Somewhere on Jerusalem’s majestic Temple Mount—the largest man-made platform in the ancient world, the size of 24 football fields, nearly 145 acres—Herod the Great (37–4 B.C.) built a new Temple to the Israelite God Yahweh,a doubtless on the very spot where the exiles returning from Babylonia more than 500 years earlier had rebuilt the original Temple, first erected in the tenth century B.C. by King Solomon. But where was that spot?
Efforts to locate it have sought clues from archaeological evidence on the Mount itself and from two famous descriptions of Herod’s Temple—one by the first-century A.D. Jewish historian Josephus and the other in a tractate of the Mishnahb called Middot. All these efforts to locate the site of the Temple have foundered, however, on the seemingly contradictory descriptions of Josephus and Middot and on the paucity of archaeological clues.
But all have focused on elements of the Temple, rather than on the Temple Mount. I would like to try a different approach. I would like to reverse the process—first to locate the Temple Mount in various periods and only then—and at present, only tentatively—to place the Temple on the Temple Mount during each period.
It is well known that Herod the Great approximately doubled the size of the Temple Mount by extending the earlier Temple Mount on the north, south and west.1 He could not extend it on the east because the land drops off steeply to the Kidron Valley beyond the wall on that side.
But where was the earlier Temple Mount, the one repaired by Nehemiah to create a level area on which to rebuild the Temple to the Lord? I believe we can now—for the first time—locate with considerable confidence the Temple Mount of the First Temple period.
The demonstration will rely on clues still visible on the site, on the text of Josephus and the tractate Middot, and especially on the work of the greatest underground explorer of Jerusalem of all times—Captain (later Sir) Charles Warren.
A brilliant engineer for the London-based Palestine Exploration Fund, Warren conducted investigations in Jerusalem between 1867–1870. His work followed the comprehensive survey and mapping of Jerusalem by Captain (later Colonel, then Sir) Charles Wilson, an engineer with the British Army who subsequently collaborated with Warren to bring out the Recovery of Jerusalem in 1871. In it we have a faithful record of the mysterious caverns, caves, tunnels and cisterns that lie underground, beneath the Temple Mount.2
Although others, such as the British Captain Claude Conder and the German architect Conrad Schick, also involved themselves with the exploration of the cisterns beneath the Temple Mount, the work of Warren, carried out while he was still in his twenties, stands as a landmark of systematic investigation and painstaking accuracy. Together with his faithful assistant, Sergeant Henry Birtles, and some local workmen, he surveyed 36 of the 37 underground structures, not to mention the extensive series of shafts and tunnels he dug outside, along the retaining walls of the Temple Mount, to assist his explorations.
One well-known scholar of Jerusalem’s archaeology has described Warren’s unique achievement this way:
“The great album of ‘Plans, Elevations, Sections, etc.’ registering his results, is an accomplishment … unequalled in the annals of Palestinian archaeology in its grandiose scale and sumptuous execution. Moreover, Warren’s explanatory records, especially his Recovery of Jerusalem, shows that every measurement and every drawing is the result of unheard of exertions and great courage.”3
A quotation from Warren’s own record of his exploration of cistern No. 17 will help explain why, as noted in the preface to the Recovery of Jerusalem, he returned to England in ill-health, suffering from fever and exhaustion:
“On sounding I found it 42 feet down to water. I tried to descend, but to no purpose, until I had nearly stripped to the skin and even then in my contortions I managed to slip the rope over one arm … On getting down to the water I found it only 3 feet deep and concluding from the size of the cistern that help would be required in measuring, I signalled for Sergeant Birtles to come down … In the meantime the excitement of our ‘find’ had begun to wear off and the water felt cold. I was just giving the sergeant some sage advice as to how he could direct his steps to the best advantage, when I 029stumbled over a large stone and fell into the water flat on my face. As just at present the weather is frosty and the rain is generally accompanied by sleet or hail, a bath in one’s clothes is anything but pleasant … We were altogether three hours in the water measuring and I took everything I could get at and have put the most important measurements on the 10 feet to an inch plan.”4
My own interest in tracing the development of the Temple Mount goes back to 1973 when I began work as field architect to the Hebrew University archaeological expedition directed by Professor Benjamin Mazar, which was excavating south and southwest of the Temple Mount. Mazar, now in his eighties, is the doyen of Israeli archaeologists. He has an uncanny intellectual intuition and an extraordinary knowledge of the archaeological as well as historical sources. I remember the moment well: Mazar and I were sitting in his room at the Hebrew University in the spring of 1980. He was reading a passage from Nehemiah 2:8 that referred to “beams for the gates of the birah that related to the Temple.”
“What and where is the birah?” Mazar asked.
Little did I suspect that this question would eventually lead to a breakthrough in the understanding of the architectural development of the Temple Mount.
My first reaction to his question was to suggest that the birah, often mistranslated as palace, might be synonymous with the 500-square-cubit measurement given for the Temple Mount in Middot 2:1.
“So,” Mazar followed up, “where is this square?”
It would have been helpful if we could have asked Charles Warren to join our “think-tank,” but although this was not possible we did have in our office his album of Plans, Elevations, Sections, etc. which contains detailed plans of the underground structures of the Temple Mount.5 Armed with this and Mazar’s knowledge of the historical sources, we set out to discover the location of this square Temple Mount.
The preliminary results were announced by Professor Mazar at the First International Congress on Biblical Archaeology held in Jerusalem in April 1984. The proposed square-shaped Temple Mount was promptly dubbed the “Ritmeyer Square.”6
I have continued to study the architectural development of the Temple Mount and believe that we can now demonstrate the results in more detail and with greater confidence.
Our demonstration begins with a flight of steps leading up from the Temple Mount to what we will call the Muslim Platform, the raised area on which the Dome of the Rock is built. Eight flights of steps lead up to the platform on which the Dome of the Rock sits; at the top of each flight of steps is an arcade that marks architecturally the entrance to the Muslim Platform.
One of these flights of steps is different, however—the one at the northwest corner. Each of the others is parallel to the wall of the Muslim Platform where it rises; that is, the bottom step of each staircase is parallel to the wall of the Muslim Platform and this determines the direction of the flight of steps.
However, the bottom step of the flight at the northwestern corner of the Muslim Platform is not exactly parallel to the wall to which it leads. This was pointed out to me in 1972 by my predecessor at the Temple Mount excavations, the Irish architect Brian Lalor. Each higher step on this flight of steps is naturally parallel to the lowest one, so the whole flight of steps is a bit off—that is, not exactly parallel to the wall to which it leads. Moreover, the construction of the bottom step is also different. It is made of a single line of large ashlarsc in contrast to the other steps of this flight, which are made up of many smaller stones. The southernmost ashlar had at the time a visible margin and boss on its front face. Lalor had suggested that this bottom step might be the remains of an early ashlar-built wall, which was why it was not parallel to the wall of the Muslim Platform.
After my talk with Mazar, I returned to this flight of steps and decided that the telltale lower step was a logical starting point for my continued investigation.
On this visit, I noticed something else about this bottom step composed of ashlars: It was exactly parallel to the eastern wall of the Temple Mount itself. When Herod enlarged the Temple Mount, he did not change the line of the eastern wall (the steep slope of the Kidron Valley was too close to the existing wall to move it). So the line of the eastern wall was pre-Herodian—perhaps even Solomonic. And now I had a wall—the lowest step on this flight of steps—that was exactly parallel to the eastern wall. I also noted that the northern end of the northernmost large stone of this step was exactly in line with the northern edge of the raised Muslim Platform.
On that later visit to the flight of steps, the sides of the ashlars that comprised the lowest step were no longer visible. The level of the adjoining pavement had been raised, concealing the sides of the ashlars. But in the office we had an early photograph that showed the boss of one of the ashlars. The boss is the central part of the ashlar that sticks out, in contrast to the margin—the edge—that encloses the boss. From the photograph, we could determine that the margin was 3.9 inches wide and the boss protruded approximately 3.1 inches. This is quite different from the typical Herodian masonry on the Temple Mount, which has a narrow margin of about 3.1 to 3.9 inches and a flat central boss that barely protrudes .4 inches. Accordingly, the ashlar in this step/wall gave a strong impression of being pre-Herodian. It looked very much like the lowest masonry in the central section of the eastern wall of the Temple Mount, near the Golden Gate. I therefore proposed that this step was actually a section of a wall—part of the western wall of the pre-Herodian, perhaps First Temple-period, Temple Mount.
If so, we had the eastern line and the western line of the pre-Herodian Temple Mount. And from Middot, 032we know that it was 500 cubits square. The next question was how far north and south did the original western wall extend?
For the northern end of the western wall, we turned to Warren for clues.
According to Warren’s records, he discovered an “excavated ditch,” 52 feet north of the step/wall we have been discussing.7 It had been “excavated” to create a fosse, or dry moat. Mazar immediately concluded that this must have been the moat described by Strabo, the Greek historian and geographer (64 B.C.–21 A.D.), who gives its measurements as 60 feet deep and 250 feet wide.8 The purpose of this moat could easily be surmised: The Temple Mount was protected by natural valleys on three sides, but not on the north. This moat protected the pre-Herodian 033Temple Mount on the north. It links the Tyropocon Valley on the west with a branch of the Bezetha Valley9 that runs into the Kidron Valley on the east. An enemy’s approach to the Temple Mount from the north over the narrow saddle that previously existed was thus effectively cut off by the moat. Incidentally, this same moat or fosse, we are told by Josephus,10 was filled in by Pompey’s soldiers in 63 B.C. to enable the Romans to storm the towers built onto the pre-Herodian northern wall of the Temple Mount.
It was thus clear that the western wall of the pre-Herodian Temple Mount terminated on the north no more than 52 feet north of the step/wall; for that is where the moat or fosse begins. The northern wall must be south of this fosse.
With this in mind we proceeded to look for possible remains of the pre-Herodian northern Temple Mount wall. Again the trail was illuminated by Warren. Underground structure No. 29 on Warren’s plan is a vaulted passage built against the northern edge of the Muslim Platform. The southern wall of this chamber is the northern wall of the Muslim Platform. Warren described the southern wall of this chamber as a quarried rockscarp. (The chamber, incidentally, may have been part of the Monastery of the Temple, built by the Crusaders.11) This rock scarp features prominently on Warren’s plans.12 We believe this scarp was cut to hold the foundation for the northern wall of the pre-Herodian Temple Mount.
A line projecting westward from this rock scarp forms a right-angled corner with the step/wall. As the northern edge of the northernmost large stone of the step/wall is exactly in line with the northern edge of the raised Muslim platform, it follows that this large stone forms the actual northwestern corner of the pre-Herodian square Temple Mount!
The 52 feet between the northern wall and the fosse is just enough room to construct the towers described by Josephus13; thus, this location of the northern wall of the pre-Herodian Temple Mount fits well with our historical information.
Continuing the line of the rock scarp eastward, we can locate the northeast corner of the pre-Herodian Temple Mount at the point where it meets the eastern wall, again forming a right-angled corner, just north of the Golden Gate. It is interesting to note that the earliest type of masonry (visible in the lowest courses) in the eastern wall is to be seen at this point.
The length of the northern wall of the pre-Herodian Temple Mount (measured along the rock scarp) located between the step/wall and the present-day eastern wall of the Temple Mount is 861 feet. According to Middot, the pre-Herodian Temple Mount was 500 cubits on a side. There are, however, at least three kinds of cubits. For two of them, 861 feet does not make 500 cubits. But for the so-called royal cubit of 20.67 inches,14 it turns out that 861 feet is exactly 500 cubits.
Naturally, in order to complete the 500-cubit-square Temple Mount referred to in Middot, we quickly measured 861 feet south from the point on the eastern wall that we had identified as the northeastern corner of the pre-Herodian Temple Mount. Here there is a slight bend in the wall, as recorded by Warren.d
To understand this bend and its significance we must look at the eastern wall more closely, assisted of course by the records of Warren. Beginning at the southeast corner of Herod’s—that is, today’s—Temple Mount, we see that the first 106 feet is clearly all Herodian masonry. At that point there is a straight joint, or seam, indicating where Herodian masonry was added to a preexisting eastern wall. North of the straight joint, or seam, on the 034eastern wall, is a distinctly different masonry that has been identified as characteristic of the Hasmonean period (142–37 B.C.).
Warren dug a shaft down along the eastern wall of the Temple Mount near the southeast corner and then began tunneling north. He tunneled 53 feet north of the straight joint. On the side of his tunnel he found, below ground, the same Hasmonean masonry that appeared above ground. Unfortunately, he stopped before reaching the limit of this Hasmonean masonry, so we don’t know how far north this Hasmonean masonry continues underground. However, Warren noted that “It is probable that below the surface the first 260 [should be 240] feet of wall [from the southeast corner] are in a straight line,”15 after which the wall changes direction slightly (to the northeast). This slight change in direction, I believe, reflects a change in masonry deep below ground.
The point at which this masonry would change, according to my speculation, based on the bend Warren observed, is exactly 861 feet, or 500 royal cubits, south of the projected northeast corner of the square, 035pre-Herodian Temple Mount. The southeastern corner of the pre-Herodian Temple Mount is probably still in existence deep below the ground.
The southern wall of this square pre-Herodian Temple Mount should be located parallel to the northern wall, beginning at the southeast corner we have just located. The southern wall should intersect with the continuation of the step/wall, the only pre-Herodian archaeological remains visible on the Temple Mount. (The masonry near the Golden Gate is also a visible remnant of the pre-Herodian Temple Mount. But the step is on the Temple Mount, while the masonry near the Golden Gate is in the outer wall of the Temple Mount.)
If our hypothesis is correct, then the description of the Temple Mount in Middot does indeed relate to the Temple Mount as it was repaired after the return from Babylon. No doubt the returning exiles, largely impoverished, did little more than repair the existing structure; they surely did not create new walls and fortifications. Accordingly, this square Temple Mount was probably the same as that which existed before the Babylonian destruction of Jerusalem in 586 B.C.—the Temple Mount of the First Temple period.
The Bible does not describe Solomon’s Temple Mount, although a Solomonic Great Court round about the Temple and the Royal Palace is mentioned in 1 Kings 7:2. A temenos, or sacred enclosure, excavated by Professor Avraham Biran at Tel Dan, is, although much smaller, also nearly square.e The temenos at Dan was probably built by Jeroboam I in the tenth century, shortly after Solomon’s death (1 Kings 12:28–31). Is it possible that this square temenos was modeled on Solomon’s Jerusalem Temple Mount?
Perhaps more relevant is the visionary temple described in Ezekiel 40–43. Adding up the dimensions of the gates and the two courts, as described in Ezekiel 40, 036this temple area also forms a square—of exactly 500 cubits to a side!
Several elements in our plan of the pre-Herodian Temple Mount are admittedly conjectural, but they tend to be confirmed by additional details.
One such detail involves the villain of the Jewish holiday of Hanukkah and a cistern investigated by Warren. Hanukkah celebrates the Maccabean victory (in 167 B.C.) over Antiochus IV Epiphanes, which returned Jerusalem to Jewish sovereignty for the first time since the Babylonian destruction of 586 B.C. The Seleucid monarch who had ruled Palestine was thereafter (in 142 B.C.) replaced with a dynasty of Jewish rulers known as Hasmoneans. The 25-year-long Maccabean revolt finally freed the Jews from foreign rule. Before his ouster the Seleucid ruler Antiochus had built a fortress called the Akra for his garrison from which the Jewish population could be controlled. After the successful Jewish revolt, Simon Maccabee razed the Akra and, according to Josephus, also leveled “the very mountain itself upon which the Akra happened to stand, that so the Temple might be higher than it.”16
The location of the Akra has been hotly debated by scholars.17 Josephus provides two clues: It stood “in the Lower City,”18 and “adjoined to and overlooked the Temple.”19 The Lower City is always understood as referring to the southeastern hill of Jerusalem, directly south of Temple Mount, known as the City of David. One would therefore expect to find the Akra south of the Temple Mount, and, if adjoined to the Temple, as Josephus says, directly adjacent to the pre-Herodian Temple Mount. South of the Temple Mount, the hill slopes away rapidly in all directions. If the Akra overlooked the Temple, this would be another reason to place it very near the southern wall of the pre-Herodian Temple Mount.20
As Professor Mazar has pointed out, there is a relatively flat area, perhaps the result of Simon Maccabee’s work, at the center of the southern wall of the pre-Herodian Temple Mount. And that is where I would locate the Akra.
The average level of the bedrock here is 2,400 feet, while the Temple Mount courts, directly to the north, are only 20 feet higher, so that it is not difficult to imagine a fortress here that was high enough to overlook the Temple.
This brings us to one of the most unusual cisterns under the Temple Mount. It is a curiously shaped 037cistern, like a letter E, unlike all the other cisterns under the Temple Mount, which are mainly irregular in shape. According to Warren, this E-shaped cistern could hold 700,000 gallons of water. It sits directly under the area where we have placed the Akra. Apparently it was especially cut to provide Antiochus’ garrisons with an ample supply of water in case of siege.
Shortly after I suggested this possibility to Mazar, he had a visit from Professor J. Schwartz of Bar-Ilan University, who was looking for the location of some cisterns on the Temple Mount mentioned in another tractate of the Mishnah, Eruvin. One of these cisterns was called “the cistern of the Akra.” Here was Professor Schwartz who had a name without a cistern, while we had a cistern without a name! Putting the two together, we realized that we had additional evidence for locating the Akra adjacent to the southern wall of the pre-Herodian Temple Mount.
This conclusion was further supported by Warren’s observation concerning the foundations of a massive wall in this cistern, apparently designed to support massive construction above.
Josephus also tells us that there was a direct entrance from the Akra to the Temple Mount, another reason to locate it where we have.f
If the E-shaped cistern is the cistern of the Akra as mentioned in the Mishnah, the fact that its northern edge is adjacent to the proposed southern line of the square Temple Mount tends to confirm our placement of that line.
Further confirmation that the square Temple Mount was where we have located it comes from the position of some of the other underground cisterns: Several similar small round cisterns appear along the northern, western and southern walls of this square Temple Mount.
Along the northern line near the eastern wall is cistern 15, on the inner edge of the platform. It would conveniently catch the rainwater runoff from the Temple Mount.
Cisterns 23 and 28 are located just outside the northern wall of the platform. They were probably located inside towers built along the northern wall to defend against an attack from that direction.
On the western line, cistern 31 has a location very similar to that of cistern 15 on the north, and probably fulfilled a similar function. The same goes for cistern 33 on the southern wall.
The next set of details leads to a completely new discovery: a Hasmonean addition to the First Temple-period square Temple Mount. That Herod enlarged the Temple Mount is well known, both from literary sources and archaeological evidence. The straight joint, or seam, in the eastern wall marks the point at which Herod began his addition; south of this straight joint is clearly Herodian masonry.
But only now have we been able to identify a Hasmonean addition to the original square Temple Mount. We have frequently referred in this article to the pre-Herodian Temple Mount. But in fact there were two pre-Herodian Temple Mounts: (1) the original Temple Mount of the First Temple period (rebuilt by the returnees from the Babylonian Exile) and (2) the Temple Mount as enlarged in the Hasmonean period in 141 B.C. by an addition on the south. It was this latter Temple Mount that Herod the Great (37–4 B.C.) enlarged.
Once we have demonstrated this, we may speak of the original square pre-Hasmonean Temple Mount, the enlarged Hasmonean Temple Mount and the further enlarged Herodian Temple Mount.
As we previously noted, the Maccabean revolt of 167 B.C. successfully expelled the Seleucid ruler Antiochus IV Epiphanes and instituted the Jewish Hasmonean dynasty. The hated Akra—the Seleucid tower south of the square Temple Mount that rose higher than the Temple and from which Antiochus could control the Jewish masses, most of whom (since the return from the Babylonian Exile) lived in the Lower City, south of this tower—was destroyed. In 141 B.C., Simon Maccabee starved out the Seleucid garrison that managed to hold out and then razed the Akra. According to 1 Maccabees 13:52, Simon Maccabee “strengthened the fortifications of the Temple Mount by the side of the Akra, and took up residence there with his men.” This seems to indicate that the area once occupied by the Akra was incorporated into the Temple Mount. Archaeology can now confirm this.
The Hasmonean masonry north of the straight joint appears to have been part of their enlargement of the square Temple Mount, in the area where the Akra 043had stood. The Hasmonean stones have been laid in a “header and stretcher”g fashion, indicating a corner construction.
There is no doubt that this masonry is earlier than the Herodian masonry south of it. We know this not only because of the obvious corner construction and the two different types of masonry on either side of the straight joint, but, in addition, a careful examination of the Hasmonean stones north of the straight joint shows that parts of their southern margins were cut away at several places to create a better key to fit into the Herodian stones south of the straight joint.
This Hasmonean masonry extends underground, I believe, for 132 feet north of the straight joint to the previously mentioned bend in the eastern wall, that is, to the southeast corner of the original, square Temple Mount.
From the straight joint, the Hasmonean wall apparently turned west until it met with the extension of the west wall of the original square Temple Mount. The Hasmonean Temple Mount was most likely embellished under Hellenistic influence with porticoes all around21 and with tunnels that gave access to the original gateways in the southern wall of the square Temple Mount—the Huldah Gates mentioned in the Mishnah.h
The E-shaped cistern, previously mentioned, helps confirm this Hasmonean extension to the square Temple Mount. Just northwest of the E-shaped cistern, the cistern of the Akra, is the largest of the cisterns under the Temple Mount, Warren’s cistern 8, known as the “Great Sea.” Unfortunately, the cisterns under the Temple Mount are inaccessible today because of Muslim religious and political sensitivities. Accordingly, we have to be content with a fairly one-dimensional view of them. However, on a recent visit to the London office of the Palestine Exploration Fund, I saw not only the original records of Warren, Wilson, Conder, etc., but also a painting of the “Great Sea” by William Simpson, who had been sent out to Jerusalem as illustrator for the Illustrated London News. (He was known as “Crimea” Simpson because of his experiences in the Crimean War). This painting is published here for the first time.22
Apart from adding to our “feel” for the underground cisterns, now tantalizingly inaccessible, careful study of the painting and a comparison with the plans of the underground cisterns led to a valuable confirmation of the Hasmonean extension.
Simpson’s painting shows a view to the north inside the gigantic cistern. The entrance was from the south, reached by a narrow staircase that begins just outside (south) of the square Temple Mount. The entrance was apparently dug on the site of the destroyed Akra. The cavern seems to be mostly rock-cut with columns left in place for support; its curious shape indicates that it may have incorporated earlier cisterns or caverns. The plans also show it to have several shafts to the surface, two of which are visible in the painting. Warren records the depth of the cistern as 43 feet. Apart from the entrance, the cistern was protected under the original Temple Mount.
If the Great Sea cistern had been in existence at the time of the building of the Akra, it would have been unnecessary to build the E-shaped cistern (cistern 11), which, as we have seen, was cut especially to supply the Seleucid garrison with water. It therefore follows that the Great Sea cistern was built after the destruction of the Akra and the subsequent Hasmonean extension of the Temple Mount to the south. (The cutting of this cistern may have supplied the stones for the Hasmonean extension.) This cistern may be one of the three described in Mishnah Eruvin 10:14:
“They [the Priests] may draw water with a wheel on the Sabbath from the Golah-cistern and from the Great Cistern, and from the Cistern of the Akra on a Festival-day.”
We have already identified the Cistern of the Akra with cistern 11 and we have reason to suggest that cistern 5 can be identified with the Golah-cistern. The cistern pictured in Simpson’s painting is therefore probably the one described in Eruvin as the Great Cistern.
Our location of the square Temple Mount and the Hasmonean extension illuminates the way in which Herod the Great extended the Temple Mount as it stood at the beginning of his reign. Herod’s extension also tends to confirm the location of the square Temple Mount.
The northern wall of Herod’s Temple Mount was completely destroyed by the Romans in 70 A.D. when they conquered Jerusalem, burned the Temple and effectively ended the First Jewish Revolt (although the Masada rebels held out for three more years). However, the northeast corner has been preserved and together with the rocky foundations of the Antonia fortress at the northwest corner23 enables us to draw the northern wall’s line.
Herod’s northern addition to the Temple Mount explains a strange notice in Middot 1:3 that “the Tadi gate [on the north] has no purpose at all.” This was probably because the Tadi gate was the northern gate of the original square Temple Mount. When Herod buried this area with fill to create his northern court, the Tadi gate was completely buried and thus rendered useless.24
At the northwest corner of his enlarged Temple Mount, on a rock scarp, Herod built the Antonia fortress, named for Mark Antony. This site is now occupied by the Omariya School. Its location, first suggested by Father Pierre Benoit,25 has been widely accepted.
044
The western wall of Herod’s extension still exists for the full length (as does the southern wall), so it is a simple matter to extend the northern wall to the point where it would meet the western wall.
From Herod’s western wall, the Temple Mount could be ascended through two lower gates, now called Warren’s Gate and Barclay’s Gate, after the scholars who discovered and identified them. (Two upper gates were built over Wilson’s Arch and Robinson’s Arch.)
In an earlier BAR article,i my wife Kathleen and I discussed the external problems relating to Barclay’s Gate. The internal problems are just as fascinating. From Barclay’s Gate, a subterranean internal stairway led up to the surface of the Temple Mount, opening on the western Temple court (see plans). Similar underground passageways led up to the Temple Mount from the Double and Triple gates on the southern wall, but these passageways are straight. The underground stairway from Barclay’s Gate is L-shaped. Starting at the gate, the passageway proceeds in an easterly direction for 84 feet inside the wall. Here, in Warren’s cistern 19, Warren describes a flat dome over the passageway at this point. The passageway then turns south; its continuation is found in Warren’s cistern 20, from which the surface of the Temple Mount was probably reached.
The eastern wall of the passageway at the point where it turns south under the flat dome has a distinct batter;26 that is, each higher course of stones in the eastern wall of cistern 19 is set in slightly from the one below. (Incidentally, the Herodian Temple Mount wall also has a batter.) The wall with a batter in cistern 19 appears to be the western wall of the Hasmonean Temple Mount built after the square platform was extended to the south in the wake of the destruction of the Akra.
The reason then for the L-shape of the underground stairway leading up from Barclay’s Gate is that Herod’s builders built the southern part of this stairway alongside the existing (Hasmonean) western wall, instead of attempting to cut through the wall, which would have been very difficult. Here again we find confirmation of the location of the square Temple Mount and its Hasmonean extension.
The other gate from Herod’s western Temple Mount wall up to the Temple Mount is Warren’s Gate, north of Barclay’s Gate. When Warren examined this gate, he described it as cistern 30, but he correctly identified it as having been a gateway tunnel before it was used as a cistern. It is 18 feet wide, the same as the passageway from Barclay’s Gate. The passageway in Warren’s Gate has a vaulted roof. I believe that this is the first part of another L-shaped subterranean stairway up to the Temple Mount. Today, the passage ends approximately 18 feet before the projected western wall of the original square Temple Mount. I believe the final 18 feet of the passageway was filled in at a later date, as was the continuation of 045the L along the original Temple Mount wall that led up to the western court of Herod’s addition.
It is interesting that even on the southern wall, the underground passageways from Herod’s Double and Triple Gates are approximately 240 feet long, opening on the Temple Mount almost precisely on the southern line of the original square Temple Mount, providing additional confirmation of our location.
Now, having located the original Temple Mount platform (as well as the Hasmonean and Herodian additions), we turn to the final question: Where, on the original Temple Mount, was the Temple located?
My research into the exact location of the Temple building is not yet complete. Many factors must be taken into consideration, but enough information has already been gathered to draw a preliminary plan.
It is generally agreed that Herod’s Temple stood at the same place as the Temple built by Solomon, which, obviously, stood within the original square Temple Mount. The square Temple Mount actually formed the great court surrounding the Temple.
One might expect the Temple to be located in the center of the square Temple Mount (in Ezekiel’s visionary Temple the altar is in the center; in the Temple Scroll the entrance to the Temple itself is located in the center). But here, that is not so. Eshakhra, the rock formation over which the Dome of the Rock is built, lies to the northwest of the center of the square Temple Mount. This mass of natural rock is the top of the middle part of the eastern hill of Jerusalem, and Josephus tells us that the Temple was built on the top of the mountain—that is, on top of es-Sakhra.27
If es-Sakhra was the location of the Temple, then the courts surrounding the Temple would all have different dimensions. That is precisely what Middot 2:1 tells us: “The Temple Mount was five hundred cubits by five hundred. Its largest [open space] was at the south, second largest, at the east, third largest, at the north, and least, at the west.”
The spaces referred to in this text are undoubtedly the different parts of the outer court that surrounded the inner court of the Temple (excluding, apparently, the Court of the Women).
Taking these guidelines into consideration, I drew the Temple according to scale, using the specific measurements given in Middot, and, while experimenting with the dimensions referred to above, hit upon an interesting picture: In order to fulfill the requirements of Middot for the southern court to have been the largest, and with the dimensions of the other parts of the outer court diminishing as one proceeds in a counterclockwise direction, the Temple must have been built around es-Sakhra.
Despite some recent proposals, which would place the Temple north of es-Sakhra,28 I have become convinced that the Temple was built over es-Sakhra, where researchers of the last century and a half have instinctively placed it.29 The difference between the approach of most previous researchers and my approach is that they used es-Sakhra as a starting point, while I reached this location after first having located the square Temple Mount, and working, so to speak, from the outside toward the center.
Another factor supporting my conclusion is the fact that, according to Middot the Temple had foundations 6 cubits high. But why would this be necessary if the Temple were built on bedrock?
Es-Sakhra now stands about 5 to 6 feet above the floor of the Dome of the Rock. A short distance outside the Dome of the Rock (in cisterns 1, 3 and 5) Warren found the difference in height from the top of es-Sakhra to bedrock to be 13 to 15 feet. Taking into consideration that the bedrock would gently slope up from the points Warren measured to the base of es-Sakhra, it seems that es-Sakhra stands about 10 feet above the bedrock immediately surrounding it. Ten feet equals 6 cubits, the height of the Temple foundation specified in Middot. It stands to reason that these foundations—6 cubits high—were necessary to “bury” es-Sakhra in order to create a level platform on which the Temple could be built.30
With this positioning of the Temple on the square Temple Mount, for the first time, all the factors—topographical, archaeological and historical—have been taken into account. And all seem to fit precisely.
While I believe that there is already enough evidence to place the Temple firmly over the top of es-Sakhra, I intend to undertake further study of the historical sources in light of the topography and architecture of the Temple Mount which will, hopefully, enable me to refine this proposal still further.
065
Drawings Available
Black-and-white copies of the drawing of “The Development of the Temple Mount During the Second Temple Period” (17” by 25”), on chrome paper, are available for $7.50 and handpainted copies for $25 (surface mail included, add $5 for airmail). Also available, for the same price, are copies of the following drawings: “The Temple Mount During the Second Temple Period” (13” by 25”) as published in “Reconstructing Herod’s Temple Mount in Jerusalem,”BAR 15:06, “The Northern Palaces of Masada” and “Qumran” (both 20” by 25”). Greeting cards (with envelopes), featuring the Jebusite Gateway, the Double Gate, the Trumpeting Stone and the Second Temple Menorah, each with an appropriate scriptural quotation, are available for $10 for a pack of 12. Please send check or money order to: Ritmeyer Archaeological Design, 49 Burtree Ave., Skelton, York YO3 6YT, England.
Somewhere on Jerusalem’s majestic Temple Mount—the largest man-made platform in the ancient world, the size of 24 football fields, nearly 145 acres—Herod the Great (37–4 B.C.) built a new Temple to the Israelite God Yahweh,a doubtless on the very spot where the exiles returning from Babylonia more than 500 years earlier had rebuilt the original Temple, first erected in the tenth century B.C. by King Solomon. But where was that spot? Efforts to locate it have sought clues from archaeological evidence on the Mount itself and from two famous descriptions of Herod’s Temple—one by the first-century A.D. Jewish […]
You have already read your free article for this month. Please join the BAS Library or become an All Access member of BAS to gain full access to this article and so much more.
The Mishnah is the collection of Jewish oral laws compiled and written down by Rabbi Judah the Prince in about 200 A.D.
3.
An ashlar is a rectangular block of hewn stone.
4.
Warren observed this slight bend with his surveying instruments. Presumably it could be confirmed by remeasurement today. However, for political reasons, it is not feasible to undertake this exercise now.
This direct entrance was probably the forerunner of the Triple Gate in the southern wall of the present Temple Mount.
7.
Stretchers are laid lengthwise in the wall, the long side facing out. Headers are laid with the shots side facing out. In “header and stretcher” construction, headers and stretchers alternate. This gives a wall much greater strength, especially where needed at corners.
8.
The Double and Triple Gates on the southern wall of the Herodian Temple Mount that can still be identified are frequently, but incorrectly, referred to as the Huldah Gates. The Huldah Gates mentioned in Middot 2:1 were located in the southern wall of the original square Temple Mount.
Charles Wilson and Charles Warren, Recovery of Jerusalem (London: Bentley, 1871), chap. 7, “Tanks and Souterrains of the Sanctuary,” pp. 204–217.
3.
Jan Simons, Jerusalem in the Old Testament (Leiden: Brill, 1952), p. 355.
4.
Wilson and Warren, Recovery of Jerusalem, pp. 209–213.
5.
Warren, Plans, Elevations, Sections, etc., showing the results of the Excavations at Jerusalem, 1867–1870, executed for Committee of the Palestine Exploration Fund (PEF) (London: PEF, 1884), Pls. VI, VII.
6.
Biblical Archaeology Today, Proceedings of the International Congress on Biblical Archaeology, Jerusalem, April 1984, ed. Janet Amitai (Jerusalem: Israel Exploration Society [IES], 1985), p. 484.
7.
Warren and Claude Conder, Survey of Western Palestine, Jerusalem (London: PEF, 1884), p. 21S. See also, Warren, Plans, Elevations, Sections, etc., Plate XII, Section O-R.
8.
Strabo, Geography 16.40.
9.
This small valley was called St. Anne’s Valley by Wilson, Chaphenatha Valley by Conrad Schick (Die Stiftshutte, Tf. 4) and more recently the Beth Zetha Valley by Dan Bahat (Carta’s Historical Atlas of Jerusalem [Jerusalem: Carta, 1983], p. 11).
10.
Josephus, Antiquities of the Jews 14.4.2.
11.
See Bahat, Carta’s Historical Atlas, pp. 55, 61.
12.
Warren, Plans, Elevations, Sections, etc., Plates II and VI.
13.
Josephus, Antiquities of the Jews 14.4.2.
14.
See George A. Smith, Jerusalem, From the Earliest Times to A.D. 70 (London: Hodder and Stoughton, 1908), vol. 2, p. 588, n. 3: 20.67 inches = sacred cubit. See also Arye Ben David, “The Hebrew-Phoenician Cubit,” Palestine Exploration Quarterly (PEQ) (1970), pp. 27–28: the Philaeterian-Ptolemaic cubit = 525 mm (20.67 inches). Joachim Jeremias (Jerusalem in the Time of Jesus [London: SCM, 1969], p. 11) defines the Philaeterian Cubit as equal to 525 mm, or almost 21 inches. He quotes Didymus (end of first century A.D.) “who calculates the Egyptian cubit of Roman times as 1 ½ Ptolemaic feet.” As this foot was 350 mm long, the cubit was 1.5 × 350 = 525 mm. David Ussishkin, “The Original Length of the Siloam Tunnel,” Levant 8 (1976), pp. 82–95: cubit = 52.5 cm (= 20.67 inches). Cf. Asher S. Kaufman, “Determining the Length of the Medium Cubit,” PEQ 116 (1984), p. 131. Kaufman is close, but his measurement is not exact (20.319 inches instead of 20.66925). My own research has shown that 500 cubits of 20.67 inches equals exactly the distance from the step to the eastern wall, and therefore I believe that this was the cubit used for laying our the 500-cubit-square Temple Mount.
This cubit originated in Egypt and is also called the Egyptian long cubit; the short cubit was only 450 mm long. The long cubit is also known as the royal cubit, and had been in use since the 15th century B.C. This cubit rarer became known as the Philaeterian cubit, after the family name of the kings of Pergamum.
15.
Warren and Conder, Survey of Western Palestine, p. 146: “[T]he general direction of the east wall with south wall, as determined by the Survey, is 92 degrees 50 minutes. The eastern wall is somewhat irregular, the first 120 feet only being in a straight line; beyond this are several bulges, but it is probable that below the surface the first 26O feet of wall are in a straight line. At this point there is a small postern on about the same level as the Single Gate on the south side. From this postern the wall takes a slight bend to north-east, so that at 650 feet from south-east angle it is about 8 feet to east of a line in production of first 260 feet.” Simons also noted this bend to the northeast, and corrected its location to 240 feet from the southeast corner (Jerusalem in the Old Testament, p. 370): “The first section of the [east] wall, from the same [S.E.] angle to a point 73.20 meters [= 240 feet] to the north, is practically straight but at this point it hends slightly outward, to the north-east, so that after 200 meters [= 650 feet] it is already 2.5 meters [= 8 feet] outside the line of the southern section.”
16.
Josephus, Antiquities of the Jews 12.6.7.
17.
See Yoram Tsafrir, “The Location of the Seleucid Akra in Jerusalem,” in Jerusalem Revealed, ed. Yigael Yadin (Jerusalem: IES, 1975), pp. 85–86. More recently, Gregory J. Wightman, “Temple Fortresses in Jerusalem, Part 1: The Ptolemaic and Seleucid Areas,” Bulletin of Anglo-Israel Archaeological Society 9 (1989–1990), pp. 29–40.
18.
Josephus, The Jewish War 1.1.4.
19.
Josephus, Antiquities of the Jews 12.9.3.
20.
Tsafrir (“Location of the Seleucid Akra,” pp. 85–86) accepts the southeastern hill of Jerusalem for the location of the Akra and correctly places it to the south of the Temple, indeed close to our suggested location. We prefer, however, to place the Akra near the center of the southern Temple Mount wall, above cistern 11, from which it would have been easier to control the ciry and the southern access to the Temple Mount, and not near the southeast corner, as Tsafrir suggests. It would have been better to construct a fortress that could overlook the Temple Mount at the summit of the southeastern hill, rather than on the much lower slopes near the southeastern corner. Tsafrir’s suggestion of identifying the masonry to the north of the “straight joint” in the eastern wall with the foundation of the Akra contradicts the statement of Josephus that the Akra was razed to the ground; it seems more logical to identify this stretch of Hellenistic masonry with the enlargement of the Temple Mount during the Hasmonean period.
21.
Josephus (Antiquities of the Jews 14.16.2) relates that when Herod, after he was made king, took Jerusalem and the Temple Mount, “the cloisters that were about the Temple were burnt.”
22.
I would like to thank Dr. Rupert Chapman, secretary of the Palestine Exploration Fund, and Shimon Gibson, the photographic officer, for their help and cooperation in obtaining a copy of Simpson’s painting of the “Great Sea.”
23.
In his recently published book, Mamluk Jerusalem, Michael Burgoyne has identified a 4-meter wide Herodian wall running east-west along the southern edge of the Antonia rock plateau. This wall, the southern wall of the Antonia fortress, is in line with the preserved northeast corner of the Temple Mount.
24.
Although the underground passageway originally leading to it from the inside was still used by the priests (see further, endnote 30).
25.
Father Pierre Benoit, O.P. “The Archaeological Reconstruction of the Antonia Fortress,” in Yadin, Jerusalem Revealed, pp. 87–89.
26.
Warren, Plans, Elevations, Sections, etc., Pl. XXXII.
27.
Josephus, The Jewish War 5.51: “At first the plain at the top was hardly sufficient for the holy house and the altar, for the ground about it was very uneven, and like a precipice.”
28.
The most famous of these is Asher Kaufman’s. See his article “Where the Ancient Temple of Jerusalem Stood,”BAR 09:02. Kaufman places the Temple over a small monument known as the Dome of the Tablets, northwest of es-Sakhra. Kaufman’s suggestion has tecently been supported by Lawrence D. Sporty (Biblical Archaeologist, March 1991, pp. 28–35). Kaufman’s theoretical location of the Temple over the Dome of the Tablets has several weaknesses: 1. He completely ignores the most important topographical data of the area north of the Muslim Platform—(1) the fosse, or moat, observed by Warren, (2) the Bezetha Valley to the east of this moat and (3) the rock scarp under the northern edge of the Muslim Platform. The northern court of Kaufman’s Temple would fall into the Bezetha valley! Near the eastern wall, this valley is 160 feet lower than the es-Sakhra! 2. The floor under the Dome of the Tablets, where Kaufman located the Temple, is a stone—apparently a large paving slab or other stone in secondary use and not bedrock. Bedrock is at least 8 feet below the floor of this small monument. Thus, this location is by no means the top of the hill, as Josephus described the Temple’s location. 3. Es-Sakhra, by contrast, is about 15 feet higher than Kaufman’s location for the Temple. On this point alone, his theory is untenable. 4. His interpretations of the bedrock formations near the northwest corner of the Muslim Platform are highly dubious. Moreover, he failed to recognize the most important remains in this area: the step/wall, which is crucial in identifying and defining the square Temple Mount.
29.
For the other theories, see Melchior de Vogue, Le Témple de Jerusalem (Paris, 1864); James Fergusson, The Temples of the Jews and the other buildings in the Haram area at Jerusalem (London, 1878); Charles Warren, The Temple or the Tomb (London, 1880), Claude R. Conder, “Statement of the Principal Controversies, II. Site of the Temple,” Survey of Western Palestine (Jerusalem and London, 1884); Conrad Schick, Die Stiftshütte, der Tempel in Jerusalem und der Tempelplatz der Jetztzeit (Berlin, 1896); Charles M. Watson, The site of the Temple (London: PEF, 1896); Carl Mommert, Topographie des alten Jerusalem, Zweiter Teil Das Salomonische Tempel—und Palast—quartier auf Moriah (Leipzig, 1903); Gustaf Dalman, “Der Zweite Tempel zu Jerusalem,” Palästina-Jahrbuch, 5 (1909); F.J. Hollis, The Archaeology of Herod’s Temple (London, 1934); Louis-H. Vincent, “Le Temple Hérodien d’après la Misnah,” Revue Biblique 61 (1954); Asher S. Kaufman, “The meaning of Har Habayit and its northern gate,” Niv Hamidrashia 18/19 (1984/1985); Benjamin Mazar, “The Temple Mount,” Biblical Archaeology Today (Jerusalem: IES, 1985).
30.
Additional support for this location can be derived from the position of the underground cisterns, as surveyed by Warren. All of the cisterns, apart from cistern 12, fall outside the sanctuary, as I have located the Temple. (One would hardly expect to find people drawing water within the sanctuary itself.) As I have located the Temple, the large cistern 5 would be situated next to the Water Gate, which was near the altar, thus providing a convenient source of water for the service of the Temple.
With this location of the Temple, one can go on to identify a few more elements of the original Temple. Middot 1:3 describes the Tadi Gate as being “on the north, serving no purpose at all.” It also mentions that if “one of [the priests] should have a nocturnal emission of semen, he goes out, proceeding along the passage that leads below the building—and lamps flicker on this side and that—until he reaches the immersion room. Rabbi Eliezer ben Jacob says, ‘He goes out by the passage which leads below the rampart (chel), and so he came to the Tadi Gate.’”
Both Warren and Conder concluded that if cisterns 1 and 3 were extended farther to the north, they would meet exactly at a point in the rock scarp where they placed the Tadi Gate (Survey of Western Palestine, Jerusalem, p. 218). Cistern 1 was probably the passageway reached by descending from the Chamber of the Hearth, which was one of the three gates on the north of the inner court of the Temple, while cistern 3 was the immersion room itself. Middot continues, that there were “four offices in the Chamber of the Hearth; … [T]hrough that on the northwestern side do they go down to the room for immersion” (Middot 1:5). Despite the fact that the Tadi Gate was put out of use by Herod’s northern extension, the underground passage was still used by the priests to visit the immersion room (see endnote 24).
It is also interesting to note that this passageway (cistern 1) is exactly in line with the rock under the Dome of the Rock, and also with the passageways of the Double Gate. Using either the northern Tadi Gate or the southern Huldah Gates, the pilgrim of the past would always see first whatever was built over that rock, whether altar, Porch, the Holy or most probably the Holy of Holies of the Temple. The importance of this architectural alignment shows that these gates were built according to a uniform plan.
This southern route passes in between cisterns 6 and 36, which according to Ronnie Reich might have been mikva’ot (“Two Possible Miqva’ot on the Temple Mount,” Israel Exploration Journal 39 [1989], pp. 63–65). Reich suggests, however, that these mikva’ot were located outside the early Temple Mount.
