On Trying Not To Be Wrong

Epistemic Status: Exhortative

I did not expect Donald Trump to become President.

I did not expect him to win, or even get very far, in the Republican primaries. Like many people, I thought the idea of Donald Trump becoming president was “weird” or “surreal” or “not a thing that happens.”  Like many people, I’ve thought 2016 was a surreal year; the Cubs won the World Series, Hillary Clinton went on television to warn people about white-supremacist memes, Elon Musk has landed rockets on ocean platforms and started an organization to develop Friendly AI.  Surreal, right?


It’s real, not surreal. If reality looks weird, this means our stories about it are wrong.

Did polls and newspapers and social media fail to see this election coming? Then those sources just took a hit in credibility.

On a longer-term note, if you know there’s a replication crisis in scientific research, that should be shaking up your trust in published papers.

There may be a crisis in politics. But before we can do anything sensible about that, we need to understand that there is a crisis in credence. If the world looks weird to you and me today, that is not a matter for rueful laughter, it is a sign that we are probably badly wrong about lots of things.

And being totally wrong about how the world works is a threat to survival.

What this election brought home to me is that I don’t want to be wrong any more.

A lot of things people say and write are not really what they think is true about the world. They’re expressions of emotion or identity or solidarity. Arthur Applebee writes about “the affective center” being the prototypical use of language: gossip, relationship-building, sharing feelings.  Evolutionary biologist Geoffrey Miller thinks that language, as well as the high intelligence needed to navigate social relations, evolved in humans through sexual selection.  We are built for these “soft” uses of language and thought: to relate, to bond, to politick. Doing that feels good and intuitive and healthy. It feels “human”, because it is.

It is, of course, also not truth-seeking. We have a host of cognitive biases, particularly in these normal, social uses of speech.  Much of it is bullshit, in Harry Frankfurt’s terminology: not so much lies as not about truth in the first place.

But it is possible for language to be used in ways that do point towards truth.  Arguments where premises follow from conclusions — when people are reading to check whether premises follow from conclusions and facts are supported by evidence — actually do hold independently of shifting social contexts.

What that requires socially is for people to do something quite unnatural. Instead of going with the flow (relating, politicking, sharing feelings, bonding), it involves breaking the flow. Nitpicking. Disagreeing. Being dry and technical. Fact-checking.

The motto of the Royal Society of London was “Nullius in verba”, or “Take nobody’s word for it.”  The modern scientific tradition was founded by a small group of people defined by the fact that they would call bullshit on things unless they were demonstrated by evidence.  This is inherently disagreeable — literally, it involves disagreeing.

The contemporary flow of social media works to prevent discussion and argument, while earlier internet formats worked to promote it.

A blog and a comments section, or a forum, or an email thread, is set up to encourage discussion.  One has space to write long-form, multi-paragraph essays, which are stored permanently; and then there is ample space for other people to write long-form, multi-paragraph responses to those essays.  One can respond to specific points separately. One can respond to responses, in long nested threads.  It is good form to cite sources (links and hat-tips).  The social reward for writing is getting a response to your writing. The world of blogs and forums provided an alternative to mass media that was more discursive and more intellectual.

Newer forms of social media inhibit discussion and promote simpler affective responses.

Facebook still allows threaded discussions, but the unit of attention is the like. You get socially rewarded when lots of people agree with you; agreement, being cheaper than response, will wind up being more abundant and hence more dominant, in a sort of Gresham’s Law phenomenon.

Twitter doesn’t allow room for long-form content, of course, and the first-class actions are liking and retweeting: approval and copying.  Tumblr allows for paragraph-long posts, but has such poor threading that it’s difficult to hold a discussion, structurally prefers images and videos to text, and makes liking and reblogging first-class actions. Snapchat and Instagram are deeply unfriendly to text.

And all of the above social media platforms have an endlessly scrolling feed, which makes conversations ephemeral and difficult to reference.

The incentives are against discussion and towards response. Instant emotional readouts, approval or disapproval.  Image rather than language. Copying rather than original writing.

The medium is designed for agreement, not disagreement. People who can’t communicate in a way that will rack up “likes” or “favorites” tend to quietly withdraw.  And that feels awkward, like they’re spoiling the party, like a kind of Puritanism.

But I also suspect that these Puritans, these disagreeable people, have something to teach us. Especially now.  Especially when it’s become clear that “believing” things out of mood affiliation leads to very wrong conclusions about the world.

Humans love to socialize. One of the things that we like to do with leisure and technology is talk to each other. I’m not denigrating this normal human urge, which finds expression in social media.

What I’m saying is that one of the remarkable things that can be done with human intelligence, language, and socializing is to have discussions. Arguments. Conversations. Science is a form of conversation, as is philosophy.  Out of the natural “affective center” of gossip can come something that is a bit more unnatural but extraordinarily powerful: you can come to reliable and referenceable common knowledge.  You can refer back to “oh, yeah, on February 15th Bob did this experiment and on February 16th Alice tried it herself and got the same result, and then Carl found a flaw in the design so it turned out Alice and Bob were wrong.”

It’s not about liking a claim, it’s about being convinced by it.

I’m not talking about something that necessarily has to be high-minded and out of reach, or something that will singlehandedly fix the problems we face.

But it seems really useful now to start having discussions again. On blogs, on forums, on email, in contexts where it’s socially rewarding to disagree and pick things apart, rather than to merge into mobs of agreement.

It’s urgent to figure out what is going on in the world and how we can keep it from hurting us.  And that means our errors need to be corrected.

If this seems kind of dutiful and unpleasant, compared to the warm rush of likes and reblogs and image-sharing, consider that it’s actually kind of fun in its own right to have arguments and discussions and to dig into the nitty-gritty. It’s the nerdy kind of fun that gets a kick out of details, out of facts, out of messing around until things click into place satisfyingly.  It’s the mental equivalent of the kind of fun it is to make things with your hands, or to play games.

The Scottish Enlightenment was a little, well, Scottish.  What are the Scottish stereotypes? A little grouchy, a little ornery, a little stingy.  Hard-headedly practical.  Blunt. In other words, there’s a kind of healthy disagreeableness, that says “don’t give me bullshit”, and “don’t rip me off”, and “I’m sure as hell not going to kneel to you” and “I may not be rich or powerful, but I’m honest”, and “I built machines that work, dammit.”  In a world where liking and agreeing is the currency, I think we are likely to underappreciate the virtues that go with disagreeableness.

I’m going to lean more into disagreement and fact-checking. I’m going to try to appreciate the people who say “not exactly” instead of going with the flow. I’m going to aim to have my discussions in contexts that are actually designed for discussion.  I care about not being wrong, now, in a way I really didn’t before.  And I encourage others to consider doing the same.


32 thoughts on “On Trying Not To Be Wrong

  1. I like this post, as it stresses some themes/discussions I’ve had with people over the years.
    I’m often seen as being, “Too critical”, “Too disagreeable”, but while I acknowledge that, the plain fact is that “Critical Thinking” requires criticism. Popular culture has moved people to imagine any criticism is wrong or at least negative. But lack of criticism is leading us towards Idiocracy!
    I did predict Trump winning the nomination and did predict him winning the election, and in fact won 5 public $100 bets against 5 thought leaders.

      • There were a few things.
        1. Democracy depends on convincing a mob to support you – so you have to be able to understand and relate to disparate groups and interests. Trump has this ability in spades, and the fact that he ran a very successful show, playing a businessman who dealt fairly with all kinds of diverse folks, really helped here.
        2. Common folk don’t want and can’t appreciate policy details, they want broad assurances, and Trump catered to that need, using simple declarative statements.
        3. Most of the US electorate are ACTUALLY hurting badly financially, and have legitimate grievences against the powers that be, who exported their jobs and future and left them to rot.

        But then I find predicting US Presidential elections easy, and have correctly predicted all presidents, since I was a 10yo, growing up in rural Ireland. I remember as a kid contradicting my aunt – who was the equivalent of Ireland’s Surgeon General – over Easter dinner one time, and insisting that Ted Kennedy would never be president. At the time she and all Irish relatives felt he was a shoe-in for president.

      • But past presidents weren’t like Trump. Is it just that populists are gradually becoming more successful and you noticed that we’ve hit a tipping point?

      • I picked Obama and Bill Clinton to win too, both over a year before they did – very easy picks!
        The most difficult recent pick was Bush v Gore, and I didn’t as much pick Bush to win, as much as I picked Gore to lose – due to turning his back on Bill Clinton and not letting him campaign for him.

      • Sarah,
        What I’m saying is that the more populist nominee wins 9 times out of 10. The problem is not this election or a previous one, the problem is endemic to Democracy. As IMO, Democracy = Mob Rule.

        I first came to that conclusion in my late teens, but since I wasn’t overly concerned with politics, I gave scant attention to the obvious question, “What system would be better than Democracy?”

        Back when I was a more regular LW attendee, a group of us read a book called “The Origins of Political Order” by Francis Fukuyama, and a very interesting book it is. Basically he surveys political systems, throughout the ages, and tries to ascertain what works and what doesn’t. He notices that there are 3 groups who make up power in a state:
        1. Military
        2. Bureaucracy
        3. Monarch

        The takeaway is that greater stability of a regime is linked to either #1 or #2 being non-native, but the greatest stability of all is when #3 is non-native. He makes the case that this is why China has had such great stability over the millennia, and adds that curiously England has had great stability without this.
        I remember trying to make a point that was either not well received or else I didn’t explain it properly; but that in fact all recent Monarchs in England had also been non-native, mostly Celtic from Wales and Scotland, then later Germanic from Holland and Germany. So essentially Fukuyma had the data and almost came to a generalized rule, but through some sort of possible ingroup bias, missed it?!

        The conclusion should have been that the optimal form of government is a neutral, absolute monarch!

      • Paul, if you’re confident in your ability to predict presidents, how would you design a career to optimize to be president? What’s an example of a famous person who would easily be able to become president according to your model? Are you sure your correct predictions are not mere survivorship bias?

      • Foobar,
        Excellent question! Off hand I’d imagine that Ivanka Trump could become the first female, jewish US President – if she so desired, and my spidey senses say she might.

  2. Not sure if the bit about Facebook’s incentives entirely matches my experience. With old-style forums, a major disincentive towards writing long thoughtful posts was that a winning argument would most frequently be met with a complete silence. If you completely made your case, then nobody could argue against you and the discussion would just stop. But on the other hand, if people got totally fed up with you and decided to disengage rather than continue beating their heads against a wall, that too would manifest as total silence.

    So the feedback that you got would be one that didn’t distinguish between a brilliant argument and a stupid argument, or for that matter your interlocutor just getting run over by a truck.

    This is a major reason why discussion on Facebook (or some other forum that implements a “like” mechanic) feels a lot more rewarding than discussion on old-style forums. In addition, the “like” mechanic can also make the conversation feel a lot more cooperative, since you can like the comments of your interlocutor and thus signal your appreciation of the conversation even as you continue disagreeing.

    • I think that’s true. A related problem is “Why Our Kind Can’t Cooperate.” If agreement is represented by silence, then it will appear that nobody agrees, when in fact there is lots of agreement. This is bad for community action, and doesn’t reward people for doing good stuff. I’m not really sure how to square this circle. It seems like you can either make it easy to say “yay good job” or make it easy to criticize, and we need both.

      These aren’t the only two means of social reward, I don’t think. Competition — prizes or honors to the *best*, with the understanding that not all things can be best — might be another potential mechanism.

    • It’s sad that the electronic forums never developed RAEBNC (read and enjoyed but no comment), a thing which was common in the apas (amateur press associations), the print versiion of internet discussion.

      Mind you, I found getting RAEBNC frustrating– I wanted discussion– but it was definitely better than nothing.

      Maybe your essay will supply some incentive towards recreating trn for the web. It’s definitely the best mode for electionic discussion I’ve seen, and somehow no one wants to rebuild it– apparently it would take a lot of hard boring work. Maybe there’s something simpler possible with similar virtues.

  3. A huge problem with people trying to adapt to reality is being unwilling to let go of “but everyone KNOWS this is terrible, right?”

    I’m not sure if this is Ra or more simple tribalism, or “Common sense” and the absurdity heuristic, but it’s unfortunately ubiquitous. It stretches from doctors not adapting modern checklists or historically not caring about handwashing all the way to disagreements about minimum wage, whether spanking is ok, and whether trump is catastrophically evil and insane. There are bubbles of millions or even billions of people who think the answer to questions like this is obviously one way, and equally large bubbles who, just as insistently, say the answer is the opposite. At this point I generally despair of raising the sanity waterline, as EY once wanted to do, because the way to avoid these problems is to actually DO the thinking, DO the research, and this is difficult enough for clever people with free time, let alone everyone else.

    • I’m not trying to make *everyone* more correct. I think there should be at least one group of people trying to be correct.

  4. There major benefit of agreeableness is coordination: when you have a large group you generally want them acting on the best known strategy. Part of the challenge of teamwork is getting people to spend X% of their resources on enacting the current best plan and Y% on improving the plan–in practice, this usually ends up with a small number of people doing all the planning, which seems quite clearly worse than all people doing some of the decision-making.

    Part of the challenge is finding group norms that get the epistemic benefits of disagreeing and the coordination benefits of agreeableness. I think I have a sense of how to train this as an individual skill, but it takes time and effort–would be much nicer to find systems that make it easy.

    • Satvik,
      The problem with widespread agreeableness in a group is it can lead to massive “Ingroup Bias”, and cognitive dissonance when the ingroup bias is not acknowledged.
      IMO, you can have either strong group cohesion or rationalism, but rarely both.

      • Paul,
        I agree that there’s tension between cohesion and rationalism, but I’m a bit more optimistic about potentially achieving both–for example, Bridgewater seems to pull it off pretty well.

  5. Great post. Looking forward to checking out your other writing now.

    The way I would put it is that while longform conversations are technically possible on Facebook, the platform doesn’t condition you for that type of behaviour. The instant gratification of ‘likes’ is definitely part of the story (who would think humans would put so much weight on a little icon turning blue?). The other important thing is that Facebook (and all other social media) is designed to trigger novelty circuits through a constant barrage of pictures and videos. I suppose someone could use Facebook for extended intellectual discussions with others, but I feel like once you fall into the habit of scrolling through your news feed (and the concomitant bursts of dopamine that go along with that), from thereafter, that is what you will want to do on Facebook. It is addictive, and I am currently addicted to it, I have to admit. I noticed falling into an addiction when I first started using Twitter, and managed to largely limit my use. Another thing I noticed is that my attention span is low on Facebook, Twitter, or online News sites, while it is higher on blogs. I literally can’t read a long form piece on news sites anymore — I have to send it to my Kindle. This conditioning makes sense considering the economics – the more pages you click through, the more chances to send ads.

    Stephen Weinberg once remarked how images are starting to dominate text. We now communicate with memes and emoticons. I don’t think this is a healthy, for a variety of reasons (that would be an entire post in itself).

  6. I think you’re right, but there are two more problems with electronic discussion as it exists now– one is that it incentivizes disliking as well as liking and another is that it incentizes speed.

    I’ve been wondering about the popularity of venues that make search difficult– not just Facebook, but also Slatestarcodex.

  7. Thank you for the essay. I think it is substantially a good approach, and I have a few minor comments.

    * You start by saying you want to be less wrong, but your suggested actions (being willing to disagree) seem to be more directed at helping others to be less wrong. I guess the idea is to surround yourself with others who practice these sorts of actions. Also, maybe you practicing it yourself can lead to you being less wrong, if you listen carefully to people’s responses when you disagree (and find the right people to disagree with).

    * I think cultivating understanding and empathy are actually helpful — even critical — to this project. It seems like a mistake to contrast agreeableness with disagreeableness in general. Maybe what you really need is to become less agreeable with your intellectual and social kin and _more_ empathetic, tolerant, and interested in your intellectual and social aliens.

    * In fact, while reading your essay, it struck me that the values and practices that you are advocating for here are similar to those advocated by a set of people who are (presumably) on the opposite end of the political spectrum from you: certain kinds of Trump supporters and alt-right thinkers. It could be that you could find epistemic allies among your political opponents. Here’s an essay about this that I love: https://jacobexmachina.blogspot.com/2016/10/election-2016.html . (Note: *not* to imply anything about the author’s political identity or label. That’s not the point!)

    * Likewise, your dawning realization that mainstream media and political discourse is confident, authoritative, reputable, and yet incorrect — that realization too is a major force in the thinking of that aforementioned social group. (I love your allusion to the parallel failure/corruption in the edifice of mainstream scientific research. That’s one of my favorite topics, as evidenced by my previous contribution to your blog — my comments in: https://srconstantin.wordpress.com/2015/12/14/contra-science-based-medicine/ )

    * You advocate for forums and tools better-suited to serious conversation. I want that, too, but I don’t know how to get it. Unfortunately, the most likely outcome of this conversation here is that I’ll never see anything further that you write, whether you wrote it in response to my comments or not. Maybe there’s some tool or service that I don’t know about that would help me manage that, but basically it probably requires some degree of _commitment_ over time from you and me both, that we’ll pay attention to the other and we’ll reply, and such a commitment seems extremely costly.

    * If you haven’t yet seen it, here is a new essay closely related to your topic that I strongly recommend: http://www.meltingasphalt.com/crony-beliefs/

    • So, I don’t naturally tend towards personal unpleasantness, nor do I think it’s a good idea. I was trying to push back against a certain kind of empathy-first social cognition that will never *notice* when a friend/ally is incorrect, and which I think has caused harm in my own community.

      I think making more discussion-friendly web tools is a UX problem, which is not my field, but in the meantime the easy thing is to shift to places that already exist, and try to poach people from FB/Twitter/etc to respond directly on blogs.

  8. >”People who can’t communicate in a way that will rack up “likes” or “favorites” tend to quietly withdraw. And that feels awkward, like they’re spoiling the party, like a kind of Puritanism”

    Ah, so very insightful. You might recall in college all I did was disagree. I tried to make it fun for people, and did a fairly good job of that I think, but I was a relentless gadfly all day every day. If I ever did agree with somebody, that was *boring* and time to move on to a new topic.
    But on facebook, I’ve been silent. I tell myself its just prudent to leave a minimal online presence, but I think you may have more accurately captured my sentiments. I’m not a “likeable” sort. I’m ornery, practical, and blunt. If I spoke my mind, apparently 30% of my facebook contacts would “unfriend” me just for voting Johnson, let alone Trump. In college perhaps I didn’t care- I could find a dozen new contacts in half an hour- but today I’m more conservative. More importantly, I have nothing to gain by speaking, and enough to lose. I’m not on particularly warm terms with the myriad fools who deliberately want to isolate their own incestuous echo chamber, but as long as they’re on my feed I get a window into what they’re thinking and avoid becoming the same.

    I recall an extremely similar situation on the OBLW list serve, actually. All contentious or discomforting discussion was relegated to the “evil” list serve- which I found pretty tremendously insulting, but I’ve learned as a lifelong gadfly that nobody cares if you’re insulted because you have no affiliated group to back you up. All the interesting discussions worth talking about were on the main list, and most people weren’t even aware of the “evil” list anyway. It sufficed to choke out contentious discussion, at any rate. Didn’t stop me from making a scene a couple times after that, one in which I recall merely insisting publicly upon my disagreement and demanding citations and evidence- nothing more. I was told to apologize by the moderator for upsetting my interlocutor. I did not apologize. The final time I participated in a thread discussion was when the group decided to kick its first “troll”. I stood up for the “troll” because his only crime was being unable to elegantly disagree as per the politeness norms of the group (norms which I’m sure even my most socially deft gadflying wouldn’t pass now, as I predict they’re more sensitive and pure following their ideological cleansing). I was given a stern warning from the moderator that I too might be kicked if I persisted in standing up for the troll. So I vowed henceforth to remain silent, and quietly enjoy what information (can hardly call it discussion) passed through the list serve.

    As you observe, I became silent. There are probably others. But here’s a bit of disagreement for you- I don’t think long-form blogs and forums will save you. On facebook my favorite threads to read are long multi-post essays and discussions from clever people arguing with each other. My experience on the list serve never involved short exchanges or likes- and I got shut out of there rather promptly when the climate changed. I’ve read plenty of anecdotes about moderators on other forums banning those they find disagreeable. Even 4chan, the septic tank of the internet and troll breeding grounds, has had moderators delete my anonymous posts for criticizing site policy (criticisms which, incidentally, were later implemented). So forums won’t save you, and is social media really worse? I don’t use Twitter, but the hashtag system should make finding those who disagree with you remarkably easy, no? Facebook does filter things based on your likes (though I never like anything), but long discussions are still possible. On Reddit you can even sort posts by most controversial.
    No, I think the only thing that can save you from an echo chamber is yourself. Being a gadfly is its own reward- there is not and I doubt there ever will be a social incentive structure to reward you for it. Indeed, I don’t think you’ll even find a place that’s penalty free! You’re either in this for the intellectual pursuit despite that, or you’re (going to end up) wrong.

  9. I appreciate this call to critical thinking and critical disagreement and think it is something we could use more of in our collective conversations. At the same time, I don’t think this alone would have allowed one to ‘be right’ about the outcome of the election.

    To get better at being right, I think more important than calling upon our own presuppositions about what would enable this (in this case rational, disagreeable discourse), it may be a more fruitful endeavor to look to those who did ‘get it right.’ Or, more clearly, those who weren’t that surprised because they considered Trump to be well-within the realm of possibilities.

    I imagine that critical thinking would be among the tools that this group of people used, but it would not be the only one. I deeply suspect that they would also hold deep understandings of psychology and politics; have an ability to naturally and intuitively think in terms of multifactor scenarios that extended across multipolar spectrums nested inside probability matrixes; and generally took a global and historical (wholistic) approach to understanding culture, values, and society. All of these things come together into a cohesive way of making sense of the world: a worldview.

    If my suspicions are correct, then more than needing more disagreement over agreement (reason over emotion), we need to find ways to help people shift into worldviews such as these. Maturer worldviews that can clearly see more of the environments they are situated in, are less likely to be surprised. Are more likely to be prepared. More able to respond. More responsible.

  10. Now, you’re asking the right questions. Question everything – absolutely everything (Epistemology/ontology). It’s the only way to get anywhere near the truth and learn to face reality as it is, not as we wish it were.

    Warning: Reality is a strange ride. 🙂

  11. So here we are in 2020, and I’m seeing a lot of people (including me for quite a while) saying that the real world seems like fiction. I thought it was fun for a while, and then it got somewhat boring, but maybe there’s an actual and possibly serious problem if the real world doesn’t seem real.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s