Abnormal Interests

The picture is from the south, with Wall 4026 running west to east (or the other way around if you prefer). You can see the north balk of Field II, Area 4, in the background and a few stones from another wall in the foreground. Don't worry about this wall in foreground. It's not important to what I have to say here. But something that is important to this post is what you can't see but likely exists behind that north balk. While it was not excavated in this part of Gezer, the archaeologists believe that the main Middle Bronze fortification wall, Macalister’s Inner Wall, was somewhere not too deep into this north bulk. To quote the forthcoming Gezer VII volume,

Though it has not yet been confirmed by excavation, it can be presumed from the Field IV evidence that the line of the Inner Wall in Field II, rising up from the glacis, also lies behind Wall 4026 still hidden by the Area 4 north balk.

Also, don't be confused by the vertical cuts and seeming irregularities in the side of Wall 4026 as seen in the above picture. These are artifacts of poor excavation technique. As such, they were created by, err, ah, err, me. In general, the exact line of mud brick walls is easier for excavators to recognize after they have been excavated than it is while they are being excavated.

There are a couple of things that I do want you to notice. First, Wall 4026 is continuous across the excavation area. Following up on the regular summer 1973 season, Seger also found that Wall 4026 continued to the west through an adjacent Field II, Area 14. Excavators in Field IV found traces of such a wall associated with a Middle Bronze gate tower. As Seger says in Gezer VII,

We postulate that Walls 14003 and 15004 formed modest parapet structures, serving as stations for forward defense, fronting the central tower and along the outer perimeter of Wall 13004 . . . A comparable structure abreast Glacis 4060 in Field II (Wall 4026) (see chap. III.B) was better preserved.

So, it is reasonable to assume that this wall extended a considerable distance around the city. All the way around the city? Only as part of the south facing defensive system of the city? Only for some smaller part of the south facing defensive system? Who knows?

Second, while gravity driven physical forces, sometimes called "slow mass movements" (creep and/or slumping), within the tell greatly distorted this wall, there does not seem to be any evidence of crenellation, "loopholes" or the like in the exposed portion.

Third, the wall appears, from the angle of the above picture, to have a stone cap and it may. Whether such a cap is original of secondary is a question for later. I'll have more to say about each of these three observations in my next post.

The drawing to the left is a highly schematic representation of the defensive configuration as I see it. This drawing illustrates only the mud brick portion of the wall (L4026A). Also, don't worry for now about the thing I have labeled "step." I was too lazy to make separate drawings of almost the same thing for this post and the next. Nothing about this drawing is to scale. You can see how Wall 4026 actually looked in section in my previous post on this wall. Wall 4060 was about 1.25 meters wide or a little wider at the base and, depending how one measures, about two meters tall, glacis plateau to top. Exact dimensions are hard to know because the wall may have been eroded in antiquity and it has certainly been distorted by slow mass movements within the tell. The distance from Wall 4026 to the main defensive wall is significantly more than one meter. In Field VI, the glacis' summit plateau was 4.5 meters wide. If one had to guess the width of the summit plateau here in Field II, it would be about the same. But it's only a guess. The height of the main fortification wall is unknown as is the nature it's battlements or crenellation if any. It was certainly much taller than Wall 4026 and much taller than it appears to be on the left side of my schematic. As a first order approximation, all this so far is nearly identical to the way I think Seger sees it. The major difference is that I see the stone portion as secondary while he is not so sure. I'll argue my case later. For now, I'll use call Wall 4026 a screening wall with the understanding that this is just the name of a short defensive wall on a relative flat surface in front of a much large defensive wall.

Are there parallels to this defensive configuration? Yes and no. A glacis running up to the main defensive wall is quite common in the Middle Bronze period. Examples of a glacis plateau with a screening wall fronting the main wall are not so common. But a couple examples are worthy of our attention. Barry Kemp, 1983, 131, refers to mid 20th century BCE defensive systems in Nubia when he says,

These forts were each defended by a massive mud-brick wall, with external towers on all sides and at the corners. On the landward side they overlooked a ditch, at Buhen with counterscarp and glacis. Their most distinctive feature was a secondary defensive line at the base of the wall, between it and the ditch. A low parapet with downward pointing loopholes ran along the inner edge of the ditch, interrupted at intervals by semicircular bastions. It seems intended to thwart a fairly sophisticated type of siege, and thus raises the possibility that it represents a form of urban fortification developed in Egypt perhaps during the civil wars of the First Intermediate Period.

Kemp associates these forts with the reign of Senusret I. In other words, they are three or four centuries older than the Middle Bronze age the installation at Gezer. On the dating and other particulars of these fortifications systems, see also Burke, 41- 42. And remember, Gezer Wall 4026 does not appear to have had loopholes or the like, at least not in the exposed portions.

The following drawing is from the Wikipedia article on Semna. As is Abnormal Interests and many other blogs, this drawing is protected by a Creative Commons License under the terms of which I reproduce it here.The Semna fort is from the same period and general area as the one at Buhen.

While an artist's reconstruction, one should note the low, outer, screening wall surrounding the main fortification wall. The Semna example appears to lack towers. At Semna there seems to have been courtyards (staging areas?) formed between the towers of the main wall. One should also note the small glacis in front of the smaller wall. Dunham and Janssen, 5, say of the fortification system at Semna,

As one approached the south or north of the gates, one saw first a glacis about 6 m. in width, an outer wall about 7.5 m. wide, and a ditch of varying width, after which came a massive main mud-brick wall of the fort which had several bastions.

This outer wall was considerably wider than our Wall 4026 at Gezer but otherwise seems a reasonable close parallel in terms of the general configuration of the fortification system.

A somewhat different but still analogous large wall fronted by a small wall arrangement was found at the funerary enclosure of Khasekhemwy in Old Kingdom fort at Abydos. See Kemp, et al (2004), 266, where Kemp provides a few other examples from Upper Egypt.

If we are willing to move from the southern most part of the Levant, some might even say south of the southern most part of the Levant, to the northern most part and from the 20th century to the 14-13th century BCE, we should look at Hattuša (Boghazköy). I found this rather bad and otherwise unattributed picture in Gorny, 73.

Yadin, I, 94, provides a schematic drawing of this complex in section that is in many of its elements identical to what we see at Gezer. Yadin, I, 92, says of the upper city fortification system at Hattuša,

These fortifications of the upper 14th-13th century city are the most powerful of the entire complex. And though similar to those of the lower city, their plan and the quality of their construction are more advanced. Here, too, they are built on a very high and wide rampart. They comprise two walls: the main wall, and, at a distance of 8.5 meters, an outer wall. This outer wall is also built on a rampart, and is 1 meter think. It is strengthened by rectangular bastions, built at a distance of 30 meters from each other and set exactly between the towers on the main wall. [reference omitted]

Yadin, I, 225, provides an artist's rendering of how the system may have looked. Not surprisingly, the artist included crenellation in his reconstruction.

I was only able to find one possible example of a wall from the Middle Bronze Age that might be similar to Wall 2026 at Gezer in some ways. Wall 1 and Wall 3 at Tell Dothan, Area D, may have some similarities to the Gezer configuration. Burke, 537, provides a section drawing that he attributes to D. Master showing a rather small, about 0.75 m thick, stone wall, Wall 3, associated by way of a flat "rock fill" with a much larger, about 4 m think, Wall 1. Wall 3 appears to be founded at the same level as the bottom of the fill and the bottom of the fill appears to be adjacent to and contiguous with the highest row of stones in the slightly wider and lower portion of Wall 1. But from the way the drawing is labeled, it is possible that the rock fill extended below the founding level of Wall 3. A stepped stone covered glacis(?) appears to descend from top of this flat "rock fill" at or within or slightly above the foundation of Wall 3 The associations between these elements is not completely clear from the section drawing. However, Wall 1, Wall 3 and the rock fill do seem to be associated each with the other in some way. I have not been able to reconcile totally this section drawing of with Free's, 14-15, discussion of the walls he uncovered in area K or Burke's own discussion of these walls. For this reason plus the inability to determine the length of Area D, Wall 3, and the difficulty of understanding the association of what I called the "stone covered glacis" with Wall 3 or the "rock fill," I have little certainty about this possible parallel. As more of an after thought than anything else, I wonder if there is a relationship between Wall 3 and Wall 4 immediately below it. If so, the Dothan Area D system is clearly not parallel with the Gezer system. Check it out and see what you think. While the drawing is on page 537 of Burke's dissertation, it will be on page 559 of your PDF reader.

While none of these parallels overwhelms me with its contribution to our understanding of Wall 4026 at Gezer, it is clear that the Gezer system not unique in its gross structure even if many of its details remain unparalleled. There is a temptation, that should be resisted without a lot of additional evidence, to see a south to north migration of the configuration over time: from mid 20th century BCE Nubia to Middle Bronze Age Gezer and Dothan(?) to Late Bronze Age Hattuša. Our sample size is just too small and the analogies between the various systems differ too much in detail to sustain such a conclusion. One must also ask why there seems to be so few, if any, parallels among the over 125 known Middle Bronze Age Near East fortification systems. Barring further evidence, my own guess it that some special and differing circumstances at each of the geographical areas showing this general fortification configuration drove their designers to independently adopt the screening wall fronting a large main wall at the top of a glacis or rampart system. And then, sometimes engineers just like to show off.

As I've said a couple of times now, in my next post, I'll discuss the construction phases of Wall 4026 and speculate on its function(s).

References: