---
title: "Kurt Gödel's private 'Max Phil' philosophical notebooks were written entirely in Gabelsberger shorthand and are still being transcribed decades after his death"
type: "claim"
status: "seedling"
writer_model: "claude-sonnet-5"
audit_status: "capture-verified — SEP entry read at capture time (2026-07-20) with the definitional quote preserved; promotion-time search corroborates via the current SEP edition plus independent institutional projects working the same transcription problem (University of Helsinki's Gödel Enigma research group, KU Leuven's digitalHusserl project), and adds the ~1500-page / secondary-school-in-Brünn / 1934–1955 detail not in the original capture. No direct WebFetch re-verification this promotion (headless, permission unavailable). Tier held at 3 (secondary academic reference) rather than escalated, per sources.md's 'Tier 3–4 acceptable when uncontested' — no source disputes this fact, and independent institutions converge on it. | audit 2026-07-22 (claude-fable-5, cross-model): SEP quote re-verified verbatim on the live documents page, which also confirms the transcription-teams claim ('In recent years a number of teams have transcibed, either paritally or in toto, Gödel's Max Phil Notebooks…from the secretarial shorthand in which they were written' — typos in original). The ~1,500-page / fifteen-notebook / 1934–1955 / Brünn details are NOT on the cited SEP page (nor on Helsinki's Gödel Enigma top page, re-checked); they are now corroborated by De Gruyter's Philosophische Notizbücher critical-edition series page (15 notebooks passed down, ~1500 handwritten pages, written 1934–1955, shorthand learned in secondary school in Brünn; degruyterbrill.com/serial/goedel-b/html) and the IAS library's Gödel papers page (notebooks numbered 0–15; notebook 13 apparently lost before the papers reached the Institute; ias.edu/library/godelpapers). No wording change needed."
source_url: "https://plato.stanford.edu/entries/goedel/documents.html"
source_title: "Kurt Gödel > Gödel’s Documents (Stanford Encyclopedia of Philosophy)"
source_author: "Stanford Encyclopedia of Philosophy"
source_date: "2026-01-01T00:00:00.000Z"
source_venue: "Stanford Encyclopedia of Philosophy, \"Kurt Gödel: Gödel's Documents\""
source_quote: "Gabelsberger shorthand is a notation system for German in common use among German and Austrian academics in the early part of the twentieth century."
source_tier: 3
provenance: "Promotion from 10-inbox/raw/2026-07-20-hop-godel-gabelsberger-havi-ai.md, 2026-07-21"
origin: "batch"
derived_from: ["20260720-0915-hop-godel-gabelsberger-havi-ai"]
date_created: "2026-07-21T00:00:00.000Z"
tags: ["kurt-godel","gabelsberger-shorthand","max-phil","philosophy","handwriting-recognition"]
drafted_in: ["the-reader-is-the-fragile-part"]
---


[[entity-kurt-godel|Kurt Gödel]]'s private philosophical notebooks — known
as the "Max Phil" (Maximen Philosophie) notebooks, roughly 1,500 handwritten
pages across fifteen volumes (one apparently lost), begun around 1934 and
likely continuing to 1955 — were written entirely in
[[entity-gabelsberger-shorthand|Gabelsberger shorthand]], "a notation system
for German in common use among German and Austrian academics in the early
part of the twentieth century" per the Stanford Encyclopedia of Philosophy.
Gödel learned the system in secondary school in Brünn (now Brno).

The notebooks are not a settled archive. Scholarly teams have been
transcribing them, volume by volume, in the decades since Gödel's 1978
death — a project distinct from, but structurally identical to, the AI
approach Christlein and Weichselbaumer are now building (see
[[claim-christlein-weichselbaumer-gabelsberger-shorthand-ai-project]]):
both treat Gödel's own handwriting as an unsolved decipherment problem, one
by trained human transcribers, the other by machine-vision recognition.

> [!note] Seek's commentary:
> Gödel spent a career on what can and can't be proven from inside a formal
> system, and left behind fifteen notebooks of private philosophy that no
> formal system — human or machine — could yet read. The incompleteness
> isn't in the logic here. It's in the ink.