The Problem of Ai - The BAS Library
Login
Home > Magazines > BAR > March/April 1985

The Problem of Ai

New theory rejects the battle as described in the Bible but explains how the story evolved

By Ziony Zevit

The problem of Ai is simply stated. The Bible tells us that an important battle in the Israelite conquest of Canaan occurred at Ai. After being defeated by the Canaanites of Ai once, the Israelites launched a second attack on Ai; this time they were victorious, destroying both the city and its inhabitants. However, extensive and sophisticated archaeological excavations have produced no evidence of human occupation at the site during the period when the Israelites supposedly settled in Canaan. In short, the evidence shows that there was no city at Ai for the Israelites to conquer. The conquest of […]

You have already read your free article for this month. Please join the BAS Library or become an All Access member of BAS to gain full access to this article and so much more.

Join the BAS Library!

Already a library member? Log in here.

Institution user? Log in with your IP address or Username

Footnotes

1.

Characteristic of a philosophical system founded by August Comte, concerned with positive fact and phenomena, and excluding speculation upon ultimate causes or origins.

2.

The Arabic name Khirbet et-Tell (literally, “the ruin of the tell”) is sometimes used to support the identification of the site as Biblical Ai. The Arabic name, it is claimed, is a translation of the Hebrew ha-‘Ai, which supposedly also refers to “the ruin,” especially as the Hebrew name always appears with the definite article, ha-‘Ai. On philological grounds, however, any connection between the Hebrew name and the Arabic name for the site is to be rejected. The Arabic term tell refers to a hill or mound on which there is a ruin, as distinguished from khirbaµ which refers to a deserted ruin, not necessarily on an elevated area.

Ha-‘Ai is commonly associated with the Hebrew words ‘iy, ‘iyyiµm, ‘iyyiµn, “ruin(s)” (Jeremiah 26:18, Micah 1:6, 3:12, Psalms 79:1), which are to be connected with ‘awwaµ, “ruin” in Ezekiel 21:32 and derived from ‘-w-h whose etymological cognate is Arabic ‘-w-y, ‘awaµ, “to bend.” This association, however, is incorrect. The Septuagint (LXX) renders the city name in Genesis 12:8 and 13:3 and throughout Joshua as Aggai. In Jeremiah 49:3 (=LXX 30:19) Gai with a Greek gamma is used for the Hebrew letter ‘ayin. In Biblical Hebrew, however, this letter was polyphonous, indicating two distinct sounds. Its rendering in Greek by gamma indicates that it was pronounced as a ghayin, gŒ, as in the name of the Philistine city ‘zh, Greek Gazza, Arabic gŒazza, and English “Gaza.” Thus, if an etymon is to be sought, the first phoneme must correspond to an Arabic gŒ and not to an Arabic ‘. Arabic gŒ-y-y (to hoist [a standard]) and gŒaµyat (extreme limit, utmost extremity) suggest that the Hebrew name gŒay refers to some topographical feature characteristic of the site. Accordingly, there is no demonstrated connection between the Arabic name of the site and Biblical ha‘ay.

Furthermore, the Arabic name is not unique. Six other sites with the name et-Tell occur in the areas of Jenin, Nablus (two sites), Jerusalem, Ramleh, and the Golan Heights (J. M. Grintz, “Ai Which Is Beside Beth-Aven: A Reexamination of the Identity of ‘Ai,” Biblica 42 [1961], p. 208; C. Epstein and S. Gutman, “The Golan,” in M. Kochavi, ed., Judea, Samaria, and The Golan: Archaeological Survey 1967–68 [Jerusalem, 1972], p. 276). Similarly, in addition to the Canaanite and later Israelite site being discussed in this paper, an Ammonite city called Ai (‘ay, in Hebrew, without the definite article) is mentioned in Jeremiah 49:3.

3.

Joshua 17:18 indicates that the area was forested. In fact, the name of the village near et-Tell, Deir Dibwan, means the “Lair of the Bears.” Folk tradition associates Deir Dibwan with the Elisha story in 2 Kings 2:23–24, in which two she-bears are said to come forth “out of the woods.”

Endnotes

1.

W. F. Albright, Annual of the American Schools of Oriental Research 4 (1924), p. 147.

2.

W. F. Albright, Bulletin of the American Schools of Oriental Research (BASOR) 74 (1939), pp. 11–23.

3.

Y. Elitsur, “The Boundary Line Between Benjamin and Ephraim” in U. Simon and M. Goshen-Gottstein, eds., Studies in Bible and Exegesis (Ramat Gan, 1980), pp. 7–14 (in Hebrew).

4.

Thus, at Khirbet Heiyaµn, on the southern edge of Deir Dibwan, an experimental sounding made in 1964 revealed that the earliest structures there were Byzantine and the earliest sherds on the site Roman (J. A. Callaway and M. B. Nicol, “A Sounding at Khirbet Heiyaµn,” BASOR 183 (1966), pp. 27–29, 12–13, 19).

Excavations at Khirbet Khudriya, approximately 750 meters east of Deir Dibwan, conducted in 1966 and 1968, revealed only a Byzantine church and an industrial area. The earliest material discovered at this site was late Hellenistic (J. A. Callaway, “The 1968–1969 ’Ai [et-Tell] Excavations,” BASOR 198 [1970], pp. 10–12).

Surface exploration of Khirbet el-Hay, approximately four miles southeast of et-Tell, near Biblical Michmash produced sherds only from the Middle Ages and the Ottoman period (Z. Kallai, “The Land of Benjamin and Mt. Ephraim,” in M. Kochavi, ed., Judea, Samaria, and The Golan Archaeological Survey 1967–68 [Jerusalem, 1972], p. 182).

An intensive survey of a fifth site, an unnamed tell located about two miles southwest of Beitin, indicated that there had been no Middle Bronze-Late Bronze occupation. Although there may have been a small Iron Age settlement on the site, its major periods of growth and development were Roman and Byzantine (R. B. Blizzard, “Intensive Systematic Surface Collection at Livingston’s Proposed Site for Biblical Ai,” Westminster Theological Journal 36 [1973–74], pp. 224–225).

5.

J. Callaway, “New Evidence on the Conquest of ‘Ai,” Journal of Biblical Literature 87 (1968), pp. 316–19.

6.

A. Kuschke, “Hiwwiter in Ha‘Ai?” Wort und geschichte, festschrift … K. Elliger (=) Alter Orient and Altes Testament 18 [1973]), pp. 117–19; Y. Yadin, “Is the Biblical Account of the Israelite Conquest of Canaan Historically Reliable?” BAR 08:02.

7.

Yadin, op. cit., p. 23.

8.

This wadi is Wadi el-Jaya. It comprises the steep paper-clip shaped valley north of the tell where contour lines 850–700 appear to bend back on themselves.

9.

J. Callaway, “Excavating at Ai (et-Tell) 1964–1972,” Biblical Archeologist (BA) 39 (1976), p. 29.

10.

Y. Shiloh, “Elements in the Development of Town Planning in the Israelite City,” Israel Exploration Journal 28 (1978), pp. 44–46.

11.

Callaway, BA 1976, p. 29.

12.

Callaway, BASOR 1970, pp. 19–24, 26.

13.

Y. Shiloh, “The Population of Iron Age Palestine in the Light of a Sample Analysis of Urban Plans, Areas, and Population Density,” BASOR 239 (1980), p. 30; cf. M. Broshi, “Estimating the Population of Ancient Jerusalem,” BAR 04:02. Callaway suggests that the population of the site was 150 (“A Visit with Ahilud,” BAR 09:05). His figure represents only the population of the 20 excavated Iron Age houses. Mine is based on a maximalist assumption that all the “acropolis” area constituted the inhabited town. The presence of public, uninhabited space within a town is built into the density coefficient so that it may be used when the inhabited area is known and is not restricted to excavated areas alone (cf. Shiloh, BASOR 1980, pp. 29–30). The most recent discussion of this topic by Gus W. Van Beek supports the plausibility of Shiloh’s figures (“A Population Estimate for Mareb: A Contemporary Tell Village in Northern Yemen,” BASOR 248 [1982], pp. 61–67, esp. pp. 64–67 and notes 4, 6).

14.

The Hebrew text does not read, “they pursued them from in front of the gate to the sebaµriµm,” i.e. mlpny hs‘r. Were this the reading, then the text would be describing the course of the chase, from x to y, which would have to be sought beyond the perimeter of the Early Bronze Age city walls. The text is describing the general locus of the pursuit, mentioning only the place at which the men of Ai stopped. Most readers of this text assume that it describes a course and, therefore, search for the sebaµriµm, which is translated variously as “ravines,” “quarries,” or “crags” some distance from the city. The root s-b-r, however, is not associated with quarrying or quarnes; the root h-s-b is often employed in such contexts, and perhaps p-s-l (Judges 3:19, 26). It is, however, associated with ruined walls: “Surely, this iniquity will work on you like a spreading breach in a high wall whose crash (sbrh) comes suddenly. And its crash (sbrh) is like the smashing (sbr) of a jug … ” (Isaiah 30:13–14). Thus, both grammatical and philological considerations support the interpretation given here.

15.

This study is based on data evaluated in Ziony Zevit, “Archaeological and Literary Stratigraphy in Joshua 7–8, ” BASOR 251 (1983), pp. 23–35.