Satoshi inquiry · Working hypothesis · Opened 7 Aug 2026

THE MICHAEL PEIRCE FILE

"A payments academic who wrote the encyclopedia, then went quiet."

STATUS: PROMISING PROFILE EVIDENCE GRADE: CIRCUMSTANTIAL ONLY NEVER PUBLICLY NAMED — KEEP IT THAT WAY

The Profile (public record only)

Michael Peirce — Dublin. PhD in computer science, University of Dublin, Trinity College: Multi-Party Electronic Payments for Mobile Communications, dated October 31st, 2000. Co-author of the standard pre-Bitcoin reference works on electronic payment systems (1997; 2nd ed. 2001) and of award-winning micropayment papers built on one-way hash chains. Maintainer of the era's best-known online catalog of digital cash systems. Public research trail fades after ~2001–02.

⚑ The thesis is dated 31 October — the whitepaper arrived 31 October 2008, eight years to the day. Filed as coincidence-grade until proven otherwise (Halloween is also just an academic deadline).

Measured Stylometry — 232-page thesis vs the Satoshi corpus

Counts are machine-measured from the full extracted thesis text (794,162 characters). Satoshi reference features are from the whitepaper and known post/email corpus.

Dimension	The thesis (measured)	Satoshi	Read
NATIONAL ENGLISH	optimise 53:2 · behaviour 7:0 · authorisation 38:7 · analyse 10:0 · favour 2:0	British throughout the forum/email corpus ("optimise", "favour", "bloody hard")	MATCH — native, not affected
REGISTER	contractions: 0 (can't 0 / cannot 93 · don't 0 · it's 0)	Whitepaper: zero contractions, formal and spare	MATCH (weak alone — academic norm)
VOICE	sentence-initial "We" 193 : "I" 7	Whitepaper written entirely in "we"	MATCH (weak alone — academic norm)
DOMAIN DNA	hash chain ×109 · electronic cash ×33 · digital signature ×30 · double spending ×26 · trusted third party ×2 · peer-to-peer ×2	The whitepaper's entire conceptual vocabulary	STRONG — he lived in Bitcoin's problem space seven years early
KNOWLEDGE CANON	Chaum ×16 · Rivest ×13 · PayWord ×17 · MicroMint ×63 — and hashcash ×0 · b-money ×0 · Wei Dai ×0 · Adam Back ×0	Satoshi demonstrably did NOT know b-money and had to ask how to cite hashcash	INTRIGUING — academic canon, not cypherpunk canon. Matches Satoshi's exact blind spots: he knew the formal literature and had to be pointed at the mailing-list schemes
THE BOOK TEST (round 2 — 7 Aug 2026)	2001 2nd edition, searched via Google Books: hashcash ×0 · Wei Dai ×0 — control term PayWord found on 14 pages, proving search-inside works on this volume	Satoshi had to be told hashcash existed (Aug 2008, by Back) and had to ask Dai for b-money's date	BLIND-SPOT MATCH STRENGTHENED — the definitive academic payments survey, seven years before the whitepaper, does not know the cypherpunk canon. The academic and mailing-list literatures were genuinely disjoint, and Satoshi's citation behavior is that of someone from the academic side
THE HYPHEN TELL	"double spending" unhyphenated — 26:0	"double-spending" hyphenated — consistently, including the abstract	MISMATCH — the single most diagnostic shared term is written with different muscle memory
COMPOUNDING	micropayment 355:4 over micro-payment · timestamp 4:1 over time-stamp	"timestamp" solid throughout	minor match
TIMEZONE	Dublin — UTC+0	Quiet hours 05:00–11:00 UTC = US-Eastern day sleeper OR a GMT night owl	COMPATIBLE — the night-owl reading reconciles the clock with the British spelling
PLATFORM (round 3 — 7 Aug 2026)	Thesis benchmarks run on a 400MHz Pentium II, 196MB RAM, Windows NT 4.0 SP3. OS mentions across 232 pages: Windows ×5 · Linux ×0 · Solaris ×0 · Unix ×1	Windows-first — "the great cypherpunk anomaly" that argues against every Unix-native candidate	MATCH — a Windows-native researcher in 2000 academia, where Solaris/Linux was the default. One of the strongest profile points found so far
LANGUAGES (round 3)	1995: PayMe e-cash prototype "implemented in a C++/Unix environment" on a Sun cluster (TCD-CS-96-21, VERIFIED verbatim) — coins, wallets, banks, signed receipts, double-spend DB. 2000: thesis prototype "implemented in Java" (RMI, JavaCard, J/Crypto); compiled & benchmarked Crypto++ v3.1 — "[Dai99] W. Dai. Crypto++…"	Windows-first C++98, self-contained	C++ HANDS CONFIRMED — he personally built a complete pre-Bitcoin electronic-cash system in C++ in 1995. Later Java-first, but the capability match is real, not inferred. He knew Dai's NAME in 1999 — as a library author, which does not imply knowing b-money. Bitcoin used OpenSSL, not Crypto++ — neutral
IDENTIFIER STYLE (round 3)	His protocol/message classes: BuyRequest, BuyReply, RedeemReply, PayEndorse ×23, PassMessageOn ×12, ChainCore ×6, UserVendorBroker — PascalCase compounds	PascalCase everywhere; C-prefixed classes (CTransaction) + n/f Hungarian	PARTIAL — PascalCase compounding matches; the C-prefix/Hungarian habit can't be tested from Java-style naming
FORMATTING HABITS	No verbatim source code in the thesis — appendices are UML class diagrams, not listings	4-space, Allman, spaced keywords/operators, error()	STILL UNTESTED — braces, indent, spacing remain the missing fingerprint
OPSEC / ARC	Public research trail fades ~2001–02; co-authors stayed publicly visible (one later engaged with Bitcoin openly)	Total anonymity, 15+ years; ~1.1M BTC untouched	NOT DISTINCTIVE (corrected 7 Aug 2026) — a low public footprint is cohort-normal for era academics who left research; treating it as Satoshi-like was an over-read. Carries no weight either way

Exhibits — verbatim from the thesis

"…with each entity requiring an X.509 certificate, issued by a trusted third party."

"The bank maintains a database of the serial numbers of all spent coins, to prevent double spending of tokens."

"…the bank must be contacted online during each purchase to prevent double spending. Recently, much work has been done to extend and improve on this basic scheme."

Read those against the whitepaper's opening argument: Bitcoin exists precisely to delete the bank, the certificate authority, and the online spent-coin database from those sentences. The thesis describes, in 2000, the exact architecture the whitepaper was written to destroy — from inside it.

Honest Verdict — recalibrated 7 Aug 2026

Classification: HIGH-INTEREST UNRESOLVED, not probable. Two evidence classes, kept separate. Capability evidence — strong: a verified 1995 C++ e-cash implementation (PayMe), a full Java micropayment system, the encyclopedic survey work, British English, and a knowledge blind-spot pattern matching Satoshi's documented one. Calling this record "zero" was wrong — it is genuine implementation history in the precise problem domain. Individualized forensics — thin: no attributable personal source corpus to style-match, no 2008-era artifact, no mailing-list presence, a failed prose round, and a hyphen mismatch now stable across three of his documents. Capability match is not identification — profile-fit is what every wrong answer in this case has ever had. The 1995 PayMe source (or full project report) is the single highest-value artifact. Keep working it.

What Would Actually Move This

[TOP TARGET] The 1995 PayMe C++ source or full B.A. project report — the single highest-value artifact on this board. It would carry his personal C++ formatting fingerprint (braces, indent, keyword spacing, naming) from his own hands, era-adjacent to Satoshi's habits. Candidate locations: TCD undergraduate project archives, Wayback captures of his pages, WWW4 conference archives, PgpTools-era Unix software collections.
Any code sample — PARTIALLY DONE (7 Aug 2026): the thesis contains no verbatim source (appendices are UML diagrams), but it establishes his stack: Java prototype (RMI/JavaCard/J-Crypto), C++ benchmarking via Wei Dai's Crypto++ v3.1, all on Windows NT 4.0; and the 1995 PayMe paper verifies earlier C++ implementation work. Still wanted: actual source — PayMe, the Java prototype, university archive tarballs. MANUAL LEAD: browse web.archive.org for ganges.cs.tcd.ie/mepeirce/ (his digital-cash pages, ~1997–2002) — blocked from my tooling, open to a human browser.
Canon refinement (7 Aug 2026): the thesis DOES cite two cypherpunk-adjacent names — "[Dai99] W. Dai, Crypto++" (as a library, not b-money) and "[Sza96] N. Szabo, The mental accounting barrier to micropayments" (the published essay). The blind-spot pattern holds precisely: he knew the citable, published layer of that world and shows zero contact with the mailing-list layer (hashcash, b-money) — exactly the boundary Satoshi's own citation behavior draws.
The book test — DONE (7 Aug 2026): the 2001 second edition contains zero mentions of hashcash and zero of Wei Dai (control term verified). The blind-spot match strengthened. Caveat: this describes the entire academic-payments community, not one man — it widens "Satoshi came from the academic side," it does not individuate.
2008–2011 incompatibility check — any verifiable professional life during the Satoshi window that couldn't coexist with Satoshi's output volume would falsify this cleanly. Falsification is progress.
Prose overlap, round two — sentence-level comparison of thesis Chapter 1 against whitepaper sections 1–2 (rhythm, connectives, argument structure), and against the 1999 papers as controls.
The co-author control — every knowledge-based point above applies equally to two other people. Any evidence that doesn't distinguish him from his co-authors isn't evidence about him.

Rules of This File

Public documents only: published papers, books, archived public webpages. No personal-life digging, no contact, no same-name conflation (multiple unrelated people share this name).
The bar for ever going further than this file: evidence that survives the Dorian test — a profile is not an accusation.

Sources

TCD tech report TCD-CS-2001-26 (the thesis, publications.scss.tcd.ie) — full text extracted & measured 7 Aug 2026
TCD tech report TCD-CS-1999-48 (multi-party payments paper)
Electronic Payment Systems (Artech House, 1997; 2nd ed. 2001)
Satoshi reference corpus: whitepaper (bitcoin.org/bitcoin.pdf), cryptography mailing-list posts (Nov 2008), published Hearn correspondence (2009–2011)
Stylometry scripts & raw counts: session scratchpad (stylometry.js, thesis.txt)