Revisiting Hypno-Bert and Political Persuasion

A little over a year ago I decided to take a look at some pseudo-science brought to us by Scott Adams of Dilbert fame. Scott likes to fancy himself knowledgeable on persuasion because he is a “trained hypnotist,” which does not follow logically. I read through this old post of mine and found this excerpt interesting:
He did pick Trump to win the nomination when many professional pundits did not, but those pundits were actually suffering from their own cognitive dissonance. The polls clearly predicted that Trump was holding a solid plurality of votes in a field with 16 opponents, and the lead did not diminish as opponents dropped out. The same polling trends have indicated a Trump loss in the general election fairly consistently for the entirety of the election season, yet Scott predicted otherwise – he predicted a landslide victory by Trump, because Trump is so good at persuading people. (That’s an extraordinarily naïve view of how Trump acquired and maintained his following, but Scott doesn’t really understand what motivates human behavior anyway, so it’s hardly surprising.) Now that he is clearly not going to be vindicated, he is changing his endorsement and blaming the most recent scandal (which undoubtedly has hurt Trump, but Trump had no clear path to 270 prior to these most recent events).
Written before the election, this prediction of mine based on the data I was watching at the time turned out to be incorrect. Does it turn out in retrospect that Scott was vindicated after all? If he had been steadfast in his belief that Trump would win, maybe. However, he was not. He repeatedly switched and hedged and allowed himself outs if his predictions were not to come true. After the 11th-hour Comey statement about Hillary’s e-mails the poling shifted and Scott went back to calling for a landslide Trump victory. Trump did eek out an electoral victory, but it was hardly a landslide; not to mention the popular loss by nearly three million votes. Certainly this doesn’t amount to a strong vindication for Scott, if vindication at all. [Of course, I was obviously wrong about who would win, but my assessment was based on the polling data available at the time, and not a belief in my own magical ability to predict things without facts.]
The original post drew a comment from a reader that described Scott’s situation with uncanny accuracy. I will share the comment here:

I myself have become fascinated by this guy, probably for the wrong reasons. I was a huge Dilbert fan back in the day and I loved his books too. After reading his blog I’ve been starting to realize that the parts I thought were satirical might have actually been 100% serious. My fascination has less to do with the things he says (which have stopped making any logical sense long ago) and more in the sheer number of new and creative ways he manages to be wrong and contradict himself. He talks about cognitive dissonance and confirmation bias as though they are concepts that he invented himself; and while he does (correctly) claim that everyone suffers from them, he seems to be blind to his own in a way that makes me conclude that he’s either A) trolling or B) not very smart.

If you read his books you might be able to glean some context. The success of Dilbert screwed with his brain in a way – he notes that he is not a talented artist, nor does he really know how to construct a joke (he does have a good sense of humor, though), and he sees Dilbert not as being in the right place at the right time but rather evidence that he has special persuasive or reality-altering powers over the universe. I am actually not making that up. Predicting that Trump would win the primary is exactly the sort of thing he’s been aiming to do his whole life: be right about something that everyone else was wrong about because you saw something they didn’t. He is not exactly a great predictor – nearly every other political prediction he makes is wrong, as is almost everything he predicts Trump will do in the future. But he got this one right, and as a result convinced himself that everything else he writes must be correct, as well. With this in mind, his latest blog posts make some kind of sense. They still strike me as incoherent, but the man’s own sense of self-importance has always been his defining quality.

Being sort-of right about Trump’s victory will only reinforce Scott’s misconception that he has a special gift for persuasion and prediction. His own biases allow him to ignore his own contradictory statements and multitude of incorrect predictions and see only the handful of coincidences where his predictions were correct, even when underpinned by faulty logic or incorrect data.
The original post is below for those interested enough to continue.

2496084215_aae4e437f3_bFor something different, I thought it might be fun to take a direct swipe at self-described minor celebrity Scott Adams. Mr. Adams, if you don’t already know, is the creator of Dilbert, a moderately funny syndicated comic that has been around since 1991-ish. Over time, Adams has successfully expanded the Dilbert empire, which now includes books – not just the ones that are full of already released cartoons that we inexplicably pay for anyway, but actual books that usually include semi-witty business insight and old cartoons – a short-lived television series, and a website that hosts all of the Dilbert cartoons and Scott Adams’ blog. The blog is what led me to decide to pick on Mr. Adams a bit. Being political season, most of his blogs over the last 13 months or so (I’m not certain of the time-frame) have been political.

I’m not going to focus on the specific political opinions expressed too much, though I will give a passing mention to his obvious fascination with Donald Trump. While he initially denied it in his blog, it has been clear for a long while that he is a strong Trump supporter. He offers an assortment of shallow apologetics for Trump’s blatant racism, misogyny, religious bigotry, xenophobia, and obvious ignorance of pretty much all policy issues, and routinely offers the standard argumentative fallacies to bolster his position. (Most frequently he employs the straw-man, ad-hominem, and false equivalence. These are very popular argumentative fallacies; Mr. Adams is in good company here.) But this isn’t what really struck me about the blog.

The running theme, and Scott’s asserted reason for being interested in the Trump candidacy, is the art of persuasion. Scott holds himself up as something of an expert in persuasion. (I’m not certain that he ever used the word expert to describe himself, but it is clear that he at the very least considers himself extremely knowledgeable.) He has gone on and on about the artistry of Trump’s persuasion skills, and on this basis has been invited to speak publicly about his thoughts on the presidential election. I suppose already being a minor celebrity paved the way – I doubt he would be on TV and radio for any opinions otherwise. But what is the basis for his claim that that he knows something about persuasion? Why, he is a trained hypnotist! Sadly, that’s really not a strong foundation for being knowledgeable in persuasion.

If you’re not familiar, hypnosis is a technique that falls under the broad category of neuro-linguistic programming, or NLP. A fun fact about NLP is that it’s 90% bullsh!t, and the other 10% isn’t well studied. I’m writing this as someone who has spent some time researching NLP as it relates specifically to influencing behavior and detecting deception. I have paid great attention to and received detailed instruction in varying applications of NLP. The problem with NLP as a proposed scientific endeavor is that it makes no testable claims. My point here isn’t related to NLP as a broad field, so I will just stick to hypnosis, as this is the foundation of Scotts’ claim to know anything special about persuasion.

Hypnosis is claimed to influence behavior through inducement of a trance-like state and then implanting suggestions. Or, without the trance, bypassing the conscious filter and pushing information straight into the unconscious, again inducing a reaction. This isn’t truly testable or falsifiable, of course. If it doesn’t work then that’s because some people aren’t susceptible to hypnosis. Are other applications of scientific or medical practice subject to individual perception the same way? Not really. If I load you up with a sedative, you’re going to sleep. Now, there are physiological differences that might impact dosage requirements, and there can also be side effects that don’t manifest in everyone, but there are identifiable physiological causes for this. If you speak soothingly to me and then suggest that I cluck like a chicken, I can assure you that I won’t. If hypnosis was a valid, testable application with repeatable results then people who were not susceptible ought to be the exception rather than the rule, and there should be a physiological explanation for why it didn’t work in those rare instances. It is this inability for the “theory” to be tested and falsified that lands it squarely in the realm of pseudo-science. Also, hypnosis isn’t specifically about persuasion in the conventional sense, so being a hypnotist – even if it was a valid practice – wouldn’t necessarily make one knowledgeable on persuasion.

I’m not saying that there aren’t studies on persuasion, or that there aren’t methods for persuading someone to do something. I am saying that hypnosis isn’t a validated method. Getting to the bottom of how and why specific methods of persuasion work requires in-depth study and understanding of behavioral psychology. Practical application, however, can be learned by observation. Some things are nearly intuitive. When Scott speaks of the intricacies of persuasion in his blog, he doesn’t refer to clinical studies or research papers. He makes mention of observables that pretty much everyone has experienced, such as the earth-shattering observation that people respond to emotional arguments more strongly than rational arguments. If we wanted to we could probably dig up dozens, maybe hundreds (or more) studies and meta-studies about this. But, we don’t really need to, because we all experience this every day (unless we never interact with other humans). Scott also likes to reference cognitive bias or cognitive dissonance regularly. He has the detailed understanding of these phenomena of a person who read about it on the internet. But, really, that’s good enough since the people that he seeks to influence are the people that are already biased toward believing him. And since this sort of bias is something that everyone sees every day, there is no need to cite scientific studies on the matter. This doesn’t mean, however, that he has any special expertise. It’s the intellectual equivalent of telling people that a strong wind will blow your toupee off your head, and then claiming to be a wind expert.

Since hypnosis isn’t based on any valid scientific principles, and hypnosis doesn’t deal with conventional persuasion in the first place, it seems unnecessary to even continue on about it. I will anyway. Even within the circles of people who purport that hypnosis is a real thing, there is no consistency on what it means to be “certified.” Some certifications require intense training for relatively long periods of time, in an effort to give the appearance of legitimacy and scientific rigor. Others you can get in an afternoon. Scott is a “certified hypnotist,” but we don’t really know what that means. If you are someone that puts stock in hypnotism, despite the lack of any evidence that you should, you are probably still going to want to determine whether someone spent hours studying and practicing or if they went to a 1 day workshop in Las Vegas. (There are online certifications too, in case you were wondering).

In summary, Scott Adams has no basis to claim expertise or even true knowledgeability in persuasion. He’s managed to get himself booked on television and make political predictions based on his claimed insight, but really he’s just guessing. And, like any good pseudoscientist, there’s nothing that can truly make his predictions wrong. In most cases he hedges his bets by making qualified predictions, allowing him to claim that he was right regardless of outcome. He did pick Trump to win the nomination when many professional pundits did not, but those pundits were actually suffering from their own cognitive dissonance. The polls clearly predicted that Trump was holding a solid plurality of votes in a field with 16 opponents, and the lead did not diminish as opponents dropped out. The same polling trends have indicated a Trump loss in the general election fairly consistently for the entirety of the election season, yet Scott predicted otherwise – he predicted a landslide victory by Trump, because Trump is so good at persuading people. (That’s an extraordinarily naïve view of how Trump acquired and maintained his following, but Scott doesn’t really understand what motivates human behavior anyway, so it’s hardly surprising.) Now that he is clearly not going to be vindicated, he is changing his endorsement and blaming the most recent scandal (which undoubtedly has hurt Trump, but Trump had no clear path to 270 prior to these most recent events).

In the meantime, I suppose Adams can continue to assert that it is some other “master persuader” that out-Trumped Trump, while claiming simultaneously that he can learn any topic in one hour and that no one knows anything about any topic. Or, perhaps, he should just stick to recycling his same seven jokes in his comic strip and writing almost-business books. Then again, no publicity is bad publicity. Bad logic and baseless assertions are getting his name out there. Good for Scott, I suppose, but not good for rational discourse.