3.1. Egyptian or Mesopotamian origins?

1It has been suggested that Greek incubation developed as a result of influence from Egyptian dream oracles.1 This assumption might be supported in Herodotus: he writes on many aspects of Greek religion, among them divination, and posits that they originated in Egypt, though he does not mention incubation as such.2 G. Zographou, after studying the evidence for cultural influence from Egypt, arrived at the conclusion that influences may have gone from Egypt to Greece and also from Greece to Egypt, but Herodotus simply chose the Egyptian origin.3 Dreams were important in Egyptian cults, as is also documented by the many inscriptions,4 but it cannot be proven that incubation was practised before the Late period in Egypt.5 Before Diodoros wrote that Isis showed herself to suppliants in their dreams, there are very few testimonia relating how Isis and Sarapis helped their worshippers in need.6 Thus, it cannot be argued that the phenomenon of incubation first appeared in Greece as a result of Egyptian influence.

2The lack of evidence for incubation in Egypt before Ptolemaic times has led Dunand and other scholars to suggest that Egyptian incubation for healing purposes did not exist before the Ptolemaic period and that it was influenced by Greek incubation.7 The rise in popular demand in Egypt for kind deities curing through dreams in the third century BC, during the Ptolemaic age, coincides with the rapid spread of the cult of Asklepios, which came to Pergamon in possibly the fourth century and to Rome in 293 BC. This influence cannot be ignored.8 Yet another hypothesis on the origin of Egyptian incubation is that several Egyptian oracles which developed during the third century BC introduced rituals that appealed to a large number of worshippers, e.g. incubation and healings, as part of the oracular agenda.9 This is highly plausible because dream interpretation and the existence of many oracles were long-standing features in Egyptian religion, and the possibility of combining dreams with cures might have been a natural development. Incubation in Egypt in all probability arose through a combination of indigenous development and external influence.

3In contrast with the theories on Egyptian origins, it has been suggested that Greek incubation was introduced as a result of influence from Mesopotamia, via the Meander River.10 This argument mainly rests on the kingly incubation enacted by Gudea of Lagash, around 2100 BC, having made its way to Greece.11 The time gap and the difference between a sought-after, divine dream by a king and the form of incubation found many years later in Greece makes the connection unlikely, and just as Wacht points out, the idea of receiving oracles through dreams may have arisen at different places independently.

4In Mesopotamia, dream books were of great importance, and their existence points to an early interest and belief in god-sent dreams. Dream books appear in the Old Babylonian Kingdom in the eighteenth century, as well as in Assyria in the seventh century BC.12 Dream interpretation of a dream which did not necessarily take place in a sanctuary and incubation are two different phenomena. The ‘Professional’ interpretation of dreams that people had dreamt at home was a widespread phenomenon, one which was much larger than the practice of incubation, where dreams were actively sought by worshippers at a sacred locality. Dreams that can be defined as incubation dreams, received in a god’s temple after due preparation, are rarely found in the evidence from Mesopotamia, and only concern kings consulting on matters of political importance.13

5It can be concluded that there is no evidence to support the idea that Greek incubation had its origins in either Egypt or Mesopotamia. Cults, as well as cultic habits, spread with people, but some common human features such as the belief in the transcendence of dreams and the need for oracles and healings, can be found all over the world. Thus, it is fully possible that incubation cults in Greece developed independently, just as it is possible that oracle cults changed their methods and targeted group of worshippers over time.

3.2. The Earth and incubation

6Many scholars view incubation as having originated in old oracles of Ge, as this goddess, the Earth, was thought to engender divinatory dreams. It has also been suggested that incubation was only practised in what some scholars have defined as Chthonian cults.14 Essentially, incubation would thus have its roots in pre-Indo-European cults associated with the Earth and then iatromantic hero cults.15

7The Olympian-Chthonian model, which dominated the study of Greek religion in the twentieth century, has been much criticized by modern scholars. Even if there is not a strict partition, especially ritually (as seen in the epigraphical evidence), certain gods of the Greek pantheon were called Chthonian by the Greeks, and many gods were associated in some way with the earth.16 Still, the presence of cults to Ge and the application of epithets like Chthonios to various gods (including Zeus), need not prove a binary model with two sets of gods who differed in the rites they required.17 In this section, I will not investigate the whole Olympian-Chthonian question; rather, I will study the testimonia relevant for incubation rituals and test whether they conform to the relevant hypotheses put forward in previous research.

8There are three presuppositions which will be investigated here. First, that the rituals of incubation were Chthonian, as defined by the followers of the Olympian-Chthonian school. Secondly, that Ge was the Sender of Dreams and that dream oracles were given in her cults. Thirdly, that incubation in historical times entailed sleeping on the tombs of heroes in order to draw on the supposedly oracular and healing powers of the Earth.

Incubation rituals as Chthonian

9It has been shown in this study that incubation rites did not conform to Chthonian rites as defined by the scholars adhering to this model (Section 2.1.1). The only case of an incubation rite that might be considered Chthonian by a scholar adhering to this model is the sacrifice of a black ram (the verb enagizein is used) in the cult of Kalchas on Mount Drion, which is attested in Strabo.18 However, as previously mentioned, this rite need not have existed in Hellenistic times, and it may have been the result of the antiquarian tendencies of the time of Strabo.

Dreams as sent from the Earth

10Secondly, we will now explore whether the Earth was considered by ancient peoples as the Sender of Dreams, and whether the existence of dream oracles might be proven or conjectured in her cults. If we look at the testimonia on the origins of incubation inferred by the researchers using this model, Homer’s is the earliest work. The Selloi, priests at the sanctuary of Zeus at Dodona, who according to Homer slept on the ground with unwashed feet, are often mentioned as the earliest example of incubants, and this also, according to some scholars, indicates that incubation originated in cults to the Earth.19 According to Parke, however, the sleeping on the ground described by Homer need not imply incubation as a means of divination.20 The passage in Homer mentions that the Selloi are Zeus’ interpreters, that they dwell around him (i.e. in the sanctuary) and that they have unwashed feet and lie, or sleep, on the ground, χαμαιεῦναι. There is nothing in the passage on receiving dreams while they slept. They were not barefoot, as has been stated in previous translations, but had unwashed feet.21 According to Georgoudi, the Selloi distinguished themselves as a group of oracle priests by living near the sanctuary and not washing their feet.22 It is likely that the mention of them sleeping on the ground was a note on how the life of the interpreters differed from normal life, and this differentiation was strengthened by their unwashed feet. This ‘primitive’ life was surely in itself a way of enhancing the oracular powers of the priests, but the sleeping as such need not have been the divinatory method.23 In Sophocles’ Trachiniae, the Selloi are described as dwelling in the hills and sleeping on the ground, but it is made clear that it is the oak of Zeus that speaks and gives the oracles.24

11At Dodona, a large collection of questions to the oracle noted down on some 1400 lead tablets has been preserved, dating from the sixth to the end of the second century BC.25 As for the technique used in obtaining the answers to these questions, several methods may have co-existed. Among those which are noted in literature are the interpretation of the rustle of oak leaves,26 possibly the interpretation of actions of doves in the sanctuary,27 the possible use of prophetically inspired priestesses,28 a lot-oracle,29 and the use of bronze tripods.30 There is thus no indication that incubation was practised at Dodona, neither by priests or priestesses, nor by worshippers.31

12Two of the tragedies of Euripides, Iphigenia in Tauris (on the takeover of Apollon at Delphi) and the Hecuba, have been cited as proof that divination via dreams was connected with the Earth.32 In the Iphigenia in Tauris, which was written before 412 BC, the Earth is angered that her daughter Themis has been removed from the oracle of Pytho at Delphi by Apollon. Earth therefore took the office of prophecy away from Apollon and started sending oracular dreams (seemingly for people sleeping on the ground).33 Zeus intervened and gave the mantic ability back to Apollon.

13The passage in the Iphigenia in Tauris refers to the myth that Ge and Themis were the first to give oracles at Delphi. No archaeological evidence for an earlier oracle of Ge has been found at Delphi, however, and it is just as likely that the account of Apollon taking over a previous oracle of the Earth at Delphi is purely mythological and has no historical foundation.34 Furthermore, caution should be advised when ascribing historical truth to information found in tragedies on Greek religion.35 Euripides wrote his play in the late fifth century, a long time after any takeover of Delphi was supposed to have occurred. Since there is no supporting evidence on incubation in earth oracles, the text of Euripides must be used with care.36 The purpose of myth is not necessarily to reflect experience, but to rework and clarify it.37 Even if oneiromantic earth-oracles existed, from which no testimonia are preserved, this need not mean that all instances of incubation originated in these. It should also be pointed out that in the Hecuba by Euripides the Earth is called the ‘mother of black-winged dreams’. These are dreams that Hecuba had at home, not oracular dreams of people sleeping on the ground in search of an oracle.38 The passage cannot support the notion that oneiromancy existed in earth cults. After all, dreams and dream oracles are two different phenomena.

14The Earth is not the only god in Greek literature described as being responsible for sending dreams. In the Iliad, Zeus sends dreams.39 According to the Olympian-Chthonian division of Greek religion into a system of two opposing spheres, the first sender of dreams was the ruler of the Olympian gods. Athena, a good Olympian, also sent dreams at her sanctuary in Lindos in 490 BC.40 In the Theogony, Dreams are referred to as the children of Earth’s daughter Night, Nyx.41 There is nothing surprising about Night being the mother of dreams, and maybe Earth became the mother of dreams in later mythology by a simple shortening of the ancestral line of the Hesiodic theology. Tragedies written in Classical times need not have made a historically reliable statement when describing Earth as the first sender of oracular dreams, and there is no cultic or ritual evidence to suggest that oracular dreams or incubation would have been associated with cults to the Earth.42

Hero cults as Chthonian cults, and the act of sleeping on the tombs of heroes in incubation

15Finally, we examine the idea that incubation originated in iatromantic hero cults, as argued in modern research by Riethmüller, Wacht and Dorati.43 Riethmüller argues that dream oracles and incubation in Greece were connected with hero worship, and that the help-seekers lay down on the graves of the heroes or at the place where they had emerged from the earth, in order to draw upon the healing and oracular powers of Ge.44 Riethmüller argues that the Asklepieia and the incubation ritual practised there exhibit a concordance with these iatromantic hero cults.

16For investigating the argument on the efficacy of incubation that came from the powers of the Earth, it suffices to determine, first, whether incubation originated in hero cults, and secondly, whether incubants slept on tombs or directly on the ground (so as to draw from supposedly mantic and healing powers of the Earth).

17Asklepios was still considered a hero, though a hero who received Panhellenic festivals at Epidauros, in the poems of Pindar, but he was perceived as a god by the time incubation is securely referenced in the iamata.45 Nowhere in the entire record of incubation in Asklepieia did incubants sleep on the earth itself, but rather they slept on makeshift beds, klinai, or occasionally on animal fells on klinai.46 There is a hypothesis that the Tholos at Epidauros was a symbolic tomb for the hero Asklepios, but it is one of many hypotheses that exist in relation to the function of this building.47 In any case, the incubants at Epidauros slept in the abaton and not in the Tholos.

18The same argument goes for Hemithea at Kastabos: she was a heroine in myth, but when worshipped in her sanctuary at Kastabos, she was, probably because of the size of her cult and sanctuary, perceived as a goddess.48 There are theories on where the dormitory may have been in her sanctuary, but the fragmentary state of both the archaeological and textual evidence makes for much uncertainty.49 There is no indication that incubants ever slept on the ground at Kastabos.

19Amphiaraos was a hero when incubation in his cult in Boeotia first appeared, but the size of his sanctuary at Oropos, which was constructed in the late fifth century BC, makes his status more that of a local god.50 In sources from the early fourth century he is also called a god.51 Amphiaraos possessed mantic powers, but his oracular powers were accounted for by his descent from the seer Melampos, or Apollon himself. Amphiaraos was traditionally swallowed by a chasm in the ground, and his tomb has not been shown to have been represented in his sanctuaries.52 Nowhere is it said that incubants in his cult slept on the ground and drew from the supposedly mantic powers of the Earth.53 The help-seekers at his sanctuary at Oropos slept on fells on klinai or just on klinai inside a dormitory.54 At his sanctuary in Rhamnous, a dormitory has been hypothesized in a portico opposite his shrine; thus, incubants would in all likelihood also have slept there in a dormitory and not on the ground.55

20In the cult of Podaleirios in Apulia, incubants consulting in order to obtain answers slept on the tomb of the hero Podaleirios (as attested by Timaeus of Tauromenion, cited in the Alexandra of Pseudo-Lycophron).56 In a later testimonium on incubation on Mount Drion, oracle-seekers slept somewhere close by or in the heroon of the seer Kalchas (as attested in Strabo, where there is nothing on incubants sleeping on Kalchas’ tomb).57 Petropolou58 argues that the origins of the hero cults of Kalchas and Podaleirios may be found in necromancy. The sacrifice of a black ram to Kalchas that appears in Strabo and the fact that the purpose of the sacrifice was to reveal the future through sleeping on a tomb have been taken as indications of such an origin. The sacrifice of a black ram is, however, not mentioned in the text of the Alexandra (on the cult of Podaleirios), and Strabo (on the cult of Kalchas) does not mention incubants sleeping on the tomb of Kalchas — he only writes of the sacrifice of a black ram and sleeping in its fell. If the tomb of Kalchas was considered empty, as stated in the Alexandra, also at the time when Strabo wrote about the cult, there would have been no point in sleeping on an empty tomb to contact an absent hero. Only in the cult of Podaleirios is it attested that incubants slept on the tombs of heroes. In any case, sleeping on fells has nothing to do with necromancy, but rather it shows a wish to communicate with the hero.59

21Incubation was a widespread phenomenon in the ancient world when Timaeus was writing. It seems logical that the seer of the Iliad, Kalchas, gave oracles, and that the son of Asklepios, Podaleirios, gave rise to dream visions as well as healing people. Maybe the unusual rite of sacrifice of a black ram was added in early Roman times (when Strabo mentions it), and thus it is a later modification that has nothing to do with the origins of the cult. In the cults of Podaleirios, incubants slept on the hero’s tomb, but this need not be because they were drawing on the mantic powers of the Earth. It could also be the case that the incubants simply wanted to come as close as possible to the hero with whom direct contact in the dream was sought.60 This wish to be close to the hero is also indicated by sleeping on the fleece of the animal sacrificed to him.

22To sum up, if we examine the rites surrounding incubation, there is nothing that would be considered Chthonian by the scholars adhering to this model, except for the enagizein-sacrifice of a black ram in the cults of Kalchas and Podaleirios. Since this black ram is attested first in Strabo, it may well have been a later addition to the ritual. As for the hypothesis that incubation cults arose from cults of the Earth, there is no evidence to support this. Neither is there any evidence to support the idea that the Earth was a first ‘Sender of Dreams’ or that incubation was ever practised in her cults in historical times. As for the theory that incubation developed from iatromantic hero cults, it is true that Hemithea, Asklepios, Amphiaraos, Kalchas and Podaleirios were heroes, although Hemithea and Asklepios were considered gods by the time that incubation can be identified in their cults.61 Incubants, however, never slept on the earth itself in these cults, with the exception of the cult of Podaleirios, where they slept on the hero’s tomb and not on the earth per se. As for sleeping on the tomb of Podaleirios, this practice is best explained by the incubants wanting to come as close as possible to the hero whom they sought to contact. Thus, there is no evidence to prove that the incubants called on the supposedly mantic and healing powers of the Earth.

23Having noted that incubation as enacted in cults had nothing to do with the supposedly mantic powers of the Earth, it should also be mentioned that incubation was, however, first practised in hero cults. There might be many reasons why incubation was not practised in the cults of the gods who were viewed as Olympians, such as Zeus or Athena. One explanation is that heroes, or deified heroes, were simply more easily approachable for the individual worshipper than Zeus or Athena. In a direct communication with a divinity, things might go wrong, as for instance in Plutarch, when a consultation continued despite the sacrificial victim not having nodded in approval, and the Pythia, who was equally unwilling to proceed, was then seized by a fit and died some days later.62 Asklepios, on the other hand, was always a gentle god who differed from most Greek divinities and had both friendly and hostile aspects.63 The many healing heroes of Greek religion are proof that worshippers felt they could easily approach locally based helpers.64

3.3. From oracle to incubation

24The origins of incubation in Greece were not Egyptian, Mesopotamian, or in pre-Hellenic earth cults, and so the focus of investigation will now shift to what is known about the beginnings of the first, known incubation cults in Greece, focusing on the cults of Amphiaraos and Asklepios because these were decidedly Greek phenomena and they also yield the oldest and largest testimonia concerning the rites and rules surrounding incubation.

25On a conceptual level, asking for an oracle and asking for a healing are closely related phenomena. Oracles were consulted to learn the will of the god or to find out the proper course of action in a difficult situation, and what question would have been more natural for an ill person than to ask what he or she might do in order to get well? In modern research, it is well accepted that iatromancy was a subdivision of oracular practices. Dreams, after all, were, according to Plutarch, the ‘oldest oracle’.65 In many oracle cults, healing was a goal of consultation,66 and incubation cults were able to help with making decisions or finding lost objects.67 In the large corpus of healings, iamata, from Epidauros, there are events that would best be defined as dream oracles given to the enquirer directly, that is, messages that were conveyed by the god in a dream inside the abaton, and also instant cures where the god and his associates operated on the patient in his or her sleep.

26An understanding of the origins and development of incubation rituals is necessary in order to explain the structure of the rituals because they must have been adapted for their enactors, and they would have developed as the composition of the worshippers, as well as their wishes and needs, evolved with the rest of society. We do not know of any cults, before the rise of the cult of Asklepios, where help-seekers en masse and indiscriminately could seek direct contact with a deity, without any medium or intermediary. It is therefore of interest to explore how direct contact with a god, something which had been reserved for only a few in Archaic times, came to be a popular healing ritual that was enacted by even women and slaves.

27Some basic definitions need to be made concerning oracle consultation. An oracle cult was a cult that offered a ritual which would enable the enquirer to consult a higher power on a matter with which he or she wanted assistance.68 A dream oracle, which offered oneiromancy, was the same type of cult, and here the technique for consulting the divinity was through a dream. An iatromantic cult primarily offered healings through oracles (these could be acquired using various techniques).

28The most generally acknowledged partition of oracular techniques is into inspired prophecy on the one hand (e.g. the Pythia at Delphi and oneiromancy) and the interpretation of signs on the other (e.g. the flight of birds, the appearance of the intestines of the sacrificial animal, and lots).69

29In inspired prophecy there are several kinds of intermediaries who can obtain a response from the deity. First, one or several priests of the cult can seek this direct contact with the deity on behalf of the enquirer. Secondly, a specially appointed priest can act as the medium, a priest who is somehow different from other cult personnel. Thirdly, the enquirer him- or herself may contact the deity directly, without an intermediary. The ‘enquirer’ is the person asking the question. The ‘consultant’ is the person going through the consultation process, be it either the enquirer him- or herself, or the priest, or some other official appointed by the cult to consult the deity on behalf of the enquirer.

3.3.1. Amphiaraos in Boeotia

30The first mention of incubation in Greece is found in Herodotus, when he describes the cult of Amphiaraos in Boeotia.70 As elsewhere in Greece, in Boeotia, direct consultation by inspired prophecy was the most common form of divination in oracular sanctuaries.71 The oracle of Amphiaraos was first situated by Thebes, but it flourished at Oropos in Classical times.72 In this cult, the enquirer him- or herself consulted by sleeping in the sanctuary and receiving a dream that was believed to have been sent by Amphiaraos.

31The two instances of consultation in the cult of Amphiaraos that are retold in Herodotus concern the Lydian King, Kroisos, who likely consulted a short time before his defeat by Kyros in 546 BC,73 and General Mardonios’ envoy Mys, who consulted the hero in 479 BC.74 The subject of Kroisos’ enquiry was whether he should attack the Persians or not (though Herodotus does not mention which consultation technique was used on this occasion). Scholars writing on the consultation of Kroisos have advised a cautious approach, since in Herodotus’ account, Kroisos consulted oracles which were famous in the fifth century BC and not geographically closer oracles which were known in the sixth century, as would perhaps have been more natural.75 Furthermore, Herodotus himself writes that there is no record of the response given to the Lydians by Amphiaraos, which may indicate a later fabrication in order to augment the fame of the oracle.76 Consequently, the first reliable account of the oracle of Amphiaraos deals with events of 479 BC.

32It is clear from the description of Mardonios’ consultation that incubation was already well established and well known when Herodotus wrote about it in the 450s or 440s BC. The Persian General Mardonios enquired through his envoy, Mys, as to the outcome of the second Persian war, preceding his final defeat at Plataiai in 479. The veracity of this consultation, 30 years prior to the text of Herodotus, has not been questioned by scholars. Mardonios was waiting for the decisive battle and enquired of a number of oracles. Mys, who was given this task, had to bribe a Greek who was a non-Theban (since apparently neither foreigners nor Thebans were allowed to consult Amphiaraos) to lie down to sleep in the sanctuary of Amphiaraos.77 The story need not imply that the oracle of Amphiaraos held international renown at the time since Mardonios consulted the oracles that were close at hand. If only Greek consultants were allowed, then the oracle cannot have been of international fame, or, at least, it could not have been sought by foreigners, which makes the story of Kroisos consulting even less likely.

33The first certain instance of incubation is thus recorded as having happened in 479 BC, and the cult, even if it was not visited by Kroisos just before 546 BC, presumably had some tradition of incubation before 479, and this possibly went as far back as 500 BC or even earlier.

34The consultants of Amphiaraos in the time of Herodotus were more eminent than the usual incubant at an Asklepieion in Classical and Hellenistic times. To compare with the oracle at Delphi, envoys from citystates and also private citizens, primarily from Delphi, consulted Apollon, but the enquirers did not consult Apollon themselves.78 The Amphiareion was a cult to a hero where the enquirers themselves consulted, but in relation to the evidence for Archaic times, it is clear that not just any enquirer was able to approach the dormitory. They were Greek, yes, and non-Thebans, but Herodotus’ account does not say that women and slaves went to consult this oracle. Mardonius consulted by proxy, using a Greek, but it was still his questions that were put to the hero. In the Herodotus’ account, all of the other oracles in the vicinity that were consulted by Mardonios were exclusive oracles: for instance, at the oracle of Ptoian Apollon above the Lake Kopais, the envoy of Mardonios was accompanied by three men chosen by the state to write down the oracles given by the diviner.79

35The very nature of any oracle consultation was that of exclusivity; the enquirers often came when specially chosen or on certain days of the month, and they enquired through officials of the cult. To be able to consult by sleeping in the sanctuary, as in the cult of Amphiaraos, was highly unusual, and there is no reason to suppose that the consultation procedure there, in Archaic times, was for everyone.

36It is not known how many visitors came to the first oracle at Thebes, or even how big the sanctuary was, because the location is not known, but since Herodotus always mentions the oracle of Amphiaraos together with the oracle of Trophonios, it might be suspected that both heroes’ sanctuaries were about the same size and enjoyed equal popularity.80 Since Herodotus only speaks of oracles in relation to obtaining answers in crisis situations, it is improbable that Amphiaraos performed healings at this time, other than what any oracle would have done.81 In the accounts of Herodotus, Mardonios camped in Thessaly in 480/479 near oracles. Hence, the oracle no doubt was a Greek, since it appears in no other sources of the time. It is unlikely, given the absence of evidence besides that in Herodotus, that worshippers came to consult the oracle in large numbers, as they did at Delphi or Dodona. As for the rites at this oracular shrine, the paying of a consultation fee, as can be noted in a later sacred law from the cult, may well have existed from the beginning.82 Also, sleeping in the fleece of the sacrificed animal may have been a remnant of this first oracle cult.

37In the first records in the Greek world concerning an incubation sanctuary, it is related that the purpose of sleeping was to receive guidance on important decisions and not to get healed. Thus, incubation probably first appeared in Greece as a method of obtaining oracular answers on the best course of action in a crisis, and the main purpose was not to be cured of illness. A move towards healings and the mass-consultation of Amphiaraos can be noted for the first time in 414 BC in a fragment of a play by Aristophanes.83 Schachter and Petropoulou argue that Amphiaraos at Oropos started as a healing deity, possibly during the plague of the Peloponnesian war, that it was modelled on the popular cult of Asklepios in Epidauros, and that it offered a local alternative before the transferral of the cult of Asklepios to Piraeus and Athens in 420/419.84 I would suggest that political considerations may also have played a part, and thus the Amphiareion was built as a competitor for the originally Epidaurian cult. Petropoulou argues convincingly that the sanctuary of Amphiaraos at Oropos was built in 420–414 BC: no archaeological structures that have been excavated there date to before the late fifth century, and Athens, which at the time ruled Oropos, did not build any constructions at all on the mainland during the Archidamian war of 431–421 BC.85

38The new focus of the cult at Oropos can also be noted in an inscription from 386–374 BC.86 Here, rules were set out to ensure the smooth running of business in the sanctuary, the proper behaviour of worshippers, and the duties of the priests, when they were present. Any animal could be sacrificed. Putting up the name of the person who had paid the consultation fee on a note by the dormitory suggests that many people were present at the sanctuary, and also that incubation was popular. There would have been no reason to set up the names if, in the past, those people who had not paid had not attempted to enter the dormitory. The inscription also states that men and women were to be separated in the dormitory, indicating that women also incubated in the sanctuary. Consequently, we can note that incubation for the sake of a cure was popular at this time. Consultation on important decisions continued in the fourth century alongside the popular healings.87 The monumentalization of the sanctuary speaks of the need to accommodate a large number of visitors. It is likely that this redirection of the purpose of consultation in the cult of Amphiaraos, from oracular responses to healings as well, prompted the need for a larger sanctuary, and such a redirection can also be noted in relation to the sanctuary of Oropos (a large incubation stoa was constructed in 360–340 BC).88

39The inclusion of many more consultants and the new aim of obtaining cures through the consultation might have changed the ritual for the consultants. The rules on the separation of men and women also indicate that many incubants were now to be accommodated. Thus, the cult of Amphiaraos underwent a development from being an oracular cult in which, at least as far as we know from Herodotus, a few Greek men consulted the hero by incubating for the sake of obtaining answers in a crisis situation, to a healing cult, where many help-seekers, both men and women, incubated.

3.3.2. Other early oracles that offered direct contact with a divinity

40How did enquirers come to believe that they could consult the deity themselves, without an intermediary? And how was a ritual formed to allow for this meeting of worlds, a meeting where a person who was not an official of the god could communicate directly with the god?89 When incubation in the cult of Amphiaraos in Boeotia was first introduced, similar consultation techniques, where enquirers themselves could come into contact with the deities, were already in use in the area, where enquirers themselves could come into contact with the deities. Incubation in the cult of Amphiaraos might have been a variation of local consultation techniques that involved the worshippers themselves.

41In an oracle contemporary with that of Amphiaraos in Boeotia, which is attested in Herodotus and Euripides, another son of Apollon, Trophonios at Lebadeia, gave oracles to single, notable consultants.90 Herodotus describes how Mys bribed one man to go down into the cave of Trophonios. Here, the consultation was such that the enquirers themselves (after careful preparation) went down into the oracle’s cave, where they were given some sort of vision, and were then to be taken up and go through rites to remember and interpret what they had experienced.91 The fact that only one person could enter at a time, and that consultation was expensive, indicates that the selection process of consultants was strict. The rites enacted by the consultant were costly, time-consuming, and they involved the active presence of the priests of the cult. The preparatory rites in particular involved parts resembling the purity instructions for Greek priests about to communicate with the divine.92

42Another contemporary religious phenomenon that might help place the cult of Amphiaraos in context is the consultation of the nymphs by several of the local inhabitants by the cave of Sphragidion, on the border between Attica and Boeotia.93 The nymphs of the cave, which are mentioned in Kleidemos — a historian of the fifth or fourth century BC, inspired many of the locals, and apparently the locals were also given oracular abilities (they were then consulted by travellers). It is not certain, but some of the people living near the cave probably thought that they had received visions from the nymphs, and as the phenomenon became famous, travellers then came with queries and asked these villagers to consult the nymphs on their behalf.

43Herodotus writes about an oracle of Dionysos in Thrace, in the land of the Satres.94 Here, the Bessoi, a clan of the Satres, served as προφητεύοντες and interpreted oracles uttered by a priestess called a πρόμαντις by Herodotus. The Bessoi were not in direct contact with the god themselves, but they were considered to possess some special abilities for which they had been chosen to interpret the utterings of the priestess. They might be compared with the Selloi at Dodona,95 or the Branchidae at Didyma,96 even though the Bessoi did not receive the oracles themselves. There is nothing on incubation in this cult, but here Herodotus highlights a variation in oracular techniques, where the mantic abilities were not exclusive to one medium but rather to a group of specially chosen.

44Another oracle where a select group of consultants were able to contact the deity themselves was the oracle of Pasiphaë in Thalamai, close to Sparta, where the Spartan ephors would go and incubate in order to obtain oracular advice on matters of the state. The sources for this oracle are much later than the events which they relate, and so prudence should be advised in recreating the oracular process. Cicero in De divinatione relates how the ephors of Sparta would sleep in the shrine of Pasiphaë near the city in order to dream and obtain oracular answers.97 Plutarch provides more information on the oracle, writing that the oracle of Pasiphaë was used at the time of Agis, King of Sparta in 262–241 BC, as well as Kleomenes, King of Sparta in 235–222 BC.98 Plutarch’s main source was the Greek historian Phylarchos, from the third century BC, so the cult of Pasiphaë could be at least Hellenistic.99 In the passage on Kleomenes, Plutarch relates how an ephor slept in the shrine of Pasiphaë and received a dream that guided the politics of Sparta. It is not clear from Plutarch whether others (aside from the ephors) were allowed to sleep in the sanctuary of Pasiphaë, and Cicero mentions only that the ephors slept there. The oracle of Pasiphaë at Thalamai shows that incubation oracles used by members of the state for consultation on matters of the state were present in more parts of Greece than just Boeotia. Just as in Boeotia, the enquirers themselves consulted, but it was still only important people, the ephors, who were allowed to incubate.

45Ritualized oneiromancy in a sanctuary involving the enquirer him- or herself could have developed in the context of other oracular consultation techniques where consultation involved the inspiration of the notable enquirer (as in the oracle of Trophonios), or where, as in the cave of the nymphs at Sphragidion, ordinary villagers who spent time by the cave could receive oracles. There is no reason to think that this oracular method developed as a result of influence from foreign cults or cults in other parts of Greece — the belief in the transcendence of dreams and the need for oracles is found in most populations. It should be noted, however, that at the time of Herodotus the incubants were all specially selected people.

3.3.3. Asklepios and Apollon at Epidauros

46The cults examined so far have been exclusive oracle cults that were used for consulting on decisions relating to important matters (with the exception of the sanctuary at Oropos in Classical times, where healings later became its speciality, under influence of the cult of Asklepios). In the earliest testimonia on incubation in the cult of Asklepios, however, all members of society, even women and slaves, incubated. In this section, the origins of incubation in the cult of Asklepios will be discussed, because it was here that a ritual was constructed, for the first time, for a much wider worshipper-clientele than the select consultants such as those in the cult of Amphiaraos (as related by Herodotus). The origins of this cult and the new type of consultation ritual it offered are vital for understanding the phenomenon of incubation as practised by the masses for the sake of healing.

47In the earliest myths about Asklepios he is a Thessalian, but this does not tell us anything about his first cult. He was a Thessalian doctor in the Iliad, with a knowledge of surgery and healing herbs.100 His two sons, Podaleirios and Machaon, who were also doctors, took part in the war against Troy. In the Hesiodic Catalogue of women from c. 700 BC, he is mentioned for the first time as the son of Apollon and Koronis from Thessaly, and after this the connection with his father is always strong, in cult as in myth.101

48In Classical times there was a struggle between the largest Asklepieia: they made claims that Asklepios’ mother came from Thessaly, Epidauros, or Messene.102 Epidauros eventually won this struggle, after the oracle of Delphi identified it as the birthplace of Asklepios.103 According to Riethmüller, the worship of Asklepios could perhaps date as far back as the eighth century and may have originated in Thessaly, possibly Larissa, close to Trikka, and from there it then spread to Delphi and Corinth.104 This worship is attested on coins from Larissa, the earliest of which dates to the second half of the fifth century. The first evidence for cult to Asklepios at Epidauros, on the other hand, is an inscription from the late sixth century.105 Incubation in the cult of Asklepios is first mentioned in the iamata from Epidauros, which were collected from earlier inscriptions and then written down in the latter half of the fourth century.106 In antiquity, Thessaly was seen as the home of many heroes, and because it was famous for its healing herbs, it would have been an ideal birthplace for Asklepios.107 The fact that Asklepios came from Thessaly in the Iliad need not mean that his first cult came from Thessaly. Epidauros provides the earliest evidence for a cult to Asklepios and solid information on the cures provided there. The origins of incubation in the cult of Asklepios would hence best be investigated in the Epidaurian material.

49In order to explore whether the origins of incubation are to be found in some cultic predecessor prior to the cult of Asklepios, the cultic traditions at Epidauros before the cult of Asklepios must first to be investigated. On Mount Kynortion (Dog’s Climb), above the later sanctuary on the plain at Epidauros, a sanctuary already existed in the Bronze Age.108 The site was abandoned during the Dark Ages, as was the settlement most likely to have been used the sanctuary, but evidence of cult reappears from the mid-eighth century.109 According to Lambrinoudakis, the dedication of medical instruments on the altar itself in the sanctuary of Kynortion, as well as lead wheels, indicate that an early Apollon, god of fertility, welfare and health, was worshipped here.110 Apollon Maleatas is first attested at Epidauros in the Hymn of Isyllos from 325–300 BC.111 It is Pausanias who locates his sanctuary on Mount Kynortion.112 Maleatas is, however, present in a lex sacra from the beginning of the fourth century BC among gods who receive a pre-sacrifice of cakes in the cult of Asklepios at Piraeus (imported from Epidauros in 420), so his roots at Epidauros seem to go further back.113 It has been speculated in previous research that Apollon Maleatas was a healing deity who influenced the character of the cult of Apollon and Asklepios on the plain, but the evidence is inconclusive.114

50In Archaic times, a sanctuary was established on the plain. As far as we can tell from the epigraphical evidence, it was dedicated jointly to Apollon Pythios and Asklepios.115 The earliest find with a secure provenience is a bronze cymbal inscribed ‘From Mikylos to Asklepios’ from c. 500 BC from Building E, a building that has been identified as an early cult complex with an altar and dining areas.116 As early as the sixth century there was a mud-brick stoa underneath the later abaton at Epidauros, facing the sixth-century well that was later incorporated into the Classical stoa.117 Festivals with competitions at Epidauros can be identified from 530–520 BC, according to poems by Pindar.118 These events, which were in all probability performed in honour of Asklepios,119 show that there was an organized cult on the plain at this time. Thus, both Apollon Pythios and Asklepios at Epidauros had an early presence on the plain, indicating that the cults developed in parallel.120 In Corinth, the cults of Apollon and Asklepios were also closely connected from the beginning. Asklepios made a first appearance in the sanctuary of Apollon around the late fifth century BC, or possibly somewhat earlier,121 and Apollon continued to receive dedications there until the late fourth century.122 At Argos as well, Asklepios was also a co-helper of Apollon.123 The cult of Asklepios seems to have grown forth from the cult of Apollon in these places. The little evidence there is on the cult of Asklepios at Trikka confirms this theory, as no one was allowed to go down to the adyton of Asklepios there without first sacrificing to Apollon Maleatas.124

The earliest epigraphical evidence for incubation: the iamata of Epidauros

51If the earliest reference to incubation is from 479 BC, when the Greek was paid by Mys to sleep in Amphiaraos’ sanctuary (as indicated by Herodotus writing in the 450s or 440s BC), the iamata of Epidauros are the next testimonia to attest to incubation.

52The iamata were most likely collected from various smaller tablets from different times that were set around the sanctuary and then re-edited on four large slabs in the latter half of the fourth century BC, the stories dating back, according to the analysis of LiDonnici, perhaps 100 years.125 This would date the earliest iamata to 450–400 BC. This is not to say that the earliest incubation cures were of this date, only the earliest evidence. The cult was famous in 420/419 for its incubation cures, when it was imported to Athens,126 and so it can be assumed that incubation was an important part of the cult at least in the early fifth century, if not earlier.

53Of the iamata inscriptions from Epidauros, the earliest stele begins, ‘Healings by Apollon and Asklepios’.127 Apollon is thus a co-healer in the earliest evidence on incubation in the cult of Asklepios, even though it must be stressed that the name of Apollon is never mentioned in the actual iamata; he is only honoured in the title. As mentioned by Wickkiser, eight out of ten of the earliest iamata describe the healer as ‘the god’, making it uncertain whether Asklepios or Apollon is meant.128 The cures on stele B are described in terms of medical procedures, and so Wickkiser argues for a decline in the role of Apollon in the healings, Asklepios being the doctor par excellence in Homer and thus contributing the medical element of the cures.129 By the early fourth century BC, Asklepios’ name appears eight times in the narratives.130 Both Apollon and Asklepios healed through incubation in the iamata, and so it is not improbable that incubation in the cult of Asklepios developed from a common cultic base and was performed within the scope of a common cult.

54Apollon healed at many of his sanctuaries, including Delphi.131 The oracle of Delphi was also consulted for the sake of offspring, just as at Epidauros.132 Some enquirers at Delphi and Epidauros thus asked for much the same things. At Epidauros, however, Apollon and Asklepios were, as far as we can tell from the evidence, worshipped and consulted jointly, since the iamata are directed at both gods. Furthermore, those who actually consulted Apollon were much more restricted at Delphi, and the enquirer contacted Apollon through an intermediary (the Pythia at Delphi), whereas at Epidauros, the enquirers were able to contact the god directly. In the iamata of Epidauros, any individual, including women and slaves, could obtain direct contact with the god by sleeping in the sanctuary. The cult at Epidauros offered worshippers, regardless of their social status, the opportunity to consult a deity directly in a dream.

55Is there anything in the nature of the cures at Epidauros that might tell us of the origins of incubation in the cult of Apollon and Asklepios? Like Delphi, Epidauros offered a mix of more traditional oracular consultations as well as consultations that pertained to healings. Some cures at Epidauros were also effected just through being present in the sanctuary, not by incubating, or through the sacred dogs and snakes of the sanctuary. The iamata can be divided into stories where the help-seeker was cured instantly inside the abaton by the god (e.g. being operated on in his or her dream);133 healings which were performed inside the sanctuary, but not in a dream;134 one healing which was performed in a dream inside the sanctuary, but outside the abaton;135 and advice given by the god in a dream inside the abaton on how to get healed.136 Aside from healings, the advice given in a dream inside the abaton could also concern other matters, such as finding a lost child or a hidden treasure.137 Other types of wonders performed by the god inside the sanctuary are also recorded — actions of the god that had nothing to do with healings.138 In sum, however, the instant healings effected during sleep inside the dormitory form the largest group of all iamata, and they are present among the cures which have been stylistically dated as the oldest and middle group.139 Incubation (for anyone in society) was thus the main consultation method in the first testimonia on cures at Epidauros. The iamata do not, however, tell us how this healing method came into being.

Apollon’s mantic techniques and Asklepios’ healings

56Given that incubation where the worshippers probably met the son of Apollon, Asklepios, in their dreams developed within the scope of a joint cult to Apollon and Asklepios, we now turn to the question of which mantic traditions in the cults of Apollon might have given rise to incubation as it is described in the iamata of Epidauros. Apollon was the main healer and one of the best-known oracle givers in the ancient world, and different oracular techniques were used in his different sanctuaries. Oracle cults of Apollon were rare on mainland Greece, probably as a result of the enormous popularity of Delphi. Boeotia, with its many oracles, was an exception, and several cults of Apollon are attested here.140 In Ionia there were, on the other hand, many oracles of Apollon, including the oracle at Patara in Lykia, which was one of the mythical homes of Apollon.

57The only (possible) occurrence of incubation being used as an oracular technique in the cult of Apollon took place at Patara.141According to Herodotus, the priestess of Apollon at Patara spent the night in the temple, acting as a promantis.142 If the priestess stayed alone in the temple at night, it must be deduced that the god did not speak through the priestess, since nobody was there to listen. It seems likely that the priestess was given some impression — a vision, a dream, or some other sign from the god conveying his answer to the enquirer’s question. Since the priestess was to stay in the temple during the whole night, it is quite possible that the answer was given through a dream. The priestess was called promantis, like the Pythia at Delphi.143 Just as in relation to the Pythia, it has been hypothesized that the priestess (promantis) should be viewed as a consort of the god (and that this would be the reason why she spent the night in the temple), but this notion has been disproven in recent research.144

58At Delphi, a single priestess, the Pythia, sat in the adyton of the temple.145 After due preparations, the enquirer at Delphi was allowed to enter and was brought before the Pythia, who gave the answer concisely and clearly for the enquirer to note down; the enquirer was then to leave.146 No one slept in the adyton at Delphi, but it may be noted that the consultation location was called adyton. The dormitory at Epidauros was called adyton (‘not to be entered, innermost sanctuary’) in one iama, as well as the more common term abaton (‘not to be trodden’).147 In the Hymn of Isyllos, a hymn from Epidauros from the last quarter of the fourth century BC, the dormitory of the Asklepieion at Trikka is mentioned, and it is called adyton.148 In the later Asklepieion at Lebena, the dormitory was also called adyton,149 possibly in analogy with Epidauros, whence it was probably founded in the fourth century BC.150 At all other incubation sites, the dormitory was called enkoimeterion or koimeterion.151 A possibility would be that the dormitories at Epidauros, Trikka and Lebena kept the first name of the location where the epiphany was originally given — possibly in a joint cult of Apollon and Asklepios at Epidauros.

59Consultation in the cult of Apollon did not only take place with one single intermediary. There were also early cults of Apollon that offered divination by inspired prophecy, where many priests or otherwise specially chosen intermediaries could act as prophets. At Didyma, in the first cult before the destruction of the Persians in 494 BC, the gift of prophecy was believed to be a hereditary trait within the Branchidae family (the oracle was later revived after the reconquest of Alexander, and was remodelled after Delphi).152 An interesting mix of the prophetic oracles, the dream oracle at Patara, and the more easily accessed lot-oracles is found in the cult of Apollon Koropaios in Thessaly. The inscription attesting to the oracular technique of this cult is dated to 116 BC, but the cult itself dates back to Archaic times.153 A number of magistrates were summoned for each consultation event. They had to prepare through abstinences from sex and wine, and sit together wearing white clothes. The enquirers, who asked for health from the god (amongst other things), would write their questions and wishes on tablets, which the magistrates accepted one after the other. All tablets were then put in a vase, sealed and put in the temple over night. In all likelihood, the vase was then taken out in the morning, the seal shown to be untouched, and the answers given by the god during the night would then be revealed.

60Another cult of Apollon where the god was thought to bestow prophetic power, and even to give answers to requests in dreams, at least to his priests, was Apollo Ptoios in Boeotia. An inscription dated to 300–275 BC from this cult mentions a priest communicating favourably with the god in a dream, asking the god to provide him with prophetic power, in the same way that his ancestors obtained this gift.154

61The inspired prophecy used in cults of Apollon was reserved for priests who were considered to have oracular capacity, for hereditary or other reasons. The presence of magistrates who acted as intermediaries in the cult of Apollon Koropaios might indicate the existence of local cults of Apollon, where the assignment of being an intermediary might have been allotted to persons with ‘secular’ positions in society. The priestess at Patara is further proof of the use of oneiromancy in the cults of Apollon.

62The cults at Delphi, Patara and Korope were cults to Apollon only. He was the main healing deity in Greece, a deity who both sent diseases, as in the Iliad, and had the power to take them away. At Epidauros, as far as we can tell from the available evidence, the cult on the plain was established for Apollon and Asklepios jointly, and it specialized in healings through dreams in the earliest iamata. There is no record of Asklepios ever having caused ill health; on the contrary, his kindness seems to have been one of the basic traits of his nature.155 This, and the fact that Asklepios was considered a hero at first, might have meant that Asklepios was easier to approach and contact directly than Apollon.

Needs of the masses

63Lot-oracles at oracular sanctuaries were a less expensive alternative for those who could not afford or make the time for a more time- and money-consuming consultation process. The presence of lot-oracles shows that there many worshippers wanted to contact the god, above all those worshippers who were not financially powerful. A promised gift of knuckle-bones to Asklepios at Epidauros recalls lot-oracles that have been noted at Delphi and tentatively identified at Didyma, where they existed as an alternative to consultation which was probably easier to use and less expensive.156 Instigating an alternate method of divination, as at Delphi and possibly Didyma, was a way of helping those who did not have time to stay for appointments with the inspired oracles, the Pythia at Delphi, or the Branchidae at Didyma, or those who simply could not afford them.157 This indicates that many people wanted to enquire of Apollon, more than just those who gained the opportunity through the Pythia and waited their turn, paid the fee and then sacrificed an animal.

64Many oracular sanctuaries in the Greek world were flexible and offered a variation of oracular techniques that was dependent, one must presume, upon the needs and (financial) abilities of the worshippers and what techniques were popular at the time and/or regionally.158 The fact that lot-oracles existed shows that more worshippers wanted to consult than the traditional oracle cults of Apollon could accommodate.

65A cult where everyone who came to the sanctuary was invited to sleep in a holy place and meet a divinity in a dream would have required planning and, not least, the funds to build the proper structures for such a ritual. Epidauros was, as far as we can tell from the available evidence, never a sanctuary for consulting on matters of state or the proper course of action in a difficult situation; rather, as suggested by the popular lot-oracles at other oracular sites, it responded to a popular need for questions on healings and fertility. It may be hypothesized that the sanctuary on the plain at Epidauros, which was a modest sanctuary to Apollon and Asklepios in the beginning, had some sort of an inspirational consultation method. Considering other known consultation techniques in cults of Apollon, one or perhaps several intermediaries deemed worthy of contacting Apollon might hypothetically have slept in a temple of Apollon and Asklepios at Epidauros, just as was the case in the cult of Apollon of Patara.

66When consultation was made for the sake of health, families with their ill members would have needed to stay for at least the night, and accommodation was at first, as in so many other sanctuaries in ancient times, arranged in allotted spaces inside or outside the temenos. An allotment of sleeping space inside the temenos area is documented in leges sacrae from the cults of Apollon and Leto, and it is clear from these sacred laws that it was seen as a privilege to be allowed to sleep inside the temenos, and it is also clear that not everyone could do this.159 In the cult of Leto at Xanthos, those who wanted to sleep in the sanctuary had to perform a sacrifice, wear simple garments, and could not carry any weapons in to the sanctuary.160 Sleeping in the dwelling place of the deity might have been seen as a way of further strengthening the prayer directed to the god or goddess at the sacrifice.

67It is possible that the consultants who slept at Epidauros, even inside the temenos, inspired by whatever other consultation method existed at the sanctuary, spontaneously received dreams in which they perceived that Asklepios spoke to them. The officials of the sanctuary might have picked up on the miraculous accounts, and constructed a ritual for enquirers to consult on their own. This is hypothetical, and as is often the case in historical processes, there need not be only one factor that contributed to the formation of a new phenomenon.

68In conclusion, at Epidauros there was an Archaic sanctuary to Apollon and Asklepios, and its festivals in 530–520 BC prove that it was important. At some point, healings for larger numbers of people were provided through a form of inspirational prophecy where any worshipper was able to sleep in the sanctuary and meet Asklepios in his or her dream. This particular form of consultation may have developed within the framework of a joint cult of two deities prone to both inspirational prophecy and healing. In Archaic times, as seen in relation to the cult of Amphiaraos in Boeotia, not everyone was allowed or deemed able to approach the divinity in this way. The accounts of the iamata, on the contrary, show that anyone, even a woman or a slave, was allowed to seek contact with a hero or god through a dream in a sanctuary. It is quite possible that the son of Apollon, who was at first a hero, was more approachable for worshippers than Apollon himself. Also, a shift in mentality relating to who could have contact with a deity through a dream may have taken place in the course of the fifth century.

3.3.4. Accessibility of significant dreams

69Oracle cults where worshippers could consult in person via inspirational prophecy existed in Greece as early as Archaic times. I will now examine how it came to be that dreams, which in Archaic times were sent by gods to significant people, in Classical times came to be considered as god-sent to all members of society, even women and slaves. I will also investigate whether there was a connection between this shift in mentality and the popularity of mass incubations in the cult of Asklepios.

70According to some scholars, inspirational prophecy was a product of the Archaic period and was characterized by a desire for oneness with the gods. It has been argued that this new religious phenomenon can also be identified in the rituals of the Dionysiac mysteries.161 This belief is mostly based on Plato’s classification of inspirational prophecy as well as inspirational poetry as mania or divine possession.162 Two opposing views are that inspired prophecy had existed in Greece since prehistoric times, and that inspired prophecy was an important part of Minoan and Mycenaean religions.163 In Early Classical times, there were cults where worshippers consulted divinities in person, such as Amphiaraos and the nymphs in the Sphragidion cave. Usually, however, in the Greek world, such direct contact with the gods was believed to be for the particularly select, the descendants of heroic seers or priests who were specially chosen by an elective process. At the end of the fifth century, however, many people, including poor old women (as in the Ploutos) and even slaves could sleep in Asklepios’ sanctuary and meet the god.

71Even if we cannot tell exactly how the cult of Asklepios developed its form of incubation, a parallel development in the way that significant dreams were perceived during Archaic times and later on might cast some light on the phenomenon of incubation for the masses. In Homer, artists were known as followers of the art of Athena and Hephaistos, and doctors were seen as members of the family of Paeon.164 However, in the fifth century, doctors were called Asklepiads, sons of Asklepios, and artists were called Daedalidae, sons of Daidalos.165 It has long been argued that the shift of patron deities for artists and physicians alike points to a shift in mentality, where craftsmen asserted their own ability to make progress and acquire knowledge. A similar process may have occurred in the belief system, and so it came about that anyone could access the gods through dreams and actively seek a cure for his or her ailment. This would explain how incubation came to be such a widespread phenomenon in Greece, from having been an exclusive consultation technique in Archaic times.

72In Homer, Zeus, Athena and Hermes appeared in dreams to Agamemnon, Rhesos, Penelope, Nausicaa and Priamos.166 The dreams in Homer are literary constructs that are used to tell the story, and they cannot be taken as indicative of how Greeks of this time experienced dreams.167 There were also plenty of low-rank characters in Homer — slaves, servants, warriors — but the gods appeared in dreams only to kings and queens. True, these are the leading characters, and their dreams would most easily advance the story. Still, one might not help but wonder whether a king’s dream was considered to be more significant than that of a servant.

73Significant dreams do not appear in Hesiod or in any other piece of Archaic Greek literature.168 This leads both Brelich and Harris to the conclusion that the interpretation of dreams was less important to Greeks of that time than other forms of divination.169

74The use of dreams in divination can be found in Pindar, who describes Bellerophon sleeping on the altar of Athena, receiving a dream vision of the goddess that helps him to capture Pegasos.170 Bellerophon was a mythical hero, and so were the heroes of Homer who, although they did not sleep in sacred places, received visions of gods talking to them. The heroes of Homer never sought to make the significant dream come to them. In the passage from Pindar, Bellerophon was told by a seer to sleep on the altar of Athena. The dream was thus actively sought, but not as a direct initiative of the sleeper himself. Still, some tradition in the age of Pindar must lie behind his text. If there was a tradition that Bellerophon slept in the temple of Athena, traditions of other people sleeping in temples and contacting the gods who dwelled there through their dreams might also have existed. In order for the passage to make sense to the listeners of Pindar, the notion of sleeping in a temple and receiving contact with the deity through a dream must have existed in his time. Since a priest tells Bellerophon to do this, thus sanctioning his action, and since Bellerophon is a hero, a notable person, the phenomenon of sleeping in a temple and contacting the deity in a dream would surely, in Pindar’s time, not have been something that was possible for every visitor to a sanctuary. Furthermore, the significant dreams that Herodotus reports were all sent to kings and rulers such as Kroisos and Kyros.171 Later on, this came to change. Rulers, however, do not stop having significant dreams after the mass-production of such dreams in incubation sanctuaries; the two continue alongside each other.172

75There are many testimonia from Classical times which confirm that people believed the gods entered anyone’s dreams. As pointed out by van Straten, the critique that Diogenes the Cynic directed towards people obsessed with dreams shows that the belief in the importance of dreams must have been a widespread phenomenon.173 The significance of dreams is attested in Aristophanes and Menander, for instance.174 In the comedies of Aristophanes, dreams of private individuals were viewed as being sent by divinities or being without meaning, but the meaningful ones needed interpretation.175 It is in the process of interpretation that charlatans appeared in the comedies. At the time of Aristophanes, when incubation had just arrived in Athens with the cult of Asklepios and was well established in the Peloponnese, the city of Athens was full of dream interpreters helping with what people believed to be god-sent dreams.176 Evidence from the early fourth century suggests that Athenians believed in the significance of god-sent signs and acted in accordance with their belief by paying speakers of oracles (manteis, chresmologes).177

76Some Classical authors were critical of significant dreams sent to private individuals, but the overwhelming evidence is that people believed that their dreams had a meaning and were sent by gods.178 Prior to this, the testimonia mention only people like Bellerophon who received god-sent dreams. Plato, although he gives little credit to the supernatural character of dreams,179 records how dreams and visions were often the cause of dedications made, especially by women and by ill people.180 The kat’ onar inscriptions are a phenomenon that affirms the ubiquity of dreams believed to be significant (in this case, the gods’ orders in dreams to make a dedication), though this need not necessarily have been an incubation dream). The majority of kat’ onar inscriptions were dedicated to healing deities such as Asklepios and his family, i.e. in cults where incubation also played an important role.181 They date, at the earliest, from the fourth century BC, and the majority are from the Hellenistic period and from the cult of Asklepios, although other gods and goddesses are represented as well.182

77A belief among all worshippers, both high and low, that gods entered their dreams, in all likelihood contributed to about the introduction of incubation for the masses. This belief is first referenced in Classical literature. The abundance of dream interpreters in Greek cities strengthens this hypothesis, as do the numerous kat’ onar inscriptions.

3.3.5. Rituals for priests, consultants and enquirers

78In my view, the development of incubation from an exclusive consultation technique, e.g. in the Archaic cult of Amphiaraos, to mass-incubations may explain the construction of the ritual itself. Most rites before and after the incubation event itself did not differ very much from common forms of worship in Greek religion, although some preparations were more elaborate in comparison to what other worshippers did before entering the temenos. The special rites designed to take place just around the dormitory were naturally exclusive to the incubants. The exceptions to this are the cults of Podaleirios and Kalchas on Mount Drion and the Plutonium at Acharaka, which were local oracle cults with smaller numbers of consultants, for whom the priests could make time.

79Rites prescribed in oracle cults may be divided into rites for the enquirers and rites for the intermediaries with the god providing the oracle. When comparing incubation rites to rites that existed within different oracle cults in the Greek world, a similar division might also be made.

80If we firstly look at rites for enquirers, the payment of an eparche at Oropos before incubation, and before any healing had taken place, which was unique in healing cults, may be compared to similar consultation fees in oracle cults.183 At Delphi, a fee, a pelanos, was sometimes paid before consultation.184 In the Manteion of Apollon Pythios at Argos, a pelanos was thrown into the thesauros.185 At Lebadeia, in order to consult the oracle of Trophonios, a fee of ten cakes worth no less than ten drachmae was to be paid.186 The wearing of wreaths, attested for incubants in the Lex sacra Hallenstrasse, was a common prescription for consultants in oracle cults, as well as in the cult of Trophonios, and thus the sanctified status of the suppliant was marked.187

81Secondly, let us now look at rites that were constructed for consulting the god directly and were used either by those who were consulting on their own behalf or by those who were consulting as an intermediary. The stricter prescription of sexual abstinence for incubants at Pergamon may be compared to prescriptions which applied to priests (including oracle priests) in Greek cults in order that they would be extra pure when they took on the role of intermediary to the divinity.188 Abstinences from certain foods have also been noted. As seen in the Lex sacra Hallenstrasse, the incubants were to observe a special abstinence from goats three days prior to incubation.189 Comparing this rule on food for incubants with those for oracular consultants, we can note that some fish were not to be eaten before consulting the oracle of Trophonios.190 Priests in any cult about to contact the deity could be subject to prohibitions on certain foods.191At Pergamon, the goat was a prohibited animal in the cult, and most probably it was prohibited in order to augment the purity of the incubant.

82Purification with water, and the notable bath in seawater in the Ploutos, may be compared to a rite for priests at Delphi, where the servants of the temple cleansed themselves in the baths of Kastalia.192 There is a parallel in the cult of Trophonios, where, at least in Roman times, consultants took a cold bath.193 Amandry, writing on the baths at Delphi, as well as Bonnechere, writing on Trophonios, see the baths as both purificatory and inspirational before the coming consultation.194

83It may be conjectured that incubants at Piraeus wore special clothes (probably white) around the dormitory.195 Several testimonia point to Greek priests wearing special clothes or being forbidden to wear certain clothes (materials such as leather, or specific colours) when interacting with the divine.196 There is, however, nothing to indicate that the preferred colour of the priests was white, although it seems to have mattered to the gods what their officiants wore.197

84Many incubation rites, especially those of preparation (abstinence and purification) have been noted as being similar to preparations made by both consultants and priests in oracle cults. It seems logical that the stricter preparations for those about to meet the god would be rites that stayed on in the incubation ritual even when more and more people were allowed to consult the god directly. A factor that surely contributed to the development of incubation for the masses is the fact that in the Classical period many people believed that their dreams could be sent from gods — something which can be noted in literature and inscriptions from this time. When constructing a consultation ritual in a sanctuary where so many were allowed to incubate, the ritual itself by necessity had to be simplified. There was no space for the complicated ritual constructed for the single consultants of the oracle of Trophonios or the rites enacted by priests about to communicate with Apollon.

85A striking similarity between oracle cults and incubation cults can also be seen in terms of locality, as well as the fact that the term adyton was used for the consultation room both at Delphi and at Epidauros. Locality was as important in oracle consultation when using a priest as the intermediary as it was when the consultant him- or herself contacted the god. The god was to be contacted in his or her dwelling place. Otherwise there could be no contact with the divinity. This condition might also explain the effectiveness of incubating by proxy — a healthy and fit relative would make the journey on behalf of the sick person to the incubation sanctuaries and obtain an answer on what remedy to use.198 An important difference between oracle and incubation cults is that when an oracle priest, or an individual, private consultant (such as in the cult of Trophonios) sought contact with the god, he or she was alone in this space, whereas incubants slept all in the same locality. The togetherness is in itself an important factor which indicates a ritual change that came about when incubation came to be practised by many worshippers. Just as being alone might effectively take the single consultant out of his or her daily context and aid an otherworldly experience, the togetherness might help consultants augment each other’s hopes and expectations, letting experiences and dreams of others augment their imagination.

3.4. One incubation ritual or many?

86Given that incubation changed from being exclusive to being more inclusive, how did the ritual develop once the enquirers had become the consultants as well? In previous research, the analysis of incubation and the rites and rules it encompassed has been based on the assumption that a general structure for the incubation ritual, varying in detail, but present in its basic form at all incubation sanctuaries, existed.199 Such a presupposition means that certain rites, which existed at only one incubation sanctuary, have been taken as representative of the whole phenomenon.200 We will now investigate the nature and development of the ritual at the different cults which offered it.

Asklepieia

87It might be expected, given the fast spread of the cult in the fourth century, that ritual characteristics were transferred from the founder cult (most often Epidauros) to the new sanctuaries, and that a coherent ritual of incubation thus might be found in the cult of Asklepios.201 The largest and most popular Asklepieia were those at Epidauros, Kos and Pergamon.202 Epidauros is the sanctuary which provides the oldest evidence for a cult of Asklepios, as well as incubation in this cult.203 Athens’ Asklepieion was founded from Epidauros in 420/19 BC, and Piraeus’ was probably founded just before that. Pergamon’s Asklepieion was founded from Epidauros in the first half of the fourth century BC204, but it is not known which cult founded the Asklepieion of Kos, which appeared at a relatively late stage.205 Lebena’s Asklepieion in Crete was founded in the fourth century BC, either directly by Epidauros, or by Balagrai near Kyrene, which was in turn founded by Epidauros.206

88The cult of Asklepios at Epidauros grew forth from a pre-existing cult of Apollon, and it is evident in sacred laws as well as the Epidaurian iamata that Apollon continued to be honoured by sacrifice before incubation, and that he was thanked for the healings effected through incubation.207 The animal sacrificed to Asklepios could presumably be of any kind, but the goat was, at least in Roman times, forbidden.208 There is no direct evidence for the sacrifice of cakes to gods (other than Apollon) who were associated with Asklepios, but there were several altars with dedicatory inscriptions to, among others, the sons of Asklepios, Machaon and Podaleirios, and so cake sacrifices to associated gods at Epidauros are sure to have existed.209 Also, an inscription from Piraeus (which was founded from Epidauros) prescribes cake sacrifices to originally Epidaurian deities such as Apollon Maleatas, and the dogs and the kynegetai.210 Surely these deities also received preliminary offerings in Epidauros. At Epidaurian festivals the participants wore white clothes, and since we know that incubants at Piraeus wore some special clothing, one might wonder whether incubants at Epidauros did the same.211 For other Asklepieia it is difficult to know whether or not this was the case, since clothes for a festival to Asklepios at Euboia were to be multicoloured.212

89The incubants at Epidauros had a special dormitory from an early stage, and it was called both adyton and abaton.213 A typical Epidaurian habit was the writing of cure tablets which related the incubation dreams and healings, and reading these tablets augmented the expectations of those about to incubate as well as promoting the cult. Incubants at Epidauros also paid a sum prescribed by the sanctuary, iatra.214

90Some of these Epidaurian rites recur at other Asklepieia, but not all, and even where they are attested, as we shall see below, they were adapted and mixed with many other local rites.

91From Epidauros, the cult spread to Attica. In 420/19 BC, the cult of Asklepios arrived at Athens from the Zea harbour in Piraeus.215 In the Ploutos by Aristophanes, incubants slept in the Asklepieion of Piraeus in 388 BC. Here, one might have expected, given the small distance, that the rites would have been more or less copied from the Epidaurian cult. On the contrary, a ritual of its own was formed at Piraeus. First of all, the incubant was bathed and purified in the sea before entering the temenos.216 It could be argued that this was a practical measure, the sanctuary being in a state of construction and the sea being close,217 but more importantly, seawater was considered to have extra purifying effects.218 The Piraean purification was thus a stricter rite, as well as a rite with a local parallel: the Athenian mystai bathed in the sea three days before the procession to Eleusis.219 Piraeus was also the only cult of Asklepios in Classical and Hellenistic times where it is known for certain that incubants wore special clothing, following the text of Aristophanes. Prior to incubation, a complicated sacrifice of cakes to many different deities associated with Asklepios took place before the main sacrifice to Asklepios.220 Maleatas, Apollon, Hermes, Iaso, Panakeia, ‘the dogs and the huntsmen’, as well as Mnemosyne and Helios who were added later, were to receive cakes as a pre-sacrifice. The selection of gods was due to a combination of factors, some of which point to Epidauros, but others can only be understood in the local context.221 First of all, cakes were given to members of Asklepios’ family: his father Apollon, both as Apollon Maleatas and as Apollon without an epithet, and also Asklepios’ daughters Iaso, Akeso and Panakeia. Apollon Maleatas, as well as the dogs and the huntsmen, were honoured at Piraeus because of their Epidaurian affiliation. Apollon Maleatas was the first deity venerated at Epidauros. The hunters were either the men who, according to a local Epidaurian myth, found the abandoned infant Asklepios on the slopes of Mount Kynortion, or Machaon and Podaleirios, the sons of Asklepios.222 The presence of Hermes and Helios is likely due to their local popularity in Attica.223 It can be argued that Mnemosyne was universally helpful at incubations (helped to remember the dream), but she also received cult in the Agora, where she received the very same cake as in the Piraean lex sacra.224 Another important local peculiarity of the cult at Piraeus was the main sacrifice to Asklepios; in the Ploutus of Aristophanes, only cakes were sacrificed to Asklepios at Piraeus.225 None of the testimonia on the cult at Piraeus contain references to an animal sacrifice, something which is in many ways exceptional.

92The evidence from the Ploutos may further attest that incubants, though exceptionally and in the absence of other locales, might have slept in the temple itself.226 The cult at Piraeus, as well as at Athens, picked up the rite of incubating on a fleece from the older incubation sanctuary of Amphiaraos at Oropos.227 In the early fourth century, Oropos was under Thebes, but it came to be occupied by the Athenians in the fourth century, and by that time had been visited by many Athenians seeking oracles and healings. The Piraean and Athenian custom of incubating on a fleece can be noted on votive reliefs, and this rite is not known from any other cult of Asklepios. The custom of sleeping on fleeces was not obligatory, and there are reliefs from both these sites with incubants either sleeping on fleeces or lying on klinai without the fleeces.228 Fashions in cultic habits thus passed back and forth with the numerous Attic worshippers travelling to Oropos. Furthermore, the cult officials at Oropos came from Athens during the periods of Athenian occupation, and they must have been influenced by the rituals of the cult.229

93Thus, contrary to what might be expected, no coherent ritual of incubation was transferred from Epidauros to Piraeus and Athens. In particular, the inclusion of sleeping on a fleece shows how independent these Attic Asklepieia were from Epidauros when it came to setting out a ritual for incubation.

94Moving on to Asia Minor and the very important Asklepieion of Pergamon, we also find evidence for notable local variation. A few ritual similarities with its founder Epidauros can be noted, such as the avoidance of goat, but at Pergamon the prohibition was taken a step further.230 While at Epidauros goat was prohibited as an offering, at Pergamon worshippers who wanted to incubate were not even allowed to eat goat meat or cheese three days before incubation. Just as at Epidauros, the father of Asklepios, Apollon, received an animal sacrifice before the sacrifice to Asklepios.231 After this, a piglet was prescribed as the animal to offer Asklepios.232 This is the only Asklepieion where a certain animal is prescribed, singling this cult out, and emphasizes how the Pergamene priests were at liberty to form a local ritual. The sacrifice in the Lex sacra Hallenstrasse may have been prescribed for incubants as well as for all other worshippers in the sanctuary, and the piglet was an inexpensive victim that was very common in Greek religion. Thus, the animal was probably not chosen solely on the basis of how successful the incubation would be, but rather the choice was a rite designed by the officials of the Asklepieion in order to create a coherent ritual at Pergamon.233 The sacrifice of cakes before the sacrifice of a piglet, as well as the offering of cakes before entering the dormitory, also differed from what was prescribed at Epidauros and other Asklepieia. Before the sacrifice, Zeus Apotropaios and Zeus Meilichios received ribbed cakes with nine buttons, while the female deities, Artemis Prothyraia and Ge, received cakes with nine buttons only. The female variant of cake was also to be given to Tyche, Mnemosyne and Themis. Zeus Apotropaios was locally revered in Asia Minor,234 and Zeus Meilichios was recognized as a beneficent deity in the Greek world.235 Tyche and Zeus Meilichios received cult also at Epidauros,236 but other deities who were very important at Epidauros, such as Apollon Maleatas, Machaon and Podaleirios, were completely left out at Pergamon. Finally, the word used for dormitory at Pergamon was enkoimeterion.237 All in all, even though the cult at Pergamon was founded from Epidauros, the ritual was created locally, as can be seen at Piraeus and Athens.

95Epidauros also had close ties to Erythrai,238 where a cult to Asklepios from around 380–360 BC can be identified.239 Only information on the sacrificial practices is available to us. It seems that Apollon, as at Epidauros, was strong at Erythrai, since he received the first sacrifice at any sacrificial occasion and was also thanked after incubation with a singing of the Paian. This might have been a more common custom than our (limited) sources suggest.240 A choice of different animals was given for any sacrifice offered at the sanctuary: an ox, a full-grown animal which was probably a sheep, or a piglet, but it seems that it was also permissible just to sacrifice cakes — a custom we have only seen at Piraeus.241

96Some exceptional ritual data has been preserved from the Asklepieion at Lebena. This was the only sanctuary of Asklepios where incubants were summoned at home to come to the sanctuary and incubate.242 The architecture of the Asklepieion at Lebena differs from all others because of the presence of a Tanzplatz, documented in an inscription from the first century BC.243 It was the custom in some cults of Egyptian gods for a worshipper to be called to come to a sanctuary in a dream, as a sort of preliminary consultation, and some Egyptian cults existed in Crete in Hellenistic times, so it may be that foreign religious practices influenced the cult of Asklepios in Crete.244 The singularities of Cretan incubation can also be seen in the choice of terminology. At Lebena, in an inscription from the second century BC, the word ἐφεύδω designated the sleeping.245 The verb ἕυδω, from which ἐφεύδω derives, is an Archaic Greek word used in Homer and the tragedies, which persisted in Crete.246 Crete is known to have had an Archaic form of Greek, which is also reflected in other words used to describe incubation. Cure tablets, for instance, were called pinakes in Epidauros, whereas at Lebena they were denoted sanides.247 It can thus be seen how incubation in Crete, both in relation to the terminology used and the rites themselves, was adapted to local Cretan customs.

97Moving lastly to sanctuaries without strong ties to Epidauros, Thessaly had an ancient claim to being the birthplace of Asklepios.248 Trikka in Thessaly was reputed in antiquity to be a very old sanctuary, but so far the sanctuary has not been found, and the origin of this cult is not known. Further north, in Macedonia in the Asklepieia of Beroia and Amphipolis, there is very little information available on the cult. All that can be said about the Asklepieion at Beroia is that the word used in inscriptions for dormitory was enkoimeterion instead of the Epidaurian term adyton used at Trikka.249 At Amphipolis, we can note that a pre-sacrifice was made ‘to all the gods’ of the sanctuary. Given that the first Asklepieia grew from cults of Apollon, it is probable that the father of Asklepios was included among these gods. All animals were allowed to become sacrificial victims at Amphipolis.250

98Comparing the rites and rules surrounding incubation at all these different Asklepieia, it seems that there is a basic structure that consisted of preparing, giving some kind of offering, sleeping in a holy place, and then offering some sort of thanks at most sanctuaries. Apart from the sleeping on holy ground, this is much too common a ritual structure to say that there was a common structure for incubation rituals which varied in detail but kept its basic outline. When considering the detail of the rites and rules, some habits are recurrent in the different cults of Asklepios, and those are primarily the inclusion of Asklepios’ relatives, above all his father Apollon, in the offerings, as well as the payment of the iatra. However, the variation in ritual detail is far greater. Not even within the cults of Asklepios founded by Epidauros was there a coherent incubation ritual. For instance, at Piraeus, the ritual might have been expected to resemble the Epidaurian one, but we have noted sizeable differences between the two. The incubants bathed in seawater, they offered cakes to local Attic deities, they sacrificed nothing but cakes, and possibly they slept in the temple, and even on fleeces. In Athens, the famous votive reliefs single this cult out, no doubt hinting at the presence of skilled craftsmen in the city working on public buildings, while Corinth seems to have specialized in anatomical terracottas.251

99If we look at the sacrificial habits before incubation, at one Asklepieion no particular animal was specified, but at another, oxen, sheep, piglets or just cakes were allowed, at yet another, only piglets were to be sacrificed, and at still other Asklepieia, goats were prohibited. When cakes were prescribed, they could be of many kinds, either plain ones or ribbed with nine buttons for male deities. Moreover, there was never one architectural design that was deemed most suitable for incubation.252 The words used for dormitory underline these variations. In the Asklepieia of Epidauros and Lebena, founded from Epidauros, the words adyton or abaton are used.253 Similarly, when Isyllos of Epidauros described the cult of Asklepios at Trikka, he used the word adyton. At the other Asklepieia, however, as well as the cult of Amphiaraos, the dormitory was designated enkoimeterion or koimeterion, words that denoted a special place for sleeping.254

100Even if the cult of Asklepios was well-known and established in the Greek world, there does not seem to have been a general rule that was valid for all Asklepieia in relation to how incubations should be performed. Just as the ritual for the healthy, or that for the ill who were making a sacrifice and a vow, varied at different Asklepieia according to the size and organizational capacities of the sanctuary as well as the wealth of the worshipper and local cultic traditions, the ritual for incubants also differed. One would have expected Asklepieia to conform to a more set model of incubation practices, given the fast spread of the cult, but this was not the case. The holy snakes signalled the arrival of the god, but the rituals of incubation were formed according to local cultic preferences. Flexibility and local adaptations seem to have been vital when establishing incubation rituals at new Asklepieia, and it was never believed that Asklepios could be contacted through one ritual only.

Other incubation cults

101Turning away from the Asklepieia, the largest non-Asklepian sanctuaries that offered incubation in the Greek world of Classical and Hellenistic times were the Amphiareion at Oropos and the cult of Hemithea at Kastabos in southern Karia. Other smaller incubation cults were the cult of Brizo on Delos, the Plutonium at Acharaka in the Meander valley, and the cults of Kalchas and Podaleirios on Mount Drion in modern-day Apulia.

102Beginning with the cult of Amphiaraos at Oropos, its rites and rules may indicate that the cult originated as an Archaic oracle cult. The cult of Amphiaraos is first attested in Herodotus, and lay at a yet undiscovered site in Boeotia close to Thebes, but nothing is known of the rites other than that the oracles were obtained through consultants sleeping in the sanctuary.255 At the time of Herodotus, only Greeks, though not Thebans, were allowed to incubate. Later, the sanctuary of Amphiaraos at Oropos was founded in 420–414 BC on the border between Attica and Boeotia.256 It is thus likely that there was an old site of worship somewhere in Boeotia, which came to be taken over by the new site at Oropos. At this new sanctuary, Amphiaraos began to deliver healings on a larger scale, probably influenced by the popularity and spread of the cult of Asklepios.257 The paying of a consultation fee, eparche, before incubation may have been an old rule that remained from the oracle cult.258 There was a great deal of choice as to which animals could be offered before incubation. One Oropian lex sacra reads that the animal could be of any kind,259 and another inscription suggests that some animals were suitable, among them chicken and cattle.260 A special building inside the sanctuary, called koimeterion, was assigned to the incubants. In the Amphiareion at Oropos, it was postulated that men and women had to sleep apart inside the dormitory, which is the only time such a rule appears.261 In the Amphiareia at Oropos and nearby Rhamnous, incubants could optionally sleep in a fleece, as depicted in reliefs where the incubants are either sleeping on fleeces or just sleeping on klinai.262 The incubation rites of paying an eparche and sleeping on fells may have originated in the first oracle cult, and they were then adapted to the wishes of worshippers and practical considerations related to the larger numbers of worshippers coming in hope of healing.263 Such adjustments could have been, for instance, allotting special areas in the dormitory for men and women respectively, giving a wide choice of sacrificial animals, and specifying that not everyone had to sleep on fells.

103Turning to the sanctuary of Hemithea at Kastabos, an account in Diodorus and the archaeological remains on the site indicate that it was large and well-visited.264 Hemithea, who was probably a Hellenized indigenous goddess, helped the sick and especially women in childbirth, and did so through dreams. The local character of her cult is evidenced by the prohibition on sacrificing pigs (and even eating or touching them before entering the sanctuary was prohibited), as well as by the libations of melikrata (honey and milk) that were given instead of wine.265 The reason for these prohibitions was probably some local tradition, and they were explained in their time by an aition myth relating the story of how Hemithea, named Molpadia as a mortal, first became a heroine and later a goddess.266 The myth relates how Molpadia was set to guard her father’s wine, but that she fell asleep and a pig broke the wine jar, whereupon she hurled herself off a cliff to escape the anger of her father.

104Turning to the smaller incubation cults, the first of these, Brizo on Delos, was a cult where dream oracles were given (though the exact technique is unknown). The evidence for this cult is dubious;267 what is known is that ‘all good things except fish’ were permitted as offerings.268 Fish may have been prohibited from being sacrificed because Brizo protected the seamen of Delos, and fish in myth ate dead seamen.269

105In the cults of Podaleirios and Kalchas on Mount Drion in Daunia, incubation was practised, but not by large amounts of worshippers and only for the sake of obtaining oracles. In the earliest testimonium on the oracular rites on Mount Drion, Timaeus of Tauromenion (c. 350–250 BC), quoted in the Alexandra, incubants slept on sheepskins on the tomb of Podaleirios, which lay close to the empty tomb of Kalchas.270 There is nothing in this first testimonium on the nature of the sacrifice to Podaleirios, and any sheep might have been offered. It seems logical, though, that the fells used for sleeping came from animals sacrificed to the hero, the communication with the hero being strengthened by the consultant sleeping in the fleece of the animal that had been sacrificed to him. Later, according to Strabo, those who consulted the oracle of Kalchas sacrificed a black ram and slept in its fleece (the verb enagizein is used).271 Sacrificial practices, as well as the hero cult offering incubation — Podaleirios or Kalchas — differ in these two testimonia which were written about 250 years apart, but the rite of sleeping in a fell when incubating on Mount Drion remained the same. Contrary to what has been stated in previous research, the enagizein sacrifice of a black ram on Mount Drion might well have been a Roman modification, since neither the cult nor the sacrifice is attested earlier.272

106The rite of sleeping on a fleece when incubating existed both in central Greece and in southern Italy, and so the question arises as to whether the rite in some way spread from the cult of Amphiaraos in Boeotia to Daunia or whether the custom in Italy was indigenous. Farnell was of the opinion that the cult of Podaleirios, as well as incubation, was brought to Mount Drion either by settlers from Kos, or by a later Achaean migration.273 This part of Italy was populated by Daunians, Peucetians and Iapygians, who all spoke Messapian but were part of a koine heavily influenced by Greece and Illyria.274 Many cities in Italy had Greek foundation legends and incorporated parts of Greek culture. It is not unlikely that colonizing Greeks brought the cult of Homeric heroes with them. Many legends tell of Homeric heroes travelling around the Mediterranean just like the colonizers. The rite of sleeping in a fell may have been incorporated into the cult by these Greek colonizers at some point. The rite would have made sense to Greek practitioners because the practice was already known from the cult of Amphiaraos.275

107Another possibility is that the rite developed in Italy of its own accord. Vergil and Ovid describe King Numa in Latium, who supposedly slept on a fleece in the incubation oracle of Faunus.276 The stories of Vergil and Ovid are possibly literary constructs, but they still show that the rite of sleeping on a fleece in order to receive oracles was known in Italy at the time of Ovid, and possibly also in an Archaic cult to Faunus.277 The use of fleeces at Mount Drion may have been driven by the fact that it was the final point of a transhumance trail for sheep and cattle — herds were led from the Abruzzi and Molise mountains down into the lowland of Apulia.278 Sheep would consequently have been in good supply on the site and this would have made for an easy enactment of the rite.

108Either of these two scenarios could be true; in any case, we know that sleeping on fleeces in Apulia developed independently of the habit in Boeotia and Attica because the cult of Amphiaraos was not a Panhellenic cult around 350 BC, when the rite is first attested on Mount Drion. The cults of Podaleirios and Kalchas were cults where only those who sought oracles came to incubate, and the worshippers performed unusual rites compared to the common forms of sacrifice and worship in Greek cults. The priests also made time and space for attending to these unusual rites, as seen especially in the account of Strabo when a black ram was to be sacrificed but not eaten (i.e. consumed in some other way). Local factors such as the presence of a hero’s tomb on the mountain top and sheep being in ample supply at this site may have helped to form the rites of these cults.

109In another cult of strong local character, the Plutonium at Acharaka in the Meander valley close to Nysa, the incubants slept in a cave that was believed to be the entrance to the underworld.279 They lay beside priests who were a sort of oracle priests and the ones who incubated most often.280 The incubants could have stayed for days in the cave, fasting while there, and the priests would then have helped them interpret their dreams. There were many Plutonia in the Meander valley at the time, and the rites at Acharaka may well have been influenced by the rituals at these shrines.281 Just as at Mount Drion, the special character of incubation at Acharaka called for the priests’ attendance and effort, and those who were allowed to incubate in the cave were specially chosen by the priests. According to Strabo, the priests were the primary incubants, and hence only a certain number were selected to accompany them into the cave.

110To sum up, variations attested in these incubation cults may be explained by their different origins as well as by adaptations to local cultic habits and cult clientele. It is striking that the smaller incubation cults on Mount Drion and at Acharaka had a more special ritual and a higher degree of attendance by the priests, whereas the cult of Amphiaraos in the early fourth century, because the cult had turned towards healings through incubation and opened itself up also to worshippers from Attica, made allowances for any sacrificial animal and constructed sacred laws in such a way so as to allow the rites to be performed even when the main priest was absent.282

Conclusion

111The rituals, and the rites and rules they consisted of, varied from sanctuary to sanctuary, even within the cult of Asklepios. Not even Epidauros exported a set ritual to its affiliated cults, such as Piraeus or Pergamon. If there was any rule, it was variation and adaptation. The interesting rite of sleeping on fleeces indicates how an incubation rite may have originated independently in local customs, at different places (Amphiaraos in Boeotia and Podaleirios in Daunia), and that an incubation rite could also spread to incubation cults of other divinities in the nearby area (from the cult of Amphiaraos in Boeotia to that of Asklepios in Attica).

112The ritual variations between various incubation sanctuaries are paralleled by the terminology used. Just as there was no coherent ritual of incubation at all sanctuaries where it was practised, neither was there a coherent terminology.283

3.5. Roman times

113The beliefs that dreams were sent by gods and that these dreams could have significant meanings persisted in Roman times, as seen in Artemidorus’ Oneirocritica.284 A brief study of incubation cults in Roman times reveals that most of the sanctuaries of this period existed in Classical and/or Hellenistic times.285

114Beginning in Greece, an interesting development in the cult of Amphiaraos at Oropos occurred in Roman times. On the rites surrounding incubation, Pausanias writes that the incubants continued to sleep on the fells of the ram that they had previously sacrificed.286 Before this, a purificatory sacrifice was performed to Amphiaraos and to several associated deities. After successful incubation, thanks were offered by throwing a silver coin into the sacred spring of Amphiaraos. Furthermore, Philostratus attests to a new rite: the incubant had to fast for one day and abstain from wine for three days before incubating at Oropos ‘so that he can absorb the messages with his soul in a state of illumination’.287 It appears from the passage of Philostratus that worshippers mostly asked the oracle questions which did not relate to cures for diseases. Tertullian attests to the renown of the cult at Oropos, listing it among famous dream oracles, and he also writes that fasting was common in dream oracles.288 Fasting was generally seen as a cathartic rite,289 and in Oropos in Roman times there seems to have been a change in incubation rituals as they were in Classical and Hellenistic times into rituals that promoted an extraordinary experience, a ‘state of illumination’ in the words of Philostratus.

115Amphiaraos was not the only dream oracle in Greece. Not far from Oropos, in Phokis, a cult of Dionysos offered cures through dreams to the Amphikleians.290 Here, a priest served as a promantis, receiving the oracles and passing them on to the enquirers. It is interesting that Dionysos, while also interacting with the human world in an epiphanic way through his mysteries, was a curer of disease here. Unfortunately, the cult is only attested in Pausanias, and so one cannot speculate further.

116Turning to the oracles of Asia Minor, in the cult of Amphilochus at Mallus, in Cilicia, Dio Cassius writes of dream oracles given by the hero, while Lucian writes of handing the questions on tablets to the priests.291 Techniques seem to have varied, but oracles via dreams were apparently among those techniques. At the oracle of Mopsos in Cilicia, these two elements were combined. The governor of Cilicia sent a servant to sleep in the sanctuary with a question for the oracle noted down on a tablet.292 The servant then dreamt of the correct answer. It is interesting to note that dream oracles here were combined with the use of questions written down on tablets, a little bit like the above discussed cult of Apollon Koropaios. In the process of communicating with the gods, different techniques were used to reach this goal, and only a few cults had a ‘set’ communication technique over a long period of time.

117To continue the survey of Asia Minor, Pausanias possibly attests to the continued popularity of Acharaka, and similar cults in the Meander valley, when he writes that the gods of the lower world in the cities above Meander sent dream-visions to all those who wished to enter their shrines.293 If the worshippers were invited to come and consult in their dreams, it is likely that oracular techniques within the sanctuaries also involved dreams.

118As for Roman incubation cults, unknown deities at Tor Tignosa294 and the cult of Faunus Albunea295 have been mentioned in previous literature, but the evidence for their existence has been questioned.296 As early as Hellenistic times, the cults of Isis and Serapis had begun to expand into the Greco-Roman world,297 and according to the textual sources, they offered, among other things, incubation cures.298 The absence of sacred laws and votive tablets, however, make it difficult to say much on the rituals surrounding these dreams.

119The cults of Asklepios continued to thrive in Roman times. On a general note, the ubiquity of cults of Asklepios and the belief in the cures they offered are attested in many Roman texts and also medical literature.299

120To start with the most important of all Asklepieia, Epidauros, there seems to have been a decline in its prosperity in the first century AD, since Livy writes that the site was full of broken votives.300 At the time of Pausanias, however, the sanctuary was a major site (this is attested also by new buildings).301 On the rituals of Roman times, we know that the sacrifice of goats was prohibited, as was taking the meat outside of the temenos.302 Reading the cure inscriptions still seems to have been an important part of the experience.303 Cultically, though, the prescription for purity of mind, as attested in a passage of Porphyry, represents a major new trend in the cult of Asklepios.304 A similar religious monition comes from the Asklepieion of Lambaesis, Roman North Africa: ‘Bonus intra, melior exi’ (‘Enter good, leave better’).305 At the Asklepieion at Aegae, the god turned away a man who was sacrificing on a lavish scale, but was not being an honest person.306 If we compare this exhortation to evaluate oneself in Roman times with the righteousness of worshippers — something which is said to be desirable in the iamata of Epidauros — it seems that Asklepios desired good deeds also in Classical times. In the world of the iamata, however, it seems that this behaviour was much easier to control, since the worshippers came from smaller communities where they, and their actions, were known to everyone. The iamata list names of those who had been cured and often their home city. For instance, in the case of Pandaros and Echedoros of Thessaly, who were two slaves,307 Echedoros was given money from Pandaros to dedicate to Asklepios at Epidauros. Echedoros kept the money from himself and was punished by the god for this. In Roman times, it seems that Asklepios was no longer content with being sent the appropriate offerings, and a new way of identifying and singling out the deserving worshippers had appeared. Maybe previously it was easier to identify who had come to the sanctuary, and who had made the promised dedication? In the new, larger Roman world, control of who was deserving of help was possibly internalized.

121From Epidauros, from Roman times, we know of dedications to Hypnos, Sleep, and possibly also to the god of Dreams, Oneiros, who was a Roman addition to the circle of Asklepios.308 In a similar manner, there is a dedicatory inscription from Lebena, where Diodorus, having been cured from blindness, dedicates two statues to Oneiros and Soter (the Saviour, i.e. Asklepios).309 These gods also had statues at the Asklepieion of Sikyon, indicating that Oneiros had become popular in Roman times.310

122The cult of Asklepios was introduced to Rome from Epidauros, allegedly after an outbreak of plague in 293.311 Four inscriptions from this cult, from Imperial times, present the character of this cult.312 They are written in Greek, and their composition alludes to the iamata of Epidauros.313 Ritually, they are unique in the context of incubation because they describe a public thanking of the god (not just communicating the miracle to others by putting up a votive telling of the cure).

123Kos seems to have been popular, but not much literature on the cult and rites survive.314 For the cult at Pergamon, however, there is ample documentation. Many Roman writers attest to it having been visited by people from all over the Roman world, including the Emperor Caracalla.315 For the rites, the Lex Sacra Hallenstrasse, which was copied in Roman times, continues to be of prime importance for the study of the cult. A fragmentary inscription from the Imperial period provides us with evidence on the different purificatory regulations inside the temenos.316 The fragmentary inscription also suggests that at least two dormitories existed, and that incubants wore white clothes but no girdles. Inscriptions from Roman times tell of groups of worshippers of semi-official status, the perithutai and therapeutai, and these worshippers seem to have stayed at the sanctuary for long times and they also took part in regular cultic activities.317 The status of these semi-official groups at the sanctuary is also confirmed by the Hieroi Logoi of Aelius Aristides. The information on cult offered by Aelius Aristides is very interesting, but we are provided with an image of regular cult practices at a big sanctuary rather than the ritual preparations and thank offerings of the singular incubant. Aristides was doubtless a part of a cultic group such as the therapeutai, and he seems to have received visionary dreams while sleeping in many places in the sanctuary.318 He describes purificatory ceremonies where everyone was clad in white,319 as well as the singing of hymns by a boys’ choir, or worshippers singing together in larger groups and among friends.320 He makes a distinction between praying and praying ‘privately’.321 Aristides mentions numerous sacrifices, but these seem to have been a common occurrence at the sanctuary, and they did not happen before every incubation.322 Apparently, ceremonies of holy lamps (presumably after sunset, when the lamps were carried into the temple) also took place.323 Dedications in gratitude to the god were also important.324 On a personal level, though, Aristides provides ample evidence of how he as an individual felt when going through the process of incubating. He compares the experience to being initiated into a mystery cult, but with the important difference that he experienced a genuine joy in the cult of Asklepios.325 Comparing this to what we know of the cult of Amphiaraos at Oropos in Roman times, there seems to be a new focus in the rituals for communication with the divine in dreams: the cults strove for the dreams themselves to be exceptional experiences, experiences that would remain with the worshippers after they had obtained that for which they had come to the sanctuary.

124In late antiquity, the importance of Asklepieia slowly dwindled. At Athens, there is an interesting testimonia on how Proclus, the Athenian philosopher of the fifth century AD, prayed in the temple of Asklepios for the health of a friend’s daughter.326 He prayed ‘in the more ancient fashion’, and the girl was healed. Sacrifice was banned in 392, and so it seems that the fashion of praying also changed in late antiquity.

125The cult of Asklepios persisted for a relatively long time into the Christian era, maybe because the kind of help offered by Asklepios was one that corresponded to an ubiquitous human need — the healing of disease — and as such the cults persisted.327 In the Christian tradition, the numerous miraculous healings offered by the cults of the martyrs soon came to fill that gap. Some cults of Early Christian saints also came to specialize in healings by incubation.328

126To sum up some general trends for Roman times, the local variations of incubation rituals continued and some new trends can be noted, trends that coincide with wider changes in Mediterranean culture. Greek religion in Classical times was a local affair, one which involved the local community, and gods were specific to a certain place. Amphiaraos at Thebes and then Oropos were Boeotian cults. The Epidaurian Asklepios and the Asklepios at Athens both had strong local identities: the Epidaurians claimed that their city was Asklepios’ birthplace, and the identities of the cults at Athens and the cult at Lebena were linked to their foundation by the locals; thus, some customs particular to the area were integrated in each Asklepion. The contest for establishing the birthplace of Asklepios reflects this concern with locality. Nonetheless, the cult of Asklepios very soon came to grow to Panhellenic status.

127In Roman times, as a result of new communicative and networking opportunities under Roman rule, Panhellenic cults increased in numbers and spread over a wider area. Above all, cults were of a large size, they administered to larger cult areas, and they had larger groups of priests (including worshippers of semi-official status, like the group which Aristides was part of at Pergamon).

128A new interest in general ‘Saviour’ gods influenced the cult of Asklepios; Asklepios is no longer the medically specialized son of a diviner — now, in Roman times, he attains the same type of appeal that initiation into the cult of Eleusis granted: the hope for a more generally improved existence. To attain this, it was, in Roman times, important not only to bring what you had promised the god and to be certain of his miraculous powers — demands were also put on the worshipper to be a moral person, to be ‘pure of mind’.