Recently, James_Miller posted a conversation between Sam Harris and Scott Adams about Donald Trump. James_Miller titled it "a model rationalist disagreement". While I agree that the tone in which the conversation was conducted was helpful, I think Scott Adams is a top practitioner of the Dark Arts. Indeed, he often prides himself on his persuasion ability. To me, he is very far from a model for a rationalist, and he is the kind of figure we rationalists should know how to fight against.
Here are some techniques that Adams uses:
- Changing the subject: (a) Harris says Trump is unethical and cites the example of Trump gate-crashing a charity event to falsely get credit for himself. Adams responds by saying that others are equally bad—that all politicians do morally dubious things. When Harris points out that Obama would never do such a thing, Adams says Trump is a very public figure and hence people have lots of dirt on him. (b) When Harris points out that almost all climate scientists agree that climate change is happening and that it is wrong for Trump to have called climate change a hoax, Adams changes the subject to how it is unclear what economic policies one ought to pursue if climate change is true.
- Motte-and-bailey: When Harris points out that the Trump University scandal and Trump's response to it means Trump is unethical, Adams says that Trump was not responsible for the university because it was only a licensing deal. Then Harris points out that Trump is unethical because he shortchanged his contractors. Adams says that that’s what happens with big construction projects. Harris tries to argue that it’s the entirety of Trump’s behavior that makes it clear that he is unethical—i.e., Trump University, his non-payment to contractors, his charity gate-crashing, and so on. At this points Adams says we ought to stop expecting ethical behavior from our Presidents. This is a classic motte-and-bailey defense. Try to defend an indefensible position (the bailey) for a while, but then once it becomes untenable to defend it, then go to the motte (something much more defensible).
- Euphemisation: (a) When Harris tells Adams that Trump lies constantly and has a dangerous disregard for the truth, Adams says, I agree that Trump doesn’t pass fact checks. Indeed, throughout the conversation Adams never refers to Trump as lying or as making false statements. Instead, Adams always says, Trump “doesn’t pass the fact checks”. This move essentially makes it sound as if there’s some organization whose arbitrary and biased standards are what Trump doesn’t pass and so downplays the much more important fact that Trump lies. (b) When Harris call Trump's actions morally wrong, Adams makes it seem as if he is agreeing with Harris but then rephrases it as: “he does things that you or I may not do in the same situation”. Indeed, that's Adams's constant euphemism for a morally wrong action. This is a very different statement compared to saying that what Trump did was wrong, and makes it seem as if Trump is just a normal person doing what normal people do.
- Diagnosis: Rather than debate the substance of Harris’s claims, Adams will often embark on a diagnosis of Harris’s beliefs or of someone else who has that belief. For example, when Harris says that Trump is not persuasive and does not seem to have any coherent views, Adams says that that's Harris's "tell" and that Harris is "triggered" by Trump's speeches. Adams constantly diagnoses Trump critics as seeing a different movie, or as being hypnotized by the mainstream media. By doing this, he moves away from the substance of the criticisms.
- Excusing: (a) When Harris says that it is wrong to not condemn, and wrong to support, the intervention of Russia in America’s election, Adams says that the US would extract revenge via its intelligence agencies and we would never know about it. He provides no evidence for the claim that Trump is indeed extracting revenge via the CIA. He also says America interferes in other elections too. (b) When Harris says that Trump degraded democratic institutions by promising to lock up his political opponent after the election, Adams says that was just a joke. (c) When Harris says Trump is using the office of the President for personal gain, Adams tries to spin the narrative as Trump trying to give as much as possible late in his life for his country.
- Cherry-picking evidence: (a) When Harris points out that seventeen different intelligence agencies agreed that Russia’s government interfered in the US elections, Adams says that the intelligence agencies have been known to be wrong before. (b) When Harris points out that almost all climate scientists agree on climate change, Adams points to some point in the 1970s where (he claims) climate scientists got something wrong, and therefore we should be skeptical about the claims of climate scientists.
Overall, I think what Adams is doing is wrong. He is an ethical and epistemological relativist: he does not seem to believe in truth or in morality. At the very least, he does not care about what is true and false and what is right and wrong. He exploits his relativism to push his agenda, which is blindingly clear: support Trump.
(Note: I wanted to work on this essay more carefully, and find out all the different ways in which Adams subverts the truth and sound reasoning. I also wanted to cite more clearly the problematic passages from the conversations. But I don't have the time. So I relied on memory and highlighted the Dark Arts moves that struck me immediately. So please, contribute in the comments with your own observations about the Dark Arts involved here.)
I think the term "Dark Arts" is used by many in the community to refer to generic, truth-agnostic ways of getting people to change their mind. I agree that Scott Adams demonstrates mastery of persuasion techniques, and that this is indeed not necessarily evidence that he is not a "rationalist".
However, the specific claim made by James_Miller is that it is a "model rationalist disagreement". I think that since Adams used the persuasion techniques that Stabilizer mentioned above, it's pretty clear that it isn't a model rationalist disagreement.
I agree, and I didn't mean to imply otherwise.