No, Nick Szabo wasn’t Satoshi Nakamoto in 2014 either

I’m presently writing an albatross of a book review (update: it’s out!), and currently have six thousand words of ranting on the topic, with a few thousand more to go. The book fingers Nick Szabo as the most likely candidate for Satoshi Nakamoto. So I thought I’d share my pain.

The book cites a 2014 Coindesk article, “Linguistic Researchers Name Nick Szabo as Author of Bitcoin Whitepaper” — which cites research from Jack Grieve at Aston University’s Centre for Forensic Linguistics, and a blog post by someone calling themselves Skye Grey.

In the blockchain world, if you see an amazing headline that would turn the world upside-down were it true … the first thing you should do is track it to its source.

Why Nick Szabo?

I didn’t cover the case for Szabo as Satoshi in detail in chapter 6 of the book — just Dorian Nakamoto and Craig Wright. Though I probably should have.

Szabo was one of the original cypherpunks. He’s an anarchocapitalist, and has long been interested in the idea of decentralised digital money. He wrote the bit gold proposal, which outlines how you might achieve such a thing. It’s about half way to Bitcoin — though with no code, just the idea.

Satoshi hadn’t heard of Szabo when he started working on Bitcoin in 2007 — he was only alerted to Szabo’s work by Wei Dai, whose b-money proposal was a direct influence. As Wei Dai puts it:

Nick considers his ideas to be at least an independent invention from b-money so why would Satoshi say expands on your ideas into a complete working system to me, and cite b-money but not Bit Gold in his paper, if Satoshi was Nick? An additional reason that I haven’t mentioned previously is that Satoshi’s writings just don’t read like Nick’s to me.

He doesn’t read like it to me either — each has a very definite personal writing style, and they just aren’t the same. I’ve never seen Satoshi not write like Satoshi, and I’ve never seen Szabo not write like Szabo.

Also, Szabo denies being Satoshi. Not that that ever stopped others’ pareidolia.

Aston University

The Aston University story got a fair bit of press at the time. With all this fuss, you might expect a published paper or something — but there only seems to be a university press release.

The press release does not link any research outputs — the “study” turns out to have been an exercise that Jack Grieve at Aston gave his final-year students.

This was then written up as a splashy bit of university press-release-ware — which is still routinely used by lazy authors to claim that Szabo is Satoshi, ‘cos linguistic analysis apparently says so.

The “tells” that Grieve’s group cites in the white paper are phrases like “trusted third party” — which is common in cryptography, and in discussions of electronic money since at least the 1990s.

Grieve’s group doesn’t show signs of knowing the topic area — at best, they’ve shown that Szabo knows the technical terms surrounding the issue, which we already knew.

More telling, though, is that I’m pretty sure they didn’t know computer science or computer scientists:

Furthermore, the researchers found that the bitcoin whitepaper was drafted using Latex, an open-source document preparation system. Latex is also used by Szabo for all his publications.

The white paper says it was written in OpenOffice 2.4 — look at the PDF properties. And the fonts are Times New Roman, Arial and Courier New — Microsoft Windows fonts, as you might expect from a Windows C++ programmer. You can see the font names in the binary file, if you do strings bitcoin.pdf | grep -i font .

It’s possible that Satoshi could have carefully faked the origins of the white paper. But I know of no reason to assume this, and the Aston University press release does not detail why the team thought it was written in LATEX.

And, you know, the reason Szabo uses LATEX for his own publications is probably that pretty much every computer scientist does.

The students don’t deserve a public earful for this, of course — it was just a class assignment. Jack Grieve and the Aston University press office really do, though.

Skye Grey

“Skye Grey” seems to be a pseudonym. Their entire visible output is two blog posts about the identity of Satoshi: “Satoshi Nakamoto is (probably) Nick Szabo” (1 December 2013) and “Occam’s Razor: who is most likely to be Satoshi Nakamoto?” (11 March 2014).

Nobody would have paid much attention — except the first post was covered in TechCrunch a few days later, including an interview with Grey.

Grey’s evidence is that Szabo uses phrases that are in the Bitcoin white paper, but don’t appear with great frequency in cryptography papers they found on Google Scholar.

The problem comes when Grey cites evidence against Szabo being Satoshi as evidence for Szabo being Satoshi:

  • Szabo uses American English, not Commonwealth English. So the paper must have had multiple authors!
  • Satoshi didn’t know about Szabo’s bit gold proposal. So Szabo must have deliberately not mentioned it, to throw people off the scent!
  • Szabo didn’t react publicly to Bitcoin until a few months after the white paper was posted. So Szabo must have deliberately not reacted, to throw people off the scent!

When someone tries to use evidence against their theory as evidence for it — you’ve got a conspiracy theorist.

So who is Satoshi, then? Huh?

I can’t see how Satoshi’s identity matters.

Bitcoin exists — it’ll linger as long as there’s a copy of the software, a copy of the blockchain and two or more interested enthusiasts. I expect there’ll be something called “Bitcoin,” clearly descended from the present software and blockchain, for decades to come. Even if it can’t be traded easily to actual money or useful things.

(Of course, should it lose hash power, there’s all that surplus mining hardware to worry about attacks from.)

If anyone claims to be Satoshi, they just have to move a Satoshi coin, or provide similar cryptographic evidence. If they can’t do that, or if they claim they have lots of other evidence but not that — they’re trying to con you.



Become a Patron!

Your subscriptions keep this site going. Sign up today!

6 Comments on “No, Nick Szabo wasn’t Satoshi Nakamoto in 2014 either”

  1. Jorge Stolfi notes on Reddit:

    “A common problem in those attempts to identify Satoshi is that they all start by assuming that he was, not only a cypherpunk, but a well-known cypherpunk.

    Imagine the police trying to catch a murderer by comparing his fingerprints to those of Charles Manson, O. J. Simpson, Ted Kaczinsky, and Ted Bundy, and closing with the best match among those.”

  2. This reminds me of the perennial discussions about whether Jesus really existed and on the authorship of Shakespeare’s works. No matter the outcome, the gospels, churches and believers will remain, for better or worse. And somebody wrote All’s Well That Ends Well and Macbeth, so that person is for all intents and purposes the one we refer to with “Shakespeare”, even if they had a different name.

  3. OpenOffice almost certainly used “pdflatex” under the hood to generate PDFs. Many tools do, such as Doxygen, the docs-from-code generator. Clicking “export as PDF” on a menu is (fortunately for almost all of us!) not anything that actually requires knowledge of LaTeX.

    So “he used LaTeX” is more like “he used an open-source word processor, which used more open-source tools to output a PDF.”

    1. I’m pretty sure OpenOffice 2.4 didn’t have pdflatex anywhere internally by default, and instead turned its printer output into a PDF. Apache OpenOffice and LibreOffice don’t go via pdflatex either. There’s nothing LaTeX-y about the bitcoin white paper at all, as far as I can tell.

      1. Yeah, you’re right! I hadn’t realized they decided to reinvent the wheel with their own PDF writer. I should’ve known…

        Maybe the students saw some Postscript/PDF-type slash commands in the PDF source and one said “oh, I’ve seen those before; they’re LaTeX.” Weak sauce no matter what.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.