Mizpah: Newly Discovered Stratum Reveals Judah’s Other Capital

028

I believe I have succeeded in identifying substantial archaeological remains from a period that is almost an archaeological blank in the history of ancient Israel—the period of the Babylonian Exile, when according to tradition, the Judeans were deported to Babylonia. All the more remarkable, I have been able to identify these remains without sinking a spade into the ground or lifting a pick.

This is the tale of my excavation into the field records of a dig that ended more than 60 years ago. Of course my own excavation, so to speak, has only been possible because the actual excavator of the site left behind such meticulous records.

William Frederic Badè excavated at Tell en-Nasbeh, just north of Jerusalem, for five seasons between 1926 and 1935 in the belief, very probably correct, that this site was Biblical Mizpah in Benjamin. The American archaeologist cleared almost two-thirds of the eight-acre site, making it the most completely excavated site in Palestine at the time. As 030work progressed, Badè employed more than 150 men, women and children who ultimately excavated 672 rooms and 387 cisterns, silos and other installations. Outside the city walls they excavated 71 tombs.

To his everlasting credit, Badè was a leading proponent of careful and complete recording.^a Over 23,000 objects were described, of which 15,000 were drawn. Many walls, structures, rooms and tombs were carefully photographed. Although Badè’s excavation method was weak stratigraphically, he was careful to record the location of each find. He clearly understood the importance of recording materials in situ.¹

Badè published a number of preliminary reports. Unfortunately, he died of overexertion in 1936, without publishing a final report. His close colleague, Chester C. McCown, and his dedicated chief recorder, Joseph C. Wampler, undertook the task, however, eventually publishing two hefty volumes that are still considered reasonably good reports for their time. McCown’s volume covers the archaeology and history;² Wampler’s, the pottery.³

My own involvement with the site began in 1986, when I became coordinator of the Badè Institute of Biblical Archaeology, part of the Pacific School of Religion in Berkeley, California. My duties as coordinator included giving tours of the institute’s small but very modern museum.⁴ To prepare myself, I read everything I could find on the site. Poring over the final report, I became more and more intrigued. Although I realized that in many ways the report, dating to 1947, was incomplete and outdated, I was drawn into a detailed study of the stratigraphy of the site proposed by the excavators. In the end, I found an occupation layer, or stratum,^b that had apparently completely escaped Badè, as well as McCown and Wampler.

Tell en-Nasbeh lies about 8 miles northwest of Jerusalem, on the southern outskirts of modern Ramallah. Its identification as Biblical Mizpah is based largely on geography. The only other serious contender is (or was) Nebi Samwil (about 5 miles northwest of Jerusalem), a site favored by the greatest of an earlier generation of Biblical archaeologists, William Foxwell Albright.

Mizpah figures prominently in the Biblical accounts of the period of the Judges, which corresponds to what archaeologists call Iron Age I (1200–1000 B.C.). The Israelite tribes convened at Mizpah to avenge the rape and murder of the Levite’s concubine at the site called Gibeah of Benjamin (Judges 19–20). Samuel, the prophet and seer, judged the people at Mizpah (1 Samuel 7:6). As a circuit-riding judge, Samuel held court at Mizpah (1 Samuel 7:15–16). The Israelite tribes mustered here for battle with the Philistines (1 Samuel 7:5–6), and Israel’s first king, Saul, was presented to the people at Mizpah (1 Samuel 10:17–24).

If a site is to be identified as Mizpah, it must have remains from Iron Age I. By this test, Nebi Samwil, the only alternative to Tell en-Nasbeh, fails. A survey of Nebi Samwil yielded no Iron Age I pottery.⁵ Moreover, the recent excavation of Nebi Samwil, though still unpublished, does not seem to have produced Iron Age I remains or even any substantial 031material from the Iron Age II through the Persian period, when according to the Bible, Mizpah was flourishing.

Tell en-Nasbeh thus remains the best candidate for Biblical Mizpah.

As I studied the careful site plan (1:400) in the published final report and the even more detailed plans (1:100) in the Badè Institute archives, I was struck by certain anomalous structures. As at many hill-country sites, Tell en-Nasbeh’s houses were arranged to follow the natural contours of the hill on which they were built. The rear walls of the outermost buildings were connected and faced out from the settlement. Up the slope, facing this first band of houses across a narrow alley, was a matching band of houses. The houses of this second band shared a rear wall with still a third band of houses. This curvilinear system of houses and streets is called a “ring-road” plan. The ring-road arrests the eye of anyone examining the 1:400 plan (see Mizpah: The Missing Years). In my revised stratigraphic scheme for Tell en-Nasbeh, this ring-road phase is designated Stratum 3.⁶ Only when studying the unpublished 1:100 plans in the archives, however, do the anomalies I mentioned earlier spring out. About 16 buildings were constructed over the ring-road town. The remains indicated on the plans were not just fragments of walls but structures larger than any of the buildings of Stratum 3. I have designated the stratum that sits over Stratum 3 as Stratum 2.

In the plan (see Mizpah: The Missing Years), Stratum 3 is in black, and Stratum 2 is in red. Building A on the plan is a typical four-room house, consisting of three long rooms and one broad room across the three long rooms. Building A was clearly built over the back wall of a Stratum 3 building along the periphery of the town.

Building C, another four-room house, sits directly in front of the entrance to the inner gatehouse of Stratum 3. This is certainly an odd place to put a residence: It must date later than the Stratum 3 gatehouse.

A third four-room house, labeled Building B on the 034plan, sits over the remains of a wall that once connected the outer gate of Stratum 3 with the inner gate.

On the plan, I have made it as easy as possible to distinguish between the buildings of Stratum 3 and the four-room houses that cut across Stratum 3 (Building A and Building B) or block a Stratum 3 building (Building C). Detecting these structures in the original plans, however, was not so easy. Nevertheless, having identified these buildings, I knew there had to have been a previously unrecognized stratum above Stratum 3, what I call Stratum 2.

These three buildings provided the first clues to the town’s post-ring-road phase. As I continued to study the plans, I was able to identify four more four-room houses in Stratum 2.⁷ At this point I felt I was on to something solid. My search widened as I looked for other building remains from my newly found Stratum 2. I was ultimately able to identify at least nine more buildings as probably belonging to Stratum 2.

There are several constructional differences between the Stratum 2 and Stratum 3 buildings. The houses of Stratum 3 are quite small (averaging 700 square feet); even the largest four-room houses of Stratum 3 average only 860 square feet. In contrast, the four-room houses of Stratum 2 average 1,430 square feet.

Most of the Stratum 3 walls (especially the earliest ones) are only a single stone wide; the Stratum 2 walls, however, are a mix of large single stones (especially at corners) and sections of double-stone construction.

Stone-paved floors are more common in Stratum 2 than in Stratum 3, as are monolithic pillars. In Stratum 3, only one out of six buildings with pillared walls has pillars made of single large stones; the pillars in the other five buildings are made of stone drums, which are less costly and less stable. In Stratum 2, however, all the pillars are monoliths.

The houses in Stratum 2 are not only more spacious and better constructed than those of Stratum 3, but as the plan shows, they also stand further apart. In short, there was less urban crowding in Stratum 2. These differences confirm the existence of a separate, additional stratum.

I am not sure why these buildings were not recognized as part of a separate stratum when the final report was prepared in 1947. The cramped design of the published 1:400 plan probably kept later scholars from recognizing the occupation layer immediately above Stratum 3.

Having identified a distinct new stratum with the aid of the unpublished plans, I then faced the task of dating it. The best way to do this would have been to study the pottery. But at the time, I was working on my dissertation and was eager to finish my research. A thorough ceramic examination would have taken several years, so I took a bit of a short cut. It was easy to date Stratum 5 to the Early Bronze Age I; the 035unmistakable, rough, handmade pottery of that era allowed me to date that stratum to around 3100 B.C. For the later strata, I decided to construct a model of the site’s settlement history based on the Biblical sources and to see how well the excavated materials could be integrated with this historical model. The fit between the historical sources and archaeological material was extremely good. After the Early Bronze I period, the site was abandoned until about 1200 B.C., the beginning of Iron Age I (1200–1000 B.C.), the period of the Israelite settlement. This stratum (Stratum 4) included numerous easily identifiable Iron Age I pottery forms, such as collar-rim jars and Philistine bichrome ware, which attested to a major settlement at the time. This clearly conformed to the Biblical account.

For most of Iron Age II (1000–586 B.C.), the period after the United Monarchy under David and Solomon split into two kingdoms, Mizpah was an important fortified town on the border between Israel in the north and Judah in the south. The Bible describes how the town came to be fortified. According to the books of Kings and Chronicles, Baasha, ruler of the northern kingdom of Israel from 906 to 883 B.C., invaded Judah as far south as Ramah, which he proceeded to fortify. In response, Asa, king of Judah from 908 to 867 B.C., sent a gift of silver and gold to the king of Damascus to induce him to attack Israel from the north. When he did so, Baasha discontinued his fortification efforts and apparently withdrew from Judah. Asa then took the stones, timber and other supplies that Baasha had assembled at Ramah and used them to fortify Mizpah and another site, thus protecting his northern border (1 Kings 15:16–22; 2 Chronicles 16:1–6). Based on these passages, we would expect to find a fortified occupation at this time at Biblical Mizpah. That is exactly what we do find in Stratum 3: A strong offset-inset wall (a major feature of the 1947 report on Badè’s excavation) protected the ring-road town.^c The massive wall was built around the pre-existing settlement, whose original defenses consisted simply of the linked back walls of the houses along the periphery of the town—a common form of defense in Iron Age I and early Iron Age II. The new offset-inset wall was built a few meters downslope from the earlier settlement so that its construction would not require the destruction of any existing homes. The new wall was, on average, 15 feet thick—wider than the contemporaneous offset-inset city wall at the major site of Megiddo.

Most scholars identify the offset-inset wall at Tell en-Nasbeh as King Asa’s wall, mentioned in the Bible (although the pottery excavated from cuts through the wall—which would firmly date the wall—was never published and appears to have been lost). In short, the ring-road city with the offset-inset wall was the Iron Age II city, Stratum 3. My newly found Stratum 2 lay above this.

Let’s skip Stratum 2 for the moment. We would expect some occupation in the Hellenistic period (332–37 B.C.) because Mizpah is mentioned in 1 Maccabees 3:46 as the site where Judah Maccabeus’s forces gathered before a battle with the Seleucids. Indeed, a Hellenistic coin was found in the foundation trench of a Stratum 1 wall.

So Stratum 2 had to fall between Iron Age II (Stratum 3) and the Hellenistic period (Stratum 1). The intermediate interval includes the Babylonian and Persian periods. The Babylonian period began with the Babylonian destruction of Jerusalem in 586 B.C. and ended in 539 B.C., when the Persians assumed hegemony over virtually all the ancient Near Eastern world and the Persian king Cyrus the Great permitted the exiles to return and rebuild their Temple and their land. Persian rule ended 036with the conquest of Alexander the Great in 332 B.C.

According to the Bible, after the destruction of Jerusalem, the Babylonians appointed Gedaliah as the new puppet ruler of Judah. He governed not from Jerusalem, however, but from Mizpah (2 Kings 25:22–25; Jeremiah 40–41). Gedaliah lived there with all the trappings of a de facto king: eunuchs, royal princesses and a detachment of Babylonian soldiers as a bodyguard. Gedaliah was murdered, however, by a disaffected member of the Davidic royal line, who no doubt felt he should have held Gedaliah’s position. When world hegemony passed from the Babylonians to the Persians in the sixth century B.C. and the exiles were permitted to return, Mizpah remained an important administrative center; several sections of Jerusalem’s wall were rebuilt by the men of Mizpah and their rulers (Nehemiah 3).

One would expect that at this time Tell en-Nasbeh would have been transformed from the border fortress created by Asa to a Babylonian administrative center. If so, the plan of the site would doubtless have changed to include a residence for the governor, houses for officials, and storage depots. The buildings of Stratum 2 match this description perfectly.

This wasn’t the best way to date Stratum 2, but it served my purposes at the time. Several years later, I was able to undertake a more thorough study of the pottery, which confirmed my earlier stratigraphic hypothesis.⁸

I found it easier to date the end of Stratum 2 than the beginning. On the floor of one of the four-room houses of Stratum 2 (Building B) were some storage jars that had been crushed, apparently in the destruction of the Stratum 2 city. (An exact parallel to one of the storage jars, found at Tel Michal on the Mediterranean coast, dates to the end of the fifth century B.C.)⁹ This dates the destruction of Building B, and thus probably all of Stratum 2, to about the end of the fifth century. This is in accord with some imported Greek pottery found at the site, the latest of which also dates to about the end of the fifth century. Who destroyed the Stratum 2 site remains a mystery; there are no ancient literary references. It was probably connected in some way with the political turmoil that rocked the Persian empire in the late fifth and early fourth centuries B.C.

The date of the beginning of Stratum 2 can be estimated from three huge sack-shaped pithoi found on the floor of one of the four-room houses. These large vessels are known only from Jerusalem, Lachish, Arad and other sites destroyed in the Babylonian invasion of 586 B.C.¹⁰ Generally dated to the end of the seventh or early sixth century B.C., these vessels probably remained in use through the Persian period. Such enormous vessels were seldom moved once put into position and filled. Since they remained stationary, there was very little chance for them to break, except from deliberate destruction or through a major accident, such as an earthquake; examples of pithoi like this lasting more than a century are known from Cyprus.¹¹ These pithoi were apparently installed soon after the construction of the Stratum 2 city in the early sixth century B.C. and continued in use until the end of Stratum 0372 around 400 B.C. Stratum 2 thus begins shortly after 586 B.C. and ends about 400 B.C.

Once Stratum 2 was dated with some confidence, I could place numerous unstratified artifacts into this stratum. A delicate onyx seal, found in a tomb that otherwise only contained Byzantine material, bears the inscription “(Belonging) to Ya’azaniah the servant of the king.” The seal may have belonged to an officer named Ya’azaniah who, according to 2 Kings 25:23 and Jeremiah 40:8, came to the Babylonian-appointed ruler Gedaliah at Mizpah after the fall of Jerusalem. At the bottom of the seal is one of the earliest known depictions of a rooster from ancient Israel.

A fairly uncommon stamp impression—found on jar handles and occasionally on the sides of vessels from the site—probably dates to the Babylonian phase of Stratum 2 as well.¹² These impressions bear the Hebrew letters M(W)S|H (Motzeh) and most likely originated from estates surrounding the village of Mozah, southwest of Mizpah. The Mozah estates may have produced the wine stored in these stamped jars for use in the courts at Mizpah. Of the 42 known M(W)S|H impressions, 30 come from Tell en-Nasbeh, which suggests that this site was a center for the governmental production and distribution of these jars and their contents. The limited distribution of the stamp impressions (about 18 miles east-west and 12 miles north-south) may mark the approximate limits of the area administered by the Babylonian governor Gedaliah.

The 24 Yahud (YHWD) stamp impressions found at the site also undoubtedly belong to Stratum 2: Yahud was the name of the province in the Persian period.

038

Several somewhat surprising conclusions can be drawn from this study of Stratum 2. The first is the obvious prosperity of Mizpah throughout the Babylonian Exile and the following Persian period. Writing from the perspective of the exiles, the Biblical authors indicate that during the Babylonian period, everyone had been exiled except “the poorest of the land to be vinedressers and tillers of the soil” (2 Kings 25:12). That some of the wealthy remained is now clearly demonstrated not only by the evidence from Tell en-Nasbeh, but also by Gabriel Barkay’s excavation of an elaborate tomb used by a prosperous Jerusalem family throughout the Exilic period.

As often happens, archaeology shows us we must read the Bible more carefully. According to Ezra 2:1–63, many of those who returned from the Exile settled in the area of Benjamin, which is approximately the same area that I have suggested was administered by Gedaliah. That so many of the returning exiles settled in Benjamin presupposes some sort of infrastructure there to absorb them. Gedaliah’s choice to settle in Mizpah (Jeremiah 40:6, 10) further suggests that this area was in one piece at the time of the Exile. Hebrew University scholar Abraham Malamat long ago suggested that the Babylonians had not devastated Benjamin as they had Judah.¹³

Perhaps the continuity of Jewish life in Judah during the period of the Babylonian Exile is reflected in the somewhat surprising use of Iron Age building-forms in the Persian period: Four-room houses were still in use only a couple of generations before the birth of Alexander the Great (356 B.C.). In short, archaeologically, the Babylonian period is a continuation of the Iron Age,¹⁴ which is understandable now that we know the Babylonians did not exile the entire population of Judah. Unfortunately, the other major late Iron Age sites in this area—el-Jib (Gibeon), Bethel, Tell el-Ful, Anathoth, Ramah and Nebi Samwil—have been poorly excavated or poorly published, or remain unexcavated. If the Babylonians left this area in peace, these sites should yield a Babylonian era occupation parallel to Tell en-Nasbeh. Until these sites are either excavated or reexamined, Tell en-Nasbeh will remain our primary example of the material culture of the Babylonian period.

The archaeological evidence does reflect some Mesopotamian influence during the Exile. Fragments of a Mesopotamian-style bathtub-shaped ceramic coffin (and possibly the remains of two others) were found 066at Tell en-Nasbeh.¹⁵ These coffin fragments come from the tell, not from the cemeteries in the surrounding hills. In Mesopotamian practice, the deceased were buried below the floors of their dwellings; Judean burials were traditionally outside the city wall. These coffin fragments at Tell en-Nasbeh fall most naturally in the Babylonian period. They indicate either that Babylonian officials were buried here or that Judeans chose to copy Babylonian fashion.

Perhaps the most intriguing find from Tell en-Nasbeh—probably belonging to Stratum 2—is a fragment of a slender bronze circlet bearing a cuneiform inscription that originally contained 30 to 35 characters. Only 11 characters are preserved. Nothing indicates the function of the object it was attached to, or how it was attached. Since only a small part of the inscription survives its translation is problematic. It may have read “ … Ayadara, king of the world, for (the preservation of) his life and … ” This is clearly a dedicatory inscription of sorts, but the words indicating what is being dedicated, and to whom, have been lost. Even the identification of Ayadara is unknown; no one with his name bearing the title “king of the world” is known from any period. What is remarkable is that such a dedicatory inscription should turn up on a small tell in ancient Judah. Was it dedicated to a Mesopotamian deity who was worshiped by a Babylonian stationed at Mizpah? Was it dedicated to the god of Israel? One thing is clear: At least an echo of Mesopotamian cultic life reverberated in the hills of Judah.

That so much has been recovered by sifting through the old records of a site excavated so long ago bodes well for this kind of research in the future. Clearly the time has come to reopen the files of other excavations. Perhaps they too will enable us to fill in the gaps in our understanding of Israel’s glorious past.

I believe I have succeeded in identifying substantial archaeological remains from a period that is almost an archaeological blank in the history of ancient Israel—the period of the Babylonian Exile, when according to tradition, the Judeans were deported to Babylonia. All the more remarkable, I have been able to identify these remains without sinking a spade into the ground or lifting a pick. This is the tale of my excavation into the field records of a dig that ended more than 60 years ago. Of course my own excavation, so to speak, has only been possible because the actual […]