A Second Temple Non-Temple Liturgy
Joseph Tabory
Bar Ilan University

A: Introduction

The biblical picture of sacrificial ritual presents the priests as the servants of God who perform this ritual. The Levites have no status in the ritual but they do have a function: they serve basically as caretakers of the Tabernacle/Temple. The rest of Israel had neither status nor function within the sacrificial service. However, rabbinic sources present us with a temple liturgy in which all three sections of the Jewish people participate. The priests offer the sacrifices, the Levites sing during part of this ritual (as instituted by David according to the later books of the bible), and representatives of the rest of Israel stood at the gate of the Temple, the Nicanor Gate, and observed what the priests (and the Levites) were doing. This observation is considered as their participation in the ritual and is known as ma'amad,1 presumably due to the standing position taken by the representatives. This participation of the rest of Israel in the sacrificial ritual is apparently referred to in the War Scroll which describes "the chiefs of tribes and father of the congregation...taking their stand at the gates of the sanctuary...these shall take their stand at the burnt offerings and sacrifices" (1QM 2:3-5).2The ma'amad of Israel has been considered as a rabbinical institution meant to make people feel that all the people of Israel are participants in the sacrificial ritual and it is not the prerogative solely of the priests. It is, perhaps, signficant that the term ma'amad is also used to describe the status and position of the Levites in their relationship to the priests in Chronicles I (23:28) "ki ma'amadam leyad b'nei Aharon" which JPS translates as "For their appointment was alongside the Aaronites" (cf. Chronicles 2, 35:15; "And the singers the sons of Asaph were in their place [bema'amadam], according to the commandment of David").3 The term ma'amad is also used to describe the status of the levites in the War Scroll (1QM 2:3) where it has been translated as "station" or "office".4 It is significant that the reference to the participation of the laypeople of Israel in the sacrificial ritual is missing in an earlier copy of the War Scroll. Davies has argued that this shows that this reference is a later addition to the text of the war scroll.5 It may be that this is due to the fact that this institution was not originally part of the Temple service and it was added at a later time.

The main rabbinic source for the ma'amad is the mishnah:

(M. Ta'anit 4) On three occasions of the year, on fast-days, on ma'amadoth, and on the day of atonement do the priests lift up their hands to bless [the people] four times during the day, namely at the shaharith [service], at musaf [=noon],6 at minhah and at the closing of the gates [ne'ilah].

The following are [the details concerning] the ma'amadoth. Because it is said, command the children of Israel [and say unto them]: "my food which is presented unto me". Now how can a man's offering be brought [on the altar] and he is not present? [therefore] the earlier prophets instituted twenty-four mishmaroth, and each mishmar was represented [at the temple] in Jerusalem by its own ma'amad of priests, levites and israelites. When the time came for the mishmar to go up [to Jerusalem] the priests and levites went up to Jerusalem and the Israelites of that mishmar assembled in their cities and read [from the law] the story of creation. The men of the [Israelite] ma'amad fasted on four days of that week, from Monday to Thursday; they did not fast on Friday out of respect for the Sabbath nor on Sunday in order not to change over [without a break] from the rest and delight [of the sabbath] to weariness and fasting and so [perhaps] die.

On Sunday [they read], in the beginning, and, let there be a firmament; on Monday, let there be a firmament, and, let the waters be gathered together; on Tuesday, let the waters be gathered together, and, let there be lights; on Wednesday, let there be lights, and, let the waters swarm; on Thursday, let the waters swarm, and, let the earth bring forth; on Friday, let the earth bring forth, and, and the heavens [and the earth] were finished.

This text presents an institution which is connected to the Temple sacrifice but, at the same time, has an independent liturgy which consists of reading of the Torah, the priestly blessing, and perhaps other elements which are not mentioned. An analysis of the rabbinic theory of this institution and its origin shows it to contain a number of inconsistencies. I have discussed this at length in a Hebrew paper7 and here I shall summarize and refine the conclusions of that paper. Rabbinic sources refer to the ma'amad an institution which consisted of two parts: a group of Israelites who went to Jerusalem attended the sacrificial ritual in the Temple and a group of Israelites who gathered in various cities outside of Jerusalem to read the Torah at specific times of the day. For purposes of distinction between them, we shall refer to the first group as the sacrificial ma'amad and to the second group as the liturgical ma'amad. There are a number of anomalies connected with these two groups and it is my contention is that these anomalies may be best explained if we assume that the liturgical ma'amad was actually an earlier institution than the sacrificial ma'amad8 and that it was originally created as a substitute for sacrifice. At some later time, as part of the continuing struggle of the sages to stress the power of the people as opposed to the elitistic approaches, the sages decided that the representatives of the people should have an official task in the sacrificial ritual. They coopted some of the people who attended the ritual and sent them to Jerusalem to be observers during the sacrificial ritual. Although these observers tried to conduct the liturgical ritual in their free time, the need for them to present at the sacrificial ritual forced them, at times, to forego the reading of the Torah. Presumably, thos who conducted the ritual liturgy outside of Jerusalem were able to conduct their liturgy on a regular basis.

It is not possible to date these events. I would suggest that the liturgical ma'amad was originally founded as a substitute for the sacrificial ritual during the time that the Temple was polluted by the Syrians and proper sacrifices were not offered there. It is worthy of notice that there is no knowledge of how the Jews did during these years. We do know that that the destruction of the Second Temple was a severe crisis for Temple-centered Judaism. The challenge of this crisis brought about the revolutions of R. Johanan b. Zakai and it may have been the stimulus for R. Gamliel to institute the amidah. The pollution of the Temple in the time of the Syrians should have presented a similar challenge. It is more or less accepted that the group in Qumran developed their liturgy as a substitute for sacrifice, although sacrifices were being offered in the Temple, because they considered the ritual conducted in the Jerusalem Temple by their opponents as being an unworthy ritual.9 It is reasonable to assume that the Hasidim who rebelled against the Syrian occupation of the Temple might have used similar means to create a ritual which would substitute, temporarily, for the lack of proper Temple sacrifice. However, possibly due to the nature of "temporary" institutions, this institution was not cancelled when the sacrificial ritual had been restored. It was decided that the insitution would continue in its original sites, the cities of the ma'amad, but some of the participants would go to Jerusalem and be present at the offering of the sacrifices. This achieved two goals. On the one hand, this changed the nature of the ma'amad while continuing its outer form, which is very often the way that religious reforms are accomplished. Since some of the members of the ma'mad were in the Jerusalem Temple, the members of the group who stayed at home felt that they were connected to the Temple. The ma'amad was no longer considered a substitute for sacrifice but rather as an auxiliary to sacrifice. On the other hand, it also changed, to a lesser extent, the nature of sacrifice. Sacrifice was no longer the fief of the priests; non-Israelites were actually to be considered as those who were bringing the sacrifices and the priests were merely serving as their agents.10 This fit in well with the pharisaic approach that tried to play down the importance of the priestly class and emphasize the importance and status of the non-priests.11 The people of the ma'amad who were co-opted to Jerusalem for this purpose tried to fulfill the liturgical aspects of the ma'mad while they were attending the Temple sacrifices, but they could not always fulfill all of them due to their need to be present at the sacrifices.

B: The Sacrificial Ma'amad

In order to clarify the content and significance of the liturgical ma'amad, it is first necessary to understand exactly what was the sacrificial ma'amad. The sacrificial ritual, according to the rabbinic report in the mishnah of Tamid,12 was divided into two sections. First, the priests prepared everything for offering the meat on the altar. This included slaughtering the animals, sprinkling the blood on the altar, cutting the meat into pieces, and carrying the pieces towards the altar in a ritual procession. At this stage, they did not actually place the meat on the altar but they lined it up on the ramp which led up to the altar. The next step should have been to place the cuts of meat on the altar to be consumed by the fire. However, at this point they made a break in the ritual and the priests went to the Chamber of Hewn Stone where they recited the sh'ma and accompanying blessings.

After this prayer break, they returned to the Temple courtyard to resume the main ritual of the sacrifice. Some sort of a signal13 was made which announced that the public part of the ceremony was about to begin. Upon hearing the signal, the priests, the levites and the Israelites assembled for their part in the ceremony. The priests could not yet take their place on the steps of the sanctuary for no one was allowed there while the incense was being offered in the sanctuary. At this point, the Israelites of the ma'amad assembled at the Nicanor gates to participate in the ceremony by observing it.14 After a priest had offered the incense on the golden altar in the sanctuary, the priests assembled on the steps of the sanctuary from where they would recite the priestly blessing and the levites assembled on a podium from where they would sing the Temple song. The priests would then bless the assembled gathering, after which the meat would be ceremoniously taken from the ramp and placed on the fire on the altar. At the end of this process, the wine would be poured on its appointed place on the altar and this was a sign for the levites to begin their song, which was the closing act of the ceremony.

It is significant that the more ceremonial aspect of the sacrificial ritual is offering the meat. This is in contrast to rabbinical theology and halakha, which stresses the blood part of the ritual as being of primary importance. I am not going to discuss the biblical theology of sacrifice15 but I think that it is clear that, according to the rabbinical portrayal, the emphasis of the sacrificial ritual in the Second Temple is on the offering of the meat as the important part of the sacrifice.16 This is not to deny the importance of a third element of the sacrificial ritual, the incense offering. Indeed, the description of the incense offering in the mishnah (Tamid 5:2-6:3) is much more detailed than the description of the offering of the meat (op. cit. 7:3). The moment of the incense offering, at least as far as the evening is concerned, was considered the optimal moment for private prayer and there is evidence that crowds gathered in the Temple to pray at this auspicious moment.17 However, the incense offering was not done in public. Rabbinic law was very strict in forbidding anyone to be present at this offering besides the priest who was actually burning the incense. Indeed, no one was allowed to be present even in the open area adjacent to the sanctuary in which the incense was offered (Mishnah Kelim 1:9). The crowds gathered at the Temple knew that the incense was being offered, and it was a solemn moment for them, but they couldn't see anything. It was only after the incense was offered that the public ceremony began with the priests blessing the people. This was immediately followed by the ceremonial transferring of the meat to the fire on the altar, from the ramp upon which it had been set earlier. Although this was an honor divided among a number of priests, the mishnah discusses how this was done if the High Priest wished to do it himself (Tamid 7:3). The mishnah also discusses what the proper procedure was in the event that the High Priest wished to worship God by prostrating himself in the Temple (op. cit. 7:1) but it does not mention what should be done if he wished to offer the incense. It is possible that this is not mentioned because there was no need to do so, but I suspect that the High Priest wished to utilize his prerogatives only at the offering of the meat as this was the most public part of the ceremony. The importance of the meat is significant for pinpointing what was considered the time of the sacrifice, as we shall discuss further on..

C: The Liturgical Ma'amad

1. The Liturgy

I will now turn to the liturgical ma'amad and I would like to focus on three points about this institution as it is portrayed in rabbinic sources. The first point deals with what was done at the ma'amad; the second deals with the time of the ma'amad and the third is the aspect of community connected with the ma'amad.

The mishnah cited above portrays the liturgy of the ma'amad as consisting of reading of the Torah. However, the context of this statement shows clearly that the priestly blessing was also part of the liturgy - although it gives us no idea of where this blessing was included in the liturgy. The command of the Torah that the priests bless the people of Israel (Numbers 6:22-27) is somewhat ambiguous. It does not order this blessing to be used as part of a sacrificial ritual and it does not tell us when this blessing is to be used. It is true that the Torah reports that Aaron blessed the people after offering sacrifices (Lev. 9:22), a sort of kiss of peace, but this blessing seems to be an ad hoc blessing which is not presented as being a fulfillment of the biblical command for the priests to bless Israel. It is hard to find any direct connection between Aaron's blessing and the blessing prescribed in Numbers18 and it is thus difficult to determine whether the priestly blessing was really meant to be part of the sacrificial ritual or whether it was to be used in some other context. If we assume that it was originally thought to be part of the sacrificial ritual, its may have been incorporated in the ma'amad in order to reflect the sacrificial liturgy. Any attempt to replace the sacrificial ritual with a non-Temple liturgy would have done well to use the priestly blessing as part of the liturgy. This is the only priestly liturgical function which was not prohibited by the Torah from being used outside of the Temple. As this blessing was normally recited in the Temple, using it in non-Temple liturgy would stress the importance of this liturgy as a sacrificial surrogate. However, the place of this blessing within the liturgy may have been different than its place within the sacrificial context.

In Second Temple times we find the priestly blessing used both in the context of the daily sacrifice, as noted above, and in two instances of non-sacrificial context: the ma'amadot and the reading of the sh'ma conducted outside the area of sacrifice (in the Chamber of the Hewn Stones; Mishnah Tamid 5:1). The Mishnah mentions that there were actually two ways of reciting the priestly blessing. In the Temple it was recited as "one blessing" while outside the Temple it was recited as "three blessings" (Tamid 7:2). According to rabbinic tradition, the reason for this distinction is that outside the Temple it was customary to respond "amen" at the end or each verse, which effectively broke up the blessing into three units. In the Temple however, the "amen" response was interdicted,19 and this meant that the three verses were recited as a single unit. In the non-sacrificial ritual in which the priestly blessing is mentioned, the reading of the sh'ma conducted outside the area of sacrifice (in the Chamber of the Hewn Stones), the priestly blessing ends the ritual (Mishnah Tamid 5:1). However, I would hesitantly suggest that the order to the priests to begin the reading of sh'ma with "one blessing" is referring to the priestly blessing which is to begin the ritual.20 This liturgy therefore, both opened and closed with the priestly blessing. As far as the ma'amadot are concerned, there is no indication at which point in the ritual the priestly blessing was recited. The only clear point to be made is that in the one example cited by the mishnah of the use of the priestly blessing in non-sacrificial context, the blessing appears in the context of a series of blessings and at the end of the series.21 This is, of course, also its place in the amidah. We might assume that this would be true also for the ma'amadot but there is no evidence for this whatsoever.

In the sacrificial context, the priestly blessing was recited at the beginning of the public ritual, just after the incense had been placed on the golden altar in a ritual which the public could not see. It was then that the priests blessed the people and this blessing served as the opening of the ritual rather than as a sign of its end. It was thus a greeting rather than farewell.22

In the non-sacrificial liturgy we find the priestly blessing serving as the conclusion of the ritual - in the same way that the Torah reports that Aaron blessed the people. The most well known example of this is to be found in the amidah. The priestly blessing is included in the last blessing of the daily amidah and this last blessing is also known as the blessing of peace. The sages considered the closing phrase of the priestly blessing, the blessing for peace, as the main theme of the priestly blessing. The last blessing of the amidah has been described by Fleisher as a receptacle for the priestly blessing. This fits in well with Bilha Nizan's discussion of the use of the priestly blessing and prayers for peace in Qumran.23

The main part of the liturgy of the ma'amad is the reading of the Torah. This is the main part of the other liturgies which are known from the Second Temple period. The other examples are the reading of the Torah by the king during the hakhel ceremony and the liturgy of the High Priest on the Day of Atonement The sh'ma is also basically a recital of biblical passages. All three of these readings are accompanied by blessings recited after the reading, blessings which have been considered by Heineman as prototypes of the Amidah. There is no mention of any blessings in connection with the reading of the Torah by the people of the ma'amad. Is this just an example of the silence of the sources or does this silence represent reality? If the latter is true, it is very likely that the liturgy of the ma'amad, reading of the Torah, was instituted before it was thought proper to accompany the reading of the Torah with blessings. The blessing of the priests may have been thought of as the closing ceremony of the reading. Of course, all this stands in stark contrast to the Qumranic practice which never instituted any public reading of the Torah. The writings found in Qumran do stress the importance of Torah study but not as ritual or liturgy.

Let us now turn to the passages selected for the reading of the ma'amad. We do not know when the continuous reading of the Torah became an established practice. We know that in other circumstances that the Torah was read publicly, they picked passages which were relevant to the day. The earliest depicted reading, that of Ezra in the beginning of Tishri, was taken from passages of the Torah which talked about the festivals of that month. The reading of the High Priest on the Day of Atonement was the passage from Leviticus which contained the prescription for what he had just done (M. Yoma 7:1). It is even more remarkable that he alst read the other biblical passage which commanded the sacrificial order of the Day of Atonement, Numbers 28, although he read this by heart to avoid the delay in rolling the scroll to this passage (loc. cit.).24 Thus, if reading the Torah was to substitute for sacrifices, it would seem natural to pick the passage which deal with the daily sacrifice (Numbers 28). This was indeed done on festivals but not for the Shabbat reading.25 The choice of the creation passages is somewhat remarkable. It could be explained in two ways. One possibility is that they rejected the idea of reading the same passage every day as they sought some measure of variety.26 I would suggest another reason. One of the earliest Second Temple sages, Simon the Pious, had declared that the sacrificial ritual, together with Torah and gemilut hesed (generally translated as "loving kindness")27 were the tripod on which the world stood (Mishnah Avot 1:2). If sacrifices could not be offered properly, creation might be in danger. Perhaps reading about the creation could protect the creation in place of the sacrifices.28 In this way, they were short-circuiting the Temple sacrifices or bypassing them, going, as if, directly to the source. This would be, perhaps, a more suitable replacement of sacrifice in the case of a polluted Temple than reading the passages about the Temple.29

Although the theory behind the reading of the Torah was based on the idea that on each day they read about the creation of that day, the practice was to read the passage about that day and the one about the next day. Thus, on Sunday they read about Sunday and Monday; on Monday they read about Monday and Tuesday...on Friday they read about Friday and Sabbath. On Sabbath there was no reading and no ma'amad at all. The reason for the expanded reading was to make the reading of sufficient length so that it could be divided between three people: a priest, a levite and an Israelite. It is reasonable to assume that the reason for requiring three people to participate in the reading was to enable all three classes of Israel to share in the reading.30 Thus, this reading was a forerunner of the idea that non-priests had an equal share in the liturgy.

No other prayer is mentioned in the context of the ma'amad. Fleisher has already remarked on this in developing his theory about the absence of prayer in the Second Temple period. It is true that later rabbinic sources state that there was a supplicatory aspect of the ma'amad. The talmudim state: "On Monday [they fasted] for those that go down to the sea; on Tuesday for those who travel in the deserts; on Wednesday that croup may not attack children; on Thursday for pregnant women and nursing mothers, that pregnant women should not suffer a miscarriage, and that nursing mothers may be able to nurse their infants" (BT Ta'anit 27b//JT Ta'anit 21b). However, Fleischer has pointed out that the supplicatory aspect is expressed by fasting, not by prayer. We might add that no supplicatory aspect is reputed to Fridays and Sundays - days on which they did not fast.31 However, this does not necessarily mean that there were no blessings at all accompanying the reading of the Torah. There may have been blessings which did not have a supplicatory aspect. Other Torah readings found in the Second Temple period, the reading of the Day of Atonement ritual by the High Priest on the Day of Atonement and the reading of selections from Deuteronomy by the king during the hakhel ceremony, were followed by blessings. The reading of the sh'ma was also accompanied by blessings. It is possible that the ma'amadot were instituted at a time when blessings were not yet considered a necessary adjunct of Torah reading. However, the earliest portrayal of a public reading of the Torah, by Ezra at the beginning of the Second Temple period, tells us that Ezra began the rite with a blessing of God (Ezra 8:6). We have already noted that we only know that the priestly blessing was part of the ritual of the ma'amadot through the context of the discussion. It is thus possible that there were blessings associated with the ritual which have not been mentioned in the reports. It is instructive to compare the ma'madot with the prayers for fast days as described in the Mishnah (Ta'anit 2). Here we find a detailed report about the series of blessings said on fast days but, as detailed as this description is, we know that it is not complete. For example, it contains no indication of the priestly blessing. Now, we know from the mishnah which discusses the priestly blessing (Ta'anit 4:2) that this blessing was part of the rite of fast days - although we have no idea where it was included in the ritual. The report of the rite of fast days (M. Ta'anit 2) also has no reference to the reading of the Torah. However, the Mishnah (Megillah ) prescribes what text is to be read from the Torah on fast days. It is possible that this mishnah reflects a later custom but I think it is more reasonable to assume that there was a reading of the Torah on fast days from the earliest times. Perhaps even the report of the reading of the Torah by Ezra reflects the tradition that Torah is read on fast days. If the fast day ritual included both Torah reading and blessings, besides the priestly blessing, it is tempting to assume that the ma'amadot ritual did the same. Since the existence of these blessings is speculative, we have no way of surmising what there content was. One might suggest that this was the provenance of some of those prayers found in Qumran which are non-sectarian.32

2. The Time of the Ma'amad

The second point I would like to discuss is the time of the ma'amadot. The ma'amadot are portrayed as taking place four times a day: morning, noon, minhah and ne'ilah. These times are supposed to coincide with the sacrifices. Before we can determine if they really coincide with the time of sacrifice, we must determine when the sacrifices were actually offered. The sacrifice is called tamid which is translated, in the King James version, as "continual [burnt offering]". It is clear that "continual" does not mean "constantly" (continuing without cessation). Bringing the sacrifice twice a day, regularly, was considered tamid.33 The rabbis did have another concept of tamid. It was apparently thought that the candles of the temple should be perpetually lit, without cessation, even in the day time. Less noticed is the fact that the rabbis thought that the altar should never be without some sacrificial meat burning on it. For this purpose, they ordained special sacrifices which were offered only when there was nothing else on the altar. But the tamid was only twice a day. The Torah prescribed morning and evening (bein ha'arbayim) for the time of the perpetual sacrifice. It may have meant "sunrise and sunset" or even "sunrise and dusk". Rabbinic sources do not give a clear picture of exactly what time the morning sacrifice was offered. The slaughter of the animal could take place shortly before sunrise (M. Yoma) but, as there was a break during the rite, it is not clear when the meat was actually brought to the altar. The evidence for the afternoon sacrifice is clear. The mishnah reports that it was usually slaughtered at 8:30 (three and a half hours before sunset; roughly equivalent to 2:30 p.m.) and the meat was brought to the altar and hour later, at 9:30 (roughly equivalent to 3:30 p.m.).34 At the risk of being a little too anthropomorphic, I will state that the times of the tamid sacrifices according to the rabbinic picture coincide with mealtimes: a morning meal and dinner in the late afternoon.35 We may thus understand something about the ritual of the sacrifice. Rabbinic literature gives greatest importance to the sprinkling of the blood of the animal on the altar, considering this the moment of atonement and atonement can be achieved only if everything revolved about the blood is done properly. The offering of the meat was secondary and did not affect the atonement. But when we examine the temple ritual we find that the actual bringing of the meat to the altar was the central feature. We might thus expect the ma'amadot, a ritual replacing sacrifice, to take place at the hours of sacrifice. The actual picture is much more complex and to clarify this issue it will be necessary to look at the more general issue of times of prayer during the day. There are two distinct traditions, one which refers to prayer three times a day and one which refers to prayer twice a day. We shall begin with the three times a day tradition.

The statement of the Psalmist "Evening, morning and noon I complain and moan and He will hear my voice" (55:17) has often been evinced as evidence for prayer three times a day. Marvin E. Tate writes "The suppliant prays three times a day (Cf. Dan. 6:10; Acts 10:9, 30) - at set times?".36 John J. Collins is a bit more skeptic, agreeing that "the psalmist says that he utters his complaint to God "Evening, morning and noon" but adds that "it is not clear that a fixed, mandatory observance is presupposed".37 I would argue that this passage does not really mean that the psalmist prayed three time a day. This expression should rather be understood as a merism signifying that he complains and moans constantly. It would certainly seem that this is true for the psalmist who says "Seven times a day I praise" (Ps. 119:164).38

There are, however, two points of interest in this statement for our subject. One is that his merism is not "night and day" but rather "Evening, morning and noon", which has some significance for his understanding of the structure of the day. The second point of interest is that the Psalmist begins his day in the evening rather than in the morning. Beginning the day in the evening is, of course, the traditional way within Rabbinic Judaism but it was not the way that the day was divided as far as Temple worship is concerned - as was recognized by later Rabbis.39 It is noteworthy that the Slavonic Enoch orders his children "In the morning and at noon and in the evening of the day it is good to go to the Lord's temple to glorify the Author of all things".40 He accepts the tri-partite division but he begins it with the morning rather than the evening.

Of particular significance in the history of prayer, and especially the times of prayer, is a passage found in the Book of Daniel (6:10-17). Prayer is presented here as a regular custom of Daniel, so regular that his opponents assume that he would be willing to give up his life rather than give up his prayer. Norman W. Porteous states "Probably we should infer that by the time of the Book of Daniel was written the later Jewish practice had already established itself".41

The report that Daniel directed his prayers towards Jerusalem has a clear antecedent in the prayer of Solomon (Kings I 8:35, 38, 44, 48) that prayers be directed through the Temple. Later Rabbinic rulings declared that orientation was a matter of distance. A person living some distance from Jerusalem could not pinpoint the location of the Temple and it was deemed sufficient if he would direct his prayers towards Jerusalem. It is not clear why Daniel's prayers were directed towards Jerusalem rather than towards the Temple. It is possible that it was physically and technically difficult to pray towards the Temple and so he prayed to Jerusalem - in the spirit of later Rabbinic rulings. It is also possible, and to my mind somewhat more likely, that he prayed to Jerusalem for there was no Temple.

We do not know at what times of the day Daniel offered his prayers. It would seem natural to assume that prayer, if it was to be offered twice a day, would be at daybreak and at sunset, the changes of the day. If we added a third time of the day for prayer, the natural time would be at noon. This is exemplified by the statement in the Jerusalem Talmud:

What is the source for three prayers a day? R. Samuel bar Nahmani said: according to the three times a day that people sense the change of times. In the morning one should say "I am grateful to You, my Lord and the Lord of my fathers, that you have brought me forth from darkness to light". At minha one should say "I am grateful to You, my Lord and the Lord of my fathers, that just as I was privileged to see the sun in the east, so I was privileged to see it in the west". In the evening one should say "May it be Thy will, my Lord and the Lord of my fathers, that just as you brought me forth from darkness to light, so may You bring me from darkness to light".

Evidence of this division as a Jewish practice may be found in the words of Epiphanius (Salamis, 315-403) who reports that Jews gather in their synagogue to curse the Christians three times a day, morning, noon and evening.42 For these times of day as an expression of human understanding of the division of the day we may turn to Julian the apostate who stated "We ought also to pray often to the gods, both in private and public, if possible three times a day, but if not so often, certainly at dawn and in the evening".43 The statement of Julian opens another way to understand the significance of three times a day prayer. Dawn and evening are breaks in the day; three times a day symbolizes the constancy of "throughout the day". A statement of Origen in regard to thrice daily prayer is particularly instructive in this sense. He states that one should spend one's life in unbroken prayer and the only way to do this is "if the entire life of the saint is a single unbroken prayer and if part of the prayer is prayer in the stricter sense".44

There is another division of the day into three times of prayer. Acts of the Apostles mentions as hours of prayer,45 the sixth hour (Acts 10:9) and the ninth hour (Acts 3:1-2). The third hour also seems to be of significance for it was at this hour, on Pentecost, that the followers of Jesus spoke in tongues (Acts 2:14). The custom of praying three times a day is mentioned also in Didascalia Apostolorum and is continued by Tertullian who specifies the third, sixth and ninth hour as the three times of the day for prayer. The third and ninth hour are not natural hours of prayer. However, the ninth hour was the hour in which the sacrifices were offered in the Temple. To be more precise, nine o' clock was the time that the incense would be offered in the evening as part of the eveining sacrifice.46 It would thus seem that this division is an attempt to reconcile the idea of prayers as adjuncts to sacrifices with the three-times-a-day tradition of prayer.47

Let us now turn to the twice a day tradition. Twice a day, in the context of praise and prayer to the creator, generally means sunrise and sunset.48 This is, of course, the sh'ma tradition which declares that the sh'ma is to be recited when one rises and before one goes to bed. Although this is based on a commandment in the Torah, this understanding of the passage is not obvious. The Torah states that one should "talk of them when thou sittest in thine house, and when thou walkest by the way, and when thou liest down, and when thou risest up" (Deuteronomy 6:7). It would seem just as reasonable to understand this verse as calling for a third recital of the sh'ma sometime during the day, "when thou walkest by the way", or, perhaps, even a fourth time - "when thou sittest in thine house". It would seem that the times for sh'ma are rooted in a twice a day tradition rather than in actual biblical exegesis. The sages utilised the biblical terminology for this tradition, getting up and lying down, to give a somewhat more liberal definition of morning and evening. The morning sh'ma could be said as long as people were still getting up and the evening sh'ma could be said throughout the night, as long as there were people who had not yet gone to bed. This was necessary to enable ordinary people to participate in the sh'ma. Exceptionally pious people, such as those known as vatikin, could be expected to pray at actual daybreak; ordinary people needed some leeway.

Most of the sources which deal with times of prayer in the Second Temple period follow the twice a day pattern. The testimonies of Philo and Josephus,49 all testify to prayers at daybreak and at sunset. The testimonies about the times of prayer in Qumran are not conclusive. On the one hand, the evidence of Serekh ha-yahad and megillat ha-hodayot may imply prayer three times a day. On the other hand, the War Scroll and other Qumranic evidence states clearly that prayer takes place in the morning and the evening. Collins summarizes the "the norm of three times of daily prayer did not yet prevail at Qumran".50 What is clear is that prayer took place at sunrise and sunset and not at the hours of sacrifice.51

We may now turn to the times of the ma'amadot. The mishnah reports that the ma'amadot took place four times a day: morning, noon, minha and ne'ilat she'arim ["closing of gates", commonly shortened to ne'ila, "closing"] - times which do not fit into either of the above mentioned patterns. Morning and noon may be recognized as part of the thrice a day pattern. The "closing" is the subject of rabbinic disagreement whether the reference is to the closing of the gates of heaven (sunset and the beginning of night) or to the closing of the gates of the Temple which was done somewhat earlier. The suggestion that it was related to the closing of the Temple gates is somewhat remarkable as there is no evidence that this closing had any particular significance in the Temple ritual. In our most detailed description of the sacrificial ritual, the description of the Day of Atonement ceremony, which ends with conducting the High Priest to his home at the end of the ritual, the closing of the gates is not even mentioned.52 I would suggest that the original meaning of this term was the closing of the gates of heaven and it thus completed the three times a day pattern which marked the stages of the sun: sunrise, the zenith at noon, and sunset. It is worthy of notice that the daily ritual found in Qumran (4Q503) uses the term "gates of light" although it refers to the light of the moon. The identification of this closing prayer with the closing of the gates of the Temple may have been a later attempt to create a correlation with the Temple worship.

This identification emphasizes the anomaly of the ma'amad which is called minhah and presumably took place at the time of the evening sacrifice. I would suggest that this ma'amad was a later addition to the original institution of three times a day ma'amadot. It may have been added as part of the attempt to connect the ma'amad with the Temple ritual. If this is true, we may now make an attempt to understand the theology of the ma'amadot. They were intended as a replacement for sacrifices. However, they were not meant to be a literal substitute for sacrifice. They were not meant to replace the animal sacrifices with listening to the word of God at the time of sacrifice. They were rather meant to provide a different method of perpetual worship of God. The Torah considered the twice daily sacrifice as being a "continual offering" or "perpetual offering" (Numbers 28:3).53 The founders of the ma'amad chose the three times a day pattern, as expressed in the book of Psalms, to be their way of offering perpetual worship. By reading the story of God's creation constantly (three times a day), they were supporting God's creation.

3. The Ma'amad as Community

Finally, I would like to refer to the issue of community. We usually think of communal prayer as being a gathering of the community. This is expressed particularly well when a community gathers in fast and prayer to avert some calamity. Here we find that if one of the members of the community does not participate, it is not a true communal prayer.54 However, the communal service in the Temple was not done by the community but by its representatives. A small number of people performed a liturgy for the entire community. The sages later argued whether the priests themselves were the servants of God or were they to be considered the agents of the people of Israel in service of God. There is certainly an aspect of Jewish prayer which is representative. The idea is that a gathering of people pray and their prayer is representative of the community. This has been developed in the theology of R. Solovetitchik but it has clear antecedents in rabbinic sources. The idea of the ma'amadot centers on this. There is no indication of the number of people required for the ma'amad ritual but there is no doubt that it was not the prayer of an individual. The gathering of the ma'amad was considered a microcosmos of the Jewish people. This explains why there was no liturgical ma'amad on Sabbath. It is well known that Jews were accustomed to gather together on Sabbath in their communities all over the world for the study of Torah.55 There was thus no need for a representative gathering. It was only during weekdays that such a gathering was necessary,56 until R. Gamliel instituted the idea that every individual should also offer prayer as a substitute for the sacrifices which could no longer be offered.57