On The Gezer Water System

There are two new and interesting news items on the Gezer Water system (one link serves both). Here’s a sample,

Gezer Water System Expedition, a joint project of NOBTS [New Orleans Baptist Theological Seminary] and the Israel Nature and Parks Authority, requires a different, more physical approach in excavating an ancient water system that could date to the time of Abraham.
Because the massive tunnel, cut by the Canaanites by hand using flint tools, was excavated by Irish archaeologist R.A.S. Macalister from 1906-08, the New Orleans team did not sift the dirt during the first two dig seasons.
The massive water system measures 13 feet wide by 24 feet high and stretches 150 feet into the ground at a 38-degree angle. When NOBTS started the dig, nearly 65 percent of the tunnel was filled with dirt. During his dig 100 years ago, Macalister built a rock retaining wall to hold back the debris he removed. The wall, located near the mouth of the tunnel, collapsed after heavy rains in 1908 sending tons of rocks and dirt back into the water system. Over the years more dirt and debris has collected in the tunnel.

In large part because I worked three seasons at Gezer in the early ‘70s and walked past the water system every working day for one of those seasons, I find all this abnormally interesting. I do wonder why, (well, I know why) the report says, “could date to the time of Abraham.” Anytime one reads “could” or the like and a biblical personal name in the same sentence it should be a strong warning that one of the goals of the piece is to tie some ancient artifact to something in the Bible even if there is no reason for such a connection. Let me be clear, from a technical point of view I think the recent ongoing excavations at Gezer are top notch. It’s the public reporting on those excavations that makes me twitch.
Via Bible Places

8 thoughts on “On The Gezer Water System”

  1. I re-excavated a site in Georgia, USA. I screened all the back filled units we found, and was facinated at what the earlier archaeologists had neglected to collect.
    I think the folks at NOBTS are missing a chance for interesting data. I am very hesitant to give much credence to any ANE archaeology done by a seminary anyway, and even less to a creationist gang.

  2. Gary,
    Good to hear from you. What is so weird about all this is that the field work and the primary formal publications of this excavation team is to the highest contemporary standards (I’ll address the issue of screening below). By their formal publications no one could tell that many, but not all, of them worked for a rather conservative seminary. I’m not sure but I’d be really surprised if the directors of this excavation were creationist in any way other than some vague “God directed evolution” sort of way. Not that I approve of that but . . .
    While screening is far more common now than it was even a few decades ago, 100% screening is virtually unheard of in any other than Neolithic or older sites in the Near East. There are exceptions but most screening is sample screening or of loci that appear to requite it on some contextual grounds, living services etc. Remember, the matrix at Gezer and many other sites is almost exclusively undifferentiated eroded mud brick detritus. I’m not necessarily claiming that anything less than 100% screening is good practice but I don’t think the water system crew is doing anything that other Syro-Palestinian field archeologists would think all that wrong. My guess is that Cole and Warner will start some sample screening as they get closer to the lower extend of Macalister’s exposer. I doubt that the co-sponsors of the excavation, the Israel Nature and Parks Authority or the Israel Antiquities Authority would put up with wholesale bad practice. I believe that these days, 100% careful dipping of pot shreds in water rather that scrubbing them is far more common that 100% matrix screening. Everyone hopes that they will find an ostracon and not just a pot shred. No telling how many inscriptions were scrubbed off shreds n the old days.
    Based on what I remember of the water system, the fill is not only eroded mud brick detritus but mud brick detritus that was deposed in the system by the last 100 years of site erosion. I don’t think Macalister used it as a dump and I am completely sure that it was not used as a dump by either Dever or Seger in the late ‘60s and early ‘70s. I do know someone who peed in it.
    I think most screening these days is through a 1 cm (1/2 inch) mesh as it was in the old days also. But when we were preparing samples for seed analysis we would sometimes use a .5 cm mesh and before a sample was submitted for pollen recovery it was, if I remember correctly, passed through a .25 cm mesh. I’m not sure of current practice in these matters. Technology may no longer require such fine prescreening for pollen recovery. Anyway, at what level of 100% screening does one no longer get meaningful returns for one’s effort?

  3. Howdy Duane,
    I did a few experiments on prehistoric shell middens of screening with different mesh screens. I ended up using 1/4 inch and 100% screening. Bone, and lithic debitage counts increased 4 fold from 1/2 inch mesh. I contrasted dry sorting with water screening, and sorting. Counts for all small artifacts (beads, utilized flakes) and all shell and bone doubled again.
    I also would do columns of 1 liter soil samples, sampled every 5 cm. These I wet screened from 1/2 inch to a #30 mesh (~0.5 mm) in about 4 steps. The small fractions were sorted under 30X magnification. Entire groups of data were collected that we had otherwise missed, esp. fish bone, insect frass, and seeds. I also took out a 10 gram pollen sample from these columns before screening. I used these column methods on some ‘living floor’ soils from an Olmec site and recovered a lot of fish bone, and scales. The fish had not been identified earlier as part of the local diet.
    In huge urban, historic sites, the architecture seems to me to overwhelm the research design. At one point, I used small front-loaders called “bobcats” to haul soil to a mechanical conveyor belt sifter (mostly used in the rock and gravel industry). The smallest mesh on those was 1/2 inch. Each load was a 1X1 meter by 20cm sample, and we could maintain provenience. This was a bit rough on the material. But, we could process ~10 tons an hour with a crew of 5. The lab cost jumped as the field costs dropped, even if I just had crew “cherry pick” projectile points, sherds, etc…
    The matrix you describe would be a real pain in the backside to work. I have done some Mission Period work with adobe bricks and “melt,” as well as Olmec mound samples. It was much slower.

  4. Gary,
    How did you maintain locus control with the bobcats and the conveyor belts? It must have been a real pain to sustain the associate between recovered material and its excavation context. By the way, Bill Dever used a bulldozer at Gezer to move one of his own dumps. He bulldozed down to a few 10s of cms from his pre-dump survey levels and then started controlled excavation until (and after) he hit virgin material. At a large scale excavation, dump management a big issue.

  5. Duane, I had 2 guys at the midden, the bobcat driver, and 2 guys at the shaker. The bobcat bucket was ~90 cm. I also had hand excavators working around the site on other units.
    The excavators would mark the 20cm “floor” with pin flags, and cut the 1X1 meter edge. The bobcat would cut the floor. The excavators would trim the unit, check the floor depth and level, and then add this soil into the bobcat bucket. The unit designation, and floor depth was ready on a card stock, with a space for brief notes on the back. The excavators put the card into a zip-lock bag. The bobcat operator took the provenience card, and the load to the shaker. There the card is handed to the shaker crew, who boxed the sample, copied the provenience on the box, and placed the unit card into the box with the screened sample.
    Every hour or two, the bobcat driver would clear out under the shaker, and take the soil to the spoil pile.
    On one project we used a dump truck to haul spoil off site.
    When we found a feature, we switched back to hand excavation, and the bobcat worked a different part of the site.
    There were sometimes problems. For example, one idiot started messing with my provenience cards, and it took three days to straighten out, and some data was lost- no provenience, no data. (The idiot was fired on the spot).

Comments are closed.