← All conjectures · Jewish book cultures

The book named by a formula

Cross-period Dated c. 400 CE – 900 CE Levant & Near East

Status: Supported

The verdict’s fine print — quoted from the resolution record: “(a) Margin: the MAIN Jaccard (0.5843 = 52/89, parasha + within-parasha ordinal alignment) sits close to the registered 0.6 line — margin ~0.016 — but the verdict is robust because all three alignment operationalizations agree below 0.6: strict section-index ma…” Read the full caveats ↓

Status is derived only from the shepherd-authored triage/prediction data above -- community submissions and claims are a separate overlay and can never change it (see the participation panel below).

This is a conjecture imagined by a language model — drawn from its trained weights and held to falsifiability, novelty, and plausibility, not to any one method: it may join two or more fields, or none. It is not an article and not evidence: it sits below the evidence/publication boundary. A quantitative prediction and a named kill-dataset are attached (when registered) so the claim stays falsifiable rather than merely evocative.

Claim (verbatim)

Scattered through the medieval sources are quotations opened by "Yelammedenu Rabbenu" ("let our master teach us") - the signature incipit of a homiletic midrash, the Tanhuma-Yelammedenu, that poses a halakhic question and unfolds an aggadic sermon from it. The problem is that no single book called Yelammedenu survives cleanly: the tradition comes down in overlapping, divergent recensions - the printed (standard) Midrash Tanhuma and the Buber recension edited from manuscripts differ substantially in which homilies they carry and in what order - and medieval citations of a "Yelammedenu" sometimes match neither. The formula itself is the tracer. Mechanism: a homiletic genre organized around a recurring rhetorical opening leaves that opening as a countable fossil, so the density and distribution of the yelammedenu-formula across the surviving recensions measures how much of a once-larger Yelammedenu each recension actually preserves. If the recensions are independent siftings of a lost fuller corpus, their formula-bearing homilies should overlap only partially, and the formula-rate should differ recension to recension. Prediction restated: the "yelammedenu rabbenu" opening appears in a large but recension-dependent share of Tanhuma sections, and the two recensions' sets of formula-bearing homilies overlap well below unity - two partial nets thrown over a vanished book.

Prediction clause (verbatim)

Prediction: string-counting the "ילמדנו רבנו" (yelammedenu rabbenu) opening formula across Sefaria's Midrash Tanchuma (standard) and Midrash Tanchuma Buber recensions, the per-section occurrence rate will differ between the two recensions by at least 50% (a rate ratio outside the band 0.67-1.5), and the Jaccard overlap of the two recensions' sets of formula-bearing homilies (matched by parasha and pericope) will fall below 0.6 - primary clause: the sub-0.6 cross-recension overlap of yelammedenu homilies; the verdict follows it. The test voids for coverage if fewer than 50 formula instances are found across both recensions combined.

Kill-dataset (verbatim)

Kill: Sefaria's Midrash Tanchuma and Midrash Tanchuma Buber - string-match the "ילמדנו רבנו" opening, tabulate per-section rates and the parasha-aligned overlap of formula-bearing homilies between the recensions, with Buber's printed edition and the Mann-Sonne studies of the synagogue homily as the philological controls.

Provenance

Run: Fresh agent generation · model: claude-fable-5

Fresh blind generation by claude-fable-5, 2026-07-17, Jewish text-culture wave instrument-anchored on the open Sefaria corpus and its cross-reference link data, with standard critical editions as controls: every kill names a real corpus and a countable operation (coverage maps, citation-formula counts, link-orphan shares, digest-fraction, recension divergence, citation-decay), thresholds far from 1 with explicit coverage guards distinguishing what Sefaria holds from what existed. Ground is CITATION-GEOMETRIC and disjoint from the 2026-07-10 w19 Jewish wave, which was material-culture-of-loss (colophons, parchment, genizah, masora, binding fragments): no material-culture re-posing here. Two candidates were dropped after a grep of all fresh packets: a Tosafot-density-by-tractate survival item (pre-empted by w02-philosophy #18, which already correlates per-folio Tosafot density with manuscript survival), and a piyyut liturgy-vs-anthology survival item (pre-empted by w01-literature #28 'Liturgy out-survives fame', keyed to Davidson's Thesaurus). The Mishneh Torah item here tests EXTERNAL source-attribution erasure vs the Tur (loss of source-geometry), deliberately distinct from minds-w02 #25, which tests the code's INTERNAL recall/promise cross-references. Two items are marked Kill (not yet built) where the deciding corpus (Kohut's Arukh apparatus; Lewin's parallel Iggeret recensions) is not yet digitized.

Novelty / leakage triage

anticipated in the literature — this exact test has never been run

The Tanhuma-Yelammedenu problem is thoroughly published: Buber's introduction set out the two recensions' divergence, Strack-Stemberger states the standard datum that the Buber recension differs substantially in Genesis and Exodus while running close to the printed Tanhuma for the other books, and Bregman's monograph is a full stratigraphy of the versions, with the Nikolsky-Atzmon volume carrying the field forward. But the specific instrument - string-counting the yelammedenu incipit across both Sefaria recensions, per-section rates, and a parasha-matched Jaccard overlap with its 0.6 threshold - has never been run; Bregman's analysis is philological, not tabulated at formula level.

Sources cited by the triage

S. Buber (ed.), Midrasch Tanchuma (Vilna, 1885), introduction
M. Bregman, The Tanhuma-Yelammedenu Literature: Studies in the Evolution of the Versions (Hebrew; Piscataway: Gorgias Press, 2003)
R. Nikolsky & A. Atzmon (eds.), Studies in the Tanhuma-Yelammedenu Literature (Leiden: Brill, 2021)
H. L. Strack & G. Stemberger, Introduction to the Talmud and Midrash (2nd ed., 1996), Tanhuma-Yelammedenu section

Predictions

Supported registered 2026-07-19 calibration prediction (parent triage: leaked/adjacent)

Resolution: Supported

Caveats: (a) Margin: the MAIN Jaccard (0.5843 = 52/89, parasha + within-parasha ordinal alignment) sits close to the registered 0.6 line — margin ~0.016 — but the verdict is robust because all three alignment operationalizations agree below 0.6: strict section-index match 0.1371 (a naive floor, since siman numbering is recension-internal), parasha-level presence-only 0.5588 (a naive ceiling, ignoring within-parasha counts), and the ordinal-aligned MAIN 0.5843 between them. The alignment ambiguity spans floor to ceiling without ever crossing 0.6, so it cannot flip the verdict. (b) The non-verdict SECONDARY clause also held: per-section bearing-rate ratio 1.829 (std 10.5% of 769 simanim vs Buber 5.8% of 1042), outside the predicted 0.67-1.5 band — both registered clauses landed. (c) CALIBRATION CONTEXT: this SUPPORT breaks the recent three-kill run of direction-right/magnitude-overshot resolutions, and it refines that finding: the three overshoots were all bold RATIO-MULTIPLES (>=3:1, >2.0, <two-thirds), whereas the threshold that held here is a PROPORTION (Jaccard < 0.6). The signal is that the noetome over-pins multiplicative magnitudes specifically, not all quantitative thresholds. (d) Incidental structural finding, no bearing on the verdict: every formula hit in BOTH recensions falls in Genesis-Numbers — zero yelammedenu openings anywhere in Deuteronomy in either recension (141 instances >= 50, so the registered void rule was not triggered).

Registered against Sefaria's open Midrash Tanchuma (standard) and Midrash Tanchuma Buber (verified-real Sefaria indices). PREDICTION VERBATIM: Prediction: string-counting the "ילמדנו רבנו" (yelammedenu rabbenu) opening formula across Sefaria's Midrash Tanchuma (standard) and Midrash Tanchuma Buber recensions, the per-section occurrence rate will differ between the two recensions by at least 50% (a rate ratio outside the band 0.67-1.5), and the Jaccard overlap of the two recensions' sets of formula-bearing homilies (matched by parasha and pericope) will fall below 0.6 - primary clause: the sub-0.6 cross-recension overlap of yelammedenu homilies; the verdict follows it. The test voids for coverage if fewer than 50 formula instances are found across both recensions combined.

Resolution criteria — the registered fine print

Resolution criteria: DENOMINATOR (verbatim): the two recensions' SETS of formula-bearing homilies (homilies opening with 'ילמדנו רבנו'/yelammedenu rabbenu), matched by parasha and pericope. DATA: fetch Sefaria's Midrash Tanchuma and Midrash Tanchuma Buber; string-match the yelammedenu-rabbenu opening; per section record whether it bears the formula; align by (parasha, pericope). PRIMARY R = Jaccard overlap of the two formula-bearing-homily sets. SECONDARY (non-verdict): per-section occurrence-rate ratio between recensions (predicted outside 0.67-1.5). Report total formula instances found. CLAUSE PRECEDENCE: (1) INCONCLUSIVE_BY_DESIGN (the conjecture's void rule) if fewer than 50 formula instances across both recensions combined; (2) SUPPORTED if Jaccard < 0.6; (3) REFUTED if Jaccard >= 0.6. Compute firewalled from threshold and direction.

Known-priors disclosure — what the registrant already knew

Known priors disclosure: Triage (2026-07-17) graded ADJACENT: Tanhuma-Yelammedenu recension divergence is textbook (Buber 1885; Bregman 2003) but the formula-rate + parasha-aligned Jaccard over Sefaria is un-run. The <50-instance void is a live risk. Direction open. CALIBRATION NOTE: the <0.6 Jaccard is a bold pin — recent resolutions show conjectures over-pin magnitudes; the overlap may land above 0.6.

Method and dataset — how it was measured

Register-before-compute: prediction and resolution criteria registered verbatim in docs/generated/conjecture_prediction_batch5_20260718.json (2026-07-18) before any measurement. Firewalled compute by a separate agent (claude-sonnet) blind to threshold and direction, no conjecture files consulted: both recensions fetched from the Sefaria API (113 parasha-node text calls, all HTTP 200, sha256 manifests recorded); Hebrew text normalized (HTML tags, bidi controls, nikud/cantillation stripped, maqaf preserved); formula matched by regex — yelammedenu then 1+ of {whitespace, maqaf, hyphen} then rabbenu/rabbeinu (optional medial yod) — over joined paragraphs per siman, siman = homily-level section. Alignment: MAIN = 53 shared parshiot with within-parasha ordinal pairing of formula-bearing siman lists (intersection 52 / union 89), unshared nodes (std-only Vayeilech, 6 Buber-only Appendix nodes) union-only; sensitivities = strict literal (parasha, siman-index) match (17/124) and parasha-level presence-only (19/34). Counting logic run twice against two independently fetched corpus copies, 0 per-siman mismatches. computed_at 2026-07-19T09:31:00Z postdates registered_at (batch5 registration, 2026-07-18).

Dataset: Sefaria Midrash Tanchuma (std) vs Tanchuma Buber; 141 yelammedenu-rabbenu instances; formula-bearing-homily set Jaccard

computed 2026-07-19

Weigh in

No community feedback yet.

New here? Create an account first

Create an account or sign in and your feedback is tied to you — you can track it, get replies, and claim this conjecture so others know you’re working on it. Prefer not to? Just leave your take below as a guest — only the name you type is shown.

Add your take

Posted immediately (spam is removed). Community feedback is never an adjudicated verdict and never changes this conjecture's triage label or status above.