{"id":11369,"date":"2026-03-24T11:49:34","date_gmt":"2026-03-24T10:49:34","guid":{"rendered":"https:\/\/www.ulclegal.com\/?p=11369"},"modified":"2026-03-27T09:36:38","modified_gmt":"2026-03-27T08:36:38","slug":"satoshi-nakamoto-found-cypherpunk-bitcoin","status":"publish","type":"post","link":"https:\/\/www.ulclegal.com\/en\/satoshi-nakamoto-found-cypherpunk-bitcoin\/","title":{"rendered":"Satoshi Nakamoto Found? A Blind Search Across 104,901 Cypherpunk Archive Data Reveals the Writing Behind Bitcoin"},"content":{"rendered":"\n<h2 class=\"wp-block-heading has-text-align-center\">A Blind Stylometric Search Across 12,979 Authors and 104,901 Texts<\/h2>\n\n\n\n<p class=\"has-text-align-center\">with Multi-Baseline Analysis, Legal Motive Verification, and Candidate Counter-Evidence Assessment<\/p>\n\n\n\n<p class=\"has-text-align-center\">Prepared by JUDr. Mag. J\u00e1n \u010carnogursk\u00fd, MBA<\/p>\n\n\n\n<p class=\"has-text-align-center\">(ulclegal.com)<\/p>\n\n\n\n<p class=\"has-text-align-center\">March 2026<\/p>\n\n\n\n<p><em>Original research. All scripts, corpora, and data available on request.<\/em> <em>The relevant stylometry results and sources are in the <a href=\"https:\/\/ulclegal.com\/ADDENDUM_ALL_RESULTS.txt\" target=\"_blank\" rel=\"noopener\">Appendix<\/a><\/em>.<\/p>\n\n\n\n<h1 class=\"wp-block-heading\">1. Executive Summary<\/h1>\n\n\n\n<p>This report presents the results of the <strong>largest stylometric attribution study ever conducted<\/strong> on the question of who created Bitcoin under the pseudonym <strong>Satoshi Nakamoto<\/strong>. Using custom-built software, we analysed the writing patterns of <strong>104,901 unique email authors<\/strong> across two mailing lists \u2013 the <strong>Cypherpunks<\/strong> list (12,708 authors, 92,193 emails) and the Cryptography list at <strong>metzdowd.com<\/strong> (271 authors) \u2013 and <strong>compared<\/strong> their writing fingerprints to Satoshi\u2019s 68,000 words of known writing.<\/p>\n\n\n\n<p><strong>What is stylometry? <\/strong>Every person has <strong>unconscious habits<\/strong> in how they use language: which small words they prefer (\u2018the\u2019, \u2018of\u2019, \u2018and\u2019), how long their sentences tend to be, whether they use British or American spellings. These patterns are as <strong>distinctive<\/strong> as a fingerprint and remarkably <strong>stable<\/strong> across different topics. Stylometry measures these patterns <strong>mathematically<\/strong>. Courts have accepted stylometric evidence in fraud and authorship dispute cases.<\/p>\n\n\n\n<p><strong>Our principal stylometry findings:<\/strong><\/p>\n\n\n\n<p><strong>(1) Ray Dillinger is the strongest overall match. <\/strong>From 104,901 authors, Ray Dillinger \u2013 the cypherpunk who reviewed Bitcoin\u2019s source code before its launch in January 2009 \u2013 scored the <strong>lowest distance from Satoshi\u2019s<\/strong> writing on our validated statistical test (Delta 0.77 on a scale where below 1.0 means \u2018same author\u2019). He also <strong>matches Satoshi\u2019s exclusive<\/strong> use of the word \u2018cannot\u2019 (never \u2018can not\u2019), with <strong>86% double-spacing<\/strong> after full stops (Satoshi: ~81%), and scores in the \u2018same author\u2019 range on <strong>all four forum<\/strong>-based baselines.<\/p>\n\n\n\n<p><strong>(2) The whitepaper was likely not written by the same person as the forum posts. <\/strong>When we used the Bitcoin whitepaper as a separate baseline, a completely different set of candidates ranked highest than when we used the forum posts as baseline. There is <strong>zero overlap <\/strong>between the top three candidates for each baseline. The <strong>whitepaper&#8217;s writing<\/strong> style is statistically closest to <strong>Craig Wright&#8217;s<\/strong> post-2016 blog posts and <strong>Nick Szabo&#8217;s<\/strong> academic writing, while the forum posts match the cypherpunks group (<strong>Dillinger, Gutmann, Back, Rosing<\/strong>). This represents a notable shift from our earlier <a href=\"https:\/\/www.ulclegal.com\/buletin\/bitcoin-turns-15-today-who-is-satoshi-nakamoto-can-stolen-bitcoins-be-returned-to-the-owner\/\">ulclegal.com article<\/a> (October 2024), where the Naive Bayes method ranked <strong>Phil Wilson<\/strong> as the closest match to both the whitepaper and Satoshi&#8217;s emails. The change results from the different methodology: Naive Bayes classification with the <strong>50 most common<\/strong> Satoshi words is <strong>sensitive to topic overlap<\/strong> (Wilson writes extensively about Bitcoin&#8217;s technical design, as does Satoshi), while <strong>Burrows&#8217; Delta<\/strong> measures <strong>unconscious function word patterns<\/strong> that are <strong>independent<\/strong> of topic. When the comparison shifts from <strong>&#8216;what you write about&#8217;<\/strong> to <strong>&#8216;how you unconsciously structure sentences<\/strong>&#8216;, Wilson moves from first place to the uncertain boundary (Delta 1.01), and the cypherpunks mailing list authors \u2013 who were not included in our earlier study&#8217;s candidate pool \u2013 emerge as the strongest matches. Both methods have merit, but <strong>Delta<\/strong> is considered <strong>more robust<\/strong> against topic contamination in the stylometry literature.<\/p>\n\n\n\n<p><strong>(3) Adam Back has a notable inconsistency. <\/strong>Back scored 0.91 (same author range) on <strong>function word analysis<\/strong> and has the strongest <strong>circumstantial<\/strong> profile: Scottish, 12 British spellings, inventor of Hashcash. However, he writes \u2018<strong>can not<\/strong>\u2019 (two words) 17 times versus only 4 times \u2018cannot\u2019 \u2013 the <strong>opposite of Satoshi<\/strong>, who exclusively uses \u2018cannot\u2019 (12 times, never \u2018can not\u2019). This does not exclude Back, but it represents a <strong>measurable stylistic inconsistency<\/strong> that other top candidates (Dillinger, Gutmann, Finney) do <strong>not<\/strong> share. He also stated in 2013 that he was <strong>still learning basic<\/strong> Bitcoin concepts on IRC, logged and witnessed by developers.<\/p>\n\n\n\n<p><strong>(4) Nick Szabo is not the forum Satoshi. <\/strong>Szabo ranked 24<sup>th<\/sup> out of 29 candidates against the forum posts (Delta 1.33). His writing is closer to the <strong>whitepaper\u2019s academic<\/strong> register but <strong>far from the conversational <\/strong>Satoshi. This <strong>contradicts<\/strong> the 2014 Aston University study which used a much smaller candidate pool. Our 210,000-word Szabo corpus \u2013 scraped from his Unenumerated blog \u2013 provides a <strong>far more reliable<\/strong> comparison.<\/p>\n\n\n\n<p><strong>(5) Three previously proposed candidates were assessed. Phil Wilson&#8217;s<\/strong> pre-Bitcoin technical writing (scrontsoft.com) scores Delta 1.85 (&#8216;different author&#8217;). However, the scrontsoft.com archive consists primarily of <strong>readme files and code documentation<\/strong> \u2013 a fundamentally <strong>different genre<\/strong> from Satoshi&#8217;s conversational forum posts. Genre mismatch alone <strong>can inflate<\/strong> Delta by 0.3-0.5 points, so this result should be interpreted with <strong>caution<\/strong>. It <strong>excludes<\/strong> Wilson as the only <strong>forum post author<\/strong>, but does not rule out other forms of involvement. Notably, Wilson&#8217;s post-2016 narrative prose on <strong>vu.hn<\/strong> (60,614 words) scores <strong>Delta 1.01<\/strong> \u2013 on the boundary of <strong>&#8216;same author&#8217;<\/strong>. While this text was written after Satoshi&#8217;s style had been publicly documented, our BCNext mimicry analysis demonstrates that sustaining a convincing stylometric disguise across 60,000 words is <strong>extremely difficult<\/strong>: BCNext&#8217;s mimicry broke down on deep features (exclusive use of &#8216;can not&#8217;, article frequency) despite surface-level success. Wilson&#8217;s vu.hn text shows <strong>no such breakdowns<\/strong> \u2013 his &#8216;cannot&#8217; usage (34 times, never &#8216;can not&#8217;) <strong>matches Satoshi perfectly<\/strong>, and his <strong>function word distribution<\/strong> remains <strong>consistent<\/strong> throughout. This <strong>does not<\/strong> prove Wilson&#8217;s involvement, but it means his prose cannot be dismissed as easily as other excluded candidates. <strong>Sergey Ivancheglo<\/strong> (CFB) scores Delta 1.60, with involuntary Russian-language article underuse that cannot be faked across 134,000 words, which <strong>clearly disqualifies him<\/strong> as a Satoshi candidate. <strong>Craig Wright<\/strong> scores Delta 1.38 and was ruled &#8216;not Satoshi&#8217; by the UK High Court in March 2024 with <strong>extensive evidence<\/strong> of forgery.<\/p>\n\n\n\n<p><strong>(6) The UIGEA gambling motive explains Satoshi\u2019s extreme anonymity. <\/strong>Bitcoin\u2019s whitepaper appeared 12 days <strong>before the final<\/strong> regulations of the Unlawful Internet Gambling Enforcement Act (<strong>UIGEA<\/strong>). The Genesis Block was mined <strong>16 days before<\/strong> the UIGEA compliance deadline. Bitcoin&#8217;s original code contained <strong>GUI code for a poker client<\/strong> and for a <strong>peer-to-peer marketplace<\/strong>. This <strong>timeline and code<\/strong> inclusion provides the only candidate-independent explanation for why Satoshi employed <strong>extreme operational security<\/strong> from the very beginning \u2013 before Bitcoin had any users or monetary value.<\/p>\n\n\n\n<p><em>Methodological note: this study <strong>does not prove<\/strong> that any specific individual <strong>is<\/strong> Satoshi Nakamoto. Stylometric analysis measures statistical <strong>similarity, not identity<\/strong>. The absence of cryptographic proof (a signed message from Satoshi\u2019s known private keys) means all attribution remains <strong>probabilistic<\/strong>.<\/em><\/p>\n\n\n\n<h1 class=\"wp-block-heading\">2. How We Did It \u2013Methodology in Plain Language<\/h1>\n\n\n\n<h2 class=\"wp-block-heading\">2.1 The Writing Fingerprint<\/h2>\n\n\n\n<p>Imagine sorting every word a person has ever written and counting how often they use each one. Most people use the word <strong>\u2018the\u2019<\/strong> about 5-7% of the time, <strong>\u2018of\u2019<\/strong> about 2-3%, <strong>\u2018and\u2019<\/strong> about 2-3%, and so on. These percentages form a <strong>unique pattern<\/strong> \u2013 a writing fingerprint. Even when writing about completely different topics, a person\u2019s use of these small \u2018function words\u2019 stays <strong>remarkably constant<\/strong>, because these are <strong>unconscious<\/strong> choices. Nobody thinks about how often they use the word \u2018the\u2019 \u2013 but everybody uses it at a slightly different rate.<\/p>\n\n\n\n<p>We <strong>measured 50+ such function words<\/strong> in every text we analyzed, creating a numerical <strong>fingerprint<\/strong> for each author. We also measured <strong>several other features<\/strong>: <strong>double-spacing<\/strong> after full stops (a legacy typing habit from the typewriter era), <strong>British versus American<\/strong> spelling forms, <strong>contraction usage<\/strong> (\u2018don\u2019t\u2019 vs \u2018do not\u2019), and the specific choice between \u2018<strong>cannot\u2019<\/strong> and \u2018<strong>can not<\/strong>\u2019.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">2.2 Burrows\u2019 Delta \u2013 How We Measure Similarity<\/h2>\n\n\n\n<p>The core statistical tool we used is called <strong>Burrows\u2019 Delta<\/strong>, a standard method in computational stylometry <strong>since 2002<\/strong>. In simple terms, it works like this: first, we calculate how each author uses each function word <strong>compared to the average<\/strong> of all authors in our dataset. If an author uses \u2018the\u2019 <strong>much more<\/strong> than average, that\u2019s a <strong>notable feature<\/strong> of their writing. Then we compare two authors by adding up <strong>how different<\/strong> their word-usage patterns are across all 50+ function words. A small total difference (below 1.0) suggests the texts were likely written by the <strong>same person<\/strong>. A large total difference (above 1.5) suggests <strong>different authors<\/strong>. Scores between 1.0 and 1.5 are in the <strong>uncertain<\/strong> zone.<\/p>\n\n\n\n<p><em>Critical implementation detail: <\/em>the \u2018proper\u2019 version of this test requires comparing all authors simultaneously, because the statistical spread across the entire population is what gives the individual comparisons meaning. Earlier studies (and our own initial attempts) that compared only two authors at a time produced meaningless results \u2013 everything scored at the ceiling. Our proper <strong>multi-author<\/strong> implementation, validated by correctly identifying Satoshi\u2019s own texts as same-author, produces <strong>meaningful differentiation<\/strong>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">2.3 The Multi-Baseline Approach<\/h2>\n\n\n\n<p>Most prior Satoshi attribution studies used a <strong>single baseline<\/strong>: either all of Satoshi\u2019s writing combined, or just the whitepaper. We used <strong>five separate baselines<\/strong>, each representing a <strong>different segment<\/strong> of Satoshi\u2019s output: <strong>(1)<\/strong> all forum posts combined, <strong>(2)<\/strong> metzdowd mailing list emails from 2008-09, <strong>(3)<\/strong> BitcoinTalk posts from approximately 2009, <strong>(4)<\/strong> BitcoinTalk posts from approximately 2010, and <strong>(5)<\/strong> the Bitcoin whitepaper. This approach, which we developed from the dual-baseline method used in our earlier ulclegal.com article, reveals whether the same candidates rank <strong>highly across all baselines<\/strong> (supporting <strong>single authorship<\/strong>) or whether different candidates <strong>match different segments<\/strong> (supporting <strong>multi-authorship<\/strong>).<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">2.4 Self-Validation: Proving the Method Works<\/h2>\n\n\n\n<p>Before comparing any external candidates, we tested whether our method could <strong>correctly<\/strong> identify <strong>Satoshi within Satoshi\u2019s own<\/strong> texts. We split the forum posts into <strong>time-based<\/strong> segments and compared them against each other. The results <strong>confirmed<\/strong> the method <strong>works<\/strong>: Satoshi\u2019s BitcoinTalk 2009 vs 2010 scored <strong>Delta 0.31-0.39<\/strong> (correctly identified as same author), and the Satoshi\u2019s metzdowd emails scored <strong>0.83<\/strong> (same author, slightly different platform). The Whitepaper scored <strong>1.34<\/strong> against the forum posts \u2013 notably <strong>more distant<\/strong>, which raises the question of whether <strong>the same<\/strong> person wrote both.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">2.5 What We Searched<\/h2>\n\n\n\n<p>We searched two major mailing list archives: the <strong>Cypherpunks<\/strong> mailing list (92,193 emails from 12,708 unique authors, spanning 2000\u20132016) and the <strong>Cryptography<\/strong> mailing list at metzdowd.com (approximately 12,000 emails from 271 unique authors, spanning 2006\u20132008 \u2014 the exact mailing list where Satoshi first announced Bitcoin on 31 October 2008). Combined, we processed over <strong>104,000 individual emails<\/strong> from approximately <strong>12,979 unique<\/strong> authors. To our knowledge, this is the largest blind stylometric search ever conducted in the Satoshi attribution context.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">2.6 The \u2018Cannot\u2019 vs \u2018Can Not\u2019 Discovery<\/h2>\n\n\n\n<p>During our analysis, we discovered that Satoshi exclusively uses the <strong>single-word form \u2018cannot\u2019<\/strong> (12 occurrences) and <strong>never<\/strong> the two-word form \u2018<strong>can not<\/strong>\u2019 (0 occurrences). This is a distinctive and easily verifiable habit. We tested all candidates for consistency with this pattern. Some candidates, such as <strong>Ray Dillinger<\/strong> (0 \u2018can not\u2019, 30 \u2018cannot\u2019) and <strong>Peter Gutmann<\/strong> (0 \u2018can not\u2019, 3 \u2018cannot\u2019), <strong>perfectly match<\/strong> Satoshi\u2019s usage. Others, notably <strong>Adam Back<\/strong> (17 \u2018can not\u2019 vs 4 \u2018cannot\u2019), <strong>diverge<\/strong> significantly.<\/p>\n\n\n\n<p><em>Note on &#8216;can not&#8217;:<\/em> some <strong>prior<\/strong> Satoshi attribution studies and online analyses have treated the two-word form &#8216;can not&#8217; as a distinctive Satoshi marker. Our corpus analysis of 595 original Satoshi posts and the whitepaper shows this is <strong>incorrect<\/strong> \u2013 Satoshi uses <strong>&#8216;cannot<\/strong>&#8216; <strong>exclusively<\/strong> (12 occurrences across the forum posts, 3 in the whitepaper) and never &#8216;can not&#8217;. This distinction matters: candidates like Adam Back (17\u00d7 &#8216;can not&#8217;) and the anonymous remailer &#8216;admin&#8217; (19\u00d7 &#8216;can not&#8217;) diverge from Satoshi on this specific feature, while Ray Dillinger (0\u00d7 &#8216;can not&#8217;, 30\u00d7 &#8216;cannot&#8217;) and Hal Finney (0\u00d7 &#8216;can not&#8217;, 2\u00d7 &#8216;cannot&#8217;) match perfectly.<\/p>\n\n\n\n<h1 class=\"wp-block-heading\">3. Main Results<\/h1>\n\n\n\n<h2 class=\"wp-block-heading\">3.1 How We Narrowed 12,979 Authors to 29 \u2013 and How Delta Works<\/h2>\n\n\n\n<p>Searching 12,979 email authors one by one using the full Burrows&#8217; Delta method would be <strong>computationally impractical<\/strong> and statistically <strong>noisy<\/strong> \u2013 most of these authors wrote only a few words, and meaningful stylometric comparison requires at least several thousand words. Instead, we used a <strong>two-stage funnel<\/strong>:<\/p>\n\n\n\n<p><strong>Stage 1:<\/strong> Fingerprint Screening (12,979 \u2192 ~50 candidates). We built a quick scoring system based on Satoshi&#8217;s most distinctive and easily measurable writing habits.<\/p>\n\n\n\n<p>Each author received <strong>0-8 points<\/strong> based on: <strong>double-spacing<\/strong> after full stops (0-3 points based on how closely their ratio matched Satoshi&#8217;s ~81%), <strong>British spelling<\/strong> forms like &#8216;colour&#8217; and &#8216;favour&#8217; (0-2 points), frequency of the word <strong>&#8216;the<\/strong>&#8216; in the normal native English range of 4-7% (0-1 point \u2013 this filter alone eliminates most non-native English speakers, as writers whose first language lacks articles, such as Russian or Chinese, typically use &#8216;the&#8217; at only 1-3%), <strong>contraction rate<\/strong> in the moderate range (0-1 point), and use of the word <strong>&#8216;cannot<\/strong>&#8216; (0-1 point).<\/p>\n\n\n\n<p>Authors needed at least <strong>1,000 words<\/strong> for a reliable score. On the Cypherpunks list, only 3 authors out of 12,708 scored the maximum 8\/8: <strong>Adam Back, Mike Rosing<\/strong>, and an <strong>anonymous remailer<\/strong>. On the Cryptography list, <strong>7 out of 271 scored 7\/7<\/strong>. We retained all authors scoring 6 or above, plus any author with notable individual features, yielding approximately 50 candidates.<\/p>\n\n\n\n<p><strong>Stage 2:<\/strong> Full <strong>Burrows&#8217; Delta<\/strong> with z-scores (50 \u2192 29 candidates). We then ran the full proper multi-author Delta on these candidates plus named candidates from prior research (Szabo, Wright, Wilson, CFB, BCNext, Finney). Here is how <strong>the z-score<\/strong> calculation works in plain language:<\/p>\n\n\n\n<p>Imagine that across all 29 authors in our final population, the average usage of the word &#8216;the&#8217; is 5.2%, and the typical spread (standard deviation) is 1.1 percentage points. If Satoshi uses &#8216;the&#8217; at 5.9%, that is <strong>0.64 standard deviations above<\/strong> average \u2013 his <strong>z-score<\/strong> for &#8216;the&#8217; is +0.64. If Adam Back uses &#8216;the&#8217; at 5.3%, his <strong>z-score<\/strong> is +0.09. The <strong>difference<\/strong> between their z-scores for &#8216;the&#8217; is |0.64 \u2212 0.09| = 0.55. We <strong>repeat<\/strong> this calculation for all 50+ function words and take the <strong>average of all the differences<\/strong>. That average is the <strong>Delta score<\/strong>.<\/p>\n\n\n\n<p>The key insight is that z-scores <strong>normalize for the natural variation<\/strong> in each word. Some function words (like &#8216;the&#8217;) <strong>vary a lot<\/strong> between authors, while others (like &#8216;very&#8217;) vary <strong>little<\/strong>. Without z-scores, the words with the most variation would dominate the comparison and drown out the subtler signals. By converting everything to <strong>standard deviations<\/strong> first, every function word contributes <strong>equally<\/strong> to the final score regardless of its absolute frequency.<\/p>\n\n\n\n<p>This is why our <strong>&#8216;proper<\/strong>&#8216; multi-author Delta <strong>produces<\/strong> meaningful results while the <strong>simpler two-sample comparison that some studies use does not<\/strong>: the two-sample version has <strong>no population<\/strong> to calculate standard deviations from, so the z-scores <strong>degenerate<\/strong>. In our initial attempts, every two-sample comparison scored at the ceiling (2.0), producing no differentiation between candidates. The multi-author version, with 29 authors providing robust standard deviations, produces a <strong>meaningful spread<\/strong> from 0.31 (Satoshi versus himself) to 1.85 (Wilson&#8217;s technical writing).<\/p>\n\n\n\n<p>Authors with <strong>fewer<\/strong> than 4,000 words in the archive <strong>were excluded<\/strong> as unreliable \u2013 function word frequencies fluctuate too much in small samples. Several cypherpunks authors who scored well on the fingerprint but had <strong>very high Delta were dropped<\/strong> as false positives: their fingerprint match was <strong>coincidental<\/strong>, not reflecting genuine stylistic similarity across the full 50-word function word distribution. The final population of 29 includes all candidates who either (a) scored in the <strong>top tier<\/strong> on the fingerprint screening, (b) were <strong>named<\/strong> in prior attribution research, or (c) represent <strong>Satoshi&#8217;s own writing<\/strong> segments for <strong>self-validation<\/strong>.<\/p>\n\n\n\n<p>This funnel ensures we <strong>did not miss any strong<\/strong> candidate while keeping the Delta calculation statistically <strong>meaningful<\/strong> \u2013 the more authors in the population, the more reliable the standard deviations that underpin the z-scores.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">3.2 The Final Ranking (Forum Baseline)<\/h2>\n\n\n\n<p>In the table below, the reference <strong>baseline<\/strong> is <strong>Satoshi&#8217;s combined forum posts<\/strong> (61,438 words from both metzdowd and BitcoinTalk). The <strong>whitepaper is deliberately excluded<\/strong> from this baseline in order to <strong>test it independently<\/strong> \u2013 if it were included, we could not measure how far its writing style diverges from the forum posts. Each candidate&#8217;s Delta score measures how far their function word distribution is <strong>from this combined forum baseline<\/strong>. Satoshi&#8217;s own time-based segments (BitcoinTalk 2010, BitcoinTalk 2009, Metzdowd) do not score exactly zero because they are <strong>subsets<\/strong> of the combined baseline text, not the combined text itself \u2013 each subset has slightly different word frequencies due to topic variation and the natural randomness of smaller samples. Their scores of 0.31-0.83 establish the <strong>&#8216;same author&#8217;<\/strong> <strong>calibration<\/strong> range: any external candidate scoring in this range has a function word distribution <strong>as close to Satoshi as Satoshi&#8217;s own writing<\/strong> is to itself across different time periods and platforms. The <strong>whitepaper<\/strong> scored as a separate candidate against this forum baseline, <strong>lands at Delta 1.34<\/strong> \u2013 notably more <strong>distant<\/strong> than any of the forum segments, and almost identical to Nick Szabo&#8217;s score of 1.33. Candidates scoring <strong>below approximately 1.0<\/strong> are in the &#8216;same author&#8217; range; between 1.0 and 1.5 is <strong>uncertain<\/strong>; above 1.5 suggests a <strong>different<\/strong> author.<\/p>\n\n\n\n<p>Reference: Satoshi forum posts (61,438 words). Population: 29 authors. Method: Proper multi-author Burrows\u2019 Delta with corpus-wide z-scores. The \u2018cannot\u2019 column shows whether the candidate matches Satoshi\u2019s exclusive use of \u2018cannot\u2019 (never \u2018can not\u2019).<\/p>\n\n\n\n<figure class=\"wp-block-table is-style-stripes satoshi-table-small\"><table class=\"has-fixed-layout\"><tbody><tr><td><strong>Author<\/strong><\/td><td><strong>Delta<\/strong><\/td><td><strong>Words<\/strong><\/td><td><strong>DblSp<\/strong><\/td><td><strong>the%<\/strong><\/td><td><strong>Brit<\/strong><\/td><td><strong>cn!<\/strong><\/td><td><strong>cannot<\/strong><\/td><td><strong>Range<\/strong><\/td><td><strong>Identity<\/strong><\/td><\/tr><tr><td>Satoshi:BTCTalk 2010<\/td><td>0.31<\/td><td>29,839<\/td><td>*<\/td><td>6.2<\/td><td>5<\/td><td>0<\/td><td>2<\/td><td>\u2605<\/td><td>(self-test)<\/td><\/tr><tr><td>Satoshi:BTC 2009<\/td><td>0.39<\/td><td>22,775<\/td><td>*<\/td><td>5.3<\/td><td>7<\/td><td>0<\/td><td>1<\/td><td>\u2605<\/td><td>(self-test)<\/td><\/tr><tr><td><strong>cp:bear<\/strong><\/td><td><strong>0.77<\/strong><\/td><td>67,615<\/td><td>86%<\/td><td>4.9<\/td><td>5<\/td><td>0\u2713<\/td><td>30<\/td><td>\u2605<\/td><td><strong>R.Dillinger<\/strong><\/td><\/tr><tr><td>Satoshi:Metzdowd<\/td><td>0.83<\/td><td>8,742<\/td><td>81%<\/td><td>6.2<\/td><td>1<\/td><td>0<\/td><td>6<\/td><td>\u2605<\/td><td>(self-test)<\/td><\/tr><tr><td><strong>cp:pgut001<\/strong><\/td><td><strong>0.86<\/strong><\/td><td>30,456<\/td><td>82%<\/td><td>5.7<\/td><td>15<\/td><td>0\u2713<\/td><td>3<\/td><td>\u2605<\/td><td><strong>P.Gutmann<\/strong><\/td><\/tr><tr><td>cp:eresrch<\/td><td>0.87<\/td><td>19,050<\/td><td>97%<\/td><td>5.0<\/td><td>3<\/td><td>4\u2717<\/td><td>1<\/td><td>\u2605<\/td><td>M.Rosing<\/td><\/tr><tr><td>cp:adam<\/td><td>0.91<\/td><td>90,584<\/td><td>77%<\/td><td>5.3<\/td><td>12<\/td><td>17\u2717<\/td><td>4<\/td><td>\u2605<\/td><td>A.Back<\/td><\/tr><tr><td>cp:honig<\/td><td>0.96<\/td><td>44,772<\/td><td>77%<\/td><td>4.8<\/td><td>8<\/td><td>0\u2713<\/td><td>5<\/td><td>\u2605<\/td><td>D.Honig<\/td><\/tr><tr><td>cp:hal<\/td><td>0.99<\/td><td>7,061<\/td><td>100%<\/td><td>6.7<\/td><td>0<\/td><td>0\u2713<\/td><td>2<\/td><td>\u2605<\/td><td>H.Finney<\/td><\/tr><tr><td>Wilson(vu.hn)<\/td><td>1.01<\/td><td>60,614<\/td><td>**<\/td><td>6.7<\/td><td>15<\/td><td>0\u2713<\/td><td>34<\/td><td>?<\/td><td>P.Wilson<\/td><\/tr><tr><td>Nick Szabo<\/td><td>1.33<\/td><td>210,313<\/td><td>45%<\/td><td>6.5<\/td><td>12<\/td><td>1\u2717<\/td><td>82<\/td><td>?<\/td><td>&nbsp;<\/td><\/tr><tr><td>Satoshi:Whitepaper<\/td><td>1.34<\/td><td>3,076<\/td><td>97%<\/td><td>7.7<\/td><td>1<\/td><td>0<\/td><td>3<\/td><td>?<\/td><td>&nbsp;<\/td><\/tr><tr><td>Craig Wright<\/td><td>1.40<\/td><td>6,902<\/td><td>**<\/td><td>6.2<\/td><td>1<\/td><td>0\u2713<\/td><td>6<\/td><td>?<\/td><td>&nbsp;<\/td><\/tr><tr><td>CFB\/Ivancheglo<\/td><td>1.60<\/td><td>134,315<\/td><td>0%<\/td><td>2.1<\/td><td>3<\/td><td>9\u2717<\/td><td>20<\/td><td>DIFF<\/td><td>&nbsp;<\/td><\/tr><tr><td>Wilson(scronty)<\/td><td>1.85<\/td><td>23,665<\/td><td>17%<\/td><td>7.6<\/td><td>3<\/td><td>&nbsp;<\/td><td>&nbsp;<\/td><td>DIFF<\/td><td>&nbsp;<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p><em>* 0% = BitcoinTalk HTML strips double-spaces (actual ~81%).<\/em><\/p>\n\n\n\n<p><em>** 0% = HTML source strips double-spaces. cn! = uses of \u2018can not\u2019 (Satoshi NEVER uses this; \u2713=matches, \u2717=diverges).<\/em><\/p>\n\n\n\n<p><em>\u2605 = same author (&lt;1.0). ? = uncertain (1.0\u20131.5). DIFF = different (&gt;1.5).<\/em><\/p>\n\n\n\n<h2 class=\"wp-block-heading\">3.3 The Multi-Baseline Discovery<\/h2>\n\n\n\n<p>When we changed the baseline from forum posts to the whitepaper, the ranking changed completely:<\/p>\n\n\n\n<p><strong>Closest to FORUM Satoshi (conversational writing): <\/strong>Ray Dillinger (0.77), Peter Gutmann (0.86), Mike Rosing (0.87), Adam Back (0.91)<\/p>\n\n\n\n<p><strong>Closest to WHITEPAPER Satoshi (academic writing): <\/strong>Craig Wright (1.06), Anonymous remailer (1.12), Nick Szabo (1.23), Adam Back (1.22)<\/p>\n\n\n\n<p>There is <strong>zero overlap<\/strong> between the forum top 3 and the whitepaper top 3. Ray Dillinger, the strongest forum match, drops to rank 16 against the whitepaper (Delta 1.37). Craig Wright, the closest whitepaper match, is rank 27 against the forum (Delta 1.40). This divergence is consistent with the hypothesis that the <strong>whitepaper and forum posts were drafted by different people <\/strong>\u2013 or by the same person in <strong>dramatically different writing registers<\/strong>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">3.4 Satoshi\u2019s Internal Consistency<\/h2>\n\n\n\n<p>The Satoshi corpus shows a clear gradient from <strong>formal to informal<\/strong>:<\/p>\n\n\n\n<figure class=\"wp-block-table is-style-stripes satoshi-table-small\"><table class=\"has-fixed-layout\"><tbody><tr><td><strong>Segment<\/strong><\/td><td><strong>Words<\/strong><\/td><td><strong>\u2018the\u2019%<\/strong><\/td><td><strong>\u2018a\u2019%<\/strong><\/td><td><strong>\u2018of\u2019%<\/strong><\/td><td><strong>\u2018it\u2019%<\/strong><\/td><td><strong>Contr%<\/strong><\/td><td><strong>British%<\/strong><\/td><\/tr><tr><td>Whitepaper<\/td><td>3,076<\/td><td>7.74<\/td><td>3.38<\/td><td>2.76<\/td><td>1.17<\/td><td>0.88<\/td><td>33%<\/td><\/tr><tr><td>Metzdowd 08-09<\/td><td>8,742<\/td><td>6.17<\/td><td>3.29<\/td><td>2.44<\/td><td>1.46<\/td><td>1.60<\/td><td>6.2%<\/td><\/tr><tr><td>BTC ~2009<\/td><td>22,775<\/td><td>5.28<\/td><td>2.39<\/td><td>1.33<\/td><td>2.78<\/td><td>3.35<\/td><td>13%<\/td><\/tr><tr><td>BTC ~2010<\/td><td>29,839<\/td><td>6.20<\/td><td>2.54<\/td><td>1.47<\/td><td>2.24<\/td><td>2.93<\/td><td>5.8%<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p>The whitepaper uses \u2018the\u2019 at 7.74% while the forum posts average 5.3-6.2%. Contractions increase fourfold from whitepaper (0.88%) to BitcoinTalk 2009 (3.35%). The pronoun \u2018it\u2019 (conversational) doubles from whitepaper to forum. The whitepaper also contains mixed British-American spellings within a single 3,076-word document: \u2018favour\u2019 (British) alongside \u2018characterized\u2019 (American).<\/p>\n\n\n\n<h1 class=\"wp-block-heading\">4. Candidate-by-Candidate Assessment<\/h1>\n\n\n\n<h2 class=\"wp-block-heading\">4.1 Ray Dillinger<\/h2>\n\n\n\n<p><strong>Stylometric evidence FOR being Satoshi: <\/strong>Lowest Delta of any external candidate (0.77 \u2013 firmly in \u2018same author\u2019 range). Scores \u2605 SAME on all four forum-based baselines \u2013 the only external candidate with this consistency. 86% double-spacing (Satoshi: ~81%). Exclusively uses \u2018cannot\u2019 (30 times), never \u2018can not\u2019 \u2013 matching Satoshi perfectly. 5 British spellings.<\/p>\n\n\n\n<p><strong>Circumstantial evidence: <\/strong>Dillinger (known online as \u2018bear\u2019) was one of the first people Satoshi contacted in late 2008. He reviewed Bitcoin\u2019s source code before its public release, providing security feedback. He is a veteran cypherpunk and privacy advocate active on the mailing lists throughout the relevant period with 67,615 words of archived writing.<\/p>\n\n\n\n<p><strong>Evidence AGAINST being Satoshi: <\/strong>American \u2013 the 5 British spellings may come from quoting others. Has publicly described his role as a reviewer, not as the creator. Has neither confirmed nor denied any deeper involvement. Not previously discussed in attribution research.<\/p>\n\n\n\n<p><strong>Motive to hide: <\/strong>As a privacy-focused cypherpunk, Dillinger would have ideological reasons for anonymity. However, no specific legal or financial motive has been identified.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">4.2 Adam Back<\/h2>\n\n\n\n<p><strong>Stylometric evidence FOR being Satoshi: <\/strong>Delta 0.91 (\u2605 same author range). 12 British spellings (Satoshi: ~11). 77% double-spacing. Scottish \u2013 naturally writes British English. 90,584 words of archived writing. Invented Hashcash, the proof-of-work system cited in the Bitcoin whitepaper. First person Satoshi contacted by email (August 2008).<\/p>\n\n\n\n<p><strong>Evidence AGAINST being Satoshi: <\/strong>Writes \u2018can not\u2019 17 times versus \u2018cannot\u2019 only 4 times \u2013 the opposite of Satoshi\u2019s exclusive \u2018cannot\u2019 usage. This is a specific, verifiable inconsistency. In 2013, joined the bitcoin-wizards IRC channel and asked basic questions about UTXO and address balances \u2013 logged and witnessed by developers, suggesting he did not have deep knowledge of Bitcoin\u2019s internals at that time (<a href=\"https:\/\/www.chaincatcher.com\/en\/article\/2146231\" target=\"_blank\" rel=\"noopener\">source<\/a>). Consistently denied being Satoshi. His company Blockstream promotes off-chain scaling solutions that some argue contradict Satoshi\u2019s on-chain scaling vision. Openly acknowledged the coincidences: \u2018I do use double-space and native spelling British. Can code C++\u2019 (<a href=\"https:\/\/decrypt.co\/28542\/new-theory-claims-adam-back-is-satoshi-nakamoto\" target=\"_blank\" rel=\"noopener\">source<\/a>).<\/p>\n\n\n\n<p><strong>Motive to hide: <\/strong>If Back were Satoshi, his position as CEO of Blockstream would create a massive undisclosed conflict of interest. However, Hashcash was published openly under his real name, suggesting anonymity was not a concern for him prior to Bitcoin.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">4.3 Hal Finney<\/h2>\n\n\n\n<p><strong>Stylometric evidence FOR being Satoshi: <\/strong>Delta 0.99 (\u2605 same author, barely). 100% double-spacing \u2013 the highest of any candidate. Exclusively uses \u2018cannot\u2019 (never \u2018can not\u2019). Created RPOW (Reusable Proof of Work), a direct predecessor technology. Received the first Bitcoin transaction.<\/p>\n\n\n\n<p><strong>Evidence AGAINST being Satoshi: <\/strong>Jameson Lopp documented that on April 18, 2009, Finney was competing in a 10-mile race in Santa Barbara while Satoshi was simultaneously confirming transactions and sending emails (<a href=\"https:\/\/blog.lopp.net\/hal-finney-was-not-satoshi-nakamoto\/\" target=\"_blank\" rel=\"noopener\">source<\/a>). Finney and Satoshi ran nodes from different IP addresses on different ISPs. By August 2010, his wife reported his typing had slowed to a \u2018sluggish finger peck\u2019 due to ALS, yet Satoshi was still actively posting (<a href=\"https:\/\/cointelegraph.com\/news\/bitcoin-pioneer-hal-finney-not-satoshi-nakamoto-argues-jameson-lopp\" target=\"_blank\" rel=\"noopener\">source<\/a>). Finney openly shared his Satoshi correspondence with journalists. Died of ALS in 2014.<\/p>\n\n\n\n<p><strong>Motive to hide: <\/strong>Finney was a privacy advocate, but his openness about his Bitcoin involvement \u2013 publishing debug logs, showing emails to reporters \u2013 is inconsistent with Satoshi\u2019s extreme operational security.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">4.4 Nick Szabo<\/h2>\n\n\n\n<p><strong>Stylometric evidence FOR being Satoshi: <\/strong>Closest match to the whitepaper baseline (Delta 1.23). Created Bit Gold, a theoretical predecessor to Bitcoin. The 2014 Aston University study ranked him first based on linguistic analysis of the whitepaper.<\/p>\n\n\n\n<p><strong>Evidence AGAINST being Satoshi: <\/strong>Delta 1.33 against forum Satoshi \u2013 ranked 24th out of 29 candidates. Satoshi was apparently unaware of Wei Dai\u2019s b-money when contacting Adam Back in August 2008 \u2013 Szabo would have known about it. Wei Dai himself stated: \u2018Why would Satoshi cite b-money but not Bit Gold if Satoshi was Nick?\u2019 (<a href=\"https:\/\/davidgerard.co.uk\/blockchain\/2018\/12\/16\/no-nick-szabo-wasnt-satoshi-in-2014-either\/\" target=\"_blank\" rel=\"noopener\">source<\/a>). Uses American English (he is American of Hungarian descent). In April 2008, publicly asked for help coding Bit Gold \u2013 six months before the whitepaper (<a href=\"https:\/\/coinmarketcap.com\/academy\/article\/satoshi-files-nick-szabo\" target=\"_blank\" rel=\"noopener\">source<\/a>). Uses \u2018can not\u2019 once (Satoshi: never).<\/p>\n\n\n\n<p><strong>Motive to hide: <\/strong>As a legal scholar, Szabo would understand regulatory risks. However, Bit Gold was proposed openly under his real name.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">4.5 Craig Wright<\/h2>\n\n\n\n<p><strong>Stylometric evidence being Satoshi: <\/strong>Delta 1.40 against forum Satoshi. Only 1 British spelling despite being Australian. However, closest external match to the whitepaper baseline (Delta 1.06). Uses \u2018cannot\u2019 exclusively (matching Satoshi on this specific feature).<\/p>\n\n\n\n<p><strong>Evidence AGAINST being Satoshi: <\/strong>The UK High Court ruled convincingly in March 2024 (COPA v. Wright) that Wright \u2018is not Satoshi Nakamoto,\u2019 that he \u2018lied to the court extensively and repeatedly,\u2019 and committed \u2018forgery on a grand scale.\u2019 Sentenced December 2024 to 12 months suspended for contempt (<a href=\"https:\/\/en.wikipedia.org\/wiki\/Craig_Steven_Wright\" target=\"_blank\" rel=\"noopener\">source<\/a>).<\/p>\n\n\n\n<p><strong>Motive context: <\/strong>Extensive gambling industry ties: Lasseter\u2019s Online Casino (1998), Centrebet, Playboy Gaming. Told the ATO he \u2018had been producing software for online casinos when he was writing code that later helped develop Bitcoin.\u2019 His whitepaper-proximity is notable, but the court definitively excluded him.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">4.6 Phil Wilson<\/h2>\n\n\n\n<p><strong>Stylometric evidence: <\/strong>His vu.hn narrative prose (post-2016) scores 1.01 \u2013 on the author\/uncertain boundary. His confirmed pre-Bitcoin writing from scrontsoft.com scores 1.85 \u2013 definitively different than Satoshi. However, he uses \u2018cannot\u2019 exclusively (34 times, matching Satoshi).<\/p>\n\n\n\n<p><strong>The vu.hn narrative: <\/strong>Wilson\u2019s Bitcoin Origins account on vu.hn provides the most detailed insider narrative of Bitcoin\u2019s creation, including step-by-step instructions for constructing the Bitcoin logo. His narrative explicitly claims multiple team members drafted Satoshi posts, with one person posting them to avoid contradictions. He also claims the original concept involved online gambling payment infrastructure.<\/p>\n\n\n\n<p><strong>Assessment: <\/strong>Wilson\u2019s pre-Bitcoin technical writing is stylistically incompatible with Satoshi. His post-2016 prose is closer but was written after Satoshi\u2019s style was publicly documented. The UIGEA motive argument he provides is compelling and independently corroborated by the poker code in Bitcoin v0.1.<\/p>\n\n\n\n<h1 class=\"wp-block-heading\">5. Why Would Satoshi Hide? \u2013 The Legal Motive Analysis<\/h1>\n\n\n\n<p>One of the <strong>most puzzling<\/strong> aspects of the Satoshi mystery is the <strong>extreme operational security<\/strong> (OPSEC) employed from the <strong>very beginning<\/strong>. OPSEC means the practices used to protect one\u2019s identity from detection. Satoshi used Tor (an anonymising network that routes internet traffic through multiple relays to hide the user\u2019s location), registered the bitcoin.org domain through an anonymous registration service, and created a <strong>completely fictional identity<\/strong> with a Japanese name. This level of pre-emptive anonymity \u2013 before Bitcoin had any value, users, or law enforcement attention \u2013 <strong>requires a concrete reason, a strong motive<\/strong>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">5.1 The UIGEA Timeline<\/h2>\n\n\n\n<p>The Unlawful Internet Gambling Enforcement Act (UIGEA) provides the most compelling timeline match:<\/p>\n\n\n\n<p><strong>October 13, 2006: <\/strong>UIGEA signed into law by President Bush. The law makes it a federal crime for financial institutions to process payments connected to illegal online gambling.<\/p>\n\n\n\n<p><strong>August 18, 2008: <\/strong>The domain bitcoin.org is registered anonymously.<\/p>\n\n\n\n<p><strong>October 31, 2008: <\/strong>Bitcoin whitepaper published on the metzdowd.com mailing list.<\/p>\n\n\n\n<p><strong>November 12, 2008: <\/strong>UIGEA final regulations released \u2013 12 days after the whitepaper.<\/p>\n\n\n\n<p><strong>January 3, 2009: <\/strong>Bitcoin Genesis Block mined.<\/p>\n\n\n\n<p><strong>January 19, 2009: <\/strong>UIGEA compliance deadline \u2013 16 days after the Genesis Block.<\/p>\n\n\n\n<p><strong>April 15, 2011: <\/strong>\u2018Black Friday\u2019 \u2013 founders of PokerStars, Full Tilt Poker, and Absolute Poker indicted under UIGEA.<\/p>\n\n\n\n<p><strong>April 26, 2011: <\/strong>Satoshi\u2019s last known communication \u2013 11 days after Black Friday.<\/p>\n\n\n\n<p>Bitcoin\u2019s original v0.1 source code <strong>contained GUI code for a poker client<\/strong> (lines 1573\u20131788). The poker client GUI is a result of the <strong>intentional coding instructions of the developer<\/strong>. Satoshi never explained why this code was included. The creation of an anonymous, censorship-resistant payment system would <strong>directly address<\/strong> UIGEA\u2019s <strong>payment-blocking<\/strong> mechanism.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">5.2 Motive Assessment by Candidate<\/h2>\n\n\n\n<p><strong>Ray Dillinger: <\/strong>No known gambling industry connection. Privacy-focused cypherpunk with ideological reasons for anonymity, but no specific legal pressure from UIGEA.<\/p>\n\n\n\n<p><strong>Adam Back: <\/strong>No gambling industry connection. Hashcash was published openly. No UIGEA exposure.<\/p>\n\n\n\n<p><strong>Hal Finney: <\/strong>No gambling connection. Publicly transparent about Bitcoin involvement. Weak motive to hide.<\/p>\n\n\n\n<p><strong>Nick Szabo: <\/strong>Legal scholar who understands regulatory risk. Published Bit Gold openly. No gambling connection.<\/p>\n\n\n\n<p><strong>Craig Wright: <\/strong>Extensive gambling industry ties: Lasseter\u2019s Online Casino (Australia\u2019s first government-licensed online casino), Centrebet, Playboy Gaming, MGM, Bodog. Told the Australian Tax Office he produced software for online casinos while writing code that helped develop Bitcoin. If true, UIGEA would provide a direct motive for anonymity. However, the COPA court found his Satoshi claims fraudulent based on extensive evidence.<\/p>\n\n\n\n<p><strong>Phil Wilson: <\/strong>His <strong>vu.hn narrative<\/strong> explicitly cites online gambling as Bitcoin\u2019s original use case and UIGEA as the reason for anonymity. This is the only candidate narrative that <strong>coherently explains the extreme pre-emptive OPSEC<\/strong>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">5.3 The 1.1 Million Unmoved Bitcoin \u2013 and Recent Movements<\/h2>\n\n\n\n<p>Approximately 1.1 million BTC in addresses associated with the earliest mining period (2009-2011) were long considered permanently dormant. However, 2024 and 2025 saw unprecedented movement from <strong>&#8216;Satoshi-era&#8217;<\/strong> wallets. In July 2025, <strong>eight wallets<\/strong> moved <strong>80,000 BTC (over $8 billion)<\/strong> in the largest such transfer on record (source: <a href=\"https:\/\/www.coindesk.com\/markets\/2025\/07\/05\/eight-bitcoin-wallets-move-80000-btc-in-largest-ever-satoshi-era-transfers\" target=\"_blank\" rel=\"noopener\">CoinDesk<\/a>). <strong>Galaxy Digital helped a Satoshi-era investor sell<\/strong> more than $9 billion in Bitcoin in <strong>July 2025<\/strong>. In total, over 270,000 BTC aged 7+ years moved on-chain in 2025 alone \u2013 a new record.<\/p>\n\n\n\n<p>Critically, blockchain analysis firms such as Arkham <strong>have confirmed<\/strong> that the moved coins are <strong>not linked to Satoshi Nakamoto&#8217;s own addresses<\/strong>. The wallets that moved belong to <strong>early miners<\/strong> \u2013 people who ran mining software in 2009-2011 \u2013 not to the creator&#8217;s known addresses. Satoshi&#8217;s own estimated holdings of approximately <strong>1.1 million BTC remain untouched<\/strong>.<\/p>\n\n\n\n<p>The distinction <strong>matters<\/strong> for our analysis: the early miners who are now selling mined Bitcoin when it was nearly worthless and are taking profits at <strong>historically high prices<\/strong>. This is <strong>rational<\/strong> economic behaviour. Satoshi&#8217;s continued inaction \u2013 holding tens of billions in value without moving a single coin for over 15 years \u2013 is <strong>not rational<\/strong> unless there is a <strong>compelling reason<\/strong> not to spend. That reason could be <strong>death<\/strong> (as in Hal Finney&#8217;s case), <strong>loss of private keys<\/strong>, or <strong>fear of legal consequences<\/strong> that would follow from revealing one&#8217;s identity through a traceable transaction. The UIGEA liability hypothesis <strong>provides the most specific<\/strong> version of the <strong>third<\/strong> explanation.<\/p>\n\n\n\n<h1 class=\"wp-block-heading\">6. Definitively Excluded Candidates<\/h1>\n\n\n\n<h2 class=\"wp-block-heading\">6.1 Sergey Ivancheglo (CFB) and the BCNext Mimicry<\/h2>\n\n\n\n<p>Ivancheglo\u2019s 134,315 words of English writing show systematic <strong>underuse of articles<\/strong>: \u2018the\u2019 at 2.1% (Satoshi: 5.9%). This is an <strong>involuntary marker<\/strong> of writers whose first language lacks articles (Russian, Belarusian). It cannot be consistently controlled across 134,000 words. Delta: 1.60 (\u2018different author\u2019).<\/p>\n\n\n\n<p>However, Ivancheglo \u2013 who cryptographically proved in 2017 that he was also the anonymous \u2018BCNext\u2019 who founded the NXT cryptocurrency \u2013 <strong>successfully mimicked<\/strong> Satoshi\u2019s writing patterns for 56,606 words as BCNext, including 96.7% double-spacing and near-native article usage. BCNext used \u2018can not\u2019 45 times (never \u2018cannot\u2019), <strong>revealing the mimicry<\/strong> was imperfect. This case proves that short-term stylometric mimicry is possible, but that deep features (article frequency, specific word-form choices) <strong>expose the disguise<\/strong> over a large enough corpus.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">6.2 Craig Wright<\/h2>\n\n\n\n<p>Excluded by the UK High Court in COPA v. Wright (March 2024). Justice Mellor found Wright \u2018<strong>lied to the court extensively and repeatedly<\/strong>\u2019 and committed \u2018<strong>forgery on a grand scale<\/strong>\u2019. Our stylometry independently confirms: Delta 1.40 against forum Satoshi, only 1 British spelling (<a href=\"https:\/\/en.wikipedia.org\/wiki\/Craig_Steven_Wright\" target=\"_blank\" rel=\"noopener\">Source<\/a>).<\/p>\n\n\n\n<h1 class=\"wp-block-heading\">7. Phil Wilson \u2013 A Complex Case<\/h1>\n\n\n\n<p>Wilson&#8217;s confirmed <strong>pre-Bitcoin writing from scrontsoft.com<\/strong> (23,665 words of DirectX programming <strong>tutorials and readme<\/strong> files, 2000-2008) scores Delta 1.85 \u2013 the second-worst match of any reasonable candidate. <strong>However<\/strong>, as noted in Section 1, this corpus consists of <strong>code documentation<\/strong>, a fundamentally <strong>different genre<\/strong> from Satoshi&#8217;s conversational forum posts, and genre <strong>mismatch<\/strong> alone can inflate Delta by 0.3-0.5 points. Therefore, the scrontsoft.com divergence should <strong>not be interpreted as conclusive<\/strong>.<\/p>\n\n\n\n<p>Wilson&#8217;s post-2016 narrative prose on vu.hn (60,614 words) scores Delta 1.01 \u2013 on the <strong>boundary of being the same author as Satoshi<\/strong>. Critically, this large corpus shows <strong>no signs of the stylometric breakdowns <\/strong>we documented in BCNext&#8217;s mimicry: Wilson&#8217;s &#8216;cannot&#8217; usage (34 times, never &#8216;can not&#8217;) matches Satoshi perfectly, and his function word distribution remains internally consistent across 60,000 words.<\/p>\n\n\n\n<p>What makes Wilson <strong>uniquely difficult to dismiss<\/strong> is the <strong>legal and circumstantial<\/strong> dimension. As discussed in our earlier article (<a href=\"https:\/\/www.ulclegal.com\/buletin\/bitcoin-turns-15-today-who-is-satoshi-nakamoto-can-stolen-bitcoins-be-returned-to-the-owner\/\">ulclegal.com: &#8216;Bitcoin Turns 15 Today. Who is Satoshi Nakamoto?&#8217;<\/a>, October 2024), Wilson provides the only candidate narrative that <strong>coherently<\/strong> <strong>explains<\/strong> several otherwise puzzling <strong>facts<\/strong>:<\/p>\n\n\n\n<ol style=\"list-style-type:lower-roman\" class=\"wp-block-list\">\n<li>the extreme pre-emptive OPSEC before Bitcoin had any value;<\/li>\n\n\n\n<li>the poker client and marketplace GUI code in Bitcoin v0.1;<\/li>\n\n\n\n<li>the precise correlation between UIGEA enforcement dates and Bitcoin&#8217;s creation timeline; and<\/li>\n\n\n\n<li>Satoshi&#8217;s permanent withdrawal after the Silk Road attracted law enforcement attention.<\/li>\n<\/ol>\n\n\n\n<p>Wilson also occupies a <strong>unique position<\/strong> regarding <strong>Craig Wright<\/strong>. Wright has consistently <strong>denied<\/strong> Wilson&#8217;s involvement and refused to name him as the &#8216;third member&#8217; of the alleged Satoshi team \u2013 even in court proceedings where Wright claimed to be Satoshi himself. Yet Wilson is <strong>the only person in the world<\/strong> who has publicly and consistently <strong>confirmed<\/strong> Wright&#8217;s partial involvement in Bitcoin&#8217;s creation, while simultaneously <strong>explaining<\/strong> Wright&#8217;s inability to produce valid cryptographic proof: according to Wilson, Wright was responsible for <strong>infrastructure<\/strong> and <strong>business<\/strong> aspects, not for the cryptographic keys, which Wilson claims were <strong>destroyed<\/strong> in 2011.<\/p>\n\n\n\n<p>This creates a <strong>paradox<\/strong>: Wright, who used forged documents to support his own Satoshi claims, <strong>denies the involvement<\/strong> of the one person who <strong>actually supports<\/strong> Wright&#8217;s partial involvement. If Wilson&#8217;s account is <strong>fabricated<\/strong>, there would be <strong>no reason<\/strong> for Wright to deny it \u2013 Wilson&#8217;s narrative is, in fact, the <strong>most favorable publicly<\/strong> <strong>available<\/strong> account of Wright&#8217;s role. Wright&#8217;s denial of Wilson is therefore <strong>either evidence<\/strong> that Wilson&#8217;s account is <strong>false<\/strong>, or evidence that Wright <strong>fears<\/strong> Wilson&#8217;s account because it contradicts Wright&#8217;s claim to be the <strong>sole<\/strong> or primary creator.<\/p>\n\n\n\n<p>There is a further <strong>logical inference<\/strong> to be drawn. The COPA judgment established, with extensive documentary evidence, that Wright is a habitual and deliberate <strong>liar<\/strong> who fabricated evidence over a period of years. If we accept the court&#8217;s finding that Wright <strong>lies systematically<\/strong> to advance his interests, then his <strong>resolute denial<\/strong> of Wilson&#8217;s involvement must itself be treated with <strong>suspicion<\/strong>. A truthful person denying Wilson would be <strong>evidence<\/strong> against Wilson. But a <strong>proven liar<\/strong> denying Wilson \u2013 particularly when Wilson&#8217;s account is the only public narrative that partially supports Wright&#8217;s own claim to have been involved with Bitcoin&#8217;s creation \u2013 points in the <strong>opposite<\/strong> direction. If Wright were telling the truth about Wilson, it would be one of the few instances where his word <strong>aligned with verifiable<\/strong> reality, and there would be <strong>no<\/strong> strategic <strong>reason<\/strong> for the denial. The more parsimonious explanation is that <strong>Wright denies<\/strong> Wilson precisely because Wilson&#8217;s account is <strong>substantially true<\/strong>, and it threatens Wright&#8217;s narrative of <strong>sole or primary<\/strong> authorship. Put simply: <strong>if a convicted liar says that someone was not involved, that person probably was involved<\/strong>.<\/p>\n\n\n\n<p>The COPA v. Wright judgment (2024) found that Wright forged documents and <strong>lied extensively<\/strong>. Wright claimed to demonstrate the ability to sign with Satoshi&#8217;s key in private sessions with Gavin Andresen in 2016, though the public proof he subsequently provided was <strong>shown to be fraudulent<\/strong>.<\/p>\n\n\n\n<p>Our stylometric analysis <strong>cannot resolve this paradox<\/strong>. Wilson&#8217;s pre-Bitcoin writing <strong>excludes<\/strong> him as the author of Satoshi&#8217;s <strong>forum posts<\/strong>, but genre mismatch <strong>weakens this conclusion<\/strong>. His post-2016 prose is stylistically compatible over a large corpus. And his narrative provides the <strong>only coherent explanation<\/strong> for the UIGEA motive, the poker code, the marketplace code, and the extreme OPSEC. He remains a candidate <strong>who cannot be excluded<\/strong> on the balance of evidence.<\/p>\n\n\n\n<h1 class=\"wp-block-heading\">8. Conclusions<\/h1>\n\n\n\n<p><strong>First<\/strong>, <strong>Ray Dillinger<\/strong> emerges as the strongest composite match to Satoshi\u2019s forum writing: lowest Delta (0.77), consistent \u2018cannot\u2019 usage (never \u2018can not\u2019), high double-spacing (86%), and the only external candidate scoring \u2605 SAME on all four forum baselines. He reviewed Bitcoin\u2019s code before launch.<\/p>\n\n\n\n<p><strong>Second<\/strong>, the multi-baseline analysis reveals that the Whitepaper and forum posts have completely <strong>different candidate<\/strong> profiles, with zero overlap between the top three for each. This is the study\u2019s <strong>most significant methodological finding<\/strong>.<\/p>\n\n\n\n<p><strong>Third<\/strong>, <strong>Adam Back<\/strong> \u2013 previously the strongest candidate based on fingerprint criteria \u2013 has a notable \u2018can not\u2019 divergence from Satoshi\u2019s exclusive \u2018cannot\u2019 usage. His 2013 IRC learning logs also suggest he <strong>did not have<\/strong> deep Bitcoin knowledge at that time.<\/p>\n\n\n\n<p><strong>Fourth<\/strong>, Nick Szabo ranks 24<sup>th<\/sup> out of 29 candidates against the forum posts, contradicting the 2014 Aston University study. His writing is closer to the whitepaper\u2019s academic register but <strong>far from the conversational Satoshi<\/strong>.<\/p>\n\n\n\n<p><strong>Fifth<\/strong>, the UIGEA gambling motive provides the <strong>only candidate-independent<\/strong> explanation for Satoshi\u2019s extreme pre-emptive anonymity. The timeline correlation and the poker code in Bitcoin v0.1 are <strong>independently verifiable facts<\/strong>.<\/p>\n\n\n\n<p><strong>Sixth<\/strong>, we documented the first proven case of <strong>deliberate stylometric mimicry<\/strong> (BCNext), establishing that superficial feature matching alone is insufficient for attribution.<\/p>\n\n\n\n<p><strong>Seventh<\/strong>, <strong>Phil Wilson cannot be excluded<\/strong> on the balance of <strong>stylometric<\/strong> evidence. His pre-Bitcoin technical writing <strong>does not match<\/strong> Satoshi&#8217;s forum style, but <strong>genre mismatch<\/strong> weakens this finding. His post-2016 prose is stylistically compatible across 60,000 words <strong>without the mimicry breakdowns <\/strong>we documented in BCNext. His narrative provides the on<strong>ly coherent explanation <\/strong>for the UIGEA motive, the code artefacts, the extreme OPSEC, and the Wright paradox. As we concluded in our earlier ulclegal.com article, the <strong>circumstantial<\/strong> and legal evidence <strong>warrants continued investigation.<\/strong><\/p>\n\n\n\n<p><strong>Finally<\/strong>, a candid assessment of our own methodology&#8217;s limitations. Stylometry is a powerful forensic tool, validated in courtrooms and academic research. However, in this specific case, it produces an a<strong>bundance of candidates <\/strong>rather than a single identification: <strong>nine <\/strong>authors scored in the &#8216;same author&#8217; range (Delta below 1.0), and <strong>three <\/strong>scored a perfect 8\/8 on the initial fingerprint screening. They cannot all be Satoshi. Stylometry can <strong>narrow <\/strong>a field and exclude candidates, but it cannot, on its own, identify a single author from a population of linguistically similar cypherpunks who shared a generation, a culture, a technical vocabulary, and typing habits learned in the same era.<\/p>\n\n\n\n<p>In forensic investigation, when physical or statistical evidence produces <strong>multiple <\/strong>plausible matches, the traditional criteria of <strong>motive, method, and causality<\/strong> become decisive. Who had a concrete reason to build an anonymous payment system? Who had the technical capability to build it? Who can explain the specific artefacts in the code \u2013 the poker client, the marketplace GUI \u2013 and the precise correlation with the UIGEA enforcement timeline? Who can explain the extreme operational security from day one, and the permanent withdrawal after law enforcement attention? Who can explain why the Satoshi&#8217;s bitcoins were never spent?<\/p>\n\n\n\n<p>Of all the candidates examined in this study, only one has publicly provided <strong>coherent answers<\/strong> to all of these questions: Phil Wilson. However, his account remains <strong>unproven <\/strong>\u2013 he has produced no cryptographic signature, no verifiable technical evidence, and no independent corroboration beyond the circumstantial matches documented here and in our earlier article. But he is the only candidate whose narrative addresses motive, method, and causality as a complete and <strong>internally consistent whole<\/strong>. He also fits the established Satoshi <strong>profile<\/strong>: native English speaker of non-US and non-British origin. The stylometric evidence is ambiguous for Wilson; the forensic logic is not.<\/p>\n\n\n\n<p><em>The absence of cryptographic proof remains the fundamental barrier to definitive attribution. Until someone produces a valid digital signature from Satoshi\u2019s known keys, all attribution analysis is necessarily probabilistic.<\/em><\/p>\n\n\n\n<h1 class=\"wp-block-heading\">9. Data Availability and Reproducibility<\/h1>\n\n\n\n<p>All scripts, corpus files, and intermediate results generated during this research are available from ULC \u010carnogursk\u00fd upon request. The Satoshi reference corpus (68,175 words, 595 posts, 88.5% double-spacing) was built from metzdowd.com monthly archives and direct BitcoinTalk scraping, with whitespace preservation verified. The whitepaper was extracted from the original PDF using PyMuPDF with image blocks excluded. The cypherpunks archive was sourced from github.com\/cryptoanarchywiki\/2000-to-2016-raw-cypherpunks-archive. Nick Szabo\u2019s corpus (217,687 words, 306 posts) was scraped from unenumerated.blogspot.com. Craig Wright\u2019s corpus (6,902 words) was manually collected from his Medium blog. All analysis tools were built in Python and are fully reproducible.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img decoding=\"async\" width=\"1024\" height=\"683\" data-src=\"https:\/\/www.ulclegal.com\/wp-content\/uploads\/2026\/03\/cyberpunk_hacker_bitcoin-1024x683.webp\" alt=\"\" class=\"wp-image-11393 lazyload\" data-srcset=\"https:\/\/www.ulclegal.com\/wp-content\/uploads\/2026\/03\/cyberpunk_hacker_bitcoin-1024x683.webp 1024w, https:\/\/www.ulclegal.com\/wp-content\/uploads\/2026\/03\/cyberpunk_hacker_bitcoin-300x200.webp 300w, https:\/\/www.ulclegal.com\/wp-content\/uploads\/2026\/03\/cyberpunk_hacker_bitcoin-768x512.webp 768w, https:\/\/www.ulclegal.com\/wp-content\/uploads\/2026\/03\/cyberpunk_hacker_bitcoin-480x320.webp 480w, https:\/\/www.ulclegal.com\/wp-content\/uploads\/2026\/03\/cyberpunk_hacker_bitcoin-992x661.webp 992w, https:\/\/www.ulclegal.com\/wp-content\/uploads\/2026\/03\/cyberpunk_hacker_bitcoin-1200x800.webp 1200w, https:\/\/www.ulclegal.com\/wp-content\/uploads\/2026\/03\/cyberpunk_hacker_bitcoin-272x182.webp 272w, https:\/\/www.ulclegal.com\/wp-content\/uploads\/2026\/03\/cyberpunk_hacker_bitcoin.webp 1536w\" data-sizes=\"(max-width: 1024px) 100vw, 1024px\" src=\"data:image\/svg+xml;base64,PHN2ZyB3aWR0aD0iMSIgaGVpZ2h0PSIxIiB4bWxucz0iaHR0cDovL3d3dy53My5vcmcvMjAwMC9zdmciPjwvc3ZnPg==\" style=\"--smush-placeholder-width: 1024px; --smush-placeholder-aspect-ratio: 1024\/683;\" \/><\/figure>\n\n\n\n<p><em>\u2013 End of Report \u2013<\/em><\/p>\n","protected":false},"excerpt":{"rendered":"<p>A Blind Stylometric Search Across 12,979 Authors and 104,901 Texts with Multi-Baseline Analysis, Legal Motive Verification, and Candidate Counter-Evidence Assessment Prepared by JUDr. Mag. J\u00e1n \u010carnogursk\u00fd, MBA (ulclegal.com) March 2026 Original research. All scripts, corpora, and data available on request. The relevant stylometry results and sources are in the Appendix. 1. Executive Summary This report [&hellip;]<\/p>\n","protected":false},"author":6,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[94,39,40],"tags":[],"class_list":["post-11369","post","type-post","status-publish","format-standard","hentry","category-crypto","category-news","category-press"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.ulclegal.com\/en\/wp-json\/wp\/v2\/posts\/11369","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.ulclegal.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.ulclegal.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.ulclegal.com\/en\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/www.ulclegal.com\/en\/wp-json\/wp\/v2\/comments?post=11369"}],"version-history":[{"count":27,"href":"https:\/\/www.ulclegal.com\/en\/wp-json\/wp\/v2\/posts\/11369\/revisions"}],"predecessor-version":[{"id":11461,"href":"https:\/\/www.ulclegal.com\/en\/wp-json\/wp\/v2\/posts\/11369\/revisions\/11461"}],"wp:attachment":[{"href":"https:\/\/www.ulclegal.com\/en\/wp-json\/wp\/v2\/media?parent=11369"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.ulclegal.com\/en\/wp-json\/wp\/v2\/categories?post=11369"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.ulclegal.com\/en\/wp-json\/wp\/v2\/tags?post=11369"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}