Temple Mount Sifting Project

Nine Thousand Tons, and No Record

In November 1999 the Islamic religious trust that administers the Temple Mount carried out works to open a new entrance to the underground space known as Solomon's Stables, converting it into the el-Marwani Mosque. The work was done with heavy earth-moving machinery, without a preceding salvage excavation and without archaeological supervision. Something on the order of nine thousand tons of earth were removed and trucked away — an estimated 350 lorry-loads — and dumped in the Kidron Valley near the north-eastern corner of the Old City.

In 2004 G. Barkay and Z. Dvira were licensed to recover it, under Bar-Ilan University auspices and with funding raised through the Israel Exploration Society. The debris was hauled to Emek Tzurim, dry-sifted into buckets, soaked, and then washed over screens by hand — a wet-sifting method the project adapted and scaled for this material, and which other excavations in Israel have since taken up. Finds are sorted into pottery, glass, bone and shell, mosaic tesserae, metal, and worked stone and plaster, then classified in the project's laboratory.

The funding history matters, because it is the ground of one of the objections below. In 2005, in financial difficulty, the project accepted Elad — the foundation that runs the City of David site and also promotes Jewish settlement in Silwan — as administrator of the sifting site, with G. Barkay and Z. Dvira keeping scientific control. That arrangement ended in 2017; sifting stopped, Elad kept the Emek Tzurim site, and the project resumed at Mount Scopus in June 2019. G. Barkay has said plainly that nobody told him what to find or what to ignore.

Volunteers do most of the washing, which is unusual and deliberate: the project has run on public participation from the start, and members of the public rather than staff have handled the great majority of the material.

What Has Come Out

The material is genuine and some of it is important. More than five hundred coins of the Second Temple period. About a hundred and thirty fragments of Iron Age Judahite clay figurines — human, animal and horse-and-rider types, the small figures that turn up all over Iron Age Judah. A further thirty came not from the platform debris but from an ancient rubbish dump on the eastern slope below it, and the two groups should not be added together, as they often are. A clay sealing — a bulla — of the seventh or sixth century BC bearing the name of the priestly family of Immer, a family named at Jeremiah 20:1 as holding office in the temple. Opus sectile floor tile fragments that have allowed a reconstruction of the paving patterns of the Herodian courts. Crusader armour scales, silver coins, Byzantine tesserae, arrowheads, weights, and a very large quantity of pottery running from the Bronze Age to the Ottoman period.

This is very nearly the only sample of material from the Temple Mount platform there is. Not quite the only one: R. W. Hamilton, director of the Mandate antiquities department, cut trenches under the Aqsa Mosque during the repairs of 1938-1942 and found a mosaic floor and a third-century relief of a centaur, most of which he never published. But no systematic excavation of the platform has ever been made, and none is going to be — the reasons are religious and political and they are not going to change. Everything else the app can say about the Temple Mount comes from digging around its base: the southern wall, the Ophel, the western wall tunnels, never the surface itself.

That scarcity is exactly why the caveat below is not a footnote to the article. It is the article.

Objects Without a Layer

Archaeology's evidence is not really objects; it is objects in relationships. A cooking pot means little on its own and a great deal if it lies on a floor, under a collapse, above a coin. Those relationships — stratigraphy — are how anything gets dated and how one find is connected to another. The 1999 works destroyed every relationship in nine thousand tons of earth in a matter of days. Nothing sifted from it can be dated by its context, assigned to a building, or connected to any other find, because it no longer has a context to be dated by.

That limit has been pressed hard, by three critics who press it in three different ways. I. Finkelstein doubts the premise that the debris all came from the platform's own deposits, and holds that only an extraordinary find — an inscription — could carry weight from material like this. M. Abu-Khalaf, of Al-Quds University, points out that the area worked in 1999 was itself a dump used to build up the platform, so some of the earth may have come from the surrounding valleys rather than from the Mount. Y. Mizrachi, of the organisation Emek Shaveh, argues the project "doesn't have any archaeological value" at all — and his objection has a second half that ought to be reported with the first, because leaving it out makes his position look merely dismissive: he holds that a project with little scientific value nonetheless has "a lot of impact" on the modern narrative around the Mount, and that even an inscription naming Solomon and a temple would, on this material, "raise more questions than answers". None of the three is a fringe voice, and this article does not brush them aside.

The project's own answer is worth stating in its own terms rather than in a caricature. G. Barkay compares the work to an archaeological surface survey, a long-accepted method in which unstratified surface material is treated as a statistical sample of what lies below. The project argues that out-of-context finds still carry typological and quantitative information; that comparison with parallel objects from well-dated contexts elsewhere in Jerusalem lets much of the material be dated typologically; and that because the bulldozers created discrete heaps rather than mixing everything uniformly, statistical clustering within heaps may partly recover which finds came from the same original deposit. That last claim is the project's own proposal and, by its own account, awaits full classification before it can be tested.

So the balance. This material can support statements of the form "objects of this kind and this period were present on the Temple Mount, in roughly these proportions." That is real knowledge and it is not nothing. It cannot support statements of the form "this object came from this building, in this year, and therefore that event happened." Any claim of the second kind made from sifted material — and such claims have been made, in headlines, about several of these finds — is claiming more than the method can deliver. The badge on this article says "no excavation context", and it means it.

Scripture references

• Mark 13:1-2 — The buildings whose debris this is, and what was said of them
• Micah 3:12 — "Zion for your sake will be plowed like a field" — a prophet's image, and in 1999 a literal one
• Jeremiah 26:18 — The same oracle quoted back a century later in Jeremiah's defence
• Matthew 24:1-2 — Stones thrown down — the destruction layer this debris partly represents

Sources

• Holding institutionTemple Mount Sifting Project — about the project — Gabriel Barkay and Zachi Dvira, Temple Mount Sifting Project · The project's own account of the 1999 works, the tonnage removed, and the sifting operationSupports: The founding, the directors and the quantity of debris
• Holding institutionTemple Mount Sifting Project — methodology — Temple Mount Sifting Project · The project's statement of its own method, including its acknowledgement that major archaeological data was lost and its case that unstratified finds retain typological and statistical valueSupports: The project's answer to the stratigraphy objection, in its own words
• ReferenceThe Times of Israel: "Temple Mount archaeological project yields treasure, unearths conflict" — The Times of Israel · Reporting that quotes I. Finkelstein, M. Abu-Khalaf and Y. Mizrachi against the project and G. Barkay in reply, and sets out Elad's adoption of the sifting site in 2005Supports: The three criticisms in the third section, G. Barkay's answer, and the funding history
• EncyclopediaWikipedia: Temple Mount Sifting Project — Wikipedia · The find counts and the project's history. Note that this article carries no criticism section — the objections in the third section here are cited to the Times of Israel report above, not to itSupports: The find counts and the operational history
• ScholarshipG. Barkay and Z. Dvira, reports of the Temple Mount Sifting Project — Gabriel Barkay and Zachi Dvira, Bar-Ilan University · Published across New Studies on Jerusalem and related conference volumes from 2006 onward. Print onlySupports: The published finds

Further reading

• Temple Mount Sifting Project: about the project
• Wikipedia: Temple Mount Sifting Project

Reflect

Somebody's careless afternoon destroyed information nobody can ever recover, and other people have spent twenty years salvaging what they can. Which of those two are you being asked to be this week?