Here’s a Claude.ai summary of Emanuel Tov’s article, Searching for the “Original” Bible: Do the Dead Sea Scrolls help? Originally published in Biblical Archaeology Review (40.4, 2014). A PDF copy is in his directory in Dropbox.
Central Question
Tov asks how reliable the Dead Sea Scrolls’ biblical texts are for reconstructing an “original” Hebrew Bible, compared to the two traditional touchstones: the Masoretic Text (MT, finalized ~1000 CE) and the Septuagint (LXX, third century BCE onward). His overarching thesis is that textual criticism is inescapably subjective — there’s rarely a definitively “right” answer, only more or less contextually plausible ones, and the Scrolls should be used judiciously rather than automatically privileged for their age.
Four Case Studies
Genesis 2:2 — sixth day or seventh day? The MT has God finishing creation on the seventh day (theologically awkward, since it implies God worked on the day of rest), while the Samaritan Pentateuch, LXX, and Peshitta read “sixth day.” The verse isn’t preserved among the Scrolls, and Tov says he’d still favor “seventh” on contextual grounds even if a Scroll reading existed — a deliberate illustration that manuscript evidence doesn’t automatically settle the question.
1 Samuel 1:24 — one bull or three? The MT has Hannah bringing “three bulls” to Shiloh; the LXX and a Qumran fragment (4QSam^a) read “a three-year-old bull” (singular). Tov argues the MT reading arose from a word-division error in the unspaced Hebrew original, and prefers the Scroll/LXX reading, partly because the next verse in the MT itself refers to “the bull” in the singular.
Deuteronomy 32:8 — sons of Israel or sons of God? This is the most striking example. The MT has the Most High dividing the nations “according to the number of the sons of Israel,” but 4QDeut^j and the LXX read “sons of Elohim” (God) — pointing to an even earlier stage where the text likely read “sons of El,” reflecting a Canaanite/Ugaritic divine-council setting. Tov reconstructs a probable original in which “the Most High” and “El” function as distinct deities in a polytheistic framework, later flattened by scribes uncomfortable with that theology — first to “Elohim,” then to “Israel.”
Jeremiah 10 — praise alongside mockery of idols? The MT’s mockery of idol-worship is interspersed with verses praising the Lord’s incomparability, but these praise-verses are absent from both the LXX and 4QJer. Tov suggests both versions may be “original” in a diachronic sense: the shorter text reflecting an earlier compositional stage, later expanded — illustrating how the Scrolls can illuminate a book’s literary growth rather than just settle a single “correct” reading.
Conclusion
Tov closes by noting that even a text as complete and admired as the Great Isaiah Scroll (1QIsa^a) is actually textually inferior in many respects — full of harmonizations, grammatical smoothing, and scribal errors — yet still preserves valuable ancient readings. His final point is methodological: the “original text” is often simply unrecoverable, and scholars must combine the MT, LXX, and Qumran evidence with contextual judgment and appropriate humility, rather than treating any one source as automatically authoritative.
Given your ongoing work on Marcion priority and second-century textual history, Tov’s argument here (that textual plurality and even directional theological editing — as in Deut 32:8 — long predates any notion of a fixed “original”) pairs interestingly with the Vinzent/Klinghardt debates about how fluid the Pauline text itself may have been in the second century. Let me know if you’d like me to pull more on the Deut 32:8 divine-council question specifically, since that connects to your interest in intermediary cosmology.
