In a new article in the Journal of Hebrew Scriptures (11/12), Israel Finkelstein, Ido Koch, and Oded Lipschits propose that the city of Jerusalem only rarely included the “City of David” ridge south of the Temple Mount. The article, available in pdf format, begins with these paragraphs:

The conventional wisdom regards the City of David ridge as the original mound of Jerusalem. Yet, intensive archaeological research in the last century—with excavations in many parts of the ca. six hectares ridge (see Fig. 1), has proven that between the Middle Bronze Age and Roman times, this site was fully occupied only in two relatively short periods: in the Iron Age IIB-C (between ca. the mid-eighth century and 586 B.C.E.) and in the late Hellenistic period (starting in the second half of the second century B.C.E.). Occupation in other periods was partial and sparse—and concentrated mainly in the central sector of the ridge, near and above the Gihon spring. This presented scholars with a problem regarding periods for which there is either textual documentation or circumstantial evidence for significant occupation in Jerusalem; we refer mainly to the Late Bronze Age, the Iron IIA and the Persian and early Hellenistic periods.
Scholars attempted to address this problem in regard to a specific period. Na’aman (2010a) argued that the Late Bronze city-states are underrepresented in the archaeological record also in other places; A. Mazar (2006; 2010) advocated the “glass half full” approach, according to which with all difficulties, the fragmentary evidence in the City of David is enough to attest to a meaningful settlement even in periods of weak activity; one of us (Lipschits 2009) argued for enough spots with Persian Period finds on the ridge; another author of this paper (Finkelstein 2008) maintained that the weak archaeological signal from the late Iron I—early Iron IIA (the tenth century B.C.E.) and the Persian and early Hellenistic periods reflects the actual situation in Jerusalem—which was only sparsely populated in these periods. Still one must admit that the bigger problem—of many centuries in the history of Jerusalem with only meager finds—has not been resolved.
In what follows we wish to put forward a solution to this riddle. Following the suggestion of Knauf (2000) regarding the Late Bronze Age and Iron Age I, we raise the possibility that similar to other hilly sites, the mound of Jerusalem was located on the summit of the ridge, in the center of the area that was boxed-in under the Herodian platform in the late first century B.C.E. Accordingly, in most periods until the second century B.C.E. the City of David ridge was outside the city. Remains representing the Late Bronze, Iron I, Iron IIA, and the Persian and early Hellenistic periods were found mainly in the central part of this ridge. They include scatters of sherds but seldom the remains of buildings, and hence seem to represent no more than (usually ephemeral) activity near the spring. In two periods—in the second half of the eighth century and in the second half of the second century B.C.E.—the settlement rapidly (and simultaneously) expanded from the mound on the Temple Mount to both the southeastern ridge (the City of David) and the southwestern hill (today’s Jewish and Armenian quarters).
The theory of “the mound on the Mount” cannot be proven without excavations on the Temple Mount or its eastern slope—something that is not feasible in the foreseen future….In other words, for clear reasons—the inability to check our hypothesis in the field—we cannot present a well-based solution for the “problem with Jerusalem.” Rather, our goal in this paper is to put this theory on the table of scholarly discussion.

Three objections come immediately to mind: (1) the biblical problem is that 2 Samuel and 1 Kings indicate that the city was near the Gihon spring and later expanded to include the Temple Mount (2 Sam 5:8; 24:18-25; 1 Kgs 1:33-45); (2) the logical problem is that such a proposed city would not have included a water source; (3) the archaeological problem is that the Gihon Spring is surrounded by massive fortifications. The authors dismiss these finds as a “riddle,” but in fact they are compelling evidence against the thesis of this article. The traditional view that the City of David was the original core of Jerusalem explains all of the evidence in a more satisfactory way.