ScratchPad: 2024-05-30 Th: How Guilty Is Trump, Really?; & Stalin Was too Very Much a Naïf with Respect to Hitler; & MOAR...
A scratchpad…
A scratchpad…
Neofascism: Let me see if I understand this: If Trump is found guilty in his current New York state criminal trial, it will be because the jury concludes that:
even though the federal prosectors were not confident that they could convict Trump of consciously and knowingly violating campaign finance laws,
Trump did in fact seek to hide this hush-money transaction—and not because he was scared of his wife—
and he knew it was the kind of expenditure related to his campaign that should not be hidden,
and so the falsification of business records was in pursuit of the crime of “promot[ing] … the election of any person by unlawful means…”
Is that right?:
Quinta Jurecic: Why Did Federal Prosecutors Drop Trump's Hush Money Case?: ‘In 2012, the Justice Department failed to win a conviction <politico.com/story/2012/06/justice-dept… against former Democratic Sen. John Edwards on campaign finance charges related to a similar scheme <politico.com/story/2012/06/justice-dept…, concerning money paid by wealthy backers of Edwards to support his pregnant mistress during his presidential campaign…. The link… is more direct than in the Edwards prosecution, in which Edwards successfully argued that the payments were made to avoid alerting his dying wife to the affair…. Cohen’s testimony… rebut[s] the… argument that Trump had sought to silence Daniels… to spare his family. (Trump’s New York defense has gestured at this argument, though not in as much depth as many commentators expected before the trial.)… As New York courts have interpreted the statute, prosecutors don’t need to prove a FECA violation beyond a reasonable doubt <lawfaremedia.org/article/what-must-pros… only the intent to commit or cover up a violation… <lawfaremedia.org/article/why-did-federa…>
WAIT!! WTF!?!?: Phil Zelikow is a very smart and thoughtful guy—most of the time. But this!
Phil Zelikow: Confronting Another Axis? History, Humility, and Wishful Thinking: ‘Stalin was not naïve about Hitler. But, as Stalin explained to his colleagues at the time, he was coming to regard the Nazi leader as a strategic partner in a wider effort for the ‘have-nots’ to take down the great European powers, including the British Empire.… <tnsr.org/2024/05/confronting-another-ax…>
As Hitler wrote in Mein Kampf:
Adolf Hitler: Mein Kampf: ‘German policy… must not be directed by cosmopolitan folkish drivel…. The right to possess soil can become a duty if without extension of its soil a great nation seems doomed to destruction. And most especially when not some little nigger nation or other is involved, but the Germanic mother of life, which has given the present-day world its cultural picture. Germany will either be a world power or there will be no Germany…. And so we National Socialists consciously draw a line beneath the foreign policy tendency of our pre-War period. We take up where we broke off six hundred years ago. We stop the endless German movement to the south and west, and turn our gaze toward the land in the east. At long last we break of the colonial and commercial policy of the pre-War period and shift to the soil policy of the future. If we speak of soil in Europe today, we can primarily have in mind only Russia and her vassal border states. Here Fate itself seems desirous of giving us a sign. By handing Russia to Bolshevism, it robbed the Russian nation of that intelligentsia which previously brought about and guaranteed its existence as a state. For the organization of a Russian state formation was not the result of the political abilities of the Slavs in Russia, but only a wonderful example of the state-forming efficacity of the German element in an inferior race… <ia801905.us.archive.org/6/items/mein-ka…>
Hitler did not see himself as leader of a “have-not” power that needed to take down the British Empire. Hitler saw himself as honor-bound to conquer all of Eastern Europe up to the Urals, divide the land among German farmers, and herd whoever survived of the Slavic population onto reservations.
In believing that Hitler could be a strategic partner, Stalin was mindbogglingly, extraordinarily naïve about what Hitler was.
MAMLMs: The very sharp Henry Farrell has a nice take on Google’s current headlong charge forward into the unsurveyed and unscouted ground of using GPT LLMs as a natural-language interface to search. I share his belief that such will flop when applied to low-quality but might prove very useful in summarizing central tendencies of high-quality data.
But, then, this gives me considerable pause:
I asked ChatGPT4o for a Chicago Manual citation for the paper at the url <https://pubs.aeaweb.org/doi/pdfplus/10.1257/jep.12.4.151>. Its reply? This: “Samuelson, Paul A. 1998. ‘The Historical Background of the Communist Manifesto’. Journal of Economic Perspectives 12 (4): 151-174. Accessed May 25, 2024. <https://web.archive.org/web/*/https://pubs.aeaweb.org/doi/pdfplus/10.1257/jep.12.4.151>. Please let me know if you need any further assistance!” The “pp. 151-174” is right. The “12 (4):” is right. For some reason or other “(Fall)”) is omitted. The “Journal of Economic Perspectives” is right. The “‘The Historical Background of the Communist Manifesto’” is right. The “1998” is right. But the “Samuelson, Paul A.” is wrong—the author is “Boyer, George R.”
You can understand how a GPT LLM might do this: Paul A. Samuelson published a lot of economics papers and appears in a hugely larger share of the training corpus than does George R. Boyer. But that means that even restricting the training corpus and the RAG context to the highest possible data is not going to fix things:
Henry Farrell: ‘Large language models are engines of summarization…potentially valuable where [(a)] you… want… summarizations, (b) the data… is fairly high quality, (c) you are prepared to tolerate… slop, and (d) you aren’t too worried if… outputs… drift towards [the] central… training corpus…. The “certain amount of slop” could end up being a real problem… even when the slop isn’t that much worse in absolute terms… than Google Original Flavor. You can tolerate slop as long as people don’t directly attribute it to you…. The outliers and soon-forthcoming New-Googlebombs that we’re focusing on right now are probably less consequential than the models’ convergence on central cultural tendencies… <https://www.programmablemutter.com/p/google-ai-fails-the-taste-test>
I uploaded a 92k word, unpublished sci book that I wrote to GPT4, then asked it basic plot questions. Who is the protagonist, did this character die, etc? The LLM usually failed, I think because fiction writing tends to imply rather than state important points. It even had trouble identifying dialogue because it didn't identify smart quotes as quotation marks - until I corrected the model. I'd like to see a larger test on understanding fiction.