Shame. Sorrow.

Let’s take a break from programming for a little bit and change tack. Thank god I’ve got years worth of half-outlined blog posts which have been biding their time, getting ready to strike.

I

Today I want to talk about my greatest shame.

A few years ago some of my rationalist friends and I were out for dinner, talking about cognitive biases and saving the world from itself and all the things we usually talked about. We met up once every week or two, so we were pretty well-rehearsed in these conversations. In particular, we studied cognitive biases. We were familiar with them all, and regularly had a cognitive-bias-of-the-week to study and with which to familiarize ourselves.

As we left dinner on our way to the park to walk around, we came across a woman passed out on the sidewalk. This was exceptionally peculiar, as we lived in a city with no homeless population to speak of, and besides, she looked relatively well-to-do.

We paused in our tracks. None of us knew what to do. We looked at one another. Was this our responsibility? Maybe she was just napping¹. We (shudder) ensured that she was still breathing, and then that was that. We walked away, went on with our conversation, and didn’t think about it again. Personally, I didn’t recall this memory until a few weeks ago.

Yes, it’s my biggest regret that we left that woman there, alone, unconscious on the sidewalk. We could have helped her. Maybe somebody else did, but I doubt it. The reason I doubt it is this: the bystander effect. If you’re not familiar with it, it goes something like this:

People are less likely to help others (or otherwise take action) if they are in a group.

You know that advice you were given as a kid, where if you were in trouble, you should run into a crowded place, where surely – they say – you will be safe. This actually turns out to be pretty bad advice, due to the bystander effect. As a matter of fact, the chance of getting help is inversely related to the number of bystanders present – which is to say, the more crowded a place is, the less likely you are to get help.

My friends and I knew about the bystander effect. It had been a cognitive-bias-of-the-week. We even knew that even when people know about the bystander effect, they still won’t help by default². We knew all of that, and we still didn’t help this woman.

I have no excuses for this behavior. I failed myself, that woman, my art, and a huge number of my personal ideals. Like I said, it’s my biggest point of shame.

I’m not entirely sure what the takeaways are, or should be, from this. Since then I’ve been better, much better, about identifying potential situations in which the bystander effect would take place, and making an effort to do something about it. It’s a start, and it’s progress, but it’s never going to help that woman. I guess there’s something to be said about never letting it happen again, but it’s still a pretty shitty feeling.

Maybe the takeaway is this: knowledge isn’t enough to make a difference. It can help, and it’s a step in the right direction, but it isn’t enough in and of itself. I guess it’s a warning that hubris is still something I need to worry about.

II

The reason all of this unpleasantness resurfaced was due to my having recently read a book on persuasion: Influence by Robert Cialdini. It made me regret having preemptively awarded my Book of the Year award, though I’m sure I can probably find a way to fix that if I try hard enough.

Influence was a book I had to read very, very carefully. Once every few pages I found myself lost in thought, my mind churning, working through all of the implications of what I was reading. The implications I was working through, it might surprise you to hear, were more about its implications to me, rather than how I could go about influencing other people.

What I liked most about Influence was how much it predicted of the psychology I knew. What were once odd bits and ends were finally given some underlying theory. For example, one of the chapters in Influence is on humans’ need for internal consistency. From my notes on the book:

Steven J. Sherman. He simply called a sample of Bloomington, Indiana, residents as part of a survey he was taking and asked them to predict what they would say if asked to spend three hours collecting money for the American Cancer Society. Of course, not wanting to seem uncharitable to the survey taker or to themselves, many of these people said that they would volunteer. The consequence of this sly commitment procedure was a 700 percent increase in volunteers when, a few days later, a representative of the American Cancer Society did call and ask for neighborhood canvassers.

and perhaps, more frighteningly, the slippery slope:

For instance, prisoners were frequently asked to make statements so mildly anti-American or pro-Communist as to seem inconsequential (“The United States is not perfect.” “In a Communist country, unemployment is not a problem.”). But once these minor requests were complied with, the men found themselves pushed to submit to related yet more substantive requests. A man who had just agreed with his Chinese interrogator that the United States is not perfect might then be asked to indicate some of the ways in which he thought this was the case. Once he had so explained himself, he might be asked to make a list of these “problems with America” and to sign his name to it. Later he might be asked to read his list in a discussion group with other prisoners. “After all, it’s what you really believe, isn’t it?” Still later he might be asked to write an essay expanding on his list and discussing these problems in greater detail.

The book paints dozens of other examples of seemingly irrational behavior in humans, tied together only by the desire of internal consistency. Cialdini makes a strong argument that such a need is prevalent enough to be considered ubiquitous. This was new to me, and immediately reconciled anchoring and not proposing solutions until you’ve thought about them for five minutes together, among other things which I regrettably didn’t write down.

The other recurring thought I had during Influence was that it seems like I have some natural defenses to a lot of these tactics. A good way to combat forced reciprocity is to do a lot of ruthless calculations about the people you spend time with. I avoid object-level consistency by staying outside of my comfort zones (which achieves this by being meta-level stable) and always seeking the truth.

It’s kind of interesting that I’ve already written about my tactics for avoiding a lot of compliance techniques. At the time it felt just like things that were useful and/or proper for other people to know.

It’s nice to think that there are secret themes running through my blog posts, ones that I’ll unearth presumably only years after wrapping my mind around these things. Honestly, that’s a big part of why I write; getting thoughts out on the screen makes me flesh them out in greater detail than I would if they remained floating around inside of my head. Also I get feedback on these crazy thoughts from smart people like you.

But I digress.

III

So, if you’re anything like me, you might be wondering in what I have to flesh out by writing about the skeletons in my closet. I’m not entirely sure either, but I have some thoughts. The straightforward answer is that I need to write about something for NaBloWriMo, but that’s phoning it in.

Part of it is that I want to exploit my internal consistency; hopefully by writing about it, by owning up to it publicly, and by signaling to myself that I’ll always call myself out on behavior that violates my ethics. There’s great value here. I think if I can convince myself of this, my past self can wield the proverbial Pavlovian stick against my future. I haven’t yet sorted out how to preserve values across time, but this seems like a road worthy of exploring.

The other major part is that I’m currently reading Ayn Rand’s Atlas Shrugged. Take away from it what you will, but the novel is certainly a warning against societies in which people hesitate to own up to their actions. Because Newcomb-like problems are the norm, my best way to avoid Randian dystopias is to cooperate against things I find horrifying. This is a new principle of mine; I suspect it is due to the larger topic at discussion.

Honestly, here at the bottom of the post, I’m still a little hesitant to actually go ahead and publish this. It’s not necessary the world’s coolest feeling to put your heart on your sleeve like this, but I think it’s the right thing to do. At my new job, there is a culture of writing post mortems whenever you go ahead and break something critical. The idea is twinfold: to figure out why, in depth, the problem happened the first time around, and to disseminate the knowledge to ensure it doesn’t happen again.

So, let’s consider this my post mortem. I think the world would be a better place if we wrote more things like this.

Thanks to Alex Guzey for catching some typos in the published version of this essay.