How many of Bebai's offspring returned from Babylon?

What follows is one scholar’s way of reading this disagreement — a critical-historical view that often concludes the texts genuinely disagree. It draws on the peer-reviewed sources listed below and stands as one position among the answers above, presented for transparency, not as GEBible’s position.

How critical scholarship weighs this: Widely regarded as a genuine contradiction

This summarizes where critical scholarship tends to land on this tension — a report on the field’s center of gravity, not GEBible’s own ruling.

The Bebai entry is one node in a systemic problem: across roughly two dozen family entries in the parallel rosters of Ezra 2 and Nehemiah 7, the totals disagree in more than half the cases. Ezra 2:11 gives 623 children of Bebai; Nehemiah 7:16 gives 628 — a five-person gap. Crucially, the two lists are not independent. Nehemiah 7:5 states that the chapter reproduces "the book of the genealogy of those who came up at the first," so both texts depend on one underlying register (the Golah list). The disagreements are therefore disagreements within a single document's transmission history, which reframes the whole question from "two events" to "two copies."

The most durable harmonization, traceable at least to Keil and Delitzsch in the nineteenth century and repeated in much conservative literature since, holds that the two lists were taken at different moments — that deaths, defections, late arrivals, and births in the interval account for the divergences. The proposal is tidy until tested against the data. If genuine attrition or growth were the cause, the discrepancies should trend in a consistent direction or at least correlate with something. They do not. Some families gain people in Nehemiah relative to Ezra; others lose people; the magnitudes range from one to over a thousand with no demographic logic linking them. That is the profile of a random walk produced by independent copying errors, not of a population changing between two reliable censuses. The time-snapshot theory also has to be invoked dozens of times in a single chapter, and it never explains why Nehemiah, who claims to be reproducing an older document, would silently substitute later figures for the ones in his source.

The versional evidence sharpens the diagnosis. 1 Esdras 5 reproduces essentially the same list in Greek and functions as a third, partly independent witness. In the standard versification of 1 Esdras 5:13, the sons of Bebai are given as 623 — agreeing with Ezra against Nehemiah's 628. That two-against-one alignment is precisely how textual critics triangulate an archetype: when two witnesses converge and a third diverges, the divergent reading is the more likely corruption — here pointing to Nehemiah 7:16 as the site of the slip. The caveat worth stating plainly is that 1 Esdras's own numerals are not perfectly stable across its manuscript streams, so the alignment is a strong indicator rather than a closed case. Williamson in WBC 16 and Blenkinsopp in the OTL volume both treat these numerical divergences as transmissional, and Sara Japhet's studies of the restoration period reach the same conclusion: the lists are paired copies, not paired censuses.

The underlying mechanism is well understood. Hebrew numerical material is the most error-prone category in any manuscript, because numbers carry no semantic redundancy. The words around a misread verb or name constrain the possible readings and let a careful scribe catch a slip; a misread number is invisible to such checks. Whether the figures were copied from a damaged exemplar or taken by dictation, the units and tens positions are exactly where transmission fails first, a pattern Tov documents across the canon and which recurs in the Kings-versus-Chronicles totals and the Genesis genealogies across MT, LXX, and the Samaritan Pentateuch.

The apologetic position requires both numbers to be original and correct, since both books are held to be inspired and both claim the same source. The critical position requires only that these be human documents whose numerals drifted in copying — which is what the three-witness evidence directly shows. One account supplies a mechanism that fits the data; the other supplies a harmonization that the data contradict.

Where the critical reading lands: genuine discrepancy, scribal corruption, with 1 Esdras pointing to Nehemiah as the corrupt witness.