The Mistress of Stratigraphy Had Clay Feet, Hershel Shanks, BAR 29:03, May-Jun 2003.

Bible and Beyond
Kathleen Kenyon’s Flawed Jerusalem Excavation

Click here to view the original article.

Excavations by Kathleen M. Kenyon in Jerusalem 1961–1967, Volume III—The Settlement in the Bronze and Iron Ages

M.L. Steiner

(London- Sheffield Academic Press, 2001) 158 pp., $105, hardback

Much as armies through the centuries have battled over Jerusalem, so scholars have fought over its history. What, for example, was Jerusalem like in the tenth century B.C.E., a period that the Bible describes as the time of Israel’s glory, the age of the United Monarchy, when David and Solomon ruled over Israel and Judah? Despite the Biblical picture of a period of wealth and power, the archaeological evidence from that time is sparse. Some so-called Biblical minimalists have questioned the very existence of the United Monarchy and even of David and Solomon.

Between 1961 and 1967 Dame Kathleen Kenyon, one of the leading archaeologists of her day, led an important excavation in the area of Jerusalem known as the City of David, where the original city was located. She wrote two popular accounts of the work she did there,1 but, unfortunately, she died in 1978 without publishing her final report. In 1982, a committee from the British School of Archaeology in Jerusalem assigned the task of completing a critical part of Kenyon’s excavation report to H.J. Franken of Leiden University in Holland. He, in turn, assigned the stratigraphy and pottery analysis to Margreet Steiner, his assistant, who produced the volume under review. With glee she leapt at the opportunity, although she now ruefully writes, “It is a good thing I didn’t know then that it would take me almost 20 years to finish the project.”2

One’s first impression of this short book (the actual text, not counting appendices, is only 115 pages, more than 65 of which are taken up by pictures, plans and drawings) is of the censorious light it casts on the great British archaeologist who led the excavation. Steiner “wonder[s] how Kathleen Kenyon would have responded to [this] work,” adding, “I hope she smiles upon this book.” If Kenyon is indeed looking down from her perch in heaven, she must be furious, for this book documents how incomplete, illegible, sloppy and just plain wrong her excavation records are.

Kenyon has long been known as the “Mistress of Stratigraphy” for her careful, painstaking excavation methods and record keeping, and for having continually drawn cross-sections (known as sections) to preserve a progressive picture of the strata within an excavation square. But some of her field notebooks, we learn from Steiner’s volume, are nearly useless because they don’t contain the locations of the layers that Kenyon excavated. Deposit numbers and levels are missing from section drawings, making them difficult, if not impossible, to use. Directional arrows on some of the plans point south instead of north.

Steiner attributes these omissions and mistakes to “slovenliness,” adding, however, “that there were more fundamental errors.” Kenyon waited until the end of each excavation season, after the field supervisors had already left, to make section drawings (she drew the main sections herself), weakening the connection between her drawings and the data in the field. Plans, too, were drawn only at the end of the season—by a surveyor who plotted the location of every stone without having any notion if those stones had once been part of walls or if they were merely rubble.
It goes on. The pottery was noted on “pottery phasing cards,” so that the pottery could be connected to the various phases of each layer or stratum. But often when the phasing was changed, no correction was made on the cards. “This made the use of the phasing and pottery cards for the study of the pottery almost impossible,” Steiner writes. She decided simply to “ignore these systems.” Furthermore, Kenyon’s own handwritten notes are “largely undecipherable.”

Early Bronze Age 3300–2200 B.C.E.

Middle Bronze Age 2200–1550 B.C.E.

Late Bronze Age 1550–1200 B.C.E.

Iron Age I 1200–1000 B.C.E.

Iron Age II 1000–587 B.C.E.

Kenyon’s conclusions are also often wrong. For example, a house Kenyon attributed to the Middle Bronze Age (2200–1550 B.C.E.) was really built over half a millennium later, in the Iron Age (1200–587 B.C.E.). A pottery group that was supposed to be “all M[iddle] B[ronze]” actually included Early Bronze (3300–2200 B.C.E.) pottery.

Worst of all, most of the pottery sherds, including diagnostic handles, bases and rims, were simply thrown out. Kenyon saved only “small amounts.”

Steiner has re-analyzed the entire stratigraphy of the site. But how much confidence can we have in the re-analysis, and particularly the conclusions, when the basic data are so shaky?

I am also a little uncomfortable with Steiner’s consistently negative conclusions based on the absence of evidence. For example, she determines that “Jerusalem was not permanently occupied in the Early Bronze Age.” Yet Kenyon found some pottery from this period, as have other excavators before and after her, in cracks and caves. Kenyon herself excavated an Early Bronze Age burial place on the Mount of Olives. Israeli archaeologist Yigal Shiloh, who followed Kenyon in excavating the City of David, even found two Early Bronze Age buildings near the Gihon Spring, the settlement’s water supply. “One would suppose,” Steiner concedes, “that Jerusalem, with its rich spring and fertile wadis, offered an excellent location for a permanent village.” Yet Steiner refutes this possibility simply because more evidence hasn’t been found—and in this way she does not allow for what may still lie underground or what may have been destroyed by later settlements. All this may not seem important to people interested in the Biblical periods, but it is indicative of a mindset.

In the Middle Bronze Age II period (1800–1550 B.C.E.), Jerusalem was protected by a very substantial wall (about ten feet thick), stretches of which were found by Kenyon (42 feet) and later by Shiloh (88 feet). The existence and date of this wall are among the few matters on which there is widespread agreement. According to an appendix to the Steiner volume by Diny Boas-Vedder, it probably took about two-and-a-half years to build. Recently, excavations by Israeli archaeologists Ronny Reich and Eli Shukron found more stretches of this wall, together with imposing towers that were built around the Gihon Spring to protect the settlement’s water supply.

The settlement enclosed by this fortification wall raises a major question for those who would conclude that there was a substantial settlement in Jerusalem during the time of David and Solomon despite the fact that so little has been found from their time- If so much has been found from the much earlier Middle Bronze Age, why hasn’t more been found from the tenth century B.C.E., the time of David and Solomon?

The answer of the nay-sayers is simply that no significant settlement existed here in the tenth century B.C.E. The argument on the other side is that Kenyon did not find that much, either, other than the fortification wall, in the Middle Bronze Age settlement—no domestic structures, no public structures, indeed no buildings at all. Yet here, because of the existence of the wall, Steiner is willing to assume that “a large part of available space inside the fortifications will have been filled by administrative buildings and storage facilities. The existence of one or more temples and several open markets may be assumed as well.” If asked to explain why she was willing to make these assumptions, Steiner would reply that, in addition to the wall, a lot of Middle Bronze Age pottery has been found in the excavation, two-thirds of which is represented by storage jars (implying storage facilities) and only one-third by cooking pots. It seems to me that she might as easily have concluded that domestic buildings also existed here, based on the cooking pots. Or she could conclude that the fortifications and quite remarkable water system would imply an administration and organization that would have to be housed in public buildings. In any case, in the Middle Bronze Age, Steiner is willing to imply structures from pots, but not in the Early Bronze Age.
Readers can decide who has the better side of this argument and what relevance it has for the Jerusalem of David and Solomon.

Whatever settlement existed here in the Middle Bronze Age was, according to Steiner, soon abandoned. “For reasons unknown, the town ceased to exist after a mere hundred years.” Steiner believes that Jerusalem was unoccupied until about the beginning of the 12th century B.C.E. This is the time of transition to the Iron Age. (The Late Bronze Age gave way to Iron Age I, sometimes called the Early Iron Age, in 1200 B.C.E.) In short, no settlement in the Late Bronze Age, says Steiner.

It is true that very little from the Late Bronze Age (1550–1200 B.C.E.) has been found in Jerusalem. Steiner argues, “Not only architecture, but also pottery and other small finds are missing.” But this is not quite true; there is some pottery from the Late Bronze Age, though not very much. This paucity of Late Bronze Age evidence, however, has been used by those who argue in favor of a substantial city here at the time of David and Solomon. Although the absence of archaeological evidence in the 14th century B.C.E. would suggest the absence of a settlement, we know that there was indeed a settlement in Jerusalem at this time, one ruled by a king. We know this from the Amarna letters, a hoard of cuneiform tablets found at Tell el-Amarna in Egypt consisting of diplomatic correspondence to and from pharaohs Amenophis III and his son Akhenaten (Amenophis IV), dynasts who reigned in the 14th century B.C.E. Seven of these letters are from Abdi-Heba, ruler of Jerusalem. So there must have been a settlement here in the 14th century B.C.E. despite the lack of archaeological evidence for such a settlement. In one of the letters Abdi-Heba even speaks of another “town” that belongs to Jerusalem. Some scholars have argued that if Jerusalem was a city-state in the 14th century B.C.E., as the Amarna letters show—despite the paucity of finds from the 14th century B.C.E.—the same may be true of the tenth century B.C.E., the time of David and Solomon. We just haven’t found much of that city. So the argument runs.

Steiner is not convinced, however, that a substantial settlement existed at Jerusalem in the 14th century B.C.E. For one thing, in the cuneiform of the Amarna letters, the city is spelled Urusalim. Maybe this isn’t Jerusalem, she suggests.3 But even if it is, she argues, Jerusalem itself is never referred to as a “town.” It could be “an estate, a royal dominion,” most likely of the Egyptians. Perhaps, she suggests, it was garrisoned by mercenaries, the very people referred to in the Bible as the Jebusites, from whom David captured the place in about 1000 B.C.E. Abdi-Heba would then simply be the “manager of the estate.” In two letters, he refers to himself as “a soldier for the king.” According to Steiner, “Abdi-Heba is not so much a local prince as an Egyptian official managing a royal estate for the pharaoh.” Jerusalem was simply a “fortified house” in a “rather marginal area, where the pharaoh stationed a military commander to protect the route to Beth Shan, then an important Egyptian garrison town, as well as the route to Moab.” Steiner is absolutely sure of her final analysis- “The conclusion that no ‘city’ existed in Jerusalem during the Late Bronze Age seems unavoidable to me.”

Steiner has made these same arguments in the past, and Israel’s leading authority on the Amarna letters, Nadav Na’aman, has responded to them on at least two occasions, once in BAR.a Na’aman notes that in the Amarna letters, Abdi-Heba is referred to as hazannu, which is the title that all the local rulers in the land of Canaan bore. Unless there were no local rulers anywhere, and only Egyptian estate managers, Abdi-Heba was the ruler of Jerusalem. Moreover, Abdi-Heba took his title as a right of dynastic ascent, although doubtless with the support of the Egyptian pharaoh. As Abdi-Heba writes to the pharaoh, “The strong arm of the king brought me into my father’s house.” Moreover, Abdi-Heba refers to his “house” and to another “house,” in which 50 Egyptian soldiers were temporarily garrisoned. In addition, Abdi-Heba sends exceptionally rich caravans to the pharaoh, including 5,000 objects (probably silver shekels) and also prisoners to be used as slaves.
There is more- In cuneiform, signs called determinatives (with no phonetic value) tell you what kind of a word they are attached to. A number of different determinatives are used in the Amarna letters to signify that various sites were towns. Jerusalem, too, is frequently designated with one or more of these determinatives. Moreover, Abdi-Heba was apparently quite a mover and shaker. In one letter, he is compared to the infamous king of Shechem, Lab’ayu, who was a prominent statesman and leader bent on military expansion. All this, Na’aman argues, suggests that Jerusalem was much more than a royal estate managed by an Egyptian-appointed steward.b

Strangely, Steiner does not even attempt to answer Na’aman’s arguments in any detail. Although in her bibliography she does list his articles interpreting the Amarna letters, she does not cite them in her text (at least as far as I can find; she cites only another Na’aman article, on the proposition that Egypt had grain storage facilities in the Jezreel Valley). She does concede, as noted above, that one Amarna letter speaks of another town that belongs to Jerusalem, but, for her, that is not enough to indicate that Jerusalem was a city-state at this time.

In about 1200 B.C.E. a major change occurred in Jerusalem. This is the time of the beginning of the emergence of Israel in Canaan. In Biblical terms, the period between 1200 and 1000 B.C.E. is the period of the Judges. This is followed by the United Monarchy of David and Solomon, beginning in about 1000 B.C.E.

Steiner dates Jerusalem’s Stepped Stone Structure—a mammoth installation (“by far the largest and most impressive structure of its kind”)—to 1200 B.C.E., the beginning of the period of the Judges. Kenyon and Shiloh, who both excavated the structure, suggest different dates. The Stepped Stone Structure is built against the eastern slope of the City of David above the Kidron Valley and consists of at least seven terraces. Kenyon thought terraces like these covered the entire east slope of the hill, but, as Steiner notes, in this she was wrong. Rather, the Stepped Stone Structure “must have looked like a stepped pyramid, its ‘steps’ descending from the top of the hill.” More than 65 feet in height and twice that in width, it represents what remains of the base of the structure that was built on top—probably some sort of fortress. An outpost from which to guard the approach may have been set up on Terrace 3, the only one wide enough (25 feet) to accommodate a building.

Doesn’t this huge structure imply a quite substantial settlement in the 12th century B.C.E.? This thought also occurred to Steiner- “The question must be raised why such an enormous task was undertaken.” She has an answer, however- This massive structure was constructed to fill in a gully. That, she says, is her “inevitable conclusion.” But why did the 12th-century B.C.E. inhabitants of Jerusalem want to fill in a gully? To answer that question, she concedes, would involve pure “speculation.” Perhaps, she says, this area was the northern defense line of the city. The settlement itself “consisted of only one large building.” Steiner’s confidence is hardly justified.
Steiner’s bottom line on the tenth century B.C.E., the time of David and Solomon and the United Monarchy, is somewhat left of center. She disagrees with the minimalists. But she also finds the Biblical characterization a gross exaggeration- The Jerusalem of David and Solomon was “neither a large city nor a small provincial town. It rather seems that Jerusalem was an administrative centre of at least regional importance, the newly built capital of a newly established political entity.” Elsewhere she calls it “a small fortified town.”

One serious problem in identifying tenth-century B.C.E. material is that it is often indistinguishable from ninth-century B.C.E. material. As Steiner tells us, “Many [pottery] shapes were produced over centuries.” Moreover, “The introduction of a new type [of pottery] took place in a centre, from where it dispersed slowly. Thus, a type may have occurred in one place in great quantities, while it was still unknown at other sites.” Add to this the fact that one leading Israeli archaeologist, Israel Finkelstein, has argued for down-dating traditional tenth-century B.C.E. pottery to the ninth.c In these circumstances, it would seem speculative to date Jerusalem pottery conclusively to the tenth century B.C.E. rather than the ninth century B.C.E.

With this caveat in mind, Steiner ascribes numerous significant finds to the tenth century B.C.E. One of these is an imposing casemate wall (only a small section has been found). A casemate wall consists of two parallel walls divided by cross-walls forming casemates, or small rooms. Each of the parallel walls at Jerusalem is more than 6 feet wide. The area between the wall is more than 4 feet wide. On the fill between the walls some sherds were found, “all dating to the tenth/ninth century B.C.” If actually tenth-century B.C.E., it may be that Solomon built a new defensive wall to protect the northern part of Jerusalem, connecting with the palace and Temple on the Temple Mount. Steiner asks whether markets and festivals may not have been held in this new northern section of the city, south of the Temple Mount, protected by the new wall.

Among the debris from this period, Kenyon also found some fairly fancy ashlars (squared stones) that were typical of public buildings of the tenth/ninth century B.C.E. She even found the remains of a wall built of such ashlars. In addition, a large and elegant proto-Aeolic capital from this period obviously came from a very impressive public building. Shiloh, however, dated this capital to the ninth rather than the tenth century B.C.E.

The tenth/ninth century B.C.E. also saw some reworking of the Stepped Stone Structure of the 12th century B.C.E. “Whoever came to Jerusalem from the east saw this 27 meter [88 feet] high ‘tower’ in front of them”—a scary sight indeed. “This bastion protected this vulnerable [northeast] corner of the town [the northern side being the only one not protected by deep valleys], while at the same time guarding the [water source at the Gihon] spring.” This Stepped Stone Structure may be the mysterious Millo that the Bible says both David and Solomon fortified (see 2 Samuel 5-9 and 1 Kings 9-15, 24, among other passages).

That the pottery Kenyon found and dated to the tenth century B.C.E. may well date to the ninth century B.C.E. is only one problem, says Steiner. There simply is not much of it. Whether it wasn’t there to begin with or was thrown out, however, we’re not sure. “Almost none [of it] bears the dark red slip layers,” Steiner notes, “traditionally ascribed to the tenth century.”

Steiner also mentions several “luxury items” from this period discovered in Shiloh’s later excavation of the area, for example, a bronze fist, perhaps belonging to a god, and a large pottery stand portraying a bearded man.

But neither Kenyon nor Shiloh discovered any domestic structures. Because of this, Steiner finds that the town must have functioned only “as a regional administrative centre or as the capital of a small, newly established state.” Again, a shaky conclusion based on the absence of evidence.

In the tenth or ninth century B.C.E. only about 5,000 people lived in Jerusalem, according to a study by Shiloh and Israeli archaeologist Magen Broshi. Steiner says it was closer to 2,000.
Finally, without qualification, she states- “The United Monarchy is not an historical fact.” She bases this conclusion on a comparison of Jerusalem with other contemporaneous sites that have been excavated—Megiddo, Hazor, Gezer and Lachish. All could be capitals of regional states. Therefore, she reasons, Jerusalem was not the capital of the United Monarchy any more than these other towns. Her conclusion, however, once again does not follow from the evidence.

This is an important but disappointing book. Kenyon’s excavation of the City of David was the most extensive and significant since the First World War. All who are intrigued by the history of the Holy City must come to terms with the information her excavation has to give us. It is disappointing, however, for several reasons, most of them not Steiner’s fault. The major disappointment is that the revered master of archaeological method and technique was, in fact, a poor excavator and record keeper. The emperor (or empress) had clay feet, to coin a phrase.
In many cases, Kenyon also reached erroneous substantive conclusions. Perhaps this is more forgivable. It sometimes seems like the major work of current archaeologists is to point out the errors of their predecessors.

Some of my disappointment comes not from Kenyon’s errors, but simply from the nature of her task. In many cases, for example, even experts cannot tell the difference between pottery from the tenth and ninth centuries B.C.E.

Steiner herself is a careful and dedicated scholar. We feel her pain at having to work for nearly 20 years to squeeze so little juice out of the lemon. But she also comes to negative conclusions without due circumspection. She is too ready to say that it—whatever it is—didn’t exist just because so little has been found. She admits that much evidence may be under the Temple Mount, which is obviously not available to archaeologists. But, in addition, much of the City of David, almost the entire top of the ridge, remains unexcavated because modern houses stand upon it. To reach the conclusion that there were no domestic dwellings at a certain period in Jerusalem, despite the fact that most of the site remains unexcavated, seems unjustified. In short, Steiner gives much too much weight to the absence of evidence. As the old aphorism has it, the absence of evidence is not evidence of absence—at least not in circumstances like these.

Steiner also has a knack for explaining what evidence there is in such a way as to deny Jerusalem’s prominence—for example, her assertion that the mammoth Stepped Stone Structure was built to fill a gully.

Steiner sometimes errs simply because later evidence was unavailable when she wrote this final report. She tries to take account of the evidence from Shiloh’s subsequent excavation of the area, but since then Israeli archaeologists Ronny Reich and Eli Shukron have also dug near the Gihon Spring. Steiner tells us with some certainty, for example, “that the building of Warren’s Shaft must be assigned to [Middle Bronze IIB],” while noting that others have assigned it many other dates (Shiloh places it in the tenth/ninth century B.C.E.). Steiner reasons that because the shaft makes the water from the spring (which is outside the city, near the bottom of the slope) accessible from inside the city (by means of an underground tunnel system), the shaft was built at a time when a wall protected the city. What seems like sensible reasoning leads her to the conclusion that the shaft was built in Middle Bronze IIB, when the city was enclosed by a massive wall. But since Steiner wrote her report, Reich and Shukron have shown that Warren’s Shaft is a natural karstic chimney that was never used to draw water (though this conclusion is also now disputed). The underground tunnel system at the top of the shaft led not to the shaft, but to towers near the spring. And it was not opened until the eighth century B.C.E.

If you’re looking for certainty, stay away from Jerusalem archaeology in general and this book in particular. But if you can’t stay away, read and enjoy. It also helps if you like crossword puzzles.

a. See Nadav Na’aman, “It Is There- Ancient Texts Prove It,” BAR 10-01. See also Margreet Steiner, “It’s Not There- Archaeology Proves a Negative,” BAR 24-04 and Jane Cahill, “It Is There- The Archaeological Evidence Proves It,” BAR 24-04.

b. In addition, as Richard Hess recently pointed out, Abdi-Heba’s letters sometimes display “an extraordinary quantity of rhetorical features when compared with the remaining letters in this collection.” This would hardly be expected of the “manager” of an estate.

1. Kathleen Kenyon, Jerusalem—Excavating 3,000 Years of History (New York- McGraw-Hill, 1967) and Digging Up Jerusalem (New York- Praeger, 1974).

2. The project also includes this previously published work- H.J. Franken and M.L. Steiner, Excavations in Jerusalem 1961–1967, vol. 2, The Extramural Quarter on the South-East Hill (Oxford- Oxford Univ. Press, 1990). Publication of the coins, inscriptions, seals, flint, some closed pottery groups, human and animal bones, shells, stone weights, hammer stones and figurines have been assigned to specialists in these various fields. Most of the work is “still to be done” and little of it is published.

3. Steiner makes the same kind of argument with the mention of Jerusalem in “Egyptian Execration” texts from the 19th or 18th century B.C.E., when, again, Steiner finds no evidence for a settlement at Jerusalem. “The mentioning of this name alone cannot be used as ‘proof’ that Jerusalem was an important town then, as the name need not specify a town—it could as easily indicate a region or a tribe.”