The strongest conclusion at present is also the least cinematic: Bardeen and Brattain really were running an experimental program that changed in response to unexpected results, and Bell really did inherit a deep body of semiconductor craft from the wartime detector program. Those two facts can coexist with important gaps in the later origin story.
What now looks ordinary
High-back-voltage germanium no longer looks like a material that appeared mysteriously just in time. Scaff and Theuerer's wartime work describes controlled germanium purification, donor impurities, high-back-voltage n-type regions, point-contact rectifiers and stabilization procedures well before December 1947. The later Bardeen–Brattain transistor patent explicitly points back to that process lineage.
Likewise, the idea of putting two contacts close together did not suddenly spring into existence on December 16. Brattain's December 4 notebook already records close two-point configurations under consideration. That does not settle why the successful geometry was chosen, but it changes the shape of the question.
What remains genuinely awkward
Shockley's January 23 junction-transistor note deserves more attention than the phrase “brilliant theoretical anticipation” gives it. It precedes Shive's later experiment that made bulk transport dramatically visible, was developed privately, and was not immediately shared with the wider group. The right next step is not suspicion by default; it is a complete ancestry of the assumptions and geometry on the page.
A second awkward boundary is interpretive. Bardeen and Brattain's December reasoning leaned strongly on surface effects. Their mature bulk minority-carrier explanation belongs to a later information state, after more experiments and after Shockley's work. A later correct mechanism cannot be used as evidence for what guided an earlier choice unless the earlier record says so.

The hypothesis map
A Bell-local developmental origin has substantial support: there is a real chain of experiments, mistakes and theoretical work. Ordinary wartime-network inheritance is even more strongly documented than many popular accounts imply, especially for material preparation and detector practice. Restricted technical transfer remains possible but currently lacks a distinctive transfer artifact. A complete working-device or blueprint transfer is weakened by the messy experimental record and by later evidence that important materials knowledge was still being learned. Retrospective compression and legal/publicity curation are plainly real. Deliberate origin laundering is not established.
A replacement story is not required for a provenance gap to remain unresolved.
What would change the picture
The most valuable contrary evidence would be a pre-1947 restricted report, sample specification, procedure or measurement containing an obscure transistor-relevant fingerprint before Bell's notebooks do—and a plausible channel by which it reached the relevant person. Conversely, dense contemporaneous derivations and failure records leading into the successful choices would strengthen the ordinary-development model.
- Scaff & Theuerer, US2602211A — wartime germanium rectifier process, priority 1945.
- Bardeen & Brattain, US2524035A — continuation-in-part and later point-contact description.
- Brattain notebook, Dec. 4, 1947 — direct notebook reproduction.
- Computer History Museum: Shockley's junction conception.
- Purdue post-war semiconductor history — an interested institutional source, useful but not neutral.