It’s not a good time to be Newsweek. After originally reporting that Dorian Satoshi Nakamoto was Bitcoin’s creator, Nakamoto reportedly refuted the claim, saying that he has nothing to do with the cryptocurrency. Unfortunately, if Forbes‘ commissioned linguistic analysis is any indication, things won’t get easier for Newsweek and the reporter who broke the story, Leah McGrath Goodman.
Forbes’ tapped analysis firm Juola & Associates, which specializes in stylometry, a technique that compares people’s patterns of language use in text as a way to identify the original author of any document. Developed by Duquesne University associate professor Patrick Juola, the technique looks at whether someone uses long or short words, as well as what the most common words used are. This technique was used to identify J.K. Rowling as the author of The Cuckoo’s Calling, a book which was written under a pseudonym.
The firm’s chief scientific officer, John Noecker Jr., attempted to identify Satoshi Nakamoto by comparing writing samples posted online by Bitcoin’s creator to that of Michael Clear, Neal J. King, Shinichi Mochizuki, and Vili Lehdonvirta, all of whom were identified as the possible creator of Bitcoin. Noecker used average word length, common words used, groups of consecutive words, and parts of speech they had a tendency to use.
King was identified as the better match, though Noecker won’t assume that King is the Satoshi Nakamoto since we don’t know if those men are the only candidates. After adding Dorian Nakamoto’s writing samples, which include letters sent to model train magazines, an email sent to his local government, and message board postings, King was still the better match.
As for the possibility that someone else did the writing and Dorian Nakamoto simply provided an outline, Noecker admitted that, while that could have a significant effect on the final results, the firm can still identify both the editor and the author. “If it was edited really significantly, if Dorian had written up an outline and someone else had followed it, then it is possible that could ‘wash out’ the effects,” said Noecker.
McGrath Goodman responded to Forbes’ inquiry via Twitter, saying that there is “NOT good enough data for good statistical analysis” and that “my understanding of stylometry is that short passages do not yield great analysis.”
Either way, we don’t see this saga coming to an end anytime soon.