was given up to them.”178 In 70 A.D., a senatorial commission was set up in order to purge the overloaded calendar of unwanted or outdated ‘holidays’.179 During the same period, the arvals, no doubt following imperial directives, restricted sacrifices on behalf of the emperor.180
Generally speaking, during the imperial period, the clearly defined temporal foci of a number of the most important Republican gods lost their distinctive focal nature thanks to the infiltration of the imperial cult and its disintegrative impact.
There were no competing modes of time-reckoning in the Republic to any significant degree. Even cults that were considered a threat to society, such as the cult of Bacchus at the beginning of the second century B.C., appear to have adhered to the traditional fasti, albeit with some unavoidable modifications in detail due to specific ritual requirements (initiations etc.). During the Augustan period and perhaps earlier, Magna Mater was officially worshipped on March 27 (lavatio), a day still vacant on the festive calendar, apart from her temple anniversary and Games held at the beginning of March (see above).
During the imperial period, this situation changed dramatically. Competing systems of time-reckoning emerged which ignored the temporal foci of traditional gods. For instance, from the first century A.D. the official worship of Magna Mater was gradually extended to a cycle of six days that included March 15, 22, 24–27. The cycle took no longer account of the ‘fixed holidays’ of Iuppiter and Anna Perenna on March 15 or the Tubilustrium of Mars on March 23. One may also refer to the Christian time-reckoning, which was revolutionary in replacing the Republican week consisting of eight days (nundinum) by the hebdomadal week with Sunday as the basic temporal focus, quite apart from the fact that all Christian temporal foci referred more or less to a single annual event, namely Easter Sunday.
Despite the dearth of relevant material for the city of Rome itself, it is a fair guess that the lunar calendar of the Jews was still in use even in profane matters in the Roman period (in religious matters it never lost its importance), and perhaps was instrumentalized as a token of Jewishness in opposition to the Julian solar calendar. At least, such a deliberate instrumentalization can be plausibly postulated in the case of the Jews of the eastern Empire, in order to effect and advertise cultural distinctiveness.181 Besides, Jews everywhere—and hence also in Rome—observed biblical festivals, most characteristically the Sabbath.182 The latter was a common target of pagan mockery,183 although it was explicitly tolerated by Augustus and Tiberius.184
Let us turn to the Christians. Sunday was already of special importance for the community in the first century A.D. It became the firmly established date of the weekly celebrated eucharist not later than the second century A.D., perhaps in deliberate contraposition to Judaism.185 At the same time the observance of the Sabbath lost its importance among the Christians, especially in the West. The observance of Sunday became all the more a genuine mark of distinction from Judaism and was eagerly advertised by the Christians as such.186 It was rendered compulsory by Constantine in 321 A.D.187
Easter was the only annual festival celebrated consistently by Christians during the first three centuries. It developed from the Jewish Pascha, since it was during this period that Jesus had died, according to the canonical scriptures. What remained in doubt was the question how the date of Easter should be calculated. On this question, the Roman see took a position against the Christian communities of Asia and Syria.188
The relative lack of temporal foci in the Christian church during the first three centuries of its existence, apart from the observance of Sunday and Easter, is undeniably impressive. It is only partly compensated for by the veneration of defunct bishops and martyrs which began to develop in the capital from the middle of the third century A.D. onwards.189 Such memorial cycles and martyrologies are first attested by two famous sections of burials (so-called ‘depositions’) of bishops and martyrs in the Calendar of Philocalus, listing the dates of death of the bishops of Rome and Roman martyrs from the mid-third to the mid-fourth century A.D.190
If we want to characterize the conceptualization of the ‘divine’ in relation to the constituent concept of time in more general terms, we have to begin with the observation that there was a clear line between those days in the calendar that served this conceptualization (feriae) and those that did not. The relation of the various feriae publicae to each other was—generally speaking—well defined and restrictive (e.g. ‘fixed holidays’ on uneven days, no ‘fixed holidays’ on successive days, etc.). These definitions applied to all divine concepts alike and thus reflect the same lack of individuality of temporal foci as was the case in the employment of spatial foci: for instance, just as the formation of ‘Iuppiter’ was spatially marked by the size of his temple rather than its architecture and layout, so too it was temporally marked by the number of ‘holidays’, not the rituals performed on them. As in the case of spatial conceptualization, we find in the calendar a sharp distinction between divine and human concepts, while within the category of ‘divinity’ all gods were treated as essentially being the same. This balance was challenged by the imperial cult, which actually blurred the existing dichotomy between ‘divine’ and ‘human’. By doing so, it became a much more disintegrative force than, say, most foreign divine concepts which arrived in Rome in the imperial period. For the latter did not come anywhere close to challenging the dichotomy between ‘divinity’ and ‘humanity’ in terms of their temporal conceptualization. Christianity, of course, differed, on this as on other points. As indicated by Beard,191 while the pagan calendar was ‘polycentric’, i.e. a conglomeration of various temporal foci unrelated to each other, the Christian calendar centered around one single historical event, the crucifixion of Jesus on Nisan 14. Gradually, the whole Christian year was constituted around this date. Both in its ‘monocentric’ outlook and in its emphasis on a specific moment of human history (Jesus’ death), it differed substantially from all modes of temporal conceptualization of the divine known up to then in the Mediterranean.
See endnotes and bibliography at source.
Chapter 1.2 (30-50) from Roman Gods: A Conceptual Approach, by Michael Lipka (Brill, 04.30.2009), published by OAPEN under the terms of a Creative Commons Attribution-NonCommercial 2.0 Generic license.