Netzer’s Antonia; by Jerry Murphy-O’Connor

[Admin1]

Netzer’s Antonia

I read with great interest Ehud Netzer’s reconstruction of the Antonia fortress in a recent issue of BAR [ BAR 35:01, Jan/Feb 2009]. My appreciation, however, was greatly diminished by the fact that he failed to take into account five archaeological elements that must be accounted for in any adequate hypothesis regarding the appearance of this important Herodian building. These were discovered by the eminent architectural historian Michael Burgoyne and published in his Mamluk Jerusalem (1987).

      (1) Roughly 7 m north of the Ghawanima Minaret in the north-west corner of the Haram esh-Sharif lies a 5 x 1.5 m stone with distinctive Herodian marginal drafting on the long northern and short western faces. If, as it would appear, this block is in situ, it can only belong to a corner of the Herodian esplanade. In Netzer’s reconstruction it is buried somewhere inside the building.

      (2) Within the Omariyya School at the top of the escarpment at the north-west corner of the Haram esh-Sharif are three rooms of the theological college built by the Mamluk Sanjar al-Jawili (Al-Jawiliyya) in the second decade of the fourteenth century. The southern part of the three rooms at the very edge of the escarpment is cut into a wall 4 m thick, which is visible for some 17 m. The obvious hypothesis that this was the south wall of the fortress is confirmed by Burgoyne’s discovery of drafted ashlars in the Herodian style at the bottom of the external revetment.

      These elements admirably illustrate JW 5.238-239, according to which Herod built the Antonia on a rocky outcrop 50 cubits high, whose natural defensive advantages he improved by facing all sides with smooth stones, thereby creating a classic glacis. The rock elevation, on which Josephus lays such emphasis, is still visible. There is the 12.5 m drop off from the 4 m wall into the Haram and 40 m further north a similar drop (but shallower) into the street Tarik al-Saray al-Qadim. This little hill is completely ignored by Netzer in his reconstruction of the floor plan of the Antonia. Even though Josephus greatly inflated the height of the hill, a the hill from the Haram side is the height of a modern three-storey house.

      (3) A 25 m extension to the west of this thick wall can be assumed with a fair degree of probability. An extension to the east of some 70 m is a certainty, because traces of the same wall have been found in a series of mid-fourteenth century Mamluk buildings built along the north wall of the Haram esh-Sharif. The easternmost building in which the wall appears is the Al-Farisiyya, which is also the end of the rocky outcrop on which the Antonia was built. No trace of the wall was found in Al-Aminiyya. Thus, the wall is certain for 87 m and probable for 25 m, which suggests that the south wall of the Antonia measured 112 m. I could find no basis for Netzer’s estimate of 86 m, which moreover conflicts with his scaled drawing measurement of approximately 100 m.

      (4) Burgoyne’s most significant discovery was in the tomb chamber of Al-Isardiyya. He concluded that the size and tooling of the masonry of the south wall (the interior face of the 4 m wall) was Herodian and in situ up to a height of roughly 2 m. He further pointed out that the stepped plan of the upper masonry suggested a series of pilasters. Pilasters immediately evoke the Herodian enclosures in Jerusalem and Hebron. Here, however, they faced an internal space, probably a courtyard.

      (5) In the rock face below the 4 m wall two sets of cuttings are still visible. The trapezoid lower ones took vaulting springers, which Netzer no doubt would have attributed to the Umayyad caliph Mu’awiyah (661-680) had he noticed them. Much more important, however, is the much higher series of square beam sockets (0.48 m). Burgoyne estimates that these are 8.37 m above ground level, and resonably sugests that the beams supported the roof of the cloister with which Herod surrounded the Temple Mount. According to JW 5.190, the columns of the cloister were 25 cubits high (11.5 m). Exaggeration, however, is one of Josephus’ trademarks.

      Unfortuately inaccuracy is also typical of Josephus, and he leads Netzer astray by saying that the Antonia was built on the site of the Hasmonean Baris. This cannot be correct. The two fortresses did serve the same defensive function regarding the Temple, and this perhaps explains Josephus’s mistake. The snag is the ditch that ditch that Pompey had to cross to reach the Temple (Strabo, Geography 16.40). It has been located just inside Bab al-Nazir. The function of this ditch can only have been to protect the Baris, which in consequence must have been sited just south of it. Ritmeyer’s reconstruction drawings show that there was adequate space between the ditch and the square pre-Herodian Temple mount. Herod opted for a much more defensible site some 80 m further north.

      In sum, therefore, I think that Netzer is slightly inaccurate in his history of the Antonia, and would suggest that the Antonia was only about half the size of the one that he postulates (3300 against 7396 sq. m). 

Jerry Murphy-O’Connor

Ecole Biblique

Jerusalem