Layers of Ancient Jerusalem

037

If you go to the famous Western Wall in Jerusalem, which is actually the western retaining wall of Herod’s Temple Mount and Judaism’s holiest prayer site, and then turn around, you will see at the other side of the plaza an area of less than half an acre that has recently been excavated. Large-scale archaeological excavations were conducted here between 2005 and 2010 on behalf of the Israel Antiquities Authority, initiated and underwritten by the Western Wall Heritage Foundation and directed by the authors of this article.¹ What we have found sheds light on important transitional phases in Jerusalem’s history, but it also raises fascinating new questions.

Actually, the site of the excavation was outside the city until the eighth century B.C.E. It is located on the northeastern slope of the so-called “western hill.” In King David’s time, Jerusalem was confined to the 10–12-acre ridge south of the Old City known even today as the City of David. Then Solomon extended the city northward, where he built the Temple on a much smaller enclosure than the one built by Herod the Great a thousand years later. Not until King Hezekiah’s time, 038 in the eighth century B.C.E., was the city extended west to the western hill—the present area of Mount Zion, the Armenian Quarter and the Jewish Quarter of the Old City. We know all about this western extension because excavations in the Jewish Quarter, directed by the late Professor Nahman Avigad in the 1970s and 1980s, have unearthed the imposing “Broad Wall,” as it is known, built by King Hezekiah when he extended the boundary of the city to the western hill. Our site was most probably enclosed within the Broad Wall, although the exact course of the wall between today’s Jewish Quarter and the Temple enclosure is not yet known.

Since our site was outside the city until the eighth century B.C.E., it is not surprising that at the very bottom of our excavation, we found a stone quarry, the earliest evidence of human activity at the site. It had been used to create building blocks of meleke, the hard limestone for which Jerusalem is still famous.

The zigzagging line of quarries cuts across the slope of the hill. The quarrying also left upright stone walls as high as 12 feet. In the smaller hewn steps are separation trenches used to remove the quarried stones. We can still see the diagonal chisel marks of the stone cutters.

In the seventh century B.C.E., not long after this area was enclosed within the city wall, an impressively large structure was erected on top of the old quarry. Although we have not excavated it completely, the part of the building that has been exposed suggests it is a so-called “four-room house,” the typical architecture of an Israelite house during the First Temple period, consisting of three long rooms and a fourth, broad room extending across the other three. Each of these rooms could be further subdivided. The outside walls of the structure we revealed are about 25 feet long, much of which is preserved to a height of more than 15 feet. The base of the walls was founded on the quarry, and the floors were then laid on an 8-foot-high fill of dirt and stone, abutting the walls.

This house continued to stand until at least the early sixth century B.C.E. Then it was destroyed. A great pile of fallen stones in one of the rooms evidences collapse of a second floor. The domino effect of the fallen stones indicates that the collapse was a one-time event, the outcome of a sudden, violent destruction. After the destruction, a dirt fill gradually accumulated in the structure to the height of the walls; this fill probably slid down naturally from other buildings farther up the slope that had been destroyed by the Babylonians. In this 039 fill we found much pottery from the last centuries of the First Temple period (eighth–sixth centuries B.C.E.).

Now we come to a major mystery: Who or what destroyed the building?

Theoretically there are at least two possibilities: (1) an earthquake or (2) the Babylonians, when they burned Solomon’s Temple and destroyed Jerusalem in 586 B.C.E. The absence of any pottery vessels in the destruction collapse weighs heavily in favor of the second possibility. This indicates that the building was abandoned in an orderly fashion before its destruction and that its occupants took their household goods and personal belongings with them (probably as the Babylonians drew near).

Yet we found in this structure several personal seals bearing Hebrew names (all studied by Tallay Ornan, Benjamin Sass, Yuval Goren, Baruch Brandl and Othmar Keel). Perhaps the most intriguing one is a black elliptical phosphorite seal engraved with the image of an Assyrian-style archer and the Hebrew inscription l’Hagav (לחגב), “[belonging] to 040 Hagav.” The lettering indicates that this was a local Judahite seal, but the image of the archer identifies the owner as someone of Assyrian cultural background or inspiration and of military status. How to explain it? We know of nothing comparable that contains all these features.²

Another scarab-like limestone seal is elliptical and measures 0.4 by 0.5 inches. The surface of the seal is divided into three registers separated by double lines: In the upper register is a chain decoration of four pomegranates; in the two lower registers is the name of the owner of the seal, engraved in ancient Hebrew script (the kind used before the Babylonian destruction). It reads: “[belonging] to Netanyahu ben Yaush” (לנתניהו בנ יאש). Each of these names is known from ancient sources: The name Netanyahu (נטניהו), or Nethaniah in English, is mentioned in the Bible a number of times (Jeremiah 40–41; 2 Kings 25; 1 Chronicles 25; 2 Chronicles 17), and Yaush is mentioned in the Lachish letters.^a However, this combination of names—Netanyahu the son of Yaush—was not known previously.

Yet another seal, this one made of bone, was incised “[belonging] to Yeda’ayahu Usha” ( לנתניהו אושא). Still another well-worn seal was decorated with the image of a roaring lion and, beneath it, the owner’s name: N’wa or N’ra (לנרא/לנוא) “[belonging] to N’ra.”³

Even more surprising—and puzzling—is a seal or amulet of pink limestone decorated with the image of a winged snake, a cobra-like hybrid, possibly to be identified with the Biblical saraf (the “fiery serpent” of Isaiah 30:6). It is being studied by Tallay Ornan.⁴ And a broken Egyptian scarab of the 22nd Egyptian dynasty (945–713 B.C.E.) is being studied by Othmar Keel.⁵ What is it doing here?

Only the scarab and the N’ra seal were found 041 outside the building; all the others were found inside. The question arises: If the inhabitants took everything with them before abandoning the structure, why didn’t they take these seals? There is no easy answer. Perhaps the owners lost them; they are quite tiny. The largest is only 0.6 by 0.5 by 0.3 inches; the smallest, 0.5 by 0.4 by 0.2 inches. Another possibility is that some of the seals were deposited at the site following its destruction, by erosion from the houses farther upslope.

While the likelihood does seem to be that the building was destroyed by the Babylonians, there is another strange fact that suggests caution: We found no evidence of fire, as might be expected if the building was destroyed by the Babylonians. It is not clear, then, that the building was a casualty of the Babylonian destruction. Was it perhaps destroyed in an earthquake after all?

Among the other finds from the First Temple period were ten so-called l’melekh handles from the eighth century B.C.E. L’melekh means “[belonging] to the king” in Hebrew, and these handles identify large storage jars presumably certified by the government with this impression concerning the contents. Three of the handles carried city names—one from Hebron and two from Soccoh. Others were too damaged to determine whether they carried any inscription at all. These l’melekh handles are usually associated with Hezekiah’s preparations for the Assyrian onslaught that came to Judah at the end of the eighth century B.C.E. Eleven other jar handles feature two small concentric circles as was used in the late eighth century B.C.E.

Other potsherds were inscribed with a few Hebrew letters, and one had a palm tree, a common symbol of Judah.

In the dirt fill that accumulated under and on top of the floors of the building, we found hundreds of fragments of clay zoomorphic and anthropomorphic figurines. Some were of women, which we interpreted as fertility figurines; others, featuring men riders and animals (horses, bulls), were also found. All of the figurines were broken, probably due to the poor quality of local industrial manufacture. Such figurines are common finds in Judah in the late Iron Age (eighth to sixth centuries B.C.E.) and must have been part of the Judahite household. Their use—whether as objects of a religious ritual or as personal amulets—is not known.

We believe that this four-room house and/or the neighboring buildings were inhabited by members of Judah’s social elite, as reflected in the seals. Perhaps they were members of the ruling class in Judah’s capital. The seals also hint at a cultural connection between the kingdom of Judah and the kingdoms of Assyria and Egypt at the end of the First Temple period.

After the Babylonian destruction, this part of Jerusalem was uninhabited for centuries. Even when some of the Babylonian exiles began returning in the late sixth century B.C.E., Jerusalem was a much smaller place, occupying mainly the City of David area. We did not find anything from the early part of the Second Temple period, and the reason is clear: There was probably very little, if anything, built here.

Eventually, however, the settled area of Jerusalem expanded, culminating in the prosperous city of the Hasmonean (167–37 B.C.E.) and Herodian periods (37 B.C.E.–70 C.E.). Even so, we found almost nothing in our excavation between the end of the First Temple period (586 B.C.E.) and the beginning of the Late Roman period (early second century C.E.). The only significant exception was a 35-foot section of the Lower Aqueduct on top of the rock cliff, to the west, that carried water from the so-called “Solomon’s Pools” south of Bethlehem to the Temple Mount in the latter part of the 042 Second Temple period. From this period we also discovered part of a small mikveh, or ritual bath, suggesting that at one time there were other buildings here. But that’s all.

Directly on top of the First Temple period fill were paving stones of the Roman colonnaded street, commonly known as the eastern cardo (one of the city’s two main north-south thoroughfares). When the Romans laid the cardo in the second century C.E., they sealed and preserved the four-room house below until our own time, but they shaved off everything else that may have been built during the Second Temple period, forever destroying this history of the site.

The common street design in Roman cities of the eastern part of the Roman Empire featured two main thoroughfares—the north-south cardo and the east-west decumanus—that formed the basis for an orthogonal plan, a square grid of streets set at right angles. This layout is well known at Jerash/Gerasa in Jordan, Hippos/Sussita overlooking the Sea of Galilee and at sites such as Beth-Shean/Scythopolis and Sepphoris/Diocaesarea in Israel, to name just a few. However, the square grid of these cities was forced uncomfortably on the steep topography of Jerusalem. In order to retain the overall north-south/east-west network of more or less level streets, Jerusalem’s eastern cardo cut its way through the slope of the western hill while pointing south. It thus created a high vertical, rocky cliff (35 ft high) along the western side of the street. This manmade cliff revealed in our excavation runs north to south, separating the current Jewish Quarter of the Old City from the Western Wall plaza.

What was this thoroughfare like? In the center was the street itself, 26 feet wide, paved with large, hard mizi-hilu limestone slabs set diagonally across the street, creating a handsome pattern. On both sides were open sidewalks, each 5 feet wide, paved with the same sort of flagstones, laid in a parallel direction to the street, and raised one or two stair heights above the street’s surface.

Although they varied somewhat in size, each flagstone was about 5 feet long, 3 feet wide and a foot thick. These street flagstones were smoothed from many years of use and were on occasion rutted with parallel grooves, probably meant to prevent slipping. Beneath the paving stones was a well-developed drainage system with deep channels to carry sewage.

On either side of the street was a row of columns that served as part of a pedestrian colonnade (20 ft wide, including the columns). Only one column base from the western colonnade was found in situ in our excavation. We reconstruct the colonnades on the basis of the many columns and bases that were incorporated in secondary use in the Islamic buildings that were built on top of the cardo in the late seventh or early eighth centuries C.E. For example, on top of the column base that was found in situ, a broken shaft of an original Roman cardo column (about 6.5 ft long and 2 ft in diameter) was set in the Early Islamic period and was later embedded in a Late Islamic square pilaster. Two Corinthian capitals were reused in Early Islamic walls, but they probably originated in buildings adjacent to the cardo and not in the colonnade itself.

043

Adjacent to the western side of the western colonnade, a row of eight shops was hewn from the bedrock at the foot of the rocky manmade cliff. Seven were identical, measuring about 11 by 15 feet with walls up to 14 feet high. The southernmost shop was longer and wider than the others and its ceiling was cut into the rock. One shop had a ceiling of wooden beams, indicated by square niches in the walls. Three adjacent shops north of the southern shop were later joined to this one to form a single space, perhaps presaging the modern tendency to combine smaller establishments into increasingly larger ones.

We date the construction of the eastern cardo according to the date of the latest finds discovered under the paving stones. A rich assemblage of small finds (potsherds, glass vessels, bones and coins) was found within dirt and dump fills that accumulated in the ancient quarries that were sealed under the cardo’s flagstones. The latest coin in this fill was minted in Antioch during the reign of the Roman emperor Hadrian (117–138 C.E.). So the cardo above must have been built during or after that time.

The Hadrianic date of the cardo is further supported by a unique assemblage of clay vessels (studied by Renate Rosenthal-Heginbottom) that can safely be dated to the late first–early second centuries (c. 70–130 C.E.). Three types of oil lamps coexist within this assemblage: Roman moldmade lamps produced in the Levant as copies of imperial prototypes and imported into Jerusalem mainly in the first century C.E.; Levantine lamps with a round nozzle and decorated discus that replaced the imperial lamps in the late first century; and 046 wheel-made lamps with spatulate nozzle, characteristic of the Jewish population of the Jerusalem area in the Second Temple period and produced in the vicinity of Jerusalem until the Bar-Kokhba Revolt (132–135 C.E.). Another special group of vessels reflects the Roman pagan culture typical of the Roman army: relief-decorated drinking vessels and jugs, imitating silver and bronze tableware, depicting deities and their attributes, for example, Eros, the god of love, holding a hare, as well as a seated god identified as Neptune.

The fill under the street of the cardo also included dozens of ceramic roof tiles. Oddly, however, none of these was stamped with the Latin abbreviation LXF (Legio X Fretensis), which is so common in archaeological layers of the second–fourth centuries, identifying the Tenth Roman Legion that was stationed in the city, then called Aelia Capitolina. The absence of these LXF tiles in our excavation may be simply accidental or it may be chronologically significant. Perhaps the Tenth Roman Legion workshop did not stamp its products prior to the foundation of Aelia Capitolina c. 130 C.E. and started doing so in connection with that event.

Another special find, clearly of a military nature, are three bread stamps used by the Roman soldiers. Two were complete and another one was broken. All were inscribed with: (1) the symbol for a Roman military century followed by (2) the centurion’s name and then (3) the name of the soldier in charge of baking for this particular army unit. Here is the text of one such bread stamp (studied by Leah Di Segni):

CASPE CANIN
(Centuria) Caspe(rii) (Opus) Canin(ii)
“(Century) of Casperius, (Work) of Caninius”

Quantities of animal bones (studied by Liora Kolska-Horwitz) were also recovered from the dump beneath the cardo’s eastern colonnade. Pigs 047 constitute the most common species, representing over 60 percent of the bones. This, of course, is quite unusual for Jerusalem, especially if compared with Second Temple period Judean sites (pre-70 C.E.), where no pig bones have been found. Kolska-Horwitz concludes that this high frequency of pig bones may thus be considered a hallmark of the Roman army’s dietary debris.

In addition to the wheel-made oil lamps with spatulate nozzle mentioned above, the dirt and dump fills contained other Jewish artifacts from the Second Temple period (together with the Roman, military artifacts) such as chalk vessels, which are not subject to impurity.^b These remains either come from the rich Jewish culture before 70 C.E. or they reflect the life of Jews who continued to live in the city until the Romans expelled them after the Second Jewish Revolt (132–135 C.E.), the so-called Bar-Kokhba Revolt.

In 130 C.E. the emperor Hadrian visited the eastern part of the Roman Empire on a royal tour. Around the time of this visit, Hadrian gave Jerusalem a new name, Aelia Capitolina. (Aelius was one of Hadrian’s own names; Capitolina refers to the triad of gods of the Capitolium, the main temple of a Roman city: Jupiter, Juno and Minerva.) With this renaming of the city, it attained the status of a Roman colony. Exactly when this occurred is a matter of scholarly disagreement. Some (relying on the writings of the Roman historian Dio Cassius) contend that the name change, and the beginning of the construction of Aelia Capitolina, occurred in connection with Hadrian’s visit in 130 C.E., perhaps even setting off the Second Jewish Revolt. Others (relying on the writings of the church father Eusebius) say this name change occurred only after the Second Jewish Revolt was suppressed in 135 C.E.

Our conclusion is somewhat different. Based on finds beneath the cardo, we can date the eastern thoroughfares of Aelia Capitolina to the early years of Hadrian’s reign, probably in the 120s, long before his famous visit to the east in 130 and the “official” founding of Aelia Capitolina. Whether Hadrian did this to restore the city to its previous splendor after the Roman destruction of 70 C.E. or due to some other reason is a question that remains to be explored, but it is clear to us that when he was appointed emperor, or soon after that, he had already begun the planning and rebuilding of Jerusalem as a Roman city.

The eastern cardo continued in use with no significant changes in the Late Roman and Byzantine periods (second–sixth centuries C.E.). The width and level of the street remained the same despite occasional repairs. One such repair is a sixth-century pavement of the eastern colonnade with a white mosaic floor decorated with small flowers.

The famous Madaba mosaic map of the sixth century (located in a church in Madaba, Jordan) pictures two colonnaded streets across the city of Jerusalem. Both start from a plaza inside the northern, Damascus Gate. The western cardo is shown with a pillared colonnade on both sides of this straight north-south route, while the eastern cardo is shown narrower, with a colonnade on only its eastern side. In the Byzantine period, when the city was Christian, the western cardo was no doubt the most important thoroughfare in the city, passing in front of the Holy Sepulchre Church (the most prominent building on the map, marking the tomb of Jesus) and extending to the Nea 069 Church in the south. By this time, the eastern cardo (the one we excavated) had most likely become less important. Nevertheless, the eastern cardo was just as wide as the western cardo, as we know by a comparison of our results with a long section of the western cardo revealed in Nahman Avigad’s excavation in the Jewish Quarter.^c

One final observation: the direction of the eastern cardo is aligned quite precisely with the western wall of Herod’s Temple Mount, today’s Western Wall. This wall is generally thought to proceed in a north-south direction. But it actually runs in a slight north-northwest to south-southeast direction, and so does the Roman eastern cardo, 300 feet to the west. Moreover, two side streets run from the eastern cardo toward the Temple Mount. This street layout suggests that the Temple Mount formed an integral part of Aelia Capitolina.

Until now, scholars have questioned the role of the Temple Mount in the Roman city: Was there a Temple to Jupiter there in the Roman period? Did statues of the Capitoline deities—Jupiter, Juno and Minerva—adorn the former Temple Mount? Or was it left empty to demonstrate Roman might? The fact that this main street was exactly aligned with the Temple Mount suggests that there must have been something on the Temple Mount in the Roman period—but what?

It is often supposed that the Temple Mount was empty in the Byzantine period, based largely on the fact that Byzantine pilgrim texts do not record Christian visits to the Temple Mount. And it is not portrayed on the Madaba mosaic map either. Our finds, however, suggest that the area around the Temple Mount did not change much when Christians controlled the city. Perhaps, as one text has it, it served as a garbage dump in Christian times.⁶ What, if anything, was on the Temple Mount in the Byzantine period remains a puzzle.

Gradually, at various stages during the subsequent Islamic period, the level of the street rose by more than 12 feet above the Roman cardo. And its overall width narrowed—from 35 feet in the Roman and Byzantine periods to 16 feet in the Early Islamic period. Then, as structures continued to be built along the street, it narrowed further to 13 feet. In the Ottoman period, the street became even narrower—as little as 8 feet in places.

Nevertheless Rehov Ha-Gāy/El-Wad (“the valley street” in Hebrew and Arabic) still follows the line of the eastern cardo. It remains one of the principal arteries in the Old City of Jerusalem, leading from 070 Damascus Gate in the north to the Dung Gate in the south.

If you go to the famous Western Wall in Jerusalem, which is actually the western retaining wall of Herod’s Temple Mount and Judaism’s holiest prayer site, and then turn around, you will see at the other side of the plaza an area of less than half an acre that has recently been excavated. Large-scale archaeological excavations were conducted here between 2005 and 2010 on behalf of the Israel Antiquities Authority, initiated and underwritten by the Western Wall Heritage Foundation and directed by the authors of this article.1 What we have found sheds light on important transitional phases in Jerusalem’s […]